0s autopkgtest [19:34:10]: starting date and time: 2025-07-07 19:34:10+0000 0s autopkgtest [19:34:10]: git checkout: 508d4a25 a-v-ssh wait_for_ssh: demote "ssh connection failed" to a debug message 0s autopkgtest [19:34:10]: host juju-7f2275-prod-proposed-migration-environment-2; command line: /home/ubuntu/autopkgtest/runner/autopkgtest --output-dir /tmp/autopkgtest-work.sp9g8x8j/out --timeout-copy=6000 --setup-commands /home/ubuntu/autopkgtest-cloud/worker-config-production/setup-canonical.sh --apt-pocket=proposed=src:glibc --apt-upgrade ggml --timeout-short=300 --timeout-copy=20000 --timeout-build=20000 --env=ADT_TEST_TRIGGERS=glibc/2.41-9ubuntu1 -- ssh -s /home/ubuntu/autopkgtest/ssh-setup/nova -- --flavor autopkgtest --security-groups autopkgtest-juju-7f2275-prod-proposed-migration-environment-2@bos03-arm64-10.secgroup --name adt-questing-arm64-ggml-20250707-193409-juju-7f2275-prod-proposed-migration-environment-2-954b271b-a1cd-49e8-988e-09a11edc8a51 --image adt/ubuntu-questing-arm64-server --keyname testbed-juju-7f2275-prod-proposed-migration-environment-2 --net-id=net_prod-proposed-migration -e TERM=linux --mirror=http://ftpmaster.internal/ubuntu/ 3s Creating nova instance adt-questing-arm64-ggml-20250707-193409-juju-7f2275-prod-proposed-migration-environment-2-954b271b-a1cd-49e8-988e-09a11edc8a51 from image adt/ubuntu-questing-arm64-server-20250707.img (UUID 5a1f42c1-3f6f-4084-8e72-afd42fd719a8)... 76s autopkgtest [19:35:26]: testbed dpkg architecture: arm64 76s autopkgtest [19:35:26]: testbed apt version: 3.1.3 76s autopkgtest [19:35:26]: @@@@@@@@@@@@@@@@@@@@ test bed setup 77s autopkgtest [19:35:27]: testbed release detected to be: None 78s autopkgtest [19:35:28]: updating testbed package index (apt update) 78s Get:1 http://ftpmaster.internal/ubuntu questing-proposed InRelease [249 kB] 78s Hit:2 http://ftpmaster.internal/ubuntu questing InRelease 78s Hit:3 http://ftpmaster.internal/ubuntu questing-updates InRelease 78s Hit:4 http://ftpmaster.internal/ubuntu questing-security InRelease 79s Get:5 http://ftpmaster.internal/ubuntu questing-proposed/universe Sources [177 kB] 79s Get:6 http://ftpmaster.internal/ubuntu questing-proposed/multiverse Sources [15.8 kB] 79s Get:7 http://ftpmaster.internal/ubuntu questing-proposed/main Sources [27.5 kB] 79s Get:8 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 Packages [38.5 kB] 79s Get:9 http://ftpmaster.internal/ubuntu questing-proposed/universe arm64 Packages [154 kB] 79s Get:10 http://ftpmaster.internal/ubuntu questing-proposed/multiverse arm64 Packages [15.7 kB] 79s Fetched 677 kB in 1s (761 kB/s) 80s Reading package lists... 81s autopkgtest [19:35:31]: upgrading testbed (apt dist-upgrade and autopurge) 81s Reading package lists... 81s Building dependency tree... 81s Reading state information... 82s Calculating upgrade... 83s The following packages will be upgraded: 83s apparmor libapparmor1 libc-bin libc-dev-bin libc6 libc6-dev libnss-systemd 83s libpam-systemd libsystemd-shared libsystemd0 libudev1 libxml2-16 locales 83s systemd systemd-cryptsetup systemd-resolved systemd-sysv systemd-timesyncd 83s udev 83s 19 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. 83s Need to get 19.1 MB of archives. 83s After this operation, 34.8 kB disk space will be freed. 83s Get:1 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc6-dev arm64 2.41-9ubuntu1 [1749 kB] 83s Get:2 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc-dev-bin arm64 2.41-9ubuntu1 [22.7 kB] 83s Get:3 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc6 arm64 2.41-9ubuntu1 [2919 kB] 84s Get:4 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc-bin arm64 2.41-9ubuntu1 [599 kB] 84s Get:5 http://ftpmaster.internal/ubuntu questing/main arm64 libsystemd0 arm64 257.7-1ubuntu1 [522 kB] 84s Get:6 http://ftpmaster.internal/ubuntu questing/main arm64 libnss-systemd arm64 257.7-1ubuntu1 [170 kB] 84s Get:7 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-sysv arm64 257.7-1ubuntu1 [11.9 kB] 84s Get:8 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-resolved arm64 257.7-1ubuntu1 [324 kB] 84s Get:9 http://ftpmaster.internal/ubuntu questing/main arm64 libpam-systemd arm64 257.7-1ubuntu1 [254 kB] 84s Get:10 http://ftpmaster.internal/ubuntu questing/main arm64 libsystemd-shared arm64 257.7-1ubuntu1 [2239 kB] 84s Get:11 http://ftpmaster.internal/ubuntu questing/main arm64 systemd arm64 257.7-1ubuntu1 [2939 kB] 84s Get:12 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-timesyncd arm64 257.7-1ubuntu1 [41.0 kB] 84s Get:13 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-cryptsetup arm64 257.7-1ubuntu1 [121 kB] 84s Get:14 http://ftpmaster.internal/ubuntu questing/main arm64 udev arm64 257.7-1ubuntu1 [1407 kB] 84s Get:15 http://ftpmaster.internal/ubuntu questing/main arm64 libudev1 arm64 257.7-1ubuntu1 [203 kB] 84s Get:16 http://ftpmaster.internal/ubuntu questing/main arm64 libapparmor1 arm64 4.1.1-0ubuntu3 [56.2 kB] 84s Get:17 http://ftpmaster.internal/ubuntu questing/main arm64 libxml2-16 arm64 2.14.4+dfsg-0exp1 [595 kB] 84s Get:18 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 locales all 2.41-9ubuntu1 [4253 kB] 85s Get:19 http://ftpmaster.internal/ubuntu questing/main arm64 apparmor arm64 4.1.1-0ubuntu3 [666 kB] 85s Preconfiguring packages ... 86s Fetched 19.1 MB in 2s (10.4 MB/s) 86s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 86s Preparing to unpack .../libc6-dev_2.41-9ubuntu1_arm64.deb ... 86s Unpacking libc6-dev:arm64 (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 86s Preparing to unpack .../libc-dev-bin_2.41-9ubuntu1_arm64.deb ... 86s Unpacking libc-dev-bin (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 86s Preparing to unpack .../libc6_2.41-9ubuntu1_arm64.deb ... 87s Unpacking libc6:arm64 (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 87s Setting up libc6:arm64 (2.41-9ubuntu1) ... 87s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 87s Preparing to unpack .../libc-bin_2.41-9ubuntu1_arm64.deb ... 87s Unpacking libc-bin (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 87s Setting up libc-bin (2.41-9ubuntu1) ... 87s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 87s Preparing to unpack .../libsystemd0_257.7-1ubuntu1_arm64.deb ... 87s Unpacking libsystemd0:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 87s Setting up libsystemd0:arm64 (257.7-1ubuntu1) ... 87s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 88s Preparing to unpack .../libnss-systemd_257.7-1ubuntu1_arm64.deb ... 88s Unpacking libnss-systemd:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 88s Preparing to unpack .../systemd-sysv_257.7-1ubuntu1_arm64.deb ... 88s Unpacking systemd-sysv (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 88s Preparing to unpack .../systemd-resolved_257.7-1ubuntu1_arm64.deb ... 88s Unpacking systemd-resolved (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 88s Preparing to unpack .../libpam-systemd_257.7-1ubuntu1_arm64.deb ... 88s Unpacking libpam-systemd:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 88s Preparing to unpack .../libsystemd-shared_257.7-1ubuntu1_arm64.deb ... 88s Unpacking libsystemd-shared:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 88s Setting up libsystemd-shared:arm64 (257.7-1ubuntu1) ... 88s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 88s Preparing to unpack .../systemd_257.7-1ubuntu1_arm64.deb ... 88s Unpacking systemd (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 89s Preparing to unpack .../systemd-timesyncd_257.7-1ubuntu1_arm64.deb ... 89s Unpacking systemd-timesyncd (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 89s Preparing to unpack .../systemd-cryptsetup_257.7-1ubuntu1_arm64.deb ... 89s Unpacking systemd-cryptsetup (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 89s Preparing to unpack .../udev_257.7-1ubuntu1_arm64.deb ... 89s Unpacking udev (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 89s Preparing to unpack .../libudev1_257.7-1ubuntu1_arm64.deb ... 89s Unpacking libudev1:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 89s Setting up libudev1:arm64 (257.7-1ubuntu1) ... 89s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127318 files and directories currently installed.) 89s Preparing to unpack .../libapparmor1_4.1.1-0ubuntu3_arm64.deb ... 89s Unpacking libapparmor1:arm64 (4.1.1-0ubuntu3) over (4.1.1-0ubuntu2) ... 89s Preparing to unpack .../libxml2-16_2.14.4+dfsg-0exp1_arm64.deb ... 89s Unpacking libxml2-16:arm64 (2.14.4+dfsg-0exp1) over (2.14.3+dfsg-0exp3) ... 89s Preparing to unpack .../locales_2.41-9ubuntu1_all.deb ... 89s Unpacking locales (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 89s Preparing to unpack .../apparmor_4.1.1-0ubuntu3_arm64.deb ... 91s Unpacking apparmor (4.1.1-0ubuntu3) over (4.1.1-0ubuntu2) ... 91s Setting up libapparmor1:arm64 (4.1.1-0ubuntu3) ... 91s Setting up libxml2-16:arm64 (2.14.4+dfsg-0exp1) ... 91s Setting up systemd (257.7-1ubuntu1) ... 91s /usr/lib/tmpfiles.d/legacy.conf:14: Duplicate line for path "/run/lock", ignoring. 91s /usr/lib/tmpfiles.d/legacy.conf:14: Duplicate line for path "/run/lock", ignoring. 92s Setting up locales (2.41-9ubuntu1) ... 93s Generating locales (this might take a while)... 96s en_US.UTF-8... done 96s Generation complete. 96s Setting up apparmor (4.1.1-0ubuntu3) ... 96s Installing new version of config file /etc/apparmor.d/lsusb ... 96s Installing new version of config file /etc/apparmor.d/mbsync ... 98s Reloading AppArmor profiles 99s Setting up systemd-cryptsetup (257.7-1ubuntu1) ... 99s Setting up systemd-timesyncd (257.7-1ubuntu1) ... 100s systemd-time-wait-sync.service is a disabled or a static unit not running, not starting it. 100s Setting up udev (257.7-1ubuntu1) ... 101s Creating group 'clock' with GID 988. 101s Setting up libc-dev-bin (2.41-9ubuntu1) ... 101s Setting up systemd-resolved (257.7-1ubuntu1) ... 102s Setting up systemd-sysv (257.7-1ubuntu1) ... 102s Setting up libnss-systemd:arm64 (257.7-1ubuntu1) ... 102s Setting up libpam-systemd:arm64 (257.7-1ubuntu1) ... 102s Setting up libc6-dev:arm64 (2.41-9ubuntu1) ... 102s Processing triggers for libc-bin (2.41-9ubuntu1) ... 102s Processing triggers for man-db (2.13.1-1) ... 104s Processing triggers for dbus (1.16.2-2ubuntu1) ... 104s Processing triggers for shared-mime-info (2.4-5build2) ... 104s Processing triggers for procps (2:4.0.4-8ubuntu2) ... 104s Processing triggers for initramfs-tools (0.149ubuntu1) ... 104s update-initramfs: Generating /boot/initrd.img-6.15.0-3-generic 104s W: No lz4 in /usr/bin:/sbin:/bin, using gzip 106s W: Possible missing firmware /lib/firmware/apple/dfrmtfw-*.bin for built-in driver apple_z2 132s System running in EFI mode, skipping. 132s Reading package lists... 132s Building dependency tree... 132s Reading state information... 132s Solving dependencies... 133s 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. 133s autopkgtest [19:36:23]: rebooting testbed after setup commands that affected boot 163s autopkgtest [19:36:53]: testbed running kernel: Linux 6.15.0-3-generic #3-Ubuntu SMP PREEMPT_DYNAMIC Wed Jun 4 08:41:23 UTC 2025 166s autopkgtest [19:36:56]: @@@@@@@@@@@@@@@@@@@@ apt-source ggml 171s Get:1 http://ftpmaster.internal/ubuntu questing/universe ggml 0.0~git20250618.9e4bee1-1 (dsc) [2567 B] 171s Get:2 http://ftpmaster.internal/ubuntu questing/universe ggml 0.0~git20250618.9e4bee1-1 (tar) [1404 kB] 171s Get:3 http://ftpmaster.internal/ubuntu questing/universe ggml 0.0~git20250618.9e4bee1-1 (diff) [10.2 kB] 171s gpgv: Signature made Fri Jun 20 09:29:08 2025 UTC 171s gpgv: using RSA key 419F7E9A47C3AB9519F1B0A339953A37DE4EB36B 171s gpgv: Can't check signature: No public key 171s dpkg-source: warning: cannot verify inline signature for ./ggml_0.0~git20250618.9e4bee1-1.dsc: no acceptable signature found 171s autopkgtest [19:37:01]: testing package ggml version 0.0~git20250618.9e4bee1-1 175s autopkgtest [19:37:05]: build not needed 178s autopkgtest [19:37:08]: test libggml-cpu: preparing testbed 178s Reading package lists... 178s Building dependency tree... 178s Reading state information... 178s Solving dependencies... 179s The following NEW packages will be installed: 179s libggml libggml-cpu libgomp1 179s 0 upgraded, 3 newly installed, 0 to remove and 0 not upgraded. 179s Need to get 729 kB of archives. 179s After this operation, 3740 kB of additional disk space will be used. 179s Get:1 http://ftpmaster.internal/ubuntu questing/main arm64 libgomp1 arm64 15.1.0-8ubuntu1 [147 kB] 179s Get:2 http://ftpmaster.internal/ubuntu questing/universe arm64 libggml-cpu arm64 0.0~git20250618.9e4bee1-1 [300 kB] 179s Get:3 http://ftpmaster.internal/ubuntu questing/universe arm64 libggml arm64 0.0~git20250618.9e4bee1-1 [282 kB] 180s Fetched 729 kB in 1s (1075 kB/s) 180s Selecting previously unselected package libgomp1:arm64. 181s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127320 files and directories currently installed.) 181s Preparing to unpack .../libgomp1_15.1.0-8ubuntu1_arm64.deb ... 181s Unpacking libgomp1:arm64 (15.1.0-8ubuntu1) ... 181s Selecting previously unselected package libggml-cpu:arm64. 181s Preparing to unpack .../libggml-cpu_0.0~git20250618.9e4bee1-1_arm64.deb ... 181s Unpacking libggml-cpu:arm64 (0.0~git20250618.9e4bee1-1) ... 181s Selecting previously unselected package libggml:arm64. 181s Preparing to unpack .../libggml_0.0~git20250618.9e4bee1-1_arm64.deb ... 181s Unpacking libggml:arm64 (0.0~git20250618.9e4bee1-1) ... 181s Setting up libgomp1:arm64 (15.1.0-8ubuntu1) ... 181s Setting up libggml-cpu:arm64 (0.0~git20250618.9e4bee1-1) ... 181s Setting up libggml:arm64 (0.0~git20250618.9e4bee1-1) ... 181s Processing triggers for libc-bin (2.41-9ubuntu1) ... 182s autopkgtest [19:37:12]: test libggml-cpu: /usr/libexec/ggml/runtests 182s autopkgtest [19:37:12]: test libggml-cpu: [----------------------- 182s dpkg-architecture: warning: cannot determine CC system type, falling back to default (native compilation) 182s 0.00 1.00 2.00 182s Testing 1 devices 182s 182s Backend 1/1: CPU 182s Skipping CPU backend 182s 1/1 backends passed 182s OK 182s main: compute buffer size: 0.19 MB 182s load_model: ggml tensor size = 336 bytes 182s load_model: backend buffer size = 0.09 MB 182s 182s Performing test: 182s ggml_conv_1d_transpose (4): PASSED 182s ggml_conv_1d_transpose (5): PASSED 182s ggml_conv_1d_transpose (10): PASSED 182s ggml_conv_1d_transpose (14): PASSED 182s ggml_conv_1d_transpose (12): PASSED 182s ggml_conv_1d_transpose (18): PASSED 182s ggml_conv_1d_transpose (24): PASSED 182s ggml_conv_1d_transpose (50688): PASSED 182s load_model: ggml tensor size = 336 bytes 182s load_model: backend buffer size = 0.00 MB 182s 182s Performing test: 182s ggml_im2col (240): PASSED 182s ggml_conv1d (80): PASSED 182s load_model: ggml tensor size = 336 bytes 182s load_model: backend buffer size = 0.00 MB 182s 182s Performing test: 182s ggml_im2col (4320): PASSED 182s ggml_conv2d (480): PASSED 182s ggml_conv_2d_dw(channels=3, kernel=1x1, stride=1, pad=0, dilation=1, layout=WHCN): PASSED 182s ggml_conv_2d_dw(channels=3, kernel=1x1, stride=1, pad=0, dilation=1, layout=CWHN): PASSED 182s ggml_conv_2d_dw(channels=42, kernel=3x3, stride=2, pad=1, dilation=1, layout=WHCN): PASSED 182s ggml_conv_2d_dw(channels=42, kernel=3x3, stride=2, pad=1, dilation=1, layout=CWHN): PASSED 182s ggml_conv_2d_dw(channels=8, kernel=5x5, stride=1, pad=2, dilation=2, layout=WHCN): PASSED 182s ggml_conv_2d_dw(channels=8, kernel=5x5, stride=1, pad=2, dilation=2, layout=CWHN): PASSED 182s main: compute buffer size: 0.00 MB 182s main: compute buffer size: 0.02 MB 182s Testing dup on i16 -> i16 copy 182s Testing dup on i32 -> i32 copy 182s Testing dup on f16 -> f16 copy 182s Testing dup on f16 -> f32 copy 182s Testing dup on f32 -> f16 copy 182s Testing dup on f32 -> f32 copy 182s 182s Performing gemm_f16_out_f32 test: 182s 1224.0f,1023.0f,1158.0f,1259.0f,1359.0f,1194.0f,1535.0f,1247.0f,1185.0f,1029.0f,889.0f,1182.0f,955.0f,1179.0f,1147.0f,1048.0f, 182s 1216.0f,1087.0f,1239.0f,1361.0f,1392.0f,1260.0f,1247.0f,1563.0f,1167.0f,1052.0f,942.0f,1214.0f,1045.0f,1134.0f,1264.0f,1126.0f, 182s 1125.0f,966.0f,1079.0f,1333.0f,1287.0f,1101.0f,1185.0f,1167.0f,1368.0f,990.0f,967.0f,1121.0f,971.0f,1086.0f,1130.0f,980.0f, 182s 999.0f,902.0f,1020.0f,1056.0f,1076.0f,929.0f,1029.0f,1052.0f,990.0f,1108.0f,823.0f,989.0f,759.0f,1041.0f,1003.0f,870.0f, 182s gemm_mult (64): PASSED 182s load_model: ggml tensor size = 336 bytes 182s load_model: backend buffer size = 3584 bytes 182s Matrix A: [36, 4] 182s Matrix B: [36, 16] 182s 182s Performing ggml_mul_mat test: 182s 1224.0 1023.0 1158.0 1259.0 1359.0 1194.0 1535.0 1247.0 1185.0 1029.0 889.0 1182.0 955.0 1179.0 1147.0 1048.0 182s 1216.0 1087.0 1239.0 1361.0 1392.0 1260.0 1247.0 1563.0 1167.0 1052.0 942.0 1214.0 1045.0 1134.0 1264.0 1126.0 182s 1125.0 966.0 1079.0 1333.0 1287.0 1101.0 1185.0 1167.0 1368.0 990.0 967.0 1121.0 971.0 1086.0 1130.0 980.0 182s 999.0 902.0 1020.0 1056.0 1076.0 929.0 1029.0 1052.0 990.0 1108.0 823.0 989.0 759.0 1041.0 1003.0 870.0 182s ggml_mul_mat (64): PASSED 182s main: compute buffer size: 0.00 MB 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s test-mul-mat0: iter:0/500 182s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 1, 1, 1] 182s 0.435 -0.717 0.214 -0.967 182s 182s x1: [4, 3, 1, 1] 182s 0.823 -0.605 -0.330 0.536 182s -0.444 0.108 -0.045 0.258 182s -0.270 0.027 0.904 0.832 182s 182s y: [1, 3, 1, 1] 182s 0.202 182s -0.530 182s -0.748 182s testing: mul_mat, [4, 1, 2, 1] = [4, 4, 2, 1] * [4, 1, 2, 1] 182s x0: [4, 4, 2, 1] 182s 0.026 0.678 0.225 -0.408 182s 0.275 0.049 -0.013 0.946 182s -0.415 0.543 0.053 0.540 182s -0.200 0.783 -0.433 -0.295 182s 182s x1: [4, 1, 2, 1] 182s -0.514 -0.726 0.608 -0.687 182s 182s y: [4, 1, 2, 1] 182s -0.088 -0.834 -0.519 -0.527 182s testing: mul_mat, [2, 4, 2, 4] = [4, 2, 2, 4] * [4, 4, 2, 4] 182s x0: [4, 2, 2, 4] 182s 0.755 0.496 0.258 -0.929 182s 0.496 0.666 0.851 0.747 182s 182s x1: [4, 4, 2, 4] 182s 0.804 0.702 -0.467 0.080 182s -0.250 0.520 0.025 0.335 182s 0.063 -0.921 -0.125 0.864 182s 0.862 0.442 -0.431 0.477 182s 182s y: [2, 4, 2, 4] 182s 0.761 0.529 182s -0.236 0.495 182s -1.244 -0.044 182s 0.315 0.711 182s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 182s x0: [4, 1, 1, 1] 182s 0.065 -0.825 -0.479 0.755 182s 182s x1: [4, 2, 1, 1] 182s 0.254 0.461 -0.343 0.481 182s -0.596 0.842 0.370 0.306 182s 182s y: [1, 2, 1, 1] 182s 0.163 182s -0.679 182s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 182s x0: [1, 1, 2, 1] 182s -0.423 182s 182s x1: [1, 4, 2, 1] 182s 0.372 182s -0.813 182s -0.777 182s -0.277 182s 182s y: [1, 4, 2, 1] 182s -0.158 182s 0.344 182s 0.329 182s 0.117 182s testing: mul_mat, [2, 1, 2, 4] = [1, 2, 2, 4] * [1, 1, 2, 4] 182s x0: [1, 2, 2, 4] 182s 0.530 182s 0.398 182s 182s x1: [1, 1, 2, 4] 182s -0.620 182s 182s y: [2, 1, 2, 4] 182s -0.329 -0.247 182s test-mul-mat0: iter:1/500 182s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 1, 1, 1] 182s -0.418 182s 182s x1: [1, 4, 1, 1] 182s 0.729 182s -0.015 182s -0.073 182s 0.698 182s 182s y: [1, 4, 1, 1] 182s -0.305 182s 0.006 182s 0.030 182s -0.292 182s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 182s x0: [1, 1, 2, 1] 182s -0.722 182s 182s x1: [1, 1, 2, 1] 182s -0.639 182s 182s y: [1, 1, 2, 1] 182s 0.461 182s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 182s x0: [1, 2, 2, 1] 182s -0.644 182s -0.556 182s 182s x1: [1, 1, 2, 1] 182s -0.015 182s 182s y: [2, 1, 2, 1] 182s 0.010 0.008 182s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 2, 1, 1] 182s 0.864 182s 0.817 182s 182s x1: [1, 2, 1, 1] 182s -0.724 182s -0.279 182s 182s y: [2, 2, 1, 1] 182s -0.625 -0.591 182s -0.241 -0.228 182s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 182s x0: [2, 1, 2, 1] 182s -0.211 0.318 182s 182s x1: [2, 1, 2, 1] 182s 0.244 0.674 182s 182s y: [1, 1, 2, 1] 182s 0.162 182s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 182s x0: [1, 1, 2, 1] 182s -0.423 182s 182s x1: [1, 2, 2, 1] 182s -0.698 182s -0.855 182s 182s y: [1, 2, 2, 1] 182s 0.295 182s 0.362 182s test-mul-mat0: iter:2/500 182s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 182s x0: [1, 2, 1, 1] 182s 0.524 182s -0.025 182s 182s x1: [1, 1, 1, 1] 182s -0.469 182s 182s y: [2, 1, 1, 1] 182s -0.246 0.012 182s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 3, 1, 1] 182s 0.035 182s -0.584 182s 0.115 182s 182s x1: [1, 2, 1, 1] 182s -0.685 182s 0.766 182s 182s y: [3, 2, 1, 1] 182s -0.024 0.401 -0.079 182s 0.027 -0.448 0.088 182s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 2, 1, 1] 182s -0.348 182s 0.459 182s 182s x1: [1, 3, 1, 1] 182s -0.148 182s 0.660 182s -0.211 182s 182s y: [2, 3, 1, 1] 182s 0.051 -0.068 182s -0.230 0.303 182s 0.073 -0.097 182s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 4, 1, 1] 182s 0.795 182s -0.728 182s -0.178 182s -0.989 182s 182s x1: [1, 2, 1, 1] 182s 0.277 182s 0.970 182s 182s y: [4, 2, 1, 1] 182s 0.220 -0.202 -0.049 -0.274 182s 0.771 -0.706 -0.173 -0.959 182s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 182s x0: [4, 1, 1, 1] 182s 0.442 -0.902 -0.102 0.973 182s 182s x1: [4, 1, 1, 1] 182s 0.567 0.549 -0.413 -0.771 182s 182s y: [1, 1, 1, 1] 182s -0.952 182s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 2, 1, 1] 182s -0.801 182s -0.234 182s 182s x1: [1, 4, 1, 1] 182s 0.416 182s -0.578 182s -0.052 182s 0.730 182s 182s y: [2, 4, 1, 1] 182s -0.333 -0.097 182s 0.463 0.135 182s 0.042 0.012 182s -0.585 -0.171 182s test-mul-mat0: iter:3/500 182s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 182s x0: [3, 1, 1, 1] 182s 0.777 -0.047 -0.660 182s 182s x1: [3, 4, 1, 1] 182s -0.897 -0.893 -0.085 182s 0.562 0.384 -0.115 182s -0.762 0.179 0.157 182s 0.060 0.190 -0.276 182s 182s y: [1, 4, 1, 1] 182s -0.600 182s 0.495 182s -0.704 182s 0.220 182s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 182s x0: [3, 1, 1, 1] 182s -0.533 0.660 -0.860 182s 182s x1: [3, 1, 1, 1] 182s 0.219 0.051 0.238 182s 182s y: [1, 1, 1, 1] 182s -0.287 182s testing: mul_ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s mat, [4, 1, 1, 3] = [3, 4, 1, 3] * [3, 1, 1, 3] 182s x0: [3, 4, 1, 3] 182s 0.630 0.318 -0.927 182s -0.485 0.557 0.252 182s 0.672 -0.384 -0.558 182s -0.604 0.225 -0.781 182s 182s x1: [3, 1, 1, 3] 182s -0.802 0.847 -0.661 182s 182s y: [4, 1, 1, 3] 182s 0.377 0.694 -0.496 1.191 182s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 182s x0: [3, 1, 1, 1] 182s -0.694 -0.934 -0.873 182s 182s x1: [3, 4, 1, 1] 182s -0.376 0.292 -0.988 182s 0.066 0.688 0.237 182s 0.285 0.037 -0.199 182s -0.276 0.438 0.604 182s 182s y: [1, 4, 1, 1] 182s 0.851 182s -0.895 182s -0.059 182s -0.745 182s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 3, 1, 1] 182s 0.136 182s -0.998 182s -0.989 182s 182s x1: [1, 3, 1, 1] 182s 0.371 182s -0.625 182s 0.238 182s 182s y: [3, 3, 1, 1] 182s 0.050 -0.371 -0.367 182s -0.085 0.623 0.618 182s 0.032 -0.237 -0.235 182s testing: mul_mat, [1, 1, 1, 3] = [3, 1, 1, 3] * [3, 1, 1, 3] 182s x0: [3, 1, 1, 3] 182s 0.770 -0.627 -0.686 182s 182s x1: [3, 1, 1, 3] 182s -0.390 -0.477 0.311 182s 182s y: [1, 1, 1, 3] 182s -0.214 182s test-mul-mat0: iter:4/500 182s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 182s x0: [2, 3, 1, 1] 182s -0.011 0.105 182s 0.878 0.599 182s 0.628 0.189 182s 182s x1: [2, 2, 1, 1] 182s 0.865 -0.534 182s 0.853 0.103 182s 182s y: [3, 2, 1, 1] 182s -0.066 0.440 0.443 182s 0.001 0.811 0.555 182s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 182s x0: [2, 3, 2, 1] 182s -0.666 0.591 182s 0.730 0.746 182s 0.329 -0.175 182s 182s x1: [2, 3, 2, 1] 182s 0.314 0.991 182s 0.872 -0.351 182s 0.749 0.178 182s 182s y: [3, 3, 2, 1] 182s 0.376 0.969 -0.070 182s -0.788 0.375 0.348 182s -0.393 0.680 0.215 182s testing: mul_mat, [4, 3, 2, 3] = [2, 4, 2, 3] * [2, 3, 2, 3] 182s x0: [2, 4, 2, 3] 182s 0.905 0.898 182s 0.798 0.534 182s -0.333 0.073 182s -0.562 -0.045 182s 182s x1: [2, 3, 2, 3] 182s -0.934 0.402 182s 0.036 0.665 182s 0.030 -0.775 182s 182s y: [4, 3, 2, 3] 182s -0.485 -0.531 0.340 0.507 182s 0.630 0.384 0.037 -0.050 182s -0.668 -0.390 -0.067 0.018 182s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 1, 1, 1] 182s 0.931 -0.689 182s 182s x1: [2, 4, 1, 1] 182s 0.203 0.541 182s 0.774 0.867 182s -0.654 -0.104 182s -0.025 0.590 182s 182s y: [1, 4, 1, 1] 182s -0.184 182s 0.124 182s -0.537 182s -0.430 182s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 182s x0: [1, 3, 2, 1] 182s -0.048 182s -0.454 182s 0.893 182s 182s x1: [1, 2, 2, 1] 182s -0.414 182s 0.764 182s 182s y: [3, 2, 2, 1] 182s 0.020 0.188 -0.370 182s -0.037 -0.347 0.683 182s testing: mul_mat, [2, 1, 2, 3] = [3, 2, 2, 3] * [3, 1, 2, 3] 182s x0: [3, 2, 2, 3] 182s -0.859 -0.007 -0.414 182s -0.213 -0.127 -0.512 182s 182s x1: [3, 1, 2, 3] 182s 0.437 0.849 -0.632 182s 182s y: [2, 1, 2, 3] 182s -0.120 0.123 182s test-mul-mat0: iter:5/500 182s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 182s x0: [4, 1, 1, 1] 182s 0.174 -0.662 0.169 -0.047 182s 182s x1: [4, 2, 1, 1] 182s -0.728 0.178 -0.884 0.779 182s 0.891 -0.888 0.850 -0.062 182s 182s y: [1, 2, 1, 1] 182s -0.431 182s 0.890 182s testing: mul_mat, [4, 1, 4, 1] = [4, 4, 4, 1] * [4, 1, 4, 1] 182s x0: [4, 4, 4, 1] 182s 0.290 -0.113 -0.462 0.406 182s -0.334 -0.571 0.518 -0.484 182s 0.367 -0.968 0.690 0.705 182s 0.202 -0.357 0.336 0.054 182s 182s x1: [4, 1, 4, 1] 182s 0.631 0.852 0.053 0.165 182s 182s y: [4, 1, 4, 1] 182s 0.129 -0.750 -0.440 -0.150 182s testing: mul_mat, [4, 4, 4, 2] = [4, 4, 4, 2] * [4, 4, 4, 2] 182s x0: [4, 4, 4, 2] 182s -0.642 -0.280 -0.114 -0.243 182s 0.295 -0.799 -0.349 0.739 182s 0.215 -0.792 0.612 0.499 182s -0.202 -0.266 -0.212 -0.456 182s 182s x1: [4, 4, 4, 2] 182s -0.521 0.619 -0.810 0.493 182s -0.446 -0.653 0.875 0.522 182s -0.807 0.962 0.691 -0.317 182s 0.385 -0.087 -0.131 0.308 182s 182s y: [4, 4, 4, 2] 182s 0.134 -0.002 -0.851 -0.113 182s 0.243 0.471 1.217 -0.159 182s 0.246 -1.481 -0.671 -0.095 182s -0.283 0.456 0.225 -0.167 182s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 182s x0: [4, 2, 1, 1] 182s 0.843 -0.821 0.937 -0.991 182s 0.857 0.642 0.018 -0.624 182s 182s x1: [4, 4, 1, 1] 182s 0.204 0.671 0.127 -0.596 182s 0.606 0.345 -0.857 0.925 182s -0.050 -0.231 -0.284 0.862 182s 0.834 -0.794 0.802 0.751 182s 182s y: [2, 4, 1, 1] 182s 0.329 0.979 182s -1.492 0.150 182s -0.971 -0.733 182s 1.361 -0.249 182s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 182s x0: [2, 1, 4, 1] 182s -0.704 0.851 182s 182s x1: [2, 4, 4, 1] 182s -0.426 0.255 182s -0.476 0.497 182s -0.927 0.444 182s -0.299 0.744 182s 182s y: [1, 4, 4, 1] 182s 0.516 182s 0.758 182s 1.030 182s 0.843 182s testing: mul_mat, [3, 2, 4, 2] = [1, 3, 4, 2] * [1, 2, 4, 2] 182s x0: [1, 3, 4, 2] 182s 0.231 182s 0.035 182s -0.242 182s 182s x1: [1, 2, 4, 2] 182s -0.554 182s -0.774 182s 182s y: [3, 2, 4, 2] 182s -0.128 -0.019 0.134 182s -0.179 -0.027 0.188 182s test-mul-mat0ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s : iter:6/500 182s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 182s x0: [2, 1, 1, 1] 182s 0.047 -0.506 182s 182s x1: [2, 2, 1, 1] 182s -0.741 -0.974 182s -0.710 0.490 182s 182s y: [1, 2, 1, 1] 182s 0.458 182s -0.282 182s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 182s x0: [2, 2, 1, 1] 182s 0.688 -0.529 182s -0.166 0.479 182s 182s x1: [2, 1, 1, 1] 182s -0.551 0.083 182s 182s y: [2, 1, 1, 1] 182s -0.423 0.131 182s testing: mul_mat, [4, 2, 1, 2] = [2, 4, 1, 2] * [2, 2, 1, 2] 182s x0: [2, 4, 1, 2] 182s -0.103 -0.083 182s -0.395 0.173 182s 0.752 0.030 182s 0.944 0.308 182s 182s x1: [2, 2, 1, 2] 182s -0.046 -0.815 182s -0.073 0.882 182s 182s y: [4, 2, 1, 2] 182s 0.073 -0.123 -0.059 -0.294 182s -0.066 0.182 -0.029 0.202 182s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 2, 1, 1] 182s 0.035 -0.145 182s 0.848 0.568 182s 182s x1: [2, 4, 1, 1] 182s 0.728 0.067 182s 0.169 -0.801 182s 0.902 -0.352 182s 0.153 -0.913 182s 182s y: [2, 4, 1, 1] 182s 0.016 0.656 182s 0.122 -0.312 182s 0.083 0.565 182s 0.138 -0.389 182s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 182s x0: [2, 4, 1, 1] 182s 0.638 -0.746 182s -0.054 -0.666 182s 0.045 0.552 182s 0.582 0.851 182s 182s x1: [2, 2, 1, 1] 182s -0.722 0.411 182s -0.535 0.194 182s 182s y: [4, 2, 1, 1] 182s -0.767 -0.234 0.194 -0.071 182s -0.486 -0.100 0.083 -0.146 182s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 182s x0: [4, 4, 1, 2] 182s -0.820 -0.237 -0.220 0.417 182s 0.569 0.735 0.865 0.814 182s 0.719 -0.716 0.963 -0.772 182s -0.355 0.709 0.115 -1.000 182s 182s x1: [4, 2, 1, 2] 182s 0.114 -0.417 -0.504 -0.613 182s -0.937 -0.776 0.455 0.232 182s 182s y: [4, 2, 1, 2] 182s -0.139 -1.177 0.368 0.219 182s 0.948 -0.521 0.141 -0.398 182s test-mul-mat0: iter:7/500 182s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 182s x0: [2, 2, 1, 1] 182s 0.538 0.979 182s -0.487 0.538 182s 182s x1: [2, 3, 1, 1] 182s -0.610 -0.745 182s -0.798 0.170 182s -0.879 -0.835 182s 182s y: [2, 3, 1, 1] 182s -1.057 -0.104 182s -0.263 0.480 182s -1.290 -0.021 182s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 182s x0: [2, 4, 2, 1] 182s -0.685 0.494 182s -0.301 0.461 182s 0.655 0.635 182s -0.212 0.385 182s 182s x1: [2, 2, 2, 1] 182s -0.711 0.129 182s 0.602 -0.177 182s 182s y: [4, 2, 2, 1] 182s 0.551 0.273 -0.384 0.200 182s -0.499 -0.263 0.282 -0.196 182s testing: mul_mat, [2, 4, 2, 4] = [2, 2, 2, 4] * [2, 4, 2, 4] 182s x0: [2, 2, 2, 4] 182s -0.939 0.612 182s -0.225 0.137 182s 182s x1: [2, 4, 2, 4] 182s -0.755 -0.865 182s -0.326 -0.776 182s -0.352 -0.787 182s -0.487 0.777 182s 182s y: [2, 4, 2, 4] 182s 0.179 0.052 182s -0.169 -0.033 182s -0.152 -0.029 182s 0.933 0.216 182s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 182s x0: [2, 2, 1, 1] 182s 0.270 0.090 182s 0.194 -0.650 182s 182s x1: [2, 2, 1, 1] 182s 0.498 -0.052 182s 0.721 0.609 182s 182s y: [2, 2, 1, 1] 182s 0.130 0.130 182s 0.250 -0.256 182s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 182s x0: [2, 4, 2, 1] 182s 0.226 0.064 182s -0.114 0.987 182s 0.138 -0.882 182s 0.092 -0.940 182s 182s x1: [2, 2, 2, 1] 182s 0.853 0.949 182s -0.609 -0.320 182s 182s y: [4, 2, 2, 1] 182s 0.254 0.839 -0.719 -0.813 182s -0.158 -0.246 0.198 0.244 182s testing: mul_mat, [2, 2, 2, 4] = [4, 2, 2, 4] * [4, 2, 2, 4] 182s x0: [4, 2, 2, 4] 182s 0.218 0.277 -0.239 0.011 182s 0.639 -0.276 0.482 0.000 182s 182s x1: [4, 2, 2, 4] 182s 0.054 0.003 -0.782 -0.752 182s -0.907 -0.432 -0.899 -0.958 182s 182s y: [2, 2, 2, 4] 182s 0.192 -0.343 182s -0.113 -0.894 182s test-mul-mat0: iter:8/500 182s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 4, 1, 1] 182s -0.236 0.357 0.853 0.689 182s 0.244 -0.226 0.368 -0.607 182s -0.700 0.613 0.360 -0.539 182s 0.642 0.786 -0.464 -0.819 182s 182s x1: [4, 3, 1, 1] 182s -0.942 0.863 0.800 -0.918 182s 0.023 0.488 -0.465 -0.347 182s 0.065 -0.130 0.936 0.159 182s 182s y: [4, 3, 1, 1] 182s 0.580 0.427 1.971 0.454 182s -0.467 -0.065 0.302 0.898 182s 0.846 0.293 0.126 -0.624 182s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 182s x0: [4, 1, 3, 1] 182s -0.407 0.816 -0.495 -0.683 182s 182s x1: [4, 4, 3, 1] 182s -0.570 -0.994 -0.761 -0.708 182s 0.807 -0.680 0.315 0.295 182s -0.145 0.968 -0.640 0.726 182s 0.904 0.519 0.908 -0.332 182s 182s y: [1, 4, 3, 1] 182s 0.281 182s -1.241 182s 0.670 182s -0.167 182s testing: mul_mat, [2, 1, 3, 1] = [4, 2, 3, 1] * [4, 1, 3, 1] 182s x0: [4, 2, 3, 1] 182s 0.261 0.015 0.994 0.497 182s -0.289 0.964 -0.910 -0.473 182s 182s x1: [4, 1, 3, 1] 182s 0.793 0.951 -0.692 0.192 182s 182s y: [2, 1, 3, 1] 182s -0.371 1.226 182s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 182s x0: [4, 4, 1, 1] 182s 0.855 0.772 0.375 -0.457 182s 0.682 -0.482 -0.903 -0.065 182s 0.991 -0.712 -0.760 0.356 182s 0.819 -0.821 -0.565 0.457 182s 182s x1: [4, 2, 1, 1] 182s -0.409 0.232 -0.930 -0.924 182s -0.076 -0.735 0.861 -0.815 182s 182s y: [4, 2, 1, 1] 182s -0.098 0.508 -0.194 -0.423 182s 0.063 -0.422 -0.497 -0.319 182s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 182s x0: [4, 3, 3, 1] 182s -0.008 -0.415 -0.724 -0.490 182s 0.517 -0.077 -0.429 0.153 182s 0.254 -0.390 -0.600 0.546 182s 182s x1: [4, 4, 3, 1] 182s 0.423 -0.772 -0.321 0.103 182s 0.432 0.855 0.693 -0.336 182s 0.925 0.769 0.588 -0.810 182s 0.630 0.773 0.471 0.486 182s 182s y: [3, 4, 3, 1] 182s 0.498 0.431 0.657 182s -0.696 -0.191 -0.823 182s -0.356 0.042 -0.860 182s -0.905 0.139 -0.159 182s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 182s x0: [3, 2, 3, 1] 182s 0.883 -0.871 -0.332 182s -0.993 -0.009 0.099 182s 182s x1: [3, 4, 3, 1] 182s -0.758 0.377 -0.608 182s -0.358 0.887 0.545 182s -0.812 -0.911 -0.296 182s 0.108 -0.339 -0.030 182s 182s y: [2, 4, 3, 1] 182s -0.796 0.690 182s -1.271 0.402 182s 0.175 0.786 182s 0.401 -0.107 182s test-mul-mat0: iter:9/500 182s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 182s x0: [3, 4, 1, 1] 182s 0.515 -0.231 0.660 182s 0.522 -0.102 -0.349 182s -0.810 0.996 -0.522 182s -0.683 0.132 -0.661 182s 182s x1: [3, 2, 1, 1] 182s 0.029 0.566 0.464 182s -0.917 0.597 0.344 182s 182s y: [4, 2, 1, 1] 182s 0.191 -0.205 0.298 -0.252 182s -0.383 -0.660 1.158 0.478 182s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 182s x0: [3, 2, 2, 1] 182s -0.073 -0.203 -0.981 182s 0.243 0.928 -0.642 182s 182s x1: [3, 4, 2, 1] 182s -0.148 0.942 -0.505 182s 0.086 -0.735 -0.962 182s -0.876 0.624 -0.652 182s -0.659 -0.463 -0.675 182s 182s y: [2, 4, 2, 1] 182s 0.314 1.162 182s 1.087 -0.044 182s 0.576 0.784 182s 0.805 -0.158 182s testing: mul_mat, [4, 2, 2, 1] = [3, 4, 2, 1] * [3, 2, 2, 1] 182s x0: [3, 4, 2, 1] 182s 0.314 -0.609 -0.412 182s -0.949 0.191 -0.871 182s -0.023 0.988 -0.852 182s -0.779 0.916 -0.493 182s 182s x1: [3, 2, 2, 1] 182s 0.305 0.760 0.240 182s 0.646 -0.704 0.565 182s 182s y: [4, 2, 2, 1] 182s -0.466 -0.354 0.539 0.340 182s 0.398 -1.240 -1.191 -1.427 182s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 182s x0: [3, 2, 1, 1] 182s -0.182 -0.505 -0.615 182s -0.531 0.039 0.001 182s 182s x1: [3, 4, 1, 1] 182s 0.464 -0.622 -0.588 182s 0.932 -0.524 -0.100 182s 0.630 0.790 0.291 182s -0.782 0.841 -0.518 182s 182s y: [2, 4, 1, 1] 182s 0.592 -0.271 182s 0.157 -0.515 182s -0.692 -0.303 182s 0.036 0.447 182s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 182s x0: [2, 3, 2, 1] 182s -0.868 -0.392 182s 0.575 -0.524 182s 0.499 0.129 182s 182s x1: [2, 3, 2, 1] 182s 0.355 -0.828 182s -0.638 -0.147 182s -0.906 -0.386 182s 182s y: [3, 3, 2, 1] 182s 0.016 0.639 0.071 182s 0.612 -0.290 -0.338 182s 0.938 -0.319 -0.502 182s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 182s x0: [3, 2, 2, 1] 182s 0.146 -0.936 0.434 182s -0.729 -0.100 0.632 182s 182s x1: [3, 2, 2, 1] 182s 0.058 -0.282 -0.812 182s -0.903 0.103 0.190 182s 182s y: [2, 2, 2, 1] 182s -0.080 -0.527 182s -0.146 0.768 182s test-mul-mat0: iter:10/500 182s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 182s x0: [1, 2, 1, 1] 182s 0.335 182s -0.781 182s 182s x1: [1, 1, 1, 1] 182s 0.501 182s 182s y: [2, 1, 1, 1] 182s 0.168 -0.391 182s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 182s x0: [1, 1, 2, 1] 182s -0.403 182s 182s x1: [1, 2, 2, 1] 182s 0.488 182s 0.432 182s 182s y: [1, 2, 2, 1] 182s -0.197 182s -0.174 182s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 182s x0: [1, 2, 2, 1] 182s 0.150 182s 0.113 182s 182s x1: [1, 1, 2, 1] 182s 0.190 182s 182s y: [2, 1, 2, 1] 182s 0.029 0.021 182s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 4, 1, 1] 182s 0.044 182s 0.110 182s 0.978 182s -0.317 182s 182s x1: [1, 2, 1, 1] 182s -0.988 182s -0.900 182s 182s y: [4, 2, 1, 1] 182s -0.043 -0.108 -0.966 0.313 182s -0.039 -0.099 -0.880 0.285 182s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 182s x0: [4, 1, 2, 1] 182s 0.567 -0.366 0.766 -0.111 182s 182s x1: [4, 1, 2, 1] 182s -0.405 -0.762 -0.279 0.848 182s 182s y: [1, 1, 2, 1] 182s -0.258 182s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 182s x0: [1, 4, 2, 1] 182s 0.052 182s 0.622 182s 0.810 182s 0.030 182s 182s x1: [1, 4, 2, 1] 182s 0.840 182s 0.453 182s 0.935 182s -0.009 182s 182s y: [4, 4, 2, 1] 182s 0.044 0.522 0.681 0.025 182s 0.024 0.282 0.367 0.014 182s 0.049 0.581 0.758 0.028 182s -0.000 -0.006 -0.008 -0.000 182s test-mul-mat0: iter:11/500 182s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 2, 1, 1] 182s -0.943 182s -0.795 182s 182s x1: [1, 4, 1, 1] 182s -0.741 182s -0.642 182s 0.168 182s -0.107 182s 182s y: [2, 4, 1, 1] 182s 0.698 0.589 182s 0.605 0.510 182s -0.159 -0.134 182s 0.101 0.085 182s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 182s x0: [1, 3, 3, 1] 182s 0.562 182s -0.719 182s 0.811 182s 182s x1: [1, 2, 3, 1] 182s -0.225 182s 0.543 182s 182s y: [3, 2, 3, 1] 182s -0.127 0.162 -0.183 182s 0.305 -0.390 0.440 182s testing: mul_mat, [4, 3, 3, 2] = [1, 4, 3, 2] * [1, 3, 3, 2] 182s x0: [1, 4, 3, 2] 182s -0.183 182s 0.060 182s -0.406 182s -0.621 182s 182s x1: [1, 3, 3, 2] 182s -0.934 182s 0.198 182s 0.162 182s 182s y: [4, 3, 3, 2] 182s 0.171 -0.056 0.379 0.580 182s -0.036 0.012 -0.080 -0.123 182s -0.030 0.010 -0.066 -0.100 182s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 1, 1, 1] 182s 0.574 182s 182s x1: [1, 4, 1, 1] 182s 0.630 182s 0.313 182s -0.805 182s -0.996 182s 182s y: [1, 4, 1, 1] 182s 0.362 182s 0.180 182s -0.462 182s -0.572 182s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 182s x0: [1, 2, 3, 1] 182s -0.671 182s 0.975 182s 182s x1: [1, 1, 3, 1] 182s -0.007 182s 182s y: [2, 1, 3, 1] 182s 0.005 -0.007 182s testing: mul_mat, [3, 1, 3, 2] = [2, 3, 3, 2] * [2, 1, 3, 2] 182s x0: [2, 3, 3, 2] 182s 0.340 -0.347 182s 0.035 0.230 182s 0.130 0.135 182s 182s x1: [2, 1, 3, 2] 182s 0.128 0.413 182s 182s y: [3, 1, 3, 2] 182s -0.100 0.099 0.072 182s test-mul-mat0: iter:12/500 182s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 182s x0: [4, 2, 1, 1] 182s 0.152 0.027 -0.478 0.091 182s -0.574 -0.496 -0.936 -0.044 182s 182s x1: [4, 4, 1, 1] 182s -0.841 -0.556 0.506 -0.775 182s 0.857 0.212 0.139 0.568 182s -0.824 0.181 -0.308 0.561 182s -0.302 0.541 0.686 0.128 182s 182s y: [2, 4, 1, 1] 182s -0.455 0.319 182s 0.121 -0.752 182s 0.078 0.646 182s -0.348 -0.743 182s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 182s x0: [4, 2, 1, 1] 182s 0.944 0.575 0.129 0.083 182s 0.144 0.306 -0.736 0.835 182s 182s x1: [4, 2, 1, 1] 182s 0.139 -0.733 -0.380 0.561 182s 0.274 -0.588 0.719 0.718 182s 182s y: [2, 2, 1, 1] 182s -0.293 0.544 182s 0.072 -0.070 182s testing: mul_mat, [1, 2, 1, 4] = [4, 1, 1, 4] * [4, 2, 1, 4] 182s x0: [4, 1, 1, 4] 182s -0.950 0.302 0.044 0.306 182s 182s x1: [4, 2, 1, 4] 182s -0.134 -0.038 0.377 -0.447 182s -0.910 0.256 0.705 0.117 182s 182s y: [1, 2, 1, 4] 182s -0.005 182s 1.009 182s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 182s x0: [4, 1, 1, 1] 182s 0.119 -0.999 -0.125 -0.152 182s 182s x1: [4, 1, 1, 1] 182s -0.812 -0.149 -0.527 -0.693 182s 182s y: [1, 1, 1, 1] 182s 0.224 182s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 1, 1, 1] 182s 0.795 182s 182s x1: [1, 4, 1, 1] 182s -0.162 182s -0.242 182s -0.160 182s 0.511 182s 182s y: [1, 4, 1, 1] 182s -0.129 182s -0.192 182s -0.127 182s 0.407 182s testing: mul_mat, [4, 1, 1, 4] = [1, 4, 1, 4] * [1, 1, 1, 4] 182s x0: [1, 4, 1, 4] 182s 0.000 182s 0.952 182s 0.616 182s -0.296 182s 182s x1: [1, 1, 1, 4] 182s 0.561 182s 182s y: [4, 1, 1, 4] 182s 0.000 0.534 0.346 -0.166 182s test-mul-mat0: iter:13/500 182s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 3, 1, 1] 182s 0.138 -0.208 0.291 182s -0.245 0.496 -0.390 182s 0.032 -0.598 -0.283 182s 182s x1: [3, 3, 1, 1] 182s 0.400 0.009 0.158 182s -0.089 -0.484 -0.047 182s -0.528 0.324 0.792 182s 182s y: [3, 3, 1, 1] 182s 0.099 -0.155 -0.037 182s 0.075 -0.200 0.300 182s 0.090 -0.019 -0.435 182s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 3, 1, 1] 182s 0.332 -0.266 0.873 182s -0.660 0.891 -0.216 182s -0.144 -0.156 0.256 182s 182s x1: [3, 3, 1, 1] 182s 0.098 -0.372 -0.404 182s 0.352 0.705 -0.053 182s 0.912 0.081 -0.584 182s 182s y: [3, 3, 1, 1] 182s -0.221 -0.308 -0.059 182s -0.118 0.408 -0.174 182s -0.228 -0.403 -0.293 182s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 4, 1, 1] 182s 0.326 0.214 -0.797 182s 0.954 0.810 0.555 182s 0.659 -0.243 0.467 182s -0.260 0.173 0.940 182s 182s x1: [3, 3, 1, 1] 182s -0.820 -0.364 -0.405 182s 0.318 0.428 0.887 182s -0.927 -0.076 -0.503 182s 182s y: [4, 3, 1, 1] 182s -0.023 -1.302 -0.641 -0.230 182s -0.511 1.142 0.519 0.824 182s 0.083 -1.225 -0.827 -0.245 182s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 182s x0: [3, 4, 1, 1] 182s -0.294 -0.170 -0.852 182s -0.666 0.944 -0.090 182s -0.161 -0.044 0.854 182s 0.630 -0.955 0.702 182s 182s x1: [3, 4, 1, 1] 182s -0.928 0.907 0.812 182s -0.588 0.798 -0.404 182s 0.268 -0.358 0.852 182s 0.448 0.278 -0.553 182s 182s y: [4, 4, 1, 1] 182s -0.573 1.402 0.803 -0.880 182s 0.382 1.182 -0.286 -1.416 182s -0.744 -0.594 0.700 1.109 182s 0.293 0.014 -0.557 -0.372 182s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 2, 1, 1] 182s -0.870 -0.492 0.816 -0.886 182s 0.369 0.408 0.274 -0.851 182s 182s x1: [4, 3, 1, 1] 182s -0.431 -0.389 0.169 0.309 182s 0.784 0.109 0.380 0.691 182s -0.079 0.792 0.488 0.517 182s 182s y: [2, 3, 1, 1] 182s 0.431 -0.534 182s -1.038 -0.150 182s -0.381 -0.012 182s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 4, 1, 1] 182s 0.907 0.486 182s -0.822 0.274 182s 0.489 -0.815 182s 0.076 -0.134 182s 182s x1: [2, 4, 1, 1] 182s -0.887 0.744 182s -0.021 -0.943 182s -0.301 0.024 182s -0.795 0.609 182s 182s y: [4, 4, 1, 1] 182s -0.443 0.933 -1.040 -0.167 182s -0.477 -0.241 0.758 0.124 182s -0.261 0.254 -0.167 -0.026 182s -0.425 0.820 -0.885 -0.142 182s test-mul-mat0: iter:14/500 182s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 182s x0: [1, 1, 1, 1] 182s -0.417 182s 182s x1: [1, 1, 1, 1] 182s -0.377 182s 182s y: [1, 1, 1, 1] 182s 0.157 182s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 182s x0: [1, 1, 4, 1] 182s -0.896 182s 182s x1: [1, 1, 4, 1] 182s 0.992 182s 182s y: [1, 1, 4, 1] 182s -0.888 182s testing: mul_mat, [1, 1, 4, 4] = [1, 1, 4, 4] * [1, 1, 4, 4] 182s x0: [1, 1, 4, 4] 182s 0.346 182s 182s x1: [1, 1, 4, 4] 182s -0.064 182s 182s y: [1, 1, 4, 4] 182s -0.022 182s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 182s x0: [1, 4, 1, 1] 182s 0.951 182s -0.753 182s -0.738 182s 0.734 182s 182s x1: [1, 1, 1, 1] 182s -0.909 182s 182s y: [4, 1, 1, 1] 182s -0.865 0.684 0.670 -0.667 182s testing: mul_mat, [4, 1, 4, 1] = [4, 4, 4, 1] * [4, 1, 4, 1] 182s x0: [4, 4, 4, 1] 182s 0.902 0.986 -0.705 0.584 182s 0.581 -0.796 0.321 0.746 182s -0.340 0.896 -0.988 0.124 182s -0.569 -0.939 -0.458 -0.827 182s 182s x1: [4, 1, 4, 1] 182s -0.418 -0.410 -0.407 -0.598 182s 182s y: [4, 1, 4, 1] 182s -0.844 -0.493 0.102 1.304 182s testing: mul_mat, [4, 4, 4, 4] = [4, 4, 4, 4] * [4, 4, 4, 4] 182s x0: [4, 4, 4, 4] 182s -0.478 0.013 -0.869 0.974 182s -0.615 0.129 -0.643 0.983 182s -0.833 0.941 -0.785 0.662 182s -0.938 0.475 0.312 -0.355 182s 182s x1: [4, 4, 4, 4] 182s -0.266 -0.658 -0.041 -0.169 182s -0.771 0.925 0.223 0.437 182s -0.097 -0.844 -0.772 0.626 182s -0.786 0.193 -0.312 0.797 182s 182s y: [4, 4, 4, 4] 182s -0.011 -0.061 -0.478 -0.017 182s 0.612 0.879 1.627 1.078 182s 1.316 1.063 0.307 -0.772 182s 1.425 1.492 1.608 0.449 182s test-mul-mat0: iter:15/500 182s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 4, 1, 1] 182s 0.574 182s 0.840 182s 0.803 182s 0.902 182s 182s x1: [1, 4, 1, 1] 182s 0.623 182s 0.952 182s 0.923 182s 0.902 182s 182s y: [4, 4, 1, 1] 182s 0.358 0.524 0.501 0.562 182s 0.546 0.799 0.764 0.858 182s 0.530 0.775 0.741 0.832 182s 0.518 0.757 0.724 0.813 182s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 182s x0: [1, 1, 2, 1] 182s -0.895 182s 182s x1: [1, 4, 2, 1] 182s 0.097 182s -0.027 182s 0.511 182s -0.836 182s 182s y: [1, 4, 2, 1] 182s -0.087 182s 0.024 182s -0.457 182s 0.749 182s testing: mul_mat, [2, 1, 2, 3] = [1, 2, 2, 3] * [1, 1, 2, 3] 182s x0: [1, 2, 2, 3] 182s -0.061 182s -0.019 182s 182s x1: [1, 1, 2, 3] 182s 0.624 182s 182s y: [2, 1, 2, 3] 182s -0.038 -0.012 182s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 3, 1, 1] 182s 0.292 182s 0.607 182s -0.197 182s 182s x1: [1, 2, 1, 1] 182s 0.781 182s 0.444 182s 182s y: [3, 2, 1, 1] 182s 0.228 0.475 -0.154 182s 0.130 0.269 -0.087 182s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 182s x0: [3, 4, 2, 1] 182s 0.186 0.683 0.017 182s -0.472 0.533 0.813 182s -0.720 0.430 0.269 182s -0.509 -0.336 -0.547 182s 182s x1: [3, 1, 2, 1] 182s -0.021 -0.061 0.296 182s 182s y: [4, 1, 2, 1] 182s -0.040 0.218 0.069 -0.131 182s testing: mul_mat, [2, 3, 2, 3] = [4, 2, 2, 3] * [4, 3, 2, 3] 182s x0: [4, 2, 2, 3] 182s -0.190 -0.626 -0.456 -0.768 182s 0.574 0.385 -0.937 -0.483 182s 182s x1: [4, 3, 2, 3] 182s -0.949 -0.888 0.317 -0.927 182s 0.156 -0.873 0.719 -0.658 182s -0.345 0.999 -0.167 -0.662 182s 182s y: [2, 3, 2, 3] 182s 1.304 -0.736 182s 0.695 -0.602 182s 0.025 0.663 182s test-mul-mat0: iter:16/500 182s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 2, 1, 1] 182s -0.366 182s 0.518 182s 182s x1: [1, 2, 1, 1] 182s -0.666 182s 0.232 182s 182s y: [2, 2, 1, 1] 182s 0.243 -0.345 182s -0.085 0.120 182s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 182s x0: [1, 3, 3, 1] 182s 0.666 182s 0.093 182s 0.475 182s 182s x1: [1, 2, 3, 1] 182s 0.291 182s -0.654 182s 182s y: [3, 2, 3, 1] 182s 0.194 0.027 0.138 182s -0.436 -0.061 -0.311 182s testing: mul_mat, [3, 3, 3, 4] = [1, 3, 3, 4] * [1, 3, 3, 4] 182s x0: [1, 3, 3, 4] 182s 0.405 182s -0.384 182s -0.712 182s 182s x1: [1, 3, 3, 4] 182s -0.680 182s 0.379 182s -0.392 182s 182s y: [3, 3, 3, 4] 182s -0.275 0.261 0.484 182s 0.154 -0.146 -0.270 182s -0.159 0.151 0.279 182s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 4, 1, 1] 182s 0.393 182s -0.758 182s -0.179 182s 0.518 182s 182s x1: [1, 3, 1, 1] 182s -0.728 182s 0.255 182s 0.353 182s 182s y: [4, 3, 1, 1] 182s -0.287 0.552 0.130 -0.377 182s 0.100 -0.194 -0.046 0.132 182s 0.139 -0.268 -0.063 0.183 182s testing: mul_mat, [1, 1, 3, 1] = [4, 1, 3, 1] * [4, 1, 3, 1] 182s x0: [4, 1, 3, 1] 182s -0.173 -0.932 0.600 0.040 182s 182s x1: [4, 1, 3, 1] 182s 0.816 -0.378 -0.178 0.743 182s 182s y: [1, 1, 3, 1] 182s 0.134 182s testing: mul_mat, [4, 4, 3, 4] = [1, 4, 3, 4] * [1, 4, 3, 4] 182s x0: [1, 4, 3, 4] 182s -0.021 182s -0.828 182s 0.722 182s -0.925 182s 182s x1: [1, 4, 3, 4] 182s 0.553 182s -0.904 182s 0.178 182s 0.794 182s 182s y: [4, 4, 3, 4] 182s -0.012 -0.458 0.399 -0.511 182s 0.019 0.749 -0.652 0.836 182s -0.004 -0.147 0.128 -0.164 182s -0.017 -0.658 0.573 -0.735 182s test-mul-mat0: iter:17/500 182s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 1, 1, 1] 182s -0.004 182s 182s x1: [1, 2, 1, 1] 182s -0.143 182s 0.623 182s 182s y: [1, 2, 1, 1] 182s 0.001 182s -0.003 182s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 182s x0: [1, 3, 1, 1] 182s 0.075 182s -0.077 182s 0.254 182s 182s x1: [1, 1, 1, 1] 182s 0.532 182s 182s y: [3, 1, 1, 1] 182s 0.040 -0.041 0.135 182s testing: mul_mat, [1, 3, 1, 4] = [1, 1, 1, 4] * [1, 3, 1, 4] 182s x0: [1, 1, 1, 4] 182s 0.189 182s 182s x1: [1, 3, 1, 4] 182s -0.048 182s 0.293 182s -0.636 182s 182s y: [1, 3, 1, 4] 182s -0.009 182s 0.055 182s -0.120 182s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 182s x0: [1, 3, 1, 1] 182s -0.897 182s 0.536 182s 0.068 182s 182s x1: [1, 1, 1, 1] 182s -0.888 182s 182s y: [3, 1, 1, 1] 182s 0.797 -0.476 -0.061 182s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 182s x0: [3, 1, 1, 1] 182s -0.309 -0.404 0.032 182s 182s x1: [3, 1, 1, 1] 182s -0.039 0.159 -0.720 182s 182s y: [1, 1, 1, 1] 182s -0.075 182s testing: mul_mat, [2, 3, 1, 4] = [1, 2, 1, 4] * [1, 3, 1, 4] 182s x0: [1, 2, 1, 4] 182s 0.903 182s 0.410 182s 182s x1: [1, 3, 1, 4] 182s 0.614 182s 0.851 182s 0.984 182s 182s y: [2, 3, 1, 4] 182s 0.555 0.251 182s 0.769 0.348 182s 0.889 0.403 182s test-mul-mat0: iter:18/500 182s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 182s x0: [2, 4, 1, 1] 182s -0.471 0.499 182s -0.190 -0.857 182s 0.350 -0.206 182s 0.050 0.565 182s 182s x1: [2, 1, 1, 1] 182s 0.838 -0.098 182s 182s y: [4, 1, 1, 1] 182s -0.443 -0.075 0.313 -0.013 182s testing: mul_mat, [4, 4, 3, 1] = [2, 4, 3, 1] * [2, 4, 3, 1] 182s x0: [2, 4, 3, 1] 182s -0.885 0.793 182s -0.832 -0.091 182s -0.157 0.733 182s -0.763 -0.464 182s 182s x1: [2, 4, 3, 1] 182s 0.328 0.693 182s -0.196 -0.454 182s -0.430 -0.475 182s 0.917 -0.267 182s 182s y: [4, 4, 3, 1] 182s 0.259 -0.337 0.456 -0.572 182s -0.187 0.204 -0.302 0.360 182s 0.004 0.401 -0.281 0.549 182s -1.023 -0.739 -0.340 -0.575 182s testing: mul_mat, [4, 4, 3, 4] = [2, 4, 3, 4] * [2, 4, 3, 4] 182s x0: [2, 4, 3, 4] 182s 0.570 0.862 182s -0.545 0.675 182s -0.919 -0.964 182s 0.703 -0.567 182s 182s x1: [2, 4, 3, 4] 182s -0.523 0.997 182s 0.954 0.796 182s 0.302 0.604 182s -0.386 0.417 182s 182s y: [4, 4, 3, 4] 182s 0.561 0.958 -0.481 -0.933 182s 1.230 0.017 -1.645 0.220 182s 0.693 0.243 -0.860 -0.131 182s 0.139 0.492 -0.047 -0.508 182s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 4, 1, 1] 182s 0.032 0.120 182s 0.142 -0.502 182s -0.080 -0.993 182s -0.282 0.433 182s 182s x1: [2, 4, 1, 1] 182s -0.147 -0.647 182s -0.666 -0.222 182s -0.214 0.791 182s 0.982 -0.533 182s 182s y: [4, 4, 1, 1] 182s -0.082 0.304 0.655 -0.239 182s -0.048 0.017 0.274 0.092 182s 0.088 -0.427 -0.768 0.403 182s -0.033 0.407 0.450 -0.508 182s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 182s x0: [4, 2, 3, 1] 182s -0.932 0.539 -0.086 -0.388 182s -0.093 -0.434 -0.028 0.243 182s 182s x1: [4, 2, 3, 1] 182s -0.955 -0.670 -0.602 0.494 182s -0.733 0.067 -0.514 0.107 182s 182s y: [2, 2, 3, 1] 182s 0.388 0.517 182s 0.722 0.080 182s testing: mul_mat, [4, 4, 3, 4] = [2, 4, 3, 4] * [2, 4, 3, 4] 182s x0: [2, 4, 3, 4] 182s -0.468 -0.049 182s -0.193 0.995 182s 0.338 0.117 182s 0.320 0.332 182s 182s x1: [2, 4, 3, 4] 182s 0.295 -0.212 182s -0.734 -0.428 182s 0.936 -0.544 182s -0.010 -0.995 182s 182s y: [4, 4, 3, 4] 182s -0.128 -0.268 0.075 0.024 182s 0.364 -0.284 -0.298 -0.377 182s -0.411 -0.722 0.253 0.119 182s 0.053 -0.989 -0.119 -0.333 182s test-mul-mat0: iter:19/500 182s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 2, 1, 1] 182s 0.122 -0.698 -0.011 182s -0.949 0.958 0.025 182s 182s x1: [3, 3, 1, 1] 182s -0.022 0.587 0.161 182s -0.016 -0.000 0.189 182s 0.294 -0.191 0.480 182s 182s y: [2, 3, 1, 1] 182s -0.414 0.586 182s -0.004 0.019 182s 0.164 -0.450 182s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 182s x0: [3, 4, 1, 1] 182s -0.263 -0.148 0.480 182s -0.741 0.941 -0.639 182s 0.918 0.928 0.340 182s 0.504 0.088 -0.676 182s 182s x1: [3, 2, 1, 1] 182s 0.294 0.556 0.789 182s 0.579 -0.284 0.486 182s 182s y: [4, 2, 1, 1] 182s 0.219 -0.198 1.054 -0.336 182s 0.123 -1.007 0.433 -0.062 182s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 182s x0: [3, 4, 1, 1] 182s 0.574 0.084 0.359 182s -0.711 -0.430 0.614 182s 0.026 0.422 0.094 182s 0.285 0.364 0.456 182s 182s x1: [3, 4, 1, 1] 182s -0.496 -0.722 0.618 182s 0.313 0.757 0.281 182s -0.565 -0.940 -0.730 182s -0.514 -0.983 0.295 182s 182s y: [4, 4, 1, 1] 182s -0.123 1.043 -0.259 -0.122 182s 0.344 -0.375 0.354 0.492 182s -0.665 0.357 -0.480 -0.836 182s -0.272 0.969 -0.400 -0.369 182s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 182s x0: [3, 3, 1, 1] 182s -0.526 -0.509 -0.935 182s -0.712 0.583 -0.333 182s 0.444 0.224 -0.417 182s 182s x1: [3, 4, 1, 1] 182s 0.202 0.291 -0.205 182s -0.294 -0.621 0.119 182s 0.210 -0.343 -0.263 182s -0.477 -0.586 -0.982 182s 182s y: [3, 4, 1, 1] 182s -0.063 0.094 0.240 182s 0.359 -0.193 -0.319 182s 0.309 -0.262 0.126 182s 1.467 0.325 0.066 182s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 2, 1, 1] 182s 0.315 -0.427 -0.338 182s 0.543 0.694 0.783 182s 182s x1: [3, 3, 1, 1] 182s -0.646 0.238 -0.803 182s 0.380 -0.340 0.292 182s -0.336 -0.977 -0.253 182s 182s y: [2, 3, 1, 1] 182s -0.034 -0.814 182s 0.166 0.198 182s 0.397 -1.058 182s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 182s x0: [2, 4, 1, 1] 182s 0.706 0.932 182s -0.292 0.795 182s 0.731 0.956 182s 0.289 -0.484 182s 182s x1: [2, 3, 1, 1] 182s -0.649 0.399 182s -0.694 -0.234 182s 0.418 0.264 182s 182s y: [4, 3, 1, 1] 182s -0.086 0.507 -0.092 -0.381 182s -0.708 0.016 -0.731 -0.087 182s 0.541 0.089 0.558 -0.007 182s test-mul-mat0: iter:20/500 182s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 4, 1, 1] 182s -0.248 182s -0.941 182s 0.191 182s -0.705 182s 182s x1: [1, 3, 1, 1] 182s 0.853 182s -0.995 182s -0.807 182s 182s y: [4, 3, 1, 1] 182s -0.212 -0.803 0.163 -0.602 182s 0.247 0.936 -0.190 0.702 182s 0.200 0.759 -0.154 0.569 182s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 2, 1, 1] 182s -0.644 182s -0.278 182s 182s x1: [1, 4, 1, 1] 182s -0.368 182s -0.115 182s -0.043 182s -0.585 182s 182s y: [2, 4, 1, 1] 182s 0.237 0.102 182s 0.074 0.032 182s 0.028 0.012 182s 0.376 0.163 182s testing: mul_mat, [2, 2, 1, 3] = [1, 2, 1, 3] * [1, 2, 1, 3] 182s x0: [1, 2, 1, 3] 182s -0.232 182s -0.373 182s 182s x1: [1, 2, 1, 3] 182s -0.998 182s 0.774 182s 182s y: [2, 2, 1, 3] 182s 0.231 0.372 182s -0.179 -0.289 182s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 2, 1, 1] 182s 0.771 182s 0.860 182s 182s x1: [1, 2, 1, 1] 182s 0.918 182s 0.856 182s 182s y: [2, 2, 1, 1] 182s 0.708 0.790 182s 0.660 0.736 182s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 182s x0: [2, 4, 1, 1] 182s -0.302 0.725 182s -0.160 0.863 182s -0.093 0.746 182s 0.330 -0.039 182s 182s x1: [2, 1, 1, 1] 182s -0.361 0.648 182s 182s y: [4, 1, 1, 1] 182s 0.579 0.618 0.517 -0.145 182s testing: mul_mat, [4, 2, 1, 3] = [4, 4, 1, 3] * [4, 2, 1, 3] 182s x0: [4, 4, 1, 3] 182s -0.847 0.720 -0.208 0.701 182s 0.974 0.829 -0.771 0.790 182s -0.132 -0.765 -0.359 -0.826 182s -0.984 -0.813 -0.003 0.942 182s 182s x1: [4, 2, 1, 3] 182s -0.781 -0.532 -0.037 0.993 182s -0.546 -0.795 0.473 0.149 182s 182s y: [4, 2, 1, 3] 182s 0.982 -0.389 -0.297 2.137 182s -0.104 -1.439 0.388 1.322 182s test-mul-mat0: iter:21/500 182s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 3, 1, 1] 182s 0.203 -0.189 182s 0.891 0.963 182s -0.883 -0.541 182s 182s x1: [2, 4, 1, 1] 182s -0.887 -0.508 182s 0.935 -0.440 182s -0.616 -0.833 182s 0.310 0.521 182s 182s y: [3, 4, 1, 1] 182s -0.084 -1.279 1.058 182s 0.273 0.409 -0.588 182s 0.033 -1.352 0.995 182s -0.036 0.778 -0.555 182s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 182s x0: [2, 2, 4, 1] 182s 0.819 -0.006 182s -0.783 0.783 182s 182s x1: [2, 3, 4, 1] 182s -0.447 0.658 182s 0.215 0.934 182s 0.428 -0.388 182s 182s y: [2, 3, 4, 1] 182s -0.371 0.865 182s 0.170 0.563 182s 0.353 -0.639 182s testing: mul_mat, [4, 2, 4, 4] = [2, 4, 4, 4] * [2, 2, 4, 4] 182s x0: [2, 4, 4, 4] 182s -0.077 0.364 182s 0.857 0.085 182s -0.751 0.231 182s 0.344 -0.290 182s 182s x1: [2, 2, 4, 4] 182s 0.408 -0.148 182s -0.019 -0.958 182s 182s y: [4, 2, 4, 4] 182s -0.085 0.337 -0.341 0.183 182s -0.347 -0.098 -0.206 0.271 182s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 4, 1, 1] 182s -0.813 0.748 182s 0.386 0.038 182s 0.452 -0.327 182s -0.950 0.986 182s 182s x1: [2, 4, 1, 1] 182s -0.238 -0.631 182s -0.254 -0.834 182s 0.709 -0.256 182s -0.111 0.988 182s 182s y: [4, 4, 1, 1] 182s -0.278 -0.116 0.098 -0.395 182s -0.418 -0.129 0.158 -0.581 182s -0.768 0.264 0.404 -0.926 182s 0.829 -0.005 -0.373 1.079 182s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 182s x0: [4, 1, 4, 1] 182s 0.348 -0.949 -0.048 -0.628 182s 182s x1: [4, 2, 4, 1] 182s -0.080 -0.249 0.656 -0.543 182s 0.694 0.024 0.224 0.179 182s 182s y: [1, 2, 4, 1] 182s 0.517 182s 0.096 182s testing: mul_mat, [3, 4, 4, 4] = [1, 3, 4, 4] * [1, 4, 4, 4] 182s x0: [1, 3, 4, 4] 182s 0.977 182s -0.652 182s -0.000 182s 182s x1: [1, 4, 4, 4] 182s -0.984 182s -0.035 182s 0.451 182s -0.412 182s 182s y: [3, 4, 4, 4] 182s -0.962 0.641 0.000 182s -0.034 0.023 0.000 182s 0.441 -0.294 -0.000 182s -0.403 0.268 0.000 182s test-mul-mat0: iter:22/500 182s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 182s x0: [4, 2, 1, 1] 182s -0.019 -0.533 0.959 0.615 182s -0.650 -0.993 -0.403 0.233 182s 182s x1: [4, 1, 1, 1] 182s 0.789 -0.229 -0.211 -0.203 182s 182s y: [2, 1, 1, 1] 182s -0.220 -0.248 182s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 182s x0: [4, 3, 3, 1] 182s -0.802 -0.976 -0.098 -0.795 182s -0.378 -0.865 0.806 0.241 182s 0.104 -0.630 0.070 -0.911 182s 182s x1: [4, 2, 3, 1] 182s 0.602 -0.381 -0.031 -0.436 182s 0.829 -0.014 -0.339 0.430 182s 182s y: [3, 2, 3, 1] 182s 0.239 -0.028 0.697 182s -0.960 -0.471 -0.320 182s testing: mul_mat, [1, 3, 3, 4] = [4, 1, 3, 4] * [4, 3, 3, 4] 182s x0: [4, 1, 3, 4] 182s -0.791 0.304 -0.679 -0.247 182s 182s x1: [4, 3, 3, 4] 182s -0.394 -0.986 0.033 0.710 182s -0.616 -0.898 0.799 0.415 182s -0.398 0.431 0.398 -0.171 182s 182s y: [1, 3, 3, 4] 182s -0.186 182s -0.430 182s 0.218 182s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 182s x0: [4, 2, 1, 1] 182s 0.624 0.659 0.558 -0.280 182s 0.138 -0.216 0.089 0.294 182s 182s x1: [4, 1, 1, 1] 182s -0.979 0.924 -0.711 0.276 182s 182s y: [2, 1, 1, 1] 182s -0.476 -0.317 182s testing: mul_mat, [1, 4, 3, 1] = [2, 1, 3, 1] * [2, 4, 3, 1] 182s x0: [2, 1, 3, 1] 182s 0.959 0.901 182s 182s x1: [2, 4, 3, 1] 182s 0.954 -0.136 182s -0.668 0.139 182s 0.761 0.018 182s 0.717 -0.535 182s 182s y: [1, 4, 3, 1] 182s 0.792 182s -0.515 182s 0.746 182s 0.206 182s testing: mul_mat, [4, 2, 3, 4] = [1, 4, 3, 4] * [1, 2, 3, 4] 182s x0: [1, 4, 3, 4] 182s 0.835 182s -0.764 182s 0.253 182s -0.117 182s 182s x1: [1, 2, 3, 4] 182s 0.002 182s 0.485 182s 182s y: [4, 2, 3, 4] 182s 0.001 -0.001 0.000 -0.000 182s 0.405 -0.371 0.123 -0.057 182s test-mul-mat0: iter:23/500 182s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 2, 1, 1] 182s -0.234 182s 0.299 182s 182s x1: [1, 3, 1, 1] 182s -0.912 182s -0.581 182s -0.047 182s 182s y: [2, 3, 1, 1] 182s 0.213 -0.273 182s 0.136 -0.174 182s 0.011 -0.014 182s testing: mul_mat, [4, 2, 3, 1] = [1, 4, 3, 1] * [1, 2, 3, 1] 182s x0: [1, 4, 3, 1] 182s 0.990 182s -0.327 182s 0.616 182s -0.184 182s 182s x1: [1, 2, 3, 1] 182s -0.267 182s -0.010 182s 182s y: [4, 2, 3, 1] 182s -0.264 0.087 -0.164 0.049 182s -0.010 0.003 -0.006 0.002 182s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 182s x0: [1, 3, 3, 1] 182s 0.828 182s -0.529 182s -0.212 182s 182s x1: [1, 4, 3, 1] 182s 0.681 182s -0.994 182s -0.587 182s -0.129 182s 182s y: [3, 4, 3, 1] 182s 0.564 -0.360 -0.144 182s -0.824 0.526 0.211 182s -0.486 0.310 0.124 182s -0.107 0.068 0.027 182s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 1, 1, 1] 182s 0.321 182s 182s x1: [1, 3, 1, 1] 182s 0.682 182s -0.170 182s 0.052 182s 182s y: [1, 3, 1, 1] 182s 0.219 182s -0.054 182s 0.017 182s testing: mul_mat, [4, 1, 3, 1] = [1, 4, 3, 1] * [1, 1, 3, 1] 182s x0: [1, 4, 3, 1] 182s 0.431 182s -0.180 182s 0.903 182s -0.699 182s 182s x1: [1, 1, 3, 1] 182s 0.499 182s 182s y: [4, 1, 3, 1] 182s 0.215 -0.090 0.451 -0.349 182s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 182s x0: [4, 3, 3, 1] 182s 0.986 0.736 0.166 0.467 182s 0.855 -0.331 0.489 0.961 182s -0.388 -0.898 -0.995 -0.855 182s 182s x1: [4, 1, 3, 1] 182s 0.165 0.211 -0.708 0.953 182s 182s y: [3, 1, 3, 1] 182s 0.645 0.641 -0.363 182s test-mul-mat0: iter:24/500 182s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 182s x0: [4, 2, 1, 1] 182s 0.404 -0.566 0.859 -0.299 182s -0.828 -0.041 -0.710 0.863 182s 182s x1: [4, 1, 1, 1] 182s 0.135 0.464 0.928 0.862 182s 182s y: [2, 1, 1, 1] 182s 0.331 -0.046 182s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 182s x0: [4, 3, 3, 1] 182s -0.859 -0.966 0.605 0.100 182s -0.676 0.468 0.246 0.866 182s 0.220 0.535 -0.638 -0.077 182s 182s x1: [4, 2, 3, 1] 182s -0.854 0.541 0.752 -0.711 182s -0.504 0.703 0.068 0.986 182s 182s y: [3, 2, 3, 1] 182s 0.595 0.400 -0.323 182s -0.107 1.539 0.145 182s testing: mul_mat, [3, 3, 3, 3] = [4, 3, 3, 3] * [4, 3, 3, 3] 182s x0: [4, 3, 3, 3] 182s -0.440 -0.372 0.374 -0.476 182s -0.564 -0.652 -0.125 -0.939 182s 0.894 0.852 0.873 0.627 182s 182s x1: [4, 3, 3, 3] 182s 0.418 -0.048 0.836 -0.362 182s -0.513 -0.802 0.560 0.090 182s -0.455 0.249 0.527 -0.776 182s 182s y: [3, 3, 3, 3] 182s 0.319 0.031 0.835 182s 0.691 0.658 -0.597 182s 0.674 0.757 -0.220 182s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 3, 1, 1] 182s -0.835 -0.869 0.079 -0.775 182s 0.154 0.779 0.926 0.315 182s -0.862 0.357 -0.211 -0.047 182s 182s x1: [4, 3, 1, 1] 182s -0.026 0.825 -0.533 -0.107 182s 0.383 -0.790 -0.711 -0.670 182s -0.722 -0.806 -0.162 -0.605 182s 182s y: [3, 3, 1, 1] 182s -0.654 0.111 0.434 182s 0.830 -1.425 -0.431 182s 1.759 -1.080 0.397 182s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 182s x0: [3, 3, 3, 1] 182s 0.811 -0.280 0.918 182s -0.649 0.695 0.370 182s 0.981 0.848 0.994 182s 182s x1: [3, 4, 3, 1] 182s -0.360 0.538 0.994 182s 0.042 -0.863 0.191 182s -0.984 0.962 0.658 182s -0.091 0.345 0.868 182s 182s y: [3, 4, 3, 1] 182s 0.470 0.976 1.092 182s 0.451 -0.557 -0.501 182s -0.463 1.551 0.505 182s 0.626 0.620 1.065 182s testing: mul_mat, [1, 3, 3, 3] = [3, 1, 3, 3] * [3, 3, 3, 3] 182s x0: [3, 1, 3, 3] 182s 0.987 0.471 0.533 182s 182s x1: [3, 3, 3, 3] 182s 0.097 0.917 -0.544 182s 0.419 0.728 -0.194 182s 0.400 -0.552 -0.498 182s 182s y: [1, 3, 3, 3] 182s 0.238 182s 0.654 182s -0.131 182s test-mul-mat0: iter:25/500 182s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 182s x0: [2, 1, 1, 1] 182s -0.546 0.533 182s 182s x1: [2, 3, 1, 1] 182s 0.619 -0.924 182s -0.011 -0.384 182s 0.819 -0.509 182s 182s y: [1, 3, 1, 1] 182s -0.831 182s -0.199 182s -0.718 182s testing: mul_mat, [1, 1, 4, 1] = [2, 1, 4, 1] * [2, 1, 4, 1] 182s x0: [2, 1, 4, 1] 182s -0.570 0.968 182s 182s x1: [2, 1, 4, 1] 182s -0.015 -0.330 182s 182s y: [1, 1, 4, 1] 182s -0.310 182s testing: mul_mat, [3, 1, 4, 2] = [2, 3, 4, 2] * [2, 1, 4, 2] 182s x0: [2, 3, 4, 2] 182s -0.215 0.696 182s 0.567 -0.777 182s -0.077 0.230 182s 182s x1: [2, 1, 4, 2] 182s -0.540 0.981 182s 182s y: [3, 1, 4, 2] 182s 0.799 -1.068 0.268 182s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 182s x0: [2, 2, 1, 1] 182s 0.644 -0.700 182s -0.538 -0.802 182s 182s x1: [2, 3, 1, 1] 182s 0.140 -0.646 182s 0.199 -0.586 182s 0.540 -0.504 182s 182s y: [2, 3, 1, 1] 182s 0.542 0.443 182s 0.538 0.362 182s 0.701 0.114 182s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 182s x0: [2, 3, 4, 1] 182s -0.790 0.350 182s 0.199 0.553 182s -0.780 0.419 182s 182s x1: [2, 2, 4, 1] 182s -0.916 -0.964 182s 0.988 -0.509 182s 182s y: [3, 2, 4, 1] 182s 0.386 -0.715 0.310 182s -0.958 -0.085 -0.984 182s testing: mul_mat, [3, 2, 4, 2] = [3, 3, 4, 2] * [3, 2, 4, 2] 182s x0: [3, 3, 4, 2] 182s -0.757 0.676 -0.939 182s 0.070 -0.062 0.023 182s 0.120 -0.027 0.094 182s 182s x1: [3, 2, 4, 2] 182s 0.106 0.464 0.810 182s -0.535 -0.333 0.414 182s 182s y: [3, 2, 4, 2] 182s -0.526 -0.002 0.076 182s -0.209 -0.007 -0.016 182s test-mul-mat0: iter:26/500 182s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 2, 1, 1] 182s -0.898 182s 0.042 182s 182s x1: [1, 2, 1, 1] 182s -0.796 182s 0.779 182s 182s y: [2, 2, 1, 1] 182s 0.715 -0.033 182s -0.699 0.033 182s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 182s x0: [1, 3, 2, 1] 182s 0.887 182s -0.312 182s 0.885 182s 182s x1: [1, 2, 2, 1] 182s -0.339 182s -0.020 182s 182s y: [3, 2, 2, 1] 182s -0.300 0.106 -0.300 182s -0.018 0.006 -0.018 182s testing: mul_mat, [3, 3, 2, 3] = [1, 3, 2, 3] * [1, 3, 2, 3] 182s x0: [1, 3, 2, 3] 182s 0.644 182s -0.225 182s 0.912 182s 182s x1: [1, 3, 2, 3] 182s 0.762 182s -0.301 182s -0.960 182s 182s y: [3, 3, 2, 3] 182s 0.491 -0.171 0.695 182s -0.194 0.068 -0.275 182s -0.619 0.216 -0.876 182s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 3, 1, 1] 182s -0.097 182s 0.179 182s -0.122 182s 182s x1: [1, 3, 1, 1] 182s 0.582 182s -0.048 182s -0.208 182s 182s y: [3, 3, 1, 1] 182s -0.056 0.104 -0.071 182s 0.005 -0.009 0.006 182s 0.020 -0.037 0.025 182s testing: mul_mat, [2, 1, 2, 1] = [3, 2, 2, 1] * [3, 1, 2, 1] 182s x0: [3, 2, 2, 1] 182s -0.012 0.049 -0.144 182s -0.448 -0.303 -0.987 182s 182s x1: [3, 1, 2, 1] 182s 0.726 -0.901 0.708 182s 182s y: [2, 1, 2, 1] 182s -0.155 -0.751 182s testing: mul_mat, [2, 3, 2, 3] = [2, 2, 2, 3] * [2, 3, 2, 3] 182s x0: [2, 2, 2, 3] 182s -0.401 0.140 182s 0.231 -0.403 182s 182s x1: [2, 3, 2, 3] 182s -0.023 -0.713 182s -0.280 -0.857 182s -0.363 -0.699 182s 182s y: [2, 3, 2, 3] 182s -0.090 0.282 182s -0.007 0.281 182s 0.048 0.198 182s test-mul-mat0: iter:27/500 182s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 3, 1, 1] 182s 0.520 182s -0.683 182s -0.196 182s 182s x1: [1, 2, 1, 1] 182s -0.079 182s 0.086 182s 182s y: [3, 2, 1, 1] 182s -0.041 0.054 0.015 182s 0.045 -0.059 -0.017 182s testing: mul_mat, [2, 3, 2, 1] = [1, 2, 2, 1] * [1, 3, 2, 1] 182s x0: [1, 2, 2, 1] 182s -0.837 182s -0.426 182s 182s x1: [1, 3, 2, 1] 182s -0.883 182s -0.845 182s 0.375 182s 182s y: [2, 3, 2, 1] 182s 0.740 0.377 182s 0.707 0.360 182s -0.314 -0.160 182s testing: mul_mat, [1, 2, 2, 2] = [1, 1, 2, 2] * [1, 2, 2, 2] 182s x0: [1, 1, 2, 2] 182s -0.244 182s 182s x1: [1, 2, 2, 2] 182s -0.607 182s -0.313 182s 182s y: [1, 2, 2, 2] 182s 0.148 182s 0.077 182s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 182s x0: [1, 4, 1, 1] 182s 0.125 182s -0.990 182s 0.095 182s 0.442 182s 182s x1: [1, 1, 1, 1] 182s -0.654 182s 182s y: [4, 1, 1, 1] 182s -0.082 0.647 -0.062 -0.289 182s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 182s x0: [4, 1, 2, 1] 182s 0.847 0.672 0.993 -0.759 182s 182s x1: [4, 1, 2, 1] 182s -0.186 0.211 0.597 -0.811 182s 182s y: [1, 1, 2, 1] 182s 1.193 182s testing: mul_mat, [1, 4, 2, 2] = [1, 1, 2, 2] * [1, 4, 2, 2] 182s x0: [1, 1, 2, 2] 182s -0.288 182s 182s x1: [1, 4, 2, 2] 182s -0.319 182s -0.722 182s -0.916 182s 0.917 182s 182s y: [1, 4, 2, 2] 182s 0.092 182s 0.208 182s 0.264 182s -0.264 182s test-mul-mat0: iter:28/500 182s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 3, 1, 1] 182s -0.901 0.335 -0.198 -0.703 182s 0.045 -0.958 0.938 0.568 182s -0.506 0.176 -0.721 -0.406 182s 182s x1: [4, 3, 1, 1] 182s 0.988 -0.927 -0.410 0.488 182s 0.713 -0.802 -0.831 0.992 182s -0.717 -0.914 0.026 0.636 182s 182s y: [3, 3, 1, 1] 182s -1.463 0.825 -0.566 182s -1.444 0.583 -0.306 182s -0.112 1.228 -0.074 182s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 2, 1, 1] 182s 0.875 -0.542 -0.563 -0.099 182s -0.906 0.094 -0.000 0.429 182s 182s x1: [4, 3, 1, 1] 182s -0.289 -0.735 0.155 -0.294 182s -0.903 0.387 -0.306 -0.830 182s 0.978 -0.818 0.884 -0.824 182s 182s y: [2, 3, 1, 1] 182s 0.087 0.067 182s -0.745 0.498 182s 0.883 -1.316 182s testing: mul_mat, [4, 2, 1, 3] = [4, 4, 1, 3] * [4, 2, 1, 3] 182s x0: [4, 4, 1, 3] 182s 0.493 0.513 -0.726 -0.414 182s -0.488 0.703 -0.518 0.810 182s -0.823 -0.580 0.045 -0.782 182s -0.148 0.456 -0.461 0.878 182s 182s x1: [4, 2, 1, 3] 182s 0.896 0.297 -0.526 0.938 182s 0.235 -0.959 -0.568 -0.589 182s 182s y: [4, 2, 1, 3] 182s 0.587 0.805 -1.666 1.069 182s 0.280 -0.971 0.797 -0.727 182s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 182s x0: [4, 3, 1, 1] 182s 0.536 -0.214 -0.543 -0.021 182s -0.462 -0.619 0.268 0.612 182s 0.634 -0.136 -0.687 -0.699 182s 182s x1: [4, 4, 1, 1] 182s 0.113 -0.540 0.330 -0.080 182s -0.913 -0.740 0.580 0.261 182s 0.894 -0.024 -0.844 -0.416 182s -0.016 0.231 -0.233 -0.495 182s 182s y: [3, 4, 1, 1] 182s -0.002 0.322 -0.026 182s -0.651 1.195 -1.059 182s 0.951 -0.880 1.441 182s 0.079 -0.502 0.465 182s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 182s x0: [3, 3, 1, 1] 182s -0.396 0.371 -0.093 182s 0.872 -0.623 -0.085 182s -0.425 0.156 -0.210 182s 182s x1: [3, 4, 1, 1] 182s 0.380 -0.465 0.413 182s 0.839 0.866 -0.667 182s 0.927 -0.874 0.912 182s 0.187 -0.980 -0.112 182s 182s y: [3, 4, 1, 1] 182s -0.361 0.585 -0.321 182s 0.052 0.250 -0.082 182s -0.776 1.275 -0.722 182s -0.428 0.783 -0.209 182s testing: mul_mat, [2, 3, 1, 3] = [3, 2, 1, 3] * [3, 3, 1, 3] 182s x0: [3, 2, 1, 3] 182s 0.605 0.325 0.273 182s 0.978 0.298 -0.021 182s 182s x1: [3, 3, 1, 3] 182s 0.693 0.295 0.654 182s -0.850 -0.437 0.967 182s 0.129 -0.824 -0.733 182s 182s y: [2, 3, 1, 3] 182s 0.694 0.752 182s -0.392 -0.982 182s -0.390 -0.104 182s test-mul-mat0: iter:29/500 182s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 182s x0: [3, 3, 1, 1] 182s -0.667 0.356 -0.250 182s 0.397 0.892 -0.645 182s 0.375 0.217 0.654 182s 182s x1: [3, 1, 1, 1] 182s -0.275 0.647 0.976 182s 182s y: [3, 1, 1, 1] 182s 0.170 -0.162 0.675 182s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 2, 1, 1] 182s 0.248 -0.800 -0.480 182s -0.845 0.910 -0.478 182s 182s x1: [3, 3, 1, 1] 182s -0.352 -0.804 0.102 182s -0.515 0.141 0.680 182s -0.503 -0.646 -0.234 182s 182s y: [2, 3, 1, 1] 182s 0.507 -0.483 182s -0.567 0.239 182s 0.504 -0.050 182s testing: mul_mat, [1, 2, 1, 2] = [3, 1, 1, 2] * [3, 2, 1, 2] 182s x0: [3, 1, 1, 2] 182s 0.522 -0.943 -0.613 182s 182s x1: [3, 2, 1, 2] 182s -0.013 0.081 0.247 182s -0.366 0.057 0.298 182s 182s y: [1, 2, 1, 2] 182s -0.234 182s -0.428 182s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 182s x0: [3, 1, 1, 1] 182s 0.605 -0.412 -0.539 182s 182s x1: [3, 1, 1, 1] 182s -0.141 -0.161 0.368 182s 182s y: [1, 1, 1, 1] 182s -0.218 182s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 4, 1, 1] 182s 0.630 182s 0.603 182s -0.204 182s -0.123 182s 182s x1: [1, 3, 1, 1] 182s 0.805 182s 0.108 182s -0.384 182s 182s y: [4, 3, 1, 1] 182s 0.507 0.485 -0.165 -0.099 182s 0.068 0.065 -0.022 -0.013 182s -0.242 -0.232 0.079 0.047 182s testing: mul_mat, [2, 1, 1, 2] = [4, 2, 1, 2] * [4, 1, 1, 2] 182s x0: [4, 2, 1, 2] 182s 0.306 -0.317 0.810 -0.437 182s -0.907 0.692 0.843 -0.331 182s 182s x1: [4, 1, 1, 2] 182s -0.764 0.853 -0.825 0.203 182s 182s y: [2, 1, 1, 2] 182s -1.261 0.520 182s test-mul-mat0: iter:30/500 182s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 4, 1, 1] 182s 0.293 0.158 182s -0.126 0.386 182s 0.841 -0.434 182s 0.196 0.684 182s 182s x1: [2, 4, 1, 1] 182s -0.397 -0.866 182s -0.202 -0.544 182s -0.691 -0.999 182s -0.278 -0.468 182s 182s y: [4, 4, 1, 1] 182s -0.253 -0.284 0.042 -0.670 182s -0.145 -0.184 0.066 -0.412 182s -0.360 -0.298 -0.148 -0.818 182s -0.155 -0.145 -0.031 -0.375 182s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 1, 1, 1] 182s 0.208 0.048 182s 182s x1: [2, 4, 1, 1] 182s 0.130 0.864 182s -0.634 0.061 182s 0.746 -0.347 182s 0.581 -0.911 182s 182s y: [1, 4, 1, 1] 182s 0.068 182s -0.129 182s 0.139 182s 0.077 182s testing: mul_mat, [3, 1, 1, 2] = [2, 3, 1, 2] * [2, 1, 1, 2] 182s x0: [2, 3, 1, 2] 182s -0.593 0.611 182s -0.091 -0.592 182s -0.667 0.441 182s 182s x1: [2, 1, 1, 2] 182s -0.448 -0.534 182s 182s y: [3, 1, 1, 2] 182s -0.060 0.357 0.064 182s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 182s x0: [2, 1, 1, 1] 182s 0.817 0.143 182s 182s x1: [2, 3, 1, 1] 182s 0.416 -0.794 182s -0.876 -0.455 182s -0.929 -0.510 182s 182s y: [1, 3, 1, 1] 182s 0.227 182s -0.781 182s -0.832 182s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 2, 1, 1] 182s -0.604 182s -0.045 182s 182s x1: [1, 2, 1, 1] 182s 0.187 182s 0.907 182s 182s y: [2, 2, 1, 1] 182s -0.113 -0.008 182s -0.548 -0.041 182s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 182s x0: [2, 4, 1, 2] 182s -0.030 0.538 182s 0.205 0.812 182s -0.629 0.354 182s 0.378 0.163 182s 182s x1: [2, 1, 1, 2] 182s -0.561 -0.138 182s 182s y: [4, 1, 1, 2] 182s -0.058 -0.227 0.304 -0.234 182s test-mul-mat0: iter:31/500 182s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 182s x0: [3, 3, 1, 1] 182s 0.354 0.130 -0.065 182s -0.784 0.694 -0.471 182s 0.678 -0.336 0.734 182s 182s x1: [3, 1, 1, 1] 182s -0.392 -0.041 -0.825 182s 182s y: [3, 1, 1, 1] 182s -0.090 0.667 -0.858 182s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 182s x0: [3, 2, 3, 1] 182s 0.364 0.372 0.864 182s 0.413 0.414 0.136 182s 182s x1: [3, 3, 3, 1] 182s -0.951 -0.958 0.272 182s 0.861 0.396 -0.565 182s -0.728 0.600 -0.391 182s 182s y: [2, 3, 3, 1] 182s -0.467 -0.752 182s -0.028 0.443 182s -0.380 -0.106 182s testing: mul_mat, [3, 2, 3, 4] = [3, 3, 3, 4] * [3, 2, 3, 4] 182s x0: [3, 3, 3, 4] 182s -0.865 0.953 -0.520 182s -0.476 -0.135 0.724 182s 0.624 0.676 -0.541 182s 182s x1: [3, 2, 3, 4] 182s -0.668 -0.830 -0.681 182s -0.497 -0.715 -0.797 182s 182s y: [3, 2, 3, 4] 182s 0.141 -0.063 -0.610 182s 0.163 -0.244 -0.362 182s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 1, 1, 1] 182s 0.477 -0.755 0.040 182s 182s x1: [3, 3, 1, 1] 182s -0.923 0.834 -0.745 182s -0.654 -0.899 0.852 182s 0.088 0.776 -0.175 182s 182s y: [1, 3, 1, 1] 182s -1.101 182s 0.401 182s -0.551 182s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 182s x0: [1, 3, 3, 1] 182s -0.435 182s 0.270 182s 0.186 182s 182s x1: [1, 3, 3, 1] 182s -0.135 182s -0.012 182s 0.077 182s 182s y: [3, 3, 3, 1] 182s 0.059 -0.036 -0.025 182s 0.005 -0.003 -0.002 182s -0.034 0.021 0.014 182s testing: mul_mat, [1, 1, 3, 4] = [3, 1, 3, 4] * [3, 1, 3, 4] 182s x0: [3, 1, 3, 4] 182s -0.108 0.010 0.077 182s 182s x1: [3, 1, 3, 4] 182s 0.117 -0.742 -0.117 182s 182s y: [1, 1, 3, 4] 182s -0.029 182s test-mul-mat0: iter:32/500 182s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 182s x0: [2, 1, 1, 1] 182s -0.138 -0.727 182s 182s x1: [2, 2, 1, 1] 182s 0.448 -0.384 182s -0.117 -0.295 182s 182s y: [1, 2, 1, 1] 182s 0.217 182s 0.231 182s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 182s x0: [2, 4, 1, 1] 182s -0.434 -0.029 182s -0.640 -0.293 182s 0.584 0.876 182s 0.527 -0.837 182s 182s x1: [2, 1, 1, 1] 182s 0.987 -0.935 182s 182s y: [4, 1, 1, 1] 182s -0.401 -0.357 -0.243 1.303 182s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 3, 1, 1] 182s 0.489 -0.499 182s -0.456 -0.655 182s 0.949 0.160 182s 182s x1: [2, 4, 1, 1] 182s -0.231 -0.756 182s -0.184 0.009 182s 0.854 -0.520 182s -0.674 -0.173 182s 182s y: [3, 4, 1, 1] 182s 0.264 0.600 -0.341 182s -0.094 0.078 -0.173 182s 0.677 -0.049 0.727 182s -0.243 0.421 -0.668 182s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 182s x0: [2, 1, 1, 1] 182s -0.735 0.400 182s 182s x1: [2, 3, 1, 1] 182s 0.228 -0.346 182s -0.153 -0.909 182s -0.073 -0.166 182s 182s y: [1, 3, 1, 1] 182s -0.306 182s -0.251 182s -0.013 182s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 4, 1, 1] 182s -0.289 182s -0.499 182s 0.633 182s -0.127 182s 182s x1: [1, 2, 1, 1] 182s 0.126 182s -0.375 182s 182s y: [4, 2, 1, 1] 182s -0.037 -0.063 0.080 -0.016 182s 0.108 0.187 -0.237 0.047 182s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 182s x0: [4, 3, 1, 1] 182s -0.831 -0.498 0.046 -0.794 182s -0.395 -0.906 -0.560 0.668 182s -0.442 0.059 0.008 0.663 182s 182s x1: [4, 1, 1, 1] 182s 0.270 0.877 0.690 -0.721 182s 182s y: [3, 1, 1, 1] 182s -0.056 -1.769 -0.540 182s test-mul-mat0: iter:33/500 182s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 1, 1, 1] 182s 0.393 182s 182s x1: [1, 4, 1, 1] 182s 0.914 182s -0.145 182s 0.286 182s 0.040 182s 182s y: [1, 4, 1, 1] 182s 0.359 182s -0.057 182s 0.112 182s 0.016 182s testing: mul_mat, [3, 1, 3, 1] = [1, 3, 3, 1] * [1, 1, 3, 1] 182s x0: [1, 3, 3, 1] 182s 0.251 182s -0.097 182s -0.686 182s 182s x1: [1, 1, 3, 1] 182s 0.751 182s 182s y: [3, 1, 3, 1] 182s 0.189 -0.073 -0.515 182s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 182s x0: [1, 3, 3, 1] 182s 0.512 182s 0.177 182s -0.991 182s 182s x1: [1, 3, 3, 1] 182s -0.771 182s -0.749 182s -0.968 182s 182s y: [3, 3, 3, 1] 182s -0.394 -0.137 0.764 182s -0.383 -0.133 0.743 182s -0.496 -0.172 0.960 182s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 3, 1, 1] 182s 0.785 182s 0.737 182s -0.391 182s 182s x1: [1, 3, 1, 1] 182s 0.828 182s -0.118 182s 0.423 182s 182s y: [3, 3, 1, 1] 182s 0.650 0.610 -0.323 182s -0.093 -0.087 0.046 182s 0.332 0.312 -0.165 182s testing: mul_mat, [4, 1, 3, 1] = [3, 4, 3, 1] * [3, 1, 3, 1] 182s x0: [3, 4, 3, 1] 182s 0.476 -0.926 0.922 182s 0.979 0.556 -0.593 182s -0.956 0.944 0.132 182s -0.234 -0.917 0.769 182s 182s x1: [3, 1, 3, 1] 182s -0.581 -0.780 -0.256 182s 182s y: [4, 1, 3, 1] 182s 0.210 -0.851 -0.214 0.654 182s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 182s x0: [4, 4, 3, 1] 182s 0.998 0.754 0.840 0.961 182s 0.998 0.549 0.135 0.470 182s -0.651 -0.660 -0.525 0.561 182s -0.170 -0.561 0.442 0.791 182s 182s x1: [4, 3, 3, 1] 182s 0.544 0.940 -0.216 0.467 182s -0.653 0.915 0.236 -0.389 182s 0.846 -0.167 0.756 0.200 182s 182s y: [4, 3, 3, 1] 182s 1.518 1.249 -0.600 -0.346 182s -0.138 -0.300 -0.522 -0.606 182s 1.545 0.948 -0.725 0.442 182s test-mul-mat0: iter:34/500 182s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 182s x0: [3, 1, 1, 1] 182s -0.696 0.363 0.754 182s 182s x1: [3, 2, 1, 1] 182s -0.448 0.345 -0.382 182s 0.727 0.444 0.296 182s 182s y: [1, 2, 1, 1] 182s 0.148 182s -0.121 182s testing: mul_mat, [2, 1, 2, 1] = [3, 2, 2, 1] * [3, 1, 2, 1] 182s x0: [3, 2, 2, 1] 182s 0.501 -0.466 -0.957 182s -0.768 0.979 0.167 182s 182s x1: [3, 1, 2, 1] 182s 0.163 0.426 -0.632 182s 182s y: [2, 1, 2, 1] 182s 0.488 0.187 182s testing: mul_mat, [1, 2, 2, 4] = [3, 1, 2, 4] * [3, 2, 2, 4] 182s x0: [3, 1, 2, 4] 182s -0.709 0.344 0.232 182s 182s x1: [3, 2, 2, 4] 182s -0.533 -0.596 0.815 182s 0.812 0.022 0.541 182s 182s y: [1, 2, 2, 4] 182s 0.362 182s -0.443 182s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 182s x0: [3, 2, 1, 1] 182s 0.637 -0.696 -0.376 182s 0.280 0.928 0.536 182s 182s x1: [3, 1, 1, 1] 182s 0.960 0.801 0.788 182s 182s y: [2, 1, 1, 1] 182s -0.241 1.434 182s testing: mul_mat, [2, 3, 2, 1] = [2, 2, 2, 1] * [2, 3, 2, 1] 182s x0: [2, 2, 2, 1] 182s 0.713 -0.022 182s 0.280 0.713 182s 182s x1: [2, 3, 2, 1] 182s -0.285 -0.947 182s -0.036 0.166 182s -1.000 0.574 182s 182s y: [2, 3, 2, 1] 182s -0.182 -0.755 182s -0.029 0.109 182s -0.725 0.129 182s testing: mul_mat, [3, 2, 2, 4] = [2, 3, 2, 4] * [2, 2, 2, 4] 182s x0: [2, 3, 2, 4] 182s -0.759 0.850 182s 0.484 0.798 182s -0.211 0.407 182s 182s x1: [2, 2, 2, 4] 182s -0.281 0.053 182s -0.671 -0.990 182s 182s y: [3, 2, 2, 4] 182s 0.258 -0.093 0.081 182s -0.333 -1.114 -0.261 182s test-mul-mat0: iter:35/500 182s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 1, 1, 1] 182s 0.346 0.005 -0.682 -0.491 182s 182s x1: [4, 3, 1, 1] 182s 0.001 -0.027 -0.375 -0.427 182s -0.952 0.923 0.189 0.037 182s -0.305 0.904 -0.734 -0.049 182s 182s y: [1, 3, 1, 1] 182s 0.466 182s -0.472 182s 0.423 182s testing: mul_mat, [4, 1, 4, 1] = [4, 4, 4, 1] * [4, 1, 4, 1] 182s x0: [4, 4, 4, 1] 182s -0.453 0.214 0.022 0.737 182s -0.749 0.718 0.641 -0.483 182s -0.331 0.048 0.863 0.673 182s 0.366 -0.628 -0.621 0.336 182s 182s x1: [4, 1, 4, 1] 182s -0.294 0.970 -0.958 -0.131 182s 182s y: [4, 1, 4, 1] 182s 0.223 0.366 -0.771 -0.166 182s testing: mul_mat, [4, 4, 4, 3] = [4, 4, 4, 3] * [4, 4, 4, 3] 182s x0: [4, 4, 4, 3] 182s -0.176 0.174 -0.362 -0.758 182s -0.155 -0.788 -0.774 0.357 182s 0.415 0.698 0.995 0.590 182s 0.711 0.269 -0.808 0.400 182s 182s x1: [4, 4, 4, 3] 182s -0.244 0.320 -0.904 -0.473 182s -0.411 0.492 -0.033 -0.094 182s -0.249 0.080 -0.411 -0.617 182s 0.572 0.607 -0.572 0.313 182s 182s y: [4, 4, 4, 3] 182s 0.785 0.316 -1.056 0.454 182s 0.241 -0.332 0.085 -0.171 182s 0.675 0.073 -0.821 -0.070 182s -0.025 -0.013 0.277 1.158 182s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 182s x0: [4, 2, 1, 1] 182s 0.279 0.936 0.918 0.773 182s -0.475 -0.201 0.952 -0.055 182s 182s x1: [4, 4, 1, 1] 182s 0.040 -0.821 0.262 0.146 182s -0.539 -0.937 -0.201 0.310 182s -0.812 -0.619 0.888 -0.804 182s 0.949 -0.253 -0.560 -0.301 182s 182s y: [2, 4, 1, 1] 182s -0.404 0.388 182s -0.973 0.236 182s -0.612 1.400 182s -0.719 -0.916 182s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 182s x0: [2, 4, 4, 1] 182s -0.303 0.009 182s -0.398 0.345 182s 0.122 -0.932 182s -0.248 -0.437 182s 182s x1: [2, 4, 4, 1] 182s -0.675 0.024 182s 0.928 0.368 182s -0.955 -0.945 182s -0.593 -0.777 182s 182s y: [4, 4, 4, 1] 182s 0.205 0.277 -0.105 0.157 182s -0.278 -0.243 -0.230 -0.391 182s 0.281 0.054 0.765 0.651 182s 0.172 -0.032 0.651 0.487 182s testing: mul_mat, [2, 2, 4, 3] = [4, 2, 4, 3] * [4, 2, 4, 3] 182s x0: [4, 2, 4, 3] 182s -0.092 0.009 -0.303 -0.002 182s -0.261 -0.002 0.730 0.028 182s 182s x1: [4, 2, 4, 3] 182s 0.257 -0.006 0.661 -0.734 182s -0.661 0.729 -0.171 -0.632 182s 182s y: [2, 2, 4, 3] 182s -0.222 0.394 182s 0.121 0.029 182s test-mul-mat0: iter:36/500 182s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 182s x0: [3, 3, 1, 1] 182s 0.084 -0.671 0.317 182s 0.870 -0.249 0.021 182s 0.748 0.900 0.418 182s 182s x1: [3, 2, 1, 1] 182s 0.813 -0.358 -0.359 182s 0.784 0.647 -0.021 182s 182s y: [3, 2, 1, 1] 182s 0.194 0.789 0.137 182s -0.375 0.521 1.161 182s testing: mul_mat, [1, 3, 2, 1] = [3, 1, 2, 1] * [3, 3, 2, 1] 182s x0: [3, 1, 2, 1] 182s -0.301 -0.492 0.944 182s 182s x1: [3, 3, 2, 1] 182s 0.161 -0.369 0.545 182s -0.255 -0.172 -0.225 182s -0.900 0.735 0.389 182s 182s y: [1, 3, 2, 1] 182s 0.647 182s -0.052 182s 0.277 182s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 182s x0: [3, 4, 2, 1] 182s -0.882 0.245 0.308 182s -0.053 -0.980 0.408 182s -0.319 0.409 -0.030 182s -0.718 -0.924 0.138 182s 182s x1: [3, 1, 2, 1] 182s 0.653 0.588 -0.718 182s 182s y: [4, 1, 2, 1] 182s -0.653 -0.904 0.054 -1.110 182s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 182s x0: [3, 3, 1, 1] 182s -0.342 -0.059 0.632 182s -0.255 -0.178 0.786 182s 0.526 -0.336 0.782 182s 182s x1: [3, 4, 1, 1] 182s 0.876 0.425 -0.628 182s 0.277 -0.088 -0.928 182s 0.840 0.030 0.317 182s 0.148 0.977 0.336 182s 182s y: [3, 4, 1, 1] 182s -0.722 -0.793 -0.174 182s -0.676 -0.784 -0.551 182s -0.089 0.029 0.680 182s 0.104 0.052 0.013 182s testing: mul_mat, [2, 3, 2, 1] = [3, 2, 2, 1] * [3, 3, 2, 1] 182s x0: [3, 2, 2, 1] 182s -0.540 -0.263 0.425 182s -0.321 0.118 -0.737 182s 182s x1: [3, 3, 2, 1] 182s 0.374 -0.722 -0.334 182s 0.815 0.278 0.494 182s -0.962 0.115 0.583 182s 182s y: [2, 3, 2, 1] 182s -0.154 0.041 182s -0.304 -0.592 182s 0.737 -0.108 182s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 182s x0: [2, 1, 2, 1] 182s -0.304 -0.568 182s 182s x1: [2, 3, 2, 1] 182s 0.064 -0.690 182s 0.375 -0.121 182s 0.588 -0.131 182s 182s y: [1, 3, 2, 1] 182s 0.372 182s -0.045 182s -0.104 182s test-mul-mat0: iter:37/500 182s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 182s x0: [4, 1, 1, 1] 182s -0.874 0.163 -0.494 -0.843 182s 182s x1: [4, 1, 1, 1] 182s 0.974 -0.243 0.879 -0.763 182s 182s y: [1, 1, 1, 1] 182s -0.682 182s testing: mul_mat, [4, 1, 3, 1] = [4, 4, 3, 1] * [4, 1, 3, 1] 182s x0: [4, 4, 3, 1] 182s -0.777 -0.300 0.405 0.280 182s 0.879 0.181 -0.483 0.518 182s 0.636 -0.509 -0.724 0.515 182s -0.273 0.092 0.641 0.891 182s 182s x1: [4, 1, 3, 1] 182s 0.872 0.570 -0.533 0.247 182s 182s y: [4, 1, 3, 1] 182s -0.996 1.256 0.777 -0.307 182s testing: mul_mat, [2, 4, 3, 3] = [4, 2, 3, 3] * [4, 4, 3, 3] 182s x0: [4, 2, 3, 3] 182s 0.048 0.527 -0.941 -0.601 182s -0.239 -0.581 0.748 -0.731 182s 182s x1: [4, 4, 3, 3] 182s 0.842 -0.924 0.314 -0.892 182s 0.086 -0.068 -0.571 -0.787 182s 0.917 0.510 0.906 0.102 182s -0.406 -0.428 -0.491 -0.623 182s 182s y: [2, 4, 3, 3] 182s -0.206 1.223 182s 0.978 0.167 182s -0.600 0.088 182s 0.590 0.433 182s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 182s x0: [4, 1, 1, 1] 182s 0.677 -0.040 -0.264 0.564 182s 182s x1: [4, 2, 1, 1] 182s -0.661 0.033 0.004 0.410 182s 0.920 0.481 -0.707 0.026 182s 182s y: [1, 2, 1, 1] 182s -0.219 182s 0.805 182s testing: mul_mat, [2, 4, 3, 1] = [1, 2, 3, 1] * [1, 4, 3, 1] 182s x0: [1, 2, 3, 1] 182s -0.485 182s -0.712 182s 182s x1: [1, 4, 3, 1] 182s 0.014 182s 0.395 182s -0.505 182s -0.414 182s 182s y: [2, 4, 3, 1] 182s -0.007 -0.010 182s -0.191 -0.281 182s 0.245 0.360 182s 0.201 0.295 182s testing: mul_mat, [2, 1, 3, 3] = [2, 2, 3, 3] * [2, 1, 3, 3] 182s x0: [2, 2, 3, 3] 182s -0.419 0.453 182s 0.695 -0.998 182s 182s x1: [2, 1, 3, 3] 182s 0.909 -0.749 182s 182s y: [2, 1, 3, 3] 182s -0.720 1.380 182s test-mul-mat0: iter:38/500 182s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 182s x0: [2, 3, 1, 1] 182s -0.789 0.010 182s -0.182 -0.008 182s -0.005 -0.977 182s 182s x1: [2, 2, 1, 1] 182s -0.421 -0.338 182s -0.373 0.881 182s 182s y: [3, 2, 1, 1] 182s 0.329 0.079 0.332 182s 0.304 0.061 -0.858 182s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 182s x0: [2, 1, 1, 1] 182s 0.459 -0.094 182s 182s x1: [2, 3, 1, 1] 182s -0.354 -0.207 182s -0.774 -0.262 182s 0.419 -0.180 182s 182s y: [1, 3, 1, 1] 182s -0.143 182s -0.331 182s 0.209 182s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 182s x0: [2, 4, 1, 2] 182s -0.759 0.173 182s -0.800 0.427 182s -0.126 -0.856 182s -0.994 0.536 182s 182s x1: [2, 1, 1, 2] 182s -0.374 -0.275 182s 182s y: [4, 1, 1, 2] 182s 0.236 0.181 0.282 0.224 182s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 1, 1, 1] 182s -0.798 0.747 182s 182s x1: [2, 4, 1, 1] 182s -0.749 -0.380 182s -0.315 -0.524 182s 0.358 -0.895 182s 0.296 0.120 182s 182s y: [1, 4, 1, 1] 182s 0.314 182s -0.140 182s -0.954 182s -0.147 182s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 2, 1, 1] 182s 0.733 182s -0.983 182s 182s x1: [1, 2, 1, 1] 182s -0.712 182s 0.775 182s 182s y: [2, 2, 1, 1] 182s -0.522 0.700 182s 0.568 -0.762 182s testing: mul_mat, [1, 1, 1, 2] = [2, 1, 1, 2] * [2, 1, 1, 2] 182s x0: [2, 1, 1, 2] 182s 0.450 -0.485 182s 182s x1: [2, 1, 1, 2] 182s -0.894 0.933 182s 182s y: [1, 1, 1, 2] 182s -0.854 182s test-mul-mat0: iter:39/500 182s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 182s x0: [3, 3, 1, 1] 182s -0.188 -0.120 0.782 182s -0.084 -0.823 -0.098 182s -0.520 -0.621 -0.351 182s 182s x1: [3, 1, 1, 1] 182s 0.804 -0.874 -0.596 182s 182s y: [3, 1, 1, 1] 182s -0.513 0.710 0.334 182s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 2, 1, 1] 182s 0.064 0.182 0.358 182s -0.599 0.108 0.188 182s 182s x1: [3, 3, 1, 1] 182s -0.232 -0.846 0.582 182s -0.499 -0.829 0.688 182s -0.567 0.614 -0.333 182s 182s y: [2, 3, 1, 1] 182s 0.039 0.157 182s 0.063 0.339 182s -0.044 0.343 182s testing: mul_mat, [4, 2, 1, 4] = [3, 4, 1, 4] * [3, 2, 1, 4] 182s x0: [3, 4, 1, 4] 182s -0.897 0.869 -0.461 182s -0.495 -0.023 0.727 182s 0.140 -0.763 -0.281 182s 0.902 0.902 -0.864 182s 182s x1: [3, 2, 1, 4] 182s -0.365 0.261 -0.008 182s -0.238 0.664 0.416 182s 182s y: [4, 2, 1, 4] 182s 0.557 0.169 -0.248 -0.087 182s 0.599 0.405 -0.657 0.024 182s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 182s x0: [3, 2, 1, 1] 182s 0.919 -0.792 0.224 182s -0.369 0.217 -0.723 182s 182s x1: [3, 4, 1, 1] 182s -0.350 -0.994 -0.911 182s -0.164 -0.324 0.611 182s 0.354 0.583 -0.443 182s 0.891 0.161 -0.458 182s 182s y: [2, 4, 1, 1] 182s 0.262 0.572 182s 0.243 -0.452 182s -0.236 0.316 182s 0.589 0.037 182s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 182s x0: [2, 2, 1, 1] 182s 0.701 0.080 182s 0.445 -0.225 182s 182s x1: [2, 3, 1, 1] 182s -0.758 -0.043 182s -0.417 0.986 182s -0.408 -0.209 182s 182s y: [2, 3, 1, 1] 182s -0.535 -0.328 182s -0.213 -0.408 182s -0.303 -0.135 182s testing: mul_mat, [2, 2, 1, 4] = [2, 2, 1, 4] * [2, 2, 1, 4] 182s x0: [2, 2, 1, 4] 182s 0.336 0.412 182s 0.921 0.578 182s 182s x1: [2, 2, 1, 4] 182s 0.267 0.731 182s -0.219 0.356 182s 182s y: [2, 2, 1, 4] 182s 0.391 0.668 182s 0.073 0.004 182s test-mul-mat0: iter:40/500 182s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 182s x0: [1, 4, 1, 1] 182s 0.223 182s 0.347 182s -0.459 182s 0.423 182s 182s x1: [1, 1, 1, 1] 182s 0.822 182s 182s y: [4, 1, 1, 1] 182s 0.183 0.285 -0.377 0.348 182s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 182s x0: [1, 4, 3, 1] 182s -0.203 182s 0.451 182s 0.551 182s 0.029 182s 182s x1: [1, 4, 3, 1] 182s 0.414 182s 0.469 182s 0.543 182s -0.889 182s 182s y: [4, 4, 3, 1] 182s -0.084 0.186 0.228 0.012 182s -0.095 0.211 0.258 0.013 182s -0.110 0.245 0.299 0.016 182s 0.180 -0.401 -0.490 -0.026 182s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 182s x0: [1, 4, 3, 1] 182s -0.632 182s 0.933 182s 0.018 182s -0.621 182s 182s x1: [1, 4, 3, 1] 182s 0.204 182s 0.738 182s 0.630 182s -0.449 182s 182s y: [4, 4, 3, 1] 182s -0.129 0.191 0.004 -0.127 182s -0.467 0.689 0.013 -0.459 182s -0.398 0.588 0.011 -0.391 182s 0.284 -0.419 -0.008 0.279 182s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 1, 1, 1] 182s 0.076 182s 182s x1: [1, 4, 1, 1] 182s -0.268 182s 0.568 182s -0.331 182s -0.886 182s 182s y: [1, 4, 1, 1] 182s -0.020 182s 0.043 182s -0.025 182s -0.067 182s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 182s x0: [1, 2, 3, 1] 182s -0.405 182s 0.758 182s 182s x1: [1, 1, 3, 1] 182s 0.318 182s 182s y: [2, 1, 3, 1] 182s -0.129 0.242 182s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 182s x0: [2, 2, 3, 1] 182s 0.718 -0.382 182s -0.567 0.124 182s 182s x1: [2, 1, 3, 1] 182s -0.305 0.830 182s 182s y: [2, 1, 3, 1] 182s -0.536 0.275 182s test-mul-mat0: iter:41/500 182s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 182s x0: [2, 4, 1, 1] 182s -0.779 -0.381 182s 0.645 0.985 182s 0.655 0.087 182s 0.703 -0.913 182s 182s x1: [2, 4, 1, 1] 182s -0.723 -0.143 182s 0.544 -0.964 182s -0.308 0.887 182s -0.609 -0.398 182s 182s y: [4, 4, 1, 1] 182s 0.617 -0.607 -0.485 -0.378 182s -0.057 -0.598 0.272 1.262 182s -0.098 0.675 -0.124 -1.026 182s 0.626 -0.786 -0.434 -0.065 182s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 182s x0: [2, 1, 4, 1] 182s -0.278 0.937 182s 182s x1: [2, 4, 4, 1] 182s 0.706 -0.173 182s 0.469 -0.471 182s 0.316 0.297 182s -0.916 0.374 182s 182s y: [1, 4, 4, 1] 182s -0.359 182s -0.572 182s 0.190 182s 0.605 182s testing: mul_mat, [4, 1, 4, 2] = [2, 4, 4, 2] * [2, 1, 4, 2] 182s x0: [2, 4, 4, 2] 182s 0.969 -0.364 182s 0.111 0.222 182s -0.956 0.916 182s 0.944 0.981 182s 182s x1: [2, 1, 4, 2] 182s 0.502 0.178 182s 182s y: [4, 1, 4, 2] 182s 0.422 0.095 -0.317 0.648 182s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1,ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s 1] * [2, 4, 1, 1] 182s x0: [2, 3, 1, 1] 182s 0.269 0.730 182s -0.248 0.476 182s 0.697 0.476 182s 182s x1: [2, 4, 1, 1] 182s -0.916 -0.053 182s -0.214 0.944 182s -0.018 -0.504 182s 0.841 -0.020 182s 182s y: [3, 4, 1, 1] 182s -0.285 0.202 -0.664 182s 0.632 0.503 0.300 182s -0.373 -0.235 -0.252 182s 0.211 -0.218 0.577 182s testing: mul_mat, [3, 2, 4, 1] = [3, 3, 4, 1] * [3, 2, 4, 1] 182s x0: [3, 3, 4, 1] 182s -0.062 -0.332 -0.647 182s -0.775 0.701 0.674 182s -0.765 0.711 0.368 182s 182s x1: [3, 2, 4, 1] 182s 0.684 0.973 0.657 182s -0.910 -0.258 -0.358 182s 182s y: [3, 2, 4, 1] 182s -0.790 0.594 0.410 182s 0.373 0.283 0.381 182s testing: mul_mat, [4, 3, 4, 2] = [3, 4, 4, 2] * [3, 3, 4, 2] 182s x0: [3, 4, 4, 2] 182s -0.619 0.003 -0.940 182s -0.262 0.910 -0.860 182s 0.832 0.892 -0.065 182s -0.382 -0.602 -0.166 182s 182s x1: [3, 3, 4, 2] 182s 0.039 0.568 0.087 182s -0.594 -0.989 0.503 182s -0.583 -0.250 -0.729 182s 182s y: [4, 3, 4, 2] 182s -0.104 0.432 0.533 -0.371 182s -0.109 -1.177 -1.408 0.739 182s 1.046 0.552 -0.661 0.494 182s test-mul-mat0: iter:42/500 182s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 182s x0: [4, 2, 1, 1] 182s 0.110 -0.315 -0.224 0.451 182s 0.649 -0.962 0.996 0.800 182s 182s x1: [4, 4, 1, 1] 182s 0.103 -0.917 -0.438 -0.132 182s -0.603 -0.302 0.512 -0.303 182s -0.724 -0.351 0.851 -0.461 182s 0.157 -0.656 -0.052 -0.977 182s 182s y: [2, 4, 1, 1] 182s 0.338 0.407 182s -0.223 0.167 182s -0.368 0.346 182s -0.205 -0.100 182s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 182s x0: [4, 2, 3, 1] 182s 0.958 0.165 -0.233 0.996 182s 0.160 -0.433 0.002 0.760 182s 182s x1: [4, 2, 3, 1] 182s 0.005 -0.400 -0.355 0.120 182s 0.030 0.580 -0.777 0.113 182s 182s y: [2, 2, 3, 1] 182s 0.141 0.265 182s 0.418 -0.161 182s testing: mul_mat, [1, 2, 3, 3] = [4, 1, 3, 3] * [4, 2, 3, 3] 182s x0: [4, 1, 3, 3] 182s -0.437 0.808 0.855 0.442 182s 182s x1: [4, 2, 3, 3] 182s 0.702 -0.918 0.145 -0.865 182s 0.264 -0.142 0.253 0.221 182s 182s y: [1, 2, 3, 3] 182s -1.307 182s 0.085 182s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 182s x0: [4, 3, 1, 1] 182s 0.010 -0.271 -0.008 -0.474 182s 0.732 -0.173 0.037 0.515 182s 0.141 -0.286 -0.393 0.253 182s 182s x1: [4, 1, 1, 1] 182s -0.633 0.235 0.651 -0.702 182s 182s y: [3, 1, 1, 1] 182s 0.257 -0.841 -0.590 182s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 182s x0: [3, 4, 3, 1] 182s 0.217 0.118 0.792 182s -0.948 -0.157 -0.178 182s -0.740 0.736 0.611 182s 0.886 0.539 0.040 182s 182s x1: [3, 4, 3, 1] 182s 0.203 0.548 -0.653 182s 0.754 0.678 -0.193 182s -0.070 -0.532 -0.233 182s -0.331 0.066 -0.417 182s 182s y: [4, 4, 3, 1] 182s -0.408 -0.163 -0.146 0.450 182s 0.090 -0.787 -0.177 1.026 182s -0.262 0.191 -0.482 -0.358 182s -0.394 0.377 0.039 -0.274 182s testing: mul_mat, [4, 3, 3, 3] = [4, 4, 3, 3] * [4, 3, 3, 3] 182s x0: [4, 4, 3, 3] 182s 0.952 -0.672 -0.698 -0.558 182s -0.500 -0.564 -0.978 -0.398 182s 0.166 0.882 -0.175 -0.298 182s -0.989 0.345 0.563 0.551 182s 182s x1: [4, 3, 3, 3] 182s -0.603 -0.705 0.228 -0.811 182s 0.117 -0.161 0.229 -0.437 182s 0.531 0.248 0.775 -0.965 182s 182s y: [4, 3, 3, 3] 182s 0.193 0.799 -0.521 0.035 182s 0.304 -0.018 -0.033 -0.283 182s 0.337 -0.780 0.459 -0.536 182s test-mul-mat0: iter:43/500 182s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 2, 1, 1] 182s 0.386 -0.427 0.803 182s -0.777 0.222 -0.052 182s 182s x1: [3, 3, 1, 1] 182s 0.238 0.024 -0.141 182s -0.895 0.908 -0.113 182s -0.162 -0.536 0.676 182s 182s y: [2, 3, 1, 1] 182s -0.032 -0.172 182s -0.823 0.903 182s 0.709 -0.028 182s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 182s x0: [3, 1, 1, 1] 182s -0.304 0.561 -0.031 182s 182s x1: [3, 2, 1, 1] 182s 0.410 -0.618 0.647 182s -0.295 -0.464 0.598 182s 182s y: [1, 2, 1, 1] 182s -0.491 182s -0.189 182s testing: mul_mat, [3, 1, 1, 2] = [3, 3, 1, 2] * [3, 1, 1, 2] 182s x0: [3, 3, 1, 2] 182s 0.728 -0.534 -0.884 182s -0.385 0.303 -0.419 182s -0.710 0.449 0.966 182s 182s x1: [3, 1, 1, 2] 182s 0.084 0.675 0.123 182s 182s y: [3, 1, 1, 2] 182s -0.408 0.121 0.362 182s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 3, 1, 1] 182s 0.551 0.279 0.894 182s 0.496 0.962 0.265 182s -0.823 -0.707 -0.126 182s 182s x1: [3, 3, 1, 1] 182s 0.004 0.446 0.044 182s 0.700 0.007 -0.987 182s -0.216 -0.318 0.136 182s 182s y: [3, 3, 1, 1] 182s 0.166 0.442 -0.324 182s -0.495 0.092 -0.456 182s -0.086 -0.377 0.385 182s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 2, 1, 1] 182s -0.985 -0.395 0.976 182s -0.068 0.461 -0.695 182s 182s x1: [3, 3, 1, 1] 182s -0.816 -0.286 -0.160 182s 0.048 0.967 0.029 182s 0.134 -0.833 -0.372 182s 182s y: [2, 3, 1, 1] 182s 0.760 0.035 182s -0.400 0.422 182s -0.166 -0.135 182s testing: mul_mat, [4, 3, 1, 2] = [2, 4, 1, 2] * [2, 3, 1, 2] 182s x0: [2, 4, 1, 2] 182s 0.848 0.242 182s -0.598 0.104 182s -0.943 -0.471 182s -0.437 -0.729 182s 182s x1: [2, 3, 1, 2] 182s -0.533 0.989 182s 0.089 0.149 182s 0.125 0.058 182s 182s y: [4, 3, 1, 2] 182s -0.212 0.422 0.037 -0.488 182s 0.112 -0.038 -0.154 -0.148 182s 0.120 -0.069 -0.145 -0.097 182s test-mul-mat0: iter:44/500 182s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 2, 1, 1] 182s 0.544 182s -0.869 182s 182s x1: [1, 4, 1, 1] 182s 0.839 182s 0.789 182s -0.077 182s -0.104 182s 182s y: [2, 4, 1, 1] 182s 0.456 -0.729 182s 0.429 -0.686 182s -0.042 0.067 182s -0.056 0.090 182s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 182s x0: [1, 1, 2, 1] 182s -0.939 182s 182s x1: [1, 2, 2, 1] 182s -0.532 182s -0.873 182s 182s y: [1, 2, 2, 1] 182s 0.500 182s 0.820 182s testing: mul_mat, [4, 1, 2, 4] = [1, 4, 2, 4] * [1, 1, 2, 4] 182s x0: [1, 4, 2, 4] 182s -0.524 182s -0.694 182s -0.976 182s -0.352 182s 182s x1: [1, 1, 2, 4] 182s -0.462 182s 182s y: [4, 1, 2, 4] 182s 0.242 0.320 0.451 0.162 182s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 4, 1, 1] 182s -0.735 182s 0.042 182s 0.001 182s 0.342 182s 182s x1: [1, 4, 1, 1] 182s 0.015 182s 0.617 182s -0.024 182s 0.130 182s 182s y: [4, 4, 1, 1] 182s -0.011 0.001 0.000 0.005 182s -0.453 0.026 0.001 0.211 182s 0.018 -0.001 -0.000 -0.008 182s -0.095 0.005 0.000 0.044 182s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 182s x0: [4, 3, 2, 1] 182s 0.243 -0.737 -0.649 -0.797 182s -0.196 0.465 0.700 -0.591 182s -0.907 -0.633 -0.619 0.974 182s 182s x1: [4, 1, 2, 1] 182s 0.876 -0.491 -0.091 -0.745 182s 182s y: [3, 1, 2, 1] 182s 1.227 -0.023 -1.154 182s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 182s x0: [3, 2, 2, 4] 182s -0.331 0.784 -0.972 182s 0.337 0.565 -0.548 182s 182s x1: [3, 4, 2, 4] 182s -0.977 0.727 0.639 182s -0.033 -0.511 0.084 182s -0.713 0.732 0.888 182s -0.620 0.995 0.352 182s 182s y: [2, 4, 2, 4] 182s 0.273 -0.269 182s -0.472 -0.346 182s -0.053 -0.313 182s 0.643 0.160 182s test-mul-mat0: iter:45/500 182s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 2, 1, 1] 182s 0.465 -0.568 0.137 182s -0.043 0.188 0.837 182s 182s x1: [3, 3, 1, 1] 182s 0.437 0.248 0.326 182s -0.795 -0.720 -0.687 182s -0.576 0.538 0.704 182s 182s y: [2, 3, 1, 1] 182s 0.107 0.301 182s -0.055 -0.676 182s -0.477 0.715 182s testing: mul_mat, [1, 2, 2, 1] = [3, 1, 2, 1] * [3, 2, 2, 1] 182s x0: [3, 1, 2, 1] 182s 0.840 -0.012 -0.543 182s 182s x1: [3, 2, 2, 1] 182s 0.359 0.773 0.534 182s -0.189 -0.851 0.775 182s 182s y: [1, 2, 2, 1] 182s 0.002 182s -0.569 182s testing: mul_mat, [3, 1, 2, 2] = [3, 3, 2, 2] * [3, 1, 2, 2] 182s x0: [3, 3, 2, 2] 182s 0.725 -0.518 0.497 182s 0.873 -0.743 -0.705 182s 0.772 -0.529 0.524 182s 182s x1: [3, 1, 2, 2] 182s 0.368 -0.372 0.086 182s 182s y: [3, 1, 2, 2] 182s 0.502 0.537 0.526 182s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 182s x0: [3, 4, 1, 1] 182s 0.562 0.337 -0.529 182s 0.626 0.748 0.298 182s -0.001 -0.163 -0.814 182s -0.253 0.380 -0.694 182s 182s x1: [3, 3, 1, 1] 182s -0.733 -0.871 0.993 182s 0.791 0.017 0.773 182s -0.665 0.746 -0.458 182s 182s y: [4, 3, 1, 1] 182s -1.230 -0.815 -0.665 -0.834 182s 0.042 0.738 -0.633 -0.730 182s 0.119 0.005 0.252 0.770 182s testing: mul_mat, [3, 3, 2, 1] = [4, 3, 2, 1] * [4, 3, 2, 1] 182s x0: [4, 3, 2, 1] 182s 0.821 0.119 -0.794 0.767 182s -0.474 -0.288 0.656 0.083 182s -0.519 -0.213 0.875 -0.771 182s 182s x1: [4, 3, 2, 1] 182s 0.087 -0.056 0.089 0.562 182s 0.427 -0.647 0.534 0.783 182s 0.875 0.801 0.912 0.867 182s 182s y: [3, 3, 2, 1] 182s 0.425 0.080 -0.388 182s 0.451 0.399 -0.221 182s 0.754 0.026 -0.495 182s testing: mul_mat, [2, 4, 2, 2] = [3, 2, 2, 2] * [3, 4, 2, 2] 182s x0: [3, 2, 2, 2] 182s 0.282 -0.425 -0.318 182s 0.760 -0.090 -0.053 182s 182s x1: [3, 4, 2, 2] 182s 0.882 -0.378 0.603 182s -0.857 0.607 -0.252 182s -0.848 0.428 0.274 182s -0.367 -0.453 0.986 182s 182s y: [2, 4, 2, 2] 182s 0.218 0.673 182s -0.419 -0.693 182s -0.508 -0.698 182s -0.225 -0.290 182s test-mul-mat0: iter:46/500 182s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 4, 1, 1] 182s -0.302 0.209 0.723 0.344 182s 0.790 0.051 -0.749 0.066 182s 0.733 0.568 -0.536 0.044 182s 0.931 -0.469 -0.030 -0.244 182s 182s x1: [4, 3, 1, 1] 182s -0.416 0.429 0.165 0.866 182s 0.189 0.740 -0.224 0.871 182s -0.313 0.284 -0.833 0.715 182s 182s y: [4, 3, 1, 1] 182s 0.633 -0.373 -0.111 -0.804 182s 0.236 0.413 0.718 -0.377 182s -0.202 0.439 0.410 -0.573 182s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 182s x0: [4, 4, 4, 1] 182s -0.787 0.908 -0.581 0.837 182s -0.776 -0.828 0.969 0.285 182s 0.191 -0.225 -0.194 -0.521 182s -0.047 0.996 0.758 0.848 182s 182s x1: [4, 4, 4, 1] 182s -0.ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s 981 0.152 0.341 0.448 182s -0.683 0.207 -0.363 -0.943 182s 0.983 -0.493 -0.256 0.267 182s -0.326 -0.540 -0.732 0.372 182s 182s y: [4, 4, 4, 1] 182s 1.087 1.092 -0.521 0.835 182s 0.147 -0.263 0.385 -0.837 182s -0.849 -0.526 0.209 -0.504 182s 0.503 0.097 0.007 -0.762 182s testing: mul_mat, [3, 4, 4, 4] = [4, 3, 4, 4] * [4, 4, 4, 4] 182s x0: [4, 3, 4, 4] 182s -0.406 -0.789 -0.729 -0.089 182s 0.370 -0.192 0.631 0.140 182s 0.397 -0.766 -0.113 0.991 182s 182s x1: [4, 4, 4, 4] 182s -0.835 -0.259 0.245 -0.539 182s 0.657 -0.892 0.947 0.348 182s 0.778 0.913 0.992 0.443 182s 0.636 -0.517 0.754 0.438 182s 182s y: [3, 4, 4, 4] 182s 0.412 -0.181 -0.696 182s -0.284 1.061 1.182 182s -1.798 0.800 -0.064 182s -0.439 0.872 0.998 182s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 3, 1, 1] 182s 0.016 -0.898 0.063 -0.288 182s -0.768 0.826 -0.929 0.015 182s 0.556 -0.893 -0.281 -0.854 182s 182s x1: [4, 3, 1, 1] 182s -0.655 -0.401 0.877 -0.888 182s -0.011 0.539 -0.132 0.845 182s -0.070 0.426 0.762 -0.767 182s 182s y: [3, 3, 1, 1] 182s 0.660 -0.655 0.505 182s -0.735 0.588 -1.171 182s -0.116 -0.313 0.021 182s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 182s x0: [3, 2, 4, 1] 182s -0.877 -0.698 0.263 182s 0.256 0.864 0.730 182s 182s x1: [3, 4, 4, 1] 182s -0.632 0.556 0.811 182s -0.017 0.693 -0.208 182s 0.327 -0.708 -0.331 182s 0.439 0.281 -0.793 182s 182s y: [2, 4, 4, 1] 182s 0.379 0.911 182s -0.523 0.442 182s 0.121 -0.770 182s -0.789 -0.224 182s testing: mul_mat, [2, 3, 4, 4] = [2, 2, 4, 4] * [2, 3, 4, 4] 182s x0: [2, 2, 4, 4] 182s 0.265 0.579 182s 0.598 -0.320 182s 182s x1: [2, 3, 4, 4] 182s 0.010 0.959 182s 0.431 -0.661 182s 0.899 -0.217 182s 182s y: [2, 3, 4, 4] 182s 0.558 -0.301 182s -0.268 0.469 182s 0.112 0.607 182s test-mul-mat0: iter:47/500 182s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 182s x0: [4, 1, 1, 1] 182s -0.113 -0.815 0.221 0.828 182s 182s x1: [4, 1, 1, 1] 182s 0.925 0.158 0.018 0.911 182s 182s y: [1, 1, 1, 1] 182s 0.524 182s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 182s x0: [4, 2, 2, 1] 182s -0.826 0.402 -0.393 -0.119 182s -0.834 -0.355 0.208 0.904 182s 182s x1: [4, 1, 2, 1] 182s -0.290 -0.975 0.424 -0.045 182s 182s y: [2, 1, 2, 1] 182s -0.313 0.636 182s testing: mul_mat, [3, 2, 2, 2] = [4, 3, 2, 2] * [4, 2, 2, 2] 182s x0: [4, 3, 2, 2] 182s 0.273 -0.275 -0.248 0.421 182s -0.247 -0.567 -0.007 0.736 182s -0.631 -0.693 0.841 -0.347 182s 182s x1: [4, 2, 2, 2] 182s 0.132 -0.566 -0.679 0.318 182s 0.655 -0.852 -0.972 0.680 182s 182s y: [3, 2, 2, 2] 182s 0.493 0.527 -0.374 182s 0.940 0.829 -0.877 182s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 3, 1, 1] 182s -0.301 -0.117 0.500 -0.513 182s 0.532 0.448 -0.642 0.575 182s -0.522 -0.363 -0.467 -0.730 182s 182s x1: [4, 3, 1, 1] 182s -0.831 -0.119 -0.210 -0.885 182s 0.261 0.543 0.115 -0.068 182s 0.814 -0.611 0.267 0.036 182s 182s y: [3, 3, 1, 1] 182s 0.613 -0.869 1.221 182s -0.050 0.270 -0.337 182s -0.059 0.008 -0.355 182s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 182s x0: [3, 3, 2, 1] 182s 0.811 -0.702 -0.576 182s 0.789 0.364 -0.824 182s -0.388 -0.118 0.011 182s 182s x1: [3, 4, 2, 1] 182s -0.874 -0.189 -0.872 182s -0.896 -0.613 0.268 182s -0.727 0.268 -0.942 182s -0.612 -0.471 0.601 182s 182s y: [3, 4, 2, 1] 182s -0.075 -0.040 0.353 182s -0.451 -1.151 0.423 182s -0.235 0.300 0.241 182s -0.512 -1.150 0.299 182s testing: mul_mat, [2, 3, 2, 2] = [3, 2, 2, 2] * [3, 3, 2, 2] 182s x0: [3, 2, 2, 2] 182s -0.834 0.066 0.941 182s 0.652 0.155 -0.716 182s 182s x1: [3, 3, 2, 2] 182s -0.307 0.822 0.352 182s -0.415 -0.449 0.803 182s 0.548 0.978 0.786 182s 182s y: [2, 3, 2, 2] 182s 0.642 -0.326 182s 1.073 -0.916 182s 0.348 -0.054 182s test-mul-mat0: iter:48/500 182s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 182s x0: [2, 1, 1, 1] 182s -0.079 -0.991 182s 182s x1: [2, 1, 1, 1] 182s -0.245 -0.644 182s 182s y: [1, 1, 1, 1] 182s 0.657 182s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 182s x0: [2, 3, 3, 1] 182s -0.802 -0.567 182s 0.252 -0.341 182s 0.695 0.460 182s 182s x1: [2, 1, 3, 1] 182s -0.984 -0.924 182s 182s y: [3, 1, 3, 1] 182s 1.313 0.067 -1.109 182s testing: mul_mat, [3, 3, 3, 4] = [2, 3, 3, 4] * [2, 3, 3, 4] 182s x0: [2, 3, 3, 4] 182s -0.528 -0.137 182s -0.385 -0.297 182s 0.013 0.471 182s 182s x1: [2, 3, 3, 4] 182s -0.494 -0.931 182s 0.634 0.456 182s -0.010 0.642 182s 182s y: [3, 3, 3, 4] 182s 0.388 0.467 -0.445 182s -0.397 -0.379 0.223 182s -0.083 -0.187 0.302 182s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 182s x0: [2, 3, 1, 1] 182s 0.481 -0.955 182s -0.386 -0.525 182s -0.524 0.622 182s 182s x1: [2, 3, 1, 1] 182s 0.137 -0.281 182s 0.719 0.133 182s -0.019 0.216 182s 182s y: [3, 3, 1, 1] 182s 0.335 0.095 -0.247 182s 0.219 -0.347 -0.294 182s -0.216 -0.106 0.145 182s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 182s x0: [3, 3, 3, 1] 182s -0.914 0.066 -0.453 182s -0.346 0.870 -0.515 182s -0.053 -0.208 0.269 182s 182s x1: [3, 2, 3, 1] 182s 0.427 0.998 -0.086 182s -0.217 -0.905 -0.389 182s 182s y: [3, 2, 3, 1] 182s -0.286 0.766 -0.253 182s 0.314 -0.512 0.095 182s testing: mul_mat, [1, 3, 3, 4] = [3, 1, 3, 4] * [3, 3, 3, 4] 182s x0: [3, 1, 3, 4] 182s 0.341 -0.576 0.745 182s 182s x1: [3, 3, 3, 4] 182s 0.448 -0.242 0.594 182s 0.262 -0.157 -0.752 182s -0.791 0.910 -0.881 182s 182s y: [1, 3, 3, 4] 182s 0.735 182s -0.381 182s -1.450 182s test-mul-mat0: iter:49/500 182s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 182s x0: [4, 4, 1, 1] 182s -0.020 -0.660 0.251 -0.026 182s 0.386 0.840 0.554 -0.951 182s -0.626 0.475 0.137 -0.346 182s -0.426 -0.741 0.567 0.887 182s 182s x1: [4, 3, 1, 1] 182s -0.618 -0.523 0.061 0.514 182s -0.242 0.144 -0.239 -0.279 182s 0.511 -0.727 -0.972 0.337 182s 182s y: [4, 3, 1, 1] 182s 0.360 -1.133 -0.031 1.142 182s -0.143 0.160 0.284 -0.387 182s 0.218 -1.272 -0.914 0.068 182s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 182s x0: [4, 2, 3, 1] 182s 0.291 0.755 -0.626 0.989 182s -0.341 0.168 -0.838 -0.301 182s 182s x1: [4, 4, 3, 1] 182s 0.205 -0.923 -0.731 0.682 182s 0.138 0.783 -0.560 -0.719 182s -0.456 0.161 0.792 -0.183 182s 0.189 0.129 0.329 -0.831 182s 182s y: [2, 4, 3, 1] 182s 0.495 0.182 182s 0.270 0.770 182s -0.688 -0.426 182s -0.875 -0.069 182s testing: mul_mat, [2, 2, 3, 2] = [4, 2, 3, 2] * [4, 2, 3, 2] 182s x0: [4, 2, 3, 2] 182s -0.193 0.556 -0.785 0.057 182s -0.875 -0.385 -0.741 0.824 182s 182s x1: [4, 2, 3, 2] 182s -0.592 -0.916 0.197 0.783 182s -0.768 -0.903 0.738 0.523 182s 182s y: [2, 2, 3, 2] 182s -0.505 1.370 182s -0.903 0.904 182s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 182s x0: [4, 3, 1, 1] 182s 0.556 -0.122 0.757 -0.240 182s 0.767 -0.061 -0.674 -0.673 182s -0.607 -0.443 0.777 0.256 182s 182s x1: [4, 2, 1, 1] 182s -0.242 0.446 -0.736 -0.376 182s 0.729 -0.041 0.665 -0.900 182s 182s y: [3, 2, 1, 1] 182s -0.656 0.537 -0.719 182s 1.129 0.719 -0.138 182s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 182s x0: [3, 1, 3, 1] 182s -0.040 -0.690 -0.493 182s 182s x1: [3, 4, 3, 1] 182s 0.588 -0.338 0.659 182s -0.699 0.153 0.416 182s 0.367 0.707 -0.846 182s 0.037 -0.535 0.600 182s 182s y: [1, 4, 3, 1] 182s -0.115 182s -0.282 182s -0.086 182s 0.072 182s testing: mul_mat, [2, 3, 3, 2] = [1, 2, 3, 2] * [1, 3, 3, 2] 182s x0: [1, 2, 3, 2] 182s 0.648 182s 0.498 182s 182s x1: [1, 3, 3, 2] 182s 0.555 182s -0.384 182s -0.135 182s 182s y: [2, 3, 3, 2] 182s 0.360 0.277 182s -0.248 -0.191 182s -0.088 -0.067 182s test-mul-mat0: iter:50/500 182s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 182s x0: [1, 4, 1, 1] 182s -0.899 182s -0.452 182s 0.631 182s 0.158 182s 182s x1: [1, 2, 1, 1] 182s 0.988 182s -0.649 182s 182s y: [4, 2, 1, 1] 182s -0.888 -0.447 0.624 0.156 182s 0.584 0.294 -0.410 -0.102 182s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 182s x0: [1, 4, 2, 1] 182s 0.045 182s -0.077 182s -0.771 182s -0.147 182s 182s x1: [1, 4, 2, 1] 182s 0.205 182s -0.561 182s 0.920 182s -0.349 182s 182s y: [4, 4, 2, 1] 182s 0.009 -0.016 -0.158 -0.030 182s -0.025 0.043 0.433 0.082 182s 0.041 -0.071 -0.709 -0.135 182s -0.016 0.027 0.269 0.051 182s testing: mul_mat, [3, 4, 2, 4] = [1, 3, 2, 4] * [1, 4, 2, 4] 182s x0: [1, 3, 2, 4] 182s -0.769 182s 0.692 182s -0.029 182s 182s x1: [1, 4, 2, 4] 182s -0.266 182s 0.882 182s -0.779 182s -0.810 182s 182s y: [3, 4, 2, 4] 182s 0.205 -0.184 0.008 182s -0.678 0.610 -0.025 182s 0.599 -0.539 0.022 182s 0.623 -0.560 0.023 182s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 3, 1, 1] 182s -0.021 182s 0.228 182s -0.554 182s 182s x1: [1, 3, 1, 1] 182s -0.728 182s -0.031 182s 0.726 182s 182s y: [3, 3, 1, 1] 182s 0.015 -0.166 0.404 182s 0.001 -0.007 0.017 182s -0.015 0.165 -0.403 182s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 182s x0: [3, 4, 2, 1] 182s -0.358 0.379 0.086 182s -0.745 -0.806 -0.070 182s -0.433 0.071 0.933 182s 0.149 -0.386 -0.209 182s 182s x1: [3, 1, 2, 1] 182s 0.210 -0.080 0.417 182s 182s y: [4, 1, 2, 1] 182s -0.070 -0.121 0.292 -0.025 182s testing: mul_mat, [4, 3, 2, 4] = [4, 4, 2, 4] * [4, 3, 2, 4] 182s x0: [4, 4, 2, 4] 182s -0.582 0.638 0.731 -0.137 182s 0.736 -0.733 0.989 -0.025 182s -0.492 -0.738 -0.181 -0.276 182s -0.009 0.842 0.439 0.847 182s 182s x1: [4, 3, 2, 4] 182s 0.187 0.788 -0.096 0.139 182s -0.103 -0.776 -0.307 0.539 182s -0.521 0.260 -0.312 0.858 182s 182s y: [4, 3, 2, 4] 182s 0.305 -0.538 -0.695 0.738 182s -0.734 0.176 0.531 -0.331 182s 0.124 -0.905 -0.116 0.814 182s test-mul-mat0: iter:51/500 182s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 182s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 182s , 1, 1] 182s x0: [1, 3, 1, 1] 182s 0.654 182s 0.527 182s -0.578 182s 182s x1: [1, 3, 1, 1] 182s -0.040 182s -0.557 182s -0.412 182s 182s y: [3, 3, 1, 1] 182s -0.026 -0.021 0.023 182s -0.364 -0.294 0.322 182s -0.269 -0.217 0.238 182s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 182s x0: [1, 4, 2, 1] 182s 0.825 182s 0.380 182s -0.512 182s -0.302 182s 182s x1: [1, 3, 2, 1] 182s 0.132 182s -0.412 182s 0.731 182s 182s y: [4, 3, 2, 1] 182s 0.109 0.050 -0.068 -0.040 182s -0.340 -0.157 0.211 0.124 182s 0.603 0.278 -0.374 -0.220 182s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 182s x0: [1, 3, 2, 1] 182s -0.563 182s -0.151 182s 0.512 182s 182s x1: [1, 4, 2, 1] 182s -0.942 182s 0.280 182s 0.164 182s -0.023 182s 182s y: [3, 4, 2, 1] 182s 0.530 0.143 -0.482 182s -0.158 -0.042 0.143 182s -0.092 -0.025 0.084 182s 0.013 0.003 -0.012 182s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 1, 1, 1] 182s 0.858 182s 182s x1: [1, 3, 1, 1] 182s 0.954 182s 0.635 182s 0.627 182s 182s y: [1, 3, 1, 1] 182s 0.819 182s 0.545 182s 0.538 182s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 182s x0: [1, 1, 2, 1] 182s 0.554 182s 182s x1: [1, 1, 2, 1] 182s 0.729 182s 182s y: [1, 1, 2, 1] 182s 0.404 182s testing: mul_mat, [3, 1, 2, 1] = [1, 3, 2, 1] * [1, 1, 2, 1] 182s x0: [1, 3, 2, 1] 182s -0.064 182s -0.044 182s -0.150 182s 182s x1: [1, 1, 2, 1] 182s 0.742 182s 182s y: [3, 1, 2, 1] 182s -0.047 -0.033 -0.111 182s test-mul-mat0: iter:52/500 182s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 182s x0: [4, 3, 1, 1] 182s 0.942 -0.581 0.056 0.801 182s -0.852 0.930 0.054 0.702 182s -0.815 -0.204 0.954 -0.485 182s 182s x1: [4, 2, 1, 1] 182s -0.043 -0.758 0.886 0.805 182s 0.753 0.418 -0.692 -0.208 182s 182s y: [3, 2, 1, 1] 182s 1.094 -0.055 0.644 182s 0.262 -0.437 -1.259 182s testing: mul_mat, [1, 3, 2, 1] = [4, 1, 2, 1] * [4, 3, 2, 1] 182s x0: [4, 1, 2, 1] 182s 0.565 0.752 0.991 -0.639 182s 182s x1: [4, 3, 2, 1] 182s 0.732 -0.090 0.364 -0.274 182s 0.146 0.378 -0.304 0.698 182s -0.304 -0.855 -0.346 -0.062 182s 182s y: [1, 3, 2, 1] 182s 0.882 182s -0.379 182s -1.117 182s testing: mul_mat, [4, 1, 2, 2] = [4, 4, 2, 2] * [4, 1, 2, 2] 182s x0: [4, 4, 2, 2] 182s -0.947 -0.041 0.833 -0.436 182s 0.009 0.639 -0.871 -0.239 182s 0.630 -0.511 -0.532 -0.865 182s -0.418 0.084 -0.995 0.400 182s 182s x1: [4, 1, 2, 2] 182s 0.870 -0.182 0.763 0.248 182s 182s y: [4, 1, 2, 2] 182s -0.288 -0.833 0.020 -1.039 182s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 182s x0: [4, 3, 1, 1] 182s -0.862 -0.293 -0.726 -0.816 182s -0.331 0.798 -0.423 -0.936 182s 0.168 -0.520 0.195 0.271 182s 182s x1: [4, 4, 1, 1] 182s 0.269 0.263 0.828 -0.898 182s -0.649 0.265 -0.435 0.928 182s 0.169 0.162 -0.914 -0.843 182s 0.764 0.646 0.994 0.619 182s 182s y: [3, 4, 1, 1] 182s -0.178 0.611 -0.173 182s 0.041 -0.258 -0.081 182s 1.158 1.249 -0.462 182s -2.074 -0.737 0.154 182s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 182s x0: [3, 3, 2, 1] 182s -0.251 -0.067 0.421 182s -0.204 -0.140 -0.924 182s -0.275 0.995 -0.465 182s 182s x1: [3, 4, 2, 1] 182s -0.511 0.217 -0.460 182s 0.752 0.045 -0.358 182s -0.897 -0.690 0.207 182s -0.970 0.479 -0.631 182s 182s y: [3, 4, 2, 1] 182s -0.080 0.499 0.571 182s -0.343 0.171 0.004 182s 0.359 0.089 -0.536 182s -0.055 0.714 1.037 182s testing: mul_mat, [3, 3, 2, 2] = [3, 3, 2, 2] * [3, 3, 2, 2] 182s x0: [3, 3, 2, 2] 182s -0.049 0.067 -0.288 182s 0.137 0.556 -0.276 182s 0.791 0.067 0.140 182s 182s x1: [3, 3, 2, 2] 182s 0.423 -0.454 -0.115 182s 0.185 -0.824 -0.363 182s 0.066 0.944 0.058 182s 182s y: [3, 3, 2, 2] 182s -0.018 -0.163 0.288 182s 0.041 -0.333 0.040 182s 0.043 0.518 0.124 182s test-mul-mat0: iter:53/500 182s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 182s x0: [1, 3, 1, 1] 182s -0.956 182s -0.929 182s -0.520 182s 182s x1: [1, 3, 1, 1] 182s -0.311 182s 0.697 182s -0.009 182s 182s y: [3, 3, 1, 1] 182s 0.297 0.289 0.162 182s -0.666 -0.648 -0.362 182s 0.009 0.009 0.005 182s testing: mul_mat, [4, 3, 3, 1] = [1, 4, 3, 1] * [1, 3, 3, 1] 182s x0: [1, 4, 3, 1] 182s 0.556 182s -0.938 182s -0.736 182s 0.011 182s 182s x1: [1, 3, 3, 1] 182s 0.087 182s 0.281 182s -0.098 182s 182s y: [4, 3, 3, 1] 182s 0.048 -0.082 -0.064 0.001 182s 0.156 -0.264 -0.207 0.003 182s -0.055 0.092 0.072 -0.001 182s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 182s x0: [1, 4, 3, 1] 182s -0.211 182s -0.415 182s 0.946 182s -0.915 182s 182s x1: [1, 4, 3, 1] 182s -0.378 182s -0.606 182s 0.285 182s -0.681 182s 182s y: [4, 4, 3, 1] 182s 0.080 0.157 -0.357 0.346 182s 0.128 0.252 -0.573 0.554 182s -0.060 -0.118 0.270 -0.261 182s 0.143 0.283 -0.644 0.623 182s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 182s x0: [1, 2, 1, 1] 182s 0.273 182s -0.373 182s 182s x1: [1, 4, 1, 1] 182s -0.746 182s 0.201 182s -0.323 182s -0.995 182s 182s y: [2, 4, 1, 1] 182s -0.204 0.278 182s 0.055 -0.075 182s -0.088 0.121 182s -0.272 0.372 182s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 182s x0: [2, 3, 3, 1] 182s 0.879 -0.319 182s -0.240 -0.687 182s -0.353 0.437 182s 182s x1: [2, 1, 3, 1] 182s 0.070 -0.442 182s 182s y: [3, 1, 3, 1] 182s 0.203 0.287 -0.218 182s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 182s x0: [3, 2, 3, 1] 182s 0.187 -0.332 0.879 182s -0.712 -0.089 0.832 182s 182s x1: [3, 2, 3, 1] 182s -0.049 0.552 -0.925 182s 0.627 -0.175 -0.299 182s 182s y: [2, 2, 3, 1] 182s -1.006 -0.784 182s -0.088 -0.679 182s test-mul-mat0: iter:54/500 182s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 182s x0: [4, 1, 1, 1] 182s 0.412 0.133 0.917 -0.677 182s 182s x1: [4, 2, 1, 1] 182s 0.064 0.098 -0.606 -0.695 182s -0.185 -0.897 -0.774 -0.155 182s 182s y: [1, 2, 1, 1] 182s -0.046 182s -0.801 182s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 182s x0: [4, 1, 2, 1] 182s 0.514 -0.094 0.726 0.833 182s 182s x1: [4, 1, 2, 1] 182s 0.013 0.749 0.479 -0.921 182s 182s y: [1, 1, 2, 1] 182s -0.484 182s testing: mul_mat, [4, 1, 2, 2] = [4, 4, 2, 2] * [4, 1, 2, 2] 182s x0: [4, 4, 2, 2] 182s -0.477 -0.952 -0.269 -0.457 182s 0.162 -0.027 -0.943 -0.932 182s -0.300 0.890 0.467 -0.394 182s 0.427 0.007 0.845 -0.973 182s 182s x1: [4, 1, 2, 2] 182s 0.239 -0.400 -0.362 0.633 182s 183s y: [4, 1, 2, 2] 183s 0.075 -0.199 -0.846 -0.823 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.633 0.282 0.760 0.511 183s 0.949 0.514 -0.819 -0.397 183s 0.435 -0.223 0.593 -0.773 183s 183s x1: [4, 4, 1, 1] 183s 0.183 0.887 0.189 -0.294 183s -0.716 -0.778 -0.071 -0.429 183s -0.278 0.757 -0.715 -0.771 183s -0.569 -0.974 0.382 -0.417 183s 183s y: [3, 4, 1, 1] 183s 0.360 0.592 0.222 183s -0.946 -0.851 0.151 183s -0.900 1.017 -0.118 183s -0.558 -1.187 0.518 183s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 2, 2, 1] 183s 0.247 -0.385 -0.935 183s 0.462 -0.242 -0.158 183s 183s x1: [3, 4, 2, 1] 183s 0.059 -0.173 0.411 183s -0.054 -0.983 -0.883 183s 0.230 -0.761 0.046 183s 0.801 -0.039 -0.197 183s 183s y: [2, 4, 2, 1] 183s -0.303 0.004 183s 1.191 0.352 183s 0.307 0.283 183s 0.397 0.410 183s testing: mul_mat, [2, 3, 2, 2] = [2, 2, 2, 2] * [2, 3, 2, 2] 183s x0: [2, 2, 2, 2] 183s 0.382 0.776 183s -0.263 -0.856 183s 183s x1: [2, 3, 2, 2] 183s 0.991 0.148 183s 0.415 0.792 183s -0.892 -0.782 183s 183s y: [2, 3, 2, 2] 183s 0.493 -0.387 183s 0.773 -0.788 183s -0.947 0.904 183s test-mul-mat0: iter:55/500 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.859 0.701 183s 183s x1: [2, 1, 1, 1] 183s 0.541 0.636 183s 183s y: [1, 1, 1, 1] 183s 0.910 183s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.780 0.681 183s 0.602 0.866 183s 0.827 0.737 183s 183s x1: [2, 1, 3, 1] 183s -0.489 0.639 183s 183s y: [3, 1, 3, 1] 183s 0.816 0.259 0.067 183s testing: mul_mat, [2, 3, 3, 3] = [2, 2, 3, 3] * [2, 3, 3, 3] 183s x0: [2, 2, 3, 3] 183s -0.548 0.526 183s 0.044 0.908 183s 183s x1: [2, 3, 3, 3] 183s 0.972 -0.554 183s 0.490 0.676 183s -0.695 0.191 183s 183s y: [2, 3, 3, 3] 183s -0.824 -0.461 183s 0.087 0.636 183s 0.481 0.143 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.994 0.891 183s 0.564 -0.710 183s -0.284 0.980 183s -0.719 0.186 183s 183s x1: [2, 2, 1, 1] 183s 0.592 -0.314 183s -0.110 -0.221 183s 183s y: [4, 2, 1, 1] 183s -0.869 0.557 -0.476 -0.484 183s -0.088 0.095 -0.186 0.038 183s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 3, 3, 1] 183s 0.226 0.515 -0.580 -0.861 183s 0.658 0.638 -0.537 -0.397 183s 0.456 -0.357 0.739 -0.127 183s 183s x1: [4, 2, 3, 1] 183s -0.095 -0.175 0.097 0.719 183s -0.860 0.134 0.717 -0.012 183s 183s y: [3, 2, 3, 1] 183s -0.787 -0.512 -0.001 183s -0.531 -0.861 0.092 183s testing: mul_mat, [3, 4, 3, 3] = [3, 3, 3, 3] * [3, 4, 3, 3] 183s x0: [3, 3, 3, 3] 183s -0.731 -0.772 -0.397 183s 0.887 -0.402 -0.994 183s -0.951 0.291 -0.651 183s 183s x1: [3, 4, 3, 3] 183s 0.577 0.934 -0.093 183s 0.316 -0.927 0.511 183s -0.811 0.930 0.102 183s -0.467 0.059 -0.211 183s 183s y: [3, 4, 3, 3] 183s -1.106 0.228 -0.216 183s 0.281 0.146 -0.903 183s -0.165 -1.195 0.975 183s 0.380 -0.228 0.599 183s test-mul-mat0: iter:56/500 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.384 -0.647 0.775 0.177 183s -0.869 0.252 -0.860 0.407 183s 183s x1: [4, 4, 1, 1] 183s 0.441 -0.627 -0.060 -0.468 183s -0.624 -0.293 0.367 0.987 183s 0.891 0.833 0.255 -0.283 183s -0.473 0.470 0.790 0.456 183s 183s y: [2, 4, 1, 1] 183s 0.106 -0.680 183s 0.889 0.554 183s -0.734 -0.898 183s 0.571 0.035 183s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.180 0.875 0.745 0.661 183s 183s x1: [4, 2, 4, 1] 183s -0.868 -0.012 -0.838 0.972 183s -0.702 -0.520 0.414 -0.329 183s 183s y: [1, 2, 4, 1] 183s -0.149 183s -0.491 183s testing: mul_mat, [1, 1, 4, 4] = [4, 1, 4, 4] * [4, 1, 4, 4] 183s x0: [4, 1, 4, 4] 183s 0.414 0.038 0.993 0.374 183s 183s x1: [4, 1, 4, 4] 183s 0.366 -0.451 -0.511 -0.506 183s 183s y: [1, 1, 4, 4] 183s -0.563 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.067 0.535 -0.239 0.676 183s 0.471 -0.267 0.394 0.757 183s 0.098 0.782 -0.940 0.377 183s 0.460 -0.894 0.093 -0.561 183s 183s x1: [4, 1, 1, 1] 183s 0.028 -0.081 0.355 0.055 183s 183s y: [4, 1, 1, 1] 183s -0.093 0.216 -0.374 0.088 183s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 4, 4, 1] 183s 0.090 -0.103 -0.061 0.182 183s -0.025 -0.465 0.277 -0.626 183s 0.980 0.986 -0.667 0.807 183s -0.625 -0.002 0.663 0.161 183s 183s x1: [4, 4, 4, 1] 183s 0.534 -0.778 0.618 -0.498 183s 0.164 -0.356 0.086 0.152 183s -0.071 -0.674 -0.820 0.848 183s 0.681 0.234 -0.679 -0.386 183s 183s y: [4, 4, 4, 1] 183s 0.000 0.831 -1.058 -0.002 183s 0.074 0.090 -0.125 -0.020 183s 0.268 -0.443 0.498 -0.362 183s 0.008 -0.072 1.039 -0.939 183s testing: mul_mat, [3, 4, 4, 4] = [4, 3, 4, 4] * [4, 4, 4, 4] 183s x0: [4, 3, 4, 4] 183s 0.052 0.331 0.096 -0.870 183s -0.512 -0.876 -0.506 -0.331 183s 0.130 -0.513 -0.122 -0.524 183s 183s x1: [4, 4, 4, 4] 183s 0.684 -0.568 -0.327 -0.990 183s -0.716 -0.119 0.274 0.412 183s 0.421 -0.797 0.723 -0.664 183s -0.062 0.194 0.906 0.776 183s 183s y: [3, 4, 4, 4] 183s 0.677 0.640 0.939 183s -0.409 0.196 -0.282 183s 0.405 0.337 0.724 183s -0.527 -0.854 -0.625 183s test-mul-mat0: iter:57/500 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.607 -0.385 183s 183s x1: [2, 3, 1, 1] 183s -0.162 0.510 183s 0.988 0.570 183s -0.602 0.217 183s 183s y: [1, 3, 1, 1] 183s -0.295 183s 0.380 183s -0.449 183s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.978 0.485 183s 183s x1: [2, 1, 2, 1] 183s -0.839 0.035 183s 183s y: [1, 1, 2, 1] 183s -0.803 183s testing: mul_mat, [2, 1, 2, 4] = [2, 2, 2, 4] * [2, 1, 2, 4] 183s x0: [2, 2, 2, 4] 183s 0.142 -0.761 183s 0.601 -0.251 183s 183s x1: [2, 1, 2, 4] 183s 0.213 -0.215 183s 183s y: [2, 1, 2, 4] 183s 0.194 0.182 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.836 -0.533 183s 0.401 -0.222 183s -0.803 0.798 183s -0.040 -0.327 183s 183s x1: [2, 2, 1, 1] 183s 0.458 0.340 183s -0.948 -0.687 183s 183s y: [4, 2, 1, 1] 183s 0.202 0.108 -0.097 -0.129 183s -0.426 -0.227 0.212 0.263 183s testing: mul_mat, [4, 2, 2, 1] = [4, 4, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 4, 2, 1] 183s -0.231 -0.239 0.551 0.415 183s -0.552 0.353 0.617 0.302 183s -0.748 -0.342 -0.371 -0.101 183s -0.179 -0.838 -0.672 -0.164 183s 183s x1: [4, 2, 2, 1] 183s -0.429 0.421 -0.343 -0.744 183s -0.889 0.706 -0.588 0.976 183s 183s y: [4, 2, 2, 1] 183s -0.499 -0.051 0.379 0.076 183s 0.118 0.672 0.543 -0.198 183s testing: mul_mat, [4, 4, 2, 4] = [4, 4, 2, 4] * [4, 4, 2, 4] 183s x0: [4, 4, 2, 4] 183s -0.401 -0.883 -0.558 0.762 183s 0.909 -0.011 -0.175 -0.370 183s -0.266 0.944 0.924 -0.809 183s -0.666 0.076 -0.637 0.859 183s 183s x1: [4, 4, 2, 4] 183s -0.648 0.311 0.916 0.113 183s -0.335 -0.426 0.275 -0.785 183s -0.809 0.904 -0.456 0.607 183s 0.206 0.443 -0.557 0.166 183s 183s y: [4, 4, 2, 4] 183s -0.440 -0.795 1.222 -0.031 183s -0.241 -0.059 0.576 -0.658 183s 0.244 -0.889 0.156 1.419 183s -0.037 0.219 -0.285 0.394 183s test-mul-mat0: iter:58/500 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.094 -0.879 183s -0.285 -0.037 183s 0.643 -0.185 183s -0.780 -0.818 183s 183s x1: [2, 3, 1, 1] 183s 0.769 -0.829 183s 0.736 0.785 183s -0.638 0.017 183s 183s y: [4, 3, 1, 1] 183s 0.657 -0.188 0.647 0.079 183s -0.759 -0.239 0.328 -1.215 183s 0.045 0.181 -0.413 0.484 183s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.116 0.202 183s 0.708 -0.942 183s -0.024 -0.611 183s 183s x1: [2, 4, 1, 1] 183s -0.800 0.624 183s -0.115 -0.968 183s 0.939 0.228 183s -0.423 -0.248 183s 183s y: [3, 4, 1, 1] 183s 0.034 -1.155 -0.362 183s -0.209 0.830 0.595 183s 0.154 0.451 -0.162 183s -0.099 -0.066 0.162 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.893 0.827 183s -0.104 -0.392 183s 183s x1: [2, 3, 1, 1] 183s -0.977 -0.256 183s -0.441 0.758 183s -0.471 -0.079 183s 183s y: [2, 3, 1, 1] 183s -1.084 0.202 183s 0.234 -0.252 183s -0.485 0.080 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.623 0.607 183s -0.365 0.655 183s 183s x1: [2, 2, 1, 1] 183s -0.210 -0.461 183s 0.423 0.011 183s 183s y: [2, 2, 1, 1] 183s -0.410 -0.226 183s 0.270 -0.147 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.348 -0.844 183s -0.472 0.406 183s 183s x1: [2, 2, 1, 1] 183s -0.425 -0.166 183s -0.767 0.326 183s 183s y: [2, 2, 1, 1] 183s -0.008 0.133 183s -0.543 0.494 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.187 0.962 183s -0.222 -0.329 183s -0.975 0.852 183s 183s x1: [2, 2, 1, 1] 183s 0.504 -0.456 183s 0.429 -0.751 183s 183s y: [3, 2, 1, 1] 183s -0.344 0.038 -0.880 183s -0.643 0.152 -1.059 183s test-mul-mat0: iter:59/500 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.810 0.007 0.561 183s -0.282 0.163 -0.033 183s -0.777 0.707 -0.604 183s -0.529 -0.189 0.583 183s 183s x1: [3, 3, 1, 1] 183s -0.299 0.654 -0.881 183s -0.675 -0.711 0.726 183s 0.980 -0.136 -0.440 183s 183s y: [4, 3, 1, 1] 183s -0.248 0.220 1.227 -0.479 183s 0.949 0.050 -0.417 0.914 183s -1.042 -0.284 -0.593 -0.749 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.274 0.645 -0.795 183s -0.863 -0.795 -0.583 183s -0.673 0.212 0.978 183s 183s x1: [3, 4, 1, 1] 183s 0.249 -0.164 0.545 183s 0.920 -0.312 -0.597 183s 0.198 0.331 0.801 183s 0.900 -0.015 0.919 183s 183s y: [3, 4, 1, 1] 183s -0.471 -0.402 0.331 183s 0.525 -0.198 -1.269 183s -0.369 -0.901 0.720 183s -0.495 -1.300 0.291 183s testing: mul_mat, [2, 3, 1, 2] = [3, 2, 1, 2] * [3, 3, 1, 2] 183s x0: [3, 2, 1, 2] 183s -0.266 -0.409 -0.770 183s -0.993 -0.764 -0.565 183s 183s x1: [3, 3, 1, 2] 183s 0.046 -0.625 -0.055 183s 0.268 -0.917 0.341 183s 0.739 -0.106 -0.077 183s 183s y: [2, 3, 1, 2] 183s 0.286 0.463 183s 0.041 0.242 183s -0.094 -0.609 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.005 -0.266 0.152 183s -0.961 -0.493 0.379 183s 0.514 -0.578 -0.471 183s 183s x1: [3, 2, 1, 1] 183s 0.518 -0.971 0.776 183s -0.214 -0.889 0.116 183s 183s y: [3, 2, 1, 1] 183s 0.379 0.275 0.462 183s 0.253 0.687 0.349 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.319 -0.117 -0.437 183s -0.947 0.777 0.112 183s -0.741 0.001 0.775 183s -0.028 -0.341 0.089 183s 183s x1: [3, 3, 1, 1] 183s -0.099 -0.950 -0.465 183s 0.635 -0.359 -0.235 183s 0.642 -0.122 0.200 183s 183s y: [4, 3, 1, 1] 183s 0.283 -0.696 -0.288 0.286 183s 0.347 -0.906 -0.653 0.084 183s 0.132 -0.681 -0.321 0.042 183s testing: mul_mat, [2, 3, 1, 2] = [4, 2, 1, 2] * [4, 3, 1, 2] 183s x0: [4, 2, 1, 2] 183s -0.437 -0.433 0.423 0.792 183s 0.648 0.675 0.656 0.238 183s 183s x1: [4, 3, 1, 2] 183s -0.883 0.814 -0.397 -0.148 183s -0.678 0.025 -0.996 0.700 183s 0.554 -0.095 0.751 -0.912 183s 183s y: [2, 3, 1, 2] 183s -0.252 -0.318 183s 0.419 -0.909 183s -0.606 0.570 183s test-mul-mat0: iter:60/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.759 0.262 0.938 183s -0.532 -0.177 -0.073 183s -0.528 0.386 -0.424 183s 0.039 0.061 0.999 183s 183s x1: [3, 1, 1, 1] 183s 0.778 -0.727 -0.327 183s 183s y: [4, 1, 1, 1] 183s -1.088 -0.262 -0.552 -0.341 183s testing: mul_mat, [1, 4, 2, 1] = [3, 1, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 1, 2, 1] 183s 0.242 -0.387 0.049 183s 183s x1: [3, 4, 2, 1] 183s -0.306 -0.147 0.237 183s -0.047 0.497 -0.204 183s -0.764 -0.856 -0.944 183s 0.073 0.696 0.067 183s 183s y: [1, 4, 2, 1] 183s -0.005 183s -0.214 183s 0.101 183s -0.249 183s testing: mul_mat, [2, 1, 2, 1] = [3, 2, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 2, 2, 1] 183s -0.833 0.136 0.237 183s 0.863 -0.797 0.235 183s 183s x1: [3, 1, 2, 1] 183s -0.025 -0.474 -0.716 183s 183s y: [2, 1, 2, 1] 183s -0.214 0.188 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.940 -0.693 -0.267 183s 0.686 0.734 0.282 183s 183s x1: [3, 2, 1, 1] 183s -0.986 -0.780 -0.836 183s -0.216 0.447 -0.363 183s 183s y: [2, 2, 1, 1] 183s 1.691 -1.484 183s -0.010 0.078 183s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 3, 2, 1] 183s -0.892 0.924 183s -0.144 0.475 183s 0.677 0.691 183s 183s x1: [2, 3, 2, 1] 183s -0.740 0.017 183s 0.210 -0.717 183s -0.903 0.374 183s 183s y: [3, 3, 2, 1] 183s 0.675 0.114 -0.489 183s -0.850 -0.370 -0.353 183s 1.151 0.307 -0.353 183s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 2, 2, 1] 183s 0.040 -0.309 0.130 183s 0.284 -0.524 -0.344 183s 183s x1: [3, 2, 2, 1] 183s 0.651 0.961 -0.193 183s 0.385 -0.306 -0.911 183s 183s y: [2, 2, 2, 1] 183s -0.296 -0.253 183s -0.008 0.583 183s test-mul-mat0: iter:61/500 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.741 0.194 183s -0.149 0.548 183s 0.867 0.229 183s -0.412 0.151 183s 183s x1: [2, 4, 1, 1] 183s 0.375 -0.001 183s 0.894 0.336 183s 0.806 0.279 183s -0.970 0.895 183s 183s y: [4, 4, 1, 1] 183s 0.277 -0.056 0.325 -0.154 183s 0.727 0.051 0.852 -0.318 183s 0.651 0.033 0.763 -0.290 183s -0.545 0.635 -0.636 0.535 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.066 -0.192 183s -0.174 0.433 183s 183s x1: [2, 4, 1, 1] 183s 0.921 0.064 183s -0.720 -0.423 183s 0.648 -0.734 183s -0.090 -0.844 183s 183s y: [2, 4, 1, 1] 183s -0.073 -0.133 183s 0.128 -0.058 183s 0.098 -0.431 183s 0.168 -0.350 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.761 -0.776 183s 0.566 -0.866 183s 183s x1: [2, 2, 1, 1] 183s 0.051 -0.800 183s -0.567 -0.055 183s 183s y: [2, 2, 1, 1] 183s 0.582 0.721 183s 0.474 -0.274 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.195 -0.654 183s 0.228 0.149 183s 0.443 -0.493 183s 183s x1: [2, 2, 1, 1] 183s 0.147 0.307 183s 0.328 0.999 183s 183s y: [3, 2, 1, 1] 183s -0.172 0.079 -0.086 183s -0.589 0.223 -0.347 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.570 0.445 -0.829 183s 0.120 0.004 0.645 183s 183s x1: [3, 2, 1, 1] 183s -0.373 0.726 -0.845 183s -0.107 -0.364 -0.689 183s 183s y: [2, 2, 1, 1] 183s 0.811 -0.587 183s 0.348 -0.459 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.513 -0.091 183s -0.946 -0.666 183s 0.139 0.714 183s -0.514 -0.648 183s 183s x1: [2, 3, 1, 1] 183s 0.437 0.116 183s 0.181 0.676 183s 0.340 -0.252 183s 183s y: [4, 3, 1, 1] 183s -0.234 -0.490 0.143 -0.299 183s -0.154 -0.622 0.508 -0.531 183s -0.151 -0.153 -0.133 -0.011 183s test-mul-mat0: iter:62/500 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.839 -0.280 0.558 0.893 183s 183s x1: [4, 4, 1, 1] 183s -0.412 0.014 -0.800 0.224 183s 0.325 -0.181 -0.206 -0.555 183s -0.736 0.798 -0.384 0.909 183s 0.234 0.731 0.091 -0.090 183s 183s y: [1, 4, 1, 1] 183s -0.596 183s -0.287 183s -0.243 183s -0.038 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.155 -0.126 -0.380 -0.986 183s -0.819 0.208 0.028 -0.619 183s 183s x1: [4, 1, 1, 1] 183s 0.859 -0.956 -0.198 -0.807 183s 183s y: [2, 1, 1, 1] 183s 1.124 -0.409 183s testing: mul_mat, [2, 2, 1, 2] = [4, 2, 1, 2] * [4, 2, 1, 2] 183s x0: [4, 2, 1, 2] 183s -0.804 0.748 0.818 -0.002 183s 0.941 0.576 -0.847 -0.185 183s 183s x1: [4, 2, 1, 2] 183s -0.568 -0.647 0.200 0.226 183s -0.202 0.464 0.024 0.414 183s 183s y: [2, 2, 1, 2] 183s 0.136 -1.118 183s 0.528 -0.019 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.877 0.513 0.122 -0.820 183s 183s x1: [4, 2, 1, 1] 183s -0.424 0.063 -0.654 -0.960 183s -0.913 0.760 0.413 -0.655 183s 183s y: [1, 2, 1, 1] 183s 0.368 183s 0.177 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.374 183s 0.090 183s 183s x1: [1, 4, 1, 1] 183s 0.402 183s -0.103 183s -0.624 183s 0.149 183s 183s y: [2, 4, 1, 1] 183s 0.150 0.036 183s -0.039 -0.009 183s -0.233 -0.056 183s 0.056 0.013 183s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 183s x0: [2, 4, 1, 2] 183s -0.100 0.730 183s 0.893 -0.564 183s -0.112 -0.148 183s -0.724 -0.206 183s 183s x1: [2, 1, 1, 2] 183s -0.708 0.527 183s 183s y: [4, 1, 1, 2] 183s 0.456 -0.930 0.001 0.404 183s test-mul-mat0: iter:63/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.637 -0.716 -0.038 183s 0.331 0.184 -0.145 183s 183s x1: [3, 2, 1, 1] 183s 0.467 -0.263 0.290 183s -0.818 -0.889 -0.620 183s 183s y: [2, 2, 1, 1] 183s 0.475 0.064 183s 0.139 -0.344 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.203 -0.740 0.978 183s 0.507 0.486 0.426 183s 0.688 -0.775 0.996 183s -0.951 -0.634 0.463 183s 183s x1: [3, 2, 1, 1] 183s -0.782 0.460 -0.414 183s -0.346 -0.688 0.379 183s 183s y: [4, 2, 1, 1] 183s -0.904 -0.349 -1.307 0.260 183s 0.810 -0.348 0.673 0.942 183s testing: mul_mat, [2, 4, 1, 3] = [3, 2, 1, 3] * [3, 4, 1, 3] 183s x0: [3, 2, 1, 3] 183s 0.937 -0.752 0.881 183s 0.801 0.751 0.818 183s 183s x1: [3, 4, 1, 3] 183s -0.214 0.655 0.645 183s -0.104 -0.965 -0.881 183s -0.466 -0.681 0.081 183s 0.865 0.503 0.937 183s 183s y: [2, 4, 1, 3] 183s -0.125 0.848 183s -0.147 -1.528 183s 0.147 -0.819 183s 1.256 1.837 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.649 0.753 0.690 183s -0.943 0.985 -0.767 183s 0.808 -0.392 -0.512 183s 183s x1: [3, 2, 1, 1] 183s -0.480 -0.876 0.925 183s -0.908 0.844 -0.837 183s 183s y: [3, 2, 1, 1] 183s 0.289 -1.120 -0.517 183s 0.648 2.330 -0.637 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.158 0.609 0.733 183s 0.292 0.344 0.269 183s 0.734 -0.746 -0.654 183s 183s x1: [3, 3, 1, 1] 183s 0.491 0.985 -0.694 183s 0.376 -0.300 -0.354 183s 0.997 -0.310 -0.112 183s 183s y: [3, 3, 1, 1] 183s 0.169 0.296 0.080 183s -0.383 -0.089 0.731 183s -0.114 0.154 1.036 183s testing: mul_mat, [1, 3, 1, 3] = [3, 1, 1, 3] * [3, 3, 1, 3] 183s x0: [3, 1, 1, 3] 183s 0.582 -0.877 0.359 183s 183s x1: [3, 3, 1, 3] 183s 0.577 0.432 0.474 183s 0.928 0.488 0.282 183s 0.682 0.474 0.889 183s 183s y: [1, 3, 1, 3] 183s 0.127 183s 0.212 183s 0.300 183s test-mul-mat0: iter:64/500 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.243 183s -0.694 183s -0.606 183s 183s x1: [1, 4, 1, 1] 183s -0.777 183s -0.619 183s 0.004 183s 0.915 183s 183s y: [3, 4, 1, 1] 183s -0.189 0.539 0.471 183s -0.150 0.429 0.375 183s 0.001 -0.003 -0.002 183s 0.222 -0.634 -0.554 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.627 183s -0.878 183s 0.262 183s 183s x1: [1, 3, 1, 1] 183s 0.478 183s 0.388 183s -0.388 183s 183s y: [3, 3, 1, 1] 183s 0.300 -0.420 0.125 183s 0.243 -0.341 0.102 183s -0.243 0.341 -0.102 183s testing: mul_mat, [1, 3, 1, 3] = [1, 1, 1, 3] * [1, 3, 1, 3] 183s x0: [1, 1, 1, 3] 183s 0.695 183s 183s x1: [1, 3, 1, 3] 183s 0.476 183s -0.024 183s -0.725 183s 183s y: [1, 3, 1, 3] 183s 0.331 183s -0.017 183s -0.504 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.787 183s 183s x1: [1, 1, 1, 1] 183s -0.789 183s 183s y: [1, 1, 1, 1] 183s 0.621 183s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.217 183s -0.652 183s -0.970 183s 183s x1: [1, 1, 1, 1] 183s -0.566 183s 183s y: [3, 1, 1, 1] 183s -0.123 0.369 0.549 183s testing: mul_mat, [2, 1, 1, 3] = [3, 2, 1, 3] * [3, 1, 1, 3] 183s x0: [3, 2, 1, 3] 183s 0.146 0.978 0.232 183s -0.866 -0.797 0.612 183s 183s x1: [3, 1, 1, 3] 183s -0.541 -0.345 -0.576 183s 183s y: [2, 1, 1, 3] 183s -0.550 0.391 183s test-mul-mat0: iter:65/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.435 -0.569 0.691 -0.904 183s -0.774 0.481 0.342 -0.766 183s 0.705 0.553 0.459 -0.722 183s 0.795 0.143 0.505 -0.539 183s 183s x1: [4, 4, 1, 1] 183s 0.227 -0.334 0.622 0.270 183s 0.702 -0.938 -0.060 -0.140 183s 0.765 -0.914 -0.006 0.743 183s -0.711 -0.774 0.355 -0.291 183s 183s y: [4, 4, 1, 1] 183s 0.474 -0.331 0.066 0.301 183s 0.924 -0.907 0.050 0.469 183s 0.176 -1.602 -0.505 0.073 183s 0.640 0.522 -0.556 -0.339 183s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 2, 3, 1] 183s -0.149 -0.112 0.739 0.414 183s -0.420 -0.030 0.531 0.731 183s 183s x1: [4, 4, 3, 1] 183s -0.235 -0.224 -0.838 -0.173 183s 0.716 0.022 -0.408 0.802 183s -0.984 -0.665 -0.909 -0.758 183s 0.690 -0.200 -0.647 0.125 183s 183s y: [2, 4, 3, 1] 183s -0.630 -0.465 183s -0.079 0.069 183s -0.763 -0.603 183s -0.506 -0.535 183s testing: mul_mat, [2, 2, 3, 3] = [4, 2, 3, 3] * [4, 2, 3, 3] 183s x0: [4, 2, 3, 3] 183s -0.904 0.667 0.260 0.861 183s 0.948 -0.707 0.976 -0.414 183s 183s x1: [4, 2, 3, 3] 183s 0.294 -0.231 0.694 0.463 183s -0.602 0.124 0.676 0.249 183s 183s y: [2, 2, 3, 3] 183s 0.159 0.928 183s 1.018 -0.102 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.020 -0.325 -0.654 0.584 183s 0.955 -0.716 -0.674 0.379 183s 0.077 -0.370 -0.571 0.884 183s -0.601 -0.917 0.126 -0.737 183s 183s x1: [4, 2, 1, 1] 183s -0.359 -0.819 0.410 -0.792 183s 0.333 0.116 -0.429 -0.294 183s 183s y: [4, 2, 1, 1] 183s -0.457 -0.333 -0.659 1.601 183s 0.065 0.412 -0.032 -0.144 183s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.853 -0.638 -0.932 -0.929 183s -0.561 0.304 -0.054 0.057 183s 0.666 -0.916 -0.555 -0.723 183s -0.195 -0.286 -0.284 0.538 183s 183s x1: [4, 4, 3, 1] 183s -0.119 -0.087 -0.659 0.522 183s -0.010 0.344 -0.837 0.171 183s -0.247 -0.629 -0.496 0.870 183s -0.058 0.210 0.931 0.922 183s 183s y: [4, 4, 3, 1] 183s 0.083 0.106 -0.012 0.516 183s 0.394 0.166 0.019 0.234 183s -0.154 0.024 0.059 0.837 183s -1.908 0.099 -1.414 0.183 183s testing: mul_mat, [2, 4, 3, 3] = [4, 2, 3, 3] * [4, 4, 3, 3] 183s x0: [4, 2, 3, 3] 183s 0.233 0.024 -0.862 0.606 183s -0.531 -0.322 -0.264 0.225 183s 183s x1: [4, 4, 3, 3] 183s 0.944 -0.058 -0.495 -0.507 183s -0.145 -0.340 0.267 -0.799 183s -0.575 0.476 -0.087 0.721 183s -0.716 0.997 -0.129 0.530 183s 183s y: [2, 4, 3, 3] 183s 0.338 -0.466 183s -0.756 -0.064 183s 0.390 0.338 183s 0.290 0.213 183s test-mul-mat0: iter:66/500 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.222 -0.767 0.203 183s 0.093 0.139 -0.973 183s 0.963 0.206 0.983 183s -0.433 -0.958 -0.701 183s 183s x1: [3, 4, 1, 1] 183s -0.241 -0.366 -0.274 183s -0.170 -0.321 -0.790 183s -0.390 -0.400 -0ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s .361 183s -0.925 0.325 0.811 183s 183s y: [4, 4, 1, 1] 183s 0.171 0.193 -0.577 0.646 183s 0.048 0.709 -1.007 0.934 183s 0.147 0.259 -0.813 0.805 183s -0.290 -0.830 -0.027 -0.479 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.654 -0.804 -0.490 183s 0.671 -0.994 0.886 183s 183s x1: [3, 4, 1, 1] 183s -0.961 0.502 0.054 183s -0.765 -0.567 -0.380 183s -0.006 0.067 0.347 183s 0.825 0.746 0.557 183s 183s y: [2, 4, 1, 1] 183s 0.198 -1.096 183s 1.142 -0.285 183s -0.220 0.237 183s -1.412 0.305 183s testing: mul_mat, [2, 2, 1, 3] = [3, 2, 1, 3] * [3, 2, 1, 3] 183s x0: [3, 2, 1, 3] 183s 0.628 -0.580 0.610 183s 0.374 0.977 -0.954 183s 183s x1: [3, 2, 1, 3] 183s -0.107 -0.761 0.089 183s 0.987 0.378 0.116 183s 183s y: [2, 2, 1, 3] 183s 0.429 -0.868 183s 0.471 0.627 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.134 0.596 -0.916 183s 183s x1: [3, 2, 1, 1] 183s -0.779 -0.620 0.745 183s -0.262 0.006 0.144 183s 183s y: [1, 2, 1, 1] 183s -1.157 183s -0.163 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.089 183s 0.480 183s -0.303 183s 183s x1: [1, 3, 1, 1] 183s 0.695 183s -0.331 183s 0.870 183s 183s y: [3, 3, 1, 1] 183s 0.062 0.334 -0.211 183s -0.030 -0.159 0.100 183s 0.078 0.418 -0.264 183s testing: mul_mat, [3, 1, 1, 3] = [3, 3, 1, 3] * [3, 1, 1, 3] 183s x0: [3, 3, 1, 3] 183s -0.947 -0.301 -0.556 183s -0.637 -0.416 0.321 183s 0.631 -0.990 0.016 183s 183s x1: [3, 1, 1, 3] 183s 0.066 0.526 -0.584 183s 183s y: [3, 1, 1, 3] 183s 0.103 -0.449 -0.488 183s test-mul-mat0: iter:67/500 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.762 0.151 -1.000 183s -0.669 -0.681 -0.057 183s 183s x1: [3, 4, 1, 1] 183s -0.768 -0.457 -0.870 183s -0.069 0.126 -0.859 183s 0.375 -0.552 0.156 183s -0.037 0.062 0.306 183s 183s y: [2, 4, 1, 1] 183s 0.215 0.875 183s 0.826 0.009 183s 0.046 0.116 183s -0.325 -0.035 183s testing: mul_mat, [3, 2, 4, 1] = [3, 3, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.672 0.249 0.440 183s -0.997 0.568 -0.616 183s -0.617 0.752 0.336 183s 183s x1: [3, 2, 4, 1] 183s -0.620 -0.817 -0.048 183s 0.956 -0.536 -0.800 183s 183s y: [3, 2, 4, 1] 183s -0.641 0.183 -0.247 183s 0.156 -0.765 -1.262 183s testing: mul_mat, [4, 3, 4, 1] = [3, 4, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 4, 4, 1] 183s 0.954 -0.459 -0.842 183s -0.164 -0.438 -0.008 183s 0.482 -0.651 -0.401 183s 0.386 -0.142 0.380 183s 183s x1: [3, 3, 4, 1] 183s 0.610 0.989 -0.647 183s -0.054 -0.672 -0.432 183s -0.519 -0.722 0.335 183s 183s y: [4, 3, 4, 1] 183s 0.673 -0.529 -0.091 -0.151 183s 0.621 0.307 0.584 -0.090 183s -0.446 0.399 0.085 0.029 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.632 -0.626 -0.576 183s -0.302 0.740 -0.126 183s 183s x1: [3, 4, 1, 1] 183s 0.444 -0.615 -0.528 183s -0.300 -0.378 0.408 183s 0.901 -0.239 -0.129 183s 0.037 0.153 0.807 183s 183s y: [2, 4, 1, 1] 183s 0.408 -0.523 183s 0.191 -0.240 183s -0.346 -0.433 183s -0.584 0.000 183s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 2, 4, 1] 183s 0.744 0.829 183s 0.269 0.484 183s 183s x1: [2, 3, 4, 1] 183s -0.896 0.632 183s 0.724 0.068 183s -0.773 0.432 183s 183s y: [2, 3, 4, 1] 183s -0.142 0.064 183s 0.595 0.228 183s -0.217 0.001 183s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 2, 4, 1] 183s 0.765 -0.889 183s -0.856 -0.532 183s 183s x1: [2, 2, 4, 1] 183s 0.014 0.432 183s 0.391 0.854 183s 183s y: [2, 2, 4, 1] 183s -0.374 -0.242 183s -0.461 -0.789 183s test-mul-mat0: iter:68/500 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.274 0.910 183s 0.269 -0.809 183s 0.058 0.205 183s -0.148 0.475 183s 183s x1: [2, 1, 1, 1] 183s 0.733 0.335 183s 183s y: [4, 1, 1, 1] 183s 0.506 -0.073 0.111 0.050 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.089 0.575 183s 0.558 0.262 183s -0.960 -0.880 183s -0.132 0.896 183s 183s x1: [2, 4, 1, 1] 183s -0.030 -0.004 183s 0.585 0.438 183s -0.271 0.022 183s -0.666 0.118 183s 183s y: [4, 4, 1, 1] 183s -0.005 -0.018 0.032 0.001 183s 0.304 0.442 -0.948 0.315 183s -0.012 -0.146 0.241 0.055 183s 0.009 -0.341 0.536 0.194 183s testing: mul_mat, [1, 4, 1, 2] = [2, 1, 1, 2] * [2, 4, 1, 2] 183s x0: [2, 1, 1, 2] 183s -0.102 -0.689 183s 183s x1: [2, 4, 1, 2] 183s -0.464 -0.609 183s 0.226 -0.731 183s 0.725 -0.984 183s 0.543 0.636 183s 183s y: [1, 4, 1, 2] 183s 0.467 183s 0.480 183s 0.604 183s -0.493 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.004 0.583 183s -0.656 0.044 183s 183s x1: [2, 1, 1, 1] 183s 0.429 -0.214 183s 183s y: [2, 1, 1, 1] 183s -0.123 -0.291 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.834 -0.818 183s 0.269 -0.188 183s -0.433 -0.797 183s 183s x1: [2, 2, 1, 1] 183s -0.536 -0.550 183s -0.061 -0.001 183s 183s y: [3, 2, 1, 1] 183s 0.897 -0.040 0.671 183s 0.051 -0.016 0.027 183s testing: mul_mat, [1, 2, 1, 2] = [3, 1, 1, 2] * [3, 2, 1, 2] 183s x0: [3, 1, 1, 2] 183s -0.546 -0.620 0.866 183s 183s x1: [3, 2, 1, 2] 183s 0.467 -0.454 -0.104 183s 0.957 0.132 0.065 183s 183s y: [1, 2, 1, 2] 183s -0.063 183s -0.548 183s test-mul-mat0: iter:69/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.307 0.789 -0.733 183s 183s x1: [3, 1, 1, 1] 183s -0.813 -0.259 0.607 183s 183s y: [1, 1, 1, 1] 183s -0.899 183s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 1, 2, 1] 183s -0.372 -0.341 -0.710 183s 183s x1: [3, 1, 2, 1] 183s 0.510 0.255 -0.187 183s 183s y: [1, 1, 2, 1] 183s -0.144 183s testing: mul_mat, [4, 1, 2, 3] = [3, 4, 2, 3] * [3, 1, 2, 3] 183s x0: [3, 4, 2, 3] 183s -0.809 -0.102 0.667 183s 0.403 -0.157 0.138 183s -0.969 0.503 0.428 183s -0.027 0.187 0.311 183s 183s x1: [3, 1, 2, 3] 183s -0.512 -0.861 -0.736 183s 183s y: [4, 1, 2, 3] 183s 0.011 -0.173 -0.252 -0.377 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.260 0.957 -0.823 183s -0.735 0.011 -0.463 183s 0.954 0.578 -0.963 183s 183s x1: [3, 4, 1, 1] 183s -0.451 -0.112 -0.087 183s 0.311 -0.215 0.433 183s -0.004 0.245 0.573 183s 0.704 -0.558 0.036 183s 183s y: [3, 4, 1, 1] 183s 0.081 0.371 -0.412 183s -0.643 -0.432 -0.245 183s -0.237 -0.260 -0.414 183s -0.746 -0.541 0.315 183s testing: mul_mat, [4, 3, 2, 1] = [3, 4, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 4, 2, 1] 183s 0.986 0.130 0.345 183s -0.135 0.767 -0.141 183s 0.813 -0.317 -0.322 183s -0.274 -0.859 0.382 183s 183s x1: [3, 3, 2, 1] 183s -0.702 -0.493 -0.265 183s 0.164 -0.939 0.701 183s -0.722 -0.374 0.911 183s 183s y: [4, 3, 2, 1] 183s -0.848 -0.246 -0.329 0.514 183s 0.282 -0.840 0.205 1.029 183s -0.446 -0.317 -0.761 0.866 183s testing: mul_mat, [3, 3, 2, 3] = [4, 3, 2, 3] * [4, 3, 2, 3] 183s x0: [4, 3, 2, 3] 183s 0.903 -0.677 -0.803 -0.291 183s 0.492 0.217 -0.215 -0.687 183s -0.483 0.840 0.803 -0.271 183s 183s x1: [4, 3, 2, 3] 183s 0.566 0.686 -0.191 -0.300 183s 0.229 -0.361 0.538 0.215 183s 0.504 0.351 0.941 -0.366 183s 183s y: [3, 3, 2, 3] 183s 0.287 0.674 0.230 183s -0.043 -0.229 -0.040 183s -0.432 0.373 0.906 183s test-mul-mat0: iter:70/500 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.933 0.951 0.655 183s 183s x1: [3, 2, 1, 1] 183s -0.319 -0.850 0.881 183s -0.492 -0.237 0.392 183s 183s y: [1, 2, 1, 1] 183s 0.065 183s 0.490 183s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 1, 2, 1] 183s -0.919 -0.736 0.405 183s 183s x1: [3, 1, 2, 1] 183s -0.750 -0.194 -0.648 183s 183s y: [1, 1, 2, 1] 183s 0.570 183s testing: mul_mat, [2, 1, 2, 3] = [3, 2, 2, 3] * [3, 1, 2, 3] 183s x0: [3, 2, 2, 3] 183s 0.560 0.325 0.754 183s 0.776 -0.284 -0.925 183s 183s x1: [3, 1, 2, 3] 183s -0.654 0.946 0.509 183s 183s y: [2, 1, 2, 3] 183s 0.324 -1.246 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.835 -0.890 -0.537 183s 0.497 -0.255 -0.974 183s 183s x1: [3, 2, 1, 1] 183s -0.573 -0.344 0.219 183s 0.908 0.238 -0.614 183s 183s y: [2, 2, 1, 1] 183s 0.666 -0.411 183s -0.639 0.989 183s testing: mul_mat, [2, 3, 2, 1] = [2, 2, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.936 -0.238 183s -0.457 -0.136 183s 183s x1: [2, 3, 2, 1] 183s 0.074 -0.041 183s 0.684 0.302 183s -0.134 0.593 183s 183s y: [2, 3, 2, 1] 183s -0.059 -0.028 183s -0.712 -0.354 183s -0.016 -0.019 183s testing: mul_mat, [1, 2, 2, 3] = [2, 1, 2, 3] * [2, 2, 2, 3] 183s x0: [2, 1, 2, 3] 183s 0.218 0.716 183s 183s x1: [2, 2, 2, 3] 183s 0.097 -0.242 183s -0.612 -0.649 183s 183s y: [1, 2, 2, 3] 183s -0.153 183s -0.598 183s test-mul-mat0: iter:71/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.649 0.573 183s 183s x1: [2, 4, 1, 1] 183s 0.325 -0.138 183s -0.026 0.368 183s 0.035 -0.262 183s 0.603 -0.216 183s 183s y: [1, 4, 1, 1] 183s 0.132 183s 0.194 183s -0.127 183s 0.268 183s testing: mul_mat, [1, 1, 4, 1] = [2, 1, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 1, 4, 1] 183s -0.072 0.233 183s 183s x1: [2, 1, 4, 1] 183s -0.077 -0.895 183s 183s y: [1, 1, 4, 1] 183s -0.203 183s testing: mul_mat, [3, 1, 4, 3] = [2, 3, 4, 3] * [2, 1, 4, 3] 183s x0: [2, 3, 4, 3] 183s -0.420 -0.596 183s -0.703 -0.131 183s 0.857 0.699 183s 183s x1: [2, 1, 4, 3] 183s -0.126 0.008 183s 183s y: [3, 1, 4, 3] 183s 0.048 0.088 -0.103 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.747 -0.258 183s -0.457 0.100 183s -0.438 0.298 183s 183s x1: [2, 3, 1, 1] 183s 0.729 -0.676 183s 0.767 -0.747 183s -0.833 -0.593 183s 183s y: [3, 3, 1, 1] 183s -0.371 -0.401 -0.521 183s -0.381 -0.426 -0.559 183s 0.775 0.322ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.do 0.188 183s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.754 0.104 0.488 183s 183s x1: [3, 2, 4, 1] 183s -0.133 -0.674 -0.853 183s 0.854 -0.803 -0.682 183s 183s y: [1, 2, 4, 1] 183s -0.586 183s 0.227 183s testing: mul_mat, [3, 3, 4, 3] = [1, 3, 4, 3] * [1, 3, 4, 3] 183s x0: [1, 3, 4, 3] 183s -0.571 183s -0.811 183s 0.905 183s 183s x1: [1, 3, 4, 3] 183s 0.075 183s -0.352 183s 0.640 183s 183s y: [3, 3, 4, 3] 183s -0.043 -0.061 0.068 183s 0.201 0.285 -0.318 183s -0.365 -0.519 0.579 183s test-mul-mat0: iter:72/500 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.266 0.799 0.691 0.498 183s -0.109 -0.942 -0.655 -0.801 183s -0.121 0.622 0.695 0.606 183s -0.493 -0.635 -0.924 -0.253 183s 183s x1: [4, 3, 1, 1] 183s -0.846 0.471 -0.500 0.557 183s -0.645 -0.549 0.935 0.309 183s -0.734 -0.925 0.696 -0.527 183s 183s y: [4, 3, 1, 1] 183s 0.532 -0.470 0.385 0.438 183s 0.534 -0.273 0.574 -0.276 183s -0.324 0.917 -0.321 0.440 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.517 0.904 -0.397 -0.585 183s 183s x1: [4, 4, 3, 1] 183s -0.475 -0.716 -0.099 0.996 183s -0.216 -0.542 -0.650 0.235 183s -0.607 0.659 0.501 -0.532 183s 0.355 0.974 0.379 -0.911 183s 183s y: [1, 4, 3, 1] 183s -0.944 183s -0.259 183s 1.022 183s 1.079 183s testing: mul_mat, [1, 1, 3, 1] = [4, 1, 3, 1] * [4, 1, 3, 1] 183s x0: [4, 1, 3, 1] 183s 0.191 0.591 -0.296 -0.912 183s 183s x1: [4, 1, 3, 1] 183s -0.162 0.695 0.883 -0.454 183s 183s y: [1, 1, 3, 1] 183s 0.532 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.526 0.409 -0.952 -0.321 183s 0.822 -0.632 0.080 -0.649 183s -0.231 0.907 -0.161 0.223 183s 183s x1: [4, 1, 1, 1] 183s -0.046 -0.174 0.615 -0.016 183s 183s y: [3, 1, 1, 1] 183s -0.676 0.132 -0.250 183s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.409 0.466 0.159 183s 0.151 -0.854 0.114 183s -0.913 -0.171 -0.503 183s 0.369 0.174 0.157 183s 183s x1: [3, 4, 3, 1] 183s -0.197 -0.351 -0.585 183s -0.605 0.352 -0.498 183s 0.225 0.532 0.073 183s 0.959 -0.685 -0.941 183s 183s y: [4, 4, 3, 1] 183s -0.337 0.203 0.534 -0.225 183s -0.163 -0.449 0.743 -0.240 183s 0.352 -0.412 -0.334 0.187 183s -0.077 0.623 -0.284 0.087 183s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.547 -0.749 -0.996 0.499 183s 0.588 -0.081 0.191 -0.229 183s 0.743 -0.065 -0.642 -0.347 183s 0.112 0.550 0.986 -0.874 183s 183s x1: [4, 3, 3, 1] 183s 0.365 -0.525 0.730 -0.476 183s 0.589 -0.773 0.682 -0.937 183s 0.572 -0.342 -0.337 0.176 183s 183s y: [4, 3, 3, 1] 183s -0.771 0.506 0.001 0.888 183s -0.890 0.754 0.376 1.132 183s 0.366 0.260 0.603 -0.610 183s test-mul-mat0: iter:73/500 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.038 -0.703 183s 183s x1: [2, 2, 1, 1] 183s 0.226 0.560 183s 0.639 0.790 183s 183s y: [1, 2, 1, 1] 183s -0.385 183s -0.531 183s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.242 -0.265 183s 0.640 0.557 183s 183s x1: [2, 1, 3, 1] 183s -0.478 0.964 183s 183s y: [2, 1, 3, 1] 183s -0.371 0.231 183s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.621 0.605 183s 0.709 -0.144 183s 183s x1: [2, 2, 3, 1] 183s -0.581 0.564 183s 0.316 -0.869 183s 183s y: [2, 2, 3, 1] 183s 0.701 -0.492 183s -0.721 0.349 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.484 -0.890 183s 0.978 0.917 183s -0.821 -0.626 183s 0.832 0.800 183s 183s x1: [2, 2, 1, 1] 183s 0.886 0.223 183s -0.923 0.559 183s 183s y: [4, 2, 1, 1] 183s -0.628 1.071 -0.867 0.915 183s -0.051 -0.391 0.408 -0.321 183s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 3, 3, 1] 183s 0.508 0.425 -0.282 -0.573 183s -0.428 -0.055 0.651 -0.014 183s -0.768 -0.968 0.178 -0.809 183s 183s x1: [4, 2, 3, 1] 183s 0.294 -0.294 0.146 0.709 183s 0.335 0.013 -0.912 -0.060 183s 183s y: [3, 2, 3, 1] 183s -0.423 -0.024 -0.489 183s 0.467 -0.737 -0.383 183s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.622 -0.630 0.667 183s -0.462 -0.026 0.371 183s 183s x1: [3, 4, 3, 1] 183s -0.684 -0.274 -0.778 183s 0.493 0.154 0.208 183s 0.684 0.669 0.134 183s -0.404 0.128 0.001 183s 183s y: [2, 4, 3, 1] 183s 0.079 0.034 183s -0.265 -0.155 183s -0.757 -0.284 183s 0.171 0.183 183s test-mul-mat0: iter:74/500 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.970 0.797 0.213 0.268 183s 0.711 -0.447 0.397 0.201 183s -0.864 -0.855 -0.988 -0.724 183s 0.442 0.222 -0.974 0.339 183s 183s x1: [4, 3, 1, 1] 183s -0.218 0.180 -0.208 0.982 183s -0.104 -0.180 -0.324 0.765 183s -0.944 0.054 -0.697 -0.574 183s 183s y: [4, 3, 1, 1] 183s 0.150 -0.121 -0.472 0.479 183s -0.109 0.031 0.010 0.489 183s -1.175 -1.087 1.874 0.078 183s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 4, 4, 1] 183s 0.030 -0.675 0.692 0.575 183s 0.830 0.814 0.375 -0.984 183s 0.768 -0.269 0.309 0.582 183s 0.616 0.507 -0.531 0.295 183s 183s x1: [4, 4, 4, 1] 183s 0.413 0.012 -0.879 -0.407 183s 0.805 -0.897 0.489 -0.375 183s -0.221 0.254 -0.319 0.832 183s 0.557 0.108 0.860 0.528 183s 183s y: [4, 4, 4, 1] 183s -0.838 0.424 -0.195 0.607 183s 0.752 0.490 0.792 -0.330 183s 0.080 -0.916 0.148 0.408 183s 0.842 0.353 0.972 0.097 183s testing: mul_mat, [4, 4, 4, 4] = [4, 4, 4, 4] * [4, 4, 4, 4] 183s x0: [4, 4, 4, 4] 183s 0.271 0.256 -0.850 0.715 183s -0.806 0.583 0.184 -0.461 183s -0.363 0.938 0.351 -0.963 183s 0.373 0.658 -0.665 0.091 183s 183s x1: [4, 4, 4, 4] 183s -0.758 -0.710 -0.508 -0.430 183s 0.165 -0.958 -0.333 -0.229 183s -0.100 -0.654 0.993 0.266 183s -0.507 -0.490 0.952 0.621 183s 183s y: [4, 4, 4, 4] 183s -0.263 0.302 -0.155 -0.452 183s -0.082 -0.648 -0.856 -0.369 183s -0.849 -0.241 -0.484 -1.104 183s -0.628 0.011 -0.540 -1.088 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.298 -0.397 0.156 -0.057 183s -0.807 -0.764 -0.308 -0.646 183s -0.228 -0.206 -0.388 0.142 183s -0.217 -0.859 0.169 -0.592 183s 183s x1: [4, 4, 1, 1] 183s 0.973 0.465 0.623 -0.787 183s 0.151 -0.225 0.438 0.937 183s 0.083 -0.831 0.918 -0.472 183s 0.540 -0.968 0.234 -0.214 183s 183s y: [4, 4, 1, 1] 183s -0.332 -0.824 -0.671 -0.040 183s 0.059 -0.691 -0.025 -0.320 183s 0.475 0.590 -0.272 1.130 183s 0.271 0.369 -0.046 0.880 183s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.699 0.414 0.714 -0.828 183s -0.883 -0.945 0.800 0.816 183s 183s x1: [4, 4, 4, 1] 183s 0.976 0.356 0.559 -0.249 183s -0.206 0.496 0.835 -0.037 183s 0.413 -0.638 -0.496 0.446 183s 0.596 0.289 -0.991 -0.702 183s 183s y: [2, 4, 4, 1] 183s 0.070 -0.954 183s 0.976 0.351 183s -1.276 0.204 183s -0.423 -2.165 183s testing: mul_mat, [3, 4, 4, 4] = [2, 3, 4, 4] * [2, 4, 4, 4] 183s x0: [2, 3, 4, 4] 183s -0.231 -0.207 183s 0.828 0.398 183s 0.740 -0.950 183s 183s x1: [2, 4, 4, 4] 183s 0.944 0.352 183s -0.804 0.658 183s 0.152 -0.633 183s 0.474 -0.216 183s 183s y: [3, 4, 4, 4] 183s -0.291 0.922 0.364 183s 0.050 -0.404 -1.221 183s 0.096 -0.126 0.714 183s -0.065 0.306 0.556 183s test-mul-mat0: iter:75/500 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.256 -0.673 -0.023 183s -0.059 -0.159 0.299 183s 0.050 0.195 0.333 183s -0.854 0.372 -0.972 183s 183s x1: [3, 3, 1, 1] 183s 0.823 0.537 -0.410 183s -0.867 0.633 -0.574 183s 0.901 -0.761 -0.513 183s 183s y: [4, 3, 1, 1] 183s -0.142 -0.257 0.009 -0.104 183s -0.635 -0.221 -0.111 1.534 183s 0.755 -0.085 -0.274 -0.554 183s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 3, 2, 1] 183s -0.420 -0.521 0.909 183s -0.274 0.851 0.938 183s 0.905 0.229 0.040 183s 183s x1: [3, 4, 2, 1] 183s 0.179 0.378 0.102 183s -0.075 -0.036 0.064 183s 0.560 0.926 -0.386 183s 0.383 0.463 0.204 183s 183s y: [3, 4, 2, 1] 183s -0.179 0.369 0.253 183s 0.109 0.050 -0.074 183s -1.068 0.273 0.704 183s -0.216 0.480 0.461 183s testing: mul_mat, [3, 3, 2, 4] = [3, 3, 2, 4] * [3, 3, 2, 4] 183s x0: [3, 3, 2, 4] 183s 0.400 -0.776 0.471 183s -0.506 0.472 0.885 183s 0.314 0.426 0.643 183s 183s x1: [3, 3, 2, 4] 183s 0.289 -0.324 0.552 183s -0.295 -0.989 0.669 183s 0.006 -0.399 -0.888 183s 183s y: [3, 3, 2, 4] 183s 0.626 0.189 0.307 183s 0.965 0.274 -0.084 183s -0.106 -0.977 -0.739 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.477 0.807 -0.809 183s 183s x1: [3, 3, 1, 1] 183s 0.237 0.622 -0.772 183s -0.832 -0.126 0.329 183s 0.893 0.023 0.669 183s 183s y: [1, 3, 1, 1] 183s 1.014 183s 0.029 183s -0.948 183s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.243 183s -0.892 183s -0.873 183s 0.609 183s 183s x1: [1, 3, 2, 1] 183s -0.836 183s 0.838 183s 0.774 183s 183s y: [4, 3, 2, 1] 183s 0.203 0.745 0.730 -0.509 183s -0.204 -0.748 -0.732 0.511 183s -0.188 -0.690 -0.676 0.471 183s testing: mul_mat, [1, 1, 2, 4] = [4, 1, 2, 4] * [4, 1, 2, 4] 183s x0: [4, 1, 2, 4] 183s 0.766 -0.619 0.544 -0.481 183s 183s x1: [4, 1, 2, 4] 183s -0.042 -0.132 -0.475 -0.806 183s 183s y: [1, 1, 2, 4] 183s 0.178 183s test-mul-mat0: iter:76/500 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.884 -0.654 183s 0.065 0.191 183s 0.317 0.489 183s 183s x1: [2, 3, 1, 1] 183s -0.719 0.622 183s -0t.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s .659 0.577 183s 0.974 -0.388 183s 183s y: [3, 3, 1, 1] 183s -1.042 0.072 0.076 183s -0.959 0.067 0.073 183s 1.115 -0.011 0.119 183s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.709 -0.625 183s 0.149 -0.735 183s 183s x1: [2, 3, 3, 1] 183s 0.885 0.472 183s 0.039 -0.277 183s -0.080 0.841 183s 183s y: [2, 3, 3, 1] 183s -0.922 -0.215 183s 0.145 0.210 183s -0.469 -0.631 183s testing: mul_mat, [1, 2, 3, 2] = [2, 1, 3, 2] * [2, 2, 3, 2] 183s x0: [2, 1, 3, 2] 183s 0.626 0.752 183s 183s x1: [2, 2, 3, 2] 183s 0.224 -0.878 183s -0.475 0.947 183s 183s y: [1, 2, 3, 2] 183s -0.519 183s 0.415 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.250 0.623 183s 183s x1: [2, 1, 1, 1] 183s 0.399 0.020 183s 183s y: [1, 1, 1, 1] 183s -0.087 183s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.284 183s -0.990 183s 183s x1: [1, 2, 3, 1] 183s 0.161 183s -0.980 183s 183s y: [2, 2, 3, 1] 183s -0.046 -0.159 183s 0.278 0.971 183s testing: mul_mat, [4, 1, 3, 2] = [2, 4, 3, 2] * [2, 1, 3, 2] 183s x0: [2, 4, 3, 2] 183s 0.805 0.976 183s -0.648 -0.072 183s 0.767 -0.425 183s -0.445 0.851 183s 183s x1: [2, 1, 3, 2] 183s -0.786 0.113 183s 183s y: [4, 1, 3, 2] 183s -0.522 0.501 -0.651 0.446 183s test-mul-mat0: iter:77/500 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.685 -0.939 0.132 0.609 183s 0.979 -0.051 0.254 0.665 183s 0.321 -0.125 -0.722 -0.732 183s 183s x1: [4, 1, 1, 1] 183s -0.556 0.515 -0.646 0.079 183s 183s y: [3, 1, 1, 1] 183s -0.141 -0.682 0.166 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.465 -0.427 0.857 -0.112 183s 183s x1: [4, 3, 1, 1] 183s -0.653 -0.999 -0.629 -0.534 183s -0.925 -0.652 0.243 0.170 183s -0.804 -0.836 -0.624 -0.871 183s 183s y: [1, 3, 1, 1] 183s 0.251 183s 0.898 183s 0.294 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.188 0.692 0.059 -0.934 183s 0.356 -0.620 -0.059 0.634 183s 183s x1: [4, 1, 1, 1] 183s 0.652 -0.023 0.204 0.713 183s 183s y: [2, 1, 1, 1] 183s -0.792 0.686 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.562 0.257 0.504 -0.736 183s 0.162 0.691 0.792 0.361 183s 183s x1: [4, 2, 1, 1] 183s -0.353 0.288 0.635 0.019 183s 0.755 0.711 0.367 -0.002 183s 183s y: [2, 2, 1, 1] 183s 0.182 0.652 183s 0.793 0.902 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.238 0.886 183s -0.324 0.704 183s 183s x1: [2, 4, 1, 1] 183s -0.320 0.916 183s -0.662 0.884 183s 0.628 -0.552 183s -0.304 0.320 183s 183s y: [2, 4, 1, 1] 183s 0.888 0.748 183s 0.941 0.836 183s -0.639 -0.592 183s 0.356 0.324 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.253 -0.081 183s 0.657 0.745 183s 183s x1: [2, 2, 1, 1] 183s -0.690 -0.466 183s -0.008 0.946 183s 183s y: [2, 2, 1, 1] 183s 0.212 -0.800 183s -0.074 0.699 183s test-mul-mat0: iter:78/500 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.564 0.770 183s -0.732 -0.648 183s 183s x1: [2, 1, 1, 1] 183s -0.828 -0.590 183s 183s y: [2, 1, 1, 1] 183s 0.012 0.989 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.301 -0.937 183s 0.610 0.977 183s 183s x1: [2, 2, 1, 1] 183s -0.891 -0.848 183s 0.980 -0.444 183s 183s y: [2, 2, 1, 1] 183s 0.527 -1.373 183s 0.710 0.165 183s testing: mul_mat, [2, 2, 1, 2] = [2, 2, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 2, 1, 2] 183s -0.045 -0.202 183s -0.573 0.437 183s 183s x1: [2, 2, 1, 2] 183s 0.949 0.314 183s -0.713 -0.517 183s 183s y: [2, 2, 1, 2] 183s -0.106 -0.406 183s 0.136 0.182 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.993 -0.622 183s -0.771 -0.451 183s 183s x1: [2, 2, 1, 1] 183s 0.809 -0.955 183s 0.884 -0.923 183s 183s y: [2, 2, 1, 1] 183s 1.398 -0.194 183s 1.453 -0.266 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.438 0.942 183s -0.999 0.045 183s 183s x1: [2, 2, 1, 1] 183s 0.077 0.678 183s -0.388 -0.313 183s 183s y: [2, 2, 1, 1] 183s 0.605 -0.046 183s -0.125 0.373 183s testing: mul_mat, [2, 2, 1, 2] = [2, 2, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 2, 1, 2] 183s 0.862 0.082 183s -0.285 0.976 183s 183s x1: [2, 2, 1, 2] 183s -0.693 -0.825 183s 0.081 0.360 183s 183s y: [2, 2, 1, 2] 183s -0.665 -0.607 183s 0.099 0.328 183s test-mul-mat0: iter:79/500 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.099 0.353 0.701 183s 183s x1: [3, 3, 1, 1] 183s -0.598 0.906 0.580 183s 0.014 -0.407 0.235 183s 0.576 -0.406 0.177 183s 183s y: [1, 3, 1, 1] 183s 0.786 183s 0.020 183s -0.076 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.910 0.211 -0.719 183s -0.375 -0.707 -0.743 183s 0.149 0.421 -0.883 183s 183s x1: [3, 1, 1, 1] 183s -0.739 0.418 0.738 183s 183s y: [3, 1, 1, 1] 183s -1.115 -0.567 -0.586 183s testing: mul_mat, [3, 3, 1, 2] = [3, 3, 1, 2] * [3, 3, 1, 2] 183s x0: [3, 3, 1, 2] 183s 0.611 0.021 0.255 183s 0.521 -0.768 0.536 183s -0.853 -0.475 0.793 183s 183s x1: [3, 3, 1, 2] 183s 0.750 -0.055 0.970 183s -0.420 -0.152 -0.628 183s -0.514 -0.572 0.386 183s 183s y: [3, 3, 1, 2] 183s 0.705 0.953 0.154 183s -0.420 -0.439 -0.068 183s -0.227 0.378 1.016 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.137 -0.525 0.996 183s 0.463 -0.516 -0.284 183s -0.723 0.532 0.068 183s 183s x1: [3, 3, 1, 1] 183s -0.887 -0.833 0.637 183s 0.193 0.830 -0.402 183s -0.134 -0.330 -0.820 183s 183s y: [3, 3, 1, 1] 183s 1.193 -0.162 0.242 183s -0.863 -0.225 0.275 183s -0.625 0.341 -0.134 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.819 0.986 0.816 183s -0.099 -0.462 0.136 183s -0.776 0.417 -0.717 183s 183s x1: [3, 3, 1, 1] 183s -0.857 0.242 -0.139 183s 0.438 0.188 -0.230 183s -0.515 -0.920 -0.350 183s 183s y: [3, 3, 1, 1] 183s -0.578 -0.046 0.866 183s 0.357 -0.162 -0.097 183s -1.616 0.429 0.267 183s testing: mul_mat, [2, 3, 1, 2] = [3, 2, 1, 2] * [3, 3, 1, 2] 183s x0: [3, 2, 1, 2] 183s -0.510 0.998 0.782 183s -0.119 -0.691 -0.778 183s 183s x1: [3, 3, 1, 2] 183s -0.513 0.316 0.724 183s 0.350 -0.221 -1.000 183s 0.825 0.263 0.532 183s 183s y: [2, 3, 1, 2] 183s 1.142 -0.720 183s -1.181 0.889 183s 0.257 -0.694 183s test-mul-mat0: iter:80/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.986 0.237 183s -0.041 -0.576 183s 183s x1: [2, 4, 1, 1] 183s 0.550 0.645 183s 0.774 -0.187 183s 0.177 0.594 183s -0.207 0.776 183s 183s y: [2, 4, 1, 1] 183s -0.390 -0.394 183s -0.807 0.076 183s -0.034 -0.350 183s 0.388 -0.439 183s testing: mul_mat, [3, 2, 3, 1] = [2, 3, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.375 -0.772 183s -0.325 -0.730 183s -0.823 -0.775 183s 183s x1: [2, 2, 3, 1] 183s 0.467 -0.154 183s -0.087 -0.652 183s 183s y: [3, 2, 3, 1] 183s -0.056 -0.039 -0.265 183s 0.536 0.504 0.577 183s testing: mul_mat, [2, 3, 3, 2] = [2, 2, 3, 2] * [2, 3, 3, 2] 183s x0: [2, 2, 3, 2] 183s 0.930 -0.909 183s 0.792 -0.572 183s 183s x1: [2, 3, 3, 2] 183s 0.571 -0.092 183s 0.182 0.919 183s 0.751 -0.596 183s 183s y: [2, 3, 3, 2] 183s 0.615 0.505 183s -0.666 -0.382 183s 1.240 0.936 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.598 -0.507 183s 0.357 0.149 183s -0.630 -0.202 183s 183s x1: [2, 2, 1, 1] 183s -0.980 -0.442 183s 0.127 0.486 183s 183s y: [3, 2, 1, 1] 183s 0.810 -0.416 0.707 183s -0.322 0.118 -0.178 183s testing: mul_mat, [1, 2, 3, 1] = [3, 1, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 1, 3, 1] 183s -0.299 0.766 0.807 183s 183s x1: [3, 2, 3, 1] 183s -0.896 -0.508 0.628 183s 0.860 0.614 0.535 183s 183s y: [1, 2, 3, 1] 183s 0.385 183s 0.645 183s testing: mul_mat, [2, 3, 3, 2] = [1, 2, 3, 2] * [1, 3, 3, 2] 183s x0: [1, 2, 3, 2] 183s 0.439 183s 0.372 183s 183s x1: [1, 3, 3, 2] 183s -0.276 183s 0.759 183s 0.414 183s 183s y: [2, 3, 3, 2] 183s -0.121 -0.103 183s 0.333 0.282 183s 0.182 0.154 183s test-mul-mat0: iter:81/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.559 183s -0.125 183s 183s x1: [1, 1, 1, 1] 183s 0.010 183s 183s y: [2, 1, 1, 1] 183s -0.006 -0.001 183s testing: mul_mat, [2, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.856 183s -0.557 183s 183s x1: [1, 2, 4, 1] 183s 0.296 183s 0.159 183s 183s y: [2, 2, 4, 1] 183s 0.254 -0.165 183s 0.136 -0.089 183s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 3, 4, 1] 183s -0.571 183s 0.527 183s 0.738 183s 183s x1: [1, 2, 4, 1] 183s 0.553 183s -0.243 183s 183s y: [3, 2, 4, 1] 183s -0.315 0.291 0.408 183s 0.138 -0.128 -0.179 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.053 183s -0.034 183s 0.011 183s 183s x1: [1, 3, 1, 1] 183s -0.467 183s 0.504 183s 0.738 183s 183s y: [3, 3, 1, 1] 183s 0.025 0.016 -0.005 183s -0.026 -0.017 0.006 183s -0.039 -0.025 0.008 183s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.892 0.626 0.813 183s 183s x1: [3, 1, 4, 1] 183s -0.469 0.332 -0.554 183s 183s y: [1, 1, 4, 1] 183s -0.661 183s testing: mul_mat, [2, 3, 4, 1] = [1, 2, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 2, 4, 1] 183s -0.833 183s 0.640 183s 183s x1: [1, 3, 4, 1] 183s 0.131 183s 0.932 183s 0.788 183s 183s y: [2, 3, 4, 1] 183s -0.109 0.084 183s -0.776 0.596 183s -0.656 0.504 183s test-mul-mat0: iter:82/500 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.749 -0.479 183s 183s x1: [2, 3, 1, 1] 183s -0.154 -0.875 183s 0.922 -0.527 183s -0.681 -0.267 183s 183s y: [1, 3, 1, 1] 183s 0.535 183s -0.438 183s 0.638 183s testing: mul_mat, [3, 1, 4, 1] = [2, 3, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.516 0.864 183s -0.026 -0.440 183s -0.604 -0.498 183s 183s x1: [2, 1, 4, 1] 183s 0.682 0.148 183s 183s y: [3, 1, 4, 1] 183s 0.480 -0.083 -0.486 183s testing: mul_mat, [1, 3, 4, 2] = [2, 1, 4, 2] * [2, 3, 4, 2] 183s x0: [2, 1, 4, 2] 183s 0.737 0.914 183s 183s x1: [2, 3, 4, 2] 183s -0.515 -0.568 183s -0.309 0.049 183s 0.417 -0.611 183s 183s y: [1, 3, 4, 2] 183s -0.898 183s -0.183 183s -0.251 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.461 -0.551 183s 0.152 -0.165 183s 183s x1: [2, 1, 1, 1] 183s 0.445 0.532 183s 183s y: [2, 1, 1, 1] 183s -0.088 -0.020 183s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 2, 4, 1] 183s 0.152 -0.602 183s 0.435 -0.411 183s 183s x1: [2, 2, 4, 1] 183s -0.734 -0.600 183s -0.930 0.418 183s 183s y: [2, 2, 4, 1] 183s 0.250 -0.072 183s -0.393 -0.576 183s testing: mul_mat, [1, 2, 4, 2] = [2, 1, 4, 2] * [2, 2, 4, 2] 183s x0: [2, 1, 4, 2] 183s -0.275 -0.834 183s 183s x1: [2, 2, 4, 2] 183s 0.057 0.612 183s -0.499 0.879 183s 183s y: [1, 2, 4, 2] 183s -0.527 183s -0.596 183s test-mul-mat0: iter:83/500 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.529 -0.607 -0.037 -0.363 183s -0.200 -0.574 0.240 -0.353 183s -0.332 0.154 0.005 0.772 183s -0.936 -0.921 -0.439 0.122 183s 183s x1: [4, 2, 1, 1] 183s -0.288 -0.701 -0.526 -0.046 183s -0.670 -0.986 -0.816 -0.445 183s 183s y: [4, 2, 1, 1] 183s 0.614 0.350 -0.050 1.141 183s 1.144 0.661 -0.277 1.839 183s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.430 0.506 -0.782 -0.826 183s 0.741 -0.662 0.372 -0.768 183s 183s x1: [4, 4, 4, 1] 183s 0.304 0.431 -0.084 0.926 183s 0.628 -0.867 -0.362 0.927 183s -0.394 0.593 -0.743 -0.380 183s 0.777 -0.188 -0.142 -0.752 183s 183s y: [2, 4, 4, 1] 183s -0.611 -0.803 183s -1.191 0.193 183s 1.364 -0.669 183s 0.302 1.224 183s testing: mul_mat, [1, 2, 4, 3] = [4, 1, 4, 3] * [4, 2, 4, 3] 183s x0: [4, 1, 4, 3] 183s -0.755 0.194 0.565 -0.961 183s 183s x1: [4, 2, 4, 3] 183s -0.158 -0.476 0.462 -0.417 183s -0.138 -0.166 -0.185 0.942 183s 183s y: [1, 2, 4, 3] 183s 0.688 183s -0.938 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.867 -0.938 -0.823 0.690 183s -0.320 0.672 -0.348 0.135 183s 0.522 0.652 -0.661 0.610 183s -0.102 -0.397 -0.753 -0.546 183s 183s x1: [4, 1, 1, 1] 183s 0.682 0.479 0.368 0.075 183s 183s y: [4, 1, 1, 1] 183s -0.109 -0.014 0.471 -0.577 183s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.331 -0.689 -0.678 -0.420 183s -0.081 -0.250 -0.533 0.372 183s 183s x1: [4, 4, 4, 1] 183s 0.439 -0.086 0.246 -0.294 183s -0.324 0.467 0.731 0.966 183s -0.548 -0.589 0.648 0.931 183s 0.779 -0.277 -0.383 0.645 183s 183s y: [2, 4, 4, 1] 183s 0.160 -0.255 183s -1.330 -0.120 183s -0.605 0.192 183s 0.438 0.450 183s testing: mul_mat, [2, 4, 4, 3] = [2, 2, 4, 3] * [2, 4, 4, 3] 183s x0: [2, 2, 4, 3] 183s -0.164 0.258 183s 0.947 0.936 183s 183s x1: [2, 4, 4, 3] 183s 0.400 -0.263 183s 0.173 0.722 183s 0.204 0.754 183s 0.094 0.759 183s 183s y: [2, 4, 4, 3] 183s -0.134 0.133 183s 0.158 0.840 183s 0.161 0.899 183s 0.180 0.799 183s test-mul-mat0: iter:84/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.852 0.157 183s 0.691 0.340 183s 0.756 -0.172 183s 183s x1: [2, 2, 1, 1] 183s -0.047 -0.847 183s 0.268 -0.688 183s 183s y: [3, 2, 1, 1] 183s -0.093 -0.321 0.110 183s -0.336 -0.049 0.321 183s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 1, 2, 1] 183s -0.453 -0.027 183s 183s x1: [2, 3, 2, 1] 183s -0.521 0.613 183s 0.901 0.019 183s 0.709 -0.646 183s 183s y: [1, 3, 2, 1] 183s 0.220 183s -0.408 183s -0.304 183s testing: mul_mat, [2, 1, 2, 4] = [2, 2, 2, 4] * [2, 1, 2, 4] 183s x0: [2, 2, 2, 4] 183s 0.255 0.428 183s 0.177 -0.125 183s 183s x1: [2, 1, 2, 4] 183s 0.929 -0.140 183s 183s y: [2, 1, 2, 4] 183s 0.177 0.182 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.214 -0.593 183s 183s x1: [2, 2, 1, 1] 183s 0.207 -0.320 183s 0.139 0.692 183s 183s y: [1, 2, 1, 1] 183s 0.145 183s -0.440 183s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.005 183s 183s x1: [1, 2, 2, 1] 183s 0.573 183s 0.091 183s 183s y: [1, 2, 2, 1] 183s -0.003 183s -0.000 183s testing: mul_mat, [3, 1, 2, 4] = [1, 3, 2, 4] * [1, 1, 2, 4] 183s x0: [1, 3, 2, 4] 183s 0.414 183s 0.706 183s -0.284 183s 183s x1: [1, 1, 2, 4] 183s -0.272 183s 183s y: [3, 1, 2, 4] 183s -0.113 -0.192 0.077 183s test-mul-mat0: iter:85/500 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.652 183s 0.850 183s -0.641 183s 183s x1: [1, 3, 1, 1] 183s -0.151 183s -0.359 183s 0.134 183s 183s y: [3, 3, 1, 1] 183s 0.098 -0.128 0.097 183s 0.234 -0.305 0.230 183s -0.088 0.114 -0.086 183s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 2, 3, 1] 183s 0.240 183s -0.733 183s 183s x1: [1, 3, 3, 1] 183s 0.063 183s -0.656 183s 0.958 183s 183s y: [2, 3, 3, 1] 183s 0.015 -0.046 183s -0.157 0.480 183s 0.229 -0.702 183s testing: mul_mat, [1, 2, 3, 2] = [1, 1, 3, 2] * [1, 2, 3, 2] 183s x0: [1, 1, 3, 2] 183s -0.858 183s 183s x1: [1, 2, 3, 2] 183s 0.173 183s -0.092 183s 183s y: [1, 2, 3, 2] 183s -0.149 183s 0.079 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.078 183s 0.971 183s 183s x1: [1, 1, 1, 1] 183s -0.468 183s 183s y: [2, 1, 1, 1] 183s 0.036 -0.455 183s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 4, 3, 1] 183s -0.417 -0.946 183s 0.093 0.158 183s -0.522 0.255 183s 0.322 -0.279 183s 183s x1: [2, 1, 3, 1] 183s 0.064 0.987 183s 183s y: [4, 1, 3, 1] 183s -0.961 0.161 0.219 -0.254 183s testing: mul_mat, [1, 2, 3, 2] = [4, 1, 3, 2] * [4, 2, 3, 2] 183s x0: [4, 1, 3, 2] 183s 0.258 -0.918 0.189 0.977 183s 183s x1: [4, 2, 3, 2] 183s -0.486 -0.921 -0.053 0.846 183s 0.126 -0.134 0.416 0.709 183s 183s y: [1, 2, 3, 2] 183s 1.537 183s 0.927 183s test-mul-mat0: iter:86/500 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.435 0.467 0.166 0.204 183s 0.868 -0.327 0.401 0.594 183s 0.447 -0.427 -0.645 0.448 183s -0.387 0.182 -0.312 -0.428 183s 183s x1: [4, 3, 1, 1] 183s 0.245 0.852 0.950 -0.497 183s 0.934 0.139 -0.520 0.913 183s 0.188 -0.589 0.640 0.267 183s 183s y: [4, 3, 1, 1] 183s 0.348 0.019 -1.090 -0.024 183s -0.241 1.099 1.103 -0.565 183s -0.196 0.771 0.043 -0.494 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.561 0.468 -0.221 0.112 183s 183s x1: [4, 4, 1, 1] 183s -0.641 0.900 0.817 -0.789 183s 0.850 -0.680 -0.854 -0.011 183s -0.200 -0.941 -0.823 0.212 183s 0.699 0.444 0.273 -0.736 183s 183s y: [1, 4, 1, 1] 183s 0.511 183s -0.607 183s -0.123 183s -0.327 183s testing: mul_mat, [3, 1, 1, 3] = [4, 3, 1, 3] * [4, 1, 1, 3] 183s x0: [4, 3, 1, 3] 183s -0.520 -0.555 -0.236 -0.374 183s 0.434 0.564 -0.316 0.610 183s -0.224 -0.617 0.054 -0.951 183s 183s x1: [4, 1, 1, 3] 183s -0.131 -0.627 -0.441 0.442 183s 183s y: [3, 1, 1, 3] 183s 0.355 -0.002 -0.028 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.944 0.670 0.980 0.513 183s 0.825 0.118 0.961 0.947 183s -0.049 -0.148 0.445 0.340 183s -0.629 0.423 0.437 0.829 183s 183s x1: [4, 3, 1, 1] 183s 0.880 0.812 -0.054 -0.344 183s -0.805 -1.000 -0.295 -0.157 183s 0.966 0.192 -0.041 -0.291 183s 183s y: [4, 3, 1, 1] 183s 1.145 0.445 -0.304 -0.519 183s -1.799 -1.215 0.002 -0.176 183s 0.850 0.504 -0.193 -0.786 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.994 0.197 0.792 0.124 183s 183s x1: [4, 4, 1, 1] 183s 0.075 -0.324 0.710 -0.113 183s 0.622 -0.634 0.082 0.622 183s 0.070 0.925 0.588 -0.737 183s -0.117 -0.703 0.055 -0.173 183s 183s y: [1, 4, 1, 1] 183s 0.410 183s -0.602 183s 0.487 183s -0.001 183s testing: mul_mat, [4, 4, 1, 3] = [1, 4, 1, 3] * [1, 4, 1, 3] 183s x0: [1, 4, 1, 3] 183s -0.431 183s -0.822 183s 0.428 183s 0.651 183s 183s x1: [1, 4, 1, 3] 183s -0.950 183s 0.215 183s 0.104 183s -0.990 183s 183s y: [4, 4, 1, 3] 183s 0.409 0.781 -0.407 -0.619 183s -0.093 -0.177 0.092 0.140 183s -0.045 -0.085 0.044 0.067 183s 0.426 0.814 -0.424 -0.645 183s test-mul-mat0: iter:87/500 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.457 183s 183s x1: [1, 4, 1, 1] 183s 0.599 183s -0.055 183s -0.467 183s -0.186 183s 183s y: [1, 4, 1, 1] 183s -0.273 183s 0.025 183s 0.213 183s 0.085 183s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 1, 4, 1] 183s 0.419 183s 183s x1: [1, 1, 4, 1] 183s 0.475 183s 183s y: [1, 1, 4, 1] 183s 0.199 183s testing: mul_mat, [4, 1, 4, 2] = [1, 4, 4, 2] * [1, 1, 4, 2] 183s x0: [1, 4, 4, 2] 183s -0.978 183s -0.924 183s -0.015 183s 0.308 183s 183s x1: [1, 1, 4, 2] 183s -0.665 183s 183s y: [4, 1, 4, 2] 183s 0.650 0.615 0.010 -0.205 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.377 183s 0.922 183s 0.865 183s -0.053 183s 183s x1: [1, 4, 1, 1] 183s -0.442 183s 0.307 183s -0.168 183s -0.559 183s 183s y: [4, 4, 1, 1] 183s 0.166 -0.407 -0.382 0.023 183s -0.116 0.283 0.266 -0.016 183s 0.063 -0.155 -0.145 0.009 183s 0.211 -0.516 -0.483 0.030 183s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.836 -0.142 -0.541 -0.842 183s 183s x1: [4, 1, 4, 1] 183s -0.782 0.664 0.815 0.988 183s 183s y: [1, 1, 4, 1] 183s -2.021 183s testing: mul_mat, [3, 4, 4, 2] = [1, 3, 4, 2] * [1, 4, 4, 2] 183s x0: [1, 3, 4, 2] 183s 0.823 183s -0.319 183s 0.661 183s 183s x1: [1, 4, 4, 2] 183s -0.401 183s 0.587 183s 0.161 183s 0.348 183s 183s y: [3, 4, 4, 2] 183s -0.330 0.128 -0.265 183s 0.483 -0.187 0.388 183s 0.133 -0.051 0.106 183s 0.286 -0.111 0.230 183s test-mul-mat0: iter:88/500 183s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.243 183s 0.246 183s -0.267 183s 183s x1: [1, 1, 1, 1] 183s 0.423 183s 183s y: [3, 1, 1, 1] 183s 0.103 0.104 -0.113 183s testing: mul_mat, [1, 3, 3, 1] = [1, 1, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 1, 3, 1] 183s 0.682 183s 183s x1: [1, 3, 3, 1] 183s -0.096 183s 0.142 183s 0.865 183s 183s y: [1, 3, 3, 1] 183s -0.066 183s 0.097 183s 0.590 183s testing: mul_mat, [1, 1, 3, 2] = [1, 1, 3, 2] * [1, 1, 3, 2] 183s x0: [1, 1, 3, 2] 183s -0.649 183s 183s x1: [1, 1, 3, 2] 183s 0.126 183s 183s y: [1, 1, 3, 2] 183s -0.082 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.286 183s 183s x1: [1, 1, 1, 1] 183s -0.464 183s 183s y: [1, 1, 1, 1] 183s 0.133 183s testing: mul_mat, [1, 1, 3, 1] = [1, 1, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 1, 3, 1] 183s 0.148 183s 183s x1: [1, 1, 3, 1] 183s 0.782 183s 183s y: [1, 1, 3, 1] 183s 0.116 183s testing: mul_mat, [2, 1, 3, 2] = [1, 2, 3, 2] * [1, 1, 3, 2] 183s x0: [1, 2, 3, 2] 183s -0.970 183s 0.072 183s 183s x1: [1, 1, 3, 2] 183s -0.466 183s 183s y: [2, 1, 3, 2] 183s 0.452 -0.033 183s test-mul-mat0: iter:89/500 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.280 0.386 183s -0.347 -0.017 183s 0.899 -0.157 183s -0.071 0.891 183s 183s x1: [2, 3, 1, 1] 183s -0.411 -0.039 183s 0.703 -0.793 183s 0.886 -0.149 183s 183s y: [4, 3, 1, 1] 183s 0.100 0.143 -0.363 -0.005 183s -0.503 -0.231 0.756 -0.756 183s -0.306 -0.305 0.820 -0.195 183s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.570 0.181 183s 0.790 -0.323 183s 0.435 0.583 183s 0.315 0.059 183s 183s x1: [2, 4, 4, 1] 183s -0.127 -0.999 183s 0.928 -0.335 183s 0.849 0.107 183s -0.746 0.793 183s 183s y: [4, 4, 4, 1] 183s -0.253 0.222 -0.638 -0.099 183s 0.468 0.841 0.208 0.272 183s 0.503 0.636 0.432 0.274 183s -0.282 -0.845 0.138 -0.188 183s testing: mul_mat, [2, 4, 4, 2] = [2, 2, 4, 2] * [2, 4, 4, 2] 183s x0: [2, 2, 4, 2] 183s 0.551 0.384 183s -0.481 -0.595 183s 183s x1: [2, 4, 4, 2] 183s 0.668 -0.509 183s -0.959 0.103 183s -0.926 0.356 183s -0.838 0.855 183s 183s y: [2, 4, 4, 2] 183s 0.172 -0.019 183s -0.488 0.400 183s -0.373 0.233 183s -0.133 -0.105 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.436 0.025 183s 183s x1: [2, 2, 1, 1] 183s 0.209 0.631 183s 0.463 -0.626 183s 183s y: [1, 2, 1, 1] 183s 0.107 183s 0.186 183s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 3, 4, 1] 183s 0.025 183s 0.973 183s 0.220 183s 183s x1: [1, 2, 4, 1] 183s -0.729 183s -0.844 183s 183s y: [3, 2, 4, 1] 183s -0.018 -0.709 -0.161 183s -0.021 -0.821 -0.186 183s testing: mul_mat, [2, 1, 4, 2] = [3, 2, 4, 2] * [3, 1, 4, 2] 183s x0: [3, 2, 4, 2] 183s 0.740 0.798 -0.714 183s -0.586 -0.602 -0.443 183s 183s x1: [3, 1, 4, 2] 183s -0.517 -0.016 0.817 183s 183s y: [2, 1, 4, 2] 183s -0.978 -0.050 183s test-mul-mat0: iter:90/500 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.099 -0.437 -0.019 183s 0.689 -0.377 -0.702 183s 183s x1: [3, 3, 1, 1] 183s -0.631 0.449 -0.883 183s -0.159 -0.547 0.428 183s 0.257 -0.094 -0.211 183s 183s y: [2, 3, 1, 1] 183s -0.241 0.016 183s 0.215 -0.204 183s 0.071 0.361 183s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.683 0.494 0.395 183s 0.223 -0.717 0.614 183s 183s x1: [3, 2, 3, 1] 183s 0.161 0.953 0.989 183s 0.207 0.170 0.333 183s 183s y: [2, 2, 3, 1] 183s 0.751 -0.040 183s 0.074 0.129 183s testing: mul_mat, [1, 2, 3, 3] = [3, 1, 3, 3] * [3, 2, 3, 3] 183s x0: [3, 1, 3, 3] 183s -0.383 0.019 0.420 183s 183s x1: [3, 2, 3, 3] 183s -0.729 -0.506 0.260 183s -0.423 -0.448 0.260 183s 183s y: [1, 2, 3, 3] 183s 0.380 183s 0.263 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.609 -0.567 0.119 183s 183s x1: [3, 1, 1, 1] 183s 0.067 0.775 0.414 183s 183s y: [1, 1, 1, 1] 183s -0.431 183s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 2, 3, 1] 183s 0.929 183s -0.812 183s 183s x1: [1, 3, 3, 1] 183s 0.328 183s 0.708 183s 0.919 183s 183s y: [2, 3, 3, 1] 183s 0.305 -0.266 183s 0.658 -0.574 183s 0.854 -0.746 183s testing: mul_mat, [1, 1, 3, 3] = [2, 1, 3, 3] * [2, 1, 3, 3] 183s x0: [2, 1, 3, 3] 183s -0.829 0.737 183s 183s x1: [2, 1, 3, 3] 183s -0.871 -0.001 183s 183s y: [1, 1, 3, 3] 183s 0.720 183s test-mul-mat0: iter:91/500 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.059 183s -0.885 183s 183s x1: [1, 2, 1, 1] 183s 0.987 183s -0.109 183s 183s y: [2, 2, 1, 1] 183s -0.058 -0.873 183s 0.006 0.097 183s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.252 183s 183s x1: [1, 2, 1, 1] 183s 0.768 183s 0.603 183s 183s y: [1, 2, 1, 1] 183s -0.194 183s -0.152 183s testing: mul_mat, [2, 1, 1, 2] = [1, 2, 1, 2] * [1, 1, 1, 2] 183s x0: [1, 2, 1, 2] 183s -0.640 183s -0.572 183s 183s x1: [1, 1, 1, 2] 183s 0.774 183s 183s y: [2, 1, 1, 2] 183s -0.495 -0.443 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.425 183s 0.126 183s 0.744 183s 183s x1: [1, 2, 1, 1] 183s -0.952 183s 0.625 183s 183s y: [3, 2, 1, 1] 183s 0.404 -0.120 -0.708 183s -0.265 0.079 0.465 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.459 -0.655 -0.145 183s 0.609 0.914 -0.668 183s 0.927 -0.783 0.702 183s 0.326 0.933 -0.543 183s 183s x1: [3, 1, 1, 1] 183s -0.899 -0.836 0.333 183s 183s y: [4, 1, 1, 1] 183s 0.912 -1.534 0.056 -1.255 183s testing: mul_mat, [1, 3, 1, 2] = [4, 1, 1, 2] * [4, 3, 1, 2] 183s x0: [4, 1, 1, 2] 183s -0.538 -0.324 -0.105 -0.206 183s 183s x1: [4, 3, 1, 2] 183s 0.591 -0.551 -0.769 -0.412 183s -0.759 -0.409 0.016 0.464 183s 0.467 0.064 0.088 0.712 183s 183s y: [1, 3, 1, 2] 183s 0.026 183s 0.443 183s -0.427 183s test-mul-mat0: iter:92/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.620 0.353 -0.308 183s 0.835 -0.191 -0.568 183s 183s x1: [3, 1, 1, 1] 183s 0.989 -0.468 0.641 183s 183s y: [2, 1, 1, 1] 183s 0.251 0.551 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.261 -0.201 0.839 183s 183s x1: [3, 2, 1, 1] 183s -0.788 0.599 -0.368 183s 0.131 0.997 -0.242 183s 183s y: [1, 2, 1, 1] 183s -0.224 183s -0.438 183s testing: mul_mat, [1, 1, 1, 4] = [3, 1, 1, 4] * [3, 1, 1, 4] 183s x0: [3, 1, 1, 4] 183s -0.554 0.036 0.841 183s 183s x1: [3, 1, 1, 4] 183s 0.415 0.694 -0.366 183s 183s y: [1, 1, 1, 4] 183s -0.514 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.651 -0.131 -0.228 183s -0.765 -0.849 0.718 183s -0.642 0.089 0.881 183s -0.957 0.888 -0.540 183s 183s x1: [3, 1, 1, 1] 183s -0.797 0.550 -0.604 183s 183s y: [4, 1, 1, 1] 183s 0.584 -0.291 0.028 1.577 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.578 0.502 0.086 -0.391 183s 0.106 -0.273 0.520 -0.506 183s 0.785 0.477 -0.394 -0.991 183s 0.128 0.737 0.237 -0.673 183s 183s x1: [4, 3, 1, 1] 183s -0.478 0.863 -0.319 -0.032 183s -0.101 -0.478 -0.233 -0.374 183s -0.802 0.375 -0.214 -0.914 183s 183s y: [4, 3, 1, 1] 183s 0.142 -0.435 0.195 0.521 183s -0.172 0.187 0.155 -0.169 183s 0.064 0.165 0.540 0.738 183s testing: mul_mat, [4, 4, 1, 4] = [4, 4, 1, 4] * [4, 4, 1, 4] 183s x0: [4, 4, 1, 4] 183s 0.528 0.742 -0.398 -0.362 183s -0.934 -0.795 0.050 0.359 183s 0.767 0.589 -0.208 0.800 183s -0.322 0.288 -0.462 0.481 183s 183s x1: [4, 4, 1, 4] 183s -0.625 -0.531 -0.151 -0.762 183s 0.150 0.817 0.136 0.673 183s -0.416 0.763 0.871 0.959 183s -0.451 0.957 0.540 -0.873 183s 183s y: [4, 4, 1, 4] 183s -0.388 0.725 -1.371 -0.248 183s 0.388 -0.542 1.107 0.448 183s -0.349 0.170 0.716 0.413 183s 0.572 -0.626 -0.592 -0.248 183s test-mul-mat0: iter:93/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.034 -0.547 183s 0.353 -0.427 183s 183s x1: [2, 4, 1, 1] 183s 0.189 -0.417 183s 0.213 0.552 183s -0.863 0.689 183s -0.447 0.050 183s 183s y: [2, 4, 1, 1] 183s 0.235 0.245 183s -0.294 -0.161 183s -0.406 -0.599 183s -0.043 -0.179 183s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 3, 4, 1] 183s -0.538 0.280 183s 0.972 0.676 183s -0.168 -0.891 183s 183s x1: [2, 2, 4, 1] 183s 0.779 -0.692 183s -0.005 0.366 183s 183s y: [3, 2, 4, 1] 183s -0.612 0.290 0.486 183s 0.105 0.243 -0.325 183s testing: mul_mat, [4, 3, 4, 2] = [2, 4, 4, 2] * [2, 3, 4, 2] 183s x0: [2, 4, 4, 2] 183s -0.949 0.553 183s -0.902 0.724 183s 0.902 0.071 183s -0.846 -0.104 183s 183s x1: [2, 3, 4, 2] 183s -0.529 -0.339 183s 0.990 -0.960 183s 0.847 -0.919 183s 183s y: [4, 3, 4, 2] 183s 0.314 0.231 -0.501 0.482 183s -1.470 -1.588 0.825 -0.737 183s -1.312 -1.430 0.699 -0.621 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.309 -0.261 183s -0.080 -0.705 183s 0.993 0.408 183s -0.092 -0.330 183s 183s x1: [2, 4, 1, 1] 183s 0.067 -0.440 183s 0.778 0.466 183s 0.743 -0.530 183s -0.114 0.224 183s 183s y: [4, 4, 1, 1] 183s 0.094 0.305 -0.113 0.139 183s -0.362 -0.391 0.963 -0.226 183s -0.092 0.314 0.522 0.106 183s -0.023 -0.149 -0.022 -0.063 183s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 3, 4, 1] 183s 0.527 0.978 -0.060 0.640 183s -0.096 -0.243 -0.164 -0.554 183s 0.473 0.191 -0.003 -0.216 183s 183s x1: [4, 2, 4, 1] 183s 0.395 -0.626 -0.940 -0.118 183s 0.972 -0.548 0.853 0.718 183s 183s y: [3, 2, 4, 1] 183s -0.423 0.333 0.095 183s 0.384 -0.498 0.197 183s testing: mul_mat, [4, 4, 4, 2] = [3, 4, 4, 2] * [3, 4, 4, 2] 183s x0: [3, 4, 4, 2] 183s -0.002 0.301 0.278 183s 0.816 0.071 0.830 183s -0.814 -0.864 -0.927 183s -0.449 -0.014 0.686 183s 183s x1: [3, 4, 4, 2] 183s 0.097 -0.118 0.461 183s 0.821 -0.546 0.183 183s -0.123 0.744 0.703 183s -0.439 -0.840 0.689 183s 183s y: [4, 4, 4, 2] 183s 0.092 0.452 -0.403 0.274 183s -0.115 0.783 -0.366 -0.235 183s 0.419 0.536 -1.194 0.527 183s -0.060 0.154 0.445 0.681 183s test-mul-mat0: iter:94/500 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.655 -0.917 183s -0.411 0.628 183s 183s x1: [2, 3, 1, 1] 183s -0.104 0.125 183s -0.413 0.089 183s -0.722 0.317 183s 183s y: [2, 3, 1, 1] 183s -0.046 0.121 183s 0.188 0.226 183s 0.182 0.495 183s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.728 0.800 183s 183s x1: [2, 2, 2, 1] 183s 0.004 -0.989 183s -0.808 0.564 183s 183s y: [1, 2, 2, 1] 183s -0.788 183s -0.137 183s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.856 0.467 183s 183s x1: [2, 1, 2, 1] 183s 0.378 0.297 183s 183s y: [1, 1, 2, 1] 183s 0.462 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.372 0.603 183s -0.679 0.454 183s 183s x1: [2, 1, 1, 1] 183s -0.945 0.885 183s 183s y: [2, 1, 1, 1] 183s 0.183 1.043 183s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 4, 2, 1] 183s 0.812 0.484 183s -0.679 0.564 183s -0.710 -0.010 183s 0.612 -0.219 183s 183s x1: [2, 2, 2, 1] 183s -0.089 0.231 183s -0.542 -0.078 183s 183s y: [4, 2, 2, 1] 183s 0.039 0.191 0.061 -0.105 183s -0.478 0.324 0.386 -0.315 183s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.772 0.792 -0.129 0.806 183s -0.084 0.906 -0.544 -0.528 183s -0.099 0.682 -0.637 0.839 183s 183s x1: [4, 2, 2, 1] 183s 0.362 0.174 0.443 0.683 183s -0.223 -0.103 -0.406 -0.992 183s 183s y: [3, 2, 2, 1] 183s 0.352 -0.475 0.374 183s -0.657 0.669 -0.622 183s test-mul-mat0: iter:95/500 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.069 -0.430 -0.750 183s 0.542 -0.591 0.971 183s -0.915 0.611 -0.134 183s -0.928 -0.504 -0.463 183s 183s x1: [3, 3, 1, 1] 183s 0.553 0.791 0.624 183s 0.781 -0.293 -0.475 183s 0.573 -0.386 -0.793 183s 183s y: [4, 3, 1, 1] 183s -0.770 0.438 -0.106 -1.200 183s 0.536 0.134 -0.829 -0.357 183s 0.801 -0.231 -0.653 0.030 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.957 0.684 0.715 183s -0.429 0.891 -0.842 183s 0.640 -0.539 -0.592 183s 183s x1: [3, 4, 1, 1] 183s -0.044 0.975 -0.783 183s 0.194 -0.029 0.328 183s -0.191 0.458 0.535 183s -0.639 0.250 0.159 183s 183s y: [3, 4, 1, 1] 183s 0.066 1.547 -0.090 183s 0.401 -0.386 -0.054 183s 0.513 0.040 -0.686 183s -0.327 0.363 -0.637 183s testing: mul_mat, [3, 3, 1, 4] = [3, 3, 1, 4] * [3, 3, 1, 4] 183s x0: [3, 3, 1, 4] 183s -0.414 -0.613 -0.512 183s -0.119 0.339 -0.513 183s -0.630 0.263 0.814 183s 183s x1: [3, 3, 1, 4] 183s 0.181 -0.130 -0.621 183s 0.267 -0.519 0.245 183s 0.339 -0.023 0.782 183s 183s y: [3, 3, 1, 4] 183s 0.322 0.253 -0.654 183s 0.082 -0.333 -0.105 183s -0.526 -0.449 0.417 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.009 0.141 -0.808 183s 0.170 0.654 0.143 183s 183s x1: [3, 3, 1, 1] 183s 0.545 -0.117 0.388 183s 0.360 0.552 -0.968 183s -0.963 -0.121 0.307 183s 183s y: [2, 3, 1, 1] 183s -0.326 0.072 183s 0.863 0.283 183s -0.273 -0.199 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.141 0.809 183s -0.407 -0.509 183s 183s x1: [2, 3, 1, 1] 183s -0.027 -0.988 183s 0.204 0.375 183s 0.812 -0.806 183s 183s y: [2, 3, 1, 1] 183s -0.795 0.514 183s 0.274 -0.274 183s -0.767 0.080 183s testing: mul_mat, [4, 2, 1, 4] = [2, 4, 1, 4] * [2, 2, 1, 4] 183s x0: [2, 4, 1, 4] 183s -0.019 0.607 183s 0.595 -0.966 183s -0.170 0.329 183s 0.954 0.419 183s 183s x1: [2, 2, 1, 4] 183s 0.380 -0.133 183s 0.004 0.506 183s 183s y: [4, 2, 1, 4] 183s -0.088 0.354 -0.108 0.307 183s 0.307 -0.486 0.166 0.216 183s test-mul-mat0: iter:96/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.583 0.199 183s 183s x1: [2, 4, 1, 1] 183s -0.667 0.011 183s 0.784 -0.439 183s 0.579 -0.939 183s 0.382 0.956 183s 183s y: [1, 4, 1, 1] 183s -0.387 183s 0.370 183s 0.151 183s 0.413 183s testing: mul_mat, [2, 1, 2, 1] = [2, 2, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.716 0.364 183s 0.417 0.522 183s 183s x1: [2, 1, 2, 1] 183s -0.397 0.290 183s 183s y: [2, 1, 2, 1] 183s 0.390 -0.014 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.247 0.968 183s 0.532 0.308 183s 183s x1: [2, 2, 2, 1] 183s -0.661 -0.225 183s 0.259 0.098 183s 183s y: [2, 2, 2, 1] 183s -0.381 -0.421 183s 0.158 0.168 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.887 -0.631 183s 0.825 0.530 183s 183s x1: [2, 2, 1, 1] 183s -0.314 -0.259 183s -0.777 0.542 183s 183s y: [2, 2, 1, 1] 183s 0.442 -0.396 183s 0.347 -0.354 183s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 1, 2, 1] 183s -0.499 -0.117 183s 183s x1: [2, 2, 2, 1] 183s 0.347 0.153 183s -0.293 -0.348 183s 183s y: [1, 2, 2, 1] 183s -0.191 183s 0.187 183s testing: mul_mat, [2, 2, 2, 1] = [1, 2, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.225 183s 0.855 183s 183s x1: [1, 2, 2, 1] 183s -0.164 183s -0.147 183s 183s y: [2, 2, 2, 1] 183s -0.037 -0.140 183s -0.033 -0.126 183s test-mul-mat0: iter:97/500 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.255 183s 183s x1: [1, 4, 1, 1] 183s -0.955 183s -0.505 183s 0.908 183s -0.586 183s 183s y: [1, 4, 1, 1] 183s -0.244 183s -0.129 183s 0.232 183s -0.150 183s testing: mul_mat, [3, 1, 3, 1] = [1, 3, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 3, 3, 1] 183s 0.778 183s 0.335 183s -0.808 183s 183s x1: [1, 1, 3, 1] 183s 0.566 183s 183s y: [3, 1, 3, 1] 183s 0.440 0.190 -0.458 183s testing: mul_mat, [3, 3, 3, 2] = [1, 3, 3, 2] * [1, 3, 3, 2] 183s x0: [1, 3, 3, 2] 183s -0.364 183s 0.829 183s -0.845 183s 183s x1: [1, 3, 3, 2] 183s -0.619 183s 0.620 183s -0.721 183s 183s y: [3, 3, 3, 2] 183s 0.226 -0.513 0.523 183s -0.226 0.514 -0.524 183s 0.263 -0.598 0.609 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.598 183s 0.178 183s 0.432 183s 183s x1: [1, 3, 1, 1] 183s 0.625 183s -0.787 183s -0.580 183s 183s y: [3, 3, 1, 1] 183s 0.374 0.111 0.270 183s -0.471 -0.140 -0.340 183s -0.346 -0.103 -0.250 183s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.243 -0.234 0.601 183s -0.821 0.757 -0.372 183s 183s x1: [3, 1, 3, 1] 183s 0.477 0.843 0.772 183s 183s y: [2, 1, 3, 1] 183s 0.150 -0.041 183s testing: mul_mat, [2, 3, 3, 2] = [2, 2, 3, 2] * [2, 3, 3, 2] 183s x0: [2, 2, 3, 2] 183s 0.317 0.057 183s -0.115 -0.922 183s 183s x1: [2, 3, 3, 2] 183s 0.884 0.890 183s -0.781 0.361 183s 0.733 0.991 183s 183s y: [2, 3, 3, 2] 183s 0.331 -0.923 183s -0.227 -0.243 183s 0.289 -0.998 183s test-mul-mat0: iter:98/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.263 183s 0.976 183s 183s x1: [1, 1, 1, 1] 183s -0.341 183s 183s y: [2, 1, 1, 1] 183s 0.090 -0.333 183s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 4, 2, 1] 183s 0.323 183s 0.034 183s -0.411 183s -0.451 183s 183s x1: [1, 2, 2, 1] 183s 0.236 183s 0.794 183s 183s y: [4, 2, 2, 1] 183s 0.076 0.008 -0.097 -0.106 183s 0.256 0.027 -0.326 -0.358 183s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 3, 2, 1] 183s 0.189 183s -0.276 183s -0.003 183s 183s x1: [1, 4, 2, 1] 183s 0.840 183s 0.627 183s -0.505 183s -0.666 183s 183s y: [3, 4, 2, 1] 183s 0.159 -0.232 -0.003 183s 0.119 -0.173 -0.002 183s -0.096 0.139 0.002 183s -0.126 0.184 0.002 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.398 183s 0.685 183s -0.048 183s -0.046 183s 183s x1: [1, 3, 1, 1] 183s -0.496 183s -0.192 183s 0.631 183s 183s y: [4, 3, 1, 1] 183s 0.197 -0.339 0.024 0.023 183s 0.076 -0.131 0.009 0.009 183s -0.251 0.432 -0.031 -0.029 183s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 2, 2, 1] 183s 0.543 0.824 -0.025 -0.677 183s -0.064 0.549 0.703 0.887 183s 183s x1: [4, 1, 2, 1] 183s 0.007 0.986 0.543 0.556 183s 183s y: [2, 1, 2, 1] 183s 0.426 1.415 183s testing: mul_mat, [4, 4, 2, 1] = [2, 4, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 4, 2, 1] 183s 0.991 0.750 183s 0.426 -0.086 183s 0.028 -0.413 183s 0.694 -0.203 183s 183s x1: [2, 4, 2, 1] 183s -0.830 0.291 183s -0.471 0.854 183s -0.758 0.483 183s -0.138 -0.772 183s 183s y: [4, 4, 2, 1] 183s -0.604 -0.379 -0.143 -0.635 183s 0.174 -0.274 -0.366 -0.501 183s -0.389 -0.364 -0.220 -0.624 183s -0.716 0.007 0.315 0.061 183s test-mul-mat0: iter:99/500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.121 183s 0.640 183s 0.621 183s 0.555 183s 183s x1: [1, 1, 1, 1] 183s 0.222 183s 183s y: [4, 1, 1, 1] 183s -0.027 0.142 0.138 0.123 183s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 1, 2, 1] 183s 0.027 183s 183s x1: [1, 4, 2, 1] 183s 0.974 183s -0.263 183s 0.721 183s 0.852 183s 183s y: [1, 4, 2, 1] 183s 0.027 183s -0.007 183s 0.020 183s 0.023 183s testing: mul_mat, [3, 1, 2, 4] = [1, 3, 2, 4] * [1, 1, 2, 4] 183s x0: [1, 3, 2, 4] 183s 0.284 183s 0.952 183s 0.730 183s 183s x1: [1, 1, 2, 4] 183s 0.993 183s 183s y: [3, 1, 2, 4] 183s 0.282 0.945 0.725 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.505 183s 0.869 183s 0.405 183s -0.212 183s 183s x1: [1, 3, 1, 1] 183s 0.170 183s -0.305 183s -0.713 183s 183s y: [4, 3, 1, 1] 183s 0.086 0.148 0.069 -0.036 183s -0.154 -0.265 -0.124 0.064 183s -0.360 -0.619 -0.289 0.151 183s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.454 -0.494 -0.823 0.253 183s 0.246 0.664 0.359 -0.918 183s 0.605 0.698 0.608 -0.750 183s 183s x1: [4, 1, 2, 1] 183s 0.820 0.136 -0.211 0.625 183s 183s y: [3, 1, 2, 1] 183s -0.107 -0.357 -0.006 183s testing: mul_mat, [1, 4, 2, 4] = [3, 1, 2, 4] * [3, 4, 2, 4] 183s x0: [3, 1, 2, 4] 183s 0.003 -0.138 0.991 183s 183s x1: [3, 4, 2, 4] 183s -0.900 -0.769 -0.915 183s 0.835 -0.460 -0.235 183s -0.863 0.086 -0.989 183s 0.742 0.591 0.675 183s 183s y: [1, 4, 2, 4] 183s -0.804 183s -0.167 183s -0.995 183s 0.589 183s test-mul-mat0: iter:100/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.241 -0.852 0.192 183s 183s x1: [3, 1, 1, 1] 183s -0.470 0.897 -0.843 183s 183s y: [1, 1, 1, 1] 183s -0.814 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.048 -0.746 -0.447 183s 0.714 0.220 0.641 183s 183s x1: [3, 1, 1, 1] 183s 0.272 0.929 0.008 183s 183s y: [2, 1, 1, 1] 183s -0.710 0.405 183s testing: mul_mat, [4, 2, 1, 3] = [3, 4, 1, 3] * [3, 2, 1, 3] 183s x0: [3, 4, 1, 3] 183s -0.356 -0.589 0.746 183s -0.427 0.419 -0.252 183s 0.525 0.673 0.301 183s 0.240 -0.107 -0.058 183s 183s x1: [3, 2, 1, 3] 183s -0.986 0.545 0.732 183s -0.635 0.226 -0.046 183s 183s y: [4, 2, 1, 3] 183s 0.576 0.465 0.069 -0.337 183s 0.058 0.377 -0.195 -0.174 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.973 0.088 -0.327 183s 0.882 0.321 0.628 183s 0.655 0.995 -0.324 183s -0.924 -0.366 0.244 183s 183s x1: [3, 4, 1, 1] 183s -0.399 -0.216 -0.926 183s 0.902 -0.976 -0.033 183s -0.156 0.277 0.405 183s 0.517 -0.105 0.069 183s 183s y: [4, 4, 1, 1] 183s -0.104 -1.003 -0.176 0.222 183s 0.803 0.462 -0.369 -0.485 183s -0.260 0.206 0.042 0.142 183s 0.471 0.466 0.212 -0.423 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.294 -0.547 -0.857 0.575 183s 0.073 0.602 0.597 -0.975 183s 183s x1: [4, 3, 1, 1] 183s 0.735 0.362 -0.774 0.588 183s -0.379 -0.373 -0.810 0.405 183s -0.299 -0.908 0.429 0.668 183s 183s y: [2, 3, 1, 1] 183s 0.589 -0.764 183s 1.243 -1.131 183s 0.601 -0.963 183s testing: mul_mat, [2, 4, 1, 3] = [2, 2, 1, 3] * [2, 4, 1, 3] 183s x0: [2, 2, 1, 3] 183s -0.732 -0.074 183s 0.549 0.865 183s 183s x1: [2, 4, 1, 3] 183s 0.252 0.651 183s 0.113 -0.427 183s 0.646 0.747 183s 0.246 0.274 183s 183s y: [2, 4, 1, 3] 183s -0.232 0.701 183s -0.051 -0.308 183s -0.528 1.001 183s -0.200 0.372 183s test-mul-mat0: iter:101/500 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.763 0.525 183s -0.033 0.478 183s 0.235 -0.217 183s 183s x1: [2, 3, 1, 1] 183s -0.296 -0.980 183s -0.860 0.301 183s -0.574 0.482 183s 183s y: [3, 3, 1, 1] 183s -0.288 -0.459 0.143 183s 0.814 0.172 -0.268 183s 0.691 0.249 -0.240 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.896 -0.967 183s -0.127 0.960 183s 0.723 0.294 183s 0.771 -0.341 183s 183s x1: [2, 3, 1, 1] 183s 0.425 -0.981 183s 0.051 -0.026 183s -0.055 -0.084 183s 183s y: [4, 3, 1, 1] 183s 1.329 -0.996 0.019 0.663 183s 0.071 -0.032 0.029 0.048 183s 0.032 -0.073 -0.064 -0.014 183s testing: mul_mat, [2, 4, 1, 2] = [2, 2, 1, 2] * [2, 4, 1, 2] 183s x0: [2, 2, 1, 2] 183s -0.514 0.882 183s -0.434 0.431 183s 183s x1: [2, 4, 1, 2] 183s -0.012 -0.994 183s -0.916 0.692 183s -0.974 -0.776 183s -0.007 -0.548 183s 183s y: [2, 4, 1, 2] 183s -0.871 -0.423 183s 1.082 0.695 183s -0.183 0.088 183s -0.480 -0.233 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.214 0.064 183s 0.820 0.298 183s 183s x1: [2, 2, 1, 1] 183s 0.538 0.287 183s -0.445 -0.169 183s 183s y: [2, 2, 1, 1] 183s 0.133 0.526 183s -0.106 -0.415 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.228 -0.769 183s 0.586 -0.540 183s 183s x1: [2, 2, 1, 1] 183s 0.512 0.090 183s 0.522 -0.495 183s 183s y: [2, 2, 1, 1] 183s 0.048 0.251 183s 0.500 0.573 183s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 1, 1, 2] 183s 0.146 0.221 183s 183s x1: [2, 2, 1, 2] 183s 0.634 -0.602 183s -0.074 -0.535 183s 183s y: [1, 2, 1, 2] 183s -0.040 183s -0.129 183s test-mul-mat0: iter:102/500 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.112 -0.171 -0.408 0.459 183s 183s x1: [4, 4, 1, 1] 183s -0.485 0.747 0.410 0.579 183s 0.044 -0.079 -0.331 -0.434 183s 0.426 -0.790 0.794 0.012 183s -0.559 -0.746 -0.354 -0.161 183s 183s y: [1, 4, 1, 1] 183s -0.083 183s -0.046 183s -0.136 183s 0.135 183s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 1, 4, 1] 183s -0.527 -0.287 -0.965 -0.733 183s 183s x1: [4, 1, 4, 1] 183s 0.386 0.435 -0.395 -0.393 183s 183s y: [1, 1, 4, 1] 183s 0.340 183s testing: mul_mat, [1, 1, 4, 3] = [4, 1, 4, 3] * [4, 1, 4, 3] 183s x0: [4, 1, 4, 3] 183s 0.639 -0.130 0.211 -0.085 183s 183s x1: [4, 1, 4, 3] 183s 0.547 -0.862 0.332 -0.870 183s 183s y: [1, 1, 4, 3] 183s 0.605 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.931 0.731 0.128 -0.208 183s -0.741 -0.094 -0.149 -0.363 183s -0.566 0.824 -0.817 0.981 183s -0.564 0.522 -0.439 -0.375 183s 183s x1: [4, 1, 1, 1] 183s 0.681 -0.167 0.821 -0.720 183s 183s y: [4, 1, 1, 1] 183s 0.767 -0.351 -1.900 -0.562 183s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.481 0.055 0.944 -0.716 183s 183s x1: [4, 4, 4, 1] 183s 0.287 -0.691 0.618 0.480 183s 0.996 -0.234 -0.890 0.633 183s -0.603 -0.799 0.315 0.230 183s -0.978 0.595 0.956 0.953 183s 183s y: [1, 4, 4, 1] 183s 0.339 183s -0.827 183s -0.202 183s -0.217 183s testing: mul_mat, [4, 4, 4, 3] = [1, 4, 4, 3] * [1, 4, 4, 3] 183s x0: [1, 4, 4, 3] 183s 0.027 183s 0.814 183s -0.135 183s 0.357 183s 183s x1: [1, 4, 4, 3] 183s -0.090 183s 0.122 183s -0.761 183s 0.268 183s 183s y: [4, 4, 4, 3] 183s -0.002 -0.073 0.012 -0.032 183s 0.003 0.099 -0.017 0.044 183s -0.020 -0.620 0.103 -0.272 183s 0.007 0.218 -0.036 0.096 183s test-mul-mat0: iter:103/500 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.561 183s 0.479 183s 0.673 183s 0.586 183s 183s x1: [1, 3, 1, 1] 183s 0.505 183s 0.520 183s -0.633 183s 183s y: [4, 3, 1, 1] 183s 0.283 0.242 0.340 0.296 183s 0.292 0.249 0.350 0.305 183s -0.355 -0.303 -0.426 -0.371 183s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.223 183s 0.389 183s 0.803 183s -0.664 183s 183s x1: [1, 4, 2, 1] 183s 0.871 183s -0.518 183s -0.362 183s -0.922 183s 183s y: [4, 4, 2, 1] 183s -0.195 0.339 0.700 -0.578 183s 0.116 -0.201 -0.416 0.344 183s 0.081 -0.141 -0.291 0.240 183s 0.206 -0.359 -0.741 0.612 183s testing: mul_mat, [1, 4, 2, 2] = [1, 1, 2, 2] * [1, 4, 2, 2] 183s x0: [1, 1, 2, 2] 183s -0.741 183s 183s x1: [1, 4, 2, 2] 183s -0.076 183s -0.398 183s -0.495 183s -0.750 183s 183s y: [1, 4, 2, 2] 183s 0.056 183s 0.295 183s 0.367 183s 0.556 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.847 183s 183s x1: [1, 1, 1, 1] 183s -0.813 183s 183s y: [1, 1, 1, 1] 183s -0.689 183s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.620 183s 183s x1: [1, 1, 2, 1] 183s 0.576 183s 183s y: [1, 1, 2, 1] 183s -0.357 183s testing: mul_mat, [4, 1, 2, 2] = [1, 4, 2, 2] * [1, 1, 2, 2] 183s x0: [1, 4, 2, 2] 183s -0.226 183s -0.143 183s 0.117 183s -0.471 183s 183s x1: [1, 1, 2, 2] 183s -0.052 183s 183s y: [4, 1, 2, 2] 183s 0.012 0.007 -0.006 0.025 183s test-mul-mat0: iter:104/500 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.209 -0.085 -0.478 0.204 183s 183s x1: [4, 4, 1, 1] 183s 0.618 -0.982 -0.643 0.802 183s -0.602 0.958 -0.250 -0.454 183s -0.518 -0.700 0.198 0.256 183s 0.157 -0.685 0.784 -0.159 183s 183s y: [1, 4, 1, 1] 183s 0.684 183s -0.181 183s -0.091 183s -0.316 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.494 0.945 0.434 -0.052 183s 0.382 0.053 -0.034 0.740 183s -0.145 0.364 0.698 0.605 183s 183s x1: [4, 1, 1, 1] 183s -0.037 -0.037 -0.749 0.624 183s 183s y: [3, 1, 1, 1] 183s -0.374 0.471 -0.153 183s testing: mul_mat, [3, 3, 1, 4] = [4, 3, 1, 4] * [4, 3, 1, 4] 183s x0: [4, 3, 1, 4] 183s -0.989 0.342 -0.999 -0.721 183s -0.320 -0.268 0.600 -0.015 183s 0.203 0.776 -0.347 0.371 183s 183s x1: [4, 3, 1, 4] 183s 0.910 -0.820 0.905 0.108 183s 0.436 0.062 0.423 0.220 183s 0.903 0.686 -0.571 -0.182 183s 183s y: [3, 3, 1, 4] 183s -2.163 0.470 -0.726 183s -0.991 0.095 0.071 183s 0.043 -0.813 0.846 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.439 -0.433 0.304 0.244 183s 0.480 0.777 0.673 0.346 183s 0.248 0.228 -0.807 0.811 183s 183s x1: [4, 3, 1, 1] 183s -0.466 -0.541 0.799 0.042 183s 0.098 -0.375 -0.018 0.203 183s 0.456 0.989 -0.444 0.463 183s 183s y: [3, 3, 1, 1] 183s 0.692 -0.091 -0.850 183s 0.164 -0.186 0.118 183s -0.650 0.849 1.072 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.807 0.775 0.329 183s -0.688 -0.918 0.492 183s -0.866 -0.075 0.031 183s -0.637 0.255 -0.443 183s 183s x1: [3, 4, 1, 1] 183s -0.444 -0.424 0.661 183s 0.587 0.047 -0.569 183s -0.879 0.506 -0.769 183s 0.163 -0.396 -0.144 183s 183s y: [4, 4, 1, 1] 183s -0.469 1.020 0.437 -0.118 183s 0.322 -0.727 -0.529 -0.110 183s -0.570 -0.239 0.698 1.029 183s -0.223 0.181 -0.116 -0.141 183s testing: mul_mat, [1, 3, 1, 4] = [4, 1, 1, 4] * [4, 3, 1, 4] 183s x0: [4, 1, 1, 4] 183s 0.372 0.726 -0.424 0.118 183s 183s x1: [4, 3, 1, 4] 183s -0.204 -0.295 -0.250 -0.959 183s -0.949 -0.439 -0.404 -0.373 183s -0.778 -0.817 0.674 -0.347 183s 183s y: [1, 3, 1, 4] 183s -0.298 183s -0.545 183s -1.209 183s test-mul-mat0: iter:105/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.392 -0.703 183s 0.246 0.151 183s -0.569 0.771 183s 183s x1: [2, 2, 1, 1] 183s -0.223 -0.441 183s 0.099 0.947 183s 183s y: [3, 2, 1, 1] 183s 0.223 -0.121 -0.213 183s -0.627 0.167 0.673 183s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.847 -0.699 183s -0.329 -0.054 183s 183s x1: [2, 3, 3, 1] 183s -0.477 -0.843 183s -0.653 0.641 183s 0.838 -0.250 183s 183s y: [2, 3, 3, 1] 183s 0.185 0.203 183s -1.001 0.181 183s 0.884 -0.263 183s testing: mul_mat, [3, 2, 3, 2] = [2, 3, 3, 2] * [2, 2, 3, 2] 183s x0: [2, 3, 3, 2] 183s 0.992 -0.833 183s -0.463 0.480 183s 0.142 -0.315 183s 183s x1: [2, 2, 3, 2] 183s -0.989 0.133 183s -0.505 0.652 183s 183s y: [3, 2, 3, 2] 183s -1.091 0.521 -0.183 183s -1.045 0.547 -0.278 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.850 -0.712 183s -0.730 -0.532 183s 183s x1: [2, 3, 1, 1] 183s -0.216 0.622 183s -0.588 0.963 183s -0.234 0.657 183s 183s y: [2, 3, 1, 1] 183s -0.259 -0.173 183s -0.186 -0.083 183s -0.269 -0.178 183s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.885 0.584 183s 183s x1: [2, 2, 3, 1] 183s 0.581 0.496 183s -0.766 0.018 183s 183s y: [1, 2, 3, 1] 183s -0.225 183s 0.688 183s testing: mul_mat, [2, 2, 3, 2] = [1, 2, 3, 2] * [1, 2, 3, 2] 183s x0: [1, 2, 3, 2] 183s -0.399 183s -0.001 183s 183s x1: [1, 2, 3, 2] 183s 0.246 183s -0.532 183s 183s y: [2, 2, 3, 2] 183s -0.098 -0.000 183s 0.213 0.001 183s test-mul-mat0: iter:106/500 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.126 -0.990 0.250 0.979 183s 183s x1: [4, 2, 1, 1] 183s 0.158 -0.814 -0.306 0.626 183s 0.567 -0.649 -0.944 0.714 183s 183s y: [1, 2, 1, 1] 183s 1.322 183s 1.034 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.433 0.021 -0.902 -0.630 183s 183s x1: [4, 1, 1, 1] 183s -0.225 -0.585 -0.420 0.774 183s 183s y: [1, 1, 1, 1] 183s -0.219 183s testing: mul_mat, [4, 1, 1, 3] = [4, 4, 1, 3] * [4, 1, 1, 3] 183s x0: [4, 4, 1, 3] 183s -0.104 0.426 0.094 -0.390 183s 0.091 0.968 -0.380 -0.659 183s 0.947 0.395 -0.243 -0.474 183s 0.169 0.356 0.960 -0.809 183s 183s x1: [4, 1, 1, 3] 183s -0.111 0.918 -0.406 -0.221 183s 183s y: [4, 1, 1, 3] 183s 0.451 1.178 0.461 0.097 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.497 0.853 0.451 0.265 183s 0.280 0.186 0.422 -0.622 183s 183s x1: [4, 4, 1, 1] 183s -0.615 -0.482 0.109 0.308 183s -0.623 0.196 0.164 -0.511 183s -0.574 0.401 -0.987 -0.015 183s -0.287 -0.425 -0.561 0.451 183s 183s y: [2, 4, 1, 1] 183s 0.026 -0.408 183s 0.416 0.248 183s 0.178 -0.493 183s -0.353 -0.677 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.265 -0.154 183s 183s x1: [2, 4, 1, 1] 183s 0.517 -0.487 183s -0.339 0.573 183s 0.951 -0.602 183s 0.957 -0.531 183s 183s y: [1, 4, 1, 1] 183s 0.212 183s -0.178 183s 0.344 183s 0.335 183s testing: mul_mat, [3, 2, 1, 3] = [1, 3, 1, 3] * [1, 2, 1, 3] 183s x0: [1, 3, 1, 3] 183s -0.459 183s 0.923 183s -0.447 183s 183s x1: [1, 2, 1, 3] 183s -0.297 183s -0.572 183s 183s y: [3, 2, 1, 3] 183s 0.136 -0.274 0.133 183s 0.262 -0.528 0.256 183s test-mul-mat0: iter:107/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.483 0.972 183s -0.839 -0.526 183s -0.559 0.668 183s 183s x1: [2, 1, 1, 1] 183s -0.055 -0.980 183s 183s y: [3, 1, 1, 1] 183s -0.925 0.562 -0.624 183s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.058 -0.035 183s 0.551 0.416 183s 0.406 0.220 183s 0.155 -0.307 183s 183s x1: [2, 3, 4, 1] 183s 0.739 0.287 183s -0.628 -0.833 183s -0.377 0.501 183s 183s y: [4, 3, 4, 1] 183s -0.053 0.527 0.363 0.026 183s 0.066 -0.693 -0.438 0.159 183s 0.004 0.000 -0.043 -0.212 183s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 3, 4, 1] 183s -0.495 -0.089 183s -0.683 0.202 183s 0.508 -0.592 183s 183s x1: [2, 4, 4, 1] 183s -0.502 0.280 183s 0.542 0.904 183s -0.501 -0.303 183s -0.403 -0.909 183s 183s y: [3, 4, 4, 1] 183s 0.224 0.399 -0.421 183s -0.349 -0.188 -0.259 183s 0.275 0.281 -0.075 183s 0.280 0.092 0.333 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.452 -0.037 183s 183s x1: [2, 3, 1, 1] 183s -0.637 0.985 183s 0.044 -0.099 183s -0.053 -0.948 183s 183s y: [1, 3, 1, 1] 183s -0.324 183s 0.024 183s 0.011 183s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 4, 4, 1] 183s -0.481 183s -0.823 183s 0.513 183s -0.229 183s 183s x1: [1, 2, 4, 1] 183s 0.035 183s -0.347 183s 183s y: [4, 2, 4, 1] 183s -0.017 -0.029 0.018 -0.008 183s 0.167 0.285 -0.178 0.079 183s testing: mul_mat, [3, 1, 4, 1] = [4, 3, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 3, 4, 1] 183s -0.566 -0.444 -0.823 -0.053 183s -0.081 -0.188 -0.736 -0.750 183s 0.785 -0.635 -0.502 -0.006 183s 183s x1: [4, 1, 4, 1] 183s 0.397 0.701 0.082 -0.885 183s 183s y: [3, 1, 4, 1] 183s -0.557 0.439 -0.170 183s test-mul-mat0: iter:108/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.420 -0.234 -0.415 183s 183s x1: [3, 1, 1, 1] 183s -0.610 0.247 -0.113 183s 183s y: [1, 1, 1, 1] 183s 0.246 183s testing: mul_mat, [2, 1, 2, 1] = [3, 2, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 2, 2, 1] 183s 0.498 0.837 -0.813 183s 0.520 -0.076 -0.445 183s 183s x1: [3, 1, 2, 1] 183s -0.166 -0.207 0.024 183s 183s y: [2, 1, 2, 1] 183s -0.275 -0.082 183s testing: mul_mat, [3, 2, 2, 1] = [3, 3, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 3, 2, 1] 183s 0.540 -0.444 0.763 183s 0.280 0.804 -0.471 183s -0.223 0.641 -0.284 183s 183s x1: [3, 2, 2, 1] 183s 0.259 -0.542 -0.876 183s 0.582 0.094 -0.486 183s 183s y: [3, 2, 2, 1] 183s -0.288 0.049 -0.156 183s -0.098 0.467 0.069 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.330 -0.248 0.663 183s 0.107 -0.130 -0.485 183s 183s x1: [3, 3, 1, 1] 183s -0.308 0.508 -0.886 183s -0.726 -0.398 -0.371 183s 0.102 -0.417 0.575 183s 183s y: [2, 3, 1, 1] 183s -0.815 0.330 183s -0.387 0.154 183s 0.518 -0.214 183s testing: mul_mat, [2, 3, 2, 1] = [2, 2, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.706 0.438 183s -0.084 0.802 183s 183s x1: [2, 3, 2, 1] 183s -0.850 0.467 183s 0.045 -0.072 183s 0.108 0.761 183s 183s y: [2, 3, 2, 1] 183s 0.804 0.446 183s -0.063 -0.062 183s 0.257 0.601 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.409 0.471 183s 0.338 0.636 183s 183s x1: [2, 2, 2, 1] 183s -0.186 -0.013 183s 0.235 0.324 183s 183s y: [2, 2, 2, 1] 183s 0.070 -0.071 183s 0.057 0.285 183s test-mul-mat0: iter:109/500 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.012 183s 0.593 183s 0.300 183s 183s x1: [1, 3, 1, 1] 183s 0.353 183s 0.550 183s 0.791 183s 183s y: [3, 3, 1, 1] 183s -0.004 0.210 0.106 183s -0.006 0.326 0.165 183s -0.009 0.469 0.238 183s testing: mul_mat, [2, 3, 2, 1] = [1, 2, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.301 183s -0.464 183s 183s x1: [1, 3, 2, 1] 183s -0.300 183s 0.997 183s -0.040 183s 183s y: [2, 3, 2, 1] 183s -0.090 0.139 183s 0.300 -0.463 183s -0.012 0.019 183s testing: mul_mat, [3, 2, 2, 4] = [1, 3, 2, 4] * [1, 2, 2, 4] 183s x0: [1, 3, 2, 4] 183s -0.077 183s 0.723 183s 0.888 183s 183s x1: [1, 2, 2, 4] 183s -0.460 183s 0.505 183s 183s y: [3, 2, 2, 4] 183s 0.035 -0.333 -0.408 183s -0.039 0.365 0.448 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.685 183s -0.968 183s 0.942 183s 0.238 183s 183s x1: [1, 3, 1, 1] 183s 0.036 183s -0.620 183s 0.902 183s 183s y: [4, 3, 1, 1] 183s -0.025 -0.035 0.034 0.009 183s 0.425 0.600 -0.584 -0.148 183s -0.618 -0.873 0.849 0.215 183s testing: mul_mat, [4, 1, 2, 1] = [4, 4, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 4, 2, 1] 183s -0.814 0.416 -0.206 0.408 183s 0.298 0.259 0.414 0.829 183s -0.135 0.002 -0.734 0.902 183s -0.319 0.546 -0.134 0.788 183s 183s x1: [4, 1, 2, 1] 183s 0.756 0.830 -0.138 0.476 183s 183s y: [4, 1, 2, 1] 183s -0.047 0.777 0.430 0.605 183s testing: mul_mat, [2, 4, 2, 4] = [4, 2, 2, 4] * [4, 4, 2, 4] 183s x0: [4, 2, 2, 4] 183s -0.556 -0.396 -0.664 0.449 183s 0.754 -0.437 -0.287 -0.574 183s 183s x1: [4, 4, 2, 4] 183s -0.278 0.042 -0.957 -0.862 183s -0.699 0.044 0.684 0.095 183s -0.542 0.950 0.961 0.867 183s 0.778 0.863 0.655 0.509 183s 183s y: [2, 4, 2, 4] 183s 0.387 0.542 183s -0.041 -0.797 183s -0.325 -1.596 183s -0.982 -0.270 183s test-mul-mat0: iter:110/500 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.792 0.563 183s -0.563 -0.376 183s -0.371 0.355 183s 183s x1: [2, 3, 1, 1] 183s 0.092 0.179 183s -0.886 -0.255 183s 0.979 0.153 183s 183s y: [3, 3, 1, 1] 183s 0.028 -0.119 0.029 183s 0.558 0.595 0.238 183s -0.690 -0.609 -0.309 183s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.856 -0.153 183s 0.765 -0.233 183s 0.476 0.120 183s -0.530 -0.322 183s 183s x1: [2, 3, 4, 1] 183s 0.703 0.202 183s 0.118 -0.296 183s 0.937 0.744 183s 183s y: [4, 3, 4, 1] 183s -0.633 0.491 0.359 -0.437 183s -0.056 0.159 0.021 0.032 183s -0.916 0.544 0.535 -0.736 183s testing: mul_mat, [4, 4, 4, 3] = [2, 4, 4, 3] * [2, 4, 4, 3] 183s x0: [2, 4, 4, 3] 183s 0.847 0.948 183s -0.692 0.272 183s -0.006 -0.222 183s 0.419 -0.605 183s 183s x1: [2, 4, 4, 3] 183s -0.258 0.496 183s -0.389 -0.782 183s -0.384 0.081 183s -0.104 -0.146 183s 183s y: [4, 4, 4, 3] 183s 0.252 0.314 -0.109 -0.408 183s -1.070 0.056 0.176 0.310 183s -0.248 0.288 -0.016 -0.210 183s -0.226 0.032 0.033 0.045 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.434 -0.633 183s 183s x1: [2, 4, 1, 1] 183s 0.185 0.121 183s -0.935 -0.439 183s -0.244 0.800 183s -0.654 0.065 183s 183s y: [1, 4, 1, 1] 183s 0.004 183s -0.129 183s -0.612 183s -0.326 183s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 3, 4, 1] 183s 0.975 183s 0.824 183s -0.654 183s 183s x1: [1, 2, 4, 1] 183s -0.342 183s -0.877 183s 183s y: [3, 2, 4, 1] 183s -0.334 -0.282 0.224 183s -0.855 -0.722 0.573 183s testing: mul_mat, [4, 1, 4, 3] = [3, 4, 4, 3] * [3, 1, 4, 3] 183s x0: [3, 4, 4, 3] 183s 0.420 -0.280 0.936 183s 0.106 -0.787 0.556 183s -0.464 0.090 -0.707 183s -0.501 -0.433 0.124 183s 183s x1: [3, 1, 4, 3] 183s -0.110 -0.761 0.774 183s 183s y: [4, 1, 4, 3] 183s 0.892 1.018 -0.565 0.480 183s test-mul-mat0: iter:111/500 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.093 -0.529 -0.654 -0.424 183s -0.914 -0.711 -0.504 -0.587 183s 183s x1: [4, 3, 1, 1] 183s -0.855 0.586 0.746 0.154 183s 0.296 -0.482 -0.519 0.243 183s -0.906 -0.513 -0.095 -0.734 183s 183s y: [2, 3, 1, 1] 183s -0.782 -0.101 183s 0.463 0.191 183s 0.729 1.672 183s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.520 -0.547 -0.533 -0.231 183s -0.051 -0.119 0.116 0.343 183s 0.364 0.236 -0.779 -0.390 183s 0.462 0.269 -0.449 -0.252 183s 183s x1: [4, 2, 3, 1] 183s -0.653 -0.606 -0.517 -0.880 183s -0.122 0.246 -0.774 0.049 183s 183s y: [4, 2, 3, 1] 183s 1.149 -0.256 0.365 -0.012 183s 0.330 -0.096 0.598 0.346 183s testing: mul_mat, [1, 4, 3, 3] = [4, 1, 3, 3] * [4, 4, 3, 3] 183s x0: [4, 1, 3, 3] 183s -0.207 0.234 0.732 -0.926 183s 183s x1: [4, 4, 3, 3] 183s 0.218 -0.319 0.322 -0.884 183s -0.987 -0.038 0.080 -0.012 183s 0.917 0.015 0.874 0.006 183s -0.172 0.142 -0.425 0.093 183s 183s y: [1, 4, 3, 3] 183s 0.934 183s 0.265 183s 0.448 183s -0.328 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.711 -0.365 0.712 0.168 183s 183s x1: [4, 1, 1, 1] 183s 0.526 0.518 -0.056 -0.502 183s 183s y: [1, 1, 1, 1] 183s 0.061 183s testing: mul_mat, [2, 4, 3, 1] = [1, 2, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.376 183s -0.215 183s 183s x1: [1, 4, 3, 1] 183s -0.648 183s -0.860 183s -0.248 183s 0.099 183s 183s y: [2, 4, 3, 1] 183s 0.244 0.139 183s 0.324 0.185 183s 0.093 0.053 183s -0.037 -0.021 183s testing: mul_mat, [4, 1, 3, 3] = [2, 4, 3, 3] * [2, 1, 3, 3] 183s x0: [2, 4, 3, 3] 183s 0.672 -0.140 183s 0.172 0.196 183s -0.083 0.684 183s 0.060 0.645 183s 183s x1: [2, 1, 3, 3] 183s 0.978 -0.361 183s 183s y: [4, 1, 3, 3] 183s 0.708 0.097 -0.329 -0.174 183s test-mul-mat0: iter:112/500 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.587 -0.562 -0.181 183s -0.623 -0.932 -0.604 183s 183s x1: [3, 4, 1, 1] 183s -0.792 -0.179 -0.749 183s -0.607 -0.613 -0.606 183s -0.598 0.194 0.705 183s -0.624 -0.769 -0.793 183s 183s y: [2, 4, 1, 1] 183s -0.228 1.112 183s 0.098 1.316 183s -0.587 -0.234 183s 0.210 1.584 183s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.849 0.052 0.612 183s -0.618 0.259 -0.186 183s 0.968 0.697 0.634 183s 183s x1: [3, 2, 3, 1] 183s 0.520 -0.842 -0.205 183s -0.790 -0.608 0.659 183s 183s y: [3, 2, 3, 1] 183s -0.611 -0.502 -0.214 183s 1.043 0.208 -0.771 183s testing: mul_mat, [4, 3, 3, 2] = [3, 4, 3, 2] * [3, 3, 3, 2] 183s x0: [3, 4, 3, 2] 183s 0.470 -0.598 0.706 183s 0.331 0.861 0.189 183s -0.360 -0.743 -0.850 183s -0.253 -0.175 -0.261 183s 183s x1: [3, 3, 3, 2] 183s -0.063 0.478 -0.203 183s -0.827 0.628 0.848 183s 0.785 -0.990 0.108 183s 183s y: [4, 3, 3, 2] 183s -0.459 0.352 -0.160 -0.014 183s -0.166 0.428 -0.891 -0.122 183s 1.037 -0.573 0.362 -0.054 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.980 0.324 -0.659 183s 0.832 0.721 -0.234 183s 0.977 0.266 -0.560 183s -0.529 -0.709 0.168 183s 183s x1: [3, 4, 1, 1] 183s -0.463 -0.698 0.561 183s 0.364 -0.099 0.621 183s -0.678 0.415 0.312 183s -0.402 -0.255 0.073 183s 183s y: [4, 4, 1, 1] 183s -1.049 -1.020 -0.952 0.834 183s -0.084 0.087 -0.018 -0.018 183s -0.735 -0.337 -0.726 0.116 183s -0.525 -0.536 -0.502 0.406 183s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.336 -0.963 0.889 -0.045 183s -0.170 0.289 -0.226 0.373 183s -0.999 -0.768 -0.974 0.068 183s -0.713 -0.341 0.562 0.112 183s 183s x1: [4, 3, 3, 1] 183s -0.167 -0.039 -0.160 0.722 183s -0.997 -0.894 -0.741 -0.695 183s 0.667 0.623 0.207 0.288 183s 183s y: [4, 3, 3, 1] 183s -0.081 0.323 0.402 0.123 183s 0.569 -0.181 2.357 0.522 183s -0.653 0.127 -1.326 -0.540 183s testing: mul_mat, [2, 4, 3, 2] = [4, 2, 3, 2] * [4, 4, 3, 2] 183s x0: [4, 2, 3, 2] 183s 0.283 0.653 0.980 -0.385 183s -0.706 -0.906 0.784 0.715 183s 183s x1: [4, 4, 3, 2] 183s -0.134 -0.772 -0.982 0.884 183s 0.438 -0.855 -0.537 -0.965 183s -0.045 0.563 -0.140 -0.830 183s 0.188 -0.981 -0.950 0.644 183s 183s y: [2, 4, 3, 2] 183s -1.845 0.656 183s -0.590 -0.646 183s 0.537 -1.181 183s -1.767 0.470 183s test-mul-mat0: iter:113/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.862 0.819 0.567 -0.428 183s 0.359 0.714 0.871 -0.555 183s 0.941 0.192 0.223 -0.807 183s 183s x1: [4, 3, 1, 1] 183s 0.401 0.031 0.532 0.618 183s -0.848 -0.402 -0.514 -0.406 183s -0.851 -0.517 -0.681 0.994 183s 183s y: [3, 3, 1, 1] 183s -0.283 0.287 0.003 183s 0.285 -0.814 -0.662 183s -0.502 -1.820 -1.854 183s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.960 0.621 0.365 -0.187 183s 183s x1: [4, 3, 4, 1] 183s -0.775 -0.237 -0.719 0.164 183s -0.154 -0.655 -0.434 0.877 183s 0.878 -0.816 -0.971 -0.524 183s 183s y: [1, 3, 4, 1] 183s -1.184 183s -0.877 183s 0.079 183s testing: mul_mat, [1, 1, 4, 3] = [4, 1, 4, 3] * [4, 1, 4, 3] 183s x0: [4, 1, 4, 3] 183s -0.165 0.299 -0.764 0.251 183s 183s x1: [4, 1, 4, 3] 183s -0.453 -0.825 -0.106 0.216 183s 183s y: [1, 1, 4, 3] 183s -0.037 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.515 -0.224 0.835 0.730 183s -0.109 -0.264 -0.264 -0.965 183s 0.182 0.754 0.024 -0.469 183s 183s x1: [4, 1, 1, 1] 183s 0.794 -0.999 0.827 0.327 183s 183s y: [3, 1, 1, 1] 183s 1.563 -0.357 -0.741 183s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.535 -0.289 0.641 183s 0.517 0.454 0.226 183s -0.732 -0.918 -0.369 183s 183s x1: [3, 4, 4, 1] 183s -0.125 -0.428 0.217 183s 0.574 0.789 -0.104 183s -0.284 0.576 -0.518 183s -0.240 -0.631 0.065 183s 183s y: [3, 4, 4, 1] 183s 0.196 -0.210 0.405 183s 0.012 0.631 -1.106 183s -0.650 -0.002 -0.130 183s 0.096 -0.396 0.731 183s testing: mul_mat, [4, 3, 4, 3] = [3, 4, 4, 3] * [3, 3, 4, 3] 183s x0: [3, 4, 4, 3] 183s -0.496 -0.966 -0.399 183s -0.531 -0.822 0.109 183s -0.814 0.187 -0.615 183s 0.112 -0.083 -0.732 183s 183s x1: [3, 3, 4, 3] 183s -0.849 0.060 0.994 183s -0.217 -0.344 -0.277 183s 0.999 -0.263 0.828 183s 183s y: [4, 3, 4, 3] 183s -0.034 0.509 0.090 -0.827 183s 0.551 0.368 0.283 0.207 183s -0.571 -0.223 -1.371 -0.472 183s test-mul-mat0: iter:114/500 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.938 183s 0.822 183s 183s x1: [1, 4, 1, 1] 183s 0.886 183s 0.706 183s -0.432 183s 0.615 183s 183s y: [2, 4, 1, 1] 183s 0.831 0.728 183s 0.662 0.581 183s -0.405 -0.355 183s 0.577 0.506 183s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.560 183s -0.906 183s 0.587 183s 183s x1: [1, 2, 2, 1] 183s 0.842 183s 0.036 183s 183s y: [3, 2, 2, 1] 183s -0.472 -0.763 0.494 183s -0.020 -0.032 0.021 183s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.395 183s 183s x1: [1, 3, 2, 1] 183s 0.784 183s 0.809 183s -0.160 183s 183s y: [1, 3, 2, 1] 183s -0.310 183s -0.320 183s 0.063 183s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.316 183s 0.406 183s -0.593 183s 183s x1: [1, 1, 1, 1] 183s -0.966 183s 183s y: [3, 1, 1, 1] 183s 0.305 -0.392 0.572 183s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 4, 2, 1] 183s -0.053 0.453 -0.824 183s 0.013 0.179 0.274 183s 0.085 0.957 -0.619 183s -0.629 -0.403 0.985 183s 183s x1: [3, 1, 2, 1] 183s -0.609 0.974 -0.977 183s 183s y: [4, 1, 2, 1] 183s 1.279 -0.101 1.486 -0.972 183s testing: mul_mat, [3, 3, 2, 1] = [4, 3, 2, 1] * [4, 3, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.176 -0.749 -0.931 0.042 183s -0.762 0.941 0.617 0.205 183s 0.653 0.029 -0.572 0.671 183s 183s x1: [4, 3, 2, 1] 183s 0.818 0.676 0.398 -0.027 183s 0.588 0.243 0.970 -0.465 183s -0.743 0.055 -0.094 0.710 183s 183s y: [3, 3, 2, 1] 183s -1.023 0.254 0.308 183s -1.208 0.284 -0.476 183s 0.207 0.706 0.046 183s test-mul-mat0: iter:115/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.252 -0.188 0.178 -0.007 183s -0.260 0.246 0.639 -0.623 183s 0.548 0.148 -0.679 -0.739 183s 183s x1: [4, 3, 1, 1] 183s 0.979 -0.749 -0.969 -0.197 183s -0.511 0.684 0.054 -0.570 183s -0.287 0.123 -0.953 0.141 183s 183s y: [3, 3, 1, 1] 183s 0.217 -0.935 1.230 183s -0.244 0.691 0.206 183s -0.266 -0.592 0.404 183s testing: mul_mat, [1, 3, 2, 1] = [4, 1, 2, 1] * [4, 3, 2, 1] 183s x0: [4, 1, 2, 1] 183s -0.639 -0.787 0.789 0.509 183s 183s x1: [4, 3, 2, 1] 183s -0.205 -0.048 0.369 -0.307 183s 0.262 -0.285 -0.328 0.512 183s -0.254 0.475 -0.999 -0.570 183s 183s y: [1, 3, 2, 1] 183s 0.304 183s 0.059 183s -1.290 183s testing: mul_mat, [3, 1, 2, 4] = [4, 3, 2, 4] * [4, 1, 2, 4] 183s x0: [4, 3, 2, 4] 183s -0.346 0.772 -0.535 -0.085 183s -0.347 0.610 0.176 0.963 183s 0.284 0.869 -0.582 -0.390 183s 183s x1: [4, 1, 2, 4] 183s 0.419 -0.002 -0.252 -0.865 183s 183s y: [3, 1, 2, 4] 183s 0.062 -1.025 0.602 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.801 0.356 0.146 0.077 183s 183s x1: [4, 3, 1, 1] 183s 0.180 0.682 0.148 0.075 183s -0.953 0.013 -0.541 0.512 183s -0.522 -0.632 0.231 0.295 183s 183s y: [1, 3, 1, 1] 183s 0.125 183s 0.729 183s 0.250 183s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 1, 2, 1] 183s 0.616 183s 183s x1: [1, 4, 2, 1] 183s -0.071 183s 0.120 183s -0.583 183s -0.547 183s 183s y: [1, 4, 2, 1] 183s -0.044 183s 0.074 183s -0.360 183s -0.337 183s testing: mul_mat, [4, 1, 2, 4] = [1, 4, 2, 4] * [1, 1, 2, 4] 183s x0: [1, 4, 2, 4] 183s 0.877 183s -0.145 183s 0.745 183s -0.755 183s 183s x1: [1, 1, 2, 4] 183s 0.163 183s 183s y: [4, 1, 2, 4] 183s 0.143 -0.024 0.121 -0.123 183s test-mul-mat0: iter:116/500 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.702 -0.215 -0.222 0.822 183s 0.169 -0.628 -0.548 0.897 183s 183s x1: [4, 3, 1, 1] 183s -0.064 0.258 0.457 0.332 183s 0.393 -0.180 -0.343 -0.351 183s -0.402 -0.565 0.608 0.563 183s 183s y: [2, 3, 1, 1] 183s 0.161 -0.126 183s -0.450 0.053 183s 0.731 0.458 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.753 -0.681 0.592 0.210 183s 0.651 -0.015 -0.970 -0.692 183s 183s x1: [4, 2, 1, 1] 183s 0.261 0.336 -0.957 0.493 183s 0.487 -0.934 -0.420 0.382 183s 183s y: [2, 2, 1, 1] 183s -0.495 0.752 183s 0.834 0.473 183s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 183s x0: [4, 4, 1, 2] 183s 0.446 -0.365 -0.849 -0.974 183s -0.982 -0.515 0.778 -0.663 183s -0.923 -0.011 0.987 0.063 183s 0.018 -0.705 -0.303 0.646 183s 183s x1: [4, 2, 1, 2] 183s 0.634 -0.372 -0.257 0.242 183s -0.809 0.542 0.540 -0.024 183s 183s y: [4, 2, 1, 2] 183s 0.401 -0.792 -0.819 0.508 183s -0.993 0.952 1.272 -0.576 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.717 -0.026 0.590 -0.945 183s -0.309 0.097 0.814 0.181 183s 183s x1: [4, 4, 1, 1] 183s -0.563 0.236 0.014 -0.545 183s 0.721 -0.207 -0.209 0.799 183s 0.781 -0.221 -0.139 -0.201 183s 0.074 0.558 -0.554 -0.888 183s 183s y: [2, 4, 1, 1] 183s 0.921 0.110 183s -1.389 -0.268 183s -0.446 -0.412 183s 0.446 -0.580 183s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.160 0.952 183s 0.097 -0.104 183s 0.519 0.300 183s 183s x1: [2, 4, 1, 1] 183s -0.533 0.055 183s -0.276 -0.732 183s -0.860 0.712 183s -0.294 0.376 183s 183s y: [3, 4, 1, 1] 183s -0.033 -0.058 -0.260 183s -0.741 0.049 -0.363 183s 0.540 -0.158 -0.233 183s 0.311 -0.068 -0.040 183s testing: mul_mat, [2, 2, 1, 2] = [3, 2, 1, 2] * [3, 2, 1, 2] 183s x0: [3, 2, 1, 2] 183s 0.600 0.378 0.035 183s 0.980 -0.934 -0.899 183s 183s x1: [3, 2, 1, 2] 183s -0.270 0.757 -0.901 183s 0.804 0.315 -0.455 183s 183s y: [2, 2, 1, 2] 183s 0.093 -0.161 183s 0.585 0.903 183s test-mul-mat0: iter:117/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.456 -0.313 0.617 183s 183s x1: [3, 1, 1, 1] 183s -0.456 0.357 -0.117 183s 183s y: [1, 1, 1, 1] 183s -0.392 183s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.217 0.879 0.769 183s -0.116 -0.786 -0.056 183s 183s x1: [3, 1, 3, 1] 183s -0.999 -0.398 -0.572 183s 183s y: [2, 1, 3, 1] 183s -0.573 0.460 183s testing: mul_mat, [3, 2, 3, 2] = [3, 3, 3, 2] * [3, 2, 3, 2] 183s x0: [3, 3, 3, 2] 183s 0.097 -0.093 0.342 183s -0.210 0.302 0.402 183s 0.218 -0.520 0.341 183s 183s x1: [3, 2, 3, 2] 183s -0.892 0.177 0.606 183s -0.891 0.780 -0.966 183s 183s y: [3, 2, 3, 2] 183s 0.104 0.484 -0.080 183s -0.489 0.034 -0.929 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.337 0.816 -0.671 183s 0.974 0.849 0.197 183s 0.911 0.576 -0.118 183s 0.449 0.072 -0.888 183s 183s x1: [3, 3, 1, 1] 183s 0.713 -0.714 -0.222 183s 0.045 0.235 0.515 183s -0.705 0.278 0.349 183s 183s y: [4, 3, 1, 1] 183s -0.674 0.045 0.264 0.466 183s -0.170 0.345 0.115 -0.421 183s 0.230 -0.381 -0.523 -0.607 183s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.823 -0.052 -0.461 0.307 183s -0.582 0.184 0.276 0.462 183s 0.345 0.698 0.756 0.919 183s -0.586 -0.287 0.242 0.248 183s 183s x1: [4, 3, 3, 1] 183s 0.719 0.881 0.192 -0.995 183s 0.846 -0.138 -0.476 -0.755 183s 0.061 -0.762 -0.469 0.839 183s 183s y: [4, 3, 3, 1] 183s -1.032 -0.663 0.094 -0.874 183s -0.702 -0.998 -0.858 -0.758 183s 0.463 0.082 -0.095 0.277 183s testing: mul_mat, [1, 4, 3, 2] = [4, 1, 3, 2] * [4, 4, 3, 2] 183s x0: [4, 1, 3, 2] 183s 0.581 0.967 0.284 0.576 183s 183s x1: [4, 4, 3, 2] 183s 0.738 0.732 -0.819 0.714 183s -0.110 0.456 -0.864 -0.722 183s -0.622 -0.115 0.090 -0.048 183s 0.806 -0.814 0.703 0.145 183s 183s y: [1, 4, 3, 2] 183s 1.315 183s -0.285 183s -0.475 183s -0.036 183s test-mul-mat0: iter:118/500 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.319 -0.367 183s 0.044 -0.829 183s 0.957 -0.206 183s -0.018 0.589 183s 183s x1: [2, 3, 1, 1] 183s -0.683 0.548 183s 0.758 0.898 183s 0.515 0.042 183s 183s y: [4, 3, 1, 1] 183s -0.419 -0.484 -0.767 0.335 183s -0.088 -0.711 0.541 0.515 183s 0.149 -0.012 0.484 0.015 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.881 0.552 183s 0.764 -0.614 183s 0.372 -0.074 183s 0.543 0.689 183s 183s x1: [2, 4, 1, 1] 183s 0.538 0.533 183s 0.531 -0.660 183s 0.337 -0.777 183s -0.315 0.777 183s 183s y: [4, 4, 1, 1] 183s 0.768 0.083 0.160 0.659 183s 0.104 0.812 0.247 -0.166 183s -0.132 0.735 0.183 -0.352 183s 0.152 -0.718 -0.175 0.364 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.478 -0.067 183s -0.102 -0.541 183s 183s x1: [2, 4, 1, 1] 183s -0.527 0.301 183s 0.586 0.988 183s -0.657 0.060 183s 0.307 -0.024 183s 183s y: [2, 4, 1, 1] 183s -0.272 -0.109 183s 0.214 -0.594 183s -0.318 0.035 183s 0.148 -0.018 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.330 0.461 183s -0.724 0.107 183s 183s x1: [2, 2, 1, 1] 183s -0.478 -0.564 183s 0.992 -0.947 183s 183s y: [2, 2, 1, 1] 183s -0.418 0.286 183s -0.109 -0.820 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.968 -0.272 183s -0.745 -0.428 183s 0.271 0.307 183s 0.720 -0.284 183s 183s x1: [2, 2, 1, 1] 183s 0.565 0.342 183s -0.340 0.329 183s 183s y: [4, 2, 1, 1] 183s -0.640 -0.567 0.258 0.309 183s 0.240 0.113 0.009 -0.339 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.346 0.804 0.952 0.399 183s 183s x1: [4, 2, 1, 1] 183s -0.084 -0.634 -0.977 0.892 183s -0.530 0.500 -0.175 0.368 183s 183s y: [1, 2, 1, 1] 183s -1.113 183s 0.199 183s test-mul-mat0: iter:119/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.177 -0.617 -0.200 183s 0.551 -0.045 -0.893 183s 183s x1: [3, 2, 1, 1] 183s -0.611 0.240 0.383 183s 0.049 -0.431 0.111 183s 183s y: [2, 2, 1, 1] 183s -0.117 -0.690 183s 0.235 -0.053 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.412 -0.688 0.749 183s 0.934 -0.884 0.701 183s 0.333 -0.304 -0.018 183s 0.008 -0.263 0.371 183s 183s x1: [3, 2, 1, 1] 183s -0.733 0.871 -0.527 183s -0.711 0.763 -0.056 183s 183s y: [4, 2, 1, 1] 183s -0.691 -1.823 -0.500 -0.431 183s -0.274 -1.377 -0.468 -0.227 183s testing: mul_mat, [4, 4, 1, 2] = [3, 4, 1, 2] * [3, 4, 1, 2] 183s x0: [3, 4, 1, 2] 183s -0.942 -0.050 0.969 183s 0.067 0.687 0.340 183s 0.315 0.807 0.760 183s -0.868 0.039 0.260 183s 183s x1: [3, 4, 1, 2] 183s -0.752 -0.879 -0.580 183s -0.183 0.232 0.500 183s 0.640 0.615 -0.699 183s 0.191 -0.430 -0.592 183s 183s y: [4, 4, 1, 2] 183s 0.190 -0.851 -1.388 0.467 183s 0.645 0.317 0.510 0.298 183s -1.311 0.227 0.166 -0.714 183s -0.732 -0.484 -0.737 -0.337 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.826 0.509 0.047 183s 183s x1: [3, 4, 1, 1] 183s 0.706 -0.340 -0.188 183s -0.008 0.386 0.065 183s 0.978 0.444 -0.985 183s 0.947 -0.489 0.702 183s 183s y: [1, 4, 1, 1] 183s 0.401 183s 0.193 183s 0.987 183s 0.566 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.202 183s 0.869 183s -0.308 183s -0.573 183s 183s x1: [1, 3, 1, 1] 183s 0.957 183s -0.452 183s -0.692 183s 183s y: [4, 3, 1, 1] 183s 0.194 0.832 -0.295 -0.549 183s -0.091 -0.393 0.139 0.259 183s -0.140 -0.601 0.213 0.397 183s testing: mul_mat, [4, 1, 1, 2] = [4, 4, 1, 2] * [4, 1, 1, 2] 183s x0: [4, 4, 1, 2] 183s -0.825 0.958 0.962 -0.534 183s 0.980 0.451 -0.043 -0.760 183s -0.993 0.255 0.192 0.746 183s 0.240 0.905 0.788 0.014 183s 183s x1: [4, 1, 1, 2] 183s 0.838 0.114 -0.908 -0.312 183s 183s y: [4, 1, 1, 2] 183s -1.290 1.149 -1.210 -0.416 183s test-mul-mat0: iter:120/500 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.934 183s 183s x1: [1, 3, 1, 1] 183s -0.223 183s -0.023 183s 0.395 183s 183s y: [1, 3, 1, 1] 183s -0.208 183s -0.022 183s 0.369 183s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 1, 4, 1] 183s 0.414 183s 183s x1: [1, 1, 4, 1] 183s -0.413 183s 183s y: [1, 1, 4, 1] 183s -0.171 183s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.010 183s -0.011 183s 183s x1: [1, 1, 4, 1] 183s 0.792 183s 183s y: [2, 1, 4, 1] 183s 0.008 -0.009 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.668 183s 0.127 183s -0.527 183s 183s x1: [1, 2, 1, 1] 183s -0.157 183s -0.650 183s 183s y: [3, 2, 1, 1] 183s 0.105 -0.020 0.083 183s 0.434 -0.082 0.342 183s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.671 0.135 -0.361 183s -0.875 0.344 -0.405 183s -0.645 0.284 0.866 183s 0.525 -0.159 -0.586 183s 183s x1: [3, 1, 4, 1] 183s 0.728 0.866 -0.593 183s 183s y: [4, 1, 4, 1] 183s -0.157 -0.098 -0.737 0.592 183s testing: mul_mat, [2, 3, 4, 1] = [4, 2, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.165 -0.678 0.570 0.197 183s -0.301 -0.637 0.067 -0.028 183s 183s x1: [4, 3, 4, 1] 183s -0.173 0.122 -0.597 0.618 183s -0.065 -0.995 0.216 0.889 183s -0.727 0.034 0.360 -0.043 183s 183s y: [2, 3, 4, 1] 183s -0.330 -0.083 183s 0.962 0.643 183s 0.054 0.222 183s test-mul-mat0: iter:121/500 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.924 0.210 0.155 0.596 183s 183s x1: [4, 2, 1, 1] 183s 0.279 0.519 0.079 -0.054 183s 0.402 0.895 -0.162 0.165 183s 183s y: [1, 2, 1, 1] 183s -0.168 183s -0.110 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.494 -0.152 0.306 -0.601 183s -0.255 -0.900 -0.763 -0.364 183s -0.940 -0.226 -0.152 -0.618 183s -0.118 -0.873 0.902 0.961 183s 183s x1: [4, 1, 1, 1] 183s 0.543 -0.567 0.286 -0.563 183s 183s y: [4, 1, 1, 1] 183s 0.244 0.359 -0.079 0.147 183s testing: mul_mat, [2, 4, 1, 2] = [4, 2, 1, 2] * [4, 4, 1, 2] 183s x0: [4, 2, 1, 2] 183s 0.482 0.846 -0.403 -0.282 183s 0.832 -0.724 0.110 0.414 183s 183s x1: [4, 4, 1, 2] 183s 0.073 0.304 0.856 0.910 183s -0.531 0.236 0.986 0.679 183s -0.609 0.582 0.223 -0.175 183s -0.133 0.659 0.074 0.374 183s 183s y: [2, 4, 1, 2] 183s -0.310 0.311 183s -0.646 -0.224 183s 0.158 -0.976 183s 0.359 -0.425 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.885 -0.595 -0.847 0.512 183s 183s x1: [4, 2, 1, 1] 183s 0.427 -0.082 0.868 0.436 183s -0.194 -0.592 -0.780 0.542 183s 183s y: [1, 2, 1, 1] 183s -0.085 183s 1.120 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.712 183s -0.982 183s 0.153 183s 183s x1: [1, 4, 1, 1] 183s -0.659 183s -0.720 183s -0.006 183s -0.814 183s 183s y: [3, 4, 1, 1] 183s -0.470 0.648 -0.101 183s -0.513 0.707 -0.110 183s -0.004 0.006 -0.001 183s -0.580 0.799 -0.124 183s testing: mul_mat, [4, 1, 1, 2] = [3, 4, 1, 2] * [3, 1, 1, 2] 183s x0: [3, 4, 1, 2] 183s 0.367 -0.607 0.428 183s 0.475 0.021 -0.550 183s 0.153 -0.770 -0.258 183s -0.206 0.200 -0.242 183s 183s x1: [3, 1, 1, 2] 183s -0.178 0.431 -0.348 183s 183s y: [4, 1, 1, 2] 183s -0.476 0.116 -0.270 0.207 183s test-mul-mat0: iter:122/500 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.550 -0.930 0.226 183s -0.250 0.498 0.676 183s 0.492 -0.744 -0.989 183s -0.361 0.167 0.858 183s 183s x1: [3, 2, 1, 1] 183s -0.164 0.682 0.896 183s -0.798 0.157 0.049 183s 183s y: [4, 2, 1, 1] 183s -0.522 0.987 -1.475 0.942 183s -0.574 0.311 -0.558 0.356 183s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 3, 2, 1] 183s -0.850 0.528 0.563 183s -0.211 -0.305 0.421 183s -0.084 0.053 -0.880 183s 183s x1: [3, 4, 2, 1] 183s -0.873 -0.642 -0.301 183s 0.801 -0.574 0.252 183s 0.492 0.647 -0.373 183s -0.384 0.519 -0.242 183s 183s y: [3, 4, 2, 1] 183s 0.233 0.253 0.304 183s -0.842 0.112 -0.319 183s -0.287 -0.458 0.321 183s 0.465 -0.179 0.272 183s testing: mul_mat, [2, 3, 2, 2] = [3, 2, 2, 2] * [3, 3, 2, 2] 183s x0: [3, 2, 2, 2] 183s 0.166 -0.146 -0.340 183s 0.623 0.627 -0.178 183s 183s x1: [3, 3, 2, 2] 183s -0.724 0.847 0.412 183s 0.930 -0.795 0.116 183s -0.420 -0.886 -0.110 183s 183s y: [2, 3, 2, 2] 183s -0.384 0.006 183s 0.231 0.061 183s 0.097 -0.797 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.424 -0.801 0.831 183s -0.884 0.743 -0.708 183s -0.391 0.431 0.944 183s 183s x1: [3, 2, 1, 1] 183s 0.524 -0.176 -0.478 183s -0.822 0.409 0.262 183s 183s y: [3, 2, 1, 1] 183s -0.478 -0.255 -0.732 183s 0.238 0.845 0.746 183s testing: mul_mat, [4, 3, 2, 1] = [3, 4, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 4, 2, 1] 183s 0.867 0.619 -0.687 183s 0.503 -0.118 -0.528 183s -0.727 -0.359 0.654 183s -0.558 0.362 -0.744 183s 183s x1: [3, 3, 2, 1] 183s 0.035 -0.447 -0.158 183s 0.233 0.443 0.179 183s -0.455 0.677 -0.566 183s 183s y: [4, 3, 2, 1] 183s -0.137 0.154 0.031 -0.064 183s 0.353 -0.030 -0.211 -0.103 183s 0.414 -0.010 -0.282 0.919 183s testing: mul_mat, [2, 3, 2, 2] = [4, 2, 2, 2] * [4, 3, 2, 2] 183s x0: [4, 2, 2, 2] 183s -0.646 0.158 -0.711 0.724 183s 0.734 0.900 0.557 -0.870 183s 183s x1: [4, 3, 2, 2] 183s -0.749 -0.573 -0.837 -0.602 183s -0.115 -0.596 -0.277 -0.249 183s 0.907 -0.004 0.194 0.526 183s 183s y: [2, 3, 2, 2] 183s 0.552 -1.008 183s -0.003 -0.559 183s -0.344 0.313 183s test-mul-mat0: iter:123/500 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.975 183s -0.775 183s -0.548 183s 183s x1: [1, 3, 1, 1] 183s 0.365 183s 0.714 183s -0.374 183s 183s y: [3, 3, 1, 1] 183s 0.355 -0.283 -0.200 183s 0.696 -0.554 -0.392 183s -0.365 0.290 0.205 183s testing: mul_mat, [2, 3, 2, 1] = [1, 2, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.258 183s -0.051 183s 183s x1: [1, 3, 2, 1] 183s 0.217 183s 0.544 183s 0.202 183s 183s y: [2, 3, 2, 1] 183s 0.056 -0.011 183s 0.140 -0.028 183s 0.052 -0.010 183s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.060 183s -0.082 183s -0.813 183s 0.031 183s 183s x1: [1, 2, 2, 1] 183s 0.891 183s 0.299 183s 183s y: [4, 2, 2, 1] 183s -0.053 -0.073 -0.724 0.027 183s -0.018 -0.025 -0.243 0.009 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.529 183s -0.541 183s 183s x1: [1, 4, 1, 1] 183s 0.359 183s -0.265 183s -0.754 183s -0.993 183s 183s y: [2, 4, 1, 1] 183s -0.190 -0.194 183s 0.140 0.143 183s 0.399 0.408 183s 0.526 0.537 183s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 4, 2, 1] 183s 0.445 0.995 183s 0.067 0.626 183s -0.747 -0.714 183s -0.606 0.293 183s 183s x1: [2, 1, 2, 1] 183s 0.926 -0.986 183s 183s y: [4, 1, 2, 1] 183s -0.568 -0.555 0.012 -0.850 183s testing: mul_mat, [4, 2, 2, 1] = [4, 4, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 4, 2, 1] 183s -0.630 -0.375 0.908 0.396 183s -0.661 0.077 -0.404 0.775 183s 0.783 -0.174 0.003 0.940 183s 0.996 0.355 -0.073 0.217 183s 183s x1: [4, 2, 2, 1] 183s 0.013 0.821 -0.608 -0.627 183s -0.444 -0.363 -0.620 0.266 183s 183s y: [4, 2, 2, 1] 183s -1.117 -0.186 -0.725 0.213 183s -0.042 0.723 -0.036 -0.468 183s test-mul-mat0: iter:124/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.061 0.298 -0.564 183s -0.665 0.074 -0.623 183s 0.552 0.241 -0.310 183s 0.325 -0.955 0.382 183s 183s x1: [3, 1, 1, 1] 183s 0.601 -0.656 -0.705 183s 183s y: [4, 1, 1, 1] 183s 0.165 -0.009 0.392 0.552 183s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 4, 3, 1] 183s -0.819 0.221 0.587 183s -0.531 -0.115 -0.177 183s 0.282 -0.933 -0.001 183s 0.695 0.593 0.469 183s 183s x1: [3, 4, 3, 1] 183s -0.637 0.957 -0.086 183s -0.406 -0.008 -0.055 183s 0.547 -0.857 -0.079 183s -0.712 0.664 0.236 183s 183s y: [4, 4, 3, 1] 183s 0.683 0.243 -1.072 0.084 183s 0.298 0.226 -0.106 -0.313 183s -0.684 -0.178 0.954 -0.165 183s 0.868 0.260 -0.820 0.010 183s testing: mul_mat, [3, 4, 3, 2] = [3, 3, 3, 2] * [3, 4, 3, 2] 183s x0: [3, 3, 3, 2] 183s -0.115 -0.832 0.964 183s -0.402 0.035 -0.980 183s -0.451 0.893 -0.347 183s 183s x1: [3, 4, 3, 2] 183s 0.743 0.490 0.633 183s -0.259 0.186 0.226 183s -0.789 -0.196 0.284 183s 0.449 -0.219 -0.858 183s 183s y: [3, 4, 3, 2] 183s 0.117 -0.901 -0.117 183s 0.093 -0.111 0.204 183s 0.528 0.032 0.083 183s -0.698 0.653 -0.100 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.337 0.913 0.878 183s -0.754 0.795 -0.348 183s 183s x1: [3, 3, 1, 1] 183s -0.035 0.494 0.079 183s 0.611 -0.573 -0.321 183s 0.255 -0.100 0.163 183s 183s y: [2, 3, 1, 1] 183s 0.508 0.392 183s -0.599 -0.805 183s 0.138 -0.328 183s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 4, 3, 1] 183s -0.073 -0.482 183s 0.355 0.300 183s -0.767 0.164 183s 0.826 -0.236 183s 183s x1: [2, 3, 3, 1] 183s -0.094 0.170 183s -0.485 -0.453 183s 0.955 -0.436 183s 183s y: [4, 3, 3, 1] 183s -0.075 0.018 0.100 -0.118 183s 0.254 -0.308 0.298 -0.294 183s 0.140 0.208 -0.804 0.892 183s testing: mul_mat, [2, 2, 3, 2] = [4, 2, 3, 2] * [4, 2, 3, 2] 183s x0: [4, 2, 3, 2] 183s 0.873 0.088 0.376 -0.086 183s -0.408 0.922 0.133 0.741 183s 183s x1: [4, 2, 3, 2] 183s -0.289 0.131 -0.489 -0.921 183s 0.631 0.187 0.057 -0.442 183s 183s y: [2, 2, 3, 2] 183s -0.345 -0.509 183s 0.627 -0.406 183s test-mul-mat0: iter:125/500 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.484 0.581 -0.497 183s 0.485 -0.829 -0.036 183s 183s x1: [3, 3, 1, 1] 183s -0.574 0.199 0.340 183s 0.509 0.199 0.743 183s -0.063 0.405 -0.634 183s 183s y: [2, 3, 1, 1] 183s -0.331 -0.455 183s -0.007 0.055 183s 0.520 -0.343 183s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.500 0.935 -0.698 183s 0.985 -0.894 0.266 183s -0.616 -0.391 -0.841 183s -0.078 0.860 -0.090 183s 183s x1: [3, 2, 4, 1] 183s -0.600 -0.497 -0.107 183s -0.462 0.251 -0.248 183s 183s y: [4, 2, 4, 1] 183s -0.089 -0.175 0.654 -0.371 183s 0.639 -0.745 0.396 0.274 183s testing: mul_mat, [1, 4, 4, 3] = [3, 1, 4, 3] * [3, 4, 4, 3] 183s x0: [3, 1, 4, 3] 183s -0.802 -0.071 0.527 183s 183s x1: [3, 4, 4, 3] 183s -0.060 -0.035 -0.282 183s 0.144 -0.179 0.614 183s 0.595 -0.638 -0.880 183s -0.966 -0.474 0.642 183s 183s y: [1, 4, 4, 3] 183s -0.098 183s 0.221 183s -0.896 183s 1.146 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.167 -0.846 0.559 183s 0.003 0.919 -0.384 183s 183s x1: [3, 1, 1, 1] 183s -0.322 0.181 -0.458 183s 183s y: [2, 1, 1, 1] 183s -0.463 0.341 183s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.676 0.595 183s 0.172 -0.269 183s -0.283 0.333 183s 183s x1: [2, 3, 4, 1] 183s -0.345 0.899 183s 0.012 0.239 183s 0.970 0.202 183s 183s y: [3, 3, 4, 1] 183s 0.302 -0.301 0.397 183s 0.150 -0.062 0.076 183s 0.777 0.113 -0.207 183s testing: mul_mat, [4, 2, 4, 3] = [3, 4, 4, 3] * [3, 2, 4, 3] 183s x0: [3, 4, 4, 3] 183s 0.359 0.206 -0.696 183s -0.763 0.697 0.702 183s -0.932 -0.614 -0.195 183s 0.382 0.621 -0.136 183s 183s x1: [3, 2, 4, 3] 183s 0.697 0.360 0.677 183s -0.059 0.869 0.846 183s 183s y: [4, 2, 4, 3] 183s -0.146 0.194 -1.002 0.398 183s -0.431 1.245 -0.643 0.402 183s test-mul-mat0: iter:126/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.906 0.888 183s 0.061 -0.208 183s 0.238 0.671 183s 183s x1: [2, 2, 1, 1] 183s 0.483 0.611 183s 0.662 -0.582 183s 183s y: [3, 2, 1, 1] 183s 0.980 -0.098 0.525 183s 0.083 0.161 -0.233 183s testing: mul_mat, [2, 3, 2, 1] = [2, 2, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.101 -0.329 183s -0.054 0.025 183s 183s x1: [2, 3, 2, 1] 183s -0.728 0.925 183s 0.572 -0.710 183s 0.823 -0.339 183s 183s y: [2, 3, 2, 1] 183s -0.378 0.063 183s 0.291 -0.049 183s 0.195 -0.053 183s testing: mul_mat, [3, 2, 2, 2] = [2, 3, 2, 2] * [2, 2, 2, 2] 183s x0: [2, 3, 2, 2] 183s 0.685 0.698 183s 0.747 -0.727 183s -0.481 0.585 183s 183s x1: [2, 2, 2, 2] 183s 0.242 0.162 183s -0.274 -0.078 183s 183s y: [3, 2, 2, 2] 183s 0.279 0.063 -0.021 183s -0.242 -0.148 0.086 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.420 0.293 183s -0.226 0.882 183s -0.703 -0.773 183s -0.147 -0.517 183s 183s x1: [2, 3, 1, 1] 183s -0.878 0.294 183s 0.189 -0.456 183s -0.105 0.076 183s 183s y: [4, 3, 1, 1] 183s 0.456 0.458 0.390 -0.023 183s -0.213 -0.445 0.219 0.208 183s 0.067 0.091 0.015 -0.024 183s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.650 0.244 0.824 -0.323 183s 0.766 -0.158 0.616 -0.090 183s -0.627 -0.077 0.219 0.101 183s 183s x1: [4, 2, 2, 1] 183s -0.587 -0.373 -0.501 -0.425 183s -0.474 -0.177 0.978 0.588 183s 183s y: [3, 2, 2, 1] 183s 0.016 -0.661 0.244 183s 0.881 0.215 0.584 183s testing: mul_mat, [2, 4, 2, 2] = [3, 2, 2, 2] * [3, 4, 2, 2] 183s x0: [3, 2, 2, 2] 183s -0.855 0.159 -0.340 183s -0.050 -0.826 0.138 183s 183s x1: [3, 4, 2, 2] 183s 0.363 -0.874 -0.016 183s 0.070 0.181 0.561 183s -0.100 0.531 0.326 183s 0.273 -0.225 -0.832 183s 183s y: [2, 4, 2, 2] 183s -0.444 0.702 183s -0.222 -0.076 183s 0.059 -0.389 183s 0.013 0.058 183s test-mul-mat0: iter:127/500 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.999 -0.571 183s 0.678 -0.573 183s -0.740 -0.848 183s 183s x1: [2, 3, 1, 1] 183s 0.070 -0.611 183s -0.868 0.215 183s 0.339 0.291 183s 183s y: [3, 3, 1, 1] 183s 0.419 0.397 0.466 183s -0.990 -0.712 0.460 183s 0.172 0.063 -0.498 183s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.632 0.374 183s 0.507 0.707 183s 183s x1: [2, 3, 3, 1] 183s -0.732 0.373 183s 0.179 -0.115 183s 0.893 0.745 183s 183s y: [2, 3, 3, 1] 183s 0.602 -0.108 183s -0.156 0.010 183s -0.286 0.980 183s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.785 -0.262 183s 183s x1: [2, 2, 3, 1] 183s -0.598 -0.274 183s 0.905 0.287 183s 183s y: [1, 2, 3, 1] 183s 0.542 183s -0.786 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.732 0.974 183s -0.982 0.112 183s -0.110 0.464 183s 183s x1: [2, 1, 1, 1] 183s -0.439 0.353 183s 183s y: [3, 1, 1, 1] 183s 0.665 0.471 0.212 183s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.697 0.432 -0.845 183s -0.922 0.265 -0.500 183s 0.271 -0.474 -0.439 183s -0.742 0.450 0.618 183s 183s x1: [3, 2, 3, 1] 183s -0.892 -0.535 -0.810 183s -0.490 0.192 -0.905 183s 183s y: [4, 2, 3, 1] 183s -0.168 1.086 0.367 -0.080 183s 0.506 0.955 0.173 -0.110 183s testing: mul_mat, [2, 3, 3, 1] = [4, 2, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 2, 3, 1] 183s -0.777 -0.648 -0.854 0.032 183s 0.180 -0.324 0.030 -0.975 183s 183s x1: [4, 3, 3, 1] 183s -ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s 0.128 -0.260 -0.129 0.026 183s 0.240 0.432 -0.355 0.848 183s -0.541 -0.547 -0.034 -0.322 183s 183s y: [2, 3, 3, 1] 183s 0.379 0.032 183s -0.136 -0.934 183s 0.793 0.392 183s test-mul-mat0: iter:128/500 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.013 0.289 0.171 -0.977 183s -0.463 0.082 0.514 0.971 183s 0.919 0.272 0.142 0.940 183s 0.209 -0.614 0.100 0.315 183s 183s x1: [4, 2, 1, 1] 183s 0.056 0.937 -0.950 0.279 183s 0.117 -0.598 0.955 0.263 183s 183s y: [4, 2, 1, 1] 183s -0.166 -0.166 0.434 -0.570 183s -0.268 0.643 0.328 0.570 183s testing: mul_mat, [3, 4, 2, 1] = [4, 3, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.590 0.960 -0.906 0.577 183s 0.804 -0.752 -0.206 -0.072 183s 0.520 -0.587 -0.994 -0.945 183s 183s x1: [4, 4, 2, 1] 183s 0.311 0.613 0.011 -0.211 183s -0.783 0.167 0.844 -0.846 183s 0.217 0.123 0.272 0.619 183s 0.079 -0.465 0.052 -0.934 183s 183s y: [3, 4, 2, 1] 183s 0.273 -0.197 -0.009 183s -0.630 -0.868 -0.546 183s 0.101 -0.019 -0.815 183s -1.078 0.470 1.145 183s testing: mul_mat, [3, 3, 2, 3] = [4, 3, 2, 3] * [4, 3, 2, 3] 183s x0: [4, 3, 2, 3] 183s -0.844 0.991 0.467 -0.045 183s 0.072 0.576 0.528 0.592 183s -0.537 0.947 0.430 0.816 183s 183s x1: [4, 3, 2, 3] 183s 0.380 -0.277 0.905 0.644 183s -0.878 0.076 0.521 -0.469 183s 0.036 0.615 -0.892 -0.160 183s 183s y: [3, 3, 2, 3] 183s -0.202 0.727 0.448 183s 1.081 -0.022 0.385 183s 0.169 -0.209 0.049 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.036 0.279 -0.974 -0.658 183s 183s x1: [4, 3, 1, 1] 183s 0.297 -0.727 -0.674 0.847 183s 0.584 0.743 -0.251 -0.995 183s -0.661 0.700 0.077 0.853 183s 183s y: [1, 3, 1, 1] 183s -0.093 183s 1.128 183s -0.464 183s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.905 183s -0.580 183s 183s x1: [1, 4, 2, 1] 183s 0.396 183s 0.460 183s 0.287 183s -0.124 183s 183s y: [2, 4, 2, 1] 183s 0.358 -0.230 183s 0.416 -0.267 183s 0.260 -0.167 183s -0.112 0.072 183s testing: mul_mat, [1, 1, 2, 3] = [2, 1, 2, 3] * [2, 1, 2, 3] 183s x0: [2, 1, 2, 3] 183s -0.211 0.467 183s 183s x1: [2, 1, 2, 3] 183s 0.898 -0.989 183s 183s y: [1, 1, 2, 3] 183s -0.652 183s test-mul-mat0: iter:129/500 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.234 0.444 0.597 183s 0.197 0.604 -0.614 183s -0.335 0.042 0.201 183s -0.526 0.876 0.476 183s 183s x1: [3, 3, 1, 1] 183s -0.186 -0.704 -0.054 183s -0.176 -0.938 0.170 183s -0.318 0.412 0.138 183s 183s y: [4, 3, 1, 1] 183s -0.301 -0.428 0.022 -0.544 183s -0.274 -0.705 0.054 -0.648 183s 0.339 0.101 0.151 0.593 183s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 3, 2, 1] 183s 0.686 0.709 0.423 183s -0.840 0.585 -0.101 183s 0.730 0.172 0.589 183s 183s x1: [3, 4, 2, 1] 183s 0.570 0.587 -0.311 183s -0.470 -0.515 0.318 183s -0.370 0.288 -0.635 183s 0.443 0.584 0.312 183s 183s y: [3, 4, 2, 1] 183s 0.676 -0.104 0.334 183s -0.554 0.061 -0.245 183s -0.318 0.543 -0.594 183s 0.851 -0.062 0.608 183s testing: mul_mat, [2, 3, 2, 1] = [3, 2, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 2, 2, 1] 183s 0.656 -0.118 0.759 183s 0.916 -0.462 -0.335 183s 183s x1: [3, 3, 2, 1] 183s -0.399 -0.825 0.065 183s 0.551 0.233 -0.315 183s 0.791 0.057 0.749 183s 183s y: [2, 3, 2, 1] 183s -0.115 -0.006 183s 0.095 0.503 183s 1.080 0.447 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.524 -0.566 0.828 183s -0.981 -0.443 -0.753 183s -0.291 -0.679 -0.731 183s -0.444 -0.441 0.998 183s 183s x1: [3, 2, 1, 1] 183s 0.740 -0.814 0.585 183s 0.291 0.419 -0.730 183s 183s y: [4, 2, 1, 1] 183s 0.557 -0.806 -0.090 0.614 183s -0.994 0.078 0.164 -1.043 183s testing: mul_mat, [3, 3, 2, 1] = [4, 3, 2, 1] * [4, 3, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.859 -0.613 -0.710 -0.891 183s 0.745 -0.985 0.166 -0.144 183s -0.029 -0.031 -0.274 0.148 183s 183s x1: [4, 3, 2, 1] 183s 0.903 0.151 0.757 0.819 183s 0.689 -0.578 -0.375 0.172 183s -0.642 0.764 0.183 -0.122 183s 183s y: [3, 3, 2, 1] 183s -2.136 0.531 -0.117 183s -0.125 0.996 0.126 183s 0.063 -1.183 -0.073 183s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 4, 2, 1] 183s -0.393 -0.520 -0.135 183s 0.114 -0.466 0.337 183s -0.046 -0.922 0.559 183s 0.404 0.110 -0.618 183s 183s x1: [3, 4, 2, 1] 183s -0.326 -0.370 -0.669 183s 0.260 -0.551 0.517 183s -0.598 -0.410 0.262 183s 0.373 -0.023 0.277 183s 183s y: [4, 4, 2, 1] 183s 0.411 -0.090 -0.018 0.241 183s 0.115 0.461 0.785 -0.275 183s 0.412 0.212 0.552 -0.449 183s -0.172 0.147 0.159 -0.023 183s test-mul-mat0: iter:130/500 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.229 -0.074 0.910 183s -0.212 0.308 0.650 183s 183s x1: [3, 3, 1, 1] 183s -0.516 -0.475 -0.427 183s 0.091 0.639 0.527 183s -0.505 -0.881 -0.939 183s 183s y: [2, 3, 1, 1] 183s -0.471 -0.314 183s 0.454 0.520 183s -0.905 -0.774 183s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 4, 4, 1] 183s 0.402 0.040 -0.347 183s -0.810 -0.652 -0.697 183s 0.126 0.301 0.066 183s 0.189 0.174 0.284 183s 183s x1: [3, 2, 4, 1] 183s -0.063 -0.048 -0.237 183s -0.937 0.874 -0.781 183s 183s y: [4, 2, 4, 1] 183s 0.055 0.248 -0.038 -0.088 183s -0.071 0.735 0.093 -0.247 183s testing: mul_mat, [4, 4, 4, 2] = [3, 4, 4, 2] * [3, 4, 4, 2] 183s x0: [3, 4, 4, 2] 183s -0.205 0.704 -0.579 183s -0.496 0.865 0.213 183s -0.569 -0.653 -0.391 183s 0.139 -0.507 0.461 183s 183s x1: [3, 4, 4, 2] 183s 0.992 -0.765 -0.046 183s 0.459 -0.212 0.573 183s 0.451 0.018 -0.753 183s 0.496 0.418 0.093 183s 183s y: [4, 4, 4, 2] 183s -0.715 -1.164 -0.047 0.504 183s -0.574 -0.288 -0.347 0.435 183s 0.356 -0.369 0.027 -0.294 183s 0.139 0.135 -0.591 -0.100 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.980 0.723 0.856 183s 183s x1: [3, 4, 1, 1] 183s 0.847 0.154 0.091 183s 0.702 0.996 0.141 183s 0.403 0.306 -0.487 183s 0.933 0.163 0.377 183s 183s y: [1, 4, 1, 1] 183s -0.641 183s 0.153 183s -0.591 183s -0.473 183s testing: mul_mat, [1, 3, 4, 1] = [1, 1, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 1, 4, 1] 183s -0.219 183s 183s x1: [1, 3, 4, 1] 183s 0.582 183s 0.736 183s 0.388 183s 183s y: [1, 3, 4, 1] 183s -0.128 183s -0.161 183s -0.085 183s testing: mul_mat, [1, 1, 4, 2] = [1, 1, 4, 2] * [1, 1, 4, 2] 183s x0: [1, 1, 4, 2] 183s 0.213 183s 183s x1: [1, 1, 4, 2] 183s -0.818 183s 183s y: [1, 1, 4, 2] 183s -0.174 183s test-mul-mat0: iter:131/500 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.139 183s 0.240 183s 0.279 183s 0.952 183s 183s x1: [1, 3, 1, 1] 183s -0.996 183s -0.330 183s 0.813 183s 183s y: [4, 3, 1, 1] 183s 0.139 -0.239 -0.278 -0.948 183s 0.046 -0.079 -0.092 -0.314 183s -0.113 0.195 0.227 0.774 183s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.819 183s 183s x1: [1, 4, 2, 1] 183s -0.486 183s 0.214 183s -0.866 183s 0.490 183s 183s y: [1, 4, 2, 1] 183s 0.399 183s -0.175 183s 0.710 183s -0.401 183s testing: mul_mat, [4, 1, 2, 3] = [1, 4, 2, 3] * [1, 1, 2, 3] 183s x0: [1, 4, 2, 3] 183s 0.377 183s 0.387 183s -0.331 183s 0.820 183s 183s x1: [1, 1, 2, 3] 183s -0.494 183s 183s y: [4, 1, 2, 3] 183s -0.186 -0.191 0.164 -0.405 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.620 183s -0.753 183s 183s x1: [1, 4, 1, 1] 183s -0.309 183s 0.791 183s 0.175 183s -0.867 183s 183s y: [2, 4, 1, 1] 183s 0.192 0.233 183s -0.491 -0.596 183s -0.109 -0.132 183s 0.538 0.653 183s testing: mul_mat, [2, 1, 2, 1] = [2, 2, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.588 -0.476 183s 0.578 -0.241 183s 183s x1: [2, 1, 2, 1] 183s 0.268 0.766 183s 183s y: [2, 1, 2, 1] 183s -0.207 -0.030 183s testing: mul_mat, [1, 2, 2, 3] = [2, 1, 2, 3] * [2, 2, 2, 3] 183s x0: [2, 1, 2, 3] 183s -0.072 0.077 183s 183s x1: [2, 2, 2, 3] 183s -0.913 -0.128 183s 0.667 0.616 183s 183s y: [1, 2, 2, 3] 183s 0.056 183s -0.000 183s test-mul-mat0: iter:132/500 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.060 183s -0.024 183s 0.158 183s 183s x1: [1, 4, 1, 1] 183s -0.443 183s 0.774 183s -0.333 183s -0.423 183s 183s y: [3, 4, 1, 1] 183s 0.027 0.010 -0.070 183s -0.046 -0.018 0.123 183s 0.020 0.008 -0.053 183s 0.025 0.010 -0.067 183s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 2, 3, 1] 183s 0.452 183s 0.571 183s 183s x1: [1, 3, 3, 1] 183s -0.518 183s 0.142 183s 0.889 183s 183s y: [2, 3, 3, 1] 183s -0.234 -0.296 183s 0.064 0.081 183s 0.402 0.508 183s testing: mul_mat, [1, 2, 3, 4] = [1, 1, 3, 4] * [1, 2, 3, 4] 183s x0: [1, 1, 3, 4] 183s -0.992 183s 183s x1: [1, 2, 3, 4] 183s 0.197 183s 0.569 183s 183s y: [1, 2, 3, 4] 183s -0.195 183s -0.565 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.371 183s 0.629 183s 183s x1: [1, 1, 1, 1] 183s 0.787 183s 183s y: [2, 1, 1, 1] 183s -0.292 0.495 183s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.449 0.856 183s -0.951 0.711 183s 183s x1: [2, 1, 3, 1] 183s 0.850 -0.466 183s 183s y: [2, 1, 3, 1] 183s -0.781 -1.140 183s testing: mul_mat, [4, 2, 3, 4] = [2, 4, 3, 4] * [2, 2, 3, 4] 183s x0: [2, 4, 3, 4] 183s 0.250 -0.915 183s 0.027 0.748 183s 0.178 0.640 183s -0.761 0.300 183s 183s x1: [2, 2, 3, 4] 183s -0.080 -0.309 183s -0.334 0.236 183s 183s y: [4, 2, 3, 4] 183s 0.262 -0.233 -0.212 -0.032 183s -0.299 0.167 0.091 0.325 183s test-mul-mat0: iter:133/500 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.439 -0.610 -0.220 183s 183s x1: [3, 4, 1, 1] 183s -0.390 0.736 0.010 183s -0.487 0.525 -0.823 183s -0.676 0.065 -0.745 183s -0.415 -0.108 0.612 183s 183s y: [1, 4, 1, 1] 183s -0.280 183s 0.074 183s 0.420 183s 0.113 183s testing: mul_mat, [3, 1, 3, 1] = [3, 3, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.776 0.230 -0.273 183s -0.627 -0.671 0.337 183s -0.890 0.339 0.851 183s 183s x1: [3, 1, 3, 1] 183s 0.316 0.199 -0.401 183s 183s y: [3, 1, 3, 1] 183s -0.090 -0.467 -0.554 183s testing: mul_mat, [2, 3, 3, 4] = [3, 2, 3, 4] * [3, 3, 3, 4] 183s x0: [3, 2, 3, 4] 183s -0.722 -0.961 -0.617 183s 0.828 0.335 -0.483 183s 183s x1: [3, 3, 3, 4] 183s 0.022 0.189 0.485 183s -0.866 0.413 -0.284 183s -0.139 0.786 0.045 183s 183s y: [2, 3, 3, 4] 183s -0.497 -0.153 183s 0.405 -0.442 183s -0.682 0.127 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.259 -0.616 0.223 183s 183s x1: [3, 2, 1, 1] 183s 0.396 0.936 0.558 183s 0.494 0.306 -0.448 183s 183s y: [1, 2, 1, 1] 183s -0.350 183s -0.160 183s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 3, 3, 1] 183s -0.592 183s -0.373 183s 0.514 183s 183s x1: [1, 3, 3, 1] 183s 0.365 183s -0.748 183s 0.473 183s 183s y: [3, 3, 3, 1] 183s -0.216 -0.136 0.188 183s 0.442 0.279 -0.384 183s -0.280 -0.177 0.243 183s testing: mul_mat, [2, 1, 3, 4] = [3, 2, 3, 4] * [3, 1, 3, 4] 183s x0: [3, 2, 3, 4] 183s -0.564 -0.962 -0.848 183s 0.802 -0.946 0.002 183s 183s x1: [3, 1, 3, 4] 183s 0.117 0.856 0.985 183s 183s y: [2, 1, 3, 4] 183s -1.724 -0.713 183s test-mul-mat0: iter:134/500 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.743 183s 183s x1: [1, 1, 1, 1] 183s -0.467 183s 183s y: [1, 1, 1, 1] 183s -0.347 183s testing: mul_mat, [4, 1, 4, 1] = [1, 4, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 4, 4, 1] 183s -0.097 183s -0.539 183s 0.561 183s -0.409 183s 183s x1: [1, 1, 4, 1] 183s -0.242 183s 183s y: [4, 1, 4, 1] 183s 0.024 0.131 -0.136 0.099 183s testing: mul_mat, [4, 4, 4, 3] = [1, 4, 4, 3] * [1, 4, 4, 3] 183s x0: [1, 4, 4, 3] 183s -0.479 183s -0.317 183s 0.554 183s -0.142 183s 183s x1: [1, 4, 4, 3] 183s -0.956 183s 0.598 183s 0.932 183s -0.303 183s 183s y: [4, 4, 4, 3] 183s 0.458 0.303 -0.529 0.136 183s -0.286 -0.189 0.331 -0.085 183s -0.446 -0.295 0.516 -0.132 183s 0.145 0.096 -0.168 0.043 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.041 183s 0.437 183s -0.249 183s 183s x1: [1, 4, 1, 1] 183s -0.685 183s 0.644 183s -0.910 183s -0.866 183s 183s y: [3, 4, 1, 1] 183s -0.028 -0.299 0.171 183s 0.026 0.281 -0.161 183s -0.037 -0.397 0.227 183s -0.035 -0.378 0.216 183s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.572 0.711 0.829 183s 0.431 0.698 0.773 183s -0.917 -0.514 0.228 183s 183s x1: [3, 1, 4, 1] 183s -0.366 -0.993 0.574 183s 183s y: [3, 1, 4, 1] 183s -0.020 -0.407 0.977 183s testing: mul_mat, [3, 3, 4, 3] = [3, 3, 4, 3] * [3, 3, 4, 3] 183s x0: [3, 3, 4, 3] 183s -0.934 -0.293 0.499 183s -0.652 -0.985 0.987 183s -0.029 -0.985 0.226 183s 183s x1: [3, 3, 4, 3] 183s 0.318 0.166 -0.610 183s -0.454 -0.961 -0.523 183s 0.419 0.001 -0.301 183s 183s y: [3, 3, 4, 3] 183s -0.651 -0.973 -0.311 183s 0.445 0.727 0.842 183s -0.542 -0.572 -0.081 183s test-mul-mat0: iter:135/500 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.267 0.658 183s 0.564 0.346 183s 183s x1: [2, 1, 1, 1] 183s 0.819 -0.818 183s 183s y: [2, 1, 1, 1] 183s -0.319 0.179 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.737 -0.136 183s 0.022 0.341 183s -0.516 -0.764 183s -0.039 -0.966 183s 183s x1: [2, 2, 1, 1] 183s 0.846 0.475 183s 0.756 0.786 183s 183s y: [4, 2, 1, 1] 183s -0.688 0.181 -0.800 -0.492 183s -0.664 0.285 -0.991 -0.789 183s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 183s x0: [2, 4, 1, 2] 183s 0.561 -0.865 183s -0.487 0.317 183s 0.921 -0.786 183s 0.580 -0.215 183s 183s x1: [2, 4, 1, 2] 183s 0.363 0.393 183s -0.629 0.033 183s -0.313 0.352 183s -0.335 -0.140 183s 183s y: [4, 4, 1, 2] 183s -0.136 -0.052 0.025 0.126 183s -0.382 0.317 -0.606 -0.372 183s -0.480 0.264 -0.565 -0.257 183s -0.067 0.119 -0.198 -0.164 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.102 0.610 183s 0.611 0.316 183s 183s x1: [2, 4, 1, 1] 183s 0.674 0.868 183s 0.962 0.026 183s -0.467 -0.178 183s -0.846 -0.019 183s 183s y: [2, 4, 1, 1] 183s 0.598 0.687 183s 0.114 0.596 183s -0.156 -0.342 183s -0.097 -0.523 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.806 -0.259 183s 0.956 -0.389 183s 0.375 -0.306 183s 0.773 -0.989 183s 183s x1: [2, 2, 1, 1] 183s 0.900 -0.352 183s -0.123 -0.965 183s 183s y: [4, 2, 1, 1] 183s -0.635 0.997 0.445 1.044 183s 0.349 0.258 0.249 0.860 183s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 183s x0: [4, 4, 1, 2] 183s -0.414 0.844 0.566 -0.876 183s 0.809 -0.809 0.430 -0.448 183s 0.030 0.657 -0.459 -0.741 183s 0.463 0.800 -0.823 -0.299 183s 183s x1: [4, 2, 1, 2] 183s -0.917 -0.424 0.314 -0.083 183s 0.851 -0.012 -0.216 0.812 183s 183s y: [4, 2, 1, 2] 183s 0.272 -0.227 -0.388 -0.997 183s -1.196 0.241 -0.485 0.319 183s test-mul-mat0: iter:136/500 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.479 183s 183s x1: [1, 4, 1, 1] 183s 0.372 183s -0.988 183s -0.062 183s -0.062 183s 183s y: [1, 4, 1, 1] 183s 0.178 183s -0.474 183s -0.030 183s -0.030 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.969 183s 0.374 183s -0.733 183s 0.897 183s 183s x1: [1, 1, 1, 1] 183s 0.115 183s 183s y: [4, 1, 1, 1] 183s -0.111 0.043 -0.084 0.103 183s testing: mul_mat, [4, 4, 1, 4] = [1, 4, 1, 4] * [1, 4, 1, 4] 183s x0: [1, 4, 1, 4] 183s -0.793 183s 0.279 183s -0.345 183s -0.781 183s 183s x1: [1, 4, 1, 4] 183s -0.450 183s -0.181 183s -0.250 183s -0.490 183s 183s y: [4, 4, 1, 4] 183s 0.357 -0.126 0.155 0.352 183s 0.144 -0.051 0.063 0.142 183s 0.198 -0.070 0.086 0.195 183s 0.388 -0.137 0.169 0.383 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.676 183s -0.354 183s -0.709 183s 183s x1: [1, 4, 1, 1] 183s -0.626 183s 0.141 183s 0.496 183s 0.992 183s 183s y: [3, 4, 1, 1] 183s -0.423 0.221 0.444 183s 0.096 -0.050 -0.100 183s 0.335 -0.175 -0.351 183s 0.671 -0.351 -0.703 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.458 0.279 0.010 183s 183s x1: [3, 1, 1, 1] 183s 0.627 0.606 0.138 183s 183s y: [1, 1, 1, 1] 183s 0.458 183s testing: mul_mat, [2, 3, 1, 4] = [1, 2, 1, 4] * [1, 3, 1, 4] 183s x0: [1, 2, 1, 4] 183s 0.823 183s -0.322 183s 183s x1: [1, 3, 1, 4] 183s 0.665 183s -0.441 183s 0.665 183s 183s y: [2, 3, 1, 4] 183s 0.548 -0.214 183s -0.364 0.142 183s 0.547 -0.214 183s test-mul-mat0: iter:137/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.985 0.783 0.451 183s -0.209 0.040 -0.004 183s 183s x1: [3, 2, 1, 1] 183s -0.557 -0.823 0.108 183s 0.901 0.457 -0.882 183s 183s y: [2, 2, 1, 1] 183s -0.046 0.083 183s -0.927 -0.167 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.730 -0.381 0.483 183s 0.549 -0.304 0.673 183s 183s x1: [3, 2, 1, 1] 183s 0.263 0.005 0.770 183s 0.462 0.052 -0.962 183s 183s y: [2, 2, 1, 1] 183s 0.562 0.661 183s -0.148 -0.410 183s testing: mul_mat, [4, 2, 1, 3] = [3, 4, 1, 3] * [3, 2, 1, 3] 183s x0: [3, 4, 1, 3] 183s -0.117 -0.364 -0.499 183s 0.935 -0.327 0.787 183s 0.665 0.292 0.270 183s 0.213 0.988 -0.057 183s 183s x1: [3, 2, 1, 3] 183s -0.703 -0.551 -0.491 183s 0.336 0.829 -0.048 183s 183s y: [4, 2, 1, 3] 183s 0.527 -0.864 -0.761 -0.667 183s -0.317 0.005 0.452 0.894 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.385 -0.287 0.027 183s -0.090 0.168 -0.736 183s -0.255 -0.579 0.968 183s 183s x1: [3, 4, 1, 1] 183s 0.417 0.457 0.014 183s -0.313 -0.329 0.002 183s 0.630 -0.819 0.439 183s 0.083 0.026 0.706 183s 183s y: [3, 4, 1, 1] 183s 0.029 0.029 -0.357 183s -0.026 -0.028 0.272 183s 0.489 -0.518 0.739 183s 0.043 -0.523 0.647 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.654 0.166 -0.309 183s -0.836 0.608 -0.777 183s 0.978 0.197 -0.689 183s 0.016 -0.395 -0.822 183s 183s x1: [3, 3, 1, 1] 183s 0.116 0.238 -0.620 183s -0.755 0.096 0.428 183s 0.875 -0.112 0.150 183s 183s y: [4, 3, 1, 1] 183s 0.155 0.529 0.588 0.417 183s 0.378 0.357 -1.015 -0.402 183s -0.637 -0.916 0.730 -0.065 183s testing: mul_mat, [1, 3, 1, 3] = [4, 1, 1, 3] * [4, 3, 1, 3] 183s x0: [4, 1, 1, 3] 183s 0.933 -0.222 0.908 -0.603 183s 183s x1: [4, 3, 1, 3] 183s 0.608 0.221 -0.077 -0.679 183s -0.611 0.344 0.347 -0.347 183s 0.312 -0.536 0.891 0.693 183s 183s y: [1, 3, 1, 3] 183s 0.858 183s -0.122 183s 0.802 183s test-mul-mat0: iter:138/500 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.654 0.485 183s -0.397 -0.068 183s 0.610 0.199 183s -0.135 -0.612 183s 183s x1: [2, 4, 1, 1] 183s -0.010 0.027 183s -0.127 0.424 183s 0.232 0.024 183s -0.919 0.748 183s 183s y: [4, 4, 1, 1] 183s 0.020 0.002 -0.001 -0.015 183s 0.288 0.021 0.007 -0.242 183s -0.140 -0.094 0.146 -0.046 183s 0.964 0.314 -0.412 -0.334 183s testing: mul_mat, [1, 4, 3, 1] = [2, 1, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.117 0.770 183s 183s x1: [2, 4, 3, 1] 183s 0.107 0.262 183s 0.057 -0.292 183s -0.633 -0.287 183s 0.429 -0.945 183s 183s y: [1, 4, 3, 1] 183s 0.189 183s -0.232 183s -0.147 183s -0.777 183s testing: mul_mat, [1, 1, 3, 4] = [2, 1, 3, 4] * [2, 1, 3, 4] 183s x0: [2, 1, 3, 4] 183s -0.036 0.850 183s 183s x1: [2, 1, 3, 4] 183s 0.189 0.897 183s 183s y: [1, 1, 3, 4] 183s 0.755 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.710 0.589 183s -0.3ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png75 -0.460 183s -0.310 -0.205 183s -0.933 -0.365 183s 183s x1: [2, 1, 1, 1] 183s 0.854 -0.456 183s 183s y: [4, 1, 1, 1] 183s -0.875 -0.111 -0.171 -0.631 183s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 3, 3, 1] 183s 0.390 0.930 -0.045 -0.638 183s 0.954 -0.252 0.610 -0.066 183s 0.294 0.929 -0.200 0.803 183s 183s x1: [4, 2, 3, 1] 183s 0.025 -0.139 -0.129 0.407 183s 0.324 0.003 -0.947 -0.103 183s 183s y: [3, 2, 3, 1] 183s -0.373 -0.047 0.231 183s 0.238 -0.262 0.205 183s testing: mul_mat, [2, 4, 3, 4] = [3, 2, 3, 4] * [3, 4, 3, 4] 183s x0: [3, 2, 3, 4] 183s 0.455 0.053 -0.939 183s 0.191 -0.327 -0.410 183s 183s x1: [3, 4, 3, 4] 183s 0.690 0.540 0.939 183s -0.510 0.902 -0.127 183s -0.707 -0.684 -0.296 183s -0.443 0.749 -0.642 183s 183s y: [2, 4, 3, 4] 183s -0.538 -0.430 183s -0.065 -0.340 183s -0.080 0.210 183s 0.440 -0.066 183s test-mul-mat0: iter:139/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.928 -0.217 -0.205 -0.499 183s -0.431 -0.544 0.294 0.582 183s 0.977 -0.104 -0.717 -0.645 183s 183s x1: [4, 3, 1, 1] 183s -0.974 0.790 0.851 0.838 183s 0.419 0.073 0.868 -0.707 183s 0.412 0.676 -0.480 0.549 183s 183s y: [3, 3, 1, 1] 183s -1.668 0.728 -2.184 183s 0.547 -0.376 0.235 183s 0.060 -0.367 0.323 183s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 1, 4, 1] 183s -0.921 0.304 -0.546 -0.621 183s 183s x1: [4, 3, 4, 1] 183s 0.129 0.314 0.429 0.165 183s 0.950 -0.288 0.191 0.740 183s -0.437 0.029 0.160 0.636 183s 183s y: [1, 3, 4, 1] 183s -0.360 183s -1.525 183s -0.072 183s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.927 -0.179 -0.410 -0.670 183s 0.955 0.992 -0.736 0.822 183s 183s x1: [4, 1, 4, 1] 183s -0.397 0.953 -0.300 -0.320 183s 183s y: [2, 1, 4, 1] 183s -0.201 0.524 183s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.260 -0.013 0.934 0.373 183s 0.272 0.578 -0.435 0.337 183s -0.349 0.492 0.103 0.462 183s 183s x1: [4, 2, 1, 1] 183s -0.813 0.590 0.487 -0.858 183s 0.582 0.751 0.964 0.569 183s 183s y: [3, 2, 1, 1] 183s 0.339 -0.381 0.228 183s 0.952 0.365 0.529 183s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 2, 4, 1] 183s 0.705 0.794 -0.697 183s -0.903 -0.762 -0.500 183s 183s x1: [3, 4, 4, 1] 183s -0.267 0.897 0.998 183s -0.509 -0.180 -0.885 183s 0.722 -0.814 0.115 183s -0.121 -0.627 -0.295 183s 183s y: [2, 4, 4, 1] 183s -0.171 -0.941 183s 0.116 1.039 183s -0.218 -0.089 183s -0.378 0.734 183s testing: mul_mat, [1, 3, 4, 1] = [2, 1, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 1, 4, 1] 183s 0.686 -0.773 183s 183s x1: [2, 3, 4, 1] 183s 0.612 -0.556 183s -0.993 -0.578 183s -0.732 -0.115 183s 183s y: [1, 3, 4, 1] 183s 0.850 183s -0.234 183s -0.413 183s test-mul-mat0: iter:140/500 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.954 -0.582 0.275 183s -0.540 0.245 0.907 183s -0.854 0.472 -0.737 183s -0.079 -0.563 -0.773 183s 183s x1: [3, 4, 1, 1] 183s 0.898 0.650 0.836 183s -0.219 0.861 0.121 183s -0.836 -0.665 -0.152 183s 0.319 -0.800 -0.468 183s 183s y: [4, 4, 1, 1] 183s -1.005 0.434 -1.076 -1.084 183s -0.259 0.439 0.504 -0.561 183s 1.143 0.151 0.512 0.558 183s 0.032 -0.793 -0.305 0.787 183s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.415 0.506 -0.056 183s 0.381 -0.458 -0.047 183s -0.658 0.913 0.810 183s 183s x1: [3, 4, 3, 1] 183s 0.308 0.847 -0.803 183s -0.883 0.524 0.052 183s -0.985 0.174 -0.112 183s -0.204 0.036 -0.991 183s 183s y: [3, 4, 3, 1] 183s 0.346 -0.232 -0.080 183s 0.629 -0.579 1.102 183s 0.503 -0.450 0.717 183s 0.159 -0.047 -0.636 183s testing: mul_mat, [2, 3, 3, 2] = [3, 2, 3, 2] * [3, 3, 3, 2] 183s x0: [3, 2, 3, 2] 183s 0.964 -0.510 0.577 183s -0.009 -0.694 -0.069 183s 183s x1: [3, 3, 3, 2] 183s -0.133 0.568 -0.315 183s -0.078 -0.847 -0.808 183s 0.866 0.534 -0.266 183s 183s y: [2, 3, 3, 2] 183s -0.599 -0.371 183s -0.109 0.645 183s 0.409 -0.360 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.450 0.548 0.219 183s -0.137 -0.763 0.568 183s 183s x1: [3, 2, 1, 1] 183s 0.232 0.417 -0.797 183s 0.084 0.855 -0.202 183s 183s y: [2, 2, 1, 1] 183s -0.050 -0.802 183s 0.386 -0.778 183s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.268 0.425 183s 183s x1: [2, 3, 3, 1] 183s 0.563 -0.506 183s -0.430 -0.343 183s -0.483 -0.198 183s 183s y: [1, 3, 3, 1] 183s -0.366 183s -0.030 183s 0.046 183s testing: mul_mat, [4, 2, 3, 2] = [1, 4, 3, 2] * [1, 2, 3, 2] 183s x0: [1, 4, 3, 2] 183s 0.440 183s -0.222 183s -0.399 183s 0.656 183s 183s x1: [1, 2, 3, 2] 183s 0.137 183s 0.085 183s 183s y: [4, 2, 3, 2] 183s 0.060 -0.030 -0.055 0.090 183s 0.038 -0.019 -0.034 0.056 183s test-mul-mat0 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s : iter:141/500 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.500 -0.637 -0.512 183s -0.970 0.334 0.047 183s -0.217 0.914 -0.257 183s 183s x1: [3, 4, 1, 1] 183s -0.745 0.855 0.947 183s 0.695 -0.367 -0.452 183s 0.350 -0.490 -0.872 183s 0.275 -0.606 -0.011 183s 183s y: [3, 4, 1, 1] 183s -1.402 1.053 0.700 183s 0.812 -0.817 -0.370 183s 0.933 -0.544 -0.300 183s 0.529 -0.470 -0.611 183s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.577 -0.351 0.072 183s -0.802 -0.700 -0.114 183s 183s x1: [3, 3, 3, 1] 183s -0.379 0.651 0.814 183s -0.237 -0.886 0.822 183s 0.704 0.871 0.424 183s 183s y: [2, 3, 3, 1] 183s -0.389 -0.245 183s 0.234 0.716 183s 0.130 -1.222 183s testing: mul_mat, [1, 2, 3, 4] = [3, 1, 3, 4] * [3, 2, 3, 4] 183s x0: [3, 1, 3, 4] 183s -0.994 -0.409 0.414 183s 183s x1: [3, 2, 3, 4] 183s 0.279 0.238 0.470 183s 0.557 -0.499 -0.494 183s 183s y: [1, 2, 3, 4] 183s -0.180 183s -0.554 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.413 -0.696 -0.891 183s 0.256 -0.380 0.442 183s 0.027 -0.629 0.296 183s 0.642 0.090 0.131 183s 183s x1: [3, 1, 1, 1] 183s -0.634 -0.090 0.684 183s 183s y: [4, 1, 1, 1] 183s -0.285 0.174 0.242 -0.325 183s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.506 0.972 0.505 0.852 183s 183s x1: [4, 3, 3, 1] 183s -0.241 -0.925 -0.119 -0.273 183s 0.088 -0.095 0.017 0.900 183s -0.619 0.358 -0.282 -0.552 183s 183s y: [1, 3, 3, 1] 183s -1.069 183s 0.639 183s 0.048 183s testing: mul_mat, [2, 4, 3, 4] = [1, 2, 3, 4] * [1, 4, 3, 4] 183s x0: [1, 2, 3, 4] 183s -0.086 183s 0.544 183s 183s x1: [1, 4, 3, 4] 183s -0.107 183s -0.786 183s -0.590 183s -0.630 183s 183s y: [2, 4, 3, 4] 183s 0.009 -0.058 183s 0.068 -0.428 183s 0.051 -0.321 183s 0.054 -0.343 183s test-mul-mat0: iter:142/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.670 0.946 183s -0.992 -0.500 183s -0.468 0.525 183s 183s x1: [2, 2, 1, 1] 183s 0.259 0.415 183s 0.328 -0.827 183s 183s y: [3, 2, 1, 1] 183s 0.219 -0.464 0.097 183s -1.002 0.088 -0.588 183s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.439 -0.016 183s -0.274 -0.233 183s 183s x1: [2, 3, 3, 1] 183s 0.569 0.161 183s 0.955 0.214 183s -0.975 -0.810 183s 183s y: [2, 3, 3, 1] 183s 0.247 -0.193 183s 0.416 -0.311 183s -0.415 0.456 183s testing: mul_mat, [2, 2, 3, 2] = [2, 2, 3, 2] * [2, 2, 3, 2] 183s x0: [2, 2, 3, 2] 183s -0.242 -0.387 183s 0.496 -0.645 183s 183s x1: [2, 2, 3, 2] 183s 0.166 -0.205 183s 0.174 -0.620 183s 183s y: [2, 2, 3, 2] 183s 0.039 0.215 183s 0.198 0.486 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.713 -0.674 183s -0.101 -0.606 183s -0.202 0.153 183s 183s x1: [2, 2, 1, 1] 183s -0.021 -0.622 183s -0.798 0.096 183s 183s y: [3, 2, 1, 1] 183s 0.405 0.379 -0.091 183s -0.633 0.023 0.176 183s testing: mul_mat, [1, 2, 3, 1] = [3, 1, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 1, 3, 1] 183s -0.369 0.801 -0.216 183s 183s x1: [3, 2, 3, 1] 183s -0.427 0.109 -0.327 183s 0.615 -0.002 -0.426 183s 183s y: [1, 2, 3, 1] 183s 0.316 183s -0.136 183s testing: mul_mat, [4, 3, 3, 2] = [1, 4, 3, 2] * [1, 3, 3, 2] 183s x0: [1, 4, 3, 2] 183s 0.100 183s -0.701 183s -0.540 183s 0.731 183s 183s x1: [1, 3, 3, 2] 183s -0.128 183s -0.810 183s -0.226 183s 183s y: [4, 3, 3, 2] 183s -0.013 0.090 0.069 -0.094 183s -0.081 0.568 0.438 -0.593 183s -0.023 0.159 0.122 -0.166 183s test-mul-mat0: iter:143/500 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.641 0.896 0.945 183s 0.340 -0.643 0.372 183s 0.619 -0.735 0.859 183s 183s x1: [3, 4, 1, 1] 183s 0.455 -0.825 -0.561 183s -0.445 -0.526 -0.101 183s -0.713 -0.427 -0.858 183s -0.149 0.568 -0.791 183s 183s y: [3, 4, 1, 1] 183s -1.561 0.476 0.407 183s -0.282 0.150 0.025 183s -0.736 -0.287 -0.864 183s -0.143 -0.710 -1.189 183s testing: mul_mat, [3, 3, 3, 1] = [3, 3, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 3, 3, 1] 183s 0.452 -0.269 -0.257 183s 0.647 -0.075 0.624 183s -0.033 0.572 0.761 183s 183s x1: [3, 3, 3, 1] 183s -0.804 -0.806 -0.119 183s 0.319 -0.354 -0.863 183s -0.543 0.344 -0.478 183s 183s y: [3, 3, 3, 1] 183s -0.116 -0.535 -0.524 183s 0.461 -0.306 -0.870 183s -0.215 -0.676 -0.149 183s testing: mul_mat, [4, 3, 3, 2] = [3, 4, 3, 2] * [3, 3, 3, 2] 183s x0: [3, 4, 3, 2] 183s -0.257 0.781 -0.765 183s -0.940 -0.684 -0.135 183s -0.197 -0.906 -0.055 183s 0.440 -0.998 0.926 183s 183s x1: [3, 3, 3, 2] 183s 0.244 0.450 0.893 183s -0.871 -0.099 -0.376 183s -0.128 -0.451 0.549 183s 183s y: [4, 3, 3, 2] 183s -0.395 -0.658 -0.504 0.486 183s 0.434 0.937 0.281 -0.633 183s -0.740 0.355 0.404 0.902 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.269 0.837 0.507 183s -0.553 -0.ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png622 0.190 183s -0.744 -0.370 -0.949 183s 0.917 0.627 0.703 183s 183s x1: [3, 4, 1, 1] 183s 0.236 -0.260 0.385 183s -0.724 -0.433 -0.094 183s 0.371 -0.105 0.867 183s -0.947 0.709 -0.765 183s 183s y: [4, 4, 1, 1] 183s 0.042 0.104 -0.445 0.325 183s -0.606 0.652 0.789 -1.002 183s 0.451 0.025 -1.059 0.883 183s -0.050 -0.062 1.169 -0.962 183s testing: mul_mat, [3, 3, 3, 1] = [4, 3, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.812 -0.410 -0.638 -0.813 183s -0.096 -0.444 0.102 0.356 183s 0.506 -0.060 -0.759 0.776 183s 183s x1: [4, 3, 3, 1] 183s 0.911 0.976 0.768 -0.110 183s -0.655 0.143 -0.874 0.085 183s -0.472 -0.598 0.652 0.434 183s 183s y: [3, 3, 3, 1] 183s -1.541 -0.483 -0.267 183s 0.961 -0.059 0.390 183s -0.140 0.532 -0.360 183s testing: mul_mat, [1, 4, 3, 2] = [3, 1, 3, 2] * [3, 4, 3, 2] 183s x0: [3, 1, 3, 2] 183s -0.485 0.236 -0.782 183s 183s x1: [3, 4, 3, 2] 183s 0.425 -0.971 -0.623 183s 0.666 -0.784 0.733 183s 0.442 -0.375 -0.386 183s 0.201 0.511 -0.064 183s 183s y: [1, 4, 3, 2] 183s 0.052 183s -1.081 183s -0.001 183s 0.073 183s test-mul-mat0: iter:144/500 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.555 -0.835 -0.585 -0.370 183s -0.232 -0.915 -0.222 -0.147 183s -0.626 0.031 -0.074 0.059 183s -0.752 0.978 0.584 -0.434 183s 183s x1: [4, 3, 1, 1] 183s 0.931 0.928 -0.159 0.790 183s -0.067 -0.843 -0.491 -0.337 183s 0.298 0.025 0.899 0.516 183s 183s y: [4, 3, 1, 1] 183s -1.491 -1.146 -0.496 -0.228 183s 1.153 0.945 0.033 -0.914 183s -0.903 -0.367 -0.223 0.100 183s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.421 -0.778 -0.650 -0.426 183s -0.670 0.477 -0.480 -0.844 183s 183s x1: [4, 4, 4, 1] 183s -0.573 0.422 -0.503 -0.645 183s -0.737 -0.713 0.288 -0.580 183s -0.204 0.951 0.718 0.821 183s 0.849 0.234 -0.513 -0.706 183s 183s y: [2, 4, 4, 1] 183s 0.514 1.371 183s 0.925 0.505 183s -1.471 -0.448 183s 0.094 0.385 183s testing: mul_mat, [4, 2, 4, 4] = [4, 4, 4, 4] * [4, 2, 4, 4] 183s x0: [4, 4, 4, 4] 183s -0.200 0.394 0.239 -0.144 183s 0.304 -0.730 -0.901 0.207 183s -0.671 -0.871 -0.520 -0.406 183s 0.425 0.315 -0.715 -0.078 183s 183s x1: [4, 2, 4, 4] 183s 0.177 -0.645 -0.907 0.506 183s 0.832 -0.387 0.662 -0.742 183s 183s y: [4, 2, 4, 4] 183s -0.579 1.447 0.709 0.481 183s -0.053 -0.215 -0.264 -0.184 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.408 -0.425 -0.870 -0.159 183s -0.308 0.425 0.383 0.833 183s -0.521 0.941 0.641 -0.434 183s -0.195 0.842 0.430 0.206 183s 183s x1: [4, 4, 1, 1] 183s 0.135 -0.635 -0.031 0.230 183s 0.844 0.909 -0.646 -0.807 183s 0.885 0.005 -0.632 -0.042 183s 0.066 -0.129 0.419 0.575 183s 183s y: [4, 4, 1, 1] 183s 0.316 -0.132 -0.788 -0.527 183s 0.648 -0.793 0.352 0.156 183s 0.915 -0.548 -0.844 -0.450 183s -0.373 0.564 -0.137 0.177 183s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.367 -0.425 0.724 0.893 183s -0.177 0.853 0.036 -0.153 183s 183s x1: [4, 4, 4, 1] 183s -0.199 0.796 0.050 -0.289 183s -0.850 0.243 -0.404 0.155 183s 0.611 0.554 -0.779 -0.518 183s -0.028 0.796 0.771 -0.620 183s 183s y: [2, 4, 4, 1] 183s -0.633 0.760 183s -0.570 0.320 183s -1.039 0.415 183s -0.344 0.806 183s testing: mul_mat, [3, 4, 4, 4] = [2, 3, 4, 4] * [2, 4, 4, 4] 183s x0: [2, 3, 4, 4] 183s -0.260 0.620 183s -0.918 0.848 183s -0.102 -0.214 183s 183s x1: [2, 4, 4, 4] 183s 0.788 -0.227 183s 0.923 0.512 183s 0.809 0.816 183s -0.641 0.787 183s 183s y: [3, 4, 4, 4] 183s -0.346 -0.916 -0.032 183s 0.077 -0.414 -0.204 183s 0.296 -0.051 -0.257 183s 0.655 1.256 -0.103 183s test-mul-mat0: iter:145/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.420 -0.973 -0.403 183s -0.725 -0.230 0.558 183s -0.112 0.576 -0.102 183s 0.169 -0.156 0.924 183s 183s x1: [3, 1, 1, 1] 183s -0.225 -0.896 -0.931 183s 183s y: [4, 1, 1, 1] 183s 1.342 -0.151 -0.396 -0.759 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.710 -0.488 -0.281 183s 0.814 -0.419 0.797 183s -0.606 -0.393 -0.605 183s 183s x1: [3, 4, 1, 1] 183s 0.103 -0.391 0.038 183s -0.411 -0.873 -0.681 183s -0.631 0.411 -0.414 183s 0.263 0.373 -0.723 183s 183s y: [3, 4, 1, 1] 183s 0.253 0.279 0.068 183s 0.326 -0.511 1.004 183s -0.532 -1.017 0.472 183s 0.208 -0.518 0.131 183s testing: mul_mat, [3, 3, 1, 2] = [3, 3, 1, 2] * [3, 3, 1, 2] 183s x0: [3, 3, 1, 2] 183s 0.317 -0.004 0.994 183s 0.027 0.508 -0.287 183s -0.158 -0.912 -0.490 183s 183s x1: [3, 3, 1, 2] 183s -0.331 0.377 0.953 183s 0.557 -0.047 -0.149 183s -0.273 0.797 -0.224 183s 183s y: [3, 3, 1, 2] 183s 0.841 -0.091 -0.758 183s 0.029 0.034 0.027 183s -0.313 0.462 -0.574 183s testing: m 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.pnul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.706 -0.853 0.983 183s 0.151 0.271 -0.108 183s 0.276 -0.950 0.743 183s 0.023 0.654 -0.506 183s 183s x1: [3, 3, 1, 1] 183s -0.811 -0.177 -0.515 183s -0.982 0.230 -0.701 183s -0.563 0.762 -0.568 183s 183s y: [4, 3, 1, 1] 183s -0.928 -0.114 -0.439 0.127 183s -1.578 -0.010 -1.010 0.483 183s -1.606 0.183 -1.301 0.773 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.590 -0.862 0.184 0.058 183s -0.249 0.352 0.364 -0.666 183s 183s x1: [4, 3, 1, 1] 183s -0.873 -0.562 0.399 -0.968 183s -0.489 -0.743 0.494 0.537 183s 0.967 0.683 -0.640 -0.548 183s 183s y: [2, 3, 1, 1] 183s -0.014 0.809 183s 0.473 -0.318 183s -0.167 0.132 183s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 183s x0: [2, 4, 1, 2] 183s -0.636 -0.538 183s -0.990 -0.941 183s -0.642 -0.686 183s -0.276 0.213 183s 183s x1: [2, 4, 1, 2] 183s -0.360 -0.919 183s -0.519 0.910 183s -0.869 -0.865 183s 0.894 0.023 183s 183s y: [4, 4, 1, 2] 183s 0.724 1.221 0.861 -0.097 183s -0.160 -0.342 -0.291 0.337 183s 1.018 1.674 1.151 0.055 183s -0.580 -0.906 -0.589 -0.241 183s test-mul-mat0: iter:146/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.710 -0.631 -0.777 183s -0.034 -0.267 -0.767 183s 183s x1: [3, 2, 1, 1] 183s -0.995 0.959 0.225 183s -0.607 0.108 0.542 183s 183s y: [2, 2, 1, 1] 183s -0.074 -0.395 183s -0.058 -0.423 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.438 -0.695 0.143 183s -0.837 0.036 0.344 183s 183s x1: [3, 2, 1, 1] 183s 0.324 0.457 -0.305 183s 0.384 0.772 0.908 183s 183s y: [2, 2, 1, 1] 183s -0.219 -0.360 183s -0.238 0.020 183s testing: mul_mat, [1, 2, 1, 3] = [3, 1, 1, 3] * [3, 2, 1, 3] 183s x0: [3, 1, 1, 3] 183s -0.079 0.186 -0.065 183s 183s x1: [3, 2, 1, 3] 183s -0.032 -0.752 0.456 183s 0.904 0.596 0.461 183s 183s y: [1, 2, 1, 3] 183s -0.167 183s 0.009 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.316 -0.720 0.101 183s -0.540 0.922 0.776 183s 0.166 0.029 -1.000 183s 183s x1: [3, 1, 1, 1] 183s -0.706 -0.564 0.004 183s 183s y: [3, 1, 1, 1] 183s 0.629 -0.136 -0.138 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.744 0.273 -0.509 183s 0.975 -0.162 0.404 183s 183s x1: [3, 3, 1, 1] 183s 0.464 0.810 0.261 183s -0.805 -0.932 0.488 183s 0.659 -0.442 0.040 183s 183s y: [2, 3, 1, 1] 183s 0.433 0.426 183s -1.102 -0.436 183s 0.349 0.730 183s testing: mul_mat, [4, 3, 1, 3] = [2, 4, 1, 3] * [2, 3, 1, 3] 183s x0: [2, 4, 1, 3] 183s 0.191 -0.054 183s 0.906 -0.365 183s -0.945 0.492 183s 0.748 -0.080 183s 183s x1: [2, 3, 1, 3] 183s -0.763 -0.849 183s 0.699 -0.327 183s 0.155 -0.039 183s 183s y: [4, 3, 1, 3] 183s -0.099 -0.382 0.303 -0.503 183s 0.151 0.752 -0.821 0.549 183s 0.032 0.155 -0.166 0.119 183s test-mul-mat0: iter:147/500 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.682 -0.740 -0.265 183s -0.398 -0.832 0.064 183s -0.986 0.481 0.631 183s 183s x1: [3, 2, 1, 1] 183s -0.696 -0.170 0.570 183s 0.494 -0.264 0.625 183s 183s y: [3, 2, 1, 1] 183s -0.500 0.455 0.965 183s 0.367 0.063 -0.220 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.818 -0.327 0.673 183s -0.770 0.846 0.977 183s 183s x1: [3, 3, 1, 1] 183s 0.576 -0.508 -0.129 183s -0.707 -0.536 0.467 183s -0.376 -0.437 0.701 183s 183s y: [2, 3, 1, 1] 183s 0.550 -1.000 183s -0.089 0.547 183s 0.308 0.604 183s testing: mul_mat, [4, 2, 1, 3] = [3, 4, 1, 3] * [3, 2, 1, 3] 183s x0: [3, 4, 1, 3] 183s 0.906 -0.518 -0.414 183s -0.531 -0.817 0.078 183s -0.712 -0.144 -0.248 183s -0.482 -0.297 -0.271 183s 183s x1: [3, 2, 1, 3] 183s 0.060 0.416 0.471 183s 0.796 0.041 -0.286 183s 183s y: [4, 2, 1, 3] 183s -0.357 -0.335 -0.219 -0.280 183s 0.818 -0.478 -0.502 -0.319 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.305 -0.765 0.257 183s -0.998 0.933 -0.186 183s 0.275 -0.848 0.173 183s 183s x1: [3, 4, 1, 1] 183s -0.683 -0.013 0.558 183s 0.069 0.505 -0.739 183s 0.797 0.082 -0.620 183s 0.998 -0.545 0.540 183s 183s y: [3, 4, 1, 1] 183s -0.055 0.566 -0.080 183s -0.556 0.539 -0.537 183s 0.021 -0.604 0.042 183s 0.861 -1.605 0.831 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.056 -0.440 0.642 183s 0.989 -0.750 -0.370 183s 0.226 0.023 0.021 183s 183s x1: [3, 3, 1, 1] 183s 0.949 0.138 0.210 183s -0.398 0.369 0.729 183s 0.840 -0.932 -0.569 183s 183s y: [3, 3, 1, 1] 183s 0.127 0.757 0.222 183s 0.284 -0.940 -0.066 183s 0.091 1.740 0.156 183s testing: mul_mat, [4, 3, 1, 3] = [3, 4, 1, 3] * [3, 3, 1, 3] 183s x0: [3, 4, 1, 3] 183s -0.247 -0g 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.p.056 0.854 183s -0.227 -0.354 0.890 183s -0.287 -0.747 0.221 183s -0.149 0.116 0.215 183s 183s x1: [3, 3, 1, 3] 183s -0.903 -0.830 0.933 183s 0.402 -0.828 0.209 183s 0.637 -0.895 0.361 183s 183s y: [4, 3, 1, 3] 183s 1.066 1.329 1.085 0.239 183s 0.125 0.388 0.549 -0.111 183s 0.201 0.494 0.565 -0.121 183s test-mul-mat0: iter:148/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.101 -0.663 -0.058 183s 0.299 0.831 -0.985 183s -0.075 -0.203 -0.628 183s -0.955 0.666 -0.827 183s 183s x1: [3, 1, 1, 1] 183s -0.471 -0.040 -0.412 183s 183s y: [4, 1, 1, 1] 183s 0.003 0.232 0.302 0.763 183s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 4, 2, 1] 183s 0.658 -0.881 -0.768 183s 0.704 0.785 -0.596 183s 0.420 -0.786 0.046 183s 0.918 0.547 0.847 183s 183s x1: [3, 4, 2, 1] 183s 0.716 -0.572 -0.359 183s -0.501 0.334 -0.198 183s -0.966 0.013 0.675 183s -0.580 0.635 0.193 183s 183s y: [4, 4, 2, 1] 183s 1.251 0.269 0.734 0.041 183s -0.472 0.028 -0.482 -0.446 183s -1.165 -1.072 -0.385 -0.309 183s -1.090 -0.024 -0.734 -0.021 183s testing: mul_mat, [4, 4, 2, 2] = [3, 4, 2, 2] * [3, 4, 2, 2] 183s x0: [3, 4, 2, 2] 183s 0.459 -0.933 -0.412 183s -0.863 0.164 -0.738 183s 0.530 0.693 0.504 183s -0.795 0.242 -0.349 183s 183s x1: [3, 4, 2, 2] 183s 0.924 0.338 0.929 183s 0.768 -0.929 0.252 183s 0.628 0.730 0.371 183s 0.859 0.433 0.157 183s 183s y: [4, 4, 2, 2] 183s -0.274 -1.428 1.192 -0.977 183s 1.114 -1.002 -0.109 -0.924 183s -0.546 -0.696 1.026 -0.452 183s -0.075 -0.787 0.835 -0.633 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.714 0.484 -0.027 183s -0.471 -0.262 -0.675 183s 183s x1: [3, 4, 1, 1] 183s -0.735 -0.497 0.272 183s -0.760 -0.183 -0.173 183s 0.934 0.952 0.373 183s -0.801 0.278 -0.778 183s 183s y: [2, 4, 1, 1] 183s -0.773 0.292 183s -0.626 0.523 183s 1.118 -0.941 183s -0.416 0.829 183s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.021 0.669 183s -0.158 0.267 183s 0.333 0.748 183s 0.796 0.042 183s 183s x1: [2, 3, 2, 1] 183s -0.977 -0.053 183s -0.142 -0.187 183s 0.127 -0.367 183s 183s y: [4, 3, 2, 1] 183s -0.015 0.140 -0.365 -0.780 183s -0.122 -0.028 -0.187 -0.121 183s -0.248 -0.118 -0.232 0.085 183s testing: mul_mat, [2, 2, 2, 2] = [4, 2, 2, 2] * [4, 2, 2, 2] 183s x0: [4, 2, 2, 2] 183s 0.207 0.614 0.199 -0.912 183s -0.642 0.521 -0.812 -0.338 183s 183s x1: [4, 2, 2, 2] 183s -0.233 0.084 0.501 0.714 183s 0.942 -0.686 -0.159 -0.425 183s 183s y: [2, 2, 2, 2] 183s -0.549 -0.454 183s 0.130 -0.689 183s test-mul-mat0: iter:149/500 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.070 0.738 -0.465 183s -0.493 0.924 -0.520 183s 183s x1: [3, 4, 1, 1] 183s 0.329 -0.061 -0.315 183s 0.216 -0.578 0.286 183s -0.630 -0.606 -0.194 183s 0.633 -0.813 0.040 183s 183s y: [2, 4, 1, 1] 183s 0.124 -0.055 183s -0.544 -0.789 183s -0.401 -0.148 183s -0.574 -1.084 183s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 4, 3, 1] 183s -0.605 -0.610 -0.155 183s -0.417 0.430 0.519 183s 0.652 0.168 -0.946 183s -0.841 0.092 -0.466 183s 183s x1: [3, 2, 3, 1] 183s 0.774 -0.661 -0.698 183s 0.629 -0.539 -0.232 183s 183s y: [4, 2, 3, 1] 183s 0.043 -0.970 1.054 -0.386 183s -0.016 -0.615 0.539 -0.470 183s testing: mul_mat, [2, 4, 3, 2] = [3, 2, 3, 2] * [3, 4, 3, 2] 183s x0: [3, 2, 3, 2] 183s 0.122 -0.171 0.402 183s 0.142 -0.201 0.370 183s 183s x1: [3, 4, 3, 2] 183s 0.061 -0.465 0.681 183s 0.115 -0.306 -0.227 183s 0.649 -0.373 -0.796 183s -0.515 -0.811 -0.904 183s 183s y: [2, 4, 3, 2] 183s 0.361 0.354 183s -0.025 -0.006 183s -0.178 -0.128 183s -0.288 -0.244 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.779 0.779 -0.162 183s 0.421 0.439 0.213 183s 183s x1: [3, 2, 1, 1] 183s 0.577 -0.028 -0.989 183s 0.479 -0.641 -0.863 183s 183s y: [2, 2, 1, 1] 183s 0.588 0.020 183s 0.013 -0.264 183s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.027 0.115 183s 0.858 0.332 183s 183s x1: [2, 3, 3, 1] 183s -0.569 0.878 183s 0.614 -0.621 183s 0.709 0.448 183s 183s y: [2, 3, 3, 1] 183s 0.116 -0.197 183s -0.088 0.321 183s 0.032 0.757 183s testing: mul_mat, [1, 2, 3, 2] = [2, 1, 3, 2] * [2, 2, 3, 2] 183s x0: [2, 1, 3, 2] 183s 0.545 -0.881 183s 183s x1: [2, 2, 3, 2] 183s -0.019 -0.868 183s 0.686 0.360 183s 183s y: [1, 2, 3, 2] 183s 0.755 183s 0.057 183s test-mul-mat0: iter:150/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.193 -0.441 183s 0.620 -0.250 183s 0.661 -0.671 183s 183s x1: [2, 2, 1, 1] 183s -0.699 0.765 183s 0.271 -0.570 183s 183s y: [3, 2, 1, 1] 183s -0.472 -0.625 -0.976 183s 0.304 0.311 0.562 183s testing: mul_mat, [2, 3, 1, 1] ng 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.737 0.524 183s -0.537 -0.186 183s 183s x1: [2, 3, 1, 1] 183s 0.422 -0.905 183s 0.032 0.173 183s 0.641 0.151 183s 183s y: [2, 3, 1, 1] 183s -0.163 -0.058 183s 0.114 -0.049 183s 0.551 -0.372 183s testing: mul_mat, [2, 2, 1, 2] = [2, 2, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 2, 1, 2] 183s 0.136 0.553 183s 0.330 0.565 183s 183s x1: [2, 2, 1, 2] 183s -0.972 -0.171 183s 0.251 -0.627 183s 183s y: [2, 2, 1, 2] 183s -0.226 -0.418 183s -0.312 -0.271 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.303 0.651 183s 0.868 0.258 183s 0.522 0.977 183s -0.546 -0.886 183s 183s x1: [2, 2, 1, 1] 183s 0.273 0.785 183s -0.729 -0.305 183s 183s y: [4, 2, 1, 1] 183s 0.594 0.440 0.910 -0.845 183s -0.419 -0.711 -0.678 0.668 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.968 -0.388 0.914 0.885 183s -0.651 0.362 -0.066 -0.962 183s 183s x1: [4, 2, 1, 1] 183s -0.928 -0.995 -0.057 0.323 183s -0.622 0.272 0.973 0.213 183s 183s y: [2, 2, 1, 1] 183s 1.518 -0.063 183s 1.575 0.234 183s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 183s x0: [2, 4, 1, 2] 183s 0.879 -0.521 183s 0.266 -0.240 183s 0.701 0.884 183s 0.852 -0.171 183s 183s x1: [2, 4, 1, 2] 183s -0.123 0.158 183s 0.824 0.148 183s 0.853 0.704 183s -0.549 0.721 183s 183s y: [4, 4, 1, 2] 183s -0.191 -0.071 0.053 -0.132 183s 0.647 0.184 0.708 0.676 183s 0.383 0.058 1.221 0.606 183s -0.858 -0.319 0.253 -0.591 183s test-mul-mat0: iter:151/500 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.236 183s 183s x1: [1, 3, 1, 1] 183s -0.218 183s 0.864 183s 0.535 183s 183s y: [1, 3, 1, 1] 183s -0.051 183s 0.204 183s 0.126 183s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 1, 4, 1] 183s 0.151 183s 183s x1: [1, 1, 4, 1] 183s -0.981 183s 183s y: [1, 1, 4, 1] 183s -0.148 183s testing: mul_mat, [3, 1, 4, 1] = [1, 3, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 3, 4, 1] 183s -0.016 183s 0.437 183s -0.226 183s 183s x1: [1, 1, 4, 1] 183s -0.694 183s 183s y: [3, 1, 4, 1] 183s 0.011 -0.303 0.157 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.900 183s -0.978 183s 0.843 183s -0.178 183s 183s x1: [1, 3, 1, 1] 183s 0.586 183s -0.136 183s -0.997 183s 183s y: [4, 3, 1, 1] 183s -0.527 -0.573 0.494 -0.104 183s 0.123 0.133 -0.115 0.024 183s 0.897 0.974 -0.840 0.178 183s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.182 -0.589 0.748 0.986 183s 183s x1: [4, 1, 4, 1] 183s -0.403 0.301 0.973 0.052 183s 183s y: [1, 1, 4, 1] 183s 0.528 183s testing: mul_mat, [1, 4, 4, 1] = [1, 1, 4, 1] * [1, 4, 4, 1] 183s x0: [1, 1, 4, 1] 183s 0.870 183s 183s x1: [1, 4, 4, 1] 183s 0.169 183s 0.892 183s -0.987 183s 0.994 183s 183s y: [1, 4, 4, 1] 183s 0.147 183s 0.776 183s -0.858 183s 0.864 183s test-mul-mat0: iter:152/500 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.469 -0.924 -0.401 183s 0.330 -0.424 -0.552 183s 0.114 0.109 0.007 183s 183s x1: [3, 3, 1, 1] 183s 0.810 -0.450 -0.168 183s 0.007 0.389 -0.179 183s -0.824 0.281 -0.165 183s 183s y: [3, 3, 1, 1] 183s 0.104 0.551 0.042 183s -0.291 -0.065 0.042 183s 0.193 -0.300 -0.065 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.495 0.918 -0.393 183s -0.956 -0.250 0.613 183s 0.434 0.572 0.789 183s 183s x1: [3, 3, 1, 1] 183s -0.548 0.231 -0.252 183s -0.799 0.101 -0.525 183s 0.361 0.188 0.774 183s 183s y: [3, 3, 1, 1] 183s 0.040 0.312 -0.304 183s -0.096 0.417 -0.703 183s 0.046 0.082 0.874 183s testing: mul_mat, [3, 3, 1, 3] = [3, 3, 1, 3] * [3, 3, 1, 3] 183s x0: [3, 3, 1, 3] 183s -0.342 0.250 -0.491 183s 0.373 0.656 -0.532 183s -0.381 -0.861 0.026 183s 183s x1: [3, 3, 1, 3] 183s -0.285 -0.594 0.959 183s 0.246 -0.518 -0.442 183s -0.424 0.058 0.006 183s 183s y: [3, 3, 1, 3] 183s -0.521 -1.006 0.645 183s 0.004 -0.013 0.341 183s 0.157 -0.124 0.112 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.999 -0.294 0.401 183s 183s x1: [3, 3, 1, 1] 183s -0.321 0.349 0.661 183s 0.399 -0.993 -0.089 183s 0.908 0.380 -0.432 183s 183s y: [1, 3, 1, 1] 183s -0.158 183s 0.654 183s 0.622 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.266 183s 0.479 183s 0.102 183s 183s x1: [1, 3, 1, 1] 183s 0.193 183s 0.903 183s 0.433 183s 183s y: [3, 3, 1, 1] 183s 0.051 0.092 0.020 183s 0.240 0.432 0.092 183s 0.115 0.207 0.044 183s testing: mul_mat, [1, 1, 1, 3] = [3, 1, 1, 3] * [3, 1, 1, 3] 183s x0: [3, 1, 1, 3] 183s 0.367 0.793 0.704 183s 183s x1: [3, 1, 1, 3] 183s -0.973 -0.714 -0.530 183s 183s y: [1, 1, 1, 3] 183s -1.297 183s test-mul-mat0: iter:153/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.475 0.925 0.788 -0.411 183s -0.720 0.156 -0.618 0.984 183s -0.816 0.574 0.695 -0.876 183s 183s x1: [4, 3, 1, 1] 183s -0.123 -0.717 0.117 -0.690 183s -0.640 -0.617 0.789 0.462 183s -0.591 -0.925 0.933 0.695 183s 183s y: [3, 3, 1, 1] 183s -0.346 -0.774 0.374 183s -0.443 0.332 0.311 183s -0.687 0.389 -0.009 183s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.054 0.969 -0.780 -0.553 183s 183s x1: [4, 3, 4, 1] 183s -0.326 -0.213 -0.368 -0.850 183s 0.878 0.846 0.027 -0.839 183s -0.038 0.337 -0.479 0.345 183s 183s y: [1, 3, 4, 1] 183s 0.533 183s 1.310 183s 0.508 183s testing: mul_mat, [2, 1, 4, 2] = [4, 2, 4, 2] * [4, 1, 4, 2] 183s x0: [4, 2, 4, 2] 183s -0.473 0.394 0.102 -0.764 183s 0.919 -0.095 0.589 -0.306 183s 183s x1: [4, 1, 4, 2] 183s 0.616 -0.036 0.412 0.402 183s 183s y: [2, 1, 4, 2] 183s -0.570 0.689 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.200 0.800 -0.458 0.009 183s -0.081 0.708 -0.602 0.604 183s 183s x1: [4, 2, 1, 1] 183s -0.431 0.507 0.085 -0.740 183s 0.107 -0.354 0.322 0.046 183s 183s y: [2, 2, 1, 1] 183s 0.446 -0.104 183s -0.452 -0.426 183s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.700 0.405 183s -0.994 -0.544 183s 0.366 0.213 183s 183s x1: [2, 4, 4, 1] 183s 0.640 0.977 183s 0.309 0.978 183s -0.869 0.705 183s -0.550 0.847 183s 183s y: [3, 4, 4, 1] 183s 0.844 -1.168 0.442 183s 0.613 -0.840 0.321 183s -0.323 0.480 -0.168 183s -0.042 0.086 -0.021 183s testing: mul_mat, [2, 2, 4, 2] = [3, 2, 4, 2] * [3, 2, 4, 2] 183s x0: [3, 2, 4, 2] 183s 0.196 -0.182 0.989 183s -0.948 -0.749 0.143 183s 183s x1: [3, 2, 4, 2] 183s 0.905 -0.901 0.715 183s 0.163 -0.510 0.743 183s 183s y: [2, 2, 4, 2] 183s 1.049 -0.081 183s 0.860 0.334 183s test-mul-mat0: iter:154/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.774 -0.590 -0.865 183s -0.628 0.742 -0.560 183s 183s x1: [3, 2, 1, 1] 183s 0.255 0.196 0.341 183s 0.395 -0.628 0.131 183s 183s y: [2, 2, 1, 1] 183s -0.213 -0.206 183s 0.562 -0.787 183s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 4, 3, 1] 183s -0.233 0.763 0.661 183s 0.140 -0.106 -0.691 183s -0.086 0.304 -0.557 183s 0.285 0.046 -0.117 183s 183s x1: [3, 2, 3, 1] 183s 0.221 0.062 0.888 183s -0.475 0.424 0.873 183s 183s y: [4, 2, 3, 1] 183s 0.583 -0.590 -0.495 -0.038 183s 1.011 -0.715 -0.316 -0.218 183s testing: mul_mat, [1, 4, 3, 4] = [3, 1, 3, 4] * [3, 4, 3, 4] 183s x0: [3, 1, 3, 4] 183s 0.953 -0.168 0.108 183s 183s x1: [3, 4, 3, 4] 183s 0.713 -0.893 0.485 183s -0.844 0.392 -0.469 183s 0.039 0.899 -0.361 183s 0.811 -0.069 -0.829 183s 183s y: [1, 4, 3, 4] 183s 0.882 183s -0.921 183s -0.153 183s 0.695 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.315 -0.363 0.623 183s -0.004 0.235 -0.341 183s 183s x1: [3, 1, 1, 1] 183s 0.193 0.916 0.471 183s 183s y: [2, 1, 1, 1] 183s -0.100 0.054 183s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 4, 3, 1] 183s -0.830 -0.145 183s 0.384 0.577 183s 0.247 -0.229 183s -0.914 0.851 183s 183s x1: [2, 3, 3, 1] 183s -0.721 -0.644 183s 0.762 -0.678 183s 0.405 0.024 183s 183s y: [4, 3, 3, 1] 183s 0.692 -0.649 -0.031 0.112 183s -0.534 -0.099 0.344 -1.273 183s -0.340 0.170 0.095 -0.350 183s testing: mul_mat, [4, 2, 3, 4] = [4, 4, 3, 4] * [4, 2, 3, 4] 183s x0: [4, 4, 3, 4] 183s -0.842 -0.832 0.976 -0.243 183s -0.320 0.972 0.469 0.596 183s 0.165 -0.327 -0.805 0.566 183s 0.768 0.442 0.456 0.422 183s 183s x1: [4, 2, 3, 4] 183s -0.339 -0.649 0.020 0.187 183s 0.569 0.603 0.241 0.739 183s 183s y: [4, 2, 3, 4] 183s 0.799 -0.402 0.245 -0.459 183s -0.925 0.958 0.121 1.126 183s test-mul-mat0: iter:155/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.733 -0.551 0.879 -0.974 183s -0.778 -0.782 -0.168 0.193 183s -0.185 -0.989 0.999 0.716 183s 0.589 -0.464 0.784 -0.242 183s 183s x1: [4, 4, 1, 1] 183s -0.933 0.169 0.566 -0.884 183s -0.160 -0.755 -0.531 -0.460 183s 0.630 0.713 0.115 0.601 183s -0.710 0.947 -0.856 0.732 183s 183s y: [4, 4, 1, 1] 183s 1.949 0.329 -0.063 0.030 183s 0.515 0.716 -0.083 -0.048 183s -1.339 -0.951 -0.275 -0.015 183s -1.467 0.098 -1.135 -1.705 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 1.000 -0.985 -0.226 -0.782 183s 183s x1: [4, 4, 1, 1] 183s -0.898 0.900 0.598 -0.653 183s -0.631 -0.862 0.977 -0.918 183s 0.253 0.578 -0.629 0.200 183s 0.722 -0.896 -0.879 0.989 183s 183s y: [1, 4, 1, 1] 183s -1.408 183s 0.716 183s -0.330 183s 1.030 183s testing: mul_mat, [2, 1, 1, 2] = [4, 2, 1, 2] * [4, 1, 1, 2] 183s x0: [4, 2, 1, 2] 183s 0.205 -0.397 -0.567 -0.197 183s -0.050 -0.198 -0.058 -0.073 183s 183s x1: [4, 1, 1, 2] 183s -0.153 0.967 0.033 -0.142 183s 183s y: [2, 1, 1, 2] 183s -0.406 -0.175 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.936 -0.383 0.517 -0.618 183s 183s x1: [4, 2, 1, 1] 183s -0.784 0.911 0.516 -0.769 183s -0.315 0.733 0.077 -0.348 183s 183s y: [1, 2, 1, 1] 183s -0.341 183s -0.320 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.609 183s 183s x1: [1, 4, 1, 1] 183s 0.118 183s -0.950 183s 0.587 183s 0.722 183s 183s y: [1, 4, 1, 1] 183s -0.072 183s 0.579 183s -0.358 183s -0.440 183s testing: mul_mat, [1, 1, 1, 2] = [1, 1, 1, 2] * [1, 1, 1, 2] 183s x0: [1, 1, 1, 2] 183s 0.599 183s 183s x1: [1, 1, 1, 2] 183s -0.328 183s 183s y: [1, 1, 1, 2] 183s -0.196 183s test-mul-mat0: iter:156/500 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.954 -0.223 0.380 0.687 183s 0.854 -0.969 -0.546 0.790 183s -0.352 0.970 -0.828 0.767 183s -0.980 0.760 0.488 -0.497 183s 183s x1: [4, 1, 1, 1] 183s -0.670 -0.217 -0.562 -0.454 183s 183s y: [4, 1, 1, 1] 183s -1.116 -0.414 0.142 0.443 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.629 0.770 0.623 -0.164 183s 183s x1: [4, 4, 3, 1] 183s -0.850 -0.840 0.789 0.483 183s 0.759 0.958 0.010 0.862 183s 0.382 0.932 -0.808 -0.835 183s -0.630 -0.262 -0.140 -0.676 183s 183s y: [1, 4, 3, 1] 183s 0.300 183s 0.126 183s 0.111 183s 0.218 183s testing: mul_mat, [4, 1, 3, 2] = [4, 4, 3, 2] * [4, 1, 3, 2] 183s x0: [4, 4, 3, 2] 183s 0.590 -0.821 -0.659 0.318 183s 0.530 0.744 -0.995 -0.879 183s 0.480 0.440 0.437 0.310 183s 0.582 -0.514 0.095 0.811 183s 183s x1: [4, 1, 3, 2] 183s -0.313 -0.408 0.737 0.435 183s 183s y: [4, 1, 3, 2] 183s -0.196 -1.585 0.127 0.451 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.552 -0.213 0.621 -0.074 183s 0.822 0.113 0.444 0.336 183s -0.385 0.952 -0.363 0.123 183s 183s x1: [4, 4, 1, 1] 183s -0.942 -0.753 0.354 0.404 183s -0.632 -0.524 0.966 0.735 183s -0.003 0.325 -0.169 -0.990 183s 0.292 -0.143 -0.455 -0.839 183s 183s y: [3, 4, 1, 1] 183s 0.870 -0.567 -0.433 183s 1.006 0.097 -0.515 183s -0.099 -0.374 0.250 183s -0.352 -0.260 -0.187 183s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.743 0.465 0.504 183s 0.198 0.681 -0.826 183s 0.306 0.183 -0.647 183s 183s x1: [3, 4, 3, 1] 183s -0.141 0.656 0.181 183s 0.105 0.010 -0.414 183s 0.473 0.486 -0.448 183s 0.208 -0.517 0.877 183s 183s y: [3, 4, 3, 1] 183s 0.502 0.268 -0.041 183s -0.283 0.370 0.302 183s -0.351 0.795 0.524 183s 0.047 -1.036 -0.599 183s testing: mul_mat, [1, 3, 3, 2] = [3, 1, 3, 2] * [3, 3, 3, 2] 183s x0: [3, 1, 3, 2] 183s -0.872 -0.403 0.066 183s 183s x1: [3, 3, 3, 2] 183s -0.417 -0.852 -0.726 183s 0.337 -0.595 0.472 183s -0.357 0.871 0.153 183s 183s y: [1, 3, 3, 2] 183s 0.659 183s -0.023 183s -0.030 183s test-mul-mat0: iter:157/500 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.028 -0.684 183s 183s x1: [2, 2, 1, 1] 183s 0.096 -0.202 183s 0.786 -0.196 183s 183s y: [1, 2, 1, 1] 183s 0.141 183s 0.156 183s testing: mul_mat, [1, 1, 3, 1] = [2, 1, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 1, 3, 1] 183s 0.590 -0.308 183s 183s x1: [2, 1, 3, 1] 183s -0.068 0.339 183s 183s y: [1, 1, 3, 1] 183s -0.145 183s testing: mul_mat, [2, 1, 3, 3] = [2, 2, 3, 3] * [2, 1, 3, 3] 183s x0: [2, 2, 3, 3] 183s -0.077 0.530 183s -0.565 0.329 183s 183s x1: [2, 1, 3, 3] 183s 0.163 0.394 183s 183s y: [2, 1, 3, 3] 183s 0.196 0.037 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.663 0.976 183s -0.815 -0.927 183s -0.422 0.528 183s 183s x1: [2, 2, 1, 1] 183s 0.964 0.494 183s -0.257 -0.860 183s 183s y: [3, 2, 1, 1] 183s 1.121 -1.243 -0.147 183s -1.010 1.007 -0.345 183s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.809 0.331 0.235 183s 0.095 -0.448 -0.214 183s 183s x1: [3, 2, 3, 1] 183s 0.365 -0.255 -0.116 183s -0.224 -0.714 -0.081 183s 183s y: [2, 2, 3, 1] 183s 0.184 0.174 183s -0.437 0.316 183s testing: mul_mat, [3, 3, 3, 3] = [2, 3, 3, 3] * [2, 3, 3, 3] 183s x0: [2, 3, 3, 3] 183s -0.807 -0.556 183s 0.875 -0.301 183s 0.207 0.688 183s 183s x1: [2, 3, 3, 3] 183s 0.290 -0.681 183s 0.124 -0.014 183s -0.607 0.039 183s 183s y: [3, 3, 3, 3] 183s 0.144 0.459 -0.408 183s -0.092 0.113 0.016 183s 0.468 -0.543 -0.099 183s test-mul-mat0: iter:158/500 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.707 0.479 -0.662 183s -0.032 0.403 -0.359 183s -0.337 0.712 -0.152 183s -0.692 -0.118 -0.869 183s 183s x1: [3, 4, 1, 1] 183s 0.127 0.465 0.902 183s 0.660 -0.721 -0.612 183s 0.399 0.537 0.573 183s -0.173 -0.815 0.570 183s 183s y: [4, 4, 1, 1] 183s -0.464 -0.140 0.151 -0.927 183s -0.406 -0.092 -0.642 0.160 183s -0.404 -0.002 0.161 -0.837 183s -0.645 -0.527 -0.609 -0.279 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.050 0.346 0.255 183s 0.136 -0.084 0.997 183s 0.428 -0.606 -0.666 183s -0.604 0.797 -0.025 183s 183s x1: [3, 4, 1, 1] 183s -0.601 -0.177 -0.806 183s -0.757 0.668 0.484 183s 0.370 0.133 0.386 183s 0.029 0.413 0.774 183s 183s y: [4, 4, 1, 1] 183s -0.237 -0.870 0.386 0.242 183s 0.393 0.324 -1.051 0.978 183s 0.126 0.424 -0.179 -0.126 183s 0.339 0.741 -0.753 0.292 183s testing: mul_mat, [2, 4, 1, 4] = [3, 2, 1, 4] * [3, 4, 1, 4] 183s x0: [3, 2, 1, 4] 183s -0.837 0.071 -0.102 183s -0.745 -0.414 0.952 183s 183s x1: [3, 4, 1, 4] 183s 0.059 0.510 0.823 183s 0.367 -0.608 0.954 183s 0.766 0.215 -0.853 183s -0.991 -0.117 0.631 183s 183s y: [2, 4, 1, 4] 183s -0.097 0.528 183s -0.447 0.886 183s -0.539 -1.472 183s 0.757 1.388 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.482 0.761 0.812 183s 0.175 -0.587 0.880 183s -0.539 0.115 -0.840 183s -0.227 0.291 -0.543 183s 183s x1: [3, 2, 1, 1] 183s -0.168 0.027 0.413 183s -0.198 -0.645 -0.896 183s 183s y: [4, 2, 1, 1] 183s 0.437 0.318 -0.253 -0.178 183s -1.123 -0.444 0.785 0.344 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.836 -0.127 0.497 -0.482 183s 183s x1: [4, 3, 1, 1] 183s -0.957 0.428 0.856 0.427 183s -0.538 -0.599 0.984 0.029 183s 0.308 0.864 0.136 0.459 183s 183s y: [1, 3, 1, 1] 183s 0.966 183s 1.001 183s -0.521 183s testing: mul_mat, [4, 4, 1, 4] = [1, 4, 1, 4] * [1, 4, 1, 4] 183s x0: [1, 4, 1, 4] 183s -0.726 183s 0.256 183s -0.541 183s -0.741 183s 183s x1: [1, 4, 1, 4] 183s -0.023 183s 0.060 183s 0.036 183s -0.847 183s 183s y: [4, 4, 1, 4] 183s 0.017 -0.006 0.012 0.017 183s -0.043 0.015 -0.032 -0.044 183s -0.026 0.009 -0.020 -0.027 183s 0.615 -0.217 0.459 0.628 183s test-mul-mat0: iter:159/500 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.494 -0.165 0.800 0.761 183s 0.293 -0.941 -0.954 0.060 183s 183s x1: [4, 2, 1, 1] 183s 0.142 -0.171 0.873 -0.653 183s -0.425 -0.322 0.712 -0.223 183s 183s y: [2, 2, 1, 1] 183s 0.160 -0.669 183s 0.664 -0.514 183s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 2, 3, 1] 183s 0.121 -0.845 0.893 0.180 183s -0.799 -0.047 0.362 0.669 183s 183s x1: [4, 2, 3, 1] 183s -0.819 0.467 -0.715 -0.001 183s 0.052 -0.616 0.314 0.155 183s 183s y: [2, 2, 3, 1] 183s -1.132 0.372 183s 0.835 0.205 183s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 2, 3, 1] 183s 0.657 -0.736 0.863 -0.982 183s -0.205 0.637 0.366 0.122 183s 183s x1: [4, 2, 3, 1] 183s 0.348 -0.673 0.662 0.113 183s 0.648 -0.944 0.712 -0.231 183s 183s y: [2, 2, 3, 1] 183s 1.185 -0.244 183s 1.963 -0.501 183s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.806 0.831 0.129 -0.517 183s 0.203 -0.431 0.047 0.238 183s 0.447 -0.431 0.339 -0.123 183s 183s x1: [4, 2, 1, 1] 183s 0.567 0.818 0.538 -0.935 183s 0.528 -0.612 -0.669 0.185 183s 183s y: [3, 2, 1, 1] 183s 0.775 -0.435 0.198 183s -1.116 0.384 0.250 183s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.299 -0.082 0.290 183s -0.816 -0.757 -0.629 183s 183s x1: [3, 4, 3, 1] 183s -0.952 0.034 -0.635 183s -0.974 0.389 0.164 183s -0.916 0.539 0.635 183s -0.056 0.106 0.452 183s 183s y: [2, 4, 3, 1] 183s -0.471 1.150 183s -0.275 0.398 183s -0.134 -0.060 183s 0.106 -0.319 183s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 1, 3, 1] 183s 0.657 0.867 183s 183s x1: [2, 3, 3, 1] 183s 0.627 0.845 183s 0.806 -0.496 183s -0.015 -0.972 183s 183s y: [1, 3, 3, 1] 183s 1.144 183s 0.100 183s -0.852 183s test-mul-mat0: iter:160/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.330 -0.407 0.507 183s 0.358 -0.148 0.681 183s -0.356 0.230 0.070 183s -0.425 -0.108 0.758 183s 183s x1: [3, 1, 1, 1] 183s -0.323 0.524 -0.911 183s 183s y: [4, 1, 1, 1] 183s -0.782 -0.814 0.172 -0.610 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.246 -0.565 0.959 183s -0.230 -0.477 -0.519 183s 183s x1: [3, 4, 1, 1] 183s -0.310 0.473 -0.311 183s -0.904 -0.576 -0.654 183s 0.962 -0.274 0.600 183s 0.792 -0.456 0.973 183s 183s y: [2, 4, 1, 1] 183s -0.642 0.008 183s -0.524 0.822 183s 0.967 -0.402 183s 1.386 -0.469 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.198 -0.260 -0.324 183s -0.910 -0.503 0.366 183s 0.563 0.186 0.462 183s 183s x1: [3, 2, 1, 1] 183s -0.900 0.116 0.988 183s 0.458 0.968 0.670 183s 183s y: [3, 2, 1, 1] 183s -0.529 1.122 -0.029 183s -0.378 -0.658 0.747 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.539 -0.937 0.179 183s -0.544 -0.062 -0.074 183s 0.272 0.372 -0.170 183s 183s x1: [3, 3, 1, 1] 183s 0.987 0.532 0.424 183s -0.288 0.131 0.216 183s 0.256 0.105 -0.503 183s 183s y: [3, 3, 1, 1] 183s 0.110 -0.601 0.394 183s -0.240 0.132 -0.066 183s -0.050 -0.109 0.194 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.019 0.859 0.954 183s -0.565 -0.302 -0.213 183s 0.986 0.682 0.571 183s -0.449 0.074 -0.544 183s 183s x1: [3, 3, 1, 1] 183s 0.147 -0.404 -0.069 183s -0.655 0.336 0.607 183s -0.564 0.833 -0.027 183s 183s y: [4, 3, 1, 1] 183s -0.410 0.054 -0.170 -0.059 183s 0.855 0.139 -0.069 -0.011 183s 0.679 0.073 -0.003 0.330 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.823 0.496 0.442 -0.325 183s -0.531 0.656 0.495 0.877 183s 183s x1: [4, 3, 1, 1] 183s -0.674 -0.972 0.728 -0.611 183s -0.035 0.100 0.568 0.891 183s 0.930 -0.285 -0.513 -0.139 183s 183s y: [2, 3, 1, 1] 183s -0.516 -0.454 183s -0.018 1.147 183s 0.442 -1.057 183s test-mul-mat0: iter:161/500 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.265 0.348 -0.308 0.752 183s 183s x1: [4, 3, 1, 1] 183s -0.092 -0.191 -0.867 0.694 183s -0.619 -0.411 -0.980 -0.592 183s -0.682 -0.591 0.373 0.418 183s 183s y: [1, 3, 1, 1] 183s 0.698 183s -0.450 183s -0.187 183s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.595 -0.963 -0.458 -0.462 183s 0.731 -0.078 0.127 0.752 183s 0.331 0.445 -0.839 -0.296 183s 183s x1: [4, 1, 3, 1] 183s -0.790 0.753 0.574 -0.322 183s 183s y: [3, 1, 3, 1] 183s -0.369 -0.806 -0.313 183s testing: mul_mat, [3, 3, 3, 4] = [4, 3, 3, 4] * [4, 3, 3, 4] 183s x0: [4, 3, 3, 4] 183s 0.430 -0.032 -0.746 -0.241 183s -0.069 0.852 0.598 0.186 183s 0.675 -0.002 -0.072 0.134 183s 183s x1: [4, 3, 3, 4] 183s -0.588 -0.128 0.078 0.743 183s -0.683 0.240 -0.553 0.180 183s 0.378 -0.584 0.391 -0.791 183s 183s y: [3, 3, 3, 4] 183s -0.486 0.117 -0.302 183s 0.068 -0.046 -0.398 183s 0.080 -0.437 0.122 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.254 -0.853 -0.606 0.930 183s 0.590 0.965 0.190 -0.589 183s -0.096 -0.450 0.484 0.551 183s 183s x1: [4, 3, 1, 1] 183s -0.908 0.708 -0.333 -0.574 183s -0.814 0.428 -0.861 -0.283 183s 0.546 0.564 0.393 -0.344 183s 183s y: [3, 3, 1, 1] 183s -1.167 0.422 -0.709 183s -0.313 -0.064 -0.687 183s -0.900 1.143 -0.305 183s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.221 -0.696 0.480 183s 0.377 -0.167 0.031 183s 0.731 -0.328 -0.014 183s 0.877 0.949 -0.215 183s 183s x1: [3, 4, 3, 1] 183s -0.910 0.142 -0.051 183s -0.861 0.033 0.007 183s -0.770 -0.259 0.674 183s -0.343 -0.073 0.102 183s 183s y: [4, 4, 3, 1] 183s -0.324 -0.368 -0.711 -0.652 183s -0.210 -0.330 -0.640 -0.726 183s 0.334 -0.226 -0.487 -1.066 183s 0.024 -0.114 -0.229 -0.392 183s testing: mul_mat, [2, 3, 3, 4] = [4, 2, 3, 4] * [4, 3, 3, 4] 183s x0: [4, 2, 3, 4] 183s -0.445 -0.299 -0.502 -0.821 183s 0.585 -0.677 0.689 -0.491 183s 183s x1: [4, 3, 3, 4] 183s -0.453 -0.080 -0.085 -0.972 183s 0.952 0.900 -0.188 -0.406 183s 0.178 -0.509 0.753 -0.107 183s 183s y: [2, 3, 3, 4] 183s 1.066 0.207 183s -0.266 0.017 183s -0.217 1.020 183s test-mul-mat0: iter:162/500 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.232 0.880 183s 0.533 0.677 183s 183s x1: [2, 2, 1, 1] 183s -0.259 -0.373 183s 0.096 0.831 183s 183s y: [2, 2, 1, 1] 183s -0.388 -0.390 183s 0.754 0.614 183s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 1, 3, 1] 183s 0.082 -0.876 183s 183s x1: [2, 2, 3, 1] 183s -0.922 0.036 183s 0.340 -0.842 183s 183s y: [1, 2, 3, 1] 183s -0.107 183s 0.765 183s testing: mul_mat, [4, 1, 3, 2] = [2, 4, 3, 2] * [2, 1, 3, 2] 183s x0: [2, 4, 3, 2] 183s -0.617 -0.273 183s -0.403 -0.940 183s 0.041 0.800 183s 0.514 0.288 183s 183s x1: [2, 1, 3, 2] 183s 0.908 -0.898 183s 183s y: [4, 1, 3, 2] 183s -0.316 0.478 -0.681 0.208 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.764 0.664 183s -0.609 0.634 183s 0.495 0.567 183s 0.302 0.534 183s 183s x1: [2, 4, 1, 1] 183s 0.214 0.090 183s -0.266 0.175 183s 0.083 0.329 183s 0.614 0.557 183s 183s y: [4, 4, 1, 1] 183s 0.223 -0.073 0.157 0.113 183s -0.087 0.273 -0.033 0.013 183s 0.283 0.158 0.228 0.201 183s 0.840 -0.021 0.620 0.483 183s testing: mul_mat, [1, 2, 3, 1] = [4, 1, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.476 0.164 -0.142 -0.842 183s 183s x1: [4, 2, 3, 1] 183s 0.833 -0.127 0.204 0.546 183s 0.468 0.090 0.016 0.255 183s 183s y: [1, 2, 3, 1] 183s -0.906 183s -0.425 183s testing: mul_mat, [1, 4, 3, 2] = [1, 1, 3, 2] * [1, 4, 3, 2] 183s x0: [1, 1, 3, 2] 183s -0.134 183s 183s x1: [1, 4, 3, 2] 183s -0.448 183s -0.673 183s -0.854 183s 0.878 183s 183s y: [1, 4, 3, 2] 183s 0.060 183s 0.090 183s 0.115 183s -0.118 183s test-mul-mat0: iter:163/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.822 -0.107 0.367 183s 0.856 0.248 -0.470 183s 183s x1: [3, 1, 1, 1] 183s 0.580 -0.728 0.782 183s 183s y: [2, 1, 1, 1] 183s 0.841 -0.052 183s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 2, 2, 1] 183s -0.238 0.583 0.090 183s -0.354 -0.462 0.984 183s 183s x1: [3, 2, 2, 1] 183s 0.987 -0.594 -0.277 183s 0.981 -0.244 0.440 183s 183s y: [2, 2, 2, 1] 183s -0.606 -0.347 183s -0.336 0.198 183s testing: mul_mat, [4, 2, 2, 3] = [3, 4, 2, 3] * [3, 2, 2, 3] 183s x0: [3, 4, 2, 3] 183s 0.025 -0.701 -0.968 183s 0.618 -0.245 0.149 183s 0.350 0.176 -0.419 183s -0.889 0.161 0.194 183s 183s x1: [3, 2, 2, 3] 183s -0.230 -0.954 -0.577 183s -0.375 0.294 -0.046 183s 183s y: [4, 2, 2, 3] 183s 1.221 0.005 -0.007 -0.061 183s -0.171 -0.311 -0.060 0.372 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.867 0.898 0.055 183s 0.655 0.129 0.382 183s 0.454 0.982 -0.876 183s 183s x1: [3, 4, 1, 1] 183s 0.701 0.301 -0.259 183s -0.607 -0.573 -0.403 183s -0.611 0.679 -0.421 183s 0.758 0.161 -0.395 183s 183s y: [3, 4, 1, 1] 183s -0.352 0.398 0.841 183s -0.010 -0.625 -0.485 183s 1.116 -0.474 0.758 183s -0.535 0.366 0.848 183s testing: mul_mat, [1, 3, 2, 1] = [3, 1, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 1, 2, 1] 183s 0.604 -0.369 0.874 183s 183s x1: [3, 3, 2, 1] 183s 0.904 -0.554 0.499 183s 0.199 0.471 0.457 183s -0.884 0.036 -0.891 183s 183s y: [1, 3, 2, 1] 183s 1.187 183s 0.346 183s -1.327 183s testing: mul_mat, [3, 3, 2, 3] = [1, 3, 2, 3] * [1, 3, 2, 3] 183s x0: [1, 3, 2, 3] 183s -0.148 183s -0.204 183s -0.466 183s 183s x1: [1, 3, 2, 3] 183s 0.635 183s 0.414 183s 0.310 183s 183s y: [3, 3, 2, 3] 183s -0.094 -0.129 -0.296 183s -0.061 -0.084 -0.193 183s -0.046 -0.063 -0.144 183s test-mul-mat0: iter:164/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.339 -0.019 0.741 183s -0.990 -0.084 -0.408 183s -0.194 0.450 0.887 183s 0.628 -0.658 -0.213 183s 183s x1: [3, 1, 1, 1] 183s 0.904 -0.838 0.256 183s 183s y: [4, 1, 1, 1] 183s -0.101 -0.929 -0.326 1.065 183s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.067 0.488 0.823 183s 0.475 0.535 -0.466 183s -0.524 -0.612 -0.305 183s 183s x1: [3, 4, 4, 1] 183s 0.081 0.107 0.423 183s -0.181 0.201 0.694 183s 0.972 -0.395 -0.345 183s 0.920 0.443 -0.790 183s 183s y: [3, 4, 4, 1] 183s 0.406 -0.101 -0.237 183s 0.658 -0.302 -0.240 183s -0.412 0.411 -0.162 183s -0.373 1.043 -0.512 183s testing: mul_mat, [2, 3, 4, 2] = [3, 2, 4, 2] * [3, 3, 4, 2] 183s x0: [3, 2, 4, 2] 183s 0.548 -0.843 0.354 183s -0.103 -0.289 -0.948 183s 183s x1: [3, 3, 4, 2] 183s 0.638 -0.762 -0.169 183s -0.667 0.603 0.701 183s 0.531 -0.358 -0.868 183s 183s y: [2, 3, 4, 2] 183s 0.932 0.314 183s -0.625 -0.769 183s 0.286 0.871 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.511 0.874 0.417 183s 0.500 0.380 -0.162 183s 0.501 0.912 -0.386 183s -0.660 0.796 -0.835 183s 183s x1: [3, 2, 1, 1] 183s 0.389 0.261 0.386 183s 0.935 -0.475 0.436 183s 183s y: [4, 2, 1, 1] 183s 0.588 0.231 0.284 -0.372 183s 0.244 0.217 -0.133 -1.359 183s testing: mul_mat, [3, 3, 4, 1] = [4, 3, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 3, 4, 1] 183s 0.237 0.727 0.597 -0.116 183s -0.498 -0.130 0.151 -0.730 183s -0.677 -0.472 -0.471 0.144 183s 183s x1: [4, 3, 4, 1] 183s 0.816 0.226 0.067 0.669 183s -0.100 -0.576 -0.406 -0.509 183s -0.184 0.446 -0.268 -0.992 183s 183s y: [3, 3, 4, 1] 183s 0.320 -0.914 -0.595 183s -0.625 0.435 0.458 183s 0.235 0.718 -0.102 183s testing: mul_mat, [4, 4, 4, 2] = [3, 4, 4, 2] * [3, 4, 4, 2] 183s x0: [3, 4, 4, 2] 183s -0.877 -0.829 -0.579 183s 0.660 -0.430 0.702 183s 0.748 0.111 -0.357 183s -0.520 -0.142 -0.006 183s 183s x1: [3, 4, 4, 2] 183s 0.722 0.813 -0.836 183s -0.319 0.375 0.812 183s 0.604 -0.294 -0.062 183s -0.516 0.199 -0.687 183s 183s y: [4, 4, 4, 2] 183s -0.823 -0.460 0.929 -0.486 183s -0.501 0.198 -0.487 0.108 183s -0.250 0.482 0.441 -0.272 183s 0.686 -0.909 -0.119 0.244 183s test-mul-mat0: iter:165/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.446 0.410 183s -0.186 -0.554 183s 183s x1: [2, 4, 1, 1] 183s 0.980 -0.613 183s -0.678 -0.662 183s 0.590 0.255 183s -0.336 0.315 183s 183s y: [2, 4, 1, 1] 183s 0.185 0.158 183s -0.574 0.492 183s 0.368 -0.251 183s -0.021 -0.112 183s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 2, 3,ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s 1] 183s -0.499 -0.400 183s -0.271 0.500 183s 183s x1: [2, 2, 3, 1] 183s -0.257 -0.801 183s 0.435 -0.143 183s 183s y: [2, 2, 3, 1] 183s 0.449 -0.331 183s -0.159 -0.190 183s testing: mul_mat, [3, 2, 3, 1] = [2, 3, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 3, 3, 1] 183s 0.915 -0.846 183s -0.840 0.984 183s -0.056 -0.912 183s 183s x1: [2, 2, 3, 1] 183s -0.208 0.224 183s 0.156 -0.798 183s 183s y: [3, 2, 3, 1] 183s -0.380 0.396 -0.193 183s 0.818 -0.917 0.719 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.009 -0.921 183s 183s x1: [2, 3, 1, 1] 183s -0.858 -0.267 183s 0.439 -0.656 183s -0.228 0.042 183s 183s y: [1, 3, 1, 1] 183s 0.238 183s 0.608 183s -0.040 183s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 3, 3, 1] 183s 0.019 183s -0.676 183s 0.725 183s 183s x1: [1, 2, 3, 1] 183s 0.091 183s 0.120 183s 183s y: [3, 2, 3, 1] 183s 0.002 -0.061 0.066 183s 0.002 -0.081 0.087 183s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.642 -0.511 0.481 183s 0.402 0.368 0.579 183s 183s x1: [3, 1, 3, 1] 183s 0.364 0.344 -0.361 183s 183s y: [2, 1, 3, 1] 183s -0.583 0.063 183s test-mul-mat0: iter:166/500 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.479 -0.654 -0.840 0.888 183s 0.827 0.739 -0.624 -0.467 183s -0.205 0.900 -0.381 0.487 183s -0.557 0.029 0.904 -0.510 183s 183s x1: [4, 2, 1, 1] 183s 0.722 0.664 0.616 -0.424 183s -0.415 -0.529 -0.838 -0.055 183s 183s y: [4, 2, 1, 1] 183s -1.674 0.902 0.008 0.389 183s 1.200 -0.186 -0.098 -0.513 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.152 -0.447 0.253 -0.413 183s -0.071 0.786 0.382 -0.171 183s 183s x1: [4, 4, 1, 1] 183s 0.306 -0.138 -0.971 0.684 183s 0.887 -0.069 0.406 0.550 183s -0.453 0.982 -0.865 0.018 183s -0.856 0.080 -0.311 -0.335 183s 183s y: [2, 4, 1, 1] 183s -0.513 -0.618 183s -0.228 -0.056 183s -0.596 0.471 183s 0.154 0.062 183s testing: mul_mat, [4, 2, 1, 4] = [4, 4, 1, 4] * [4, 2, 1, 4] 183s x0: [4, 4, 1, 4] 183s -0.795 -0.409 0.434 0.967 183s 0.331 0.069 0.758 0.755 183s -0.410 -0.520 0.585 0.661 183s 0.749 0.398 -0.132 -0.965 183s 183s x1: [4, 2, 1, 4] 183s -0.595 -0.131 0.272 0.434 183s -0.227 0.762 -0.260 0.635 183s 183s y: [4, 2, 1, 4] 183s 1.064 0.328 0.758 -0.952 183s 0.369 0.259 -0.035 -0.445 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.773 0.097 -0.915 0.795 183s 0.926 0.577 0.091 -0.833 183s -0.892 -0.223 -0.656 0.632 183s -0.351 -0.148 -0.708 -0.145 183s 183s x1: [4, 4, 1, 1] 183s 0.341 0.060 0.482 0.543 183s 0.175 -0.103 -0.665 0.770 183s 0.897 -0.212 -0.886 -0.706 183s -0.890 -0.507 -0.954 -0.415 183s 183s y: [4, 4, 1, 1] 183s 0.259 -0.058 -0.290 -0.549 183s 1.346 -0.600 0.790 0.313 183s 0.922 1.215 -0.619 0.446 183s -0.194 -0.857 1.270 1.124 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.602 0.981 0.212 0.984 183s 0.204 -0.133 -0.648 0.723 183s -0.984 -0.940 -0.133 0.027 183s 0.975 0.073 0.379 0.159 183s 183s x1: [4, 4, 1, 1] 183s 0.649 0.175 -0.970 -0.454 183s 0.510 0.800 -0.558 -0.701 183s 0.914 -0.263 -0.592 -0.593 183s -0.217 -0.007 -0.710 -0.445 183s 183s y: [4, 4, 1, 1] 183s -0.871 0.410 -0.687 0.206 183s -0.330 -0.147 -1.199 0.233 183s -1.517 0.177 -0.590 0.554 183s -0.464 0.095 0.302 -0.552 183s testing: mul_mat, [4, 4, 1, 4] = [4, 4, 1, 4] * [4, 4, 1, 4] 183s x0: [4, 4, 1, 4] 183s 0.759 0.043 -0.505 0.110 183s -0.087 0.018 -0.200 0.511 183s 0.911 0.053 -0.538 0.329 183s 0.469 -0.293 0.260 0.956 183s 183s x1: [4, 4, 1, 4] 183s 0.753 0.573 -0.330 0.553 183s -0.985 -0.032 0.467 -0.248 183s 0.377 0.874 0.535 -0.630 183s -0.836 -0.910 0.288 -0.438 183s 183s y: [4, 4, 1, 4] 183s 0.824 0.293 1.075 0.628 183s -1.012 -0.135 -1.231 -0.568 183s -0.016 -0.446 -0.106 -0.543 183s -0.867 -0.225 -1.108 -0.470 183s test-mul-mat0: iter:167/500 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.827 0.005 183s 183s x1: [2, 3, 1, 1] 183s -0.908 0.494 183s -0.129 0.909 183s 0.905 -0.726 183s 183s y: [1, 3, 1, 1] 183s -0.748 183s -0.102 183s 0.745 183s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.497 0.863 183s 0.277 -0.619 183s -0.320 0.011 183s 183s x1: [2, 1, 3, 1] 183s 0.720 0.583 183s 183s y: [3, 1, 3, 1] 183s 0.145 -0.162 -0.224 183s testing: mul_mat, [3, 3, 3, 2] = [2, 3, 3, 2] * [2, 3, 3, 2] 183s x0: [2, 3, 3, 2] 183s -0.351 -0.983 183s -0.901 -0.611 183s 0.302 0.621 183s 183s x1: [2, 3, 3, 2] 183s 0.180 -0.970 183s -0.582 -0.984 183s 0.857 0.423 183s 183s y: [3, 3, 3, 2] 183s 0.891 0.431 -0.548 183s 1.171 1.125 -0.787 183s -0.716 -1.030 0.521 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.640 -0.373 183s -0.496 -0.802 183s 0.055 -0.093 183s 183s x1: [2, 3, 1, 1] 183s 0.063 -0.084 183s -0.181 -0.932 183s -0.776 -0.921 183s 183s y: [3, 3, 1, 1] 183s 0.072 0.036 0.011 183s 0.232 0.838 0.076 183s -0.153 1.123 0.043 183s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 4, 3, 1] 183s -0.444 0.725 -0.682 183s 0.646 0.264 0.046 183s -0.874 -0.581 -0.573 183s -0.221 -0.771 0.226 183s 183s x1: [3, 2, 3, 1] 183s 0.729 -0.662 0.326 183s -0.782 -0.804 -0.564 183s 183s y: [4, 2, 3, 1] 183s -1.026 0.311 -0.439 0.423 183s 0.149 -0.743 1.473 0.665 183s testing: mul_mat, [1, 3, 3, 2] = [4, 1, 3, 2] * [4, 3, 3, 2] 183s x0: [4, 1, 3, 2] 183s 0.633 0.223 0.371 -0.937 183s 183s x1: [4, 3, 3, 2] 183s 0.440 -0.142 0.455 0.758 183s 0.904 0.882 -0.016 -0.322 183s -0.352 0.536 0.670 0.610 183s 183s y: [1, 3, 3, 2] 183s -0.295 183s 1.064 183s -0.426 183s test-mul-mat0: iter:168/500 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.672 -0.999 -0.930 -0.230 183s -0.454 -0.855 0.621 0.086 183s -0.768 -0.788 -0.759 -0.080 183s 183s x1: [4, 1, 1, 1] 183s -0.696 -0.948 -0.043 0.937 183s 183s y: [3, 1, 1, 1] 183s 1.240 1.181 1.239 183s testing: mul_mat, [4, 3, 2, 1] = [4, 4, 2, 1] * [4, 3, 2, 1] 183s x0: [4, 4, 2, 1] 183s -0.731 0.289 0.215 -0.519 183s 0.530 -0.866 -0.599 0.818 183s -0.568 -0.297 0.761 -0.889 183s 0.052 -0.458 0.945 0.870 183s 183s x1: [4, 3, 2, 1] 183s 0.920 -0.712 -0.703 -0.698 183s 0.943 0.679 -0.650 -0.219 183s 0.834 -0.182 -0.721 0.415 183s 183s y: [4, 3, 2, 1] 183s -0.667 0.954 -0.225 -0.897 183s -0.519 0.122 -1.038 -1.067 183s -1.033 1.371 -1.338 -0.194 183s testing: mul_mat, [3, 4, 2, 3] = [4, 3, 2, 3] * [4, 4, 2, 3] 183s x0: [4, 3, 2, 3] 183s 0.037 -0.748 0.672 -0.492 183s -0.248 -0.907 0.546 -0.326 183s -0.047 0.383 -0.719 -0.395 183s 183s x1: [4, 4, 2, 3] 183s 0.883 -0.391 -0.763 0.413 183s -0.257 -0.362 0.230 0.174 183s 0.341 -0.009 0.285 -0.607 183s 0.533 0.230 -0.737 -0.647 183s 183s y: [3, 4, 2, 3] 183s -0.391 -0.415 0.194 183s 0.330 0.461 -0.361 183s 0.510 0.277 0.015 183s -0.329 -0.532 0.848 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.927 0.332 -0.503 0.810 183s 0.561 0.931 0.369 0.677 183s 0.839 -0.340 0.507 -0.882 183s -0.070 -0.910 -0.956 -0.344 183s 183s x1: [4, 3, 1, 1] 183s 0.532 -0.813 -0.498 0.368 183s 0.314 0.909 -0.138 -0.244 183s 0.950 -0.886 -0.130 -0.381 183s 183s y: [4, 3, 1, 1] 183s 0.773 -0.393 0.146 1.052 183s 0.466 0.807 0.099 -0.634 183s 0.343 -0.597 1.368 0.994 183s testing: mul_mat, [3, 4, 2, 1] = [4, 3, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.780 0.777 -0.804 -0.271 183s -0.458 0.595 0.405 -0.915 183s -0.150 0.363 0.826 -0.091 183s 183s x1: [4, 4, 2, 1] 183s 0.553 -0.898 -0.811 0.739 183s -0.396 0.557 0.054 -0.487 183s -0.581 0.810 -0.537 -0.467 183s -0.320 0.082 0.716 -0.393 183s 183s y: [3, 4, 2, 1] 183s -0.677 -1.793 -1.147 183s 0.830 0.980 0.350 183s 1.641 0.958 -0.020 183s -0.156 0.845 0.705 183s testing: mul_mat, [3, 4, 2, 3] = [3, 3, 2, 3] * [3, 4, 2, 3] 183s x0: [3, 3, 2, 3] 183s 0.649 0.831 -0.141 183s 0.162 -0.366 -0.064 183s 0.321 0.202 0.849 183s 183s x1: [3, 4, 2, 3] 183s 0.492 0.367 -0.584 183s 0.636 0.399 0.219 183s 0.882 0.619 0.760 183s -0.269 0.396 0.355 183s 183s y: [3, 4, 2, 3] 183s 0.707 -0.017 -0.263 183s 0.713 -0.057 0.471 183s 0.980 -0.132 1.054 183s 0.105 -0.211 0.296 183s test-mul-mat0: iter:169/500 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.748 -0.605 183s 183s x1: [2, 1, 1, 1] 183s -0.796 0.416 183s 183s y: [1, 1, 1, 1] 183s 0.343 183s testing: mul_mat, [4, 1, 4, 1] = [2, 4, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.010 -0.703 183s -0.177 0.657 183s -0.098 0.968 183s -0.383 0.807 183s 183s x1: [2, 1, 4, 1] 183s 0.893 0.621 183s 183s y: [4, 1, 4, 1] 183s -0.428 0.251 0.514 0.160 183s testing: mul_mat, [3, 4, 4, 2] = [2, 3, 4, 2] * [2, 4, 4, 2] 183s x0: [2, 3, 4, 2] 183s 0.910 -0.919 183s -0.161 -0.895 183s -0.509 -0.162 183s 183s x1: [2, 4, 4, 2] 183s -0.362 -0.589 183s 0.834 0.540 183s -0.621 -0.549 183s 0.347 -0.504 183s 183s y: [3, 4, 4, 2] 183s 0.212 0.586 0.279 183s 0.262 -0.618 -0.512 183s -0.061 0.591 0.405 183s 0.780 0.395 -0.095 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.634 0.054 183s 183s x1: [2, 3, 1, 1] 183s -1.000 -0.340 183s 0.377 0.732 183s -0.310 0.338 183s 183s y: [1, 3, 1, 1] 183s -0.652 183s 0.278 183s -0.179 183s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 4, 4, 1] 183s 0.584 183s 0.847 183s 0.345 183s 0.125 183s 183s x1: [1, 2, 4, 1] 183s -0.165 183s -0.905 183s 183s y: [4, 2, 4, 1] 183s -0.097 -0.140 -0.057 -0.021 183s -0.529 -0.767 -0.313 -0.113 183s testing: mul_mat, [2, 1, 4, 2] = [4, 2, 4, 2] * [4, 1, 4, 2] 183s x0: [4, 2, 4, 2] 183s 0.191 0.231 -0.469 0.399 183s -0.292 0.380 -0.340 0.222 183s 183s x1: [4, 1, 4, 2] 183s -0.980 0.700 0.135 -0.011 183s 183s y: [2, 1, 4, 2] 183s -0.093 0.503 183s test-mul-mat0: iter:170/500 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.631 -0.975 -0.982 -0.788 183s 0.351 -0.762 0.471 -0.036 183s 183s x1: [4, 2, 1, 1] 183s -0.695 -0.408 0.049 -0.187 183s -0.204 -0.480 0.104 0.147 183s 183s y: [2, 2, 1, 1] 183s 0.936 0.097 183s 0.379 0.338 183s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 1, 4, 1] 183s -0.230 -0.304 0.775 -0.342 183s 183s x1: [4, 2, 4, 1] 183s 0.889 0.332 0.124 -0.329 183s 0.991 0.888 -0.697 0.344 183s 183s y: [1, 2, 4, 1] 183s -0.096 183s -1.155 183s testing: mul_mat, [4, 1, 4, 2] = [4, 4, 4, 2] * [4, 1, 4, 2] 183s x0: [4, 4, 4, 2] 183s 0.845 -0.257 0.655 -0.929 183s -0.830 0.275 0.277 -0.722 183s 0.031 -0.064 -0.967 -0.187 183s -0.621 -0.443 0.291 0.596 183s 183s x1: [4, 1, 4, 2] 183s -0.212 0.839 0.457 -0.905 183s 183s y: [4, 1, 4, 2] 183s 0.746 1.187 -0.334 -0.647 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.064 0.503 -0.474 -0.497 183s -0.024 0.746 -1.000 -0.447 183s 183s x1: [4, 4, 1, 1] 183s -0.988 -0.177 0.712 0.347 183s 0.643 -0.633 -0.139 0.707 183s 0.026 -0.187 0.439 0.105 183s 0.200 -0.965 0.954 -0.952 183s 183s y: [2, 4, 1, 1] 183s -0.536 -0.975 183s -0.645 -0.665 183s -0.356 -0.626 183s -0.479 -1.254 183s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 3, 4, 1] 183s -0.323 0.987 183s 0.765 -0.949 183s -0.849 -0.225 183s 183s x1: [2, 4, 4, 1] 183s 0.116 0.690 183s -0.932 -0.726 183s 0.801 0.310 183s -0.714 -0.376 183s 183s y: [3, 4, 4, 1] 183s 0.643 -0.565 -0.253 183s -0.416 -0.024 0.954 183s 0.048 0.319 -0.750 183s -0.141 -0.190 0.690 183s testing: mul_mat, [1, 2, 4, 2] = [3, 1, 4, 2] * [3, 2, 4, 2] 183s x0: [3, 1, 4, 2] 183s -0.612 0.341 0.945 183s 183s x1: [3, 2, 4, 2] 183s -0.622 -0.866 0.502 183s 0.038 -0.257 -0.007 183s 183s y: [1, 2, 4, 2] 183s 0.560 183s -0.118 183s test-mul-mat0: iter:171/500 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.699 -0.744 183s -0.103 -0.607 183s -0.863 0.980 183s 0.178 0.455 183s 183s x1: [2, 2, 1, 1] 183s -0.327 -0.751 183s -0.644 0.061 183s 183s y: [4, 2, 1, 1] 183s 0.787 0.489 -0.454 -0.400 183s 0.404 0.029 0.616 -0.086 183s testing: mul_mat, [2, 4, 2, 1] = [2, 2, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.124 -0.612 183s -0.103 0.480 183s 183s x1: [2, 4, 2, 1] 183s 0.059 -0.268 183s 0.337 -0.487 183s 0.189 0.284 183s 0.380 0.293 183s 183s y: [2, 4, 2, 1] 183s 0.172 -0.135 183s 0.340 -0.268 183s -0.150 0.117 183s -0.132 0.102 183s testing: mul_mat, [2, 2, 2, 2] = [2, 2, 2, 2] * [2, 2, 2, 2] 183s x0: [2, 2, 2, 2] 183s 0.319 -0.584 183s 0.139 0.878 183s 183s x1: [2, 2, 2, 2] 183s 0.384 -0.825 183s 0.842 0.364 183s 183s y: [2, 2, 2, 2] 183s 0.604 -0.671 183s 0.057 0.436 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.875 0.280 183s 0.448 0.148 183s 183s x1: [2, 2, 1, 1] 183s 0.516 0.920 183s 0.791 -0.186 183s 183s y: [2, 2, 1, 1] 183s 0.709 0.367 183s 0.640 0.327 183s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.228 -0.974 183s 0.507 -0.647 183s 0.668 -0.744 183s 0.774 0.994 183s 183s x1: [2, 2, 2, 1] 183s 0.367 0.597 183s 0.713 0.575 183s 183s y: [4, 2, 2, 1] 183s -0.665 -0.200 -0.199 0.877 183s -0.723 -0.011 0.049 1.124 183s testing: mul_mat, [1, 2, 2, 2] = [4, 1, 2, 2] * [4, 2, 2, 2] 183s x0: [4, 1, 2, 2] 183s -0.064 0.379 0.702 -0.292 183s 183s x1: [4, 2, 2, 2] 183s 0.998 -0.656 -0.376 0.446 183s 0.624 0.772 -0.187 0.221 183s 183s y: [1, 2, 2, 2] 183s -0.707 183s 0.057 183s test-mul-mat0: iter:172/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.821 0.997 -0.416 183s 0.556 -0.691 -0.482 183s 0.633 0.245 0.897 183s 0.335 0.952 -0.911 183s 183s x1: [3, 1, 1, 1] 183s 0.896 -0.912 -0.565 183s 183s y: [4, 1, 1, 1] 183s 0.062 1.401 -0.163 -0.054 183s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 4, 2, 1] 183s -0.071 -0.992 -0.902 183s -0.736 0.960 -0.813 183s -0.478 -0.895 -0.145 183s 0.882 -0.330 0.385 183s 183s x1: [3, 4, 2, 1] 183s -0.743 -0.855 -0.332 183s 0.360 -0.435 -0.470 183s -0.980 0.093 -0.553 183s 0.742 0.398 -0.808 183s 183s y: [4, 4, 2, 1] 183s 1.200 -0.004 1.169 -0.500 183s 0.830 -0.300 0.285 0.280 183s 0.476 1.261 0.466 -1.108 183s 0.282 0.493 -0.594 0.212 183s testing: mul_mat, [3, 4, 2, 1] = [3, 3ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s , 2, 1] * [3, 4, 2, 1] 183s x0: [3, 3, 2, 1] 183s 0.289 -0.622 -0.940 183s 0.938 0.290 -0.437 183s 0.687 0.942 0.521 183s 183s x1: [3, 4, 2, 1] 183s 0.749 -0.348 -0.043 183s 0.432 0.949 0.720 183s 0.633 -0.122 0.728 183s 0.731 0.143 0.688 183s 183s y: [3, 4, 2, 1] 183s 0.472 0.621 0.165 183s -1.143 0.366 1.566 183s -0.426 0.241 0.699 183s -0.524 0.427 0.995 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.658 0.334 -0.099 183s -0.477 -0.053 0.394 183s 183s x1: [3, 3, 1, 1] 183s -0.599 0.576 -0.550 183s 0.174 0.121 -0.088 183s 0.332 -0.432 0.137 183s 183s y: [2, 3, 1, 1] 183s -0.147 0.038 183s 0.164 -0.124 183s 0.060 -0.082 183s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 3, 2, 1] 183s -0.295 -0.719 183s -0.655 -0.206 183s 0.330 -0.496 183s 183s x1: [2, 3, 2, 1] 183s -0.114 -0.809 183s 0.957 -0.427 183s -0.867 0.478 183s 183s y: [3, 3, 2, 1] 183s 0.616 0.242 0.364 183s 0.025 -0.539 0.528 183s -0.088 0.470 -0.523 183s testing: mul_mat, [3, 2, 2, 1] = [3, 3, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 3, 2, 1] 183s -0.387 0.406 -0.889 183s -0.962 0.726 -0.929 183s -0.829 0.099 -0.571 183s 183s x1: [3, 2, 2, 1] 183s 0.629 0.366 -0.509 183s -0.424 -0.734 0.885 183s 183s y: [3, 2, 2, 1] 183s 0.358 0.134 -0.194 183s -0.921 -0.947 -0.226 183s test-mul-mat0: iter:173/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.502 -0.857 183s 0.464 -0.908 183s 0.463 0.190 183s 183s x1: [2, 1, 1, 1] 183s 0.045 -0.699 183s 183s y: [3, 1, 1, 1] 183s 0.622 0.656 -0.111 183s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 3, 2, 1] 183s -0.266 -0.410 183s -0.828 -0.853 183s -0.485 0.564 183s 183s x1: [2, 3, 2, 1] 183s -0.295 0.501 183s 0.360 -0.888 183s 0.228 -0.540 183s 183s y: [3, 3, 2, 1] 183s -0.127 -0.184 0.426 183s 0.268 0.459 -0.676 183s 0.161 0.272 -0.415 183s testing: mul_mat, [1, 3, 2, 4] = [2, 1, 2, 4] * [2, 3, 2, 4] 183s x0: [2, 1, 2, 4] 183s 0.962 0.197 183s 183s x1: [2, 3, 2, 4] 183s 0.474 0.376 183s -0.575 0.566 183s -0.161 0.615 183s 183s y: [1, 3, 2, 4] 183s 0.531 183s -0.442 183s -0.034 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.006 -0.054 183s -0.245 0.510 183s 183s x1: [2, 1, 1, 1] 183s -0.116 -0.215 183s 183s y: [2, 1, 1, 1] 183s 0.011 -0.082 183s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 4, 2, 1] 183s 0.331 -0.053 183s -0.818 0.810 183s -0.319 -0.618 183s -0.562 0.895 183s 183s x1: [2, 2, 2, 1] 183s -0.706 0.255 183s -0.378 -0.301 183s 183s y: [4, 2, 2, 1] 183s -0.247 0.784 0.067 0.625 183s -0.109 0.066 0.307 -0.057 183s testing: mul_mat, [4, 2, 2, 4] = [4, 4, 2, 4] * [4, 2, 2, 4] 183s x0: [4, 4, 2, 4] 183s 0.945 0.364 0.624 -0.391 183s -0.694 -0.928 0.359 0.055 183s 0.667 0.520 -0.686 0.600 183s 0.759 -0.523 0.240 0.706 183s 183s x1: [4, 2, 2, 4] 183s -0.803 0.966 -0.941 -0.047 183s -0.088 0.569 0.247 -0.833 183s 183s y: [4, 2, 2, 4] 183s -0.976 -0.680 0.584 -1.374 183s 0.604 -0.424 -0.432 -0.894 183s test-mul-mat0: iter:174/500 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.185 0.171 183s -0.014 -0.749 183s 0.621 0.891 183s -0.130 0.051 183s 183s x1: [2, 1, 1, 1] 183s -0.252 -0.710 183s 183s y: [4, 1, 1, 1] 183s -0.168 0.535 -0.789 -0.004 183s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 1, 4, 1] 183s 0.832 0.908 183s 183s x1: [2, 4, 4, 1] 183s -0.300 0.788 183s -0.410 0.002 183s -0.003 0.315 183s -0.517 0.177 183s 183s y: [1, 4, 4, 1] 183s 0.465 183s -0.340 183s 0.283 183s -0.269 183s testing: mul_mat, [2, 1, 4, 2] = [2, 2, 4, 2] * [2, 1, 4, 2] 183s x0: [2, 2, 4, 2] 183s 0.534 0.387 183s -0.922 -0.224 183s 183s x1: [2, 1, 4, 2] 183s -0.769 -0.961 183s 183s y: [2, 1, 4, 2] 183s -0.782 0.924 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.720 -0.058 183s 183s x1: [2, 2, 1, 1] 183s -0.553 0.524 183s 0.112 -0.260 183s 183s y: [1, 2, 1, 1] 183s -0.429 183s 0.096 183s testing: mul_mat, [2, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 2, 4, 1] 183s -0.498 183s 0.034 183s 183s x1: [1, 2, 4, 1] 183s 0.530 183s 0.476 183s 183s y: [2, 2, 4, 1] 183s -0.264 0.018 183s -0.237 0.016 183s testing: mul_mat, [3, 1, 4, 2] = [2, 3, 4, 2] * [2, 1, 4, 2] 183s x0: [2, 3, 4, 2] 183s 0.494 0.458 183s -0.785 0.641 183s -0.582 -0.707 183s 183s x1: [2, 1, 4, 2] 183s 0.337 -0.129 183s 183s y: [3, 1, 4, 2] 183s 0.107 -0.348 -0.105 183s test-mul-mat0: iter:175/500 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.384 0.373 183s 0.167 0.231 183s 183s x1: [2, 3, 1, 1] 183s 0.776 -0.459 183s -0.824 0.003 183s 0.032 -0.617 183s 183s y: [2, 3, 1, 1] 183s 0.127 0.024 183s -0.316 -0.137 183s -0.218 -0.137 183s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 183s x0:ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png [2, 4, 2, 1] 183s 0.508 0.560 183s 0.136 0.387 183s -0.310 0.071 183s 0.391 -0.345 183s 183s x1: [2, 2, 2, 1] 183s -0.423 -0.536 183s -0.465 0.423 183s 183s y: [4, 2, 2, 1] 183s -0.515 -0.265 0.093 0.019 183s 0.001 0.101 0.174 -0.328 183s testing: mul_mat, [1, 4, 2, 2] = [2, 1, 2, 2] * [2, 4, 2, 2] 183s x0: [2, 1, 2, 2] 183s 0.261 0.920 183s 183s x1: [2, 4, 2, 2] 183s -0.509 0.805 183s -0.269 0.864 183s -0.028 0.962 183s -0.559 0.437 183s 183s y: [1, 4, 2, 2] 183s 0.608 183s 0.725 183s 0.878 183s 0.256 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.402 -0.223 183s 0.311 0.745 183s 0.352 0.056 183s 183s x1: [2, 1, 1, 1] 183s 0.979 -0.986 183s 183s y: [3, 1, 1, 1] 183s -0.174 -0.429 0.289 183s testing: mul_mat, [3, 2, 2, 1] = [3, 3, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 3, 2, 1] 183s -0.399 0.652 -0.860 183s 0.992 0.215 -0.067 183s -0.828 -0.467 -0.242 183s 183s x1: [3, 2, 2, 1] 183s 0.647 0.249 -0.850 183s 0.171 0.492 0.018 183s 183s y: [3, 2, 2, 1] 183s 0.636 0.752 -0.447 183s 0.237 0.274 -0.376 183s testing: mul_mat, [1, 3, 2, 2] = [3, 1, 2, 2] * [3, 3, 2, 2] 183s x0: [3, 1, 2, 2] 183s -0.370 -0.357 0.275 183s 183s x1: [3, 3, 2, 2] 183s 0.140 0.253 -0.650 183s -0.688 -0.255 0.369 183s -0.059 -0.006 0.570 183s 183s y: [1, 3, 2, 2] 183s -0.321 183s 0.447 183s 0.181 183s test-mul-mat0: iter:176/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.659 0.412 183s 183s x1: [2, 4, 1, 1] 183s 0.982 0.209 183s -0.710 -0.274 183s -0.143 -0.924 183s -0.037 -0.492 183s 183s y: [1, 4, 1, 1] 183s -0.561 183s 0.355 183s -0.286 183s -0.179 183s testing: mul_mat, [1, 1, 3, 1] = [2, 1, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.575 0.864 183s 183s x1: [2, 1, 3, 1] 183s -0.601 -0.970 183s 183s y: [1, 1, 3, 1] 183s -0.493 183s testing: mul_mat, [3, 1, 3, 4] = [2, 3, 3, 4] * [2, 1, 3, 4] 183s x0: [2, 3, 3, 4] 183s 0.271 -0.361 183s 0.999 -0.153 183s -0.220 -0.176 183s 183s x1: [2, 1, 3, 4] 183s -0.715 -0.813 183s 183s y: [3, 1, 3, 4] 183s 0.100 -0.590 0.300 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.868 -0.116 183s -0.125 0.658 183s -0.241 -0.657 183s 183s x1: [2, 3, 1, 1] 183s -0.948 -0.868 183s -0.148 0.598 183s 0.816 -0.673 183s 183s y: [3, 3, 1, 1] 183s -0.722 -0.452 0.798 183s -0.198 0.412 -0.357 183s 0.787 -0.544 0.245 183s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.102 0.971 -0.868 183s 0.805 -0.082 0.200 183s 183s x1: [3, 2, 3, 1] 183s -0.129 -0.764 -0.834 183s -0.100 -0.725 0.073 183s 183s y: [2, 2, 3, 1] 183s -0.031 -0.208 183s -0.777 -0.007 183s testing: mul_mat, [1, 3, 3, 4] = [2, 1, 3, 4] * [2, 3, 3, 4] 183s x0: [2, 1, 3, 4] 183s 0.601 -0.051 183s 183s x1: [2, 3, 3, 4] 183s 0.185 0.995 183s 0.878 0.230 183s 0.861 -0.321 183s 183s y: [1, 3, 3, 4] 183s 0.060 183s 0.515 183s 0.534 183s test-mul-mat0: iter:177/500 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.395 -0.244 0.073 0.843 183s 183s x1: [4, 4, 1, 1] 183s -0.669 0.549 -0.353 0.932 183s -0.502 0.172 -0.413 -0.684 183s -0.091 0.465 -0.319 0.737 183s -0.258 0.225 0.203 0.259 183s 183s y: [1, 4, 1, 1] 183s 0.362 183s -0.847 183s 0.448 183s 0.076 183s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 1, 4, 1] 183s -0.058 0.112 0.740 0.623 183s 183s x1: [4, 1, 4, 1] 183s -0.849 -0.272 0.259 -0.645 183s 183s y: [1, 1, 4, 1] 183s -0.192 183s testing: mul_mat, [3, 1, 4, 3] = [4, 3, 4, 3] * [4, 1, 4, 3] 183s x0: [4, 3, 4, 3] 183s 0.660 0.068 0.327 -0.459 183s 0.844 0.646 -0.842 -0.933 183s -0.537 0.583 0.580 0.927 183s 183s x1: [4, 1, 4, 3] 183s -0.451 0.004 0.214 -0.866 183s 183s y: [3, 1, 4, 3] 183s 0.169 0.248 -0.434 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.676 -0.229 0.120 0.516 183s 0.014 0.722 -0.949 0.342 183s -0.233 -0.317 0.782 0.894 183s -0.614 -0.098 -0.802 -0.690 183s 183s x1: [4, 3, 1, 1] 183s -0.296 0.807 -0.956 0.403 183s -0.922 0.384 -0.583 -0.368 183s 0.727 -0.063 0.461 -0.914 183s 183s y: [4, 3, 1, 1] 183s 0.108 1.624 -0.574 0.591 183s 0.275 0.691 -0.692 1.250 183s -0.894 -0.786 -0.606 -0.179 183s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 4, 4, 1] 183s -0.936 0.368 -0.455 -0.964 183s -0.251 0.663 0.298 0.007 183s -0.449 -0.538 -0.053 0.047 183s 0.204 -0.261 -0.916 0.063 183s 183s x1: [4, 4, 4, 1] 183s 0.600 0.597 0.014 0.407 183s 0.640 -0.584 0.485 0.024 183s -0.167 -0.883 -0.249 0.770 183s 0.578 -0.163 -0.343 0.902 183s 183s y: [4, 4, 4, 1] 183s -0.741 0.251 -0.572 -0.020 183s -1.059 -0.403 0.002 -0.160 183s -0.798 -0.612 0.599 0.473 183s -1.315 -0.350 -0.112 0.532 183s testing: mul_mat, [4, 4, 4, 3] = [4, 4, 4, 3] * [4, 4, 4, 3] 183s x0: [4, 4, 4, 3] 183s -0.139 -0.9 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.p44 0.920 0.210 183s 0.581 0.069 -0.056 0.876 183s 0.618 -0.045 -0.482 0.093 183s 0.233 -0.310 -0.180 -0.497 183s 183s x1: [4, 4, 4, 3] 183s -0.848 0.335 0.863 -0.308 183s 0.039 0.438 0.777 -0.960 183s 0.961 -0.054 0.077 0.491 183s 0.162 0.178 -0.291 -0.467 183s 183s y: [4, 4, 4, 3] 183s 0.531 -0.787 -0.983 -0.304 183s 0.095 -0.832 -0.460 0.210 183s 0.091 0.980 0.604 -0.018 183s -0.557 -0.286 0.189 0.267 183s test-mul-mat0: iter:178/500 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.597 -0.982 0.508 0.585 183s -0.756 0.107 0.028 0.764 183s 183s x1: [4, 1, 1, 1] 183s -0.827 0.929 -0.990 -0.336 183s 183s y: [2, 1, 1, 1] 183s -1.118 0.440 183s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 3, 4, 1] 183s -0.646 0.712 0.052 0.782 183s 0.108 -0.905 0.607 -0.847 183s -0.017 0.100 0.842 -0.144 183s 183s x1: [4, 2, 4, 1] 183s 0.894 -0.683 0.530 0.800 183s -0.923 0.428 0.396 0.043 183s 183s y: [3, 2, 4, 1] 183s -0.411 0.358 0.248 183s 0.955 -0.284 0.385 183s testing: mul_mat, [1, 3, 4, 2] = [4, 1, 4, 2] * [4, 3, 4, 2] 183s x0: [4, 1, 4, 2] 183s -0.357 0.278 -0.645 -0.446 183s 183s x1: [4, 3, 4, 2] 183s -0.102 0.118 -0.200 0.562 183s -0.588 -0.647 0.138 0.336 183s -0.033 -0.268 -0.083 -0.595 183s 183s y: [1, 3, 4, 2] 183s -0.053 183s -0.209 183s 0.256 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.770 0.512 -0.832 -0.318 183s -0.617 0.252 0.347 -0.112 183s 0.682 0.639 0.440 0.803 183s 183s x1: [4, 1, 1, 1] 183s 0.027 0.260 -0.378 0.473 183s 183s y: [3, 1, 1, 1] 183s 0.277 -0.135 0.398 183s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.302 -0.555 0.781 183s 0.825 -0.645 0.198 183s 183s x1: [3, 4, 4, 1] 183s 0.009 0.333 -0.972 183s -0.882 0.632 -0.876 183s 0.627 -0.902 -0.130 183s -0.303 0.524 -0.381 183s 183s y: [2, 4, 4, 1] 183s -0.947 -0.400 183s -0.769 -1.309 183s 0.210 1.073 183s -0.497 -0.663 183s testing: mul_mat, [3, 3, 4, 2] = [2, 3, 4, 2] * [2, 3, 4, 2] 183s x0: [2, 3, 4, 2] 183s 0.707 -0.761 183s 0.069 0.436 183s 0.627 0.575 183s 183s x1: [2, 3, 4, 2] 183s 0.926 -0.866 183s 0.361 0.752 183s -0.144 -0.258 183s 183s y: [3, 3, 4, 2] 183s 1.314 -0.314 0.083 183s -0.317 0.353 0.658 183s 0.094 -0.123 -0.239 183s test-mul-mat0: iter:179/500 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.270 -0.303 0.926 -0.082 183s 183s x1: [4, 3, 1, 1] 183s 0.413 0.229 -0.892 0.128 183s -0.647 0.215 -0.983 0.261 183s -0.483 0.044 0.876 -0.111 183s 183s y: [1, 3, 1, 1] 183s -1.017 183s -0.823 183s 0.938 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.664 0.180 -0.054 -0.266 183s -0.730 -0.968 0.353 -0.476 183s 183s x1: [4, 1, 1, 1] 183s -0.505 0.695 -0.848 -0.420 183s 183s y: [2, 1, 1, 1] 183s 0.618 -0.403 183s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 183s x0: [4, 4, 1, 2] 183s -0.107 0.205 -0.727 0.045 183s 0.785 0.516 0.381 -0.035 183s -0.538 -0.885 0.235 -0.506 183s 0.467 0.760 0.778 -0.767 183s 183s x1: [4, 2, 1, 2] 183s 0.284 -0.235 0.754 0.392 183s 0.893 -0.894 -0.393 0.910 183s 183s y: [4, 2, 1, 2] 183s -0.609 0.375 0.034 0.240 183s 0.048 0.058 -0.242 -1.266 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.464 0.707 0.687 0.499 183s 183s x1: [4, 4, 1, 1] 183s -0.171 -0.509 -0.857 -0.386 183s -0.993 0.524 0.579 -0.531 183s 0.638 -0.186 -0.037 0.106 183s -0.426 -0.259 0.338 -0.913 183s 183s y: [1, 4, 1, 1] 183s -1.221 183s 0.043 183s 0.192 183s -0.604 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.872 183s 0.131 183s 183s x1: [1, 4, 1, 1] 183s -0.305 183s -0.413 183s -0.512 183s -0.074 183s 183s y: [2, 4, 1, 1] 183s 0.266 -0.040 183s 0.360 -0.054 183s 0.447 -0.067 183s 0.065 -0.010 183s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 183s x0: [2, 4, 1, 2] 183s -0.628 -0.854 183s -0.433 0.312 183s -0.157 0.534 183s 0.896 -0.040 183s 183s x1: [2, 1, 1, 2] 183s 0.124 -0.654 183s 183s y: [4, 1, 1, 2] 183s 0.481 -0.258 -0.369 0.137 183s test-mul-mat0: iter:180/500 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.902 -0.690 183s 0.779 -0.261 183s 183s x1: [2, 3, 1, 1] 183s 0.201 0.429 183s -0.050 -0.298 183s 0.557 -0.919 183s 183s y: [2, 3, 1, 1] 183s -0.114 0.045 183s 0.160 0.039 183s 1.136 0.673 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.247 -0.014 183s -0.814 -0.269 183s 183s x1: [2, 2, 2, 1] 183s -0.318 -0.654 183s 0.582 -0.421 183s 183s y: [2, 2, 2, 1] 183s 0.088 0.434 183s -0.138 -0.360 183s testing: mul_mat, [4, 2, 2, 2] = [2, 4, 2, 2] * [2, 2, 2, 2] 183s x0: [2,ng 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot. 4, 2, 2] 183s -0.438 -0.928 183s 0.216 -0.325 183s 0.524 -0.614 183s 0.428 -0.490 183s 183s x1: [2, 2, 2, 2] 183s 0.728 0.114 183s 0.960 -0.071 183s 183s y: [4, 2, 2, 2] 183s -0.425 0.120 0.311 0.255 183s -0.355 0.231 0.547 0.446 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.468 -0.574 183s 0.013 -0.199 183s 0.355 0.536 183s -0.478 -0.231 183s 183s x1: [2, 4, 1, 1] 183s -0.683 -0.632 183s 0.435 0.841 183s -0.246 -0.137 183s -0.649 0.327 183s 183s y: [4, 4, 1, 1] 183s 0.043 0.118 -0.581 0.473 183s -0.279 -0.162 0.605 -0.403 183s -0.037 0.024 -0.160 0.149 183s -0.491 -0.073 -0.055 0.235 183s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 3, 2, 1] 183s 0.240 0.590 0.058 0.580 183s -0.932 0.394 -0.593 -0.167 183s 0.849 -0.128 -0.774 0.027 183s 183s x1: [4, 2, 2, 1] 183s -0.211 0.460 -0.258 -0.123 183s -0.625 -0.829 -0.035 -0.409 183s 183s y: [3, 2, 2, 1] 183s 0.135 0.552 -0.042 183s -0.878 0.344 -0.408 183s testing: mul_mat, [3, 4, 2, 2] = [3, 3, 2, 2] * [3, 4, 2, 2] 183s x0: [3, 3, 2, 2] 183s 0.536 -0.806 0.544 183s 0.650 -0.359 -0.416 183s 0.536 0.896 -0.230 183s 183s x1: [3, 4, 2, 2] 183s -0.668 -0.568 -0.602 183s 0.883 0.831 0.712 183s -0.131 0.070 0.780 183s -0.282 -0.339 0.174 183s 183s y: [3, 4, 2, 2] 183s -0.227 0.020 -0.729 183s 0.190 -0.020 1.055 183s 0.297 -0.435 -0.186 183s 0.218 -0.134 -0.495 183s test-mul-mat0: iter:181/500 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.512 0.934 -0.838 183s 183s x1: [3, 2, 1, 1] 183s -0.160 -0.962 -0.861 183s 0.852 0.925 0.793 183s 183s y: [1, 2, 1, 1] 183s -0.259 183s 0.636 183s testing: mul_mat, [4, 1, 3, 1] = [3, 4, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 4, 3, 1] 183s -0.492 0.627 0.000 183s 0.252 -0.896 0.745 183s -0.362 -0.523 -0.967 183s -0.393 0.620 -0.126 183s 183s x1: [3, 1, 3, 1] 183s 0.668 0.635 -0.048 183s 183s y: [4, 1, 3, 1] 183s 0.069 -0.436 -0.528 0.138 183s testing: mul_mat, [4, 4, 3, 3] = [3, 4, 3, 3] * [3, 4, 3, 3] 183s x0: [3, 4, 3, 3] 183s -0.691 0.150 0.395 183s 0.864 0.751 -0.827 183s -0.611 0.013 0.450 183s 0.937 0.565 -0.550 183s 183s x1: [3, 4, 3, 3] 183s 0.493 -0.511 0.722 183s 0.526 0.096 0.342 183s -0.600 0.742 0.102 183s 0.126 -0.688 -0.346 183s 183s y: [4, 4, 3, 3] 183s -0.132 -0.555 0.017 -0.224 183s -0.214 0.244 -0.166 0.359 183s 0.566 -0.046 0.422 -0.199 183s -0.327 -0.122 -0.241 -0.080 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.745 0.964 0.921 183s -0.698 -0.621 -0.945 183s -0.699 -0.367 0.020 183s 183s x1: [3, 4, 1, 1] 183s -0.493 0.845 0.958 183s 0.865 -0.068 -0.839 183s 0.722 -0.754 0.633 183s 0.974 -0.770 0.059 183s 183s y: [3, 4, 1, 1] 183s 1.329 -1.086 0.054 183s -0.194 0.231 -0.597 183s 0.395 -0.634 -0.216 183s 0.038 -0.257 -0.398 183s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.493 1.000 0.577 183s 183s x1: [3, 3, 3, 1] 183s 0.042 0.110 0.912 183s -0.312 -0.115 -0.791 183s -0.470 -0.390 -0.547 183s 183s y: [1, 3, 3, 1] 183s 0.656 183s -0.725 183s -0.937 183s testing: mul_mat, [3, 3, 3, 3] = [1, 3, 3, 3] * [1, 3, 3, 3] 183s x0: [1, 3, 3, 3] 183s -0.626 183s 0.214 183s 0.782 183s 183s x1: [1, 3, 3, 3] 183s 0.753 183s -0.395 183s -0.850 183s 183s y: [3, 3, 3, 3] 183s -0.471 0.161 0.589 183s 0.247 -0.084 -0.309 183s 0.532 -0.181 -0.665 183s test-mul-mat0: iter:182/500 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.738 0.214 0.079 -0.244 183s 0.823 -0.595 -0.618 -0.249 183s 183s x1: [4, 1, 1, 1] 183s -0.830 -0.618 -0.671 -0.703 183s 183s y: [2, 1, 1, 1] 183s 0.600 0.274 183s testing: mul_mat, [2, 2, 4, 1] = [4, 2, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.380 -0.359 -0.316 -0.641 183s -0.294 0.324 -0.815 -0.203 183s 183s x1: [4, 2, 4, 1] 183s 0.812 0.346 -0.405 0.163 183s -0.090 -0.021 -0.935 -0.360 183s 183s y: [2, 2, 4, 1] 183s 0.208 0.171 183s 0.500 0.855 183s testing: mul_mat, [1, 2, 4, 2] = [4, 1, 4, 2] * [4, 2, 4, 2] 183s x0: [4, 1, 4, 2] 183s -0.264 -0.697 -0.970 0.587 183s 183s x1: [4, 2, 4, 2] 183s 0.036 0.983 -0.464 0.741 183s 0.307 -0.280 -0.461 0.110 183s 183s y: [1, 2, 4, 2] 183s 0.191 183s 0.626 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.136 -0.748 -0.304 0.816 183s -0.355 0.058 -0.608 -0.914 183s 0.992 -0.764 -0.626 0.442 183s -0.295 0.837 0.070 0.158 183s 183s x1: [4, 1, 1, 1] 183s 0.250 -0.122 -0.374 -0.965 183s 183s y: [4, 1, 1, 1] 183s -0.548 1.014 0.149 -0.354 183s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.444 -0.865 -0.140 0.885 183s 183s x1: [4, 4, 4, 1] 183s -0.887 -0.406 0.866 0.993 183s -0.389 0.565 -0.704 0.559 183s -0.586 0.335 -0.192 png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot0.292 183s 0.961 -0.156 0.677 0.097 183s 183s y: [1, 4, 4, 1] 183s 0.714 183s -0.068 183s -0.265 183s 0.553 183s testing: mul_mat, [4, 4, 4, 2] = [1, 4, 4, 2] * [1, 4, 4, 2] 183s x0: [1, 4, 4, 2] 183s 0.859 183s 0.328 183s -0.210 183s -0.496 183s 183s x1: [1, 4, 4, 2] 183s -0.884 183s 0.369 183s 0.540 183s 0.276 183s 183s y: [4, 4, 4, 2] 183s -0.760 -0.290 0.185 0.439 183s 0.318 0.121 -0.078 -0.183 183s 0.464 0.177 -0.113 -0.268 183s 0.237 0.091 -0.058 -0.137 183s test-mul-mat0: iter:183/500 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.082 0.468 183s -0.095 0.604 183s 183s x1: [2, 1, 1, 1] 183s 0.955 -0.778 183s 183s y: [2, 1, 1, 1] 183s -0.286 -0.560 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.694 0.772 183s 0.729 0.346 183s 183s x1: [2, 2, 2, 1] 183s 0.486 0.278 183s 0.657 0.136 183s 183s y: [2, 2, 2, 1] 183s -0.123 0.450 183s -0.351 0.526 183s testing: mul_mat, [2, 2, 2, 3] = [2, 2, 2, 3] * [2, 2, 2, 3] 183s x0: [2, 2, 2, 3] 183s 0.369 0.610 183s 0.698 0.561 183s 183s x1: [2, 2, 2, 3] 183s -0.292 -0.162 183s 0.396 -0.447 183s 183s y: [2, 2, 2, 3] 183s -0.207 -0.295 183s -0.127 0.025 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.505 -0.895 183s 0.734 0.432 183s -0.445 -0.883 183s 183s x1: [2, 2, 1, 1] 183s 0.738 0.545 183s -0.770 -0.150 183s 183s y: [3, 2, 1, 1] 183s -0.861 0.777 -0.810 183s 0.524 -0.630 0.475 183s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 2, 2, 1] 183s 0.232 0.585 -0.484 183s 0.707 -0.929 -0.309 183s 183s x1: [3, 2, 2, 1] 183s -0.886 0.304 -0.614 183s -0.038 -0.428 0.209 183s 183s y: [2, 2, 2, 1] 183s 0.269 -0.719 183s -0.360 0.306 183s testing: mul_mat, [2, 3, 2, 3] = [2, 2, 2, 3] * [2, 3, 2, 3] 183s x0: [2, 2, 2, 3] 183s 0.537 -0.737 183s -0.892 0.784 183s 183s x1: [2, 3, 2, 3] 183s 0.066 -0.162 183s 0.978 0.171 183s -0.730 -0.905 183s 183s y: [2, 3, 2, 3] 183s 0.155 -0.186 183s 0.400 -0.739 183s 0.275 -0.058 183s test-mul-mat0: iter:184/500 183s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.306 0.516 0.480 -0.859 183s -0.326 0.112 -0.252 -0.137 183s -0.706 -0.287 -0.848 0.544 183s 183s x1: [4, 2, 1, 1] 183s -0.921 0.793 0.476 0.617 183s 0.901 0.739 0.401 -0.845 183s 183s y: [3, 2, 1, 1] 183s -0.174 0.184 0.353 183s 1.575 -0.196 -1.648 183s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 4, 4, 1] 183s -0.244 0.501 -0.160 0.871 183s 0.502 -0.751 -0.902 -0.749 183s -0.479 -0.381 0.713 -0.056 183s -0.316 0.901 -0.749 0.863 183s 183s x1: [4, 3, 4, 1] 183s -0.274 -0.116 0.114 0.025 183s 0.630 -0.890 0.033 0.458 183s 0.370 -0.273 0.537 0.163 183s 183s y: [4, 3, 4, 1] 183s 0.013 -0.172 0.255 -0.082 183s -0.206 0.612 0.034 -0.631 183s -0.170 -0.216 0.300 -0.623 183s testing: mul_mat, [3, 4, 4, 3] = [4, 3, 4, 3] * [4, 4, 4, 3] 183s x0: [4, 3, 4, 3] 183s 0.956 -0.864 0.744 0.831 183s 0.311 -0.373 0.859 0.893 183s -0.424 -0.951 0.105 -0.428 183s 183s x1: [4, 4, 4, 3] 183s 0.727 -0.283 0.556 0.441 183s -0.924 0.053 0.676 -0.946 183s -0.738 -0.146 -0.284 0.372 183s -0.716 0.241 0.597 -0.097 183s 183s y: [3, 4, 4, 3] 183s 1.721 1.204 -0.169 183s -1.213 -0.572 0.818 183s -0.481 -0.087 0.263 183s -0.530 0.113 0.179 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.231 0.225 0.086 -0.387 183s 0.041 0.479 -0.092 -0.202 183s -0.722 -0.278 0.913 0.625 183s 0.206 0.850 0.213 -0.561 183s 183s x1: [4, 3, 1, 1] 183s 0.072 -0.770 0.507 -0.669 183s 0.603 0.408 -0.797 0.406 183s 0.274 0.608 -0.015 -0.695 183s 183s y: [4, 3, 1, 1] 183s 0.145 -0.277 0.207 -0.157 183s 0.006 0.211 -1.022 0.073 183s 0.468 0.445 -0.815 0.961 183s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.729 0.541 -0.062 -0.762 183s 183s x1: [4, 4, 4, 1] 183s -0.462 0.511 0.511 -0.232 183s 0.018 0.842 -0.629 -0.574 183s -0.954 0.777 0.700 0.654 183s -0.238 -0.995 0.574 0.993 183s 183s y: [1, 4, 4, 1] 183s 0.085 183s 0.945 183s -0.817 183s -1.504 183s testing: mul_mat, [2, 4, 4, 3] = [1, 2, 4, 3] * [1, 4, 4, 3] 183s x0: [1, 2, 4, 3] 183s -0.851 183s 0.828 183s 183s x1: [1, 4, 4, 3] 183s 0.767 183s 0.712 183s 0.014 183s 0.121 183s 183s y: [2, 4, 4, 3] 183s -0.653 0.635 183s -0.606 0.590 183s -0.012 0.011 183s -0.103 0.100 183s test-mul-mat0: iter:185/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.050 -0.703 183s 183s x1: [2, 4, 1, 1] 183s -0.403 0.467 183s 0.660 -0.254 183s 0.296 0.158 183s 0.707 -0.619 183s 183s y: [1, 4, 1, 1] 183s -0.349 183s 0.211 183s -0.096 183s 0.470 183s testing: mul_mat, [3, 1, 2, 1] = [2, 3, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 3, 2, 1] 183s 0.993 -0.512 183s -0.367 -0.340 183s 0.733 -.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.do0.446 183s 183s x1: [2, 1, 2, 1] 183s -0.905 0.390 183s 183s y: [3, 1, 2, 1] 183s -1.099 0.199 -0.838 183s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.962 0.143 183s 0.239 0.617 183s -0.642 -0.181 183s 0.610 -0.154 183s 183s x1: [2, 3, 2, 1] 183s 0.748 -0.404 183s -0.185 -0.654 183s -0.936 -0.525 183s 183s y: [4, 3, 2, 1] 183s -0.778 -0.070 -0.408 0.519 183s 0.085 -0.448 0.237 -0.012 183s 0.826 -0.548 0.696 -0.490 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.534 0.358 183s 0.748 -0.793 183s 183s x1: [2, 4, 1, 1] 183s -0.553 -0.174 183s 0.122 0.043 183s 0.641 0.468 183s 0.107 -0.885 183s 183s y: [2, 4, 1, 1] 183s 0.233 -0.276 183s -0.050 0.057 183s -0.174 0.108 183s -0.374 0.781 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.142 0.742 183s -0.283 0.721 183s 183s x1: [2, 2, 2, 1] 183s -0.204 0.207 183s -0.755 0.939 183s 183s y: [2, 2, 2, 1] 183s 0.124 0.207 183s 0.589 0.890 183s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 1, 2, 1] 183s -0.375 0.870 183s 183s x1: [2, 2, 2, 1] 183s 0.476 0.254 183s 0.330 -0.402 183s 183s y: [1, 2, 2, 1] 183s 0.043 183s -0.474 183s test-mul-mat0: iter:186/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.574 0.087 0.145 0.639 183s -0.509 -0.769 -0.736 -0.639 183s 0.065 -0.753 -0.562 -0.305 183s 0.436 -0.239 -0.737 0.071 183s 183s x1: [4, 4, 1, 1] 183s 0.568 0.635 -0.815 0.865 183s -0.100 0.657 0.007 0.617 183s 0.399 -0.272 -0.660 0.501 183s 0.535 -0.155 -0.023 -0.210 183s 183s y: [4, 4, 1, 1] 183s 0.816 -0.731 -0.248 0.758 183s 0.395 -0.853 -0.694 -0.162 183s 0.429 0.171 0.449 0.761 183s 0.156 -0.002 0.229 0.273 183s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.402 -0.902 -0.662 -0.447 183s -0.776 0.728 -0.245 -0.213 183s 0.813 -0.607 -0.888 -0.839 183s 0.102 -0.618 -0.364 0.782 183s 183s x1: [4, 4, 3, 1] 183s -0.054 -0.872 0.971 -0.398 183s 0.135 0.588 -0.999 0.863 183s 0.929 0.502 0.398 -0.226 183s -0.522 -0.812 -0.052 -0.948 183s 183s y: [4, 4, 3, 1] 183s 0.343 -0.747 -0.043 -0.131 183s -0.309 0.385 -0.085 0.689 183s -0.988 -0.405 0.287 -0.537 183s 1.400 0.027 0.910 -0.274 183s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.145 0.545 -0.595 -0.647 183s -0.449 -0.480 0.055 -0.249 183s -0.698 0.344 -0.219 0.532 183s -0.525 -0.113 -0.738 -0.424 183s 183s x1: [4, 4, 3, 1] 183s -0.923 0.500 -0.029 0.797 183s 0.749 0.476 -0.613 -0.031 183s -0.165 -0.963 -0.891 0.100 183s -0.980 0.262 0.963 -0.899 183s 183s y: [4, 4, 3, 1] 183s -0.092 -0.025 1.247 0.113 183s 0.535 -0.591 -0.241 0.018 183s -0.035 0.462 0.033 0.811 183s 0.293 0.592 0.085 0.155 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.622 -0.150 -0.334 -0.043 183s -0.288 0.693 0.454 -0.897 183s 0.770 0.404 -0.430 0.998 183s 183s x1: [4, 4, 1, 1] 183s -0.831 0.021 -0.478 -0.643 183s 0.587 0.932 -0.897 0.665 183s 0.690 -0.334 0.218 0.898 183s -0.199 0.625 -0.038 -0.519 183s 183s y: [3, 4, 1, 1] 183s 0.701 0.613 -1.068 183s -0.234 -0.528 1.878 183s -0.490 -1.137 1.199 183s 0.065 0.938 -0.401 183s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.500 -0.404 0.911 183s -0.858 0.829 0.414 183s -0.565 0.727 0.821 183s 183s x1: [3, 4, 3, 1] 183s 0.824 0.543 -0.834 183s -0.155 -0.935 -0.476 183s -0.568 0.997 -0.374 183s -0.903 0.687 0.292 183s 183s y: [3, 4, 3, 1] 183s -1.391 -0.603 -0.755 183s 0.021 -0.839 -0.983 183s -0.459 1.159 0.739 183s 0.441 1.466 1.250 183s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.610 -0.583 0.866 183s -0.925 0.842 -0.728 183s 183s x1: [3, 3, 3, 1] 183s 0.452 0.792 0.855 183s 0.261 -0.709 0.996 183s 0.696 -0.113 0.825 183s 183s y: [2, 3, 3, 1] 183s 0.003 -0.374 183s 1.117 -1.564 183s 0.356 -1.340 183s test-mul-mat0: iter:187/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.152 -0.849 -0.293 0.245 183s -0.626 -0.450 -0.424 0.841 183s -0.959 0.108 -0.053 0.408 183s 0.725 0.375 -0.975 0.089 183s 183s x1: [4, 4, 1, 1] 183s -0.383 0.364 -0.590 0.793 183s 0.213 0.487 -0.983 0.563 183s -0.806 -0.593 0.638 -0.389 183s -0.751 0.504 -0.117 0.352 183s 183s y: [4, 4, 1, 1] 183s 0.000 0.993 0.761 0.504 183s 0.045 0.538 0.131 1.345 183s 0.099 0.174 0.516 -1.463 183s -0.421 0.589 0.924 -0.211 183s testing: mul_mat, [4, 4, 2, 1] = [4, 4, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 4, 2, 1] 183s -0.843 0.589 0.055 -0.978 183s 0.056 0.297 0.525 0.133 183s 0.788 -0t.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.d.467 -0.905 0.632 183s 0.202 -0.793 0.627 -0.422 183s 183s x1: [4, 4, 2, 1] 183s 0.785 0.819 -0.698 0.272 183s 0.835 0.865 0.466 -0.758 183s 0.503 -0.922 -0.509 0.007 183s -0.039 0.844 -0.331 -0.887 183s 183s y: [4, 4, 2, 1] 183s -0.484 -0.044 1.041 -1.043 183s 0.572 0.447 -0.647 0.095 183s -1.002 -0.512 1.293 0.510 183s 1.379 -0.044 -0.686 -0.509 183s testing: mul_mat, [3, 4, 2, 4] = [4, 3, 2, 4] * [4, 4, 2, 4] 183s x0: [4, 3, 2, 4] 183s -0.243 0.217 0.111 -0.974 183s 0.034 -0.333 -0.674 -0.636 183s 0.701 -0.263 0.621 -0.165 183s 183s x1: [4, 4, 2, 4] 183s 0.800 0.521 0.343 -0.144 183s -0.182 -0.132 0.989 -0.394 183s 0.401 -0.916 -0.762 -0.398 183s -0.709 0.865 0.180 0.856 183s 183s y: [3, 4, 2, 4] 183s 0.097 -0.285 0.660 183s 0.509 -0.379 0.586 183s 0.006 1.085 0.114 183s -0.454 -0.977 -0.754 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.203 -0.610 -0.716 -0.842 183s 0.517 0.710 0.132 0.296 183s -0.128 -0.907 0.897 -0.543 183s 183s x1: [4, 3, 1, 1] 183s 0.923 -0.692 0.492 0.759 183s -0.378 -0.607 0.743 0.045 183s -0.949 0.896 -0.854 0.227 183s 183s y: [3, 3, 1, 1] 183s -0.757 0.275 0.539 183s -0.123 -0.515 1.241 183s 0.067 0.101 -1.581 183s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 4, 2, 1] 183s -0.852 0.549 0.980 183s -0.747 0.764 -0.568 183s 0.307 -0.379 -0.821 183s -0.693 0.624 -0.290 183s 183s x1: [3, 4, 2, 1] 183s -0.686 -0.926 -0.784 183s 0.811 -0.444 -0.269 183s 0.734 -0.136 -0.777 183s 0.492 0.486 -0.385 183s 183s y: [4, 4, 2, 1] 183s -0.692 0.251 0.784 0.125 183s -1.199 -0.792 0.639 -0.761 183s -1.461 -0.210 0.915 -0.368 183s -0.529 0.222 0.282 0.073 183s testing: mul_mat, [1, 3, 2, 4] = [4, 1, 2, 4] * [4, 3, 2, 4] 183s x0: [4, 1, 2, 4] 183s 0.664 0.079 -0.537 -0.324 183s 183s x1: [4, 3, 2, 4] 183s -0.531 0.212 0.518 0.786 183s 0.076 -0.361 0.196 0.225 183s -0.109 -0.497 0.531 -0.559 183s 183s y: [1, 3, 2, 4] 183s -0.868 183s -0.156 183s -0.215 183s test-mul-mat0: iter:188/500 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.851 183s -0.710 183s 0.101 183s 183s x1: [1, 2, 1, 1] 183s -0.374 183s -0.544 183s 183s y: [3, 2, 1, 1] 183s -0.318 0.265 -0.038 183s -0.463 0.386 -0.055 183s testing: mul_mat, [3, 3, 2, 1] = [1, 3, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.923 183s -0.964 183s -0.292 183s 183s x1: [1, 3, 2, 1] 183s -0.977 183s 0.007 183s -0.926 183s 183s y: [3, 3, 2, 1] 183s 0.902 0.942 0.285 183s -0.007 -0.007 -0.002 183s 0.854 0.893 0.270 183s testing: mul_mat, [3, 3, 2, 4] = [1, 3, 2, 4] * [1, 3, 2, 4] 183s x0: [1, 3, 2, 4] 183s 0.093 183s 0.360 183s -0.995 183s 183s x1: [1, 3, 2, 4] 183s -0.790 183s 0.208 183s 0.846 183s 183s y: [3, 3, 2, 4] 183s -0.073 -0.285 0.786 183s 0.019 0.075 -0.206 183s 0.078 0.305 -0.842 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.821 183s -0.116 183s 0.918 183s 0.271 183s 183s x1: [1, 3, 1, 1] 183s 0.418 183s 0.164 183s 0.559 183s 183s y: [4, 3, 1, 1] 183s -0.343 -0.049 0.384 0.113 183s -0.134 -0.019 0.150 0.044 183s -0.459 -0.065 0.514 0.152 183s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.622 -0.455 -0.778 0.866 183s -0.355 0.068 -0.599 -0.615 183s -0.788 0.588 -0.597 -0.492 183s 183s x1: [4, 1, 2, 1] 183s -0.756 0.924 0.202 -0.175 183s 183s y: [3, 1, 2, 1] 183s -0.258 0.318 1.105 183s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 183s x0: [3, 2, 2, 4] 183s -0.709 0.818 0.914 183s 0.153 0.933 -0.469 183s 183s x1: [3, 4, 2, 4] 183s 0.581 0.631 0.781 183s -0.078 -0.754 -0.513 183s -0.430 -0.376 0.132 183s -0.218 0.169 -0.800 183s 183s y: [2, 4, 2, 4] 183s 0.819 0.312 183s -1.031 -0.474 183s 0.118 -0.479 183s -0.439 0.499 183s test-mul-mat0: iter:189/500 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.780 0.429 0.476 183s 0.029 0.137 0.682 183s -0.652 0.497 0.610 183s 183s x1: [3, 4, 1, 1] 183s -0.145 -0.949 -0.437 183s 0.911 -0.623 0.926 183s -0.540 -0.994 0.686 183s 0.410 0.927 -0.400 183s 183s y: [3, 4, 1, 1] 183s -0.728 -0.433 -0.644 183s 0.884 0.572 -0.340 183s -0.520 0.316 0.276 183s 0.527 -0.134 -0.050 183s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.326 0.014 -0.967 183s 0.134 -0.542 0.982 183s 0.504 -0.723 0.857 183s 183s x1: [3, 3, 4, 1] 183s -0.116 -0.927 0.789 183s 0.275 0.325 0.672 183s 0.605 0.393 0.386 183s 183s y: [3, 3, 4, 1] 183s -0.738 1.262 1.287 183s -0.735 0.521 0.480 183s -0.565 0.247 0.352 183s testing: mul_mat, [2, 3, 4, 3] = [3, 2, 4, 3] * [3, 3, 4, 3] 183s x0: [3, 2, 4, 3] 183s -0.519 -0.413 0.218 183s -0.204 -0.621 -0.166 183s 183s x1: [3, 3, 4, 3] 183s 0.285 0.033 0.786 183s 0.141 0.368 0.8ot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward89 183s -0.441 0.943 0.745 183s 183s y: [2, 3, 4, 3] 183s 0.010 -0.209 183s -0.032 -0.405 183s 0.002 -0.620 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.070 -0.746 0.554 183s 183s x1: [3, 2, 1, 1] 183s 0.726 -0.059 0.967 183s -0.764 0.541 0.030 183s 183s y: [1, 2, 1, 1] 183s 0.630 183s -0.441 183s testing: mul_mat, [2, 3, 4, 1] = [1, 2, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.274 183s -0.144 183s 183s x1: [1, 3, 4, 1] 183s 0.426 183s -0.129 183s 0.791 183s 183s y: [2, 3, 4, 1] 183s 0.117 -0.061 183s -0.035 0.019 183s 0.216 -0.114 183s testing: mul_mat, [4, 1, 4, 3] = [2, 4, 4, 3] * [2, 1, 4, 3] 183s x0: [2, 4, 4, 3] 183s -0.564 -0.898 183s 0.642 0.118 183s -0.935 0.216 183s 0.963 0.504 183s 183s x1: [2, 1, 4, 3] 183s 0.253 0.401 183s 183s y: [4, 1, 4, 3] 183s -0.503 0.210 -0.150 0.446 183s test-mul-mat0: iter:190/500 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.111 -0.991 183s -0.226 0.852 183s 0.461 0.977 183s 183s x1: [2, 3, 1, 1] 183s 0.794 -0.773 183s -0.279 -0.166 183s 0.570 0.768 183s 183s y: [3, 3, 1, 1] 183s 0.678 -0.838 -0.389 183s 0.195 -0.078 -0.291 183s -0.824 0.526 1.013 183s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 4, 3, 1] 183s -0.020 0.877 183s 0.636 -0.450 183s 0.645 -0.836 183s 0.439 0.654 183s 183s x1: [2, 3, 3, 1] 183s -0.151 0.599 183s -0.064 -0.077 183s 0.198 -0.966 183s 183s y: [4, 3, 3, 1] 183s 0.528 -0.365 -0.598 0.326 183s -0.066 -0.006 0.023 -0.078 183s -0.851 0.560 0.935 -0.545 183s testing: mul_mat, [2, 4, 3, 1] = [2, 2, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.582 -0.267 183s -0.605 -0.936 183s 183s x1: [2, 4, 3, 1] 183s -0.459 0.091 183s -0.459 0.549 183s 0.843 0.698 183s -0.649 -0.177 183s 183s y: [2, 4, 3, 1] 183s -0.292 0.193 183s -0.413 -0.236 183s 0.305 -1.164 183s -0.331 0.558 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.148 0.381 183s -0.356 -0.977 183s 183s x1: [2, 2, 1, 1] 183s -0.495 0.195 183s 0.927 -0.508 183s 183s y: [2, 2, 1, 1] 183s 0.001 -0.014 183s -0.057 0.166 183s testing: mul_mat, [3, 2, 3, 1] = [2, 3, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.316 0.289 183s 0.507 -0.964 183s 0.845 -0.502 183s 183s x1: [2, 2, 3, 1] 183s 0.734 0.485 183s -0.987 0.739 183s 183s y: [3, 2, 3, 1] 183s -0.092 -0.096 0.377 183s 0.525 -1.212 -1.204 183s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.841 -0.586 0.637 183s -0.966 0.788 0.191 183s 183s x1: [3, 2, 3, 1] 183s -0.184 0.038 0.961 183s -0.445 -0.232 -0.321 183s 183s y: [2, 2, 3, 1] 183s 0.435 0.391 183s -0.443 0.186 183s test-mul-mat0: iter:191/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.315 0.776 -0.235 -0.506 183s -0.109 -0.926 0.857 0.953 183s -0.801 -0.742 -0.024 -0.121 183s 183s x1: [4, 3, 1, 1] 183s 0.400 0.799 0.540 -0.489 183s 0.389 -0.394 -0.091 0.014 183s -0.829 -0.250 0.048 -0.415 183s 183s y: [3, 3, 1, 1] 183s 0.614 -0.787 -0.868 183s -0.414 0.258 -0.018 183s 0.266 -0.033 0.898 183s testing: mul_mat, [2, 3, 4, 1] = [4, 2, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.425 0.929 0.889 0.849 183s -0.250 0.486 0.197 0.427 183s 183s x1: [4, 3, 4, 1] 183s 0.426 -0.323 0.710 -0.600 183s -0.607 -0.972 0.801 -0.809 183s 0.568 -0.689 0.580 -0.825 183s 183s y: [2, 3, 4, 1] 183s -0.359 -0.380 183s -0.619 -0.508 183s -1.067 -0.716 183s testing: mul_mat, [4, 2, 4, 3] = [4, 4, 4, 3] * [4, 2, 4, 3] 183s x0: [4, 4, 4, 3] 183s 0.056 -0.500 0.142 0.142 183s -0.778 0.981 -0.298 -0.313 183s -0.553 0.022 0.817 -0.705 183s 0.755 0.011 0.327 -0.304 183s 183s x1: [4, 2, 4, 3] 183s 0.006 -0.562 0.504 0.756 183s 0.924 -0.299 0.184 -0.039 183s 183s y: [4, 2, 4, 3] 183s 0.461 -0.944 -0.137 -0.066 183s 0.222 -1.054 -0.340 0.766 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.657 -0.710 -0.410 0.135 183s 0.841 -0.261 0.024 -0.895 183s -0.871 -0.116 0.592 0.711 183s 183s x1: [4, 4, 1, 1] 183s 0.147 0.865 0.181 -0.207 183s -0.441 -0.452 -0.948 0.417 183s 0.558 0.777 0.252 -0.839 183s 0.688 0.218 0.841 0.319 183s 183s y: [3, 4, 1, 1] 183s -0.813 0.087 -0.268 183s 1.055 -0.648 0.172 183s -1.135 1.022 -1.022 183s -0.908 0.256 0.101 183s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.404 0.439 -0.642 183s -0.079 0.151 0.972 183s 0.899 -0.033 0.058 183s 183s x1: [3, 4, 4, 1] 183s 0.001 -0.692 -0.142 183s -0.133 0.489 0.651 183s 0.426 -0.963 0.703 183s -0.158 0.595 0.481 183s 183s y: [3, 4, 4, 1] 183s -0.212 -0.243 0.016 183s -0.258 0.717 -0.098 183s -0.702 0.504 0.456 183s -0.111 0.569 -0.134 183s testing: mul_mat, [3, 3, 4, 3] = [3, 3, 4, 3] * [3, 3, 4, 3] 183s x0: [3, 3, 4, 3] 183s -.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backwar0.118 -0.599 -0.714 183s 0.283 0.953 -0.466 183s 0.478 -0.039 0.201 183s 183s x1: [3, 3, 4, 3] 183s -0.233 -0.067 0.816 183s 0.825 0.801 0.743 183s 0.079 -0.109 -0.373 183s 183s y: [3, 3, 4, 3] 183s -0.515 -0.511 0.056 183s -1.108 0.651 0.512 183s 0.322 0.093 -0.033 183s test-mul-mat0: iter:192/500 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.263 183s -0.579 183s -0.602 183s -0.085 183s 183s x1: [1, 4, 1, 1] 183s -0.549 183s 0.386 183s 0.891 183s -0.308 183s 183s y: [4, 4, 1, 1] 183s 0.144 0.318 0.331 0.047 183s -0.102 -0.224 -0.233 -0.033 183s -0.234 -0.516 -0.536 -0.076 183s 0.081 0.178 0.185 0.026 183s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 3, 3, 1] 183s -0.934 183s 0.650 183s -0.512 183s 183s x1: [1, 4, 3, 1] 183s -0.267 183s -0.546 183s -0.842 183s 0.609 183s 183s y: [3, 4, 3, 1] 183s 0.250 -0.174 0.137 183s 0.510 -0.355 0.279 183s 0.787 -0.548 0.431 183s -0.569 0.396 -0.312 183s testing: mul_mat, [4, 3, 3, 4] = [1, 4, 3, 4] * [1, 3, 3, 4] 183s x0: [1, 4, 3, 4] 183s 0.367 183s 0.459 183s -0.848 183s -0.545 183s 183s x1: [1, 3, 3, 4] 183s 0.243 183s 0.228 183s 0.447 183s 183s y: [4, 3, 3, 4] 183s 0.089 0.111 -0.206 -0.132 183s 0.083 0.105 -0.193 -0.124 183s 0.164 0.205 -0.379 -0.244 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.971 183s 183s x1: [1, 4, 1, 1] 183s -0.969 183s 0.744 183s -0.550 183s 0.166 183s 183s y: [1, 4, 1, 1] 183s -0.941 183s 0.722 183s -0.534 183s 0.161 183s testing: mul_mat, [1, 1, 3, 1] = [1, 1, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 1, 3, 1] 183s -0.495 183s 183s x1: [1, 1, 3, 1] 183s -0.307 183s 183s y: [1, 1, 3, 1] 183s 0.152 183s testing: mul_mat, [2, 1, 3, 4] = [1, 2, 3, 4] * [1, 1, 3, 4] 183s x0: [1, 2, 3, 4] 183s 0.149 183s 0.429 183s 183s x1: [1, 1, 3, 4] 183s -0.801 183s 183s y: [2, 1, 3, 4] 183s -0.120 -0.344 183s test-mul-mat0: iter:193/500 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.340 -0.178 0.519 -0.764 183s -0.025 -0.839 -0.702 0.221 183s -0.225 0.054 0.333 0.955 183s 183s x1: [4, 1, 1, 1] 183s 0.264 0.898 -0.525 -0.587 183s 183s y: [3, 1, 1, 1] 183s 0.106 -0.522 -0.746 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.157 -0.597 -0.289 0.318 183s 183s x1: [4, 3, 1, 1] 183s 0.650 0.950 0.311 0.575 183s -0.234 0.815 0.774 0.626 183s -0.103 -0.774 0.785 0.140 183s 183s y: [1, 3, 1, 1] 183s -0.576 183s -0.474 183s 0.295 183s testing: mul_mat, [2, 1, 1, 3] = [4, 2, 1, 3] * [4, 1, 1, 3] 183s x0: [4, 2, 1, 3] 183s -0.552 0.897 0.116 0.759 183s 0.472 0.882 0.574 0.246 183s 183s x1: [4, 1, 1, 3] 183s -0.184 -0.962 -0.342 0.638 183s 183s y: [2, 1, 1, 3] 183s -0.317 -0.975 183s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.198 0.080 -0.411 0.882 183s 0.010 -0.416 0.055 -0.334 183s 0.814 0.060 -0.468 -0.184 183s 183s x1: [4, 2, 1, 1] 183s 0.081 -0.691 -0.801 -0.085 183s -0.198 -0.555 -0.918 0.251 183s 183s y: [3, 2, 1, 1] 183s 0.215 0.273 0.415 183s 0.515 0.095 0.189 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.976 0.237 0.797 183s -0.454 -0.942 0.579 183s 183s x1: [3, 4, 1, 1] 183s -0.981 -0.320 0.812 183s 0.473 0.511 -0.155 183s 0.586 0.980 0.435 183s -0.408 0.061 0.744 183s 183s y: [2, 4, 1, 1] 183s -0.387 1.217 183s 0.459 -0.786 183s 1.150 -0.937 183s 0.209 0.558 183s testing: mul_mat, [3, 3, 1, 3] = [2, 3, 1, 3] * [2, 3, 1, 3] 183s x0: [2, 3, 1, 3] 183s 0.282 -0.657 183s 0.893 0.637 183s -0.522 0.269 183s 183s x1: [2, 3, 1, 3] 183s 0.256 -0.194 183s 0.394 -0.664 183s 0.390 -0.546 183s 183s y: [3, 3, 1, 3] 183s 0.199 0.105 -0.186 183s 0.547 -0.071 -0.384 183s 0.469 0.000 -0.351 183s test-mul-mat0: iter:194/500 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.173 -0.174 -0.208 -0.288 183s -0.892 -0.315 0.190 -0.549 183s 183s x1: [4, 1, 1, 1] 183s 0.732 -0.280 -0.472 0.524 183s 183s y: [2, 1, 1, 1] 183s -0.131 -0.942 183s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 2, 3, 1] 183s 0.223 0.414 -0.741 0.908 183s -0.397 -0.289 -0.769 0.062 183s 183s x1: [4, 2, 3, 1] 183s -0.677 -0.541 -0.229 -0.494 183s -0.036 0.148 -0.514 0.049 183s 183s y: [2, 2, 3, 1] 183s -0.654 0.571 183s 0.479 0.370 183s testing: mul_mat, [3, 2, 3, 4] = [4, 3, 3, 4] * [4, 2, 3, 4] 183s x0: [4, 3, 3, 4] 183s -0.911 -0.347 0.129 0.126 183s 0.327 0.623 -0.013 -0.441 183s 0.897 0.002 0.159 0.689 183s 183s x1: [4, 2, 3, 4] 183s 0.812 0.539 0.029 0.137 183s 0.970 0.731 0.288 0.193 183s 183s y: [3, 2, 3, 4] 183s -0.906 0.541 0.828 183s -1.076 0.684 1.050 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.402 -0.482 -0.021 0.773 183s -0.011 0.125 -0.945 -d.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backwa0.652 183s -0.122 -0.940 0.784 0.865 183s 183s x1: [4, 3, 1, 1] 183s -0.718 -0.957 -0.817 0.928 183s 0.262 -0.643 0.082 -0.321 183s 0.551 -0.877 0.264 0.472 183s 183s y: [3, 3, 1, 1] 183s 0.908 0.056 1.150 183s 0.165 0.048 0.359 183s 1.003 -0.673 1.372 183s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 1, 3, 1] 183s -0.520 -0.886 0.531 183s 183s x1: [3, 4, 3, 1] 183s 0.891 0.967 -0.960 183s 0.038 -0.689 0.402 183s 0.320 -0.646 0.586 183s 0.248 0.615 0.943 183s 183s y: [1, 4, 3, 1] 183s -1.829 183s 0.805 183s 0.717 183s -0.174 183s testing: mul_mat, [3, 3, 3, 4] = [1, 3, 3, 4] * [1, 3, 3, 4] 183s x0: [1, 3, 3, 4] 183s -0.579 183s -0.183 183s -0.532 183s 183s x1: [1, 3, 3, 4] 183s -0.489 183s 0.584 183s 0.934 183s 183s y: [3, 3, 3, 4] 183s 0.283 0.090 0.260 183s -0.338 -0.107 -0.311 183s -0.541 -0.171 -0.497 183s test-mul-mat0: iter:195/500 183s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.492 0.907 183s 0.412 0.317 183s 0.694 0.611 183s 183s x1: [2, 4, 1, 1] 183s -0.689 -0.784 183s -0.006 -0.508 183s 0.061 -0.019 183s -0.625 -0.618 183s 183s y: [3, 4, 1, 1] 183s -1.050 -0.533 -0.958 183s -0.464 -0.164 -0.315 183s 0.013 0.019 0.031 183s -0.868 -0.453 -0.811 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.804 0.351 183s 0.610 -0.126 183s 183s x1: [2, 3, 1, 1] 183s -0.183 -0.746 183s -0.942 0.926 183s -0.986 0.296 183s 183s y: [2, 3, 1, 1] 183s -0.115 -0.018 183s 1.083 -0.691 183s 0.896 -0.639 183s testing: mul_mat, [4, 2, 1, 3] = [2, 4, 1, 3] * [2, 2, 1, 3] 183s x0: [2, 4, 1, 3] 183s -0.089 0.431 183s 0.640 -0.677 183s -0.252 0.334 183s 0.935 0.565 183s 183s x1: [2, 2, 1, 3] 183s -0.295 0.257 183s -0.239 -0.927 183s 183s y: [4, 2, 1, 3] 183s 0.137 -0.362 0.160 -0.131 183s -0.379 0.475 -0.249 -0.747 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.572 -0.793 183s -0.728 0.753 183s -0.811 0.350 183s 0.762 -0.800 183s 183s x1: [2, 4, 1, 1] 183s 0.706 -0.996 183s -0.930 0.056 183s 0.038 -0.991 183s -0.211 0.949 183s 183s y: [4, 4, 1, 1] 183s 0.386 -1.264 -0.921 1.335 183s 0.487 0.719 0.775 -0.754 183s 0.764 -0.775 -0.378 0.822 183s -0.632 0.869 0.503 -0.921 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.798 -0.110 0.363 0.754 183s -0.143 0.749 -0.404 0.433 183s 183s x1: [4, 2, 1, 1] 183s 0.244 -0.048 0.488 0.270 183s -0.250 -0.872 -0.095 0.423 183s 183s y: [2, 2, 1, 1] 183s 0.581 -0.151 183s 0.181 -0.397 183s testing: mul_mat, [2, 4, 1, 3] = [2, 2, 1, 3] * [2, 4, 1, 3] 183s x0: [2, 2, 1, 3] 183s 0.399 0.326 183s 0.791 0.995 183s 183s x1: [2, 4, 1, 3] 183s 0.653 -0.208 183s 0.442 -0.559 183s -0.259 -0.118 183s -0.130 0.014 183s 183s y: [2, 4, 1, 3] 183s 0.192 0.309 183s -0.006 -0.206 183s -0.142 -0.322 183s -0.047 -0.089 183s test-mul-mat0: iter:196/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.323 -0.371 183s 0.294 0.186 183s 183s x1: [2, 4, 1, 1] 183s 0.312 -0.022 183s 0.572 -0.268 183s 0.046 -0.495 183s 0.641 -0.615 183s 183s y: [2, 4, 1, 1] 183s -0.093 0.088 183s -0.085 0.119 183s 0.169 -0.078 183s 0.021 0.074 183s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.658 0.544 183s -0.683 0.419 183s -0.373 0.276 183s 0.510 -0.219 183s 183s x1: [2, 2, 2, 1] 183s -0.330 0.257 183s -0.415 -0.539 183s 183s y: [4, 2, 2, 1] 183s 0.357 0.333 0.194 -0.225 183s -0.020 0.058 0.006 -0.094 183s testing: mul_mat, [1, 4, 2, 2] = [2, 1, 2, 2] * [2, 4, 2, 2] 183s x0: [2, 1, 2, 2] 183s 0.666 -0.866 183s 183s x1: [2, 4, 2, 2] 183s 0.190 0.133 183s 0.871 0.818 183s -0.573 0.057 183s -0.511 0.685 183s 183s y: [1, 4, 2, 2] 183s 0.011 183s -0.129 183s -0.430 183s -0.933 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.415 0.798 183s 0.033 -0.461 183s 0.824 -0.649 183s -0.430 0.839 183s 183s x1: [2, 1, 1, 1] 183s 0.804 -0.415 183s 183s y: [4, 1, 1, 1] 183s 0.002 0.218 0.931 -0.694 183s testing: mul_mat, [2, 2, 2, 1] = [4, 2, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 2, 2, 1] 183s 0.192 0.005 0.544 -0.590 183s -0.041 -0.005 0.532 -0.424 183s 183s x1: [4, 2, 2, 1] 183s -0.235 -0.086 -0.254 0.812 183s 0.235 -0.908 -0.588 -0.639 183s 183s y: [2, 2, 2, 1] 183s -0.663 -0.469 183s 0.098 -0.047 183s testing: mul_mat, [1, 4, 2, 2] = [2, 1, 2, 2] * [2, 4, 2, 2] 183s x0: [2, 1, 2, 2] 183s 0.526 0.234 183s 183s x1: [2, 4, 2, 2] 183s -0.041 -0.528 183s 0.479 0.051 183s -0.116 0.840 183s 0.433 0.091 183s 183s y: [1, 4, 2, 2] 183s -0.146 183s 0.264 183s 0.136 183s 0.249 183s test-mul-mat0: iter:197/500 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.733 183s 0.970 183s 183s x1: [1, 2, 1, 1] 183s 0.968 183s -0.112 183s 183s y: [2, 2, 1, 1] 183s 0.709 0.939 183s -0.082 -0.108 183s testing: mul_mat, [2rd.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backw, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.634 183s 0.439 183s 183s x1: [1, 2, 4, 1] 183s -0.242 183s 0.030 183s 183s y: [2, 2, 4, 1] 183s -0.153 -0.106 183s 0.019 0.013 183s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 4, 4, 1] 183s 0.732 183s 0.316 183s -0.168 183s 0.464 183s 183s x1: [1, 2, 4, 1] 183s -0.326 183s 0.720 183s 183s y: [4, 2, 4, 1] 183s -0.239 -0.103 0.055 -0.151 183s 0.527 0.227 -0.121 0.334 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.352 183s -0.817 183s 183s x1: [1, 4, 1, 1] 183s -0.799 183s -0.152 183s -0.035 183s 0.509 183s 183s y: [2, 4, 1, 1] 183s 0.281 0.653 183s 0.053 0.124 183s 0.012 0.028 183s -0.179 -0.416 183s testing: mul_mat, [4, 1, 4, 1] = [2, 4, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.011 -0.500 183s -0.090 -0.808 183s 0.698 0.024 183s -0.725 0.487 183s 183s x1: [2, 1, 4, 1] 183s 0.620 0.986 183s 183s y: [4, 1, 4, 1] 183s -0.499 -0.852 0.457 0.031 183s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.039 -0.589 0.585 -0.469 183s 183s x1: [4, 2, 4, 1] 183s 0.153 0.191 -0.532 0.653 183s 0.384 0.492 0.140 0.753 183s 183s y: [1, 2, 4, 1] 183s -0.724 183s -0.546 183s test-mul-mat0: iter:198/500 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.327 -0.621 -0.928 183s -0.742 -0.090 0.623 183s 183s x1: [3, 3, 1, 1] 183s 0.933 0.944 -0.608 183s 0.296 0.461 -0.230 183s -0.098 -0.529 0.341 183s 183s y: [2, 3, 1, 1] 183s -0.327 -1.156 183s -0.170 -0.404 183s 0.044 0.332 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.606 -0.080 -0.760 183s -0.421 -0.508 -0.025 183s -0.735 0.709 -0.091 183s 183s x1: [3, 2, 1, 1] 183s 0.596 -0.360 0.198 183s 0.463 0.693 -0.291 183s 183s y: [3, 2, 1, 1] 183s -0.482 -0.073 -0.711 183s -0.115 -0.540 0.177 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.409 -0.160 0.074 183s 0.502 -0.537 -0.330 183s -0.858 0.660 -0.867 183s 0.834 -0.631 -0.385 183s 183s x1: [3, 3, 1, 1] 183s -0.791 -0.899 -0.795 183s 0.671 -0.129 0.107 183s -0.858 -0.788 -0.232 183s 183s y: [4, 3, 1, 1] 183s 0.409 0.349 0.775 0.214 183s -0.246 0.370 -0.754 0.599 183s 0.460 0.070 0.417 -0.129 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.538 -0.326 0.265 183s -0.948 -0.180 0.659 183s 0.095 0.910 -0.017 183s 183s x1: [3, 4, 1, 1] 183s -0.772 0.289 -0.145 183s -0.193 0.781 0.830 183s 0.072 0.490 -0.261 183s 0.282 0.591 -0.055 183s 183s y: [3, 4, 1, 1] 183s 0.282 0.584 0.192 183s 0.069 0.590 0.678 183s -0.268 -0.329 0.457 183s -0.359 -0.410 0.565 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.440 0.255 0.778 183s 183s x1: [3, 3, 1, 1] 183s -0.233 -0.878 0.653 183s -0.092 0.782 0.786 183s -0.257 -0.849 -0.600 183s 183s y: [1, 3, 1, 1] 183s 0.181 183s 0.771 183s -0.797 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.825 183s 183s x1: [1, 3, 1, 1] 183s 0.221 183s 0.086 183s -0.150 183s 183s y: [1, 3, 1, 1] 183s 0.182 183s 0.071 183s -0.123 183s test-mul-mat0: iter:199/500 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.205 -0.489 0.121 183s 0.356 -0.089 -0.908 183s -0.204 -0.834 0.870 183s -0.983 0.252 -0.280 183s 183s x1: [3, 4, 1, 1] 183s -0.236 -0.178 -0.821 183s 0.438 0.641 -0.911 183s -0.297 0.301 0.072 183s 0.470 0.423 -0.275 183s 183s y: [4, 4, 1, 1] 183s -0.060 0.678 -0.517 0.417 183s -0.334 0.927 -1.417 -0.014 183s -0.199 -0.198 -0.128 0.348 183s -0.144 0.380 -0.688 -0.278 183s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 2, 2, 1] 183s -0.610 0.589 -0.135 183s -0.593 -0.159 0.585 183s 183s x1: [3, 4, 2, 1] 183s -0.273 0.078 -0.148 183s 0.029 0.847 0.937 183s 0.793 -0.331 -0.885 183s 0.232 -0.690 -0.796 183s 183s y: [2, 4, 2, 1] 183s 0.232 0.063 183s 0.354 0.396 183s -0.559 -0.935 183s -0.440 -0.493 183s testing: mul_mat, [1, 2, 2, 1] = [3, 1, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 1, 2, 1] 183s 0.629 -0.519 0.213 183s 183s x1: [3, 2, 2, 1] 183s 0.957 -0.565 -0.511 183s 0.188 -0.255 -0.307 183s 183s y: [1, 2, 2, 1] 183s 0.786 183s 0.185 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.666 -0.747 -0.981 183s -0.342 0.095 0.468 183s 0.155 0.318 0.275 183s 183s x1: [3, 1, 1, 1] 183s 0.614 0.493 0.073 183s 183s y: [3, 1, 1, 1] 183s -0.849 -0.129 0.272 183s testing: mul_mat, [1, 3, 2, 1] = [3, 1, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 1, 2, 1] 183s -0.534 -0.082 0.433 183s 183s x1: [3, 3, 2, 1] 183s -0.546 0.957 -0.537 183s 0.199 -0.350 0.586 183s 0.556 -0.381 -0.887 183s 183s y: [1, 3, 2, 1] 183s -0.019 183s 0.176 183s -0.650 183s testing: mul_mat, [3,ard.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-back 3, 2, 1] = [1, 3, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.904 183s 0.644 183s -0.276 183s 183s x1: [1, 3, 2, 1] 183s 0.053 183s -0.329 183s 0.906 183s 183s y: [3, 3, 2, 1] 183s -0.047 0.034 -0.014 183s 0.297 -0.212 0.091 183s -0.819 0.583 -0.250 183s test-mul-mat0: iter:200/500 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.403 183s 0.600 183s 0.998 183s 183s x1: [1, 2, 1, 1] 183s 0.766 183s 0.802 183s 183s y: [3, 2, 1, 1] 183s 0.308 0.460 0.764 183s 0.323 0.481 0.800 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.554 183s 0.264 183s 0.743 183s 183s x1: [1, 3, 1, 1] 183s 0.248 183s 0.013 183s -0.536 183s 183s y: [3, 3, 1, 1] 183s -0.137 0.065 0.184 183s -0.007 0.003 0.010 183s 0.297 -0.141 -0.398 183s testing: mul_mat, [2, 3, 1, 4] = [1, 2, 1, 4] * [1, 3, 1, 4] 183s x0: [1, 2, 1, 4] 183s 0.917 183s -0.239 183s 183s x1: [1, 3, 1, 4] 183s 0.034 183s -0.684 183s -0.586 183s 183s y: [2, 3, 1, 4] 183s 0.031 -0.008 183s -0.627 0.163 183s -0.538 0.140 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.247 183s 0.558 183s -0.798 183s -0.081 183s 183s x1: [1, 2, 1, 1] 183s 0.505 183s -0.455 183s 183s y: [4, 2, 1, 1] 183s -0.125 0.282 -0.403 -0.041 183s 0.112 -0.254 0.363 0.037 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.075 0.170 -0.900 -0.723 183s 0.979 0.628 -0.917 0.864 183s 183s x1: [4, 1, 1, 1] 183s -0.995 0.466 -0.338 0.039 183s 183s y: [2, 1, 1, 1] 183s 0.281 -0.338 183s testing: mul_mat, [3, 4, 1, 4] = [2, 3, 1, 4] * [2, 4, 1, 4] 183s x0: [2, 3, 1, 4] 183s 0.673 -0.134 183s -0.309 0.471 183s -0.777 0.374 183s 183s x1: [2, 4, 1, 4] 183s 0.720 0.673 183s -0.142 0.637 183s -0.565 0.194 183s -0.219 0.151 183s 183s y: [3, 4, 1, 4] 183s 0.394 0.095 -0.308 183s -0.181 0.344 0.349 183s -0.406 0.266 0.512 183s -0.168 0.139 0.227 183s test-mul-mat0: iter:201/500 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.901 0.674 183s 183s x1: [2, 3, 1, 1] 183s -0.380 0.739 183s -0.923 -0.707 183s -0.570 -0.700 183s 183s y: [1, 3, 1, 1] 183s 0.156 183s -1.309 183s -0.986 183s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 3, 3, 1] 183s 0.863 -0.444 183s 0.795 -0.007 183s -0.422 0.921 183s 183s x1: [2, 1, 3, 1] 183s 0.322 0.098 183s 183s y: [3, 1, 3, 1] 183s 0.234 0.255 -0.046 183s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.697 0.352 183s -0.888 -0.847 183s 0.384 -0.459 183s 183s x1: [2, 3, 3, 1] 183s -0.431 -0.114 183s 0.094 0.729 183s -0.213 -0.233 183s 183s y: [3, 3, 3, 1] 183s 0.260 0.479 -0.113 183s 0.191 -0.700 -0.299 183s 0.066 0.386 0.025 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.120 -0.078 183s -0.750 -0.296 183s -0.152 -0.510 183s -0.813 -0.726 183s 183s x1: [2, 3, 1, 1] 183s 0.434 -0.884 183s -0.382 -0.229 183s -0.433 -0.844 183s 183s y: [4, 3, 1, 1] 183s 0.017 -0.064 0.385 0.289 183s 0.063 0.354 0.175 0.477 183s 0.118 0.575 0.496 0.965 183s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.543 -0.769 0.601 0.035 183s 0.740 0.943 -0.580 0.867 183s 0.669 0.063 0.328 0.678 183s -0.753 -0.125 0.361 0.701 183s 183s x1: [4, 2, 3, 1] 183s -0.184 -0.356 0.658 0.357 183s -0.547 0.759 -0.385 -0.461 183s 183s y: [4, 2, 3, 1] 183s 0.583 -0.544 0.312 0.671 183s -1.129 0.135 -0.756 -0.146 183s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 2, 3, 1] 183s 0.449 0.787 -0.375 0.935 183s -0.697 -0.275 -0.528 -0.656 183s 183s x1: [4, 4, 3, 1] 183s -0.488 0.651 0.500 0.940 183s -0.113 0.150 0.474 -0.535 183s -0.416 0.680 -0.460 -0.288 183s 0.440 0.362 -0.616 -0.060 183s 183s y: [2, 4, 3, 1] 183s 0.984 -0.719 183s -0.610 0.138 183s 0.252 0.535 183s 0.657 -0.041 183s test-mul-mat0: iter:202/500 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.135 183s 0.113 183s 183s x1: [1, 2, 1, 1] 183s 0.416 183s -0.190 183s 183s y: [2, 2, 1, 1] 183s -0.056 0.047 183s 0.026 -0.021 183s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 4, 4, 1] 183s 0.039 183s 0.286 183s 0.659 183s -0.196 183s 183s x1: [1, 2, 4, 1] 183s -0.879 183s 0.590 183s 183s y: [4, 2, 4, 1] 183s -0.035 -0.252 -0.579 0.172 183s 0.023 0.169 0.389 -0.116 183s testing: mul_mat, [2, 4, 4, 1] = [1, 2, 4, 1] * [1, 4, 4, 1] 183s x0: [1, 2, 4, 1] 183s -0.766 183s -0.469 183s 183s x1: [1, 4, 4, 1] 183s 0.454 183s -0.399 183s -0.576 183s 0.789 183s 183s y: [2, 4, 4, 1] 183s -0.348 -0.213 183s 0.306 0.187 183s 0.442 0.270 183s -0.604 -0.370 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.609 183s 0.054 183s 0.526 183s 183s x1: [1, 2, 1, 1] 183s 0.985 183s 0.869 183s 183s y: [3, 2, 1, 1] 183s 0.600 0.053 0.5ward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-bac18 183s 0.529 0.047 0.457 183s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.810 -0.851 0.960 183s -0.308 -0.074 -0.497 183s 0.677 -0.792 0.106 183s 183s x1: [3, 1, 4, 1] 183s 0.063 0.655 0.949 183s 183s y: [3, 1, 4, 1] 183s 0.405 -0.540 -0.376 183s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 1, 4, 1] 183s -0.498 -0.896 -0.462 183s 183s x1: [3, 3, 4, 1] 183s 0.319 -0.620 -0.933 183s -0.575 -0.668 0.893 183s -0.379 -0.731 0.589 183s 183s y: [1, 3, 4, 1] 183s 0.827 183s 0.473 183s 0.572 183s test-mul-mat0: iter:203/500 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.454 0.622 0.645 -0.573 183s 183s x1: [4, 2, 1, 1] 183s 0.928 -0.708 -0.825 -0.047 183s -0.005 0.600 0.090 -0.881 183s 183s y: [1, 2, 1, 1] 183s -1.367 183s 0.939 183s testing: mul_mat, [1, 1, 3, 1] = [4, 1, 3, 1] * [4, 1, 3, 1] 183s x0: [4, 1, 3, 1] 183s 0.517 0.834 -0.609 0.376 183s 183s x1: [4, 1, 3, 1] 183s 0.458 -0.406 -0.072 0.562 183s 183s y: [1, 1, 3, 1] 183s 0.153 183s testing: mul_mat, [1, 1, 3, 4] = [4, 1, 3, 4] * [4, 1, 3, 4] 183s x0: [4, 1, 3, 4] 183s -0.285 -0.832 -0.016 0.893 183s 183s x1: [4, 1, 3, 4] 183s -0.218 0.585 -0.091 -0.595 183s 183s y: [1, 1, 3, 4] 183s -0.955 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.279 -0.453 -0.245 0.859 183s 183s x1: [4, 1, 1, 1] 183s 0.885 0.999 0.896 0.965 183s 183s y: [1, 1, 1, 1] 183s 0.405 183s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 3, 3, 1] 183s -0.903 183s 0.884 183s -0.001 183s 183s x1: [1, 4, 3, 1] 183s -0.404 183s 0.398 183s -0.632 183s -0.296 183s 183s y: [3, 4, 3, 1] 183s 0.365 -0.357 0.000 183s -0.360 0.352 -0.000 183s 0.571 -0.559 0.001 183s 0.267 -0.262 0.000 183s testing: mul_mat, [2, 1, 3, 4] = [3, 2, 3, 4] * [3, 1, 3, 4] 183s x0: [3, 2, 3, 4] 183s 0.129 0.432 -0.358 183s 0.009 -0.884 0.217 183s 183s x1: [3, 1, 3, 4] 183s 0.878 0.133 -0.721 183s 183s y: [2, 1, 3, 4] 183s 0.429 -0.266 183s test-mul-mat0: iter:204/500 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.768 0.526 0.258 183s 0.513 0.877 -0.650 183s 183s x1: [3, 4, 1, 1] 183s -0.410 -0.473 0.347 183s 0.603 0.658 0.814 183s -0.187 -0.924 0.428 183s -0.618 0.961 0.086 183s 183s y: [2, 4, 1, 1] 183s -0.473 -0.851 183s 1.019 0.357 183s -0.519 -1.185 183s 0.052 0.470 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.006 -0.441 0.959 183s 0.883 0.570 -0.451 183s -0.589 -0.083 -0.848 183s -0.932 -0.269 -0.035 183s 183s x1: [3, 2, 1, 1] 183s -0.372 -0.743 0.522 183s -0.670 -0.530 -0.218 183s 183s y: [4, 2, 1, 1] 183s 0.830 -0.987 -0.161 0.528 183s 0.029 -0.796 0.623 0.775 183s testing: mul_mat, [1, 4, 1, 4] = [3, 1, 1, 4] * [3, 4, 1, 4] 183s x0: [3, 1, 1, 4] 183s -0.845 -0.890 -0.379 183s 183s x1: [3, 4, 1, 4] 183s -0.855 -0.841 0.347 183s -0.895 0.245 -0.048 183s 0.873 -0.230 -0.790 183s 0.386 -0.352 -0.440 183s 183s y: [1, 4, 1, 4] 183s 1.341 183s 0.557 183s -0.234 183s 0.154 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.609 0.011 0.150 183s 183s x1: [3, 1, 1, 1] 183s 0.138 -0.371 -0.034 183s 183s y: [1, 1, 1, 1] 183s 0.075 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.289 183s 0.095 183s -0.044 183s 183s x1: [1, 3, 1, 1] 183s -0.538 183s 0.826 183s -0.158 183s 183s y: [3, 3, 1, 1] 183s -0.155 -0.051 0.024 183s 0.239 0.078 -0.036 183s -0.046 -0.015 0.007 183s testing: mul_mat, [4, 1, 1, 4] = [3, 4, 1, 4] * [3, 1, 1, 4] 183s x0: [3, 4, 1, 4] 183s -0.614 -0.792 -0.118 183s 0.070 0.273 -0.472 183s -0.355 0.517 0.026 183s -0.449 0.876 0.345 183s 183s x1: [3, 1, 1, 4] 183s 0.656 0.429 0.554 183s 183s y: [4, 1, 1, 4] 183s -0.808 -0.098 0.003 0.272 183s test-mul-mat0: iter:205/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.020 183s -0.692 183s 183s x1: [1, 1, 1, 1] 183s -0.896 183s 183s y: [2, 1, 1, 1] 183s 0.018 0.620 183s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 2, 3, 1] 183s 0.213 183s -0.050 183s 183s x1: [1, 2, 3, 1] 183s 0.449 183s -0.376 183s 183s y: [2, 2, 3, 1] 183s 0.096 -0.022 183s -0.080 0.019 183s testing: mul_mat, [1, 2, 3, 2] = [1, 1, 3, 2] * [1, 2, 3, 2] 183s x0: [1, 1, 3, 2] 183s -0.269 183s 183s x1: [1, 2, 3, 2] 183s 0.164 183s 0.594 183s 183s y: [1, 2, 3, 2] 183s -0.044 183s -0.160 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.947 183s 0.718 183s 183s x1: [1, 1, 1, 1] 183s -0.666 183s 183s y: [2, 1, 1, 1] 183s 0.630 -0.478 183s testing: mul_mat, [1, 1, 3, 1] = [2, 1, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.963 0.345 183s 183s x1: [2, 1, 3, 1] 183s 0.961 -0.856 183s 183s y: [1, 1, 3, 1] 183s -1.221 183s testing: mul_mat, [4, 2, 3, 2] = [1, 4, 3, 2kward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-ba] * [1, 2, 3, 2] 183s x0: [1, 4, 3, 2] 183s 0.392 183s 0.688 183s 0.227 183s 0.445 183s 183s x1: [1, 2, 3, 2] 183s -0.805 183s -0.640 183s 183s y: [4, 2, 3, 2] 183s -0.315 -0.554 -0.183 -0.358 183s -0.251 -0.440 -0.145 -0.285 183s test-mul-mat0: iter:206/500 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.267 0.356 183s 183s x1: [2, 2, 1, 1] 183s -0.481 0.022 183s -0.961 0.911 183s 183s y: [1, 2, 1, 1] 183s -0.120 183s 0.068 183s testing: mul_mat, [3, 1, 2, 1] = [2, 3, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 3, 2, 1] 183s -0.167 -0.030 183s 0.394 -0.242 183s -0.167 -0.430 183s 183s x1: [2, 1, 2, 1] 183s -0.883 -0.546 183s 183s y: [3, 1, 2, 1] 183s 0.164 -0.215 0.382 183s testing: mul_mat, [3, 3, 2, 3] = [2, 3, 2, 3] * [2, 3, 2, 3] 183s x0: [2, 3, 2, 3] 183s 0.709 -0.015 183s -0.234 0.426 183s -0.449 -0.910 183s 183s x1: [2, 3, 2, 3] 183s 0.189 -0.467 183s 0.923 -0.752 183s 0.136 0.591 183s 183s y: [3, 3, 2, 3] 183s 0.141 -0.243 0.340 183s 0.665 -0.536 0.271 183s 0.088 0.220 -0.599 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.163 0.403 183s 183s x1: [2, 3, 1, 1] 183s -0.872 0.476 183s -0.979 0.658 183s 0.324 0.362 183s 183s y: [1, 3, 1, 1] 183s 0.050 183s 0.106 183s 0.199 183s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.575 183s 0.183 183s 0.485 183s 183s x1: [1, 2, 2, 1] 183s 0.370 183s -0.613 183s 183s y: [3, 2, 2, 1] 183s -0.213 0.068 0.179 183s 0.353 -0.112 -0.297 183s testing: mul_mat, [2, 1, 2, 3] = [3, 2, 2, 3] * [3, 1, 2, 3] 183s x0: [3, 2, 2, 3] 183s -0.305 0.432 -0.934 183s -0.289 0.596 -0.002 183s 183s x1: [3, 1, 2, 3] 183s -0.557 -0.647 0.163 183s 183s y: [2, 1, 2, 3] 183s -0.261 -0.226 183s test-mul-mat0: iter:207/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.347 0.210 -0.216 183s 0.650 0.806 0.261 183s 183s x1: [3, 2, 1, 1] 183s 0.002 -0.256 0.399 183s -0.992 0.016 0.087 183s 183s y: [2, 2, 1, 1] 183s -0.141 -0.101 183s 0.329 -0.609 183s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.907 0.234 -0.792 183s -0.076 -0.678 0.597 183s 0.577 0.532 -0.620 183s 0.227 0.338 0.641 183s 183s x1: [3, 2, 3, 1] 183s -0.444 0.003 -0.704 183s 0.391 -0.541 0.213 183s 183s y: [4, 2, 3, 1] 183s 0.155 -0.388 0.182 -0.552 183s 0.060 0.464 -0.194 0.042 183s testing: mul_mat, [3, 4, 3, 3] = [3, 3, 3, 3] * [3, 4, 3, 3] 183s x0: [3, 3, 3, 3] 183s -0.186 0.842 0.630 183s -0.601 -0.028 -0.540 183s -0.583 -0.918 0.483 183s 183s x1: [3, 4, 3, 3] 183s -0.009 -0.269 -0.223 183s 0.372 0.958 -0.885 183s 0.013 0.741 -0.544 183s -0.050 -0.085 0.256 183s 183s y: [3, 4, 3, 3] 183s -0.365 0.133 0.144 183s 0.181 0.227 -1.523 183s 0.279 0.265 -0.950 183s 0.099 -0.105 0.231 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.685 -0.880 0.335 183s -0.078 0.176 0.124 183s -0.979 -0.075 0.898 183s 183s x1: [3, 3, 1, 1] 183s 0.844 -0.366 -0.465 183s 0.211 -0.596 -0.518 183s -0.230 0.061 0.901 183s 183s y: [3, 3, 1, 1] 183s 0.744 -0.188 -1.216 183s 0.495 -0.186 -0.627 183s 0.091 0.140 1.030 183s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.546 0.844 -0.960 183s 0.793 -0.039 -0.745 183s 183s x1: [3, 3, 3, 1] 183s 0.415 -0.752 -0.588 183s 0.377 0.028 -0.979 183s -0.995 0.949 -0.612 183s 183s y: [2, 3, 3, 1] 183s 0.156 0.796 183s 1.169 1.027 183s 0.845 -0.370 183s testing: mul_mat, [4, 3, 3, 3] = [2, 4, 3, 3] * [2, 3, 3, 3] 183s x0: [2, 4, 3, 3] 183s 0.379 -0.683 183s 0.068 0.194 183s 0.783 0.893 183s 0.384 -0.978 183s 183s x1: [2, 3, 3, 3] 183s -0.545 0.609 183s 0.173 -0.315 183s -0.985 0.744 183s 183s y: [4, 3, 3, 3] 183s -0.622 0.081 0.118 -0.805 183s 0.281 -0.049 -0.146 0.375 183s -0.881 0.078 -0.106 -1.106 183s test-mul-mat0: iter:208/500 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.198 -0.086 -0.338 183s 183s x1: [3, 3, 1, 1] 183s 0.348 0.420 0.376 183s 0.982 0.511 0.064 183s 0.788 -0.898 0.355 183s 183s y: [1, 3, 1, 1] 183s -0.094 183s 0.129 183s 0.114 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.720 0.553 -0.924 183s -0.149 -0.922 0.104 183s 183s x1: [3, 1, 1, 1] 183s 0.611 0.429 0.854 183s 183s y: [2, 1, 1, 1] 183s -0.992 -0.398 183s testing: mul_mat, [3, 2, 1, 2] = [3, 3, 1, 2] * [3, 2, 1, 2] 183s x0: [3, 3, 1, 2] 183s -0.108 -0.044 0.004 183s -0.753 0.867 -0.798 183s 0.161 -0.470 0.813 183s 183s x1: [3, 2, 1, 2] 183s 0.034 0.450 0.275 183s -0.594 -0.920 0.839 183s 183s y: [3, 2, 1, 2] 183s -0.023 0.145 0.018 183s 0.108 -1.021 1.019 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.824 -0.284 -0.037 183s -0.564 0.391 0.259 183s 0.423 -0.573 -0.371 183s 183s x1: [3,ckward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-b 3, 1, 1] 183s -0.245 0.464 -0.757 183s 0.161 0.544 -0.918 183s 0.131 0.996 0.268 183s 183s y: [3, 3, 1, 1] 183s -0.306 0.124 -0.089 183s 0.012 -0.116 0.097 183s -0.185 0.385 -0.615 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.877 -0.941 0.307 183s 0.063 0.587 0.220 183s 183s x1: [3, 3, 1, 1] 183s -0.876 0.789 -0.558 183s -0.286 -0.828 -0.769 183s 0.584 -0.891 -0.905 183s 183s y: [2, 3, 1, 1] 183s -0.145 0.284 183s 0.794 -0.674 183s 0.049 -0.686 183s testing: mul_mat, [1, 3, 1, 2] = [2, 1, 1, 2] * [2, 3, 1, 2] 183s x0: [2, 1, 1, 2] 183s 0.742 0.352 183s 183s x1: [2, 3, 1, 2] 183s 0.130 0.388 183s -0.648 0.126 183s -0.344 -0.127 183s 183s y: [1, 3, 1, 2] 183s 0.233 183s -0.437 183s -0.300 183s test-mul-mat0: iter:209/500 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.183 -0.584 0.521 -0.511 183s 0.636 -0.349 0.878 0.988 183s 183s x1: [4, 1, 1, 1] 183s 0.052 0.811 -0.881 -0.644 183s 183s y: [2, 1, 1, 1] 183s -0.595 -1.658 183s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 2, 3, 1] 183s -0.649 -0.159 -0.294 0.002 183s -0.281 -0.306 -0.221 0.253 183s 183s x1: [4, 2, 3, 1] 183s 0.777 -0.466 -0.139 -0.272 183s 0.626 0.157 -0.606 -0.891 183s 183s y: [2, 2, 3, 1] 183s -0.390 -0.114 183s -0.255 -0.315 183s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.802 0.157 -0.099 -0.853 183s -0.539 -0.200 -0.607 -0.340 183s 0.869 0.293 0.690 0.290 183s 183s x1: [4, 2, 3, 1] 183s -0.754 -0.800 0.069 -0.100 183s 0.030 0.421 0.969 0.380 183s 183s y: [3, 2, 3, 1] 183s 0.558 0.559 -0.871 183s -0.378 -0.818 0.928 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.885 0.302 0.713 -0.779 183s -0.981 0.736 -0.924 -0.949 183s 0.857 0.982 -0.496 -0.301 183s -0.884 0.155 -0.372 0.267 183s 183s x1: [4, 3, 1, 1] 183s 0.846 -0.914 0.192 0.715 183s 0.379 -0.118 0.005 -0.359 183s 0.952 -0.443 -0.422 0.405 183s 183s y: [4, 3, 1, 1] 183s 0.053 -2.359 -0.483 -0.770 183s 0.583 -0.123 0.314 -0.451 183s 0.093 -1.255 0.468 -0.645 183s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 2, 3, 1] 183s 0.499 -0.961 -0.211 -0.601 183s 0.652 0.911 -0.174 -0.571 183s 183s x1: [4, 4, 3, 1] 183s 0.744 0.231 0.114 0.831 183s -0.577 -0.171 0.210 0.305 183s 0.834 0.851 0.257 -0.609 183s -0.571 -0.339 0.482 -0.686 183s 183s y: [2, 4, 3, 1] 183s -0.373 0.202 183s -0.351 -0.743 183s -0.090 1.622 183s 0.351 -0.373 183s testing: mul_mat, [3, 4, 3, 1] = [2, 3, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 3, 3, 1] 183s 0.719 0.566 183s -0.806 0.817 183s 0.753 -0.378 183s 183s x1: [2, 4, 3, 1] 183s 0.992 -0.622 183s 0.859 0.619 183s 0.448 0.530 183s -0.416 -0.053 183s 183s y: [3, 4, 3, 1] 183s 0.361 -1.308 0.982 183s 0.968 -0.188 0.413 183s 0.622 0.072 0.136 183s -0.329 0.292 -0.293 183s test-mul-mat0: iter:210/500 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.250 -0.722 0.429 183s 0.067 -0.101 -0.012 183s 0.262 0.380 -0.835 183s 183s x1: [3, 3, 1, 1] 183s 0.070 -0.916 0.567 183s -0.480 -0.840 0.312 183s -0.856 0.057 -0.475 183s 183s y: [3, 3, 1, 1] 183s 0.922 0.090 -0.804 183s 0.621 0.049 -0.706 183s -0.459 -0.057 0.194 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.311 0.662 -0.286 183s 183s x1: [3, 3, 1, 1] 183s -0.096 0.391 0.925 183s 0.493 -0.416 -0.692 183s -0.059 0.747 -0.860 183s 183s y: [1, 3, 1, 1] 183s -0.035 183s 0.075 183s 0.722 183s testing: mul_mat, [3, 1, 1, 2] = [3, 3, 1, 2] * [3, 1, 1, 2] 183s x0: [3, 3, 1, 2] 183s 0.066 0.241 0.862 183s 0.344 -0.330 -0.071 183s -0.757 0.658 -0.809 183s 183s x1: [3, 1, 1, 2] 183s 0.395 0.229 0.234 183s 183s y: [3, 1, 1, 2] 183s 0.283 0.044 -0.338 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.628 -0.782 0.742 183s -0.499 -0.306 -0.920 183s 183s x1: [3, 3, 1, 1] 183s 0.345 -0.803 0.174 183s -0.344 0.858 0.888 183s -0.948 0.087 0.122 183s 183s y: [2, 3, 1, 1] 183s 0.974 -0.086 183s -0.228 -0.909 183s -0.573 0.334 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.893 -0.706 183s 183s x1: [2, 3, 1, 1] 183s -0.963 -0.588 183s 0.008 -0.719 183s -0.930 0.199 183s 183s y: [1, 3, 1, 1] 183s -0.444 183s 0.515 183s -0.971 183s testing: mul_mat, [3, 2, 1, 2] = [1, 3, 1, 2] * [1, 2, 1, 2] 183s x0: [1, 3, 1, 2] 183s -0.589 183s 0.861 183s 0.894 183s 183s x1: [1, 2, 1, 2] 183s -0.082 183s 0.641 183s 183s y: [3, 2, 1, 2] 183s 0.048 -0.071 -0.073 183s -0.377 0.552 0.573 183s test-mul-mat0: iter:211/500 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.482 0.479 0.321 -0.319 183s 183s x1: [4, 2, 1, 1] 183s -0.824 -0.ackward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s 533 0.097 -0.606 183s 0.161 -0.162 -0.527 0.198 183s 183s y: [1, 2, 1, 1] 183s -0.428 183s -0.232 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.858 -0.183 -0.448 -0.731 183s -0.322 -0.555 0.854 0.196 183s -0.802 0.328 0.765 0.037 183s 183s x1: [4, 1, 1, 1] 183s 0.383 0.214 -0.025 -0.699 183s 183s y: [3, 1, 1, 1] 183s 0.811 -0.400 -0.282 183s testing: mul_mat, [2, 3, 1, 4] = [4, 2, 1, 4] * [4, 3, 1, 4] 183s x0: [4, 2, 1, 4] 183s 0.741 0.576 0.693 -0.209 183s 0.649 -0.787 -0.713 0.576 183s 183s x1: [4, 3, 1, 4] 183s 0.924 0.941 0.504 0.021 183s -0.665 -0.336 0.859 -0.192 183s 0.863 0.110 -0.710 0.341 183s 183s y: [2, 3, 1, 4] 183s 1.572 -0.488 183s -0.051 -0.891 183s 0.139 1.177 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.237 0.457 -0.426 0.163 183s -0.374 -0.268 0.726 0.202 183s 0.788 0.458 0.196 0.882 183s 0.248 -0.956 0.177 -0.802 183s 183s x1: [4, 2, 1, 1] 183s 0.181 0.076 -0.189 -0.170 183s 0.289 0.097 -0.594 0.217 183s 183s y: [4, 2, 1, 1] 183s 0.131 -0.260 -0.010 0.075 183s 0.402 -0.522 0.348 -0.301 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.269 -0.536 -0.002 -0.425 183s 0.540 -0.999 -0.492 -0.276 183s 183s x1: [4, 4, 1, 1] 183s 0.035 0.414 -0.524 -0.361 183s -0.202 -0.504 0.820 0.874 183s 0.306 -0.351 0.163 -0.596 183s 0.055 -0.620 0.481 -0.708 183s 183s y: [2, 4, 1, 1] 183s -0.058 -0.038 183s -0.158 -0.250 183s 0.524 0.600 183s 0.647 0.608 183s testing: mul_mat, [3, 4, 1, 4] = [2, 3, 1, 4] * [2, 4, 1, 4] 183s x0: [2, 3, 1, 4] 183s 0.546 0.631 183s 0.853 0.794 183s 0.478 -0.474 183s 183s x1: [2, 4, 1, 4] 183s 0.704 0.751 183s 0.887 -0.093 183s 0.634 -0.915 183s 0.942 0.048 183s 183s y: [3, 4, 1, 4] 183s 0.859 1.198 -0.019 183s 0.425 0.682 0.468 183s -0.231 -0.186 0.736 183s 0.544 0.841 0.428 183s test-mul-mat0: iter:212/500 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.084 183s -0.966 183s 183s x1: [1, 4, 1, 1] 183s -0.142 183s 0.425 183s 0.606 183s -0.597 183s 183s y: [2, 4, 1, 1] 183s -0.012 0.138 183s 0.036 -0.410 183s 0.051 -0.585 183s -0.050 0.576 183s testing: mul_mat, [4, 2, 3, 1] = [1, 4, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 4, 3, 1] 183s -0.927 183s 0.092 183s -0.953 183s 0.047 183s 183s x1: [1, 2, 3, 1] 183s 0.072 183s 0.611 183s 183s y: [4, 2, 3, 1] 183s -0.067 0.007 -0.069 0.003 183s -0.566 0.056 -0.582 0.028 183s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 4, 3, 1] 183s 0.984 183s 0.289 183s -0.275 183s 0.967 183s 183s x1: [1, 4, 3, 1] 183s 0.024 183s 0.922 183s 0.766 183s 0.503 183s 183s y: [4, 4, 3, 1] 183s 0.024 0.007 -0.007 0.023 183s 0.908 0.266 -0.254 0.891 183s 0.754 0.221 -0.211 0.741 183s 0.495 0.145 -0.138 0.486 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.843 183s 183s x1: [1, 4, 1, 1] 183s -0.631 183s 0.438 183s 0.275 183s -0.845 183s 183s y: [1, 4, 1, 1] 183s -0.532 183s 0.369 183s 0.232 183s -0.712 183s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 2, 3, 1] 183s 0.298 183s -0.767 183s 183s x1: [1, 1, 3, 1] 183s 0.179 183s 183s y: [2, 1, 3, 1] 183s 0.053 -0.137 183s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.623 -0.048 183s -0.234 -0.764 183s 183s x1: [2, 1, 3, 1] 183s -0.039 -0.678 183s 183s y: [2, 1, 3, 1] 183s 0.057 0.528 183s test-mul-mat0: iter:213/500 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.802 -0.048 0.850 0.158 183s 183s x1: [4, 1, 1, 1] 183s 0.459 -0.576 0.611 0.691 183s 183s y: [1, 1, 1, 1] 183s 0.289 183s testing: mul_mat, [4, 1, 3, 1] = [4, 4, 3, 1] * [4, 1, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.981 -0.586 -0.927 -0.974 183s -0.023 -0.228 0.875 -0.281 183s 0.085 0.334 0.144 -0.304 183s 0.025 0.178 -0.106 0.977 183s 183s x1: [4, 1, 3, 1] 183s 0.273 -0.056 -0.896 -0.116 183s 183s y: [4, 1, 3, 1] 183s 0.710 -0.746 -0.089 -0.022 183s testing: mul_mat, [2, 4, 3, 4] = [4, 2, 3, 4] * [4, 4, 3, 4] 183s x0: [4, 2, 3, 4] 183s -0.238 -0.790 0.486 -0.983 183s -0.534 0.925 -0.122 -0.460 183s 183s x1: [4, 4, 3, 4] 183s 0.529 0.056 -0.424 0.685 183s 0.190 0.310 -0.144 -0.416 183s 0.571 0.994 0.476 0.120 183s 0.139 0.433 -0.312 -0.302 183s 183s y: [2, 4, 3, 4] 183s -1.049 -0.494 183s 0.048 0.394 183s -0.808 0.501 183s -0.230 0.503 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.921 -0.819 -0.274 -0.921 183s -0.668 -0.887 -0.825 0.832 183s -0.994 0.245 -0.037 -0.433 183s -0.185 -0.316 -0.599 -0.528 183s 183s x1: [4, 2, 1, 1] 183s 0.500 -0.528 0.876 -0.007 183s -0.778 -0.482 0.425 -0.191 183s 183s y: [4, 2, 1, 1] 183s 0.659 -0.593 -0.657 -0.447 183s -0.262 0.437 0.722 0.143 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.793 -0.414 0.262 0.915 183s 183s x1: [4, 4, 3, 1] 183s 0.718 -0.487 -0.374 0.808 183s 0.740 0.300 0.309 -0.788 183s 0.176 -0.698 -0.566 0.694 183s 0.727 0.243 0.303 0.647 183s 183s y: [1, 4, 3, 1] 183s 0.274 183s -1.351 183s 0.636 183s -0.005 183s testing: mul_mat, [4, 4, 3, 4] = [1, 4, 3, 4] * [1, 4, 3, 4] 183s x0: [1, 4, 3, 4] 183s 0.601 183s -0.873 183s 0.514 183s 0.618 183s 183s x1: [1, 4, 3, 4] 183s -0.806 183s -0.386 183s -0.301 183s 0.254 183s 183s y: [4, 4, 3, 4] 183s -0.485 0.704 -0.415 -0.499 183s -0.232 0.337 -0.199 -0.239 183s -0.181 0.263 -0.155 -0.186 183s 0.153 -0.222 0.131 0.157 183s test-mul-mat0: iter:214/500 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.173 0.680 0.664 183s 0.913 0.731 0.322 183s 183s x1: [3, 3, 1, 1] 183s -0.137 -0.528 0.761 183s 0.512 0.493 0.518 183s 0.602 0.359 0.536 183s 183s y: [2, 3, 1, 1] 183s 0.171 -0.266 183s 0.590 0.995 183s 0.496 0.985 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.667 -0.402 0.527 183s -0.820 -0.256 -0.735 183s 0.564 0.352 0.128 183s 183s x1: [3, 2, 1, 1] 183s -0.956 0.267 0.632 183s -0.397 0.624 -0.962 183s 183s y: [3, 2, 1, 1] 183s 0.863 0.251 -0.365 183s -0.494 0.872 -0.127 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.847 -0.218 0.627 183s 0.578 -0.896 0.671 183s 183s x1: [3, 3, 1, 1] 183s -0.964 0.113 -0.360 183s 0.529 -0.369 -0.758 183s -0.112 -0.833 0.118 183s 183s y: [2, 3, 1, 1] 183s -1.066 -0.900 183s 0.053 0.127 183s 0.161 0.761 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.628 0.848 -0.018 183s 183s x1: [3, 2, 1, 1] 183s -0.155 0.736 -0.725 183s -0.531 0.774 -0.679 183s 183s y: [1, 2, 1, 1] 183s 0.735 183s 1.002 183s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.241 183s -0.418 183s 183s x1: [1, 3, 1, 1] 183s 0.116 183s -0.887 183s -0.454 183s 183s y: [2, 3, 1, 1] 183s 0.028 -0.048 183s -0.214 0.370 183s -0.109 0.189 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.788 -0.622 183s -0.877 0.241 183s -0.001 -0.286 183s 183s x1: [2, 1, 1, 1] 183s 0.581 0.881 183s 183s y: [3, 1, 1, 1] 183s -1.005 -0.297 -0.253 183s test-mul-mat0: iter:215/500 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.863 -0.714 183s 0.116 0.638 183s -0.393 0.918 183s -0.990 -0.544 183s 183s x1: [2, 1, 1, 1] 183s -0.873 -0.988 183s 183s y: [4, 1, 1, 1] 183s -0.049 -0.731 -0.565 1.403 183s testing: mul_mat, [1, 4, 2, 1] = [2, 1, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.077 -0.806 183s 183s x1: [2, 4, 2, 1] 183s -0.100 0.125 183s -0.431 0.447 183s 0.593 0.810 183s -0.971 0.173 183s 183s y: [1, 4, 2, 1] 183s -0.108 183s -0.393 183s -0.607 183s -0.214 183s testing: mul_mat, [2, 1, 2, 3] = [2, 2, 2, 3] * [2, 1, 2, 3] 183s x0: [2, 2, 2, 3] 183s 0.820 0.432 183s 0.933 0.849 183s 183s x1: [2, 1, 2, 3] 183s -0.794 -0.562 183s 183s y: [2, 1, 2, 3] 183s -0.894 -1.218 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.096 0.191 183s -0.349 -0.472 183s 183s x1: [2, 2, 1, 1] 183s -0.028 -0.002 183s 0.257 -0.473 183s 183s y: [2, 2, 1, 1] 183s -0.003 0.011 183s -0.065 0.133 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.864 -0.381 183s 0.139 0.703 183s 183s x1: [2, 2, 2, 1] 183s -0.416 0.040 183s 0.133 -0.792 183s 183s y: [2, 2, 2, 1] 183s 0.344 -0.030 183s 0.187 -0.539 183s testing: mul_mat, [3, 2, 2, 3] = [2, 3, 2, 3] * [2, 2, 2, 3] 183s x0: [2, 3, 2, 3] 183s 0.490 0.641 183s 0.994 -0.532 183s -0.112 -0.417 183s 183s x1: [2, 2, 2, 3] 183s -0.877 -0.687 183s 0.568 0.648 183s 183s y: [3, 2, 2, 3] 183s -0.870 -0.506 0.384 183s 0.694 0.219 -0.334 183s test-mul-mat0: iter:216/500 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.413 -0.569 -0.495 -0.851 183s -0.128 -0.249 -0.418 0.604 183s 0.571 -0.495 -0.670 0.279 183s -0.931 0.524 0.570 0.397 183s 183s x1: [4, 3, 1, 1] 183s -0.260 0.612 0.133 -0.201 183s 0.362 -0.109 -0.124 -0.678 183s -0.634 -0.839 0.019 -0.856 183s 183s y: [4, 3, 1, 1] 183s -0.136 -0.296 -0.596 0.558 183s 0.551 -0.377 0.155 -0.735 183s 1.459 -0.236 -0.199 -0.179 183s testing: mul_mat, [3, 4, 4, 1] = [4, 3, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 3, 4, 1] 183s 0.573 -0.788 -0.053 0.961 183s -0.248 -0.062 0.144 0.693 183s -0.096 -0.859 0.180 -0.463 183s 183s x1: [4, 4, 4, 1] 183s 0.499 -0.987 -0.862 0.111 183s 0.145 -0.063 -0.527 -0.964 183s 0.813 -0.205 -0.598 0.974 183s 0.814 -0.455 0.668 -0.600 183s 183s y: [3, 4, 4, 1] 183s 1.216 -0.111 0.593 183s -0.766 -0.777 0.392 183s 1.595 0.399 -0.461 183s 0.213 -0.4ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s 93 0.710 183s testing: mul_mat, [1, 3, 4, 3] = [4, 1, 4, 3] * [4, 3, 4, 3] 183s x0: [4, 1, 4, 3] 183s -0.576 0.777 -0.157 -0.466 183s 183s x1: [4, 3, 4, 3] 183s -0.254 -0.816 -0.478 0.051 183s 0.465 0.788 -0.640 0.168 183s 0.907 -0.982 -0.601 0.683 183s 183s y: [1, 3, 4, 3] 183s -0.436 183s 0.366 183s -1.509 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.418 -0.240 0.621 -0.093 183s 0.186 0.840 -0.245 0.448 183s -0.530 0.233 -0.826 0.959 183s 183s x1: [4, 1, 1, 1] 183s -0.077 -0.514 0.051 0.614 183s 183s y: [3, 1, 1, 1] 183s 0.131 -0.183 0.468 183s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 2, 4, 1] 183s 0.555 -0.056 0.905 183s -0.727 -0.488 -0.266 183s 183s x1: [3, 4, 4, 1] 183s -0.273 -0.538 -0.319 183s 0.594 0.080 -0.824 183s -0.376 -0.212 0.671 183s -0.565 0.547 -0.772 183s 183s y: [2, 4, 4, 1] 183s -0.410 0.546 183s -0.421 -0.252 183s 0.410 0.199 183s -1.043 0.349 183s testing: mul_mat, [4, 3, 4, 3] = [2, 4, 4, 3] * [2, 3, 4, 3] 183s x0: [2, 4, 4, 3] 183s -0.332 -0.423 183s -0.211 0.038 183s -0.529 0.716 183s 0.882 0.322 183s 183s x1: [2, 3, 4, 3] 183s 0.502 0.394 183s -0.433 0.731 183s -0.538 0.967 183s 183s y: [4, 3, 4, 3] 183s -0.333 -0.091 0.017 0.570 183s -0.166 0.119 0.753 -0.147 183s -0.231 0.151 0.978 -0.164 183s test-mul-mat0: iter:217/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.879 0.051 0.085 183s 183s x1: [3, 1, 1, 1] 183s -0.835 -0.849 0.527 183s 183s y: [1, 1, 1, 1] 183s -0.732 183s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.835 -0.787 -0.014 183s 183s x1: [3, 1, 4, 1] 183s 0.960 0.960 0.814 183s 183s y: [1, 1, 4, 1] 183s 0.035 183s testing: mul_mat, [2, 1, 4, 1] = [3, 2, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.779 0.905 0.667 183s -0.972 0.342 -0.950 183s 183s x1: [3, 1, 4, 1] 183s 0.731 -0.903 -0.096 183s 183s y: [2, 1, 4, 1] 183s -1.451 -0.928 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.332 -0.691 0.723 183s -0.620 -0.304 -0.641 183s 183s x1: [3, 2, 1, 1] 183s 0.870 -0.069 -0.341 183s 0.505 -0.553 -0.525 183s 183s y: [2, 2, 1, 1] 183s -0.488 -0.300 183s -0.165 0.191 183s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.205 -0.657 183s -0.379 0.548 183s 0.189 -0.539 183s 0.901 0.621 183s 183s x1: [2, 3, 4, 1] 183s -0.440 0.278 183s 0.395 0.529 183s -0.929 0.480 183s 183s y: [4, 3, 4, 1] 183s -0.273 0.319 -0.233 -0.224 183s -0.266 0.140 -0.211 0.685 183s -0.505 0.615 -0.434 -0.540 183s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.746 0.001 -0.533 -0.135 183s 183s x1: [4, 2, 4, 1] 183s 0.214 0.235 -0.342 0.557 183s -0.217 0.119 0.179 0.726 183s 183s y: [1, 2, 4, 1] 183s 0.267 183s -0.356 183s test-mul-mat0: iter:218/500 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.850 183s 183s x1: [1, 1, 1, 1] 183s 0.537 183s 183s y: [1, 1, 1, 1] 183s -0.456 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.518 183s 0.404 183s 0.274 183s 0.570 183s 183s x1: [1, 1, 1, 1] 183s 0.174 183s 183s y: [4, 1, 1, 1] 183s -0.090 0.070 0.048 0.099 183s testing: mul_mat, [2, 4, 1, 2] = [1, 2, 1, 2] * [1, 4, 1, 2] 183s x0: [1, 2, 1, 2] 183s 0.079 183s 0.960 183s 183s x1: [1, 4, 1, 2] 183s -0.152 183s -0.041 183s 0.316 183s 0.849 183s 183s y: [2, 4, 1, 2] 183s -0.012 -0.146 183s -0.003 -0.040 183s 0.025 0.303 183s 0.067 0.815 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.846 183s 0.773 183s 0.144 183s 183s x1: [1, 2, 1, 1] 183s -0.707 183s -0.102 183s 183s y: [3, 2, 1, 1] 183s -0.598 -0.546 -0.102 183s -0.086 -0.079 -0.015 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.378 0.147 0.061 183s -0.335 -0.787 -0.564 183s -0.191 0.477 -0.968 183s -0.798 -0.283 -0.090 183s 183s x1: [3, 1, 1, 1] 183s -0.902 -0.044 -0.485 183s 183s y: [4, 1, 1, 1] 183s -0.376 0.610 0.621 0.776 183s testing: mul_mat, [3, 3, 1, 2] = [4, 3, 1, 2] * [4, 3, 1, 2] 183s x0: [4, 3, 1, 2] 183s 0.186 0.622 0.483 -0.037 183s -0.485 -0.453 -0.240 0.126 183s 0.393 -0.698 0.241 0.258 183s 183s x1: [4, 3, 1, 2] 183s 0.861 -0.787 -0.062 0.480 183s -0.634 -0.983 0.439 -0.498 183s -0.211 0.732 0.400 -0.673 183s 183s y: [3, 3, 1, 2] 183s -0.377 0.014 0.996 183s -0.498 0.584 0.414 183s 0.634 -0.410 -0.671 183s test-mul-mat0: iter:219/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.393 0.337 0.882 0.766 183s 0.050 -0.917 0.610 -0.562 183s -0.586 0.439 0.138 -0.748 183s 0.919 0.957 0.799 -0.803 183s 183s x1: [4, 4, 1, 1] 183s 0.547 -0.722 0.152 -0.267 183s -0.208 -0.455 -0.645 0.339 183s -0.153 0.839 -0.901 -0.912 183s -0.199 0.225 0.346 0.203 183s 183s y: [4, 4, 1, 1] 183s -0.099 0.932 -0.418 0.147 183s -0.544 -0.177 -0.421 -1.415 183s -1.271 -0.813 1.016 0.674 183s 0.459 -0.119 0.111 0.146 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.612 0.093 0.118 -0.529 183s 183s x1: [4, 4, 1, 1] 183s 0.108 -0.467 -0.011 0.653 183s -0.112 -0.672 -0.500 -0.273 183s -0.573 -0.412 0.528 0.652 183s 0.934 -0.269 -0.494 0.327 183s 183s y: [1, 4, 1, 1] 183s -0.457 183s 0.091 183s 0.030 183s -0.828 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.138 -0.938 -0.349 -0.063 183s -0.741 0.759 0.470 0.248 183s 183s x1: [4, 1, 1, 1] 183s -0.297 0.556 -0.115 -0.857 183s 183s y: [2, 1, 1, 1] 183s -0.469 0.376 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.565 0.413 0.021 0.701 183s 0.345 -0.450 0.558 -0.218 183s -0.838 -0.145 -0.912 0.044 183s 0.762 -0.468 0.907 0.763 183s 183s x1: [4, 2, 1, 1] 183s 0.412 -0.642 0.576 0.913 183s 0.085 -0.996 -0.499 -0.387 183s 183s y: [4, 2, 1, 1] 183s 0.154 0.554 -0.738 1.834 183s -0.742 0.283 0.511 -0.217 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.331 -0.734 -0.931 -0.759 183s 0.214 0.219 -0.202 0.628 183s 183s x1: [4, 4, 1, 1] 183s 0.433 0.982 -0.978 0.192 183s 0.452 0.270 -0.396 0.810 183s -0.153 -0.483 -0.105 -0.149 183s 0.018 0.508 -0.493 0.452 183s 183s y: [2, 4, 1, 1] 183s -0.099 0.626 183s -0.594 0.744 183s 0.616 -0.211 183s -0.263 0.499 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.707 -0.375 183s 183s x1: [2, 4, 1, 1] 183s -0.114 -0.853 183s 0.329 0.668 183s 0.192 0.092 183s 0.101 0.173 183s 183s y: [1, 4, 1, 1] 183s 0.400 183s -0.483 183s -0.170 183s -0.136 183s test-mul-mat0: iter:220/500 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.244 -0.281 -0.205 0.642 183s -0.134 -0.876 0.310 -0.942 183s 0.216 -0.589 0.231 -0.670 183s -0.296 0.856 -0.285 0.601 183s 183s x1: [4, 3, 1, 1] 183s 0.414 0.330 -0.917 -0.568 183s -0.162 -0.410 0.884 0.691 183s 0.259 0.098 0.957 -0.522 183s 183s y: [4, 3, 1, 1] 183s -0.371 -0.093 0.064 0.079 183s 0.417 0.004 -0.052 -0.140 183s -0.622 0.668 0.569 -0.579 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s 0.076 -0.375 0.254 -0.110 183s 183s x1: [4, 4, 3, 1] 183s -0.710 0.947 0.015 0.620 183s -0.970 0.447 -0.542 -0.381 183s 0.331 -0.851 0.878 -0.571 183s -0.894 -0.643 0.596 -0.138 183s 183s y: [1, 4, 3, 1] 183s -0.473 183s -0.337 183s 0.629 183s 0.339 183s testing: mul_mat, [2, 1, 3, 3] = [4, 2, 3, 3] * [4, 1, 3, 3] 183s x0: [4, 2, 3, 3] 183s 0.585 -0.336 -0.105 0.019 183s 0.036 -0.180 0.203 0.119 183s 183s x1: [4, 1, 3, 3] 183s -0.426 -0.682 -0.454 -0.881 183s 183s y: [2, 1, 3, 3] 183s 0.011 -0.089 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.382 0.054 -0.931 0.001 183s -0.637 -0.974 0.820 0.716 183s -0.123 -0.306 0.451 -0.601 183s -0.960 -0.215 0.228 0.257 183s 183s x1: [4, 2, 1, 1] 183s -0.820 -0.064 0.564 0.210 183s -0.904 -0.888 -0.634 -0.869 183s 183s y: [4, 2, 1, 1] 183s -0.842 1.198 0.249 0.984 183s 0.197 0.299 0.619 0.691 183s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 2, 3, 1] 183s 0.980 -0.828 -0.585 -0.922 183s 0.465 0.119 -0.145 -0.183 183s 183s x1: [4, 4, 3, 1] 183s -0.385 -0.599 0.790 0.268 183s 0.586 -0.663 0.448 -0.478 183s 0.901 -0.342 -0.382 -0.987 183s 0.023 -0.251 0.711 -0.595 183s 183s y: [2, 4, 3, 1] 183s -0.590 -0.414 183s 1.302 0.216 183s 2.299 0.614 183s 0.363 -0.013 183s testing: mul_mat, [1, 4, 3, 3] = [2, 1, 3, 3] * [2, 4, 3, 3] 183s x0: [2, 1, 3, 3] 183s 0.182 -0.976 183s 183s x1: [2, 4, 3, 3] 183s -0.892 -0.651 183s -0.472 0.077 183s 0.884 0.827 183s -0.366 0.864 183s 183s y: [1, 4, 3, 3] 183s 0.473 183s -0.161 183s -0.646 183s -0.909 183s test-mul-mat0: iter:221/500 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.750 183s -0.164 183s -0.453 183s 183s x1: [1, 4, 1, 1] 183s -0.330 183s -0.083 183s 0.305 183s 0.623 183s 183s y: [3, 4, 1, 1] 183s 0.247 0.054 0.149 183s 0.062 0.014 0.038 183s -0.229 -0.050 -0.138 183s -0.467 -0.102 -0.282 183s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 1, 2, 1] 183s 0.315 183s 183s x1: [1, 3, 2, 1] 183s 0.399 183s -0.982 183s -0.428 183s 183s y: [1, 3, 2, 1] 183s 0.126 183s -0.309 183s -0.135 183s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.068 183s 183s x1: [1, 1, 2, 1] 183s -0.906 183s 183s y: [1, 1, 2, 1] 183s 0.062 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.106 183s 183s x1: [1, 1, 1, 1] 183s -0.218 183s 183s y: [1, 1, 1, 1] 183s 0.023 183s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.319 183s 0.204 183s 183s x1: [1, 1, 2, 1] 183s -0.369 183s 183s y: [2, 1, 2, 1] 183s -0.118 -0.075 183s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.552 0.122 183s -0.449 -0.540 183s -0.353 -0.842 183s -0.534 0.505 183s 183s x1: [2, 1, 2, 1] 183s -0.879 -0.002 183s 183s y: [4, 1, 2, 1] 183s 0.485 0.396 0.313 0.468 183s test-mul-mat0: iter:222/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.124 183s -0.027 183s 183s x1: [1, 1, 1, 1] 183s 0.769 183s 183s y: [2, 1, 1, 1] 183s -0.096 -0.021 183s testing: mul_mat, [2, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.640 183s 0.391 183s 183s x1: [1, 2, 4, 1] 183s -0.252 183s -0.135 183s 183s y: [2, 2, 4, 1] 183s -0.161 -0.098 183s -0.087 -0.053 183s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 4, 4, 1] 183s 0.887 183s 0.563 183s -0.443 183s -0.237 183s 183s x1: [1, 2, 4, 1] 183s 0.261 183s -0.697 183s 183s y: [4, 2, 4, 1] 183s 0.232 0.147 -0.116 -0.062 183s -0.618 -0.392 0.309 0.165 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.009 183s -0.319 183s 0.243 183s 0.496 183s 183s x1: [1, 4, 1, 1] 183s 0.145 183s 0.785 183s 0.969 183s 0.420 183s 183s y: [4, 4, 1, 1] 183s -0.001 -0.046 0.035 0.072 183s -0.007 -0.251 0.191 0.390 183s -0.009 -0.309 0.235 0.481 183s -0.004 -0.134 0.102 0.208 183s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.487 -0.749 -0.753 0.983 183s 0.828 -0.803 0.396 -0.784 183s 183s x1: [4, 1, 4, 1] 183s -0.915 0.627 -0.947 -0.584 183s 183s y: [2, 1, 4, 1] 183s 0.115 -1.177 183s testing: mul_mat, [2, 4, 4, 1] = [2, 2, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 2, 4, 1] 183s -0.457 0.981 183s -0.512 0.544 183s 183s x1: [2, 4, 4, 1] 183s -0.312 -0.009 183s 0.019 -0.065 183s -0.614 0.001 183s 0.152 0.202 183s 183s y: [2, 4, 4, 1] 183s 0.133 0.155 183s -0.072 -0.045 183s 0.281 0.315 183s 0.129 0.032 183s test-mul-mat0: iter:223/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.036 -0.319 183s -0.333 0.829 183s 0.285 -0.809 183s 183s x1: [2, 1, 1, 1] 183s -0.041 -0.184 183s 183s y: [3, 1, 1, 1] 183s 0.060 -0.139 0.137 183s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 1, 2, 1] 183s -0.915 -0.260 183s 183s x1: [2, 3, 2, 1] 183s 0.382 -0.121 183s -0.265 0.869 183s -0.140 -0.721 183s 183s y: [1, 3, 2, 1] 183s -0.318 183s 0.016 183s 0.316 183s testing: mul_mat, [3, 1, 2, 3] = [2, 3, 2, 3] * [2, 1, 2, 3] 183s x0: [2, 3, 2, 3] 183s -0.150 -0.328 183s -0.256 0.585 183s -0.458 0.604 183s 183s x1: [2, 1, 2, 3] 183s 0.268 -0.538 183s 183s y: [3, 1, 2, 3] 183s 0.136 -0.383 -0.448 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.540 0.913 183s 183s x1: [2, 3, 1, 1] 183s -0.303 0.016 183s -0.769 -0.655 183s -0.374 0.788 183s 183s y: [1, 3, 1, 1] 183s -0.149 183s -1.013 183s 0.518 183s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 1, 2, 1] 183s 0.522 183s 183s x1: [1, 2, 2, 1] 183s 0.573 183s 0.194 183s 183s y: [1, 2, 2, 1] 183s 0.300 183s 0.101 183s testing: mul_mat, [4, 1, 2, 3] = [1, 4, 2, 3] * [1, 1, 2, 3] 183s x0: [1, 4, 2, 3] 183s -0.638 183s -0.669 183s -0.142 183s 0.929 183s 183s x1: [1, 1, 2, 3] 183s -0.880 183s 183s y: [4, 1, 2, 3] 183s 0.561 0.589 0.125 -0.817 183s test-mul-mat0: iter:224/500 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.058 0.034 183s 0.698 0.210 183s 0.683 0.794 183s 183s x1: [2, 3, 1, 1] 183s -0.674 -0.243 183s -0.775 -0.312 183s 0.088 0.083 183s 183s y: [3, 3, 1, 1] 183s -0.047 -0.521 -0.653 183s -0.056 -0.606 -0.776 183s 0.008 0.079 0.126 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.932 -0.763 183s 183s x1: [2, 3, 1, 1] 183s 0.592 -0.324 183s -0.483 0.763 183s 0.585 -0.220 183s 183s y: [1, 3, 1, 1] 183s -0.305 183s -0.132 183s -0.377 183s testing: mul_mat, [1, 1, 1, 3] = [2, 1, 1, 3] * [2, 1, 1, 3] 183s x0: [2, 1, 1, 3] 183s 0.405 0.218 183s 183s x1: [2, 1, 1, 3] 183s 0.418 -0.696 183s 183s y: [1, 1, 1, 3] 183s 0.018 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.691 0.518 183s 0.561 0.244 183s -0.480 -0.686 183s 183s x1: [2, 1, 1, 1] 183s 0.250 -0.473 183s 183s y: [3, 1, 1, 1] 183s -0.418 0.025 0.204 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.573 -0.861 0.656 183s 0.847 0.877 -0.350 183s 183s x1: [3, 2, 1, 1] 183s 0.111 0.920 -0.169 183s -0.126 0.505 0.611 183s 183s y: [2, 2, 1, 1] 183s -0.839 0.959 183s -0.106 0.122 183s testing: mul_mat, [1, 3, 1, 3] = [2, 1, 1, 3] * [2, 3, 1, 3] 183s x0: [2, 1, 1, 3] 183s 0.070 -0.948 183s 183s x1: [2, 3, 1, 3] 183s 0.888 0.157 183s 0.841 0.293 183s -0.625 0.359 183s 183s y: [1, 3, 1, 3] 183s -0.087 183s -0.219 183s -0.384 183s test-mul-mat0: iter:225/500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.912 183s -0.677 183s 0.869 183s 0.206 183s 183s x1: [1, 1, 1, 1] 183s 0.166 183s 183s y: [4, 1, 1, 1] 183s 0.151 -0.112 0.144 0.034 183s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 2, 2, 1] 183s -0.231 183s -0.915 183s 183s x1: [1, 4, 2, 1] 183s -0.302 183s 0.227 183s -0.338 183s -0.483 183s 183s y: [2, 4, 2, 1] 183s 0.070 0.276 183s -0.052 -0.208 183s 0.078 0.310 183s 0.112 0.442 183s testing: mul_mat, [2, 2, 2, 4] = [1, 2, 2, 4] * [1, 2, 2, 4] 183s x0: [1, 2, 2, 4] 183s -0.146 183s -0.284 183s 183s x1: [1, 2, 2, 4] 183s 0.306 183s -0.440 183s 183s y: [2, 2, 2, 4] 183s -0.045 -0.087 183s 0.064 0.125 183s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.338 183s 183s x1: [1, 2, 1, 1] 183s 0.773 183s 0.536 183s 183s y: [1, 2, 1, 1] 183s 0.261 183s 0.181 183s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.086 183s 0.832 183s 183s x1: [1, 1, 2, 1] 183s -0.444 183s 183s y: [2, 1, 2, 1] 183s -0.038 -0.369 183s testing: mul_mat, [4, 1, 2, 4] = [2, 4, 2, 4] * [2, 1, 2, 4] 183s x0: [2, 4, 2, 4] 183s -0.507 -0.054 183s 0.410 0.705 183s 0.884 0.604 183s -0.733 0.502 183s 183s x1: [2, 1, 2, 4] 183s 0.554 -0.566 183s 183s y: [4, 1, 2, 4] 183s -0.250 -0.171 0.148 -0.690 183s test-mul-mat0: iter:226/500 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.504 183s 183s x1: [1, 4, 1, 1] 183s -0.544 183s -0.404 183s -0.355 183s -0.537 183s 183s y: [1, 4, 1, 1] 183s -0.274 183s -0.204 183s -0.179 183s -0.271 183s testing: mul_mat, [4, 1, 3, 1] = [1, 4, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 4, 3, 1] 183s 0.343 183s -0.948 183s 0.716 183s -0.486 183s 183s x1: [1, 1, 3, 1] 183s 0.910 183s 183s y: [4, 1, 3, 1] 183s 0.312 -0.863 0.652 -0.442 183s testing: mul_mat, [1, 4, 3, 4] = [1, 1, 3, 4] * [1, 4, 3, 4] 183s x0: [1, 1, 3, 4] 183s 0.640 183s 183s x1: [1, 4, 3, 4] 183s -0.123 183s -0.005 183s 0.480 183s -0.768 183s 183s y: [1, 4, 3, 4] 183s -0.079 183s -0.003 183s 0.307 183s -0.492 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.271 183s -0.808 183s 183s x1: [1, 1, 1, 1] 183s -0.849 183s 183s y: [2, 1, 1, 1] 183s 0.230 0.686 183s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 4, 3, 1] 183s 0.599 -0.479 183s -0.080 -0.727 183s -0.970 0.649 183s 0.319 -0.081 183s 183s x1: [2, 1, 3, 1] 183s -0.907 0.873 183s 183s y: [4, 1, 3, 1] 183s -0.962 -0.561 1.447 -0.360 183s testing: mul_mat, [1, 2, 3, 4] = [4, 1, 3, 4] * [4, 2, 3, 4] 183s x0: [4, 1, 3, 4] 183s 0.676 0.841 -0.210 -0.176 183s 183s x1: [4, 2, 3, 4] 183s -0.567 0.864 -0.540 -0.440 183s -0.505 -0.016 -0.489 -0.906 183s 183s y: [1, 2, 3, 4] 183s 0.535 183s -0.093 183s test-mul-mat0: iter:227/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.426 0.791 183s 0.633 -0.546 183s -0.064 -0.158 183s 183s x1: [2, 2, 1, 1] 183s -0.533 -0.280 183s 0.054 0.464 183s 183s y: [3, 2, 1, 1] 183s 0.005 -0.184 0.079 183s 0.344 -0.219 -0.077 183s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 2, 4, 1] 183s -0.278 -0.867 183s -0.109 0.176 183s 183s x1: [2, 3, 4, 1] 183s -0.825 -0.130 183s -0.125 -0.988 183s -0.405 0.918 183s 183s y: [2, 3, 4, 1] 183s 0.342 0.067 183s 0.891 -0.160 183s -0.684 0.206 183s testing: mul_mat, [4, 2, 4, 1] = [2, 4, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.517 -0.072 183s -0.993 0.249 183s -0.721 0.945 183s 0.857 0.643 183s 183s x1: [2, 2, 4, 1] 183s -0.163 -0.943 183s -0.166 0.995 183s 183s y: [4, 2, 4, 1] 183s -0.016 -0.073 -0.774 -0.747 183s -0.158 0.412 1.060 0.498 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.559 0.393 183s -0.409 -0.951 183s -0.173 -0.912 183s 0.406 -0.810 183s 183s x1: [2, 4, 1, 1] 183s 0.172 -0.800 183s 0.422 0.452 183s -0.855 0.279 183s 0.095 -0.375 183s 183s y: [4, 4, 1, 1] 183s -0.218 0.690 0.700 0.718 183s 0.414 -0.602 -0.485 -0.194 183s -0.368 0.085 -0.106 -0.573 183s -0.094 0.318 0.326 0.342 183s testing: mul_mat, [4, 2, 4, 1] = [4, 4, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 4, 4, 1] 183s -0.754 0.063 -0.383 -0.663 183s -0.818 -0.274 -0.810 -0.944 183s 0.492 0.214 0.528 -0.832 183s -0.517 0.199 -0.319 0.937 183s 183s x1: [4, 2, 4, 1] 183s -0.552 -0.459 -0.227 -0.568 183s 0.204 -0.762 -0.119 0.234 183s 183s y: [4, 2, 4, 1] 183s 0.850 1.297 -0.017 -0.265 183s -0.312 -0.083 -0.321 0.000 183s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 4, 4, 1] 183s 0.949 0.173 0.580 0.859 183s -0.759 -0.699 -0.442 0.779 183s 0.957 0.115 0.545 -0.337 183s -0.666 0.247 -0.784 0.574 183s 183s x1: [4, 4, 4, 1] 183s -0.915 -0.428 0.499 ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s -0.117 183s 0.190 0.607 -0.701 0.997 183s 0.243 -0.604 0.142 -0.065 183s -0.339 -0.776 0.358 0.858 183s 183s y: [4, 4, 4, 1] 183s -0.753 0.682 -0.614 0.046 183s 0.735 0.519 -0.466 1.145 183s 0.152 0.124 0.262 -0.460 183s 0.489 1.310 -0.508 0.246 183s test-mul-mat0: iter:228/500 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.034 183s 183s x1: [1, 3, 1, 1] 183s 0.992 183s 0.828 183s 0.575 183s 183s y: [1, 3, 1, 1] 183s -0.034 183s -0.028 183s -0.019 183s testing: mul_mat, [1, 1, 3, 1] = [1, 1, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 1, 3, 1] 183s -0.697 183s 183s x1: [1, 1, 3, 1] 183s -0.960 183s 183s y: [1, 1, 3, 1] 183s 0.669 183s testing: mul_mat, [4, 1, 3, 4] = [1, 4, 3, 4] * [1, 1, 3, 4] 183s x0: [1, 4, 3, 4] 183s -0.659 183s -0.766 183s -0.087 183s 0.884 183s 183s x1: [1, 1, 3, 4] 183s 0.611 183s 183s y: [4, 1, 3, 4] 183s -0.403 -0.468 -0.053 0.541 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.327 183s 0.578 183s 183s x1: [1, 4, 1, 1] 183s 0.153 183s 0.507 183s -0.715 183s 0.445 183s 183s y: [2, 4, 1, 1] 183s 0.050 0.089 183s 0.166 0.293 183s -0.234 -0.413 183s 0.145 0.257 183s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 4, 3, 1] 183s 0.815 -0.418 183s 0.776 -0.019 183s 0.132 -0.674 183s -0.868 0.685 183s 183s x1: [2, 1, 3, 1] 183s -0.611 0.289 183s 183s y: [4, 1, 3, 1] 183s -0.619 -0.480 -0.275 0.728 183s testing: mul_mat, [2, 2, 3, 4] = [4, 2, 3, 4] * [4, 2, 3, 4] 183s x0: [4, 2, 3, 4] 183s 0.255 0.642 0.445 0.160 183s 0.424 0.358 0.288 -0.289 183s 183s x1: [4, 2, 3, 4] 183s 0.863 -0.532 -0.778 -0.307 183s 0.850 0.661 -0.783 0.665 183s 183s y: [2, 2, 3, 4] 183s -0.518 0.041 183s 0.399 0.180 183s test-mul-mat0: iter:229/500 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.934 -0.386 183s 0.326 0.611 183s -0.180 -0.395 183s 0.390 0.372 183s 183s x1: [2, 2, 1, 1] 183s -0.545 0.792 183s 0.682 -0.908 183s 183s y: [4, 2, 1, 1] 183s -0.815 0.306 -0.215 0.083 183s 0.988 -0.332 0.236 -0.072 183s testing: mul_mat, [4, 4, 2, 1] = [2, 4, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.592 0.829 183s 0.846 -0.910 183s 0.921 -0.177 183s -0.976 -0.465 183s 183s x1: [2, 4, 2, 1] 183s -0.657 0.045 183s -0.550 -0.186 183s -0.527 0.324 183s 0.914 -0.584 183s 183s y: [4, 4, 2, 1] 183s 0.426 -0.597 -0.613 0.620 183s 0.172 -0.296 -0.473 0.623 183s 0.581 -0.741 -0.543 0.364 183s -1.025 1.305 0.945 -0.620 183s testing: mul_mat, [4, 4, 2, 4] = [2, 4, 2, 4] * [2, 4, 2, 4] 183s x0: [2, 4, 2, 4] 183s 0.260 0.212 183s 0.716 0.011 183s -0.586 0.162 183s -0.407 0.129 183s 183s x1: [2, 4, 2, 4] 183s 0.177 0.911 183s -0.458 -0.499 183s 0.825 -0.042 183s -0.180 0.991 183s 183s y: [4, 4, 2, 4] 183s 0.239 0.137 0.043 0.045 183s -0.225 -0.333 0.188 0.122 183s 0.206 0.590 -0.490 -0.341 183s 0.163 -0.119 0.266 0.201 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.688 0.879 183s 0.277 -0.235 183s -0.442 0.015 183s -0.228 0.890 183s 183s x1: [2, 4, 1, 1] 183s -0.401 -0.700 183s -0.239 0.934 183s -0.837 0.191 183s -0.372 -0.625 183s 183s y: [4, 4, 1, 1] 183s -0.891 0.054 0.166 -0.532 183s 0.656 -0.285 0.120 0.886 183s -0.408 -0.276 0.373 0.361 183s -0.805 0.044 0.155 -0.471 183s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.416 0.212 -0.100 0.671 183s -0.774 -0.399 0.878 0.757 183s 0.288 0.639 -0.802 -0.037 183s 183s x1: [4, 2, 2, 1] 183s 0.095 0.180 0.350 -0.805 183s -0.596 0.654 0.965 0.302 183s 183s y: [3, 2, 2, 1] 183s -0.577 -0.448 -0.108 183s 0.493 1.277 -0.539 183s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 183s x0: [3, 2, 2, 4] 183s 0.375 -0.840 -0.091 183s -0.206 0.687 -0.537 183s 183s x1: [3, 4, 2, 4] 183s 0.448 -0.675 -0.284 183s -0.215 -0.025 -0.221 183s -0.118 0.559 0.005 183s -0.830 -0.229 0.606 183s 183s y: [2, 4, 2, 4] 183s 0.761 -0.403 183s -0.040 0.146 183s -0.514 0.406 183s -0.174 -0.312 183s test-mul-mat0: iter:230/500 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.326 -0.396 0.770 183s -0.445 -0.044 -0.392 183s 183s x1: [3, 4, 1, 1] 183s -0.893 0.448 0.458 183s 0.397 0.096 -0.100 183s 0.135 0.476 0.969 183s 0.752 0.231 0.832 183s 183s y: [2, 4, 1, 1] 183s 0.467 0.197 183s -0.244 -0.141 183s 0.513 -0.461 183s 0.303 -0.671 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.865 -0.504 0.567 183s 0.864 -0.085 0.674 183s 0.312 -0.627 0.071 183s 183s x1: [3, 2, 1, 1] 183s 0.739 0.595 0.015 183s -0.458 -0.488 -0.227 183s 183s y: [3, 2, 1, 1] 183s 0.348 0.599 -0.141 183s -0.279 -0.508 0.147 183s testing: mul_mat, [1, 3, 1, 4] = [3, 1, 1, 4] * [3, 3, 1, 4] 183s x0: [3, 1, 1, 4] 183s 0.188 -0.975 -0.655 183s 183s x1: [3, 3, 1, 4] 183s -0.592 0.273 -0.795 183s 0.884 0.242 0.957 183s 0.115 0.074 -0.514 183s 183s y: [1, 3, 1, 4] 183s 0.143 183s -0.697 183s 0.286 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.915 0.204 -0.317 183s 0.142 -0.013 -0.642 183s -0.638 -0.832 0.348 183s -0.615 0.639 -0.416 183s 183s x1: [3, 1, 1, 1] 183s -0.973 0.308 0.995 183s 183s y: [4, 1, 1, 1] 183s -1.143 -0.781 0.710 0.382 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.069 -0.810 0.120 -0.293 183s -0.805 -0.242 -0.461 -0.825 183s 0.346 -0.846 -0.464 -0.210 183s -0.760 0.332 -0.893 0.065 183s 183s x1: [4, 3, 1, 1] 183s -0.568 -0.701 -0.558 0.620 183s -0.676 -0.213 -0.403 -0.973 183s 0.388 0.917 -0.635 -0.148 183s 183s y: [4, 3, 1, 1] 183s 0.358 0.372 0.524 0.738 183s 0.456 1.584 0.338 0.741 183s -0.802 -0.120 -0.315 0.567 183s testing: mul_mat, [1, 4, 1, 4] = [4, 1, 1, 4] * [4, 4, 1, 4] 183s x0: [4, 1, 1, 4] 183s -0.696 -0.343 -0.339 0.068 183s 183s x1: [4, 4, 1, 4] 183s 0.523 0.373 0.496 0.822 183s 0.816 0.117 -0.854 -0.397 183s 0.714 -0.826 0.991 0.630 183s -0.462 -0.157 -0.050 0.469 183s 183s y: [1, 4, 1, 4] 183s -0.605 183s -0.345 183s -0.506 183s 0.424 183s test-mul-mat0: iter:231/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.284 -0.135 -0.803 -0.555 183s -0.130 -0.430 0.738 -0.649 183s -0.362 -0.889 -0.225 0.442 183s 183s x1: [4, 3, 1, 1] 183s -0.215 -0.484 -0.086 0.670 183s -0.455 0.806 0.624 -0.251 183s 0.493 -0.978 -0.377 0.798 183s 183s y: [3, 3, 1, 1] 183s -0.298 -0.262 0.824 183s -0.599 0.336 -0.804 183s 0.132 -0.439 1.129 183s testing: mul_mat, [2, 3, 4, 1] = [4, 2, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.778 -0.533 0.640 -0.202 183s -0.921 -0.109 -0.613 0.429 183s 183s x1: [4, 3, 4, 1] 183s 0.861 0.854 -0.789 -0.958 183s 0.450 0.504 -0.172 0.966 183s -0.582 -0.502 -0.489 -0.776 183s 183s y: [2, 3, 4, 1] 183s -0.096 -0.813 183s -0.224 0.051 183s -0.341 0.558 183s testing: mul_mat, [1, 2, 4, 2] = [4, 1, 4, 2] * [4, 2, 4, 2] 183s x0: [4, 1, 4, 2] 183s 0.960 0.992 -0.161 0.349 183s 183s x1: [4, 2, 4, 2] 183s 0.773 -0.980 -0.197 0.852 183s -0.088 0.189 0.281 0.405 183s 183s y: [1, 2, 4, 2] 183s 0.099 183s 0.199 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.343 -0.229 -0.579 -0.149 183s 0.758 0.063 -0.353 0.785 183s 0.333 -0.738 0.904 0.998 183s 0.508 0.957 -0.443 -0.665 183s 183s x1: [4, 1, 1, 1] 183s -0.597 0.102 0.748 -0.240 183s 183s y: [4, 1, 1, 1] 183s -0.216 -0.898 0.163 -0.377 183s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 4, 4, 1] 183s -0.616 -0.168 0.049 0.352 183s -0.581 -0.257 0.132 0.633 183s 0.698 0.517 -0.820 0.062 183s 0.850 0.833 -0.956 0.686 183s 183s x1: [4, 4, 4, 1] 183s -0.223 -0.606 -0.215 0.364 183s 0.773 0.894 0.803 -0.816 183s -0.843 -0.799 -0.413 0.260 183s 0.948 0.347 -0.310 -0.395 183s 183s y: [4, 4, 4, 1] 183s 0.356 0.487 -0.270 -0.239 183s -0.874 -1.089 0.293 0.075 183s 0.724 0.804 -0.646 -0.809 183s -0.796 -0.931 1.071 1.121 183s testing: mul_mat, [2, 4, 4, 2] = [4, 2, 4, 2] * [4, 4, 4, 2] 183s x0: [4, 2, 4, 2] 183s 0.324 0.954 -0.524 0.725 183s -0.855 0.815 0.925 0.878 183s 183s x1: [4, 4, 4, 2] 183s -0.697 0.931 -0.689 0.103 183s 0.847 0.768 -0.049 -0.105 183s 0.405 -0.157 0.294 -0.325 183s 0.798 0.380 -0.094 0.917 183s 183s y: [2, 4, 4, 2] 183s 1.098 0.808 183s 0.956 -0.237 183s -0.408 -0.487 183s 1.334 0.346 183s test-mul-mat0: iter:232/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.238 183s -0.909 183s 183s x1: [1, 1, 1, 1] 183s 0.430 183s 183s y: [2, 1, 1, 1] 183s -0.102 -0.391 183s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.997 183s 0.764 183s 183s x1: [1, 2, 3, 1] 183s 0.388 183s -0.810 183s 183s y: [2, 2, 3, 1] 183s -0.387 0.297 183s 0.808 -0.619 183s testing: mul_mat, [4, 2, 3, 4] = [1, 4, 3, 4] * [1, 2, 3, 4] 183s x0: [1, 4, 3, 4] 183s 0.292 183s -0.566 183s 0.431 183s -0.388 183s 183s x1: [1, 2, 3, 4] 183s -0.218 183s 0.420 183s 183s y: [4, 2, 3, 4] 183s -0.064 0.123 -0.094 0.085 183s 0.123 -0.238 0.181 -0.163 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.630 183s -0.418 183s 0.391 183s 183s x1: [1, 4, 1, 1] 183s -0.794 183s 0.398 183s -0.969 183s -0.837 183s 183s y: [3, 4, 1, 1] 183s -0.500 0.332 -0.310 183s 0.251 -0.166 0.156 183s -0.611 0.405 -0.379 183s -0.528 0.350 -0.327 183s testing: mul_mat, [1, 1, 3, 1] = [3, 1, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.258 -0.937 -0.432 183s 183s x1: [3, 1, 3, 1] 183s 0.888 -0.411 0.714 183s 183s y: [1, 1, 3, 1] 183s 0.306 183s testing: mul_mat, [2, 3, 3, 4] = [1, 2, 3, 4] *ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s [1, 3, 3, 4] 183s x0: [1, 2, 3, 4] 183s -0.339 183s 0.596 183s 183s x1: [1, 3, 3, 4] 183s -0.816 183s 0.801 183s 0.357 183s 183s y: [2, 3, 3, 4] 183s 0.276 -0.486 183s -0.271 0.477 183s -0.121 0.213 183s test-mul-mat0: iter:233/500 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.739 0.046 183s 183s x1: [2, 3, 1, 1] 183s 0.385 -0.682 183s -0.967 -0.954 183s 0.914 0.817 183s 183s y: [1, 3, 1, 1] 183s -0.316 183s 0.671 183s -0.638 183s testing: mul_mat, [4, 1, 4, 1] = [2, 4, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.425 0.058 183s 0.995 0.730 183s -0.843 0.848 183s 0.444 0.534 183s 183s x1: [2, 1, 4, 1] 183s 0.494 0.950 183s 183s y: [4, 1, 4, 1] 183s -0.154 1.185 0.389 0.726 183s testing: mul_mat, [3, 4, 4, 3] = [2, 3, 4, 3] * [2, 4, 4, 3] 183s x0: [2, 3, 4, 3] 183s -0.383 -0.363 183s -0.594 -0.922 183s -0.892 0.753 183s 183s x1: [2, 4, 4, 3] 183s 0.489 -0.598 183s 0.716 0.646 183s -0.750 0.160 183s 0.180 0.571 183s 183s y: [3, 4, 4, 3] 183s 0.030 0.260 -0.886 183s -0.508 -1.021 -0.152 183s 0.229 0.299 0.789 183s -0.276 -0.633 0.269 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.690 -0.254 183s 0.551 -0.322 183s 0.472 0.402 183s -0.832 -0.548 183s 183s x1: [2, 3, 1, 1] 183s 0.156 0.958 183s -0.148 -0.508 183s -0.510 0.466 183s 183s y: [4, 3, 1, 1] 183s -0.351 -0.222 0.459 -0.655 183s 0.231 0.082 -0.274 0.402 183s 0.234 -0.431 -0.053 0.169 183s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 3, 4, 1] 183s -0.101 0.955 0.685 -0.773 183s 0.860 0.469 -0.903 -0.222 183s -0.407 0.921 0.102 -0.832 183s 183s x1: [4, 2, 4, 1] 183s 0.981 -0.188 -0.235 0.419 183s 0.241 0.056 0.609 0.329 183s 183s y: [3, 2, 4, 1] 183s -0.764 0.875 -0.946 183s 0.192 -0.389 -0.258 183s testing: mul_mat, [2, 4, 4, 3] = [3, 2, 4, 3] * [3, 4, 4, 3] 183s x0: [3, 2, 4, 3] 183s -0.430 -0.418 -0.600 183s -0.318 -0.401 0.205 183s 183s x1: [3, 4, 4, 3] 183s -0.456 -0.014 0.066 183s 0.959 0.223 0.154 183s 0.155 -0.403 -0.075 183s -0.188 -0.489 -0.169 183s 183s y: [2, 4, 4, 3] 183s 0.162 0.164 183s -0.598 -0.363 183s 0.147 0.097 183s 0.387 0.222 183s test-mul-mat0: iter:234/500 183s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.345 -0.436 183s -0.675 0.495 183s 0.102 -0.703 183s 183s x1: [2, 4, 1, 1] 183s -0.647 -0.414 183s -0.227 0.894 183s -0.072 -0.393 183s 0.265 -0.836 183s 183s y: [3, 4, 1, 1] 183s -0.043 0.232 0.225 183s -0.468 0.596 -0.652 183s 0.146 -0.146 0.269 183s 0.455 -0.593 0.615 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.861 0.338 183s 0.573 -0.700 183s 183s x1: [2, 3, 1, 1] 183s -0.244 0.339 183s -0.549 0.808 183s 0.857 0.389 183s 183s y: [2, 3, 1, 1] 183s -0.096 -0.377 183s -0.200 -0.880 183s 0.869 0.219 183s testing: mul_mat, [2, 2, 1, 3] = [2, 2, 1, 3] * [2, 2, 1, 3] 183s x0: [2, 2, 1, 3] 183s -0.877 -0.312 183s -0.554 -0.552 183s 183s x1: [2, 2, 1, 3] 183s 0.476 -0.679 183s 0.830 -0.797 183s 183s y: [2, 2, 1, 3] 183s -0.205 0.112 183s -0.479 -0.020 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.096 -0.266 183s 0.397 0.600 183s 0.692 -0.533 183s 183s x1: [2, 2, 1, 1] 183s -0.416 0.798 183s -0.396 0.922 183s 183s y: [3, 2, 1, 1] 183s -0.172 0.313 -0.713 183s -0.207 0.396 -0.765 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.987 0.025 -0.257 183s -0.705 -0.567 0.208 183s 183s x1: [3, 2, 1, 1] 183s -0.083 -0.611 -0.984 183s -0.005 -0.894 -0.828 183s 183s y: [2, 2, 1, 1] 183s 0.156 0.200 183s 0.185 0.339 183s testing: mul_mat, [3, 3, 1, 3] = [2, 3, 1, 3] * [2, 3, 1, 3] 183s x0: [2, 3, 1, 3] 183s -0.425 -0.319 183s -0.763 -0.591 183s 0.765 -0.897 183s 183s x1: [2, 3, 1, 3] 183s 0.980 0.487 183s -0.853 -0.133 183s -0.318 -0.107 183s 183s y: [3, 3, 1, 3] 183s -0.572 -1.036 0.313 183s 0.405 0.730 -0.533 183s 0.169 0.306 -0.147 183s test-mul-mat0: iter:235/500 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.418 -0.662 0.279 -0.902 183s -0.253 0.382 -0.234 0.451 183s 183s x1: [4, 2, 1, 1] 183s -0.929 -0.872 -0.109 -0.343 183s 0.881 -0.256 -0.157 -0.899 183s 183s y: [2, 2, 1, 1] 183s 0.468 -0.227 183s 1.304 -0.689 183s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 1, 4, 1] 183s -0.687 0.392 0.000 -0.855 183s 183s x1: [4, 2, 4, 1] 183s -0.490 0.867 -0.102 -0.154 183s -0.052 0.832 0.329 -0.104 183s 183s y: [1, 2, 4, 1] 183s 0.809 183s 0.452 183s testing: mul_mat, [1, 1, 4, 3] = [4, 1, 4, 3] * [4, 1, 4, 3] 183s x0: [4, 1, 4, 3] 183s -0.662 0.500 0.892 0.738 183s 183s x1: [4, 1, 4, 3] 183s 0.872 0.555 0.290 -0.249 183s 183s y: [1, 1, 4, 3] 183s -0.224 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.476 0.107 -0.779 0.673 183s 0.962 0.851 -0.044 0.561 183s 0.011 0.655 0.026 -0.711 183s 183s x1: [4, 1, 1, 1] 183s -0.028 0.828 -0.189 0.726 183s 183s y: [3, 1, 1, 1] 183s 0.737 1.094 0.021 183s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 2, 4, 1] 183s 0.461 -0.125 -0.418 183s -0.841 0.728 -0.954 183s 183s x1: [3, 4, 4, 1] 183s -0.138 0.375 -0.040 183s 0.114 -0.697 0.129 183s -0.920 0.342 -0.418 183s -0.201 0.361 -0.341 183s 183s y: [2, 4, 4, 1] 183s -0.093 0.428 183s 0.086 -0.727 183s -0.292 1.421 183s 0.005 0.757 183s testing: mul_mat, [1, 3, 4, 3] = [2, 1, 4, 3] * [2, 3, 4, 3] 183s x0: [2, 1, 4, 3] 183s 0.637 0.318 183s 183s x1: [2, 3, 4, 3] 183s 0.836 0.330 183s 0.120 -0.965 183s 0.534 0.963 183s 183s y: [1, 3, 4, 3] 183s 0.638 183s -0.230 183s 0.647 183s test-mul-mat0: iter:236/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.731 0.109 183s 183s x1: [2, 4, 1, 1] 183s -0.130 -0.181 183s -0.578 -0.493 183s -0.864 0.999 183s 0.817 -0.949 183s 183s y: [1, 4, 1, 1] 183s 0.075 183s 0.369 183s 0.740 183s -0.701 183s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 1, 2, 1] 183s -0.160 -0.340 183s 183s x1: [2, 1, 2, 1] 183s 0.187 -0.946 183s 183s y: [1, 1, 2, 1] 183s 0.292 183s testing: mul_mat, [2, 1, 2, 1] = [2, 2, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.650 0.025 183s 0.176 0.326 183s 183s x1: [2, 1, 2, 1] 183s 0.277 0.530 183s 183s y: [2, 1, 2, 1] 183s -0.167 0.221 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.444 -0.447 183s 0.193 -0.620 183s -0.211 -0.156 183s 183s x1: [2, 2, 1, 1] 183s -0.456 -0.297 183s 0.393 -0.457 183s 183s y: [3, 2, 1, 1] 183s -0.070 0.096 0.142 183s 0.379 0.359 -0.011 183s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 2, 2, 1] 183s 0.413 -0.876 0.847 183s 0.725 0.851 -0.681 183s 183s x1: [3, 2, 2, 1] 183s 0.337 -0.165 0.219 183s -0.823 0.495 0.039 183s 183s y: [2, 2, 2, 1] 183s 0.470 -0.045 183s -0.741 -0.202 183s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.609 -0.997 183s 183s x1: [2, 3, 2, 1] 183s 0.946 -0.872 183s -0.668 -0.501 183s -0.493 0.176 183s 183s y: [1, 3, 2, 1] 183s 1.446 183s 0.092 183s -0.475 183s test-mul-mat0: iter:237/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.317 -0.914 0.998 0.153 183s 0.428 -0.607 0.167 -0.734 183s 0.827 0.402 0.875 0.830 183s 183s x1: [4, 3, 1, 1] 183s 0.605 0.262 -0.876 0.924 183s -0.630 -0.484 -0.053 -0.128 183s 0.451 -0.510 -0.183 0.579 183s 183s y: [3, 3, 1, 1] 183s -0.780 -0.725 0.607 183s 0.170 0.109 -0.868 183s 0.516 0.048 0.488 183s testing: mul_mat, [2, 3, 3, 1] = [4, 2, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 2, 3, 1] 183s -0.537 0.681 0.224 0.558 183s -0.404 -0.476 0.600 -0.618 183s 183s x1: [4, 3, 3, 1] 183s -0.915 0.872 -0.618 0.566 183s 0.282 -0.341 0.171 -0.456 183s -0.217 0.095 -0.086 0.300 183s 183s y: [2, 3, 3, 1] 183s 1.263 -0.766 183s -0.600 0.433 183s 0.329 -0.194 183s testing: mul_mat, [2, 2, 3, 4] = [4, 2, 3, 4] * [4, 2, 3, 4] 183s x0: [4, 2, 3, 4] 183s -0.919 -0.167 -0.954 0.800 183s -0.370 0.253 0.773 0.238 183s 183s x1: [4, 2, 3, 4] 183s 0.552 0.586 0.809 0.335 183s -0.122 -0.054 0.776 0.341 183s 183s y: [2, 2, 3, 4] 183s -1.109 0.649 183s -0.346 0.712 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.408 0.781 -0.174 0.695 183s 0.340 -0.206 0.504 0.580 183s -0.939 0.730 0.747 -0.831 183s -0.777 -0.907 -0.340 0.143 183s 183s x1: [4, 2, 1, 1] 183s 0.761 -0.339 -0.207 0.787 183s 0.331 -0.849 -0.475 0.461 183s 183s y: [4, 2, 1, 1] 183s 0.008 0.681 -1.771 -0.101 183s -0.395 0.316 -1.668 0.741 183s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 2, 3, 1] 183s -0.286 -0.919 -0.002 0.652 183s 0.287 0.431 -0.051 -0.545 183s 183s x1: [4, 4, 3, 1] 183s -0.849 0.853 -0.122 -0.902 183s 0.045 0.978 0.859 0.705 183s -0.230 0.646 0.036 -0.079 183s -0.829 -0.503 0.186 -0.236 183s 183s y: [2, 4, 3, 1] 183s -1.128 0.621 183s -0.453 0.006 183s -0.579 0.253 183s 0.545 -0.335 183s testing: mul_mat, [4, 4, 3, 4] = [2, 4, 3, 4] * [2, 4, 3, 4] 183s x0: [2, 4, 3, 4] 183s 0.088 -0.927 183s 0.743 0.863 183s -0.511 -0.694 183s 0.258 -0.995 183s 183s x1: [2, 4, 3, 4] 183s -0.224 -0.129 183s 0.783 0.569 183s -0.402 0.452 183s 0.609 0.312 183s 183s y: [4, 4, 3, 4] 183s 0.100 -0.278 0.204 0.071 183s -0.459 1.073 -0.794 -0.364 183s -0.454 0.092 -0.108 -0.554 183s -0.235 0.721 -0.527 -0.153 183s test-mul-mat0: iter:238/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.419 0.435 0.938 0.909 183s -0.666 0.873 -0.125 0.975 183s -0.687 -0.842 -0.386 0.318 183s 183s x1: [4, 3, 1, 1] 183s 0.950 -0.985 -0.ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s 840 -0.443 183s 0.238 -0.649 0.492 0.568 183s -0.475 -0.255 -0.154 -0.408 183s 183s y: [3, 3, 1, 1] 183s -1.220 -1.821 0.359 183s 0.794 -0.233 0.374 183s -0.824 -0.284 0.470 183s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.197 0.049 0.383 0.781 183s -0.976 -0.497 0.857 -0.433 183s -0.853 -0.321 0.957 -0.617 183s 0.036 -0.234 -0.945 0.330 183s 183s x1: [4, 3, 3, 1] 183s 0.341 -0.006 -0.791 -0.923 183s 0.723 -0.392 -0.972 0.738 183s -0.232 -0.416 -0.024 0.119 183s 183s y: [4, 3, 3, 1] 183s -0.957 -0.609 -0.477 0.457 183s 0.326 -1.663 -1.877 1.280 183s 0.018 0.360 0.234 0.151 183s testing: mul_mat, [2, 4, 3, 2] = [4, 2, 3, 2] * [4, 4, 3, 2] 183s x0: [4, 2, 3, 2] 183s -0.698 -0.599 0.087 -0.139 183s 0.028 -0.915 -0.705 -0.007 183s 183s x1: [4, 4, 3, 2] 183s -0.071 -0.537 -0.546 -0.031 183s -0.560 0.281 0.603 -0.785 183s 0.199 0.825 0.998 -0.714 183s 0.357 -0.605 -0.431 0.938 183s 183s y: [2, 4, 3, 2] 183s 0.328 0.874 183s 0.384 -0.692 183s -0.446 -1.447 183s -0.054 0.861 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.023 -0.085 -0.349 -0.213 183s -0.785 0.933 0.987 -0.215 183s -0.924 -0.821 -0.424 0.674 183s 0.400 -0.297 0.664 -0.551 183s 183s x1: [4, 2, 1, 1] 183s -0.477 -0.623 -0.748 0.021 183s 0.741 0.587 0.471 0.500 183s 183s y: [4, 2, 1, 1] 183s 0.299 -0.949 1.284 -0.514 183s -0.304 0.322 -1.030 0.160 183s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 3, 3, 1] 183s 0.603 -0.065 -0.746 -0.640 183s -0.995 -0.373 0.824 0.807 183s 0.804 0.925 0.709 0.167 183s 183s x1: [4, 4, 3, 1] 183s -0.227 -0.174 0.386 0.715 183s 0.171 0.102 -0.762 0.548 183s 0.353 0.259 0.289 -0.060 183s -0.270 -0.211 0.461 0.753 183s 183s y: [3, 4, 3, 1] 183s -0.872 1.186 0.050 183s 0.314 -0.394 -0.217 183s 0.019 -0.258 0.719 183s -0.975 1.335 0.041 183s testing: mul_mat, [2, 4, 3, 2] = [3, 2, 3, 2] * [3, 4, 3, 2] 183s x0: [3, 2, 3, 2] 183s -0.736 -0.106 -0.434 183s -0.018 -0.978 -0.315 183s 183s x1: [3, 4, 3, 2] 183s -0.511 0.648 0.830 183s -0.802 -0.992 0.636 183s 0.366 -0.078 -0.802 183s -0.182 0.777 0.231 183s 183s y: [2, 4, 3, 2] 183s -0.053 -0.886 183s 0.418 0.785 183s 0.088 0.322 183s -0.048 -0.829 183s test-mul-mat0: iter:239/500 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.582 0.245 0.929 183s 0.568 0.258 -0.445 183s 0.478 -0.912 -0.896 183s 183s x1: [3, 3, 1, 1] 183s -0.882 0.962 0.753 183s 0.566 0.585 0.429 183s -0.120 -0.839 -0.353 183s 183s y: [3, 3, 1, 1] 183s 0.422 -0.589 -1.974 183s 0.871 0.282 -0.647 183s -0.603 -0.127 1.025 183s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.959 0.127 0.331 183s 183s x1: [3, 3, 3, 1] 183s -0.135 0.954 -0.277 183s -0.483 -0.198 -0.763 183s -0.808 -0.784 -0.564 183s 183s y: [1, 3, 3, 1] 183s -0.099 183s -0.740 183s -1.061 183s testing: mul_mat, [2, 1, 3, 3] = [3, 2, 3, 3] * [3, 1, 3, 3] 183s x0: [3, 2, 3, 3] 183s 0.328 0.236 0.522 183s -0.139 0.336 -0.749 183s 183s x1: [3, 1, 3, 3] 183s 0.533 -0.900 -0.272 183s 183s y: [2, 1, 3, 3] 183s -0.180 -0.173 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.532 -0.515 0.050 183s -0.436 -0.140 -0.659 183s 0.730 0.531 -0.212 183s -0.594 0.125 0.608 183s 183s x1: [3, 2, 1, 1] 183s -0.827 -0.941 0.773 183s 0.378 -0.820 0.504 183s 183s y: [4, 2, 1, 1] 183s 0.083 -0.016 -1.267 0.843 183s 0.649 -0.382 -0.266 -0.021 183s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.657 0.675 0.388 0.563 183s 0.226 0.068 -0.098 0.089 183s -0.480 -0.823 -0.130 0.601 183s -0.169 -0.818 0.213 0.953 183s 183s x1: [4, 3, 3, 1] 183s -0.267 0.786 0.014 -0.510 183s 0.319 0.077 0.936 0.334 183s -0.479 0.174 -0.552 -0.452 183s 183s y: [4, 3, 3, 1] 183s 0.073 -0.054 -0.826 -1.080 183s 0.813 0.016 -0.138 0.400 183s -0.666 -0.083 -0.113 -0.610 183s testing: mul_mat, [3, 4, 3, 3] = [4, 3, 3, 3] * [4, 4, 3, 3] 183s x0: [4, 3, 3, 3] 183s -0.137 -0.440 -0.840 -0.867 183s -0.347 0.028 -0.297 -0.395 183s -0.743 -0.083 0.506 -0.250 183s 183s x1: [4, 4, 3, 3] 183s -0.403 0.608 -0.645 0.071 183s 0.111 -0.276 -0.668 -0.477 183s -0.676 -0.834 -0.144 0.645 183s 0.892 0.189 -0.378 -0.155 183s 183s y: [3, 4, 3, 3] 183s 0.268 0.321 -0.095 183s 1.082 0.341 -0.279 183s 0.022 -0.001 0.338 183s 0.247 -0.131 -0.831 183s test-mul-mat0: iter:240/500 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.471 0.535 0.662 0.201 183s -0.532 0.064 -0.958 -0.787 183s 183s x1: [4, 1, 1, 1] 183s -0.141 0.244 0.610 0.887 183s 183s y: [2, 1, 1, 1] 183s 0.779 -1.192 183s testing: mul_mat, [2, 2, 1, ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.000 -0.463 0.679 0.162 183s 0.056 -0.646 -0.233 -0.806 183s 183s x1: [4, 2, 1, 1] 183s -0.249 0.728 -0.145 -0.472 183s 0.849 -0.359 -0.093 0.503 183s 183s y: [2, 2, 1, 1] 183s -0.512 -0.070 183s 0.185 -0.104 183s testing: mul_mat, [4, 2, 1, 4] = [4, 4, 1, 4] * [4, 2, 1, 4] 183s x0: [4, 4, 1, 4] 183s -0.366 -0.289 -0.818 0.836 183s 0.796 0.230 -0.300 -0.918 183s -0.254 -0.372 -0.589 -0.526 183s 0.307 -0.533 0.058 -0.485 183s 183s x1: [4, 2, 1, 4] 183s -0.053 0.626 0.439 -0.443 183s 0.513 0.202 0.086 0.048 183s 183s y: [4, 2, 1, 4] 183s -0.891 0.376 -0.245 -0.110 183s -0.276 0.385 -0.281 0.031 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.901 -0.952 0.814 -0.150 183s -0.659 0.112 0.981 -0.195 183s 183s x1: [4, 4, 1, 1] 183s -0.353 -0.399 0.226 0.792 183s 0.689 0.488 0.684 0.406 183s -0.840 -0.950 -0.570 0.299 183s 0.473 0.969 -0.191 0.179 183s 183s y: [2, 4, 1, 1] 183s 0.763 0.255 183s -0.590 0.192 183s 1.153 -0.171 183s -1.530 -0.425 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.851 -0.585 183s 0.009 0.387 183s 0.227 0.882 183s 0.833 -0.155 183s 183s x1: [2, 4, 1, 1] 183s -0.021 -0.492 183s 0.468 0.710 183s -0.281 -0.486 183s -0.643 0.320 183s 183s y: [4, 4, 1, 1] 183s 0.306 -0.191 -0.439 0.058 183s -0.813 0.279 0.733 0.280 183s 0.523 -0.191 -0.493 -0.159 183s 0.360 0.118 0.136 -0.585 183s testing: mul_mat, [2, 2, 1, 4] = [4, 2, 1, 4] * [4, 2, 1, 4] 183s x0: [4, 2, 1, 4] 183s 0.187 0.779 0.607 0.136 183s -0.987 -0.863 -0.345 -0.686 183s 183s x1: [4, 2, 1, 4] 183s -0.314 0.356 -0.186 0.496 183s -0.466 -0.050 0.594 -0.125 183s 183s y: [2, 2, 1, 4] 183s 0.173 -0.273 183s 0.218 0.383 183s test-mul-mat0: iter:241/500 183s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.220 183s -0.138 183s -0.590 183s 183s x1: [1, 1, 1, 1] 183s 0.518 183s 183s y: [3, 1, 1, 1] 183s 0.114 -0.071 -0.305 183s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.501 183s -0.774 183s 183s x1: [1, 3, 1, 1] 183s 0.090 183s 0.700 183s -0.516 183s 183s y: [2, 3, 1, 1] 183s 0.045 -0.070 183s 0.351 -0.542 183s -0.258 0.399 183s testing: mul_mat, [3, 2, 1, 2] = [1, 3, 1, 2] * [1, 2, 1, 2] 183s x0: [1, 3, 1, 2] 183s 0.936 183s -0.502 183s -0.792 183s 183s x1: [1, 2, 1, 2] 183s 0.742 183s -0.222 183s 183s y: [3, 2, 1, 2] 183s 0.695 -0.373 -0.588 183s -0.208 0.112 0.176 183s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.182 183s 0.560 183s 183s x1: [1, 3, 1, 1] 183s 0.723 183s 0.796 183s -0.739 183s 183s y: [2, 3, 1, 1] 183s -0.131 0.404 183s -0.145 0.445 183s 0.134 -0.414 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.187 0.686 183s -0.728 -0.951 183s 183s x1: [2, 1, 1, 1] 183s 0.669 0.977 183s 183s y: [2, 1, 1, 1] 183s 0.795 -1.417 183s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 1, 1, 2] 183s 0.387 -0.894 183s 183s x1: [2, 2, 1, 2] 183s -0.318 -0.225 183s 0.749 0.166 183s 183s y: [1, 2, 1, 2] 183s 0.078 183s 0.141 183s test-mul-mat0: iter:242/500 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.258 183s 0.735 183s 0.179 183s 183s x1: [1, 2, 1, 1] 183s -0.519 183s -0.071 183s 183s y: [3, 2, 1, 1] 183s 0.134 -0.381 -0.093 183s 0.018 -0.052 -0.013 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.408 183s 0.428 183s -0.547 183s 183s x1: [1, 3, 1, 1] 183s -0.698 183s 0.405 183s 0.156 183s 183s y: [3, 3, 1, 1] 183s 0.285 -0.299 0.382 183s -0.165 0.173 -0.221 183s -0.064 0.067 -0.085 183s testing: mul_mat, [2, 3, 1, 4] = [1, 2, 1, 4] * [1, 3, 1, 4] 183s x0: [1, 2, 1, 4] 183s 0.378 183s 0.282 183s 183s x1: [1, 3, 1, 4] 183s -0.359 183s -0.890 183s 0.229 183s 183s y: [2, 3, 1, 4] 183s -0.136 -0.101 183s -0.336 -0.251 183s 0.086 0.064 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.998 183s -0.347 183s -0.458 183s -0.234 183s 183s x1: [1, 2, 1, 1] 183s 0.696 183s 0.248 183s 183s y: [4, 2, 1, 1] 183s 0.695 -0.242 -0.319 -0.163 183s 0.248 -0.086 -0.114 -0.058 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.552 0.356 0.916 -0.952 183s -0.724 0.477 -0.975 0.585 183s 183s x1: [4, 1, 1, 1] 183s 0.245 0.970 0.220 0.886 183s 183s y: [2, 1, 1, 1] 183s -0.432 0.589 183s testing: mul_mat, [4, 4, 1, 4] = [2, 4, 1, 4] * [2, 4, 1, 4] 183s x0: [2, 4, 1, 4] 183s -0.085 0.001 183s -0.041 -0.782 183s -0.636 0.916 183s 0.939 -0.444 183s 183s x1: [2, 4, 1, 4] 183s -0.919 -0.275 183s -0.449 0.458 183s -0.993 0.024 183s -0.958 0.854 183s 183s y: [4, 4, 1, 4] 183s 0.078 0.253 0.333 -0.741 183s 0.038 -0.340 0.705 -0.625 183s 0.084 0.022 0.653 -0.943 183s 0.082 -0.628 1.392 -1.279 183s test-mul-mat0: iter:243/500 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.337 0.436 -0.765 0.011 183s -0.205 -0.660 -0.277 -0.739 183s 183s x1: [4, 2, 1, 1] 183s 0.619 -0.118 -0.456 0.897 183s -0.527 0.903 0.623 0.735 183s 183s y: [2, 2, 1, 1] 183s 0.516 -0.586 183s -0.252 -1.204 183s testing: mul_mat, [4, 2, 2, 1] = [4, 4, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 4, 2, 1] 183s 0.945 -0.486 0.808 0.568 183s -0.751 -0.473 -0.095 0.685 183s -0.238 0.916 -0.519 0.102 183s -0.360 -0.258 -0.591 -0.530 183s 183s x1: [4, 2, 2, 1] 183s 0.307 0.831 0.310 0.054 183s -0.642 0.990 0.166 -0.613 183s 183s y: [4, 2, 2, 1] 183s 0.168 -0.616 0.533 -0.537 183s -1.302 -0.423 0.912 0.202 183s testing: mul_mat, [3, 4, 2, 1] = [4, 3, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.545 -0.292 0.293 -0.580 183s 0.179 -0.752 -0.698 -0.869 183s -0.695 -0.177 0.824 -0.230 183s 183s x1: [4, 4, 2, 1] 183s -0.463 0.666 0.060 -0.214 183s -0.807 0.965 -0.529 -0.045 183s 0.882 -0.048 -0.943 -0.479 183s 0.694 -0.534 -0.008 0.745 183s 183s y: [3, 4, 2, 1] 183s 0.200 -0.439 0.303 183s 0.029 -0.461 -0.035 183s -0.465 1.269 -1.272 183s -0.657 -0.116 -0.566 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.046 0.869 0.683 0.318 183s 0.324 0.148 -0.282 -0.925 183s -0.746 0.025 0.724 -0.018 183s 183s x1: [4, 3, 1, 1] 183s 0.874 -0.526 0.386 0.308 183s 0.869 0.308 -0.591 -0.676 183s -0.985 0.702 -0.256 0.194 183s 183s y: [3, 3, 1, 1] 183s -0.055 -0.189 -0.391 183s -0.311 1.120 -1.056 183s 0.451 -0.323 0.563 183s testing: mul_mat, [1, 4, 2, 1] = [3, 1, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 1, 2, 1] 183s -0.130 -0.286 -0.469 183s 183s x1: [3, 4, 2, 1] 183s 0.988 -0.711 0.620 183s -0.715 -0.140 0.621 183s -0.841 0.335 0.007 183s 0.467 0.204 -0.686 183s 183s y: [1, 4, 2, 1] 183s -0.216 183s -0.159 183s 0.010 183s 0.203 183s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 1, 2, 1] 183s 0.273 183s 183s x1: [1, 3, 2, 1] 183s -0.824 183s 0.078 183s 0.132 183s 183s y: [1, 3, 2, 1] 183s -0.225 183s 0.021 183s 0.036 183s test-mul-mat0: iter:244/500 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.216 -0.229 183s 0.655 0.655 183s 0.486 -0.814 183s 0.842 -0.725 183s 183s x1: [2, 4, 1, 1] 183s -0.994 -0.596 183s 0.540 -0.721 183s 0.927 0.067 183s 0.597 -0.225 183s 183s y: [4, 4, 1, 1] 183s 0.351 -1.042 0.003 -0.405 183s 0.049 -0.119 0.850 0.978 183s -0.215 0.651 0.396 0.732 183s -0.077 0.244 0.473 0.666 183s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 3, 4, 1] 183s -0.592 -0.027 183s 0.704 0.699 183s -0.443 -0.197 183s 183s x1: [2, 4, 4, 1] 183s 0.698 -0.981 183s 0.353 -0.170 183s -0.520 0.292 183s 0.584 0.100 183s 183s y: [3, 4, 4, 1] 183s -0.387 -0.195 -0.116 183s -0.204 0.130 -0.123 183s 0.300 -0.162 0.173 183s -0.348 0.480 -0.278 183s testing: mul_mat, [2, 3, 4, 3] = [2, 2, 4, 3] * [2, 3, 4, 3] 183s x0: [2, 2, 4, 3] 183s -0.317 0.834 183s -0.398 0.395 183s 183s x1: [2, 3, 4, 3] 183s 0.675 0.486 183s -0.113 -0.120 183s -0.459 0.030 183s 183s y: [2, 3, 4, 3] 183s 0.191 -0.077 183s -0.064 -0.003 183s 0.170 0.195 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.749 0.405 183s 0.023 0.042 183s 0.702 0.232 183s 0.275 0.371 183s 183s x1: [2, 2, 1, 1] 183s -0.735 0.927 183s -0.035 -0.413 183s 183s y: [4, 2, 1, 1] 183s -0.176 0.022 -0.302 0.142 183s -0.193 -0.018 -0.120 -0.163 183s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.952 -0.825 -0.289 0.607 183s 183s x1: [4, 2, 4, 1] 183s -0.473 0.872 0.171 -0.387 183s 0.907 0.655 0.979 0.827 183s 183s y: [1, 2, 4, 1] 183s -1.454 183s 0.543 183s testing: mul_mat, [1, 4, 4, 3] = [1, 1, 4, 3] * [1, 4, 4, 3] 183s x0: [1, 1, 4, 3] 183s 0.964 183s 183s x1: [1, 4, 4, 3] 183s -0.647 183s 0.061 183s -0.530 183s 0.335 183s 183s y: [1, 4, 4, 3] 183s -0.624 183s 0.058 183s -0.511 183s 0.323 183s test-mul-mat0: iter:245/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.187 0.707 0.380 183s -0.827 -0.217 -0.228 183s 183s x1: [3, 1, 1, 1] 183s -0.594 0.832 -0.841 183s 183s y: [2, 1, 1, 1] 183s 0.380 0.502 183s testing: mul_mat, [1, 2, 3, 1] = [3, 1, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 1, 3, 1] 183s -0.333 0.388 0.922 183s 183s x1: [3, 2, 3, 1] 183s 0.476 -0.808 0.647 183s -0.105 -0.355 0.611 183s 183s y: [1, 2, 3, 1] 183s 0.125 183s 0.461 183s testing: mul_mat, [3, 1, 3, 4] = [3, 3, 3, 4] * [3, 1, 3, 4] 183s x0: [3, 3, 3, 4] 183s 0.238 -0.473 0.956 183s 0.678 -0.919 -0.513 183s -0.312 0.509 -0.524 183s 183s x1: [3, 1, 3, 4] 183s 0.485 -0.962 0.592 183s 183s y: [3, 1, 3, 4] 183s 1.137 0.910 -0.951 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.823 -0.785 -0.701 183s 183s x1: [3, 3, 1, 1] 183s 0.551 0.712 -0.831 183s 0.917 0.622 0.786 183s 0.323 -0.581 0.732 183s 183s y: [1, 3, 1, 1] 183s -0.431 183s -1.795 183s -0.322 183s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 2, 3, 1] 183s 0.875 183s 0.572 183s 183s x1: [1, 3, 3, 1] 183s -0.649 183s 0.098 183s -0.699 183s 183s y: [2, 3, 3, 1] 183s -0.568 -0.371 183s 0.086 0.056 183s -0.612 -0.399 183s testing: mul_mat, [1, 1, 3, 4] = [2, 1, 3, 4] * [2, 1, 3, 4] 183s x0: [2, 1, 3, 4] 183s 0.170 0.049 183s 183s x1: [2, 1, 3, 4] 183s 0.800 0.819 183s 183s y: [1, 1, 3, 4] 183s 0.176 183s test-mul-mat0: iter:246/500 183s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.621 183s -0.652 183s 183s x1: [1, 3, 1, 1] 183s 0.285 183s 0.330 183s 0.451 183s 183s y: [2, 3, 1, 1] 183s -0.177 -0.186 183s -0.205 -0.215 183s -0.280 -0.294 183s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.672 183s -0.937 183s 183s x1: [1, 2, 3, 1] 183s -0.614 183s 0.154 183s 183s y: [2, 2, 3, 1] 183s 0.412 0.575 183s -0.103 -0.144 183s testing: mul_mat, [2, 2, 3, 4] = [1, 2, 3, 4] * [1, 2, 3, 4] 183s x0: [1, 2, 3, 4] 183s 0.330 183s 0.904 183s 183s x1: [1, 2, 3, 4] 183s -0.456 183s 0.511 183s 183s y: [2, 2, 3, 4] 183s -0.150 -0.412 183s 0.169 0.462 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.173 183s 0.843 183s 0.907 183s 0.292 183s 183s x1: [1, 2, 1, 1] 183s -0.585 183s -0.159 183s 183s y: [4, 2, 1, 1] 183s -0.101 -0.493 -0.531 -0.171 183s -0.027 -0.134 -0.144 -0.046 183s testing: mul_mat, [4, 1, 3, 1] = [4, 4, 3, 1] * [4, 1, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.187 -0.390 -0.347 -0.508 183s 0.985 -0.122 0.651 0.031 183s 0.357 -0.937 -0.035 -0.759 183s -0.423 -0.180 0.068 -0.665 183s 183s x1: [4, 1, 3, 1] 183s 0.173 0.811 0.152 0.187 183s 183s y: [4, 1, 3, 1] 183s -0.497 0.176 -0.845 -0.333 183s testing: mul_mat, [1, 4, 3, 4] = [4, 1, 3, 4] * [4, 4, 3, 4] 183s x0: [4, 1, 3, 4] 183s 0.610 -0.210 0.514 0.247 183s 183s x1: [4, 4, 3, 4] 183s -0.770 0.694 0.510 -0.086 183s -0.611 0.714 -0.718 -0.172 183s -0.473 0.199 0.695 -0.218 183s 0.612 -0.804 0.849 -0.990 183s 183s y: [1, 4, 3, 4] 183s -0.375 183s -0.935 183s -0.027 183s 0.735 183s test-mul-mat0: iter:247/500 183s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.091 0.296 -0.127 -0.622 183s 0.650 -0.694 -0.985 -0.035 183s -0.183 0.070 -0.166 0.235 183s 183s x1: [4, 2, 1, 1] 183s 0.283 -0.170 -0.326 0.297 183s -0.308 -0.606 -0.046 -0.811 183s 183s y: [3, 2, 1, 1] 183s -0.220 0.613 0.060 183s 0.359 0.295 -0.169 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.846 0.248 0.404 -0.200 183s 0.437 0.642 0.709 -0.267 183s -0.486 -0.913 -0.616 -0.180 183s 183s x1: [4, 3, 1, 1] 183s 0.427 -0.841 0.864 0.953 183s -0.377 0.919 -0.750 0.267 183s 0.726 -0.664 -0.451 -0.444 183s 183s y: [3, 3, 1, 1] 183s 0.311 0.006 -0.143 183s -0.447 -0.177 -0.242 183s 0.356 -0.310 0.612 183s testing: mul_mat, [1, 3, 1, 2] = [4, 1, 1, 2] * [4, 3, 1, 2] 183s x0: [4, 1, 1, 2] 183s 0.136 -0.645 0.708 0.223 183s 183s x1: [4, 3, 1, 2] 183s -0.898 0.349 0.637 0.172 183s -0.817 -0.128 -0.401 -0.658 183s -0.264 -0.448 -0.035 -0.345 183s 183s y: [1, 3, 1, 2] 183s 0.142 183s -0.460 183s 0.151 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.388 -0.571 0.588 0.216 183s 0.774 -0.550 -0.411 0.931 183s 0.648 0.808 0.747 0.288 183s 0.577 -0.190 0.737 -0.356 183s 183s x1: [4, 1, 1, 1] 183s -0.834 -0.503 -0.730 0.038 183s 183s y: [4, 1, 1, 1] 183s 0.190 -0.034 -1.481 -0.937 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.569 0.191 -0.875 -0.981 183s 183s x1: [4, 4, 1, 1] 183s 0.553 -0.518 0.780 0.908 183s -0.810 0.003 -0.354 -0.282 183s -0.672 -0.266 -0.116 -0.175 183s 0.003 0.921 0.921 0.615 183s 183s y: [1, 4, 1, 1] 183s -1.358 183s 0.126 183s -0.160 183s -1.231 183s testing: mul_mat, [2, 4, 1, 2] = [1, 2, 1, 2] * [1, 4, 1, 2] 183s x0: [1, 2, 1, 2] 183s 0.448 183s -0.834 183s 183s x1: [1, 4, 1, 2] 183s -0.000 183s -0.065 183s 0.607 183s 0.588 183s 183s y: [2, 4, 1, 2] 183s -0.000 0.000 183s -0.029 0.054 183s 0.272 -0.506 183s 0.264 -0.491 183s test-mul-mat0: iter:248/500 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.543 -0.173 0.305 -0.928 183s 183s x1: [4, 1, 1, 1] 183s -0.979 -0.441 0.454 -0.095 183s 183s y: [1, 1, 1, 1] 183s 0.834 183s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.332 0.116 -0.108 0.563 183s 183s x1: [4, 1, 4, 1] 183s -0.478 -0.126 0.263 -0.353 183s 183s y: [1, 1, 4, 1] 183s -0.400 183s testing: mul_mat, [3, 1, 4, 3] = [4, 3, 4, 3] * [4, 1, 4, 3] 183s x0: [4, 3, 4, 3] 183s -0.040 -0.370 -0.391 0.933 183s -0.549 0.220 -0.588 -0.981 183s -0.523 0.703 0.406 -0.099 183s 183s x1: [4, 1, 4, 3] 183s 0.082 0.814 -0.422 0.345 183s 183s y: [3, 1, 4, 3] 183s 0.183 0.043 0.324 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.348 0.512 0.849 0.369 183s 0.488 0.315 0.876 0.447 183s 0.513 -0.676 -0.660 -0.010 183s 0.247 -0.185 -0.282 -0.482 183s 183s x1: [4, 3, 1, 1] 183s -0.164 0.901 -0.453 -0.243 183s 0.900 0.278 -0.579 0.719 183s 0.732 -0.269 -0.117 0.982 183s 183s y: [4, 3, 1, 1] 183s 0.045 -0.302 -0.392 0.037 183s -0.397 0.341 0.649 -0.012 183s -0.130 0.609 0.625 -0.209 183s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.061 0.129 0.732 0.408 183s 0.358 -0.975 -0.707 0.575 183s 183s x1: [4, 4, 4, 1] 183s -0.223 0.910 0.354 -0.322 183s -0.542 -0.889 -0.422 0.736 183s -0.468 -0.703 0.468 0.263 183s 0.180 0.449 0.886 0.833 183s 183s y: [2, 4, 4, 1] 183s 0.259 -1.402 183s -0.091 1.393 183s 0.388 0.338 183s 1.036 -0.521 183s testing: mul_mat, [4, 4, 4, 3] = [2, 4, 4, 3] * [2, 4, 4, 3] 183s x0: [2, 4, 4, 3] 183s -0.428 -0.481 183s 0.372 -0.774 183s 0.443 0.901 183s -0.032 0.776 183s 183s x1: [2, 4, 4, 3] 183s 0.040 0.295 183s 0.269 -0.227 183s 0.589 -0.323 183s -0.652 -0.690 183s 183s y: [4, 4, 4, 3] 183s -0.159 -0.214 0.284 0.228 183s -0.006 0.276 -0.086 -0.185 183s -0.097 0.469 -0.030 -0.270 183s 0.612 0.292 -0.911 -0.515 183s test-mul-mat0: iter:249/500 183s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.266 -0.944 183s -0.127 -0.650 183s -0.494 -0.106 183s 183s x1: [2, 4, 1, 1] 183s 0.510 0.336 183s 0.726 -0.584 183s -0.168 -0.244 183s 0.290 0.705 183s 183s y: [3, 4, 1, 1] 183s -0.453 -0.283 -0.288 183s 0.358 0.287 -0.297 183s 0.275 0.180 0.109 183s -0.743 -0.495 -0.218 183s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.651 -0.405 183s 0.042 0.638 183s 183s x1: [2, 3, 3, 1] 183s 0.769 0.535 183s 0.026 0.671 183s 0.334 0.196 183s 183s y: [2, 3, 3, 1] 183s -0.718 0.374 183s -0.288 0.429 183s -0.297 0.139 183s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.795 0.490 183s -0.479 -0.862 183s 183s x1: [2, 2, 3, 1] 183s -0.509 -0.602 183s 0.124 -0.838 183s 183s y: [2, 2, 3, 1] 183s 0.109 0.762 183s -0.509 0.663 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.481 0.766 183s 0.710 -0.395 183s 183s x1: [2, 2, 1, 1] 183s -0.503 -0.495 183s 0.384 0.219 183s 183s y: [2, 2, 1, 1] 183s -0.621 -0.162 183s 0.352 0.186 183s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.234 -0.098 183s 183s x1: [2, 2, 3, 1] 183s 0.872 0.498 183s -0.076 -0.863 183s 183s y: [1, 2, 3, 1] 183s -0.253 183s 0.102 183s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 3, 3, 1] 183s -0.228 183s -0.727 183s -0.196 183s 183s x1: [1, 2, 3, 1] 183s 0.378 183s -0.101 183s 183s y: [3, 2, 3, 1] 183s -0.086 -0.275 -0.074 183s 0.023 0.073 0.020 183s test-mul-mat0: iter:250/500 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.639 -0.865 183s 0.124 0.260 183s 0.543 -0.001 183s -0.336 -0.665 183s 183s x1: [2, 4, 1, 1] 183s 0.058 -0.046 183s 0.727 -0.292 183s 0.527 -0.507 183s 0.610 -0.093 183s 183s y: [4, 4, 1, 1] 183s 0.002 -0.005 0.032 0.011 183s -0.212 0.015 0.395 -0.050 183s 0.101 -0.066 0.287 0.160 183s -0.309 0.052 0.332 -0.143 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.614 -0.313 183s 183s x1: [2, 4, 1, 1] 183s -0.907 0.436 183s -0.392 -0.103 183s 0.974 0.486 183s 0.573 -0.990 183s 183s y: [1, 4, 1, 1] 183s 0.420 183s 0.273 183s -0.750 183s -0.041 183s testing: mul_mat, [4, 1, 1, 3] = [2, 4, 1, 3] * [2, 1, 1, 3] 183s x0: [2, 4, 1, 3] 183s 0.771 -0.993 183s 0.564 -0.323 183s 0.753 0.925 183s -0.188 -0.123 183s 183s x1: [2, 1, 1, 3] 183s -0.813 0.289 183s 183s y: [4, 1, 1, 3] 183s -0.914 -0.552 -0.344 0.117 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.024 -0.193 183s 183s x1: [2, 4, 1, 1] 183s -0.739 -0.885 183s 0.741 0.507 183s -0.641 0.255 183s 0.460 -0.871 183s 183s y: [1, 4, 1, 1] 183s 0.153 183s -0.080 183s -0.065 183s 0.179 183s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.891 183s 183s x1: [1, 2, 1, 1] 183s 0.014 183s -0.051 183s 183s y: [1, 2, 1, 1] 183s 0.013 183s -0.045 183s testing: mul_mat, [2, 1, 1, 3] = [1, 2, 1, 3] * [1, 1, 1, 3] 183s x0: [1, 2, 1, 3] 183s -0.334 183s 0.737 183s 183s x1: [1, 1, 1, 3] 183s -0.866 183s 183s y: [2, 1, 1, 3] 183s 0.289 -0.639 183s test-mul-mat0: iter:251/500 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.022 -0.973 -0.103 183s -0.087 -0.839 -0.128 183s 183s x1: [3, 4, 1, 1] 183s 0.868 0.751 0.087 183s 0.983 0.492 -0.406 183s -0.658 -0.253 -0.946 183s -0.528 -0.992 0.078 183s 183s y: [2, 4, 1, 1] 183s -0.720 -0.716 183s -0.415 -0.446 183s 0.329 0.390 183s 0.945 0.868 183s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.723 -0.184 -0.270 183s 0.731 0.895 -0.992 183s -0.247 0.922 -0.095 183s 0.666 -0.917 0.778 183s 183s x1: [3, 2, 3, 1] 183s -0.319 -0.856 0.538 183s -0.582 0.411 0.504 183s 183s y: [4, 2, 3, 1] 183s -0.219 -1.533 -0.761 0.992 183s -0.632 -0.557 0.475 -0.373 183s testing: mul_mat, [1, 4, 3, 2] = [3, 1, 3, 2] * [3, 4, 3, 2] 183s x0: [3, 1, 3, 2] 183s -0.655 0.030 0.197 183s 183s x1: [3, 4, 3, 2] 183s -0.304 0.972 0.720 183s 0.602 0.638 0.803 183s 0.379 0.985 -0.971 183s -0.305 -0.249 0.666 183s 183s y: [1, 4, 3, 2] 183s 0.370 183s -0.217 183s -0.410 183s 0.323 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.962 0.391 0.096 183s 183s x1: [3, 1, 1, 1] 183s -0.336 0.488 0.861 183s 183s y: [1, 1, 1, 1] 183s -0.049 183s testing: mul_mat, [1, 3, 3, 1] = [1, 1, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 1, 3, 1] 183s -0.202 183s 183s x1: [1, 3, 3, 1] 183s 0.819 183s 0.119 183s 0.434 183s 183s y: [1, 3, 3, 1] 183s -0.165 183s -0.024 183s -0.088 183s testing: mul_mat, [2, 1, 3, 2] = [1, 2, 3, 2] * [1, 1, 3, 2] 183s x0: [1, 2, 3, 2] 183s 0.306 183s 0.343 183s 183s x1: [1, 1, 3, 2] 183s -0.431 183s 183s y: [2, 1, 3, 2] 183s -0.132 -0.148 183s test-mul-mat0: iter:252/500 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.437 -0.424 0.097 0.906 183s 183s x1: [4, 3, 1, 1] 183s -0.362 0.095 0.339 0.767 183s 0.189 -0.863 0.181 -0.189 183s -0.293 -0.857 -0.834 0.862 183s 183s y: [1, 3, 1, 1] 183s 0.846 183s 0.129 183s 1.192 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.910 0.439 0.578 0.363 183s 183s x1: [4, 1, 1, 1] 183s 0.884 0.553 0.914 -0.320 183s 183s y: [1, 1, 1, 1] 183s -0.149 183s testing: mul_mat, [3, 1, 1, 4] = [4, 3, 1, 4] * [4, 1, 1, 4] 183s x0: [4, 3, 1, 4] 183s 0.590 0.505 0.595 -0.313 183s 0.411 0.479 -0.760 0.325 183s -0.840 -0.890 0.415 0.598 183s 183s x1: [4, 1, 1, 4] 183s -0.984 -0.595 -0.931 0.681 183s 183s y: [3, 1, 1, 4] 183s -1.648 0.240 1.377 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.165 0.682 -0.383 0.850 183s -0.143 -0.233 -0.648 -0.384 183s 183s x1: [4, 3, 1, 1] 183s 0.750 -0.042 -0.412 0.821 183s -0.662 -0.384 0.858 -0.831 183s -0.699 -0.330 0.157 0.784 183s 183s y: [2, 3, 1, 1] 183s 0.951 -0.146 183s -1.405 -0.053 183s 0.265 -0.226 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.023 0.373 183s -0.864 0.564 183s 183s x1: [2, 4, 1, 1] 183s -0.654 0.091 183s -0.274 0.756 183s 0.931 -0.476 183s 0.943 -0.614 183s 183s y: [2, 4, 1, 1] 183s 0.049 0.616 183s 0.288 0.663 183s -0.199 -1.073 183s -0.251 -1.161 183s testing: mul_mat, [4, 2, 1, 4] = [2, 4, 1, 4] * [2, 2, 1, 4] 183s x0: [2, 4, 1, 4] 183s -0.645 0.045 183s 0.290 0.842 183s 0.911 0.604 183s 0.081 -0.012 183s 183s x1: [2, 2, 1, 4] 183s 0.958 0.711 183s -0.819 0.815 183s 183s y: [4, 2, 1, 4] 183s -0.586 0.876 1.302 0.069 183s 0.566 0.449 -0.254 -0.076 183s test-mul-mat0: iter:253/500 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.994 183s -0.625 183s 183s x1: [1, 2, 1, 1] 183s 0.765 183s -0.644 183s 183s y: [2, 2, 1, 1] 183s 0.760 -0.478 183s -0.640 0.403 183s testing: mul_mat, [2, 2, 2, 1] = [1, 2, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 2, 2, 1] 183s -0.131 183s 0.594 183s 183s x1: [1, 2, 2, 1] 183s -0.645 183s -0.641 183s 183s y: [2, 2, 2, 1] 183s 0.085 -0.383 183s 0.084 -0.381 183s testing: mul_mat, [4, 2, 2, 3] = [1, 4, 2, 3] * [1, 2, 2, 3] 183s x0: [1, 4, 2, 3] 183s -0.479 183s 0.222 183s 0.559 183s 0.923 183s 183s x1: [1, 2, 2, 3] 183s 0.786 183s 0.443 183s 183s y: [4, 2, 2, 3] 183s -0.376 0.174 0.439 0.725 183s -0.212 0.098 0.247 0.408 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.327 183s 0.938 183s 183s x1: [1, 4, 1, 1] 183s 0.994 183s -0.645 183s -0.972 183s 0.131 183s 183s y: [2, 4, 1, 1] 183s 0.326 0.933 183s -0.211 -0.605 183s -0.318 -0.911 183s 0.043 0.123 183s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.412 0.276 183s 183s x1: [2, 1, 2, 1] 183s 0.041 -0.451 183s 183s y: [1, 1, 2, 1] 183s -0.108 183s testing: mul_mat, [4, 2, 2, 3] = [1, 4, 2, 3] * [1, 2, 2, 3] 183s x0: [1, 4, 2, 3] 183s -0.743 183s -0.153 183s -0.256 183s -0.388 183s 183s x1: [1, 2, 2, 3] 183s 0.944 183s 0.856 183s 183s y: [4, 2, 2, 3] 183s -0.702 -0.144 -0.241 -0.367 183s -0.637 -0.131 -0.219 -0.333 183s test-mul-mat0: iter:254/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.274 0.566 0.123 183s 183s x1: [3, 1, 1, 1] 183s -0.745 -0.573 -0.178 183s 183s y: [1, 1, 1, 1] 183s -0.551 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.254 -0.573 0.105 183s -0.435 0.562 0.394 183s 183s x1: [3, 1, 1, 1] 183s -0.850 0.441 -0.745 183s 183s y: [2, 1, 1, 1] 183s -0.548 0.325 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.183 0.400 -0.247 183s -0.321 -0.345 0.074 183s -0.249 -0.089 0.501 183s 183s x1: [3, 2, 1, 1] 183s 0.286 0.974 -0.801 183s 0.710 -0.670 -0.139 183s 183s y: [3, 2, 1, 1] 183s 0.640 -0.487 -0.560 183s -0.104 -0.007 -0.187 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.227 -0.211 -0.236 183s -0.617 0.776 0.976 183s -0.602 0.684 0.394 183s 183s x1: [3, 3, 1, 1] 183s 0.572 0.422 -0.225 183s 0.138 -0.455 -0.076 183s -0.421 -0.200 0.277 183s 183s y: [3, 3, 1, 1] 183s 0.094 -0.245 -0.144 183s 0.145 -0.512 -0.424 183s -0.119 0.375 0.226 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.803 -0.387 0.279 183s 183s x1: [3, 3, 1, 1] 183s 0.094 -0.163 0.284 183s -0.723 -0.763 -0.962 183s -0.044 -0.108 0.112 183s 183s y: [1, 3, 1, 1] 183s 0.218 183s -0.553 183s 0.038 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.688 183s -0.004 183s -0.429 183s 183s x1: [1, 3, 1, 1] 183s 0.225 183s 0.387 183s -0.583 183s 183s y: [3, 3, 1, 1] 183s -0.155 -0.001 -0.097 183s -0.267 -0.002 -0.166 183s 0.401 0.002 0.250 183s test-mul-mat0: iter:255/500 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.555 -0.439 183s 0.591 -0.277 183s -0.202 0.629 183s 183s x1: [2, 3, 1, 1] 183s 0.228 -0.413 183s 0.748 -0.088 183s 0.351 0.724 183s 183s y: [3, 3, 1, 1] 183s 0.055 0.249 -0.305 183s -0.376 0.466 -0.206 183s -0.513 0.007 0.384 183s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 3, 3, 1] 183s 0.446 -0.638 183s 0.047 -0.203 183s -0.914 0.354 183s 183s x1: [2, 3, 3, 1] 183s 0.679 0.691 183s -0.260 0.386 183s 0.494 0.353 183s 183s y: [3, 3, 3, 1] 183s -0.138 -0.108 -0.376 183s -0.362 -0.091 0.374 183s -0.005 -0.048 -0.326 183s testing: mul_mat, [3, 3, 3, 4] = [2, 3, 3, 4] * [2, 3, 3, 4] 183s x0: [2, 3, 3, 4] 183s -0.812 0.200 183s -0.194 -0.256 183s 0.006 0.868 183s 183s x1: [2, 3, 3, 4] 183s 0.695 0.348 183s 0.407 -0.224 183s 0.838 -0.540 183s 183s y: [3, 3, 3, 4] 183s -0.495 -0.224 0.307 183s -0.376 -0.022 -0.192 183s -0.789 -0.024 -0.464 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.948 -0.815 183s -0.452 -0.978 183s 183s x1: [2, 3, 1, 1] 183s -0.667 -0.721 183s 0.299 0.690 183s 0.972 0.345 183s 183s y: [2, 3, 1, 1] 183s -0.045 1.007 183s -0.278 -0.810 183s 0.641 -0.777 183s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.638 -0.302 183s 183s x1: [2, 2, 3, 1] 183s 0.475 -0.311 183s -0.641 0.903 183s 183s y: [1, 2, 3, 1] 183s -0.209 183s 0.136 183s testing: mul_mat, [1, 2, 3, 4] = [1, 1, 3, 4] * [1, 2, 3, 4] 183s x0: [1, 1, 3, 4] 183s 0.703 183s 183s x1: [1, 2, 3, 4] 183s 0.476 183s 0.621 183s 183s y: [1, 2, 3, 4] 183s 0.335 183s 0.436 183s test-mul-mat0: iter:256/500 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.872 -0.670 183s 0.200 -0.339 183s -0.968 -0.042 183s 0.229 -0.987 183s 183s x1: [2, 1, 1, 1] 183s -0.003 -0.040 183s 183s y: [4, 1, 1, 1] 183s 0.029 0.013 0.005 0.039 183s testing: mul_mat, [2, 4, 3, 1] = [2, 2, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.609 -0.538 183s -0.664 -0.432 183s 183s x1: [2, 4, 3, 1] 183s 0.359 -0.430 183s 0.393 0.263 183s -0.728 0.471 183s 0.328 -0.760 183s 183s y: [2, 4, 3, 1] 183s 0.450 -0.052 183s 0.098 -0.375 183s -0.696 0.280 183s 0.608 0.111 183s testing: mul_mat, [2, 2, 3, 3] = [2, 2, 3, 3] * [2, 2, 3, 3] 183s x0: [2, 2, 3, 3] 183s 0.582 -0.312 183s -0.288 0.447 183s 183s x1: [2, 2, 3, 3] 183s -0.483 -0.226 183s -0.950 -0.301 183s 183s y: [2, 2, 3, 3] 183s -0.210 0.038 183s -0.459 0.139 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.187 0.423 183s 0.289 -0.504 183s 183s x1: [2, 2, 1, 1] 183s 0.697 0.911 183s 0.758 0.421 183s 183s y: [2, 2, 1, 1] 183s 0.515 -0.257 183s 0.320 0.007 183s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 1, 3, 1] 183s 0.088 -0.588 183s 183s x1: [2, 2, 3, 1] 183s 0.763 0.248 183s -0.746 -0.465 183s 183s y: [1, 2, 3, 1] 183s -0.078 183s 0.207 183s testing: mul_mat, [4, 2, 3, 3] = [1, 4, 3, 3] * [1, 2, 3, 3] 183s x0: [1, 4, 3, 3] 183s -0.121 183s 0.260 183s 0.866 183s 0.082 183s 183s x1: [1, 2, 3, 3] 183s 0.007 183s -0.866 183s 183s y: [4, 2, 3, 3] 183s -0.001 0.002 0.006 0.001 183s 0.104 -0.225 -0.750 -0.071 183s test-mul-mat0: iter:257/500 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.117 -0.976 183s -0.397 -0.493 183s 183s x1: [2, 3, 1, 1] 183s 0.465 -0.351 183s -0.325 0.334 183s -0.139 0.920 183s 183s y: [2, 3, 1, 1] 183s 0.397 -0.012 183s -0.364 -0.036 183s -0.915 -0.398 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.793 -0.650 183s -0.004 -0.322 183s 0.016 0.570 183s 183s x1: [2, 2, 1, 1] 183s -0.827 -0.287 183s -0.545 -0.136 183s 183s y: [3, 2, 1, 1] 183s 0.843 0.096 -0.177 183s 0.521 0.046 -0.087 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.443 0.429 183s 183s x1: [2, 3, 1, 1] 183s -0.972 -0.057 183s -0.989 -0.539 183s 0.986 -0.102 183s 183s y: [1, 3, 1, 1] 183s 0.406 183s 0.206 183s -0.480 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.864 0.184 183s -0.933 -0.951 183s -0.975 -0.627 183s 0.770 0.691 183s 183s x1: [2, 1, 1, 1] 183s -0.471 0.206 183s 183s y: [4, 1, 1, 1] 183s -0.369 0.243 0.330 -0.220 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.547 0.650 0.059 0.396 183s 183s x1: [4, 2, 1, 1] 183s -0.778 0.086 -0.854 0.086 183s 0.084 -0.647 0.436 -0.920 183s 183s y: [1, 2, 1, 1] 183s 0.465 183s -0.805 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.939 183s 183s x1: [1, 4, 1, 1] 183s 0.661 183s 0.520 183s 0.382 183s -0.441 183s 183s y: [1, 4, 1, 1] 183s 0.621 183s 0.488 183s 0.359 183s -0.414 183s test-mul-mat0: iter:258/500 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.767 0.477 0.290 183s -0.414 -0.087 0.370 183s 0.616 0.366 0.020 183s -0.325 -0.239 -0.319 183s 183s x1: [3, 3, 1, 1] 183s -0.023 0.212 -0.881 183s -0.253 -0.604 -0.832 183s 0.120 -0.913 -0.610 183s 183s y: [4, 3, 1, 1] 183s -0.137 -0.335 0.046 0.238 183s -0.335 -0.150 -0.393 0.492 183s -0.704 -0.196 -0.272 0.374 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.164 0.608 -0.263 183s -0.077 0.999 -0.057 183s 0.156 0.475 -0.767 183s 0.742 -0.612 0.602 183s 183s x1: [3, 4, 1, 1] 183s -0.805 -0.857 0.240 183s -0.221 -0.918 0.228 183s 0.893 0.228 0.322 183s -0.130 -0.560 0.440 183s 183s y: [4, 4, 1, 1] 183s -0.452 -0.808 -0.717 0.072 183s -0.582 -0.912 -0.646 0.535 183s -0.092 0.141 0.001 0.717 183s -0.435 -0.574 -0.624 0.511 183s testing: mul_mat, [2, 4, 1, 3] = [3, 2, 1, 3] * [3, 4, 1, 3] 183s x0: [3, 2, 1, 3] 183s -0.313 -0.071 0.369 183s 0.695 0.670 0.140 183s 183s x1: [3, 4, 1, 3] 183s 0.358 0.753 -0.378 183s -0.966 -0.485 0.303 183s -0.771 -0.342 -0.457 183s 0.008 -0.259 0.771 183s 183s y: [2, 4, 1, 3] 183s -0.305 0.701 183s 0.449 -0.954 183s 0.097 -0.829 183s 0.301 -0.060 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.727 -0.333 -0.307 183s -0.578 -0.269 0.494 183s 183s x1: [3, 2, 1, 1] 183s -0.057 -0.217 -0.338 183s -0.758 0.849 -0.335 183s 183s y: [2, 2, 1, 1] 183s 0.217 -0.075 183s 0.371 0.045 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.070 0.002 183s -0.764 0.827 183s -0.926 -0.732 183s -0.023 0.910 183s 183s x1: [2, 3, 1, 1] 183s 0.418 0.622 183s -0.137 0.113 183s 0.292 -0.997 183s 183s y: [4, 3, 1, 1] 183s -0.028 0.196 -0.842 0.556 183s 0.010 0.198 0.044 0.106 183s -0.023 -1.047 0.459 -0.914 183s testing: mul_mat, [2, 2, 1, 3] = [4, 2, 1, 3] * [4, 2, 1, 3] 183s x0: [4, 2, 1, 3] 183s 0.676 -0.701 -0.511 0.207 183s -0.779 0.503 -0.791 -0.784 183s 183s x1: [4, 2, 1, 3] 183s 0.704 0.727 -0.043 -0.353 183s -0.491 0.620 -0.111 -0.642 183s 183s y: [2, 2, 1, 3] 183s -0.085 0.128 183s -0.842 1.285 183s test-mul-mat0: iter:259/500 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.382 -0.468 183s -0.532 0.591 183s 0.739 -0.317 183s -0.243 0.593 183s 183s x1: [2, 2, 1, 1] 183s 0.353 0.758 183s 0.083 0.028 183s 183s y: [4, 2, 1, 1] 183s -0.220 0.260 0.020 0.364 183s 0.018 -0.027 0.052 -0.003 183s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.474 0.071 183s 0.739 -0.414 183s 0.277 -0.869 183s 0.174 -0.439 183s 183s x1: [2, 4, 4, 1] 183s 0.408 0.543 183s -0.664 -0.508 183s -0.386 0.112 183s -0.794 -0.608 183s 183s y: [4, 4, 4, 1] 183s 0.232 0.077 -0.359 -0.167 183s -0.350 -0.280 0.258 0.108 183s -0.175 -0.332 -0.204 -0.116 183s -0.419 -0.335 0.308 0.129 183s testing: mul_mat, [1, 4, 4, 2] = [2, 1, 4, 2] * [2, 4, 4, 2] 183s x0: [2, 1, 4, 2] 183s 0.595 0.792 183s 183s x1: [2, 4, 4, 2] 183s 0.123 0.539 183s -0.339 -0.599 183s 0.670 0.835 183s -0.038 0.372 183s 183s y: [1, 4, 4, 2] 183s 0.501 183s -0.676 183s 1.060 183s 0.272 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.621 0.222 183s -0.920 0.921 183s -0.535 0.173 183s 183s x1: [2, 1, 1, 1] 183s 0.518 -0.826 183s 183s y: [3, 1, 1, 1] 183s -0.505 -1.237 -0.420 183s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.595 -0.172 -0.322 183s 183s x1: [3, 2, 4, 1] 183s -0.900 0.496 -0.841 183s -0.864 -0.655 -0.913 183s 183s y: [1, 2, 4, 1] 183s -0.350 183s -0.107 183s testing: mul_mat, [1, 3, 4, 2] = [1, 1, 4, 2] * [1, 3, 4, 2] 183s x0: [1, 1, 4, 2] 183s 0.255 183s 183s x1: [1, 3, 4, 2] 183s -0.215 183s -0.273 183s -0.114 183s 183s y: [1, 3, 4, 2] 183s -0.055 183s -0.070 183s -0.029 183s test-mul-mat0: iter:260/500 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.612 183s -0.847 183s -0.950 183s 183s x1: [1, 3, 1, 1] 183s 0.506 183s 0.375 183s -0.310 183s 183s y: [3, 3, 1, 1] 183s -0.310 -0.429 -0.481 183s -0.229 -0.318 -0.356 183s 0.190 0.262 0.294 183s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.417 183s 183s x1: [1, 3, 2, 1] 183s 0.099 183s -0.811 183s -0.832 183s 183s y: [1, 3, 2, 1] 183s -0.041 183s 0.338 183s 0.347 183s testing: mul_mat, [4, 1, 2, 1] = [1, 4, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.860 183s 0.549 183s 0.239 183s -0.551 183s 183s x1: [1, 1, 2, 1] 183s -0.116 183s 183s y: [4, 1, 2, 1] 183s 0.099 -0.063 -0.028 0.064 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.068 183s 183s x1: [1, 4, 1, 1] 183s -0.941 183s 0.269 183s 0.557 183s -0.668 183s 183s y: [1, 4, 1, 1] 183s 0.064 183s -0.018 183s -0.038 183s 0.045 183s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.175 183s 0.007 183s 183s x1: [1, 1, 2, 1] 183s 0.022 183s 183s y: [2, 1, 2, 1] 183s 0.004 0.000 183s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 1, 2, 1] 183s -0.081 0.182 183s 183s x1: [2, 1, 2, 1] 183s 0.175 0.445 183s 183s y: [1, 1, 2, 1] 183s 0.067 183s test-mul-mat0: iter:261/500 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.007 -0.181 183s 0.056 -0.725 183s -0.624 0.388 183s 0.050 0.309 183s 183s x1: [2, 1, 1, 1] 183s 0.464 -0.846 183s 183s y: [4, 1, 1, 1] 183s 0.156 0.639 -0.618 -0.238 183s testing: mul_mat, [2, 4, 2, 1] = [2, 2, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.623 0.477 183s 0.386 0.841 183s 183s x1: [2, 4, 2, 1] 183s -0.589 -0.993 183s -0.371 0.586 183s 0.014 0.049 183s -0.050 -0.812 183s 183s y: [2, 4, 2, 1] 183s -0.107 -1.063 183s 0.510 0.350 183s 0.015 0.047 183s -0.356 -0.702 183s testing: mul_mat, [3, 2, 2, 3] = [2, 3, 2, 3] * [2, 2, 2, 3] 183s x0: [2, 3, 2, 3] 183s -0.727 -0.626 183s -0.562 -0.344 183s 0.223 -0.112 183s 183s x1: [2, 2, 2, 3] 183s 0.439 0.124 183s -0.174 -0.173 183s 183s y: [3, 2, 2, 3] 183s -0.397 -0.289 0.084 183s 0.234 0.157 -0.019 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.810 0.387 183s 183s x1: [2, 3, 1, 1] 183s -0.458 -0.272 183s -0.836 -0.743 183s 0.650 -0.315 183s 183s y: [1, 3, 1, 1] 183s -0.476 183s -0.964 183s 0.404 183s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.434 183s 0.888 183s 0.366 183s 183s x1: [1, 2, 2, 1] 183s -0.278 183s -0.207 183s 183s y: [3, 2, 2, 1] 183s 0.121 -0.247 -0.102 183s 0.090 -0.184 -0.076 183s testing: mul_mat, [4, 1, 2, 3] = [3, 4, 2, 3] * [3, 1, 2, 3] 183s x0: [3, 4, 2, 3] 183s 0.975 -0.497 -0.877 183s 0.352 -0.243 -0.081 183s 0.207 0.826 0.811 183s 0.260 -0.609 -0.432 183s 183s x1: [3, 1, 2, 3] 183s 0.911 0.107 -0.146 183s 183s y: [4, 1, 2, 3] 183s 0.964 0.307 0.158 0.235 183s test-mul-mat0: iter:262/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.368 0.220 183s -0.542 -0.445 183s 0.715 0.630 183s 183s x1: [2, 2, 1, 1] 183s -0.598 -0.184 183s -0.385 -0.796 183s 183s y: [3, 2, 1, 1] 183s 0.180 0.406 -0.544 183s -0.034 0.563 -0.776 183s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.682 -0.479 183s 0.531 -0.764 183s -0.764 0.161 183s 183s x1: [2, 3, 4, 1] 183s -0.715 -0.790 183s -0.375 -0.588 183s 0.336 0.857 183s 183s y: [3, 3, 4, 1] 183s -0.109 0.224 0.419 183s 0.026 0.250 0.192 183s -0.181 -0.476 -0.120 183s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 2, 4, 1] 183s 0.541 0.532 183s 0.549 0.638 183s 183s x1: [2, 3, 4, 1] 183s 0.849 -0.120 183s 0.977 -0.583 183s 0.930 0.277 183s 183s y: [2, 3, 4, 1] 183s 0.396 0.389 183s 0.218 0.164 183s 0.651 0.687 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.438 -0.744 183s 0.920 -0.830 183s 183s x1: [2, 2, 1, 1] 183s -0.752 0.515 183s -0.405 -0.383 183s 183s y: [2, 2, 1, 1] 183s -0.053 -1.119 183s 0.463 -0.055 183s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 2, 4, 1] 183s 0.382 0.058 183s 0.060 -0.137 183s 183s x1: [2, 2, 4, 1] 183s 0.067 -0.255 183s 0.953 0.099 183s 183s y: [2, 2, 4, 1] 183s 0.011 0.039 183s 0.370 0.044 183s testing: mul_mat, [4, 2, 4, 1] = [2, 4, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.497 -0.450 183s -0.828 -0.042 183s -0.110 0.731 183s 0.808 -0.078 183s 183s x1: [2, 2, 4, 1] 183s -0.247 -0.992 183s -0.126 -0.931 183s 183s y: [4, 2, 4, 1] 183s 0.569 0.246 -0.697 -0.123 183s 0.482 0.144 -0.667 -0.030 183s test-mul-mat0: iter:263/500 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.533 -0.200 -0.649 183s -0.050 0.724 0.492 183s 0.442 0.522 -0.298 183s 183s x1: [3, 1, 1, 1] 183s -0.624 -0.106 0.249 183s 183s y: [3, 1, 1, 1] 183s 0.192 0.077 -0.406 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.894 0.115 0.043 183s 0.460 -0.108 -0.539 183s -0.377 0.218 -0.163 183s 0.517 -0.533 -0.200 183s 183s x1: [3, 3, 1, 1] 183s -0.418 0.225 -0.155 183s -0.352 0.846 0.215 183s 0.403 0.254 0.511 183s 183s y: [4, 3, 1, 1] 183s 0.393 -0.133 0.232 -0.305 183s 0.422 -0.369 0.282 -0.676 183s -0.309 -0.117 -0.180 -0.028 183s testing: mul_mat, [2, 4, 1, 2] = [3, 2, 1, 2] * [3, 4, 1, 2] 183s x0: [3, 2, 1, 2] 183s 0.027 -0.859 0.868 183s -0.456 -0.392 -0.332 183s 183s x1: [3, 4, 1, 2] 183s 0.985 0.267 0.152 183s -0.065 -0.009 -0.356 183s -0.623 -0.487 0.346 183s -0.041 0.738 -0.809 183s 183s y: [2, 4, 1, 2] 183s -0.071 -0.604 183s -0.303 0.151 183s 0.702 0.360 183s -1.337 -0.002 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.298 0.406 0.417 183s 183s x1: [3, 2, 1, 1] 183s 0.841 0.380 0.810 183s -0.200 0.118 -1.000 183s 183s y: [1, 2, 1, 1] 183s 0.241 183s -0.310 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.647 183s 0.512 183s -0.114 183s 0.276 183s 183s x1: [1, 3, 1, 1] 183s 0.539 183s -0.678 183s -0.057 183s 183s y: [4, 3, 1, 1] 183s -0.349 0.276 -0.061 0.149 183s 0.439 -0.347 0.077 -0.187 183s 0.037 -0.029 0.006 -0.016 183s testing: mul_mat, [1, 1, 1, 2] = [4, 1, 1, 2] * [4, 1, 1, 2] 183s x0: [4, 1, 1, 2] 183s -0.102 -0.550 -0.232 0.362 183s 183s x1: [4, 1, 1, 2] 183s -0.457 0.913 0.417 -0.590 183s 183s y: [1, 1, 1, 2] 183s -0.766 183s test-mul-mat0: iter:264/500 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.989 0.253 183s 0.370 -0.124 183s -0.471 0.912 183s -0.211 0.946 183s 183s x1: [2, 2, 1, 1] 183s -0.422 -0.085 183s 0.506 -0.100 183s 183s y: [4, 2, 1, 1] 183s -0.439 -0.145 0.121 0.008 183s 0.476 0.200 -0.330 -0.202 183s testing: mul_mat, [4, 4, 2, 1] = [2, 4, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.981 -0.416 183s -0.522 0.526 183s 0.484 -0.664 183s 0.515 -0.263 183s 183s x1: [2, 4, 2, 1] 183s -0.677 0.247 183s 0.971 0.401 183s -0.767 0.872 183s -0.700 -0.316 183s 183s y: [4, 4, 2, 1] 183s 0.561 0.483 -0.492 -0.414 183s -1.119 -0.296 0.203 0.395 183s 0.389 0.859 -0.950 -0.624 183s 0.818 0.199 -0.129 -0.278 183s testing: mul_mat, [3, 4, 2, 4] = [2, 3, 2, 4] * [2, 4, 2, 4] 183s x0: [2, 3, 2, 4] 183s -0.600 -0.605 183s 0.181 0.756 183s -0.290 0.094 183s 183s x1: [2, 4, 2, 4] 183s 0.182 0.342 183s -0.340 0.054 183s 0.642 0.344 183s 0.694 0.304 183s 183s y: [3, 4, 2, 4] 183s -0.316 0.292 -0.021 183s 0.171 -0.021 0.104 183s -0.593 0.376 -0.154 183s -0.600 0.355 -0.173 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.525 0.658 183s 183s x1: [2, 3, 1, 1] 183s 0.095 0.655 183s 0.893 -0.254 183s -0.282 0.711 183s 183s y: [1, 3, 1, 1] 183s 0.481 183s 0.302 183s 0.319 183s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 4, 2, 1] 183s 0.430 183s -0.485 183s -0.695 183s -0.727 183s 183s x1: [1, 2, 2, 1] 183s -0.959 183s 0.668 183s 183s y: [4, 2, 2, 1] 183s -0.413 0.466 0.666 0.698 183s 0.287 -0.324 -0.464 -0.486 183s testing: mul_mat, [4, 1, 2, 4] = [4, 4, 2, 4] * [4, 1, 2, 4] 183s x0: [4, 4, 2, 4] 183s 0.606 0.665 -0.738 -0.805 183s 0.580 0.044 -0.377 0.082 183s 0.482 -0.238 -0.229 0.658 183s -0.205 -0.155 -0.803 0.425 183s 183s x1: [4, 1, 2, 4] 183s 0.704 0.107 0.034 -0.262 183s 183s y: [4, 1, 2, 4] 183s 0.683 0.379 0.134 -0.300 183s test-mul-mat0: iter:265/500 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.664 -0.527 183s 0.662 0.346 183s 183s x1: [2, 2, 1, 1] 183s -0.506 0.638 183s 0.790 -0.251 183s 183s y: [2, 2, 1, 1] 183s -0.000 -0.114 183s -0.392 0.436 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.027 0.652 183s -0.677 0.465 183s 183s x1: [2, 2, 1, 1] 183s 0.446 -0.650 183s 0.185 -0.239 183s 183s y: [2, 2, 1, 1] 183s -0.412 -0.604 183s -0.151 -0.237 183s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 1, 1, 2] 183s -0.373 -0.589 183s 183s x1: [2, 2, 1, 2] 183s -0.368 -0.788 183s 0.080 -0.665 183s 183s y: [1, 2, 1, 2] 183s 0.602 183s 0.362 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.654 -0.037 183s 0.302 -0.693 183s -0.039 0.573 183s -0.873 0.313 183s 183s x1: [2, 1, 1, 1] 183s -0.095 0.080 183s 183s y: [4, 1, 1, 1] 183s -0.065 -0.084 0.050 0.108 183s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.065 0.111 -0.961 0.481 183s -0.048 -0.949 -0.323 0.513 183s -0.842 0.856 -0.733 0.172 183s 183s x1: [4, 2, 1, 1] 183s 0.492 -0.666 0.993 -0.480 183s 0.986 -0.684 0.985 -0.383 183s 183s y: [3, 2, 1, 1] 183s -1.228 0.042 -1.795 183s -1.143 0.088 -2.203 183s testing: mul_mat, [2, 4, 1, 2] = [3, 2, 1, 2] * [3, 4, 1, 2] 183s x0: [3, 2, 1, 2] 183s 0.617 0.258 -0.344 183s 0.143 0.098 -0.570 183s 183s x1: [3, 4, 1, 2] 183s -0.439 -0.863 -0.174 183s 0.863 0.098 -0.047 183s -0.174 0.405 -0.474 183s -0.861 -0.102 -0.141 183s 183s y: [2, 4, 1, 2] 183s -0.433 -0.048 183s 0.574 0.160 183s 0.161 0.285 183s -0.509 -0.053 183s test-mul-mat0: iter:266/500 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.482 -0.684 0.180 0.060 183s -0.219 -0.829 -0.324 0.924 183s 183s x1: [4, 2, 1, 1] 183s -0.806 0.594 -0.380 0.612 183s -0.561 0.068 0.015 0.901 183s 183s y: [2, 2, 1, 1] 183s -0.826 0.373 183s -0.261 0.894 183s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 2, 3, 1] 183s -0.297 0.923 0.087 -0.126 183s -0.400 0.011 -0.698 -0.626 183s 183s x1: [4, 2, 3, 1] 183s 0.429 0.774 -0.420 0.858 183s 0.433 0.373 0.114 -0.045 183s 183s y: [2, 2, 3, 1] 183s 0.442 -0.407 183s 0.231 -0.220 183s testing: mul_mat, [2, 2, 3, 4] = [4, 2, 3, 4] * [4, 2, 3, 4] 183s x0: [4, 2, 3, 4] 183s 0.665 0.962 -0.388 0.418 183s 0.231 -0.159 -0.203 -0.472 183s 183s x1: [4, 2, 3, 4] 183s -0.606 0.743 0.294 -0.616 183s 0.266 -0.843 -0.310 0.969 183s 183s y: [2, 2, 3, 4] 183s -0.060 -0.027 183s -0.108 -0.199 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.389 0.543 0.197 0.269 183s 183s x1: [4, 2, 1, 1] 183s -0.575 -0.250 -0.388 0.809 183s -0.391 -0.374 0.478 0.796 183s 183s y: [1, 2, 1, 1] 183s 0.229 183s 0.258 183s testing: mul_mat, [2, 4, 3, 1] = [1, 2, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.464 183s -0.721 183s 183s x1: [1, 4, 3, 1] 183s -0.026 183s -0.920 183s -0.763 183s -0.638 183s 183s y: [2, 4, 3, 1] 183s 0.012 0.018 183s 0.427 0.663 183s 0.354 0.550 183s 0.296 0.460 183s testing: mul_mat, [3, 1, 3, 4] = [2, 3, 3, 4] * [2, 1, 3, 4] 183s x0: [2, 3, 3, 4] 183s 0.312 0.157 183s -0.644 0.561 183s 0.141 0.222 183s 183s x1: [2, 1, 3, 4] 183s -0.404 0.817 183s 183s y: [3, 1, 3, 4] 183s 0.002 0.718 0.125 183s test-mul-mat0: iter:267/500 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.640 -0.142 183s 0.901 0.083 183s -0.636 -0.564 183s 0.171 0.571 183s 183s x1: [2, 1, 1, 1] 183s -0.165 0.254 183s 183s y: [4, 1, 1, 1] 183s 0.070 -0.128 -0.039 0.117 183s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.551 -0.838 183s 0.847 0.646 183s -0.162 -0.155 183s 0.887 -0.181 183s 183s x1: [2, 4, 4, 1] 183s -0.863 -0.245 183s -0.312 0.779 183s 0.912 0.197 183s -0.325 -0.002 183s 183s y: [4, 4, 4, 1] 183s 0.681 -0.890 0.178 -0.721 183s -0.481 0.239 -0.070 -0.418 183s -0.668 0.900 -0.179 0.773 183s 0.181 -0.276 0.053 -0.287 183s testing: mul_mat, [1, 4, 4, 3] = [2, 1, 4, 3] * [2, 4, 4, 3] 183s x0: [2, 1, 4, 3] 183s -0.089 -0.850 183s 183s x1: [2, 4, 4, 3] 183s -0.526 0.834 183s -0.136 0.312 183s -0.320 -0.249 183s -0.869 0.099 183s 183s y: [1, 4, 4, 3] 183s -0.662 183s -0.253 183s 0.240 183s -0.007 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.976 0.962 183s -0.165 0.745 183s 0.936 -0.914 183s 183s x1: [2, 1, 1, 1] 183s 0.909 0.169 183s 183s y: [3, 1, 1, 1] 183s -0.724 -0.024 0.696 183s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.103 0.056 0.136 183s 183s x1: [3, 2, 4, 1] 183s -0.478 0.909 0.021 183s -0.094 0.869 -0.102 183s 183s y: [1, 2, 4, 1] 183s 0.004 183s 0.025 183s testing: mul_mat, [2, 3, 4, 3] = [1, 2, 4, 3] * [1, 3, 4, 3] 183s x0: [1, 2, 4, 3] 183s -0.514 183s 0.256 183s 183s x1: [1, 3, 4, 3] 183s 0.525 183s 0.867 183s 0.702 183s 183s y: [2, 3, 4, 3] 183s -0.270 0.135 183s -0.446 0.222 183s -0.361 0.180 183s test-mul-mat0: iter:268/500 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.501 0.558 0.823 183s -0.804 -0.400 0.168 183s 0.500 0.670 -0.007 183s 183s x1: [3, 1, 1, 1] 183s 0.002 -0.755 0.492 183s 183s y: [3, 1, 1, 1] 183s -0.016 0.383 -0.508 183s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 3, 2, 1] 183s 0.280 -0.430 -0.064 183s -0.219 -0.872 -0.241 183s -0.022 -0.272 0.927 183s 183s x1: [3, 3, 2, 1] 183s -0.510 -0.011 -0.502 183s -0.731 0.613 -0.180 183s 0.903 -0.658 0.180 183s 183s y: [3, 3, 2, 1] 183s -0.106 0.242 -0.451 183s -0.457 -0.331 -0.318 183s 0.525 0.333 0.326 183s testing: mul_mat, [4, 3, 2, 3] = [3, 4, 2, 3] * [3, 3, 2, 3] 183s x0: [3, 4, 2, 3] 183s 0.478 0.997 -0.058 183s 0.435 -0.190 0.357 183s 0.629 -0.237 0.114 183s 0.242 -0.169 -0.900 183s 183s x1: [3, 3, 2, 3] 183s 0.139 0.341 0.418 183s -0.885 0.229 0.240 183s -0.902 -0.940 0.128 183s 183s y: [4, 3, 2, 3] 183s 0.382 0.145 0.054 -0.401 183s -0.208 -0.342 -0.584 -0.469 183s -1.376 -0.168 -0.330 -0.174 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.793 0.987 -0.343 183s -0.454 0.904 0.165 183s -0.056 0.146 -0.847 183s -0.174 0.542 -0.362 183s 183s x1: [3, 4, 1, 1] 183s 0.900 -0.821 0.131 183s -0.001 -0.049 0.402 183s 0.807 0.009 0.935 183s 0.028 -0.130 0.695 183s 183s y: [4, 4, 1, 1] 183s -1.569 -1.129 -0.281 -0.649 183s -0.185 0.022 -0.347 -0.172 183s -0.951 -0.203 -0.836 -0.474 183s -0.389 -0.016 -0.609 -0.327 183s testing: mul_mat, [4, 3, 2, 1] = [4, 4, 2, 1] * [4, 3, 2, 1] 183s x0: [4, 4, 2, 1] 183s -0.081 -0.195 0.143 -0.346 183s -0.349 0.667 0.265 0.624 183s 0.029 0.775 0.714 0.101 183s -0.916 0.983 0.557 0.042 183s 183s x1: [4, 3, 2, 1] 183s 0.692 0.178 0.476 0.489 183s 0.731 -0.587 0.389 0.910 183s 0.545 -0.612 -0.139 -0.054 183s 183s y: [4, 3, 2, 1] 183s -0.192 0.309 0.548 -0.172 183s -0.204 0.025 -0.064 -0.991 183s 0.074 -0.668 -0.563 -1.180 183s testing: mul_mat, [4, 4, 2, 3] = [4, 4, 2, 3] * [4, 4, 2, 3] 183s x0: [4, 4, 2, 3] 183s 0.620 0.254 0.585 0.930 183s 0.611 -0.988 0.054 -0.393 183s -0.213 0.613 -0.421 -0.867 183s -0.759 0.725 -0.494 -0.463 183s 183s x1: [4, 4, 2, 3] 183s -0.609 0.850 -0.182 0.196 183s 0.517 -0.406 0.179 -0.340 183s 0.858 -0.107 -0.783 -0.487 183s -0.483 0.318 0.555 0.002 183s 183s y: [4, 4, 2, 3] 183s -0.085 -1.298 0.557 1.077 183s 0.006 0.860 -0.140 -0.618 183s -0.407 0.779 0.504 -0.117 183s 0.107 -0.580 0.063 0.323 183s test-mul-mat0: iter:269/500 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.313 0.321 183s 0.881 -0.250 183s -0.912 -0.498 183s -0.073 0.257 183s 183s x1: [2, 4, 1, 1] 183s -0.280 0.879 183s -0.316 0.793 183s -0.645 0.698 183s -0.905 0.196 183s 183s y: [4, 4, 1, 1] 183s 0.370 -0.467 -0.182 0.247 183s 0.354 -0.477 -0.107 0.227 183s 0.426 -0.743 0.241 0.227 183s 0.346 -0.847 0.727 0.117 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.577 -0.844 183s 0.895 -0.942 183s 0.783 -0.386 183s 0.936 -0.533 183s 183s x1: [2, 4, 1, 1] 183s -0.918 0.185 183s 0.999 -0.430 183s 0.023 0.768 183s 0.307 0.561 183s 183s y: [4, 4, 1, 1] 183s 0.374 -0.996 -0.790 -0.958 183s -0.214 1.300 0.948 1.165 183s -0.661 -0.703 -0.278 -0.388 183s -0.650 -0.253 0.024 -0.011 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.940 -0.103 183s 0.754 -0.134 183s 183s x1: [2, 4, 1, 1] 183s -0.592 -0.709 183s -0.835 -0.497 183s 0.487 0.370 183s 0.190 -0.191 183s 183s y: [2, 4, 1, 1] 183s -0.484 -0.352 183s -0.734 -0.563 183s 0.420 0.318 183s 0.198 0.169 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.075 0.712 183s 0.922 -0.365 183s 183s x1: [2, 2, 1, 1] 183s -0.845 0.835 183s -0.949 -0.846 183s 183s y: [2, 2, 1, 1] 183s 0.532 -1.084 183s -0.673 -0.566 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.426 0.041 183s -0.942 -0.534 183s -0.871 -0.579 183s 183s x1: [2, 2, 1, 1] 183s 0.549 -0.865 183s -0.209 0.443 183s 183s y: [3, 2, 1, 1] 183s -0.269 -0.055 0.023 183s 0.107 -0.040 -0.075 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.902 0.614 0.581 183s 183s x1: [3, 2, 1, 1] 183s 0.206 -0.032 0.908 183s -0.424 -0.842 -0.283 183s 183s y: [1, 2, 1, 1] 183s 0.323 183s -0.299 183s test-mul-mat0: iter:270/500 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.164 0.862 183s -0.908 -0.106 183s 0.992 0.133 183s 0.359 -0.588 183s 183s x1: [2, 4, 1, 1] 183s 0.922 0.822 183s -0.909 0.844 183s 0.534 -0.274 183s 0.393 0.669 183s 183s y: [4, 4, 1, 1] 183s 0.557 -0.925 1.024 -0.152 183s 0.877 0.736 -0.790 -0.823 183s -0.324 -0.456 0.493 0.353 183s 0.513 -0.428 0.478 -0.252 183s testing: mul_mat, [2, 4, 4, 1] = [2, 2, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 2, 4, 1] 183s 0.353 0.377 183s 0.481 -0.609 183s 183s x1: [2, 4, 4, 1] 183s -0.661 -0.672 183s -0.680 -0.085 183s -0.514 0.037 183s -0.257 -0.416 183s 183s y: [2, 4, 4, 1] 183s -0.487 0.092 183s -0.272 -0.275 183s -0.168 -0.270 183s -0.248 0.130 183s testing: mul_mat, [3, 2, 4, 2] = [2, 3, 4, 2] * [2, 2, 4, 2] 183s x0: [2, 3, 4, 2] 183s 0.505 0.033 183s -0.547 0.487 183s 0.277 -0.443 183s 183s x1: [2, 2, 4, 2] 183s -0.152 -0.785 183s 0.567 0.615 183s 183s y: [3, 2, 4, 2] 183s -0.103 -0.299 0.306 183s 0.307 -0.011 -0.115 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.688 -0.547 183s -0.022 -0.372 183s 0.284 -0.553 183s 183s x1: [2, 3, 1, 1] 183s -0.994 -0.430 183s 0.501 -0.637 183s 0.296 0.265 183s 183s y: [3, 3, 1, 1] 183s 0.919 0.182 -0.044 183s 0.004 0.226 0.495 183s -0.349 -0.105 -0.063 183s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.964 0.964 0.843 183s 183s x1: [3, 2, 4, 1] 183s 0.932 0.782 -0.357 183s 0.864 -0.576 -0.883 183s 183s y: [1, 2, 4, 1] 183s 1.352 183s -0.466 183s testing: mul_mat, [3, 3, 4, 2] = [1, 3, 4, 2] * [1, 3, 4, 2] 183s x0: [1, 3, 4, 2] 183s -0.010 183s -0.636 183s -0.484 183s 183s x1: [1, 3, 4, 2] 183s -0.434 183s -0.359 183s -0.218 183s 183s y: [3, 3, 4, 2] 183s 0.004 0.276 0.210 183s 0.004 0.228 0.174 183s 0.002 0.139 0.106 183s test-mul-mat0: iter:271/500 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.148 -0.650 0.258 -0.187 183s -0.211 0.775 0.336 0.693 183s -0.200 0.073 -0.151 -0.449 183s -0.320 -0.387 -0.251 0.180 183s 183s x1: [4, 2, 1, 1] 183s -0.111 0.021 -0.172 0.879 183s 0.385 0.344 0.082 -0.453 183s 183s y: [4, 2, 1, 1] 183s -0.206 0.591 -0.345 0.229 183s -0.175 -0.101 0.139 -0.358 183s testing: mul_mat, [1, 4, 2, 1] = [4, 1, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 1, 2, 1] 183s 0.800 -0.819 -0.615 -0.824 183s 183s x1: [4, 4, 2, 1] 183s -0.363 -0.618 0.131 -0.115 183s 0.413 -0.625 0.774 -0.566 183s 0.203 0.653 0.818 -0.453 183s -0.264 -0.634 0.457 0.587 183s 183s y: [1, 4, 2, 1] 183s 0.230 183s 0.833 183s -0.502 183s -0.457 183s testing: mul_mat, [3, 1, 2, 4] = [4, 3, 2, 4] * [4, 1, 2, 4] 183s x0: [4, 3, 2, 4] 183s -0.487 -0.878 0.697 -0.556 183s 0.840 0.794 -0.225 0.063 183s -0.620 -0.157 0.484 -0.744 183s 183s x1: [4, 1, 2, 4] 183s -0.363 0.925 -0.656 0.003 183s 183s y: [3, 1, 2, 4] 183s -1.094 0.577 -0.239 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.600 -0.296 -0.146 -0.825 183s 183s x1: [4, 3, 1, 1] 183s 0.249 0.731 -0.796 -0.586 183s -0.333 -0.422 -0.013 -0.065 183s -0.247 -0.311 -0.513 -0.298 183s 183s y: [1, 3, 1, 1] 183s 0.234 183s 0.380 183s 0.562 183s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.918 183s -0.672 183s -0.223 183s -0.275 183s 183s x1: [1, 4, 2, 1] 183s -0.683 183s -0.605 183s 0.917 183s -0.238 183s 183s y: [4, 4, 2, 1] 183s 0.627 0.459 0.152 0.188 183s 0.555 0.406 0.135 0.166 183s -0.842 -0.616 -0.204 -0.252 183s 0.219 0.160 0.053 0.066 183s testing: mul_mat, [3, 1, 2, 4] = [4, 3, 2, 4] * [4, 1, 2, 4] 183s x0: [4, 3, 2, 4] 183s 0.205 -0.558 -0.618 0.024 183s -0.875 -0.760 0.959 -0.467 183s 0.906 -0.550 0.395 -0.485 183s 183s x1: [4, 1, 2, 4] 183s -0.561 0.122 -0.515 0.427 183s 183s y: [3, 1, 2, 4] 183s 0.146 -0.296 -0.985 183s test-mul-mat0: iter:272/500 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.532 183s -0.112 183s -0.188 183s 0.738 183s 183s x1: [1, 4, 1, 1] 183s -0.550 183s -0.517 183s 0.031 183s 0.728 183s 183s y: [4, 4, 1, 1] 183s 0.292 0.061 0.104 -0.405 183s 0.275 0.058 0.097 -0.381 183s -0.017 -0.003 -0.006 0.023 183s -0.387 -0.081 -0.137 0.537 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.992 183s -0.436 183s 183s x1: [1, 4, 1, 1] 183s 0.598 183s -0.081 183s -0.992 183s -0.696 183s 183s y: [2, 4, 1, 1] 183s 0.593 -0.261 183s -0.080 0.035 183s -0.985 0.433 183s -0.691 0.303 183s testing: mul_mat, [4, 2, 1, 2] = [1, 4, 1, 2] * [1, 2, 1, 2] 183s x0: [1, 4, 1, 2] 183s 0.947 183s -0.338 183s -0.173 183s -0.545 183s 183s x1: [1, 2, 1, 2] 183s 0.473 183s -0.324 183s 183s y: [4, 2, 1, 2] 183s 0.447 -0.160 -0.082 -0.258 183s -0.307 0.110 0.056 0.177 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.099 183s -0.410 183s -0.769 183s 183s x1: [1, 4, 1, 1] 183s 0.147 183s -0.269 183s 0.173 183s 0.630 183s 183s y: [3, 4, 1, 1] 183s -0.014 -0.060 -0.113 183s 0.027 0.110 0.207 183s -0.017 -0.071 -0.133 183s -0.062 -0.259 -0.485 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.680 0.968 0.003 183s 0.976 -0.745 -0.278 183s 0.691 0.410 -0.587 183s 0.937 -0.356 -0.050 183s 183s x1: [3, 1, 1, 1] 183s 0.789 0.401 0.968 183s 183s y: [4, 1, 1, 1] 183s -0.145 0.203 0.141 0.549 183s testing: mul_mat, [2, 3, 1, 2] = [4, 2, 1, 2] * [4, 3, 1, 2] 183s x0: [4, 2, 1, 2] 183s -0.009 0.476 -0.890 0.097 183s 0.697 0.713 -0.574 0.350 183s 183s x1: [4, 3, 1, 2] 183s 0.384 0.240 0.405 -0.447 183s -0.445 -0.585 0.276 0.702 183s 0.146 -0.551 0.332 -0.091 183s 183s y: [2, 3, 1, 2] 183s -0.293 0.050 183s -0.451 -0.639 183s -0.568 -0.513 183s test-mul-mat0: iter:273/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.392 0.409 -0.313 183s 0.081 -0.601 -0.616 183s 183s x1: [3, 2, 1, 1] 183s 0.441 0.864 -0.815 183s -0.450 -0.610 0.828 183s 183s y: [2, 2, 1, 1] 183s 0.435 0.019 183s -0.332 -0.180 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.141 0.750 0.059 183s 0.784 0.357 -0.548 183s 183s x1: [3, 2, 1, 1] 183s -0.444 -0.888 -0.506 183s -0.018 0.209 0.844 183s 183s y: [2, 2, 1, 1] 183s -0.633 -0.387 183s 0.209 -0.402 183s testing: mul_mat, [4, 2, 1, 4] = [3, 4, 1, 4] * [3, 2, 1, 4] 183s x0: [3, 4, 1, 4] 183s 0.159 0.538 0.236 183s -0.057 -0.105 0.687 183s -0.899 0.192 -0.560 183s 0.983 -0.650 -0.365 183s 183s x1: [3, 2, 1, 4] 183s 0.158 -0.703 -0.247 183s 0.882 0.158 -0.805 183s 183s y: [4, 2, 1, 4] 183s -0.411 -0.105 -0.139 0.702 183s 0.036 -0.620 -0.311 1.059 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.976 -0.378 -0.588 183s -0.295 -0.217 -0.446 183s -0.655 -0.161 -0.358 183s 0.564 -0.360 -0.608 183s 183s x1: [3, 4, 1, 1] 183s 0.486 -0.797 0.981 183s -0.691 -0.365 -0.597 183s -0.418 0.448 0.917 183s -0.201 0.624 0.731 183s 183s y: [4, 4, 1, 1] 183s 0.200 -0.408 -0.541 -0.035 183s -0.186 0.549 0.725 0.105 183s -1.116 -0.383 -0.128 -0.955 183s -0.862 -0.402 -0.231 -0.783 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.747 -0.365 -0.043 -0.744 183s -0.097 -0.388 0.417 -0.398 183s 0.809 0.853 0.961 -0.848 183s 0.877 0.340 0.740 -0.190 183s 183s x1: [4, 3, 1, 1] 183s -0.097 -0.628 -0.935 -0.368 183s 0.967 -0.398 -0.881 -0.830 183s -0.417 -0.572 -0.195 -0.014 183s 183s y: [4, 3, 1, 1] 183s 0.615 0.010 -1.201 -0.920 183s 0.079 0.024 0.298 0.219 183s 0.539 0.187 -1.001 -0.702 183s testing: mul_mat, [2, 4, 1, 4] = [4, 2, 1, 4] * [4, 4, 1, 4] 183s x0: [4, 2, 1, 4] 183s -0.851 -0.988 0.417 0.220 183s 0.881 -0.191 0.556 -0.253 183s 183s x1: [4, 4, 1, 4] 183s 0.243 -0.456 0.713 0.615 183s -0.391 -0.655 0.582 0.211 183s -0.536 0.751 0.794 -0.108 183s -0.444 -0.220 0.902 -0.191 183s 183s y: [2, 4, 1, 4] 183s 0.676 0.542 183s 1.269 0.051 183s 0.021 -0.148 183s 0.929 0.201 183s test-mul-mat0: iter:274/500 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.750 -0.044 183s 0.359 -0.421 183s 183s x1: [2, 1, 1, 1] 183s -0.742 -0.231 183s 183s y: [2, 1, 1, 1] 183s 0.567 -0.169 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.702 -0.222 183s 0.297 -0.903 183s 0.169 0.212 183s 0.160 -0.267 183s 183s x1: [2, 2, 1, 1] 183s -0.080 -0.991 183s -0.045 0.347 183s 183s y: [4, 2, 1, 1] 183s 0.164 0.872 -0.223 0.252 183s -0.109 -0.327 0.066 -0.100 183s testing: mul_mat, [1, 4, 1, 4] = [2, 1, 1, 4] * [2, 4, 1, 4] 183s x0: [2, 1, 1, 4] 183s -0.220 0.823 183s 183s x1: [2, 4, 1, 4] 183s -0.267 -0.937 183s -0.507 0.326 183s -0.479 0.184 183s -0.933 -0.140 183s 183s y: [1, 4, 1, 4] 183s -0.713 183s 0.379 183s 0.257 183s 0.090 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.934 -0.610 183s 183s x1: [2, 1, 1, 1] 183s -0.053 0.996 183s 183s y: [1, 1, 1, 1] 183s -0.557 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.493 183s 0.031 183s 183s x1: [1, 2, 1, 1] 183s 0.232 183s -0.456 183s 183s y: [2, 2, 1, 1] 183s 0.115 0.007 183s -0.225 -0.014 183s testing: mul_mat, [3, 1, 1, 4] = [2, 3, 1, 4] * [2, 1, 1, 4] 183s x0: [2, 3, 1, 4] 183s 0.536 -0.515 183s 0.330 -0.890 183s 0.100 -0.077 183s 183s x1: [2, 1, 1, 4] 183s 0.292 0.102 183s 183s y: [3, 1, 1, 4] 183s 0.104 0.006 0.021 183s test-mul-mat0: iter:275/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.897 0.703 -0.440 183s 183s x1: [3, 1, 1, 1] 183s -0.461 -0.433 -0.398 183s 183s y: [1, 1, 1, 1] 183s -0.543 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.438 -0.197 -0.993 183s 183s x1: [3, 1, 1, 1] 183s -0.993 -0.374 -0.484 183s 183s y: [1, 1, 1, 1] 183s 0.119 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.335 -0.761 0.406 183s 183s x1: [3, 1, 1, 1] 183s 0.756 -0.964 0.928 183s 183s y: [1, 1, 1, 1] 183s 1.364 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.119 0.213 0.901 183s -0.623 -0.229 0.213 183s 183s x1: [3, 1, 1, 1] 183s 0.178 0.914 -0.659 183s 183s y: [2, 1, 1, 1] 183s -0.421 -0.460 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.628 -0.546 183s -0.857 0.582 183s -0.539 0.264 183s 183s x1: [2, 3, 1, 1] 183s 0.310 -0.532 183s 0.815 0.385 183s -0.635 0.518 183s 183s y: [3, 3, 1, 1] 183s 0.096 -0.575 -0.307 183s -0.722 -0.474 -0.337 183s 0.116 0.845 0.479 183s testing: mul_mat, [1, 2, 1, 1]ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.291 0.814 -0.457 183s 183s x1: [3, 2, 1, 1] 183s -0.539 0.338 0.300 183s -0.611 0.303 -0.365 183s 183s y: [1, 2, 1, 1] 183s 0.294 183s 0.591 183s test-mul-mat0: iter:276/500 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.109 -0.839 0.624 183s -0.164 -0.467 0.768 183s 0.298 -0.013 0.349 183s -0.438 0.449 -0.313 183s 183s x1: [3, 2, 1, 1] 183s 0.079 -0.805 0.328 183s -0.708 0.506 0.796 183s 183s y: [4, 2, 1, 1] 183s 0.871 0.614 0.148 -0.498 183s 0.150 0.491 0.061 0.288 183s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 1, 3, 1] 183s -0.610 -0.258 0.520 183s 183s x1: [3, 4, 3, 1] 183s 0.862 0.838 0.990 183s -0.503 -0.534 -0.301 183s -0.689 0.009 0.748 183s -0.823 -0.567 0.175 183s 183s y: [1, 4, 3, 1] 183s -0.228 183s 0.288 183s 0.807 183s 0.740 183s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.982 -0.074 0.747 183s 0.702 -0.645 -0.170 183s 183s x1: [3, 1, 3, 1] 183s 0.371 0.937 0.495 183s 183s y: [2, 1, 3, 1] 183s 0.665 -0.429 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.973 -0.159 0.482 183s -0.465 -0.204 -0.807 183s 183s x1: [3, 2, 1, 1] 183s -0.859 0.074 0.086 183s 0.512 0.011 -0.419 183s 183s y: [2, 2, 1, 1] 183s -0.806 0.315 183s 0.294 0.098 183s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 4, 3, 1] 183s -0.821 0.097 183s -0.484 0.386 183s -0.151 0.163 183s 0.190 0.632 183s 183s x1: [2, 3, 3, 1] 183s -0.222 -0.592 183s 0.940 -0.521 183s -0.237 -0.230 183s 183s y: [4, 3, 3, 1] 183s 0.125 -0.121 -0.063 -0.416 183s -0.822 -0.656 -0.227 -0.151 183s 0.172 0.026 -0.002 -0.190 183s testing: mul_mat, [1, 2, 3, 1] = [4, 1, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.233 0.956 0.687 -0.129 183s 183s x1: [4, 2, 3, 1] 183s -0.019 -0.758 0.691 -0.823 183s 0.347 0.121 0.514 0.526 183s 183s y: [1, 2, 3, 1] 183s -0.140 183s 0.320 183s test-mul-mat0: iter:277/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.387 0.630 0.909 183s -0.133 -0.975 -0.517 183s 183s x1: [3, 1, 1, 1] 183s -0.805 0.020 0.654 183s 183s y: [2, 1, 1, 1] 183s 0.296 -0.251 183s testing: mul_mat, [3, 2, 4, 1] = [3, 3, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.058 -0.478 0.694 183s 0.926 -0.452 -0.824 183s 0.795 0.328 -0.552 183s 183s x1: [3, 2, 4, 1] 183s 0.869 -0.219 -0.729 183s 0.156 -0.334 0.039 183s 183s y: [3, 2, 4, 1] 183s -0.350 1.504 1.021 183s 0.196 0.264 -0.007 183s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.314 0.824 0.401 183s -0.439 -0.902 0.092 183s 183s x1: [3, 3, 4, 1] 183s -0.456 0.720 -0.187 183s -0.008 -0.329 -0.193 183s -0.522 -0.266 0.154 183s 183s y: [2, 3, 4, 1] 183s 0.661 -0.466 183s -0.346 0.283 183s 0.007 0.483 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.013 0.104 -0.183 183s 0.005 0.781 -0.256 183s -0.311 0.168 -0.340 183s 0.574 0.101 -0.586 183s 183s x1: [3, 2, 1, 1] 183s 0.602 -0.880 -0.908 183s 0.041 -0.673 0.182 183s 183s y: [4, 2, 1, 1] 183s 0.083 -0.452 -0.026 0.787 183s -0.103 -0.572 -0.187 -0.151 183s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 1, 4, 1] 183s -0.886 -0.082 -0.999 0.762 183s 183s x1: [4, 3, 4, 1] 183s -0.413 -0.042 0.844 -0.782 183s -0.114 -0.954 -0.947 0.118 183s 0.022 0.028 -0.978 -0.304 183s 183s y: [1, 3, 4, 1] 183s -1.070 183s 1.216 183s 0.724 183s testing: mul_mat, [2, 4, 4, 1] = [1, 2, 4, 1] * [1, 4, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.079 183s 0.686 183s 183s x1: [1, 4, 4, 1] 183s -0.508 183s 0.583 183s 0.155 183s -0.682 183s 183s y: [2, 4, 4, 1] 183s -0.040 -0.348 183s 0.046 0.400 183s 0.012 0.106 183s -0.054 -0.468 183s test-mul-mat0: iter:278/500 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.002 -0.825 183s 183s x1: [2, 1, 1, 1] 183s 0.415 0.020 183s 183s y: [1, 1, 1, 1] 183s -0.017 183s testing: mul_mat, [4, 1, 4, 1] = [2, 4, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.036 -0.608 183s 0.030 -0.040 183s -0.075 -0.891 183s -0.354 0.433 183s 183s x1: [2, 1, 4, 1] 183s 0.401 0.385 183s 183s y: [4, 1, 4, 1] 183s -0.220 -0.003 -0.373 0.025 183s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 1, 4, 1] 183s 0.967 0.152 183s 183s x1: [2, 4, 4, 1] 183s -0.235 0.235 183s -0.123 0.690 183s 0.344 0.523 183s 0.123 -0.389 183s 183s y: [1, 4, 4, 1] 183s -0.191 183s -0.013 183s 0.412 183s 0.060 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.663 -0.018 183s -0.435 -0.770 183s -0.322 0.503 183s 0.776 -0.149 183s 183s x1: [2, 1, 1, 1] 183s -0.834 0.423 183s 183s y: [4, 1, 1, 1] 183s -0.560 0.038 0.481 -0.710 183s testing: mul_mat, [2, 2, 4, 1] = [4, 2, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.695 0.171 0.746 -0.916 183s 0.418 0.854 -0.578 0.165 183s 183s x1: [4, 2, 4, 1] 183s -0.134 -0.770 0.073 0.157 183s -0.446 0.549 -0.671 0.252 183s 183s y: [2, 2, 4, 1] 183s -0.127 -0.729 183s -0.328 0.712 183s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.976 -0.866 183s 0.675 0.501 183s 0.422 -0.943 183s -0.099 0.144 183s 183s x1: [2, 4, 4, 1] 183s 0.548 -0.480 183s -0.158 0.295 183s -0.057 -0.074 183s -0.541 -0.921 183s 183s y: [4, 4, 4, 1] 183s 0.950 0.130 0.684 -0.123 183s -0.409 0.041 -0.345 0.058 183s 0.009 -0.076 0.046 -0.005 183s 0.270 -0.826 0.641 -0.079 183s test-mul-mat0: iter:279/500 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.643 0.931 -0.505 183s 0.664 0.196 -0.756 183s -0.706 -0.394 -0.176 183s -0.118 0.150 0.785 183s 183s x1: [3, 4, 1, 1] 183s 0.989 0.477 0.914 183s -0.429 -0.674 0.491 183s 0.378 -0.120 -0.660 183s 0.234 0.497 -0.870 183s 183s y: [4, 4, 1, 1] 183s 0.619 0.059 -1.047 0.672 183s -1.152 -0.789 0.482 0.335 183s 0.464 0.727 -0.104 -0.581 183s 1.053 0.911 -0.208 -0.636 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.772 -0.065 0.611 183s 0.269 0.065 0.059 183s 183s x1: [3, 4, 1, 1] 183s 0.949 0.669 -0.580 183s -0.562 0.090 0.190 183s -0.573 -0.433 0.104 183s -0.002 -0.108 -0.405 183s 183s y: [2, 4, 1, 1] 183s 0.335 0.265 183s -0.324 -0.134 183s -0.351 -0.176 183s -0.242 -0.032 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.202 -0.378 0.752 183s 183s x1: [3, 2, 1, 1] 183s -0.088 -0.004 0.554 183s -0.424 -0.808 0.798 183s 183s y: [1, 2, 1, 1] 183s 0.436 183s 0.991 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.173 -0.983 0.121 183s -0.861 -0.434 0.564 183s -0.293 0.274 -0.834 183s 183s x1: [3, 1, 1, 1] 183s 0.948 -0.593 0.702 183s 183s y: [3, 1, 1, 1] 183s 0.504 -0.163 -1.025 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.836 -0.096 0.566 183s -0.304 0.962 0.702 183s 183s x1: [3, 3, 1, 1] 183s 0.716 0.940 0.938 183s -0.349 0.550 0.208 183s 0.716 -0.390 -0.880 183s 183s y: [2, 3, 1, 1] 183s -0.157 1.346 183s 0.357 0.781 183s -1.060 -1.212 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.160 0.726 183s 0.974 0.247 183s 0.835 -0.045 183s 183s x1: [2, 3, 1, 1] 183s -0.416 0.319 183s 0.650 -0.009 183s 0.020 0.267 183s 183s y: [3, 3, 1, 1] 183s 0.165 -0.326 -0.361 183s 0.097 0.631 0.543 183s 0.197 0.086 0.005 183s test-mul-mat0: iter:280/500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.220 183s -0.441 183s 0.039 183s 0.389 183s 183s x1: [1, 1, 1, 1] 183s 0.352 183s 183s y: [4, 1, 1, 1] 183s -0.077 -0.155 0.014 0.137 183s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.740 183s -0.992 183s 0.090 183s 0.280 183s 183s x1: [1, 4, 2, 1] 183s -0.321 183s 0.751 183s 0.553 183s 0.375 183s 183s y: [4, 4, 2, 1] 183s 0.238 0.318 -0.029 -0.090 183s -0.556 -0.745 0.068 0.210 183s -0.410 -0.549 0.050 0.155 183s -0.278 -0.372 0.034 0.105 183s testing: mul_mat, [3, 4, 2, 4] = [1, 3, 2, 4] * [1, 4, 2, 4] 183s x0: [1, 3, 2, 4] 183s -0.477 183s -0.539 183s -0.582 183s 183s x1: [1, 4, 2, 4] 183s -0.258 183s -0.834 183s -0.503 183s 0.698 183s 183s y: [3, 4, 2, 4] 183s 0.123 0.139 0.150 183s 0.398 0.450 0.486 183s 0.240 0.271 0.293 183s -0.333 -0.376 -0.406 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.329 183s 183s x1: [1, 3, 1, 1] 183s -0.859 183s -0.356 183s -0.092 183s 183s y: [1, 3, 1, 1] 183s 0.283 183s 0.117 183s 0.030 183s testing: mul_mat, [3, 1, 2, 1] = [1, 3, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.867 183s -0.683 183s 0.627 183s 183s x1: [1, 1, 2, 1] 183s 0.672 183s 183s y: [3, 1, 2, 1] 183s -0.583 -0.459 0.421 183s testing: mul_mat, [3, 1, 2, 4] = [3, 3, 2, 4] * [3, 1, 2, 4] 183s x0: [3, 3, 2, 4] 183s -0.637 0.326 -0.189 183s -0.469 0.951 0.937 183s 0.707 0.029 0.866 183s 183s x1: [3, 1, 2, 4] 183s 0.485 0.987 -0.163 183s 183s y: [3, 1, 2, 4] 183s 0.043 0.558 0.231 183s test-mul-mat0: iter:281/500 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.238 0.820 0.485 0.525 183s 0.032 -0.750 -0.649 -0.909 183s -0.241 0.946 0.208 0.093 183s -0.328 -0.084 -0.041 0.857 183s 183s x1: [4, 1, 1, 1] 183s -0.390 -0.691 -0.067 -0.862 183s 183s y: [4, 1, 1, 1] 183s -1.144 1.333 -0.653 -0.550 183s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 3, 3, 1] 183s 0.993 -0.547 0.464 0.448 183s -0.620 0.536 -0.152 0.514 183s -0.300 -0.079 0.566 -0.176 183s 183s x1: [4, 4, 3, 1] 183s 0.166 0.385 -0.690 0.896 183s -0.762 0.581 -0.551 -0.764 183s 0.691 0.408 -0.154 -1.000 183s -0.660 -0.016 0.577 0.579 183s 183s y: [3, 4, 3, 1] 183s 0.035 0.668 -0.628 183s -1.673 0.476 0.005 183s -0.056 -0.701 -0.150 183s -0.119 0.611 0.423 183s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.359 -0.717 0.909 -0.327 183s 183s x1: [4, 3, 3, 1] 183s -0.824 -0.085 0.665 -0.124 183s 0.836 0.230 0.700 -0.314 183s -0.987 -0.968 0.777 0.910 183s 183s y: [1, 3, 3, 1] 183s 1.001 183s 0.273 183s 1.456 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.326 -0.770 0.703 -0.701 183s -0.147 0.453 -0.882 -0.297 183s 183s x1: [4, 1, 1, 1] 183s -0.986 -0.091 -0.485 0.774 183s 183s y: [2, 1, 1, 1] 183s -0.492 0.301 183s testing: mul_mat, [2, 4, 3, 1] = [2, 2, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.269 -0.480 183s 0.800 -0.151 183s 183s x1: [2, 4, 3, 1] 183s -0.817 -0.940 183s 0.368 0.123 183s 0.657 0.263 183s 0.764 0.941 183s 183s y: [2, 4, 3, 1] 183s 0.672 -0.512 183s -0.158 0.276 183s -0.304 0.486 183s -0.658 0.470 183s testing: mul_mat, [3, 2, 3, 1] = [2, 3, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.361 0.140 183s 0.024 0.371 183s 0.905 -0.279 183s 183s x1: [2, 2, 3, 1] 183s 0.919 0.758 183s 0.026 -0.643 183s 183s y: [3, 2, 3, 1] 183s -0.226 0.303 0.621 183s -0.099 -0.238 0.203 183s test-mul-mat0: iter:282/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.322 -0.240 -0.666 0.786 183s -0.489 0.765 -0.160 -0.583 183s 0.014 0.750 -0.757 -0.980 183s 0.685 -0.574 -0.216 0.638 183s 183s x1: [4, 4, 1, 1] 183s -0.235 0.952 0.760 0.273 183s -0.854 0.513 -0.608 0.497 183s 0.455 0.031 -0.479 0.360 183s -0.829 0.892 -0.919 -0.959 183s 183s y: [4, 4, 1, 1] 183s -0.596 0.564 -0.132 -0.698 183s 0.397 0.618 0.346 -0.432 183s 0.741 -0.332 0.039 0.627 183s -0.623 1.793 2.293 -1.494 183s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.354 0.217 -0.668 0.737 183s 0.877 -0.187 0.342 -0.812 183s 183s x1: [4, 4, 4, 1] 183s -0.814 -0.943 0.783 0.700 183s -0.551 0.280 0.155 0.480 183s 0.801 -0.486 0.651 0.693 183s -0.405 0.692 -0.244 0.917 183s 183s y: [2, 4, 4, 1] 183s 0.075 -0.837 183s 0.506 -0.873 183s -0.313 0.454 183s 1.132 -1.312 183s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 3, 4, 1] 183s -0.907 0.992 -0.641 -0.708 183s -0.191 0.532 -0.967 0.817 183s -0.379 0.433 0.473 0.644 183s 183s x1: [4, 2, 4, 1] 183s -0.249 0.374 -0.143 -0.372 183s -0.813 -0.801 -0.183 0.173 183s 183s y: [3, 2, 4, 1] 183s 0.951 0.082 -0.050 183s -0.062 0.048 -0.014 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.708 0.429 -0.317 -0.442 183s -0.643 0.397 -0.530 -0.024 183s -0.024 0.278 0.398 -0.009 183s 183s x1: [4, 3, 1, 1] 183s 0.866 -0.870 -0.769 -0.685 183s -0.235 0.534 0.841 0.156 183s -0.047 0.980 -0.655 0.608 183s 183s y: [3, 3, 1, 1] 183s 0.785 -0.478 -0.563 183s -0.272 -0.087 0.488 183s 0.326 0.751 0.008 183s testing: mul_mat, [4, 4, 4, 1] = [3, 4, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.499 -0.621 -0.875 183s 0.189 0.226 0.054 183s -0.203 -0.155 -0.316 183s -0.966 -0.748 0.999 183s 183s x1: [3, 4, 4, 1] 183s -0.923 0.796 0.778 183s 0.073 0.657 -0.651 183s -0.061 0.787 -0.420 183s 0.254 -0.448 -0.886 183s 183s y: [4, 4, 4, 1] 183s -0.715 0.047 -0.182 1.074 183s 0.125 0.127 0.089 -1.213 183s -0.091 0.144 0.023 -0.949 183s 0.927 -0.101 0.298 -0.795 183s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 4, 4, 1] 183s -0.433 -0.175 0.569 0.912 183s 0.625 0.401 0.374 0.340 183s 0.801 0.889 0.349 0.860 183s -0.316 0.510 0.536 -0.205 183s 183s x1: [4, 3, 4, 1] 183s -0.848 -0.643 -0.654 -0.734 183s -0.178 0.413 -0.953 0.308 183s 0.349 0.203 -0.118 0.230 183s 183s y: [4, 3, 4, 1] 183s -0.562 -1.283 -2.111 -0.260 183s -0.257 -0.198 0.156 -0.307 183s -0.044 0.334 0.617 -0.118 183s test-mul-mat0: iter:283/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.146 0.274 -0.020 -0.332 183s 0.313 0.196 0.181 0.703 183s 0.511 0.248 -0.648 -0.560 183s 183s x1: [4, 3, 1, 1] 183s 0.686 0.462 -0.456 0.754 183s 0.031 0.464 0.735 0.941 183s -0.757 -0.490 -0.553 -0.971 183s 183s y: [3, 3, 1, 1] 183s -0.215 0.753 0.338 183s -0.205 0.895 -0.872 183s 0.310 -1.116 0.394 183s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 4, 4, 1] 183s 0.006 -0.473 -0.311 -0.525 183s 0.059 -0.501 0.139 -0.228 183s -0.989 0.314 0.734 0.876 183s 0.929 -0.870 -0.375 -0.673 183s 183s x1: [4, 3, 4, 1] 183s 0.435 -0.115 0.783 -0.573 183s -0.872 -0.057 -0.887 0.590 183s 0.486 0.867 -0.378 -0.050 183s 183s y: [4, 3, 4, 1] 183s 0.114 0.323 -0.393 0.596 183s -0.012 -0.280 0.710 -0.825 183s -0.263 -0.447 -0.530 -0.127 183s testing: mul_mat, [3, 4, 4, 3] = [4, 3, 4, 3] * [4, 4, 4, 3] 183s x0: [4, 3, 4, 3] 183s -0.989 -0.880 -0.796 -0.862 183s 0.779 -0.126 0.909 -0.603 183s 0.978 -0.226 -0.402 0.067 183s 183s x1: [4, 4, 4, 3] 183s 0.042 0.573 -0.390 0.664 183s 0.810 -0.658 -0.703 0.361 183s 0.807 -0.194 -0.159 0.403 183s 0.940 -0.353 0.953 -0.424 183s 183s y: [3, 4, 4, 3] 183s -0.807 -0.795 0.112 183s 0.027 -0.143 1.247 183s -0.848 0.265 0.924 183s -1.012 1.898 0.587 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.030 -0.922 -0.367 -0.929 183s -0.114 -0.211 -0.806 0.268 183s -0.696 0.400 0.354 0.299 183s 183s x1: [4, 3, 1, 1] 183s 0.092 0.260 -0.503 -0.537 183s 0.412 0.480 0.450 0.401 183s 0.641 0.524 0.815 -0.294 183s 183s y: [3, 3, 1, 1] 183s 0.446 0.196 -0.299 183s -0.968 -0.404 0.185 183s -0.489 -0.920 -0.036 183s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.142 -0.936 0.889 183s -0.239 -0.996 0.571 183s 183s x1: [3, 4, 4, 1] 183s -0.076 -0.807 -0.808 183s 0.416 -0.869 0.507 183s -0.492 0.391 -0.995 183s -0.029 -0.197 0.485 183s 183s y: [2, 4, 4, 1] 183s 0.048 0.361 183s 1.205 1.056 183s -1.181 -0.840 183s 0.619 0.480 183s testing: mul_mat, [4, 3, 4, 3] = [2, 4, 4, 3] * [2, 3, 4, 3] 183s x0: [2, 4, 4, 3] 183s -0.034 -0.021 183s 0.872 0.402 183s -0.062 -0.136 183s 0.890 0.327 183s 183s x1: [2, 3, 4, 3] 183s 0.231 -0.189 183s -0.443 -0.255 183s -0.172 -0.731 183s 183s y: [4, 3, 4, 3] 183s -0.004 0.125 0.011 0.144 183s 0.020 -0.488 0.062 -0.477 183s 0.021 -0.444 0.110 -0.392 183s test-mul-mat0: iter:284/500 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.513 -0.088 0.905 -0.930 183s -0.355 0.046 0.909 -0.601 183s 0.623 -0.768 0.896 0.333 183s 183s x1: [4, 4, 1, 1] 183s 0.866 0.300 -0.136 0.364 183s -0.627 -0.301 -0.153 -0.508 183s 0.161 -0.996 0.164 -0.650 183s 0.151 -0.072 0.973 0.369 183s 183s y: [3, 4, 1, 1] 183s -0.044 -0.636 0.308 183s 0.039 0.374 -0.466 183s 0.923 0.437 0.796 183s 0.621 0.606 1.144 183s testing: mul_mat, [1, 3, 2, 1] = [4, 1, 2, 1] * [4, 3, 2, 1] 183s x0: [4, 1, 2, 1] 183s -0.741 0.890 0.952 -0.509 183s 183s x1: [4, 3, 2, 1] 183s 0.641 0.720 0.199 -0.059 183s -0.416 -0.437 0.314 0.283 183s 0.409 0.806 -0.556 0.413 183s 183s y: [1, 3, 2, 1] 183s 0.386 183s 0.074 183s -0.326 183s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 3, 2, 1] 183s 0.869 0.161 -0.389 -0.566 183s 0.059 0.378 -0.633 -0.563 183s 0.657 0.213 0.720 -0.993 183s 183s x1: [4, 1, 2, 1] 183s -0.516 0.072 -0.910 0.047 183s 183s y: [3, 1, 2, 1] 183s -0.109 0.546 -1.026 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.945 0.195 -0.187 0.158 183s 183s x1: [4, 3, 1, 1] 183s 0.977 -0.182 -0.654 0.350 183s 0.274 -0.463 0.167 0.144 183s 0.699 0.778 0.577 -0.243 183s 183s y: [1, 3, 1, 1] 183s 1.065 183s 0.161 183s 0.665 183s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.746 183s 183s x1: [1, 4, 2, 1] 183s -0.085 183s -0.180 183s 0.862 183s -0.035 183s 183s y: [1, 4, 2, 1] 183s 0.063 183s 0.135 183s -0.643 183s 0.026 183s testing: mul_mat, [4, 1, 2, 1] = [1, 4, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.299 183s -0.157 183s 0.034 183s -0.948 183s 183s x1: [1, 1, 2, 1] 183s -0.276 183s 183s y: [4, 1, 2, 1] 183s 0.082 0.043 -0.009 0.262 183s test-mul-mat0: iter:285/500 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.863 0.785 0.803 0.102 183s -0.130 -0.741 -0.931 -0.289 183s 0.096 -0.678 0.440 0.821 183s 183s x1: [4, 1, 1, 1] 183s -0.848 -0.216 0.223 -0.035 183s 183s y: [3, 1, 1, 1] 183s 0.737 0.073 0.134 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.967 -0.059 -0.807 0.255 183s 0.906 -0.865 0.393 0.691 183s 0.938 -0.505 -0.439 -0.803 183s 183s x1: [4, 3, 1, 1] 183s -0.653 -0.871 -0.478 0.191 183s 0.163 -0.426 0.309 -0.265 183s 0.792 -0.430 -0.995 -0.211 183s 183s y: [3, 3, 1, 1] 183s 1.117 0.107 -0.115 183s -0.449 0.454 0.446 183s 0.008 0.553 1.567 183s testing: mul_mat, [3, 3, 1, 2] = [4, 3, 1, 2] * [4, 3, 1, 2] 183s x0: [4, 3, 1, 2] 183s 0.424 0.329 0.978 0.918 183s 0.890 -0.825 -0.518 0.161 183s 0.468 0.367 0.873 -0.418 183s 183s x1: [4, 3, 1, 2] 183s -0.437 0.272 0.293 -0.114 183s -0.288 0.114 0.233 -0.160 183s 0.636 -0.576 -0.997 -0.790 183s 183s y: [3, 3, 1, 2] 183s 0.086 -0.783 0.199 183s -0.004 -0.497 0.177 183s -1.621 1.431 -0.454 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.174 -0.154 0.137 -0.425 183s -0.027 -0.081 0.818 0.729 183s -0.663 -0.206 -0.725 0.994 183s -0.806 0.559 0.564 0.634 183s 183s x1: [4, 3, 1, 1] 183s -0.228 -0.028 -0.047 -0.323 183s 0.003 -0.408 -0.283 -0.574 183s 0.921 -0.305 -0.655 0.811 183s 183s y: [4, 3, 1, 1] 183s 0.175 -0.265 -0.130 -0.063 183s 0.268 -0.617 -0.283 -0.754 183s -0.547 0.054 0.734 -0.769 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.700 -0.689 -0.516 0.756 183s 0.690 -0.381 0.870 -0.697 183s 183s x1: [4, 4, 1, 1] 183s 0.925 0.215 -0.594 -0.103 183s -0.831 0.083 0.900 -0.239 183s 0.800 -0.674 -0.318 -0.506 183s -0.330 -0.507 0.364 0.497 183s 183s y: [2, 4, 1, 1] 183s 0.728 0.111 183s -1.284 0.344 183s 0.806 0.885 183s 0.306 -0.065 183s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 183s x0: [2, 4, 1, 2] 183s -0.706 0.088 183s 0.592 -0.900 183s -0.365 0.358 183s 0.657 -0.221 183s 183s x1: [2, 4, 1, 2] 183s -0.855 0.320 183s -0.123 0.874 183s 0.315 -0.489 183s 0.799 -0.470 183s 183s y: [4, 4, 1, 2] 183s 0.632 -0.794 0.427 -0.633 183s 0.164 -0.859 0.358 -0.274 183s -0.265 0.626 -0.290 0.315 183s -0.606 0.896 -0.460 0.629 183s test-mul-mat0: iter:286/500 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.597 183s 183s x1: [1, 3, 1, 1] 183s -0.900 183s -0.102 183s 0.871 183s 183s y: [1, 3, 1, 1] 183s -0.537 183s -0.061 183s 0.520 183s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.198 183s 0.964 183s 183s x1: [1, 1, 4, 1] 183s 0.871 183s 183s y: [2, 1, 4, 1] 183s 0.173 0.840 183s testing: mul_mat, [1, 2, 4, 2] = [1, 1, 4, 2] * [1, 2, 4, 2] 183s x0: [1, 1, 4, 2] 183s -0.394 183s 183s x1: [1, 2, 4, 2] 183s 0.773 183s -0.655 183s 183s y: [1, 2, 4, 2] 183s -0.305 183s 0.258 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.002 183s 183s x1: [1, 1, 1, 1] 183s 0.864 183s 183s y: [1, 1, 1, 1] 183s -0.002 183s testing: mul_mat, [3, 1, 4, 1] = [1, 3, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 3, 4, 1] 183s -0.066 183s 0.668 183s 0.599 183s 183s x1: [1, 1, 4, 1] 183s -0.141 183s 183s y: [3, 1, 4, 1] 183s 0.009 -0.094 -0.085 183s testing: mul_mat, [2, 1, 4, 2] = [3, 2, 4, 2] * [3, 1, 4, 2] 183s x0: [3, 2, 4, 2] 183s -0.386 -0.835 0.236 183s 0.923 -0.909 0.626 183s 183s x1: [3, 1, 4, 2] 183s -0.514 0.184 0.713 183s 183s y: [2, 1, 4, 2] 183s 0.213 -0.195 183s test-mul-mat0: iter:287/500 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.891 -0.169 -0.562 183s -0.519 -0.169 0.077 183s 0.429 0.250 0.055 183s -0.953 0.015 0.728 183s 183s x1: [3, 3, 1, 1] 183s -0.298 0.744 -0.007 183s 0.887 -0.837 0.226 183s -0.423 0.829 -0.933 183s 183s y: [4, 3, 1, 1] 183s 0.144 0.028 0.057 0.290 183s -0.775 -0.301 0.184 -0.692 183s 0.760 0.007 -0.025 -0.264 183s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.701 0.839 -0.309 183s -0.083 -0.585 -0.981 183s -0.388 -0.580 -0.403 183s 183s x1: [3, 4, 4, 1] 183s 0.088 -0.539 -0.589 183s -0.834 0.302 -0.717 183s 0.589 -0.750 0.381 183s -0.709 0.994 -0.626 183s 183s y: [3, 4, 4, 1] 183s -0.333 0.885 0.515 183s 1.060 0.595 0.437 183s -1.160 0.016 0.053 183s 1.525 0.092 -0.049 183s testing: mul_mat, [4, 3, 4, 3] = [3, 4, 4, 3] * [3, 3, 4, 3] 183s x0: [3, 4, 4, 3] 183s 0.807 0.259 0.804 183s -0.953 0.544 0.474 183s -0.139 0.825 0.521 183s -0.745 -0.445 -0.579 183s 183s x1: [3, 3, 4, 3] 183s -0.186 -0.684 -0.018 183s 0.411 0.678 0.455 183s 0.601 0.519 0.068 183s 183s y: [4, 3, 4, 3] 183s -0.342 -0.203 -0.548 0.454 183s 0.873 0.193 0.740 -0.872 183s 0.675 -0.258 0.380 -0.719 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.562 -0.915 -0.587 183s -0.252 -0.915 0.084 183s 183s x1: [3, 4, 1, 1] 183s 0.326 0.604 -0.644 183s 0.554 -0.428 -0.728 183s -0.260 -0.288 0.390 183s -0.015 0.506 0.108 183s 183s y: [2, 4, 1, 1] 183s 0.009 -0.689 183s 1.130 0.191 183s -0.111 0.362 183s -0.535 -0.451 183s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.723 0.017 183s -0.396 -0.340 183s 0.576 -0.795 183s 0.807 -0.736 183s 183s x1: [2, 3, 4, 1] 183s -0.602 0.246 183s 0.811 -0.635 183s -0.542 -0.496 183s 183s y: [4, 3, 4, 1] 183s -0.431 0.155 -0.543 -0.668 183s 0.575 -0.105 0.972 1.123 183s -0.400 0.383 0.082 -0.072 183s testing: mul_mat, [1, 2, 4, 3] = [4, 1, 4, 3] * [4, 2, 4, 3] 183s x0: [4, 1, 4, 3] 183s 0.983 -0.582 0.198 0.785 183s 183s x1: [4, 2, 4, 3] 183s 0.968 -0.341 -0.730 -0.015 183s 0.319 -0.525 0.249 0.791 183s 183s y: [1, 2, 4, 3] 183s 0.995 183s 1.289 183s test-mul-mat0: iter:288/500 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png0.748 183s -0.609 183s -0.874 183s 183s x1: [1, 2, 1, 1] 183s -0.301 183s 0.461 183s 183s y: [3, 2, 1, 1] 183s -0.225 0.184 0.263 183s 0.345 -0.281 -0.403 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.973 183s 183s x1: [1, 3, 1, 1] 183s -0.455 183s 0.190 183s 0.140 183s 183s y: [1, 3, 1, 1] 183s -0.443 183s 0.185 183s 0.137 183s testing: mul_mat, [1, 1, 1, 2] = [1, 1, 1, 2] * [1, 1, 1, 2] 183s x0: [1, 1, 1, 2] 183s 0.160 183s 183s x1: [1, 1, 1, 2] 183s -0.427 183s 183s y: [1, 1, 1, 2] 183s -0.069 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.933 183s 183s x1: [1, 1, 1, 1] 183s 0.148 183s 183s y: [1, 1, 1, 1] 183s 0.138 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.626 183s 183s x1: [1, 1, 1, 1] 183s -0.143 183s 183s y: [1, 1, 1, 1] 183s -0.090 183s testing: mul_mat, [4, 1, 1, 2] = [1, 4, 1, 2] * [1, 1, 1, 2] 183s x0: [1, 4, 1, 2] 183s 0.288 183s 0.045 183s -0.183 183s -0.424 183s 183s x1: [1, 1, 1, 2] 183s -0.793 183s 183s y: [4, 1, 1, 2] 183s -0.228 -0.036 0.145 0.336 183s test-mul-mat0: iter:289/500 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.736 0.802 -0.213 0.425 183s -0.553 0.935 0.191 -0.620 183s -0.208 0.356 -0.993 -0.001 183s 0.478 -0.342 -0.713 -0.477 183s 183s x1: [4, 1, 1, 1] 183s 0.567 -0.727 0.692 0.428 183s 183s y: [4, 1, 1, 1] 183s -0.966 -1.127 -1.064 -0.178 183s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 3, 3, 1] 183s 0.470 -0.747 -0.723 0.216 183s 0.038 0.838 -0.683 0.828 183s -0.601 -0.103 0.277 0.246 183s 183s x1: [4, 4, 3, 1] 183s 0.475 -0.137 0.753 -0.802 183s -0.862 -0.555 -0.377 0.160 183s 0.527 0.174 -0.272 0.800 183s -0.134 -0.844 -0.953 0.130 183s 183s y: [3, 4, 3, 1] 183s -0.392 -1.275 -0.260 183s 0.317 -0.108 0.511 183s 0.487 1.015 -0.214 183s 1.285 0.046 -0.064 183s testing: mul_mat, [3, 3, 3, 1] = [4, 3, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.921 0.114 0.548 -0.414 183s -0.447 -0.239 0.900 -0.855 183s 0.043 0.028 -0.282 0.310 183s 183s x1: [4, 3, 3, 1] 183s 0.905 0.689 -0.587 -0.696 183s -0.414 0.690 0.550 -0.415 183s 0.891 -0.493 0.439 0.214 183s 183s y: [3, 3, 3, 1] 183s -0.788 -0.502 0.008 183s 0.933 0.870 -0.282 183s -0.725 -0.069 -0.033 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.082 0.303 0.078 -0.182 183s -0.914 -0.979 -0.817 -0.127 183s -0.566 0.642 -0.207 -0.528 183s 183s x1: [4, 3, 1, 1] 183s 0.705 0.255 0.491 -0.252 183s -0.716 -0.791 -0.942 0.993 183s 0.108 0.406 0.828 -0.530 183s 183s y: [3, 3, 1, 1] 183s 0.219 -1.265 -0.204 183s -0.553 2.072 -0.431 183s 0.293 -1.106 0.308 183s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.125 0.879 -0.514 183s 0.948 -0.513 -0.808 183s 183s x1: [3, 4, 3, 1] 183s -0.786 0.642 0.266 183s 0.158 -0.002 -0.388 183s -0.136 -0.746 -0.896 183s 0.612 -0.463 -0.687 183s 183s y: [2, 4, 3, 1] 183s 0.527 -1.290 183s 0.178 0.464 183s -0.179 0.978 183s -0.130 1.373 183s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.600 0.632 183s -0.442 0.932 183s 0.358 -0.137 183s 183s x1: [2, 3, 3, 1] 183s 0.411 0.379 183s 0.662 -0.164 183s -0.983 -0.284 183s 183s y: [3, 3, 3, 1] 183s -0.007 0.171 0.096 183s -0.501 -0.445 0.260 183s 0.410 0.169 -0.313 183s test-mul-mat0: iter:290/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.250 183s 0.209 183s 183s x1: [1, 1, 1, 1] 183s 0.452 183s 183s y: [2, 1, 1, 1] 183s -0.113 0.095 183s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 4, 4, 1] 183s 0.811 183s -0.731 183s -0.095 183s -0.443 183s 183s x1: [1, 2, 4, 1] 183s 0.779 183s -0.697 183s 183s y: [4, 2, 4, 1] 183s 0.631 -0.569 -0.074 -0.345 183s -0.565 0.509 0.066 0.309 183s testing: mul_mat, [3, 4, 4, 3] = [1, 3, 4, 3] * [1, 4, 4, 3] 183s x0: [1, 3, 4, 3] 183s 0.872 183s 0.992 183s 0.447 183s 183s x1: [1, 4, 4, 3] 183s 0.748 183s 0.418 183s -0.733 183s -0.529 183s 183s y: [3, 4, 4, 3] 183s 0.652 0.742 0.334 183s 0.364 0.414 0.187 183s -0.639 -0.727 -0.327 183s -0.461 -0.524 -0.236 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.720 183s -0.770 183s 0.495 183s 183s x1: [1, 3, 1, 1] 183s 0.827 183s -0.516 183s -0.498 183s 183s y: [3, 3, 1, 1] 183s 0.595 -0.637 0.409 183s -0.372 0.397 -0.255 183s -0.358 0.383 -0.246 183s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 4, 4, 1] 183s 0.094 -0.634 0.305 183s 0.513 0.915 -0.737 183s 0.972 0.549 0.295 183s 0.386 0.210 0.131 183s 183s x1: [3, 1, 4, 1] 183s 0.237 0.014 -0.516 183s 183s y: [4, 1, 4, 1] 183s -0.144 0.515 0.086 0.027 183s testing: mul_mat, [3, 3, 4, 3] = [4, 3, 4, 3] * [4, 3, 4, 3] 183s x0: [4, 3, 4, 3] 183s -0.371 -0.522 -0.234 -0.063 183s 0.700 0.874 -0.654 -0.728 183s 0.569 0.441 -0.751 -0.549 183s 183s x1: [4, 3, 4, 3] 183s -0.070 0.853 0.335 -0.545 183s -0.142 0.703 0.531 0.250 183s 0.490 0.772 -0.931 -0.865 183s 183s y: [3, 3, 4, 3] 183s -0.464 0.874 0.385 183s -0.454 -0.015 -0.307 183s -0.313 2.256 1.793 183s test-mul-mat0: iter:291/500 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.941 183s 0.722 183s 0.535 183s 183s x1: [1, 4, 1, 1] 183s -0.504 183s 0.982 183s -0.843 183s 0.058 183s 183s y: [3, 4, 1, 1] 183s 0.475 -0.364 -0.270 183s -0.924 0.709 0.525 183s 0.793 -0.608 -0.451 183s -0.055 0.042 0.031 183s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.166 183s -0.573 183s 183s x1: [1, 3, 1, 1] 183s -0.136 183s 0.523 183s 0.169 183s 183s y: [2, 3, 1, 1] 183s -0.023 0.078 183s 0.087 -0.299 183s 0.028 -0.097 183s testing: mul_mat, [4, 2, 1, 3] = [1, 4, 1, 3] * [1, 2, 1, 3] 183s x0: [1, 4, 1, 3] 183s -0.115 183s 0.644 183s -0.660 183s -0.669 183s 183s x1: [1, 2, 1, 3] 183s 0.206 183s 0.275 183s 183s y: [4, 2, 1, 3] 183s -0.024 0.132 -0.136 -0.138 183s -0.032 0.177 -0.182 -0.184 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.210 183s 0.638 183s -0.235 183s -0.621 183s 183s x1: [1, 4, 1, 1] 183s -0.284 183s 0.520 183s 0.675 183s -0.226 183s 183s y: [4, 4, 1, 1] 183s -0.060 -0.181 0.067 0.176 183s 0.109 0.332 -0.122 -0.323 183s 0.142 0.431 -0.159 -0.419 183s -0.047 -0.144 0.053 0.140 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.898 0.163 0.264 0.948 183s 0.887 -0.067 0.982 -0.117 183s -0.952 0.241 0.073 0.438 183s 0.597 0.503 -0.120 -0.590 183s 183s x1: [4, 1, 1, 1] 183s 0.389 0.931 -0.194 -0.405 183s 183s y: [4, 1, 1, 1] 183s -0.632 0.140 -0.337 0.963 183s testing: mul_mat, [4, 4, 1, 3] = [4, 4, 1, 3] * [4, 4, 1, 3] 183s x0: [4, 4, 1, 3] 183s 0.613 -0.068 -0.308 -0.398 183s 0.828 -0.506 -0.421 0.901 183s 0.177 -0.035 -0.832 -0.883 183s 0.298 0.108 -0.672 -0.445 183s 183s x1: [4, 4, 1, 3] 183s -0.326 -0.846 0.930 -0.651 183s -0.072 0.172 0.559 -0.434 183s 0.936 0.938 0.955 0.867 183s -0.256 -0.450 0.073 -0.154 183s 183s y: [4, 4, 1, 3] 183s -0.169 -0.821 -0.227 -0.524 183s -0.055 -0.772 -0.101 -0.186 183s -0.129 0.679 -1.428 -0.647 183s -0.088 -0.154 0.046 -0.106 183s test-mul-mat0: iter:292/500 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.064 -0.932 183s -0.653 -0.560 183s -0.140 -0.344 183s 0.644 -0.172 183s 183s x1: [2, 2, 1, 1] 183s -0.271 -0.876 183s 0.914 -0.602 183s 183s y: [4, 2, 1, 1] 183s 0.799 0.667 0.339 -0.024 183s 0.619 -0.260 0.079 0.692 183s testing: mul_mat, [4, 4, 3, 1] = [2, 4, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 4, 3, 1] 183s -0.913 0.938 183s -0.912 -0.257 183s 0.582 -0.083 183s 0.006 0.362 183s 183s x1: [2, 4, 3, 1] 183s -0.737 0.780 183s 0.291 -0.165 183s 0.024 -0.554 183s 0.985 -0.564 183s 183s y: [4, 4, 3, 1] 183s 1.404 0.472 -0.494 0.278 183s -0.421 -0.223 0.183 -0.058 183s -0.541 0.121 0.060 -0.200 183s -1.428 -0.753 0.620 -0.198 183s testing: mul_mat, [4, 4, 3, 4] = [2, 4, 3, 4] * [2, 4, 3, 4] 183s x0: [2, 4, 3, 4] 183s -0.141 0.187 183s 0.441 0.614 183s 0.766 0.822 183s 0.569 -0.554 183s 183s x1: [2, 4, 3, 4] 183s -0.597 0.670 183s 0.031 0.563 183s 0.411 -0.472 183s -0.208 0.498 183s 183s y: [4, 4, 3, 4] 183s 0.210 0.148 0.093 -0.711 183s 0.101 0.359 0.486 -0.295 183s -0.147 -0.109 -0.073 0.495 183s 0.123 0.214 0.249 -0.395 183s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.746 0.280 183s -0.656 0.051 183s -0.719 0.103 183s 183s x1: [2, 4, 1, 1] 183s 0.367 -0.899 183s -0.479 0.882 183s 0.712 -0.643 183s -0.293 0.749 183s 183s y: [3, 4, 1, 1] 183s 0.022 -0.286 -0.357 183s -0.111 0.359 0.436 183s 0.351 -0.500 -0.579 183s -0.009 0.230 0.288 183s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.334 0.636 0.768 183s 0.887 0.487 0.066 183s 0.402 0.787 0.031 183s -0.959 0.980 -0.129 183s 183s x1: [3, 2, 3, 1] 183s -0.109 0.948 -0.293 183s -0.537 0.551 0.095 183s 183s y: [4, 2, 3, 1] 183s 0.341 0.345 0.693 1.072 183s 0.244 -0.202 0.221 1.044 183s testing: mul_mat, [4, 3, 3, 4] = [4, 4, 3, 4] * [4, 3, 3, 4] 183s x0: [4, 4, 3, 4] 183s -0.056 -0.377 -0.830 -0.293 183s 0.243 0.742 0.676 -0.702 183s 0.059 0.340 0.741 -0.381 183s 0.968 0.021 -0.040 -0.481 183s 183s x1: [4, 3, 3, 4] 183s -0.187 0.218 -0.789 -0.420 183s -0.716 0.242 0.452 -0.759 183s -0.779 0.030 -0.339 -0.249 183s 183s y: [4, 3, 3, 4] 183s 0.706 -0.123 -0.362 0.057 183s -0.204 0.844 0.664 -0.340 183s 0.387 -0.222 -0.192 -0.620 183s test-mul-mat0: iter:293/500 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.898 -0.354 0.184 -0.986 183s 0.842 0.098 0.575 -0.074 183s 183s x1: [4, 1, 1, 1] 183s 0.265 0.012 -0.370 0.050 183s 183s y: [2, 1, 1, 1] 183s 0.116 0.008 183s testing: mul_mat, [4, 2, 2, 1] = [4, 4, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 4, 2, 1] 183s -0.453 0.088 0.243 0.177 183s -0.862 0.387 0.075 -0.216 183s -0.429 0.089 -0.375 0.668 183s -0.336 0.551 -0.331 -0.611 183s 183s x1: [4, 2, 2, 1] 183s 0.001 0.726 0.299 -0.213 183s 0.557 -0.703 0.690 -0.876 183s 183s y: [4, 2, 2, 1] 183s 0.098 0.349 -0.190 0.431 183s -0.301 -0.512 -1.145 -0.268 183s testing: mul_mat, [3, 4, 2, 4] = [4, 3, 2, 4] * [4, 4, 2, 4] 183s x0: [4, 3, 2, 4] 183s 0.097 0.836 0.193 -0.264 183s -0.794 -0.780 0.836 0.277 183s -0.919 -0.010 -0.580 0.903 183s 183s x1: [4, 4, 2, 4] 183s -0.152 -0.140 -0.509 -0.015 183s -0.753 0.566 0.769 -0.182 183s -0.345 -0.605 -0.514 0.319 183s 0.946 0.155 0.708 -0.204 183s 183s y: [3, 4, 2, 4] 183s -0.226 -0.200 0.423 183s 0.597 0.750 0.076 183s -0.723 0.405 0.909 183s 0.412 -0.337 -1.465 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.956 0.437 -0.016 0.840 183s -0.835 0.913 0.858 -0.176 183s -0.757 -0.034 -0.614 0.764 183s 183s x1: [4, 3, 1, 1] 183s -0.522 -0.771 0.524 0.495 183s -0.895 -0.437 -0.551 -0.863 183s -0.254 0.785 0.273 -0.357 183s 183s y: [3, 3, 1, 1] 183s -0.429 0.094 0.478 183s -1.763 0.028 0.371 183s -0.205 1.226 -0.274 183s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 4, 2, 1] 183s 0.430 -0.196 0.783 183s 0.981 -0.063 -0.753 183s 0.174 -0.445 0.337 183s 0.270 -0.972 -0.649 183s 183s x1: [3, 4, 2, 1] 183s 0.285 0.224 -0.155 183s 0.691 0.309 -0.431 183s -0.831 0.538 -0.908 183s 0.664 0.643 -0.344 183s 183s y: [4, 4, 2, 1] 183s -0.042 0.382 -0.102 -0.040 183s -0.101 0.983 -0.163 0.166 183s -1.173 -0.165 -0.690 -0.159 183s -0.109 0.871 -0.287 -0.223 183s testing: mul_mat, [2, 3, 2, 4] = [4, 2, 2, 4] * [4, 3, 2, 4] 183s x0: [4, 2, 2, 4] 183s 0.019 0.853 0.323 -0.422 183s 0.288 0.303 -0.621 -0.369 183s 183s x1: [4, 3, 2, 4] 183s 0.670 -0.638 -0.244 0.492 183s -0.192 -0.121 -0.720 -0.762 183s -0.141 0.454 0.508 0.664 183s 183s y: [2, 3, 2, 4] 183s -0.819 -0.030 183s -0.018 0.636 183s 0.268 -0.463 183s test-mul-mat0: iter:294/500 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.196 -0.091 -0.228 -0.306 183s -0.317 0.719 -0.382 0.605 183s 0.181 -0.207 -0.460 -0.527 183s -0.762 -0.224 0.448 -0.744 183s 183s x1: [4, 3, 1, 1] 183s 0.168 -0.886 -0.616 0.225 183s -0.725 -0.524 0.882 0.610 183s 0.851 -0.077 -0.711 0.095 183s 183s y: [4, 3, 1, 1] 183s 0.119 -0.319 0.379 -0.373 183s -0.198 -0.115 -0.750 0.612 183s -0.026 0.003 0.448 -1.020 183s testing: mul_mat, [4, 4, 2, 1] = [4, 4, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 4, 2, 1] 183s -0.249 -0.813 0.430 0.399 183s -0.789 -0.226 0.930 0.845 183s 0.970 0.384 0.768 -0.740 183s 0.865 -0.505 0.460 0.119 183s 183s x1: [4, 4, 2, 1] 183s 0.635 -0.515 0.424 0.749 183s -0.131 -0.351 -0.976 0.345 183s -0.469 0.634 0.196 0.454 183s 0.923 -0.710 -0.519 -0.273 183s 183s y: [4, 4, 2, 1] 183s 0.742 0.642 0.189 1.093 183s 0.037 -0.434 -1.266 -0.344 183s -0.133 0.792 -0.397 -0.582 183s 0.015 -1.282 0.426 0.887 183s testing: mul_mat, [3, 4, 2, 4] = [4, 3, 2, 4] * [4, 4, 2, 4] 183s x0: [4, 3, 2, 4] 183s 0.618 0.541 0.273 -0.822 183s -0.362 0.829 -0.409 0.872 183s -0.910 -0.494 0.761 0.788 183s 183s x1: [4, 4, 2, 4] 183s -0.718 -0.061 0.336 -0.507 183s 0.714 0.266 -0.662 0.684 183s -0.349 -0.895 0.944 -0.484 183s -0.400 0.404 0.635 -0.483 183s 183s y: [3, 4, 2, 4] 183s 0.032 -0.370 0.539 183s -0.158 0.829 -0.746 183s -0.044 -1.423 1.097 183s 0.542 -0.202 0.267 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.836 -0.091 0.239 -0.548 183s -0.411 -0.755 0.826 0.632 183s -0.430 0.539 0.846 0.262 183s 183s x1: [4, 3, 1, 1] 183s -0.956 0.017 0.623 -0.170 183s -0.471 0.121 0.034 -0.268 183s -0.195 0.952 -0.076 -0.092 183s 183s y: [3, 3, 1, 1] 183s -0.558 0.787 0.903 183s -0.250 -0.039 0.227 183s -0.217 -0.760 0.508 183s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 3, 2, 1] 183s -0.838 -0.386 0.686 183s -0.367 -0.736 -0.194 183s -0.053 -0.791 0.202 183s 183s x1: [3, 4, 2, 1] 183s 0.072 -0.458 -0.007 183s 0.244 -0.024 0.516 183s 0.288 0.993 0.139 183s -0.882 -0.478 -0.740 183s 183s y: [3, 4, 2, 1] 183s 0.112 0.312 0.357 183s 0.159 -0.172 0.110 183s -0.530 -0.863 -0.772 183s 0.417 0.819 0.276 183s testing: mul_mat, [1, 3, 2, 4] = [3, 1, 2, 4] * [3, 3, 2, 4] 183s x0: [3, 1, 2, 4] 183s 0.945 -0.811 0.013 183s 183s x1: [3, 3, 2, 4] 183s 0.486 0.551 0.182 183s -0.410 0.728 0.155 183s 0.581 0.742 0.717 183s 183s y: [1, 3, 2, 4] 183s 0.015 183s -0.976 183s -0.043 183s test-mul-mat0: iter:295/500 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.630 183s 183s x1: [1, 4, 1, 1] 183s -0.858 183s 0.718 183s -0.383 183s -0.913 183s 183s y: [1, 4, 1, 1] 183s -0.541 183s 0.452 183s -0.241 183s -0.575 183s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.793 183s -0.119 183s 183s x1: [1, 1, 3, 1] 183s 0.059 183s 183s y: [2, 1, 3, 1] 183s -0.046 -0.007 183s testing: mul_mat, [4, 2, 3, 3] = [1, 4, 3, 3] * [1, 2, 3, 3] 183s x0: [1, 4, 3, 3] 183s -0.767 183s -0.590 183s 0.132 183s 0.170 183s 183s x1: [1, 2, 3, 3] 183s -0.406 183s 0.203 183s 183s y: [4, 2, 3, 3] 183s 0.312 0.240 -0.054 -0.069 183s -0.156 -0.120 0.027 0.034 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.642 183s 0.099 183s -0.825 183s 183s x1: [1, 4, 1, 1] 183s 0.364 183s -0.813 183s 0.375 183s 0.552 183s 183s y: [3, 4, 1, 1] 183s 0.234 0.036 -0.301 183s -0.522 -0.080 0.670 183s 0.241 0.037 -0.309 183s 0.354 0.055 -0.455 183s testing: mul_mat, [1, 1, 3, 1] = [3, 1, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.820 -0.119 0.386 183s 183s x1: [3, 1, 3, 1] 183s 0.111 0.260 -0.330 183s 183s y: [1, 1, 3, 1] 183s -0.067 183s testing: mul_mat, [4, 3, 3, 3] = [1, 4, 3, 3] * [1, 3, 3, 3] 183s x0: [1, 4, 3, 3] 183s 0.413 183s -0.556 183s -0.748 183s -0.398 183s 183s x1: [1, 3, 3, 3] 183s 0.189 183s 0.801 183s 0.245 183s 183s y: [4, 3, 3, 3] 183s 0.078 -0.105 -0.141 -0.075 183s 0.331 -0.445 -0.599 -0.319 183s 0.101 -0.136 -0.183 -0.097 183s test-mul-mat0: iter:296/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.483 -0.463 183s -0.683 0.380 183s 0.552 -0.737 183s 183s x1: [2, 2, 1, 1] 183s -0.758 0.347 183s 0.748 0.027 183s 183s y: [3, 2, 1, 1] 183s -0.527 0.649 -0.674 183s 0.349 -0.500 0.393 183s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.890 0.171 183s -0.830 0.490 183s -0.277 -0.566 183s -0.466 -0.376 183s 183s x1: [2, 3, 4, 1] 183s 0.044 0.901 183s 0.982 -0.205 183s 0.886 0.293 183s 183s y: [4, 3, 4, 1] 183s 0.115 0.405 -0.522 -0.359 183s -0.909 -0.915 -0.156 -0.381 183s -0.739 -0.591 -0.412 -0.524 183s testing: mul_mat, [2, 4, 4, 4] = [2, 2, 4, 4] * [2, 4, 4, 4] 183s x0: [2, 2, 4, 4] 183s -0.386 0.511 183s 0.874 0.160 183s 183s x1: [2, 4, 4, 4] 183s -0.651 0.456 183s -0.594 0.071 183s 0.890 -0.060 183s 0.695 -0.694 183s 183s y: [2, 4, 4, 4] 183s 0.485 -0.496 183s 0.266 -0.507 183s -0.374 0.768 183s -0.623 0.496 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.746 0.722 183s -0.987 0.110 183s 183s x1: [2, 2, 1, 1] 183s -0.220 -0.358 183s 0.608 -0.111 183s 183s y: [2, 2, 1, 1] 183s -0.422 0.177 183s 0.374 -0.612 183s testing: mul_mat, [1, 2, 4, 1] = [2, 1, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 1, 4, 1] 183s -0.361 -0.905 183s 183s x1: [2, 2, 4, 1] 183s -0.089 0.593 183s -0.028 -0.540 183s 183s y: [1, 2, 4, 1] 183s -0.504 183s 0.499 183s testing: mul_mat, [4, 2, 4, 4] = [1, 4, 4, 4] * [1, 2, 4, 4] 183s x0: [1, 4, 4, 4] 183s 0.797 183s -0.400 183s -0.502 183s 0.828 183s 183s x1: [1, 2, 4, 4] 183s 0.686 183s 0.453 183s 183s y: [4, 2, 4, 4] 183s 0.547 -0.275 -0.345 0.568 183s 0.361 -0.181 -0.228 0.375 183s test-mul-mat0: iter:297/500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.086 183s 0.546 183s 0.747 183s -0.375 183s 183s x1: [1, 1, 1, 1] 183s -0.532 183s 183s y: [4, 1, 1, 1] 183s 0.046 -0.290 -0.397 0.199 183s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 3, 3, 1] 183s -0.522 183s 0.669 183s 0.102 183s 183s x1: [1, 4, 3, 1] 183s 0.885 183s 0.618 183s -0.226 183s -0.327 183s 183s y: [3, 4, 3, 1] 183s -0.462 0.592 0.090 183s -0.323 0.413 0.063 183s 0.118 -0.151 -0.023 183s 0.171 -0.219 -0.033 183s testing: mul_mat, [1, 3, 3, 2] = [1, 1, 3, 2] * [1, 3, 3, 2] 183s x0: [1, 1, 3, 2] 183s 0.416 183s 183s x1: [1, 3, 3, 2] 183s -0.257 183s 0.462 183s 0.854 183s 183s y: [1, 3, 3, 2] 183s -0.107 183s 0.192 183s 0.355 183s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.327 183s -0.217 183s 0.464 183s 183s x1: [1, 1, 1, 1] 183s -0.401 183s 183s y: [3, 1, 1, 1] 183s 0.131 0.087 -0.186 183s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.863 0.005 -0.831 183s 0.968 0.452 -0.963 183s 183s x1: [3, 1, 3, 1] 183s -0.176 0.526 -0.074 183s 183s y: [2, 1, 3, 1] 183s 0.217 0.139 183s testing: mul_mat, [1, 3, 3, 2] = [2, 1, 3, 2] * [2, 3, 3, 2] 183s x0: [2, 1, 3, 2] 183s -0.280 0.621 183s 183s x1: [2, 3, 3, 2] 183s -0.432 -0.428 183s 0.370 0.392 183s -0.902 -0.704 183s 183s y: [1, 3, 3, 2] 183s -0.145 183s 0.140 183s -0.185 183s test-mul-mat0: iter:298/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.031 -0.294 -0.975 183s 0.085 0.260 0.283 183s 0.773 0.183 0.933 183s -0.441 0.403 -0.347 183s 183s x1: [3, 1, 1, 1] 183s -0.460 -0.358 -0.941 183s 183s y: [4, 1, 1, 1] 183s 1.037 -0.399 -1.300 0.385 183s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.948 -0.744 -0.119 183s -0.984 0.215 -0.233 183s 0.367 0.948 0.836 183s 183s x1: [3, 4, 4, 1] 183s -0.820 0.487 -0.113 183s -0.870 0.395 0.474 183s -0.454 0.097 0.419 183s -0.121 -0.240 -0.242 183s 183s y: [3, 4, 4, 1] 183s -1.126 0.938 0.066 183s -1.175 0.830 0.452 183s -0.552 0.370 0.275 183s 0.093 0.123 -0.474 183s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 1, 4, 1] 183s -0.842 -0.217 -0.738 183s 183s x1: [3, 3, 4, 1] 183s -0.855 0.920 -0.446 183s 0.981 -0.428 0.793 183s -0.712 0.161 -0.495 183s 183s y: [1, 3, 4, 1] 183s 0.849 183s -1.318 183s 0.929 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.143 -0.656 0.194 183s -0.383 0.222 0.149 183s 183s x1: [3, 1, 1, 1] 183s -0.347 -0.011 -0.423 183s 183s y: [2, 1, 1, 1] 183s -0.025 0.068 183s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 2, 4, 1] 183s 0.990 0.542 183s -0.701 0.212 183s 183s x1: [2, 3, 4, 1] 183s -0.034 -0.677 183s -0.441 -0.145 183s 0.373 -0.477 183s 183s y: [2, 3, 4, 1] 183s -0.400 -0.120 183s -0.515 0.279 183s 0.111 -0.363 183s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.339 -0.970 183s -0.643 -0.464 183s -0.223 -0.407 183s 183s x1: [2, 2, 4, 1] 183s 0.962 -0.405 183s -0.761 0.097 183s 183s y: [3, 2, 4, 1] 183s 0.720 -0.430 -0.050 183s -0.353 0.444 0.130 183s test-mul-mat0: iter:299/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.270 -0.837 0.196 0.756 183s -0.923 -0.872 -0.340 -0.812 183s 0.394 0.585 0.602 0.261 183s 0.666 -0.998 -0.075 -0.108 183s 183s x1: [4, 4, 1, 1] 183s 0.665 0.226 0.684 0.004 183s 0.584 -0.539 0.034 -0.881 183s 0.053 -0.688 0.688 -0.165 183s 0.363 0.016 -0.198 -0.535 183s 183s y: [4, 4, 1, 1] 183s -0.231 -1.047 0.807 0.166 183s -0.365 0.635 -0.295 1.019 183s 0.572 0.450 -0.010 0.688 183s -0.554 0.153 -0.107 0.298 183s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.935 0.707 -0.842 0.919 183s 0.238 0.161 0.489 -0.629 183s 0.425 -0.275 0.506 -0.386 183s -0.147 -0.672 0.733 0.458 183s 183s x1: [4, 4, 3, 1] 183s 0.686 0.930 -0.524 -0.853 183s -0.037 -0.405 0.200 0.275 183s -0.717 -0.964 -0.362 0.299 183s -0.162 0.103 -0.714 0.568 183s 183s y: [4, 4, 3, 1] 183s 0.956 0.594 0.100 -1.500 183s -0.236 -0.149 0.091 0.550 183s -0.773 -0.691 -0.338 0.625 183s 1.045 -0.729 -0.678 -0.309 183s testing: mul_mat, [1, 4, 3, 4] = [4, 1, 3, 4] * [4, 4, 3, 4] 183s x0: [4, 1, 3, 4] 183s -0.789 0.030 0.427 -0.440 183s 183s x1: [4, 4, 3, 4] 183s 0.444 -0.642 -0.458 -0.384 183s -0.503 -0.789 0.783 0.418 183s -0.078 -0.951 0.407 -0.757 183s -0.321 0.945 0.500 0.018 183s 183s y: [1, 4, 3, 4] 183s -0.396 183s 0.523 183s 0.539 183s 0.487 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.557 0.240 0.862 -0.476 183s 0.408 -0.346 -0.307 -0.924 183s 0.536 0.421 -0.279 -0.199 183s -0.639 -0.535 -0.502 0.148 183s 183s x1: [4, 1, 1, 1] 183s -0.399 -0.756 -0.900 0.449 183s 183s y: [4, 1, 1, 1] 183s -0.949 -0.039 -0.370 1.178 183s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.117 -0.310 0.359 -0.101 183s 0.294 -0.765 -0.901 0.777 183s 0.609 -0.553 -0.002 -0.926 183s 0.318 -0.194 0.283 -0.250 183s 183s x1: [4, 4, 3, 1] 183s -0.414 -0.133 0.050 -0.976 183s -0.467 -0.459 -0.909 -0.768 183s 0.763 -0.889 -0.166 -0.992 183s -0.789 -0.717 -0.426 -0.347 183s 183s y: [4, 4, 3, 1] 183s 0.206 -0.822 0.725 0.152 183s -0.052 0.436 0.682 -0.125 183s 0.227 0.284 1.876 0.617 183s 0.197 0.432 0.238 -0.146 183s testing: mul_mat, [1, 4, 3, 4] = [4, 1, 3, 4] * [4, 4, 3, 4] 183s x0: [4, 1, 3, 4] 183s 0.751 -0.612 -0.239 -0.475 183s 183s x1: [4, 4, 3, 4] 183s 0.001 0.100 0.999 0.956 183s -0.602 -0.523 0.302 0.857 183s -0.830 0.321 -0.946 0.732 183s 0.703 0.096 -0.823 0.379 183s 183s y: [1, 4, 3, 4] 183s -0.754 183s -0.612 183s -0.941 183s 0.486 183s test-mul-mat0: iter:300/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.468 0.785 183s -0.862 0.166 183s 0.120 -0.483 183s 183s x1: [2, 1, 1, 1] 183s -0.306 -0.046 183s 183s y: [3, 1, 1, 1] 183s -0.179 0.256 -0.015 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.315 0.618 183s 183s x1: [2, 3, 1, 1] 183s -0.774 -0.794 183s 0.284 -0.953 183s 0.816 -0.512 183s 183s y: [1, 3, 1, 1] 183s -0.735 183s -0.500 183s -0.059 183s testing: mul_mat, [1, 1, 1, 4] = [2, 1, 1, 4] * [2, 1, 1, 4] 183s x0: [2, 1, 1, 4] 183s 0.099 0.481 183s 183s x1: [2, 1, 1, 4] 183s -0.722 0.530 183s 183s y: [1, 1, 1, 4] 183s 0.184 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.549 -0.343 183s 0.093 -0.622 183s 0.036 0.083 183s 183s x1: [2, 1, 1, 1] 183s 0.331 -0.424 183s 183s y: [3, 1, 1, 1] 183s -0.036 0.295 -0.023 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.186 0.103 0.911 183s -0.199 0.359 -0.542 183s 0.265 -0.133 -0.651 183s 183s x1: [3, 2, 1, 1] 183s -0.527 -0.134 -0.539 183s 0.789 -0.516 -0.261 183s 183s y: [3, 2, 1, 1] 183s -0.407 0.348 0.229 183s -0.437 -0.201 0.447 183s testing: mul_mat, [2, 3, 1, 4] = [3, 2, 1, 4] * [3, 3, 1, 4] 183s x0: [3, 2, 1, 4] 183s -0.249 -0.783 0.116 183s 0.848 -0.419 -0.771 183s 183s x1: [3, 3, 1, 4] 183s -0.498 0.160 -0.476 183s -0.952 0.598 -0.833 183s 0.379 -0.825 -0.023 183s 183s y: [2, 3, 1, 4] 183s -0.057 -0.122 183s -0.328 -0.416 183s 0.549 0.685 183s test-mul-mat0: iter:301/500 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.608 0.891 0.712 -0.340 183s 0.068 -0.444 -0.263 0.342 183s 183s x1: [4, 3, 1, 1] 183s -0.491 0.393 0.238 0.260 183s 0.241 0.455 0.841 -0.643 183s 0.685 -0.091 0.119 0.009 183s 183s y: [2, 3, 1, 1] 183s 0.730 -0.181 183s 1.076 -0.627 183s -0.416 0.059 183s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.570 -0.590 0.193 0.549 183s 0.420 -0.871 0.941 0.310 183s 0.841 -0.399 -0.622 -0.604 183s 183s x1: [4, 2, 2, 1] 183s -0.268 0.726 -0.927 0.109 183s 0.537 0.719 -0.966 -0.327 183s 183s y: [3, 2, 2, 1] 183s -0.394 -1.583 -0.004 183s -1.095 -1.412 0.962 183s testing: mul_mat, [2, 3, 2, 1] = [4, 2, 2, 1] * [4, 3, 2, 1] 183s x0: [4, 2, 2, 1] 183s 0.382 -0.155 -0.621 -0.343 183s 0.797 -0.318 0.113 0.581 183s 183s x1: [4, 3, 2, 1] 183s -0.544 0.784 0.574 -0.101 183s 0.857 -0.701 0.183 -0.712 183s -0.291 -0.624 0.837 -0.871 183s 183s y: [2, 3, 2, 1] 183s -0.651 -0.676 183s 0.567 0.513 183s -0.235 -0.444 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.436 0.521 0.517 0.370 183s -0.811 0.139 -0.744 0.744 183s 0.031 0.600 -0.837 -0.545 183s -0.782 -0.634 -0.801 0.382 183s 183s x1: [4, 2, 1, 1] 183s 0.562 0.393 -0.925 0.340 183s -0.167 -0.580 0.719 -0.350 183s 183s y: [4, 2, 1, 1] 183s 0.098 0.539 0.842 0.181 183s -0.132 -0.739 -0.765 -0.211 183s testing: mul_mat, [3, 4, 2, 1] = [4, 3, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.633 -0.485 -0.261 0.382 183s 0.223 0.692 0.502 0.031 183s 0.431 0.099 0.600 -0.249 183s 183s x1: [4, 4, 2, 1] 183s 0.182 0.010 -0.423 -0.093 183s -0.725 -0.852 -0.531 0.668 183s -0.777 0.810 -0.499 -0.357 183s 0.529 0.151 0.805 -0.035 183s 183s y: [3, 4, 2, 1] 183s -0.045 -0.168 -0.151 183s 1.266 -0.997 -0.882 183s 0.093 0.126 -0.465 183s -0.631 0.625 0.734 183s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 4, 2, 1] 183s 0.410 -0.396 -0.146 183s 0.603 -0.848 0.406 183s -0.417 -0.921 0.215 183s -0.101 -0.986 0.872 183s 183s x1: [3, 4, 2, 1] 183s 0.180 0.410 -0.298 183s 0.327 0.200 0.018 183s -0.517 0.568 -0.759 183s 0.914 -0.917 0.933 183s 183s y: [4, 4, 2, 1] 183s -0.045 -0.360 -0.517 -0.682 183s 0.052 0.034 -0.317 -0.215 183s -0.326 -1.102 -0.470 -1.170 183s 0.601 1.708 0.663 1.625 183s test-mul-mat0: iter:302/500 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.485 -0.889 -0.624 183s 183s x1: [3, 4, 1, 1] 183s -0.307 -0.950 -0.350 183s -0.897 0.653 0.233 183s 0.002 -0.743 0.385 183s 0.081 -0.729 -0.761 183s 183s y: [1, 4, 1, 1] 183s 0.914 183s -1.161 183s 0.422 183s 1.163 183s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 1, 2, 1] 183s 0.750 -0.575 -0.289 183s 183s x1: [3, 1, 2, 1] 183s -0.732 -0.776 -0.820 183s 183s y: [1, 1, 2, 1] 183s 0.134 183s testing: mul_mat, [3, 1, 2, 2] = [3, 3, 2, 2] * [3, 1, 2, 2] 183s x0: [3, 3, 2, 2] 183s -0.030 0.359 0.821 183s 0.081 0.735 -0.911 183s 0.304 0.915 0.222 183s 183s x1: [3, 1, 2, 2] 183s -0.894 0.463 -0.314 183s 183s y: [3, 1, 2, 2] 183s -0.065 0.555 0.082 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.479 -0.308 -0.836 183s 183s x1: [3, 3, 1, 1] 183s 0.399 0.562 0.891 183s -0.379 0.135 -0.198 183s -0.336 -0.541 -0.862 183s 183s y: [1, 3, 1, 1] 183s -0.726 183s -0.058 183s 0.726 183s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.686 183s 0.828 183s -0.870 183s -0.479 183s 183s x1: [1, 3, 2, 1] 183s -0.395 183s -0.292 183s 0.597 183s 183s y: [4, 3, 2, 1] 183s 0.271 -0.327 0.343 0.189 183s 0.200 -0.242 0.254 0.140 183s -0.409 0.494 -0.519 -0.285 183s testing: mul_mat, [4, 1, 2, 2] = [4, 4, 2, 2] * [4, 1, 2, 2] 183s x0: [4, 4, 2, 2] 183s 0.159 -0.390 -0.810 -0.468 183s -0.207 -0.940 -0.167 -0.898 183s 0.084 -0.689 0.149 -0.261 183s 0.997 -0.562 0.809 -0.512 183s 183s x1: [4, 1, 2, 2] 183s -0.070 0.957 -0.071 -0.671 183s 183s y: [4, 1, 2, 2] 183s -0.012 -0.269 -0.500 -0.322 183s test-mul-mat0: iter:303/500 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.159 183s 183s x1: [1, 3, 1, 1] 183s 0.049 183s -0.692 183s 0.630 183s 183s y: [1, 3, 1, 1] 183s 0.008 183s -0.110 183s 0.100 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.888 183s 0.785 183s -0.451 183s -0.905 183s 183s x1: [1, 1, 1, 1] 183s -0.662 183s 183s y: [4, 1, 1, 1] 183s 0.588 -0.520 0.298 0.599 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.232 183s -0.001 183s -0.327 183s 0.767 183s 183s x1: [1, 4, 1, 1] 183s -0.699 183s 0.559 183s 0.928 183s -0.166 183s 183s y: [4, 4, 1, 1] 183s -0.162 0.001 0.229 -0.536 183s 0.129 -0.001 -0.183 0.428 183s 0.215 -0.001 -0.303 0.711 183s -0.038 0.000 0.054 -0.127 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.044 183s 183s x1: [1, 4, 1, 1] 183s 0.828 183s -0.517 183s 0.985 183s -0.482 183s 183s y: [1, 4, 1, 1] 183s 0.037 183s -0.023 183s 0.043 183s -0.021 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.057 183s 183s x1: [1, 1, 1, 1] 183s 0.861 183s 183s y: [1, 1, 1, 1] 183s -0.049 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.991 183s 183s x1: [1, 1, 1, 1] 183s -0.532 183s 183s y: [1, 1, 1, 1] 183s -0.528 183s test-mul-mat0: iter:304/500 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.323 -0.930 0.498 183s 0.906 0.302 -0.503 183s -0.421 0.068 -0.674 183s 0.062 0.053 -0.156 183s 183s x1: [3, 2, 1, 1] 183s -0.422 -0.430 -0.952 183s -0.636 0.119 -0.858 183s 183s y: [4, 2, 1, 1] 183s 0.062 -0.033 0.790 0.100 183s -0.332 -0.110 0.854 0.101 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.376 0.932 -0.409 183s 183s x1: [3, 4, 1, 1] 183s -0.081 -0.903 -0.295 183s -0.202 0.040 0.173 183s 0.408 0.031 0.949 183s 0.649 -0.946 0.884 183s 183s y: [1, 4, 1, 1] 183s -0.690 183s 0.042 183s -0.513 183s -1.487 183s testing: mul_mat, [2, 1, 1, 2] = [3, 2, 1, 2] * [3, 1, 1, 2] 183s x0: [3, 2, 1, 2] 183s 0.446 0.251 -0.095 183s -0.485 0.576 0.967 183s 183s x1: [3, 1, 1, 2] 183s 0.742 -0.926 0.255 183s 183s y: [2, 1, 1, 2] 183s 0.075 -0.647 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.869 0.559 -0.425 183s -0.032 -0.844 -0.718 183s -0.587 -0.863 0.568 183s -0.937 -0.186 -0.098 183s 183s x1: [3, 2, 1, 1] 183s -0.296 -0.702 0.092 183s 0.735 -0.754 -0.259 183s 183s y: [4, 2, 1, 1] 183s -0.175 0.536 0.832 0.398 183s -0.950 0.799 0.071 -0.524 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.763 0.914 -0.822 -0.386 183s 0.437 -0.517 0.704 0.146 183s 183s x1: [4, 3, 1, 1] 183s 0.728 -0.182 0.807 -0.758 183s -0.606 0.774 0.810 -0.186 183s -0.340 0.475 0.939 -0.656 183s 183s y: [2, 3, 1, 1] 183s -1.093 0.869 183s 0.576 -0.122 183s 0.174 0.172 183s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 183s x0: [2, 4, 1, 2] 183s 0.114 -0.664 183s 0.991 -0.470 183s -0.649 0.290 183s 0.053 0.772 183s 183s x1: [2, 4, 1, 2] 183s -0.883 -0.323 183s -0.295 -0.728 183s -0.186 0.519 183s -0.152 0.096 183s 183s y: [4, 4, 1, 2] 183s 0.114 -0.723 0.480 -0.296 183s 0.450 0.050 -0.020 -0.578 183s -0.366 -0.428 0.271 0.391 183s -0.081 -0.196 0.127 0.066 183s test-mul-mat0: iter:305/500 183s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.979 0.742 183s 0.727 -0.988 183s -0.908 -0.220 183s 183s x1: [2, 4, 1, 1] 183s 0.384 -0.024 183s -0.592 0.134 183s 0.800 -0.687 183s -0.310 -0.134 183s 183s y: [3, 4, 1, 1] 183s -0.395 0.304 -0.343 183s 0.679 -0.562 0.508 183s -1.292 1.260 -0.574 183s 0.204 -0.093 0.311 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.213 -0.769 183s 0.166 0.478 183s 183s x1: [2, 3, 1, 1] 183s -0.653 -0.378 183s -0.930 -0.880 183s 0.066 0.844 183s 183s y: [2, 3, 1, 1] 183s 0.151 -0.289 183s 0.478 -0.575 183s -0.634 0.414 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.399 -0.849 183s -0.303 -0.468 183s 183s x1: [2, 2, 1, 1] 183s 0.296 -0.032 183s -0.986 0.176 183s 183s y: [2, 2, 1, 1] 183s 0.145 -0.075 183s -0.542 0.217 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.243 0.350 183s -0.442 0.255 183s 183s x1: [2, 2, 1, 1] 183s 0.951 0.010 183s 0.222 -0.183 183s 183s y: [2, 2, 1, 1] 183s 0.235 -0.417 183s -0.010 -0.145 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.722 -0.957 183s -0.661 0.792 183s 183s x1: [2, 2, 1, 1] 183s -0.349 -0.871 183s 0.602 0.273 183s 183s y: [2, 2, 1, 1] 183s 0.581 -0.458 183s 0.174 -0.182 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.926 0.044 183s 183s x1: [2, 2, 1, 1] 183s 0.552 -0.725 183s -0.042 0.030 183s 183s y: [1, 2, 1, 1] 183s 0.479 183s -0.037 183s test-mul-mat0: iter:306/500 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.440 -0.776 183s -0.107 -0.305 183s -0.126 0.023 183s 183s x1: [2, 3, 1, 1] 183s -0.089 -0.025 183s -0.151 0.921 183s -0.803 0.665 183s 183s y: [3, 3, 1, 1] 183s -0.019 0.017 0.011 183s -0.782 -0.265 0.040 183s -0.870 -0.117 0.116 183s testing: mul_mat, [1, 3, 4, 1] = [2, 1, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 1, 4, 1] 183s 0.165 0.493 183s 183s x1: [2, 3, 4, 1] 183s -0.703 -0.853 183s 0.223 -0.981 183s -0.514 0.266 183s 183s y: [1, 3, 4, 1] 183s -0.537 183s -0.447 183s 0.046 183s testing: mul_mat, [1, 1, 4, 1] = [2, 1, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 1, 4, 1] 183s -0.657 0.452 183s 183s x1: [2, 1, 4, 1] 183s -0.575 -0.824 183s 183s y: [1, 1, 4, 1] 183s 0.005 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.975 0.325 183s 0.832 0.392 183s 0.212 0.050 183s 0.190 0.693 183s 183s x1: [2, 1, 1, 1] 183s -0.700 -0.438 183s 183s y: [4, 1, 1, 1] 183s 0.541 -0.754 -0.170 -0.437 183s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.064 -0.733 -0.501 -0.727 183s 183s x1: [4, 2, 4, 1] 183s -0.406 0.208 -0.792 0.020 183s 0.384 0.209 -0.289 0.371 183s 183s y: [1, 2, 4, 1] 183s 0.204 183s -0.254 183s testing: mul_mat, [3, 4, 4, 1] = [1, 3, 4, 1] * [1, 4, 4, 1] 183s x0: [1, 3, 4, 1] 183s -0.518 183s -0.664 183s -0.901 183s 183s x1: [1, 4, 4, 1] 183s 0.137 183s 0.644 183s -0.566 183s 0.143 183s 183s y: [3, 4, 4, 1] 183s -0.071 -0.091 -0.123 183s -0.333 -0.427 -0.580 183s 0.293 0.376 0.510 183s -0.074 -0.095 -0.128 183s test-mul-mat0: iter:307/500 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.177 183s 183s x1: [1, 3, 1, 1] 183s -0.594 183s -0.803 183s -0.213 183s 183s y: [1, 3, 1, 1] 183s -0.105 183s -0.142 183s -0.038 183s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.348 183s -0.665 183s 183s x1: [1, 1, 2, 1] 183s 0.741 183s 183s y: [2, 1, 2, 1] 183s 0.258 -0.493 183s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 1, 2, 1] 183s 0.421 183s 183s x1: [1, 2, 2, 1] 183s -0.682 183s -0.617 183s 183s y: [1, 2, 2, 1] 183s -0.287 183s -0.260 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.505 183s 0.310 183s -0.335 183s 0.435 183s 183s x1: [1, 1, 1, 1] 183s 0.577 183s 183s y: [4, 1, 1, 1] 183s 0.291 0.179 -0.193 0.251 183s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 3, 2, 1] 183s 0.376 -0.157 0.178 0.495 183s 0.012 -0.668 -0.348 0.035 183s -0.493 0.854 -0.944 0.379 183s 183s x1: [4, 1, 2, 1] 183s -0.510 -0.417 -0.115 -0.538 183s 183s y: [3, 1, 2, 1] 183s -0.413 0.293 -0.200 183s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 2, 2, 1] 183s -0.950 0.760 0.849 183s -0.282 -0.561 -0.751 183s 183s x1: [3, 4, 2, 1] 183s 0.274 0.593 -0.874 183s 0.253 -0.539 0.961 183s -0.981 0.836 -0.027 183s -0.474 -0.321 0.305 183s 183s y: [2, 4, 2, 1] 183s -0.552 0.246 183s 0.164 -0.491 183s 1.545 -0.172 183s 0.466 0.085 183s test-mul-mat0: iter:308/500 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.671 0.627 183s 0.562 -0.856 183s 0.274 -0.692 183s -0.241 -0.630 183s 183s x1: [2, 2, 1, 1] 183s -0.582 -0.387 183s 0.892 -0.725 183s 183s y: [4, 2, 1, 1] 183s 0.148 0.004 0.108 0.384 183s -1.053 1.122 0.746 0.242 183s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.581 0.183 183s -0.852 0.278 183s -0.866 -0.229 183s 0.849 -0.472 183s 183s x1: [2, 4, 4, 1] 183s -0.615 -0.386 183s -0.580 0.103 183s -0.625 -0.141 183s -0.048 -0.377 183s 183s y: [4, 4, 4, 1] 183s 0.286 0.417 0.620 -0.340 183s 0.356 0.523 0.479 -0.541 183s 0.337 0.494 0.574 -0.464 183s -0.041 -0.063 0.128 0.137 183s testing: mul_mat, [4, 4, 4, 4] = [2, 4, 4, 4] * [2, 4, 4, 4] 183s x0: [2, 4, 4, 4] 183s -0.271 0.434 183s 0.102 0.837 183s -0.958 0.898 183s -0.137 0.050 183s 183s x1: [2, 4, 4, 4] 183s 0.164 0.611 183s 0.719 0.299 183s -0.618 0.569 183s 0.827 -0.848 183s 183s y: [4, 4, 4, 4] 183s 0.220 0.528 0.391 0.008 183s -0.065 0.323 -0.420 -0.084 183s 0.414 0.413 1.102 0.113 183s -0.592 -0.626 -1.553 -0.156 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.783 -0.013 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.pn 183s 0.998 -0.202 183s -0.760 0.544 183s 0.913 0.515 183s 183s x1: [2, 4, 1, 1] 183s -0.936 -0.062 183s -0.867 0.094 183s -0.426 -0.305 183s -0.930 -0.293 183s 183s y: [4, 4, 1, 1] 183s -0.733 -0.922 0.678 -0.887 183s -0.680 -0.884 0.710 -0.742 183s -0.330 -0.363 0.158 -0.546 183s -0.725 -0.869 0.548 -1.000 183s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 1, 4, 1] 183s -0.331 -0.223 0.838 -0.040 183s 183s x1: [4, 2, 4, 1] 183s -0.476 0.993 0.870 -0.272 183s -0.465 -0.708 0.511 0.096 183s 183s y: [1, 2, 4, 1] 183s 0.676 183s 0.736 183s testing: mul_mat, [4, 4, 4, 4] = [1, 4, 4, 4] * [1, 4, 4, 4] 183s x0: [1, 4, 4, 4] 183s 0.801 183s 0.727 183s 0.209 183s -0.603 183s 183s x1: [1, 4, 4, 4] 183s -0.699 183s -0.071 183s 0.618 183s -0.402 183s 183s y: [4, 4, 4, 4] 183s -0.560 -0.508 -0.146 0.421 183s -0.057 -0.051 -0.015 0.043 183s 0.495 0.450 0.129 -0.373 183s -0.322 -0.292 -0.084 0.242 183s test-mul-mat0: iter:309/500 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.054 0.839 0.459 183s 0.536 0.762 0.292 183s 0.775 -0.075 0.539 183s 183s x1: [3, 4, 1, 1] 183s 0.026 0.634 0.486 183s 0.168 0.521 0.051 183s 0.619 -0.811 -0.244 183s -0.625 0.335 0.782 183s 183s y: [3, 4, 1, 1] 183s 0.756 0.639 0.234 183s 0.470 0.502 0.118 183s -0.759 -0.357 0.409 183s 0.606 0.149 -0.087 183s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.882 0.497 0.734 183s 183s x1: [3, 3, 3, 1] 183s 0.868 -0.831 -0.830 183s 0.617 0.221 -0.606 183s 0.419 -0.162 0.895 183s 183s y: [1, 3, 3, 1] 183s -0.257 183s 0.209 183s 0.947 183s testing: mul_mat, [2, 1, 3, 4] = [3, 2, 3, 4] * [3, 1, 3, 4] 183s x0: [3, 2, 3, 4] 183s -0.572 0.713 0.941 183s -0.495 -0.914 0.137 183s 183s x1: [3, 1, 3, 4] 183s -0.214 0.725 -0.806 183s 183s y: [2, 1, 3, 4] 183s -0.119 -0.667 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.913 0.440 -0.409 183s 183s x1: [3, 2, 1, 1] 183s -0.764 -0.677 -0.150 183s 0.497 -0.321 0.376 183s 183s y: [1, 2, 1, 1] 183s -0.934 183s 0.159 183s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 3, 3, 1] 183s 0.720 183s 0.622 183s -0.809 183s 183s x1: [1, 3, 3, 1] 183s -0.261 183s -0.094 183s -0.810 183s 183s y: [3, 3, 3, 1] 183s -0.188 -0.162 0.211 183s -0.067 -0.058 0.076 183s -0.583 -0.504 0.656 183s testing: mul_mat, [2, 1, 3, 4] = [3, 2, 3, 4] * [3, 1, 3, 4] 183s x0: [3, 2, 3, 4] 183s 0.616 0.076 0.416 183s 0.914 -0.773 0.952 183s 183s x1: [3, 1, 3, 4] 183s 0.695 0.837 -0.427 183s 183s y: [2, 1, 3, 4] 183s 0.314 -0.418 183s test-mul-mat0: iter:310/500 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.350 0.356 0.966 0.485 183s 183s x1: [4, 3, 1, 1] 183s 0.153 -0.631 -0.307 -0.184 183s 0.418 -0.402 0.198 -0.929 183s 0.407 0.626 -0.525 -0.979 183s 183s y: [1, 3, 1, 1] 183s -0.557 183s -0.257 183s -0.617 183s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.570 -0.234 -0.550 -0.233 183s -0.163 0.857 -0.607 0.312 183s 183s x1: [4, 1, 4, 1] 183s 0.031 0.643 0.956 0.270 183s 183s y: [2, 1, 4, 1] 183s -0.722 0.050 183s testing: mul_mat, [1, 2, 4, 4] = [4, 1, 4, 4] * [4, 2, 4, 4] 183s x0: [4, 1, 4, 4] 183s 0.732 0.596 0.943 -0.435 183s 183s x1: [4, 2, 4, 4] 183s -0.859 0.107 -0.712 -0.022 183s -0.036 -0.319 -0.711 -0.159 183s 183s y: [1, 2, 4, 4] 183s -1.226 183s -0.817 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.421 0.288 -0.367 0.014 183s 0.326 0.122 0.581 0.905 183s 0.380 0.582 0.982 -0.792 183s 0.919 0.354 0.184 0.428 183s 183s x1: [4, 1, 1, 1] 183s -0.397 -0.209 -0.973 -0.514 183s 183s y: [4, 1, 1, 1] 183s 0.457 -1.186 -0.821 -0.838 183s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 1, 4, 1] 183s -0.622 -0.197 -0.169 0.887 183s 183s x1: [4, 4, 4, 1] 183s -0.566 -0.782 -0.536 0.414 183s 0.121 -0.924 0.813 -0.097 183s 0.995 0.227 0.506 -0.213 183s 0.254 0.992 -0.544 0.833 183s 183s y: [1, 4, 4, 1] 183s 0.963 183s -0.117 183s -0.938 183s 0.478 183s testing: mul_mat, [3, 4, 4, 4] = [1, 3, 4, 4] * [1, 4, 4, 4] 183s x0: [1, 3, 4, 4] 183s 0.266 183s -0.756 183s -0.094 183s 183s x1: [1, 4, 4, 4] 183s -0.822 183s 0.445 183s -0.079 183s -0.212 183s 183s y: [3, 4, 4, 4] 183s -0.218 0.622 0.077 183s 0.118 -0.337 -0.042 183s -0.021 0.060 0.007 183s -0.056 0.160 0.020 183s test-mul-mat0: iter:311/500 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.767 -0.313 183s 0.898 -0.896 183s 183s x1: [2, 3, 1, 1] 183s -0.487 -0.709 183s -0.921 -0.030 183s -0.304 0.246 183s 183s y: [2, 3, 1, 1] 183s -0.151 0.199 183s -0.697 -0.800 183s -0.311 -0.494 183s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 2, 4, 1] 183s -0.069 -0.171 183s -0.430 0.44g 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.p5 183s 183s x1: [2, 2, 4, 1] 183s -0.629 0.701 183s 0.647 0.536 183s 183s y: [2, 2, 4, 1] 183s -0.077 0.582 183s -0.136 -0.040 183s testing: mul_mat, [1, 2, 4, 3] = [2, 1, 4, 3] * [2, 2, 4, 3] 183s x0: [2, 1, 4, 3] 183s -0.301 0.319 183s 183s x1: [2, 2, 4, 3] 183s 0.333 -0.512 183s 0.962 0.411 183s 183s y: [1, 2, 4, 3] 183s -0.264 183s -0.158 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.619 -0.521 183s -0.142 0.557 183s 0.320 -0.360 183s 183s x1: [2, 1, 1, 1] 183s -0.224 0.962 183s 183s y: [3, 1, 1, 1] 183s -0.640 0.568 -0.418 183s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 1, 4, 1] 183s -0.836 0.645 0.723 183s 183s x1: [3, 2, 4, 1] 183s -0.996 -0.527 -0.607 183s -0.910 0.022 -0.838 183s 183s y: [1, 2, 4, 1] 183s 0.054 183s 0.169 183s testing: mul_mat, [3, 3, 4, 3] = [1, 3, 4, 3] * [1, 3, 4, 3] 183s x0: [1, 3, 4, 3] 183s 0.400 183s -0.062 183s -0.942 183s 183s x1: [1, 3, 4, 3] 183s -0.945 183s 0.213 183s -0.447 183s 183s y: [3, 3, 4, 3] 183s -0.378 0.058 0.891 183s 0.085 -0.013 -0.201 183s -0.179 0.028 0.422 183s test-mul-mat0: iter:312/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.008 -0.638 183s 183s x1: [2, 4, 1, 1] 183s 0.680 -0.807 183s 0.275 0.010 183s 0.336 0.767 183s 0.332 -0.490 183s 183s y: [1, 4, 1, 1] 183s 0.520 183s -0.004 183s -0.487 183s 0.315 183s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 4, 3, 1] 183s -0.990 0.272 183s -0.335 -0.317 183s 0.576 -0.276 183s 0.414 0.995 183s 183s x1: [2, 1, 3, 1] 183s 0.362 -0.206 183s 183s y: [4, 1, 3, 1] 183s -0.414 -0.056 0.265 -0.055 183s testing: mul_mat, [4, 4, 3, 3] = [2, 4, 3, 3] * [2, 4, 3, 3] 183s x0: [2, 4, 3, 3] 183s -0.223 0.376 183s -0.389 0.582 183s 0.596 0.387 183s -0.684 -0.789 183s 183s x1: [2, 4, 3, 3] 183s -0.079 0.437 183s 0.104 0.758 183s 0.047 0.473 183s 0.303 0.057 183s 183s y: [4, 4, 3, 3] 183s 0.182 0.285 0.122 -0.291 183s 0.262 0.401 0.355 -0.670 183s 0.167 0.257 0.211 -0.406 183s -0.046 -0.085 0.202 -0.252 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.726 -0.322 183s 0.145 0.353 183s 0.755 0.069 183s 0.150 0.498 183s 183s x1: [2, 4, 1, 1] 183s -0.152 0.416 183s -0.930 0.300 183s 0.082 0.926 183s -0.324 -0.375 183s 183s y: [4, 4, 1, 1] 183s -0.244 0.125 -0.086 0.184 183s -0.772 -0.029 -0.681 0.010 183s -0.239 0.339 0.126 0.473 183s -0.114 -0.179 -0.270 -0.235 183s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.252 -0.536 -0.187 0.725 183s 0.395 -0.271 0.424 0.712 183s 0.231 0.291 -0.730 -0.097 183s -0.395 -0.765 -0.375 -0.694 183s 183s x1: [4, 2, 3, 1] 183s -0.314 0.771 -0.916 -0.868 183s -0.965 0.812 -0.698 0.442 183s 183s y: [4, 2, 3, 1] 183s -0.950 -1.339 0.905 0.480 183s -0.227 -0.582 0.480 -0.284 183s testing: mul_mat, [4, 4, 3, 3] = [4, 4, 3, 3] * [4, 4, 3, 3] 183s x0: [4, 4, 3, 3] 183s -0.621 0.257 -0.734 0.247 183s -0.501 -0.646 -0.425 -0.995 183s 0.129 -0.255 -0.277 -0.573 183s -0.655 -0.722 -0.191 0.254 183s 183s x1: [4, 4, 3, 3] 183s 0.937 0.487 -0.236 0.076 183s 0.780 -0.339 -0.943 -0.144 183s 0.457 0.557 0.342 0.749 183s 0.302 -0.619 0.294 -0.559 183s 183s y: [4, 4, 3, 3] 183s -0.265 -0.759 0.019 -0.900 183s 0.085 0.372 0.531 -0.122 183s -0.206 -1.480 -0.607 -0.576 183s -0.701 0.680 0.435 0.051 183s test-mul-mat0: iter:313/500 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.902 0.166 183s 0.573 0.522 183s 183s x1: [2, 1, 1, 1] 183s -0.244 0.548 183s 183s y: [2, 1, 1, 1] 183s -0.129 0.146 183s testing: mul_mat, [3, 2, 3, 1] = [2, 3, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 3, 3, 1] 183s 0.248 0.364 183s -0.319 -0.599 183s 0.862 -0.136 183s 183s x1: [2, 2, 3, 1] 183s 0.147 0.061 183s -0.943 -0.177 183s 183s y: [3, 2, 3, 1] 183s 0.059 -0.083 0.118 183s -0.299 0.406 -0.789 183s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.179 0.550 183s 0.373 0.351 183s 183s x1: [2, 3, 3, 1] 183s -0.968 -0.250 183s -0.420 -0.145 183s -0.960 -0.280 183s 183s y: [2, 3, 3, 1] 183s -0.311 -0.449 183s -0.155 -0.208 183s -0.326 -0.456 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.298 -0.482 183s 0.150 -0.118 183s 183s x1: [2, 2, 1, 1] 183s 0.986 0.424 183s -0.422 -0.159 183s 183s y: [2, 2, 1, 1] 183s 0.090 0.098 183s -0.049 -0.045 183s testing: mul_mat, [4, 2, 3, 1] = [2, 4, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 4, 3, 1] 183s -0.557 0.389 183s -0.595 -0.741 183s -0.215 0.468 183s 0.513 -0.938 183s 183s x1: [2, 2, 3, 1] 183s -0.739 -0.983 183s 0.054 0.745 183s 183s y: [4, 2, 3, 1] 183s 0.029 1.168 -0.301 0.542 183s 0.260 -0.584 0.337 -0.671 183s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.582 0.675 0.270 0.310 183s 0.425 -0.375 -0.303 -0.204 183s 0.967 -0.604 0.096 -0.964 183s 0.034 0.839 0.451 -0.292 183s 183s x1: [4, 2, 3, 1] 183s 0.362 -0.836 -0.356 -0.699 183s 0.257 -0.819 -0.387 0.700 183s 183s y: [4, 2, 3, 1] 183s -0.666 0.718 1.495 -0.645 183s -0.291 0.391 0.032 -1.058 183s test-mul-mat0: iter:314/500 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.186 0.298 -0.573 0.329 183s 183s x1: [4, 3, 1, 1] 183s 0.501 -0.663 -0.214 -0.470 183s 0.593 -0.172 -0.559 -0.276 183s -0.191 -0.750 0.438 -0.784 183s 183s y: [1, 3, 1, 1] 183s -0.136 183s 0.288 183s -0.768 183s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 2, 2, 1] 183s 0.872 0.546 -0.604 0.384 183s -0.238 0.897 0.721 0.548 183s 183s x1: [4, 1, 2, 1] 183s -0.780 -0.902 -0.832 -0.763 183s 183s y: [2, 1, 2, 1] 183s -0.963 -1.642 183s testing: mul_mat, [3, 2, 2, 2] = [4, 3, 2, 2] * [4, 2, 2, 2] 183s x0: [4, 3, 2, 2] 183s -0.326 -0.220 0.072 -0.336 183s 0.699 -0.260 -0.177 -0.145 183s 0.628 0.299 0.060 -0.107 183s 183s x1: [4, 2, 2, 2] 183s 0.401 0.144 -0.338 -0.301 183s 0.571 0.991 -0.081 0.668 183s 183s y: [3, 2, 2, 2] 183s -0.086 0.346 0.307 183s -0.634 0.059 0.578 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.655 0.981 -0.343 0.824 183s 183s x1: [4, 3, 1, 1] 183s 0.718 0.171 -0.435 -0.887 183s -0.519 -0.157 -0.997 0.031 183s 0.657 -0.426 0.937 -0.518 183s 183s y: [1, 3, 1, 1] 183s -0.885 183s 0.553 183s -1.596 183s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 2, 2, 1] 183s -0.350 183s 0.004 183s 183s x1: [1, 4, 2, 1] 183s -0.856 183s 0.788 183s -0.957 183s 0.130 183s 183s y: [2, 4, 2, 1] 183s 0.299 -0.004 183s -0.276 0.003 183s 0.335 -0.004 183s -0.045 0.001 183s testing: mul_mat, [4, 1, 2, 2] = [2, 4, 2, 2] * [2, 1, 2, 2] 183s x0: [2, 4, 2, 2] 183s -0.148 -0.296 183s 0.479 0.846 183s 0.501 0.314 183s 0.333 0.361 183s 183s x1: [2, 1, 2, 2] 183s 0.787 -0.211 183s 183s y: [4, 1, 2, 2] 183s -0.054 0.198 0.328 0.186 183s test-mul-mat0: iter:315/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.442 183s -0.301 183s 183s x1: [1, 1, 1, 1] 183s 0.029 183s 183s y: [2, 1, 1, 1] 183s 0.013 -0.009 183s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.987 183s 0.631 183s 0.259 183s 183s x1: [1, 2, 2, 1] 183s 0.040 183s -0.409 183s 183s y: [3, 2, 2, 1] 183s -0.039 0.025 0.010 183s 0.403 -0.258 -0.106 183s testing: mul_mat, [4, 3, 2, 2] = [1, 4, 2, 2] * [1, 3, 2, 2] 183s x0: [1, 4, 2, 2] 183s -0.607 183s -0.285 183s 0.987 183s 0.422 183s 183s x1: [1, 3, 2, 2] 183s 0.546 183s 0.820 183s -0.458 183s 183s y: [4, 3, 2, 2] 183s -0.332 -0.156 0.539 0.231 183s -0.498 -0.234 0.809 0.346 183s 0.278 0.130 -0.452 -0.193 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.973 183s 183s x1: [1, 4, 1, 1] 183s -0.119 183s -0.199 183s -0.665 183s -0.299 183s 183s y: [1, 4, 1, 1] 183s -0.116 183s -0.194 183s -0.647 183s -0.291 183s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.634 183s -0.714 183s 183s x1: [1, 1, 2, 1] 183s -0.892 183s 183s y: [2, 1, 2, 1] 183s -0.566 0.637 183s testing: mul_mat, [1, 1, 2, 2] = [2, 1, 2, 2] * [2, 1, 2, 2] 183s x0: [2, 1, 2, 2] 183s 0.082 -0.525 183s 183s x1: [2, 1, 2, 2] 183s -0.267 0.040 183s 183s y: [1, 1, 2, 2] 183s -0.043 183s test-mul-mat0: iter:316/500 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.722 183s 0.558 183s 0.966 183s 183s x1: [1, 3, 1, 1] 183s -0.776 183s -0.051 183s -0.047 183s 183s y: [3, 3, 1, 1] 183s -0.560 -0.433 -0.750 183s -0.037 -0.029 -0.050 183s -0.034 -0.026 -0.045 183s testing: mul_mat, [2, 3, 4, 1] = [1, 2, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.688 183s -0.205 183s 183s x1: [1, 3, 4, 1] 183s -0.992 183s -0.309 183s -0.190 183s 183s y: [2, 3, 4, 1] 183s -0.683 0.203 183s -0.212 0.063 183s -0.131 0.039 183s testing: mul_mat, [3, 2, 4, 2] = [1, 3, 4, 2] * [1, 2, 4, 2] 183s x0: [1, 3, 4, 2] 183s 0.967 183s 0.308 183s 0.651 183s 183s x1: [1, 2, 4, 2] 183s 0.226 183s -0.892 183s 183s y: [3, 2, 4, 2] 183s 0.219 0.070 0.147 183s -0.862 -0.275 -0.580 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.486 183s -0.200 183s 0.046 183s 183s x1: [1, 3, 1, 1] 183s 0.801 183s 0.304 183s -0.804 183s 183s y: [3, 3, 1, 1] 183s -0.389 -0.160 0.037 183s -0.148 -0.061 0.014 183s 0.390 0.161 -0.037 183s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.741 0.285 -0.408 183s 0.178 0.578 -0.067 183s 0.343 0.107 -0.186 183s -0.931 0.175 -0.607 183s 183s x1: [3, 1, 4, 1] 183s -0.519 -0.891 -0.303 183s 183s y: [4, 1, 4, 1] 183s 0.253 -0.587 -0.217 0.510 183s testing: mul_mat, [2, 3, 4, 2] = [4, 2, 4, 2] * [4, 3, 4, 2] 183s x0: [4, 2, 4, 2] 183s -0.283 0.071 -0.565 -0.438 183s -0.741 -0.814ng 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot. 0.154 0.090 183s 183s x1: [4, 3, 4, 2] 183s 0.084 -0.374 0.632 -0.450 183s 0.520 -0.548 0.379 -0.055 183s 0.759 0.298 0.465 0.793 183s 183s y: [2, 3, 4, 2] 183s -0.210 0.299 183s -0.376 0.114 183s -0.803 -0.662 183s test-mul-mat0: iter:317/500 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.727 -0.010 0.143 -0.505 183s 0.816 -0.287 -0.216 -0.857 183s 183s x1: [4, 4, 1, 1] 183s -0.716 0.224 -0.475 -0.713 183s 0.382 0.915 -0.045 -0.924 183s -0.340 0.979 -0.537 0.515 183s -0.942 -0.059 0.348 0.475 183s 183s y: [2, 4, 1, 1] 183s 0.810 0.064 183s 0.173 0.849 183s -0.099 -0.884 183s 0.496 -1.234 183s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.886 -0.602 0.662 -0.173 183s 0.182 0.805 -0.631 -0.422 183s -0.153 -0.814 0.348 -0.486 183s 183s x1: [4, 2, 3, 1] 183s 0.542 0.397 0.042 0.817 183s -0.230 0.667 -0.899 0.994 183s 183s y: [3, 2, 3, 1] 183s -0.833 0.047 -0.788 183s -0.965 0.644 -1.304 183s testing: mul_mat, [2, 3, 3, 3] = [4, 2, 3, 3] * [4, 3, 3, 3] 183s x0: [4, 2, 3, 3] 183s -0.190 0.673 -0.535 0.502 183s 0.113 -0.098 -0.567 0.607 183s 183s x1: [4, 3, 3, 3] 183s -0.109 0.221 0.550 0.739 183s 0.407 -0.102 -0.747 0.694 183s -0.760 -0.042 -0.631 -0.042 183s 183s y: [2, 3, 3, 3] 183s 0.246 0.103 183s 0.602 0.901 183s 0.433 0.251 183s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.223 0.211 -0.161 0.985 183s 0.602 0.804 0.323 0.381 183s -0.533 0.358 0.197 0.423 183s 183s x1: [4, 2, 1, 1] 183s -0.291 -0.808 -0.869 0.859 183s -0.317 0.841 0.951 0.105 183s 183s y: [3, 2, 1, 1] 183s 0.880 -0.778 0.058 183s 0.198 0.832 0.701 183s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.655 0.536 0.628 183s 183s x1: [3, 4, 3, 1] 183s 0.459 0.653 -0.498 183s -0.239 0.250 0.335 183s -0.675 0.780 -0.236 183s 0.210 -0.511 -0.044 183s 183s y: [1, 4, 3, 1] 183s 0.337 183s 0.188 183s -0.172 183s -0.164 183s testing: mul_mat, [3, 3, 3, 3] = [1, 3, 3, 3] * [1, 3, 3, 3] 183s x0: [1, 3, 3, 3] 183s -0.107 183s 0.984 183s -0.121 183s 183s x1: [1, 3, 3, 3] 183s 0.386 183s -0.328 183s -0.034 183s 183s y: [3, 3, 3, 3] 183s -0.041 0.380 -0.047 183s 0.035 -0.323 0.040 183s 0.004 -0.034 0.004 183s test-mul-mat0: iter:318/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.518 0.238 -0.198 183s 0.921 -0.606 -0.937 183s -0.642 0.011 -0.255 183s 0.955 -0.439 -0.547 183s 183s x1: [3, 1, 1, 1] 183s -0.722 -0.138 0.392 183s 183s y: [4, 1, 1, 1] 183s -0.484 -0.949 0.362 -0.843 183s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.716 -0.058 0.479 183s 0.813 0.420 -0.329 183s 0.296 -0.224 -0.625 183s 0.057 -0.006 0.472 183s 183s x1: [3, 4, 3, 1] 183s 0.821 0.899 -0.788 183s -0.474 0.531 0.551 183s 0.400 0.212 0.335 183s -0.902 -0.522 0.191 183s 183s y: [4, 4, 3, 1] 183s 0.157 1.305 0.534 -0.331 183s -0.106 -0.344 -0.604 0.230 183s 0.435 0.304 -0.138 0.180 183s -0.524 -1.016 -0.270 0.042 183s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.852 -0.262 0.137 183s 183s x1: [3, 4, 3, 1] 183s -0.125 0.300 0.641 183s 0.250 -0.642 -0.365 183s -0.278 0.733 0.364 183s -0.472 -0.772 0.749 183s 183s y: [1, 4, 3, 1] 183s -0.098 183s 0.331 183s -0.379 183s -0.097 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.879 0.382 0.190 183s -0.770 -0.441 0.766 183s 0.373 -0.240 -0.928 183s 0.957 -0.275 -0.313 183s 183s x1: [3, 1, 1, 1] 183s -0.221 -0.619 0.553 183s 183s y: [4, 1, 1, 1] 183s -0.325 0.866 -0.447 -0.214 183s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.903 0.524 0.886 0.107 183s 183s x1: [4, 3, 3, 1] 183s 0.647 -0.280 0.824 -0.222 183s 0.725 0.597 -0.370 -0.537 183s -0.266 0.010 0.428 0.620 183s 183s y: [1, 3, 3, 1] 183s -0.024 183s -0.727 183s 0.692 183s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 3, 3, 1] 183s -0.981 183s 0.126 183s -0.803 183s 183s x1: [1, 4, 3, 1] 183s -0.284 183s -0.507 183s -0.193 183s -0.989 183s 183s y: [3, 4, 3, 1] 183s 0.278 -0.036 0.228 183s 0.498 -0.064 0.408 183s 0.189 -0.024 0.155 183s 0.970 -0.124 0.794 183s test-mul-mat0: iter:319/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.761 -0.658 183s -0.921 0.773 183s 0.351 -0.294 183s 183s x1: [2, 2, 1, 1] 183s 0.917 -0.097 183s 0.045 0.849 183s 183s y: [3, 2, 1, 1] 183s 0.761 -0.919 0.351 183s -0.525 0.615 -0.234 183s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.005 0.273 183s 0.616 -0.961 183s 0.122 -0.112 183s 183s x1: [2, 3, 3, 1] 183s 0.638 0.638 183s 0.878 -0.452 183s -0.033 0.667 183s 183s y: [3, 3, 3, 1] 183s 0.171 -0.219 0.006 183s -0.128 0.975 0.158 183s 0.182 -0.661 -0.079 183s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.492 -0.103 183s 0.225 -0.919 183s 183s x1: [2, 3, 3, 1] 183s -0.051 -0.595 183s -0.990 -0.665 183s -0.124 -0.010 183s 183s y: [2, 3, 3, 1] 183s 0.087 0.535 183s 0.555 0.388 183s 0.062 -0.019 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.333 0.793 183s 0.408 -0.651 183s 183s x1: [2, 2, 1, 1] 183s -0.998 0.988 183s 0.677 -0.663 183s 183s y: [2, 2, 1, 1] 183s 1.116 -1.050 183s -0.751 0.707 183s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.432 0.956 183s 0.739 -0.275 183s 183s x1: [2, 2, 3, 1] 183s -0.550 0.467 183s 0.387 -0.796 183s 183s y: [2, 2, 3, 1] 183s 0.209 -0.535 183s -0.594 0.505 183s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.425 -0.020 183s 183s x1: [2, 2, 3, 1] 183s -0.884 -0.662 183s -0.586 0.524 183s 183s y: [1, 2, 3, 1] 183s 0.389 183s 0.239 183s test-mul-mat0: iter:320/500 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.863 0.077 183s 183s x1: [2, 1, 1, 1] 183s -0.349 0.006 183s 183s y: [1, 1, 1, 1] 183s 0.302 183s testing: mul_mat, [3, 1, 4, 1] = [2, 3, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.252 0.678 183s -0.847 0.430 183s 0.411 -0.998 183s 183s x1: [2, 1, 4, 1] 183s 0.638 0.611 183s 183s y: [3, 1, 4, 1] 183s 0.576 -0.278 -0.348 183s testing: mul_mat, [2, 3, 4, 4] = [2, 2, 4, 4] * [2, 3, 4, 4] 183s x0: [2, 2, 4, 4] 183s 0.645 -0.548 183s 0.200 0.706 183s 183s x1: [2, 3, 4, 4] 183s -0.076 -0.396 183s 0.086 -0.837 183s 0.883 0.166 183s 183s y: [2, 3, 4, 4] 183s 0.168 -0.294 183s 0.514 -0.574 183s 0.478 0.293 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.459 -0.367 183s 0.370 -0.147 183s 183s x1: [2, 2, 1, 1] 183s 0.128 0.079 183s 0.366 0.832 183s 183s y: [2, 2, 1, 1] 183s -0.088 0.036 183s -0.473 0.013 183s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 3, 4, 1] 183s -0.459 -0.044 183s -0.324 -0.819 183s 0.045 -0.827 183s 183s x1: [2, 2, 4, 1] 183s 0.305 -0.473 183s 0.946 -0.244 183s 183s y: [3, 2, 4, 1] 183s -0.119 0.289 0.405 183s -0.423 -0.107 0.244 183s testing: mul_mat, [1, 2, 4, 4] = [3, 1, 4, 4] * [3, 2, 4, 4] 183s x0: [3, 1, 4, 4] 183s -0.752 0.163 0.580 183s 183s x1: [3, 2, 4, 4] 183s -0.748 -0.379 0.222 183s 0.345 0.707 0.088 183s 183s y: [1, 2, 4, 4] 183s 0.630 183s -0.093 183s test-mul-mat0: iter:321/500 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.507 0.710 0.514 -0.755 183s 0.415 0.387 -0.437 -0.004 183s 183s x1: [4, 3, 1, 1] 183s -0.171 0.358 0.205 -0.577 183s 0.939 -0.281 0.085 -0.200 183s 0.625 -0.813 0.909 0.694 183s 183s y: [2, 3, 1, 1] 183s 0.882 -0.019 183s -0.481 0.244 183s -0.951 -0.456 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.397 -0.863 -0.335 0.808 183s -0.440 -0.397 -0.472 0.645 183s 183s x1: [4, 2, 1, 1] 183s -0.698 -0.647 -0.330 0.636 183s 0.041 0.201 -0.954 0.307 183s 183s y: [2, 2, 1, 1] 183s 1.461 1.130 183s 0.378 0.551 183s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 183s x0: [4, 4, 1, 2] 183s 0.541 -0.993 0.839 0.588 183s 0.314 0.145 -0.809 0.451 183s 0.810 0.999 -0.989 -0.587 183s -0.473 0.655 0.817 -0.320 183s 183s x1: [4, 2, 1, 2] 183s 0.403 -0.847 0.832 0.312 183s 0.847 -0.297 0.805 0.557 183s 183s y: [4, 2, 1, 2] 183s 1.942 -0.529 -1.526 -0.166 183s 1.757 -0.177 -0.735 -0.116 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.031 0.053 -0.890 0.134 183s -0.322 0.084 0.094 0.010 183s -0.251 0.733 0.330 0.892 183s 183s x1: [4, 4, 1, 1] 183s -0.431 0.501 0.044 0.883 183s -0.353 0.235 0.334 -0.543 183s 0.235 0.344 -0.130 0.762 183s -0.000 -0.314 -0.558 -0.513 183s 183s y: [3, 4, 1, 1] 183s 0.119 0.194 1.279 183s -0.347 0.160 -0.114 183s 0.229 -0.051 0.831 183s 0.411 -0.084 -0.871 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.743 -0.490 0.522 183s -0.977 -0.227 -0.793 183s -0.042 0.154 -0.728 183s -0.815 0.188 0.255 183s 183s x1: [3, 4, 1, 1] 183s -0.203 -0.533 -0.539 183s -0.702 0.511 -0.655 183s -0.055 -0.254 0.678 183s 0.401 0.981 0.023 183s 183s y: [4, 4, 1, 1] 183s -0.171 0.747 0.319 -0.072 183s -1.114 1.090 0.585 0.501 183s 0.437 -0.426 -0.530 0.170 183s -0.171 -0.633 0.117 -0.137 183s testing: mul_mat, [4, 3, 1, 2] = [4, 4, 1, 2] * [4, 3, 1, 2] 183s x0: [4, 4, 1, 2] 183s -0.838 0.977 -0.744 -0.615 183s 0.705 0.671 -0.863 -0.437 183s -0.584 -0.314 0.720 0.673 183s -0.631 0.344 -0.266 -0.070 183s 183s x1: [4, 3, 1, 2] 183s -0.683 png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.d0.366 -0.415 0.451 183s -0.624 -0.522 -0.752 -0.157 183s -0.061 -0.454 -0.646 0.284 183s 183s y: [4, 3, 1, 2] 183s 0.962 -0.075 0.289 0.636 183s 0.669 -0.072 -0.119 0.425 183s -0.087 0.086 -0.096 0.034 183s test-mul-mat0: iter:322/500 183s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.455 183s 0.834 183s -0.719 183s 183s x1: [1, 1, 1, 1] 183s 0.578 183s 183s y: [3, 1, 1, 1] 183s 0.263 0.482 -0.415 183s testing: mul_mat, [1, 3, 4, 1] = [1, 1, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 1, 4, 1] 183s 0.021 183s 183s x1: [1, 3, 4, 1] 183s 0.175 183s -0.432 183s -0.334 183s 183s y: [1, 3, 4, 1] 183s 0.004 183s -0.009 183s -0.007 183s testing: mul_mat, [2, 1, 4, 4] = [1, 2, 4, 4] * [1, 1, 4, 4] 183s x0: [1, 2, 4, 4] 183s 0.975 183s -0.977 183s 183s x1: [1, 1, 4, 4] 183s -0.598 183s 183s y: [2, 1, 4, 4] 183s -0.582 0.584 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.283 183s 0.118 183s 0.957 183s 183s x1: [1, 2, 1, 1] 183s 0.496 183s 0.248 183s 183s y: [3, 2, 1, 1] 183s -0.140 0.058 0.474 183s -0.070 0.029 0.237 183s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.132 -0.546 -0.730 183s 183s x1: [3, 1, 4, 1] 183s 0.056 -0.683 0.022 183s 183s y: [1, 1, 4, 1] 183s 0.364 183s testing: mul_mat, [1, 3, 4, 4] = [1, 1, 4, 4] * [1, 3, 4, 4] 183s x0: [1, 1, 4, 4] 183s 0.634 183s 183s x1: [1, 3, 4, 4] 183s -0.729 183s 0.782 183s -0.168 183s 183s y: [1, 3, 4, 4] 183s -0.462 183s 0.496 183s -0.106 183s test-mul-mat0: iter:323/500 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.421 -0.837 -0.266 183s -0.107 -0.250 -0.181 183s 0.310 -0.616 0.336 183s 183s x1: [3, 1, 1, 1] 183s -0.567 -0.321 0.108 183s 183s y: [3, 1, 1, 1] 183s 0.001 0.121 0.058 183s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.191 -0.707 -0.108 183s 0.755 0.514 0.136 183s -0.601 -0.085 0.179 183s 183s x1: [3, 3, 4, 1] 183s 0.127 0.335 0.895 183s -0.653 0.691 0.800 183s -0.892 0.595 0.826 183s 183s y: [3, 3, 4, 1] 183s -0.358 0.390 0.055 183s -0.450 -0.029 0.477 183s -0.339 -0.255 0.634 183s testing: mul_mat, [2, 3, 4, 4] = [3, 2, 4, 4] * [3, 3, 4, 4] 183s x0: [3, 2, 4, 4] 183s 0.499 0.641 -0.379 183s 0.563 0.357 0.599 183s 183s x1: [3, 3, 4, 4] 183s -0.495 0.630 -0.258 183s 0.685 -0.870 -0.418 183s -0.870 0.163 -0.975 183s 183s y: [2, 3, 4, 4] 183s 0.255 -0.208 183s -0.058 -0.176 183s 0.040 -1.016 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.351 0.855 -0.039 183s -0.492 -0.673 -0.020 183s 183s x1: [3, 2, 1, 1] 183s -0.696 0.735 0.421 183s 0.683 -0.280 -0.220 183s 183s y: [2, 2, 1, 1] 183s 0.856 -0.161 183s -0.471 -0.143 183s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.445 -0.822 183s 0.217 -0.391 183s 0.629 0.317 183s -0.118 -0.835 183s 183s x1: [2, 3, 4, 1] 183s -0.565 0.987 183s -0.142 0.171 183s -0.490 -0.008 183s 183s y: [4, 3, 4, 1] 183s -0.559 -0.509 -0.043 -0.757 183s -0.077 -0.098 -0.035 -0.126 183s 0.225 -0.103 -0.311 0.064 183s testing: mul_mat, [1, 2, 4, 4] = [4, 1, 4, 4] * [4, 2, 4, 4] 183s x0: [4, 1, 4, 4] 183s 0.317 0.542 0.796 -0.290 183s 183s x1: [4, 2, 4, 4] 183s 0.053 0.896 -0.987 0.232 183s -0.495 0.329 0.397 0.972 183s 183s y: [1, 2, 4, 4] 183s -0.351 183s 0.056 183s test-mul-mat0: iter:324/500 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.568 -0.302 -0.226 0.868 183s 183s x1: [4, 3, 1, 1] 183s 0.223 0.438 -0.671 0.983 183s -0.677 0.009 -0.309 0.003 183s -0.590 -0.151 -0.597 0.333 183s 183s y: [1, 3, 1, 1] 183s 1.000 183s -0.315 183s 0.134 183s testing: mul_mat, [4, 1, 2, 1] = [4, 4, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 4, 2, 1] 183s 0.440 -0.474 0.179 0.689 183s 0.790 -0.598 0.127 -0.881 183s -0.615 0.450 0.128 0.077 183s -0.547 0.538 0.925 -0.143 183s 183s x1: [4, 1, 2, 1] 183s -0.785 -0.378 0.239 -0.636 183s 183s y: [4, 1, 2, 1] 183s -0.562 0.197 0.295 0.538 183s testing: mul_mat, [1, 4, 2, 4] = [4, 1, 2, 4] * [4, 4, 2, 4] 183s x0: [4, 1, 2, 4] 183s 0.544 -0.303 -0.548 -0.806 183s 183s x1: [4, 4, 2, 4] 183s 0.750 -0.938 -0.798 0.540 183s -0.536 0.328 0.659 -0.151 183s -0.222 -0.213 0.925 0.231 183s -0.675 0.851 -0.912 0.195 183s 183s y: [1, 4, 2, 4] 183s 0.694 183s -0.630 183s -0.750 183s -0.283 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.207 -0.375 -0.059 0.522 183s 0.119 0.093 0.805 0.266 183s -0.624 -0.871 0.628 0.950 183s -0.924 -0.806 0.216 0.602 183s 183s x1: [4, 1, 1, 1] 183s 0.734 0.695 -0.343 0.713 183s 183s y: [4, 1, 1, 1] 183s -0.020 0.065 -0.602 -0.884 183s testing: mul_mat, [2, 4, 2, 1] = [4, 2, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 2, 2, 1] 183s 0.758 -0.065 -0.429 0.105 183s 0.353 -0.190 0.964 -0.748 183s 183s x1ot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward: [4, 4, 2, 1] 183s 0.747 0.186 0.493 -0.701 183s 0.992 0.052 0.218 0.029 183s -0.705 0.681 -0.237 0.990 183s -0.662 -0.524 -0.819 0.131 183s 183s y: [2, 4, 2, 1] 183s 0.269 1.228 183s 0.658 0.529 183s -0.373 -1.348 183s -0.102 -1.022 183s testing: mul_mat, [1, 4, 2, 4] = [2, 1, 2, 4] * [2, 4, 2, 4] 183s x0: [2, 1, 2, 4] 183s 0.349 -0.437 183s 183s x1: [2, 4, 2, 4] 183s -0.072 0.016 183s 0.482 -0.736 183s 0.429 -0.926 183s 0.404 0.470 183s 183s y: [1, 4, 2, 4] 183s -0.032 183s 0.490 183s 0.555 183s -0.065 183s test-mul-mat0: iter:325/500 183s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.397 0.363 183s 0.637 0.166 183s -0.211 0.900 183s 183s x1: [2, 4, 1, 1] 183s 0.525 0.952 183s 0.856 -0.995 183s 0.724 -0.551 183s -0.307 0.700 183s 183s y: [3, 4, 1, 1] 183s 0.138 0.493 0.746 183s -0.701 0.380 -1.076 183s -0.487 0.369 -0.649 183s 0.376 -0.079 0.695 183s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.291 0.199 183s 0.700 -0.543 183s 0.988 0.601 183s 183s x1: [2, 3, 4, 1] 183s 0.852 0.856 183s 0.144 0.438 183s -0.795 0.698 183s 183s y: [3, 3, 4, 1] 183s 0.418 0.132 1.356 183s 0.129 -0.137 0.405 183s -0.092 -0.936 -0.366 183s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.785 0.806 183s 0.061 -0.365 183s -0.265 0.935 183s 183s x1: [2, 3, 4, 1] 183s -0.824 -0.076 183s 0.010 0.952 183s 0.611 -0.154 183s 183s y: [3, 3, 4, 1] 183s -0.708 -0.023 0.147 183s 0.775 -0.347 0.887 183s 0.356 0.094 -0.306 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.725 -0.641 183s -0.265 0.469 183s -0.368 -0.433 183s 183s x1: [2, 3, 1, 1] 183s -0.627 0.534 183s -0.938 -0.029 183s 0.940 -0.071 183s 183s y: [3, 3, 1, 1] 183s -0.797 0.417 -0.000 183s -0.661 0.235 0.358 183s 0.727 -0.283 -0.315 183s testing: mul_mat, [3, 2, 4, 1] = [3, 3, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.117 0.476 -0.353 183s -0.848 0.622 -0.250 183s -0.570 -0.003 -0.491 183s 183s x1: [3, 2, 4, 1] 183s 0.170 0.128 0.512 183s 0.933 -0.602 0.304 183s 183s y: [3, 2, 4, 1] 183s -0.100 -0.193 -0.349 183s -0.285 -1.242 -0.679 183s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.172 0.204 0.542 183s -0.909 0.581 -0.420 183s 0.653 0.910 0.288 183s 183s x1: [3, 3, 4, 1] 183s 0.100 0.124 0.575 183s 0.609 0.704 -0.278 183s 0.422 -0.005 -0.561 183s 183s y: [3, 3, 4, 1] 183s 0.320 -0.261 0.344 183s -0.112 -0.028 0.958 183s -0.377 -0.152 0.109 183s test-mul-mat0: iter:326/500 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.401 -0.973 0.131 -0.193 183s 183s x1: [4, 1, 1, 1] 183s -0.405 -0.815 -0.213 0.393 183s 183s y: [1, 1, 1, 1] 183s 0.528 183s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 1, 2, 1] 183s -0.302 -0.684 -0.705 -0.072 183s 183s x1: [4, 1, 2, 1] 183s 0.546 0.655 -0.994 0.156 183s 183s y: [1, 1, 2, 1] 183s 0.077 183s testing: mul_mat, [4, 1, 2, 3] = [4, 4, 2, 3] * [4, 1, 2, 3] 183s x0: [4, 4, 2, 3] 183s -0.208 -0.956 0.010 0.137 183s -0.146 0.034 0.113 0.459 183s 0.054 0.301 0.145 0.670 183s 0.674 0.618 0.679 0.246 183s 183s x1: [4, 1, 2, 3] 183s 0.976 -0.396 -0.980 -0.812 183s 183s y: [4, 1, 2, 3] 183s 0.054 -0.639 -0.752 -0.451 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.349 0.453 -0.904 -0.944 183s 0.488 0.330 -0.201 0.944 183s 183s x1: [4, 4, 1, 1] 183s -0.310 0.251 -0.722 -0.230 183s -0.138 -0.900 -0.286 0.540 183s -0.203 -0.805 -0.095 -0.951 183s -0.541 -0.552 0.568 0.455 183s 183s y: [2, 4, 1, 1] 183s 1.092 -0.140 183s -0.610 0.203 183s 0.690 -1.243 183s -1.004 -0.131 183s testing: mul_mat, [3, 4, 2, 1] = [2, 3, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 3, 2, 1] 183s -0.975 0.593 183s 0.902 0.357 183s -0.036 0.307 183s 183s x1: [2, 4, 2, 1] 183s 0.902 -0.604 183s 0.140 0.074 183s 0.959 -0.199 183s 0.765 0.210 183s 183s y: [3, 4, 2, 1] 183s -1.238 0.598 -0.218 183s -0.092 0.153 0.018 183s -1.053 0.794 -0.096 183s -0.621 0.765 0.037 183s testing: mul_mat, [1, 2, 2, 3] = [3, 1, 2, 3] * [3, 2, 2, 3] 183s x0: [3, 1, 2, 3] 183s -0.860 -0.696 -0.721 183s 183s x1: [3, 2, 2, 3] 183s 0.631 -0.116 0.461 183s -0.973 -0.976 -0.465 183s 183s y: [1, 2, 2, 3] 183s -0.794 183s 1.851 183s test-mul-mat0: iter:327/500 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.242 183s 183s x1: [1, 4, 1, 1] 183s 0.532 183s 0.135 183s 0.313 183s 0.660 183s 183s y: [1, 4, 1, 1] 183s -0.129 183s -0.033 183s -0.076 183s -0.160 183s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.017 183s 0.291 183s 183s x1: [1, 1, 4, 1] 183s 0.814 183s 183s y: [2, 1, 4, 1] 183s 0.014 0.237 183s testing: mul_mat, [2, 2, 4, 3] = [1, 2, 4, 3] * [1, 2, 4, 3] 183s x0: [1, 2, 4, 3] 183s -0.552 183s 0.681 183s 183s x1: [1, 2, 4, 3] 183s -0.777 183s -0.950 183s 183s y: [2, 2, 4, 3] 183s 0.429 -0.529 183s 0.524 -0.647 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.293 183s 0.721 183s 0.113 183s 183s x1: [1, 2, 1, 1] 183s -0.062 183s 0.853 183s 183s y: [3, 2, 1, 1] 183s 0.018 -0.045 -0.007 183s -0.250 0.615 0.096 183s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.208 -0.488 -0.812 183s 183s x1: [3, 1, 4, 1] 183s 0.767 -0.831 -0.206 183s 183s y: [1, 1, 4, 1] 183s 0.733 183s testing: mul_mat, [1, 3, 4, 3] = [1, 1, 4, 3] * [1, 3, 4, 3] 183s x0: [1, 1, 4, 3] 183s 0.642 183s 183s x1: [1, 3, 4, 3] 183s 0.173 183s 0.751 183s -0.791 183s 183s y: [1, 3, 4, 3] 183s 0.111 183s 0.482 183s -0.508 183s test-mul-mat0: iter:328/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.276 -0.845 -0.035 183s -0.521 0.977 -0.040 183s 183s x1: [3, 2, 1, 1] 183s 0.202 -0.729 0.578 183s 0.909 0.825 -0.457 183s 183s y: [2, 2, 1, 1] 183s 0.652 -0.841 183s -0.429 0.351 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.556 -0.953 0.523 183s 183s x1: [3, 2, 1, 1] 183s 0.154 0.121 -0.398 183s 0.469 -0.620 -0.387 183s 183s y: [1, 2, 1, 1] 183s -0.238 183s 0.650 183s testing: mul_mat, [2, 1, 1, 4] = [3, 2, 1, 4] * [3, 1, 1, 4] 183s x0: [3, 2, 1, 4] 183s -0.841 0.097 0.987 183s -0.685 -0.938 -0.534 183s 183s x1: [3, 1, 1, 4] 183s 0.510 -0.461 0.528 183s 183s y: [2, 1, 1, 4] 183s 0.047 -0.199 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.556 -0.064 0.501 183s -0.126 0.881 -0.537 183s 0.415 0.050 0.582 183s -0.241 0.958 -0.411 183s 183s x1: [3, 2, 1, 1] 183s 0.832 -0.331 0.529 183s -0.282 -0.715 0.776 183s 183s y: [4, 2, 1, 1] 183s -0.177 -0.681 0.637 -0.735 183s 0.591 -1.011 0.299 -0.936 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.774 -0.105 -0.771 -0.489 183s 183s x1: [4, 3, 1, 1] 183s -0.745 0.819 0.593 0.604 183s -0.897 -0.880 -0.538 -0.074 183s -0.825 0.679 0.557 0.366 183s 183s y: [1, 3, 1, 1] 183s -0.262 183s 1.238 183s -0.041 183s testing: mul_mat, [3, 4, 1, 4] = [1, 3, 1, 4] * [1, 4, 1, 4] 183s x0: [1, 3, 1, 4] 183s 0.085 183s 0.943 183s -0.243 183s 183s x1: [1, 4, 1, 4] 183s -0.329 183s 0.657 183s -0.045 183s 0.545 183s 183s y: [3, 4, 1, 4] 183s -0.028 -0.310 0.080 183s 0.056 0.620 -0.160 183s -0.004 -0.043 0.011 183s 0.046 0.513 -0.132 183s test-mul-mat0: iter:329/500 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.831 0.889 183s 183s x1: [2, 1, 1, 1] 183s -0.883 0.408 183s 183s y: [1, 1, 1, 1] 183s -0.371 183s testing: mul_mat, [1, 1, 4, 1] = [2, 1, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 1, 4, 1] 183s -0.216 -0.612 183s 183s x1: [2, 1, 4, 1] 183s -0.350 -0.405 183s 183s y: [1, 1, 4, 1] 183s 0.324 183s testing: mul_mat, [3, 1, 4, 1] = [2, 3, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.357 0.235 183s 0.185 0.188 183s 0.124 0.835 183s 183s x1: [2, 1, 4, 1] 183s 0.425 -0.229 183s 183s y: [3, 1, 4, 1] 183s 0.098 0.036 -0.138 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.214 0.684 183s 183s x1: [2, 3, 1, 1] 183s -0.712 -0.382 183s -0.378 0.152 183s 0.858 -0.551 183s 183s y: [1, 3, 1, 1] 183s -0.414 183s 0.023 183s -0.193 183s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 4, 4, 1] 183s 0.688 183s -0.306 183s -0.952 183s 0.559 183s 183s x1: [1, 2, 4, 1] 183s 0.826 183s -0.598 183s 183s y: [4, 2, 4, 1] 183s 0.569 -0.253 -0.787 0.462 183s -0.411 0.183 0.569 -0.334 183s testing: mul_mat, [3, 1, 4, 1] = [4, 3, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 3, 4, 1] 183s 0.449 0.614 -0.103 -0.280 183s 0.024 0.996 -0.146 -0.630 183s -0.429 0.788 0.395 0.289 183s 183s x1: [4, 1, 4, 1] 183s 0.047 -0.700 -0.414 0.688 183s 183s y: [3, 1, 4, 1] 183s -0.559 -1.069 -0.536 183s test-mul-mat0: iter:330/500 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.461 0.522 0.557 -0.394 183s 0.352 -0.225 -0.529 0.552 183s 183s x1: [4, 3, 1, 1] 183s -0.640 0.610 0.120 0.353 183s 0.394 0.611 -0.766 -0.421 183s 0.566 0.698 0.213 -0.454 183s 183s y: [2, 3, 1, 1] 183s -0.049 -0.231 183s 0.240 0.174 183s 0.923 -0.321 183s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.607 -0.910 0.117 0.168 183s -0.439 -0.330 -0.549 -0.939 183s -0.071 -0.423 -0.950 -0.009 183s 183s x1: [4, 2, 3, 1] 183s 0.283 0.499 -0.740 -0.874 183s 0.990 -0.939 -0.006 0.457 183s 183s y: [3, 2, 3, 1] 183s -0.859 0.938 0.480 183s 0.329 -0.550 0.328 183s testing: mul_mat, [1, 3, 3, 2] = [4, 1, 3, 2] * [4, 3, 3, 2] 183s x0: [4, 1, 3, 2] 183s -0.637 -0.510 0.719 0.200 183s 183s x1: [4, 3, 3, 2] 183s 0.114 0.847 -0.681 -0.957 183s -0.576 -0.631 0.034 0.995 183s 0.876 0.252 -0.033 0.200 183s 183s y: [1, 3, 3, 2] 183s -1.186 183s 0.913 183s -0.671 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.616 0.167 0.469 0.698 183s 0.447 0.839 0.775 -0.601 183s -0.361 0.439 0.465 -0.786 183s -0.126 -0.293 -0.882 -0.368 183s 183s x1: [4, 1, 1, 1] 183s -0.222 0.057 -0.626 0.470 183s 183s y: [4, 1, 1, 1] 183s 0.182 -0.819 -0.555 0.390 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s 0.675 -0.747 -0.887 0.281 183s 183s x1: [4, 4, 3, 1] 183s 0.794 -0.588 0.696 -0.620 183s -0.237 -0.729 0.998 0.166 183s -0.947 -0.465 0.944 0.110 183s -0.091 0.414 0.386 0.293 183s 183s y: [1, 4, 3, 1] 183s 0.183 183s -0.454 183s -1.099 183s -0.630 183s testing: mul_mat, [2, 4, 3, 2] = [1, 2, 3, 2] * [1, 4, 3, 2] 183s x0: [1, 2, 3, 2] 183s -0.628 183s 0.443 183s 183s x1: [1, 4, 3, 2] 183s 0.452 183s -0.819 183s -0.367 183s -0.141 183s 183s y: [2, 4, 3, 2] 183s -0.284 0.200 183s 0.515 -0.363 183s 0.231 -0.163 183s 0.089 -0.063 183s test-mul-mat0: iter:331/500 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.085 -0.923 0.631 -0.642 183s 183s x1: [4, 3, 1, 1] 183s 0.121 0.594 -0.668 -0.633 183s -0.618 0.664 0.121 0.394 183s 0.245 0.549 -0.340 -0.061 183s 183s y: [1, 3, 1, 1] 183s -0.573 183s -0.737 183s -0.703 183s testing: mul_mat, [3, 1, 4, 1] = [4, 3, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 3, 4, 1] 183s -0.055 -0.236 0.913 -0.934 183s -0.842 0.158 0.615 -0.182 183s -0.903 -0.125 0.733 -0.826 183s 183s x1: [4, 1, 4, 1] 183s -0.890 0.743 0.589 -0.300 183s 183s y: [3, 1, 4, 1] 183s 0.691 1.284 1.390 183s testing: mul_mat, [3, 3, 4, 2] = [4, 3, 4, 2] * [4, 3, 4, 2] 183s x0: [4, 3, 4, 2] 183s -0.980 0.137 0.466 0.634 183s 0.667 -0.192 0.639 0.976 183s -0.533 0.231 0.351 -0.909 183s 183s x1: [4, 3, 4, 2] 183s -0.247 0.894 -0.380 0.660 183s 0.181 -0.332 0.185 0.158 183s -0.827 0.497 0.052 -0.743 183s 183s y: [3, 3, 4, 2] 183s 0.606 0.065 -0.395 183s -0.037 0.457 -0.252 183s 0.432 -1.338 1.248 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.403 0.930 0.165 -0.003 183s 0.207 0.185 0.108 0.487 183s 183s x1: [4, 3, 1, 1] 183s 0.412 -0.662 -0.945 0.049 183s -0.706 -0.925 0.287 0.782 183s -0.321 0.786 -0.027 -0.186 183s 183s y: [2, 3, 1, 1] 183s -0.606 -0.115 183s -1.100 0.094 183s 0.598 -0.015 183s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 1, 4, 1] 183s -0.466 -0.197 183s 183s x1: [2, 4, 4, 1] 183s 0.315 0.169 183s 0.443 -0.944 183s 0.779 -0.566 183s 0.082 -0.076 183s 183s y: [1, 4, 4, 1] 183s -0.180 183s -0.021 183s -0.251 183s -0.023 183s testing: mul_mat, [4, 2, 4, 2] = [1, 4, 4, 2] * [1, 2, 4, 2] 183s x0: [1, 4, 4, 2] 183s -0.413 183s -0.160 183s 0.448 183s 0.678 183s 183s x1: [1, 2, 4, 2] 183s -0.824 183s 0.298 183s 183s y: [4, 2, 4, 2] 183s 0.340 0.132 -0.369 -0.559 183s -0.123 -0.048 0.134 0.202 183s test-mul-mat0: iter:332/500 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.586 0.288 183s 0.471 0.357 183s 183s x1: [2, 2, 1, 1] 183s -0.552 0.146 183s 0.310 0.949 183s 183s y: [2, 2, 1, 1] 183s -0.281 -0.207 183s 0.455 0.484 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.888 0.188 183s -0.010 0.252 183s 0.188 0.419 183s -0.063 -0.728 183s 183s x1: [2, 2, 1, 1] 183s -0.197 0.913 183s -0.714 0.879 183s 183s y: [4, 2, 1, 1] 183s 0.346 0.232 0.346 -0.652 183s 0.800 0.228 0.235 -0.595 183s testing: mul_mat, [1, 4, 1, 4] = [2, 1, 1, 4] * [2, 4, 1, 4] 183s x0: [2, 1, 1, 4] 183s -0.583 0.475 183s 183s x1: [2, 4, 1, 4] 183s -0.644 0.607 183s -0.235 -0.228 183s -0.137 -0.903 183s -0.364 0.798 183s 183s y: [1, 4, 1, 4] 183s 0.664 183s 0.029 183s -0.349 183s 0.591 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.661 -0.450 183s 0.771 0.080 183s 183s x1: [2, 1, 1, 1] 183s 0.068 0.858 183s 183s y: [2, 1, 1, 1] 183s -0.432 0.121 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.671 0.183 183s -0.736 -0.750 183s -0.130 -0.054 183s -0.401 0.040 183s 183s x1: [2, 2, 1, 1] 183s 0.041 0.674 183s 0.734 -0.403 183s 183s y: [4, 2, 1, 1] 183s 0.151 -0.536 -0.042 0.010 183s 0.419 -0.238 -0.073 -0.311 183s testing: mul_mat, [4, 2, 1, 4] = [4, 4, 1, 4] * [4, 2, 1, 4] 183s x0: [4, 4, 1, 4] 183s -0.380 0.732 -0.174 -0.214 183s -0.100 -0.604 -0.575 -0.691 183s 0.162 -0.597 -0.684 -0.307 183s 0.450 -0.905 0.485 -0.774 183s 183s x1: [4, 2, 1, 4] 183s 0.902 0.648 -0.497 -0.682 183s 0.123 -0.171 -0.168 0.234 183s 183s y: [4, 2, 1, 4] 183s 0.365 0.275 0.308 0.106 183s -0.193 0.025 0.164 -0.053 183s test-mul-mat0: iter:333/500 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.763 0.717 0.522 0.439 183s 0.108 -0.105 0.470 -0.383 183s 183s x1: [4, 4, 1, 1] 183s -0.273 -0.224 -0.033 0.689 183s -0.352 -0.045 0.437 -0.171 183s 0.634 0.411 -0.647 0.890 183s -0.214 -0.151 0.280 -0.594 183s 183s y: [2, 4, 1, 1] 183s 0.333 -0.285 183s 0.389 0.238 183s -0.137 -0.620 183s -0.059 0.352 183s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 2, 3, 1] 183s 0.638 -0.728 -0.579 -0.467 183s 0.742 0.037 0.484 -0.029 183s 183s x1: [4, 2, 3, 1] 183s -0.049 0.229 0.406 0.141 183s -0.552 0.871 0.868 0.224 183s 183s y: [2, 2, 3, 1] 183s -0.498 0.164 183s -1.593 0.036 183s testing: mul_mat, [4, 2, 3, 4] = [4, 4, 3, 4] * [4, 2, 3, 4] 183s x0: [4, 4, 3, 4] 183s 0.221 -0.913 -0.155 0.117 183s 0.433 -0.022 -0.978 -0.378 183s -0.901 0.878 0.758 0.655 183s 0.249 -0.635 0.408 0.196 183s 183s x1: [4, 2, 3, 4] 183s -0.363 -0.828 -0.715 -0.766 183s -0.713 0.118 0.547 0.925 183s 183s y: [4, 2, 3, 4] 183s 0.697 0.851 -1.444 -0.007 183s -0.242 -1.195 1.767 0.152 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.319 0.670 0.175 -0.657 183s 183s x1: [4, 4, 1, 1] 183s 0.213 0.933 -0.194 0.976 183s -0.756 0.537 -0.053 -0.907 183s 0.093 -0.906 -0.507 0.178 183s -0.175 0.172 -0.572 -0.243 183s 183s y: [1, 4, 1, 1] 183s 0.018 183s 0.706 183s -0.784 183s 0.119 183s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 4, 3, 1] 183s 0.614 183s 0.257 183s -0.339 183s -0.439 183s 183s x1: [1, 4, 3, 1] 183s 0.200 183s 0.231 183s -0.102 183s -0.033 183s 183s y: [4, 4, 3, 1] 183s 0.123 0.051 -0.068 -0.088 183s 0.142 0.059 -0.078 -0.101 183s -0.063 -0.026 0.035 0.045 183s -0.020 -0.009 0.011 0.015 183s testing: mul_mat, [4, 1, 3, 4] = [4, 4, 3, 4] * [4, 1, 3, 4] 183s x0: [4, 4, 3, 4] 183s 0.124 -0.263 -0.032 -0.735 183s 0.015 0.529 -0.945 0.068 183s 0.933 0.205 -0.027 -0.828 183s 0.287 0.115 0.423 0.155 183s 183s x1: [4, 1, 3, 4] 183s -0.762 -0.707 -0.321 0.908 183s 183s y: [4, 1, 3, 4] 183s -0.566 -0.020 -1.599 -0.295 183s test-mul-mat0: iter:334/500 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.193 0.395 0.690 183s 0.506 0.923 -0.784 183s 183s x1: [3, 4, 1, 1] 183s -0.353 -0.821 0.623 183s 0.670 0.575 0.283 183s 0.541 -0.781 -0.569 183s 0.252 -0.560 -0.418 183s 183s y: [2, 4, 1, 1] 183s 0.038 -1.425 183s 0.551 0.648 183s -0.596 -0.001 183s -0.461 -0.061 183s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 3, 3, 1] 183s 0.709 -0.885 0.220 183s -0.851 -0.303 0.318 183s 0.342 -0.909 0.008 183s 183s x1: [3, 2, 3, 1] 183s 0.798 0.874 -0.944 183s -0.928 -0.188 -0.953 183s 183s y: [3, 2, 3, 1] 183s -0.416 -1.244 -0.528 183s -0.702 0.543 -0.155 183s testing: mul_mat, [4, 3, 3, 2] = [3, 4, 3, 2] * [3, 3, 3, 2] 183s x0: [3, 4, 3, 2] 183s 0.653 -0.304 -0.488 183s 0.392 0.114 0.839 183s 0.248 0.273 0.851 183s -0.095 0.143 -0.743 183s 183s x1: [3, 3, 3, 2] 183s 0.742 -0.316 0.801 183s 0.710 -0.607 0.916 183s -0.070 -0.457 -0.387 183s 183s y: [4, 3, 3, 2] 183s 0.189 0.927 0.779 -0.711 183s 0.201 0.978 0.790 -0.835 183s 0.282 -0.404 -0.472 0.229 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.194 -0.621 0.925 183s 183s x1: [3, 4, 1, 1] 183s -0.483 0.529 0.680 183s 0.877 -0.524 -0.736 183s -0.773 0.462 -0.599 183s 0.870 -0.948 0.701 183s 183s y: [1, 4, 1, 1] 183s 0.207 183s -0.185 183s -0.991 183s 1.406 183s testing: mul_mat, [1, 3, 3, 1] = [1, 1, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 1, 3, 1] 183s 0.792 183s 183s x1: [1, 3, 3, 1] 183s 0.936 183s 0.962 183s -0.490 183s 183s y: [1, 3, 3, 1] 183s 0.742 183s 0.762 183s -0.388 183s testing: mul_mat, [2, 1, 3, 2] = [1, 2, 3, 2] * [1, 1, 3, 2] 183s x0: [1, 2, 3, 2] 183s -0.212 183s -0.764 183s 183s x1: [1, 1, 3, 2] 183s 0.578 183s 183s y: [2, 1, 3, 2] 183s -0.123 -0.441 183s test-mul-mat0: iter:335/500 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.948 -0.970 -0.654 183s -0.939 0.613 0.134 183s -0.703 -0.343 -0.677 183s 183s x1: [3, 3, 1, 1] 183s 0.124 0.789 0.554 183s 0.894 -0.695 0.346 183s -0.021 0.920 -0.076 183s 183s y: [3, 3, 1, 1] 183s -1.245 0.442 -0.733 183s -0.400 -1.218 -0.624 183s -0.822 0.574 -0.249 183s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.010 0.967 -0.126 183s -0.143 -0.666 -0.352 183s 0.828 -0.509 0.976 183s 183s x1: [3, 3, 4, 1] 183s -0.597 0.753 0.213 183s -0.903 -0.956 -0.519 183s 0.717 -0.729 0.713 183s 183s y: [3, 3, 4, 1] 183s 0.695 -0.491 -0.670 183s -0.868 0.948 -0.767 183s -0.787 0.132 1.660 183s testing: mul_mat, [3, 3, 4, 4] = [3, 3, 4, 4] * [3, 3, 4, 4] 183s x0: [3, 3, 4, 4] 183s 0.612 0.766 -0.100 183s 0.090 -0.241 -0.111 183s 0.124 -0.757 0.102 183s 183s x1: [3, 3, 4, 4] 183s -0.812 0.243 0.157 183s -0.836 0.935 -0.889 183s -0.185 -0.531 0.447 183s 183s y: [3, 3, 4, 4] 183s -0.326 -0.149 -0.268 183s 0.294 -0.202 -0.902 183s -0.565 0.062 0.425 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.043 0.140 0.149 183s 0.341 -0.312 -0.313 183s -0.095 -0.232 -0.285 183s 183s x1: [3, 3, 1, 1] 183s 0.133 1.000 -0.130 183s 0.332 0.474 -0.132 183s 0.072 -0.964 0.101 183s 183s y: [3, 3, 1, 1] 183s 0.126 -0.226 -0.207 183s 0.061 0.007 -0.104 183s -0.117 0.294 0.188 183s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.538 -0.076 -0.442 183s 0.596 0.179 0.912 183s 183s x1: [3, 3, 4, 1] 183s 0.172 0.538 -0.298 183s 0.118 0.196 0.386 183s -0.632 -0.681 0.423 183s 183s y: [2, 3, 4, 1] 183s -0.002 -0.073 183s -0.249 0.458 183s 0.204 -0.112 183s testing: mul_mat, [1, 3, 4, 4] = [2, 1, 4, 4] * [2, 3, 4, 4] 183s x0: [2, 1, 4, 4] 183s 0.279 0.853 183s 183s x1: [2, 3, 4, 4] 183s 0.528 0.377 183s 0.969 0.643 183s 0.472 0.246 183s 183s y: [1, 3, 4, 4] 183s 0.469 183s 0.818 183s 0.341 183s test-mul-mat0: iter:336/500 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.858 -0.644 0.706 -0.305 183s 0.529 0.602 -0.308 0.889 183s 183s x1: [4, 1, 1, 1] 183s 0.076 -0.327 0.189 0.605 183s 183s y: [2, 1, 1, 1] 183s 0.095 0.323 183s testing: mul_mat, [1, 2, 3, 1] = [4, 1, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.907 -0.659 -0.687 0.695 183s 183s x1: [4, 2, 3, 1] 183s -0.342 0.215 -0.736 -0.648 183s 0.189 -0.117 -0.997 0.781 183s 183s y: [1, 2, 3, 1] 183s 0.224 183s 1.134 183s testing: mul_mat, [1, 1, 3, 4] = [4, 1, 3, 4] * [4, 1, 3, 4] 183s x0: [4, 1, 3, 4] 183s 0.416 0.204 -0.808 0.441 183s 183s x1: [4, 1, 3, 4] 183s -0.292 0.218 -0.379 0.004 183s 183s y: [1, 1, 3, 4] 183s 0.231 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.368 -0.134 -0.845 0.846 183s 0.758 -0.040 0.011 -0.289 183s -0.706 0.618 -0.705 0.816 183s 0.641 -0.022 0.542 -0.650 183s 183s x1: [4, 1, 1, 1] 183s -0.482 -0.094 0.871 0.549 183s 183s y: [4, 1, 1, 1] 183s -0.436 -0.511 0.116 -0.192 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s 0.413 -0.649 -0.592 0.118 183s 183s x1: [4, 4, 3, 1] 183s -0.527 0.920 0.216 -0.464 183s -0.766 -0.514 -0.441 -0.646 183s -0.177 0.207 -0.129 0.730 183s 0.078 -0.580 -0.436 -0.554 183s 183s y: [1, 4, 3, 1] 183s -0.998 183s 0.202 183s -0.046 183s 0.602 183s testing: mul_mat, [2, 4, 3, 4] = [1, 2, 3, 4] * [1, 4, 3, 4] 183s x0: [1, 2, 3, 4] 183s -0.080 183s 0.697 183s 183s x1: [1, 4, 3, 4] 183s 0.308 183s 0.217 183s -0.525 183s 0.436 183s 183s y: [2, 4, 3, 4] 183s -0.025 0.214 183s -0.017 0.151 183s 0.042 -0.366 183s -0.035 0.304 183s test-mul-mat0: iter:337/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.398 -0.872 0.316 183s -0.090 0.635 0.046 183s 183s x1: [3, 2, 1, 1] 183s -0.981 -0.181 0.485 183s -0.060 -0.483 -0.025 183s 183s y: [2, 2, 1, 1] 183s 0.701 -0.004 183s 0.437 -0.303 183s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.700 -0.853 -0.926 183s -0.183 0.121 0.319 183s 0.419 0.250 -0.365 183s 183s x1: [3, 2, 3, 1] 183s 0.835 0.212 0.662 183s -0.316 -0.926 0.638 183s 183s y: [3, 2, 3, 1] 183s -1.378 0.084 0.161 183s 0.420 0.149 -0.597 183s testing: mul_mat, [1, 3, 3, 4] = [3, 1, 3, 4] * [3, 3, 3, 4] 183s x0: [3, 1, 3, 4] 183s 0.407 0.164 -0.446 183s 183s x1: [3, 3, 3, 4] 183s -0.586 0.477 0.292 183s 0.548 0.777 0.438 183s 0.622 -0.406 -0.440 183s 183s y: [1, 3, 3, 4] 183s -0.291 183s 0.155 183s 0.383 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.199 0.963 -0.961 183s -0.769 -0.359 0.232 183s 0.001 0.199 0.397 183s 0.172 -0.759 0.789 183s 183s x1: [3, 1, 1, 1] 183s 0.724 -0.651 0.484 183s 183s y: [4, 1, 1, 1] 183s -1.236 -0.210 0.063 1.000 183s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.267 0.360 0.595 -0.269 183s 183s x1: [4, 3, 3, 1] 183s 0.860 0.120 0.157 -0.151 183s -0.526 0.885 -0.282 -0.762 183s -0.107 0.998 0.409 0.063 183s 183s y: [1, 3, 3, 1] 183s -0.053 183s 0.496 183s 0.614 183s testing: mul_mat, [1, 4, 3, 4] = [1, 1, 3, 4] * [1, 4, 3, 4] 183s x0: [1, 1, 3, 4] 183s -0.209 183s 183s x1: [1, 4, 3, 4] 183s -0.691 183s 0.929 183s -0.505 183s 0.501 183s 183s y: [1, 4, 3, 4] 183s 0.145 183s -0.195 183s 0.106 183s -0.105 183s test-mul-mat0: iter:338/500 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.417 0.694 -0.344 183s -0.408 0.715 0.446 183s -0.297 -0.090 -0.786 183s 0.645 -0.950 0.618 183s 183s x1: [3, 3, 1, 1] 183s 0.001 0.547 -0.494 183s 0.659 -0.349 0.975 183s 0.522 0.941 0.488 183s 183s y: [4, 3, 1, 1] 183s 0.549 0.170 0.339 -0.824 183s -0.853 -0.083 -0.931 1.359 183s 0.268 0.678 -0.623 -0.256 183s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.334 0.891 -0.020 183s 0.179 0.114 -0.450 183s 0.351 -0.011 -0.906 183s 183s x1: [3, 4, 3, 1] 183s 0.982 -0.544 -0.472 183s 0.835 -0.644 0.512 183s 0.511 0.677 0.064 183s -0.488 0.223 0.570 183s 183s y: [3, 4, 3, 1] 183s -0.803 0.326 0.778 183s -0.863 -0.155 -0.163 183s 0.431 0.140 0.114 183s 0.350 -0.318 -0.689 183s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.785 -0.032 -0.400 183s 0.952 -0.333 0.725 183s 183s x1: [3, 3, 3, 1] 183s -0.833 0.698 0.125 183s -0.917 -0.635 0.016 183s 0.063 0.543 -0.870 183s 183s y: [2, 3, 3, 1] 183s -0.726 -0.935 183s -0.706 -0.650 183s 0.380 -0.752 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.024 0.562 0.982 183s 0.557 -0.001 0.251 183s 183s x1: [3, 2, 1, 1] 183s 0.251 -0.668 0.038 183s 0.940 -0.160 -0.645 183s 183s y: [2, 2, 1, 1] 183s -0.344 0.150 183s -0.746 0.362 183s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.570 -0.869 183s 183s x1: [2, 3, 3, 1] 183s -0.216 -0.049 183s 0.851 -0.264 183s 0.617 0.577 183s 183s y: [1, 3, 3, 1] 183s 0.166 183s -0.256 183s -0.852 183s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.024 183s 0.778 183s 183s x1: [1, 2, 3, 1] 183s -0.753 183s -0.956 183s 183s y: [2, 2, 3, 1] 183s 0.018 -0.586 183s 0.022 -0.744 183s test-mul-mat0: iter:339/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.018 0.255 0.135 183s 0.229 0.685 0.267 183s 0.415 -0.045 0.753 183s -0.472 0.202 0.797 183s 183s x1: [3, 1, 1, 1] 183s -0.138 0.985 0.156 183s 183s y: [4, 1, 1, 1] 183s 0.275 0.685 0.016 0.389 183s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 2, 2, 1] 183s 0.798 -0.849 -0.007 183s -0.674 0.353 -0.210 183s 183s x1: [3, 4, 2, 1] 183s 0.203 -0.553 0.823 183s 0.129 -0.523 -0.201 183s -0.093 0.243 0.393 183s -0.739 -0.220 -0.365 183s 183s y: [2, 4, 2, 1] 183s 0.626 -0.504 183s 0.549 -0.229 183s -0.284 0.066 183s -0.400 0.497 183s testing: mul_mat, [3, 2, 2, 4] = [3, 3, 2, 4] * [3, 2, 2, 4] 183s x0: [3, 3, 2, 4] 183s 0.577 -0.945 0.458 183s 0.176 -0.626 -0.537 183s 0.565 -0.777 0.268 183s 183s x1: [3, 2, 2, 4] 183s 0.565 0.521 -0.194 183s 0.826 -0.699 0.441 183s 183s y: [3, 2, 2, 4] 183s -0.255 -0.122 -0.138 183s 1.339 0.346 1.128 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.546 -0.085 0.875 183s 183s x1: [3, 3, 1, 1] 183s -0.130 0.513 0.485 183s -0.723 -0.827 -0.301 183s -0.618 -0.311 -0.225 183s 183s y: [1, 3, 1, 1] 183s 0.452 183s 0.202 183s 0.168 183s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.586 183s 183s x1: [1, 3, 2, 1] 183s 0.518 183s 0.116 183s 0.081 183s 183s y: [1, 3, 2, 1] 183s -0.303 183s -0.068 183s -0.047 183s testing: mul_mat, [4, 1, 2, 4] = [1, 4, 2, 4] * [1, 1, 2, 4] 183s x0: [1, 4, 2, 4] 183s 0.454 183s -0.457 183s 0.889 183s 0.730 183s 183s x1: [1, 1, 2, 4] 183s 0.932 183s 183s y: [4, 1, 2, 4] 183s 0.423 -0.426 0.829 0.681 183s test-mul-mat0: iter:340/500 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.495 0.857 0.759 0.352 183s 183s x1: [4, 3, 1, 1] 183s -0.842 -0.547 0.181 0.521 183s 0.408 0.039 0.799 -0.762 183s -0.585 0.194 -0.851 0.536 183s 183s y: [1, 3, 1, 1] 183s -0.565 183s 0.574 183s -0.580 183s testing: mul_mat, [2, 1, 3, 1] = [4, 2, 3, 1] * [4, 1, 3, 1] 183s x0: [4, 2, 3, 1] 183s -0.133 0.383 0.243 -0.951 183s -0.096 -0.349 0.088 -0.297 183s 183s x1: [4, 1, 3, 1] 183s -0.890 -0.977 -0.302 -0.317 183s 183s y: [2, 1, 3, 1] 183s -0.028 0.494 183s testing: mul_mat, [4, 2, 3, 3] = [4, 4, 3, 3] * [4, 2, 3, 3] 183s x0: [4, 4, 3, 3] 183s -0.184 -0.515 0.021 0.601 183s 0.693 -0.475 -0.287 0.492 183s 0.154 -0.680 -0.587 0.802 183s -0.654 -0.657 0.476 -0.261 183s 183s x1: [4, 2, 3, 3] 183s 0.528 0.488 -0.786 0.977 183s -0.421 0.440 -0.232 0.447 183s 183s y: [4, 2, 3, 3] 183s 0.222 0.839 0.994 -1.294 183s 0.114 -0.214 0.131 -0.241 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.877 -0.891 -0.310 0.298 183s 0.099 -0.291 0.251 0.297 183s 183s x1: [4, 4, 1, 1] 183s 0.382 -0.394 0.721 0.165 183s -0.783 -0.428 -0.341 0.276 183s -0.718 -0.024 -0.479 -0.806 183s 0.704 0.995 0.932 -0.397 183s 183s y: [2, 4, 1, 1] 183s -0.159 0.382 183s 1.256 0.043 183s 0.559 -0.424 183s -1.911 -0.103 183s testing: mul_mat, [3, 4, 3, 1] = [2, 3, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.813 0.438 183s 0.729 0.027 183s 0.377 -0.326 183s 183s x1: [2, 4, 3, 1] 183s -0.096 0.901 183s 0.124 0.099 183s 0.893 -0.806 183s -0.519 -0.501 183s 183s y: [3, 4, 3, 1] 183s 0.473 -0.045 -0.330 183s -0.058 0.093 0.014 183s -1.079 0.629 0.599 183s 0.202 -0.392 -0.032 183s testing: mul_mat, [3, 2, 3, 3] = [3, 3, 3, 3] * [3, 2, 3, 3] 183s x0: [3, 3, 3, 3] 183s -0.874 -0.709 -0.148 183s -0.219 -0.193 -0.292 183s 0.656 0.565 0.500 183s 183s x1: [3, 2, 3, 3] 183s 0.875 -0.973 0.166 183s 0.156 -0.461 -0.870 183s 183s y: [3, 2, 3, 3] 183s -0.100 -0.052 0.108 183s 0.319 0.309 -0.593 183s test-mul-mat0: iter:341/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.744 -0.423 0.667 183s 0.619 -0.958 -0.992 183s 183s x1: [3, 2, 1, 1] 183s -0.590 0.247 -0.868 183s 0.663 -0.983 0.371 183s 183s y: [2, 2, 1, 1] 183s -0.244 0.259 183s 0.170 0.983 183s testing: mul_mat, [1, 2, 3, 1] = [3, 1, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 1, 3, 1] 183s -0.953 -0.053 -0.953 183s 183s x1: [3, 2, 3, 1] 183s 0.671 -0.381 0.551 183s 0.913 -0.114 -0.938 183s 183s y: [1, 2, 3, 1] 183s -1.145 183s 0.030 183s testing: mul_mat, [4, 1, 3, 3] = [3, 4, 3, 3] * [3, 1, 3, 3] 183s x0: [3, 4, 3, 3] 183s -0.295 0.475 -0.733 183s -0.356 -0.589 0.810 183s -0.745 0.967 -0.095 183s 0.778 -0.590 -0.938 183s 183s x1: [3, 1, 3, 3] 183s 0.939 -0.063 0.543 183s 183s y: [4, 1, 3, 3] 183s -0.705 0.144 -0.813 0.258 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.366 0.186 0.697 183s 0.252 -0.443 -0.450 183s -0.438 0.999 -0.520 183s 183s x1: [3, 4, 1, 1] 183s -0.077 0.199 0.604 183s -0.290 0.612 0.636 183s 0.799 0.355 -0.661 183s 0.740 0.465 0.941 183s 183s y: [3, 4, 1, 1] 183s 0.430 -0.380 -0.082 183s 0.451 -0.631 0.407 183s -0.103 0.342 0.348 183s 1.013 -0.443 -0.349 183s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.058 0.545 0.191 183s 0.296 -0.130 0.676 183s 183s x1: [3, 3, 3, 1] 183s 0.927 0.895 -0.870 183s -0.318 0.665 0.140 183s -0.613 0.063 0.418 183s 183s y: [2, 3, 3, 1] 183s 0.268 -0.430 183s 0.407 -0.086 183s 0.150 0.093 183s testing: mul_mat, [3, 3, 3, 3] = [2, 3, 3, 3] * [2, 3, 3, 3] 183s x0: [2, 3, 3, 3] 183s 0.699 -0.491 183s 0.224 -0.675 183s -0.938 -0.657 183s 183s x1: [2, 3, 3, 3] 183s 0.088 0.282 183s -0.089 -0.353 183s 0.976 0.849 183s 183s y: [3, 3, 3, 3] 183s -0.077 -0.171 -0.268 183s 0.111 0.218 0.315 183s 0.266 -0.354 -1.473 183s test-mul-mat0: iter:342/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.464 -0.656 0.632 183s 183s x1: [3, 1, 1, 1] 183s 0.860 -0.489 0.803 183s 183s y: [1, 1, 1, 1] 183s 0.429 183s testing: mul_mat, [2, 1, 4, 1] = [3, 2, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 2, 4, 1] 183s 0.009 -0.201 -0.644 183s 0.660 -0.753 0.965 183s 183s x1: [3, 1, 4, 1] 183s 0.070 0.210 0.891 183s 183s y: [2, 1, 4, 1] 183s -0.616 0.748 183s testing: mul_mat, [2, 2, 4, 2] = [3, 2, 4, 2] * [3, 2, 4, 2] 183s x0: [3, 2, 4, 2] 183s -0.954 0.576 0.377 183s 0.269 -0.013 0.993 183s 183s x1: [3, 2, 4, 2] 183s -0.722 -0.435 -0.372 183s -0.265 0.958 -0.075 183s 183s y: [2, 2, 4, 2] 183s 0.298 -0.557 183s 0.777 -0.158 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.503 -0.668 0.644 183s -0.028 0.581 -0.798 183s 0.095 -0.352 -0.586 183s -0.529 0.739 -0.306 183s 183s x1: [3, 2, 1, 1] 183s 0.296 -0.846 -0.313 183s 0.172 -0.348 0.816 183s 183s y: [4, 2, 1, 1] 183s 0.215 -0.251 0.510 -0.686 183s 0.672 -0.859 -0.339 -0.598 183s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 4, 4, 1] 183s -0.823 0.331 -0.258 -0.950 183s 0.547 -0.096 -0.656 -0.220 183s -0.194 0.331 -0.602 0.838 183s 0.963 0.764 -0.007 0.867 183s 183s x1: [4, 3, 4, 1] 183s 0.748 -0.406 -0.889 -0.445 183s 0.416 -0.917 0.131 -0.021 183s 0.076 -0.449 -0.989 -0.709 183s 183s y: [4, 3, 4, 1] 183s -0.098 1.130 -0.118 0.031 183s -0.660 0.234 -0.481 -0.319 183s 0.717 0.889 -0.162 -0.878 183s testing: mul_mat, [4, 4, 4, 2] = [4, 4, 4, 2] * [4, 4, 4, 2] 183s x0: [4, 4, 4, 2] 183s -0.693 0.517 -0.325 -0.188 183s 0.340 -0.410 -0.331 0.586 183s 0.327 -0.326 -0.032 0.205 183s 0.796 -0.996 -0.684 0.402 183s 183s x1: [4, 4, 4, 2] 183s -0.744 -0.791 0.753 0.260 183s -0.966 -0.366 0.006 -0.798 183s 0.912 0.744 -0.579 0.970 183s -0.678 -0.492 0.744 -0.002 183s 183s y: [4, 4, 4, 2] 183s -0.188 -0.026 0.043 -0.216 183s 0.628 -0.648 -0.360 -0.729 183s -0.241 0.765 0.273 0.770 183s -0.026 -0.277 -0.086 -0.560 183s test-mul-mat0: iter:343/500 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.543 -0.600 183s 183s x1: [2, 3, 1, 1] 183s -0.761 0.729 183s -0.473 -0.786 183s 0.254 -0.805 183s 183s y: [1, 3, 1, 1] 183s -0.025 183s 0.728 183s 0.345 183s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 4, 2, 1] 183s 0.142 0.324 183s 0.104 0.895 183s 0.220 0.765 183s 0.369 0.346 183s 183s x1: [2, 1, 2, 1] 183s 0.523 -0.080 183s 183s y: [4, 1, 2, 1] 183s 0.048 -0.017 0.054 0.165 183s testing: mul_mat, [3, 4, 2, 4] = [2, 3, 2, 4] * [2, 4, 2, 4] 183s x0: [2, 3, 2, 4] 183s -0.904 -0.144 183s -0.876 -0.741 183s -0.589 0.976 183s 183s x1: [2, 4, 2, 4] 183s -0.069 0.872 183s -0.409 0.459 183s -0.914 0.845 183s 0.654 -0.760 183s 183s y: [3, 4, 2, 4] 183s -0.064 -0.586 0.891 183s 0.303 0.018 0.688 183s 0.704 0.174 1.363 183s -0.482 -0.010 -1.127 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.913 0.780 183s 183s x1: [2, 3, 1, 1] 183s -0.251 0.929 183s 0.947 -0.222 183s -0.871 0.594 183s 183s y: [1, 3, 1, 1] 183s 0.954 183s -1.038 183s 1.260 183s testing: mul_mat, [2, 2, 2, 1] = [1, 2, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.424 183s 0.578 183s 183s x1: [1, 2, 2, 1] 183s 0.054 183s -0.084 183s 183s y: [2, 2, 2, 1] 183s 0.023 0.031 183s -0.036 -0.049 183s testing: mul_mat, [3, 1, 2, 4] = [2, 3, 2, 4] * [2, 1, 2, 4] 183s x0: [2, 3, 2, 4] 183s -0.181 -0.095 183s 0.858 0.638 183s -0.937 0.117 183s 183s x1: [2, 1, 2, 4] 183s 0.679 0.352 183s 183s y: [3, 1, 2, 4] 183s -0.156 0.807 -0.595 183s test-mul-mat0: iter:344/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.947 -0.291 0.521 183s -0.935 -0.369 -0.742 183s 183s x1: [3, 2, 1, 1] 183s 0.629 -0.962 -0.156 183s -0.524 -0.353 -0.355 183s 183s y: [2, 2, 1, 1] 183s 0.794 -0.117 183s -0.579 0.884 183s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.466 0.141 -0.769 183s -0.400 0.772 -0.511 183s 0.688 0.964 -0.399 183s 0.621 -0.092 0.571 183s 183s x1: [3, 2, 4, 1] 183s 0.089 -0.808 -0.888 183s 0.933 -0.056 -0.030 183s 183s y: [4, 2, 4, 1] 183s 0.528 -0.206 -0.364 -0.378 183s -0.419 -0.401 0.600 0.567 183s testing: mul_mat, [3, 4, 4, 4] = [3, 3, 4, 4] * [3, 4, 4, 4] 183s x0: [3, 3, 4, 4] 183s -0.151 -0.698 -0.949 183s 0.283 0.145 -0.889 183s -0.960 0.722 -0.191 183s 183s x1: [3, 4, 4, 4] 183s 0.314 0.255 0.699 183s -0.012 0.643 -0.614 183s 0.606 0.418 0.013 183s -0.989 0.056 -0.772 183s 183s y: [3, 4, 4, 4] 183s -0.888 -0.495 -0.251 183s 0.136 0.636 0.593 183s -0.396 0.220 -0.283 183s 0.843 0.414 1.138 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.584 -0.893 -0.763 183s -0.292 0.303 0.405 183s -0.964 0.085 0.115 183s 183s x1: [3, 3, 1, 1] 183s -0.945 0.803 0.021 183s 0.846 -0.652 -0.160 183s -0.317 -0.431 -0.489 183s 183s y: [3, 3, 1, 1] 183s -0.181 0.528 0.982 183s 0.210 -0.510 -0.889 183s 0.943 -0.237 0.212 183s testing: mul_mat, [4, 3, 4, 1] = [3, 4, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.376 0.303 0.695 183s 0.140 -0.391 -0.524 183s -0.456 -0.390 0.582 183s -0.410 0.269 0.622 183s 183s x1: [3, 3, 4, 1] 183s -0.412 -0.408 -0.798 183s 0.861 0.731 -0.202 183s -0.466 -0.037 0.769 183s 183s y: [4, 3, 4, 1] 183s -0.524 0.519 -0.118 -0.437 183s -0.242 -0.059 -0.795 -0.282 183s 0.699 -0.453 0.675 0.659 183s testing: mul_mat, [4, 3, 4, 4] = [4, 4, 4, 4] * [4, 3, 4, 4] 183s x0: [4, 4, 4, 4] 183s 0.856 -0.694 -0.671 -0.250 183s -0.697 0.344 0.021 0.216 183s -0.379 0.899 0.795 0.458 183s -0.817 0.912 0.789 -0.473 183s 183s x1: [4, 3, 4, 4] 183s 0.064 0.166 -0.180 -0.283 183s 0.968 0.630 -0.777 -0.453 183s -0.330 -0.824 0.102 0.801 183s 183s y: [4, 3, 4, 4] 183s 0.132 -0.053 -0.148 0.091 183s 1.027 -0.572 -0.626 -0.615 183s 0.021 0.121 -0.168 -0.781 183s test-mul-mat0: iter:345/500 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.316 0.435 0.550 183s -0.547 0.817 0.259 183s 183s x1: [3, 3, 1, 1] 183s 0.376 0.053 0.155 183s 0.529 0.112 -0.319 183s -0.980 -0.862 0.804 183s 183s y: [2, 3, 1, 1] 183s -0.011 -0.123 183s -0.294 -0.280 183s 0.377 0.040 183s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 2, 2, 1] 183s 0.169 -0.413 0.014 183s -0.778 0.742 -0.456 183s 183s x1: [3, 2, 2, 1] 183s 0.934 0.810 0.717 183s 0.251 0.892 -0.994 183s 183s y: [2, 2, 2, 1] 183s -0.166 -0.453 183s -0.340 0.920 183s testing: mul_mat, [3, 2, 2, 1] = [3, 3, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 3, 2, 1] 183s 0.353 -0.163 0.061 183s 0.164 0.350 -0.301 183s -0.667 0.937 0.713 183s 183s x1: [3, 2, 2, 1] 183s -0.844 0.661 -0.623 183s -0.391 0.478 0.636 183s 183s y: [3, 2, 2, 1] 183s -0.444 0.281 0.738 183s -0.178 -0.088 1.162 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.934 0.287 -0.549 183s -0.628 0.209 -0.441 183s 0.098 -0.842 -0.143 183s 183s x1: [3, 3, 1, 1] 183s 0.518 0.988 -0.975 183s -0.872 0.466 0.661 183s -0.329 -0.246 0.013 183s 183s y: [3, 3, 1, 1] 183s 1.303 0.311 -0.641 183s -1.043 0.354 -0.572 183s -0.385 0.149 0.173 183s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 3, 2, 1] 183s -0.189 -0.256 -0.969 183s 0.992 -0.636 0.573 183s 0.374 -0.528 -0.370 183s 183s x1: [3, 3, 2, 1] 183s -0.216 -0.504 -0.430 183s 0.339 -0.825 -0.174 183s 0.229 -0.725 0.646 183s 183s y: [3, 3, 2, 1] 183s 0.587 -0.141 0.345 183s 0.315 0.762 0.627 183s -0.484 1.058 0.229 183s testing: mul_mat, [4, 3, 2, 1] = [3, 4, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 4, 2, 1] 183s 0.914 -0.819 0.700 183s -0.952 0.644 -0.732 183s -0.843 0.663 0.612 183s -0.603 -0.191 -0.417 183s 183s x1: [3, 3, 2, 1] 183s 0.835 0.079 -0.472 183s -0.574 0.681 0.822 183s -0.464 -0.728 0.140 183s 183s y: [4, 3, 2, 1] 183s 0.369 -0.399 -0.940 -0.322 183s -0.508 0.384 1.439 -0.126 183s 0.270 -0.129 -0.005 0.361 183s test-mul-mat0: iter:346/500 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.910 183s -0.174 183s -0.742 183s 183s x1: [1, 2, 1, 1] 183s 0.996 183s -0.222 183s 183s y: [3, 2, 1, 1] 183s 0.906 -0.173 -0.739 183s -0.202 0.039 0.165 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.816 183s 183s x1: [1, 3, 1, 1] 183s -0.693 183s 0.007 183s 0.902 183s 183s y: [1, 3, 1, 1] 183s -0.566 183s 0.006 183s 0.736 183s testing: mul_mat, [2, 1, 1, 4] = [1, 2, 1, 4] * [1, 1, 1, 4] 183s x0: [1, 2, 1, 4] 183s -0.880 183s 0.296 183s 183s x1: [1, 1, 1, 4] 183s 0.602 183s 183s y: [2, 1, 1, 4] 183s -0.530 0.179 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.289 183s -0.778 183s -0.202 183s 0.863 183s 183s x1: [1, 2, 1, 1] 183s 0.771 183s -0.890 183s 183s y: [4, 2, 1, 1] 183s 0.223 -0.600 -0.156 0.666 183s -0.258 0.692 0.180 -0.768 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.778 -0.876 0.726 0.911 183s -0.565 -0.596 -0.358 -0.018 183s 0.409 0.225 -0.346 -0.227 183s 183s x1: [4, 1, 1, 1] 183s -0.782 0.576 0.964 -0.872 183s 183s y: [3, 1, 1, 1] 183s 0.009 -0.231 -0.327 183s testing: mul_mat, [4, 4, 1, 4] = [3, 4, 1, 4] * [3, 4, 1, 4] 183s x0: [3, 4, 1, 4] 183s 0.929 -0.119 -0.980 183s 0.862 0.459 -0.705 183s 0.539 0.995 -0.718 183s -0.650 0.626 0.564 183s 183s x1: [3, 4, 1, 4] 183s 0.207 0.852 0.266 183s 0.437 0.786 -0.597 183s 0.740 0.557 -0.487 183s -0.765 -0.154 -0.265 183s 183s y: [4, 4, 1, 4] 183s -0.169 0.382 0.769 0.548 183s 0.898 1.159 1.447 -0.130 183s 1.099 1.237 1.303 -0.408 183s -0.433 -0.543 -0.375 0.251 183s test-mul-mat0: iter:347/500 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.462 -0.776 0.609 183s 183s x1: [3, 4, 1, 1] 183s -0.218 -0.785 0.685 183s -0.528 -0.714 -0.557 183s 0.105 0.275 -0.832 183s 0.761 -0.807 0.755 183s 183s y: [1, 4, 1, 1] 183s 0.926 183s -0.028 183s -0.671 183s 1.437 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.305 -0.479 0.210 183s 0.216 -0.280 0.843 183s 0.183 0.060 -0.077 183s 0.636 -0.719 0.705 183s 183s x1: [3, 1, 1, 1] 183s -0.274 -0.681 0.283 183s 183s y: [4, 1, 1, 1] 183s 0.470 0.370 -0.113 0.515 183s testing: mul_mat, [4, 4, 1, 3] = [3, 4, 1, 3] * [3, 4, 1, 3] 183s x0: [3, 4, 1, 3] 183s 0.006 0.424 -0.739 183s -0.389 0.757 0.260 183s 0.678 -0.687 0.868 183s 0.471 0.188 0.758 183s 183s x1: [3, 4, 1, 3] 183s 0.851 0.966 -0.823 183s -0.863 -0.591 0.283 183s 0.412 -0.423 0.043 183s 0.605 -0.668 -0.001 183s 183s y: [4, 4, 1, 3] 183s 1.023 0.186 -0.800 -0.041 183s -0.465 -0.038 0.066 -0.303 183s -0.209 -0.469 0.608 0.147 183s -0.279 -0.741 0.868 0.159 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.282 0.299 0.074 183s 0.748 -0.132 0.445 183s 0.037 0.003 0.406 183s 183s x1: [3, 4, 1, 1] 183s -0.551 0.466 -0.255 183s -0.684 -0.063 0.933 183s -0.926 -0.516 -0.309 183s -0.756 0.318 -0.765 183s 183s y: [3, 4, 1, 1] 183s -0.035 -0.587 -0.123 183s -0.143 -0.088 0.354 183s -0.438 -0.762 -0.161 183s -0.175 -0.948 -0.338 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.953 -0.016 0.468 183s -0.489 -0.556 0.135 183s 0.665 0.739 0.983 183s 183s x1: [3, 3, 1, 1] 183s -0.217 0.616 -0.342 183s 0.711 -0.207 0.333 183s -0.100 -0.419 0.766 183s 183s y: [3, 3, 1, 1] 183s 0.036 -0.283 -0.025 183s -0.518 -0.187 0.647 183s 0.461 0.385 0.376 183s testing: mul_mat, [3, 3, 1, 3] = [3, 3, 1, 3] * [3, 3, 1, 3] 183s x0: [3, 3, 1, 3] 183s -0.081 0.706 0.775 183s -0.180 -0.810 0.308 183s -0.430 -0.903 -0.338 183s 183s x1: [3, 3, 1, 3] 183s -0.213 0.370 0.527 183s -0.931 0.118 -0.436 183s 0.368 0.986 0.567 183s 183s y: [3, 3, 1, 3] 183s 0.687 -0.099 -0.421 183s -0.179 -0.063 0.441 183s 1.106 -0.690 -1.240 183s test-mul-mat0: iter:348/500 183s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.493 0.068 183s 0.763 0.078 183s 0.793 -0.035 183s 183s x1: [2, 4, 1, 1] 183s -0.133 -0.725 183s -0.225 -0.427 183s -0.535 -0.127 183s -0.651 0.773 183s 183s y: [3, 4, 1, 1] 183s -0.115 -0.158 -0.080 183s -0.140 -0.205 -0.163 183s -0.272 -0.418 -0.420 183s -0.268 -0.436 -0.543 183s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 1, 3, 1] 183s 0.091 -0.915 183s 183s x1: [2, 3, 3, 1] 183s 0.890 0.565 183s 0.798 0.637 183s 0.176 0.596 183s 183s y: [1, 3, 3, 1] 183s -0.435 183s -0.510 183s -0.529 183s testing: mul_mat, [4, 1, 3, 4] = [2, 4, 3, 4] * [2, 1, 3, 4] 183s x0: [2, 4, 3, 4] 183s 0.629 -0.815 183s -0.782 0.069 183s -0.956 0.214 183s -0.998 -0.029 183s 183s x1: [2, 1, 3, 4] 183s -0.067 -0.073 183s 183s y: [4, 1, 3, 4] 183s 0.018 0.047 0.048 0.069 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.691 0.718 183s 183s x1: [2, 4, 1, 1] 183s 0.766 0.620 183s 0.262 0.567 183s -0.191 -0.597 183s 0.109 -0.817 183s 183s y: [1, 4, 1, 1] 183s -0.085 183s 0.226 183s -0.296 183s -0.662 183s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.831 183s -0.321 183s 183s x1: [1, 2, 3, 1] 183s 0.957 183s -0.382 183s 183s y: [2, 2, 3, 1] 183s -0.795 -0.307 183s 0.317 0.123 183s testing: mul_mat, [3, 1, 3, 4] = [2, 3, 3, 4] * [2, 1, 3, 4] 183s x0: [2, 3, 3, 4] 183s 0.172 -0.133 183s -0.667 -0.327 183s -0.055 0.063 183s 183s x1: [2, 1, 3, 4] 183s -0.684 0.689 183s 183s y: [3, 1, 3, 4] 183s -0.209 0.231 0.081 183s test-mul-mat0: iter:349/500 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.677 -0.047 0.334 0.573 183s 0.012 0.147 0.005 -0.801 183s 183s x1: [4, 2, 1, 1] 183s -0.771 -0.865 -0.907 0.555 183s 0.051 -0.315 0.662 0.133 183s 183s y: [2, 2, 1, 1] 183s -0.467 -0.585 183s 0.347 -0.150 183s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.809 0.303 0.665 0.338 183s -0.693 0.865 -0.202 -0.387 183s 0.326 0.936 -0.238 0.186 183s 183s x1: [4, 2, 3, 1] 183s 0.460 -0.694 0.461 0.138 183s -0.851 0.860 0.367 -0.104 183s 183s y: [3, 2, 3, 1] 183s -0.229 -1.065 -0.584 183s 1.158 1.300 0.422 183s testing: mul_mat, [4, 3, 3, 2] = [4, 4, 3, 2] * [4, 3, 3, 2] 183s x0: [4, 4, 3, 2] 183s 0.534 0.707 -0.895 0.918 183s 0.894 -0.938 0.717 -0.188 183s -0.665 0.446 0.872 -0.002 183s -0.526 -0.789 -0.224 -0.637 183s 183s x1: [4, 3, 3, 2] 183s -0.744 0.697 0.362 0.581 183s 0.633 -0.876 -0.232 -0.064 183s 0.782 -0.714 0.847 0.732 183s 183s y: [4, 3, 3, 2] 183s 0.304 -1.168 1.121 -0.610 183s -0.131 1.233 -1.014 0.451 183s -0.173 1.839 -0.101 -0.504 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.841 -0.018 -0.642 0.178 183s -0.976 0.920 -0.217 -0.765 183s 0.374 -0.355 -0.409 -0.701 183s 183s x1: [4, 4, 1, 1] 183s -0.524 0.395 0.870 0.968 183s 0.233 -0.903 0.628 -0.753 183s 0.445 0.931 -0.353 -0.310 183s 0.240 -0.295 0.893 0.567 183s 183s y: [3, 4, 1, 1] 183s 0.047 -0.054 -1.371 183s -0.717 -0.618 0.679 183s -0.219 0.736 0.198 183s -0.670 -1.133 -0.568 183s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.187 -0.711 0.322 183s 0.909 0.728 -0.410 183s 183s x1: [3, 4, 3, 1] 183s -0.961 -0.999 -0.225 183s 0.435 0.871 -0.257 183s -0.332 0.967 -0.630 183s -0.085 0.412 -0.699 183s 183s y: [2, 4, 3, 1] 183s 0.817 -1.509 183s -0.783 1.135 183s -0.829 0.660 183s -0.502 0.509 183s testing: mul_mat, [2, 3, 3, 2] = [2, 2, 3, 2] * [2, 3, 3, 2] 183s x0: [2, 2, 3, 2] 183s -0.055 0.053 183s -0.636 -0.955 183s 183s x1: [2, 3, 3, 2] 183s 0.422 -0.797 183s 0.259 0.524 183s -0.125 0.329 183s 183s y: [2, 3, 3, 2] 183s -0.065 0.492 183s 0.013 -0.665 183s 0.024 -0.235 183s test-mul-mat0: iter:350/500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.760 183s 0.294 183s 0.371 183s 0.293 183s 183s x1: [1, 1, 1, 1] 183s -0.651 183s 183s y: [4, 1, 1, 1] 183s 0.495 -0.191 -0.242 -0.191 183s testing: mul_mat, [1, 4, 3, 1] = [1, 1, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 1, 3, 1] 183s 0.086 183s 183s x1: [1, 4, 3, 1] 183s 0.339 183s 0.467 183s 0.689 183s -0.599 183s 183s y: [1, 4, 3, 1] 183s 0.029 183s 0.040 183s 0.059 183s -0.051 183s testing: mul_mat, [2, 1, 3, 4] = [1, 2, 3, 4] * [1, 1, 3, 4] 183s x0: [1, 2, 3, 4] 183s -0.025 183s -0.516 183s 183s x1: [1, 1, 3, 4] 183s -0.159 183s 183s y: [2, 1, 3, 4] 183s 0.004 0.082 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.995 183s -0.133 183s -0.986 183s 0.646 183s 183s x1: [1, 2, 1, 1] 183s 0.657 183s 0.263 183s 183s y: [4, 2, 1, 1] 183s -0.654 -0.088 -0.648 0.425 183s -0.262 -0.035 -0.259 0.170 183s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.908 -0.650 -0.351 0.316 183s 0.868 0.282 -0.208 -0.336 183s 0.608 -0.404 -0.745 0.912 183s 183s x1: [4, 1, 3, 1] 183s 0.842 -0.502 -0.374 -0.716 183s 183s y: [3, 1, 3, 1] 183s -0.534 0.908 0.341 183s testing: mul_mat, [1, 4, 3, 4] = [3, 1, 3, 4] * [3, 4, 3, 4] 183s x0: [3, 1, 3, 4] 183s -0.018 -0.514 0.701 183s 183s x1: [3, 4, 3, 4] 183s -0.530 0.188 0.645 183s -0.275 -0.496 -0.691 183s -0.363 0.082 0.796 183s -0.446 -0.472 0.296 183s 183s y: [1, 4, 3, 4] 183s 0.366 183s -0.225 183s 0.523 183s 0.459 183s test-mul-mat0: iter:351/500 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.618 0.077 0.632 183s -0.821 0.710 -0.274 183s 0.226 0.486 0.748 183s 0.255 0.749 -0.529 183s 183s x1: [3, 4, 1, 1] 183s 0.857 0.249 0.994 183s 0.891 -0.300 -0.147 183s 0.230 -0.925 -0.723 183s -0.474 -0.222 -0.169 183s 183s y: [4, 4, 1, 1] 183s 0.117 -0.799 1.059 -0.121 183s -0.667 -0.905 -0.054 0.080 183s -0.671 -0.647 -0.939 -0.251 183s 0.169 0.277 -0.342 -0.198 183s testing: mul_mat, [4, 4, 4, 1] = [3, 4, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 4, 4, 1] 183s 0.176 0.220 -0.535 183s 0.211 0.646 0.052 183s 0.973 0.298 0.126 183s -0.517 -0.796 -0.052 183s 183s x1: [3, 4, 4, 1] 183s -0.856 0.809 0.465 183s 0.526 0.063 0.844 183s 0.384 -0.687 0.839 183s 0.275 0.013 -0.308 183s 183s y: [4, 4, 4, 1] 183s -0.221 0.366 -0.534 -0.225 183s -0.345 0.196 0.637 -0.366 183s -0.533 -0.319 0.274 0.305 183s 0.216 0.050 0.232 -0.136 183s testing: mul_mat, [2, 4, 4, 2] = [3, 2, 4, 2] * [3, 4, 4, 2] 183s x0: [3, 2, 4, 2] 183s -0.023 0.129 -0.381 183s -0.689 0.132 -0.924 183s 183s x1: [3, 4, 4, 2] 183s 0.968 0.951 0.633 183s -0.749 -0.723 -0.338 183s -0.445 0.422 0.431 183s 0.967 -0.645 -0.029 183s 183s y: [2, 4, 4, 2] 183s -0.141 -1.127 183s 0.053 0.733 183s -0.100 -0.036 183s -0.094 -0.725 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.086 0.587 0.864 183s 0.977 -0.596 0.808 183s 0.901 -0.836 0.556 183s 183s x1: [3, 2, 1, 1] 183s 0.306 -0.287 0.035 183s 0.741 -0.837 -0.659 183s 183s y: [3, 2, 1, 1] 183s -0.112 0.498 0.535 183s -0.997 0.690 1.001 183s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 1, 4, 1] 183s -0.767 -0.510 0.034 183s 183s x1: [3, 3, 4, 1] 183s -0.240 0.523 -0.920 183s 0.732 -0.615 -0.206 183s 0.202 0.609 0.502 183s 183s y: [1, 3, 4, 1] 183s -0.114 183s -0.254 183s -0.449 183s testing: mul_mat, [1, 3, 4, 2] = [1, 1, 4, 2] * [1, 3, 4, 2] 183s x0: [1, 1, 4, 2] 183s 0.619 183s 183s x1: [1, 3, 4, 2] 183s -0.468 183s -0.887 183s -0.970 183s 183s y: [1, 3, 4, 2] 183s -0.290 183s -0.549 183s -0.600 183s test-mul-mat0: iter:352/500 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.374 0.250 0.582 183s 0.102 0.189 -0.446 183s -0.279 0.400 -0.670 183s 183s x1: [3, 4, 1, 1] 183s -0.427 0.513 0.799 183s -0.815 0.930 -0.273 183s -0.544 0.647 -0.848 183s 0.465 -0.606 -0.544 183s 183s y: [3, 4, 1, 1] 183s 0.433 -0.303 -0.211 183s -0.231 0.215 0.782 183s -0.535 0.446 0.979 183s -0.294 0.175 -0.007 183s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 2, 4, 1] 183s 0.562 -0.114 -0.612 183s 0.371 -0.811 -0.566 183s 183s x1: [3, 3, 4, 1] 183s -0.319 0.721 0.352 183s 0.174 -0.845 -0.596 183s 0.236 0.790 -0.753 183s 183s y: [2, 3, 4, 1] 183s -0.477 -0.902 183s 0.559 1.087 183s 0.503 -0.127 183s testing: mul_mat, [4, 2, 4, 3] = [3, 4, 4, 3] * [3, 2, 4, 3] 183s x0: [3, 4, 4, 3] 183s -0.950 -0.419 -0.268 183s 0.951 -0.569 0.112 183s 0.872 0.918 0.259 183s 0.391 -0.445 -0.176 183s 183s x1: [3, 2, 4, 3] 183s 0.596 0.599 -0.876 183s 0.333 0.925 0.219 183s 183s y: [4, 2, 4, 3] 183s -0.583 0.127 0.842 0.120 183s -0.763 -0.185 1.197 -0.320 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.966 0.556 -0.500 183s -0.241 -0.225 0.656 183s -0.006 -0.652 -0.570 183s 0.417 0.549 -0.630 183s 183s x1: [3, 4, 1, 1] 183s -0.694 0.435 -0.454 183s 0.024 0.936 0.577 183s -0.234 0.689 0.924 183s -0.213 0.218 -0.199 183s 183s y: [4, 4, 1, 1] 183s -0.202 -0.228 -0.021 0.235 183s 0.254 0.162 -0.939 0.160 183s -0.305 0.507 -0.975 -0.301 183s 0.015 -0.128 -0.027 0.156 183s testing: mul_mat, [3, 3, 4, 1] = [4, 3, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 3, 4, 1] 183s -0.414 0.252 -0.307 -0.366 183s -0.790 -0.170 0.798 -0.577 183s 0.253 -0.760 -0.223 0.743 183s 183s x1: [4, 3, 4, 1] 183s 0.319 0.102 -0.178 0.271 183s -0.041 0.825 0.577 -0.606 183s -0.629 -0.399 -0.670 0.948 183s 183s y: [3, 3, 4, 1] 183s -0.151 -0.568 0.244 183s 0.270 0.703 -1.216 183s 0.019 -0.517 0.998 183s testing: mul_mat, [2, 4, 4, 3] = [3, 2, 4, 3] * [3, 4, 4, 3] 183s x0: [3, 2, 4, 3] 183s 0.068 -0.610 0.428 183s 0.433 -0.280 -0.832 183s 183s x1: [3, 4, 4, 3] 183s 0.278 -0.335 -0.635 183s 0.538 0.032 -0.870 183s 0.549 -0.324 -0.347 183s 0.330 0.470 -0.910 183s 183s y: [2, 4, 4, 3] 183s -0.048 0.742 183s -0.355 0.948 183s 0.086 0.617 183s -0.653 0.769 183s test-mul-mat0: iter:353/500 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.718 0.003 -0.473 0.574 183s 183s x1: [4, 4, 1, 1] 183s -0.615 0.894 0.189 -0.069 183s 0.463 -0.003 -0.721 0.854 183s 0.131 -0.298 0.510 -0.609 183s -0.133 -0.214 -0.769 0.823 183s 183s y: [1, 4, 1, 1] 183s -0.568 183s 1.164 183s -0.498 183s 0.740 183s testing: mul_mat, [4, 1, 2, 1] = [4, 4, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 4, 2, 1] 183s 0.478 -0.992 0.395 -0.333 183s -0.061 -0.141 0.664 0.218 183s -0.287 -0.206 0.920 -0.777 183s 0.185 -0.213 0.009 0.416 183s 183s x1: [4, 1, 2, 1] 183s 0.834 0.550 -0.693 -0.031 183s 183s y: [4, 1, 2, 1] 183s -0.410 -0.595 -0.966 0.018 183s testing: mul_mat, [4, 4, 2, 3] = [4, 4, 2, 3] * [4, 4, 2, 3] 183s x0: [4, 4, 2, 3] 183s 0.644 0.163 -0.844 -0.748 183s -0.530 -0.590 -0.935 0.722 183s -0.181 0.284 -0.037 -0.012 183s -0.221 0.371 -0.862 -0.382 183s 183s x1: [4, 4, 2, 3] 183s -0.866 0.396 -0.015 0.073 183s -0.746 -0.351 -0.709 -0.033 183s 0.443 -0.789 0.190 -0.372 183s -0.002 -0.800 -0.956 0.608 183s 183s y: [4, 4, 2, 3] 183s -0.535 0.292 0.269 0.323 183s 0.085 1.241 0.062 0.658 183s 0.274 -0.216 -0.307 -0.412 183s 0.221 1.806 -0.199 0.296 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.879 0.629 0.198 0.811 183s 183s x1: [4, 4, 1, 1] 183s -0.267 0.486 -0.630 0.623 183s 0.751 0.062 0.320 0.197 183s -0.426 -0.395 0.193 0.045 183s -0.906 -0.429 0.408 0.405 183s 183s y: [1, 4, 1, 1] 183s 0.920 183s -0.399 183s 0.201 183s 0.936 183s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.147 183s 183s x1: [1, 4, 2, 1] 183s 0.256 183s 0.960 183s -0.833 183s 0.678 183s 183s y: [1, 4, 2, 1] 183s -0.038 183s -0.141 183s 0.122 183s -0.100 183s testing: mul_mat, [2, 1, 2, 3] = [1, 2, 2, 3] * [1, 1, 2, 3] 183s x0: [1, 2, 2, 3] 183s -0.141 183s 0.071 183s 183s x1: [1, 1, 2, 3] 183s -0.027 183s 183s y: [2, 1, 2, 3] 183s 0.004 -0.002 183s test-mul-mat0: iter:354/500 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.525 -0.373 183s -0.053 0.328 183s -0.850 0.920 183s -0.449 0.296 183s 183s x1: [2, 1, 1, 1] 183s 0.132 0.838 183s 183s y: [4, 1, 1, 1] 183s -0.382 0.268 0.659 0.188 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.366 -0.091 183s -0.433 -0.376 183s -0.659 -0.235 183s 0.080 0.584 183s 183s x1: [2, 4, 1, 1] 183s -0.355 -0.165 183s 0.762 -0.434 183s 0.694 -0.168 183s 0.680 0.598 183s 183s y: [4, 4, 1, 1] 183s -0.115 0.216 0.273 -0.125 183s 0.318 -0.167 -0.400 -0.192 183s 0.269 -0.237 -0.418 -0.042 183s 0.195 -0.519 -0.588 0.404 183s testing: mul_mat, [1, 4, 1, 2] = [2, 1, 1, 2] * [2, 4, 1, 2] 183s x0: [2, 1, 1, 2] 183s -0.469 -0.766 183s 183s x1: [2, 4, 1, 2] 183s -0.780 0.497 183s -0.006 0.352 183s 0.335 -0.547 183s 0.827 0.962 183s 183s y: [1, 4, 1, 2] 183s -0.015 183s -0.266 183s 0.262 183s -1.124 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.259 -0.798 183s -0.435 0.082 183s 183s x1: [2, 1, 1, 1] 183s 0.832 0.998 183s 183s y: [2, 1, 1, 1] 183s -1.013 -0.280 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.660 -0.887 183s -0.098 -0.887 183s 183s x1: [2, 2, 1, 1] 183s 0.331 0.282 183s -0.334 0.550 183s 183s y: [2, 2, 1, 1] 183s -0.032 -0.282 183s -0.708 -0.455 183s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 1, 1, 2] 183s -0.872 -0.236 183s 183s x1: [2, 2, 1, 2] 183s -0.271 -0.925 183s 0.514 0.883 183s 183s y: [1, 2, 1, 2] 183s 0.455 183s -0.656 183s test-mul-mat0: iter:355/500 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.543 0.841 0.572 0.971 183s 0.724 0.759 -0.194 0.312 183s 183s x1: [4, 3, 1, 1] 183s 0.121 -0.065 0.329 0.323 183s -0.983 -0.340 -0.395 -0.316 183s -0.790 0.383 -0.656 0.112 183s 183s y: [2, 3, 1, 1] 183s 0.381 0.075 183s -0.286 -0.992 183s 0.485 -0.119 183s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.981 -0.595 -0.477 0.741 183s 0.210 0.835 0.095 -0.184 183s 0.275 0.213 0.754 0.077 183s -0.903 -0.764 0.074 0.745 183s 183s x1: [4, 2, 3, 1] 183s 0.354 0.606 0.440 -0.881 183s -0.062 0.802 -0.116 -0.519 183s 183s y: [4, 2, 3, 1] 183s -0.876 0.784 0.491 -1.408 183s -0.867 0.741 0.027 -0.952 183s testing: mul_mat, [4, 4, 3, 4] = [4, 4, 3, 4] * [4, 4, 3, 4] 183s x0: [4, 4, 3, 4] 183s -0.743 0.777 0.403 0.184 183s -0.787 -0.646 -0.465 0.679 183s 0.405 0.732 -0.292 0.132 183s 0.114 -0.141 0.696 -0.968 183s 183s x1: [4, 4, 3, 4] 183s 0.750 -0.792 -0.981 -0.296 183s -0.434 -0.689 0.291 -0.230 183s 0.516 0.854 0.932 0.131 183s -0.149 0.893 -0.274 -0.748 183s 183s y: [4, 4, 3, 4] 183s -1.623 0.176 -0.028 -0.199 183s -0.137 0.495 -0.796 0.473 183s 0.679 -1.303 0.579 0.460 183s 0.556 -0.839 0.575 0.391 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.113 -0.979 0.662 0.442 183s 183s x1: [4, 4, 1, 1] 183s 0.136 -0.856 0.470 0.334 183s -0.450 0.874 0.378 -0.725 183s 0.117 -0.257 -0.303 -0.833 183s 0.847 -0.845 -0.946 0.803 183s 183s y: [1, 4, 1, 1] 183s 1.311 183s -0.977 183s -0.304 183s 0.652 183s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 4, 3, 1] 183s -0.285 183s 0.556 183s 0.585 183s -0.907 183s 183s x1: [1, 4, 3, 1] 183s 0.935 183s -0.213 183s 0.043 183s 0.844 183s 183s y: [4, 4, 3, 1] 183s -0.267 0.520 0.547 -0.848 183s 0.061 -0.118 -0.125 0.193 183s -0.012 0.024 0.025 -0.039 183s -0.241 0.469 0.493 -0.765 183s testing: mul_mat, [3, 1, 3, 4] = [4, 3, 3, 4] * [4, 1, 3, 4] 183s x0: [4, 3, 3, 4] 183s -0.377 -0.277 0.525 0.197 183s 0.824 0.325 0.662 -0.318 183s -0.426 0.776 0.508 0.922 183s 183s x1: [4, 1, 3, 4] 183s -0.514 -0.060 -0.963 -0.493 183s 183s y: [3, 1, 3, 4] 183s -0.393 -0.924 -0.771 183s test-mul-mat0: iter:356/500 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.280 183s 183s x1: [1, 4, 1, 1] 183s 0.245 183s 0.852 183s -0.611 183s 0.630 183s 183s y: [1, 4, 1, 1] 183s 0.069 183s 0.238 183s -0.171 183s 0.176 183s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.664 183s 183s x1: [1, 1, 2, 1] 183s 0.030 183s 183s y: [1, 1, 2, 1] 183s -0.020 183s testing: mul_mat, [1, 1, 2, 3] = [1, 1, 2, 3] * [1, 1, 2, 3] 183s x0: [1, 1, 2, 3] 183s -0.728 183s 183s x1: [1, 1, 2, 3] 183s 0.679 183s 183s y: [1, 1, 2, 3] 183s -0.495 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.722 183s 183s x1: [1, 1, 1, 1] 183s 0.528 183s 183s y: [1, 1, 1, 1] 183s -0.381 183s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.734 183s 183s x1: [1, 1, 2, 1] 183s -0.394 183s 183s y: [1, 1, 2, 1] 183s 0.289 183s testing: mul_mat, [1, 1, 2, 3] = [1, 1, 2, 3] * [1, 1, 2, 3] 183s x0: [1, 1, 2, 3] 183s 0.232 183s 183s x1: [1, 1, 2, 3] 183s -0.221 183s 183s y: [1, 1, 2, 3] 183s -0.051 183s test-mul-mat0: iter:357/500 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.295 0.361 0.471 183s 183s x1: [3, 3, 1, 1] 183s 0.296 0.403 0.401 183s 0.071 -0.866 -0.844 183s -0.401 -0.230 -0.566 183s 183s y: [1, 3, 1, 1] 183s 0.422 183s -0.690 183s -0.468 183s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 4, 2, 1] 183s 0.235 0.293 0.663 183s 0.237 -0.796 0.548 183s -0.577 -0.285 -0.427 183s 0.866 -0.920 0.869 183s 183s x1: [3, 1, 2, 1] 183s 0.065 -0.859 -0.874 183s 183s y: [4, 1, 2, 1] 183s -0.816 0.221 0.580 0.088 183s testing: mul_mat, [2, 4, 2, 3] = [3, 2, 2, 3] * [3, 4, 2, 3] 183s x0: [3, 2, 2, 3] 183s -0.387 -0.090 0.670 183s -0.455 0.933 -0.147 183s 183s x1: [3, 4, 2, 3] 183s -0.467 0.574 -0.659 183s -0.002 -0.815 -0.598 183s 0.894 0.421 0.695 183s 0.557 -0.342 0.899 183s 183s y: [2, 4, 2, 3] 183s -0.313 0.845 183s -0.327 -0.671 183s 0.082 -0.116 183s 0.417 -0.705 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.530 -0.389 0.505 183s 0.064 0.553 -0.128 183s 0.202 0.336 0.323 183s 183s x1: [3, 2, 1, 1] 183s 0.614 -0.025 -0.136 183s 0.865 -0.054 -0.576 183s 183s y: [3, 2, 1, 1] 183s 0.267 0.043 0.072 183s 0.188 0.099 -0.030 183s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 3, 2, 1] 183s 0.321 -0.148 0.463 183s 0.498 -0.438 -0.885 183s -0.395 0.807 0.143 183s 183s x1: [3, 3, 2, 1] 183s 0.104 -0.656 0.952 183s -0.516 0.693 0.905 183s -0.858 -0.662 0.422 183s 183s y: [3, 3, 2, 1] 183s 0.571 -0.503 -0.434 183s 0.150 -1.362 0.893 183s 0.017 -0.512 -0.134 183s testing: mul_mat, [2, 3, 2, 3] = [3, 2, 2, 3] * [3, 3, 2, 3] 183s x0: [3, 2, 2, 3] 183s 0.693 0.576 0.873 183s 0.905 0.140 0.973 183s 183s x1: [3, 3, 2, 3] 183s -0.302 0.363 0.744 183s 0.931 -0.040 0.594 183s 0.926 -0.640 0.804 183s 183s y: [2, 3, 2, 3] 183s 0.650 0.501 183s 1.140 1.415 183s 0.975 1.531 183s test-mul-mat0: iter:358/500 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.142 183s -0.637 183s 183s x1: [1, 4, 1, 1] 183s -0.611 183s 0.097 183s 0.611 183s 0.920 183s 183s y: [2, 4, 1, 1] 183s -0.087 0.389 183s 0.014 -0.062 183s 0.087 -0.389 183s 0.131 -0.586 183s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 3, 3, 1] 183s 0.704 183s -0.490 183s -0.721 183s 183s x1: [1, 2, 3, 1] 183s -0.527 183s -0.512 183s 183s y: [3, 2, 3, 1] 183s -0.371 0.258 0.380 183s -0.360 0.251 0.369 183s testing: mul_mat, [1, 3, 3, 3] = [1, 1, 3, 3] * [1, 3, 3, 3] 183s x0: [1, 1, 3, 3] 183s -0.898 183s 183s x1: [1, 3, 3, 3] 183s -0.955 183s 0.515 183s 0.478 183s 183s y: [1, 3, 3, 3] 183s 0.858 183s -0.462 183s -0.429 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.687 183s 0.564 183s 183s x1: [1, 1, 1, 1] 183s 0.040 183s 183s y: [2, 1, 1, 1] 183s 0.027 0.023 183s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.728 0.984 183s 0.608 0.985 183s 183s x1: [2, 1, 3, 1] 183s -0.261 0.441 183s 183s y: [2, 1, 3, 1] 183s 0.624 0.276 183s testing: mul_mat, [2, 2, 3, 3] = [2, 2, 3, 3] * [2, 2, 3, 3] 183s x0: [2, 2, 3, 3] 183s -0.064 0.819 183s 0.377 0.951 183s 183s x1: [2, 2, 3, 3] 183s 0.081 -0.223 183s -0.410 0.064 183s 183s y: [2, 2, 3, 3] 183s -0.188 -0.182 183s 0.079 -0.094 183s test-mul-mat0: iter:359/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.234 0.701 183s -0.693 -0.635 183s 0.669 0.884 183s 183s x1: [2, 1, 1, 1] 183s 0.432 0.713 183s 183s y: [3, 1, 1, 1] 183s 0.398 -0.752 0.919 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.627 -0.889 183s 183s x1: [2, 3, 1, 1] 183s -0.034 0.826 183s 0.990 -0.093 183s -0.215 -0.527 183s 183s y: [1, 3, 1, 1] 183s -0.756 183s 0.703 183s 0.333 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.405 -0.859 183s 0.440 0.898 183s 183s x1: [2, 1, 1, 1] 183s -0.269 0.416 183s 183s y: [2, 1, 1, 1] 183s -0.466 0.255 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.589 0.396 183s -0.370 -0.971 183s -0.671 -0.359 183s -0.666 0.762 183s 183s x1: [2, 2, 1, 1] 183s -0.183 -0.106 183s 0.016 0.198 183s 183s y: [4, 2, 1, 1] 183s 0.066 0.170 0.160 0.041 183s 0.069 -0.198 -0.082 0.140 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.937 0.186 -0.699 0.510 183s 0.099 -0.597 -0.689 -0.864 183s -0.002 0.494 0.521 0.729 183s 0.353 -0.374 0.318 -0.860 183s 183s x1: [4, 2, 1, 1] 183s 0.699 -0.475 -0.273 0.525 183s -0.486 0.635 -0.690 -0.012 183s 183s y: [4, 2, 1, 1] 183s 1.024 0.087 0.005 -0.114 183s 0.139 0.059 -0.054 -0.617 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.075 -0.188 0.162 -0.324 183s 0.318 0.641 -0.053 -0.227 183s 0.015 0.629 -0.367 0.763 183s 0.328 0.123 0.298 -0.261 183s 183s x1: [4, 4, 1, 1] 183s -0.535 -0.937 0.536 -0.506 183s -0.296 0.298 -0.807 0.228 183s -0.975 0.718 0.743 0.660 183s -0.972 -0.270 0.927 0.965 183s 183s y: [4, 4, 1, 1] 183s 0.467 -0.684 -1.180 0.001 183s -0.238 0.088 0.654 -0.361 183s -0.156 -0.039 0.668 -0.183 183s -0.039 -0.750 0.212 -0.328 183s test-mul-mat0: iter:360/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.987 0.472 -0.709 0.628 183s 0.101 0.414 -0.210 -0.952 183s -0.953 -0.912 -0.276 -0.180 183s 183s x1: [4, 3, 1, 1] 183s -0.549 -0.329 0.632 -0.831 183s -0.587 0.292 -0.803 0.144 183s 0.219 -0.838 -0.027 -0.856 183s 183s y: [3, 3, 1, 1] 183s -1.668 0.466 0.799 183s 0.218 0.093 0.488 183s -0.698 0.496 0.717 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.133 -0.121 0.087 -0.894 183s 0.023 -0.433 -0.907 -0.505 183s -0.141 0.722 0.597 -0.727 183s -0.488 0.645 -0.680 -0.401 183s 183s x1: [4, 3, 1, 1] 183s 0.851 0.463 0.129 -0.892 183s -0.095 -0.264 -0.441 0.576 183s -0.632 -0.272 0.990 0.660 183s 183s y: [4, 3, 1, 1] 183s 0.866 0.153 0.939 0.153 183s -0.534 0.221 -0.859 -0.055 183s -0.555 -1.127 0.003 -0.804 183s testing: mul_mat, [2, 4, 1, 4] = [4, 2, 1, 4] * [4, 4, 1, 4] 183s x0: [4, 2, 1, 4] 183s -0.420 -0.489 -0.074 -0.811 183s -0.199 0.519 -0.889 -0.537 183s 183s x1: [4, 4, 1, 4] 183s -0.631 0.140 -0.550 0.832 183s -0.731 -0.442 -0.263 0.004 183s 0.117 -0.687 0.372 0.845 183s -0.698 0.032 -0.230 0.436 183s 183s y: [2, 4, 1, 4] 183s -0.437 0.241 183s 0.539 0.148 183s -0.426 -1.164 183s -0.059 0.126 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.195 -0.928 0.090 -0.194 183s 0.842 0.056 -0.465 0.116 183s 0.859 -0.739 -0.711 0.612 183s -0.042 0.794 0.320 0.746 183s 183s x1: [4, 2, 1, 1] 183s -0.227 0.200 0.417 0.574 183s -0.281 0.528 -0.963 -0.324 183s 183s y: [4, 2, 1, 1] 183s -0.303 -0.308 -0.288 0.730 183s -0.569 0.203 -0.145 -0.118 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.710 -0.051 0.219 0.346 183s 183s x1: [4, 4, 1, 1] 183s -0.849 -0.564 -0.607 -0.587 183s -0.290 0.622 0.185 0.910 183s 0.039 -0.241 -0.371 -0.432 183s -0.204 0.306 -0.569 0.991 183s 183s y: [1, 4, 1, 1] 183s 0.295 183s 0.529 183s -0.246 183s 0.347 183s testing: mul_mat, [2, 4, 1, 4] = [1, 2, 1, 4] * [1, 4, 1, 4] 183s x0: [1, 2, 1, 4] 183s -0.576 183s 0.352 183s 183s x1: [1, 4, 1, 4] 183s 0.210 183s 0.013 183s -0.564 183s 0.746 183s 183s y: [2, 4, 1, 4] 183s -0.121 0.074 183s -0.008 0.005 183s 0.325 -0.199 183s -0.429 0.262 183s test-mul-mat0: iter:361/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.873 183s 0.739 183s 183s x1: [1, 1, 1, 1] 183s -0.274 183s 183s y: [2, 1, 1, 1] 183s -0.239 -0.203 183s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 3, 4, 1] 183s 0.349 183s 0.160 183s 0.010 183s 183s x1: [1, 2, 4, 1] 183s -0.578 183s 0.619 183s 183s y: [3, 2, 4, 1] 183s -0.202 -0.092 -0.006 183s 0.216 0.099 0.006 183s testing: mul_mat, [2, 3, 4, 1] = [1, 2, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 2, 4, 1] 183s -0.681 183s 0.865 183s 183s x1: [1, 3, 4, 1] 183s 0.700 183s 0.094 183s 0.540 183s 183s y: [2, 3, 4, 1] 183s -0.476 0.605 183s -0.064 0.082 183s -0.368 0.467 183s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.614 183s 183s x1: [1, 2, 1, 1] 183s 0.209 183s 0.075 183s 183s y: [1, 2, 1, 1] 183s -0.128 183s -0.046 183s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 1, 4, 1] 183s -0.683 183s 183s x1: [1, 1, 4, 1] 183s 0.362 183s 183s y: [1, 1, 4, 1] 183s -0.247 183s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 2, 4, 1] 183s -0.864 183s 0.210 183s 183s x1: [1, 1, 4, 1] 183s 0.608 183s 183s y: [2, 1, 4, 1] 183s -0.526 0.128 183s test-mul-mat0: iter:362/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.884 0.059 -0.725 183s -0.626 0.762 -0.192 183s 183s x1: [3, 1, 1, 1] 183s 0.520 0.497 0.697 183s 183s y: [2, 1, 1, 1] 183s -0.016 -0.081 183s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 1, 4, 1] 183s -0.798 -0.565 -0.422 183s 183s x1: [3, 2, 4, 1] 183s -0.309 -0.836 -0.564 183s -0.452 0.772 0.798 183s 183s y: [1, 2, 4, 1] 183s 0.957 183s -0.413 183s testing: mul_mat, [3, 1, 4, 4] = [3, 3, 4, 4] * [3, 1, 4, 4] 183s x0: [3, 3, 4, 4] 183s -0.842 -0.609 -0.316 183s 0.166 0.252 0.686 183s 0.559 0.379 0.049 183s 183s x1: [3, 1, 4, 4] 183s -0.576 0.446 -0.499 183s 183s y: [3, 1, 4, 4] 183s 0.371 -0.326 -0.177 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.276 0.347 -0.157 183s 0.355 -0.771 0.689 183s 183s x1: [3, 3, 1, 1] 183s -0.816 -0.674 0.213 183s -0.730 -0.328 -0.732 183s 0.650 0.985 0.149 183s 183s y: [2, 3, 1, 1] 183s -0.493 0.376 183s -0.201 -0.511 183s 0.498 -0.426 183s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.969 -0.802 183s 0.228 -0.929 183s -0.879 0.809 183s 183s x1: [2, 3, 4, 1] 183s -0.113 0.581 183s -0.993 -0.301 183s 0.326 0.773 183s 183s y: [3, 3, 4, 1] 183s -0.576 -0.566 0.570 183s -0.720 0.053 0.629 183s -0.304 -0.644 0.339 183s testing: mul_mat, [1, 2, 4, 4] = [3, 1, 4, 4] * [3, 2, 4, 4] 183s x0: [3, 1, 4, 4] 183s 0.585 0.220 -0.946 183s 183s x1: [3, 2, 4, 4] 183s -0.542 -0.552 -0.919 183s 0.268 0.141 0.954 183s 183s y: [1, 2, 4, 4] 183s 0.430 183s -0.714 183s test-mul-mat0: iter:363/500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.774 183s -0.203 183s -0.595 183s 0.195 183s 183s x1: [1, 1, 1, 1] 183s -0.838 183s 183s y: [4, 1, 1, 1] 183s 0.649 0.170 0.499 -0.163 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.225 183s -0.801 183s 183s x1: [1, 4, 1, 1] 183s -0.671 183s -0.981 183s -0.625 183s 0.237 183s 183s y: [2, 4, 1, 1] 183s 0.151 0.537 183s 0.220 0.785 183s 0.140 0.501 183s -0.053 -0.190 183s testing: mul_mat, [1, 2, 1, 2] = [1, 1, 1, 2] * [1, 2, 1, 2] 183s x0: [1, 1, 1, 2] 183s -0.211 183s 183s x1: [1, 2, 1, 2] 183s 0.581 183s -0.890 183s 183s y: [1, 2, 1, 2] 183s -0.122 183s 0.188 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.382 183s 183s x1: [1, 1, 1, 1] 183s 0.138 183s 183s y: [1, 1, 1, 1] 183s -0.053 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.813 183s -0.285 183s 183s x1: [1, 1, 1, 1] 183s -0.349 183s 183s y: [2, 1, 1, 1] 183s -0.284 0.099 183s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 183s x0: [2, 4, 1, 2] 183s 0.407 -0.269 183s 0.536 0.206 183s -0.123 0.831 183s 0.812 0.106 183s 183s x1: [2, 1, 1, 2] 183s 0.328 0.918 183s 183s y: [4, 1, 1, 2] 183s -0.114 0.365 0.723 0.364 183s test-mul-mat0: iter:364/500 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.659 -0.854 -0.327 183s 0.547 0.977 0.779 183s 183s x1: [3, 4, 1, 1] 183s -0.015 0.651 0.833 183s 0.768 0.464 -0.452 183s 0.096 0.383 -0.073 183s -0.594 0.463 -0.666 183s 183s y: [2, 4, 1, 1] 183s -0.819 1.276 183s -0.755 0.522 183s -0.367 0.370 183s 0.213 -0.390 183s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 4, 4, 1] 183s 0.846 -0.337 -0.907 183s 0.393 -0.360 0.872 183s 0.383 0.667 0.531 183s -0.984 0.326 0.181 183s 183s x1: [3, 2, 4, 1] 183s 0.990 0.027 0.659 183s -0.367 0.659 0.650 183s 183s y: [4, 2, 4, 1] 183s 0.230 0.954 0.748 -0.846 183s -1.122 0.186 0.644 0.694 183s testing: mul_mat, [4, 4, 4, 1] = [3, 4, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.899 -0.513 0.780 183s -0.013 -0.077 0.033 183s -0.416 -0.042 0.253 183s -0.272 0.467 0.210 183s 183s x1: [3, 4, 4, 1] 183s -0.775 -0.285 -0.500 183s -0.317 0.667 0.733 183s -0.603 0.691 0.008 183s -0.707 0.193 0.788 183s 183s y: [4, 4, 4, 1] 183s 0.453 0.015 0.208 -0.027 183s 0.515 -0.023 0.289 0.552 183s 0.194 -0.046 0.223 0.489 183s 1.151 0.020 0.485 0.448 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.367 -0.737 -0.504 183s 183s x1: [3, 4, 1, 1] 183s -0.088 0.186 -0.633 183s 0.162 -0.835 0.905 183s -0.365 -0.757 -0.742 183s 0.447 0.832 0.210 183s 183s y: [1, 4, 1, 1] 183s 0.150 183s 0.218 183s 0.798 183s -0.555 183s testing: mul_mat, [4, 3, 4, 1] = [1, 4, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 4, 4, 1] 183s 0.945 183s -0.797 183s 0.045 183s 0.131 183s 183s x1: [1, 3, 4, 1] 183s -0.319 183s -0.697 183s 0.187 183s 183s y: [4, 3, 4, 1] 183s -0.302 0.254 -0.014 -0.042 183s -0.659 0.555 -0.031 -0.092 183s 0.177 -0.149 0.008 0.025 183s testing: mul_mat, [4, 1, 4, 1] = [4, 4, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 4, 4, 1] 183s 0.697 -0.481 -0.261 -0.951 183s 0.222 -0.995 0.033 -0.831 183s 0.529 -0.255 0.430 -0.110 183s 0.904 -0.941 -0.317 0.561 183s 183s x1: [4, 1, 4, 1] 183s -0.793 0.512 0.419 -0.490 183s 183s y: [4, 1, 4, 1] 183s -0.443 -0.265 -0.316 -1.607 183s test-mul-mat0: iter:365/500 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.943 -0.882 -0.119 183s 183s x1: [3, 4, 1, 1] 183s 0.353 -0.742 0.993 183s 0.019 -0.722 0.948 183s -0.266 -0.533 -0.208 183s -0.990 -0.756 0.036 183s 183s y: [1, 4, 1, 1] 183s 0.203 183s 0.506 183s 0.746 183s 1.596 183s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.985 -0.278 0.303 183s -0.727 -0.285 -0.678 183s -0.449 -0.337 0.056 183s 183s x1: [3, 1, 4, 1] 183s -0.015 0.753 0.499 183s 183s y: [3, 1, 4, 1] 183s -0.043 -0.542 -0.219 183s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.174 -0.733 -0.944 183s -0.386 -0.285 0.678 183s 0.895 0.468 0.664 183s 183s x1: [3, 3, 4, 1] 183s 0.053 -0.152 0.068 183s -0.891 0.866 -0.478 183s 0.175 0.129 0.014 183s 183s y: [3, 3, 4, 1] 183s 0.038 0.069 0.021 183s -0.029 -0.227 -0.709 183s -0.138 -0.095 0.226 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.779 0.387 -0.022 183s 0.963 0.120 0.811 183s 183s x1: [3, 3, 1, 1] 183s 0.135 -0.968 0.709 183s -0.202 -0.799 0.781 183s -0.103 0.795 -0.293 183s 183s y: [2, 3, 1, 1] 183s -0.286 0.588 183s -0.484 0.343 183s 0.234 -0.242 183s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.690 -0.190 183s 0.964 -0.059 183s 0.871 0.682 183s 183s x1: [2, 3, 4, 1] 183s 0.508 -0.518 183s -0.341 0.348 183s 0.527 -0.221 183s 183s y: [3, 3, 4, 1] 183s 0.449 0.520 0.089 183s -0.301 -0.349 -0.060 183s 0.406 0.521 0.308 183s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 1, 4, 1] 183s -0.876 0.505 -0.471 183s 183s x1: [3, 2, 4, 1] 183s 0.240 -0.168 0.329 183s -0.366 -0.258 0.329 183s 183s y: [1, 2, 4, 1] 183s -0.450 183s 0.036 183s test-mul-mat0: iter:366/500 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.015 -0.341 -0.144 183s 183s x1: [3, 2, 1, 1] 183s 0.470 -0.250 0.517 183s 0.030 -0.593 -0.472 183s 183s y: [1, 2, 1, 1] 183s 0.018 183s 0.271 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.953 -0.941 0.003 183s -0.518 -0.075 0.225 183s 0.046 0.031 -0.721 183s 183s x1: [3, 1, 1, 1] 183s -0.204 0.946 0.935 183s 183s y: [3, 1, 1, 1] 183s -1.081 0.245 -0.655 183s testing: mul_mat, [1, 3, 1, 4] = [3, 1, 1, 4] * [3, 3, 1, 4] 183s x0: [3, 1, 1, 4] 183s -0.614 0.738 0.803 183s 183s x1: [3, 3, 1, 4] 183s -0.060 0.771 -0.925 183s -0.837 0.628 -0.697 183s -0.321 -0.526 0.773 183s 183s y: [1, 3, 1, 4] 183s -0.137 183s 0.418 183s 0.430 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.625 0.183 0.787 183s 0.343 -0.822 -0.957 183s 0.442 -0.183 -0.531 183s 183s x1: [3, 1, 1, 1] 183s -0.852 -0.452 0.433 183s 183s y: [3, 1, 1, 1] 183s -0.274 -0.335 -0.524 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.590 -0.698 0.538 183s -0.364 0.390 -0.370 183s -0.111 -0.918 0.318 183s -0.343 0.885 -0.168 183s 183s x1: [3, 3, 1, 1] 183s 0.332 0.680 -0.803 183s 0.039 -0.056 -0.773 183s 0.425 -0.318 -0.970 183s 183s y: [4, 3, 1, 1] 183s -1.102 0.442 -0.916 0.623 183s -0.400 0.250 -0.199 0.067 183s -0.551 0.080 -0.063 -0.263 183s testing: mul_mat, [2, 3, 1, 4] = [4, 2, 1, 4] * [4, 3, 1, 4] 183s x0: [4, 2, 1, 4] 183s -0.683 -0.762 -0.357 -0.995 183s 0.388 0.466 0.190 0.981 183s 183s x1: [4, 3, 1, 4] 183s -0.745 -0.339 -0.726 0.437 183s -0.162 0.091 0.225 -0.118 183s 0.560 -0.443 -0.439 0.757 183s 183s y: [2, 3, 1, 4] 183s 0.592 -0.156 183s 0.079 -0.094 183s -0.641 0.670 183s test-mul-mat0: iter:367/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.118 0.635 -0.329 183s -0.890 0.806 0.668 183s 183s x1: [3, 1, 1, 1] 183s -0.538 0.035 0.687 183s 183s y: [2, 1, 1, 1] 183s -0.267 0.966 183s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 2, 2, 1] 183s -0.056 -0.655 0.371 183s -0.491 0.408 -0.357 183s 183s x1: [3, 2, 2, 1] 183s -0.715 0.636 -0.682 183s -0.266 -0.650 0.504 183s 183s y: [2, 2, 2, 1] 183s -0.629 0.853 183s 0.627 -0.314 183s testing: mul_mat, [4, 2, 2, 2] = [3, 4, 2, 2] * [3, 2, 2, 2] 183s x0: [3, 4, 2, 2] 183s 0.419 0.080 -0.815 183s 0.993 0.880 0.780 183s 0.686 -0.685 -0.954 183s 0.488 -0.444 0.979 183s 183s x1: [3, 2, 2, 2] 183s 0.693 -0.565 -0.734 183s 0.803 -0.759 0.934 183s 183s y: [4, 2, 2, 2] 183s 0.843 -0.382 1.562 -0.130 183s -0.486 0.857 0.179 1.644 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.125 -0.050 0.610 183s 0.132 -0.395 -0.821 183s 183s x1: [3, 4, 1, 1] 183s 0.754 0.865 -0.421 183s 0.284 -0.169 -0.605 183s 0.028 -0.138 0.581 183s 0.331 0.329 -0.353 183s 183s y: [2, 4, 1, 1] 183s -0.395 0.103 183s -0.396 0.601 183s 0.358 -0.419 183s -0.273 0.203 183s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.854 -0.479 183s -0.874 0.041 183s -0.988 0.839 183s -0.023 -0.898 183s 183s x1: [2, 3, 2, 1] 183s -0.885 0.773 183s 0.724 0.671 183s 0.425 0.825 183s 183s y: [4, 3, 2, 1] 183s 0.386 0.804 1.522 -0.674 183s -0.940 -0.605 -0.152 -0.620 183s -0.758 -0.338 0.273 -0.750 183s testing: mul_mat, [1, 2, 2, 2] = [4, 1, 2, 2] * [4, 2, 2, 2] 183s x0: [4, 1, 2, 2] 183s 0.205 -0.309 0.585 0.640 183s 183s x1: [4, 2, 2, 2] 183s 0.492 0.100 -0.237 0.265 183s -0.175 -0.565 -0.311 -0.351 183s 183s y: [1, 2, 2, 2] 183s 0.100 183s -0.268 183s test-mul-mat0: iter:368/500 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.502 -0.011 -0.034 183s -0.842 0.431 -0.607 183s -0.081 -0.364 0.084 183s 183s x1: [3, 1, 1, 1] 183s -0.021 0.759 0.194 183s 183s y: [3, 1, 1, 1] 183s -0.004 0.227 -0.258 183s testing: mul_mat, [4, 3, 3, 1] = [3, 4, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.614 -0.806 -0.625 183s -0.882 -0.530 -0.161 183s -0.143 -0.973 -0.975 183s -0.164 0.481 0.556 183s 183s x1: [3, 3, 3, 1] 183s -0.496 -0.724 -0.536 183s -0.261 0.557 0.186 183s 0.979 0.454 0.531 183s 183s y: [4, 3, 3, 1] 183s 0.614 0.907 1.298 -0.565 183s -0.726 -0.095 -0.686 0.414 183s -0.096 -1.189 -1.099 0.352 183s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.134 0.581 0.689 183s -0.068 0.483 0.964 183s -0.570 -0.835 0.981 183s 0.573 -0.473 0.779 183s 183s x1: [3, 4, 3, 1] 183s 0.019 0.601 0.156 183s 0.044 -0.563 -0.362 183s -0.401 0.046 -0.188 183s 0.798 0.902 -0.725 183s 183s y: [4, 4, 3, 1] 183s 0.459 0.440 -0.359 -0.152 183s -0.571 -0.625 0.090 0.009 183s -0.157 -0.132 0.006 -0.398 183s 0.131 -0.318 -1.920 -0.535 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.294 -0.807 0.532 183s -0.562 -0.841 -0.568 183s -0.115 0.532 0.219 183s -0.129 0.439 0.440 183s 183s x1: [3, 4, 1, 1] 183s 0.432 -0.723 0.137 183s 0.413 0.849 0.664 183s 0.192 0.345 -0.834 183s 0.086 0.672 0.156 183s 183s y: [4, 4, 1, 1] 183s 0.783 0.288 -0.404 -0.313 183s -0.211 -1.323 0.549 0.611 183s -0.666 0.076 -0.022 -0.240 183s -0.434 -0.702 0.381 0.352 183s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.752 0.683 -0.208 -0.230 183s 183s x1: [4, 3, 3, 1] 183s -0.790 -0.866 -0.071 0.677 183s 0.980 0.901 0.109 -0.743 183s 0.039 -0.478 -0.894 -0.298 183s 183s y: [1, 3, 3, 1] 183s -0.139 183s 0.028 183s -0.101 183s testing: mul_mat, [2, 4, 3, 1] = [1, 2, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 2, 3, 1] 183s 0.602 183s -0.643 183s 183s x1: [1, 4, 3, 1] 183s -0.808 183s -0.220 183s -0.152 183s 0.985 183s 183s y: [2, 4, 3, 1] 183s -0.487 0.520 183s -0.133 0.141 183s -0.092 0.098 183s 0.593 -0.633 183s test-mul-mat0: iter:369/500 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.831 0.704 0.707 183s 0.334 0.640 -0.405 183s -0.069 0.059 -0.265 183s 183s x1: [3, 3, 1, 1] 183s 0.256 -0.766 -0.756 183s -0.504 -0.403 0.528 183s -0.141 -0.017 0.719 183s 183s y: [3, 3, 1, 1] 183s -0.860 -0.098 0.137 183s -0.329 -0.640 -0.129 183s 0.379 -0.349 -0.182 183s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 1, 3, 1] 183s -0.807 0.724 -0.262 183s 183s x1: [3, 3, 3, 1] 183s 0.841 0.401 -0.662 183s -0.802 0.596 0.079 183s 0.982 0.544 -0.930 183s 183s y: [1, 3, 3, 1] 183s -0.215 183s 1.057 183s -0.155 183s testing: mul_mat, [2, 1, 3, 3] = [3, 2, 3, 3] * [3, 1, 3, 3] 183s x0: [3, 2, 3, 3] 183s 0.930 -0.689 -0.289 183s -0.855 -0.424 -0.064 183s 183s x1: [3, 1, 3, 3] 183s -0.786 -0.735 -0.775 183s 183s y: [2, 1, 3, 3] 183s 0.000 1.034 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.153 -0.906 0.970 183s -0.475 -0.704 0.876 183s -0.501 -0.837 -0.049 183s 183s x1: [3, 2, 1, 1] 183s 0.446 0.249 0.202 183s -0.367 -0.077 -0.102 183s 183s y: [3, 2, 1, 1] 183s 0.039 -0.210 -0.441 183s -0.086 0.139 0.253 183s testing: mul_mat, [3, 3, 3, 1] = [3, 3, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.924 -0.087 0.738 183s -0.837 0.797 -0.062 183s -0.765 -0.835 0.449 183s 183s x1: [3, 3, 3, 1] 183s -0.163 0.634 0.931 183s -0.175 0.142 0.284 183s 0.664 -0.970 0.291 183s 183s y: [3, 3, 3, 1] 183s 0.782 0.584 0.013 183s 0.359 0.242 0.144 183s -0.314 -1.347 0.432 183s testing: mul_mat, [4, 3, 3, 3] = [3, 4, 3, 3] * [3, 3, 3, 3] 183s x0: [3, 4, 3, 3] 183s 0.040 0.701 -0.467 183s 0.661 -0.849 -0.036 183s -0.575 0.962 -0.985 183s -0.315 -0.611 -0.962 183s 183s x1: [3, 3, 3, 3] 183s -0.534 -0.803 -0.422 183s 0.749 -0.728 -0.259 183s 0.984 0.185 -0.462 183s 183s y: [4, 3, 3, 3] 183s -0.388 0.344 -0.051 1.065 183s -0.360 1.122 -0.876 0.458 183s 0.385 0.510 0.068 0.021 183s test-mul-mat0: iter:370/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.421 -0.202 183s -0.673 0.877 183s 183s x1: [2, 4, 1, 1] 183s -0.174 -0.412 183s 0.854 0.935 183s -0.047 -0.590 183s 0.399 0.533 183s 183s y: [2, 4, 1, 1] 183s 0.157 -0.244 183s -0.549 0.245 183s 0.139 -0.485 183s -0.276 0.199 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.130 0.205 183s 183s x1: [2, 2, 1, 1] 183s -0.122 0.923 183s -0.623 0.607 183s 183s y: [1, 2, 1, 1] 183s 0.174 183s 0.044 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.094 0.323 183s 0.112 0.182 183s 0.565 -0.230 183s 183s x1: [2, 1, 1, 1] 183s 0.179 -0.855 183s 183s y: [3, 1, 1, 1] 183s -0.293 -0.135 0.298 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.532 -0.071 183s 183s x1: [2, 3, 1, 1] 183s 0.118 0.268 183s 0.596 0.706 183s 0.122 0.530 183s 183s y: [1, 3, 1, 1] 183s 0.044 183s 0.267 183s 0.027 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.011 183s 0.125 183s 0.385 183s 183s x1: [1, 2, 1, 1] 183s -0.808 183s -0.202 183s 183s y: [3, 2, 1, 1] 183s 0.009 -0.101 -0.311 183s 0.002 -0.025 -0.078 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.584 -0.108 -0.320 183s 183s x1: [3, 1, 1, 1] 183s 0.867 0.049 0.762 183s 183s y: [1, 1, 1, 1] 183s 0.257 183s test-mul-mat0: iter:371/500 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.597 -0.667 0.846 183s -0.474 -0.475 -0.355 183s -0.966 0.514 0.770 183s 183s x1: [3, 3, 1, 1] 183s -0.640 0.701 -0.150 183s 0.925 -0.530 0.968 183s 0.193 -0.934 0.674 183s 183s y: [3, 3, 1, 1] 183s -0.975 0.024 0.863 183s 1.725 -0.531 -0.421 183s 1.308 0.113 -0.147 183s testing: mul_mat, [4, 3, 3, 1] = [3, 4, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 4, 3, 1] 183s -0.482 0.700 0.165 183s 0.937 0.081 0.983 183s 0.118 0.937 -0.614 183s 0.191 0.473 0.553 183s 183s x1: [3, 3, 3, 1] 183s 0.419 0.381 -0.181 183s 0.181 -0.144 -0.597 183s -0.927 0.536 0.166 183s 183s y: [4, 3, 3, 1] 183s 0.035 0.245 0.518 0.160 183s -0.286 -0.429 0.253 -0.364 183s 0.849 -0.661 0.291 0.169 183s testing: mul_mat, [4, 4, 3, 4] = [3, 4, 3, 4] * [3, 4, 3, 4] 183s x0: [3, 4, 3, 4] 183s 0.275 0.343 -0.753 183s 0.852 0.871 0.091 183s -0.994 0.524 -0.940 183s 0.887 -0.429 -0.551 183s 183s x1: [3, 4, 3, 4] 183s 0.952 -0.906 0.620 183s -0.661 0.285 0.093 183s 0.891 0.586 -0.310 183s -0.720 0.358 -0.887 183s 183s y: [4, 4, 3, 4] 183s -0.516 0.079 -2.003 0.892 183s -0.154 -0.306 0.720 -0.760 183s 0.679 1.242 -0.288 0.709 183s 0.593 -0.383 1.738 -0.304 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.102 0.916 -0.600 183s -0.182 -0.398 0.442 183s 0.843 -0.975 -0.161 183s 183s x1: [3, 4, 1, 1] 183s -0.523 -0.912 0.716 183s -0.785 -0.791 -0.928 183s 0.357 -0.035 -0.639 183s -0.385 -0.764 -0.354 183s 183s y: [3, 4, 1, 1] 183s -1.212 0.775 0.333 183s -0.089 0.048 0.259 183s 0.315 -0.334 0.439 183s -0.449 0.218 0.478 183s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.848 -0.759 0.355 183s -0.447 -0.790 -0.216 183s 183s x1: [3, 3, 3, 1] 183s -0.941 -0.198 -0.457 183s -0.380 0.968 0.693 183s 0.440 0.440 -0.103 183s 183s y: [2, 3, 3, 1] 183s 0.787 0.676 183s -0.167 -0.745 183s -0.743 -0.522 183s testing: mul_mat, [3, 3, 3, 4] = [2, 3, 3, 4] * [2, 3, 3, 4] 183s x0: [2, 3, 3, 4] 183s 0.834 0.257 183s 0.758 0.657 183s 0.699 0.192 183s 183s x1: [2, 3, 3, 4] 183s 0.808 -0.273 183s 0.719 0.913 183s 0.664 0.410 183s 183s y: [3, 3, 3, 4] 183s 0.604 0.433 0.513 183s 0.834 1.145 0.678 183s 0.659 0.773 0.543 183s test-mul-mat0: iter:372/500 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.609 183s 183s x1: [1, 1, 1, 1] 183s -0.720 183s 183s y: [1, 1, 1, 1] 183s -0.439 183s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.876 183s 183s x1: [1, 1, 2, 1] 183s -0.735 183s 183s y: [1, 1, 2, 1] 183s 0.644 183s testing: mul_mat, [2, 1, 2, 2] = [1, 2, 2, 2] * [1, 1, 2, 2] 183s x0: [1, 2, 2, 2] 183s 0.280 183s 0.577 183s 183s x1: [1, 1, 2, 2] 183s 0.144 183s 183s y: [2, 1, 2, 2] 183s 0.040 0.083 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.741 183s -0.833 183s 0.640 183s 0.955 183s 183s x1: [1, 2, 1, 1] 183s -0.492 183s -0.628 183s 183s y: [4, 2, 1, 1] 183s 0.365 0.410 -0.315 -0.470 183s 0.465 0.523 -0.402 -0.600 183s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 3, 2, 1] 183s 0.213 0.799 -0.433 0.561 183s 0.889 0.154 -0.566 -0.477 183s 0.463 0.315 -0.108 0.046 183s 183s x1: [4, 1, 2, 1] 183s -0.455 0.883 0.251 -0.175 183s 183s y: [3, 1, 2, 1] 183s 0.402 -0.326 0.032 183s testing: mul_mat, [2, 4, 2, 2] = [3, 2, 2, 2] * [3, 4, 2, 2] 183s x0: [3, 2, 2, 2] 183s 0.651 0.776 -0.526 183s -0.248 0.990 0.139 183s 183s x1: [3, 4, 2, 2] 183s -0.444 0.820 0.506 183s 0.416 -0.091 0.169 183s 0.735 -0.878 0.058 183s 0.198 0.921 -0.788 183s 183s y: [2, 4, 2, 2] 183s 0.081 0.993 183s 0.111 -0.169 183s -0.234 -1.043 183s 1.259 0.753 183s test-mul-mat0: iter:373/500 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.716 183s 0.204 183s -0.415 183s 183s x1: [1, 4, 1, 1] 183s 0.563 183s 0.359 183s -0.492 183s 0.214 183s 183s y: [3, 4, 1, 1] 183s -0.403 0.115 -0.233 183s -0.257 0.073 -0.149 183s 0.352 -0.100 0.204 183s -0.153 0.044 -0.089 183s testing: mul_mat, [3, 3, 4, 1] = [1, 3, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 3, 4, 1] 183s -0.047 183s 0.402 183s -0.954 183s 183s x1: [1, 3, 4, 1] 183s 0.424 183s -0.322 183s -0.868 183s 183s y: [3, 3, 4, 1] 183s -0.020 0.171 -0.405 183s 0.015 -0.130 0.308 183s 0.041 -0.349 0.829 183s testing: mul_mat, [1, 3, 4, 1] = [1, 1, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 1, 4, 1] 183s -0.488 183s 183s x1: [1, 3, 4, 1] 183s -0.041 183s 0.112 183s 0.926 183s 183s y: [1, 3, 4, 1] 183s 0.020 183s -0.055 183s -0.452 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.685 183s 0.126 183s -0.785 183s -0.959 183s 183s x1: [1, 1, 1, 1] 183s 0.724 183s 183s y: [4, 1, 1, 1] 183s 0.496 0.091 -0.568 -0.694 183s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.047 0.479 0.676 0.943 183s 0.165 -0.770 0.088 -0.592 183s 183s x1: [4, 1, 4, 1] 183s 0.097 0.430 -0.319 0.432 183s 183s y: [2, 1, 4, 1] 183s 0.402 -0.599 183s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 3, 4, 1] 183s -0.141 0.456 183s 0.996 0.995 183s 0.570 -0.162 183s 183s x1: [2, 4, 4, 1] 183s 0.914 0.368 183s -0.345 0.590 183s -0.544 -0.403 183s 0.999 0.268 183s 183s y: [3, 4, 4, 1] 183s 0.039 1.277 0.461 183s 0.318 0.243 -0.293 183s -0.107 -0.943 -0.245 183s -0.019 1.261 0.526 183s test-mul-mat0: iter:374/500 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.118 183s 0.240 183s -0.923 183s 183s x1: [1, 3, 1, 1] 183s 0.762 183s -0.115 183s 0.670 183s 183s y: [3, 3, 1, 1] 183s -0.090 0.183 -0.704 183s 0.014 -0.028 0.106 183s -0.079 0.161 -0.618 183s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 2, 3, 1] 183s 0.396 183s 0.860 183s 183s x1: [1, 3, 3, 1] 183s -0.123 183s -0.922 183s -0.759 183s 183s y: [2, 3, 3, 1] 183s -0.049 -0.106 183s -0.365 -0.794 183s -0.301 -0.653 183s testing: mul_mat, [1, 2, 3, 1] = [1, 1, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 1, 3, 1] 183s -0.615 183s 183s x1: [1, 2, 3, 1] 183s 0.136 183s 0.019 183s 183s y: [1, 2, 3, 1] 183s -0.084 183s -0.012 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.368 183s 0.295 183s 183s x1: [1, 1, 1, 1] 183s -0.945 183s 183s y: [2, 1, 1, 1] 183s -0.348 -0.279 183s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 4, 3, 1] 183s -0.513 0.294 183s 0.898 0.128 183s 0.268 -0.899 183s 0.121 -0.813 183s 183s x1: [2, 1, 3, 1] 183s -0.847 -0.755 183s 183s y: [4, 1, 3, 1] 183s 0.213 -0.857 0.452 0.511 183s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.866 -0.720 0.882 -0.529 183s -0.624 -0.269 -0.725 0.630 183s -0.047 0.915 0.786 0.067 183s 0.948 -0.572 -0.474 0.934 183s 183s x1: [4, 2, 3, 1] 183s 0.967 -0.851 0.151 -0.639 183s 0.358 -0.356 -0.805 0.844 183s 183s y: [4, 2, 3, 1] 183s 0.247 -0.886 -0.749 0.735 183s -1.209 0.988 -0.918 1.713 183s test-mul-mat0: iter:375/500 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.315 -0.341 -0.665 -0.541 183s -0.567 -0.102 -0.036 -0.877 183s 0.889 0.351 -0.913 -0.334 183s -0.737 -0.478 0.072 -0.758 183s 183s x1: [4, 2, 1, 1] 183s 0.205 0.845 -0.920 0.514 183s 0.141 -0.855 0.865 -0.167 183s 183s y: [4, 2, 1, 1] 183s -0.019 -0.620 1.147 -1.010 183s -0.237 0.122 -0.908 0.493 183s testing: mul_mat, [2, 4, 2, 1] = [4, 2, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 2, 2, 1] 183s 0.456 0.055 -0.121 -0.612 183s -0.955 0.460 0.399 -0.451 183s 183s x1: [4, 4, 2, 1] 183s -0.109 -0.926 0.124 0.785 183s -0.264 -0.068 -0.010 -0.419 183s 0.011 -0.497 0.722 0.157 183s -0.632 -0.445 0.209 0.053 183s 183s y: [2, 4, 2, 1] 183s -0.597 -0.627 183s 0.134 0.406 183s -0.206 -0.022 183s -0.371 0.458 183s testing: mul_mat, [4, 2, 2, 3] = [4, 4, 2, 3] * [4, 2, 2, 3] 183s x0: [4, 4, 2, 3] 183s 0.431 0.078 -0.766 -0.120 183s -0.397 0.047 -0.115 0.044 183s 0.596 -0.346 0.455 0.096 183s 0.732 0.348 0.195 -0.440 183s 183s x1: [4, 2, 2, 3] 183s 0.482 -0.402 0.751 0.958 183s -0.632 -0.918 0.785 -0.775 183s 183s y: [4, 2, 2, 3] 183s -0.514 -0.255 0.860 -0.063 183s -0.853 0.084 0.224 -0.288 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.666 -0.258 -0.608 -0.911 183s -0.833 0.566 -0.349 -0.232 183s 183s x1: [4, 4, 1, 1] 183s -0.543 0.783 -0.623 -0.922 183s -0.741 0.453 -0.005 0.840 183s -0.863 -0.061 0.873 0.622 183s -0.530 0.559 0.570 0.059 183s 183s y: [2, 4, 1, 1] 183s 1.378 1.327 183s -0.386 0.681 183s -0.507 0.236 183s -0.192 0.546 183s testing: mul_mat, [3, 4, 2, 1] = [2, 3, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 3, 2, 1] 183s -0.674 0.896 183s 0.325 -0.616 183s 0.098 -0.168 183s 183s x1: [2, 4, 2, 1] 183s -0.093 -0.173 183s -0.073 -0.616 183s 0.784 -0.263 183s -0.158 0.567 183s 183s y: [3, 4, 2, 1] 183s -0.092 0.076 0.020 183s -0.502 0.355 0.096 183s -0.764 0.417 0.121 183s 0.615 -0.401 -0.111 183s testing: mul_mat, [1, 2, 2, 3] = [3, 1, 2, 3] * [3, 2, 2, 3] 183s x0: [3, 1, 2, 3] 183s 0.098 -0.027 0.829 183s 183s x1: [3, 2, 2, 3] 183s 0.884 0.710 0.132 183s -0.289 -0.363 0.516 183s 183s y: [1, 2, 2, 3] 183s 0.177 183s 0.409 183s test-mul-mat0: iter:376/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.998 -0.714 0.059 183s 0.596 0.757 -0.843 183s 183s x1: [3, 1, 1, 1] 183s -0.436 0.036 -0.668 183s 183s y: [2, 1, 1, 1] 183s -0.500 0.331 183s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.674 -0.020 -0.348 183s 0.362 0.312 0.107 183s 0.360 0.598 -0.834 183s 183s x1: [3, 2, 3, 1] 183s -0.432 0.586 0.318 183s -0.971 -0.307 -0.632 183s 183s y: [3, 2, 3, 1] 183s 0.168 0.060 -0.070 183s 0.881 -0.515 -0.005 183s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.783 0.116 -0.539 183s 183s x1: [3, 3, 3, 1] 183s 0.641 0.569 0.677 183s 0.728 0.895 -0.344 183s -0.620 0.257 0.968 183s 183s y: [1, 3, 3, 1] 183s 0.203 183s 0.859 183s -0.978 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.546 0.247 -0.734 183s -0.673 -0.351 0.773 183s -0.328 0.123 -0.398 183s -0.974 0.108 0.891 183s 183s x1: [3, 1, 1, 1] 183s 0.532 0.837 -0.239 183s 183s y: [4, 1, 1, 1] 183s 0.091 -0.837 0.024 -0.641 183s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.102 -0.748 -0.464 -0.744 183s 0.642 -0.606 0.149 0.945 183s 0.045 0.061 -0.534 -0.330 183s -0.495 0.440 0.440 -0.964 183s 183s x1: [4, 3, 3, 1] 183s -0.477 -0.278 -0.503 -0.803 183s -0.191 -0.249 0.980 0.925 183s 0.212 0.404 -0.391 0.349 183s 183s y: [4, 3, 3, 1] 183s 0.990 -0.971 0.495 0.666 183s -0.975 1.048 -0.852 -0.475 183s -0.359 0.163 0.128 -0.435 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.258 0.925 0.888 0.167 183s 183s x1: [4, 4, 3, 1] 183s 0.892 0.100 -0.180 -0.376 183s -0.741 0.300 0.543 -0.997 183s -0.296 0.371 -0.171 0.230 183s -0.083 0.488 -0.142 -0.895 183s 183s y: [1, 4, 3, 1] 183s -0.360 183s 0.785 183s 0.306 183s 0.197 183s test-mul-mat0: iter:377/500 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.653 183s 183s x1: [1, 1, 1, 1] 183s -0.506 183s 183s y: [1, 1, 1, 1] 183s 0.330 183s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.632 183s -0.170 183s 183s x1: [1, 1, 3, 1] 183s 0.298 183s 183s y: [2, 1, 3, 1] 183s -0.189 -0.051 183s testing: mul_mat, [1, 2, 3, 2] = [1, 1, 3, 2] * [1, 2, 3, 2] 183s x0: [1, 1, 3, 2] 183s -0.600 183s 183s x1: [1, 2, 3, 2] 183s -0.557 183s 0.247 183s 183s y: [1, 2, 3, 2] 183s 0.334 183s -0.148 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.695 183s 183s x1: [1, 1, 1, 1] 183s 0.859 183s 183s y: [1, 1, 1, 1] 183s 0.597 183s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.016 183s 0.733 183s 183s x1: [1, 1, 3, 1] 183s -0.726 183s 183s y: [2, 1, 3, 1] 183s 0.012 -0.533 183s testing: mul_mat, [2, 1, 3, 2] = [2, 2, 3, 2] * [2, 1, 3, 2] 183s x0: [2, 2, 3, 2] 183s 0.299 -0.474 183s -0.878 -0.867 183s 183s x1: [2, 1, 3, 2] 183s 0.775 -0.999 183s 183s y: [2, 1, 3, 2] 183s 0.705 0.186 183s test-mul-mat0: iter:378/500 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.741 0.105 0.118 -0.098 183s 183s x1: [4, 4, 1, 1] 183s 0.458 0.157 0.640 -0.243 183s 0.279 -0.834 -0.110 0.798 183s -0.449 -0.796 -0.988 0.209 183s 0.673 0.393 -0.491 -0.875 183s 183s y: [1, 4, 1, 1] 183s -0.223 183s -0.386 183s 0.112 183s -0.429 183s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.202 0.563 0.140 0.688 183s 0.362 0.691 0.892 0.373 183s -0.100 0.565 -0.234 0.409 183s 183s x1: [4, 1, 3, 1] 183s 0.859 0.730 -0.194 0.524 183s 183s y: [3, 1, 3, 1] 183s 0.571 0.838 0.587 183s testing: mul_mat, [3, 3, 3, 2] = [4, 3, 3, 2] * [4, 3, 3, 2] 183s x0: [4, 3, 3, 2] 183s -0.340 0.659 0.335 0.391 183s 0.547 -0.109 -0.911 0.890 183s -0.703 0.475 0.381 -0.285 183s 183s x1: [4, 3, 3, 2] 183s -0.032 0.140 0.275 0.868 183s -0.295 -0.959 0.277 -0.605 183s 0.315 -0.055 0.190 0.707 183s 183s y: [3, 3, 3, 2] 183s 0.535 0.489 -0.053 183s -0.676 -0.848 0.029 183s 0.197 0.634 -0.376 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.633 -0.007 -0.518 0.076 183s -0.138 -0.587 0.271 -0.936 183s -0.874 -0.188 -0.519 -0.774 183s -0.847 -0.579 0.723 -0.884 183s 183s x1: [4, 3, 1, 1] 183s 0.657 0.940 -0.709 0.269 183s -0.243 0.500 -0.845 0.847 183s -0.023 0.158 -0.409 -0.773 183s 183s y: [4, 3, 1, 1] 183s 0.798 -1.086 -0.590 -1.851 183s 0.345 -1.282 -0.098 -1.444 183s 0.138 0.523 0.801 0.316 183s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 3, 3, 1] 183s 0.564 -0.234 0.239 -0.213 183s 0.109 0.848 0.067 -0.650 183s -0.362 -0.026 0.033 0.319 183s 183s x1: [4, 4, 3, 1] 183s 0.468 0.467 0.773 0.408 183s 0.758 0.041 -0.835 0.258 183s 0.196 -0.988 -0.765 -0.646 183s -0.397 -0.538 0.315 -0.764 183s 183s y: [3, 4, 3, 1] 183s 0.252 0.234 -0.026 183s 0.164 -0.106 -0.221 183s 0.297 -0.448 -0.277 183s 0.139 0.018 -0.076 183s testing: mul_mat, [1, 4, 3, 2] = [3, 1, 3, 2] * [3, 4, 3, 2] 183s x0: [3, 1, 3, 2] 183s -0.576 0.214 0.773 183s 183s x1: [3, 4, 3, 2] 183s 0.809 0.602 -0.364 183s -0.158 -0.610 -0.014 183s -0.839 -0.159 -0.912 183s 0.232 -0.271 0.349 183s 183s y: [1, 4, 3, 2] 183s -0.618 183s -0.050 183s -0.256 183s 0.078 183s test-mul-mat0: iter:379/500 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.142 -0.315 -0.880 183s -0.057 -0.333 0.674 183s -0.434 0.975 -0.226 183s 183s x1: [3, 4, 1, 1] 183s -0.001 -0.961 -0.911 183s -0.837 -0.950 -0.770 183s -0.717 0.755 0.639 183s -0.293 -0.031 0.411 183s 183s y: [3, 4, 1, 1] 183s 1.104 -0.293 -0.730 183s 1.095 -0.154 -0.388 183s -0.698 0.219 0.903 183s -0.311 0.304 0.004 183s testing: mul_mat, [2, 3, 2, 1] = [3, 2, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 2, 2, 1] 183s 0.966 0.203 0.193 183s -0.176 0.888 0.313 183s 183s x1: [3, 3, 2, 1] 183s 0.683 0.424 -0.238 183s 0.969 -0.337 0.346 183s -0.328 0.153 0.834 183s 183s y: [2, 3, 2, 1] 183s 0.700 0.182 183s 0.935 -0.361 183s -0.126 0.454 183s testing: mul_mat, [2, 2, 2, 3] = [3, 2, 2, 3] * [3, 2, 2, 3] 183s x0: [3, 2, 2, 3] 183s -0.255 -0.918 -0.553 183s 0.073 0.042 -0.928 183s 183s x1: [3, 2, 2, 3] 183s -0.985 0.953 -0.478 183s 0.985 -0.384 0.868 183s 183s y: [2, 2, 2, 3] 183s -0.360 0.412 183s -0.379 -0.750 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.188 0.723 -0.758 183s -0.765 0.830 0.181 183s 0.071 0.923 0.990 183s 183s x1: [3, 2, 1, 1] 183s 0.609 -0.316 -0.813 183s 0.459 -0.804 -0.371 183s 183s y: [3, 2, 1, 1] 183s 0.273 -0.876 -1.054 183s -0.386 -1.086 -1.077 183s testing: mul_mat, [4, 3, 2, 1] = [3, 4, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 4, 2, 1] 183s -0.498 0.023 -0.266 183s -0.212 -0.189 0.280 183s 0.740 0.037 -0.966 183s 0.314 -0.147 -0.276 183s 183s x1: [3, 3, 2, 1] 183s 0.356 -0.379 0.800 183s -0.176 -0.836 0.307 183s 0.168 0.796 0.127 183s 183s y: [4, 3, 2, 1] 183s -0.399 0.220 -0.524 -0.053 183s -0.013 0.281 -0.459 -0.017 183s -0.099 -0.150 0.031 -0.099 183s testing: mul_mat, [1, 3, 2, 3] = [4, 1, 2, 3] * [4, 3, 2, 3] 183s x0: [4, 1, 2, 3] 183s 0.808 0.147 -0.175 0.770 183s 183s x1: [4, 3, 2, 3] 183s -0.321 -0.590 0.961 -0.919 183s -0.284 -0.880 -0.058 0.218 183s -0.092 -0.317 -0.468 0.931 183s 183s y: [1, 3, 2, 3] 183s -1.221 183s -0.182 183s 0.678 183s test-mul-mat0: iter:380/500 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.413 0.015 0.917 183s 0.637 -0.277 0.268 183s -0.835 0.423 -0.928 183s 0.265 0.575 -0.974 183s 183s x1: [3, 3, 1, 1] 183s -0.014 0.322 -0.914 183s -0.206 -0.531 -0.089 183s -0.437 -0.818 -0.274 183s 183s y: [4, 3, 1, 1] 183s -0.840 -0.343 0.997 1.072 183s -0.174 -0.008 0.030 -0.273 183s -0.444 -0.125 0.273 -0.318 183s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 2, 2, 1] 183s -0.440 -0.196 0.234 183s 0.825 -0.622 0.260 183s 183s x1: [3, 4, 2, 1] 183s 0.192 -0.686 0.235 183s -0.836 -0.141 -0.480 183s -0.480 0.371 0.920 183s 0.506 -0.307 -0.994 183s 183s y: [2, 4, 2, 1] 183s 0.105 0.646 183s 0.283 -0.726 183s 0.353 -0.388 183s -0.395 0.350 183s testing: mul_mat, [4, 2, 2, 4] = [3, 4, 2, 4] * [3, 2, 2, 4] 183s x0: [3, 4, 2, 4] 183s -0.372 -0.612 -0.316 183s 0.264 -0.953 0.405 183s 0.159 -0.278 -0.182 183s 0.549 -0.422 0.782 183s 183s x1: [3, 2, 2, 4] 183s 0.700 -0.078 -0.065 183s 0.206 0.615 -0.059 183s 183s y: [4, 2, 2, 4] 183s -0.192 0.233 0.145 0.367 183s -0.435 -0.555 -0.127 -0.192 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.549 -0.117 0.887 183s 183s x1: [3, 4, 1, 1] 183s -0.460 0.252 -0.879 183s -0.650 0.383 -0.517 183s -0.229 -0.074 0.036 183s -0.243 -0.983 0.447 183s 183s y: [1, 4, 1, 1] 183s -1.062 183s -0.860 183s -0.085 183s 0.379 183s testing: mul_mat, [3, 3, 2, 1] = [1, 3, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.335 183s 0.904 183s -0.077 183s 183s x1: [1, 3, 2, 1] 183s -0.312 183s -0.212 183s 0.325 183s 183s y: [3, 3, 2, 1] 183s 0.104 -0.282 0.024 183s 0.071 -0.192 0.016 183s -0.109 0.294 -0.025 183s testing: mul_mat, [1, 1, 2, 4] = [3, 1, 2, 4] * [3, 1, 2, 4] 183s x0: [3, 1, 2, 4] 183s -0.694 -0.869 0.345 183s 183s x1: [3, 1, 2, 4] 183s 0.396 0.875 0.255 183s 183s y: [1, 1, 2, 4] 183s -0.948 183s test-mul-mat0: iter:381/500 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.903 183s 0.701 183s -0.160 183s 183s x1: [1, 3, 1, 1] 183s -0.213 183s 0.967 183s -0.644 183s 183s y: [3, 3, 1, 1] 183s 0.192 -0.149 0.034 183s -0.873 0.678 -0.155 183s 0.581 -0.451 0.103 183s testing: mul_mat, [2, 3, 2, 1] = [1, 2, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.996 183s 0.592 183s 183s x1: [1, 3, 2, 1] 183s 0.936 183s 0.761 183s 0.981 183s 183s y: [2, 3, 2, 1] 183s 0.933 0.554 183s 0.758 0.450 183s 0.978 0.581 183s testing: mul_mat, [2, 2, 2, 3] = [1, 2, 2, 3] * [1, 2, 2, 3] 183s x0: [1, 2, 2, 3] 183s -0.634 183s -0.613 183s 183s x1: [1, 2, 2, 3] 183s -0.630 183s 0.252 183s 183s y: [2, 2, 2, 3] 183s 0.400 0.386 183s -0.160 -0.155 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.644 183s -0.021 183s -0.999 183s -0.987 183s 183s x1: [1, 2, 1, 1] 183s -0.948 183s 0.431 183s 183s y: [4, 2, 1, 1] 183s -0.610 0.020 0.948 0.935 183s 0.277 -0.009 -0.430 -0.425 183s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.839 -0.473 0.987 -0.714 183s 0.712 -0.900 -0.416 0.005 183s -0.766 -0.592 -0.717 0.918 183s 183s x1: [4, 1, 2, 1] 183s -0.769 0.723 -0.865 -0.849 183s 183s y: [3, 1, 2, 1] 183s 0.056 -0.843 0.001 183s testing: mul_mat, [4, 4, 2, 3] = [3, 4, 2, 3] * [3, 4, 2, 3] 183s x0: [3, 4, 2, 3] 183s -0.000 -0.184 -0.965 183s 0.092 -0.707 0.248 183s -0.470 0.414 0.499 183s 0.024 -0.258 -0.627 183s 183s x1: [3, 4, 2, 3] 183s -0.589 0.868 0.673 183s -0.057 -0.549 -0.035 183s 0.359 -0.388 -0.322 183s 0.593 0.139 -0.223 183s 183s y: [4, 4, 2, 3] 183s -0.809 -0.501 0.973 -0.660 183s 0.134 0.374 -0.218 0.162 183s 0.382 0.227 -0.490 0.311 183s 0.189 -0.099 -0.332 0.118 183s test-mul-mat0: iter:382/500 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.489 -0.380 -0.423 183s 0.766 -0.508 0.611 183s 0.047 0.260 0.751 183s 183s x1: [3, 3, 1, 1] 183s 0.650 -0.910 -0.079 183s -0.619 -0.415 -0.771 183s -0.106 0.102 -0.355 183s 183s y: [3, 3, 1, 1] 183s 0.062 0.913 -0.266 183s 0.786 -0.734 -0.715 183s 0.163 -0.350 -0.245 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.202 -0.104 0.605 183s -0.112 0.817 0.986 183s 0.472 -0.953 -0.120 183s -0.426 -0.308 -0.564 183s 183s x1: [3, 3, 1, 1] 183s -0.118 0.948 0.812 183s 0.566 0.971 -0.464 183s 0.132 -0.423 0.422 183s 183s y: [4, 3, 1, 1] 183s 0.417 1.588 -1.057 -0.700 183s -0.496 0.272 -0.603 -0.278 183s 0.273 0.056 0.415 -0.164 183s testing: mul_mat, [1, 4, 1, 3] = [3, 1, 1, 3] * [3, 4, 1, 3] 183s x0: [3, 1, 1, 3] 183s 0.127 0.774 -0.649 183s 183s x1: [3, 4, 1, 3] 183s 0.086 0.312 0.013 183s -0.148 0.803 -0.376 183s 0.899 0.063 -0.626 183s -0.219 0.011 -0.814 183s 183s y: [1, 4, 1, 3] 183s 0.244 183s 0.847 183s 0.569 183s 0.509 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.682 -0.556 -0.795 183s 183s x1: [3, 1, 1, 1] 183s 0.733 0.325 0.403 183s 183s y: [1, 1, 1, 1] 183s -0.002 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.793 183s 0.134 183s -0.065 183s 0.803 183s 183s x1: [1, 3, 1, 1] 183s -0.073 183s 0.302 183s 0.229 183s 183s y: [4, 3, 1, 1] 183s 0.058 -0.010 0.005 -0.059 183s -0.240 0.040 -0.020 0.243 183s -0.182 0.031 -0.015 0.184 183s testing: mul_mat, [4, 1, 1, 3] = [4, 4, 1, 3] * [4, 1, 1, 3] 183s x0: [4, 4, 1, 3] 183s 0.829 0.072 -0.202 -0.838 183s 0.747 -0.646 -0.067 0.325 183s -0.049 -0.272 0.681 -0.693 183s 0.562 -0.246 -0.900 0.296 183s 183s x1: [4, 1, 1, 3] 183s 0.474 0.311 -0.289 -0.173 183s 183s y: [4, 1, 1, 3] 183s 0.619 0.117 -0.185 0.399 183s test-mul-mat0: iter:383/500 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.009 0.194 -0.808 -0.538 183s -0.813 0.545 -0.135 -0.140 183s -0.002 0.431 -0.142 -0.242 183s 0.293 0.116 -0.372 0.926 183s 183s x1: [4, 3, 1, 1] 183s -0.769 0.938 -0.066 -0.367 183s 0.715 -0.628 -0.811 0.067 183s 0.791 0.438 -0.752 -0.043 183s 183s y: [4, 3, 1, 1] 183s 0.441 1.197 0.504 -0.433 183s 0.490 -0.824 -0.174 0.501 183s 0.709 -0.297 0.304 0.522 183s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.853 0.484 0.318 -0.918 183s 0.098 -0.892 0.743 0.570 183s 0.657 -0.302 0.159 0.834 183s 0.578 0.690 0.330 -0.385 183s 183s x1: [4, 4, 3, 1] 183s -0.056 0.319 -0.843 -0.118 183s -0.747 -0.211 -0.403 -0.375 183s -0.021 0.664 -0.584 -0.583 183s 0.912 0.373 -0.945 -0.097 183s 183s y: [4, 4, 3, 1] 183s -0.053 -0.984 -0.366 -0.045 183s -0.523 -0.399 -0.804 -0.566 183s 0.653 -1.360 -0.793 0.478 183s 0.747 -1.001 0.255 0.510 183s testing: mul_mat, [3, 4, 3, 2] = [4, 3, 3, 2] * [4, 4, 3, 2] 183s x0: [4, 3, 3, 2] 183s -0.342 -0.813 0.269 0.988 183s 0.724 0.106 0.148 0.944 183s -0.620 -0.253 0.940 0.548 183s 183s x1: [4, 4, 3, 2] 183s -0.403 0.873 -0.645 -0.950 183s 0.120 0.143 -0.706 0.590 183s 0.516 -0.300 0.337 -0.623 183s 0.837 -0.391 0.091 -0.542 183s 183s y: [3, 4, 3, 2] 183s -1.683 -1.191 -1.098 183s 0.235 0.555 -0.451 183s -0.457 -0.197 -0.269 183s -0.479 0.066 -0.631 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.427 0.913 0.894 0.883 183s 0.147 0.163 0.882 -0.598 183s 183s x1: [4, 3, 1, 1] 183s -0.600 -0.747 0.672 -0.224 183s 0.105 0.258 -0.499 0.518 183s -0.847 0.079 0.718 0.130 183s 183s y: [2, 3, 1, 1] 183s -0.535 0.516 183s 0.292 -0.693 183s 0.467 0.443 183s testing: mul_mat, [1, 4, 3, 1] = [2, 1, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 1, 3, 1] 183s 0.100 -0.014 183s 183s x1: [2, 4, 3, 1] 183s -0.184 0.773 183s 0.939 -0.295 183s -0.444 0.612 183s -0.305 -0.822 183s 183s y: [1, 4, 3, 1] 183s -0.030 183s 0.098 183s -0.053 183s -0.019 183s testing: mul_mat, [2, 2, 3, 2] = [1, 2, 3, 2] * [1, 2, 3, 2] 183s x0: [1, 2, 3, 2] 183s 0.307 183s 0.756 183s 183s x1: [1, 2, 3, 2] 183s 0.799 183s -0.359 183s 183s y: [2, 2, 3, 2] 183s 0.245 0.603 183s -0.110 -0.271 183s test-mul-mat0: iter:384/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.163 183s 0.145 183s 183s x1: [1, 1, 1, 1] 183s -0.496 183s 183s y: [2, 1, 1, 1] 183s -0.081 -0.072 183s testing: mul_mat, [1, 2, 4, 1] = [1, 1, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 1, 4, 1] 183s -0.358 183s 183s x1: [1, 2, 4, 1] 183s -0.237 183s 0.667 183s 183s y: [1, 2, 4, 1] 183s 0.085 183s -0.239 183s testing: mul_mat, [3, 1, 4, 4] = [1, 3, 4, 4] * [1, 1, 4, 4] 183s x0: [1, 3, 4, 4] 183s 0.596 183s -0.788 183s -0.906 183s 183s x1: [1, 1, 4, 4] 183s 0.934 183s 183s y: [3, 1, 4, 4] 183s 0.556 -0.735 -0.846 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.858 183s 183s x1: [1, 3, 1, 1] 183s -0.521 183s 0.945 183s 0.145 183s 183s y: [1, 3, 1, 1] 183s 0.447 183s -0.811 183s -0.124 183s testing: mul_mat, [4, 1, 4, 1] = [1, 4, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 4, 4, 1] 183s 0.600 183s -0.707 183s -0.584 183s 0.388 183s 183s x1: [1, 1, 4, 1] 183s -0.253 183s 183s y: [4, 1, 4, 1] 183s -0.152 0.179 0.148 -0.098 183s testing: mul_mat, [3, 1, 4, 4] = [4, 3, 4, 4] * [4, 1, 4, 4] 183s x0: [4, 3, 4, 4] 183s -0.704 0.812 -0.376 -0.032 183s -0.171 0.093 -0.051 -0.307 183s -0.702 -0.676 0.717 0.571 183s 183s x1: [4, 1, 4, 4] 183s 0.943 0.849 -0.501 0.208 183s 183s y: [3, 1, 4, 4] 183s 0.208 -0.121 -1.475 183s test-mul-mat0: iter:385/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.738 0.460 0.109 183s -0.823 0.513 0.359 183s 183s x1: [3, 1, 1, 1] 183s -0.380 -0.304 0.457 183s 183s y: [2, 1, 1, 1] 183s 0.190 0.321 183s testing: mul_mat, [2, 2, 4, 1] = [3, 2, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.979 0.987 0.050 183s -0.801 0.500 -0.591 183s 183s x1: [3, 2, 4, 1] 183s 0.111 0.084 -0.085 183s -0.045 0.474 -0.387 183s 183s y: [2, 2, 4, 1] 183s -0.029 0.004 183s 0.493 0.501 183s testing: mul_mat, [2, 2, 4, 4] = [3, 2, 4, 4] * [3, 2, 4, 4] 183s x0: [3, 2, 4, 4] 183s 0.479 -0.422 0.365 183s -0.747 0.161 -0.822 183s 183s x1: [3, 2, 4, 4] 183s 0.094 -0.189 0.857 183s -0.243 -0.430 -0.616 183s 183s y: [2, 2, 4, 4] 183s 0.437 -0.805 183s -0.159 0.618 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s -1.000 0.029 0.085 183s 0.499 -0.333 0.675 183s -0.281 0.946 -0.355 183s 183s x1: [3, 2, 1, 1] 183s 0.765 0.940 0.309 183s 0.361 0.774 -0.493 183s 183s y: [3, 2, 1, 1] 183s -0.712 0.277 0.565 183s -0.381 -0.411 0.806 183s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 2, 4, 1] 183s 0.033 -0.699 -0.743 183s 0.912 -0.499 0.339 183s 183s x1: [3, 3, 4, 1] 183s 0.190 0.869 0.079 183s 0.062 -0.677 0.450 183s 0.703 -0.645 0.421 183s 183s y: [2, 3, 4, 1] 183s -0.660 -0.234 183s 0.141 0.547 183s 0.161 1.106 183s testing: mul_mat, [4, 3, 4, 4] = [2, 4, 4, 4] * [2, 3, 4, 4] 183s x0: [2, 4, 4, 4] 183s -0.566 0.267 183s -0.278 -0.244 183s -0.441 -0.064 183s 0.751 -0.787 183s 183s x1: [2, 3, 4, 4] 183s -0.183 0.968 183s -0.099 -0.753 183s 0.298 -0.543 183s 183s y: [4, 3, 4, 4] 183s 0.362 -0.185 0.018 -0.899 183s -0.145 0.211 0.092 0.518 183s -0.313 0.050 -0.096 0.650 183s test-mul-mat0: iter:386/500 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.431 183s 0.104 183s 183s x1: [1, 1, 1, 1] 183s 0.769 183s 183s y: [2, 1, 1, 1] 183s -0.331 0.080 183s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 3, 4, 1] 183s 0.375 183s -0.430 183s -0.613 183s 183s x1: [1, 2, 4, 1] 183s 0.998 183s -0.215 183s 183s y: [3, 2, 4, 1] 183s 0.375 -0.430 -0.612 183s -0.081 0.092 0.132 183s testing: mul_mat, [2, 3, 4, 2] = [1, 2, 4, 2] * [1, 3, 4, 2] 183s x0: [1, 2, 4, 2] 183s 0.589 183s 0.404 183s 183s x1: [1, 3, 4, 2] 183s 0.373 183s -0.130 183s 0.170 183s 183s y: [2, 3, 4, 2] 183s 0.220 0.151 183s -0.076 -0.052 183s 0.100 0.069 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.315 183s 0.345 183s 0.205 183s 183s x1: [1, 2, 1, 1] 183s 0.875 183s -0.062 183s 183s y: [3, 2, 1, 1] 183s -0.276 0.302 0.180 183s 0.020 -0.021 -0.013 183s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.372 -0.017 0.612 183s 0.485 0.935 -0.426 183s -0.166 -0.526 -0.071 183s 183s x1: [3, 1, 4, 1] 183s -0.654 -0.292 0.357 183s 183s y: [3, 1, 4, 1] 183s 0.467 -0.742 0.237 183s testing: mul_mat, [2, 3, 4, 2] = [3, 2, 4, 2] * [3, 3, 4, 2] 183s x0: [3, 2, 4, 2] 183s 0.260 0.099 0.703 183s -0.246 0.914 -0.613 183s 183s x1: [3, 3, 4, 2] 183s 0.149 0.710 -0.960 183s -0.921 -0.425 -0.991 183s 0.383 0.315 -0.084 183s 183s y: [2, 3, 4, 2] 183s -0.566 1.200 183s -0.979 0.446 183s 0.072 0.245 183s test-mul-mat0: iter:387/500 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.966 183s 0.990 183s 0.784 183s -0.809 183s 183s x1: [1, 2, 1, 1] 183s -0.893 183s 0.869 183s 183s y: [4, 2, 1, 1] 183s -0.863 -0.884 -0.700 0.723 183s 0.839 0.860 0.681 -0.703 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.307 183s 183s x1: [1, 4, 1, 1] 183s -0.153 183s -0.914 183s 0.892 183s -0.035 183s 183s y: [1, 4, 1, 1] 183s 0.047 183s 0.281 183s -0.274 183s 0.011 183s testing: mul_mat, [2, 1, 1, 2] = [1, 2, 1, 2] * [1, 1, 1, 2] 183s x0: [1, 2, 1, 2] 183s 0.326 183s -0.180 183s 183s x1: [1, 1, 1, 2] 183s -0.004 183s 183s y: [2, 1, 1, 2] 183s -0.001 0.001 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.752 183s -0.763 183s 183s x1: [1, 2, 1, 1] 183s 0.705 183s 0.416 183s 183s y: [2, 2, 1, 1] 183s 0.530 -0.538 183s 0.313 -0.318 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.833 -0.614 183s -0.787 0.940 183s 183s x1: [2, 1, 1, 1] 183s 0.590 0.084 183s 183s y: [2, 1, 1, 1] 183s 0.440 -0.385 183s testing: mul_mat, [3, 2, 1, 2] = [2, 3, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 3, 1, 2] 183s 0.884 -0.120 183s -0.741 -0.840 183s -0.319 -0.439 183s 183s x1: [2, 2, 1, 2] 183s -0.918 0.248 183s 0.906 -0.928 183s 183s y: [3, 2, 1, 2] 183s -0.841 0.472 0.184 183s 0.912 0.109 0.119 183s test-mul-mat0: iter:388/500 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.250 183s 0.816 183s 183s x1: [1, 4, 1, 1] 183s 0.521 183s -0.891 183s 0.310 183s 0.734 183s 183s y: [2, 4, 1, 1] 183s 0.130 0.426 183s -0.222 -0.727 183s 0.077 0.253 183s 0.183 0.600 183s testing: mul_mat, [2, 2, 2, 1] = [1, 2, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 2, 2, 1] 183s -0.193 183s 0.892 183s 183s x1: [1, 2, 2, 1] 183s 0.744 183s 0.156 183s 183s y: [2, 2, 2, 1] 183s -0.144 0.663 183s -0.030 0.139 183s testing: mul_mat, [1, 2, 2, 4] = [1, 1, 2, 4] * [1, 2, 2, 4] 183s x0: [1, 1, 2, 4] 183s -0.931 183s 183s x1: [1, 2, 2, 4] 183s -0.849 183s 0.922 183s 183s y: [1, 2, 2, 4] 183s 0.790 183s -0.858 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.820 183s 0.223 183s -0.032 183s -0.129 183s 183s x1: [1, 1, 1, 1] 183s 0.332 183s 183s y: [4, 1, 1, 1] 183s -0.272 0.074 -0.011 -0.043 183s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 1, 2, 1] 183s -0.048 -0.590 -0.103 0.365 183s 183s x1: [4, 1, 2, 1] 183s 0.374 -0.111 -0.558 0.685 183s 183s y: [1, 1, 2, 1] 183s 0.354 183s testing: mul_mat, [1, 4, 2, 4] = [1, 1, 2, 4] * [1, 4, 2, 4] 183s x0: [1, 1, 2, 4] 183s 0.599 183s 183s x1: [1, 4, 2, 4] 183s 0.726 183s -0.032 183s 0.750 183s 0.045 183s 183s y: [1, 4, 2, 4] 183s 0.435 183s -0.019 183s 0.449 183s 0.027 183s test-mul-mat0: iter:389/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.867 -0.582 0.029 0.983 183s 0.852 -0.492 0.581 -0.668 183s -0.559 0.487 -0.667 -0.801 183s 183s x1: [4, 3, 1, 1] 183s -0.359 -0.434 0.964 -0.438 183s 0.043 0.503 0.458 -0.074 183s -0.342 0.305 0.221 -0.291 183s 183s y: [3, 3, 1, 1] 183s 0.161 0.760 -0.303 183s -0.389 0.104 -0.025 183s -0.160 -0.120 0.425 183s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 4, 4, 1] 183s -0.418 -0.435 0.357 0.532 183s 0.479 0.071 -0.011 0.974 183s 0.117 -0.113 0.956 0.495 183s -0.016 -0.032 0.453 0.047 183s 183s x1: [4, 3, 4, 1] 183s 0.030 0.580 0.965 0.436 183s 0.042 -0.391 -0.923 0.608 183s -0.428 -0.361 -0.349 -0.924 183s 183s y: [4, 3, 4, 1] 183s 0.312 0.470 1.076 0.439 183s 0.146 0.595 -0.532 -0.378 183s -0.281 -1.127 -0.800 -0.184 183s testing: mul_mat, [3, 4, 4, 4] = [4, 3, 4, 4] * [4, 4, 4, 4] 183s x0: [4, 3, 4, 4] 183s 0.410 0.694 -0.258 0.495 183s -0.162 0.405 -0.200 -0.326 183s -0.890 -0.379 -0.315 -0.018 183s 183s x1: [4, 4, 4, 4] 183s 0.362 -0.663 -0.692 0.621 183s -0.533 0.119 -0.004 0.484 183s 0.887 0.867 0.485 0.623 183s -0.810 -0.467 -0.226 -0.032 183s 183s y: [3, 4, 4, 4] 183s 0.175 -0.391 0.136 183s 0.105 -0.023 0.422 183s 1.148 -0.093 -1.282 183s -0.613 -0.002 0.969 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.067 -0.944 0.300 0.723 183s -0.811 0.032 -0.497 -0.092 183s 183s x1: [4, 3, 1, 1] 183s 0.127 -0.892 0.843 0.267 183s -0.777 -0.225 0.465 0.568 183s 0.049 -0.553 0.746 0.097 183s 183s y: [2, 3, 1, 1] 183s 1.280 -0.575 183s 0.815 0.340 183s 0.813 -0.437 183s testing: mul_mat, [2, 4, 4, 1] = [2, 2, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 2, 4, 1] 183s 0.285 -0.251 183s 0.788 0.992 183s 183s x1: [2, 4, 4, 1] 183s -0.060 0.195 183s -0.610 -0.075 183s -0.004 -0.293 183s 0.852 -0.531 183s 183s y: [2, 4, 4, 1] 183s -0.066 0.146 183s -0.155 -0.556 183s 0.072 -0.294 183s 0.376 0.144 183s testing: mul_mat, [4, 2, 4, 4] = [2, 4, 4, 4] * [2, 2, 4, 4] 183s x0: [2, 4, 4, 4] 183s 0.814 -0.377 183s -0.376 -0.703 183s -0.589 -0.961 183s -0.008 -0.889 183s 183s x1: [2, 2, 4, 4] 183s -0.301 -0.221 183s -0.900 0.196 183s 183s y: [4, 2, 4, 4] 183s -0.162 0.268 0.390 0.199 183s -0.806 0.201 0.342 -0.167 183s test-mul-mat0: iter:390/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.809 0.204 -0.024 183s 183s x1: [3, 1, 1, 1] 183s 0.417 0.205 -0.163 183s 183s y: [1, 1, 1, 1] 183s 0.383 183s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.589 -0.104 -0.455 183s 0.285 0.205 -0.702 183s -0.869 -0.983 -0.736 183s -0.997 0.187 0.681 183s 183s x1: [3, 1, 4, 1] 183s -0.740 0.349 -0.035 183s 183s y: [4, 1, 4, 1] 183s 0.415 -0.115 0.326 0.779 183s testing: mul_mat, [4, 4, 4, 3] = [3, 4, 4, 3] * [3, 4, 4, 3] 183s x0: [3, 4, 4, 3] 183s 0.665 0.655 -0.936 183s -0.656 -0.970 -0.651 183s 0.612 0.768 -0.899 183s 0.362 -0.641 -0.054 183s 183s x1: [3, 4, 4, 3] 183s 0.474 -0.926 0.088 183s 0.003 0.253 -0.376 183s -0.587 -0.755 0.505 183s 0.623 -0.075 -0.753 183s 183s y: [4, 4, 4, 3] 183s -0.373 0.530 -0.500 0.761 183s 0.520 -0.002 0.534 -0.141 183s -1.358 0.789 -1.393 0.245 183s 1.070 0.154 1.000 0.314 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.491 0.419 -0.646 183s 183s x1: [3, 4, 1, 1] 183s -0.158 0.019 0.886 183s 0.462 0.425 -0.658 183s -0.994 0.927 -0.867 183s 0.272 -0.011 0.702 183s 183s y: [1, 4, 1, 1] 183s -0.642 183s 0.830 183s 0.460 183s -0.325 183s testing: mul_mat, [1, 3, 4, 1] = [1, 1, 4, 1] * [1, 3, 4, 1] 183s x0: [1, 1, 4, 1] 183s 0.029 183s 183s x1: [1, 3, 4, 1] 183s 0.086 183s 0.192 183s -0.216 183s 183s y: [1, 3, 4, 1] 183s 0.002 183s 0.006 183s -0.006 183s testing: mul_mat, [3, 1, 4, 3] = [1, 3, 4, 3] * [1, 1, 4, 3] 183s x0: [1, 3, 4, 3] 183s -0.747 183s -0.034 183s 0.603 183s 183s x1: [1, 1, 4, 3] 183s -0.184 183s 183s y: [3, 1, 4, 3] 183s 0.137 0.006 -0.111 183s test-mul-mat0: iter:391/500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.333 183s -0.710 183s 0.908 183s 0.470 183s 183s x1: [1, 1, 1, 1] 183s 0.386 183s 183s y: [4, 1, 1, 1] 183s 0.129 -0.274 0.351 0.181 183s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.680 183s -0.986 183s 183s x1: [1, 4, 2, 1] 183s -0.876 183s 0.616 183s 0.655 183s -0.226 183s 183s y: [2, 4, 2, 1] 183s -0.596 0.864 183s 0.419 -0.607 183s 0.445 -0.645 183s -0.154 0.223 183s testing: mul_mat, [4, 2, 2, 2] = [1, 4, 2, 2] * [1, 2, 2, 2] 183s x0: [1, 4, 2, 2] 183s -0.289 183s 0.244 183s -0.850 183s -0.903 183s 183s x1: [1, 2, 2, 2] 183s 0.898 183s 0.140 183s 183s y: [4, 2, 2, 2] 183s -0.259 0.219 -0.763 -0.811 183s -0.040 0.034 -0.119 -0.126 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.316 183s 0.436 183s 183s x1: [1, 4, 1, 1] 183s 0.524 183s -0.712 183s 0.545 183s 0.538 183s 183s y: [2, 4, 1, 1] 183s 0.166 0.228 183s -0.225 -0.310 183s 0.172 0.237 183s 0.170 0.234 183s testing: mul_mat, [3, 1, 2, 1] = [2, 3, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 3, 2, 1] 183s -0.323 0.388 183s 0.655 -0.100 183s 0.897 -0.953 183s 183s x1: [2, 1, 2, 1] 183s 0.241 0.591 183s 183s y: [3, 1, 2, 1] 183s 0.151 0.099 -0.347 183s testing: mul_mat, [1, 2, 2, 2] = [3, 1, 2, 2] * [3, 2, 2, 2] 183s x0: [3, 1, 2, 2] 183s -0.713 0.084 0.430 183s 183s x1: [3, 2, 2, 2] 183s -0.619 0.279 0.092 183s 0.665 -0.227 0.012 183s 183s y: [1, 2, 2, 2] 183s 0.504 183s -0.488 183s test-mul-mat0: iter:392/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.974 0.361 183s 0.180 -0.287 183s -0.325 0.134 183s 183s x1: [2, 1, 1, 1] 183s -0.295 -0.281 183s 183s y: [3, 1, 1, 1] 183s -0.389 0.028 0.058 183s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.076 0.234 183s -0.879 0.499 183s 0.252 0.855 183s -0.631 -0.430 183s 183s x1: [2, 3, 2, 1] 183s -0.871 0.645 183s -0.015 -0.068 183s 0.514 -0.116 183s 183s y: [4, 3, 2, 1] 183s 0.217 1.087 0.331 0.272 183s -0.015 -0.021 -0.062 0.039 183s -0.066 -0.510 0.030 -0.274 183s testing: mul_mat, [1, 4, 2, 1] = [2, 1, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.458 -0.374 183s 183s x1: [2, 4, 2, 1] 183s 0.324 -0.826 183s 0.891 -0.031 183s 0.159 -0.178 183s -0.517 -0.958 183s 183s y: [1, 4, 2, 1] 183s 0.458 183s 0.419 183s 0.139 183s 0.122 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.930 0.835 183s -0.071 0.118 183s 183s x1: [2, 1, 1, 1] 183s -0.005 0.640 183s 183s y: [2, 1, 1, 1] 183s 0.539 0.076 183s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 1, 2, 1] 183s -0.645 -0.750 183s 183s x1: [2, 2, 2, 1] 183s 0.578 -0.125 183s -0.120 -0.097 183s 183s y: [1, 2, 2, 1] 183s -0.279 183s 0.151 183s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.232 183s 0.006 183s 0.848 183s 183s x1: [1, 2, 2, 1] 183s -0.394 183s 0.054 183s 183s y: [3, 2, 2, 1] 183s 0.092 -0.002 -0.334 183s -0.013 0.000 0.046 183s test-mul-mat0: iter:393/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.004 0.103 183s -0.180 0.352 183s 183s x1: [2, 4, 1, 1] 183s -0.356 -0.029 183s 0.577 -0.778 183s 0.847 -0.543 183s 0.125 -0.104 183s 183s y: [2, 4, 1, 1] 183s -0.002 0.054 183s -0.082 -0.377 183s -0.059 -0.344 183s -0.011 -0.059 183s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.092 -0.021 183s 0.946 -0.956 183s 183s x1: [2, 2, 3, 1] 183s 0.353 -0.546 183s -0.212 0.959 183s 183s y: [2, 2, 3, 1] 183s -0.021 0.856 183s -0.000 -1.117 183s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.380 -0.302 183s 0.477 0.127 183s 183s x1: [2, 2, 3, 1] 183s 0.758 -0.302 183s -0.067 -0.139 183s 183s y: [2, 2, 3, 1] 183s 0.379 0.323 183s 0.017 -0.049 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.007 0.927 183s 0.714 -0.424 183s -0.110 -0.592 183s 183s x1: [2, 2, 1, 1] 183s 0.094 -0.280 183s 0.056 0.956 183s 183s y: [3, 2, 1, 1] 183s -0.260 0.186 0.155 183s 0.886 -0.366 -0.572 183s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 3, 3, 1] 183s 0.370 -0.574 0.419 183s 0.577 0.534 0.248 183s 0.465 0.656 -0.454 183s 183s x1: [3, 2, 3, 1] 183s -0.788 -0.209 -0.619 183s 0.285 -0.035 0.862 183s 183s y: [3, 2, 3, 1] 183s -0.431 -0.720 -0.223 183s 0.487 0.359 -0.282 183s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.820 -0.324 0.288 183s 183s x1: [3, 3, 3, 1] 183s -0.074 -0.769 0.696 183s 0.671 0.601 0.273 183s 0.136 -0.973 -0.194 183s 183s y: [1, 3, 3, 1] 183s 0.389 183s 0.434 183s 0.371 183s test-mul-mat0: iter:394/500 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.775 -0.111 0.746 0.063 183s 0.635 -0.347 -0.953 -0.948 183s 0.160 -0.697 0.604 0.693 183s 183s x1: [4, 4, 1, 1] 183s 0.980 0.117 0.683 -0.228 183s -0.005 0.977 0.122 0.488 183s 0.606 -0.686 0.560 -0.567 183s 0.135 0.499 0.699 0.069 183s 183s y: [3, 4, 1, 1] 183s 1.242 0.147 0.330 183s 0.010 -0.921 -0.270 183s 0.928 0.627 0.520 183s 0.575 -0.820 0.144 183s testing: mul_mat, [2, 3, 4, 1] = [4, 2, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.528 -0.214 0.989 0.314 183s -0.029 -0.403 -0.191 0.203 183s 183s x1: [4, 3, 4, 1] 183s 0.366 0.949 0.673 -0.517 183s 0.632 -0.554 0.478 0.609 183s 0.567 -0.035 0.215 0.881 183s 183s y: [2, 3, 4, 1] 183s 0.107 -0.626 183s 0.448 0.237 183s 0.197 0.135 183s testing: mul_mat, [2, 2, 4, 2] = [4, 2, 4, 2] * [4, 2, 4, 2] 183s x0: [4, 2, 4, 2] 183s 0.746 -0.121 -0.587 0.670 183s 0.243 0.210 0.983 0.165 183s 183s x1: [4, 2, 4, 2] 183s -0.523 -0.821 0.223 0.448 183s -0.224 -0.968 -0.349 -0.965 183s 183s y: [2, 2, 4, 2] 183s -0.122 -0.006 183s -0.492 -0.759 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.224 -0.093 0.693 -0.398 183s 0.670 0.865 -0.190 0.703 183s 183s x1: [4, 2, 1, 1] 183s 0.567 -0.572 -0.973 0.580 183s -0.610 -0.427 -0.096 -0.325 183s 183s y: [2, 2, 1, 1] 183s -0.980 0.478 183s 0.239 -0.988 183s testing: mul_mat, [2, 4, 4, 1] = [2, 2, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 2, 4, 1] 183s -0.585 -0.727 183s 0.370 -0.135 183s 183s x1: [2, 4, 4, 1] 183s -0.148 0.428 183s 0.357 -0.663 183s 0.966 -0.550 183s 0.912 0.213 183s 183s y: [2, 4, 4, 1] 183s -0.225 -0.112 183s 0.273 0.222 183s -0.165 0.432 183s -0.688 0.309 183s testing: mul_mat, [2, 2, 4, 2] = [2, 2, 4, 2] * [2, 2, 4, 2] 183s x0: [2, 2, 4, 2] 183s -0.236 0.484 183s -0.313 0.974 183s 183s x1: [2, 2, 4, 2] 183s -0.927 -0.075 183s -0.203 -0.935 183s 183s y: [2, 2, 4, 2] 183s 0.182 0.217 183s -0.404 -0.848 183s test-mul-mat0: iter:395/500 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.593 -0.521 -0.413 183s 0.859 -0.912 0.727 183s 0.745 -0.197 -0.172 183s 0.269 0.900 0.721 183s 183s x1: [3, 2, 1, 1] 183s 0.725 -0.927 -0.343 183s -0.087 0.903 -0.709 183s 183s y: [4, 2, 1, 1] 183s 0.194 1.219 0.781 -0.886 183s -0.126 -1.415 -0.121 0.278 183s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.986 0.150 0.321 183s 0.367 -0.926 -0.541 183s 183s x1: [3, 4, 3, 1] 183s -0.998 0.805 0.162 183s 0.904 0.677 0.225 183s 0.900 -0.687 0.975 183s 0.381 -0.076 0.138 183s 183s y: [2, 4, 3, 1] 183s -0.812 -1.199 183s 1.066 -0.417 183s 1.098 0.439 183s 0.408 0.135 183s testing: mul_mat, [4, 2, 3, 4] = [3, 4, 3, 4] * [3, 2, 3, 4] 183s x0: [3, 4, 3, 4] 183s -0.295 -0.026 0.382 183s 0.675 0.400 0.098 183s -0.764 -0.056 -0.557 183s 0.459 0.862 0.907 183s 183s x1: [3, 2, 3, 4] 183s 0.676 0.356 0.013 183s -0.086 0.536 -0.631 183s 183s y: [4, 2, 3, 4] 183s -0.203 0.600 -0.544 0.629 183s -0.229 0.094 0.388 -0.150 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.696 -0.017 0.722 183s 0.592 -0.065 0.517 183s 183s x1: [3, 4, 1, 1] 183s 0.202 0.131 0.885 183s 0.847 -0.564 0.187 183s 0.774 0.481 0.505 183s -0.044 0.544 -0.453 183s 183s y: [2, 4, 1, 1] 183s 0.496 0.569 183s -0.445 0.635 183s -0.182 0.688 183s -0.306 -0.296 183s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.571 -0.251 183s 183s x1: [2, 3, 3, 1] 183s -0.452 0.284 183s 0.492 -0.898 183s 0.206 -0.876 183s 183s y: [1, 3, 3, 1] 183s 0.187 183s -0.055 183s 0.102 183s testing: mul_mat, [4, 2, 3, 4] = [1, 4, 3, 4] * [1, 2, 3, 4] 183s x0: [1, 4, 3, 4] 183s -0.546 183s 0.879 183s -0.391 183s -0.573 183s 183s x1: [1, 2, 3, 4] 183s -0.722 183s -0.112 183s 183s y: [4, 2, 3, 4] 183s 0.395 -0.635 0.283 0.414 183s 0.061 -0.098 0.044 0.064 183s test-mul-mat0: iter:396/500 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.880 0.703 -0.010 183s -0.780 0.274 0.841 183s -0.762 0.520 -0.763 183s 183s x1: [3, 4, 1, 1] 183s -0.992 -0.251 -0.913 183s 0.810 0.240 0.410 183s -0.749 0.755 0.862 183s -0.162 -0.276 -0.805 183s 183s y: [3, 4, 1, 1] 183s 0.706 -0.063 1.322 183s -0.548 -0.221 -0.805 183s 1.181 1.516 0.305 183s -0.044 -0.627 0.594 183s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 1, 3, 1] 183s 0.198 0.063 -0.710 183s 183s x1: [3, 3, 3, 1] 183s 0.933 -0.666 0.322 183s 0.168 -0.183 -0.638 183s -0.720 -0.848 0.012 183s 183s y: [1, 3, 3, 1] 183s -0.086 183s 0.475 183s -0.204 183s testing: mul_mat, [4, 1, 3, 1] = [3, 4, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.303 -0.036 -0.499 183s 0.922 0.530 0.722 183s -0.338 0.799 0.934 183s 0.621 0.342 0.986 183s 183s x1: [3, 1, 3, 1] 183s -0.421 -0.634 -0.015 183s 183s y: [4, 1, 3, 1] 183s -0.097 -0.735 -0.378 -0.493 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.580 -0.861 0.761 183s 0.380 -0.592 -0.015 183s 183s x1: [3, 4, 1, 1] 183s -0.804 -0.059 0.817 183s -0.871 -0.437 0.159 183s -0.884 0.381 0.563 183s -0.608 0.331 0.365 183s 183s y: [2, 4, 1, 1] 183s 0.205 -0.283 183s -0.008 -0.074 183s -0.413 -0.570 183s -0.361 -0.433 183s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 4, 3, 1] 183s 0.985 0.845 183s -0.519 -0.679 183s -0.958 -0.103 183s 0.316 0.225 183s 183s x1: [2, 3, 3, 1] 183s -0.897 0.408 183s 0.915 -0.237 183s -0.537 0.438 183s 183s y: [4, 3, 3, 1] 183s -0.538 0.188 0.816 -0.192 183s 0.701 -0.314 -0.851 0.236 183s -0.159 -0.018 0.469 -0.071 183s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 2, 3, 1] 183s 0.414 0.857 0.584 0.055 183s -0.193 0.532 -0.478 -0.597 183s 183s x1: [4, 2, 3, 1] 183s 0.523 -0.181 -0.576 0.058 183s 0.423 0.341 -0.291 0.408 183s 183s y: [2, 2, 3, 1] 183s -0.272 0.044 183s 0.320 -0.004 183s test-mul-mat0: iter:397/500 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.374 183s 183s x1: [1, 1, 1, 1] 183s 0.349 183s 183s y: [1, 1, 1, 1] 183s -0.130 183s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.705 183s 183s x1: [1, 1, 2, 1] 183s -0.237 183s 183s y: [1, 1, 2, 1] 183s 0.167 183s testing: mul_mat, [3, 1, 2, 2] = [1, 3, 2, 2] * [1, 1, 2, 2] 183s x0: [1, 3, 2, 2] 183s -0.431 183s -0.764 183s -0.717 183s 183s x1: [1, 1, 2, 2] 183s 0.005 183s 183s y: [3, 1, 2, 2] 183s -0.002 -0.004 -0.004 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.847 183s 0.322 183s -0.956 183s 183s x1: [1, 3, 1, 1] 183s 0.852 183s 0.334 183s 0.830 183s 183s y: [3, 3, 1, 1] 183s -0.721 0.274 -0.815 183s -0.283 0.108 -0.320 183s -0.703 0.267 -0.794 183s testing: mul_mat, [3, 1, 2, 1] = [3, 3, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 3, 2, 1] 183s 0.437 0.387 0.623 183s -0.548 0.915 -0.802 183s 0.940 -0.491 -0.686 183s 183s x1: [3, 1, 2, 1] 183s -0.898 0.948 -0.193 183s 183s y: [3, 1, 2, 1] 183s -0.146 1.514 -1.177 183s testing: mul_mat, [2, 3, 2, 2] = [3, 2, 2, 2] * [3, 3, 2, 2] 183s x0: [3, 2, 2, 2] 183s 0.175 -0.149 -0.149 183s -0.910 -0.503 -0.229 183s 183s x1: [3, 3, 2, 2] 183s -0.790 0.059 0.968 183s -0.637 -0.619 -0.989 183s -0.535 -0.671 -0.182 183s 183s y: [2, 3, 2, 2] 183s -0.292 0.468 183s 0.128 1.117 183s 0.033 0.866 183s test-mul-mat0: iter:398/500 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.628 -0.830 -0.915 0.470 183s -0.914 0.811 -0.513 -0.459 183s 183s x1: [4, 3, 1, 1] 183s 0.476 0.005 0.974 -0.348 183s 0.095 -0.176 0.149 0.946 183s 0.596 -0.835 -0.005 -0.261 183s 183s y: [2, 3, 1, 1] 183s -0.760 -0.771 183s 0.513 -0.740 183s 0.949 -1.099 183s testing: mul_mat, [2, 2, 2, 1] = [4, 2, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 2, 2, 1] 183s 0.243 0.017 0.392 -0.762 183s 0.756 0.033 0.866 0.927 183s 183s x1: [4, 2, 2, 1] 183s -0.104 -0.234 -0.914 -0.238 183s 0.184 0.952 -0.318 0.950 183s 183s y: [2, 2, 2, 1] 183s -0.206 -1.097 183s -0.788 0.775 183s testing: mul_mat, [2, 2, 2, 3] = [4, 2, 2, 3] * [4, 2, 2, 3] 183s x0: [4, 2, 2, 3] 183s -0.222 -0.487 0.426 0.413 183s 0.124 0.421 0.384 0.748 183s 183s x1: [4, 2, 2, 3] 183s -0.360 -0.412 0.773 -0.408 183s 0.270 0.722 0.789 0.893 183s 183s y: [2, 2, 2, 3] 183s 0.441 -0.227 183s 0.294 1.308 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.384 0.692 -0.610 -0.355 183s 183s x1: [4, 2, 1, 1] 183s 0.880 -0.360 -0.189 0.420 183s 0.255 -0.651 0.951 -0.878 183s 183s y: [1, 2, 1, 1] 183s 0.054 183s -0.621 183s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 3, 2, 1] 183s 0.632 183s 0.787 183s 0.535 183s 183s x1: [1, 4, 2, 1] 183s 0.811 183s -0.807 183s -0.581 183s 0.355 183s 183s y: [3, 4, 2, 1] 183s 0.513 0.638 0.434 183s -0.510 -0.635 -0.432 183s -0.367 -0.457 -0.311 183s 0.225 0.280 0.190 183s testing: mul_mat, [1, 1, 2, 3] = [3, 1, 2, 3] * [3, 1, 2, 3] 183s x0: [3, 1, 2, 3] 183s 0.048 0.829 0.894 183s 183s x1: [3, 1, 2, 3] 183s -0.123 -0.827 0.311 183s 183s y: [1, 1, 2, 3] 183s -0.413 183s test-mul-mat0: iter:399/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.207 -0.350 0.566 183s 0.626 0.227 -0.387 183s 183s x1: [3, 1, 1, 1] 183s 0.992 -0.019 -0.507 183s 183s y: [2, 1, 1, 1] 183s -0.075 0.813 183s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.826 0.587 0.558 183s 0.800 -0.187 -0.829 183s 0.255 0.934 -0.846 183s -0.451 -0.086 0.137 183s 183s x1: [3, 2, 4, 1] 183s 0.455 0.121 0.982 183s 0.293 -0.021 -0.017 183s 183s y: [4, 2, 4, 1] 183s 0.243 -0.472 -0.602 -0.081 183s -0.264 0.253 0.070 -0.133 183s testing: mul_mat, [4, 4, 4, 4] = [3, 4, 4, 4] * [3, 4, 4, 4] 183s x0: [3, 4, 4, 4] 183s 0.482 -0.937 0.937 183s 0.523 -0.117 0.738 183s -0.098 -0.534 -0.616 183s -0.741 -0.531 -0.592 183s 183s x1: [3, 4, 4, 4] 183s 0.568 -0.032 0.057 183s 0.279 0.020 -0.036 183s -0.547 0.200 -0.144 183s -0.929 0.499 0.715 183s 183s y: [4, 4, 4, 4] 183s 0.357 0.343 -0.074 -0.438 183s 0.082 0.117 -0.016 -0.196 183s -0.586 -0.416 0.036 0.385 183s -0.246 -0.017 -0.616 0.000 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.276 0.701 0.647 183s 0.020 0.265 0.494 183s 183s x1: [3, 4, 1, 1] 183s -0.198 0.217 0.329 183s -0.546 0.522 0.885 183s 0.771 0.708 0.249 183s -0.947 0.551 -0.295 183s 183s y: [2, 4, 1, 1] 183s 0.310 0.216 183s 0.788 0.565 183s 0.870 0.326 183s -0.066 -0.019 183s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.308 -0.521 183s -0.168 0.233 183s 0.739 -0.686 183s 0.957 -0.243 183s 183s x1: [2, 3, 4, 1] 183s -0.643 -0.721 183s -0.914 -0.306 183s 0.004 -0.248 183s 183s y: [4, 3, 4, 1] 183s 0.574 -0.060 0.020 -0.440 183s 0.441 0.083 -0.466 -0.801 183s 0.128 -0.058 0.173 0.064 183s testing: mul_mat, [3, 2, 4, 4] = [4, 3, 4, 4] * [4, 2, 4, 4] 183s x0: [4, 3, 4, 4] 183s 0.601 0.559 -0.083 -0.871 183s 0.980 -0.091 -0.342 -0.905 183s 0.225 0.602 -0.394 -0.889 183s 183s x1: [4, 2, 4, 4] 183s 0.050 0.549 0.012 0.528 183s -0.218 0.326 -0.715 -0.899 183s 183s y: [3, 2, 4, 4] 183s -0.124 -0.484 -0.132 183s 0.893 0.815 1.228 183s test-mul-mat0: iter:400/500 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.595 183s 0.317 183s -0.969 183s 183s x1: [1, 3, 1, 1] 183s -0.271 183s 0.259 183s -0.567 183s 183s y: [3, 3, 1, 1] 183s -0.161 -0.086 0.262 183s 0.154 0.082 -0.251 183s -0.337 -0.180 0.549 183s testing: mul_mat, [4, 3, 3, 1] = [1, 4, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 4, 3, 1] 183s 0.697 183s 0.547 183s 0.354 183s 0.032 183s 183s x1: [1, 3, 3, 1] 183s 0.397 183s 0.128 183s 0.301 183s 183s y: [4, 3, 3, 1] 183s 0.277 0.217 0.140 0.013 183s 0.089 0.070 0.045 0.004 183s 0.210 0.165 0.106 0.010 183s testing: mul_mat, [4, 4, 3, 3] = [1, 4, 3, 3] * [1, 4, 3, 3] 183s x0: [1, 4, 3, 3] 183s 0.112 183s 0.966 183s 0.690 183s -0.483 183s 183s x1: [1, 4, 3, 3] 183s 0.155 183s 0.413 183s -0.384 183s -0.587 183s 183s y: [4, 4, 3, 3] 183s 0.017 0.149 0.107 -0.075 183s 0.046 0.399 0.285 -0.200 183s -0.043 -0.371 -0.265 0.186 183s -0.066 -0.567 -0.405 0.284 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.372 183s 183s x1: [1, 4, 1, 1] 183s 0.300 183s 0.059 183s 0.981 183s -0.095 183s 183s y: [1, 4, 1, 1] 183s 0.111 183s 0.022 183s 0.364 183s -0.035 183s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 2, 3, 1] 183s 0.365 183s -0.975 183s 183s x1: [1, 1, 3, 1] 183s -0.529 183s 183s y: [2, 1, 3, 1] 183s -0.193 0.515 183s testing: mul_mat, [1, 1, 3, 3] = [2, 1, 3, 3] * [2, 1, 3, 3] 183s x0: [2, 1, 3, 3] 183s -0.521 0.680 183s 183s x1: [2, 1, 3, 3] 183s 0.242 0.414 183s 183s y: [1, 1, 3, 3] 183s 0.155 183s test-mul-mat0: iter:401/500 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.282 -0.080 0.071 0.525 183s -0.025 0.598 0.697 0.056 183s -0.574 0.852 0.088 -0.462 183s 0.250 -0.466 -0.542 0.676 183s 183s x1: [4, 2, 1, 1] 183s 0.028 -0.395 0.974 0.104 183s -0.068 0.579 -0.761 0.240 183s 183s y: [4, 2, 1, 1] 183s 0.148 0.447 -0.315 -0.266 183s 0.045 -0.169 0.354 0.288 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.123 0.126 0.084 -0.430 183s 183s x1: [4, 4, 3, 1] 183s 0.547 0.669 -0.941 0.383 183s -0.548 0.727 -0.589 0.056 183s 0.701 0.514 0.989 0.279 183s 0.754 0.229 0.369 -0.529 183s 183s y: [1, 4, 3, 1] 183s -0.227 183s 0.085 183s -0.059 183s 0.194 183s testing: mul_mat, [4, 1, 3, 4] = [4, 4, 3, 4] * [4, 1, 3, 4] 183s x0: [4, 4, 3, 4] 183s 0.977 0.836 -0.982 -0.916 183s -0.179 -0.718 0.738 0.920 183s 0.119 0.941 0.930 -0.671 183s -0.847 0.322 0.260 0.501 183s 183s x1: [4, 1, 3, 4] 183s 0.383 0.721 0.137 0.914 183s 183s y: [4, 1, 3, 4] 183s 0.005 0.355 0.238 0.402 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.649 0.114 0.127 -0.056 183s -0.956 -0.537 -0.093 -0.546 183s 0.556 0.900 -0.434 -0.830 183s -0.820 -0.900 0.770 0.280 183s 183s x1: [4, 4, 1, 1] 183s -0.601 -0.664 0.611 -0.554 183s 0.250 -0.622 0.802 0.928 183s 0.003 0.035 0.928 -0.319 183s -0.125 0.931 -0.075 0.806 183s 183s y: [4, 4, 1, 1] 183s 0.422 1.176 -0.737 1.406 183s -0.184 -0.485 -1.540 1.233 183s 0.138 0.066 -0.105 0.592 183s 0.133 -0.814 0.132 -0.568 183s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.471 -0.988 -0.599 0.734 183s 0.989 0.381 0.339 0.751 183s 0.336 -0.033 -0.631 0.941 183s 183s x1: [4, 4, 3, 1] 183s -0.935 -0.384 -0.470 -0.557 183s -0.582 -0.543 0.446 0.453 183s -0.614 -0.873 -0.672 -0.683 183s 0.052 -0.866 -0.514 0.403 183s 183s y: [3, 4, 3, 1] 183s 0.693 -1.649 -0.529 183s 0.875 -0.292 -0.033 183s 1.053 -1.681 -0.397 183s 1.435 -0.150 0.749 183s testing: mul_mat, [4, 4, 3, 4] = [3, 4, 3, 4] * [3, 4, 3, 4] 183s x0: [3, 4, 3, 4] 183s 0.613 -0.013 -0.545 183s 0.155 0.279 -0.421 183s 0.868 0.564 -0.572 183s 0.506 0.190 0.909 183s 183s x1: [3, 4, 3, 4] 183s 0.300 0.440 0.102 183s 0.670 0.174 -0.147 183s 0.611 0.521 0.081 183s 0.739 0.011 -0.772 183s 183s y: [4, 4, 3, 4] 183s 0.123 0.126 0.450 0.328 183s 0.488 0.214 0.764 0.238 183s 0.324 0.206 0.778 0.481 183s 0.874 0.442 1.089 -0.325 183s test-mul-mat0: iter:402/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.809 0.613 183s -0.756 0.594 183s 183s x1: [2, 4, 1, 1] 183s 0.125 0.496 183s -0.211 -0.669 183s 0.174 -0.715 183s 0.404 0.289 183s 183s y: [2, 4, 1, 1] 183s 0.405 0.201 183s -0.581 -0.239 183s -0.298 -0.556 183s 0.504 -0.133 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.407 0.369 183s 0.091 0.450 183s -0.360 0.678 183s 183s x1: [2, 2, 1, 1] 183s 0.104 -0.869 183s 0.845 -0.515 183s 183s y: [3, 2, 1, 1] 183s -0.363 -0.381 -0.627 183s -0.533 -0.155 -0.654 183s testing: mul_mat, [1, 3, 1, 2] = [2, 1, 1, 2] * [2, 3, 1, 2] 183s x0: [2, 1, 1, 2] 183s -0.770 -0.273 183s 183s x1: [2, 3, 1, 2] 183s 0.223 0.746 183s -0.340 -0.216 183s 0.371 -0.232 183s 183s y: [1, 3, 1, 2] 183s -0.375 183s 0.321 183s -0.222 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.444 0.930 183s -0.707 0.246 183s 0.479 -0.927 183s 183s x1: [2, 1, 1, 1] 183s -0.660 -0.838 183s 183s y: [3, 1, 1, 1] 183s -1.072 0.260 0.461 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.325 0.055 -0.922 183s 0.530 0.443 0.551 183s 183s x1: [3, 2, 1, 1] 183s 0.298 -0.663 0.522 183s 0.938 -0.985 -0.254 183s 183s y: [2, 2, 1, 1] 183s -0.615 0.152 183s -0.125 -0.079 183s testing: mul_mat, [1, 3, 1, 2] = [2, 1, 1, 2] * [2, 3, 1, 2] 183s x0: [2, 1, 1, 2] 183s 0.412 0.760 183s 183s x1: [2, 3, 1, 2] 183s 0.707 0.416 183s -0.623 0.145 183s 0.039 -0.392 183s 183s y: [1, 3, 1, 2] 183s 0.608 183s -0.146 183s -0.282 183s test-mul-mat0: iter:403/500 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.981 -0.837 -0.901 183s 0.853 0.120 -0.255 183s -0.934 -0.761 0.026 183s 0.723 0.651 -0.214 183s 183s x1: [3, 4, 1, 1] 183s -0.177 0.011 -0.899 183s -0.492 0.686 0.631 183s 0.563 0.129 0.709 183s 0.114 -0.164 0.124 183s 183s y: [4, 4, 1, 1] 183s 0.628 0.079 0.133 0.071 183s -1.626 -0.499 -0.046 -0.044 183s -0.194 0.315 -0.605 0.340 183s 0.137 0.046 0.022 -0.051 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.130 0.414 -0.797 183s 0.966 -0.461 0.694 183s 183s x1: [3, 4, 1, 1] 183s 0.310 -0.691 -0.381 183s -0.804 0.305 0.974 183s 0.019 -0.684 -0.925 183s 0.527 -0.999 0.705 183s 183s y: [2, 4, 1, 1] 183s 0.058 0.354 183s -0.754 -0.241 183s 0.456 -0.308 183s -0.907 1.459 183s testing: mul_mat, [4, 2, 1, 3] = [3, 4, 1, 3] * [3, 2, 1, 3] 183s x0: [3, 4, 1, 3] 183s 0.641 0.019 0.642 183s 0.642 -0.276 -0.269 183s -0.228 -0.861 -0.066 183s -0.262 -0.323 -0.371 183s 183s x1: [3, 2, 1, 3] 183s 0.947 -0.298 0.793 183s 0.800 0.821 -0.462 183s 183s y: [4, 2, 1, 3] 183s 1.110 0.477 -0.011 -0.446 183s 0.232 0.412 -0.859 -0.303 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.369 -0.843 -0.387 183s 0.187 -0.232 0.353 183s 0.893 0.415 0.464 183s -0.427 0.148 0.402 183s 183s x1: [3, 4, 1, 1] 183s -0.575 -0.629 -0.329 183s 0.359 0.109 0.348 183s 0.988 0.794 -0.273 183s -0.591 -0.721 0.927 183s 183s y: [4, 4, 1, 1] 183s 0.869 -0.077 -0.927 0.020 183s -0.359 0.165 0.528 0.003 183s -0.928 -0.096 1.084 -0.414 183s 0.466 0.384 -0.396 0.518 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.706 0.280 -0.584 0.205 183s -0.511 -0.426 -0.351 0.603 183s 183s x1: [4, 3, 1, 1] 183s 0.405 0.421 0.210 0.099 183s 0.020 0.743 0.524 0.391 183s -0.586 -0.117 -0.500 0.762 183s 183s y: [2, 3, 1, 1] 183s -0.270 -0.400 183s -0.032 -0.275 183s 0.829 0.984 183s testing: mul_mat, [2, 4, 1, 3] = [2, 2, 1, 3] * [2, 4, 1, 3] 183s x0: [2, 2, 1, 3] 183s 0.609 0.357 183s 0.574 -0.743 183s 183s x1: [2, 4, 1, 3] 183s -0.459 0.778 183s 0.760 0.309 183s 0.193 -0.092 183s 0.921 -0.455 183s 183s y: [2, 4, 1, 3] 183s -0.001 -0.841 183s 0.573 0.207 183s 0.084 0.179 183s 0.398 0.867 183s test-mul-mat0: iter:404/500 183s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.392 183s -0.839 183s 183s x1: [1, 3, 1, 1] 183s -0.755 183s 0.794 183s 0.885 183s 183s y: [2, 3, 1, 1] 183s 0.296 0.634 183s -0.311 -0.666 183s -0.347 -0.743 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.325 183s 0.520 183s 183s x1: [1, 2, 1, 1] 183s 0.868 183s -0.186 183s 183s y: [2, 2, 1, 1] 183s 0.282 0.451 183s -0.060 -0.097 183s testing: mul_mat, [1, 2, 1, 2] = [1, 1, 1, 2] * [1, 2, 1, 2] 183s x0: [1, 1, 1, 2] 183s -0.906 183s 183s x1: [1, 2, 1, 2] 183s 0.302 183s 0.653 183s 183s y: [1, 2, 1, 2] 183s -0.273 183s -0.591 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.440 183s 183s x1: [1, 1, 1, 1] 183s 0.519 183s 183s y: [1, 1, 1, 1] 183s -0.228 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.280 183s 183s x1: [1, 1, 1, 1] 183s -0.682 183s 183s y: [1, 1, 1, 1] 183s -0.191 183s testing: mul_mat, [2, 1, 1, 2] = [1, 2, 1, 2] * [1, 1, 1, 2] 183s x0: [1, 2, 1, 2] 183s -0.545 183s 0.473 183s 183s x1: [1, 1, 1, 2] 183s -0.575 183s 183s y: [2, 1, 1, 2] 183s 0.313 -0.272 183s test-mul-mat0: iter:405/500 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.875 -0.297 183s -0.795 0.645 183s -0.995 0.857 183s -0.835 -0.752 183s 183s x1: [2, 1, 1, 1] 183s 0.311 0.853 183s 183s y: [4, 1, 1, 1] 183s -0.525 0.302 0.422 -0.901 183s testing: mul_mat, [2, 4, 3, 1] = [2, 2, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.029 0.287 183s 0.896 -0.114 183s 183s x1: [2, 4, 3, 1] 183s 0.119 -0.741 183s 0.848 -0.362 183s 0.132 -0.592 183s -0.045 -0.439 183s 183s y: [2, 4, 3, 1] 183s -0.209 0.191 183s -0.080 0.801 183s -0.166 0.185 183s -0.127 0.010 183s testing: mul_mat, [3, 2, 3, 4] = [2, 3, 3, 4] * [2, 2, 3, 4] 183s x0: [2, 3, 3, 4] 183s -0.906 0.409 183s -0.052 -0.143 183s 0.381 0.239 183s 183s x1: [2, 2, 3, 4] 183s 0.598 0.403 183s -0.913 -0.022 183s 183s y: [3, 2, 3, 4] 183s -0.376 -0.089 0.324 183s 0.818 0.051 -0.353 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.489 0.185 183s -0.901 0.579 183s 0.530 -0.024 183s -0.504 -0.421 183s 183s x1: [2, 3, 1, 1] 183s 0.143 0.181 183s -0.340 -0.452 183s 0.988 -0.325 183s 183s y: [4, 3, 1, 1] 183s -0.036 -0.024 0.072 -0.148 183s 0.082 0.045 -0.170 0.362 183s -0.543 -1.078 0.532 -0.361 183s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.439 0.370 -0.317 0.279 183s -0.606 0.847 0.113 0.501 183s 0.268 0.877 -0.866 0.957 183s -0.861 0.588 -0.253 -0.333 183s 183s x1: [4, 2, 3, 1] 183s 0.738 -0.782 -0.164 0.236 183s -0.455 -0.936 -0.833 0.367 183s 183s y: [4, 2, 3, 1] 183s -0.496 -1.010 -0.120 -1.132 183s 0.220 -0.428 0.131 -0.071 183s testing: mul_mat, [1, 4, 3, 4] = [4, 1, 3, 4] * [4, 4, 3, 4] 183s x0: [4, 1, 3, 4] 183s -0.999 -0.855 -0.493 -0.452 183s 183s x1: [4, 4, 3, 4] 183s 0.327 -0.611 -0.508 0.016 183s -0.635 0.197 -0.895 0.650 183s 0.592 -0.421 -0.082 0.018 183s -0.048 -0.731 0.004 0.719 183s 183s y: [1, 4, 3, 4] 183s 0.440 183s 0.613 183s -0.199 183s 0.346 183s test-mul-mat0: iter:406/500 183s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.624 183s 183s x1: [1, 2, 1, 1] 183s -0.737 183s 0.198 183s 183s y: [1, 2, 1, 1] 183s 0.460 183s -0.123 183s testing: mul_mat, [4, 1, 2, 1] = [1, 4, 2, 1] * [1, 1, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.344 183s -0.879 183s 0.863 183s -0.610 183s 183s x1: [1, 1, 2, 1] 183s -0.078 183s 183s y: [4, 1, 2, 1] 183s 0.027 0.068 -0.067 0.047 183s testing: mul_mat, [4, 4, 2, 3] = [1, 4, 2, 3] * [1, 4, 2, 3] 183s x0: [1, 4, 2, 3] 183s -0.162 183s 0.064 183s 0.232 183s 0.106 183s 183s x1: [1, 4, 2, 3] 183s 0.218 183s -0.894 183s 0.692 183s -0.881 183s 183s y: [4, 4, 2, 3] 183s -0.035 0.014 0.050 0.023 183s 0.144 -0.058 -0.207 -0.095 183s -0.112 0.045 0.160 0.073 183s 0.142 -0.057 -0.204 -0.093 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.702 183s -0.220 183s -0.925 183s -0.233 183s 183s x1: [1, 4, 1, 1] 183s -0.711 183s -0.740 183s 0.028 183s 0.474 183s 183s y: [4, 4, 1, 1] 183s -0.500 0.157 0.658 0.166 183s -0.519 0.163 0.684 0.173 183s 0.020 -0.006 -0.026 -0.007 183s 0.333 -0.104 -0.438 -0.111 183s tes.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backwarting: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 1, 2, 1] 183s -0.189 -0.473 -0.253 -0.682 183s 183s x1: [4, 1, 2, 1] 183s -0.988 0.180 -0.497 0.952 183s 183s y: [1, 1, 2, 1] 183s -0.423 183s testing: mul_mat, [4, 4, 2, 3] = [1, 4, 2, 3] * [1, 4, 2, 3] 183s x0: [1, 4, 2, 3] 183s -0.573 183s -0.782 183s 0.548 183s 0.417 183s 183s x1: [1, 4, 2, 3] 183s 0.605 183s -0.755 183s -0.002 183s -0.145 183s 183s y: [4, 4, 2, 3] 183s -0.346 -0.473 0.332 0.252 183s 0.432 0.590 -0.414 -0.315 183s 0.001 0.001 -0.001 -0.001 183s 0.083 0.114 -0.080 -0.061 183s test-mul-mat0: iter:407/500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.128 183s -0.971 183s 0.834 183s -0.324 183s 183s x1: [1, 1, 1, 1] 183s 0.601 183s 183s y: [4, 1, 1, 1] 183s 0.077 -0.584 0.502 -0.195 183s testing: mul_mat, [2, 4, 4, 1] = [1, 2, 4, 1] * [1, 4, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.627 183s 0.825 183s 183s x1: [1, 4, 4, 1] 183s 0.446 183s -0.615 183s 0.659 183s -0.532 183s 183s y: [2, 4, 4, 1] 183s 0.279 0.368 183s -0.386 -0.507 183s 0.413 0.544 183s -0.334 -0.439 183s testing: mul_mat, [1, 2, 4, 3] = [1, 1, 4, 3] * [1, 2, 4, 3] 183s x0: [1, 1, 4, 3] 183s 0.858 183s 183s x1: [1, 2, 4, 3] 183s -0.213 183s 0.583 183s 183s y: [1, 2, 4, 3] 183s -0.183 183s 0.500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.490 183s 0.635 183s -0.058 183s -0.551 183s 183s x1: [1, 1, 1, 1] 183s -0.094 183s 183s y: [4, 1, 1, 1] 183s -0.046 -0.059 0.005 0.052 183s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.711 -0.684 0.903 -0.549 183s 183s x1: [4, 1, 4, 1] 183s -0.291 0.154 -0.684 0.119 183s 183s y: [1, 1, 4, 1] 183s -0.995 183s testing: mul_mat, [4, 4, 4, 3] = [1, 4, 4, 3] * [1, 4, 4, 3] 183s x0: [1, 4, 4, 3] 183s 0.003 183s 0.903 183s 0.372 183s -0.500 183s 183s x1: [1, 4, 4, 3] 183s 0.255 183s -0.083 183s 0.589 183s -0.625 183s 183s y: [4, 4, 4, 3] 183s 0.001 0.230 0.095 -0.128 183s -0.000 -0.075 -0.031 0.042 183s 0.002 0.532 0.219 -0.295 183s -0.002 -0.564 -0.232 0.313 183s test-mul-mat0: iter:408/500 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.123 183s 0.307 183s 0.110 183s 0.879 183s 183s x1: [1, 1, 1, 1] 183s -0.899 183s 183s y: [4, 1, 1, 1] 183s 0.111 -0.276 -0.099 -0.790 183s testing: mul_mat, [3, 4, 4, 1] = [1, 3, 4, 1] * [1, 4, 4, 1] 183s x0: [1, 3, 4, 1] 183s -0.446 183s -0.904 183s -0.212 183s 183s x1: [1, 4, 4, 1] 183s -0.351 183s -0.235 183s -0.130 183s -0.528 183s 183s y: [3, 4, 4, 1] 183s 0.156 0.317 0.074 183s 0.105 0.213 0.050 183s 0.058 0.118 0.028 183s 0.236 0.478 0.112 183s testing: mul_mat, [3, 3, 4, 2] = [1, 3, 4, 2] * [1, 3, 4, 2] 183s x0: [1, 3, 4, 2] 183s -0.202 183s -0.412 183s -0.019 183s 183s x1: [1, 3, 4, 2] 183s 0.731 183s -0.560 183s 0.438 183s 183s y: [3, 3, 4, 2] 183s -0.148 -0.301 -0.014 183s 0.113 0.231 0.011 183s -0.088 -0.180 -0.008 183s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.542 183s -0.551 183s -0.233 183s 183s x1: [1, 3, 1, 1] 183s 0.150 183s -0.288 183s 0.546 183s 183s y: [3, 3, 1, 1] 183s -0.081 -0.083 -0.035 183s 0.156 0.159 0.067 183s -0.295 -0.301 -0.127 183s testing: mul_mat, [2, 1, 4, 1] = [3, 2, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.354 0.107 0.383 183s -0.465 0.163 0.972 183s 183s x1: [3, 1, 4, 1] 183s 0.257 0.037 0.748 183s 183s y: [2, 1, 4, 1] 183s 0.199 0.614 183s testing: mul_mat, [3, 3, 4, 2] = [2, 3, 4, 2] * [2, 3, 4, 2] 183s x0: [2, 3, 4, 2] 183s 0.807 0.215 183s -0.195 -0.809 183s 0.759 -0.203 183s 183s x1: [2, 3, 4, 2] 183s -0.130 0.574 183s -0.033 -0.655 183s -0.752 -0.191 183s 183s y: [3, 3, 4, 2] 183s 0.019 -0.439 -0.215 183s -0.168 0.537 0.108 183s -0.648 0.301 -0.532 183s test-mul-mat0: iter:409/500 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.134 -0.378 0.868 -0.851 183s 183s x1: [4, 3, 1, 1] 183s 0.928 -0.998 0.344 0.110 183s -0.946 -0.183 0.249 -0.286 183s -0.182 -0.316 0.182 0.625 183s 183s y: [1, 3, 1, 1] 183s 0.458 183s 0.656 183s -0.230 183s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.618 -0.823 -0.754 -0.132 183s -0.109 0.064 0.552 -0.927 183s 183s x1: [4, 1, 4, 1] 183s 0.562 0.305 -0.857 0.240 183s 183s y: [2, 1, 4, 1] 183s 0.711 -0.738 183s testing: mul_mat, [4, 2, 4, 2] = [4, 4, 4, 2] * [4, 2, 4, 2] 183s x0: [4, 4, 4, 2] 183s -0.023 -0.504 0.064 0.655 183s 0.633 -0.538 0.911 -0.978 183s 0.779 -0.168 -0.463 0.617 183s 0.649 0.293 0.215 0.792 183s 183s x1: [4, 2, 4, 2] 183s 0.674 -0.813 0.637 -0.435 183s 0.251 0.189 -0.361 0.941 183s 183s y: [4, 2, 4, 2] 183s 0.150 1.870 0.099 -0.008 183s 0.492 -1.192 0.911 0.886 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.886 -0.263 -0.139 0.489 183s 183s x1: [4, 4, 1, 1] 183s -0.566 -0.955 -0.207 -0.349 183s -0.554 0.d.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backwa621 -0.631 -0.175 183s 0.384 0.294 0.924 -0.953 183s -0.140 -0.819 -0.100 0.585 183s 183s y: [1, 4, 1, 1] 183s 0.611 183s 0.329 183s -1.013 183s 0.640 183s testing: mul_mat, [3, 4, 4, 1] = [1, 3, 4, 1] * [1, 4, 4, 1] 183s x0: [1, 3, 4, 1] 183s 0.232 183s -0.203 183s 0.093 183s 183s x1: [1, 4, 4, 1] 183s 0.794 183s 0.665 183s 0.149 183s 0.388 183s 183s y: [3, 4, 4, 1] 183s 0.184 -0.161 0.074 183s 0.154 -0.135 0.062 183s 0.035 -0.030 0.014 183s 0.090 -0.079 0.036 183s testing: mul_mat, [3, 1, 4, 2] = [3, 3, 4, 2] * [3, 1, 4, 2] 183s x0: [3, 3, 4, 2] 183s -0.668 0.169 -0.132 183s 0.323 0.181 -0.471 183s 0.345 0.621 0.690 183s 183s x1: [3, 1, 4, 2] 183s 0.698 0.347 -0.930 183s 183s y: [3, 1, 4, 2] 183s -0.285 0.727 -0.186 183s test-mul-mat0: iter:410/500 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.583 -0.659 183s 0.119 0.698 183s 183s x1: [2, 3, 1, 1] 183s -0.003 0.007 183s 0.632 0.589 183s 0.773 -0.209 183s 183s y: [2, 3, 1, 1] 183s -0.003 0.005 183s -0.756 0.486 183s -0.313 -0.054 183s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 2, 4, 1] 183s -0.856 0.691 183s 0.767 0.140 183s 183s x1: [2, 2, 4, 1] 183s -0.881 -0.272 183s 0.468 0.121 183s 183s y: [2, 2, 4, 1] 183s 0.566 -0.713 183s -0.318 0.376 183s testing: mul_mat, [2, 2, 4, 4] = [2, 2, 4, 4] * [2, 2, 4, 4] 183s x0: [2, 2, 4, 4] 183s 0.640 -0.987 183s -0.549 0.226 183s 183s x1: [2, 2, 4, 4] 183s -0.945 0.771 183s 0.617 0.590 183s 183s y: [2, 2, 4, 4] 183s -1.365 0.693 183s -0.187 -0.206 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.027 0.702 183s 0.585 -0.075 183s 0.415 -0.818 183s 183s x1: [2, 2, 1, 1] 183s -0.655 -0.016 183s -0.086 -0.347 183s 183s y: [3, 2, 1, 1] 183s 0.007 -0.382 -0.259 183s -0.241 -0.024 0.248 183s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.461 0.641 0.833 183s -0.285 -0.393 -0.914 183s -0.576 -0.376 0.900 183s -0.759 0.505 0.100 183s 183s x1: [3, 2, 4, 1] 183s -0.617 0.865 0.226 183s 0.461 -0.606 0.267 183s 183s y: [4, 2, 4, 1] 183s 1.027 -0.371 0.234 0.928 183s -0.378 -0.137 0.202 -0.630 183s testing: mul_mat, [3, 3, 4, 4] = [4, 3, 4, 4] * [4, 3, 4, 4] 183s x0: [4, 3, 4, 4] 183s 0.368 -0.426 -0.303 -0.655 183s -0.414 0.685 -0.908 -0.801 183s -0.301 -0.283 0.929 -0.451 183s 183s x1: [4, 3, 4, 4] 183s -0.511 -0.940 0.144 -0.047 183s -0.886 0.787 0.549 -0.187 183s 0.083 -0.378 0.141 0.945 183s 183s y: [3, 3, 4, 4] 183s 0.199 -0.525 0.575 183s -0.705 0.556 0.638 183s -0.471 -1.179 -0.213 183s test-mul-mat0: iter:411/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.168 -0.284 183s 0.089 -0.401 183s 0.312 0.190 183s 183s x1: [2, 2, 1, 1] 183s 0.404 0.187 183s 0.377 -0.520 183s 183s y: [3, 2, 1, 1] 183s 0.015 -0.039 0.162 183s 0.211 0.242 0.019 183s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 183s x0: [2, 3, 3, 1] 183s -0.246 -0.548 183s 0.251 -0.869 183s -0.067 -0.504 183s 183s x1: [2, 3, 3, 1] 183s 0.729 0.663 183s -0.547 0.489 183s -0.621 0.819 183s 183s y: [3, 3, 3, 1] 183s -0.543 -0.394 -0.384 183s -0.134 -0.562 -0.210 183s -0.296 -0.867 -0.371 183s testing: mul_mat, [2, 3, 3, 3] = [2, 2, 3, 3] * [2, 3, 3, 3] 183s x0: [2, 2, 3, 3] 183s 0.051 0.227 183s -0.380 -0.295 183s 183s x1: [2, 3, 3, 3] 183s 0.388 -0.955 183s -0.891 -0.204 183s -0.896 0.488 183s 183s y: [2, 3, 3, 3] 183s -0.197 0.134 183s -0.092 0.399 183s 0.065 0.196 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.160 0.084 183s -0.948 0.159 183s 183s x1: [2, 2, 1, 1] 183s -0.445 -0.650 183s 0.285 -0.573 183s 183s y: [2, 2, 1, 1] 183s 0.017 0.318 183s -0.094 -0.362 183s testing: mul_mat, [4, 2, 3, 1] = [2, 4, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 4, 3, 1] 183s 0.778 -0.242 183s -0.482 -0.338 183s 0.975 0.821 183s 0.490 0.702 183s 183s x1: [2, 2, 3, 1] 183s -0.272 -0.165 183s -0.501 0.215 183s 183s y: [4, 2, 3, 1] 183s -0.172 0.187 -0.401 -0.249 183s -0.442 0.169 -0.312 -0.095 183s testing: mul_mat, [4, 2, 3, 3] = [4, 4, 3, 3] * [4, 2, 3, 3] 183s x0: [4, 4, 3, 3] 183s -0.634 -0.844 0.172 -0.630 183s -0.718 0.893 0.944 -0.508 183s 0.746 -0.911 0.394 0.278 183s 0.103 -0.569 -0.561 0.906 183s 183s x1: [4, 2, 3, 3] 183s 0.827 -0.222 0.889 0.641 183s -0.137 0.219 -0.844 -0.360 183s 183s y: [4, 2, 3, 3] 183s -0.588 -0.278 1.347 0.293 183s -0.016 -0.320 -0.734 0.009 183s test-mul-mat0: iter:412/500 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.112 183s 183s x1: [1, 1, 1, 1] 183s 0.148 183s 183s y: [1, 1, 1, 1] 183s 0.017 183s testing: mul_mat, [1, 1, 3, 1] = [1, 1, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 1, 3, 1] 183s -0.855 183s 183s x1: [1, 1, 3, 1] 183s -0.167 183s 183s y: [1, 1, 3, 1] 183s 0.143 183s testing: mul_mat, [1, 1, 3, 2] = [1, 1, 3, 2] * [1, 1, 3, 2] 183s x0: [1, 1, 3, 2] 183s 0.562 183s 183s x1: [1, 1, 3, 2] 183s -0.249 183s 183s y: [1, 1, 3, 2] 183s -0.140 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.210 183s -0.026 183s 183s x1: [1, 1, 1, 1] 183s 0.351 183s 183s y: [2, 1, 1, 1] 183s -0.074 -0.009 183s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 2, 3, 1] 183s -0.380 0.018 183s 0.542 -0.172 183s 183s x1: [2, 1, 3, 1] 183s 0.543 -0.361 183s 183s y: [2, 1, 3, 1] 183s -0.213 0.356 183s testing: mul_mat, [4, 2, 3, 2] = [2, 4, 3, 2] * [2, 2, 3, 2] 183s x0: [2, 4, 3, 2] 183s 0.316 -0.576 183s -0.306 -0.115 183s -0.804 0.397 183s -0.063 0.458 183s 183s x1: [2, 2, 3, 2] 183s -0.540 0.034 183s -0.748 -0.898 183s 183s y: [4, 2, 3, 2] 183s -0.190 0.162 0.447 0.049 183s 0.281 0.332 0.245 -0.365 183s test-mul-mat0: iter:413/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.702 0.020 183s -0.295 0.493 183s 183s x1: [2, 4, 1, 1] 183s -0.700 0.055 183s 0.099 -0.127 183s 0.451 0.232 183s 0.300 0.578 183s 183s y: [2, 4, 1, 1] 183s 0.493 0.233 183s -0.072 -0.092 183s -0.312 -0.019 183s -0.199 0.196 183s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.914 0.207 183s -0.549 0.474 183s 183s x1: [2, 2, 3, 1] 183s 0.399 -0.787 183s 0.572 0.176 183s 183s y: [2, 2, 3, 1] 183s 0.203 -0.592 183s 0.560 -0.231 183s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.268 -0.224 183s 183s x1: [2, 2, 3, 1] 183s 0.557 -0.307 183s 0.222 -0.423 183s 183s y: [1, 2, 3, 1] 183s -0.080 183s 0.035 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.689 -0.227 183s -0.065 0.532 183s 0.908 -0.803 183s -0.782 0.845 183s 183s x1: [2, 1, 1, 1] 183s -0.784 -0.805 183s 183s y: [4, 1, 1, 1] 183s -0.358 -0.377 -0.066 -0.067 183s testing: mul_mat, [1, 2, 3, 1] = [4, 1, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 1, 3, 1] 183s 0.655 0.494 0.425 0.187 183s 183s x1: [4, 2, 3, 1] 183s -0.544 0.755 -0.462 0.678 183s -0.668 0.936 -0.607 -0.692 183s 183s y: [1, 2, 3, 1] 183s -0.054 183s -0.362 183s testing: mul_mat, [1, 4, 3, 1] = [1, 1, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 1, 3, 1] 183s 0.086 183s 183s x1: [1, 4, 3, 1] 183s 0.714 183s -0.914 183s 0.290 183s -0.606 183s 183s y: [1, 4, 3, 1] 183s 0.061 183s -0.078 183s 0.025 183s -0.052 183s test-mul-mat0: iter:414/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.341 -0.181 183s 183s x1: [2, 4, 1, 1] 183s -0.574 -1.000 183s 0.674 -0.882 183s 0.270 0.316 183s 0.564 0.077 183s 183s y: [1, 4, 1, 1] 183s -0.015 183s 0.389 183s 0.035 183s 0.178 183s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 4, 2, 1] 183s 0.737 -0.104 183s 0.587 -0.565 183s 0.582 -0.917 183s -0.496 0.789 183s 183s x1: [2, 1, 2, 1] 183s -0.650 0.427 183s 183s y: [4, 1, 2, 1] 183s -0.523 -0.622 -0.770 0.659 183s testing: mul_mat, [2, 4, 2, 2] = [2, 2, 2, 2] * [2, 4, 2, 2] 183s x0: [2, 2, 2, 2] 183s -0.353 0.325 183s 0.913 0.846 183s 183s x1: [2, 4, 2, 2] 183s -0.712 -0.837 183s 0.636 0.558 183s 0.479 0.200 183s -0.365 -0.884 183s 183s y: [2, 4, 2, 2] 183s -0.021 -1.358 183s -0.043 1.053 183s -0.104 0.607 183s -0.158 -1.082 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.782 -0.830 183s -0.346 0.999 183s 183s x1: [2, 2, 1, 1] 183s -0.986 0.502 183s -0.563 -0.746 183s 183s y: [2, 2, 1, 1] 183s 0.355 0.843 183s 1.059 -0.550 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.650 -0.355 183s -0.299 -0.963 183s 183s x1: [2, 2, 2, 1] 183s -0.854 0.649 183s -0.731 0.157 183s 183s y: [2, 2, 2, 1] 183s -0.786 -0.370 183s -0.531 0.067 183s testing: mul_mat, [4, 2, 2, 2] = [2, 4, 2, 2] * [2, 2, 2, 2] 183s x0: [2, 4, 2, 2] 183s 0.597 0.119 183s -0.622 0.758 183s 0.142 0.051 183s 0.401 0.820 183s 183s x1: [2, 2, 2, 2] 183s -0.153 0.690 183s 0.371 -0.340 183s 183s y: [4, 2, 2, 2] 183s -0.009 0.618 0.013 0.505 183s 0.181 -0.488 0.036 -0.130 183s test-mul-mat0: iter:415/500 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.882 -0.732 183s -0.225 -0.090 183s 183s x1: [2, 2, 1, 1] 183s -0.678 -0.794 183s 0.946 -0.276 183s 183s y: [2, 2, 1, 1] 183s -0.017 0.224 183s 1.036 -0.189 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.094 -0.079 183s 0.272 -0.481 183s 0.276 -0.916 183s 183s x1: [2, 2, 1, 1] 183s -0.342 0.945 183s 0.309 0.708 183s 183s y: [3, 2, 1, 1] 183s -0.107 -0.548 -0.960 183s -0.027 -0.257 -0.563 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.421 0.050 183s 0.305 -0.295 183s 0.372 0.511 183s -0.349 -0.904 183s 183s x1: [2, 3, 1, 1] 183s -0.763 -0.056 183s 0.066 -0.605 183s 0.573 -0.513 183s 183s y: [4, 3, 1, 1] 183s 0.318 -0.216 -0.313 0.317 183s -0.058 0.198 -0.285 0.524 183s -0.267 0.326 -0.049 0.264 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.844 0.650 183s 0.847 0.429 183s 183s x1: [2, 4, 1, 1] 183s -0.195 -0.467 183s -0.636 0.579 183s 0.443 0.022 183s 0.524 -0.248 183s 183s y: [2, 4, 1, 1] 183s -0.139 -0.366 183s 0.913 -0.290 183s -0.360 0.384 183s -0.603 0.338 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.422 0.146 183s -0.004 -0.978 183s 0.443 -0.885 183s 0.065 -0.549 183s 183s x1: [2, 2, 1, 1] 183s -0.635 -0.074 183s 0.513 -0.398 183s 183s y: [4, 2, 1, 1] 183s 0.257 0.074 -0.216 -0.001 183s -0.274 0.388 0.579 0.252 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.355 0.698 0.455 -0.079 183s -0.031 -0.974 -0.681 -0.855 183s 0.870 0.954 -0.725 0.546 183s -0.893 0.349 -0.032 0.918 183s 183s x1: [4, 2, 1, 1] 183s -0.273 0.487 0.961 0.379 183s 0.583 -0.234 0.912 0.947 183s 183s y: [4, 2, 1, 1] 183s 0.651 -1.444 -0.263 0.730 183s 0.384 -1.221 0.140 0.238 183s test-mul-mat0: iter:416/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.476 0.715 0.797 183s -0.758 0.627 0.745 183s 183s x1: [3, 1, 1, 1] 183s -0.273 0.514 -0.663 183s 183s y: [2, 1, 1, 1] 183s -0.291 0.036 183s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.265 0.377 0.789 183s -0.220 0.714 -0.997 183s -0.744 0.429 0.801 183s -0.503 0.057 0.545 183s 183s x1: [3, 2, 3, 1] 183s -0.413 -0.018 -0.287 183s -0.544 0.090 -0.588 183s 183s y: [4, 2, 3, 1] 183s -0.343 0.364 0.070 0.050 183s -0.575 0.770 -0.028 -0.043 183s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.825 0.101 -0.337 183s 0.603 -0.454 0.822 183s -0.949 -0.987 0.152 183s -0.069 0.442 -0.983 183s 183s x1: [3, 4, 3, 1] 183s -0.827 -0.678 -0.065 183s 0.973 -0.181 0.992 183s 0.519 0.903 -0.969 183s 0.778 0.444 0.159 183s 183s y: [4, 4, 3, 1] 183s -0.730 -0.244 1.444 -0.179 183s 0.450 1.484 -0.594 -1.121 183s 0.847 -0.894 -1.531 1.316 183s 0.633 0.398 -1.152 -0.013 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.773 -0.947 -0.696 183s -0.445 -0.530 -0.544 183s 0.015 0.233 -0.980 183s -0.350 0.884 0.335 183s 183s x1: [3, 4, 1, 1] 183s -0.660 0.616 -0.308 183s 0.492 -0.453 -0.866 183s 0.509 -0.175 0.870 183s -0.789 0.095 0.217 183s 183s y: [4, 4, 1, 1] 183s -0.880 0.135 0.435 0.672 183s 1.412 0.492 0.750 -0.862 183s -0.046 -0.607 -0.885 -0.041 183s -0.851 0.183 -0.202 0.433 183s testing: mul_mat, [3, 3, 3, 1] = [4, 3, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.945 -0.284 -0.114 0.268 183s 0.265 -0.025 0.511 0.141 183s 0.576 -0.451 0.392 0.452 183s 183s x1: [4, 3, 3, 1] 183s -0.688 0.982 0.468 -0.170 183s 0.547 -0.853 0.170 0.163 183s -0.161 -0.339 0.711 -0.027 183s 183s y: [3, 3, 3, 1] 183s 0.273 0.009 -0.733 183s -0.250 0.275 0.840 183s 0.161 0.325 0.326 183s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 1, 3, 1] 183s -0.512 -0.235 0.997 183s 183s x1: [3, 4, 3, 1] 183s -0.874 -0.773 -0.735 183s 0.518 0.494 0.406 183s -0.030 -0.239 0.951 183s 0.037 0.041 -0.450 183s 183s y: [1, 4, 3, 1] 183s -0.104 183s 0.024 183s 1.020 183s -0.477 183s test-mul-mat0: iter:417/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.941 -0.938 -0.536 183s -0.657 -0.563 -0.536 183s 183s x1: [3, 2, 1, 1] 183s 0.999 -0.203 -0.402 183s -0.706 0.116 -0.559 183s 183s y: [2, 2, 1, 1] 183s -0.534 -0.327 183s 0.855 0.698 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.120 0.994 0.430 183s 0.666 0.290 -0.959 183s -0.352 0.643 0.010 183s 0.695 -0.087 0.009 183s 183s x1: [3, 2, 1, 1] 183s -0.147 0.077 0.982 183s -0.142 0.123 -0.530 183s 183s y: [4, 2, 1, 1] 183s 0.481 -1.017 0.111 -0.100 183s -0.123 0.449 0.124 -0.115 183s testing: mul_mat, [3, 4, 1, 2] = [3, 3, 1, 2] * [3, 4, 1, 2] 183s x0: [3, 3, 1, 2] 183s -0.916 -0.797 0.825 183s 0.779 0.116 -0.166 183s -0.729 -0.373 -0.862 183s 183s x1: [3, 4, 1, 2] 183s -0.508 0.511 0.304 183s 0.608 0.952 0.894 183s 0.667 -0.987 -0.641 183s -0.990 -0.550 -0.177 183s 183s y: [3, 4, 1, 2] 183s 0.309 -0.387 -0.082 183s -0.577 0.436 -1.569 183s -0.354 0.511 0.434 183s 1.199 -0.806 1.080 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.575 0.124 0.069 183s 183s x1: [3, 3, 1, 1] 183s -0.582 -0.430 -0.981 183s -0.862 0.220 0.295 183s 0.481 -0.011 -0.436 183s 183s y: [1, 3, 1, 1] 183s 0.214 183s 0.544 183s -0.308 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.556 183s 183s x1: [1, 3, 1, 1] 183s -0.492 183s 0.328 183s -0.107 183s 183s y: [1, 3, 1, 1] 183s 0.273 183s -0.182 183s 0.059 183s testing: mul_mat, [4, 1, 1, 2] = [1, 4, 1, 2] * [1, 1, 1, 2] 183s x0: [1, 4, 1, 2] 183s 0.865 183s -0.563 183s 0.651 183s 0.897 183s 183s x1: [1, 1, 1, 2] 183s 0.727 183s 183s y: [4, 1, 1, 2] 183s 0.629 -0.410 0.473 0.653 183s test-mul-mat0: iter:418/500 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.828 -0.510 -0.495 183s 183s x1: [3, 2, 1, 1] 183s 0.521 -0.908 0.076 183s -0.184 0.574 -0.935 183s 183s y: [1, 2, 1, 1] 183s 0.857 183s 0.018 183s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.719 -0.001 0.281 183s -0.737 0.242 0.569 183s 183s x1: [3, 1, 3, 1] 183s -0.103 -0.002 0.833 183s 183s y: [2, 1, 3, 1] 183s 0.308 0.549 183s testing: mul_mat, [3, 2, 3, 2] = [3, 3, 3, 2] * [3, 2, 3, 2] 183s x0: [3, 3, 3, 2] 183s -0.163 -0.789 0.371 183s -0.708 -0.647 -0.878 183s 0.129 -0.224 0.872 183s 183s x1: [3, 2, 3, 2] 183s 0.249 -0.986 0.294 183s -0.854 0.012 0.126 183s 183s y: [3, 2, 3, 2] 183s 0.846 0.204 0.509 183s 0.176 0.486 -0.003 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.660 0.315 -0.157 183s 0.147 -0.159 0.783 183s 0.521 -0.786 -0.632 183s 183s x1: [3, 3, 1, 1] 183s 0.252 0.771 -0.126 183s -0.310 0.965 -0.690 183s -0.958 -0.873 -0.417 183s 183s y: [3, 3, 1, 1] 183s 0.429 -0.184 -0.395 183s 0.208 -0.739 -0.484 183s -0.841 -0.330 0.451 183s testing: mul_mat, [4, 3, 3, 1] = [3, 4, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.944 -0.146 -0.294 183s 0.194 0.950 -0.735 183s 0.481 0.190 0.930 183s 0.179 -0.916 -0.978 183s 183s x1: [3, 3, 3, 1] 183s 0.235 0.660 -0.531 183s -0.989 0.230 -0.244 183s 0.075 0.845 0.316 183s 183s y: [4, 3, 3, 1] 183s 0.282 1.063 -0.255 -0.043 183s -0.896 0.206 -0.658 -0.150 183s -0.146 0.585 0.491 -1.070 183s testing: mul_mat, [3, 3, 3, 2] = [4, 3, 3, 2] * [4, 3, 3, 2] 183s x0: [4, 3, 3, 2] 183s 0.304 -0.618 -0.021 0.239 183s 0.683 0.018 0.903 0.376 183s -0.517 0.367 -0.378 0.124 183s 183s x1: [4, 3, 3, 2] 183s 0.419 0.557 -0.418 -0.875 183s 0.823 -0.488 -0.853 0.877 183s -0.295 0.825 -0.028 -0.065 183s 183s y: [3, 3, 3, 2] 183s -0.417 -0.411 0.037 183s 0.779 0.114 -0.173 183s -0.615 -0.237 0.458 183s test-mul-mat0: iter:419/500 183s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.214 183s 183s x1: [1, 1, 1, 1] 183s 0.563 183s 183s y: [1, 1, 1, 1] 183s 0.121 183s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 1, 4, 1] 183s -0.736 183s 183s x1: [1, 1, 4, 1] 183s 0.338 183s 183s y: [1, 1, 4, 1] 183s -0.249 183s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.745 183s 0.568 183s 183s x1: [1, 1, 4, 1] 183s 0.513 183s 183s y: [2, 1, 4, 1] 183s 0.383 0.291 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.839 183s -0.727 183s 183s x1: [1, 2, 1, 1] 183s 0.070 183s -0.303 183s 183s y: [2, 2, 1, 1] 183s -0.058 -0.051 183s 0.254 0.220 183s testing: mul_mat, [2, 1, 4, 1] = [2, 2, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 2, 4, 1] 183s 0.702 -0.281 183s 0.345 0.295 183s 183s x1: [2, 1, 4, 1] 183s 0.282 0.724 183s 183s y: [2, 1, 4, 1] 183s -0.006 0.311 183s testing: mul_mat, [1, 2, 4, 1] = [2, 1, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 1, 4, 1] 183s -0.205 -0.296 183s 183s x1: [2, 2, 4, 1] 183s -0.846 -0.751 183s 0.137 -0.509 183s 183s y: [1, 2, 4, 1] 183s 0.396 183s 0.123 183s test-mul-mat0: iter:420/500 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.722 -0.199 183s -0.098 -0.092 183s 0.070 -0.901 183s -0.536 -0.223 183s 183s x1: [2, 3, 1, 1] 183s -0.892 0.731 183s -0.462 -0.643 183s -0.132 0.029 183s 183s y: [4, 3, 1, 1] 183s 0.500 0.020 -0.721 0.316 183s 0.461 0.104 0.547 0.391 183s 0.090 0.010 -0.035 0.065 183s testing: mul_mat, [2, 4, 2, 1] = [2, 2, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.905 -0.571 183s -0.602 -0.227 183s 183s x1: [2, 4, 2, 1] 183s 0.338 -0.461 183s -0.814 0.921 183s 0.334 -0.111 183s 0.362 0.670 183s 183s y: [2, 4, 2, 1] 183s 0.569 -0.098 183s -1.263 0.281 183s 0.366 -0.176 183s -0.055 -0.370 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.425 -0.705 183s 0.182 -0.213 183s 183s x1: [2, 2, 2, 1] 183s -0.932 0.959 183s 0.330 -0.833 183s 183s y: [2, 2, 2, 1] 183s -1.072 -0.373 183s 0.727 0.237 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.054 0.345 183s 0.352 -0.875 183s 0.315 -0.227 183s 183s x1: [2, 2, 1, 1] 183s 0.272 -0.005 183s -0.095 0.447 183s 183s y: [3, 2, 1, 1] 183s -0.016 0.100 0.087 183s 0.159 -0.424 -0.132 183s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 2, 2, 1] 183s -0.187 -0.537 -0.669 183s 0.942 -0.222 0.951 183s 183s x1: [3, 2, 2, 1] 183s -0.489 0.176 0.032 183s -0.421 0.134 -0.638 183s 183s y: [2, 2, 2, 1] 183s -0.024 -0.470 183s 0.434 -1.034 183s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 1, 2, 1] 183s -0.606 0.781 183s 183s x1: [2, 3, 2, 1] 183s 0.600 -0.598 183s 0.147 -0.055 183s -0.473 0.919 183s 183s y: [1, 3, 2, 1] 183s -0.832 183s -0.132 183s 1.005 183s test-mul-mat0: iter:421/500 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.748 -0.582 -0.931 183s -0.094 0.467 0.901 183s -0.493 0.869 0.048 183s 183s x1: [3, 1, 1, 1] 183s 0.171 -0.497 -0.185 183s 183s y: [3, 1, 1, 1] 183s 0.334 -0.415 -0.525 183s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.434 -0.963 -0.108 183s 183s x1: [3, 3, 4, 1] 183s 0.452 -0.604 -0.033 183s -0.093 -0.902 0.919 183s -0.058 -0.065 0.232 183s 183s y: [1, 3, 4, 1] 183s 0.781 183s 0.729 183s 0.012 183s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.898 0.977 0.834 183s -0.406 0.079 0.495 183s -0.971 0.116 -0.613 183s 0.186 -0.517 -0.708 183s 183s x1: [3, 1, 4, 1] 183s -0.992 -0.961 -0.646 183s 183s y: [4, 1, 4, 1] 183s -0.588 0.007 1.248 0.770 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.800 0.805 -0.141 183s -0.199 -0.121 0.430 183s 183s x1: [3, 4, 1, 1] 183s -0.743 0.322 -0.327 183s 0.863 0.567 0.132 183s 0.094 0.402 0.976 183s -0.310 -0.445 0.337 183s 183s y: [2, 4, 1, 1] 183s -0.289 -0.031 183s 1.128 -0.184 183s 0.261 0.352 183s -0.654 0.260 183s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 2, 4, 1] 183s -0.922 0.639 183s 0.600 -0.043 183s 183s x1: [2, 3, 4, 1] 183s 0.453 -0.054 183s -0.829 -0.586 183s -0.781 0.394 183s 183s y: [2, 3, 4, 1] 183s -0.452 0.274 183s 0.390 -0.472 183s 0.972 -0.486 183s testing: mul_mat, [1, 2, 4, 1] = [2, 1, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 1, 4, 1] 183s -0.177 0.753 183s 183s x1: [2, 2, 4, 1] 183s 0.644 0.022 183s -0.906 -0.603 183s 183s y: [1, 2, 4, 1] 183s -0.098 183s -0.294 183s test-mul-mat0: iter:422/500 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.624 0.185 183s -0.634 -0.603 183s 0.625 -0.310 183s 0.746 -0.656 183s 183s x1: [2, 3, 1, 1] 183s 0.106 -0.358 183s -0.931 -0.872 183s -0.263 -0.533 183s 183s y: [4, 3, 1, 1] 183s 0.000 0.148 0.178 0.314 183s -0.741 1.116 -0.311 -0.122 183s -0.263 0.489 0.001 0.153 183s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 1, 4, 1] 183s 0.772 -0.317 183s 183s x1: [2, 4, 4, 1] 183s 0.812 -0.070 183s 0.140 -0.465 183s 0.753 -0.106 183s 0.718 0.645 183s 183s y: [1, 4, 4, 1] 183s 0.650 183s 0.256 183s 0.615 183s 0.350 183s testing: mul_mat, [1, 1, 4, 4] = [2, 1, 4, 4] * [2, 1, 4, 4] 183s x0: [2, 1, 4, 4] 183s -0.841 -0.502 183s 183s x1: [2, 1, 4, 4] 183s 0.772 0.766 183s 183s y: [1, 1, 4, 4] 183s -1.034 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.436 0.010 183s -0.082 0.636 183s 183s x1: [2, 1, 1, 1] 183s -0.025 -0.986 183s 183s y: [2, 1, 1, 1] 183s -0.021 -0.625 183s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.596 -0.378 183s 0.287 0.764 183s -0.247 0.315 183s 183s x1: [2, 2, 4, 1] 183s -0.824 0.165 183s -0.146 -0.689 183s 183s y: [3, 2, 4, 1] 183s -0.554 -0.110 0.256 183s 0.173 -0.569 -0.181 183s testing: mul_mat, [4, 2, 4, 4] = [3, 4, 4, 4] * [3, 2, 4, 4] 183s x0: [3, 4, 4, 4] 183s -0.432 0.111 0.372 183s 0.853 0.147 -0.832 183s 0.923 -0.419 -0.080 183s -0.219 0.626 0.580 183s 183s x1: [3, 2, 4, 4] 183s -0.124 -0.334 -0.263 183s 0.231 -0.828 0.731 183s 183s y: [4, 2, 4, 4] 183s -0.081 0.064 0.047 -0.334 183s 0.080 -0.533 0.502 -0.145 183s test-mul-mat0: iter:423/500 183s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.430 183s 0.796 183s -0.779 183s 183s x1: [1, 1, 1, 1] 183s -0.130 183s 183s y: [3, 1, 1, 1] 183s 0.056 -0.103 0.101 183s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.477 183s 0.997 183s 0.600 183s 0.440 183s 183s x1: [1, 3, 2, 1] 183s -0.897 183s -0.347 183s 0.662 183s 183s y: [4, 3, 2, 1] 183s 0.428 -0.895 -0.538 -0.395 183s 0.165 -0.346 -0.208 -0.153 183s -0.316 0.661 0.398 0.292 183s testing: mul_mat, [4, 4, 2, 2] = [1, 4, 2, 2] * [1, 4, 2, 2] 183s x0: [1, 4, 2, 2] 183s 0.933 183s -0.598 183s -0.470 183s 0.569 183s 183s x1: [1, 4, 2, 2] 183s -0.264 183s 0.470 183s -0.087 183s -0.564 183s 183s y: [4, 4, 2, 2] 183s -0.246 0.158 0.124 -0.150 183s 0.439 -0.281 -0.221 0.268 183s -0.081 0.052 0.041 -0.049 183s -0.526 0.337 0.265 -0.321 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.997 183s 0.798 183s 0.323 183s 183s x1: [1, 4, 1, 1] 183s 0.048 183s -0.869 183s -0.351 183s 0.185 183s 183s y: [3, 4, 1, 1] 183s -0.048 0.038 0.015 183s 0.866 -0.694 -0.281 183s 0.350 -0.280 -0.113 183s -0.185 0.148 0.060 183s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 1, 2, 1] 183s -0.507 -0.018 0.549 183s 183s x1: [3, 1, 2, 1] 183s -0.406 0.281 0.841 183s 183s y: [1, 1, 2, 1] 183s 0.662 183s testing: mul_mat, [3, 3, 2, 2] = [1, 3, 2, 2] * [1, 3, 2, 2] 183s x0: [1, 3, 2, 2] 183s -0.153 183s -0.395 183s 0.187 183s 183s x1: [1, 3, 2, 2] 183s -0.845 183s 0.946 183s -0.547 183s 183s y: [3, 3, 2, 2] 183s 0.129 0.333 -0.158 183s -0.144 -0.373 0.177 183s 0.083 0.216 -0.102 183s test-mul-mat0: iter:424/500 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.514 -0.019 0.348 183s -0.880 -0.832 0.993 183s -0.952 0.950 0.920 183s 0.818 0.487 -0.592 183s 183s x1: [3, 4, 1, 1] 183s 0.797 0.306 0.198 183s 0.744 0.760 -0.114 183s -0.776 0.171 0.262 183s -0.162 -0.723 0.797 183s 183s y: [4, 4, 1, 1] 183s 0.473 -0.760 -0.286 0.685 183s 0.328 -1.400 -0.091 1.046 183s -0.311 0.801 1.142 -0.707 183s 0.208 1.535 0.201 -0.956 183s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.856 -0.756 -0.856 183s -0.904 -0.459 -0.133 183s 183s x1: [3, 4, 4, 1] 183s -0.831 -0.957 0.389 183s 0.501 -0.852 -0.568 183s 0.298 0.454 0.629 183s 0.042 0.214 -0.485 183s 183s y: [2, 4, 4, 1] 183s 1.102 1.140 183s 0.702 0.014 183s -1.137 -0.561 183s 0.218 -0.072 183s testing: mul_mat, [3, 2, 4, 3] = [3, 3, 4, 3] * [3, 2, 4, 3] 183s x0: [3, 3, 4, 3] 183s -0.992 0.482 -0.317 183s -0.528 -0.681 0.675 183s -0.791 -0.087 -0.472 183s 183s x1: [3, 2, 4, 3] 183s 0.899 0.952 -0.513 183s 0.914 0.492 -0.446 183s 183s y: [3, 2, 4, 3] 183s -0.270 -1.469 -0.551 183s -0.528 -1.119 -0.556 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.875 0.766 0.622 183s 0.577 0.595 -0.848 183s 0.424 0.345 -0.698 183s 183s x1: [3, 3, 1, 1] 183s -0.741 -0.481 -0.813 183s -0.018 0.487 -0.040 183s -0.828 0.823 -0.822 183s 183s y: [3, 3, 1, 1] 183s -1.523 -0.025 0.086 183s 0.332 0.313 0.188 183s -0.606 0.709 0.507 183s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.111 -0.953 0.137 183s 183s x1: [3, 3, 4, 1] 183s -0.954 0.487 -0.730 183s 0.845 -0.249 0.752 183s -0.572 -0.733 -0.665 183s 183s y: [1, 3, 4, 1] 183s -0.670 183s 0.434 183s 0.544 183s testing: mul_mat, [1, 3, 4, 3] = [1, 1, 4, 3] * [1, 3, 4, 3] 183s x0: [1, 1, 4, 3] 183s 0.935 183s 183s x1: [1, 3, 4, 3] 183s -0.750 183s 0.204 183s 0.363 183s 183s y: [1, 3, 4, 3] 183s -0.701 183s 0.191 183s 0.339 183s test-mul-mat0: iter:425/500 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.371 183s 183s x1: [1, 4, 1, 1] 183s 0.583 183s -0.578 183s -0.844 183s 0.440 183s 183s y: [1, 4, 1, 1] 183s -0.216 183s 0.214 183s 0.313 183s -0.163 183s testing: mul_mat, [3, 1, 4, 1] = [1, 3, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 3, 4, 1] 183s 0.233 183s 0.808 183s 0.099 183s 183s x1: [1, 1, 4, 1] 183s -0.923 183s 183s y: [3, 1, 4, 1] 183s -0.215 -0.746 -0.092 183s testing: mul_mat, [4, 3, 4, 3] = [1, 4, 4, 3] * [1, 3, 4, 3] 183s x0: [1, 4, 4, 3] 183s -0.411 183s 0.328 183s -0.506 183s -0.750 183s 183s x1: [1, 3, 4, 3] 183s -0.277 183s -0.654 183s 0.435 183s 183s y: [4, 3, 4, 3] 183s 0.114 -0.091 0.140 0.208 183s 0.269 -0.215 0.331 0.490 183s -0.179 0.143 -0.220 -0.326 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.090 183s -0.425 183s 0.277 183s 183s x1: [1, 4, 1, 1] 183s -0.786 183s -0.844 183s 0.547 183s 0.095 183s 183s y: [3, 4, 1, 1] 183s -0.071 0.334 -0.217 183s -0.076 0.359 -0.234 183s 0.049 -0.233 0.151 183s 0.009 -0.041 0.026 183s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 3, 4, 1] 183s 0.228 -0.705 0.277 183s 0.452 0.164 0.999 183s -0.625 -0.649 -0.201 183s 183s x1: [3, 1, 4, 1] 183s -0.527 -0.001 0.556 183s 183s y: [3, 1, 4, 1] 183s 0.034 0.318 0.218 183s testing: mul_mat, [2, 3, 4, 3] = [3, 2, 4, 3] * [3, 3, 4, 3] 183s x0: [3, 2, 4, 3] 183s 0.674 0.300 0.906 183s 0.802 0.217 -0.188 183s 183s x1: [3, 3, 4, 3] 183s -0.275 0.730 -0.379 183s 0.524 -0.871 -0.166 183s -0.522 -0.924 -0.857 183s 183s y: [2, 3, 4, 3] 183s -0.310 0.009 183s -0.058 0.263 183s -1.405 -0.458 183s test-mul-mat0: iter:426/500 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.879 183s 0.959 183s 0.122 183s 183s x1: [1, 4, 1, 1] 183s 0.887 183s 0.279 183s -0.659 183s -0.240 183s 183s y: [3, 4, 1, 1] 183s -0.779 0.851 0.108 183s -0.246 0.268 0.034 183s 0.579 -0.632 -0.081 183s 0.211 -0.230 -0.029 183s testing: mul_mat, [3, 3, 2, 1] = [1, 3, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 3, 2, 1] 183s 0.795 183s 0.374 183s 0.789 183s 183s x1: [1, 3, 2, 1] 183s -0.509 183s -0.599 183s -0.718 183s 183s y: [3, 3, 2, 1] 183s -0.405 -0.190 -0.402 183s -0.477 -0.224 -0.473 183s -0.571 -0.269 -0.567 183s testing: mul_mat, [2, 3, 2, 3] = [1, 2, 2, 3] * [1, 3, 2, 3] 183s x0: [1, 2, 2, 3] 183s -0.002 183s -0.519 183s 183s x1: [1, 3, 2, 3] 183s 0.661 183s 0.256 183s 0.080 183s 183s y: [2, 3, 2, 3] 183s -0.001 -0.343 183s -0.000 -0.133 183s -0.000 -0.042 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.191 183s -0.759 183s -0.826 183s 183s x1: [1, 2, 1, 1] 183s 0.637 183s 0.678 183s 183s y: [3, 2, 1, 1] 183s -0.122 -0.483 -0.526 183s -0.130 -0.515 -0.560 183s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 4, 2, 1] 183s -0.252 0.348 0.666 183s 0.149 -0.904 -0.455 183s -0.653 -0.656 0.879 183s 0.345 0.829 -0.736 183s 183s x1: [3, 1, 2, 1] 183s 0.154 -0.768 -0.517 183s 183s y: [4, 1, 2, 1] 183s -0.650 0.952 -0.050 -0.203 183s testing: mul_mat, [1, 3, 2, 3] = [4, 1, 2, 3] * [4, 3, 2, 3] 183s x0: [4, 1, 2, 3] 183s 0.123 0.773 -0.581 0.422 183s 183s x1: [4, 3, 2, 3] 183s -0.882 0.876 0.806 -0.932 183s 0.986 0.878 0.156 -0.266 183s 0.027 0.502 -0.920 -0.625 183s 183s y: [1, 3, 2, 3] 183s -0.293 183s 0.597 183s 0.663 183s test-mul-mat0: iter:427/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.423 0.023 183s 183s x1: [2, 4, 1, 1] 183s -0.772 0.609 183s -0.976 0.351 183s 0.382 -0.557 183s -0.227 -0.990 183s 183s y: [1, 4, 1, 1] 183s 0.340 183s 0.421 183s -0.174 183s 0.073 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.591 -0.409 183s 0.280 0.672 183s -0.581 -0.247 183s 183s x1: [2, 1, 1, 1] 183s 0.456 -0.853 183s 183s y: [3, 1, 1, 1] 183s 0.619 -0.446 -0.054 183s testing: mul_mat, [2, 3, 1, 2] = [2, 2, 1, 2] * [2, 3, 1, 2] 183s x0: [2, 2, 1, 2] 183s -0.871 -0.306 183s -0.954 -0.428 183s 183s x1: [2, 3, 1, 2] 183s -0.151 0.787 183s -0.974 -0.611 183s -0.276 0.916 183s 183s y: [2, 3, 1, 2] 183s -0.109 -0.193 183s 1.035 1.191 183s -0.040 -0.129 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.218 -0.626 183s -0.042 0.201 183s 183s x1: [2, 2, 1, 1] 183s 0.079 0.112 183s -0.809 0.678 183s 183s y: [2, 2, 1, 1] 183s -0.087 0.019 183s -0.248 0.170 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.712 -0.232 183s 0.653 -0.704 183s 0.241 0.064 183s -0.041 -0.889 183s 183s x1: [2, 2, 1, 1] 183s 0.711 0.224 183s -0.012 0.737 183s 183s y: [4, 2, 1, 1] 183s 0.454 0.307 0.185 -0.228 183s -0.179 -0.527 0.044 -0.655 183s testing: mul_mat, [2, 2, 1, 2] = [4, 2, 1, 2] * [4, 2, 1, 2] 183s x0: [4, 2, 1, 2] 183s -0.509 -0.169 -0.776 0.191 183s -0.513 0.479 -0.556 0.877 183s 183s x1: [4, 2, 1, 2] 183s -0.040 0.192 -0.195 0.005 183s 0.764 -0.727 0.061 0.420 183s 183s y: [2, 2, 1, 2] 183s 0.140 0.226 183s -0.234 -0.406 183s test-mul-mat0: iter:428/500 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.342 183s -0.036 183s 0.355 183s 0.948 183s 183s x1: [1, 4, 1, 1] 183s 0.007 183s 0.590 183s -0.850 183s 0.146 183s 183s y: [4, 4, 1, 1] 183s -0.002 -0.000 0.002 0.006 183s -0.202 -0.021 0.209 0.560 183s 0.290 0.030 -0.301 -0.806 183s -0.050 -0.005 0.052 0.139 183s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 2, 2, 1] 183s -0.671 183s 0.923 183s 183s x1: [1, 4, 2, 1] 183s 0.741 183s -0.038 183s -0.561 183s -0.772 183s 183s y: [2, 4, 2, 1] 183s -0.497 0.684 183s 0.025 -0.035 183s 0.376 -0.518 183s 0.518 -0.713 183s testing: mul_mat, [2, 2, 2, 2] = [1, 2, 2, 2] * [1, 2, 2, 2] 183s x0: [1, 2, 2, 2] 183s 0.796 183s -0.359 183s 183s x1: [1, 2, 2, 2] 183s 0.904 183s -0.526 183s 183s y: [2, 2, 2, 2] 183s 0.720 -0.325 183s -0.419 0.189 183s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.681 183s 0.876 183s -0.127 183s 183s x1: [1, 2, 1, 1] 183s 0.880 183s 0.648 183s 183s y: [3, 2, 1, 1] 183s -0.599 0.771 -0.112 183s -0.441 0.568 -0.082 183s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 1, 2, 1] 183s 0.631 -0.062 0.080 183s 183s x1: [3, 1, 2, 1] 183s -0.763 -0.672 0.111 183s 183s y: [1, 1, 2, 1] 183s -0.431 183s testing: mul_mat, [1, 3, 2, 2] = [1, 1, 2, 2] * [1, 3, 2, 2] 183s x0: [1, 1, 2, 2] 183s -0.134 183s 183s x1: [1, 3, 2, 2] 183s -0.349 183s -0.428 183s 0.010 183s 183s y: [1, 3, 2, 2] 183s 0.047 183s 0.058 183s -0.001 183s test-mul-mat0: iter:429/500 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.600 0.485 0.171 -0.182 183s -0.984 -0.968 0.683 0.626 183s 183s x1: [4, 4, 1, 1] 183s -0.236 0.727 -0.371 0.370 183s 0.358 0.567 -0.550 -0.282 183s 0.349 -0.590 0.369 0.921 183s 0.420 0.143 -0.711 0.072 183s 183s y: [2, 4, 1, 1] 183s 0.080 -0.493 183s 0.447 -1.453 183s -0.182 1.057 183s 0.187 -0.992 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.237 -0.421 0.900 0.213 183s 183s x1: [4, 2, 1, 1] 183s 0.212 -0.575 -0.890 -0.371 183s 0.495 0.704 0.393 0.221 183s 183s y: [1, 2, 1, 1] 183s -0.689 183s -0.012 183s testing: mul_mat, [3, 1, 1, 2] = [4, 3, 1, 2] * [4, 1, 1, 2] 183s x0: [4, 3, 1, 2] 183s 1.000 0.060 0.601 0.171 183s 0.877 0.617 0.203 0.561 183s 0.243 -0.585 0.985 0.353 183s 183s x1: [4, 1, 1, 2] 183s 0.297 0.250 0.623 -0.334 183s 183s y: [3, 1, 1, 2] 183s 0.629 0.354 0.422 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.426 0.669 0.657 0.841 183s 183s x1: [4, 3, 1, 1] 183s -0.963 0.244 0.712 0.080 183s 0.053 0.172 0.196 0.052 183s -0.769 -0.204 -0.777 -0.891 183s 183s y: [1, 3, 1, 1] 183s 0.288 183s 0.310 183s -1.723 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.322 183s 0.836 183s 183s x1: [1, 4, 1, 1] 183s 0.655 183s 0.010 183s -0.115 183s 0.135 183s 183s y: [2, 4, 1, 1] 183s 0.211 0.547 183s 0.003 0.008 183s -0.037 -0.097 183s 0.043 0.113 183s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 183s x0: [2, 4, 1, 2] 183s 0.695 0.002 183s 0.291 -0.947 183s -0.028 0.991 183s -0.061 -0.826 183s 183s x1: [2, 1, 1, 2] 183s 0.459 -0.478 183s 183s y: [4, 1, 1, 2] 183s 0.317 0.586 -0.487 0.367 183s test-mul-mat0: iter:430/500 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.261 0.926 0.401 183s 183s x1: [3, 2, 1, 1] 183s 0.125 0.418 -0.027 183s 0.193 -0.261 -0.191 183s 183s y: [1, 2, 1, 1] 183s 0.409 183s -0.268 183s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.140 -0.211 -0.028 183s 0.553 -0.028 0.087 183s -0.914 -0.189 -0.788 183s 183s x1: [3, 1, 4, 1] 183s -0.804 0.621 -0.308 183s 183s y: [3, 1, 4, 1] 183s -0.010 -0.488 0.859 183s testing: mul_mat, [4, 3, 4, 3] = [3, 4, 4, 3] * [3, 3, 4, 3] 183s x0: [3, 4, 4, 3] 183s -0.932 -0.447 -0.680 183s 0.294 -0.641 0.063 183s -0.742 0.763 0.624 183s 0.844 -0.143 -0.264 183s 183s x1: [3, 3, 4, 3] 183s 0.132 0.343 -0.441 183s 0.344 -0.154 -0.657 183s -0.251 0.089 -0.064 183s 183s y: [4, 3, 4, 3] 183s 0.023 -0.209 -0.111 0.179 183s 0.195 0.158 -0.782 0.486 183s 0.238 -0.135 0.214 -0.208 183s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.995 -0.480 -0.574 183s -0.758 -0.355 0.825 183s -0.233 0.533 -0.121 183s 0.988 0.001 0.229 183s 183s x1: [3, 4, 1, 1] 183s 0.386 -0.298 0.630 183s -0.337 0.816 0.308 183s 0.979 0.097 -0.675 183s -0.937 0.569 0.329 183s 183s y: [4, 4, 1, 1] 183s -0.602 0.333 -0.325 0.525 183s -0.232 0.220 0.476 -0.263 183s -0.633 -1.333 -0.095 0.813 183s 0.470 0.780 0.482 -0.850 183s testing: mul_mat, [3, 3, 4, 1] = [4, 3, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 3, 4, 1] 183s 0.891 -0.055 -0.356 -0.230 183s -0.082 0.699 -0.988 -0.850 183s 0.856 -0.379 -0.614 0.030 183s 183s x1: [4, 3, 4, 1] 183s -0.897 0.448 -0.218 -0.238 183s 0.603 0.410 -0.852 -0.695 183s 0.041 -0.189 -0.879 -0.652 183s 183s y: [3, 3, 4, 1] 183s -0.692 0.804 -0.811 183s 0.977 1.670 0.863 183s 0.509 1.287 0.627 183s testing: mul_mat, [2, 4, 4, 3] = [3, 2, 4, 3] * [3, 4, 4, 3] 183s x0: [3, 2, 4, 3] 183s 0.420 0.304 0.151 183s -0.322 -0.888 -0.352 183s 183s x1: [3, 4, 4, 3] 183s -0.615 0.311 0.515 183s 0.201 0.708 -0.976 183s -0.862 0.040 0.285 183s -0.586 0.982 -0.850 183s 183s y: [2, 4, 4, 3] 183s -0.086 -0.260 183s 0.152 -0.351 183s -0.307 0.141 183s -0.076 -0.385 183s test-mul-mat0: iter:431/500 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.042 -0.146 -0.560 183s 183s x1: [3, 1, 1, 1] 183s 0.879 0.136 0.131 183s 183s y: [1, 1, 1, 1] 183s -0.056 183s testing: mul_mat, [2, 1, 4, 1] = [3, 2, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.677 0.768 -0.248 183s 0.316 -0.079 0.154 183s 183s x1: [3, 1, 4, 1] 183s 0.028 0.244 0.067 183s 183s y: [2, 1, 4, 1] 183s 0.151 0.000 183s testing: mul_mat, [3, 2, 4, 2] = [3, 3, 4, 2] * [3, 2, 4, 2] 183s x0: [3, 3, 4, 2] 183s -0.659 -0.420 0.859 183s 0.459 0.407 0.865 183s 0.487 0.445 0.958 183s 183s x1: [3, 2, 4, 2] 183s -0.076 0.670 0.469 183s -0.793 0.392 0.302 183s 183s y: [3, 2, 4, 2] 183s 0.172 0.643 0.710 183s 0.617 0.057 0.077 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.089 0.099 -0.354 183s -0.986 -0.386 0.057 183s 0.146 0.084 -0.070 183s 183s x1: [3, 3, 1, 1] 183s 0.992 0.175 0.257 183s 0.788 -0.628 -0.447 183s -0.663 0.649 0.732 183s 183s y: [3, 3, 1, 1] 183s -0.162 -1.031 0.142 183s 0.026 -0.560 0.094 183s -0.136 0.445 -0.094 183s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 3, 4, 1] 183s -0.301 0.650 -0.406 183s 0.270 -0.801 -0.697 183s 0.338 -0.141 0.637 183s 183s x1: [3, 3, 4, 1] 183s 0.490 -0.961 0.026 183s -0.789 0.420 -0.170 183s -0.783 0.064 0.831 183s 183s y: [3, 3, 4, 1] 183s -0.783 0.884 0.318 183s 0.579 -0.431 -0.434 183s -0.060 -0.841 0.255 183s testing: mul_mat, [3, 3, 4, 2] = [3, 3, 4, 2] * [3, 3, 4, 2] 183s x0: [3, 3, 4, 2] 183s 0.795 0.175 0.411 183s 0.980 0.723 -0.168 183s 0.768 -0.938 -0.820 183s 183s x1: [3, 3, 4, 2] 183s 0.280 -0.452 -0.690 183s -0.083 0.874 0.089 183s 0.885 -0.731 0.992 183s 183s y: [3, 3, 4, 2] 183s -0.140 0.063 1.205 183s 0.123 0.535 -0.956 183s 0.983 0.173 0.551 183s test-mul-mat0: iter:432/500 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.621 183s -0.093 183s 0.225 183s 183s x1: [1, 4, 1, 1] 183s 0.574 183s -0.396 183s 0.721 183s 0.407 183s 183s y: [3, 4, 1, 1] 183s -0.356 -0.053 0.129 183s 0.246 0.037 -0.089 183s -0.447 -0.067 0.162 183s -0.252 -0.038 0.091 183s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.396 183s 0.321 183s 183s x1: [1, 3, 1, 1] 183s 0.494 183s -0.747 183s 0.549 183s 183s y: [2, 3, 1, 1] 183s -0.196 0.159 183s 0.296 -0.240 183s -0.218 0.176 183s testing: mul_mat, [4, 2, 1, 2] = [1, 4, 1, 2] * [1, 2, 1, 2] 183s x0: [1, 4, 1, 2] 183s -0.820 183s 0.311 183s 0.706 183s 0.096 183s 183s x1: [1, 2, 1, 2] 183s -0.572 183s -0.981 183s 183s y: [4, 2, 1, 2] 183s 0.469 -0.178 -0.404 -0.055 183s 0.805 -0.305 -0.692 -0.094 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.862 183s -0.678 183s 0.180 183s 0.774 183s 183s x1: [1, 4, 1, 1] 183s -0.399 183s -0.800 183s 0.534 183s -0.398 183s 183s y: [4, 4, 1, 1] 183s 0.344 0.271 -0.072 -0.309 183s 0.689 0.542 -0.144 -0.620 183s -0.460 -0.362 0.096 0.413 183s 0.343 0.270 -0.072 -0.309 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.452 -0.338 0.442 -0.913 183s -0.766 -0.037 -0.009 -0.177 183s -0.056 -0.188 -0.566 0.680 183s 0.870 0.123 0.147 0.958 183s 183s x1: [4, 1, 1, 1] 183s -0.298 -0.912 -0.001 -0.804 183s 183s y: [4, 1, 1, 1] 183s 1.177 0.405 -0.357 -1.142 183s testing: mul_mat, [3, 4, 1, 2] = [4, 3, 1, 2] * [4, 4, 1, 2] 183s x0: [4, 3, 1, 2] 183s 0.015 -0.053 -0.941 0.259 183s 0.200 0.016 -0.393 0.197 183s -0.122 -0.918 0.988 -0.910 183s 183s x1: [4, 4, 1, 2] 183s -0.856 -0.720 -0.842 0.678 183s -0.118 0.932 0.816 0.204 183s 0.112 0.591 0.906 0.200 183s -0.410 -0.898 0.540 0.138 183s 183s y: [3, 4, 1, 2] 183s 0.993 0.282 -0.684 183s -0.766 -0.290 -0.220 183s -0.830 -0.285 0.158 183s -0.431 -0.281 1.282 183s test-mul-mat0: iter:433/500 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.015 -0.594 0.806 183s 0.350 -0.168 0.089 183s 183s x1: [3, 4, 1, 1] 183s 0.726 -0.004 0.367 183s -0.259 -0.804 -0.755 183s 0.688 0.211 -0.673 183s 0.747 0.818 -0.685 183s 183s y: [2, 4, 1, 1] 183s 0.309 0.288 183s -0.134 -0.023 183s -0.658 0.145 183s -1.027 0.063 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.896 0.094 -0.304 183s 0.371 -0.536 -0.577 183s 183s x1: [3, 2, 1, 1] 183s -0.308 -0.745 0.820 183s 0.854 0.799 0.439 183s 183s y: [2, 2, 1, 1] 183s -0.595 -0.188 183s 0.707 -0.364 183s testing: mul_mat, [2, 2, 1, 2] = [3, 2, 1, 2] * [3, 2, 1, 2] 183s x0: [3, 2, 1, 2] 183s 0.125 0.412 0.958 183s 0.956 -0.499 -0.350 183s 183s x1: [3, 2, 1, 2] 183s -0.633 0.831 0.164 183s -0.437 -0.924 -0.148 183s 183s y: [2, 2, 1, 2] 183s 0.421 -1.077 183s -0.577 0.095 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.149 -0.387 -0.792 183s 183s x1: [3, 2, 1, 1] 183s 0.906 -0.147 0.026 183s 0.278 0.318 0.449 183s 183s y: [1, 2, 1, 1] 183s 0.171 183s -0.437 183s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.935 183s 0.575 183s 183s x1: [1, 3, 1, 1] 183s 0.225 183s 0.465 183s 0.982 183s 183s y: [2, 3, 1, 1] 183s -0.211 0.129 183s -0.435 0.267 183s -0.919 0.564 183s testing: mul_mat, [3, 1, 1, 2] = [2, 3, 1, 2] * [2, 1, 1, 2] 183s x0: [2, 3, 1, 2] 183s -0.860 -0.211 183s 0.428 0.639 183s 0.584 0.905 183s 183s x1: [2, 1, 1, 2] 183s -0.653 0.671 183s 183s y: [3, 1, 1, 2] 183s 0.420 0.149 0.226 183s test-mul-mat0: iter:434/500 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.758 -0.815 0.912 -0.576 183s 183s x1: [4, 2, 1, 1] 183s -0.522 -0.016 -0.289 -0.315 183s -0.791 -0.824 -0.333 0.838 183s 183s y: [1, 2, 1, 1] 183s 0.327 183s 0.485 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.353 -0.276 -0.518 -0.354 183s -0.627 -0.854 -0.731 0.928 183s 183s x1: [4, 1, 1, 1] 183s 0.657 -0.006 -0.436 0.085 183s 183s y: [2, 1, 1, 1] 183s 0.429 -0.010 183s testing: mul_mat, [2, 2, 1, 4] = [4, 2, 1, 4] * [4, 2, 1, 4] 183s x0: [4, 2, 1, 4] 183s -0.760 0.421 -0.974 0.479 183s -0.097 0.754 0.403 -0.405 183s 183s x1: [4, 2, 1, 4] 183s -0.336 -0.081 -0.239 -0.501 183s 0.017 0.239 0.483 0.729 183s 183s y: [2, 2, 1, 4] 183s 0.214 0.078 183s -0.034 0.078 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.121 0.106 0.807 -0.567 183s 0.243 0.901 0.590 0.222 183s 0.817 0.926 -0.435 0.786 183s -0.282 -0.979 -0.114 -0.173 183s 183s x1: [4, 2, 1, 1] 183s 0.257 -0.976 0.593 -0.840 183s 0.778 -0.004 -0.245 0.438 183s 183s y: [4, 2, 1, 1] 183s 0.820 -0.653 -1.612 0.960 183s -0.540 0.138 1.083 -0.264 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.789 -0.636 0.668 0.919 183s 0.742 -0.888 -0.615 0.258 183s 183s x1: [4, 4, 1, 1] 183s -0.864 -0.225 -0.299 0.972 183s -0.226 -0.185 0.229 -0.202 183s -0.592 0.389 -0.424 0.404 183s -0.855 -0.986 -0.606 0.024 183s 183s y: [2, 4, 1, 1] 183s 1.518 -0.007 183s 0.263 -0.197 183s 0.308 -0.420 183s 0.919 0.619 183s testing: mul_mat, [3, 4, 1, 4] = [2, 3, 1, 4] * [2, 4, 1, 4] 183s x0: [2, 3, 1, 4] 183s -0.573 -0.054 183s -0.341 -0.032 183s 0.878 0.245 183s 183s x1: [2, 4, 1, 4] 183s -0.049 -0.195 183s 0.691 -0.828 183s -0.409 -0.481 183s -0.692 0.366 183s 183s y: [3, 4, 1, 4] 183s 0.039 0.023 -0.091 183s -0.351 -0.209 0.405 183s 0.261 0.155 -0.477 183s 0.376 0.224 -0.518 183s test-mul-mat0: iter:435/500 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.799 -0.978 -0.471 183s -0.042 -0.655 0.172 183s 183s x1: [3, 3, 1, 1] 183s -0.209 0.326 0.514 183s 0.218 0.985 0.392 183s -0.836 -0.047 -0.363 183s 183s y: [2, 3, 1, 1] 183s -0.394 -0.116 183s -1.322 -0.587 183s 0.886 0.004 183s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.154 -0.552 -0.109 183s -0.893 0.085 0.220 183s 183s x1: [3, 2, 3, 1] 183s -0.385 -0.817 0.661 183s 0.316 0.549 -0.610 183s 183s y: [2, 2, 3, 1] 183s 0.320 0.420 183s -0.188 -0.370 183s testing: mul_mat, [1, 2, 3, 2] = [3, 1, 3, 2] * [3, 2, 3, 2] 183s x0: [3, 1, 3, 2] 183s -0.298 0.869 0.032 183s 183s x1: [3, 2, 3, 2] 183s 0.282 -0.116 -0.457 183s 0.944 0.889 0.858 183s 183s y: [1, 2, 3, 2] 183s -0.199 183s 0.519 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.633 0.809 0.894 183s 0.561 -0.739 -0.526 183s -0.513 0.772 0.814 183s 0.730 0.945 -0.074 183s 183s x1: [3, 1, 1, 1] 183s -0.003 -0.183 -0.802 183s 183s y: [4, 1, 1, 1] 183s -0.867 0.555 -0.793 -0.115 183s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 1, 3, 1] 183s -0.377 -0.604 -0.227 0.859 183s 183s x1: [4, 3, 3, 1] 183s 0.346 -0.415 -0.587 0.115 183s 0.412 -0.197 -0.328 -0.597 183s 0.180 0.236 0.865 -0.349 183s 183s y: [1, 3, 3, 1] 183s 0.352 183s -0.474 183s -0.706 183s testing: mul_mat, [1, 4, 3, 2] = [1, 1, 3, 2] * [1, 4, 3, 2] 183s x0: [1, 1, 3, 2] 183s -0.847 183s 183s x1: [1, 4, 3, 2] 183s -0.224 183s -0.281 183s 0.387 183s -0.927 183s 183s y: [1, 4, 3, 2] 183s 0.190 183s 0.238 183s -0.328 183s 0.786 183s test-mul-mat0: iter:436/500 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.681 0.692 0.711 0.140 183s 183s x1: [4, 1, 1, 1] 183s -0.757 -0.895 -0.401 0.223 183s 183s y: [1, 1, 1, 1] 183s -1.389 183s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 2, 2, 1] 183s 0.872 -0.731 0.888 -0.975 183s -0.375 -0.978 -0.855 -0.814 183s 183s x1: [4, 1, 2, 1] 183s -0.002 -0.908 0.389 -0.746 183s 183s y: [2, 1, 2, 1] 183s 1.734 1.163 183s testing: mul_mat, [1, 2, 2, 3] = [4, 1, 2, 3] * [4, 2, 2, 3] 183s x0: [4, 1, 2, 3] 183s 0.225 -0.024 -0.209 -0.723 183s 183s x1: [4, 2, 2, 3] 183s 0.932 0.612 -0.711 0.624 183s 0.322 0.429 -0.379 0.415 183s 183s y: [1, 2, 2, 3] 183s -0.107 183s -0.158 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.016 0.329 -0.594 0.993 183s 183s x1: [4, 1, 1, 1] 183s 0.152 -0.137 -0.946 -0.653 183s 183s y: [1, 1, 1, 1] 183s -0.129 183s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.464 183s -0.292 183s -0.253 183s 183s x1: [1, 4, 2, 1] 183s -0.880 183s -0.316 183s 0.169 183s 0.836 183s 183s y: [3, 4, 2, 1] 183s 0.409 0.257 0.223 183s 0.147 0.093 0.080 183s -0.079 -0.049 -0.043 183s -0.388 -0.244 -0.212 183s testing: mul_mat, [2, 1, 2, 3] = [3, 2, 2, 3] * [3, 1, 2, 3] 183s x0: [3, 2, 2, 3] 183s -0.861 -0.963 -0.664 183s -0.189 -0.787 -0.737 183s 183s x1: [3, 1, 2, 3] 183s -0.713 0.498 0.197 183s 183s y: [2, 1, 2, 3] 183s 0.003 -0.403 183s test-mul-mat0: iter:437/500 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.568 0.489 0.360 183s -0.647 -0.484 -0.665 183s 0.644 0.029 -0.841 183s 183s x1: [3, 1, 1, 1] 183s 0.311 -0.864 0.086 183s 183s y: [3, 1, 1, 1] 183s -0.215 0.160 0.103 183s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 3, 2, 1] 183s -0.200 -0.593 -0.283 183s -0.633 0.896 -0.923 183s -0.280 -0.588 -0.588 183s 183s x1: [3, 3, 2, 1] 183s 0.789 -0.790 0.846 183s -0.884 -0.535 0.391 183s -0.254 -0.522 0.963 183s 183s y: [3, 3, 2, 1] 183s 0.070 -1.988 -0.254 183s 0.383 -0.282 0.331 183s 0.088 -1.197 -0.188 183s testing: mul_mat, [2, 3, 2, 4] = [3, 2, 2, 4] * [3, 3, 2, 4] 183s x0: [3, 2, 2, 4] 183s 0.709 -0.749 -0.622 183s 0.015 -0.924 0.278 183s 183s x1: [3, 3, 2, 4] 183s 0.016 -0.405 0.771 183s -0.469 0.895 -0.658 183s 0.028 -0.645 0.783 183s 183s y: [2, 3, 2, 4] 183s -0.165 0.589 183s -0.593 -1.017 183s 0.015 0.814 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.783 0.612 0.353 183s -0.697 0.327 0.864 183s 183s x1: [3, 2, 1, 1] 183s 0.630 0.120 0.259 183s 0.390 0.006 0.168 183s 183s y: [2, 2, 1, 1] 183s 0.658 -0.176 183s 0.369 -0.125 183s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.014 0.478 183s 0.809 0.444 183s 0.815 -0.290 183s 0.589 -0.264 183s 183s x1: [2, 3, 2, 1] 183s -0.466 -0.889 183s -0.634 -0.880 183s 0.885 -0.054 183s 183s y: [4, 3, 2, 1] 183s -0.418 -0.772 -0.121 -0.040 183s -0.411 -0.903 -0.261 -0.142 183s -0.039 0.691 0.736 0.536 183s testing: mul_mat, [2, 2, 2, 4] = [4, 2, 2, 4] * [4, 2, 2, 4] 183s x0: [4, 2, 2, 4] 183s -0.712 0.179 -0.020 -0.121 183s -0.915 0.112 0.906 0.804 183s 183s x1: [4, 2, 2, 4] 183s 0.560 0.605 0.360 0.671 183s 0.971 0.480 0.555 -0.083 183s 183s y: [2, 2, 2, 4] 183s -0.379 0.421 183s -0.607 -0.398 183s test-mul-mat0: iter:438/500 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.212 0.180 -0.951 183s -0.566 -0.532 -0.936 183s -0.747 0.663 -0.595 183s 183s x1: [3, 1, 1, 1] 183s -0.650 -0.040 0.294 183s 183s y: [3, 1, 1, 1] 183s -0.424 0.115 0.285 183s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.925 0.476 -0.795 183s -0.282 0.980 0.841 183s 183s x1: [3, 3, 4, 1] 183s -0.776 0.804 0.178 183s 0.731 -0.411 0.090 183s -0.802 -0.879 -0.729 183s 183s y: [2, 3, 4, 1] 183s 0.959 1.156 183s -0.943 -0.533 183s 0.903 -1.249 183s testing: mul_mat, [4, 2, 4, 4] = [3, 4, 4, 4] * [3, 2, 4, 4] 183s x0: [3, 4, 4, 4] 183s 0.376 -0.719 -0.660 183s -0.289 -0.106 0.581 183s -0.880 0.876 0.711 183s -0.786 0.913 -0.650 183s 183s x1: [3, 2, 4, 4] 183s -0.409 0.868 -0.928 183s 0.436 0.746 0.427 183s 183s y: [4, 2, 4, 4] 183s -0.165 -0.513 0.460 1.717 183s -0.653 0.043 0.572 0.061 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.385 0.018 -0.087 183s -0.658 0.485 -0.216 183s -0.057 -0.027 0.595 183s 183s x1: [3, 4, 1, 1] 183s 0.581 0.299 0.776 183s 0.516 -0.671 -0.482 183s 0.979 -0.079 0.831 183s 0.417 -0.866 0.615 183s 183s y: [3, 4, 1, 1] 183s -0.286 -0.405 0.421 183s -0.169 -0.560 -0.298 183s -0.451 -0.862 0.441 183s -0.230 -0.827 0.366 183s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 2, 4, 1] 183s -0.104 0.294 0.848 183s 0.414 -0.876 0.249 183s 183s x1: [3, 3, 4, 1] 183s 0.730 0.542 -0.204 183s 0.155 -0.794 -0.930 183s 0.646 0.136 -0.820 183s 183s y: [2, 3, 4, 1] 183s -0.089 -0.224 183s -1.039 0.527 183s -0.723 -0.056 183s testing: mul_mat, [3, 3, 4, 4] = [2, 3, 4, 4] * [2, 3, 4, 4] 183s x0: [2, 3, 4, 4] 183s -0.464 -0.889 183s 0.756 -0.787 183s -0.659 0.469 183s 183s x1: [2, 3, 4, 4] 183s 0.045 -0.537 183s -0.446 -0.910 183s 0.713 -0.168 183s 183s y: [3, 3, 4, 4] 183s 0.457 0.457 -0.282 183s 1.016 0.379 -0.133 183s -0.181 0.671 -0.549 183s test-mul-mat0: iter:439/500 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.662 0.068 -0.623 0.193 183s 183s x1: [4, 1, 1, 1] 183s -0.613 -0.858 0.200 0.018 183s 183s y: [1, 1, 1, 1] 183s 0.226 183s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 1, 2, 1] 183s -0.933 -0.768 0.420 -0.890 183s 183s x1: [4, 1, 2, 1] 183s 0.034 0.329 -0.606 -0.340 183s 183s y: [1, 1, 2, 1] 183s -0.236 183s testing: mul_mat, [1, 1, 2, 4] = [4, 1, 2, 4] * [4, 1, 2, 4] 183s x0: [4, 1, 2, 4] 183s -0.137 -0.203 0.775 0.930 183s 183s x1: [4, 1, 2, 4] 183s 0.215 0.257 -0.319 -0.872 183s 183s y: [1, 1, 2, 4] 183s -1.141 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.193 -0.569 -0.718 0.978 183s 183s x1: [4, 1, 1, 1] 183s 0.457 0.944 -0.209 -0.476 183s 183s y: [1, 1, 1, 1] 183s -0.940 183s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.622 183s 0.403 183s 0.349 183s 183s x1: [1, 4, 2, 1] 183s 0.555 183s -0.624 183s 0.336 183s 0.225 183s 183s y: [3, 4, 2, 1] 183s -0.345 0.224 0.194 183s 0.388 -0.252 -0.218 183s -0.209 0.136 0.118 183s -0.140 0.091 0.079 183s testing: mul_mat, [4, 1, 2, 4] = [3, 4, 2, 4] * [3, 1, 2, 4] 183s x0: [3, 4, 2, 4] 183s 0.333 0.074 -0.322 183s 0.134 -0.791 -0.801 183s 0.084 0.306 0.597 183s -0.589 0.005 0.316 183s 183s x1: [3, 1, 2, 4] 183s -0.105 0.931 -0.531 183s 183s y: [4, 1, 2, 4] 183s 0.205 -0.325 -0.041 -0.101 183s test-mul-mat0: iter:440/500 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.934 -0.863 0.299 -0.884 183s -0.180 -0.360 0.209 -0.336 183s 183s x1: [4, 4, 1, 1] 183s 0.728 -0.661 0.008 0.346 183s -0.978 0.478 -0.016 -0.327 183s -0.840 -0.138 -0.360 -0.098 183s -0.803 0.001 -0.120 -0.232 183s 183s y: [2, 4, 1, 1] 183s 0.947 -0.008 183s -1.042 0.111 183s -0.686 0.159 183s -0.583 0.197 183s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 1, 4, 1] 183s -0.409 -0.812 0.760 -0.223 183s 183s x1: [4, 2, 4, 1] 183s -0.332 -0.920 0.930 0.155 183s 0.808 -0.815 -0.117 -0.852 183s 183s y: [1, 2, 4, 1] 183s 1.555 183s 0.433 183s testing: mul_mat, [4, 1, 4, 4] = [4, 4, 4, 4] * [4, 1, 4, 4] 183s x0: [4, 4, 4, 4] 183s 0.732 0.453 0.330 0.667 183s -0.386 -0.067 -0.322 0.177 183s 0.435 0.423 -0.823 -0.210 183s -0.504 0.799 -0.244 -0.328 183s 183s x1: [4, 1, 4, 4] 183s -0.140 -0.112 0.818 0.900 183s 183s y: [4, 1, 4, 4] 183s 0.718 -0.042 -0.971 -0.514 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.519 0.834 -0.642 0.414 183s 0.935 0.834 -0.859 0.896 183s 183s x1: [4, 4, 1, 1] 183s -0.688 -0.333 -0.150 -0.741 183s -0.566 0.692 0.645 0.585 183s 0.114 0.724 0.612 -0.074 183s 0.297 0.584 0.021 0.599 183s 183s y: [2, 4, 1, 1] 183s -0.846 -1.457 183s 0.111 0.018 183s 0.239 0.118 183s 0.876 1.284 183s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 3, 4, 1] 183s -0.938 -0.608 183s -0.690 0.981 183s 0.290 -0.812 183s 183s x1: [2, 4, 4, 1] 183s -0.095 -0.640 183s 0.680 0.359 183s -0.769 -0.670 183s 0.671 -0.102 183s 183s y: [3, 4, 4, 1] 183s 0.478 -0.562 0.492 183s -0.856 -0.117 -0.095 183s 1.129 -0.127 0.322 183s -0.567 -0.563 0.277 183s testing: mul_mat, [3, 2, 4, 4] = [3, 3, 4, 4] * [3, 2, 4, 4] 183s x0: [3, 3, 4, 4] 183s 0.053 0.185 -0.107 183s 0.701 -0.469 -0.356 183s 0.048 0.364 0.263 183s 183s x1: [3, 2, 4, 4] 183s -0.992 -0.813 0.098 183s -0.293 -0.430 -0.951 183s 183s y: [3, 2, 4, 4] 183s -0.214 -0.349 -0.318 183s 0.007 0.335 -0.420 183s test-mul-mat0: iter:441/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.206 0.292 183s -0.755 0.201 183s -0.947 0.805 183s 183s x1: [2, 1, 1, 1] 183s -0.806 -0.308 183s 183s y: [3, 1, 1, 1] 183s 0.076 0.547 0.516 183s testing: mul_mat, [1, 3, 4, 1] = [2, 1, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 1, 4, 1] 183s 0.375 -0.738 183s 183s x1: [2, 3, 4, 1] 183s -0.528 -0.251 183s 0.659 -0.565 183s 0.371 -0.113 183s 183s y: [1, 3, 4, 1] 183s -0.012 183s 0.664 183s 0.223 183s testing: mul_mat, [2, 1, 4, 1] = [2, 2, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 2, 4, 1] 183s -0.944 0.749 183s 0.025 -0.771 183s 183s x1: [2, 1, 4, 1] 183s -0.043 0.483 183s 183s y: [2, 1, 4, 1] 183s 0.402 -0.374 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.590 0.819 183s 183s x1: [2, 2, 1, 1] 183s 0.563 -0.410 183s -0.473 -0.138 183s 183s y: [1, 2, 1, 1] 183s -0.003 183s -0.392 183s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 3, 4, 1] 183s 0.096 183s -0.105 183s -0.196 183s 183s x1: [1, 2, 4, 1] 183s 0.121 183s -0.975 183s 183s y: [3, 2, 4, 1] 183s 0.012 -0.013 -0.024 183s -0.094 0.103 0.191 183s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.827 -0.973 0.596 183s -0.842 -0.208 0.901 183s -0.783 -0.384 0.488 183s 0.520 -0.078 0.105 183s 183s x1: [3, 1, 4, 1] 183s 0.790 0.279 -0.346 183s 183s y: [4, 1, 4, 1] 183s -1.131 -1.035 -0.894 0.353 183s test-mul-mat0: iter:442/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.380 -0.759 -0.581 0.417 183s 0.564 -0.912 0.568 -0.518 183s -0.721 0.096 0.437 -0.218 183s 0.280 0.083 -0.050 0.989 183s 183s x1: [4, 4, 1, 1] 183s -0.832 -0.291 0.519 -0.953 183s -0.260 -0.837 0.961 0.399 183s 0.605 0.573 -0.769 0.204 183s 0.402 -0.417 -0.169 -0.976 183s 183s y: [4, 4, 1, 1] 183s -0.795 0.584 1.007 -1.225 183s 0.144 0.955 0.440 0.204 183s 0.327 -0.724 -0.762 0.458 183s 0.160 1.017 -0.191 -0.879 183s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 4, 3, 1] 183s 0.474 0.350 0.698 -0.755 183s 0.595 -0.023 -0.888 0.277 183s 0.038 0.259 -0.958 -0.951 183s -0.606 0.410 -0.944 -0.741 183s 183s x1: [4, 4, 3, 1] 183s -0.398 -0.003 -0.271 -0.234 183s -0.042 -0.872 -0.629 -0.469 183s -0.641 0.575 0.934 -0.058 183s -0.594 0.957 -0.507 0.786 183s 183s y: [4, 4, 3, 1] 183s -0.202 -0.060 0.466 0.669 183s -0.410 0.424 0.821 0.609 183s 0.593 -1.239 -0.715 -0.214 183s -0.894 0.293 -0.037 0.649 183s testing: mul_mat, [4, 4, 3, 4] = [4, 4, 3, 4] * [4, 4, 3, 4] 183s x0: [4, 4, 3, 4] 183s 0.649 0.507 0.782 0.914 183s -0.463 0.968 -0.388 0.608 183s 0.256 -0.186 0.626 -0.261 183s 0.221 -0.004 -0.897 0.613 183s 183s x1: [4, 4, 3, 4] 183s 0.223 -0.124 -0.812 0.742 183s 0.479 -0.685 0.691 0.376 183s -0.073 0.028 0.059 0.809 183s 0.951 0.441 -0.836 -0.776 183s 183s y: [4, 4, 3, 4] 183s 0.125 0.542 -0.622 1.233 183s 0.847 -0.923 0.585 -0.281 183s 0.752 0.529 -0.198 0.426 183s -0.522 -0.161 -0.159 0.482 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.357 -0.255 -0.352 0.193 183s 0.222 -0.984 0.573 -0.041 183s -0.969 0.659 -0.460 0.491 183s 183s x1: [4, 4, 1, 1] 183s 0.973 -0.715 -0.769 -0.317 183s -0.775 -0.066 -0.192 0.760 183s 0.359 0.066 -0.761 -0.234 183s -0.721 0.680 0.434 0.928 183s 183s y: [3, 4, 1, 1] 183s 0.740 0.492 -1.215 183s -0.045 -0.248 1.168 183s 0.334 -0.413 -0.069 183s -0.405 -0.619 1.402 183s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 3, 3, 1] 183s -0.559 -0.305 0.188 183s -0.525 -0.975 -0.659 183s 0.998 -0.980 -0.678 183s 183s x1: [3, 4, 3, 1] 183s 0.984 0.026 0.464 183s -0.731 0.257 -0.853 183s -0.505 -0.809 -0.044 183s -0.745 0.550 -0.978 183s 183s y: [3, 4, 3, 1] 183s -0.471 -0.848 0.642 183s 0.170 0.695 -0.403 183s 0.521 1.083 0.319 183s 0.065 0.499 -0.620 183s testing: mul_mat, [1, 3, 3, 4] = [3, 1, 3, 4] * [3, 3, 3, 4] 183s x0: [3, 1, 3, 4] 183s -0.328 -0.791 0.388 183s 183s x1: [3, 3, 3, 4] 183s -0.515 -0.137 0.827 183s 0.528 0.303 -0.698 183s 0.526 0.998 -0.673 183s 183s y: [1, 3, 3, 4] 183s 0.598 183s -0.684 183s -1.224 183s test-mul-mat0: iter:443/500 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.554 -0.132 -0.233 183s 0.534 -0.865 0.051 183s 183s x1: [3, 1, 1, 1] 183s -0.710 -0.213 -0.652 183s 183s y: [2, 1, 1, 1] 183s -0.213 -0.228 183s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.457 0.012 0.701 183s -0.756 0.360 -0.167 183s 183s x1: [3, 2, 3, 1] 183s 0.579 -0.055 0.775 183s 0.424 0.015 -0.780 183s 183s y: [2, 2, 3, 1] 183s 0.808 -0.587 183s -0.353 -0.184 183s testing: mul_mat, [3, 2, 3, 3] = [3, 3, 3, 3] * [3, 2, 3, 3] 183s x0: [3, 3, 3, 3] 183s -0.072 -0.242 0.660 183s -0.851 0.148 -0.281 183s -0.985 -0.258 0.755 183s 183s x1: [3, 2, 3, 3] 183s 0.204 0.977 0.669 183s 0.533 -0.495 -0.730 183s 183s y: [3, 2, 3, 3] 183s 0.191 -0.217 0.052 183s -0.400 -0.322 -0.948 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.552 -0.601 -0.078 183s 0.521 -0.160 -0.918 183s -0.014 -0.850 -0.579 183s 183s x1: [3, 3, 1, 1] 183s 0.430 -0.840 -0.106 183s -0.335 -0.940 -0.160 183s 0.781 0.921 0.408 183s 183s y: [3, 3, 1, 1] 183s 0.276 0.457 0.769 183s 0.763 0.123 0.897 183s -1.017 -0.115 -1.030 183s testing: mul_mat, [4, 3, 3, 1] = [3, 4, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 4, 3, 1] 183s 0.291 -0.095 0.050 183s 0.902 -0.241 -0.198 183s 0.410 0.827 -0.479 183s -0.951 0.803 0.294 183s 183s x1: [3, 3, 3, 1] 183s -0.242 0.003 0.350 183s 0.778 -0.102 -0.709 183s -0.025 -0.282 -0.756 183s 183s y: [4, 3, 3, 1] 183s -0.053 -0.288 -0.264 0.335 183s 0.200 0.867 0.574 -1.031 183s -0.018 0.195 0.119 -0.425 183s testing: mul_mat, [1, 3, 3, 3] = [4, 1, 3, 3] * [4, 3, 3, 3] 183s x0: [4, 1, 3, 3] 183s -0.109 0.644 -0.743 0.649 183s 183s x1: [4, 3, 3, 3] 183s -0.693 -0.928 0.744 0.357 183s -0.126 -0.735 -0.350 -0.260 183s -0.833 -0.738 0.100 0.611 183s 183s y: [1, 3, 3, 3] 183s -0.844 183s -0.369 183s -0.063 183s test-mul-mat0: iter:444/500 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.290 0.264 0.439 183s 183s x1: [3, 3, 1, 1] 183s 0.286 0.224 -0.879 183s -0.419 -0.543 0.485 183s 0.721 -0.853 -0.192 183s 183s y: [1, 3, 1, 1] 183s -0.244 183s -0.052 183s -0.101 183s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.073 0.578 0.671 183s -0.558 -0.467 -0.443 183s 183s x1: [3, 1, 3, 1] 183s 0.675 0.924 0.573 183s 183s y: [2, 1, 3, 1] 183s 0.869 -1.061 183s testing: mul_mat, [3, 2, 3, 3] = [3, 3, 3, 3] * [3, 2, 3, 3] 183s x0: [3, 3, 3, 3] 183s 0.209 0.252 0.737 183s 0.680 -0.786 -0.408 183s 0.845 -0.765 -0.070 183s 183s x1: [3, 2, 3, 3] 183s 0.059 0.482 0.133 183s -0.265 0.406 -0.294 183s 183s y: [3, 2, 3, 3] 183s 0.232 -0.393 -0.328 183s -0.170 -0.380 -0.514 183s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.153 0.231 0.700 183s 183s x1: [3, 3, 1, 1] 183s -0.027 0.580 0.304 183s 0.115 -0.008 -0.908 183s 0.641 0.395 -0.632 183s 183s y: [1, 3, 1, 1] 183s 0.351 183s -0.654 183s -0.449 183s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 3, 3, 1] 183s -0.061 183s 0.965 183s -0.375 183s 183s x1: [1, 3, 3, 1] 183s -0.398 183s -0.582 183s -0.877 183s 183s y: [3, 3, 3, 1] 183s 0.024 -0.384 0.149 183s 0.036 -0.561 0.218 183s 0.054 -0.846 0.329 183s testing: mul_mat, [1, 1, 3, 3] = [3, 1, 3, 3] * [3, 1, 3, 3] 183s x0: [3, 1, 3, 3] 183s -0.660 -0.464 0.932 183s 183s x1: [3, 1, 3, 3] 183s 0.362 0.405 -0.054 183s 183s y: [1, 1, 3, 3] 183s -0.477 183s test-mul-mat0: iter:445/500 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.497 183s 0.310 183s 183s x1: [1, 2, 1, 1] 183s -0.199 183s 0.369 183s 183s y: [2, 2, 1, 1] 183s 0.099 -0.062 183s -0.184 0.115 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.868 183s -0.635 183s 183s x1: [1, 2, 1, 1] 183s -0.597 183s -0.440 183s 183s y: [2, 2, 1, 1] 183s -0.518 0.379 183s -0.383 0.280 183s testing: mul_mat, [3, 2, 1, 2] = [1, 3, 1, 2] * [1, 2, 1, 2] 183s x0: [1, 3, 1, 2] 183s -0.634 183s -0.631 183s 0.600 183s 183s x1: [1, 2, 1, 2] 183s -0.947 183s 0.353 183s 183s y: [3, 2, 1, 2] 183s 0.601 0.598 -0.569 183s -0.224 -0.223 0.212 183s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.318 183s 0.365 183s 0.406 183s 0.143 183s 183s x1: [1, 3, 1, 1] 183s 0.666 183s -0.222 183s -0.034 183s 183s y: [4, 3, 1, 1] 183s -0.212 0.243 0.271 0.095 183s 0.071 -0.081 -0.090 -0.032 183s 0.011 -0.012 -0.014 -0.005 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.553 -0.071 0.627 0.372 183s -0.122 0.007 -0.203 0.680 183s 183s x1: [4, 1, 1, 1] 183s -0.802 -0.000 0.944 0.568 183s 183s y: [2, 1, 1, 1] 183s 1.246 0.292 183s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 183s x0: [2, 4, 1, 2] 183s 0.425 0.437 183s 0.869 0.050 183s -0.475 0.817 183s 0.369 -0.685 183s 183s x1: [2, 4, 1, 2] 183s -0.850 -0.862 183s -0.867 -0.294 183s -0.496 -0.498 183s -0.694 -0.461 183s 183s y: [4, 4, 1, 2] 183s -0.738 -0.782 -0.300 0.277 183s -0.497 -0.768 0.171 -0.118 183s -0.428 -0.456 -0.171 0.158 183s -0.497 -0.626 -0.047 0.060 183s test-mul-mat0: iter:446/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.276 0.580 183s -0.319 0.582 183s 0.698 -0.036 183s 183s x1: [2, 1, 1, 1] 183s 0.937 0.376 183s 183s y: [3, 1, 1, 1] 183s 0.476 -0.080 0.641 183s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 4, 2, 1] 183s 0.553 -0.407 183s -0.564 -0.522 183s -0.074 -0.809 183s -0.891 0.763 183s 183s x1: [2, 3, 2, 1] 183s 0.252 0.224 183s -0.611 0.121 183s 0.748 0.759 183s 183s y: [4, 3, 2, 1] 183s 0.048 -0.259 -0.200 -0.054 183s -0.387 0.281 -0.053 0.636 183s 0.105 -0.818 -0.669 -0.088 183s testing: mul_mat, [1, 4, 2, 2] = [2, 1, 2, 2] * [2, 4, 2, 2] 183s x0: [2, 1, 2, 2] 183s 0.531 -0.509 183s 183s x1: [2, 4, 2, 2] 183s -0.583 0.174 183s -0.844 0.640 183s 0.564 0.277 183s 0.389 0.322 183s 183s y: [1, 4, 2, 2] 183s -0.399 183s -0.774 183s 0.158 183s 0.042 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.410 -0.170 183s -0.128 -0.458 183s 0.547 0.295 183s -0.520 -0.022 183s 183s x1: [2, 1, 1, 1] 183s 0.294 0.350 183s 183s y: [4, 1, 1, 1] 183s -0.180 -0.198 0.264 -0.160 183s testing: mul_mat, [1, 2, 2, 1] = [4, 1, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 1, 2, 1] 183s 0.015 0.626 0.313 -0.692 183s 183s x1: [4, 2, 2, 1] 183s 0.646 0.873 0.575 -0.204 183s 0.590 0.701 0.268 -0.976 183s 183s y: [1, 2, 2, 1] 183s 0.877 183s 1.205 183s testing: mul_mat, [3, 4, 2, 2] = [1, 3, 2, 2] * [1, 4, 2, 2] 183s x0: [1, 3, 2, 2] 183s -0.486 183s -0.040 183s -0.598 183s 183s x1: [1, 4, 2, 2] 183s 0.891 183s 0.378 183s -0.322 183s -0.567 183s 183s y: [3, 4, 2, 2] 183s -0.433 -0.036 -0.533 183s -0.184 -0.015 -0.226 183s 0.157 0.013 0.193 183s 0.276 0.023 0.339 183s test-mul-mat0: iter:447/500 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.952 0.957 -0.419 -0.912 183s 183s x1: [4, 1, 1, 1] 183s 0.070 0.628 0.406 -0.274 183s 183s y: [1, 1, 1, 1] 183s 0.747 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.284 -0.630 -0.532 0.686 183s 0.503 -0.590 0.334 0.650 183s 183s x1: [4, 1, 1, 1] 183s -0.112 -0.277 -0.752 0.856 183s 183s y: [2, 1, 1, 1] 183s 1.129 0.413 183s testing: mul_mat, [2, 2, 1, 3] = [4, 2, 1, 3] * [4, 2, 1, 3] 183s x0: [4, 2, 1, 3] 183s 0.409 -0.142 0.282 0.819 183s -0.809 -0.067 0.797 -0.514 183s 183s x1: [4, 2, 1, 3] 183s 0.978 -0.706 -0.894 -0.479 183s 0.598 0.461 -0.087 -0.207 183s 183s y: [2, 2, 1, 3] 183s -0.144 -1.209 183s -0.015 -0.478 183s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.388 0.545 -0.416 0.084 183s 0.735 0.532 0.851 0.082 183s 0.596 -0.918 -0.834 0.044 183s 183s x1: [4, 2, 1, 1] 183s -0.860 0.566 0.375 0.550 183s -0.493 -0.453 0.330 0.916 183s 183s y: [3, 2, 1, 1] 183s 0.533 0.033 -1.321 183s -0.116 -0.248 -0.113 183s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.505 0.825 -0.563 183s 0.976 0.892 0.627 183s 0.325 -0.270 -0.674 183s 183s x1: [3, 4, 1, 1] 183s 0.778 -0.704 -0.658 183s -0.928 -0.782 0.768 183s 0.918 -0.195 0.903 183s 0.404 -0.055 0.469 183s 183s y: [3, 4, 1, 1] 183s -0.603 -0.281 0.886 183s -0.609 -1.121 -0.609 183s -1.133 1.289 -0.257 183s -0.513 0.640 -0.170 183s testing: mul_mat, [3, 3, 1, 3] = [3, 3, 1, 3] * [3, 3, 1, 3] 183s x0: [3, 3, 1, 3] 183s 0.114 0.095 0.661 183s -0.403 0.757 -0.233 183s 0.983 -0.609 -0.034 183s 183s x1: [3, 3, 1, 3] 183s 0.982 0.160 -0.592 183s -0.434 0.010 -0.425 183s 0.650 -0.908 0.618 183s 183s y: [3, 3, 1, 3] 183s -0.264 -0.137 0.888 183s -0.330 0.282 -0.419 183s 0.396 -1.093 1.170 183s test-mul-mat0: iter:448/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.041 0.021 183s -0.824 -0.993 183s -0.667 0.035 183s 183s x1: [2, 2, 1, 1] 183s -0.735 0.652 183s -0.351 0.360 183s 183s y: [3, 2, 1, 1] 183s -0.016 -0.041 0.514 183s -0.007 -0.069 0.247 183s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 3, 2, 1] 183s 0.535 0.500 183s -0.591 -0.456 183s -0.349 0.464 183s 183s x1: [2, 3, 2, 1] 183s -0.409 0.072 183s -0.717 0.136 183s 0.312 -0.862 183s 183s y: [3, 3, 2, 1] 183s -0.183 0.209 0.176 183s -0.315 0.361 0.313 183s -0.264 0.209 -0.509 183s testing: mul_mat, [2, 3, 2, 4] = [2, 2, 2, 4] * [2, 3, 2, 4] 183s x0: [2, 2, 2, 4] 183s -0.378 -0.035 183s -0.962 -0.282 183s 183s x1: [2, 3, 2, 4] 183s -0.099 0.297 183s -0.300 -0.093 183s 0.630 0.735 183s 183s y: [2, 3, 2, 4] 183s 0.027 0.012 183s 0.117 0.315 183s -0.264 -0.813 183s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.226 -0.234 183s 183s x1: [2, 2, 1, 1] 183s -0.924 -0.494 183s -0.052 0.557 183s 183s y: [1, 2, 1, 1] 183s -0.093 183s -0.142 183s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 1, 2, 1] 183s -0.155 183s 183s x1: [1, 2, 2, 1] 183s 0.707 183s -0.821 183s 183s y: [1, 2, 2, 1] 183s -0.110 183s 0.127 183s testing: mul_mat, [3, 1, 2, 4] = [1, 3, 2, 4] * [1, 1, 2, 4] 183s x0: [1, 3, 2, 4] 183s -0.443 183s -0.134 183s 0.743 183s 183s x1: [1, 1, 2, 4] 183s -0.615 183s 183s y: [3, 1, 2, 4] 183s 0.272 0.083 -0.457 183s test-mul-mat0: iter:449/500 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.759 -0.115 0.809 0.095 183s -0.575 -0.488 -0.976 -0.530 183s -0.837 -0.490 0.795 0.040 183s 0.763 -0.173 -0.903 -0.423 183s 183s x1: [4, 1, 1, 1] 183s 0.256 -0.530 -0.502 0.813 183s 183s y: [4, 1, 1, 1] 183s -0.462 0.171 -0.321 0.397 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s 0.003 -0.946 0.187 0.191 183s 183s x1: [4, 4, 3, 1] 183s 0.920 0.486 -0.763 -0.347 183s 0.260 0.419 0.459 0.106 183s 0.696 -0.669 -0.639 -0.835 183s -0.171 -0.826 0.501 0.070 183s 183s y: [1, 4, 3, 1] 183s -0.667 183s -0.290 183s 0.356 183s 0.888 183s testing: mul_mat, [4, 1, 3, 3] = [4, 4, 3, 3] * [4, 1, 3, 3] 183s x0: [4, 4, 3, 3] 183s 0.982 0.721 -0.903 0.735 183s 0.244 -0.798 -0.956 -0.466 183s 0.817 -0.590 -0.337 -0.998 183s -0.857 0.793 -0.760 0.062 183s 183s x1: [4, 1, 3, 3] 183s 0.391 -0.172 0.927 -0.333 183s 183s y: [4, 1, 3, 3] 183s -0.822 -0.498 0.441 -1.196 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.095 -0.138 0.984 0.453 183s -0.392 0.969 -0.272 0.993 183s 183s x1: [4, 4, 1, 1] 183s -0.433 0.113 -0.586 -0.217 183s -0.828 -0.304 -0.395 0.033 183s 0.128 0.198 -0.331 0.873 183s -0.435 -0.942 -0.725 -0.201 183s 183s y: [2, 4, 1, 1] 183s -0.649 0.223 183s -0.253 0.170 183s 0.031 1.098 183s -0.633 -0.745 183s testing: mul_mat, [1, 4, 3, 1] = [2, 1, 3, 1] * [2, 4, 3, 1] 183s x0: [2, 1, 3, 1] 183s 0.265 0.719 183s 183s x1: [2, 4, 3, 1] 183s 0.487 -0.820 183s -0.107 0.400 183s -0.027 -0.541 183s 0.968 -0.915 183s 183s y: [1, 4, 3, 1] 183s -0.460 183s 0.260 183s -0.396 183s -0.401 183s testing: mul_mat, [3, 2, 3, 3] = [1, 3, 3, 3] * [1, 2, 3, 3] 183s x0: [1, 3, 3, 3] 183s 0.932 183s -0.595 183s -0.453 183s 183s x1: [1, 2, 3, 3] 183s 0.480 183s 0.352 183s 183s y: [3, 2, 3, 3] 183s 0.447 -0.285 -0.217 183s 0.328 -0.209 -0.159 183s test-mul-mat0: iter:450/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.593 -0.045 -0.361 0.655 183s 0.356 -0.564 -0.988 -0.289 183s -0.811 0.361 0.577 0.857 183s -0.729 0.071 0.457 0.586 183s 183s x1: [4, 4, 1, 1] 183s 0.600 0.315 -0.959 0.935 183s 0.470 0.114 0.536 0.344 183s -0.304 0.129 -0.611 -0.220 183s -0.798 -0.214 -0.741 -0.846 183s 183s y: [4, 4, 1, 1] 183s 1.301 0.713 -0.124 -0.306 183s 0.305 -0.526 0.264 0.112 183s -0.110 0.486 -0.248 -0.177 183s -0.751 0.813 -0.581 -0.267 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.722 0.208 -0.276 -0.842 183s 0.220 0.436 -0.652 -0.419 183s 0.012 -0.795 -0.149 -0.917 183s 183s x1: [4, 4, 1, 1] 183s 0.112 0.393 0.056 -0.775 183s -0.071 -0.601 -0.079 -0.943 183s -0.211 0.701 -0.741 0.575 183s 0.960 -0.587 0.083 0.554 183s 183s y: [3, 4, 1, 1] 183s 0.799 0.484 0.391 183s 0.638 0.169 1.353 183s -0.286 0.501 -0.977 183s 0.082 -0.331 -0.042 183s testing: mul_mat, [1, 3, 1, 3] = [4, 1, 1, 3] * [4, 3, 1, 3] 183s x0: [4, 1, 1, 3] 183s -0.527 0.238 -0.707 -0.486 183s 183s x1: [4, 3, 1, 3] 183s 0.662 -0.563 0.195 0.055 183s 0.493 0.420 0.984 0.892 183s -0.659 -0.959 -0.319 -0.958 183s 183s y: [1, 3, 1, 3] 183s -0.647 183s -1.289 183s 0.810 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.356 -0.343 0.361 0.210 183s 183s x1: [4, 1, 1, 1] 183s 0.609 -0.426 -0.341 -0.584 183s 183s y: [1, 1, 1, 1] 183s -0.317 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.520 183s 0.779 183s 0.057 183s 183s x1: [1, 4, 1, 1] 183s 0.911 183s 0.133 183s -0.374 183s 0.248 183s 183s y: [3, 4, 1, 1] 183s -0.474 0.710 0.052 183s -0.069 0.104 0.008 183s 0.194 -0.291 -0.021 183s -0.129 0.193 0.014 183s testing: mul_mat, [1, 1, 1, 3] = [3, 1, 1, 3] * [3, 1, 1, 3] 183s x0: [3, 1, 1, 3] 183s -0.303 -0.615 -0.330 183s 183s x1: [3, 1, 1, 3] 183s 0.826 0.731 0.530 183s 183s y: [1, 1, 1, 3] 183s -0.875 183s test-mul-mat0: iter:451/500 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.269 -0.918 -0.880 183s 183s x1: [3, 4, 1, 1] 183s -0.022 0.533 0.730 183s -0.774 -0.899 -0.790 183s -0.994 0.158 -0.964 183s 0.737 -0.313 0.100 183s 183s y: [1, 4, 1, 1] 183s -1.126 183s 1.729 183s 0.972 183s 0.001 183s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 183s x0: [3, 4, 2, 1] 183s 0.154 -0.342 -0.624 183s -0.807 -0.448 0.443 183s -0.407 0.656 -0.578 183s -0.874 0.386 -0.352 183s 183s x1: [3, 1, 2, 1] 183s 0.588 0.239 -0.183 183s 183s y: [4, 1, 2, 1] 183s 0.123 -0.663 0.023 -0.357 183s testing: mul_mat, [1, 4, 2, 4] = [3, 1, 2, 4] * [3, 4, 2, 4] 183s x0: [3, 1, 2, 4] 183s 0.302 0.413 0.640 183s 183s x1: [3, 4, 2, 4] 183s -0.609 -0.947 0.088 183s -0.636 0.963 -0.504 183s -0.669 0.117 0.154 183s -0.293 0.309 0.706 183s 183s y: [1, 4, 2, 4] 183s -0.519 183s -0.118 183s -0.056 183s 0.491 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.837 -0.269 0.047 183s 0.633 0.139 -0.629 183s 183s x1: [3, 1, 1, 1] 183s 0.219 0.444 0.750 183s 183s y: [2, 1, 1, 1] 183s 0.098 -0.272 183s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.944 0.624 183s 183s x1: [2, 3, 2, 1] 183s -0.993 0.412 183s -0.893 -0.597 183s 0.860 -0.603 183s 183s y: [1, 3, 2, 1] 183s -0.681 183s -1.216 183s 0.436 183s testing: mul_mat, [3, 2, 2, 4] = [1, 3, 2, 4] * [1, 2, 2, 4] 183s x0: [1, 3, 2, 4] 183s -0.616 183s -0.557 183s 0.731 183s 183s x1: [1, 2, 2, 4] 183s 0.614 183s -0.972 183s 183s y: [3, 2, 2, 4] 183s -0.378 -0.342 0.449 183s 0.599 0.542 -0.711 183s test-mul-mat0: iter:452/500 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.919 0.762 -0.718 183s 183s x1: [3, 2, 1, 1] 183s 0.458 -0.685 -0.969 183s 0.843 -0.242 0.763 183s 183s y: [1, 2, 1, 1] 183s -0.247 183s -1.506 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.324 0.153 -0.787 183s 183s x1: [3, 1, 1, 1] 183s 0.118 0.731 0.942 183s 183s y: [1, 1, 1, 1] 183s -0.591 183s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.123 -0.450 -0.084 183s 183s x1: [3, 1, 1, 1] 183s -0.469 -0.608 0.883 183s 183s y: [1, 1, 1, 1] 183s 0.257 183s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 2, 1, 1] 183s 0.530 0.656 0.526 183s 0.211 -0.011 0.687 183s 183s x1: [3, 1, 1, 1] 183s 0.490 -0.947 0.478 183s 183s y: [2, 1, 1, 1] 183s -0.110 0.443 183s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.483 0.581 183s -0.057 -0.733 183s 0.477 -0.370 183s 183s x1: [2, 3, 1, 1] 183s -0.169 -0.716 183s 0.450 0.254 183s -0.635 0.212 183s 183s y: [3, 3, 1, 1] 183s -0.498 0.534 0.184 183s 0.365 -0.212 0.121 183s -0.184 -0.119 -0.381 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.807 0.389 -0.719 183s 183s x1: [3, 2, 1, 1] 183s 0.795 -0.202 0.022 183s 0.678 0.277 0.900 183s 183s y: [1, 2, 1, 1] 183s -0.736 183s -1.086 183s test-mul-mat0: iter:453/500 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.691 183s 0.631 183s 183s x1: [1, 2, 1, 1] 183s -0.996 183s -0.200 183s 183s y: [2, 2, 1, 1] 183s -0.688 -0.629 183s -0.138 -0.126 183s testing: mul_mat, [1, 2, 3, 1] = [1, 1, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 1, 3, 1] 183s 0.295 183s 183s x1: [1, 2, 3, 1] 183s -0.308 183s 0.141 183s 183s y: [1, 2, 3, 1] 183s -0.091 183s 0.042 183s testing: mul_mat, [4, 1, 3, 1] = [1, 4, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 4, 3, 1] 183s 0.550 183s 0.594 183s 0.637 183s 0.449 183s 183s x1: [1, 1, 3, 1] 183s 0.562 183s 183s y: [4, 1, 3, 1] 183s 0.309 0.334 0.358 0.253 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.482 183s -0.978 183s 183s x1: [1, 4, 1, 1] 183s 0.190 183s -0.286 183s -0.210 183s -0.119 183s 183s y: [2, 4, 1, 1] 183s 0.092 -0.186 183s -0.138 0.280 183s -0.101 0.206 183s -0.057 0.116 183s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 2, 3, 1] 183s 0.201 -0.925 183s 0.231 0.117 183s 183s x1: [2, 1, 3, 1] 183s 0.230 -0.461 183s 183s y: [2, 1, 3, 1] 183s 0.473 -0.001 183s testing: mul_mat, [4, 2, 3, 1] = [2, 4, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 4, 3, 1] 183s 0.751 0.236 183s 0.567 -0.963 183s -0.510 -0.969 183s -0.170 -0.532 183s 183s x1: [2, 2, 3, 1] 183s -0.363 -0.441 183s 0.100 -0.449 183s 183s y: [4, 2, 3, 1] 183s -0.377 0.219 0.613 0.296 183s -0.031 0.489 0.384 0.222 183s test-mul-mat0: iter:454/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.459 -0.953 183s 0.786 0.372 183s 0.284 0.823 183s 183s x1: [2, 1, 1, 1] 183s -0.210 -0.520 183s 183s y: [3, 1, 1, 1] 183s 0.591 -0.358 -0.487 183s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.108 -0.031 183s 0.805 -0.520 183s -0.747 0.628 183s 0.883 -0.996 183s 183s x1: [2, 3, 4, 1] 183s 0.403 0.751 183s 0.936 -0.448 183s 0.811 0.674 183s 183s y: [4, 3, 4, 1] 183s 0.020 -0.066 0.171 -0.393 183s 0.115 0.987 -0.981 1.273 183s 0.067 0.302 -0.183 0.045 183s testing: mul_mat, [1, 4, 4, 4] = [2, 1, 4, 4] * [2, 4, 4, 4] 183s x0: [2, 1, 4, 4] 183s -0.403 -0.690 183s 183s x1: [2, 4, 4, 4] 183s 0.463 0.721 183s -0.158 0.715 183s 0.350 -0.276 183s 0.719 -0.086 183s 183s y: [1, 4, 4, 4] 183s -0.684 183s -0.429 183s 0.049 183s -0.231 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.832 -0.181 183s -0.680 -0.048 183s 0.860 0.931 183s 183s x1: [2, 1, 1, 1] 183s -0.391 0.547 183s 183s y: [3, 1, 1, 1] 183s -0.424 0.239 0.173 183s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.024 0.198 -0.918 183s 183s x1: [3, 2, 4, 1] 183s -0.901 0.209 0.961 183s 0.401 0.522 -0.663 183s 183s y: [1, 2, 4, 1] 183s -0.862 183s 0.721 183s testing: mul_mat, [4, 3, 4, 4] = [1, 4, 4, 4] * [1, 3, 4, 4] 183s x0: [1, 4, 4, 4] 183s -0.564 183s -0.711 183s -0.572 183s -0.159 183s 183s x1: [1, 3, 4, 4] 183s 0.426 183s 0.790 183s 0.319 183s 183s y: [4, 3, 4, 4] 183s -0.240 -0.303 -0.244 -0.068 183s -0.445 -0.562 -0.452 -0.126 183s -0.180 -0.227 -0.183 -0.051 183s test-mul-mat0: iter:455/500 183s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.174 0.907 -0.687 183s -0.113 0.547 0.282 183s 183s x1: [3, 2, 1, 1] 183s 0.509 0.891 0.478 183s 0.413 -0.858 -0.494 183s 183s y: [2, 2, 1, 1] 183s 0.392 0.565 183s -0.511 -0.656 183s testing: mul_mat, [4, 2, 2, 1] = [3, 4, 2, 1] * [3, 2, 2, 1] 183s x0: [3, 4, 2, 1] 183s -0.792 0.941 -0.090 183s 0.020 0.153 -0.580 183s -0.090 -0.369 0.833 183s 0.052 0.137 0.506 183s 183s x1: [3, 2, 2, 1] 183s -0.908 -0.079 0.966 183s -0.116 0.433 0.773 183s 183s y: [4, 2, 2, 1] 183s 0.559 -0.590 0.915 0.431 183s 0.430 -0.384 0.494 0.445 183s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 183s x0: [3, 2, 2, 4] 183s 0.451 -0.546 -0.514 183s -0.345 -0.105 -0.452 183s 183s x1: [3, 4, 2, 4] 183s 0.021 0.926 -0.692 183s -0.589 0.068 -0.643 183s -0.935 0.276 -0.703 183s -0.025 -0.705 0.451 183s 183s y: [2, 4, 2, 4] 183s -0.141 0.208 183s 0.028 0.487 183s -0.212 0.611 183s 0.142 -0.121 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.729 -0.258 -0.992 183s 0.454 0.788 -0.315 183s -0.978 -0.910 -0.124 183s 0.983 0.482 0.122 183s 183s x1: [3, 2, 1, 1] 183s 0.057 -0.401 0.055 183s -0.588 0.863 0.996 183s 183s y: [4, 2, 1, 1] 183s 0.007 -0.307 0.302 -0.130 183s -0.782 0.099 -0.333 -0.041 183s testing: mul_mat, [3, 3, 2, 1] = [4, 3, 2, 1] * [4, 3, 2, 1] 183s x0: [4, 3, 2, 1] 183s 0.128 0.813 0.640 -0.413 183s -0.222 0.654 0.661 -0.298 183s -0.381 0.741 0.288 0.042 183s 183s x1: [4, 3, 2, 1] 183s 0.827 -0.993 0.547 -0.053 183s 0.040 0.754 -0.933 0.483 183s 0.492 -0.461 -0.678 -0.952 183s 183s y: [3, 3, 2, 1] 183s -0.329 -0.456 -0.896 183s -0.179 -0.277 0.295 183s -0.352 -0.576 -0.764 183s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 183s x0: [3, 2, 2, 4] 183s -0.847 -0.779 0.809 183s -0.197 -0.705 -0.960 183s 183s x1: [3, 4, 2, 4] 183s 0.599 0.266 0.341 183s -0.577 -0.696 -0.362 183s 0.034 0.432 0.416 183s 0.653 0.244 0.070 183s 183s y: [2, 4, 2, 4] 183s -0.439 -0.633 183s 0.738 0.953 183s -0.029 -0.711 183s -0.687 -0.369 183s test-mul-mat0: iter:456/500 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.097 183s 0.748 183s 0.942 183s 0.477 183s 183s x1: [1, 2, 1, 1] 183s -0.514 183s -0.137 183s 183s y: [4, 2, 1, 1] 183s -0.050 -0.384 -0.484 -0.245 183s -0.013 -0.102 -0.129 -0.065 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.512 183s 183s x1: [1, 4, 1, 1] 183s 0.411 183s -0.895 183s 0.872 183s 0.227 183s 183s y: [1, 4, 1, 1] 183s -0.211 183s 0.458 183s -0.447 183s -0.116 183s testing: mul_mat, [1, 1, 1, 4] = [1, 1, 1, 4] * [1, 1, 1, 4] 183s x0: [1, 1, 1, 4] 183s -0.336 183s 183s x1: [1, 1, 1, 4] 183s 0.338 183s 183s y: [1, 1, 1, 4] 183s -0.113 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.772 183s -0.291 183s 183s x1: [1, 1, 1, 1] 183s -0.080 183s 183s y: [2, 1, 1, 1] 183s -0.061 0.023 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.868 0.006 183s 183s x1: [2, 1, 1, 1] 183s 0.150 -0.118 183s 183s y: [1, 1, 1, 1] 183s 0.130 183s testing: mul_mat, [1, 2, 1, 4] = [1, 1, 1, 4] * [1, 2, 1, 4] 183s x0: [1, 1, 1, 4] 183s 0.964 183s 183s x1: [1, 2, 1, 4] 183s 0.517 183s -0.268 183s 183s y: [1, 2, 1, 4] 183s 0.499 183s -0.259 183s test-mul-mat0: iter:457/500 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.521 -0.035 0.799 183s 0.485 -0.518 -0.469 183s -0.226 0.097 0.010 183s 183s x1: [3, 3, 1, 1] 183s 0.467 -0.740 -0.695 183s 0.934 0.825 0.225 183s -0.219 0.596 0.934 183s 183s y: [3, 3, 1, 1] 183s -0.773 0.936 -0.184 183s -0.336 -0.080 -0.129 183s 0.839 -0.854 0.117 183s testing: mul_mat, [4, 3, 4, 1] = [3, 4, 4, 1] * [3, 3, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.286 -0.220 -0.148 183s -0.284 0.979 0.681 183s -0.815 0.439 -0.186 183s 0.588 -0.278 0.173 183s 183s x1: [3, 3, 4, 1] 183s -0.994 0.188 -0.098 183s 0.342 0.152 -0.980 183s 0.749 0.604 -0.524 183s 183s y: [4, 3, 4, 1] 183s 0.257 0.399 0.911 -0.654 183s 0.014 -0.615 -0.030 -0.010 183s -0.270 0.023 -0.248 0.182 183s testing: mul_mat, [3, 4, 4, 2] = [3, 3, 4, 2] * [3, 4, 4, 2] 183s x0: [3, 3, 4, 2] 183s 0.719 0.626 -0.584 183s 0.461 -0.189 -0.826 183s 0.195 0.822 -0.045 183s 183s x1: [3, 4, 4, 2] 183s 0.136 -0.328 0.782 183s 0.596 -0.084 0.204 183s -0.109 -0.386 -0.751 183s -0.790 -0.332 -0.741 183s 183s y: [3, 4, 4, 2] 183s -0.564 -0.521 -0.278 183s 0.257 0.122 0.038 183s 0.119 0.644 -0.305 183s -0.342 0.311 -0.393 183s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.988 0.155 -0.380 183s 0.554 -0.989 -0.396 183s 0.875 -0.433 0.396 183s -0.628 0.443 0.420 183s 183s x1: [3, 3, 1, 1] 183s 0.628 0.442 0.344 183s -0.934 -0.856 0.939 183s 0.535 0.777 -0.006 183s 183s y: [4, 3, 1, 1] 183s -0.683 -0.225 0.494 -0.054 183s 0.433 -0.042 -0.075 0.602 183s -0.406 -0.470 0.130 0.006 183s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 4, 4, 1] 183s 0.709 0.088 -0.566 0.576 183s -0.870 -0.337 0.732 -0.019 183s -0.664 0.307 -0.257 -0.804 183s 0.503 -0.387 0.601 0.295 183s 183s x1: [4, 3, 4, 1] 183s -0.891 0.429 0.874 0.893 183s 0.715 0.175 0.777 0.211 183s 0.341 0.405 -0.347 -0.315 183s 183s y: [4, 3, 4, 1] 183s -0.574 1.254 -0.219 0.174 183s 0.204 -0.117 -0.791 0.821 183s 0.292 -0.682 0.240 -0.287 183s testing: mul_mat, [1, 4, 4, 2] = [4, 1, 4, 2] * [4, 4, 4, 2] 183s x0: [4, 1, 4, 2] 183s -0.416 0.617 -0.610 0.259 183s 183s x1: [4, 4, 4, 2] 183s 0.307 0.493 -0.310 -0.643 183s -0.082 0.238 0.840 -0.926 183s -0.471 -0.076 -0.193 -0.200 183s -0.796 -0.165 -0.815 0.991 183s 183s y: [1, 4, 4, 2] 183s 0.199 183s -0.571 183s 0.215 183s 0.984 183s test-mul-mat0: iter:458/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.171 -0.474 183s 183s x1: [2, 4, 1, 1] 183s 0.169 -0.888 183s 0.273 -0.242 183s 0.021 -0.333 183s 0.243 0.738 183s 183s y: [1, 4, 1, 1] 183s 0.450 183s 0.161 183s 0.162 183s -0.309 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.893 0.383 183s -0.600 0.221 183s 0.160 -0.749 183s 183s x1: [2, 1, 1, 1] 183s -0.734 -0.053 183s 183s y: [3, 1, 1, 1] 183s 0.636 0.429 -0.078 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.648 0.663 183s -0.951 0.585 183s 183s x1: [2, 3, 1, 1] 183s 0.798 0.604 183s 0.826 0.740 183s 0.817 0.550 183s 183s y: [2, 3, 1, 1] 183s 0.918 -0.405 183s 1.026 -0.352 183s 0.894 -0.455 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.525 -0.627 183s -0.167 -0.996 183s -0.080 0.446 183s 183s x1: [2, 2, 1, 1] 183s -0.168 -0.839 183s -0.142 0.590 183s 183s y: [3, 2, 1, 1] 183s 0.438 0.864 -0.361 183s -0.445 -0.564 0.275 183s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.494 0.098 -0.085 183s 0.962 0.788 0.338 183s 0.764 0.504 0.769 183s 183s x1: [3, 2, 1, 1] 183s -0.361 -0.049 0.311 183s -0.255 -0.666 0.711 183s 183s y: [3, 2, 1, 1] 183s -0.210 -0.281 -0.062 183s -0.251 -0.529 0.017 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.743 -0.739 0.212 183s -0.918 0.527 -0.161 183s -0.096 0.543 -0.712 183s 183s x1: [3, 3, 1, 1] 183s -0.437 0.001 0.818 183s -0.852 0.834 0.979 183s 0.006 0.424 0.161 183s 183s y: [3, 3, 1, 1] 183s 0.497 0.270 -0.540 183s 0.225 1.064 -0.163 183s -0.283 0.192 0.115 183s test-mul-mat0: iter:459/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.088 -0.229 0.897 -0.751 183s 0.302 -0.846 -0.670 -0.794 183s -0.584 0.858 0.748 0.628 183s 183s x1: [4, 3, 1, 1] 183s -0.488 -0.982 -0.861 -0.700 183s 0.522 0.054 0.638 0.291 183s 0.617 -0.361 0.109 0.765 183s 183s y: [3, 3, 1, 1] 183s -0.064 1.816 -1.641 183s 0.387 -0.546 0.401 183s -0.341 -0.189 -0.109 183s testing: mul_mat, [1, 3, 2, 1] = [4, 1, 2, 1] * [4, 3, 2, 1] 183s x0: [4, 1, 2, 1] 183s 0.897 0.069 0.273 0.754 183s 183s x1: [4, 3, 2, 1] 183s -0.304 -0.964 0.175 0.246 183s -0.709 -0.785 0.758 -0.691 183s -0.646 -0.942 0.831 0.408 183s 183s y: [1, 3, 2, 1] 183s -0.105 183s -1.005 183s -0.109 183s testing: mul_mat, [3, 1, 2, 3] = [4, 3, 2, 3] * [4, 1, 2, 3] 183s x0: [4, 3, 2, 3] 183s 0.185 0.170 -0.819 -0.914 183s 0.621 0.035 -0.838 0.318 183s -0.820 -0.547 -0.227 0.634 183s 183s x1: [4, 1, 2, 3] 183s -0.924 0.697 0.145 -0.709 183s 183s y: [3, 1, 2, 3] 183s 0.477 -0.895 -0.106 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.716 0.583 0.292 -0.954 183s 183s x1: [4, 3, 1, 1] 183s -0.059 0.937 0.217 -0.047 183s 0.074 0.703 -0.541 0.534 183s -0.580 0.000 0.953 0.236 183s 183s y: [1, 3, 1, 1] 183s 0.696 183s -0.310 183s 0.468 183s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.023 183s 0.769 183s -0.248 183s -0.305 183s 183s x1: [1, 4, 2, 1] 183s 0.755 183s -0.995 183s -0.346 183s 0.115 183s 183s y: [4, 4, 2, 1] 183s -0.017 0.581 -0.187 -0.230 183s 0.023 -0.765 0.247 0.303 183s 0.008 -0.266 0.086 0.106 183s -0.003 0.088 -0.028 -0.035 183s testing: mul_mat, [1, 1, 2, 3] = [4, 1, 2, 3] * [4, 1, 2, 3] 183s x0: [4, 1, 2, 3] 183s 0.745 -0.802 -0.874 0.439 183s 183s x1: [4, 1, 2, 3] 183s -0.353 0.706 -0.427 0.106 183s 183s y: [1, 1, 2, 3] 183s -0.411 183s test-mul-mat0: iter:460/500 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.655 183s 0.078 183s -0.130 183s 0.784 183s 183s x1: [1, 2, 1, 1] 183s -0.400 183s -0.119 183s 183s y: [4, 2, 1, 1] 183s 0.262 -0.031 0.052 -0.313 183s 0.078 -0.009 0.015 -0.093 183s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.061 183s 183s x1: [1, 4, 1, 1] 183s 0.909 183s 0.866 183s 0.870 183s -0.555 183s 183s y: [1, 4, 1, 1] 183s 0.055 183s 0.053 183s 0.053 183s -0.034 183s testing: mul_mat, [3, 1, 1, 4] = [1, 3, 1, 4] * [1, 1, 1, 4] 183s x0: [1, 3, 1, 4] 183s -0.311 183s 0.641 183s 0.394 183s 183s x1: [1, 1, 1, 4] 183s -0.778 183s 183s y: [3, 1, 1, 4] 183s 0.242 -0.499 -0.307 183s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.681 183s 0.757 183s 183s x1: [1, 3, 1, 1] 183s -0.830 183s 0.974 183s -0.321 183s 183s y: [2, 3, 1, 1] 183s 0.565 -0.628 183s -0.663 0.737 183s 0.219 -0.243 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.651 0.034 183s 0.973 -0.668 183s -0.889 0.086 183s 0.087 0.392 183s 183s x1: [2, 1, 1, 1] 183s 0.785 -0.897 183s 183s y: [4, 1, 1, 1] 183s 0.481 1.363 -0.775 -0.283 183s testing: mul_mat, [2, 2, 1, 4] = [4, 2, 1, 4] * [4, 2, 1, 4] 183s x0: [4, 2, 1, 4] 183s -0.201 0.495 0.070 0.979 183s -0.711 -0.776 -0.668 0.617 183s 183s x1: [4, 2, 1, 4] 183s 0.525 0.876 -0.919 0.166 183s 0.271 0.425 -0.219 -0.163 183s 183s y: [2, 2, 1, 4] 183s 0.427 -0.337 183s -0.019 -0.477 183s test-mul-mat0: iter:461/500 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.520 -0.219 -0.610 -0.045 183s 0.156 -0.416 -0.397 0.019 183s 0.300 -0.430 -0.962 -0.809 183s -0.171 -0.676 -0.657 -0.588 183s 183s x1: [4, 1, 1, 1] 183s -0.448 0.303 0.982 -0.860 183s 183s y: [4, 1, 1, 1] 183s -0.859 -0.602 -0.512 -0.267 183s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 183s x0: [4, 1, 3, 1] 183s 0.151 -0.120 0.325 0.082 183s 183s x1: [4, 4, 3, 1] 183s -0.793 -0.156 -0.287 0.213 183s 0.569 0.812 -0.493 0.788 183s 0.136 -0.129 -0.660 -0.562 183s -0.147 -0.520 0.373 -0.627 183s 183s y: [1, 4, 3, 1] 183s -0.177 183s -0.108 183s -0.225 183s 0.110 183s testing: mul_mat, [1, 1, 3, 2] = [4, 1, 3, 2] * [4, 1, 3, 2] 183s x0: [4, 1, 3, 2] 183s -0.779 -0.639 0.925 0.135 183s 183s x1: [4, 1, 3, 2] 183s -0.344 0.594 -0.986 0.112 183s 183s y: [1, 1, 3, 2] 183s -1.010 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.337 0.613 0.429 0.698 183s 183s x1: [4, 1, 1, 1] 183s -0.648 -0.430 -0.232 -0.503 183s 183s y: [1, 1, 1, 1] 183s -0.933 183s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 3, 3, 1] 183s -0.392 183s -0.530 183s 0.667 183s 183s x1: [1, 4, 3, 1] 183s 0.538 183s -0.436 183s 0.648 183s -0.547 183s 183s y: [3, 4, 3, 1] 183s -0.211 -0.285 0.359 183s 0.171 0.231 -0.291 183s -0.254 -0.343 0.432 183s 0.214 0.290 -0.365 183s testing: mul_mat, [1, 1, 3, 2] = [3, 1, 3, 2] * [3, 1, 3, 2] 183s x0: [3, 1, 3, 2] 183s 0.023 0.959 -0.982 183s 183s x1: [3, 1, 3, 2] 183s -0.814 -0.335 -0.910 183s 183s y: [1, 1, 3, 2] 183s 0.554 183s test-mul-mat0: iter:462/500 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.284 0.294 -0.963 183s -0.325 0.312 0.668 183s 0.105 -0.003 0.599 183s -0.385 0.490 0.364 183s 183s x1: [3, 2, 1, 1] 183s 0.311 0.233 0.046 183s 0.533 0.200 -0.287 183s 183s y: [4, 2, 1, 1] 183s -0.065 0.003 0.060 0.011 183s 0.184 -0.303 -0.117 -0.212 183s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 4, 2, 1] 183s 0.933 -0.002 0.239 183s -0.452 -0.513 -0.397 183s -0.233 0.792 0.918 183s 0.256 0.003 0.865 183s 183s x1: [3, 4, 2, 1] 183s -0.781 0.305 0.315 183s -0.511 0.211 0.947 183s -0.368 -0.456 -0.125 183s -0.923 0.003 0.636 183s 183s y: [4, 4, 2, 1] 183s -0.654 0.072 0.712 0.073 183s -0.251 -0.253 1.155 0.689 183s -0.372 0.450 -0.390 -0.204 183s -0.710 0.163 0.801 0.314 183s testing: mul_mat, [2, 4, 2, 3] = [3, 2, 2, 3] * [3, 4, 2, 3] 183s x0: [3, 2, 2, 3] 183s -0.245 0.269 -0.100 183s -0.698 -0.042 0.997 183s 183s x1: [3, 4, 2, 3] 183s -0.792 -0.623 0.143 183s -0.820 0.205 0.144 183s 0.819 0.138 -0.859 183s 0.058 0.685 -0.372 183s 183s y: [2, 4, 2, 3] 183s 0.012 0.721 183s 0.242 0.707 183s -0.078 -1.434 183s 0.208 -0.440 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.921 -0.622 0.996 183s 183s x1: [3, 2, 1, 1] 183s -0.739 -0.808 0.695 183s -0.050 0.405 0.100 183s 183s y: [1, 2, 1, 1] 183s 1.876 183s -0.106 183s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.419 183s -0.843 183s -0.499 183s -0.183 183s 183s x1: [1, 3, 2, 1] 183s -0.478 183s 0.147 183s -0.551 183s 183s y: [4, 3, 2, 1] 183s 0.200 0.403 0.239 0.088 183s -0.061 -0.124 -0.073 -0.027 183s 0.231 0.464 0.275 0.101 183s testing: mul_mat, [2, 1, 2, 3] = [4, 2, 2, 3] * [4, 1, 2, 3] 183s x0: [4, 2, 2, 3] 183s 0.408 -0.293 0.885 0.479 183s 0.204 0.762 -0.537 0.184 183s 183s x1: [4, 1, 2, 3] 183s 0.717 0.300 -0.597 0.981 183s 183s y: [2, 1, 2, 3] 183s 0.147 0.876 183s test-mul-mat0: iter:463/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.058 0.873 0.376 0.825 183s 0.871 0.729 -0.590 -0.550 183s -0.024 -0.370 -0.587 -0.802 183s -0.621 -0.056 -0.524 -0.922 183s 183s x1: [4, 4, 1, 1] 183s -0.723 0.698 0.428 -0.140 183s 0.332 -0.869 0.615 0.874 183s 0.221 -0.299 0.824 0.515 183s -0.506 0.535 0.694 -0.250 183s 183s y: [4, 4, 1, 1] 183s 0.613 -0.295 -0.379 0.315 183s 0.213 -1.188 -0.748 -1.286 183s 0.487 -0.795 -0.791 -1.028 183s 0.493 -0.322 -0.393 0.151 183s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.255 0.816 -0.360 -0.526 183s 183s x1: [4, 4, 1, 1] 183s -0.628 0.336 0.410 -0.497 183s -0.049 0.284 0.723 0.652 183s 0.108 0.239 -0.854 -0.357 183s -0.067 -0.104 0.369 0.990 183s 183s y: [1, 4, 1, 1] 183s 0.548 183s -0.359 183s 0.662 183s -0.721 183s testing: mul_mat, [1, 1, 1, 2] = [4, 1, 1, 2] * [4, 1, 1, 2] 183s x0: [4, 1, 1, 2] 183s 0.313 -0.180 0.460 -0.277 183s 183s x1: [4, 1, 1, 2] 183s -0.774 0.090 0.450 -0.144 183s 183s y: [1, 1, 1, 2] 183s -0.012 183s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.217 0.208 0.024 0.250 183s 0.103 0.872 -0.487 0.603 183s 0.127 0.872 -0.654 0.796 183s 183s x1: [4, 1, 1, 1] 183s -0.938 0.115 -0.715 -0.792 183s 183s y: [3, 1, 1, 1] 183s -0.396 -0.126 -0.182 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.258 0.500 -0.500 183s 183s x1: [3, 4, 1, 1] 183s -0.894 0.106 0.444 183s 0.342 0.554 0.568 183s -0.345 -0.626 0.028 183s 0.378 0.696 0.438 183s 183s y: [1, 4, 1, 1] 183s 0.061 183s -0.096 183s -0.238 183s 0.031 183s testing: mul_mat, [1, 3, 1, 2] = [1, 1, 1, 2] * [1, 3, 1, 2] 183s x0: [1, 1, 1, 2] 183s -0.316 183s 183s x1: [1, 3, 1, 2] 183s 0.026 183s 0.708 183s -0.742 183s 183s y: [1, 3, 1, 2] 183s -0.008 183s -0.224 183s 0.235 183s test-mul-mat0: iter:464/500 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.174 0.459 -0.588 0.829 183s 0.832 0.440 0.207 0.528 183s -0.122 -0.408 -0.730 -0.622 183s 0.092 0.297 -0.914 0.350 183s 183s x1: [4, 1, 1, 1] 183s -0.201 0.399 0.832 0.905 183s 183s y: [4, 1, 1, 1] 183s 0.409 0.658 -1.308 -0.345 183s testing: mul_mat, [3, 4, 4, 1] = [4, 3, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 3, 4, 1] 183s 0.590 0.321 -0.422 0.457 183s 0.605 0.166 -0.482 0.465 183s -0.254 -0.394 -0.196 -0.839 183s 183s x1: [4, 4, 4, 1] 183s 0.541 -0.103 -0.266 -0.008 183s 0.581 0.990 0.468 -0.223 183s 0.593 0.194 0.575 -0.008 183s 0.027 0.480 0.835 -0.799 183s 183s y: [3, 4, 4, 1] 183s 0.395 0.435 -0.038 183s 0.361 0.186 -0.443 183s 0.165 0.110 -0.333 183s -0.548 -0.678 0.311 183s testing: mul_mat, [4, 3, 4, 3] = [4, 4, 4, 3] * [4, 3, 4, 3] 183s x0: [4, 4, 4, 3] 183s 0.499 -0.984 -0.091 0.844 183s -0.704 0.607 0.296 -0.205 183s -0.969 0.373 0.659 -0.029 183s -0.689 0.400 -0.357 0.993 183s 183s x1: [4, 3, 4, 3] 183s -0.731 0.418 0.536 0.356 183s 0.573 -0.066 0.301 -0.966 183s -0.885 0.134 0.073 -0.512 183s 183s y: [4, 3, 4, 3] 183s -0.525 0.854 1.208 0.834 183s -0.492 -0.156 -0.353 -1.489 183s -1.012 0.831 0.971 0.129 183s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 3, 1, 1] 183s 0.202 -0.977 0.556 0.112 183s -0.457 0.235 0.313 -0.505 183s -0.043 0.941 0.355 0.017 183s 183s x1: [4, 4, 1, 1] 183s 0.062 0.563 0.876 -0.882 183s -0.232 -0.788 0.088 0.130 183s 0.598 -0.446 -0.623 -0.676 183s -0.932 0.139 -0.965 0.237 183s 183s y: [3, 4, 1, 1] 183s -0.150 0.824 0.823 183s 0.786 -0.117 -0.698 183s 0.134 -0.231 -0.678 183s -0.833 0.037 -0.168 183s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 183s x0: [3, 2, 4, 1] 183s 0.207 0.589 0.229 183s -0.826 -0.657 -0.489 183s 183s x1: [3, 4, 4, 1] 183s -0.811 0.784 -0.921 183s 0.752 0.660 -0.803 183s -0.480 0.873 0.285 183s 0.650 0.470 0.839 183s 183s y: [2, 4, 4, 1] 183s 0.083 0.605 183s 0.360 -0.663 183s 0.480 -0.317 183s 0.603 -1.257 183s testing: mul_mat, [3, 3, 4, 3] = [2, 3, 4, 3] * [2, 3, 4, 3] 183s x0: [2, 3, 4, 3] 183s 0.952 0.990 183s 0.896 0.442 183s 0.796 0.700 183s 183s x1: [2, 3, 4, 3] 183s -0.584 0.022 183s 0.035 0.173 183s -0.950 -0.823 183s 183s y: [3, 3, 4, 3] 183s -0.533 -0.513 -0.449 183s 0.205 0.108 0.149 183s -1.719 -1.215 -1.332 183s test-mul-mat0: iter:465/500 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.301 183s -0.189 183s 183s x1: [1, 4, 1, 1] 183s 0.075 183s -0.255 183s 0.534 183s -0.344 183s 183s y: [2, 4, 1, 1] 183s -0.023 -0.014 183s 0.077 0.048 183s -0.161 -0.101 183s 0.104 0.065 183s testing: mul_mat, [2, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.705 183s -0.986 183s 183s x1: [1, 2, 4, 1] 183s 0.022 183s -0.242 183s 183s y: [2, 2, 4, 1] 183s 0.016 -0.022 183s -0.171 0.239 183s testing: mul_mat, [4, 2, 4, 2] = [1, 4, 4, 2] * [1, 2, 4, 2] 183s x0: [1, 4, 4, 2] 183s 0.671 183s -0.044 183s 0.044 183s -0.948 183s 183s x1: [1, 2, 4, 2] 183s -0.665 183s 0.441 183s 183s y: [4, 2, 4, 2] 183s -0.446 0.029 -0.029 0.631 183s 0.296 -0.019 0.019 -0.418 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.831 183s 0.797 183s 183s x1: [1, 4, 1, 1] 183s -0.674 183s 0.713 183s -0.926 183s 0.055 183s 183s y: [2, 4, 1, 1] 183s -0.560 -0.537 183s 0.593 0.568 183s -0.770 -0.738 183s 0.046 0.044 183s testing: mul_mat, [3, 1, 4, 1] = [2, 3, 4, 1] * [2, 1, 4, 1] 183s x0: [2, 3, 4, 1] 183s 0.355 -0.916 183s -0.523 -0.838 183s 0.077 -0.408 183s 183s x1: [2, 1, 4, 1] 183s 0.997 0.099 183s 183s y: [3, 1, 4, 1] 183s 0.262 -0.604 0.036 183s testing: mul_mat, [3, 2, 4, 2] = [3, 3, 4, 2] * [3, 2, 4, 2] 183s x0: [3, 3, 4, 2] 183s -0.789 -0.210 -0.170 183s -0.042 -0.472 0.577 183s 0.386 0.599 0.677 183s 183s x1: [3, 2, 4, 2] 183s -0.466 0.987 -0.744 183s -0.853 -0.426 -0.675 183s 183s y: [3, 2, 4, 2] 183s 0.287 -0.876 -0.093 183s 0.878 -0.152 -1.041 183s test-mul-mat0: iter:466/500 183s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.938 0.814 0.353 183s 0.268 0.022 0.658 183s 183s x1: [3, 3, 1, 1] 183s 0.484 -0.099 -0.021 183s 0.212 0.321 0.896 183s 0.418 0.018 0.111 183s 183s y: [2, 3, 1, 1] 183s -0.542 0.113 183s 0.378 0.653 183s -0.339 0.185 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.856 -0.546 0.450 183s -0.590 0.214 0.252 183s 0.060 0.198 -0.264 183s 0.961 -0.823 0.948 183s 183s x1: [3, 2, 1, 1] 183s -0.670 -0.054 -0.584 183s -0.364 -0.623 -0.844 183s 183s y: [4, 2, 1, 1] 183s 0.340 0.236 0.103 -1.153 183s 0.271 -0.132 0.077 -0.637 183s testing: mul_mat, [2, 4, 1, 2] = [3, 2, 1, 2] * [3, 4, 1, 2] 183s x0: [3, 2, 1, 2] 183s -0.044 -0.640 0.819 183s -0.083 -0.463 0.768 183s 183s x1: [3, 4, 1, 2] 183s 0.282 -0.927 0.366 183s -0.700 0.184 0.910 183s -0.638 -0.003 0.263 183s 0.630 -0.981 -0.079 183s 183s y: [2, 4, 1, 2] 183s 0.881 0.687 183s 0.659 0.672 183s 0.246 0.256 183s 0.536 0.342 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.774 -0.939 0.260 183s -0.433 0.429 -0.503 183s 0.049 -0.398 0.171 183s 0.116 -0.991 -0.657 183s 183s x1: [3, 2, 1, 1] 183s -0.139 -0.210 -0.692 183s -0.509 -0.191 0.229 183s 183s y: [4, 2, 1, 1] 183s -0.091 0.318 -0.042 0.646 183s -0.155 0.023 0.090 -0.020 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.262 0.646 0.923 0.563 183s -0.583 -0.125 -0.038 -0.718 183s -0.577 0.846 -0.727 -0.387 183s -0.929 -0.788 0.353 -0.939 183s 183s x1: [4, 3, 1, 1] 183s 0.454 0.531 -0.837 -0.629 183s -0.932 0.931 -0.429 -0.322 183s 0.065 0.070 -0.528 -0.891 183s 183s y: [4, 3, 1, 1] 183s -0.666 0.153 1.040 -0.544 183s -0.221 0.675 1.761 0.283 183s -0.928 0.614 0.751 0.535 183s testing: mul_mat, [1, 4, 1, 2] = [4, 1, 1, 2] * [4, 4, 1, 2] 183s x0: [4, 1, 1, 2] 183s 0.481 0.488 -0.280 0.367 183s 183s x1: [4, 4, 1, 2] 183s 0.452 -0.044 0.515 -0.017 183s 0.119 0.886 0.052 0.050 183s -0.544 0.730 -0.886 0.526 183s -0.799 -0.776 0.458 0.464 183s 183s y: [1, 4, 1, 2] 183s 0.046 183s 0.493 183s 0.536 183s -0.722 183s test-mul-mat0: iter:467/500 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.192 -0.698 -0.845 183s 0.098 0.354 -0.201 183s 0.065 0.859 -0.601 183s 183s x1: [3, 3, 1, 1] 183s -0.193 0.935 -0.253 183s -0.312 0.470 0.033 183s 0.444 -0.150 0.531 183s 183s y: [3, 3, 1, 1] 183s -0.476 0.363 0.943 183s -0.416 0.129 0.364 183s -0.259 -0.116 -0.419 183s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 2, 3, 1] 183s -0.761 0.707 0.499 183s -0.216 0.054 -0.381 183s 183s x1: [3, 3, 3, 1] 183s -0.454 0.347 0.119 183s 0.912 0.887 -0.112 183s 0.735 -0.843 -0.228 183s 183s y: [2, 3, 3, 1] 183s 0.651 0.072 183s -0.123 -0.106 183s -1.269 -0.117 183s testing: mul_mat, [1, 2, 3, 2] = [3, 1, 3, 2] * [3, 2, 3, 2] 183s x0: [3, 1, 3, 2] 183s 0.683 0.614 -0.386 183s 183s x1: [3, 2, 3, 2] 183s 0.986 -0.676 0.445 183s -0.990 0.208 0.636 183s 183s y: [1, 2, 3, 2] 183s 0.087 183s -0.794 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.311 0.896 -0.573 183s -0.225 0.942 0.417 183s 0.093 0.734 0.840 183s -0.793 -0.774 0.042 183s 183s x1: [3, 1, 1, 1] 183s 0.070 0.158 0.977 183s 183s y: [4, 1, 1, 1] 183s -0.440 0.541 0.943 -0.137 183s testing: mul_mat, [3, 3, 3, 1] = [4, 3, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.591 -0.156 -0.112 -0.998 183s 0.314 -0.333 0.054 -0.401 183s 0.767 0.379 -0.538 0.522 183s 183s x1: [4, 3, 3, 1] 183s 0.557 0.911 0.633 0.254 183s -0.321 0.907 -0.993 -0.073 183s -0.080 -0.566 0.353 0.612 183s 183s y: [3, 3, 3, 1] 183s -0.795 -0.195 0.564 183s 0.232 -0.428 0.594 183s -0.515 -0.063 -0.146 183s testing: mul_mat, [2, 4, 3, 2] = [3, 2, 3, 2] * [3, 4, 3, 2] 183s x0: [3, 2, 3, 2] 183s -0.283 0.671 -0.769 183s 0.966 0.509 -0.888 183s 183s x1: [3, 4, 3, 2] 183s 0.043 -0.265 0.974 183s 0.505 -0.263 -0.427 183s 0.028 0.542 0.407 183s -0.615 -0.757 -0.318 183s 183s y: [2, 4, 3, 2] 183s -0.939 -0.957 183s 0.009 0.733 183s 0.043 -0.059 183s -0.090 -0.697 183s test-mul-mat0: iter:468/500 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.588 183s 0.451 183s 183s x1: [1, 4, 1, 1] 183s -0.036 183s 0.402 183s 0.738 183s 0.075 183s 183s y: [2, 4, 1, 1] 183s 0.021 -0.016 183s -0.236 0.181 183s -0.434 0.333 183s -0.044 0.034 183s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.009 183s 0.583 183s -0.568 183s 0.992 183s 183s x1: [1, 2, 2, 1] 183s -0.526 183s -0.068 183s 183s y: [4, 2, 2, 1] 183s 0.005 -0.307 0.299 -0.522 183s 0.001 -0.040 0.039 -0.068 183s testing: mul_mat, [4, 4, 2, 2] = [1, 4, 2, 2] * [1, 4, 2, 2] 183s x0: [1, 4, 2, 2] 183s 0.912 183s 0.235 183s -0.434 183s -0.182 183s 183s x1: [1, 4, 2, 2] 183s -0.893 183s -0.698 183s 0.593 183s -0.979 183s 183s y: [4, 4, 2, 2] 183s -0.814 -0.210 0.388 0.163 183s -0.637 -0.164 0.303 0.127 183s 0.541 0.140 -0.258 -0.108 183s -0.893 -0.230 0.425 0.178 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.262 183s -0.223 183s 183s x1: [1, 4, 1, 1] 183s -0.914 183s 0.936 183s -0.622 183s -0.644 183s 183s y: [2, 4, 1, 1] 183s -0.240 0.204 183s 0.246 -0.209 183s -0.163 0.139 183s -0.169 0.144 183s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 4, 2, 1] 183s -0.376 0.839 183s -0.396 -0.834 183s 0.601 0.354 183s -0.464 -0.654 183s 183s x1: [2, 1, 2, 1] 183s -0.729 -0.254 183s 183s y: [4, 1, 2, 1] 183s 0.060 0.501 -0.528 0.504 183s testing: mul_mat, [2, 2, 2, 2] = [4, 2, 2, 2] * [4, 2, 2, 2] 183s x0: [4, 2, 2, 2] 183s 0.865 0.172 -0.844 -0.432 183s -0.282 -0.415 0.967 -0.628 183s 183s x1: [4, 2, 2, 2] 183s -0.003 -0.243 -0.657 0.083 183s -0.307 -0.279 0.439 -0.204 183s 183s y: [2, 2, 2, 2] 183s 0.474 -0.586 183s -0.595 0.755 183s test-mul-mat0: iter:469/500 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.868 -0.855 -0.441 183s 183s x1: [3, 4, 1, 1] 183s -0.185 0.803 -0.841 183s 0.850 0.100 -0.090 183s 0.793 0.092 -0.336 183s 0.248 0.381 0.958 183s 183s y: [1, 4, 1, 1] 183s -0.155 183s -0.783 183s -0.619 183s -0.963 183s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 1, 4, 1] 183s -0.678 -0.196 -0.254 183s 183s x1: [3, 1, 4, 1] 183s -0.270 -0.611 0.198 183s 183s y: [1, 1, 4, 1] 183s 0.253 183s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 1, 4, 1] 183s 0.117 0.311 0.257 183s 183s x1: [3, 1, 4, 1] 183s 0.698 -0.506 0.692 183s 183s y: [1, 1, 4, 1] 183s 0.102 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.499 -0.754 -0.718 183s -0.445 0.722 -0.337 183s 0.892 0.635 0.945 183s -0.005 -0.503 0.050 183s 183s x1: [3, 1, 1, 1] 183s 0.876 0.965 -0.082 183s 183s y: [4, 1, 1, 1] 183s -1.106 0.335 1.317 -0.494 183s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.622 -0.886 -0.382 0.425 183s 183s x1: [4, 3, 4, 1] 183s 0.344 0.869 0.224 -0.539 183s 0.179 -0.520 -0.335 0.507 183s 0.668 0.190 -0.362 0.602 183s 183s y: [1, 3, 4, 1] 183s -0.871 183s 0.915 183s 0.640 183s testing: mul_mat, [2, 4, 4, 1] = [1, 2, 4, 1] * [1, 4, 4, 1] 183s x0: [1, 2, 4, 1] 183s 0.920 183s -0.376 183s 183s x1: [1, 4, 4, 1] 183s -0.239 183s -0.385 183s -0.739 183s 0.722 183s 183s y: [2, 4, 4, 1] 183s -0.220 0.090 183s -0.354 0.145 183s -0.679 0.278 183s 0.664 -0.272 183s test-mul-mat0: iter:470/500 183s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 3, 1, 1] 183s 0.910 -0.784 0.672 183s -0.599 -0.910 -0.377 183s -0.275 0.957 0.792 183s 183s x1: [3, 1, 1, 1] 183s 0.593 0.648 -0.506 183s 183s y: [3, 1, 1, 1] 183s -0.309 -0.754 0.056 183s testing: mul_mat, [3, 3, 3, 1] = [3, 3, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 3, 3, 1] 183s 0.898 -0.021 -0.962 183s -0.089 -0.487 0.915 183s 0.226 0.246 0.989 183s 183s x1: [3, 3, 3, 1] 183s 0.013 0.534 0.876 183s -0.685 -0.267 -0.926 183s -0.765 0.357 -0.421 183s 183s y: [3, 3, 3, 1] 183s -0.842 0.540 1.001 183s 0.281 -0.656 -1.136 183s -0.290 -0.490 -0.501 183s testing: mul_mat, [1, 3, 3, 2] = [3, 1, 3, 2] * [3, 3, 3, 2] 183s x0: [3, 1, 3, 2] 183s -0.234 -0.065 0.944 183s 183s x1: [3, 3, 3, 2] 183s 0.384 -0.186 0.171 183s 0.631 -0.288 -0.850 183s 0.669 0.623 -0.337 183s 183s y: [1, 3, 3, 2] 183s 0.084 183s -0.931 183s -0.515 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.721 -0.619 -0.997 183s -0.994 0.785 -0.600 183s -0.580 0.633 -0.935 183s 0.574 0.131 0.947 183s 183s x1: [3, 1, 1, 1] 183s 0.421 0.796 0.940 183s 183s y: [4, 1, 1, 1] 183s -1.126 -0.358 -0.619 1.236 183s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 4, 3, 1] 183s -0.942 -0.622 0.678 -0.804 183s -0.745 0.372 -0.836 -0.906 183s 0.333 -0.970 -0.557 -0.823 183s 0.316 -0.487 0.924 -0.352 183s 183s x1: [4, 3, 3, 1] 183s 0.090 -0.055 0.258 0.852 183s 0.155 -0.249 0.858 -0.675 183s 0.137 0.258 -0.877 -0.961 183s 183s y: [4, 3, 3, 1] 183s -0.561 -1.075 -0.763 -0.007 183s 1.134 -0.314 0.371 1.202 183s -0.111 1.597 1.075 -0.554 183s testing: mul_mat, [3, 4, 3, 2] = [4, 3, 3, 2] * [4, 4, 3, 2] 183s x0: [4, 3, 3, 2] 183s 0.593 0.244 -0.290 0.411 183s 0.090 -0.507 0.337 0.145 183s -0.303 -0.145 -0.412 -0.238 183s 183s x1: [4, 4, 3, 2] 183s -0.756 0.278 -0.715 -0.182 183s -0.598 0.719 0.221 0.713 183s 0.502 -0.602 -0.531 0.484 183s 0.505 0.420 0.524 -0.133 183s 183s y: [3, 4, 3, 2] 183s -0.248 -0.476 0.527 183s 0.050 -0.241 -0.183 183s 0.504 0.242 0.039 183s 0.196 -0.011 -0.399 183s test-mul-mat0: iter:471/500 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.072 -0.808 -0.172 183s 0.148 0.693 0.463 183s 0.256 0.236 -0.147 183s 0.584 -0.890 -0.359 183s 183s x1: [3, 1, 1, 1] 183s 0.467 -0.191 -0.882 183s 183s y: [4, 1, 1, 1] 183s 0.340 -0.472 0.204 0.760 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.677 0.073 0.721 183s 183s x1: [3, 4, 1, 1] 183s -0.260 0.907 0.916 183s 0.419 0.872 0.171 183s -0.523 -0.284 0.743 183s -0.499 0.848 -0.222 183s 183s y: [1, 4, 1, 1] 183s 0.903 183s -0.096 183s 0.870 183s 0.240 183s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 183s x0: [3, 4, 1, 1] 183s 0.383 -0.157 0.353 183s 0.076 -0.693 -0.391 183s -0.687 0.160 -0.807 183s -0.577 0.801 -0.067 183s 183s x1: [3, 1, 1, 1] 183s 0.132 0.191 -0.984 183s 183s y: [4, 1, 1, 1] 183s -0.327 0.262 0.734 0.143 183s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 2, 1, 1] 183s -0.518 -0.779 0.322 183s 0.306 0.674 -0.574 183s 183s x1: [3, 4, 1, 1] 183s -0.670 0.716 -0.648 183s -0.798 -0.113 -0.171 183s -0.082 -0.369 0.330 183s -0.234 0.409 0.585 183s 183s y: [2, 4, 1, 1] 183s -0.419 0.649 183s 0.446 -0.222 183s 0.436 -0.464 183s -0.009 -0.132 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.632 -0.132 183s -0.419 0.433 183s 183s x1: [2, 3, 1, 1] 183s 0.057 -0.834 183s 0.779 -0.867 183s -0.528 -0.612 183s 183s y: [2, 3, 1, 1] 183s 0.146 -0.385 183s 0.606 -0.701 183s -0.253 -0.044 183s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.964 -0.974 183s -0.305 0.085 183s -0.682 -0.924 183s -0.333 0.611 183s 183s x1: [2, 2, 1, 1] 183s 0.514 0.198 183s 0.149 0.866 183s 183s y: [4, 2, 1, 1] 183s -0.689 -0.140 -0.534 -0.050 183s -0.988 0.028 -0.902 0.479 183s test-mul-mat0: iter:472/500 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.123 -0.012 0.867 183s -0.490 0.568 -0.264 183s 0.942 0.083 0.934 183s 0.092 -0.051 0.334 183s 183s x1: [3, 2, 1, 1] 183s -0.769 -0.761 -0.179 183s 0.288 -0.595 -0.400 183s 183s y: [4, 2, 1, 1] 183s -0.051 -0.008 -0.954 -0.091 183s -0.375 -0.374 -0.151 -0.077 183s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 4, 3, 1] 183s -0.944 0.910 -0.457 183s -0.515 -0.892 0.851 183s 0.145 -0.455 0.979 183s 0.093 0.387 -0.292 183s 183s x1: [3, 4, 3, 1] 183s 0.128 0.644 0.653 183s 0.795 0.670 -0.263 183s 0.871 0.280 -0.088 183s -0.572 -0.802 0.307 183s 183s y: [4, 4, 3, 1] 183s 0.167 -0.085 0.365 0.070 183s -0.021 -1.230 -0.447 0.409 183s -0.526 -0.774 -0.088 0.215 183s -0.331 1.272 0.584 -0.453 183s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.216 0.183 0.772 183s -0.930 0.470 0.144 183s 183s x1: [3, 4, 3, 1] 183s 0.742 0.031 0.545 183s 0.722 -0.876 -0.069 183s -0.570 -0.413 -0.963 183s -0.146 -0.713 0.372 183s 183s y: [2, 4, 3, 1] 183s 0.586 -0.598 183s -0.058 -1.093 183s -0.941 0.197 183s 0.125 -0.146 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s 0.234 -0.899 0.376 183s 183s x1: [3, 2, 1, 1] 183s 0.056 -0.152 -0.813 183s 0.314 -0.830 0.629 183s 183s y: [1, 2, 1, 1] 183s -0.156 183s 1.056 183s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.218 183s -0.951 183s 183s x1: [1, 3, 3, 1] 183s -0.981 183s -0.303 183s 0.110 183s 183s y: [2, 3, 3, 1] 183s 0.214 0.933 183s 0.066 0.288 183s -0.024 -0.105 183s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 3, 3, 1] 183s 0.017 -0.355 183s -0.425 0.148 183s 0.530 -0.237 183s 183s x1: [2, 1, 3, 1] 183s 0.093 -0.802 183s 183s y: [3, 1, 3, 1] 183s 0.286 -0.158 0.239 183s test-mul-mat0: iter:473/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.647 0.028 183s 0.707 -0.002 183s 183s x1: [2, 4, 1, 1] 183s -0.691 -0.013 183s -0.695 -0.493 183s 0.192 -0.183 183s 0.336 -0.547 183s 183s y: [2, 4, 1, 1] 183s 0.447 -0.488 183s 0.436 -0.490 183s -0.129 0.136 183s -0.233 0.239 183s testing: mul_mat, [4, 2, 4, 1] = [2, 4, 4, 1] * [2, 2, 4, 1] 183s x0: [2, 4, 4, 1] 183s 0.858 -0.714 183s 0.309 -0.836 183s -0.208 -0.499 183s -0.019 -0.872 183s 183s x1: [2, 2, 4, 1] 183s -0.825 -0.530 183s 0.745 -0.302 183s 183s y: [4, 2, 4, 1] 183s -0.330 0.188 0.436 0.478 183s 0.855 0.482 -0.004 0.249 183s testing: mul_mat, [2, 4, 4, 2] = [2, 2, 4, 2] * [2, 4, 4, 2] 183s x0: [2, 2, 4, 2] 183s 0.172 0.725 183s -0.601 -0.665 183s 183s x1: [2, 4, 4, 2] 183s 0.681 0.932 183s -0.216 -0.527 183s -0.568 0.765 183s -0.398 0.385 183s 183s y: [2, 4, 4, 2] 183s 0.792 -1.029 183s -0.419 0.480 183s 0.457 -0.167 183s 0.211 -0.017 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.948 0.340 183s 0.892 0.933 183s 0.387 -0.216 183s 183s x1: [2, 2, 1, 1] 183s 0.015 -0.500 183s 0.367 0.427 183s 183s y: [3, 2, 1, 1] 183s -0.156 -0.453 0.114 183s 0.493 0.725 0.050 183s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 183s x0: [3, 1, 4, 1] 183s -0.782 0.805 0.659 183s 183s x1: [3, 2, 4, 1] 183s 0.328 0.925 -0.101 183s -0.884 0.429 0.299 183s 183s y: [1, 2, 4, 1] 183s 0.421 183s 1.234 183s testing: mul_mat, [4, 3, 4, 2] = [1, 4, 4, 2] * [1, 3, 4, 2] 183s x0: [1, 4, 4, 2] 183s 0.606 183s 0.213 183s -0.219 183s -0.392 183s 183s x1: [1, 3, 4, 2] 183s 0.834 183s 0.129 183s -0.360 183s 183s y: [4, 3, 4, 2] 183s 0.506 0.178 -0.183 -0.327 183s 0.078 0.028 -0.028 -0.051 183s -0.218 -0.077 0.079 0.141 183s test-mul-mat0: iter:474/500 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.433 0.874 0.734 0.194 183s -0.968 0.413 0.048 0.256 183s -0.962 -0.091 0.817 0.827 183s -0.417 0.375 -0.664 0.808 183s 183s x1: [4, 2, 1, 1] 183s 0.200 0.165 -0.823 0.263 183s -0.837 0.502 0.281 0.508 183s 183s y: [4, 2, 1, 1] 183s -0.322 -0.098 -0.663 0.738 183s 0.380 1.161 1.409 0.762 183s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.172 -0.666 0.787 0.407 183s 183s x1: [4, 4, 4, 1] 183s -0.629 -0.549 -0.753 -0.851 183s 0.174 -0.010 0.350 -0.661 183s 0.166 -0.387 -0.498 -0.332 183s 0.894 -0.990 0.399 0.327 183s 183s y: [1, 4, 4, 1] 183s -0.682 183s 0.043 183s -0.241 183s 1.261 183s testing: mul_mat, [3, 1, 4, 3] = [4, 3, 4, 3] * [4, 1, 4, 3] 183s x0: [4, 3, 4, 3] 183s 0.098 -0.434 -0.683 -0.233 183s -0.621 -0.444 0.432 0.672 183s 0.135 0.163 0.555 -0.474 183s 183s x1: [4, 1, 4, 3] 183s 0.612 0.832 0.813 0.809 183s 183s y: [3, 1, 4, 3] 183s -1.045 0.146 0.286 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.914 -0.768 -0.877 0.531 183s -0.764 -0.858 0.439 0.967 183s 0.949 0.529 0.105 -0.732 183s 0.227 0.128 -0.385 -0.256 183s 183s x1: [4, 3, 1, 1] 183s -0.844 0.346 0.760 -0.357 183s 0.080 -0.951 0.743 -0.706 183s 0.044 -0.119 0.179 -0.167 183s 183s y: [4, 3, 1, 1] 183s -1.893 0.336 -0.277 -0.348 183s -0.223 0.399 0.166 -0.209 183s -0.114 -0.014 0.120 -0.031 183s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 1, 4, 1] 183s 0.363 -0.639 0.333 0.444 183s 183s x1: [4, 4, 4, 1] 183s -0.454 -0.684 -0.100 0.892 183s -0.924 0.543 -0.028 -0.875 183s 0.286 0.267 0.169 -0.833 183s -0.554 -0.997 -0.431 -0.640 183s 183s y: [1, 4, 4, 1] 183s 0.635 183s -1.080 183s -0.380 183s 0.008 183s testing: mul_mat, [3, 4, 4, 3] = [1, 3, 4, 3] * [1, 4, 4, 3] 183s x0: [1, 3, 4, 3] 183s -0.089 183s 0.771 183s 0.046 183s 183s x1: [1, 4, 4, 3] 183s -0.089 183s -0.802 183s 0.338 183s -0.425 183s 183s y: [3, 4, 4, 3] 183s 0.008 -0.068 -0.004 183s 0.071 -0.618 -0.037 183s -0.030 0.260 0.016 183s 0.038 -0.328 -0.020 183s test-mul-mat0: iter:475/500 183s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.913 -0.332 -0.679 183s 183s x1: [3, 2, 1, 1] 183s -0.222 -0.581 -0.862 183s -0.266 -0.307 -0.485 183s 183s y: [1, 2, 1, 1] 183s 0.981 183s 0.675 183s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 2, 3, 1] 183s 0.961 -0.878 0.011 183s 0.254 -0.257 -0.691 183s 183s x1: [3, 1, 3, 1] 183s 0.544 -0.296 -0.044 183s 183s y: [2, 1, 3, 1] 183s 0.782 0.245 183s testing: mul_mat, [2, 2, 3, 3] = [3, 2, 3, 3] * [3, 2, 3, 3] 183s x0: [3, 2, 3, 3] 183s 0.279 -0.843 0.176 183s -0.205 -0.118 -0.415 183s 183s x1: [3, 2, 3, 3] 183s 0.743 -0.700 -0.735 183s 0.287 0.004 0.221 183s 183s y: [2, 2, 3, 3] 183s 0.668 0.234 183s 0.116 -0.151 183s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 183s x0: [3, 4, 1, 1] 183s -0.910 -0.128 0.446 183s 0.722 -0.186 0.941 183s 0.532 -0.067 0.659 183s -0.781 -0.724 0.183 183s 183s x1: [3, 2, 1, 1] 183s 0.739 0.585 -0.710 183s -0.108 0.138 0.366 183s 183s y: [4, 2, 1, 1] 183s -1.064 -0.245 -0.114 -1.131 183s 0.244 0.241 0.174 0.052 183s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 183s x0: [4, 1, 3, 1] 183s 0.482 0.484 -0.530 0.619 183s 183s x1: [4, 3, 3, 1] 183s 0.956 0.965 -0.890 0.745 183s 0.511 0.451 -0.433 -0.467 183s -0.967 -0.863 0.306 0.291 183s 183s y: [1, 3, 3, 1] 183s 1.861 183s 0.405 183s -0.866 183s testing: mul_mat, [3, 4, 3, 3] = [1, 3, 3, 3] * [1, 4, 3, 3] 183s x0: [1, 3, 3, 3] 183s 0.193 183s 0.283 183s 0.450 183s 183s x1: [1, 4, 3, 3] 183s -0.837 183s 0.830 183s -0.045 183s 0.596 183s 183s y: [3, 4, 3, 3] 183s -0.162 -0.237 -0.377 183s 0.161 0.235 0.373 183s -0.009 -0.013 -0.020 183s 0.115 0.168 0.268 183s test-mul-mat0: iter:476/500 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.830 0.712 0.383 -0.467 183s -0.576 0.681 0.197 -0.027 183s 183s x1: [4, 2, 1, 1] 183s 0.774 0.525 0.051 -0.498 183s 0.460 -0.714 0.638 0.218 183s 183s y: [2, 2, 1, 1] 183s -0.017 -0.065 183s -0.747 -0.631 183s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.565 -0.668 -0.863 -0.324 183s 0.933 0.912 -0.799 -0.016 183s -0.587 0.661 0.270 -0.949 183s -0.121 -0.984 -0.779 -0.409 183s 183s x1: [4, 2, 1, 1] 183s -0.571 0.923 -0.147 0.310 183s -0.948 -0.673 -0.965 0.870 183s 183s y: [4, 2, 1, 1] 183s -0.268 0.421 0.611 -0.852 183s 1.537 -0.742 -0.974 1.172 183s testing: mul_mat, [4, 4, 1, 3] = [4, 4, 1, 3] * [4, 4, 1, 3] 183s x0: [4, 4, 1, 3] 183s 0.252 0.947 -0.333 0.420 183s -0.009 0.853 0.248 0.305 183s 0.090 0.956 0.635 -0.451 183s 0.254 0.364 -0.239 0.567 183s 183s x1: [4, 4, 1, 3] 183s 0.400 -0.246 0.015 0.081 183s 0.951 0.987 0.510 0.874 183s -0.159 -0.180 0.926 0.168 183s -0.144 0.796 -0.276 0.290 183s 183s y: [4, 4, 1, 3] 183s -0.103 -0.185 -0.226 0.054 183s 1.372 1.226 0.959 0.975 183s -0.448 0.129 0.326 -0.232 183s 0.931 0.700 0.441 0.484 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.580 -0.938 -0.678 0.712 183s 0.250 0.921 -0.902 -0.694 183s -0.932 -0.950 -0.680 0.570 183s 0.257 -0.491 0.484 0.193 183s 183s x1: [4, 4, 1, 1] 183s -0.914 0.481 -0.520 0.048 183s -0.512 -0.885 0.973 0.805 183s 0.745 -0.344 -0.847 -0.338 183s 0.226 -0.604 -0.169 0.743 183s 183s y: [4, 4, 1, 1] 183s 0.466 0.650 0.776 -0.714 183s 1.040 -2.379 1.115 0.929 183s 0.223 0.868 0.015 -0.114 183s 1.078 -0.863 0.901 0.416 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.409 0.529 0.867 0.652 183s 0.816 0.330 -0.962 0.965 183s 183s x1: [4, 4, 1, 1] 183s -0.214 -0.382 0.681 -0.099 183s -0.409 0.486 -0.354 0.247 183s 0.639 0.308 -0.527 -0.965 183s -0.861 -0.784 0.341 -0.441 183s 183s y: [2, 4, 1, 1] 183s 0.236 -1.052 183s -0.056 0.405 183s -0.661 0.200 183s -0.759 -1.715 183s testing: mul_mat, [1, 4, 1, 3] = [2, 1, 1, 3] * [2, 4, 1, 3] 183s x0: [2, 1, 1, 3] 183s -0.153 0.791 183s 183s x1: [2, 4, 1, 3] 183s -0.642 0.136 183s 0.676 -0.337 183s -0.293 -0.131 183s 0.449 0.325 183s 183s y: [1, 4, 1, 3] 183s 0.206 183s -0.370 183s -0.059 183s 0.189 183s test-mul-mat0: iter:477/500 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s 0.705 -0.472 183s 0.782 -0.299 183s 0.691 0.457 183s 183s x1: [2, 2, 1, 1] 183s 0.287 0.196 183s 0.355 -0.388 183s 183s y: [3, 2, 1, 1] 183s 0.110 0.166 0.288 183s 0.434 0.394 0.068 183s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.502 -0.319 183s 183s x1: [2, 3, 1, 1] 183s 0.006 0.327 183s 0.168 0.449 183s 0.179 -0.781 183s 183s y: [1, 3, 1, 1] 183s -0.107 183s -0.228 183s 0.159 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.956 -0.770 183s 183s x1: [2, 1, 1, 1] 183s 0.270 0.345 183s 183s y: [1, 1, 1, 1] 183s -0.007 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.504 -0.068 183s 183s x1: [2, 1, 1, 1] 183s 0.916 -0.394 183s 183s y: [1, 1, 1, 1] 183s -0.435 183s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 2, 1, 1] 183s -0.266 183s -0.605 183s 183s x1: [1, 2, 1, 1] 183s 0.379 183s 0.783 183s 183s y: [2, 2, 1, 1] 183s -0.101 -0.229 183s -0.208 -0.474 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.656 0.141 183s 183s x1: [2, 1, 1, 1] 183s 0.874 -0.561 183s 183s y: [1, 1, 1, 1] 183s 0.495 183s test-mul-mat0: iter:478/500 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.653 0.273 0.193 0.569 183s 183s x1: [4, 2, 1, 1] 183s -0.719 0.595 0.121 -0.499 183s -0.953 0.619 0.181 0.317 183s 183s y: [1, 2, 1, 1] 183s -0.568 183s -0.238 183s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 183s x0: [4, 3, 3, 1] 183s -0.641 0.781 0.583 -0.495 183s 0.722 0.864 -0.899 -0.157 183s -0.635 -0.852 -0.539 0.546 183s 183s x1: [4, 1, 3, 1] 183s 0.879 -0.334 -0.935 -0.189 183s 183s y: [3, 1, 3, 1] 183s -1.276 1.216 0.127 183s testing: mul_mat, [3, 3, 3, 2] = [4, 3, 3, 2] * [4, 3, 3, 2] 183s x0: [4, 3, 3, 2] 183s 0.367 0.730 0.267 0.161 183s 0.416 0.338 -0.491 0.594 183s -0.538 0.519 0.188 0.820 183s 183s x1: [4, 3, 3, 2] 183s 0.098 -0.710 -0.564 0.463 183s -0.563 -0.102 0.009 0.902 183s -0.678 -0.792 -0.360 -0.059 183s 183s y: [3, 3, 3, 2] 183s -0.558 0.353 -0.147 183s -0.133 0.262 0.991 183s -0.932 -0.408 -0.162 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.890 0.211 -0.200 0.279 183s 183s x1: [4, 3, 1, 1] 183s -0.869 -0.376 0.489 0.431 183s -0.150 0.789 0.501 -0.496 183s 0.795 0.793 -0.830 -0.819 183s 183s y: [1, 3, 1, 1] 183s 0.717 183s 0.062 183s -0.602 183s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 183s x0: [1, 4, 3, 1] 183s 0.382 183s 0.145 183s 0.783 183s -0.129 183s 183s x1: [1, 4, 3, 1] 183s -0.364 183s -0.521 183s 0.532 183s -0.665 183s 183s y: [4, 4, 3, 1] 183s -0.139 -0.053 -0.285 0.047 183s -0.199 -0.076 -0.408 0.067 183s 0.203 0.077 0.417 -0.069 183s -0.254 -0.097 -0.521 0.086 183s testing: mul_mat, [1, 1, 3, 2] = [4, 1, 3, 2] * [4, 1, 3, 2] 183s x0: [4, 1, 3, 2] 183s -0.928 0.536 -0.196 -0.473 183s 183s x1: [4, 1, 3, 2] 183s -0.364 0.041 0.417 0.847 183s 183s y: [1, 1, 3, 2] 183s -0.123 183s test-mul-mat0: iter:479/500 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.477 0.469 183s 0.193 0.534 183s 0.756 0.416 183s 0.027 -0.636 183s 183s x1: [2, 4, 1, 1] 183s -0.238 -0.920 183s -0.250 -0.166 183s 0.616 0.554 183s 0.360 0.022 183s 183s y: [4, 4, 1, 1] 183s -0.318 -0.537 -0.563 0.579 183s 0.041 -0.137 -0.258 0.099 183s -0.033 0.414 0.696 -0.336 183s -0.162 0.081 0.282 -0.004 183s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 3, 4, 1] 183s -0.318 0.569 183s -0.901 0.199 183s -0.283 -0.058 183s 183s x1: [2, 4, 4, 1] 183s 0.868 0.338 183s -0.133 0.147 183s -0.955 -0.483 183s 0.149 -0.157 183s 183s y: [3, 4, 4, 1] 183s -0.083 -0.715 -0.265 183s 0.126 0.149 0.029 183s 0.029 0.765 0.298 183s -0.137 -0.165 -0.033 183s testing: mul_mat, [1, 3, 4, 3] = [2, 1, 4, 3] * [2, 3, 4, 3] 183s x0: [2, 1, 4, 3] 183s -0.242 0.546 183s 183s x1: [2, 3, 4, 3] 183s -0.614 0.277 183s -0.545 0.102 183s 0.092 -0.123 183s 183s y: [1, 3, 4, 3] 183s 0.300 183s 0.188 183s -0.089 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.167 0.425 183s -0.333 0.308 183s 0.762 -0.232 183s 0.437 0.749 183s 183s x1: [2, 1, 1, 1] 183s -0.541 -0.293 183s 183s y: [4, 1, 1, 1] 183s -0.215 0.090 -0.344 -0.455 183s testing: mul_mat, [2, 2, 4, 1] = [4, 2, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 2, 4, 1] 183s 0.048 -0.794 -0.653 -0.993 183s -0.146 -0.632 0.778 0.285 183s 183s x1: [4, 2, 4, 1] 183s -0.538 0.859 0.520 0.066 183s -0.956 0.320 0.494 0.571 183s 183s y: [2, 2, 4, 1] 183s -1.112 -0.041 183s -1.189 0.484 183s testing: mul_mat, [2, 4, 4, 3] = [2, 2, 4, 3] * [2, 4, 4, 3] 183s x0: [2, 2, 4, 3] 183s 0.872 -0.968 183s -0.231 0.634 183s 183s x1: [2, 4, 4, 3] 183s 0.454 -0.060 183s -0.020 0.801 183s -0.282 -0.013 183s 0.087 0.723 183s 183s y: [2, 4, 4, 3] 183s 0.453 -0.143 183s -0.793 0.513 183s -0.233 0.057 183s -0.624 0.439 183s test-mul-mat0: iter:480/500 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.296 183s 0.143 183s 183s x1: [1, 4, 1, 1] 183s 0.884 183s 0.439 183s 0.941 183s 0.364 183s 183s y: [2, 4, 1, 1] 183s 0.262 0.126 183s 0.130 0.063 183s 0.279 0.134 183s 0.108 0.052 183s testing: mul_mat, [1, 2, 3, 1] = [1, 1, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 1, 3, 1] 183s 0.335 183s 183s x1: [1, 2, 3, 1] 183s -0.910 183s 0.297 183s 183s y: [1, 2, 3, 1] 183s -0.305 183s 0.100 183s testing: mul_mat, [2, 1, 3, 4] = [1, 2, 3, 4] * [1, 1, 3, 4] 183s x0: [1, 2, 3, 4] 183s -0.633 183s -0.636 183s 183s x1: [1, 1, 3, 4] 183s 0.762 183s 183s y: [2, 1, 3, 4] 183s -0.482 -0.485 183s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.292 183s 183s x1: [1, 2, 1, 1] 183s -0.837 183s -0.930 183s 183s y: [1, 2, 1, 1] 183s 0.245 183s 0.272 183s testing: mul_mat, [4, 1, 3, 1] = [1, 4, 3, 1] * [1, 1, 3, 1] 183s x0: [1, 4, 3, 1] 183s 0.984 183s -0.723 183s -0.513 183s 0.711 183s 183s x1: [1, 1, 3, 1] 183s 0.812 183s 183s y: [4, 1, 3, 1] 183s 0.798 -0.587 -0.416 0.577 183s testing: mul_mat, [1, 1, 3, 4] = [4, 1, 3, 4] * [4, 1, 3, 4] 183s x0: [4, 1, 3, 4] 183s -0.133 0.859 0.268 -0.842 183s 183s x1: [4, 1, 3, 4] 183s 0.756 -0.683 -0.737 -0.406 183s 183s y: [1, 1, 3, 4] 183s -0.543 183s test-mul-mat0: iter:481/500 183s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.546 0.631 0.721 -0.663 183s -0.351 -0.933 0.043 -0.028 183s 183s x1: [4, 3, 1, 1] 183s -0.374 -0.186 -0.493 0.462 183s -0.024 0.332 0.305 0.201 183s 0.495 -0.853 -0.705 0.517 183s 183s y: [2, 3, 1, 1] 183s -0.983 0.270 183s 0.283 -0.293 183s -1.119 0.576 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.034 -0.994 0.658 0.541 183s 0.468 -0.366 -0.127 -0.226 183s 183s x1: [4, 2, 1, 1] 183s 0.920 0.999 -0.244 -0.784 183s 0.803 -0.248 -0.447 0.380 183s 183s y: [2, 2, 1, 1] 183s -1.546 0.273 183s 0.186 0.438 183s testing: mul_mat, [2, 2, 1, 2] = [4, 2, 1, 2] * [4, 2, 1, 2] 183s x0: [4, 2, 1, 2] 183s 0.103 -0.141 -0.355 0.657 183s -0.761 0.489 -0.309 -0.755 183s 183s x1: [4, 2, 1, 2] 183s 0.835 -0.632 -0.079 -0.870 183s 0.885 0.179 0.676 0.516 183s 183s y: [2, 2, 1, 2] 183s -0.368 -0.264 183s 0.165 -1.185 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.546 -0.823 0.548 -0.557 183s -0.508 0.340 -0.929 -0.593 183s 183s x1: [4, 2, 1, 1] 183s 0.409 0.747 0.359 -0.413 183s 0.423 -0.125 0.487 0.436 183s 183s y: [2, 2, 1, 1] 183s -0.412 -0.042 183s -0.104 -0.968 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.134 0.092 183s -0.872 -0.248 183s 183s x1: [2, 4, 1, 1] 183s -0.284 -0.900 183s -0.354 -0.795 183s -0.209 -0.109 183s 0.353 0.023 183s 183s y: [2, 4, 1, 1] 183s -0.121 0.471 183s -0.121 0.506 183s -0.038 0.209 183s 0.050 -0.313 183s testing: mul_mat, [4, 2, 1, 2] = [2, 4, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 4, 1, 2] 183s -0.125 -0.632 183s 0.028 0.398 183s 0.850 0.123 183s 0.052 -0.188 183s 183s x1: [2, 2, 1, 2] 183s 0.601 -0.500 183s -0.501 -0.040 183s 183s y: [4, 2, 1, 2] 183s 0.240 -0.182 0.450 0.125 183s 0.088 -0.030 -0.431 -0.019 183s test-mul-mat0: iter:482/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.090 0.176 -0.403 -0.693 183s -0.701 0.409 -0.888 -0.862 183s -0.679 -0.437 -0.818 -0.977 183s 183s x1: [4, 3, 1, 1] 183s 0.038 0.283 -0.545 0.538 183s 0.781 0.415 -0.374 0.703 183s 0.250 0.201 0.060 -0.876 183s 183s y: [3, 3, 1, 1] 183s -0.107 0.108 -0.229 183s -0.333 -0.652 -1.092 183s 0.595 0.609 0.549 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.124 -0.278 -0.746 0.936 183s 183s x1: [4, 3, 1, 1] 183s 0.905 0.578 -0.251 0.851 183s -0.891 0.603 -0.111 0.392 183s -0.942 -0.572 0.173 0.473 183s 183s y: [1, 3, 1, 1] 183s 0.711 183s 0.393 183s 0.589 183s testing: mul_mat, [3, 1, 1, 4] = [4, 3, 1, 4] * [4, 1, 1, 4] 183s x0: [4, 3, 1, 4] 183s 0.584 -0.671 0.184 -0.527 183s 0.721 0.242 -0.099 -0.106 183s -0.285 0.955 0.771 0.438 183s 183s x1: [4, 1, 1, 4] 183s -0.153 -0.517 -0.154 -0.977 183s 183s y: [3, 1, 1, 4] 183s 0.744 -0.118 -0.997 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.168 0.049 -0.665 0.499 183s 183s x1: [4, 3, 1, 1] 183s -0.345 -0.882 -0.370 0.285 183s -0.622 0.601 -0.626 0.447 183s -0.600 -0.895 0.349 -0.068 183s 183s y: [1, 3, 1, 1] 183s 0.403 183s 0.773 183s -0.208 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s -0.965 183s -0.425 183s 0.896 183s 183s x1: [1, 4, 1, 1] 183s 0.639 183s -0.665 183s 0.410 183s 0.129 183s 183s y: [3, 4, 1, 1] 183s -0.616 -0.272 0.572 183s 0.641 0.283 -0.596 183s -0.395 -0.174 0.367 183s -0.125 -0.055 0.116 183s testing: mul_mat, [3, 1, 1, 4] = [3, 3, 1, 4] * [3, 1, 1, 4] 183s x0: [3, 3, 1, 4] 183s -0.483 0.866 -0.302 183s 0.823 -0.245 0.804 183s -0.134 -0.078 0.256 183s 183s x1: [3, 1, 1, 4] 183s -0.901 -0.771 -0.732 183s 183s y: [3, 1, 1, 4] 183s -0.011 -1.142 -0.007 183s test-mul-mat0: iter:483/500 183s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.030 -0.385 183s 0.983 -0.654 183s 183s x1: [2, 3, 1, 1] 183s -0.007 0.251 183s 0.074 -0.400 183s 0.823 0.897 183s 183s y: [2, 3, 1, 1] 183s -0.096 -0.171 183s 0.152 0.335 183s -0.370 0.223 183s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.120 -0.013 183s 183s x1: [2, 2, 3, 1] 183s 0.601 -0.829 183s -0.702 0.780 183s 183s y: [1, 2, 3, 1] 183s -0.061 183s 0.074 183s testing: mul_mat, [3, 1, 3, 4] = [2, 3, 3, 4] * [2, 1, 3, 4] 183s x0: [2, 3, 3, 4] 183s -0.173 -0.571 183s 0.457 0.934 183s -0.788 0.723 183s 183s x1: [2, 1, 3, 4] 183s -0.411 -0.224 183s 183s y: [3, 1, 3, 4] 183s 0.199 -0.397 0.162 183s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 183s x0: [2, 4, 1, 1] 183s 0.022 -0.342 183s 0.664 0.587 183s 0.685 0.543 183s 0.085 -0.758 183s 183s x1: [2, 3, 1, 1] 183s -0.946 0.438 183s 0.878 -0.578 183s 0.457 0.923 183s 183s y: [4, 3, 1, 1] 183s -0.171 -0.371 -0.410 -0.413 183s 0.217 0.244 0.287 0.513 183s -0.306 0.845 0.815 -0.661 183s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 183s x0: [4, 2, 3, 1] 183s 0.219 -0.836 -0.137 0.256 183s 0.319 -0.195 -0.594 0.813 183s 183s x1: [4, 2, 3, 1] 183s -0.549 -0.050 -0.040 -0.776 183s -0.253 0.812 0.735 0.364 183s 183s y: [2, 2, 3, 1] 183s -0.272 -0.773 183s -0.741 -0.379 183s testing: mul_mat, [2, 4, 3, 4] = [2, 2, 3, 4] * [2, 4, 3, 4] 183s x0: [2, 2, 3, 4] 183s -0.616 0.902 183s 0.691 0.114 183s 183s x1: [2, 4, 3, 4] 183s -0.617 0.516 183s -0.273 -0.774 183s 0.050 -0.039 183s 0.732 -0.732 183s 183s y: [2, 4, 3, 4] 183s 0.846 -0.368 183s -0.530 -0.277 183s -0.066 0.030 183s -1.111 0.423 183s test-mul-mat0: iter:484/500 183s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.751 -0.854 183s 183s x1: [2, 4, 1, 1] 183s 0.151 -0.508 183s -0.137 -0.223 183s -0.646 -0.363 183s 0.387 0.058 183s 183s y: [1, 4, 1, 1] 183s 0.320 183s 0.293 183s 0.795 183s -0.340 183s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 183s x0: [2, 1, 2, 1] 183s -0.343 -0.406 183s 183s x1: [2, 1, 2, 1] 183s 0.329 0.014 183s 183s y: [1, 1, 2, 1] 183s -0.118 183s testing: mul_mat, [2, 1, 2, 3] = [2, 2, 2, 3] * [2, 1, 2, 3] 183s x0: [2, 2, 2, 3] 183s -0.052 -0.055 183s 0.770 0.585 183s 183s x1: [2, 1, 2, 3] 183s 0.788 -0.463 183s 183s y: [2, 1, 2, 3] 183s -0.015 0.336 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.301 0.326 183s -0.630 -0.754 183s 183s x1: [2, 2, 1, 1] 183s -0.970 -0.707 183s 0.133 -0.046 183s 183s y: [2, 2, 1, 1] 183s -0.522 1.145 183s 0.025 -0.049 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.267 0.952 183s 0.876 0.248 183s 183s x1: [2, 2, 2, 1] 183s -0.860 -0.089 183s -0.423 -0.031 183s 183s y: [2, 2, 2, 1] 183s -0.315 -0.775 183s -0.143 -0.378 183s testing: mul_mat, [1, 2, 2, 3] = [2, 1, 2, 3] * [2, 2, 2, 3] 183s x0: [2, 1, 2, 3] 183s -0.891 -0.887 183s 183s x1: [2, 2, 2, 3] 183s 0.949 -0.125 183s 0.192 0.689 183s 183s y: [1, 2, 2, 3] 183s -0.735 183s -0.782 183s test-mul-mat0: iter:485/500 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.946 0.332 -0.840 0.284 183s -0.263 0.758 -0.800 -0.107 183s 183s x1: [4, 1, 1, 1] 183s 0.415 0.820 0.458 0.085 183s 183s y: [2, 1, 1, 1] 183s -0.481 0.137 183s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 3, 4, 1] 183s 0.939 -0.014 0.581 -0.267 183s 0.725 -0.288 0.267 0.340 183s 0.720 0.231 -0.177 0.829 183s 183s x1: [4, 2, 4, 1] 183s -0.377 -0.008 0.002 -0.264 183s -0.555 -0.206 -0.261 0.131 183s 183s y: [3, 2, 4, 1] 183s -0.282 -0.360 -0.492 183s -0.706 -0.368 -0.293 183s testing: mul_mat, [4, 3, 4, 3] = [4, 4, 4, 3] * [4, 3, 4, 3] 183s x0: [4, 4, 4, 3] 183s -0.556 -0.406 0.874 -0.326 183s 0.073 -0.432 -0.360 -0.579 183s -0.053 0.482 0.751 -0.360 183s -0.565 -0.562 -0.490 -0.311 183s 183s x1: [4, 3, 4, 3] 183s 0.717 0.342 0.390 0.842 183s -0.636 0.857 0.836 0.331 183s -0.632 0.457 -0.558 0.815 183s 183s y: [4, 3, 4, 3] 183s -0.471 -0.723 0.117 -1.050 183s 0.628 -0.909 0.955 -0.635 183s -0.588 -0.514 -0.459 0.120 183s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.528 0.890 0.437 0.555 183s -0.749 0.054 -0.572 0.657 183s 183s x1: [4, 4, 1, 1] 183s -0.961 -0.091 -0.463 -0.748 183s 0.417 -0.364 -0.918 0.083 183s 0.007 -0.382 -0.629 0.793 183s 0.525 -0.961 -0.500 -0.615 183s 183s y: [2, 4, 1, 1] 183s -1.206 0.488 183s -0.459 0.247 183s -0.171 0.854 183s -1.137 -0.564 183s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 183s x0: [2, 4, 4, 1] 183s -0.775 0.744 183s -0.902 0.489 183s 0.847 -0.357 183s -0.834 0.640 183s 183s x1: [2, 4, 4, 1] 183s -0.256 -0.438 183s -0.627 0.606 183s -0.267 0.941 183s 0.645 0.642 183s 183s y: [4, 4, 4, 1] 183s -0.127 0.017 -0.061 -0.067 183s 0.936 0.862 -0.747 0.911 183s 0.906 0.701 -0.562 0.825 183s -0.023 -0.268 0.318 -0.127 183s testing: mul_mat, [1, 2, 4, 3] = [4, 1, 4, 3] * [4, 2, 4, 3] 183s x0: [4, 1, 4, 3] 183s -0.872 -0.073 -0.515 0.123 183s 183s x1: [4, 2, 4, 3] 183s -0.159 -0.341 -0.955 -0.416 183s -0.852 -0.312 -0.775 -0.305 183s 183s y: [1, 2, 4, 3] 183s 0.605 183s 1.128 183s test-mul-mat0: iter:486/500 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.611 -0.212 -0.118 0.239 183s -0.773 0.578 0.016 -0.849 183s 0.397 -0.537 -0.182 -0.652 183s -0.989 0.060 -0.962 0.896 183s 183s x1: [4, 1, 1, 1] 183s 0.046 -0.495 0.088 -0.913 183s 183s y: [4, 1, 1, 1] 183s -0.152 0.455 0.863 -0.978 183s testing: mul_mat, [1, 4, 2, 1] = [4, 1, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 1, 2, 1] 183s -0.290 -0.130 -0.943 0.407 183s 183s x1: [4, 4, 2, 1] 183s -0.316 -0.991 -0.526 0.715 183s 0.266 -0.303 0.255 -0.706 183s 0.721 -0.865 0.340 -0.773 183s 0.223 0.428 -0.460 0.612 183s 183s y: [1, 4, 2, 1] 183s 1.007 183s -0.566 183s -0.733 183s 0.563 183s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 2, 2, 1] 183s 0.778 -0.683 0.813 -0.629 183s -0.287 -0.858 0.783 0.458 183s 183s x1: [4, 1, 2, 1] 183s 0.692 0.356 0.483 0.119 183s 183s y: [2, 1, 2, 1] 183s 0.613 -0.070 183s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.750 0.586 0.364 0.735 183s 0.217 -0.466 -0.149 0.564 183s 0.244 0.190 0.003 0.145 183s 183s x1: [4, 2, 1, 1] 183s 0.088 -0.163 -0.595 0.439 183s 0.966 -0.335 0.131 0.321 183s 183s y: [3, 2, 1, 1] 183s -0.055 0.431 0.052 183s -0.636 0.527 0.219 183s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 183s x0: [3, 3, 2, 1] 183s 0.150 -0.115 -0.597 183s -0.499 -0.934 0.088 183s -0.883 0.262 -0.111 183s 183s x1: [3, 4, 2, 1] 183s 0.518 0.022 0.627 183s 0.772 0.262 -0.654 183s 0.181 -0.707 0.775 183s 0.461 0.381 -0.388 183s 183s y: [3, 4, 2, 1] 183s -0.300 -0.224 -0.521 183s 0.476 -0.687 -0.541 183s -0.355 0.638 -0.431 183s 0.257 -0.620 -0.265 183s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 183s x0: [3, 3, 2, 1] 183s 0.217 0.391 -0.253 183s -0.481 -0.131 0.511 183s 0.203 -0.562 0.448 183s 183s x1: [3, 3, 2, 1] 183s -0.316 -0.158 -0.800 183s 0.216 -0.207 -0.901 183s 0.896 -0.007 0.215 183s 183s y: [3, 3, 2, 1] 183s 0.072 -0.236 -0.334 183s 0.193 -0.537 -0.244 183s 0.137 -0.320 0.282 183s test-mul-mat0: iter:487/500 183s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.130 -0.184 183s 0.168 0.684 183s -0.752 0.482 183s 183s x1: [2, 1, 1, 1] 183s -0.059 0.450 183s 183s y: [3, 1, 1, 1] 183s -0.075 0.298 0.261 183s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.056 -0.087 183s 183s x1: [2, 3, 2, 1] 183s -0.099 -0.233 183s -0.315 -0.708 183s 0.636 0.123 183s 183s y: [1, 3, 2, 1] 183s 0.015 183s 0.044 183s 0.025 183s testing: mul_mat, [1, 1, 2, 3] = [2, 1, 2, 3] * [2, 1, 2, 3] 183s x0: [2, 1, 2, 3] 183s 0.626 -0.218 183s 183s x1: [2, 1, 2, 3] 183s -0.717 0.678 183s 183s y: [1, 1, 2, 3] 183s -0.597 183s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.441 0.566 183s 0.888 0.277 183s -0.109 0.710 183s -0.158 -0.575 183s 183s x1: [2, 1, 1, 1] 183s 0.646 0.571 183s 183s y: [4, 1, 1, 1] 183s 0.038 0.732 0.335 -0.431 183s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 183s x0: [4, 3, 2, 1] 183s -0.082 -0.436 0.123 0.965 183s 0.368 -0.061 -0.172 -0.606 183s 0.019 0.685 -0.424 0.853 183s 183s x1: [4, 2, 2, 1] 183s -0.397 -0.972 0.400 -0.200 183s 0.162 0.544 -0.319 -0.456 183s 183s y: [3, 2, 2, 1] 183s 0.313 -0.035 -1.014 183s -0.730 0.358 0.122 183s testing: mul_mat, [1, 4, 2, 3] = [3, 1, 2, 3] * [3, 4, 2, 3] 183s x0: [3, 1, 2, 3] 183s -0.424 -0.812 0.078 183s 183s x1: [3, 4, 2, 3] 183s -0.899 -0.587 0.935 183s -0.422 0.326 -0.415 183s 0.986 -0.756 0.953 183s 0.004 -0.193 -0.108 183s 183s y: [1, 4, 2, 3] 183s 0.931 183s -0.118 183s 0.270 183s 0.146 183s test-mul-mat0: iter:488/500 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.049 -0.683 183s 0.473 -0.126 183s 183s x1: [2, 4, 1, 1] 183s 0.017 -0.221 183s -0.993 0.426 183s 0.846 0.883 183s -0.527 -0.871 183s 183s y: [2, 4, 1, 1] 183s 0.152 0.036 183s -0.340 -0.523 183s -0.562 0.289 183s 0.570 -0.139 183s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 2, 2, 1] 183s 0.041 0.752 183s 0.763 -0.292 183s 183s x1: [2, 2, 2, 1] 183s 0.675 0.010 183s -0.938 0.323 183s 183s y: [2, 2, 2, 1] 183s 0.036 0.512 183s 0.205 -0.810 183s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.501 0.965 183s 183s x1: [2, 2, 2, 1] 183s 0.216 0.017 183s 0.744 0.223 183s 183s y: [1, 2, 2, 1] 183s 0.124 183s 0.588 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.099 -0.438 183s 0.034 -0.554 183s 183s x1: [2, 1, 1, 1] 183s -0.161 0.710 183s 183s y: [2, 1, 1, 1] 183s -0.295 -0.399 183s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 183s x0: [2, 1, 2, 1] 183s 0.618 0.768 183s 183s x1: [2, 2, 2, 1] 183s 0.869 -0.993 183s -0.080 -0.090 183s 183s y: [1, 2, 2, 1] 183s -0.225 183s -0.118 183s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 3, 2, 1] 183s 0.993 183s -0.181 183s -0.013 183s 183s x1: [1, 2, 2, 1] 183s 0.511 183s -0.620 183s 183s y: [3, 2, 2, 1] 183s 0.508 -0.092 -0.006 183s -0.616 0.112 0.008 183s test-mul-mat0: iter:489/500 183s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.065 183s -0.623 183s 183s x1: [1, 4, 1, 1] 183s 0.598 183s -0.492 183s -0.856 183s 0.467 183s 183s y: [2, 4, 1, 1] 183s 0.039 -0.372 183s -0.032 0.307 183s -0.055 0.533 183s 0.030 -0.291 183s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 183s x0: [1, 2, 3, 1] 183s -0.153 183s -0.753 183s 183s x1: [1, 2, 3, 1] 183s -0.725 183s 0.748 183s 183s y: [2, 2, 3, 1] 183s 0.111 0.546 183s -0.114 -0.563 183s testing: mul_mat, [1, 2, 3, 4] = [1, 1, 3, 4] * [1, 2, 3, 4] 183s x0: [1, 1, 3, 4] 183s 0.265 183s 183s x1: [1, 2, 3, 4] 183s -0.286 183s 0.318 183s 183s y: [1, 2, 3, 4] 183s -0.076 183s 0.084 183s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 2, 1, 1] 183s 0.740 183s -0.823 183s 183s x1: [1, 1, 1, 1] 183s -0.431 183s 183s y: [2, 1, 1, 1] 183s -0.319 0.354 183s testing: mul_mat, [1, 1, 3, 1] = [2, 1, 3, 1] * [2, 1, 3, 1] 183s x0: [2, 1, 3, 1] 183s -0.073 0.829 183s 183s x1: [2, 1, 3, 1] 183s -0.245 0.032 183s 183s y: [1, 1, 3, 1] 183s 0.045 183s testing: mul_mat, [3, 2, 3, 4] = [1, 3, 3, 4] * [1, 2, 3, 4] 183s x0: [1, 3, 3, 4] 183s -0.086 183s -0.542 183s 0.091 183s 183s x1: [1, 2, 3, 4] 183s -0.198 183s -0.362 183s 183s y: [3, 2, 3, 4] 183s 0.017 0.107 -0.018 183s 0.031 0.196 -0.033 183s test-mul-mat0: iter:490/500 183s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.758 0.463 183s 0.574 0.666 183s 183s x1: [2, 1, 1, 1] 183s -0.436 -0.726 183s 183s y: [2, 1, 1, 1] 183s -0.006 -0.734 183s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 2, 1, 1] 183s 0.461 0.172 183s -0.757 0.570 183s 183s x1: [2, 2, 1, 1] 183s -0.970 -0.498 183s 0.744 -0.981 183s 183s y: [2, 2, 1, 1] 183s -0.533 0.451 183s 0.174 -1.123 183s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 183s x0: [2, 1, 1, 2] 183s 0.327 -0.553 183s 183s x1: [2, 2, 1, 2] 183s -0.479 0.525 183s -0.257 0.857 183s 183s y: [1, 2, 1, 2] 183s -0.447 183s -0.558 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s -0.732 -0.406 183s 183s x1: [2, 1, 1, 1] 183s 0.993 0.281 183s 183s y: [1, 1, 1, 1] 183s -0.841 183s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 1, 1, 1] 183s -0.205 183s 183s x1: [1, 2, 1, 1] 183s 0.129 183s 0.731 183s 183s y: [1, 2, 1, 1] 183s -0.026 183s -0.150 183s testing: mul_mat, [1, 1, 1, 2] = [1, 1, 1, 2] * [1, 1, 1, 2] 183s x0: [1, 1, 1, 2] 183s 0.780 183s 183s x1: [1, 1, 1, 2] 183s 0.761 183s 183s y: [1, 1, 1, 2] 183s 0.593 183s test-mul-mat0: iter:491/500 183s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 4, 1, 1] 183s -0.237 -0.204 183s 0.016 0.085 183s -0.211 -0.703 183s 0.972 0.057 183s 183s x1: [2, 4, 1, 1] 183s 0.473 0.104 183s -0.687 0.329 183s -0.458 -0.059 183s 0.435 -0.651 183s 183s y: [4, 4, 1, 1] 183s -0.133 0.017 -0.173 0.465 183s 0.096 0.017 -0.087 -0.649 183s 0.121 -0.012 0.138 -0.449 183s 0.030 -0.049 0.366 0.386 183s testing: mul_mat, [2, 4, 2, 1] = [2, 2, 2, 1] * [2, 4, 2, 1] 183s x0: [2, 2, 2, 1] 183s -0.918 -0.708 183s 0.257 0.518 183s 183s x1: [2, 4, 2, 1] 183s -0.109 0.101 183s -0.212 0.059 183s 0.895 -0.451 183s 0.730 -0.566 183s 183s y: [2, 4, 2, 1] 183s 0.029 0.024 183s 0.152 -0.024 183s -0.503 -0.004 183s -0.269 -0.106 183s testing: mul_mat, [1, 2, 2, 4] = [2, 1, 2, 4] * [2, 2, 2, 4] 183s x0: [2, 1, 2, 4] 183s -0.143 -0.485 183s 183s x1: [2, 2, 2, 4] 183s -0.180 0.366 183s -0.774 0.116 183s 183s y: [1, 2, 2, 4] 183s -0.152 183s 0.054 183s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 183s x0: [2, 1, 1, 1] 183s 0.019 0.031 183s 183s x1: [2, 1, 1, 1] 183s -0.241 0.512 183s 183s y: [1, 1, 1, 1] 183s 0.011 183s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 183s x0: [1, 4, 2, 1] 183s -0.615 183s 0.509 183s 0.395 183s 0.784 183s 183s x1: [1, 2, 2, 1] 183s -0.937 183s 0.571 183s 183s y: [4, 2, 2, 1] 183s 0.576 -0.477 -0.370 -0.734 183s -0.351 0.291 0.226 0.448 183s testing: mul_mat, [2, 1, 2, 4] = [4, 2, 2, 4] * [4, 1, 2, 4] 183s x0: [4, 2, 2, 4] 183s -0.192 0.114 0.242 0.216 183s 0.803 -0.541 0.909 -0.535 183s 183s x1: [4, 1, 2, 4] 183s -0.607 -0.179 0.311 -0.394 183s 183s y: [2, 1, 2, 4] 183s 0.086 0.103 183s test-mul-mat0: iter:492/500 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.601 183s 0.369 183s 0.551 183s -0.455 183s 183s x1: [1, 2, 1, 1] 183s 0.517 183s -0.520 183s 183s y: [4, 2, 1, 1] 183s 0.311 0.191 0.285 -0.235 183s -0.312 -0.192 -0.286 0.236 183s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 2, 2, 1] 183s 0.353 183s -0.961 183s 183s x1: [1, 4, 2, 1] 183s -0.806 183s 0.716 183s 0.960 183s 0.253 183s 183s y: [2, 4, 2, 1] 183s -0.285 0.774 183s 0.253 -0.688 183s 0.339 -0.922 183s 0.089 -0.243 183s testing: mul_mat, [4, 2, 2, 2] = [1, 4, 2, 2] * [1, 2, 2, 2] 183s x0: [1, 4, 2, 2] 183s -0.592 183s -0.435 183s -0.593 183s -0.112 183s 183s x1: [1, 2, 2, 2] 183s 0.453 183s 0.919 183s 183s y: [4, 2, 2, 2] 183s -0.268 -0.197 -0.269 -0.051 183s -0.544 -0.400 -0.545 -0.103 183s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.585 183s 0.239 183s -0.556 183s 0.628 183s 183s x1: [1, 4, 1, 1] 183s 0.747 183s -0.980 183s -0.455 183s 0.786 183s 183s y: [4, 4, 1, 1] 183s 0.437 0.179 -0.415 0.469 183s -0.573 -0.235 0.545 -0.616 183s -0.266 -0.109 0.253 -0.286 183s 0.460 0.188 -0.437 0.494 183s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 183s x0: [4, 1, 2, 1] 183s -0.805 0.164 -0.894 -0.548 183s 183s x1: [4, 1, 2, 1] 183s 0.783 0.636 -0.198 -0.102 183s 183s y: [1, 1, 2, 1] 183s -0.295 183s testing: mul_mat, [4, 4, 2, 2] = [1, 4, 2, 2] * [1, 4, 2, 2] 183s x0: [1, 4, 2, 2] 183s 0.129 183s -0.358 183s -0.949 183s -0.565 183s 183s x1: [1, 4, 2, 2] 183s 0.172 183s -0.394 183s 0.674 183s -0.055 183s 183s y: [4, 4, 2, 2] 183s 0.022 -0.061 -0.163 -0.097 183s -0.051 0.141 0.374 0.222 183s 0.087 -0.241 -0.640 -0.381 183s -0.007 0.020 0.052 0.031 183s test-mul-mat0: iter:493/500 183s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 2, 1, 1] 183s -0.011 -0.424 -0.176 0.632 183s -0.373 0.259 0.911 -0.224 183s 183s x1: [4, 2, 1, 1] 183s 0.996 -0.268 0.019 0.486 183s -0.339 0.465 0.420 -0.783 183s 183s y: [2, 2, 1, 1] 183s 0.407 -0.533 183s -0.762 0.805 183s testing: mul_mat, [4, 2, 4, 1] = [4, 4, 4, 1] * [4, 2, 4, 1] 183s x0: [4, 4, 4, 1] 183s -0.446 0.077 -0.107 -0.498 183s -0.969 -0.352 -0.464 0.575 183s -0.208 0.602 0.508 -0.252 183s -0.335 -0.827 0.723 0.213 183s 183s x1: [4, 2, 4, 1] 183s 0.889 0.354 0.032 -0.030 183s 0.038 0.947 -0.046 0.754 183s 183s y: [4, 2, 4, 1] 183s -0.357 -1.018 0.053 -0.574 183s -0.314 0.085 0.349 -0.668 183s testing: mul_mat, [3, 4, 4, 1] = [4, 3, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 3, 4, 1] 183s 0.987 0.465 0.360 -0.199 183s 0.598 0.774 -0.323 -0.885 183s -0.798 0.505 0.517 -0.908 183s 183s x1: [4, 4, 4, 1] 183s -0.151 0.452 -0.979 0.476 183s -0.527 -0.712 -0.105 0.096 183s -0.936 0.062 -0.616 -0.987 183s -0.048 -0.115 0.462 0.004 183s 183s y: [3, 4, 4, 1] 183s -0.386 0.155 -0.590 183s -0.908 -0.917 -0.080 183s -0.920 0.560 1.355 183s 0.064 -0.271 0.216 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.433 0.895 0.279 -0.482 183s 183s x1: [4, 3, 1, 1] 183s -0.536 -0.637 -0.802 -0.553 183s 0.225 0.374 0.178 -0.007 183s -0.826 0.772 0.709 0.041 183s 183s y: [1, 3, 1, 1] 183s -0.759 183s 0.486 183s 0.511 183s testing: mul_mat, [3, 4, 4, 1] = [1, 3, 4, 1] * [1, 4, 4, 1] 183s x0: [1, 3, 4, 1] 183s 0.587 183s -0.828 183s 0.105 183s 183s x1: [1, 4, 4, 1] 183s 0.549 183s 0.364 183s 0.376 183s 0.881 183s 183s y: [3, 4, 4, 1] 183s 0.323 -0.455 0.058 183s 0.214 -0.301 0.038 183s 0.221 -0.311 0.040 183s 0.517 -0.729 0.093 183s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 4, 4, 1] 183s -0.096 0.782 0.671 183s -0.165 -0.434 -0.602 183s -0.165 0.632 0.120 183s -0.955 0.954 -0.050 183s 183s x1: [3, 1, 4, 1] 183s -0.389 -0.175 0.764 183s 183s y: [4, 1, 4, 1] 183s 0.413 -0.320 0.045 0.167 183s test-mul-mat0: iter:494/500 183s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 4, 1, 1] 183s -0.504 183s -0.088 183s 0.691 183s 0.705 183s 183s x1: [1, 2, 1, 1] 183s -0.895 183s -0.054 183s 183s y: [4, 2, 1, 1] 183s 0.451 0.079 -0.618 -0.631 183s 0.027 0.005 -0.037 -0.038 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.564 183s 0.345 183s -0.336 183s 183s x1: [1, 4, 1, 1] 183s 0.105 183s -0.802 183s -0.783 183s -0.798 183s 183s y: [3, 4, 1, 1] 183s 0.059 0.036 -0.035 183s -0.452 -0.276 0.270 183s -0.442 -0.270 0.263 183s -0.450 -0.275 0.268 183s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.291 183s 183s x1: [1, 3, 1, 1] 183s 0.477 183s -0.881 183s 0.742 183s 183s y: [1, 3, 1, 1] 183s 0.139 183s -0.256 183s 0.216 183s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 183s x0: [1, 4, 1, 1] 183s 0.398 183s -0.869 183s 0.402 183s 0.773 183s 183s x1: [1, 1, 1, 1] 183s 0.939 183s 183s y: [4, 1, 1, 1] 183s 0.374 -0.816 0.378 0.726 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.131 0.189 -0.331 0.627 183s 183s x1: [4, 1, 1, 1] 183s -0.377 -0.307 0.026 -0.758 183s 183s y: [1, 1, 1, 1] 183s -0.591 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.549 183s 0.408 183s -0.902 183s 183s x1: [1, 4, 1, 1] 183s -0.900 183s -0.640 183s 0.332 183s 0.205 183s 183s y: [3, 4, 1, 1] 183s -0.494 -0.367 0.812 183s -0.351 -0.261 0.578 183s 0.182 0.135 -0.300 183s 0.113 0.084 -0.185 183s test-mul-mat0: iter:495/500 183s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 4, 1, 1] 183s 0.248 0.560 -0.427 -0.350 183s 0.333 0.197 0.343 -0.641 183s 0.439 -0.423 0.490 -0.372 183s 0.246 0.117 -0.272 0.606 183s 183s x1: [4, 1, 1, 1] 183s -0.129 -0.497 0.651 -0.838 183s 183s y: [4, 1, 1, 1] 183s -0.295 0.620 0.784 -0.775 183s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 2, 4, 1] 183s -0.189 0.055 -0.713 0.357 183s -0.040 0.760 0.603 0.829 183s 183s x1: [4, 4, 4, 1] 183s -0.551 0.934 -0.836 0.998 183s 0.341 -0.738 0.111 -0.906 183s 0.024 0.701 -0.035 0.527 183s 0.352 0.127 0.970 -0.400 183s 183s y: [2, 4, 4, 1] 183s 1.108 1.055 183s -0.508 -1.259 183s 0.247 0.947 183s -0.894 0.335 183s testing: mul_mat, [3, 2, 4, 2] = [4, 3, 4, 2] * [4, 2, 4, 2] 183s x0: [4, 3, 4, 2] 183s 0.402 0.806 0.619 -0.421 183s 0.662 0.456 -0.390 0.268 183s -0.774 -0.998 0.866 -0.224 183s 183s x1: [4, 2, 4, 2] 183s -0.759 0.500 -0.269 0.200 183s 0.260 -0.666 0.029 -0.980 183s 183s y: [3, 2, 4, 2] 183s -0.153 -0.116 -0.190 183s -0.002 -0.405 0.709 183s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.956 -0.632 -0.161 -0.415 183s 0.367 0.880 0.972 0.683 183s 0.790 0.262 -0.603 -0.549 183s -0.235 0.330 -0.890 -0.806 183s 183s x1: [4, 3, 1, 1] 183s -0.722 0.149 0.068 0.229 183s 0.722 -0.593 -0.538 0.012 183s -0.419 0.945 -0.030 0.579 183s 183s y: [4, 3, 1, 1] 183s 0.490 0.088 -0.697 -0.026 183s -0.234 -0.772 0.733 0.103 183s -0.432 1.044 -0.383 -0.030 183s testing: mul_mat, [3, 4, 4, 1] = [4, 3, 4, 1] * [4, 4, 4, 1] 183s x0: [4, 3, 4, 1] 183s 0.729 -0.907 0.449 -0.233 183s 0.464 -0.625 0.581 -0.340 183s 0.719 -0.161 -0.934 0.875 183s 183s x1: [4, 4, 4, 1] 183s 0.109 -0.761 -0.528 -0.742 183s 0.307 0.701 0.980 0.713 183s -0.837 -0.008 -0.706 -0.892 183s 0.962 0.874 -0.419 -0.994 183s 183s y: [3, 4, 4, 1] 183s 0.705 0.472 0.043 183s -0.138 0.031 -0.183 183s -0.712 -0.490 -0.723 183s -0.048 -0.005 0.071 183s testing: mul_mat, [1, 4, 4, 2] = [3, 1, 4, 2] * [3, 4, 4, 2] 183s x0: [3, 1, 4, 2] 183s -0.286 0.212 0.611 183s 183s x1: [3, 4, 4, 2] 183s -0.523 0.161 0.108 183s -0.515 0.607 0.079 183s 0.215 -0.225 -0.731 183s -0.681 0.449 -0.197 183s 183s y: [1, 4, 4, 2] 183s 0.250 183s 0.324 183s -0.556 183s 0.169 183s test-mul-mat0: iter:496/500 183s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 3, 1, 1] 183s -0.336 -0.698 -0.726 -0.180 183s -0.261 0.918 -0.675 0.124 183s -0.423 0.693 -0.584 0.629 183s 183s x1: [4, 3, 1, 1] 183s 0.209 -0.734 0.065 0.923 183s 0.478 -0.323 -0.674 -0.230 183s 0.864 -0.279 0.143 -0.169 183s 183s y: [3, 3, 1, 1] 183s 0.228 -0.658 -0.055 183s 0.595 0.005 -0.177 183s -0.169 -0.599 -0.749 183s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.428 -0.408 0.407 -0.429 183s 183s x1: [4, 3, 1, 1] 183s 0.899 0.978 -0.532 0.228 183s -0.085 -0.014 -0.563 0.181 183s -0.948 -0.640 -0.342 -0.272 183s 183s y: [1, 3, 1, 1] 183s -0.329 183s -0.338 183s -0.168 183s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 1, 1, 1] 183s -0.945 -0.536 0.895 -0.620 183s 183s x1: [4, 1, 1, 1] 183s 0.423 0.702 0.235 0.725 183s 183s y: [1, 1, 1, 1] 183s -1.015 183s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 183s x0: [4, 2, 1, 1] 183s 0.971 -0.431 -0.103 0.636 183s 0.982 0.199 0.555 -0.922 183s 183s x1: [4, 1, 1, 1] 183s 0.588 -0.528 -0.928 -0.996 183s 183s y: [2, 1, 1, 1] 183s 0.260 0.876 183s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 183s x0: [2, 2, 1, 1] 183s -0.650 0.971 183s 0.792 -0.925 183s 183s x1: [2, 4, 1, 1] 183s 0.607 0.996 183s -0.264 -0.665 183s -0.318 -0.836 183s -0.073 -0.910 183s 183s y: [2, 4, 1, 1] 183s 0.572 -0.441 183s -0.474 0.406 183s -0.605 0.522 183s -0.836 0.784 183s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 183s x0: [2, 3, 1, 1] 183s -0.873 -0.869 183s -0.863 -0.762 183s -0.522 -0.551 183s 183s x1: [2, 2, 1, 1] 183s 0.769 -0.974 183s -0.461 0.665 183s 183s y: [3, 2, 1, 1] 183s 0.175 0.079 0.135 183s -0.175 -0.108 -0.125 183s test-mul-mat0: iter:497/500 183s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 183s x0: [4, 4, 1, 1] 183s -0.280 -0.806 0.279 0.125 183s -0.680 0.416 0.603 -0.549 183s 0.654 -0.949 -0.436 -0.539 183s 0.699 0.573 -0.177 -0.017 183s 183s x1: [4, 4, 1, 1] 183s -0.638 0.284 0.087 0.969 183s 0.280 0.823 -0.696 0.963 183s 0.988 0.231 -0.948 -0.277 183s 0.580 0.845 -0.306 0.655 183s 183s y: [4, 4, 1, 1] 183s 0.095 0.072 -1.247 -0.315 183s -0.816 -0.796 -0.814 0.775 183s -0.761 -0.994 0.989 0.995 183s -0.847 -0.587 -0.642 0.933 183s testing: mul_mat, [4, 4, 2, 1] = [4, 4, 2, 1] * [4, 4, 2, 1] 183s x0: [4, 4, 2, 1] 183s -0.666 0.809 0.256 -0.440 183s -0.719 0.582 -0.161 0.476 183s 0.489 -0.038 0.366 0.050 183s -0.301 0.975 -0.195 0.715 183s 183s x1: [4, 4, 2, 1] 183s -0.339 -0.208 -0.737 -0.516 183s 0.096 -0.774 -0.528 -0.674 183s -0.722 0.195 0.907 -0.877 183s 0.889 0.562 0.737 -0.391 183s 183s y: [4, 4, 2, 1] 183s 0.096 -0.004 -0.453 -0.326 183s -0.529 -0.755 -0.151 -1.162 183s 1.256 0.069 -0.073 -0.396 183s 0.223 -0.618 0.663 -0.143 183s testing: mul_mat, [2, 4, 2, 2] = [4, 2, 2, 2] * [4, 4, 2, 2] 183s x0: [4, 2, 2, 2] 183s 0.635 -0.614 0.462 0.491 183s -0.966 0.180 -0.234 0.365 183s 183s x1: [4, 4, 2, 2] 183s -0.703 0.562 -0.514 -0.422 183s 0.144 0.325 -0.946 -0.367 183s -0.714 0.420 0.683 -0.014 183s 0.394 -0.512 -0.299 0.103 183s 183s y: [2, 4, 2, 2] 183s -1.236 0.746 183s -0.725 0.007 183s -0.402 0.599 183s 0.477 -0.365 183s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 183s x0: [4, 1, 1, 1] 183s 0.092 -0.687 -0.757 0.465 183s 183s x1: [4, 2, 1, 1] 183s -0.492 0.960 -0.167 -0.458 183s 0.140 0.599 0.907 0.024 183s 183s y: [1, 2, 1, 1] 183s -0.792 183s -1.075 183s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 183s x0: [1, 3, 2, 1] 183s -0.967 183s 0.793 183s -0.474 183s 183s x1: [1, 4, 2, 1] 183s 0.853 183s 0.373 183s 0.804 183s -0.276 183s 183s y: [3, 4, 2, 1] 183s -0.824 0.676 -0.404 183s -0.361 0.296 -0.177 183s -0.777 0.637 -0.381 183s 0.267 -0.219 0.131 183s testing: mul_mat, [2, 1, 2, 2] = [3, 2, 2, 2] * [3, 1, 2, 2] 183s x0: [3, 2, 2, 2] 183s 0.195 -0.582 -0.340 183s 0.287 0.048 -0.778 183s 183s x1: [3, 1, 2, 2] 183s -0.095 -0.242 0.576 183s 183s y: [2, 1, 2, 2] 183s -0.074 -0.487 183s test-mul-mat0: iter:498/500 183s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 183s x0: [3, 1, 1, 1] 183s -0.414 -0.033 -0.655 183s 183s x1: [3, 4, 1, 1] 183s 0.186 -0.942 -0.542 183s -0.619 0.345 -0.124 183s 0.429 -0.995 0.098 183s 0.201 0.671 -0.293 183s 183s y: [1, 4, 1, 1] 183s 0.309 183s 0.326 183s -0.209 183s 0.087 183s testing: mul_mat, [4, 1, 3, 1] = [3, 4, 3, 1] * [3, 1, 3, 1] 183s x0: [3, 4, 3, 1] 183s -0.623 0.326 -0.265 183s 0.391 -0.169 0.920 183s 0.450 0.289 -0.699 183s -0.205 -0.836 0.730 183s 183s x1: [3, 1, 3, 1] 183s 0.294 0.754 0.717 183s 183s y: [4, 1, 3, 1] 183s -0.128 0.647 -0.152 -0.167 183s testing: mul_mat, [3, 4, 3, 3] = [3, 3, 3, 3] * [3, 4, 3, 3] 183s x0: [3, 3, 3, 3] 183s -0.168 0.395 -0.083 183s -0.996 -0.326 0.380 183s -0.553 0.716 -0.592 183s 183s x1: [3, 4, 3, 3] 183s 0.687 -0.947 0.872 183s 0.988 -0.152 -0.964 183s 0.718 0.648 0.299 183s -0.351 -0.881 0.269 183s 183s y: [3, 4, 3, 3] 183s -0.562 -0.045 -1.574 183s -0.147 -1.301 -0.085 183s 0.110 -0.813 -0.110 183s -0.311 0.738 -0.596 183s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 183s x0: [3, 3, 1, 1] 183s -0.494 -0.352 0.064 183s -0.305 0.945 0.394 183s -0.511 0.853 0.905 183s 183s x1: [3, 3, 1, 1] 183s -0.262 -0.188 -0.620 183s 0.001 0.626 0.736 183s -0.519 0.963 -0.735 183s 183s y: [3, 3, 1, 1] 183s 0.156 -0.343 -0.588 183s -0.174 0.882 1.199 183s -0.130 0.779 0.422 183s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 183s x0: [3, 1, 3, 1] 183s -0.603 0.406 0.957 183s 183s x1: [3, 3, 3, 1] 183s 0.469 0.451 -0.420 183s -0.207 -0.842 0.891 183s -0.644 -0.571 0.747 183s 183s y: [1, 3, 3, 1] 183s -0.502 183s 0.636 183s 0.872 183s testing: mul_mat, [1, 3, 3, 3] = [1, 1, 3, 3] * [1, 3, 3, 3] 183s x0: [1, 1, 3, 3] 183s 0.154 183s 183s x1: [1, 3, 3, 3] 183s 0.015 183s 0.445 183s -0.825 183s 183s y: [1, 3, 3, 3] 183s 0.002 183s 0.069 183s -0.127 183s test-mul-mat0: iter:499/500 183s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 183s x0: [1, 1, 1, 1] 183s 0.097 183s 183s x1: [1, 2, 1, 1] 183s -0.115 183s -0.391 183s 183s y: [1, 2, 1, 1] 183s -0.011 183s -0.038 183s testing: mul_mat, [4, 1, 4, 1] = [1, 4, 4, 1] * [1, 1, 4, 1] 183s x0: [1, 4, 4, 1] 183s -0.776 183s 0.796 183s 0.573 183s -0.148 183s 183s x1: [1, 1, 4, 1] 183s 0.623 183s 183s y: [4, 1, 4, 1] 183s -0.483 0.496 0.357 -0.092 183s testing: mul_mat, [4, 4, 4, 2] = [1, 4, 4, 2] * [1, 4, 4, 2] 183s x0: [1, 4, 4, 2] 183s -0.434 183s 0.259 183s 0.574 183s 0.023 183s 183s x1: [1, 4, 4, 2] 183s -0.122 183s 0.144 183s 0.287 183s 0.732 183s 183s y: [4, 4, 4, 2] 183s 0.053 -0.032 -0.070 -0.003 183s -0.062 0.037 0.082 0.003 183s -0.124 0.074 0.165 0.007 183s -0.318 0.190 0.420 0.017 183s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 183s x0: [1, 3, 1, 1] 183s 0.799 183s -0.840 183s -0.539 183s 183s x1: [1, 4, 1, 1] 183s -0.308 183s 0.534 183s -0.057 183s -0.431 183s 183s y: [3, 4, 1, 1] 183s -0.246 0.258 0.166 183s 0.426 -0.449 -0.288 183s -0.045 0.048 0.031 183s -0.344 0.362 0.232 183s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 183s x0: [3, 4, 4, 1] 183s 0.616 0.929 -0.493 183s -0.978 -0.453 -0.035 183s -0.949 0.044 -0.946 183s 0.691 0.362 -0.959 183s 183s x1: [3, 1, 4, 1] 183s 0.778 0.458 -0.860 183s 183s y: [4, 1, 4, 1] 183s 1.328 -0.937 0.095 1.527 183s testing: mul_mat, [4, 3, 4, 2] = [4, 4, 4, 2] * [4, 3, 4, 2] 183s x0: [4, 4, 4, 2] 183s -0.098 0.579 -0.110 0.879 183s -0.433 0.923 -0.015 0.200 183s 0.544 -0.770 0.666 0.494 183s -0.418 0.165 0.131 -0.611 183s 183s x1: [4, 3, 4, 2] 183s 0.187 -0.115 0.229 0.655 183s -0.346 -0.844 -0.215 0.320 183s 0.784 -0.766 0.291 -0.746 183s 183s y: [4, 3, 4, 2] 183s 0.466 -0.060 0.667 -0.468 183s -0.150 -0.562 0.476 -0.218 183s -1.208 -1.200 0.841 0.040 183s rd.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 183s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 185s compression: 0.000000 185s convert time: 0.000000 ms / method = 0 185s -0.040188 0.785085 185s 0.405617 0.183192 185s 0.016901 0.564737 185s 0.001560 0.452003 185s -0.111647 -0.158779 185s 0.602449 0.452541 185s 0.464777 0.596264 185s 0.031770 0.389379 185s 0.522225 -0.094984 185s 0.246030 0.173205 185s 0.322603 -0.063522 185s 0.171129 0.240454 185s 0.435216 0.577889 185s 0.286599 -0.102425 185s -0.152230 0.570174 185s -0.116195 0.552043 185s main: elapsed ticks: 1812937 185s main: elapsed us: 1815905 / 1815.905000 ms 185s 135.421090 185s Testing 1 devices 185s 185s Backend 1/1: CPU 185s Device description: CPU 185s Device memory: 3897 MB (3897 MB free) 185s 185s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=1): OK 185s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=2): OK 185s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=3): OK 185s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=4): OK 185s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=5): OK 185s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=6): OK 185s test_dataset(shuffle=no, ndata_shard=2, ndata_batch=2): OK 185s test_dataset(shuffle=no, ndata_shard=2, ndata_batch=4): OK 185s test_dataset(shuffle=no, ndata_shard=2, ndata_batch=6): OK 185s test_dataset(shuffle=no, ndata_shard=3, ndata_batch=3): OK 185s test_dataset(shuffle=no, ndata_shard=3, ndata_batch=6): OK 185s test_dataset(shuffle=no, ndata_shard=4, ndata_batch=4): OK 185s test_dataset(shuffle=no, ndata_shard=5, ndata_batch=5): OK 185s test_dataset(shuffle=no, ndata_shard=6, ndata_batch=6): OK 185s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=1): OK 185s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=2): OK 185s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=3): OK 185s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=4): OK 185s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=5): OK 185s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=6): OK 185s test_dataset(shuffle=yes, ndata_shard=2, ndata_batch=2): OK 185s test_dataset(shuffle=yes, ndata_shard=2, ndata_batch=4): OK 185s test_dataset(shuffle=yes, ndata_shard=2, ndata_batch=6): OK 185s test_dataset(shuffle=yes, ndata_shard=3, ndata_batch=3): OK 185s test_dataset(shuffle=yes, ndata_shard=3, ndata_batch=6): OK 185s test_dataset(shuffle=yes, ndata_shard=4, ndata_batch=4): OK 185s test_dataset(shuffle=yes, ndata_shard=5, ndata_batch=5): OK 185s test_dataset(shuffle=yes, ndata_shard=6, ndata_batch=6): OK 185s test_grad(): OK 185s test_forward_backward(high_level=no, shuffle=no, subtest=results_initial): OK 185s test_forward_backward(high_level=no, shuffle=no, subtest=weights_after_forward): OK 185s test_forward_backward(high_level=no, shuffle=no, subtest=results_after_forward): OK 185s test_forward_backward(high_level=no, shuffle=no, subtest=weights_after_forward_backward): OK 185s test_forward_backward(high_level=no, shuffle=no, subtest=result_after_forward_backward): OK 185s test_forward_backward(high_level=yes, shuffle=no, subtest=results_initial): OK 185s test_forward_backward(high_level=yes, shuffle=no, subtest=weights_after_forward): OK 185s test_forward_backward(high_level=yes, shuffle=no, subtest=results_after_forward): OK 185s test_forward_backward(high_level=yes, shuffle=no, subtest=weights_after_forward_backward): OK 185s test_forward_backward(high_level=yes, shuffle=no, subtest=result_after_forward_backward): OK 185s test_forward_backward(high_level=yes, shuffle=yes, subtest=results_initial): OK 185s test_forward_backward(high_level=yes, shuffle=yes, subtest=weights_after_forward): OK 185s test_forward_backward(high_level=yes, shuffle=yes, subtest=results_after_forward): OK 185s test_forward_backward(high_level=yes, shuffle=yes, subtest=weights_after_forward_backward): OK 185s test_forward_backward(high_level=yes, shuffle=yes, subtest=result_after_forward_backward): OK 185s test_epoch_vs_fit(): OK 185s test_idata_split(high_level=no, epoch=1, subtest=weights): OK 185s test_idata_split(high_level=no, epoch=1, subtest=results_backward): OK 185s test_idata_split(high_level=no, epoch=1, subtest=results_forward): OK 185s test_idata_split(high_level=no, epoch=2, subtest=weights): OK 185s test_idata_split(high_level=no, epoch=2, subtest=results_backward): OK 185s test_idata_split(high_level=no, epoch=2, subtest=results_forward): OK 185s test_idata_split(high_level=no, epoch=3, subtest=weights): OK 185s test_idata_split(high_level=no, epoch=3, subtest=results_backward): OK 185s test_idata_split(high_level=no, epoch=3, subtest=results_forward): OK 185s test_idata_split(high_level=no, epoch=4, subtest=weights): OK 185s test_idata_split(high_level=no, epoch=4, subtest=results_backward): OK 185s test_idata_split(high_level=no, epoch=4, subtest=results_forward): OK 185s test_idata_split(high_level=yes, epoch=1, subtest=weights): OK 185s test_idata_split(high_level=yes, epoch=1, subtest=results_backward): OK 185s test_idata_split(high_level=yes, epoch=1, subtest=results_forward): OK 185s test_idata_split(high_level=yes, epoch=2, subtest=weights): OK 185s test_idata_split(high_level=yes, epoch=2, subtest=results_backward): OK 185s test_idata_split(high_level=yes, epoch=2, subtest=results_forward): OK 185s test_idata_split(high_level=yes, epoch=3, subtest=weights): OK 185s test_idata_split(high_level=yes, epoch=3, subtest=results_backward): OK 185s test_idata_split(high_level=yes, epoch=3, subtest=results_forward): OK 185s test_idata_split(high_level=yes, epoch=4, subtest=weights): OK 185s test_idata_split(high_level=yes, epoch=4, subtest=results_backward): OK 185s test_idata_split(high_level=yes, epoch=4, subtest=results_forward): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=1, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=1, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=1, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=2, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=2, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=2, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=3, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=3, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=3, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=4, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=4, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=4, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=1, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=1, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=1, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=2, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=2, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=2, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=3, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=3, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=3, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=4, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=4, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=4, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=1, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=1, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=1, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=2, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=2, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=2, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=3, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=3, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=3, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=4, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=4, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=4, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=1, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=1, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=1, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=2, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=2, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=2, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=3, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=3, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=3, subtest=results): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=4, subtest=grads): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=4, subtest=weights): OK 185s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=4, subtest=results): OK 185s test_regression(subtest=weights): OK 185s 118/118 tests passed 185s Backend CPU: OK 185s 185s 1/1 backends passed 185s OK 185s test_pad_reflect_1d: using CPU backend 205s Testing f32 205s Testing f16 205s Testing q4_0 205s Testing q4_1 205s Testing q5_0 205s Testing q5_1 205s Testing q8_0 205s Testing q8_1 205s Testing q2_K 205s Testing q3_K 205s Testing q4_K 205s Testing q5_K 205s Testing q6_K 205s Testing q8_K 205s Testing iq2_xxs 205s Testing iq2_xs 205s Testing iq3_xxs 205s Testing iq1_s 205s Testing iq4_nl 205s Testing iq3_s 205s Testing iq2_s 205s Testing iq4_xs 205s Testing i8 205s Testing i16 205s Testing i32 205s Testing i64 205s Testing f64 205s Testing iq1_m 205s Testing bf16 205s Testing tq1_0 205s Testing tq2_0 205s f16 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 6.63 GB/s 205s quantized throughput : 3.32 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 6.10 GB/s 205s quantized throughput : 3.05 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 6.36 GB/s 205s quantized throughput : 3.18 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 6.36 GB/s 205s quantized throughput : 3.18 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 10.17 GB/s 205s quantized throughput : 5.09 GB/s 205s 205s q4_0 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.55 GB/s 205s quantized throughput : 0.50 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.63 GB/s 205s quantized throughput : 0.51 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 11.74 GB/s 205s quantized throughput : 1.65 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.47 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 25.43 GB/s 205s quantized throughput : 3.58 GB/s 205s 205s q4_1 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 1.77 GB/s 205s quantized throughput : 0.28 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.09 GB/s 205s quantized throughput : 0.33 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 11.74 GB/s 205s quantized throughput : 1.83 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.42 GB/s 205s quantized throughput : 0.38 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 21.80 GB/s 205s quantized throughput : 3.41 GB/s 205s 205s q5_0 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.77 GB/s 205s quantized throughput : 0.48 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.88 GB/s 205s quantized throughput : 0.49 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 8.48 GB/s 205s quantized throughput : 1.46 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.57 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 21.80 GB/s 205s quantized throughput : 3.75 GB/s 205s 205s q5_1 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 1.75 GB/s 205s quantized throughput : 0.33 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 1.51 GB/s 205s quantized throughput : 0.28 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 8.48 GB/s 205s quantized throughput : 1.59 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.68 GB/s 205s quantized throughput : 0.13 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 19.07 GB/s 205s quantized throughput : 3.58 GB/s 205s 205s q8_0 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 1.98 GB/s 205s quantized throughput : 0.53 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.88 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 12.72 GB/s 205s quantized throughput : 3.38 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.88 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 25.43 GB/s 205s quantized throughput : 6.76 GB/s 205s 205s q2_K 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.05 GB/s 205s quantized throughput : 0.00 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.05 GB/s 205s quantized throughput : 0.00 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 8.48 GB/s 205s quantized throughput : 0.70 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.27 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 30.52 GB/s 205s quantized throughput : 2.50 GB/s 205s 205s q3_K 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.55 GB/s 205s quantized throughput : 0.06 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.50 GB/s 205s quantized throughput : 0.05 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 9.54 GB/s 205s quantized throughput : 1.02 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.36 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 19.07 GB/s 205s quantized throughput : 2.05 GB/s 205s 205s q4_K 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.04 GB/s 205s quantized throughput : 0.01 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.04 GB/s 205s quantized throughput : 0.01 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 9.54 GB/s 205s quantized throughput : 1.34 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.39 GB/s 205s quantized throughput : 0.48 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 25.43 GB/s 205s quantized throughput : 3.58 GB/s 205s 205s q5_K 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.05 GB/s 205s quantized throughput : 0.01 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.05 GB/s 205s quantized throughput : 0.01 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 8.03 GB/s 205s quantized throughput : 1.38 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.57 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 21.80 GB/s 205s quantized throughput : 3.75 GB/s 205s 205s q6_K 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.07 GB/s 205s quantized throughput : 0.02 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.07 GB/s 205s quantized throughput : 0.02 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.88 GB/s 205s quantized throughput : 0.59 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.68 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 19.07 GB/s 205s quantized throughput : 3.91 GB/s 205s 205s iq4_nl 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.13 GB/s 205s quantized throughput : 0.02 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.19 GB/s 205s quantized throughput : 0.03 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 4.36 GB/s 205s quantized throughput : 0.61 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.59 GB/s 205s quantized throughput : 0.36 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 25.43 GB/s 205s quantized throughput : 3.58 GB/s 205s 205s iq4_xs 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.02 GB/s 205s quantized throughput : 0.00 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.02 GB/s 205s quantized throughput : 0.00 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 4.36 GB/s 205s quantized throughput : 0.58 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.25 GB/s 205s quantized throughput : 0.43 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 30.52 GB/s 205s quantized throughput : 4.05 GB/s 205s 205s bf16 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.28 GB/s 205s quantized throughput : 1.14 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.31 GB/s 205s quantized throughput : 1.16 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 5.09 GB/s 205s quantized throughput : 2.54 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.03 GB/s 205s quantized throughput : 1.02 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 2.18 GB/s 205s quantized throughput : 1.09 GB/s 205s 205s tq1_0 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 1.03 GB/s 205s quantized throughput : 0.05 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 1.02 GB/s 205s quantized throughput : 0.05 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 8.03 GB/s 205s quantized throughput : 0.42 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.17 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 50.86 GB/s 205s quantized throughput : 2.68 GB/s 205s 205s tq2_0 205s quantize_row_q_reference 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 1.00 GB/s 205s quantized throughput : 0.06 GB/s 205s 205s quantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 0.97 GB/s 205s quantized throughput : 0.06 GB/s 205s 205s dequantize_row_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 6.36 GB/s 205s quantized throughput : 0.41 GB/s 205s 205s quantize_row_q_dot 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 3.32 GB/s 205s quantized throughput : 0.21 GB/s 205s 205s vec_dot_q 205s 4096 values (0.02 MB) 205s min cycles/32 vals : 0.00 205s avg cycles/32 vals : 0.00 205s float32 throughput : 76.29 GB/s 205s quantized throughput : 4.92 GB/s 205s 205s ggml_roll(3(1), 7(0), 4(-1), 2(0), contiguous): PASSED 205s ggml_roll(37(-4), 42(3), 59(-7), 2(1), contiguous): PASSED 205s ggml_roll(37(-4), 42(3), 59(-7), 2(1), permuted): PASSED 205s 0.8439 -0.9970 0.6497 0.9733 0.9981 0.9999 1.0000 -0.5366 -0.0776 0.7602 0.2296 0.0621 0.0167 0.0045 0.0000 0.0000 0.4242 0.9880 -0.1558 0.8946 0.9923 0.9994 1.0000 -0.9056 0.1547 0.9878 0.4470 0.1240 0.0333 0.0089 0.0000 0.0000 205s ----------------------------------- 205s 0.8439 -0.9970 0.6497 0.9733 0.9981 0.9999 1.0000 -0.5366 -0.0776 0.7602 0.2296 0.0621 0.0167 0.0045 0.0000 0.0000 0.4242 0.9880 -0.1558 0.8946 0.9923 0.9994 1.0000 -0.9056 0.1547 0.9878 0.4470 0.1240 0.0333 0.0089 0.0000 0.0000 207s main: elapsed ticks: 1982394 207s 2951358968546669887488.000000 208s ggml_print_objects: objects in context 0xb7d2731cf930: 208s - ggml_object: type = 0, offset = 32, size = 384, next = 0xe01c756bf1e0 208s - ggml_object: type = 0, offset = 448, size = 736, next = 0xe01c756bf4e0 208s - ggml_object: type = 0, offset = 1216, size = 24336, next = (nil) 208s ggml_print_objects: --- end --- 208s autopkgtest [19:37:38]: test libggml-cpu: -----------------------] 209s autopkgtest [19:37:39]: test libggml-cpu: - - - - - - - - - - results - - - - - - - - - - 209s libggml-cpu PASS 209s autopkgtest [19:37:39]: test libggml-cpu: preparing testbed 239s Creating nova instance adt-questing-arm64-ggml-20250707-193409-juju-7f2275-prod-proposed-migration-environment-2-954b271b-a1cd-49e8-988e-09a11edc8a51 from image adt/ubuntu-questing-arm64-server-20250707.img (UUID 5a1f42c1-3f6f-4084-8e72-afd42fd719a8)... 299s autopkgtest [19:39:09]: testbed dpkg architecture: arm64 299s autopkgtest [19:39:09]: testbed apt version: 3.1.3 300s autopkgtest [19:39:10]: @@@@@@@@@@@@@@@@@@@@ test bed setup 300s autopkgtest [19:39:10]: testbed release detected to be: questing 301s autopkgtest [19:39:11]: updating testbed package index (apt update) 301s Get:1 http://ftpmaster.internal/ubuntu questing-proposed InRelease [249 kB] 302s Hit:2 http://ftpmaster.internal/ubuntu questing InRelease 302s Hit:3 http://ftpmaster.internal/ubuntu questing-updates InRelease 302s Hit:4 http://ftpmaster.internal/ubuntu questing-security InRelease 302s Get:5 http://ftpmaster.internal/ubuntu questing-proposed/universe Sources [177 kB] 302s Get:6 http://ftpmaster.internal/ubuntu questing-proposed/multiverse Sources [15.8 kB] 302s Get:7 http://ftpmaster.internal/ubuntu questing-proposed/main Sources [27.5 kB] 302s Get:8 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 Packages [38.5 kB] 302s Get:9 http://ftpmaster.internal/ubuntu questing-proposed/universe arm64 Packages [154 kB] 302s Get:10 http://ftpmaster.internal/ubuntu questing-proposed/multiverse arm64 Packages [15.7 kB] 302s Fetched 677 kB in 1s (793 kB/s) 303s Reading package lists... 304s autopkgtest [19:39:14]: upgrading testbed (apt dist-upgrade and autopurge) 304s Reading package lists... 305s Building dependency tree... 305s Reading state information... 306s Calculating upgrade... 307s The following packages will be upgraded: 307s apparmor libapparmor1 libc-bin libc-dev-bin libc6 libc6-dev libnss-systemd 307s libpam-systemd libsystemd-shared libsystemd0 libudev1 libxml2-16 locales 307s systemd systemd-cryptsetup systemd-resolved systemd-sysv systemd-timesyncd 307s udev 307s 19 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. 307s Need to get 19.1 MB of archives. 307s After this operation, 34.8 kB disk space will be freed. 307s Get:1 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc6-dev arm64 2.41-9ubuntu1 [1749 kB] 308s Get:2 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc-dev-bin arm64 2.41-9ubuntu1 [22.7 kB] 308s Get:3 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc6 arm64 2.41-9ubuntu1 [2919 kB] 308s Get:4 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc-bin arm64 2.41-9ubuntu1 [599 kB] 308s Get:5 http://ftpmaster.internal/ubuntu questing/main arm64 libsystemd0 arm64 257.7-1ubuntu1 [522 kB] 308s Get:6 http://ftpmaster.internal/ubuntu questing/main arm64 libnss-systemd arm64 257.7-1ubuntu1 [170 kB] 308s Get:7 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-sysv arm64 257.7-1ubuntu1 [11.9 kB] 308s Get:8 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-resolved arm64 257.7-1ubuntu1 [324 kB] 308s Get:9 http://ftpmaster.internal/ubuntu questing/main arm64 libpam-systemd arm64 257.7-1ubuntu1 [254 kB] 308s Get:10 http://ftpmaster.internal/ubuntu questing/main arm64 libsystemd-shared arm64 257.7-1ubuntu1 [2239 kB] 308s Get:11 http://ftpmaster.internal/ubuntu questing/main arm64 systemd arm64 257.7-1ubuntu1 [2939 kB] 308s Get:12 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-timesyncd arm64 257.7-1ubuntu1 [41.0 kB] 308s Get:13 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-cryptsetup arm64 257.7-1ubuntu1 [121 kB] 308s Get:14 http://ftpmaster.internal/ubuntu questing/main arm64 udev arm64 257.7-1ubuntu1 [1407 kB] 308s Get:15 http://ftpmaster.internal/ubuntu questing/main arm64 libudev1 arm64 257.7-1ubuntu1 [203 kB] 308s Get:16 http://ftpmaster.internal/ubuntu questing/main arm64 libapparmor1 arm64 4.1.1-0ubuntu3 [56.2 kB] 308s Get:17 http://ftpmaster.internal/ubuntu questing/main arm64 libxml2-16 arm64 2.14.4+dfsg-0exp1 [595 kB] 308s Get:18 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 locales all 2.41-9ubuntu1 [4253 kB] 308s Get:19 http://ftpmaster.internal/ubuntu questing/main arm64 apparmor arm64 4.1.1-0ubuntu3 [666 kB] 309s Preconfiguring packages ... 309s Fetched 19.1 MB in 1s (13.2 MB/s) 309s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 310s Preparing to unpack .../libc6-dev_2.41-9ubuntu1_arm64.deb ... 310s Unpacking libc6-dev:arm64 (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 310s Preparing to unpack .../libc-dev-bin_2.41-9ubuntu1_arm64.deb ... 310s Unpacking libc-dev-bin (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 310s Preparing to unpack .../libc6_2.41-9ubuntu1_arm64.deb ... 310s Unpacking libc6:arm64 (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 310s Setting up libc6:arm64 (2.41-9ubuntu1) ... 310s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 310s Preparing to unpack .../libc-bin_2.41-9ubuntu1_arm64.deb ... 310s Unpacking libc-bin (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 310s Setting up libc-bin (2.41-9ubuntu1) ... 311s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 311s Preparing to unpack .../libsystemd0_257.7-1ubuntu1_arm64.deb ... 311s Unpacking libsystemd0:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 311s Setting up libsystemd0:arm64 (257.7-1ubuntu1) ... 311s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 311s Preparing to unpack .../libnss-systemd_257.7-1ubuntu1_arm64.deb ... 311s Unpacking libnss-systemd:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 311s Preparing to unpack .../systemd-sysv_257.7-1ubuntu1_arm64.deb ... 311s Unpacking systemd-sysv (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 311s Preparing to unpack .../systemd-resolved_257.7-1ubuntu1_arm64.deb ... 311s Unpacking systemd-resolved (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 311s Preparing to unpack .../libpam-systemd_257.7-1ubuntu1_arm64.deb ... 311s Unpacking libpam-systemd:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 311s Preparing to unpack .../libsystemd-shared_257.7-1ubuntu1_arm64.deb ... 311s Unpacking libsystemd-shared:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 311s Setting up libsystemd-shared:arm64 (257.7-1ubuntu1) ... 311s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 311s Preparing to unpack .../systemd_257.7-1ubuntu1_arm64.deb ... 311s Unpacking systemd (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 312s Preparing to unpack .../systemd-timesyncd_257.7-1ubuntu1_arm64.deb ... 312s Unpacking systemd-timesyncd (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 312s Preparing to unpack .../systemd-cryptsetup_257.7-1ubuntu1_arm64.deb ... 312s Unpacking systemd-cryptsetup (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 312s Preparing to unpack .../udev_257.7-1ubuntu1_arm64.deb ... 312s Unpacking udev (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 312s Preparing to unpack .../libudev1_257.7-1ubuntu1_arm64.deb ... 312s Unpacking libudev1:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 312s Setting up libudev1:arm64 (257.7-1ubuntu1) ... 312s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127318 files and directories currently installed.) 312s Preparing to unpack .../libapparmor1_4.1.1-0ubuntu3_arm64.deb ... 312s Unpacking libapparmor1:arm64 (4.1.1-0ubuntu3) over (4.1.1-0ubuntu2) ... 312s Preparing to unpack .../libxml2-16_2.14.4+dfsg-0exp1_arm64.deb ... 312s Unpacking libxml2-16:arm64 (2.14.4+dfsg-0exp1) over (2.14.3+dfsg-0exp3) ... 312s Preparing to unpack .../locales_2.41-9ubuntu1_all.deb ... 312s Unpacking locales (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 313s Preparing to unpack .../apparmor_4.1.1-0ubuntu3_arm64.deb ... 314s Unpacking apparmor (4.1.1-0ubuntu3) over (4.1.1-0ubuntu2) ... 314s Setting up libapparmor1:arm64 (4.1.1-0ubuntu3) ... 314s Setting up libxml2-16:arm64 (2.14.4+dfsg-0exp1) ... 314s Setting up systemd (257.7-1ubuntu1) ... 314s /usr/lib/tmpfiles.d/legacy.conf:14: Duplicate line for path "/run/lock", ignoring. 314s /usr/lib/tmpfiles.d/legacy.conf:14: Duplicate line for path "/run/lock", ignoring. 315s Setting up locales (2.41-9ubuntu1) ... 315s Generating locales (this might take a while)... 318s en_US.UTF-8... done 318s Generation complete. 318s Setting up apparmor (4.1.1-0ubuntu3) ... 318s Installing new version of config file /etc/apparmor.d/lsusb ... 318s Installing new version of config file /etc/apparmor.d/mbsync ... 319s Reloading AppArmor profiles 321s Setting up systemd-cryptsetup (257.7-1ubuntu1) ... 321s Setting up systemd-timesyncd (257.7-1ubuntu1) ... 321s systemd-time-wait-sync.service is a disabled or a static unit not running, not starting it. 322s Setting up udev (257.7-1ubuntu1) ... 322s Creating group 'clock' with GID 988. 322s Setting up libc-dev-bin (2.41-9ubuntu1) ... 322s Setting up systemd-resolved (257.7-1ubuntu1) ... 323s Setting up systemd-sysv (257.7-1ubuntu1) ... 323s Setting up libnss-systemd:arm64 (257.7-1ubuntu1) ... 323s Setting up libpam-systemd:arm64 (257.7-1ubuntu1) ... 323s Setting up libc6-dev:arm64 (2.41-9ubuntu1) ... 323s Processing triggers for libc-bin (2.41-9ubuntu1) ... 323s Processing triggers for man-db (2.13.1-1) ... 325s Processing triggers for dbus (1.16.2-2ubuntu1) ... 325s Processing triggers for shared-mime-info (2.4-5build2) ... 325s Processing triggers for procps (2:4.0.4-8ubuntu2) ... 325s Processing triggers for initramfs-tools (0.149ubuntu1) ... 325s update-initramfs: Generating /boot/initrd.img-6.15.0-3-generic 325s W: No lz4 in /usr/bin:/sbin:/bin, using gzip 327s W: Possible missing firmware /lib/firmware/apple/dfrmtfw-*.bin for built-in driver apple_z2 346s System running in EFI mode, skipping. 346s Reading package lists... 347s Building dependency tree... 347s Reading state information... 347s Solving dependencies... 347s 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. 347s autopkgtest [19:39:57]: rebooting testbed after setup commands that affected boot 384s Reading package lists... 384s Building dependency tree... 384s Reading state information... 384s Solving dependencies... 385s The following NEW packages will be installed: 385s libgfortran5 libggml libggml-blas libgomp1 libopenblas0-pthread 385s 0 upgraded, 5 newly installed, 0 to remove and 0 not upgraded. 385s Need to get 5378 kB of archives. 385s After this operation, 31.3 MB of additional disk space will be used. 385s Get:1 http://ftpmaster.internal/ubuntu questing/main arm64 libgfortran5 arm64 15.1.0-8ubuntu1 [451 kB] 385s Get:2 http://ftpmaster.internal/ubuntu questing/universe arm64 libopenblas0-pthread arm64 0.3.29+ds-3 [4189 kB] 385s Get:3 http://ftpmaster.internal/ubuntu questing/main arm64 libgomp1 arm64 15.1.0-8ubuntu1 [147 kB] 385s Get:4 http://ftpmaster.internal/ubuntu questing/universe arm64 libggml-blas arm64 0.0~git20250618.9e4bee1-1 [308 kB] 385s Get:5 http://ftpmaster.internal/ubuntu questing/universe arm64 libggml arm64 0.0~git20250618.9e4bee1-1 [282 kB] 386s Fetched 5378 kB in 1s (5861 kB/s) 386s Selecting previously unselected package libgfortran5:arm64. 386s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127320 files and directories currently installed.) 386s Preparing to unpack .../libgfortran5_15.1.0-8ubuntu1_arm64.deb ... 386s Unpacking libgfortran5:arm64 (15.1.0-8ubuntu1) ... 386s Selecting previously unselected package libopenblas0-pthread:arm64. 386s Preparing to unpack .../libopenblas0-pthread_0.3.29+ds-3_arm64.deb ... 386s Unpacking libopenblas0-pthread:arm64 (0.3.29+ds-3) ... 387s Selecting previously unselected package libgomp1:arm64. 387s Preparing to unpack .../libgomp1_15.1.0-8ubuntu1_arm64.deb ... 387s Unpacking libgomp1:arm64 (15.1.0-8ubuntu1) ... 387s Selecting previously unselected package libggml-blas:arm64. 387s Preparing to unpack .../libggml-blas_0.0~git20250618.9e4bee1-1_arm64.deb ... 387s Unpacking libggml-blas:arm64 (0.0~git20250618.9e4bee1-1) ... 387s Selecting previously unselected package libggml:arm64. 387s Preparing to unpack .../libggml_0.0~git20250618.9e4bee1-1_arm64.deb ... 387s Unpacking libggml:arm64 (0.0~git20250618.9e4bee1-1) ... 387s Setting up libgomp1:arm64 (15.1.0-8ubuntu1) ... 387s Setting up libgfortran5:arm64 (15.1.0-8ubuntu1) ... 387s Setting up libopenblas0-pthread:arm64 (0.3.29+ds-3) ... 387s update-alternatives: using /usr/lib/aarch64-linux-gnu/openblas-pthread/libblas.so.3 to provide /usr/lib/aarch64-linux-gnu/libblas.so.3 (libblas.so.3-aarch64-linux-gnu) in auto mode 387s update-alternatives: using /usr/lib/aarch64-linux-gnu/openblas-pthread/liblapack.so.3 to provide /usr/lib/aarch64-linux-gnu/liblapack.so.3 (liblapack.so.3-aarch64-linux-gnu) in auto mode 387s update-alternatives: using /usr/lib/aarch64-linux-gnu/openblas-pthread/libopenblas.so.0 to provide /usr/lib/aarch64-linux-gnu/libopenblas.so.0 (libopenblas.so.0-aarch64-linux-gnu) in auto mode 387s Setting up libggml-blas:arm64 (0.0~git20250618.9e4bee1-1) ... 387s Setting up libggml:arm64 (0.0~git20250618.9e4bee1-1) ... 387s Processing triggers for libc-bin (2.41-9ubuntu1) ... 390s autopkgtest [19:40:40]: test libggml-cpu: /usr/libexec/ggml/runtests 390s autopkgtest [19:40:40]: test libggml-cpu: [----------------------- 391s dpkg-architecture: warning: cannot determine CC system type, falling back to default (native compilation) 391s 0.00 1.00 2.00 391s Testing 2 devices 391s 391s Backend 1/2: BLAS 391s Device description: BLAS 391s Device memory: 0 MB (0 MB free) 391s 391s ABS(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s ABS(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s SGN(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s SGN(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s NEG(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s NEG(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s STEP(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s STEP(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s TANH(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s TANH(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s ELU(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s ELU(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s RELU(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s RELU(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s SIGMOID(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s SIGMOID(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s GELU(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s GELU(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s GELU_QUICK(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s GELU_QUICK(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s SILU(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s SILU(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s HARDSWISH(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s HARDSWISH(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s HARDSIGMOID(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s HARDSIGMOID(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s EXP(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s EXP(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s GELU_ERF(type=f16,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s GELU_ERF(type=f16,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s ABS(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s ABS(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s SGN(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s SGN(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s NEG(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s NEG(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s STEP(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s STEP(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s TANH(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s TANH(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s ELU(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s ELU(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s RELU(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s RELU(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s SIGMOID(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s SIGMOID(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s GELU(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s GELU(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s GELU_QUICK(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s GELU_QUICK(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s SILU(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s SILU(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s HARDSWISH(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s HARDSWISH(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s HARDSIGMOID(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s HARDSIGMOID(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s EXP(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s EXP(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s GELU_ERF(type=f16,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s GELU_ERF(type=f16,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s ABS(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s ABS(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s SGN(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s SGN(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s NEG(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s NEG(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s STEP(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s STEP(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s TANH(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s TANH(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s ELU(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s ELU(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s RELU(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s RELU(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s SIGMOID(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s SIGMOID(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s GELU(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s GELU(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s GELU_QUICK(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s GELU_QUICK(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s SILU(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s SILU(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s HARDSWISH(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s HARDSWISH(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s HARDSIGMOID(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s HARDSIGMOID(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s EXP(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s EXP(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s GELU_ERF(type=f32,ne_a=[128,2,2,2],v=0): not supported [BLAS] 391s GELU_ERF(type=f32,ne_a=[5,7,11,13],v=0): not supported [BLAS] 391s ABS(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s ABS(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s SGN(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s SGN(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s NEG(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s NEG(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s STEP(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s STEP(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s TANH(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s TANH(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s ELU(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s ELU(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s RELU(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s RELU(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s SIGMOID(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s SIGMOID(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s GELU(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s GELU(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s GELU_QUICK(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s GELU_QUICK(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s SILU(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s SILU(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s HARDSWISH(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s HARDSWISH(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s HARDSIGMOID(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s HARDSIGMOID(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s EXP(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s EXP(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s GELU_ERF(type=f32,ne_a=[128,2,2,2],v=1): not supported [BLAS] 391s GELU_ERF(type=f32,ne_a=[5,7,11,13],v=1): not supported [BLAS] 391s GET_ROWS(type=f32,n=1,m=8,r=2,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=f32,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=f32,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=f32,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=f32,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=f16,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=f16,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=f16,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=f16,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=bf16,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=bf16,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=bf16,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=bf16,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q4_0,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q4_0,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q4_0,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q4_0,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q4_1,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q4_1,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q4_1,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q4_1,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q5_0,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q5_0,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q5_0,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q5_0,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q5_1,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q5_1,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q5_1,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q5_1,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q8_0,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q8_0,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q8_0,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q8_0,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q2_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q2_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q2_K,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q2_K,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q3_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q3_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q3_K,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q3_K,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q4_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q4_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q4_K,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q4_K,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q5_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q5_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q5_K,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q5_K,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=q6_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=q6_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=q6_K,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=q6_K,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=iq2_xxs,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=iq2_xxs,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=iq2_xxs,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=iq2_xxs,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=iq2_xs,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=iq2_xs,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=iq2_xs,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=iq2_xs,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=iq2_s,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=iq2_s,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=iq2_s,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=iq2_s,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=iq3_xxs,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=iq3_xxs,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=iq3_xxs,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=iq3_xxs,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=iq1_s,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=iq1_s,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=iq1_s,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=iq1_s,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=iq1_m,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=iq1_m,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=iq1_m,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=iq1_m,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=iq4_nl,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=iq4_nl,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=iq4_nl,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=iq4_nl,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=iq3_s,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=iq3_s,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=iq3_s,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=iq3_s,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=iq4_xs,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=iq4_xs,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=iq4_xs,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=iq4_xs,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS(type=i32,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS(type=i32,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS(type=i32,n=256,m=5,r=4,b=7,v=0): not supported [BLAS] 391s GET_ROWS(type=i32,n=256,m=5,r=4,b=7,v=1): not supported [BLAS] 391s GET_ROWS_BACK(type=f32,n=1,m=8,r=2,b=1,v=0): not supported [BLAS] 391s GET_ROWS_BACK(type=f32,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS_BACK(type=f32,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS_BACK(type=f16,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] 391s GET_ROWS_BACK(type=f16,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] 391s GET_ROWS_BACK(type=bf16,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=bf16,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q4_0,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q4_0,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q4_1,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q4_1,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q5_0,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q5_0,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q5_1,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q5_1,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q8_0,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q8_0,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q2_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q2_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q3_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q3_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q4_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q4_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q5_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q5_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q6_K,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=q6_K,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq2_xxs,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq2_xxs,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq2_xs,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq2_xs,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq2_s,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq2_s,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq3_xxs,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq3_xxs,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq1_s,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq1_s,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq1_m,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq1_m,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq4_nl,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq4_nl,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq3_s,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq3_s,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq4_xs,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=iq4_xs,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=i32,n=256,m=5,r=4,b=1,v=0): not supported [BLAS] not supported [CPU] 391s GET_ROWS_BACK(type=i32,n=256,m=5,r=4,b=1,v=1): not supported [BLAS] not supported [CPU] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=avg,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=1,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=1,s0=2,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=1,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=1,k1=3,s0=2,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=1,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=1,s0=2,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=1,s1=2,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=1,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=1,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=1,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=1,p0=1,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=2,p0=0,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=2,p0=0,p1=1): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=2,p0=1,p1=0): not supported [BLAS] 391s POOL_2D(pool_type=max,type_input=f32,ne_input=[10,10,3,1],k0=3,k1=3,s0=2,s1=2,p0=1,p1=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[3000,128,1,1],ne_kernel=[3,128,1280,1],s0=1,s1=0,p0=1,p1=0,d0=1,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f32,ne_input=[3000,128,1,1],ne_kernel=[3,128,1280,1],s0=1,s1=0,p0=1,p1=0,d0=1,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[3000,128,1,1],ne_kernel=[3,128,1280,1],s0=1,s1=0,p0=1,p1=0,d0=1,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,2,2,1],ne_kernel=[3,2,2,1],s0=1,s1=0,p0=0,p1=0,d0=1,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,2,2,1],ne_kernel=[3,2,2,1],s0=1,s1=0,p0=0,p1=0,d0=3,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,2,2,1],ne_kernel=[3,2,2,1],s0=1,s1=0,p0=3,p1=0,d0=1,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,2,2,1],ne_kernel=[3,2,2,1],s0=1,s1=0,p0=3,p1=0,d0=3,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,2,2,1],ne_kernel=[3,2,2,1],s0=3,s1=0,p0=0,p1=0,d0=1,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,2,2,1],ne_kernel=[3,2,2,1],s0=3,s1=0,p0=0,p1=0,d0=3,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,2,2,1],ne_kernel=[3,2,2,1],s0=3,s1=0,p0=3,p1=0,d0=1,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,2,2,1],ne_kernel=[3,2,2,1],s0=3,s1=0,p0=3,p1=0,d0=3,d1=0,is_2D=0): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[10,10,3,1],ne_kernel=[3,3,3,1],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f32,ne_input=[10,10,3,1],ne_kernel=[3,3,3,1],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[10,10,3,1],ne_kernel=[3,3,3,1],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=0,p1=0,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=0,p1=0,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=0,p1=0,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=0,p1=0,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=0,p1=3,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=0,p1=3,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=0,p1=3,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=0,p1=3,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=3,p1=0,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=3,p1=0,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=3,p1=0,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=3,p1=0,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=3,p1=3,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=3,p1=3,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=3,p1=3,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=1,p0=3,p1=3,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=0,p1=0,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=0,p1=0,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=0,p1=0,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=0,p1=0,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=0,p1=3,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=0,p1=3,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=0,p1=3,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=0,p1=3,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=3,p1=0,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=3,p1=0,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=3,p1=0,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=3,p1=0,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=3,p1=3,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=3,p1=3,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=3,p1=3,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=1,s1=3,p0=3,p1=3,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=0,p1=0,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=0,p1=0,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=0,p1=0,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=0,p1=0,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=0,p1=3,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=0,p1=3,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=0,p1=3,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=0,p1=3,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=3,p1=0,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=3,p1=0,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=3,p1=0,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=3,p1=0,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=3,p1=3,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=3,p1=3,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=3,p1=3,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=1,p0=3,p1=3,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=0,p1=0,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=0,p1=0,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=0,p1=0,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=0,p1=0,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=0,p1=3,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=0,p1=3,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=0,p1=3,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=0,p1=3,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=3,p1=0,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=3,p1=0,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=3,p1=0,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=3,p1=0,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=3,p1=3,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=3,p1=3,d0=1,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=3,p1=3,d0=3,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f32,dst_type=f32,ne_input=[20,20,2,2],ne_kernel=[3,3,2,2],s0=3,s1=3,p0=3,p1=3,d0=3,d1=3,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[12,12,1,32],ne_kernel=[3,3,1,32],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[12,12,2,32],ne_kernel=[3,3,2,32],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[12,12,1,1024],ne_kernel=[3,3,1,1024],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[12,12,2,1024],ne_kernel=[3,3,2,1024],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[12,12,1,2048],ne_kernel=[3,3,1,2048],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[12,12,2,2048],ne_kernel=[3,3,2,2048],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[12,12,1,2560],ne_kernel=[3,3,1,2560],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s IM2COL(type_input=f32,type_kernel=f16,dst_type=f16,ne_input=[12,12,2,2560],ne_kernel=[3,3,2,2560],s0=1,s1=1,p0=1,p1=1,d0=1,d1=1,is_2D=1): not supported [BLAS] 391s CONV_2D_DW(ne_input=[17,34,9,1],ne_kernel=[3,3,1,9],stride=1,padding=0,dilation=1,cwhn=0): not supported [BLAS] 391s CONV_2D_DW(ne_input=[17,34,9,1],ne_kernel=[3,3,1,9],stride=1,padding=0,dilation=1,cwhn=1): not supported [BLAS] 391s CONV_2D_DW(ne_input=[32,8,64,1],ne_kernel=[3,3,1,64],stride=2,padding=1,dilation=1,cwhn=0): not supported [BLAS] 391s CONV_2D_DW(ne_input=[32,8,64,1],ne_kernel=[3,3,1,64],stride=2,padding=1,dilation=1,cwhn=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1,1,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1,1,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1,1,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1,1,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1,1,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1,1,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[3,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[3,1,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[3,1,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[3,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[3,1,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[3,1,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[3,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[3,1,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[3,1,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1337,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1337,1,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1337,1,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1337,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1337,1,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1337,1,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1337,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1337,1,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1337,1,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1,1,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1,1,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1,1,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1,1,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1,1,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1,1,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1,1,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1,1,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1,1,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[3,1,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[3,1,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[3,1,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[3,1,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[3,1,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[3,1,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[3,1,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[3,1,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[3,1,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1337,1,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1337,1,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1337,1,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1337,1,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1337,1,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1337,1,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1337,1,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1337,1,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1337,1,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1,9,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1,9,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1,9,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1,9,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1,9,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1,9,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1,9,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1,9,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1,9,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[3,9,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[3,9,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[3,9,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[3,9,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[3,9,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[3,9,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[3,9,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[3,9,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[3,9,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1337,9,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1337,9,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,1,1,1],ne_kernel=[1337,9,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1337,9,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1337,9,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[1337,9,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1337,9,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1337,9,1,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,1,1,1],ne_kernel=[1337,9,1,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1,9,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1,9,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1,9,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1,9,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1,9,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1,9,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1,9,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1,9,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1,9,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[3,9,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[3,9,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[3,9,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[3,9,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[3,9,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[3,9,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[3,9,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[3,9,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[3,9,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1337,9,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1337,9,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[1,7,1,1],ne_kernel=[1337,9,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1337,9,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1337,9,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,7,1,1],ne_kernel=[1337,9,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1337,9,7,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1337,9,7,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[13,7,1,1],ne_kernel=[1337,9,7,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[197,32,1,1],ne_kernel=[16,32,32,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[3,2,1,1],ne_kernel=[2,3,2,1],s0=3,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[3,2,1,1],ne_kernel=[2,3,2,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[3,2,1,1],ne_kernel=[2,3,2,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[3,2,1,1],ne_kernel=[3,2,2,1],s0=2,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[3,2,1,1],ne_kernel=[3,2,2,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[3,2,1,1],ne_kernel=[3,1,2,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s CONV_TRANSPOSE_1D(ne_input=[2,1,1,1],ne_kernel=[3,1,1,1],s0=1,p0=0,d0=1): not supported [BLAS] 391s COUNT_EQUAL(type=f32,ne=[4,500,1,1]): not supported [BLAS] 391s COUNT_EQUAL(type=f32,ne=[4,5000,1,1]): not supported [BLAS] 391s ARGMAX(type=f32,ne=[32,1,1,1]): not supported [BLAS] 391s ARGMAX(type=f32,ne=[100,10,1,1]): not supported [BLAS] 391s ARGMAX(type=f32,ne=[1024,10,1,1]): not supported [BLAS] 391s ARGMAX(type=f32,ne=[1024,12,1,1]): not supported [BLAS] 391s ARGMAX(type=f32,ne=[2000,10,1,1]): not supported [BLAS] 391s ARGMAX(type=f32,ne=[5438,3,1,1]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,1],nr=[1,1,1,1]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,1],nr=[2,1,1,1]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,1],nr=[1,2,1,1]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,1],nr=[1,1,2,1]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,1],nr=[1,1,1,2]): not supported [BLAS] 391s REPEAT(type=i32,ne=[10,5,4,1],nr=[2,1,1,1]): not supported [BLAS] 391s REPEAT(type=i16,ne=[10,5,4,1],nr=[1,1,1,2]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,3],nr=[1,1,1,1]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,3],nr=[1,2,1,1]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,3],nr=[1,1,2,1]): not supported [BLAS] 391s REPEAT(type=f32,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s REPEAT(type=i32,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s REPEAT(type=i16,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[1,1,1,1],v=0): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[2,1,1,1],v=0): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[1,2,1,1],v=0): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[1,1,2,1],v=0): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[1,1,1,2],v=0): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[1,1,1,1],v=1): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[2,1,1,1],v=1): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[1,2,1,1],v=1): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[1,1,2,1],v=1): not supported [BLAS] 391s REPEAT_BACK(type=f32,ne=[8,6,4,2],nr=[1,1,1,2],v=1): not supported [BLAS] 391s DUP(type=f32,ne=[10,10,20,1]): not supported [BLAS] 391s DUP(type=f16,ne=[10,10,20,1]): not supported [BLAS] 391s DUP(type=i32,ne=[10,10,20,1]): not supported [BLAS] 391s DUP(type=i16,ne=[10,10,20,1]): not supported [BLAS] 391s DUP(type=f32,ne=[10,10,5,1],permute=[0,2,1,3]): not supported [BLAS] 391s DUP(type=f16,ne=[10,10,5,1],permute=[0,2,1,3]): not supported [BLAS] 391s DUP(type=f32,ne=[10,10,5,1],permute=[1,0,2,3]): not supported [BLAS] 391s DUP(type=f16,ne=[10,10,5,1],permute=[1,0,2,3]): not supported [BLAS] 391s DUP(type=i16,ne=[10,8,3,1],permute=[0,2,1,3]): not supported [BLAS] 391s DUP(type=i16,ne=[10,8,3,1],permute=[1,2,0,3]): not supported [BLAS] 391s SET(type_src=f32,type_dst=f32,ne=[6,5,4,3],dim=1): not supported [BLAS] 391s SET(type_src=f32,type_dst=f32,ne=[6,5,4,3],dim=2): not supported [BLAS] 391s SET(type_src=f32,type_dst=f32,ne=[6,5,4,3],dim=3): not supported [BLAS] 391s SET(type_src=i32,type_dst=i32,ne=[6,5,4,3],dim=1): not supported [BLAS] 391s SET(type_src=i32,type_dst=i32,ne=[6,5,4,3],dim=2): not supported [BLAS] 391s SET(type_src=i32,type_dst=i32,ne=[6,5,4,3],dim=3): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[1,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[1,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[1,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[2,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[2,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[2,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[3,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[3,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[3,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[1,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[1,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[1,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[2,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[2,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[2,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[3,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[3,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[3,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[1,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[1,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[1,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[2,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[2,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[2,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[3,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[3,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[3,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=q4_0,ne=[32,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=q4_0,ne=[32,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=q4_0,ne=[32,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=q4_0,ne=[64,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=q4_0,ne=[64,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=q4_0,ne=[64,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=q4_0,ne=[96,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=q4_0,ne=[96,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=q4_0,ne=[96,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=q4_1,ne=[32,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=q4_1,ne=[32,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=q4_1,ne=[32,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=q4_1,ne=[64,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=q4_1,ne=[64,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=q4_1,ne=[64,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=q4_1,ne=[96,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=q4_1,ne=[96,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=q4_1,ne=[96,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=q5_0,ne=[32,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=q5_0,ne=[32,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=q5_0,ne=[32,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=q5_0,ne=[64,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=q5_0,ne=[64,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=q5_0,ne=[64,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=q5_0,ne=[96,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=q5_0,ne=[96,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=q5_0,ne=[96,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=q5_1,ne=[32,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=q5_1,ne=[32,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=q5_1,ne=[32,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=q5_1,ne=[64,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=q5_1,ne=[64,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=q5_1,ne=[64,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=q5_1,ne=[96,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=q5_1,ne=[96,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=q5_1,ne=[96,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=q8_0,ne=[32,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=q8_0,ne=[32,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=q8_0,ne=[32,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=q8_0,ne=[64,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=q8_0,ne=[64,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=q8_0,ne=[64,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=q8_0,ne=[96,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=q8_0,ne=[96,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=q8_0,ne=[96,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=q2_K,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=q2_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=q2_K,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=q2_K,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=q2_K,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=q2_K,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=q2_K,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=q2_K,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=q2_K,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=q3_K,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=q3_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=q3_K,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=q3_K,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=q3_K,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=q3_K,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=q3_K,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=q3_K,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=q3_K,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=q4_K,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=q4_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=q4_K,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=q4_K,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=q4_K,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=q4_K,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=q4_K,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=q4_K,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=q4_K,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=q5_K,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=q5_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=q5_K,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=q5_K,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=q5_K,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=q5_K,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=q5_K,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=q5_K,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=q5_K,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=q6_K,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=q6_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=q6_K,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=q6_K,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=q6_K,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=q6_K,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=q6_K,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=q6_K,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=q6_K,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=iq2_xxs,type_dst=iq2_xxs,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xxs,type_dst=iq2_xxs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xxs,type_dst=iq2_xxs,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xxs,type_dst=iq2_xxs,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xxs,type_dst=iq2_xxs,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xxs,type_dst=iq2_xxs,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xxs,type_dst=iq2_xxs,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xxs,type_dst=iq2_xxs,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xxs,type_dst=iq2_xxs,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xs,type_dst=iq2_xs,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xs,type_dst=iq2_xs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xs,type_dst=iq2_xs,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xs,type_dst=iq2_xs,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xs,type_dst=iq2_xs,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xs,type_dst=iq2_xs,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xs,type_dst=iq2_xs,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xs,type_dst=iq2_xs,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_xs,type_dst=iq2_xs,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_s,type_dst=iq2_s,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_s,type_dst=iq2_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_s,type_dst=iq2_s,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_s,type_dst=iq2_s,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_s,type_dst=iq2_s,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_s,type_dst=iq2_s,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_s,type_dst=iq2_s,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_s,type_dst=iq2_s,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq2_s,type_dst=iq2_s,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_xxs,type_dst=iq3_xxs,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_xxs,type_dst=iq3_xxs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_xxs,type_dst=iq3_xxs,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_xxs,type_dst=iq3_xxs,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_xxs,type_dst=iq3_xxs,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_xxs,type_dst=iq3_xxs,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_xxs,type_dst=iq3_xxs,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_xxs,type_dst=iq3_xxs,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_xxs,type_dst=iq3_xxs,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_s,type_dst=iq1_s,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_s,type_dst=iq1_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_s,type_dst=iq1_s,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_s,type_dst=iq1_s,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_s,type_dst=iq1_s,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_s,type_dst=iq1_s,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_s,type_dst=iq1_s,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_s,type_dst=iq1_s,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_s,type_dst=iq1_s,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_m,type_dst=iq1_m,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_m,type_dst=iq1_m,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_m,type_dst=iq1_m,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_m,type_dst=iq1_m,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_m,type_dst=iq1_m,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_m,type_dst=iq1_m,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_m,type_dst=iq1_m,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_m,type_dst=iq1_m,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq1_m,type_dst=iq1_m,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq4_nl,type_dst=iq4_nl,ne=[32,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=iq4_nl,ne=[32,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=iq4_nl,ne=[32,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=iq4_nl,ne=[64,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=iq4_nl,ne=[64,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=iq4_nl,ne=[64,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=iq4_nl,ne=[96,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=iq4_nl,ne=[96,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=iq4_nl,ne=[96,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=iq3_s,type_dst=iq3_s,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_s,type_dst=iq3_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_s,type_dst=iq3_s,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_s,type_dst=iq3_s,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_s,type_dst=iq3_s,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_s,type_dst=iq3_s,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_s,type_dst=iq3_s,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_s,type_dst=iq3_s,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq3_s,type_dst=iq3_s,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=iq4_xs,type_dst=iq4_xs,ne=[256,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=iq4_xs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=iq4_xs,ne=[256,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=iq4_xs,ne=[512,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=iq4_xs,ne=[512,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=iq4_xs,ne=[512,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=iq4_xs,ne=[768,2,3,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=iq4_xs,ne=[768,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=iq4_xs,ne=[768,2,3,4],permute_src=[0,3,1,2],permute_dst=[0,2,1,3]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=bf16,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=bf16,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q4_0,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q4_0,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q4_1,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q4_1,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q5_0,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q5_0,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q5_1,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q5_1,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q8_0,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q8_0,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q2_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q2_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q3_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q3_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q4_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q4_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q5_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q5_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q6_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=q6_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=iq2_xxs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq2_xxs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq2_xs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq2_xs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq2_s,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq2_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq3_xxs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq3_xxs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq1_s,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq1_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq1_m,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq1_m,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq4_nl,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=iq4_nl,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=iq3_s,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq3_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f16,type_dst=iq4_xs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=iq4_xs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=f16,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=f16,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=bf16,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q4_0,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q4_0,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q4_1,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q4_1,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q5_0,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q5_0,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q5_1,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q5_1,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q8_0,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q8_0,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q2_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q2_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q3_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q3_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q4_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q4_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q5_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q5_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q6_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=q6_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=iq2_xxs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq2_xxs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq2_xs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq2_xs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq2_s,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq2_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq3_xxs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq3_xxs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq1_s,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq1_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq1_m,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq1_m,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq4_nl,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=iq4_nl,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=iq3_s,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq3_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=bf16,type_dst=iq4_xs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=iq4_xs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f16,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f16,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=bf16,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=bf16,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q4_0,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q4_0,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q4_1,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q4_1,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q5_0,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q5_0,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q5_1,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q5_1,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q8_0,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q8_0,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q2_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q2_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q3_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q3_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q4_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q4_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q5_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q5_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q6_K,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=q6_K,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=iq2_xxs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq2_xxs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq2_xs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq2_xs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq2_s,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq2_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq3_xxs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq3_xxs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq1_s,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq1_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq1_m,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq1_m,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq4_nl,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=iq4_nl,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=iq3_s,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq3_s,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] not supported [CPU] 391s CPY(type_src=f32,type_dst=iq4_xs,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=iq4_xs,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=bf16,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_0,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_1,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_0,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_1,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q8_0,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q2_K,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q3_K,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q4_K,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q5_K,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=q6_K,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq2_xxs,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq2_xxs,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq2_xs,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq2_xs,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq2_s,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq2_s,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq3_xxs,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq3_xxs,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq1_s,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq1_s,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq1_m,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq1_m,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_nl,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq3_s,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq3_s,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=f32,ne=[256,4,4,4],permute_src=[0,0,0,0],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=iq4_xs,type_dst=f32,ne=[256,2,3,4],permute_src=[0,2,1,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f16,ne=[256,2,3,4],permute_src=[1,0,2,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f16,type_dst=f32,ne=[256,2,3,4],permute_src=[1,0,2,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f16,ne=[256,2,3,4],permute_src=[1,0,2,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CPY(type_src=f32,type_dst=f32,ne=[256,2,3,4],permute_src=[1,0,2,3],permute_dst=[0,0,0,0]): not supported [BLAS] 391s CONT(type=f32,ne=[10,10,10,1]): not supported [BLAS] 391s CONT(type=f32,ne=[2,1,1,1]): not supported [BLAS] 391s CONT(type=f32,ne=[2,1,3,5]): not supported [BLAS] 391s CONT(type=f32,ne=[2,3,5,7]): not supported [BLAS] 391s CONT(type=f16,ne=[2,1,1,1]): not supported [BLAS] 391s CONT(type=f16,ne=[2,1,3,5]): not supported [BLAS] 391s CONT(type=f16,ne=[2,3,5,7]): not supported [BLAS] 391s CONT(type=bf16,ne=[2,1,1,1]): not supported [BLAS] 391s CONT(type=bf16,ne=[2,1,3,5]): not supported [BLAS] 391s CONT(type=bf16,ne=[2,3,5,7]): not supported [BLAS] 391s ADD(type=f16,ne=[1,1,8,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1,1,8,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1,1,8,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1,1,8,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1,1,1,1],nr=[32,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1,1,1,1],nr=[32,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1,1,1,1],nr=[32,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1,1,1,1],nr=[32,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1,1,320,320],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1,1,320,320],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1,1,320,320],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1,1,320,320],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,4,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,4,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,4,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,4,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,4,3],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,4,3],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,4,3],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,4,3],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,4,3],nr=[1,2,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,4,3],nr=[1,2,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,4,3],nr=[1,2,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,4,3],nr=[1,2,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,4,3],nr=[1,1,2,1]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,4,3],nr=[1,1,2,1]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,4,3],nr=[1,1,2,1]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,4,3],nr=[1,1,2,1]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,4,3],nr=[1,1,2,2]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,4,3],nr=[1,1,2,2]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,4,3],nr=[1,1,2,2]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,4,3],nr=[1,1,2,2]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,4,3],nr=[1,2,2,2]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,4,3],nr=[1,2,2,2]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,4,3],nr=[1,2,2,2]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,4,3],nr=[1,2,2,2]): not supported [BLAS] 391s ADD(type=f16,ne=[10,5,4,3],nr=[2,2,2,2]): not supported [BLAS] 391s SUB(type=f16,ne=[10,5,4,3],nr=[2,2,2,2]): not supported [BLAS] 391s MUL(type=f16,ne=[10,5,4,3],nr=[2,2,2,2]): not supported [BLAS] 391s DIV(type=f16,ne=[10,5,4,3],nr=[2,2,2,2]): not supported [BLAS] 391s ADD(type=f16,ne=[1280,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1280,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1280,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1280,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1280,1,1,1],nr=[1,16,16,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1280,1,1,1],nr=[1,16,16,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1280,1,1,1],nr=[1,16,16,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1280,1,1,1],nr=[1,16,16,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1280,16,16,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1280,16,16,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1280,16,16,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1280,16,16,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1280,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1280,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1280,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1280,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1,1,1280,1],nr=[16,16,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1,1,1280,1],nr=[16,16,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1,1,1280,1],nr=[16,16,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1,1,1280,1],nr=[16,16,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[16,16,1280,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[16,16,1280,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[16,16,1280,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[16,16,1280,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1,1,1920,1],nr=[16,16,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1,1,1920,1],nr=[16,16,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1,1,1920,1],nr=[16,16,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1,1,1920,1],nr=[16,16,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1,1,2560,1],nr=[16,16,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1,1,2560,1],nr=[16,16,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1,1,2560,1],nr=[16,16,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1,1,2560,1],nr=[16,16,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1,1,1280,1],nr=[32,32,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1,1,1280,1],nr=[32,32,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1,1,1280,1],nr=[32,32,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1,1,1280,1],nr=[32,32,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1,1,1920,1],nr=[32,32,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1,1,1920,1],nr=[32,32,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1,1,1920,1],nr=[32,32,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1,1,1920,1],nr=[32,32,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[1,1,640,1],nr=[32,32,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[1,1,640,1],nr=[32,32,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[1,1,640,1],nr=[32,32,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[1,1,640,1],nr=[32,32,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[5120,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[5120,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[5120,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[5120,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s ADD(type=f16,ne=[640,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f16,ne=[640,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f16,ne=[640,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f16,ne=[640,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1,1,8,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1,1,8,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1,1,8,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1,1,8,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1,1,1,1],nr=[32,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1,1,1,1],nr=[32,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1,1,1,1],nr=[32,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1,1,1,1],nr=[32,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1,1,320,320],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1,1,320,320],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1,1,320,320],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1,1,320,320],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,4,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,4,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,4,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,4,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,4,3],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,4,3],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,4,3],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,4,3],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,4,3],nr=[2,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,4,3],nr=[1,2,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,4,3],nr=[1,2,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,4,3],nr=[1,2,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,4,3],nr=[1,2,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,4,3],nr=[1,1,2,1]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,4,3],nr=[1,1,2,1]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,4,3],nr=[1,1,2,1]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,4,3],nr=[1,1,2,1]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,4,3],nr=[1,1,1,2]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,4,3],nr=[1,1,2,2]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,4,3],nr=[1,1,2,2]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,4,3],nr=[1,1,2,2]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,4,3],nr=[1,1,2,2]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,4,3],nr=[1,2,2,2]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,4,3],nr=[1,2,2,2]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,4,3],nr=[1,2,2,2]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,4,3],nr=[1,2,2,2]): not supported [BLAS] 391s ADD(type=f32,ne=[10,5,4,3],nr=[2,2,2,2]): not supported [BLAS] 391s SUB(type=f32,ne=[10,5,4,3],nr=[2,2,2,2]): not supported [BLAS] 391s MUL(type=f32,ne=[10,5,4,3],nr=[2,2,2,2]): not supported [BLAS] 391s DIV(type=f32,ne=[10,5,4,3],nr=[2,2,2,2]): not supported [BLAS] 391s ADD(type=f32,ne=[1280,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1280,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1280,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1280,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1280,1,1,1],nr=[1,16,16,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1280,1,1,1],nr=[1,16,16,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1280,1,1,1],nr=[1,16,16,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1280,1,1,1],nr=[1,16,16,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1280,16,16,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1280,16,16,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1280,16,16,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1280,16,16,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1280,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1280,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1280,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1280,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1,1,1280,1],nr=[16,16,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1,1,1280,1],nr=[16,16,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1,1,1280,1],nr=[16,16,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1,1,1280,1],nr=[16,16,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[16,16,1280,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[16,16,1280,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[16,16,1280,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[16,16,1280,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1,1,1920,1],nr=[16,16,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1,1,1920,1],nr=[16,16,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1,1,1920,1],nr=[16,16,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1,1,1920,1],nr=[16,16,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1,1,2560,1],nr=[16,16,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1,1,2560,1],nr=[16,16,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1,1,2560,1],nr=[16,16,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1,1,2560,1],nr=[16,16,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1,1,1280,1],nr=[32,32,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1,1,1280,1],nr=[32,32,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1,1,1280,1],nr=[32,32,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1,1,1280,1],nr=[32,32,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1,1,1920,1],nr=[32,32,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1,1,1920,1],nr=[32,32,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1,1,1920,1],nr=[32,32,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1,1,1920,1],nr=[32,32,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[1,1,640,1],nr=[32,32,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[1,1,640,1],nr=[32,32,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[1,1,640,1],nr=[32,32,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[1,1,640,1],nr=[32,32,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[5120,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[5120,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[5120,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[5120,1,1,1],nr=[1,256,1,1]): not supported [BLAS] 391s ADD(type=f32,ne=[640,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s SUB(type=f32,ne=[640,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s MUL(type=f32,ne=[640,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s DIV(type=f32,ne=[640,1,1,1],nr=[1,1,1,1]): not supported [BLAS] 391s ADD1(type=f32,ne=[10,5,4,3]): not supported [BLAS] 391s SCALE(type=f32,ne=[10,10,10,10],scale=2.000000): not supported [BLAS] 391s SILU_BACK(type=f32,ne=[64,5,4,3],eps=0.000001): not supported [BLAS] 391s NORM(type=f32,ne=[64,5,4,3],v=0,eps=0.000000): not supported [BLAS] 391s RMS_NORM(type=f32,ne=[64,5,4,3],v=0,eps=0.000000): not supported [BLAS] 391s NORM(type=f32,ne=[64,5,4,3],v=1,eps=0.000000): not supported [BLAS] 391s RMS_NORM(type=f32,ne=[64,5,4,3],v=1,eps=0.000000): not supported [BLAS] 391s RMS_NORM_BACK(type=f32,ne=[64,5,4,3],eps=0.000000): not supported [BLAS] 391s L2_NORM(type=f32,ne=[64,5,4,3]): not supported [BLAS] 391s NORM(type=f32,ne=[64,5,4,3],v=0,eps=0.000001): not supported [BLAS] 391s RMS_NORM(type=f32,ne=[64,5,4,3],v=0,eps=0.000001): not supported [BLAS] 391s NORM(type=f32,ne=[64,5,4,3],v=1,eps=0.000001): not supported [BLAS] 391s RMS_NORM(type=f32,ne=[64,5,4,3],v=1,eps=0.000001): not supported [BLAS] 391s RMS_NORM_BACK(type=f32,ne=[64,5,4,3],eps=0.000001): not supported [BLAS] 391s L2_NORM(type=f32,ne=[64,5,4,3]): not supported [BLAS] 391s NORM(type=f32,ne=[64,5,4,3],v=0,eps=0.000100): not supported [BLAS] 391s RMS_NORM(type=f32,ne=[64,5,4,3],v=0,eps=0.000100): not supported [BLAS] 391s NORM(type=f32,ne=[64,5,4,3],v=1,eps=0.000100): not supported [BLAS] 391s RMS_NORM(type=f32,ne=[64,5,4,3],v=1,eps=0.000100): not supported [BLAS] 391s RMS_NORM_BACK(type=f32,ne=[64,5,4,3],eps=0.000100): not supported [BLAS] 391s L2_NORM(type=f32,ne=[64,5,4,3]): not supported [BLAS] 391s NORM(type=f32,ne=[64,5,4,3],v=0,eps=0.100000): not supported [BLAS] 391s RMS_NORM(type=f32,ne=[64,5,4,3],v=0,eps=0.100000): not supported [BLAS] 391s NORM(type=f32,ne=[64,5,4,3],v=1,eps=0.100000): not supported [BLAS] 391s RMS_NORM(type=f32,ne=[64,5,4,3],v=1,eps=0.100000): not supported [BLAS] 391s RMS_NORM_BACK(type=f32,ne=[64,5,4,3],eps=0.100000): not supported [BLAS] 391s L2_NORM(type=f32,ne=[64,5,4,3]): not supported [BLAS] 391s L2_NORM(type=f32,ne=[64,5,4,3]): not supported [BLAS] 391s SSM_CONV(type=f32,ne_a=[4,1536,1,1],ne_b=[4,1536,1,1]): not supported [BLAS] 391s SSM_CONV(type=f32,ne_a=[8,1536,1,1],ne_b=[4,1536,1,1]): not supported [BLAS] 391s SSM_CONV(type=f32,ne_a=[4,1536,4,1],ne_b=[4,1536,1,1]): not supported [BLAS] 391s SSM_SCAN(type=f32,d_state=16,d_inner=1024,n_seq_tokens=32,n_seqs=4): not supported [BLAS] 391s RWKV_WKV6(type=f32,head_count=32,head_size=64,n_seq_tokens=1,n_seqs=1): not supported [BLAS] 391s RWKV_WKV6(type=f32,head_count=32,head_size=64,n_seq_tokens=32,n_seqs=1): not supported [BLAS] 391s RWKV_WKV6(type=f32,head_count=32,head_size=64,n_seq_tokens=32,n_seqs=4): not supported [BLAS] 391s RWKV_WKV6(type=f32,head_count=32,head_size=64,n_seq_tokens=128,n_seqs=4): not supported [BLAS] 391s RWKV_WKV7(type=f32,head_count=32,head_size=64,n_seq_tokens=1,n_seqs=1): not supported [BLAS] 391s RWKV_WKV7(type=f32,head_count=32,head_size=64,n_seq_tokens=32,n_seqs=1): not supported [BLAS] 391s RWKV_WKV7(type=f32,head_count=32,head_size=64,n_seq_tokens=32,n_seqs=4): not supported [BLAS] 391s RWKV_WKV7(type=f32,head_count=32,head_size=64,n_seq_tokens=128,n_seqs=4): not supported [BLAS] 391s GATED_LINEAR_ATTN(type=f32,head_count=32,head_size=64,n_seq_tokens=1,n_seqs=1): not supported [BLAS] 391s GATED_LINEAR_ATTN(type=f32,head_count=32,head_size=64,n_seq_tokens=32,n_seqs=1): not supported [BLAS] 391s GATED_LINEAR_ATTN(type=f32,head_count=32,head_size=64,n_seq_tokens=32,n_seqs=4): not supported [BLAS] 391s GATED_LINEAR_ATTN(type=f32,head_count=32,head_size=64,n_seq_tokens=128,n_seqs=4): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=2,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=3,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=4,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=5,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=6,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=7,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=8,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=9,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f32,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f32,type_b=f16,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f16,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q8_0,type_b=f16,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f32,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_0,type_b=f16,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f16,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f32,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_K,type_b=f16,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f32,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[1,1],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[3,1],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[1,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[3,2],nr=[2,2],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=8,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,1,3,2],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=256,bs=[2,3],nr=[1,1],per=[0,3,2,1],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=1,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=8,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=iq2_xxs,type_b=f16,m=16,n=16,k=1024,bs=[3,2],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] not supported [CPU] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=32,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q4_1,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=1,k=32,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=1,k=32,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_1,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=32,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q8_0,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q2_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q3_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q5_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=q6_K,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_xs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq2_s,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_xxs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_s,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq1_m,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=1,k=32,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_nl,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq3_s,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=iq4_xs,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=1,k=1,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=bf16,type_b=f32,m=16,n=1,k=256,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=64,n=2,k=128,bs=[8,1],nr=[1,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=83,n=2,k=128,bs=[8,1],nr=[4,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=64,n=2,k=64,bs=[8,1],nr=[4,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=83,n=2,k=64,bs=[8,1],nr=[4,1],per=[0,1,2,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=64,n=45,k=128,bs=[8,1],nr=[4,1],per=[0,1,2,3],v=0): OK 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=45,k=64,bs=[8,1],nr=[4,1],per=[0,1,2,3],v=0): OK 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=193,bs=[1,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=67,bs=[1,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=128,bs=[1,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1056,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=129,bs=[1,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1057,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=128,bs=[1,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1056,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=129,bs=[1,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1057,bs=[1,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=128,bs=[1,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1056,bs=[1,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=129,bs=[1,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1057,bs=[1,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=128,bs=[1,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1056,bs=[1,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=129,bs=[1,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1057,bs=[1,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=128,bs=[2,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1056,bs=[2,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=129,bs=[2,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1057,bs=[2,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=128,bs=[2,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1056,bs=[2,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=129,bs=[2,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1057,bs=[2,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=128,bs=[2,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1056,bs=[2,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=129,bs=[2,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1057,bs=[2,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=128,bs=[2,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1056,bs=[2,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=129,bs=[2,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1057,bs=[2,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=128,bs=[4,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1056,bs=[4,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=129,bs=[4,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1057,bs=[4,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=128,bs=[4,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1056,bs=[4,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=129,bs=[4,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1057,bs=[4,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=128,bs=[4,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1056,bs=[4,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=129,bs=[4,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1057,bs=[4,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=128,bs=[4,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1056,bs=[4,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=129,bs=[4,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1057,bs=[4,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=128,bs=[8,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1056,bs=[8,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=129,bs=[8,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1057,bs=[8,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=128,bs=[8,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1056,bs=[8,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=129,bs=[8,1],nr=[1,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1057,bs=[8,1],nr=[1,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=128,bs=[8,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1056,bs=[8,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1056,n=1,k=129,bs=[8,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=128,n=1,k=1057,bs=[8,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=128,bs=[8,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1056,bs=[8,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=1057,n=1,k=129,bs=[8,1],nr=[4,1],per=[0,2,1,3],v=0): not supported [BLAS] 391s MUL_MAT(type_a=f16,type_b=f32,m=129,n=1,k=1057,bs=[8,1],nr=[4,1],per=[0,1,2,3],v=1): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f32,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=f16,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_0,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_K,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=4,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=1,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=1,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=2,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=2,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=4,b=0,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xxs,type_b=f32,n_mats=8,n_used=4,b=1,m=512,n=129,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q4_1,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q5_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q5_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q5_1,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q5_1,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q8_0,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q2_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q2_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q3_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q3_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q5_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q5_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q6_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=q6_K,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xs,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_xs,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_s,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq2_s,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq3_xxs,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq3_xxs,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq1_s,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq1_s,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq1_m,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq1_m,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq4_nl,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq4_nl,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq3_s,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq3_s,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq4_xs,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=iq4_xs,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=bf16,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=1,k=256): not supported [BLAS] 391s MUL_MAT_ID(type_a=bf16,type_b=f32,n_mats=4,n_used=2,b=0,m=512,n=32,k=256): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): OK 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): OK 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): OK 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): OK 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f32,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=f16,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q8_0,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_amain: compute buffer size: 0.19 MB 391s main: compute buffer size: 0.00 MB 391s main: compute buffer size: 0.02 MB 391s main: compute buffer size: 0.00 MB 391s =q4_0,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_0,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not suppoggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s rted [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_1,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=q4_K,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f32,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=1,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=1,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[1,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[1,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[3,1],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[1,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,1],trans_b=0): not supported [BLAS] not supported [CPU] 391s OUT_PROD(type_a=iq2_xxs,type_b=f16,m=256,n=16,k=16,bs=[3,3],nr=[2,2],trans_b=0): not supported [BLAS] not supported [CPU] 391s SQR(type=f16,ne=[10,5,4,3]): not supported [BLAS] 391s SQRT(type=f16,ne=[10,3,3,2]): not supported [BLAS] 391s LOG(type=f16,ne=[10,5,4,3]): not supported [BLAS] 391s SIN(type=f16,ne=[10,2,2,2]): not supported [BLAS] 391s COS(type=f16,ne=[10,2,2,2]): not supported [BLAS] 391s CLAMP(type=f16,ne=[10,5,4,3],min=-0.500000,max=0.500000): not supported [BLAS] 391s SQR(type=f32,ne=[10,5,4,3]): not supported [BLAS] 391s SQRT(type=f32,ne=[10,3,3,2]): not supported [BLAS] 391s LOG(type=f32,ne=[10,5,4,3]): not supported [BLAS] 391s SIN(type=f32,ne=[10,2,2,2]): not supported [BLAS] 391s COS(type=f32,ne=[10,2,2,2]): not supported [BLAS] 391s CLAMP(type=f32,ne=[10,5,4,3],min=-0.500000,max=0.500000): not supported [BLAS] 391s DIAG_MASK_INF(type=f32,ne=[10,10,1,1],n_past=5): not supported [BLAS] 391s DIAG_MASK_INF(type=f32,ne=[10,10,3,1],n_past=5): not supported [BLAS] 391s DIAG_MASK_INF(type=f32,ne=[10,10,3,2],n_past=5): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=0,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=0,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=0,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=0,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=0,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=0,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=0,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=0,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=0,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=0,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=0,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=0,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=0,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=0,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=0,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=0,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=1,m_prec=f32,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=1,m_prec=f16,scale=1.000000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,16,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,15,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,1024,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[15,1023,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,16,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,15,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=1,m_prec=f32,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1024,1024,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[1023,1023,1,1],mask=1,m_prec=f16,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,2,32,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,2,32,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[16,2,32,1],mask=0,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[32,2,32,1],mask=1,m_prec=f32,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[32,2,32,1],mask=1,m_prec=f16,scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[32,2,32,1],mask=1,m_prec=f32,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX(type=f32,ne=[32,2,32,1],mask=1,m_prec=f16,scale=0.100000,max_bias=8.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[16,16,1,1],scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[15,15,1,1],scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[16,1024,1,1],scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[15,1023,1,1],scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[1024,16,1,1],scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[1023,15,1,1],scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[1024,1024,1,1],scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[1023,1023,1,1],scale=1.000000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[16,16,1,1],scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[15,15,1,1],scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[16,1024,1,1],scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[15,1023,1,1],scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[1024,16,1,1],scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[1023,15,1,1],scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[1024,1024,1,1],scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[1023,1023,1,1],scale=0.100000,max_bias=0.000000): not supported [BLAS] 391s SOFT_MAX_BACK(type=f32,ne=[16,16,1,1],scale=1.000000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[15,15,1,1],scale=1.000000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[16,1024,1,1],scale=1.000000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[15,1023,1,1],scale=1.000000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[1024,16,1,1],scale=1.000000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[1023,15,1,1],scale=1.000000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[1024,1024,1,1],scale=1.000000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[1023,1023,1,1],scale=1.000000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[16,16,1,1],scale=0.100000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[15,15,1,1],scale=0.100000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[16,1024,1,1],scale=0.100000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[15,1023,1,1],scale=0.100000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[1024,16,1,1],scale=0.100000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[1023,15,1,1],scale=0.100000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[1024,1024,1,1],scale=0.100000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s SOFT_MAX_BACK(type=f32,ne=[1023,1023,1,1],scale=0.100000,max_bias=8.000000): not supported [BLAS] not supported [CPU] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,40,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,52,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,64,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,1,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,71,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,8,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[80,32,2,1],n_dims=20,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[80,32,2,1],n_dims=32,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,12,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,28,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[80,16,2,1],n_dims=80,mode=24,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,40,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,52,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,64,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,1,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,71,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,8,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[80,32,2,1],n_dims=20,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[80,32,2,1],n_dims=32,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,12,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,28,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[80,16,2,1],n_dims=80,mode=24,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,40,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,52,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,64,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,1,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,71,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,8,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[80,32,2,1],n_dims=20,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[80,32,2,1],n_dims=32,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,12,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,28,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[80,16,2,1],n_dims=80,mode=24,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,40,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,52,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,64,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,1,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,71,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,8,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[80,32,2,1],n_dims=20,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[80,32,2,1],n_dims=32,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,12,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,28,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[80,16,2,1],n_dims=80,mode=24,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,40,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,52,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,64,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,1,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,71,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,8,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[80,32,2,1],n_dims=20,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[80,32,2,1],n_dims=32,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,12,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,28,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[80,16,2,1],n_dims=80,mode=24,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,40,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,52,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,64,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,1,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,71,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,8,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[80,32,2,1],n_dims=20,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[80,32,2,1],n_dims=32,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,12,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,28,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[80,16,2,1],n_dims=80,mode=24,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,40,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,52,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,64,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,1,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,71,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,8,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[80,32,2,1],n_dims=20,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[80,32,2,1],n_dims=32,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,12,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,28,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[80,16,2,1],n_dims=80,mode=24,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,40,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,52,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,64,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,1,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,71,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,8,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[80,32,2,1],n_dims=20,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[80,32,2,1],n_dims=32,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,12,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,28,2,1],n_dims=128,mode=8,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[80,16,2,1],n_dims=80,mode=24,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=0): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.000000,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.000000,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.000000,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f32,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=0,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[128,32,2,1],n_dims=128,mode=0,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s ROPE_BACK(type=f16,ne_a=[64,128,2,1],n_dims=64,mode=2,n_ctx=512,fs=1.424500,ef=0.746500,af=1.424500,ff=1,v=1): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=0,v=0): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=0,v=0): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=1,v=0): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=1,v=0): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=2,v=0): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=2,v=0): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=3,v=0): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=3,v=0): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=0,v=1): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=0,v=1): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=1,v=1): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=1,v=1): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=2,v=1): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=2,v=1): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=3,v=1): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=3,v=1): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=0,v=2): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=0,v=2): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=1,v=2): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=1,v=2): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=2,v=2): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=2,v=2): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=3,v=2): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=3,v=2): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=0,v=3): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=0,v=3): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=1,v=3): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=1,v=3): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=2,v=3): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=2,v=3): not supported [BLAS] 391s CONCAT(type=f32,ne_a=[11,12,13,14],ne_b_d=7,dim=3,v=3): not supported [BLAS] 391s CONCAT(type=i32,ne_a=[11,12,13,14],ne_b_d=7,dim=3,v=3): not supported [BLAS] 391s ARGSORT(type=f32,ne=[8,1,1,1],order=0): not supported [BLAS] 391s ARGSORT(type=f32,ne=[16,10,10,10],order=0): not supported [BLAS] 391s ARGSORT(type=f32,ne=[60,10,10,10],order=0): not supported [BLAS] 391s ARGSORT(type=f32,ne=[8,1,1,1],order=1): not supported [BLAS] 391s ARGSORT(type=f32,ne=[16,10,10,10],order=1): not supported [BLAS] 391s ARGSORT(type=f32,ne=[60,10,10,10],order=1): not supported [BLAS] 391s UPSCALE(type=f32,ne=[512,512,3,2],scale_factor=2,mode=nearest,transpose=0): not supported [BLAS] 391s UPSCALE(type=f32,ne=[512,512,3,2],scale_factor=2,mode=nearest,transpose=1): not supported [BLAS] 391s UPSCALE(type=f32,ne=[2,5,7,11],ne_tgt=[5,7,11,13],mode=nearest): not supported [BLAS] 391s UPSCALE(type=f32,ne=[512,512,3,2],scale_factor=2,mode=bilinear,transpose=0): not supported [BLAS] 391s UPSCALE(type=f32,ne=[512,512,3,2],scale_factor=2,mode=bilinear,transpose=1): not supported [BLAS] 391s UPSCALE(type=f32,ne=[2,5,7,11],ne_tgt=[5,7,11,13],mode=bilinear): not supported [BLAS] 391s SUM(type=f32,ne=[10,5,4,3]): not supported [BLAS] 391s SUM_ROWS(type=f32,ne=[10,5,4,3]): not supported [BLAS] 391s MEAN(type=f32,ne=[10,5,4,3]): not supported [BLAS] 391s GROUP_NORM(type=f32,ne=[64,64,320,1],num_groups=32,eps=0.000001): not supported [BLAS] 391s GROUP_NORM(type=f32,ne=[9,9,1280,1],num_groups=32,eps=0.000001): not supported [BLAS] 391s ACC(type=f32,ne_a=[256,17,1,1],ne_b=[256,16,1,1]): not supported [BLAS] 391s PAD(type=f32,ne_a=[512,512,1,1],pad_0=1,pad_1=1): not supported [BLAS] 391s PAD_REFLECT_1D(type=f32,ne_a=[512,34,2,1],pad_0=10,pad_1=9): not supported [BLAS] 391s ARANGE(type=f32,start=0.000000,stop=10.000000,step=1.000000): not supported [BLAS] 391s TIMESTEP_EMBEDDING(type=f32,ne_a=[2,1,1,1],dim=320,max_period=10000): not supported [BLAS] 391s LEAKY_RELU(type=f32,ne_a=[10,5,4,3],negative_slope=0.100000): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=64,hsv=64,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=80,hsv=80,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_soggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ftcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s 000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=128,hsv=128,nh=4,nr=16,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=10.000000,prec=def,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=128,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=192,hsv=192,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=256,hsv=256,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,2,1,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=1,max_bias=8.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=1,kv=1024,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=1,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=3,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=32,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=f16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=bf16,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q8_0,permute=[0,1,2,3]): not supported [BLAS] 391s FLASH_ATTN_EXT(hsk=576,hsv=512,nh=4,nr=4,kv=512,nb=35,mask=0,max_bias=0.000000,logit_softcap=0.000000,prec=f32,type_KV=q4_0,permute=[0,1,2,3]): not supported [BLAS] 391s CROSS_ENTROPY_LOSS(type=f32,ne=[10,5,4,3]): not supported [BLAS] 391s CROSS_ENTROPY_LOSS(type=f32,ne=[30000,1,1,1]): not supported [BLAS] 391s CROSS_ENTROPY_LOSS_BACK(type=f32,ne=[10,5,4,3]): not supported [BLAS] 391s CROSS_ENTROPY_LOSS_BACK(type=f32,ne=[30000,1,1,1]): not supported [BLAS] 391s OPT_STEP_ADAMW(type=f32,ne=[10,5,4,3]): not supported [BLAS] 391s 5635/5635 tests passed 391s Backend BLAS: OK 391s 391s Backend 2/2: CPU 391s Skipping CPU backend 391s 2/2 backends passed 391s OK 391s load_model: ggml tensor size = 336 bytes 391s load_model: backend buffer size = 0.09 MB 391s 391s Performing test: 391s ggml_conv_1d_transpose (4): PASSED 391s ggml_conv_1d_transpose (5): PASSED 391s ggml_conv_1d_transpose (10): PASSED 391s ggml_conv_1d_transpose (14): PASSED 391s ggml_conv_1d_transpose (12): PASSED 391s ggml_conv_1d_transpose (18): PASSED 391s ggml_conv_1d_transpose (24): PASSED 391s ggml_conv_1d_transpose (50688): PASSED 391s load_model: ggml tensor size = 336 bytes 391s load_model: backend buffer size = 0.00 MB 391s 391s Performing test: 391s ggml_im2col (240): PASSED 391s ggml_conv1d (80): PASSED 391s load_model: ggml tensor size = 336 bytes 391s load_model: backend buffer size = 0.00 MB 391s 391s Performing test: 391s ggml_im2col (4320): PASSED 391s ggml_conv2d (480): PASSED 391s ggml_conv_2d_dw(channels=3, kernel=1x1, stride=1, pad=0, dilation=1, layout=WHCN): PASSED 391s ggml_conv_2d_dw(channels=3, kernel=1x1, stride=1, pad=0, dilation=1, layout=CWHN): PASSED 391s ggml_conv_2d_dw(channels=42, kernel=3x3, stride=2, pad=1, dilation=1, layout=WHCN): PASSED 391s ggml_conv_2d_dw(channels=42, kernel=3x3, stride=2, pad=1, dilation=1, layout=CWHN): PASSED 391s ggml_conv_2d_dw(channels=8, kernel=5x5, stride=1, pad=2, dilation=2, layout=WHCN): PASSED 391s ggml_conv_2d_dw(channels=8, kernel=5x5, stride=1, pad=2, dilation=2, layout=CWHN): PASSED 391s Testing dup on i16 -> i16 copy 391s Testing dup on i32 -> i32 copy 391s Testing dup on f16 -> f16 copy 391s Testing dup on f16 -> f32 copy 391s Testing dup on f32 -> f16 copy 391s Testing dup on f32 -> f32 copy 391s 391s Performing gemm_f16_out_f32 test: 391s 1224.0f,1023.0f,1158.0f,1259.0f,1359.0f,1194.0f,1535.0f,1247.0f,1185.0f,1029.0f,889.0f,1182.0f,955.0f,1179.0f,1147.0f,1048.0f, 391s 1216.0f,1087.0f,1239.0f,1361.0f,1392.0f,1260.0f,1247.0f,1563.0f,1167.0f,1052.0f,942.0f,1214.0f,1045.0f,1134.0f,1264.0f,1126.0f, 391s 1125.0f,966.0f,1079.0f,1333.0f,1287.0f,1101.0f,1185.0f,1167.0f,1368.0f,990.0f,967.0f,1121.0f,971.0f,1086.0f,1130.0f,980.0f, 391s 999.0f,902.0f,1020.0f,1056.0f,1076.0f,929.0f,1029.0f,1052.0f,990.0f,1108.0f,823.0f,989.0f,759.0f,1041.0f,1003.0f,870.0f, 391s gemm_mult (64): PASSED 391s load_model: ggml tensor size = 336 bytes 391s load_model: backend buffer size = 3584 bytes 391s Matrix A: [36, 4] 391s Matrix B: [36, 16] 391s 391s Performing ggml_mul_mat test: 391s 1224.0 1023.0 1158.0 1259.0 1359.0 1194.0 1535.0 1247.0 1185.0 1029.0 889.0 1182.0 955.0 1179.0 1147.0 1048.0 391s 1216.0 1087.0 1239.0 1361.0 1392.0 1260.0 1247.0 1563.0 1167.0 1052.0 942.0 1214.0 1045.0 1134.0 1264.0 1126.0 391s 1125.0 966.0 1079.0 1333.0 1287.0 1101.0 1185.0 1167.0 1368.0 990.0 967.0 1121.0 971.0 1086.0 1130.0 980.0 391s 999.0 902.0 1020.0 1056.0 1076.0 929.0 1029.0 1052.0 990.0 1108.0 823.0 989.0 759.0 1041.0 1003.0 870.0 391s ggml_mul_mat (64): PASSED 391s test-mul-mat0: iter:0/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.435 -0.717 0.214 -0.967 391s 391s x1: [4, 3, 1, 1] 391s 0.823 -0.605 -0.330 0.536 391s -0.444 0.108 -0.045 0.258 391s -0.270 0.027 0.904 0.832 391s 391s y: [1, 3, 1, 1] 391s 0.202 391s -0.530 391s -0.748 391s testing: mul_mat, [4, 1, 2, 1] = [4, 4, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 4, 2, 1] 391s 0.026 0.678 0.225 -0.408 391s 0.275 0.049 -0.013 0.946 391s -0.415 0.543 0.053 0.540 391s -0.200 0.783 -0.433 -0.295 391s 391s x1: [4, 1, 2, 1] 391s -0.514 -0.726 0.608 -0.687 391s 391s y: [4, 1, 2, 1] 391s -0.088 -0.834 -0.519 -0.527 391s testing: mul_mat, [2, 4, 2, 4] = [4, 2, 2, 4] * [4, 4, 2, 4] 391s x0: [4, 2, 2, 4] 391s 0.755 0.496 0.258 -0.929 391s 0.496 0.666 0.851 0.747 391s 391s x1: [4, 4, 2, 4] 391s 0.804 0.702 -0.467 0.080 391s -0.250 0.520 0.025 0.335 391s 0.063 -0.921 -0.125 0.864 391s 0.862 0.442 -0.431 0.477 391s 391s y: [2, 4, 2, 4] 391s 0.761 0.529 391s -0.236 0.495 391s -1.244 -0.044 391s 0.315 0.711 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.065 -0.825 -0.479 0.755 391s 391s x1: [4, 2, 1, 1] 391s 0.254 0.461 -0.343 0.481 391s -0.596 0.842 0.370 0.306 391s 391s y: [1, 2, 1, 1] 391s 0.163 391s -0.679 391s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.423 391s 391s x1: [1, 4, 2, 1] 391s 0.372 391s -0.813 391s -0.777 391s -0.277 391s 391s y: [1, 4, 2, 1] 391s -0.158 391s 0.344 391s 0.329 391s 0.117 391s testing: mul_mat, [2, 1, 2, 4] = [1, 2, 2, 4] * [1, 1, 2, 4] 391s x0: [1, 2, 2, 4] 391s 0.530 391s 0.398 391s 391s x1: [1, 1, 2, 4] 391s -0.620 391s 391s y: [2, 1, 2, 4] 391s -0.329 -0.247 391s test-mul-mat0: iter:1/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.418 391s 391s x1: [1, 4, 1, 1] 391s 0.729 391s -0.015 391s -0.073 391s 0.698 391s 391s y: [1, 4, 1, 1] 391s -0.305 391s 0.006 391s 0.030 391s -0.292 391s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.722 391s 391s x1: [1, 1, 2, 1] 391s -0.639 391s 391s y: [1, 1, 2, 1] 391s 0.461 391s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 2, 2, 1] 391s -0.644 391s -0.556 391s 391s x1: [1, 1, 2, 1] 391s -0.015 391s 391s y: [2, 1, 2, 1] 391s 0.010 0.008 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.864 391s 0.817 391s 391s x1: [1, 2, 1, 1] 391s -0.724 391s -0.279 391s 391s y: [2, 2, 1, 1] 391s -0.625 -0.591 391s -0.241 -0.228 391s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.211 0.318 391s 391s x1: [2, 1, 2, 1] 391s 0.244 0.674 391s 391s y: [1, 1, 2, 1] 391s 0.162 391s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.423 391s 391s x1: [1, 2, 2, 1] 391s -0.698 391s -0.855 391s 391s y: [1, 2, 2, 1] 391s 0.295 391s 0.362 391s test-mul-mat0: iter:2/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.524 391s -0.025 391s 391s x1: [1, 1, 1, 1] 391s -0.469 391s 391s y: [2, 1, 1, 1] 391s -0.246 0.012 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.035 391s -0.584 391s 0.115 391s 391s x1: [1, 2, 1, 1] 391s -0.685 391s 0.766 391s 391s y: [3, 2, 1, 1] 391s -0.024 0.401 -0.079 391s 0.027 -0.448 0.088 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.348 391s 0.459 391s 391s x1: [1, 3, 1, 1] 391s -0.148 391s 0.660 391s -0.211 391s 391s y: [2, 3, 1, 1] 391s 0.051 -0.068 391s -0.230 0.303 391s 0.073 -0.097 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.795 391s -0.728 391s -0.178 391s -0.989 391s 391s x1: [1, 2, 1, 1] 391s 0.277 391s 0.970 391s 391s y: [4, 2, 1, 1] 391s 0.220 -0.202 -0.049 -0.274 391s 0.771 -0.706 -0.173 -0.959 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.442 -0.902 -0.102 0.973 391s 391s x1: [4, 1, 1, 1] 391s 0.567 0.549 -0.413 -0.771 391s 391s y: [1, 1, 1, 1] 391s -0.952 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.801 391s -0.234 391s 391s x1: [1, 4, 1, 1] 391s 0.416 391s -0.578 391s -0.052 391s 0.730 391s 391s y: [2, 4, 1, 1] 391s -0.333 -0.097 391s 0.463 0.135 391s 0.042 0.012 391s -0.585 -0.171 391s test-mul-mat0: iter:3/500 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.777 -0.047 -0.660 391s 391s x1: [3, 4, 1, 1] 391s -0.897 -0.893 -0.085 391s 0.562 0.384 -0.115 391s -0.762 0.179 0.157 391s 0.060 0.190 -0.276 391s 391s y: [1, 4, 1, 1] 391s -0.600 391s 0.495 391s -0.704 391s 0.220 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.533 0.660 -0.860 391s 391s x1: [3, 1, 1, 1] 391s 0.219 0.051 0.238 391s 391s y: [1, 1, 1, 1] 391s -0.287 391s testing: mul_mat, [4, 1, 1, 3] = [3, 4, 1, 3] * [3, 1, 1, 3] 391s x0: [3, 4, 1, 3] 391s 0.630 0.318 -0.927 391s -0.485 0.557 0.252 391s 0.672 -0.384 -0.558 391s -0.604 0.225 -0.781 391s 391s x1: [3, 1, 1, 3] 391s -0.802 0.847 -0.661 391s 391s y: [4, 1, 1, 3] 391s 0.377 0.694 -0.496 1.191 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.694 -0.934 -0.873 391s 391s x1: [3, 4, 1, 1] 391s -0.376 0.292 -0.988 391s 0.066 0.688 0.237 391s 0.285 0.037 -0.199 391s -0.276 0.438 0.604 391s 391s y: [1, 4, 1, 1] 391s 0.851 391s -0.895 391s -0.059 391s -0.745 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.136 391s -0.998 391s -0.989 391s 391s x1: [1, 3, 1, 1] 391s 0.371 391s -0.625 391s 0.238 391s 391s y: [3, 3, 1, 1] 391s 0.050 -0.371 -0.367 391s -0.085 0.623 0.618 391s 0.032 -0.237 -0.235 391s testing: mul_mat, [1, 1, 1, 3] = [3, 1, 1, 3] * [3, 1, 1, 3] 391s x0: [3, 1, 1, 3] 391s 0.770 -0.627 -0.686 391s 391s x1: [3, 1, 1, 3] 391s -0.390 -0.477 0.311 391s 391s y: [1, 1, 1, 3] 391s -0.214 391s test-mul-mat0: iter:4/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.011 0.105 391s 0.878 0.599 391s 0.628 0.189 391s 391s x1: [2, 2, 1, 1] 391s 0.865 -0.534 391s 0.853 0.103 391s 391s y: [3, 2, 1, 1] 391s -0.066 0.440 0.443 391s 0.001 0.811 0.555 391s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 3, 2, 1] 391s -0.666 0.591 391s 0.730 0.746 391s 0.329 -0.175 391s 391s x1: [2, 3, 2, 1] 391s 0.314 0.991 391s 0.872 -0.351 391s 0.749 0.178 391s 391s y: [3, 3, 2, 1] 391s 0.376 0.969 -0.070 391s -0.788 0.375 0.348 391s -0.393 0.680 0.215 391s testing: mul_mat, [4, 3, 2, 3] = [2, 4, 2, 3] * [2, 3, 2, 3] 391s x0: [2, 4, 2, 3] 391s 0.905 0.898 391s 0.798 0.534 391s -0.333 0.073 391s -0.562 -0.045 391s 391s x1: [2, 3, 2, 3] 391s -0.934 0.402 391s 0.036 0.665 391s 0.030 -0.775 391s 391s y: [4, 3, 2, 3] 391s -0.485 -0.531 0.340 0.507 391s 0.630 0.384 0.037 -0.050 391s -0.668 -0.390 -0.067 0.018 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.931 -0.689 391s 391s x1: [2, 4, 1, 1] 391s 0.203 0.541 391s 0.774 0.867 391s -0.654 -0.104 391s -0.025 0.590 391s 391s y: [1, 4, 1, 1] 391s -0.184 391s 0.124 391s -0.537 391s -0.430 391s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.048 391s -0.454 391s 0.893 391s 391s x1: [1, 2, 2, 1] 391s -0.414 391s 0.764 391s 391s y: [3, 2, 2, 1] 391s 0.020 0.188 -0.370 391s -0.037 -0.347 0.683 391s testing: mul_mat, [2, 1, 2, 3] = [3, 2, 2, 3] * [3, 1, 2, 3] 391s x0: [3, 2, 2, 3] 391s -0.859 -0.007 -0.414 391s -0.213 -0.127 -0.512 391s 391s x1: [3, 1, 2, 3] 391s 0.437 0.849 -0.632 391s 391s y: [2, 1, 2, 3] 391s -0.120 0.123 391s test-mul-mat0: iter:5/500 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.174 -0.662 0.169 -0.047 391s 391s x1: [4, 2, 1, 1] 391s -0.728 0.178 -0.884 0.779 391s 0.891 -0.888 0.850 -0.062 391s 391s y: [1, 2, 1, 1] 391s -0.431 391s 0.890 391s testing: mul_mat, [4, 1, 4, 1] = [4, 4, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 4, 4, 1] 391s 0.290 -0.113 -0.462 0.406 391s -0.334 -0.571 0.518 -0.484 391s 0.367 -0.968 0.690 0.705 391s 0.202 -0.357 0.336 0.054 391s 391s x1: [4, 1, 4, 1] 391s 0.631 0.852 0.053 0.165 391s 391s y: [4, 1, 4, 1] 391s 0.129 -0.750 -0.440 -0.150 391s testing: mul_mat, [4, 4, 4, 2] = [4, 4, 4, 2] * [4, 4, 4, 2] 391s x0: [4, 4, 4, 2] 391s -0.642 -0.280 -0.114 -0.243 391s 0.295 -0.799 -0.349 0.739 391s 0.215 -0.792 0.612 0.499 391s -0.202 -0.266 -0.212 -0.456 391s 391s x1: [4, 4, 4, 2] 391s -0.521 0.619 -0.810 0.493 391s -0.446 -0.653 0.875 0.522 391s -0.807 0.962 0.691 -0.317 391s 0.385 -0.087 -0.131 0.308 391s 391s y: [4, 4, 4, 2] 391s 0.134 -0.002 -0.851 -0.113 391s 0.243 0.471 1.217 -0.159 391s 0.246 -1.481 -0.671 -0.095 391s -0.283 0.456 0.225 -0.167 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.843 -0.821 0.937 -0.991 391s 0.857 0.642 0.018 -0.624 391s 391s x1: [4, 4, 1, 1] 391s 0.204 0.671 0.127 -0.596 391s 0.606 0.345 -0.857 0.925 391s -0.050 -0.231 -0.284 0.862 391s 0.834 -0.794 0.802 0.751 391s 391s y: [2, 4, 1, 1] 391s 0.329 0.979 391s -1.492 0.150 391s -0.971 -0.733 391s 1.361 -0.249 391s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.704 0.851 391s 391s x1: [2, 4, 4, 1] 391s -0.426 0.255 391s -0.476 0.497 391s -0.927 0.444 391s -0.299 0.744 391s 391s y: [1, 4, 4, 1] 391s 0.516 391s 0.758 391s 1.030 391s 0.843 391s testing: mul_mat, [3, 2, 4, 2] = [1, 3, 4, 2] * [1, 2, 4, 2] 391s x0: [1, 3, 4, 2] 391s 0.231 391s 0.035 391s -0.242 391s 391s x1: [1, 2, 4, 2] 391s -0.554 391s -0.774 391s 391s y: [3, 2, 4, 2] 391s -0.128 -0.019 0.134 391s -0.179 -0.027 0.188 391s test-mul-mat0: iter:6/500 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.047 -0.506 391s 391s x1: [2, 2, 1, 1] 391s -0.741 -0.974 391s -0.710 0.490 391s 391s y: [1, 2, 1, 1] 391s 0.458 391s -0.282 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.688 -0.529 391s -0.166 0.479 391s 391s x1: [2, 1, 1, 1] 391s -0.551 0.083 391s 391s y: [2, 1, 1, 1] 391s -0.423 0.131 391s testing: mul_mat, [4, 2, 1, 2] = [2, 4, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 4, 1, 2] 391s -0.103 -0.083 391s -0.395 0.173 391s 0.752 0.030 391s 0.944 0.308 391s 391s x1: [2, 2, 1, 2] 391s -0.046 -0.815 391s -0.073 0.882 391s 391s y: [4, 2, 1, 2] 391s 0.073 -0.123 -0.059 -0.294 391s -0.066 0.182 -0.029 0.202 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.035 -0.145 391s 0.848 0.568 391s 391s x1: [2, 4, 1, 1] 391s 0.728 0.067 391s 0.169 -0.801 391s 0.902 -0.352 391s 0.153 -0.913 391s 391s y: [2, 4, 1, 1] 391s 0.016 0.656 391s 0.122 -0.312 391s 0.083 0.565 391s 0.138 -0.389 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.638 -0.746 391s -0.054 -0.666 391s 0.045 0.552 391s 0.582 0.851 391s 391s x1: [2, 2, 1, 1] 391s -0.722 0.411 391s -0.535 0.194 391s 391s y: [4, 2, 1, 1] 391s -0.767 -0.234 0.194 -0.071 391s -0.486 -0.100 0.083 -0.146 391s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 391s x0: [4, 4, 1, 2] 391s -0.820 -0.237 -0.220 0.417 391s 0.569 0.735 0.865 0.814 391s 0.719 -0.716 0.963 -0.772 391s -0.355 0.709 0.115 -1.000 391s 391s x1: [4, 2, 1, 2] 391s 0.114 -0.417 -0.504 -0.613 391s -0.937 -0.776 0.455 0.232 391s 391s y: [4, 2, 1, 2] 391s -0.139 -1.177 0.368 0.219 391s 0.948 -0.521 0.141 -0.398 391s test-mul-mat0: iter:7/500 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.538 0.979 391s -0.487 0.538 391s 391s x1: [2, 3, 1, 1] 391s -0.610 -0.745 391s -0.798 0.170 391s -0.879 -0.835 391s 391s y: [2, 3, 1, 1] 391s -1.057 -0.104 391s -0.263 0.480 391s -1.290 -0.021 391s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.685 0.494 391s -0.301 0.461 391s 0.655 0.635 391s -0.212 0.385 391s 391s x1: [2, 2, 2, 1] 391s -0.711 0.129 391s 0.602 -0.177 391s 391s y: [4, 2, 2, 1] 391s 0.551 0.273 -0.384 0.200 391s -0.499 -0.263 0.282 -0.196 391s testing: mul_mat, [2, 4, 2, 4] = [2, 2, 2, 4] * [2, 4, 2, 4] 391s x0: [2, 2, 2, 4] 391s -0.939 0.612 391s -0.225 0.137 391s 391s x1: [2, 4, 2, 4] 391s -0.755 -0.865 391s -0.326 -0.776 391s -0.352 -0.787 391s -0.487 0.777 391s 391s y: [2, 4, 2, 4] 391s 0.179 0.052 391s -0.169 -0.033 391s -0.152 -0.029 391s 0.933 0.216 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.270 0.090 391s 0.194 -0.650 391s 391s x1: [2, 2, 1, 1] 391s 0.498 -0.052 391s 0.721 0.609 391s 391s y: [2, 2, 1, 1] 391s 0.130 0.130 391s 0.250 -0.256 391s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 4, 2, 1] 391s 0.226 0.064 391s -0.114 0.987 391s 0.138 -0.882 391s 0.092 -0.940 391s 391s x1: [2, 2, 2, 1] 391s 0.853 0.949 391s -0.609 -0.320 391s 391s y: [4, 2, 2, 1] 391s 0.254 0.839 -0.719 -0.813 391s -0.158 -0.246 0.198 0.244 391s testing: mul_mat, [2, 2, 2, 4] = [4, 2, 2, 4] * [4, 2, 2, 4] 391s x0: [4, 2, 2, 4] 391s 0.218 0.277 -0.239 0.011 391s 0.639 -0.276 0.482 0.000 391s 391s x1: [4, 2, 2, 4] 391s 0.054 0.003 -0.782 -0.752 391s -0.907 -0.432 -0.899 -0.958 391s 391s y: [2, 2, 2, 4] 391s 0.192 -0.343 391s -0.113 -0.894 391s test-mul-mat0: iter:8/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.236 0.357 0.853 0.689 391s 0.244 -0.226 0.368 -0.607 391s -0.700 0.613 0.360 -0.539 391s 0.642 0.786 -0.464 -0.819 391s 391s x1: [4, 3, 1, 1] 391s -0.942 0.863 0.800 -0.918 391s 0.023 0.488 -0.465 -0.347 391s 0.065 -0.130 0.936 0.159 391s 391s y: [4, 3, 1, 1] 391s 0.580 0.427 1.971 0.454 391s -0.467 -0.065 0.302 0.898 391s 0.846 0.293 0.126 -0.624 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.407 0.816 -0.495 -0.683 391s 391s x1: [4, 4, 3, 1] 391s -0.570 -0.994 -0.761 -0.708 391s 0.807 -0.680 0.315 0.295 391s -0.145 0.968 -0.640 0.726 391s 0.904 0.519 0.908 -0.332 391s 391s y: [1, 4, 3, 1] 391s 0.281 391s -1.241 391s 0.670 391s -0.167 391s testing: mul_mat, [2, 1, 3, 1] = [4, 2, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.261 0.015 0.994 0.497 391s -0.289 0.964 -0.910 -0.473 391s 391s x1: [4, 1, 3, 1] 391s 0.793 0.951 -0.692 0.192 391s 391s y: [2, 1, 3, 1] 391s -0.371 1.226 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.855 0.772 0.375 -0.457 391s 0.682 -0.482 -0.903 -0.065 391s 0.991 -0.712 -0.760 0.356 391s 0.819 -0.821 -0.565 0.457 391s 391s x1: [4, 2, 1, 1] 391s -0.409 0.232 -0.930 -0.924 391s -0.076 -0.735 0.861 -0.815 391s 391s y: [4, 2, 1, 1] 391s -0.098 0.508 -0.194 -0.423 391s 0.063 -0.422 -0.497 -0.319 391s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.008 -0.415 -0.724 -0.490 391s 0.517 -0.077 -0.429 0.153 391s 0.254 -0.390 -0.600 0.546 391s 391s x1: [4, 4, 3, 1] 391s 0.423 -0.772 -0.321 0.103 391s 0.432 0.855 0.693 -0.336 391s 0.925 0.769 0.588 -0.810 391s 0.630 0.773 0.471 0.486 391s 391s y: [3, 4, 3, 1] 391s 0.498 0.431 0.657 391s -0.696 -0.191 -0.823 391s -0.356 0.042 -0.860 391s -0.905 0.139 -0.159 391s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.883 -0.871 -0.332 391s -0.993 -0.009 0.099 391s 391s x1: [3, 4, 3, 1] 391s -0.758 0.377 -0.608 391s -0.358 0.887 0.545 391s -0.812 -0.911 -0.296 391s 0.108 -0.339 -0.030 391s 391s y: [2, 4, 3, 1] 391s -0.796 0.690 391s -1.271 0.402 391s 0.175 0.786 391s 0.401 -0.107 391s test-mul-mat0: iter:9/500 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.515 -0.231 0.660 391s 0.522 -0.102 -0.349 391s -0.810 0.996 -0.522 391s -0.683 0.132 -0.661 391s 391s x1: [3, 2, 1, 1] 391s 0.029 0.566 0.464 391s -0.917 0.597 0.344 391s 391s y: [4, 2, 1, 1] 391s 0.191 -0.205 0.298 -0.252 391s -0.383 -0.660 1.158 0.478 391s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.073 -0.203 -0.981 391s 0.243 0.928 -0.642 391s 391s x1: [3, 4, 2, 1] 391s -0.148 0.942 -0.505 391s 0.086 -0.735 -0.962 391s -0.876 0.624 -0.652 391s -0.659 -0.463 -0.675 391s 391s y: [2, 4, 2, 1] 391s 0.314 1.162 391s 1.087 -0.044 391s 0.576 0.784 391s 0.805 -0.158 391s testing: mul_mat, [4, 2, 2, 1] = [3, 4, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.314 -0.609 -0.412 391s -0.949 0.191 -0.871 391s -0.023 0.988 -0.852 391s -0.779 0.916 -0.493 391s 391s x1: [3, 2, 2, 1] 391s 0.305 0.760 0.240 391s 0.646 -0.704 0.565 391s 391s y: [4, 2, 2, 1] 391s -0.466 -0.354 0.539 0.340 391s 0.398 -1.240 -1.191 -1.427 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.182 -0.505 -0.615 391s -0.531 0.039 0.001 391s 391s x1: [3, 4, 1, 1] 391s 0.464 -0.622 -0.588 391s 0.932 -0.524 -0.100 391s 0.630 0.790 0.291 391s -0.782 0.841 -0.518 391s 391s y: [2, 4, 1, 1] 391s 0.592 -0.271 391s 0.157 -0.515 391s -0.692 -0.303 391s 0.036 0.447 391s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 3, 2, 1] 391s -0.868 -0.392 391s 0.575 -0.524 391s 0.499 0.129 391s 391s x1: [2, 3, 2, 1] 391s 0.355 -0.828 391s -0.638 -0.147 391s -0.906 -0.386 391s 391s y: [3, 3, 2, 1] 391s 0.016 0.639 0.071 391s 0.612 -0.290 -0.338 391s 0.938 -0.319 -0.502 391s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.146 -0.936 0.434 391s -0.729 -0.100 0.632 391s 391s x1: [3, 2, 2, 1] 391s 0.058 -0.282 -0.812 391s -0.903 0.103 0.190 391s 391s y: [2, 2, 2, 1] 391s -0.080 -0.527 391s -0.146 0.768 391s test-mul-mat0: iter:10/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.335 391s -0.781 391s 391s x1: [1, 1, 1, 1] 391s 0.501 391s 391s y: [2, 1, 1, 1] 391s 0.168 -0.391 391s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.403 391s 391s x1: [1, 2, 2, 1] 391s 0.488 391s 0.432 391s 391s y: [1, 2, 2, 1] 391s -0.197 391s -0.174 391s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.150 391s 0.113 391s 391s x1: [1, 1, 2, 1] 391s 0.190 391s 391s y: [2, 1, 2, 1] 391s 0.029 0.021 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.044 391s 0.110 391s 0.978 391s -0.317 391s 391s x1: [1, 2, 1, 1] 391s -0.988 391s -0.900 391s 391s y: [4, 2, 1, 1] 391s -0.043 -0.108 -0.966 0.313 391s -0.039 -0.099 -0.880 0.285 391s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 1, 2, 1] 391s 0.567 -0.366 0.766 -0.111 391s 391s x1: [4, 1, 2, 1] 391s -0.405 -0.762 -0.279 0.848 391s 391s y: [1, 1, 2, 1] 391s -0.258 391s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 4, 2, 1] 391s 0.052 391s 0.622 391s 0.810 391s 0.030 391s 391s x1: [1, 4, 2, 1] 391s 0.840 391s 0.453 391s 0.935 391s -0.009 391s 391s y: [4, 4, 2, 1] 391s 0.044 0.522 0.681 0.025 391s 0.024 0.282 0.367 0.014 391s 0.049 0.581 0.758 0.028 391s -0.000 -0.006 -0.008 -0.000 391s test-mul-mat0: iter:11/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.943 391s -0.795 391s 391s x1: [1, 4, 1, 1] 391s -0.741 391s -0.642 391s 0.168 391s -0.107 391s 391s y: [2, 4, 1, 1] 391s 0.698 0.589 391s 0.605 0.510 391s -0.159 -0.134 391s 0.101 0.085 391s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 3, 3, 1] 391s 0.562 391s -0.719 391s 0.811 391s 391s x1: [1, 2, 3, 1] 391s -0.225 391s 0.543 391s 391s y: [3, 2, 3, 1] 391s -0.127 0.162 -0.183 391s 0.305 -0.390 0.440 391s testing: mul_mat, [4, 3, 3, 2] = [1, 4, 3, 2] * [1, 3, 3, 2] 391s x0: [1, 4, 3, 2] 391s -0.183 391s 0.060 391s -0.406 391s -0.621 391s 391s x1: [1, 3, 3, 2] 391s -0.934 391s 0.198 391s 0.162 391s 391s y: [4, 3, 3, 2] 391s 0.171 -0.056 0.379 0.580 391s -0.036 0.012 -0.080 -0.123 391s -0.030 0.010 -0.066 -0.100 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.574 391s 391s x1: [1, 4, 1, 1] 391s 0.630 391s 0.313 391s -0.805 391s -0.996 391s 391s y: [1, 4, 1, 1] 391s 0.362 391s 0.180 391s -0.462 391s -0.572 391s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.671 391s 0.975 391s 391s x1: [1, 1, 3, 1] 391s -0.007 391s 391s y: [2, 1, 3, 1] 391s 0.005 -0.007 391s testing: mul_mat, [3, 1, 3, 2] = [2, 3, 3, 2] * [2, 1, 3, 2] 391s x0: [2, 3, 3, 2] 391s 0.340 -0.347 391s 0.035 0.230 391s 0.130 0.135 391s 391s x1: [2, 1, 3, 2] 391s 0.128 0.413 391s 391s y: [3, 1, 3, 2] 391s -0.100 0.099 0.072 391s test-mul-mat0: iter:12/500 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.152 0.027 -0.478 0.091 391s -0.574 -0.496 -0.936 -0.044 391s 391s x1: [4, 4, 1, 1] 391s -0.841 -0.556 0.506 -0.775 391s 0.857 0.212 0.139 0.568 391s -0.824 0.181 -0.308 0.561 391s -0.302 0.541 0.686 0.128 391s 391s y: [2, 4, 1, 1] 391s -0.455 0.319 391s 0.121 -0.752 391s 0.078 0.646 391s -0.348 -0.743 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.944 0.575 0.129 0.083 391s 0.144 0.306 -0.736 0.835 391s 391s x1: [4, 2, 1, 1] 391s 0.139 -0.733 -0.380 0.561 391s 0.274 -0.588 0.719 0.718 391s 391s y: [2, 2, 1, 1] 391s -0.293 0.544 391s 0.072 -0.070 391s testing: mul_mat, [1, 2, 1, 4] = [4, 1, 1, 4] * [4, 2, 1, 4] 391s x0: [4, 1, 1, 4] 391s -0.950 0.302 0.044 0.306 391s 391s x1: [4, 2, 1, 4] 391s -0.134 -0.038 0.377 -0.447 391s -0.910 0.256 0.705 0.117 391s 391s y: [1, 2, 1, 4] 391s -0.005 391s 1.009 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.119 -0.999 -0.125 -0.152 391s 391s x1: [4, 1, 1, 1] 391s -0.812 -0.149 -0.527 -0.693 391s 391s y: [1, 1, 1, 1] 391s 0.224 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.795 391s 391s x1: [1, 4, 1, 1] 391s -0.162 391s -0.242 391s -0.160 391s 0.511 391s 391s y: [1, 4, 1, 1] 391s -0.129 391s -0.192 391s -0.127 391s 0.407 391s testing: mul_mat, [4, 1, 1, 4] = [1, 4, 1, 4] * [1, 1, 1, 4] 391s x0: [1, 4, 1, 4] 391s 0.000 391s 0.952 391s 0.616 391s -0.296 391s 391s x1: [1, 1, 1, 4] 391s 0.561 391s 391s y: [4, 1, 1, 4] 391s 0.000 0.534 0.346 -0.166 391s test-mul-mat0: iter:13/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.138 -0.208 0.291 391s -0.245 0.496 -0.390 391s 0.032 -0.598 -0.283 391s 391s x1: [3, 3, 1, 1] 391s 0.400 0.009 0.158 391s -0.089 -0.484 -0.047 391s -0.528 0.324 0.792 391s 391s y: [3, 3, 1, 1] 391s 0.099 -0.155 -0.037 391s 0.075 -0.200 0.300 391s 0.090 -0.019 -0.435 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.332 -0.266 0.873 391s -0.660 0.891 -0.216 391s -0.144 -0.156 0.256 391s 391s x1: [3, 3, 1, 1] 391s 0.098 -0.372 -0.404 391s 0.352 0.705 -0.053 391s 0.912 0.081 -0.584 391s 391s y: [3, 3, 1, 1] 391s -0.221 -0.308 -0.059 391s -0.118 0.408 -0.174 391s -0.228 -0.403 -0.293 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.326 0.214 -0.797 391s 0.954 0.810 0.555 391s 0.659 -0.243 0.467 391s -0.260 0.173 0.940 391s 391s x1: [3, 3, 1, 1] 391s -0.820 -0.364 -0.405 391s 0.318 0.428 0.887 391s -0.927 -0.076 -0.503 391s 391s y: [4, 3, 1, 1] 391s -0.023 -1.302 -0.641 -0.230 391s -0.511 1.142 0.519 0.824 391s 0.083 -1.225 -0.827 -0.245 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.294 -0.170 -0.852 391s -0.666 0.944 -0.090 391s -0.161 -0.044 0.854 391s 0.630 -0.955 0.702 391s 391s x1: [3, 4, 1, 1] 391s -0.928 0.907 0.812 391s -0.588 0.798 -0.404 391s 0.268 -0.358 0.852 391s 0.448 0.278 -0.553 391s 391s y: [4, 4, 1, 1] 391s -0.573 1.402 0.803 -0.880 391s 0.382 1.182 -0.286 -1.416 391s -0.744 -0.594 0.700 1.109 391s 0.293 0.014 -0.557 -0.372 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.870 -0.492 0.816 -0.886 391s 0.369 0.408 0.274 -0.851 391s 391s x1: [4, 3, 1, 1] 391s -0.431 -0.389 0.169 0.309 391s 0.784 0.109 0.380 0.691 391s -0.079 0.792 0.488 0.517 391s 391s y: [2, 3, 1, 1] 391s 0.431 -0.534 391s -1.038 -0.150 391s -0.381 -0.012 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.907 0.486 391s -0.822 0.274 391s 0.489 -0.815 391s 0.076 -0.134 391s 391s x1: [2, 4, 1, 1] 391s -0.887 0.744 391s -0.021 -0.943 391s -0.301 0.024 391s -0.795 0.609 391s 391s y: [4, 4, 1, 1] 391s -0.443 0.933 -1.040 -0.167 391s -0.477 -0.241 0.758 0.124 391s -0.261 0.254 -0.167 -0.026 391s -0.425 0.820 -0.885 -0.142 391s test-mul-mat0: iter:14/500 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.417 391s 391s x1: [1, 1, 1, 1] 391s -0.377 391s 391s y: [1, 1, 1, 1] 391s 0.157 391s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 1, 4, 1] 391s -0.896 391s 391s x1: [1, 1, 4, 1] 391s 0.992 391s 391s y: [1, 1, 4, 1] 391s -0.888 391s testing: mul_mat, [1, 1, 4, 4] = [1, 1, 4, 4] * [1, 1, 4, 4] 391s x0: [1, 1, 4, 4] 391s 0.346 391s 391s x1: [1, 1, 4, 4] 391s -0.064 391s 391s y: [1, 1, 4, 4] 391s -0.022 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.951 391s -0.753 391s -0.738 391s 0.734 391s 391s x1: [1, 1, 1, 1] 391s -0.909 391s 391s y: [4, 1, 1, 1] 391s -0.865 0.684 0.670 -0.667 391s testing: mul_mat, [4, 1, 4, 1] = [4, 4, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 4, 4, 1] 391s 0.902 0.986 -0.705 0.584 391s 0.581 -0.796 0.321 0.746 391s -0.340 0.896 -0.988 0.124 391s -0.569 -0.939 -0.458 -0.827 391s 391s x1: [4, 1, 4, 1] 391s -0.418 -0.410 -0.407 -0.598 391s 391s y: [4, 1, 4, 1] 391s -0.844 -0.493 0.102 1.304 391s testing: mul_mat, [4, 4, 4, 4] = [4, 4, 4, 4] * [4, 4, 4, 4] 391s x0: [4, 4, 4, 4] 391s -0.478 0.013 -0.869 0.974 391s -0.615 0.129 -0.643 0.983 391s -0.833 0.941 -0.785 0.662 391s -0.938 0.475 0.312 -0.355 391s 391s x1: [4, 4, 4, 4] 391s -0.266 -0.658 -0.041 -0.169 391s -0.771 0.925 0.223 0.437 391s -0.097 -0.844 -0.772 0.626 391s -0.786 0.193 -0.312 0.797 391s 391s y: [4, 4, 4, 4] 391s -0.011 -0.061 -0.478 -0.017 391s 0.612 0.879 1.627 1.078 391s 1.316 1.063 0.307 -0.772 391s 1.425 1.492 1.608 0.449 391s test-mul-mat0: iter:15/500 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.574 391s 0.840 391s 0.803 391s 0.902 391s 391s x1: [1, 4, 1, 1] 391s 0.623 391s 0.952 391s 0.923 391s 0.902 391s 391s y: [4, 4, 1, 1] 391s 0.358 0.524 0.501 0.562 391s 0.546 0.799 0.764 0.858 391s 0.530 0.775 0.741 0.832 391s 0.518 0.757 0.724 0.813 391s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.895 391s 391s x1: [1, 4, 2, 1] 391s 0.097 391s -0.027 391s 0.511 391s -0.836 391s 391s y: [1, 4, 2, 1] 391s -0.087 391s 0.024 391s -0.457 391s 0.749 391s testing: mul_mat, [2, 1, 2, 3] = [1, 2, 2, 3] * [1, 1, 2, 3] 391s x0: [1, 2, 2, 3] 391s -0.061 391s -0.019 391s 391s x1: [1, 1, 2, 3] 391s 0.624 391s 391s y: [2, 1, 2, 3] 391s -0.038 -0.012 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.292 391s 0.607 391s -0.197 391s 391s x1: [1, 2, 1, 1] 391s 0.781 391s 0.444 391s 391s y: [3, 2, 1, 1] 391s 0.228 0.475 -0.154 391s 0.130 0.269 -0.087 391s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.186 0.683 0.017 391s -0.472 0.533 0.813 391s -0.720 0.430 0.269 391s -0.509 -0.336 -0.547 391s 391s x1: [3, 1, 2, 1] 391s -0.021 -0.061 0.296 391s 391s y: [4, 1, 2, 1] 391s -0.040 0.218 0.069 -0.131 391s testing: mul_mat, [2, 3, 2, 3] = [4, 2, 2, 3] * [4, 3, 2, 3] 391s x0: [4, 2, 2, 3] 391s -0.190 -0.626 -0.456 -0.768 391s 0.574 0.385 -0.937 -0.483 391s 391s x1: [4, 3, 2, 3] 391s -0.949 -0.888 0.317 -0.927 391s 0.156 -0.873 0.719 -0.658 391s -0.345 0.999 -0.167 -0.662 391s 391s y: [2, 3, 2, 3] 391s 1.304 -0.736 391s 0.695 -0.602 391s 0.025 0.663 391s test-mul-mat0: iter:16/500 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.366 391s 0.518 391s 391s x1: [1, 2, 1, 1] 391s -0.666 391s 0.232 391s 391s y: [2, 2, 1, 1] 391s 0.243 -0.345 391s -0.085 0.120 391s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 3, 3, 1] 391s 0.666 391s 0.093 391s 0.475 391s 391s x1: [1, 2, 3, 1] 391s 0.291 391s -0.654 391s 391s y: [3, 2, 3, 1] 391s 0.194 0.027 0.138 391s -0.436 -0.061 -0.311 391s testing: mul_mat, [3, 3, 3, 4] = [1, 3, 3, 4] * [1, 3, 3, 4] 391s x0: [1, 3, 3, 4] 391s 0.405 391s -0.384 391s -0.712 391s 391s x1: [1, 3, 3, 4] 391s -0.680 391s 0.379 391s -0.392 391s 391s y: [3, 3, 3, 4] 391s -0.275 0.261 0.484 391s 0.154 -0.146 -0.270 391s -0.159 0.151 0.279 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.393 391s -0.758 391s -0.179 391s 0.518 391s 391s x1: [1, 3, 1, 1] 391s -0.728 391s 0.255 391s 0.353 391s 391s y: [4, 3, 1, 1] 391s -0.287 0.552 0.130 -0.377 391s 0.100 -0.194 -0.046 0.132 391s 0.139 -0.268 -0.063 0.183 391s testing: mul_mat, [1, 1, 3, 1] = [4, 1, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.173 -0.932 0.600 0.040 391s 391s x1: [4, 1, 3, 1] 391s 0.816 -0.378 -0.178 0.743 391s 391s y: [1, 1, 3, 1] 391s 0.134 391s testing: mul_mat, [4, 4, 3, 4] = [1, 4, 3, 4] * [1, 4, 3, 4] 391s x0: [1, 4, 3, 4] 391s -0.021 391s -0.828 391s 0.722 391s -0.925 391s 391s x1: [1, 4, 3, 4] 391s 0.553 391s -0.904 391s 0.178 391s 0.794 391s 391s y: [4, 4, 3, 4] 391s -0.012 -0.458 0.399 -0.511 391s 0.019 0.749 -0.652 0.836 391s -0.004 -0.147 0.128 -0.164 391s -0.017 -0.658 0.573 -0.735 391s test-mul-mat0: iter:17/500 391s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.004 391s 391s x1: [1, 2, 1, 1] 391s -0.143 391s 0.623 391s 391s y: [1, 2, 1, 1] 391s 0.001 391s -0.003 391s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.075 391s -0.077 391s 0.254 391s 391s x1: [1, 1, 1, 1] 391s 0.532 391s 391s y: [3, 1, 1, 1] 391s 0.040 -0.041 0.135 391s testing: mul_mat, [1, 3, 1, 4] = [1, 1, 1, 4] * [1, 3, 1, 4] 391s x0: [1, 1, 1, 4] 391s 0.189 391s 391s x1: [1, 3, 1, 4] 391s -0.048 391s 0.293 391s -0.636 391s 391s y: [1, 3, 1, 4] 391s -0.009 391s 0.055 391s -0.120 391s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.897 391s 0.536 391s 0.068 391s 391s x1: [1, 1, 1, 1] 391s -0.888 391s 391s y: [3, 1, 1, 1] 391s 0.797 -0.476 -0.061 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.309 -0.404 0.032 391s 391s x1: [3, 1, 1, 1] 391s -0.039 0.159 -0.720 391s 391s y: [1, 1, 1, 1] 391s -0.075 391s testing: mul_mat, [2, 3, 1, 4] = [1, 2, 1, 4] * [1, 3, 1, 4] 391s x0: [1, 2, 1, 4] 391s 0.903 391s 0.410 391s 391s x1: [1, 3, 1, 4] 391s 0.614 391s 0.851 391s 0.984 391s 391s y: [2, 3, 1, 4] 391s 0.555 0.251 391s 0.769 0.348 391s 0.889 0.403 391s test-mul-mat0: iter:18/500 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.471 0.499 391s -0.190 -0.857 391s 0.350 -0.206 391s 0.050 0.565 391s 391s x1: [2, 1, 1, 1] 391s 0.838 -0.098 391s 391s y: [4, 1, 1, 1] 391s -0.443 -0.075 0.313 -0.013 391s testing: mul_mat, [4, 4, 3, 1] = [2, 4, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.885 0.793 391s -0.832 -0.091 391s -0.157 0.733 391s -0.763 -0.464 391s 391s x1: [2, 4, 3, 1] 391s 0.328 0.693 391s -0.196 -0.454 391s -0.430 -0.475 391s 0.917 -0.267 391s 391s y: [4, 4, 3, 1] 391s 0.259 -0.337 0.456 -0.572 391s -0.187 0.204 -0.302 0.360 391s 0.004 0.401 -0.281 0.549 391s -1.023 -0.739 -0.340 -0.575 391s testing: mul_mat, [4, 4, 3, 4] = [2, 4, 3, 4] * [2, 4, 3, 4] 391s x0: [2, 4, 3, 4] 391s 0.570 0.862 391s -0.545 0.675 391s -0.919 -0.964 391s 0.703 -0.567 391s 391s x1: [2, 4, 3, 4] 391s -0.523 0.997 391s 0.954 0.796 391s 0.302 0.604 391s -0.386 0.417 391s 391s y: [4, 4, 3, 4] 391s 0.561 0.958 -0.481 -0.933 391s 1.230 0.017 -1.645 0.220 391s 0.693 0.243 -0.860 -0.131 391s 0.139 0.492 -0.047 -0.508 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.032 0.120 391s 0.142 -0.502 391s -0.080 -0.993 391s -0.282 0.433 391s 391s x1: [2, 4, 1, 1] 391s -0.147 -0.647 391s -0.666 -0.222 391s -0.214 0.791 391s 0.982 -0.533 391s 391s y: [4, 4, 1, 1] 391s -0.082 0.304 0.655 -0.239 391s -0.048 0.017 0.274 0.092 391s 0.088 -0.427 -0.768 0.403 391s -0.033 0.407 0.450 -0.508 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s -0.932 0.539 -0.086 -0.388 391s -0.093 -0.434 -0.028 0.243 391s 391s x1: [4, 2, 3, 1] 391s -0.955 -0.670 -0.602 0.494 391s -0.733 0.067 -0.514 0.107 391s 391s y: [2, 2, 3, 1] 391s 0.388 0.517 391s 0.722 0.080 391s testing: mul_mat, [4, 4, 3, 4] = [2, 4, 3, 4] * [2, 4, 3, 4] 391s x0: [2, 4, 3, 4] 391s -0.468 -0.049 391s -0.193 0.995 391s 0.338 0.117 391s 0.320 0.332 391s 391s x1: [2, 4, 3, 4] 391s 0.295 -0.212 391s -0.734 -0.428 391s 0.936 -0.544 391s -0.010 -0.995 391s 391s y: [4, 4, 3, 4] 391s -0.128 -0.268 0.075 0.024 391s 0.364 -0.284 -0.298 -0.377 391s -0.411 -0.722 0.253 0.119 391s 0.053 -0.989 -0.119 -0.333 391s test-mul-mat0: iter:19/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.122 -0.698 -0.011 391s -0.949 0.958 0.025 391s 391s x1: [3, 3, 1, 1] 391s -0.022 0.587 0.161 391s -0.016 -0.000 0.189 391s 0.294 -0.191 0.480 391s 391s y: [2, 3, 1, 1] 391s -0.414 0.586 391s -0.004 0.019 391s 0.164 -0.450 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.263 -0.148 0.480 391s -0.741 0.941 -0.639 391s 0.918 0.928 0.340 391s 0.504 0.088 -0.676 391s 391s x1: [3, 2, 1, 1] 391s 0.294 0.556 0.789 391s 0.579 -0.284 0.486 391s 391s y: [4, 2, 1, 1] 391s 0.219 -0.198 1.054 -0.336 391s 0.123 -1.007 0.433 -0.062 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.574 0.084 0.359 391s -0.711 -0.430 0.614 391s 0.026 0.422 0.094 391s 0.285 0.364 0.456 391s 391s x1: [3, 4, 1, 1] 391s -0.496 -0.722 0.618 391s 0.313 0.757 0.281 391s -0.565 -0.940 -0.730 391s -0.514 -0.983 0.295 391s 391s y: [4, 4, 1, 1] 391s -0.123 1.043 -0.259 -0.122 391s 0.344 -0.375 0.354 0.492 391s -0.665 0.357 -0.480 -0.836 391s -0.272 0.969 -0.400 -0.369 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.526 -0.509 -0.935 391s -0.712 0.583 -0.333 391s 0.444 0.224 -0.417 391s 391s x1: [3, 4, 1, 1] 391s 0.202 0.291 -0.205 391s -0.294 -0.621 0.119 391s 0.210 -0.343 -0.263 391s -0.477 -0.586 -0.982 391s 391s y: [3, 4, 1, 1] 391s -0.063 0.094 0.240 391s 0.359 -0.193 -0.319 391s 0.309 -0.262 0.126 391s 1.467 0.325 0.066 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.315 -0.427 -0.338 391s 0.543 0.694 0.783 391s 391s x1: [3, 3, 1, 1] 391s -0.646 0.238 -0.803 391s 0.380 -0.340 0.292 391s -0.336 -0.977 -0.253 391s 391s y: [2, 3, 1, 1] 391s -0.034 -0.814 391s 0.166 0.198 391s 0.397 -1.058 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.706 0.932 391s -0.292 0.795 391s 0.731 0.956 391s 0.289 -0.484 391s 391s x1: [2, 3, 1, 1] 391s -0.649 0.399 391s -0.694 -0.234 391s 0.418 0.264 391s 391s y: [4, 3, 1, 1] 391s -0.086 0.507 -0.092 -0.381 391s -0.708 0.016 -0.731 -0.087 391s 0.541 0.089 0.558 -0.007 391s test-mul-mat0: iter:20/500 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.248 391s -0.941 391s 0.191 391s -0.705 391s 391s x1: [1, 3, 1, 1] 391s 0.853 391s -0.995 391s -0.807 391s 391s y: [4, 3, 1, 1] 391s -0.212 -0.803 0.163 -0.602 391s 0.247 0.936 -0.190 0.702 391s 0.200 0.759 -0.154 0.569 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.644 391s -0.278 391s 391s x1: [1, 4, 1, 1] 391s -0.368 391s -0.115 391s -0.043 391s -0.585 391s 391s y: [2, 4, 1, 1] 391s 0.237 0.102 391s 0.074 0.032 391s 0.028 0.012 391s 0.376 0.163 391s testing: mul_mat, [2, 2, 1, 3] = [1, 2, 1, 3] * [1, 2, 1, 3] 391s x0: [1, 2, 1, 3] 391s -0.232 391s -0.373 391s 391s x1: [1, 2, 1, 3] 391s -0.998 391s 0.774 391s 391s y: [2, 2, 1, 3] 391s 0.231 0.372 391s -0.179 -0.289 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.771 391s 0.860 391s 391s x1: [1, 2, 1, 1] 391s 0.918 391s 0.856 391s 391s y: [2, 2, 1, 1] 391s 0.708 0.790 391s 0.660 0.736 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.302 0.725 391s -0.160 0.863 391s -0.093 0.746 391s 0.330 -0.039 391s 391s x1: [2, 1, 1, 1] 391s -0.361 0.648 391s 391s y: [4, 1, 1, 1] 391s 0.579 0.618 0.517 -0.145 391s testing: mul_mat, [4, 2, 1, 3] = [4, 4, 1, 3] * [4, 2, 1, 3] 391s x0: [4, 4, 1, 3] 391s -0.847 0.720 -0.208 0.701 391s 0.974 0.829 -0.771 0.790 391s -0.132 -0.765 -0.359 -0.826 391s -0.984 -0.813 -0.003 0.942 391s 391s x1: [4, 2, 1, 3] 391s -0.781 -0.532 -0.037 0.993 391s -0.546 -0.795 0.473 0.149 391s 391s y: [4, 2, 1, 3] 391s 0.982 -0.389 -0.297 2.137 391s -0.104 -1.439 0.388 1.322 391s test-mul-mat0: iter:21/500 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.203 -0.189 391s 0.891 0.963 391s -0.883 -0.541 391s 391s x1: [2, 4, 1, 1] 391s -0.887 -0.508 391s 0.935 -0.440 391s -0.616 -0.833 391s 0.310 0.521 391s 391s y: [3, 4, 1, 1] 391s -0.084 -1.279 1.058 391s 0.273 0.409 -0.588 391s 0.033 -1.352 0.995 391s -0.036 0.778 -0.555 391s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.819 -0.006 391s -0.783 0.783 391s 391s x1: [2, 3, 4, 1] 391s -0.447 0.658 391s 0.215 0.934 391s 0.428 -0.388 391s 391s y: [2, 3, 4, 1] 391s -0.371 0.865 391s 0.170 0.563 391s 0.353 -0.639 391s testing: mul_mat, [4, 2, 4, 4] = [2, 4, 4, 4] * [2, 2, 4, 4] 391s x0: [2, 4, 4, 4] 391s -0.077 0.364 391s 0.857 0.085 391s -0.751 0.231 391s 0.344 -0.290 391s 391s x1: [2, 2, 4, 4] 391s 0.408 -0.148 391s -0.019 -0.958 391s 391s y: [4, 2, 4, 4] 391s -0.085 0.337 -0.341 0.183 391s -0.347 -0.098 -0.206 0.271 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.813 0.748 391s 0.386 0.038 391s 0.452 -0.327 391s -0.950 0.986 391s 391s x1: [2, 4, 1, 1] 391s -0.238 -0.631 391s -0.254 -0.834 391s 0.709 -0.256 391s -0.111 0.988 391s 391s y: [4, 4, 1, 1] 391s -0.278 -0.116 0.098 -0.395 391s -0.418 -0.129 0.158 -0.581 391s -0.768 0.264 0.404 -0.926 391s 0.829 -0.005 -0.373 1.079 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.348 -0.949 -0.048 -0.628 391s 391s x1: [4, 2, 4, 1] 391s -0.080 -0.249 0.656 -0.543 391s 0.694 0.024 0.224 0.179 391s 391s y: [1, 2, 4, 1] 391s 0.517 391s 0.096 391s testing: mul_mat, [3, 4, 4, 4] = [1, 3, 4, 4] * [1, 4, 4, 4] 391s x0: [1, 3, 4, 4] 391s 0.977 391s -0.652 391s -0.000 391s 391s x1: [1, 4, 4, 4] 391s -0.984 391s -0.035 391s 0.451 391s -0.412 391s 391s y: [3, 4, 4, 4] 391s -0.962 0.641 0.000 391s -0.034 0.023 0.000 391s 0.441 -0.294 -0.000 391s -0.403 0.268 0.000 391s test-mul-mat0: iter:22/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.019 -0.533 0.959 0.615 391s -0.650 -0.993 -0.403 0.233 391s 391s x1: [4, 1, 1, 1] 391s 0.789 -0.229 -0.211 -0.203 391s 391s y: [2, 1, 1, 1] 391s -0.220 -0.248 391s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.802 -0.976 -0.098 -0.795 391s -0.378 -0.865 0.806 0.241 391s 0.104 -0.630 0.070 -0.911 391s 391s x1: [4, 2, 3, 1] 391s 0.602 -0.381 -0.031 -0.436 391s 0.829 -0.014 -0.339 0.430 391s 391s y: [3, 2, 3, 1] 391s 0.239 -0.028 0.697 391s -0.960 -0.471 -0.320 391s testing: mul_mat, [1, 3, 3, 4] = [4, 1, 3, 4] * [4, 3, 3, 4] 391s x0: [4, 1, 3, 4] 391s -0.791 0.304 -0.679 -0.247 391s 391s x1: [4, 3, 3, 4] 391s -0.394 -0.986 0.033 0.710 391s -0.616 -0.898 0.799 0.415 391s -0.398 0.431 0.398 -0.171 391s 391s y: [1, 3, 3, 4] 391s -0.186 391s -0.430 391s 0.218 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.624 0.659 0.558 -0.280 391s 0.138 -0.216 0.089 0.294 391s 391s x1: [4, 1, 1, 1] 391s -0.979 0.924 -0.711 0.276 391s 391s y: [2, 1, 1, 1] 391s -0.476 -0.317 391s testing: mul_mat, [1, 4, 3, 1] = [2, 1, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 1, 3, 1] 391s 0.959 0.901 391s 391s x1: [2, 4, 3, 1] 391s 0.954 -0.136 391s -0.668 0.139 391s 0.761 0.018 391s 0.717 -0.535 391s 391s y: [1, 4, 3, 1] 391s 0.792 391s -0.515 391s 0.746 391s 0.206 391s testing: mul_mat, [4, 2, 3, 4] = [1, 4, 3, 4] * [1, 2, 3, 4] 391s x0: [1, 4, 3, 4] 391s 0.835 391s -0.764 391s 0.253 391s -0.117 391s 391s x1: [1, 2, 3, 4] 391s 0.002 391s 0.485 391s 391s y: [4, 2, 3, 4] 391s 0.001 -0.001 0.000 -0.000 391s 0.405 -0.371 0.123 -0.057 391s test-mul-mat0: iter:23/500 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.234 391s 0.299 391s 391s x1: [1, 3, 1, 1] 391s -0.912 391s -0.581 391s -0.047 391s 391s y: [2, 3, 1, 1] 391s 0.213 -0.273 391s 0.136 -0.174 391s 0.011 -0.014 391s testing: mul_mat, [4, 2, 3, 1] = [1, 4, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.990 391s -0.327 391s 0.616 391s -0.184 391s 391s x1: [1, 2, 3, 1] 391s -0.267 391s -0.010 391s 391s y: [4, 2, 3, 1] 391s -0.264 0.087 -0.164 0.049 391s -0.010 0.003 -0.006 0.002 391s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 3, 3, 1] 391s 0.828 391s -0.529 391s -0.212 391s 391s x1: [1, 4, 3, 1] 391s 0.681 391s -0.994 391s -0.587 391s -0.129 391s 391s y: [3, 4, 3, 1] 391s 0.564 -0.360 -0.144 391s -0.824 0.526 0.211 391s -0.486 0.310 0.124 391s -0.107 0.068 0.027 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.321 391s 391s x1: [1, 3, 1, 1] 391s 0.682 391s -0.170 391s 0.052 391s 391s y: [1, 3, 1, 1] 391s 0.219 391s -0.054 391s 0.017 391s testing: mul_mat, [4, 1, 3, 1] = [1, 4, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.431 391s -0.180 391s 0.903 391s -0.699 391s 391s x1: [1, 1, 3, 1] 391s 0.499 391s 391s y: [4, 1, 3, 1] 391s 0.215 -0.090 0.451 -0.349 391s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 3, 3, 1] 391s 0.986 0.736 0.166 0.467 391s 0.855 -0.331 0.489 0.961 391s -0.388 -0.898 -0.995 -0.855 391s 391s x1: [4, 1, 3, 1] 391s 0.165 0.211 -0.708 0.953 391s 391s y: [3, 1, 3, 1] 391s 0.645 0.641 -0.363 391s test-mul-mat0: iter:24/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.404 -0.566 0.859 -0.299 391s -0.828 -0.041 -0.710 0.863 391s 391s x1: [4, 1, 1, 1] 391s 0.135 0.464 0.928 0.862 391s 391s y: [2, 1, 1, 1] 391s 0.331 -0.046 391s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.859 -0.966 0.605 0.100 391s -0.676 0.468 0.246 0.866 391s 0.220 0.535 -0.638 -0.077 391s 391s x1: [4, 2, 3, 1] 391s -0.854 0.541 0.752 -0.711 391s -0.504 0.703 0.068 0.986 391s 391s y: [3, 2, 3, 1] 391s 0.595 0.400 -0.323 391s -0.107 1.539 0.145 391s testing: mul_mat, [3, 3, 3, 3] = [4, 3, 3, 3] * [4, 3, 3, 3] 391s x0: [4, 3, 3, 3] 391s -0.440 -0.372 0.374 -0.476 391s -0.564 -0.652 -0.125 -0.939 391s 0.894 0.852 0.873 0.627 391s 391s x1: [4, 3, 3, 3] 391s 0.418 -0.048 0.836 -0.362 391s -0.513 -0.802 0.560 0.090 391s -0.455 0.249 0.527 -0.776 391s 391s y: [3, 3, 3, 3] 391s 0.319 0.031 0.835 391s 0.691 0.658 -0.597 391s 0.674 0.757 -0.220 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.835 -0.869 0.079 -0.775 391s 0.154 0.779 0.926 0.315 391s -0.862 0.357 -0.211 -0.047 391s 391s x1: [4, 3, 1, 1] 391s -0.026 0.825 -0.533 -0.107 391s 0.383 -0.790 -0.711 -0.670 391s -0.722 -0.806 -0.162 -0.605 391s 391s y: [3, 3, 1, 1] 391s -0.654 0.111 0.434 391s 0.830 -1.425 -0.431 391s 1.759 -1.080 0.397 391s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 3, 3, 1] 391s 0.811 -0.280 0.918 391s -0.649 0.695 0.370 391s 0.981 0.848 0.994 391s 391s x1: [3, 4, 3, 1] 391s -0.360 0.538 0.994 391s 0.042 -0.863 0.191 391s -0.984 0.962 0.658 391s -0.091 0.345 0.868 391s 391s y: [3, 4, 3, 1] 391s 0.470 0.976 1.092 391s 0.451 -0.557 -0.501 391s -0.463 1.551 0.505 391s 0.626 0.620 1.065 391s testing: mul_mat, [1, 3, 3, 3] = [3, 1, 3, 3] * [3, 3, 3, 3] 391s x0: [3, 1, 3, 3] 391s 0.987 0.471 0.533 391s 391s x1: [3, 3, 3, 3] 391s 0.097 0.917 -0.544 391s 0.419 0.728 -0.194 391s 0.400 -0.552 -0.498 391s 391s y: [1, 3, 3, 3] 391s 0.238 391s 0.654 391s -0.131 391s test-mul-mat0: iter:25/500 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.546 0.533 391s 391s x1: [2, 3, 1, 1] 391s 0.619 -0.924 391s -0.011 -0.384 391s 0.819 -0.509 391s 391s y: [1, 3, 1, 1] 391s -0.831 391s -0.199 391s -0.718 391s testing: mul_mat, [1, 1, 4, 1] = [2, 1, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.570 0.968 391s 391s x1: [2, 1, 4, 1] 391s -0.015 -0.330 391s 391s y: [1, 1, 4, 1] 391s -0.310 391s testing: mul_mat, [3, 1, 4, 2] = [2, 3, 4, 2] * [2, 1, 4, 2] 391s x0: [2, 3, 4, 2] 391s -0.215 0.696 391s 0.567 -0.777 391s -0.077 0.230 391s 391s x1: [2, 1, 4, 2] 391s -0.540 0.981 391s 391s y: [3, 1, 4, 2] 391s 0.799 -1.068 0.268 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.644 -0.700 391s -0.538 -0.802 391s 391s x1: [2, 3, 1, 1] 391s 0.140 -0.646 391s 0.199 -0.586 391s 0.540 -0.504 391s 391s y: [2, 3, 1, 1] 391s 0.542 0.443 391s 0.538 0.362 391s 0.701 0.114 391s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 3, 4, 1] 391s -0.790 0.350 391s 0.199 0.553 391s -0.780 0.419 391s 391s x1: [2, 2, 4, 1] 391s -0.916 -0.964 391s 0.988 -0.509 391s 391s y: [3, 2, 4, 1] 391s 0.386 -0.715 0.310 391s -0.958 -0.085 -0.984 391s testing: mul_mat, [3, 2, 4, 2] = [3, 3, 4, 2] * [3, 2, 4, 2] 391s x0: [3, 3, 4, 2] 391s -0.757 0.676 -0.939 391s 0.070 -0.062 0.023 391s 0.120 -0.027 0.094 391s 391s x1: [3, 2, 4, 2] 391s 0.106 0.464 0.810 391s -0.535 -0.333 0.414 391s 391s y: [3, 2, 4, 2] 391s -0.526 -0.002 0.076 391s -0.209 -0.007 -0.016 391s test-mul-mat0: iter:26/500 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.898 391s 0.042 391s 391s x1: [1, 2, 1, 1] 391s -0.796 391s 0.779 391s 391s y: [2, 2, 1, 1] 391s 0.715 -0.033 391s -0.699 0.033 391s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 3, 2, 1] 391s 0.887 391s -0.312 391s 0.885 391s 391s x1: [1, 2, 2, 1] 391s -0.339 391s -0.020 391s 391s y: [3, 2, 2, 1] 391s -0.300 0.106 -0.300 391s -0.018 0.006 -0.018 391s testing: mul_mat, [3, 3, 2, 3] = [1, 3, 2, 3] * [1, 3, 2, 3] 391s x0: [1, 3, 2, 3] 391s 0.644 391s -0.225 391s 0.912 391s 391s x1: [1, 3, 2, 3] 391s 0.762 391s -0.301 391s -0.960 391s 391s y: [3, 3, 2, 3] 391s 0.491 -0.171 0.695 391s -0.194 0.068 -0.275 391s -0.619 0.216 -0.876 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.097 391s 0.179 391s -0.122 391s 391s x1: [1, 3, 1, 1] 391s 0.582 391s -0.048 391s -0.208 391s 391s y: [3, 3, 1, 1] 391s -0.056 0.104 -0.071 391s 0.005 -0.009 0.006 391s 0.020 -0.037 0.025 391s testing: mul_mat, [2, 1, 2, 1] = [3, 2, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.012 0.049 -0.144 391s -0.448 -0.303 -0.987 391s 391s x1: [3, 1, 2, 1] 391s 0.726 -0.901 0.708 391s 391s y: [2, 1, 2, 1] 391s -0.155 -0.751 391s testing: mul_mat, [2, 3, 2, 3] = [2, 2, 2, 3] * [2, 3, 2, 3] 391s x0: [2, 2, 2, 3] 391s -0.401 0.140 391s 0.231 -0.403 391s 391s x1: [2, 3, 2, 3] 391s -0.023 -0.713 391s -0.280 -0.857 391s -0.363 -0.699 391s 391s y: [2, 3, 2, 3] 391s -0.090 0.282 391s -0.007 0.281 391s 0.048 0.198 391s test-mul-mat0: iter:27/500 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.520 391s -0.683 391s -0.196 391s 391s x1: [1, 2, 1, 1] 391s -0.079 391s 0.086 391s 391s y: [3, 2, 1, 1] 391s -0.041 0.054 0.015 391s 0.045 -0.059 -0.017 391s testing: mul_mat, [2, 3, 2, 1] = [1, 2, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 2, 2, 1] 391s -0.837 391s -0.426 391s 391s x1: [1, 3, 2, 1] 391s -0.883 391s -0.845 391s 0.375 391s 391s y: [2, 3, 2, 1] 391s 0.740 0.377 391s 0.707 0.360 391s -0.314 -0.160 391s testing: mul_mat, [1, 2, 2, 2] = [1, 1, 2, 2] * [1, 2, 2, 2] 391s x0: [1, 1, 2, 2] 391s -0.244 391s 391s x1: [1, 2, 2, 2] 391s -0.607 391s -0.313 391s 391s y: [1, 2, 2, 2] 391s 0.148 391s 0.077 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.125 391s -0.990 391s 0.095 391s 0.442 391s 391s x1: [1, 1, 1, 1] 391s -0.654 391s 391s y: [4, 1, 1, 1] 391s -0.082 0.647 -0.062 -0.289 391s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 1, 2, 1] 391s 0.847 0.672 0.993 -0.759 391s 391s x1: [4, 1, 2, 1] 391s -0.186 0.211 0.597 -0.811 391s 391s y: [1, 1, 2, 1] 391s 1.193 391s testing: mul_mat, [1, 4, 2, 2] = [1, 1, 2, 2] * [1, 4, 2, 2] 391s x0: [1, 1, 2, 2] 391s -0.288 391s 391s x1: [1, 4, 2, 2] 391s -0.319 391s -0.722 391s -0.916 391s 0.917 391s 391s y: [1, 4, 2, 2] 391s 0.092 391s 0.208 391s 0.264 391s -0.264 391s test-mul-mat0: iter:28/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.901 0.335 -0.198 -0.703 391s 0.045 -0.958 0.938 0.568 391s -0.506 0.176 -0.721 -0.406 391s 391s x1: [4, 3, 1, 1] 391s 0.988 -0.927 -0.410 0.488 391s 0.713 -0.802 -0.831 0.992 391s -0.717 -0.914 0.026 0.636 391s 391s y: [3, 3, 1, 1] 391s -1.463 0.825 -0.566 391s -1.444 0.583 -0.306 391s -0.112 1.228 -0.074 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.875 -0.542 -0.563 -0.099 391s -0.906 0.094 -0.000 0.429 391s 391s x1: [4, 3, 1, 1] 391s -0.289 -0.735 0.155 -0.294 391s -0.903 0.387 -0.306 -0.830 391s 0.978 -0.818 0.884 -0.824 391s 391s y: [2, 3, 1, 1] 391s 0.087 0.067 391s -0.745 0.498 391s 0.883 -1.316 391s testing: mul_mat, [4, 2, 1, 3] = [4, 4, 1, 3] * [4, 2, 1, 3] 391s x0: [4, 4, 1, 3] 391s 0.493 0.513 -0.726 -0.414 391s -0.488 0.703 -0.518 0.810 391s -0.823 -0.580 0.045 -0.782 391s -0.148 0.456 -0.461 0.878 391s 391s x1: [4, 2, 1, 3] 391s 0.896 0.297 -0.526 0.938 391s 0.235 -0.959 -0.568 -0.589 391s 391s y: [4, 2, 1, 3] 391s 0.587 0.805 -1.666 1.069 391s 0.280 -0.971 0.797 -0.727 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.536 -0.214 -0.543 -0.021 391s -0.462 -0.619 0.268 0.612 391s 0.634 -0.136 -0.687 -0.699 391s 391s x1: [4, 4, 1, 1] 391s 0.113 -0.540 0.330 -0.080 391s -0.913 -0.740 0.580 0.261 391s 0.894 -0.024 -0.844 -0.416 391s -0.016 0.231 -0.233 -0.495 391s 391s y: [3, 4, 1, 1] 391s -0.002 0.322 -0.026 391s -0.651 1.195 -1.059 391s 0.951 -0.880 1.441 391s 0.079 -0.502 0.465 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.396 0.371 -0.093 391s 0.872 -0.623 -0.085 391s -0.425 0.156 -0.210 391s 391s x1: [3, 4, 1, 1] 391s 0.380 -0.465 0.413 391s 0.839 0.866 -0.667 391s 0.927 -0.874 0.912 391s 0.187 -0.980 -0.112 391s 391s y: [3, 4, 1, 1] 391s -0.361 0.585 -0.321 391s 0.052 0.250 -0.082 391s -0.776 1.275 -0.722 391s -0.428 0.783 -0.209 391s testing: mul_mat, [2, 3, 1, 3] = [3, 2, 1, 3] * [3, 3, 1, 3] 391s x0: [3, 2, 1, 3] 391s 0.605 0.325 0.273 391s 0.978 0.298 -0.021 391s 391s x1: [3, 3, 1, 3] 391s 0.693 0.295 0.654 391s -0.850 -0.437 0.967 391s 0.129 -0.824 -0.733 391s 391s y: [2, 3, 1, 3] 391s 0.694 0.752 391s -0.392 -0.982 391s -0.390 -0.104 391s test-mul-mat0: iter:29/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.667 0.356 -0.250 391s 0.397 0.892 -0.645 391s 0.375 0.217 0.654 391s 391s x1: [3, 1, 1, 1] 391s -0.275 0.647 0.976 391s 391s y: [3, 1, 1, 1] 391s 0.170 -0.162 0.675 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.248 -0.800 -0.480 391s -0.845 0.910 -0.478 391s 391s x1: [3, 3, 1, 1] 391s -0.352 -0.804 0.102 391s -0.515 0.141 0.680 391s -0.503 -0.646 -0.234 391s 391s y: [2, 3, 1, 1] 391s 0.507 -0.483 391s -0.567 0.239 391s 0.504 -0.050 391s testing: mul_mat, [1, 2, 1, 2] = [3, 1, 1, 2] * [3, 2, 1, 2] 391s x0: [3, 1, 1, 2] 391s 0.522 -0.943 -0.613 391s 391s x1: [3, 2, 1, 2] 391s -0.013 0.081 0.247 391s -0.366 0.057 0.298 391s 391s y: [1, 2, 1, 2] 391s -0.234 391s -0.428 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.605 -0.412 -0.539 391s 391s x1: [3, 1, 1, 1] 391s -0.141 -0.161 0.368 391s 391s y: [1, 1, 1, 1] 391s -0.218 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.630 391s 0.603 391s -0.204 391s -0.123 391s 391s x1: [1, 3, 1, 1] 391s 0.805 391s 0.108 391s -0.384 391s 391s y: [4, 3, 1, 1] 391s 0.507 0.485 -0.165 -0.099 391s 0.068 0.065 -0.022 -0.013 391s -0.242 -0.232 0.079 0.047 391s testing: mul_mat, [2, 1, 1, 2] = [4, 2, 1, 2] * [4, 1, 1, 2] 391s x0: [4, 2, 1, 2] 391s 0.306 -0.317 0.810 -0.437 391s -0.907 0.692 0.843 -0.331 391s 391s x1: [4, 1, 1, 2] 391s -0.764 0.853 -0.825 0.203 391s 391s y: [2, 1, 1, 2] 391s -1.261 0.520 391s test-mul-mat0: iter:30/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.293 0.158 391s -0.126 0.386 391s 0.841 -0.434 391s 0.196 0.684 391s 391s x1: [2, 4, 1, 1] 391s -0.397 -0.866 391s -0.202 -0.544 391s -0.691 -0.999 391s -0.278 -0.468 391s 391s y: [4, 4, 1, 1] 391s -0.253 -0.284 0.042 -0.670 391s -0.145 -0.184 0.066 -0.412 391s -0.360 -0.298 -0.148 -0.818 391s -0.155 -0.145 -0.031 -0.375 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.208 0.048 391s 391s x1: [2, 4, 1, 1] 391s 0.130 0.864 391s -0.634 0.061 391s 0.746 -0.347 391s 0.581 -0.911 391s 391s y: [1, 4, 1, 1] 391s 0.068 391s -0.129 391s 0.139 391s 0.077 391s testing: mul_mat, [3, 1, 1, 2] = [2, 3, 1, 2] * [2, 1, 1, 2] 391s x0: [2, 3, 1, 2] 391s -0.593 0.611 391s -0.091 -0.592 391s -0.667 0.441 391s 391s x1: [2, 1, 1, 2] 391s -0.448 -0.534 391s 391s y: [3, 1, 1, 2] 391s -0.060 0.357 0.064 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.817 0.143 391s 391s x1: [2, 3, 1, 1] 391s 0.416 -0.794 391s -0.876 -0.455 391s -0.929 -0.510 391s 391s y: [1, 3, 1, 1] 391s 0.227 391s -0.781 391s -0.832 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.604 391s -0.045 391s 391s x1: [1, 2, 1, 1] 391s 0.187 391s 0.907 391s 391s y: [2, 2, 1, 1] 391s -0.113 -0.008 391s -0.548 -0.041 391s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 391s x0: [2, 4, 1, 2] 391s -0.030 0.538 391s 0.205 0.812 391s -0.629 0.354 391s 0.378 0.163 391s 391s x1: [2, 1, 1, 2] 391s -0.561 -0.138 391s 391s y: [4, 1, 1, 2] 391s -0.058 -0.227 0.304 -0.234 391s test-mul-mat0: iter:31/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.354 0.130 -0.065 391s -0.784 0.694 -0.471 391s 0.678 -0.336 0.734 391s 391s x1: [3, 1, 1, 1] 391s -0.392 -0.041 -0.825 391s 391s y: [3, 1, 1, 1] 391s -0.090 0.667 -0.858 391s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.364 0.372 0.864 391s 0.413 0.414 0.136 391s 391s x1: [3, 3, 3, 1] 391s -0.951 -0.958 0.272 391s 0.861 0.396 -0.565 391s -0.728 0.600 -0.391 391s 391s y: [2, 3, 3, 1] 391s -0.467 -0.752 391s -0.028 0.443 391s -0.380 -0.106 391s testing: mul_mat, [3, 2, 3, 4] = [3, 3, 3, 4] * [3, 2, 3, 4] 391s x0: [3, 3, 3, 4] 391s -0.865 0.953 -0.520 391s -0.476 -0.135 0.724 391s 0.624 0.676 -0.541 391s 391s x1: [3, 2, 3, 4] 391s -0.668 -0.830 -0.681 391s -0.497 -0.715 -0.797 391s 391s y: [3, 2, 3, 4] 391s 0.141 -0.063 -0.610 391s 0.163 -0.244 -0.362 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.477 -0.755 0.040 391s 391s x1: [3, 3, 1, 1] 391s -0.923 0.834 -0.745 391s -0.654 -0.899 0.852 391s 0.088 0.776 -0.175 391s 391s y: [1, 3, 1, 1] 391s -1.101 391s 0.401 391s -0.551 391s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 3, 3, 1] 391s -0.435 391s 0.270 391s 0.186 391s 391s x1: [1, 3, 3, 1] 391s -0.135 391s -0.012 391s 0.077 391s 391s y: [3, 3, 3, 1] 391s 0.059 -0.036 -0.025 391s 0.005 -0.003 -0.002 391s -0.034 0.021 0.014 391s testing: mul_mat, [1, 1, 3, 4] = [3, 1, 3, 4] * [3, 1, 3, 4] 391s x0: [3, 1, 3, 4] 391s -0.108 0.010 0.077 391s 391s x1: [3, 1, 3, 4] 391s 0.117 -0.742 -0.117 391s 391s y: [1, 1, 3, 4] 391s -0.029 391s test-mul-mat0: iter:32/500 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.138 -0.727 391s 391s x1: [2, 2, 1, 1] 391s 0.448 -0.384 391s -0.117 -0.295 391s 391s y: [1, 2, 1, 1] 391s 0.217 391s 0.231 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.434 -0.029 391s -0.640 -0.293 391s 0.584 0.876 391s 0.527 -0.837 391s 391s x1: [2, 1, 1, 1] 391s 0.987 -0.935 391s 391s y: [4, 1, 1, 1] 391s -0.401 -0.357 -0.243 1.303 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.489 -0.499 391s -0.456 -0.655 391s 0.949 0.160 391s 391s x1: [2, 4, 1, 1] 391s -0.231 -0.756 391s -0.184 0.009 391s 0.854 -0.520 391s -0.674 -0.173 391s 391s y: [3, 4, 1, 1] 391s 0.264 0.600 -0.341 391s -0.094 0.078 -0.173 391s 0.677 -0.049 0.727 391s -0.243 0.421 -0.668 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.735 0.400 391s 391s x1: [2, 3, 1, 1] 391s 0.228 -0.346 391s -0.153 -0.909 391s -0.073 -0.166 391s 391s y: [1, 3, 1, 1] 391s -0.306 391s -0.251 391s -0.013 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.289 391s -0.499 391s 0.633 391s -0.127 391s 391s x1: [1, 2, 1, 1] 391s 0.126 391s -0.375 391s 391s y: [4, 2, 1, 1] 391s -0.037 -0.063 0.080 -0.016 391s 0.108 0.187 -0.237 0.047 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.831 -0.498 0.046 -0.794 391s -0.395 -0.906 -0.560 0.668 391s -0.442 0.059 0.008 0.663 391s 391s x1: [4, 1, 1, 1] 391s 0.270 0.877 0.690 -0.721 391s 391s y: [3, 1, 1, 1] 391s -0.056 -1.769 -0.540 391s test-mul-mat0: iter:33/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.393 391s 391s x1: [1, 4, 1, 1] 391s 0.914 391s -0.145 391s 0.286 391s 0.040 391s 391s y: [1, 4, 1, 1] 391s 0.359 391s -0.057 391s 0.112 391s 0.016 391s testing: mul_mat, [3, 1, 3, 1] = [1, 3, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 3, 3, 1] 391s 0.251 391s -0.097 391s -0.686 391s 391s x1: [1, 1, 3, 1] 391s 0.751 391s 391s y: [3, 1, 3, 1] 391s 0.189 -0.073 -0.515 391s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 3, 3, 1] 391s 0.512 391s 0.177 391s -0.991 391s 391s x1: [1, 3, 3, 1] 391s -0.771 391s -0.749 391s -0.968 391s 391s y: [3, 3, 3, 1] 391s -0.394 -0.137 0.764 391s -0.383 -0.133 0.743 391s -0.496 -0.172 0.960 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.785 391s 0.737 391s -0.391 391s 391s x1: [1, 3, 1, 1] 391s 0.828 391s -0.118 391s 0.423 391s 391s y: [3, 3, 1, 1] 391s 0.650 0.610 -0.323 391s -0.093 -0.087 0.046 391s 0.332 0.312 -0.165 391s testing: mul_mat, [4, 1, 3, 1] = [3, 4, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.476 -0.926 0.922 391s 0.979 0.556 -0.593 391s -0.956 0.944 0.132 391s -0.234 -0.917 0.769 391s 391s x1: [3, 1, 3, 1] 391s -0.581 -0.780 -0.256 391s 391s y: [4, 1, 3, 1] 391s 0.210 -0.851 -0.214 0.654 391s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.998 0.754 0.840 0.961 391s 0.998 0.549 0.135 0.470 391s -0.651 -0.660 -0.525 0.561 391s -0.170 -0.561 0.442 0.791 391s 391s x1: [4, 3, 3, 1] 391s 0.544 0.940 -0.216 0.467 391s -0.653 0.915 0.236 -0.389 391s 0.846 -0.167 0.756 0.200 391s 391s y: [4, 3, 3, 1] 391s 1.518 1.249 -0.600 -0.346 391s -0.138 -0.300 -0.522 -0.606 391s 1.545 0.948 -0.725 0.442 391s test-mul-mat0: iter:34/500 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.696 0.363 0.754 391s 391s x1: [3, 2, 1, 1] 391s -0.448 0.345 -0.382 391s 0.727 0.444 0.296 391s 391s y: [1, 2, 1, 1] 391s 0.148 391s -0.121 391s testing: mul_mat, [2, 1, 2, 1] = [3, 2, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.501 -0.466 -0.957 391s -0.768 0.979 0.167 391s 391s x1: [3, 1, 2, 1] 391s 0.163 0.426 -0.632 391s 391s y: [2, 1, 2, 1] 391s 0.488 0.187 391s testing: mul_mat, [1, 2, 2, 4] = [3, 1, 2, 4] * [3, 2, 2, 4] 391s x0: [3, 1, 2, 4] 391s -0.709 0.344 0.232 391s 391s x1: [3, 2, 2, 4] 391s -0.533 -0.596 0.815 391s 0.812 0.022 0.541 391s 391s y: [1, 2, 2, 4] 391s 0.362 391s -0.443 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.637 -0.696 -0.376 391s 0.280 0.928 0.536 391s 391s x1: [3, 1, 1, 1] 391s 0.960 0.801 0.788 391s 391s y: [2, 1, 1, 1] 391s -0.241 1.434 391s testing: mul_mat, [2, 3, 2, 1] = [2, 2, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.713 -0.022 391s 0.280 0.713 391s 391s x1: [2, 3, 2, 1] 391s -0.285 -0.947 391s -0.036 0.166 391s -1.000 0.574 391s 391s y: [2, 3, 2, 1] 391s -0.182 -0.755 391s -0.029 0.109 391s -0.725 0.129 391s testing: mul_mat, [3, 2, 2, 4] = [2, 3, 2, 4] * [2, 2, 2, 4] 391s x0: [2, 3, 2, 4] 391s -0.759 0.850 391s 0.484 0.798 391s -0.211 0.407 391s 391s x1: [2, 2, 2, 4] 391s -0.281 0.053 391s -0.671 -0.990 391s 391s y: [3, 2, 2, 4] 391s 0.258 -0.093 0.081 391s -0.333 -1.114 -0.261 391s test-mul-mat0: iter:35/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.346 0.005 -0.682 -0.491 391s 391s x1: [4, 3, 1, 1] 391s 0.001 -0.027 -0.375 -0.427 391s -0.952 0.923 0.189 0.037 391s -0.305 0.904 -0.734 -0.049 391s 391s y: [1, 3, 1, 1] 391s 0.466 391s -0.472 391s 0.423 391s testing: mul_mat, [4, 1, 4, 1] = [4, 4, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.453 0.214 0.022 0.737 391s -0.749 0.718 0.641 -0.483 391s -0.331 0.048 0.863 0.673 391s 0.366 -0.628 -0.621 0.336 391s 391s x1: [4, 1, 4, 1] 391s -0.294 0.970 -0.958 -0.131 391s 391s y: [4, 1, 4, 1] 391s 0.223 0.366 -0.771 -0.166 391s testing: mul_mat, [4, 4, 4, 3] = [4, 4, 4, 3] * [4, 4, 4, 3] 391s x0: [4, 4, 4, 3] 391s -0.176 0.174 -0.362 -0.758 391s -0.155 -0.788 -0.774 0.357 391s 0.415 0.698 0.995 0.590 391s 0.711 0.269 -0.808 0.400 391s 391s x1: [4, 4, 4, 3] 391s -0.244 0.320 -0.904 -0.473 391s -0.411 0.492 -0.033 -0.094 391s -0.249 0.080 -0.411 -0.617 391s 0.572 0.607 -0.572 0.313 391s 391s y: [4, 4, 4, 3] 391s 0.785 0.316 -1.056 0.454 391s 0.241 -0.332 0.085 -0.171 391s 0.675 0.073 -0.821 -0.070 391s -0.025 -0.013 0.277 1.158 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.279 0.936 0.918 0.773 391s -0.475 -0.201 0.952 -0.055 391s 391s x1: [4, 4, 1, 1] 391s 0.040 -0.821 0.262 0.146 391s -0.539 -0.937 -0.201 0.310 391s -0.812 -0.619 0.888 -0.804 391s 0.949 -0.253 -0.560 -0.301 391s 391s y: [2, 4, 1, 1] 391s -0.404 0.388 391s -0.973 0.236 391s -0.612 1.400 391s -0.719 -0.916 391s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.303 0.009 391s -0.398 0.345 391s 0.122 -0.932 391s -0.248 -0.437 391s 391s x1: [2, 4, 4, 1] 391s -0.675 0.024 391s 0.928 0.368 391s -0.955 -0.945 391s -0.593 -0.777 391s 391s y: [4, 4, 4, 1] 391s 0.205 0.277 -0.105 0.157 391s -0.278 -0.243 -0.230 -0.391 391s 0.281 0.054 0.765 0.651 391s 0.172 -0.032 0.651 0.487 391s testing: mul_mat, [2, 2, 4, 3] = [4, 2, 4, 3] * [4, 2, 4, 3] 391s x0: [4, 2, 4, 3] 391s -0.092 0.009 -0.303 -0.002 391s -0.261 -0.002 0.730 0.028 391s 391s x1: [4, 2, 4, 3] 391s 0.257 -0.006 0.661 -0.734 391s -0.661 0.729 -0.171 -0.632 391s 391s y: [2, 2, 4, 3] 391s -0.222 0.394 391s 0.121 0.029 391s test-mul-mat0: iter:36/500 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.084 -0.671 0.317 391s 0.870 -0.249 0.021 391s 0.748 0.900 0.418 391s 391s x1: [3, 2, 1, 1] 391s 0.813 -0.358 -0.359 391s 0.784 0.647 -0.021 391s 391s y: [3, 2, 1, 1] 391s 0.194 0.789 0.137 391s -0.375 0.521 1.161 391s testing: mul_mat, [1, 3, 2, 1] = [3, 1, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 1, 2, 1] 391s -0.301 -0.492 0.944 391s 391s x1: [3, 3, 2, 1] 391s 0.161 -0.369 0.545 391s -0.255 -0.172 -0.225 391s -0.900 0.735 0.389 391s 391s y: [1, 3, 2, 1] 391s 0.647 391s -0.052 391s 0.277 391s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 4, 2, 1] 391s -0.882 0.245 0.308 391s -0.053 -0.980 0.408 391s -0.319 0.409 -0.030 391s -0.718 -0.924 0.138 391s 391s x1: [3, 1, 2, 1] 391s 0.653 0.588 -0.718 391s 391s y: [4, 1, 2, 1] 391s -0.653 -0.904 0.054 -1.110 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.342 -0.059 0.632 391s -0.255 -0.178 0.786 391s 0.526 -0.336 0.782 391s 391s x1: [3, 4, 1, 1] 391s 0.876 0.425 -0.628 391s 0.277 -0.088 -0.928 391s 0.840 0.030 0.317 391s 0.148 0.977 0.336 391s 391s y: [3, 4, 1, 1] 391s -0.722 -0.793 -0.174 391s -0.676 -0.784 -0.551 391s -0.089 0.029 0.680 391s 0.104 0.052 0.013 391s testing: mul_mat, [2, 3, 2, 1] = [3, 2, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.540 -0.263 0.425 391s -0.321 0.118 -0.737 391s 391s x1: [3, 3, 2, 1] 391s 0.374 -0.722 -0.334 391s 0.815 0.278 0.494 391s -0.962 0.115 0.583 391s 391s y: [2, 3, 2, 1] 391s -0.154 0.041 391s -0.304 -0.592 391s 0.737 -0.108 391s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.304 -0.568 391s 391s x1: [2, 3, 2, 1] 391s 0.064 -0.690 391s 0.375 -0.121 391s 0.588 -0.131 391s 391s y: [1, 3, 2, 1] 391s 0.372 391s -0.045 391s -0.104 391s test-mul-mat0: iter:37/500 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.874 0.163 -0.494 -0.843 391s 391s x1: [4, 1, 1, 1] 391s 0.974 -0.243 0.879 -0.763 391s 391s y: [1, 1, 1, 1] 391s -0.682 391s testing: mul_mat, [4, 1, 3, 1] = [4, 4, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.777 -0.300 0.405 0.280 391s 0.879 0.181 -0.483 0.518 391s 0.636 -0.509 -0.724 0.515 391s -0.273 0.092 0.641 0.891 391s 391s x1: [4, 1, 3, 1] 391s 0.872 0.570 -0.533 0.247 391s 391s y: [4, 1, 3, 1] 391s -0.996 1.256 0.777 -0.307 391s testing: mul_mat, [2, 4, 3, 3] = [4, 2, 3, 3] * [4, 4, 3, 3] 391s x0: [4, 2, 3, 3] 391s 0.048 0.527 -0.941 -0.601 391s -0.239 -0.581 0.748 -0.731 391s 391s x1: [4, 4, 3, 3] 391s 0.842 -0.924 0.314 -0.892 391s 0.086 -0.068 -0.571 -0.787 391s 0.917 0.510 0.906 0.102 391s -0.406 -0.428 -0.491 -0.623 391s 391s y: [2, 4, 3, 3] 391s -0.206 1.223 391s 0.978 0.167 391s -0.600 0.088 391s 0.590 0.433 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.677 -0.040 -0.264 0.564 391s 391s x1: [4, 2, 1, 1] 391s -0.661 0.033 0.004 0.410 391s 0.920 0.481 -0.707 0.026 391s 391s y: [1, 2, 1, 1] 391s -0.219 391s 0.805 391s testing: mul_mat, [2, 4, 3, 1] = [1, 2, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.485 391s -0.712 391s 391s x1: [1, 4, 3, 1] 391s 0.014 391s 0.395 391s -0.505 391s -0.414 391s 391s y: [2, 4, 3, 1] 391s -0.007 -0.010 391s -0.191 -0.281 391s 0.245 0.360 391s 0.201 0.295 391s testing: mul_mat, [2, 1, 3, 3] = [2, 2, 3, 3] * [2, 1, 3, 3] 391s x0: [2, 2, 3, 3] 391s -0.419 0.453 391s 0.695 -0.998 391s 391s x1: [2, 1, 3, 3] 391s 0.909 -0.749 391s 391s y: [2, 1, 3, 3] 391s -0.720 1.380 391s test-mul-mat0: iter:38/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.789 0.010 391s -0.182 -0.008 391s -0.005 -0.977 391s 391s x1: [2, 2, 1, 1] 391s -0.421 -0.338 391s -0.373 0.881 391s 391s y: [3, 2, 1, 1] 391s 0.329 0.079 0.332 391s 0.304 0.061 -0.858 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.459 -0.094 391s 391s x1: [2, 3, 1, 1] 391s -0.354 -0.207 391s -0.774 -0.262 391s 0.419 -0.180 391s 391s y: [1, 3, 1, 1] 391s -0.143 391s -0.331 391s 0.209 391s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 391s x0: [2, 4, 1, 2] 391s -0.759 0.173 391s -0.800 0.427 391s -0.126 -0.856 391s -0.994 0.536 391s 391s x1: [2, 1, 1, 2] 391s -0.374 -0.275 391s 391s y: [4, 1, 1, 2] 391s 0.236 0.181 0.282 0.224 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.798 0.747 391s 391s x1: [2, 4, 1, 1] 391s -0.749 -0.380 391s -0.315 -0.524 391s 0.358 -0.895 391s 0.296 0.120 391s 391s y: [1, 4, 1, 1] 391s 0.314 391s -0.140 391s -0.954 391s -0.147 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.733 391s -0.983 391s 391s x1: [1, 2, 1, 1] 391s -0.712 391s 0.775 391s 391s y: [2, 2, 1, 1] 391s -0.522 0.700 391s 0.568 -0.762 391s testing: mul_mat, [1, 1, 1, 2] = [2, 1, 1, 2] * [2, 1, 1, 2] 391s x0: [2, 1, 1, 2] 391s 0.450 -0.485 391s 391s x1: [2, 1, 1, 2] 391s -0.894 0.933 391s 391s y: [1, 1, 1, 2] 391s -0.854 391s test-mul-mat0: iter:39/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.188 -0.120 0.782 391s -0.084 -0.823 -0.098 391s -0.520 -0.621 -0.351 391s 391s x1: [3, 1, 1, 1] 391s 0.804 -0.874 -0.596 391s 391s y: [3, 1, 1, 1] 391s -0.513 0.710 0.334 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.064 0.182 0.358 391s -0.599 0.108 0.188 391s 391s x1: [3, 3, 1, 1] 391s -0.232 -0.846 0.582 391s -0.499 -0.829 0.688 391s -0.567 0.614 -0.333 391s 391s y: [2, 3, 1, 1] 391s 0.039 0.157 391s 0.063 0.339 391s -0.044 0.343 391s testing: mul_mat, [4, 2, 1, 4] = [3, 4, 1, 4] * [3, 2, 1, 4] 391s x0: [3, 4, 1, 4] 391s -0.897 0.869 -0.461 391s -0.495 -0.023 0.727 391s 0.140 -0.763 -0.281 391s 0.902 0.902 -0.864 391s 391s x1: [3, 2, 1, 4] 391s -0.365 0.261 -0.008 391s -0.238 0.664 0.416 391s 391s y: [4, 2, 1, 4] 391s 0.557 0.169 -0.248 -0.087 391s 0.599 0.405 -0.657 0.024 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.919 -0.792 0.224 391s -0.369 0.217 -0.723 391s 391s x1: [3, 4, 1, 1] 391s -0.350 -0.994 -0.911 391s -0.164 -0.324 0.611 391s 0.354 0.583 -0.443 391s 0.891 0.161 -0.458 391s 391s y: [2, 4, 1, 1] 391s 0.262 0.572 391s 0.243 -0.452 391s -0.236 0.316 391s 0.589 0.037 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.701 0.080 391s 0.445 -0.225 391s 391s x1: [2, 3, 1, 1] 391s -0.758 -0.043 391s -0.417 0.986 391s -0.408 -0.209 391s 391s y: [2, 3, 1, 1] 391s -0.535 -0.328 391s -0.213 -0.408 391s -0.303 -0.135 391s testing: mul_mat, [2, 2, 1, 4] = [2, 2, 1, 4] * [2, 2, 1, 4] 391s x0: [2, 2, 1, 4] 391s 0.336 0.412 391s 0.921 0.578 391s 391s x1: [2, 2, 1, 4] 391s 0.267 0.731 391s -0.219 0.356 391s 391s y: [2, 2, 1, 4] 391s 0.391 0.668 391s 0.073 0.004 391s test-mul-mat0: iter:40/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.223 391s 0.347 391s -0.459 391s 0.423 391s 391s x1: [1, 1, 1, 1] 391s 0.822 391s 391s y: [4, 1, 1, 1] 391s 0.183 0.285 -0.377 0.348 391s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 4, 3, 1] 391s -0.203 391s 0.451 391s 0.551 391s 0.029 391s 391s x1: [1, 4, 3, 1] 391s 0.414 391s 0.469 391s 0.543 391s -0.889 391s 391s y: [4, 4, 3, 1] 391s -0.084 0.186 0.228 0.012 391s -0.095 0.211 0.258 0.013 391s -0.110 0.245 0.299 0.016 391s 0.180 -0.401 -0.490 -0.026 391s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 4, 3, 1] 391s -0.632 391s 0.933 391s 0.018 391s -0.621 391s 391s x1: [1, 4, 3, 1] 391s 0.204 391s 0.738 391s 0.630 391s -0.449 391s 391s y: [4, 4, 3, 1] 391s -0.129 0.191 0.004 -0.127 391s -0.467 0.689 0.013 -0.459 391s -0.398 0.588 0.011 -0.391 391s 0.284 -0.419 -0.008 0.279 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.076 391s 391s x1: [1, 4, 1, 1] 391s -0.268 391s 0.568 391s -0.331 391s -0.886 391s 391s y: [1, 4, 1, 1] 391s -0.020 391s 0.043 391s -0.025 391s -0.067 391s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.405 391s 0.758 391s 391s x1: [1, 1, 3, 1] 391s 0.318 391s 391s y: [2, 1, 3, 1] 391s -0.129 0.242 391s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.718 -0.382 391s -0.567 0.124 391s 391s x1: [2, 1, 3, 1] 391s -0.305 0.830 391s 391s y: [2, 1, 3, 1] 391s -0.536 0.275 391s test-mul-mat0: iter:41/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.779 -0.381 391s 0.645 0.985 391s 0.655 0.087 391s 0.703 -0.913 391s 391s x1: [2, 4, 1, 1] 391s -0.723 -0.143 391s 0.544 -0.964 391s -0.308 0.887 391s -0.609 -0.398 391s 391s y: [4, 4, 1, 1] 391s 0.617 -0.607 -0.485 -0.378 391s -0.057 -0.598 0.272 1.262 391s -0.098 0.675 -0.124 -1.026 391s 0.626 -0.786 -0.434 -0.065 391s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.278 0.937 391s 391s x1: [2, 4, 4, 1] 391s 0.706 -0.173 391s 0.469 -0.471 391s 0.316 0.297 391s -0.916 0.374 391s 391s y: [1, 4, 4, 1] 391s -0.359 391s -0.572 391s 0.190 391s 0.605 391s testing: mul_mat, [4, 1, 4, 2] = [2, 4, 4, 2] * [2, 1, 4, 2] 391s x0: [2, 4, 4, 2] 391s 0.969 -0.364 391s 0.111 0.222 391s -0.956 0.916 391s 0.944 0.981 391s 391s x1: [2, 1, 4, 2] 391s 0.502 0.178 391s 391s y: [4, 1, 4, 2] 391s 0.422 0.095 -0.317 0.648 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.269 0.730 391s -0.248 0.476 391s 0.697 0.476 391s 391s x1: [2, 4, 1, 1] 391s -0.916 -0.053 391s -0.214 0.944 391s -0.018 -0.504 391s 0.841 -0.020 391s 391s y: [3, 4, 1, 1] 391s -0.285 0.202 -0.664 391s 0.632 0.503 0.300 391s -0.373 -0.235 -0.252 391s 0.211 -0.218 0.577 391s testing: mul_mat, [3, 2, 4, 1] = [3, 3, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.062 -0.332 -0.647 391s -0.775 0.701 0.674 391s -0.765 0.711 0.368 391s 391s x1: [3, 2, 4, 1] 391s 0.684 0.973 0.657 391s -0.910 -0.258 -0.358 391s 391s y: [3, 2, 4, 1] 391s -0.790 0.594 0.410 391s 0.373 0.283 0.381 391s testing: mul_mat, [4, 3, 4, 2] = [3, 4, 4, 2] * [3, 3, 4, 2] 391s x0: [3, 4, 4, 2] 391s -0.619 0.003 -0.940 391s -0.262 0.910 -0.860 391s 0.832 0.892 -0.065 391s -0.382 -0.602 -0.166 391s 391s x1: [3, 3, 4, 2] 391s 0.039 0.568 0.087 391s -0.594 -0.989 0.503 391s -0.583 -0.250 -0.729 391s 391s y: [4, 3, 4, 2] 391s -0.104 0.432 0.533 -0.371 391s -0.109 -1.177 -1.408 0.739 391s 1.046 0.552 -0.661 0.494 391s test-mul-mat0: iter:42/500 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.110 -0.315 -0.224 0.451 391s 0.649 -0.962 0.996 0.800 391s 391s x1: [4, 4, 1, 1] 391s 0.103 -0.917 -0.438 -0.132 391s -0.603 -0.302 0.512 -0.303 391s -0.724 -0.351 0.851 -0.461 391s 0.157 -0.656 -0.052 -0.977 391s 391s y: [2, 4, 1, 1] 391s 0.338 0.407 391s -0.223 0.167 391s -0.368 0.346 391s -0.205 -0.100 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.958 0.165 -0.233 0.996 391s 0.160 -0.433 0.002 0.760 391s 391s x1: [4, 2, 3, 1] 391s 0.005 -0.400 -0.355 0.120 391s 0.030 0.580 -0.777 0.113 391s 391s y: [2, 2, 3, 1] 391s 0.141 0.265 391s 0.418 -0.161 391s testing: mul_mat, [1, 2, 3, 3] = [4, 1, 3, 3] * [4, 2, 3, 3] 391s x0: [4, 1, 3, 3] 391s -0.437 0.808 0.855 0.442 391s 391s x1: [4, 2, 3, 3] 391s 0.702 -0.918 0.145 -0.865 391s 0.264 -0.142 0.253 0.221 391s 391s y: [1, 2, 3, 3] 391s -1.307 391s 0.085 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.010 -0.271 -0.008 -0.474 391s 0.732 -0.173 0.037 0.515 391s 0.141 -0.286 -0.393 0.253 391s 391s x1: [4, 1, 1, 1] 391s -0.633 0.235 0.651 -0.702 391s 391s y: [3, 1, 1, 1] 391s 0.257 -0.841 -0.590 391s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.217 0.118 0.792 391s -0.948 -0.157 -0.178 391s -0.740 0.736 0.611 391s 0.886 0.539 0.040 391s 391s x1: [3, 4, 3, 1] 391s 0.203 0.548 -0.653 391s 0.754 0.678 -0.193 391s -0.070 -0.532 -0.233 391s -0.331 0.066 -0.417 391s 391s y: [4, 4, 3, 1] 391s -0.408 -0.163 -0.146 0.450 391s 0.090 -0.787 -0.177 1.026 391s -0.262 0.191 -0.482 -0.358 391s -0.394 0.377 0.039 -0.274 391s testing: mul_mat, [4, 3, 3, 3] = [4, 4, 3, 3] * [4, 3, 3, 3] 391s x0: [4, 4, 3, 3] 391s 0.952 -0.672 -0.698 -0.558 391s -0.500 -0.564 -0.978 -0.398 391s 0.166 0.882 -0.175 -0.298 391s -0.989 0.345 0.563 0.551 391s 391s x1: [4, 3, 3, 3] 391s -0.603 -0.705 0.228 -0.811 391s 0.117 -0.161 0.229 -0.437 391s 0.531 0.248 0.775 -0.965 391s 391s y: [4, 3, 3, 3] 391s 0.193 0.799 -0.521 0.035 391s 0.304 -0.018 -0.033 -0.283 391s 0.337 -0.780 0.459 -0.536 391s test-mul-mat0: iter:43/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.386 -0.427 0.803 391s -0.777 0.222 -0.052 391s 391s x1: [3, 3, 1, 1] 391s 0.238 0.024 -0.141 391s -0.895 0.908 -0.113 391s -0.162 -0.536 0.676 391s 391s y: [2, 3, 1, 1] 391s -0.032 -0.172 391s -0.823 0.903 391s 0.709 -0.028 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.304 0.561 -0.031 391s 391s x1: [3, 2, 1, 1] 391s 0.410 -0.618 0.647 391s -0.295 -0.464 0.598 391s 391s y: [1, 2, 1, 1] 391s -0.491 391s -0.189 391s testing: mul_mat, [3, 1, 1, 2] = [3, 3, 1, 2] * [3, 1, 1, 2] 391s x0: [3, 3, 1, 2] 391s 0.728 -0.534 -0.884 391s -0.385 0.303 -0.419 391s -0.710 0.449 0.966 391s 391s x1: [3, 1, 1, 2] 391s 0.084 0.675 0.123 391s 391s y: [3, 1, 1, 2] 391s -0.408 0.121 0.362 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.551 0.279 0.894 391s 0.496 0.962 0.265 391s -0.823 -0.707 -0.126 391s 391s x1: [3, 3, 1, 1] 391s 0.004 0.446 0.044 391s 0.700 0.007 -0.987 391s -0.216 -0.318 0.136 391s 391s y: [3, 3, 1, 1] 391s 0.166 0.442 -0.324 391s -0.495 0.092 -0.456 391s -0.086 -0.377 0.385 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.985 -0.395 0.976 391s -0.068 0.461 -0.695 391s 391s x1: [3, 3, 1, 1] 391s -0.816 -0.286 -0.160 391s 0.048 0.967 0.029 391s 0.134 -0.833 -0.372 391s 391s y: [2, 3, 1, 1] 391s 0.760 0.035 391s -0.400 0.422 391s -0.166 -0.135 391s testing: mul_mat, [4, 3, 1, 2] = [2, 4, 1, 2] * [2, 3, 1, 2] 391s x0: [2, 4, 1, 2] 391s 0.848 0.242 391s -0.598 0.104 391s -0.943 -0.471 391s -0.437 -0.729 391s 391s x1: [2, 3, 1, 2] 391s -0.533 0.989 391s 0.089 0.149 391s 0.125 0.058 391s 391s y: [4, 3, 1, 2] 391s -0.212 0.422 0.037 -0.488 391s 0.112 -0.038 -0.154 -0.148 391s 0.120 -0.069 -0.145 -0.097 391s test-mul-mat0: iter:44/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.544 391s -0.869 391s 391s x1: [1, 4, 1, 1] 391s 0.839 391s 0.789 391s -0.077 391s -0.104 391s 391s y: [2, 4, 1, 1] 391s 0.456 -0.729 391s 0.429 -0.686 391s -0.042 0.067 391s -0.056 0.090 391s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.939 391s 391s x1: [1, 2, 2, 1] 391s -0.532 391s -0.873 391s 391s y: [1, 2, 2, 1] 391s 0.500 391s 0.820 391s testing: mul_mat, [4, 1, 2, 4] = [1, 4, 2, 4] * [1, 1, 2, 4] 391s x0: [1, 4, 2, 4] 391s -0.524 391s -0.694 391s -0.976 391s -0.352 391s 391s x1: [1, 1, 2, 4] 391s -0.462 391s 391s y: [4, 1, 2, 4] 391s 0.242 0.320 0.451 0.162 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.735 391s 0.042 391s 0.001 391s 0.342 391s 391s x1: [1, 4, 1, 1] 391s 0.015 391s 0.617 391s -0.024 391s 0.130 391s 391s y: [4, 4, 1, 1] 391s -0.011 0.001 0.000 0.005 391s -0.453 0.026 0.001 0.211 391s 0.018 -0.001 -0.000 -0.008 391s -0.095 0.005 0.000 0.044 391s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 3, 2, 1] 391s 0.243 -0.737 -0.649 -0.797 391s -0.196 0.465 0.700 -0.591 391s -0.907 -0.633 -0.619 0.974 391s 391s x1: [4, 1, 2, 1] 391s 0.876 -0.491 -0.091 -0.745 391s 391s y: [3, 1, 2, 1] 391s 1.227 -0.023 -1.154 391s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 391s x0: [3, 2, 2, 4] 391s -0.331 0.784 -0.972 391s 0.337 0.565 -0.548 391s 391s x1: [3, 4, 2, 4] 391s -0.977 0.727 0.639 391s -0.033 -0.511 0.084 391s -0.713 0.732 0.888 391s -0.620 0.995 0.352 391s 391s y: [2, 4, 2, 4] 391s 0.273 -0.269 391s -0.472 -0.346 391s -0.053 -0.313 391s 0.643 0.160 391s test-mul-mat0: iter:45/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.465 -0.568 0.137 391s -0.043 0.188 0.837 391s 391s x1: [3, 3, 1, 1] 391s 0.437 0.248 0.326 391s -0.795 -0.720 -0.687 391s -0.576 0.538 0.704 391s 391s y: [2, 3, 1, 1] 391s 0.107 0.301 391s -0.055 -0.676 391s -0.477 0.715 391s testing: mul_mat, [1, 2, 2, 1] = [3, 1, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 1, 2, 1] 391s 0.840 -0.012 -0.543 391s 391s x1: [3, 2, 2, 1] 391s 0.359 0.773 0.534 391s -0.189 -0.851 0.775 391s 391s y: [1, 2, 2, 1] 391s 0.002 391s -0.569 391s testing: mul_mat, [3, 1, 2, 2] = [3, 3, 2, 2] * [3, 1, 2, 2] 391s x0: [3, 3, 2, 2] 391s 0.725 -0.518 0.497 391s 0.873 -0.743 -0.705 391s 0.772 -0.529 0.524 391s 391s x1: [3, 1, 2, 2] 391s 0.368 -0.372 0.086 391s 391s y: [3, 1, 2, 2] 391s 0.502 0.537 0.526 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.562 0.337 -0.529 391s 0.626 0.748 0.298 391s -0.001 -0.163 -0.814 391s -0.253 0.380 -0.694 391s 391s x1: [3, 3, 1, 1] 391s -0.733 -0.871 0.993 391s 0.791 0.017 0.773 391s -0.665 0.746 -0.458 391s 391s y: [4, 3, 1, 1] 391s -1.230 -0.815 -0.665 -0.834 391s 0.042 0.738 -0.633 -0.730 391s 0.119 0.005 0.252 0.770 391s testing: mul_mat, [3, 3, 2, 1] = [4, 3, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 3, 2, 1] 391s 0.821 0.119 -0.794 0.767 391s -0.474 -0.288 0.656 0.083 391s -0.519 -0.213 0.875 -0.771 391s 391s x1: [4, 3, 2, 1] 391s 0.087 -0.056 0.089 0.562 391s 0.427 -0.647 0.534 0.783 391s 0.875 0.801 0.912 0.867 391s 391s y: [3, 3, 2, 1] 391s 0.425 0.080 -0.388 391s 0.451 0.399 -0.221 391s 0.754 0.026 -0.495 391s testing: mul_mat, [2, 4, 2, 2] = [3, 2, 2, 2] * [3, 4, 2, 2] 391s x0: [3, 2, 2, 2] 391s 0.282 -0.425 -0.318 391s 0.760 -0.090 -0.053 391s 391s x1: [3, 4, 2, 2] 391s 0.882 -0.378 0.603 391s -0.857 0.607 -0.252 391s -0.848 0.428 0.274 391s -0.367 -0.453 0.986 391s 391s y: [2, 4, 2, 2] 391s 0.218 0.673 391s -0.419 -0.693 391s -0.508 -0.698 391s -0.225 -0.290 391s test-mul-mat0: iter:46/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.302 0.209 0.723 0.344 391s 0.790 0.051 -0.749 0.066 391s 0.733 0.568 -0.536 0.044 391s 0.931 -0.469 -0.030 -0.244 391s 391s x1: [4, 3, 1, 1] 391s -0.416 0.429 0.165 0.866 391s 0.189 0.740 -0.224 0.871 391s -0.313 0.284 -0.833 0.715 391s 391s y: [4, 3, 1, 1] 391s 0.633 -0.373 -0.111 -0.804 391s 0.236 0.413 0.718 -0.377 391s -0.202 0.439 0.410 -0.573 391s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.787 0.908 -0.581 0.837 391s -0.776 -0.828 0.969 0.285 391s 0.191 -0.225 -0.194 -0.521 391s -0.047 0.996 0.758 0.848 391s 391s x1: [4, 4, 4, 1] 391s -0.981 0.152 0.341 0.448 391s -0.683 0.207 -0.363 -0.943 391s 0.983 -0.493 -0.256 0.267 391s -0.326 -0.540 -0.732 0.372 391s 391s y: [4, 4, 4, 1] 391s 1.087 1.092 -0.521 0.835 391s 0.147 -0.263 0.385 -0.837 391s -0.849 -0.526 0.209 -0.504 391s 0.503 0.097 0.007 -0.762 391s testing: mul_mat, [3, 4, 4, 4] = [4, 3, 4, 4] * [4, 4, 4, 4] 391s x0: [4, 3, 4, 4] 391s -0.406 -0.789 -0.729 -0.089 391s 0.370 -0.192 0.631 0.140 391s 0.397 -0.766 -0.113 0.991 391s 391s x1: [4, 4, 4, 4] 391s -0.835 -0.259 0.245 -0.539 391s 0.657 -0.892 0.947 0.348 391s 0.778 0.913 0.992 0.443 391s 0.636 -0.517 0.754 0.438 391s 391s y: [3, 4, 4, 4] 391s 0.412 -0.181 -0.696 391s -0.284 1.061 1.182 391s -1.798 0.800 -0.064 391s -0.439 0.872 0.998 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.016 -0.898 0.063 -0.288 391s -0.768 0.826 -0.929 0.015 391s 0.556 -0.893 -0.281 -0.854 391s 391s x1: [4, 3, 1, 1] 391s -0.655 -0.401 0.877 -0.888 391s -0.011 0.539 -0.132 0.845 391s -0.070 0.426 0.762 -0.767 391s 391s y: [3, 3, 1, 1] 391s 0.660 -0.655 0.505 391s -0.735 0.588 -1.171 391s -0.116 -0.313 0.021 391s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.877 -0.698 0.263 391s 0.256 0.864 0.730 391s 391s x1: [3, 4, 4, 1] 391s -0.632 0.556 0.811 391s -0.017 0.693 -0.208 391s 0.327 -0.708 -0.331 391s 0.439 0.281 -0.793 391s 391s y: [2, 4, 4, 1] 391s 0.379 0.911 391s -0.523 0.442 391s 0.121 -0.770 391s -0.789 -0.224 391s testing: mul_mat, [2, 3, 4, 4] = [2, 2, 4, 4] * [2, 3, 4, 4] 391s x0: [2, 2, 4, 4] 391s 0.265 0.579 391s 0.598 -0.320 391s 391s x1: [2, 3, 4, 4] 391s 0.010 0.959 391s 0.431 -0.661 391s 0.899 -0.217 391s 391s y: [2, 3, 4, 4] 391s 0.558 -0.301 391s -0.268 0.469 391s 0.112 0.607 391s test-mul-mat0: iter:47/500 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.113 -0.815 0.221 0.828 391s 391s x1: [4, 1, 1, 1] 391s 0.925 0.158 0.018 0.911 391s 391s y: [1, 1, 1, 1] 391s 0.524 391s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 2, 2, 1] 391s -0.826 0.402 -0.393 -0.119 391s -0.834 -0.355 0.208 0.904 391s 391s x1: [4, 1, 2, 1] 391s -0.290 -0.975 0.424 -0.045 391s 391s y: [2, 1, 2, 1] 391s -0.313 0.636 391s testing: mul_mat, [3, 2, 2, 2] = [4, 3, 2, 2] * [4, 2, 2, 2] 391s x0: [4, 3, 2, 2] 391s 0.273 -0.275 -0.248 0.421 391s -0.247 -0.567 -0.007 0.736 391s -0.631 -0.693 0.841 -0.347 391s 391s x1: [4, 2, 2, 2] 391s 0.132 -0.566 -0.679 0.318 391s 0.655 -0.852 -0.972 0.680 391s 391s y: [3, 2, 2, 2] 391s 0.493 0.527 -0.374 391s 0.940 0.829 -0.877 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.301 -0.117 0.500 -0.513 391s 0.532 0.448 -0.642 0.575 391s -0.522 -0.363 -0.467 -0.730 391s 391s x1: [4, 3, 1, 1] 391s -0.831 -0.119 -0.210 -0.885 391s 0.261 0.543 0.115 -0.068 391s 0.814 -0.611 0.267 0.036 391s 391s y: [3, 3, 1, 1] 391s 0.613 -0.869 1.221 391s -0.050 0.270 -0.337 391s -0.059 0.008 -0.355 391s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.811 -0.702 -0.576 391s 0.789 0.364 -0.824 391s -0.388 -0.118 0.011 391s 391s x1: [3, 4, 2, 1] 391s -0.874 -0.189 -0.872 391s -0.896 -0.613 0.268 391s -0.727 0.268 -0.942 391s -0.612 -0.471 0.601 391s 391s y: [3, 4, 2, 1] 391s -0.075 -0.040 0.353 391s -0.451 -1.151 0.423 391s -0.235 0.300 0.241 391s -0.512 -1.150 0.299 391s testing: mul_mat, [2, 3, 2, 2] = [3, 2, 2, 2] * [3, 3, 2, 2] 391s x0: [3, 2, 2, 2] 391s -0.834 0.066 0.941 391s 0.652 0.155 -0.716 391s 391s x1: [3, 3, 2, 2] 391s -0.307 0.822 0.352 391s -0.415 -0.449 0.803 391s 0.548 0.978 0.786 391s 391s y: [2, 3, 2, 2] 391s 0.642 -0.326 391s 1.073 -0.916 391s 0.348 -0.054 391s test-mul-mat0: iter:48/500 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.079 -0.991 391s 391s x1: [2, 1, 1, 1] 391s -0.245 -0.644 391s 391s y: [1, 1, 1, 1] 391s 0.657 391s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.802 -0.567 391s 0.252 -0.341 391s 0.695 0.460 391s 391s x1: [2, 1, 3, 1] 391s -0.984 -0.924 391s 391s y: [3, 1, 3, 1] 391s 1.313 0.067 -1.109 391s testing: mul_mat, [3, 3, 3, 4] = [2, 3, 3, 4] * [2, 3, 3, 4] 391s x0: [2, 3, 3, 4] 391s -0.528 -0.137 391s -0.385 -0.297 391s 0.013 0.471 391s 391s x1: [2, 3, 3, 4] 391s -0.494 -0.931 391s 0.634 0.456 391s -0.010 0.642 391s 391s y: [3, 3, 3, 4] 391s 0.388 0.467 -0.445 391s -0.397 -0.379 0.223 391s -0.083 -0.187 0.302 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.481 -0.955 391s -0.386 -0.525 391s -0.524 0.622 391s 391s x1: [2, 3, 1, 1] 391s 0.137 -0.281 391s 0.719 0.133 391s -0.019 0.216 391s 391s y: [3, 3, 1, 1] 391s 0.335 0.095 -0.247 391s 0.219 -0.347 -0.294 391s -0.216 -0.106 0.145 391s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.914 0.066 -0.453 391s -0.346 0.870 -0.515 391s -0.053 -0.208 0.269 391s 391s x1: [3, 2, 3, 1] 391s 0.427 0.998 -0.086 391s -0.217 -0.905 -0.389 391s 391s y: [3, 2, 3, 1] 391s -0.286 0.766 -0.253 391s 0.314 -0.512 0.095 391s testing: mul_mat, [1, 3, 3, 4] = [3, 1, 3, 4] * [3, 3, 3, 4] 391s x0: [3, 1, 3, 4] 391s 0.341 -0.576 0.745 391s 391s x1: [3, 3, 3, 4] 391s 0.448 -0.242 0.594 391s 0.262 -0.157 -0.752 391s -0.791 0.910 -0.881 391s 391s y: [1, 3, 3, 4] 391s 0.735 391s -0.381 391s -1.450 391s test-mul-mat0: iter:49/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.020 -0.660 0.251 -0.026 391s 0.386 0.840 0.554 -0.951 391s -0.626 0.475 0.137 -0.346 391s -0.426 -0.741 0.567 0.887 391s 391s x1: [4, 3, 1, 1] 391s -0.618 -0.523 0.061 0.514 391s -0.242 0.144 -0.239 -0.279 391s 0.511 -0.727 -0.972 0.337 391s 391s y: [4, 3, 1, 1] 391s 0.360 -1.133 -0.031 1.142 391s -0.143 0.160 0.284 -0.387 391s 0.218 -1.272 -0.914 0.068 391s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.291 0.755 -0.626 0.989 391s -0.341 0.168 -0.838 -0.301 391s 391s x1: [4, 4, 3, 1] 391s 0.205 -0.923 -0.731 0.682 391s 0.138 0.783 -0.560 -0.719 391s -0.456 0.161 0.792 -0.183 391s 0.189 0.129 0.329 -0.831 391s 391s y: [2, 4, 3, 1] 391s 0.495 0.182 391s 0.270 0.770 391s -0.688 -0.426 391s -0.875 -0.069 391s testing: mul_mat, [2, 2, 3, 2] = [4, 2, 3, 2] * [4, 2, 3, 2] 391s x0: [4, 2, 3, 2] 391s -0.193 0.556 -0.785 0.057 391s -0.875 -0.385 -0.741 0.824 391s 391s x1: [4, 2, 3, 2] 391s -0.592 -0.916 0.197 0.783 391s -0.768 -0.903 0.738 0.523 391s 391s y: [2, 2, 3, 2] 391s -0.505 1.370 391s -0.903 0.904 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.556 -0.122 0.757 -0.240 391s 0.767 -0.061 -0.674 -0.673 391s -0.607 -0.443 0.777 0.256 391s 391s x1: [4, 2, 1, 1] 391s -0.242 0.446 -0.736 -0.376 391s 0.729 -0.041 0.665 -0.900 391s 391s y: [3, 2, 1, 1] 391s -0.656 0.537 -0.719 391s 1.129 0.719 -0.138 391s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.040 -0.690 -0.493 391s 391s x1: [3, 4, 3, 1] 391s 0.588 -0.338 0.659 391s -0.699 0.153 0.416 391s 0.367 0.707 -0.846 391s 0.037 -0.535 0.600 391s 391s y: [1, 4, 3, 1] 391s -0.115 391s -0.282 391s -0.086 391s 0.072 391s testing: mul_mat, [2, 3, 3, 2] = [1, 2, 3, 2] * [1, 3, 3, 2] 391s x0: [1, 2, 3, 2] 391s 0.648 391s 0.498 391s 391s x1: [1, 3, 3, 2] 391s 0.555 391s -0.384 391s -0.135 391s 391s y: [2, 3, 3, 2] 391s 0.360 0.277 391s -0.248 -0.191 391s -0.088 -0.067 391s test-mul-mat0: iter:50/500 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.899 391s -0.452 391s 0.631 391s 0.158 391s 391s x1: [1, 2, 1, 1] 391s 0.988 391s -0.649 391s 391s y: [4, 2, 1, 1] 391s -0.888 -0.447 0.624 0.156 391s 0.584 0.294 -0.410 -0.102 391s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 4, 2, 1] 391s 0.045 391s -0.077 391s -0.771 391s -0.147 391s 391s x1: [1, 4, 2, 1] 391s 0.205 391s -0.561 391s 0.920 391s -0.349 391s 391s y: [4, 4, 2, 1] 391s 0.009 -0.016 -0.158 -0.030 391s -0.025 0.043 0.433 0.082 391s 0.041 -0.071 -0.709 -0.135 391s -0.016 0.027 0.269 0.051 391s testing: mul_mat, [3, 4, 2, 4] = [1, 3, 2, 4] * [1, 4, 2, 4] 391s x0: [1, 3, 2, 4] 391s -0.769 391s 0.692 391s -0.029 391s 391s x1: [1, 4, 2, 4] 391s -0.266 391s 0.882 391s -0.779 391s -0.810 391s 391s y: [3, 4, 2, 4] 391s 0.205 -0.184 0.008 391s -0.678 0.610 -0.025 391s 0.599 -0.539 0.022 391s 0.623 -0.560 0.023 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.021 391s 0.228 391s -0.554 391s 391s x1: [1, 3, 1, 1] 391s -0.728 391s -0.031 391s 0.726 391s 391s y: [3, 3, 1, 1] 391s 0.015 -0.166 0.404 391s 0.001 -0.007 0.017 391s -0.015 0.165 -0.403 391s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 4, 2, 1] 391s -0.358 0.379 0.086 391s -0.745 -0.806 -0.070 391s -0.433 0.071 0.933 391s 0.149 -0.386 -0.209 391s 391s x1: [3, 1, 2, 1] 391s 0.210 -0.080 0.417 391s 391s y: [4, 1, 2, 1] 391s -0.070 -0.121 0.292 -0.025 391s testing: mul_mat, [4, 3, 2, 4] = [4, 4, 2, 4] * [4, 3, 2, 4] 391s x0: [4, 4, 2, 4] 391s -0.582 0.638 0.731 -0.137 391s 0.736 -0.733 0.989 -0.025 391s -0.492 -0.738 -0.181 -0.276 391s -0.009 0.842 0.439 0.847 391s 391s x1: [4, 3, 2, 4] 391s 0.187 0.788 -0.096 0.139 391s -0.103 -0.776 -0.307 0.539 391s -0.521 0.260 -0.312 0.858 391s 391s y: [4, 3, 2, 4] 391s 0.305 -0.538 -0.695 0.738 391s -0.734 0.176 0.531 -0.331 391s 0.124 -0.905 -0.116 0.814 391s test-mul-mat0: iter:51/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.654 391s 0.527 391s -0.578 391s 391s x1: [1, 3, 1, 1] 391s -0.040 391s -0.557 391s -0.412 391s 391s y: [3, 3, 1, 1] 391s -0.026 -0.021 0.023 391s -0.364 -0.294 0.322 391s -0.269 -0.217 0.238 391s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 4, 2, 1] 391s 0.825 391s 0.380 391s -0.512 391s -0.302 391s 391s x1: [1, 3, 2, 1] 391s 0.132 391s -0.412 391s 0.731 391s 391s y: [4, 3, 2, 1] 391s 0.109 0.050 -0.068 -0.040 391s -0.340 -0.157 0.211 0.124 391s 0.603 0.278 -0.374 -0.220 391s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.563 391s -0.151 391s 0.512 391s 391s x1: [1, 4, 2, 1] 391s -0.942 391s 0.280 391s 0.164 391s -0.023 391s 391s y: [3, 4, 2, 1] 391s 0.530 0.143 -0.482 391s -0.158 -0.042 0.143 391s -0.092 -0.025 0.084 391s 0.013 0.003 -0.012 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.858 391s 391s x1: [1, 3, 1, 1] 391s 0.954 391s 0.635 391s 0.627 391s 391s y: [1, 3, 1, 1] 391s 0.819 391s 0.545 391s 0.538 391s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 1, 2, 1] 391s 0.554 391s 391s x1: [1, 1, 2, 1] 391s 0.729 391s 391s y: [1, 1, 2, 1] 391s 0.404 391s testing: mul_mat, [3, 1, 2, 1] = [1, 3, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.064 391s -0.044 391s -0.150 391s 391s x1: [1, 1, 2, 1] 391s 0.742 391s 391s y: [3, 1, 2, 1] 391s -0.047 -0.033 -0.111 391s test-mul-mat0: iter:52/500 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.942 -0.581 0.056 0.801 391s -0.852 0.930 0.054 0.702 391s -0.815 -0.204 0.954 -0.485 391s 391s x1: [4, 2, 1, 1] 391s -0.043 -0.758 0.886 0.805 391s 0.753 0.418 -0.692 -0.208 391s 391s y: [3, 2, 1, 1] 391s 1.094 -0.055 0.644 391s 0.262 -0.437 -1.259 391s testing: mul_mat, [1, 3, 2, 1] = [4, 1, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 1, 2, 1] 391s 0.565 0.752 0.991 -0.639 391s 391s x1: [4, 3, 2, 1] 391s 0.732 -0.090 0.364 -0.274 391s 0.146 0.378 -0.304 0.698 391s -0.304 -0.855 -0.346 -0.062 391s 391s y: [1, 3, 2, 1] 391s 0.882 391s -0.379 391s -1.117 391s testing: mul_mat, [4, 1, 2, 2] = [4, 4, 2, 2] * [4, 1, 2, 2] 391s x0: [4, 4, 2, 2] 391s -0.947 -0.041 0.833 -0.436 391s 0.009 0.639 -0.871 -0.239 391s 0.630 -0.511 -0.532 -0.865 391s -0.418 0.084 -0.995 0.400 391s 391s x1: [4, 1, 2, 2] 391s 0.870 -0.182 0.763 0.248 391s 391s y: [4, 1, 2, 2] 391s -0.288 -0.833 0.020 -1.039 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.862 -0.293 -0.726 -0.816 391s -0.331 0.798 -0.423 -0.936 391s 0.168 -0.520 0.195 0.271 391s 391s x1: [4, 4, 1, 1] 391s 0.269 0.263 0.828 -0.898 391s -0.649 0.265 -0.435 0.928 391s 0.169 0.162 -0.914 -0.843 391s 0.764 0.646 0.994 0.619 391s 391s y: [3, 4, 1, 1] 391s -0.178 0.611 -0.173 391s 0.041 -0.258 -0.081 391s 1.158 1.249 -0.462 391s -2.074 -0.737 0.154 391s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 3, 2, 1] 391s -0.251 -0.067 0.421 391s -0.204 -0.140 -0.924 391s -0.275 0.995 -0.465 391s 391s x1: [3, 4, 2, 1] 391s -0.511 0.217 -0.460 391s 0.752 0.045 -0.358 391s -0.897 -0.690 0.207 391s -0.970 0.479 -0.631 391s 391s y: [3, 4, 2, 1] 391s -0.080 0.499 0.571 391s -0.343 0.171 0.004 391s 0.359 0.089 -0.536 391s -0.055 0.714 1.037 391s testing: mul_mat, [3, 3, 2, 2] = [3, 3, 2, 2] * [3, 3, 2, 2] 391s x0: [3, 3, 2, 2] 391s -0.049 0.067 -0.288 391s 0.137 0.556 -0.276 391s 0.791 0.067 0.140 391s 391s x1: [3, 3, 2, 2] 391s 0.423 -0.454 -0.115 391s 0.185 -0.824 -0.363 391s 0.066 0.944 0.058 391s 391s y: [3, 3, 2, 2] 391s -0.018 -0.163 0.288 391s 0.041 -0.333 0.040 391s 0.043 0.518 0.124 391s test-mul-mat0: iter:53/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.956 391s -0.929 391s -0.520 391s 391s x1: [1, 3, 1, 1] 391s -0.311 391s 0.697 391s -0.009 391s 391s y: [3, 3, 1, 1] 391s 0.297 0.289 0.162 391s -0.666 -0.648 -0.362 391s 0.009 0.009 0.005 391s testing: mul_mat, [4, 3, 3, 1] = [1, 4, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.556 391s -0.938 391s -0.736 391s 0.011 391s 391s x1: [1, 3, 3, 1] 391s 0.087 391s 0.281 391s -0.098 391s 391s y: [4, 3, 3, 1] 391s 0.048 -0.082 -0.064 0.001 391s 0.156 -0.264 -0.207 0.003 391s -0.055 0.092 0.072 -0.001 391s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 4, 3, 1] 391s -0.211 391s -0.415 391s 0.946 391s -0.915 391s 391s x1: [1, 4, 3, 1] 391s -0.378 391s -0.606 391s 0.285 391s -0.681 391s 391s y: [4, 4, 3, 1] 391s 0.080 0.157 -0.357 0.346 391s 0.128 0.252 -0.573 0.554 391s -0.060 -0.118 0.270 -0.261 391s 0.143 0.283 -0.644 0.623 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.273 391s -0.373 391s 391s x1: [1, 4, 1, 1] 391s -0.746 391s 0.201 391s -0.323 391s -0.995 391s 391s y: [2, 4, 1, 1] 391s -0.204 0.278 391s 0.055 -0.075 391s -0.088 0.121 391s -0.272 0.372 391s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 3, 3, 1] 391s 0.879 -0.319 391s -0.240 -0.687 391s -0.353 0.437 391s 391s x1: [2, 1, 3, 1] 391s 0.070 -0.442 391s 391s y: [3, 1, 3, 1] 391s 0.203 0.287 -0.218 391s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.187 -0.332 0.879 391s -0.712 -0.089 0.832 391s 391s x1: [3, 2, 3, 1] 391s -0.049 0.552 -0.925 391s 0.627 -0.175 -0.299 391s 391s y: [2, 2, 3, 1] 391s -1.006 -0.784 391s -0.088 -0.679 391s test-mul-mat0: iter:54/500 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.412 0.133 0.917 -0.677 391s 391s x1: [4, 2, 1, 1] 391s 0.064 0.098 -0.606 -0.695 391s -0.185 -0.897 -0.774 -0.155 391s 391s y: [1, 2, 1, 1] 391s -0.046 391s -0.801 391s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 1, 2, 1] 391s 0.514 -0.094 0.726 0.833 391s 391s x1: [4, 1, 2, 1] 391s 0.013 0.749 0.479 -0.921 391s 391s y: [1, 1, 2, 1] 391s -0.484 391s testing: mul_mat, [4, 1, 2, 2] = [4, 4, 2, 2] * [4, 1, 2, 2] 391s x0: [4, 4, 2, 2] 391s -0.477 -0.952 -0.269 -0.457 391s 0.162 -0.027 -0.943 -0.932 391s -0.300 0.890 0.467 -0.394 391s 0.427 0.007 0.845 -0.973 391s 391s x1: [4, 1, 2, 2] 391s 0.239 -0.400 -0.362 0.633 391s 391s y: [4, 1, 2, 2] 391s 0.075 -0.199 -0.846 -0.823 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.633 0.282 0.760 0.511 391s 0.949 0.514 -0.819 -0.397 391s 0.435 -0.223 0.593 -0.773 391s 391s x1: [4, 4, 1, 1] 391s 0.183 0.887 0.189 -0.294 391s -0.716 -0.778 -0.071 -0.429 391s -0.278 0.757 -0.715 -0.771 391s -0.569 -0.974 0.382 -0.417 391s 391s y: [3, 4, 1, 1] 391s 0.360 0.592 0.222 391s -0.946 -0.851 0.151 391s -0.900 1.017 -0.118 391s -0.558 -1.187 0.518 391s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.247 -0.385 -0.935 391s 0.462 -0.242 -0.158 391s 391s x1: [3, 4, 2, 1] 391s 0.059 -0.173 0.411 391s -0.054 -0.983 -0.883 391s 0.230 -0.761 0.046 391s 0.801 -0.039 -0.197 391s 391s y: [2, 4, 2, 1] 391s -0.303 0.004 391s 1.191 0.352 391s 0.307 0.283 391s 0.397 0.410 391s testing: mul_mat, [2, 3, 2, 2] = [2, 2, 2, 2] * [2, 3, 2, 2] 391s x0: [2, 2, 2, 2] 391s 0.382 0.776 391s -0.263 -0.856 391s 391s x1: [2, 3, 2, 2] 391s 0.991 0.148 391s 0.415 0.792 391s -0.892 -0.782 391s 391s y: [2, 3, 2, 2] 391s 0.493 -0.387 391s 0.773 -0.788 391s -0.947 0.904 391s test-mul-mat0: iter:55/500 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.859 0.701 391s 391s x1: [2, 1, 1, 1] 391s 0.541 0.636 391s 391s y: [1, 1, 1, 1] 391s 0.910 391s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.780 0.681 391s 0.602 0.866 391s 0.827 0.737 391s 391s x1: [2, 1, 3, 1] 391s -0.489 0.639 391s 391s y: [3, 1, 3, 1] 391s 0.816 0.259 0.067 391s testing: mul_mat, [2, 3, 3, 3] = [2, 2, 3, 3] * [2, 3, 3, 3] 391s x0: [2, 2, 3, 3] 391s -0.548 0.526 391s 0.044 0.908 391s 391s x1: [2, 3, 3, 3] 391s 0.972 -0.554 391s 0.490 0.676 391s -0.695 0.191 391s 391s y: [2, 3, 3, 3] 391s -0.824 -0.461 391s 0.087 0.636 391s 0.481 0.143 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.994 0.891 391s 0.564 -0.710 391s -0.284 0.980 391s -0.719 0.186 391s 391s x1: [2, 2, 1, 1] 391s 0.592 -0.314 391s -0.110 -0.221 391s 391s y: [4, 2, 1, 1] 391s -0.869 0.557 -0.476 -0.484 391s -0.088 0.095 -0.186 0.038 391s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 3, 3, 1] 391s 0.226 0.515 -0.580 -0.861 391s 0.658 0.638 -0.537 -0.397 391s 0.456 -0.357 0.739 -0.127 391s 391s x1: [4, 2, 3, 1] 391s -0.095 -0.175 0.097 0.719 391s -0.860 0.134 0.717 -0.012 391s 391s y: [3, 2, 3, 1] 391s -0.787 -0.512 -0.001 391s -0.531 -0.861 0.092 391s testing: mul_mat, [3, 4, 3, 3] = [3, 3, 3, 3] * [3, 4, 3, 3] 391s x0: [3, 3, 3, 3] 391s -0.731 -0.772 -0.397 391s 0.887 -0.402 -0.994 391s -0.951 0.291 -0.651 391s 391s x1: [3, 4, 3, 3] 391s 0.577 0.934 -0.093 391s 0.316 -0.927 0.511 391s -0.811 0.930 0.102 391s -0.467 0.059 -0.211 391s 391s y: [3, 4, 3, 3] 391s -1.106 0.228 -0.216 391s 0.281 0.146 -0.903 391s -0.165 -1.195 0.975 391s 0.380 -0.228 0.599 391s test-mul-mat0: iter:56/500 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.384 -0.647 0.775 0.177 391s -0.869 0.252 -0.860 0.407 391s 391s x1: [4, 4, 1, 1] 391s 0.441 -0.627 -0.060 -0.468 391s -0.624 -0.293 0.367 0.987 391s 0.891 0.833 0.255 -0.283 391s -0.473 0.470 0.790 0.456 391s 391s y: [2, 4, 1, 1] 391s 0.106 -0.680 391s 0.889 0.554 391s -0.734 -0.898 391s 0.571 0.035 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.180 0.875 0.745 0.661 391s 391s x1: [4, 2, 4, 1] 391s -0.868 -0.012 -0.838 0.972 391s -0.702 -0.520 0.414 -0.329 391s 391s y: [1, 2, 4, 1] 391s -0.149 391s -0.491 391s testing: mul_mat, [1, 1, 4, 4] = [4, 1, 4, 4] * [4, 1, 4, 4] 391s x0: [4, 1, 4, 4] 391s 0.414 0.038 0.993 0.374 391s 391s x1: [4, 1, 4, 4] 391s 0.366 -0.451 -0.511 -0.506 391s 391s y: [1, 1, 4, 4] 391s -0.563 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.067 0.535 -0.239 0.676 391s 0.471 -0.267 0.394 0.757 391s 0.098 0.782 -0.940 0.377 391s 0.460 -0.894 0.093 -0.561 391s 391s x1: [4, 1, 1, 1] 391s 0.028 -0.081 0.355 0.055 391s 391s y: [4, 1, 1, 1] 391s -0.093 0.216 -0.374 0.088 391s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 4, 4, 1] 391s 0.090 -0.103 -0.061 0.182 391s -0.025 -0.465 0.277 -0.626 391s 0.980 0.986 -0.667 0.807 391s -0.625 -0.002 0.663 0.161 391s 391s x1: [4, 4, 4, 1] 391s 0.534 -0.778 0.618 -0.498 391s 0.164 -0.356 0.086 0.152 391s -0.071 -0.674 -0.820 0.848 391s 0.681 0.234 -0.679 -0.386 391s 391s y: [4, 4, 4, 1] 391s 0.000 0.831 -1.058 -0.002 391s 0.074 0.090 -0.125 -0.020 391s 0.268 -0.443 0.498 -0.362 391s 0.008 -0.072 1.039 -0.939 391s testing: mul_mat, [3, 4, 4, 4] = [4, 3, 4, 4] * [4, 4, 4, 4] 391s x0: [4, 3, 4, 4] 391s 0.052 0.331 0.096 -0.870 391s -0.512 -0.876 -0.506 -0.331 391s 0.130 -0.513 -0.122 -0.524 391s 391s x1: [4, 4, 4, 4] 391s 0.684 -0.568 -0.327 -0.990 391s -0.716 -0.119 0.274 0.412 391s 0.421 -0.797 0.723 -0.664 391s -0.062 0.194 0.906 0.776 391s 391s y: [3, 4, 4, 4] 391s 0.677 0.640 0.939 391s -0.409 0.196 -0.282 391s 0.405 0.337 0.724 391s -0.527 -0.854 -0.625 391s test-mul-mat0: iter:57/500 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.607 -0.385 391s 391s x1: [2, 3, 1, 1] 391s -0.162 0.510 391s 0.988 0.570 391s -0.602 0.217 391s 391s y: [1, 3, 1, 1] 391s -0.295 391s 0.380 391s -0.449 391s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.978 0.485 391s 391s x1: [2, 1, 2, 1] 391s -0.839 0.035 391s 391s y: [1, 1, 2, 1] 391s -0.803 391s testing: mul_mat, [2, 1, 2, 4] = [2, 2, 2, 4] * [2, 1, 2, 4] 391s x0: [2, 2, 2, 4] 391s 0.142 -0.761 391s 0.601 -0.251 391s 391s x1: [2, 1, 2, 4] 391s 0.213 -0.215 391s 391s y: [2, 1, 2, 4] 391s 0.194 0.182 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.836 -0.533 391s 0.401 -0.222 391s -0.803 0.798 391s -0.040 -0.327 391s 391s x1: [2, 2, 1, 1] 391s 0.458 0.340 391s -0.948 -0.687 391s 391s y: [4, 2, 1, 1] 391s 0.202 0.108 -0.097 -0.129 391s -0.426 -0.227 0.212 0.263 391s testing: mul_mat, [4, 2, 2, 1] = [4, 4, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 4, 2, 1] 391s -0.231 -0.239 0.551 0.415 391s -0.552 0.353 0.617 0.302 391s -0.748 -0.342 -0.371 -0.101 391s -0.179 -0.838 -0.672 -0.164 391s 391s x1: [4, 2, 2, 1] 391s -0.429 0.421 -0.343 -0.744 391s -0.889 0.706 -0.588 0.976 391s 391s y: [4, 2, 2, 1] 391s -0.499 -0.051 0.379 0.076 391s 0.118 0.672 0.543 -0.198 391s testing: mul_mat, [4, 4, 2, 4] = [4, 4, 2, 4] * [4, 4, 2, 4] 391s x0: [4, 4, 2, 4] 391s -0.401 -0.883 -0.558 0.762 391s 0.909 -0.011 -0.175 -0.370 391s -0.266 0.944 0.924 -0.809 391s -0.666 0.076 -0.637 0.859 391s 391s x1: [4, 4, 2, 4] 391s -0.648 0.311 0.916 0.113 391s -0.335 -0.426 0.275 -0.785 391s -0.809 0.904 -0.456 0.607 391s 0.206 0.443 -0.557 0.166 391s 391s y: [4, 4, 2, 4] 391s -0.440 -0.795 1.222 -0.031 391s -0.241 -0.059 0.576 -0.658 391s 0.244 -0.889 0.156 1.419 391s -0.037 0.219 -0.285 0.394 391s test-mul-mat0: iter:58/500 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.094 -0.879 391s -0.285 -0.037 391s 0.643 -0.185 391s -0.780 -0.818 391s 391s x1: [2, 3, 1, 1] 391s 0.769 -0.829 391s 0.736 0.785 391s -0.638 0.017 391s 391s y: [4, 3, 1, 1] 391s 0.657 -0.188 0.647 0.079 391s -0.759 -0.239 0.328 -1.215 391s 0.045 0.181 -0.413 0.484 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.116 0.202 391s 0.708 -0.942 391s -0.024 -0.611 391s 391s x1: [2, 4, 1, 1] 391s -0.800 0.624 391s -0.115 -0.968 391s 0.939 0.228 391s -0.423 -0.248 391s 391s y: [3, 4, 1, 1] 391s 0.034 -1.155 -0.362 391s -0.209 0.830 0.595 391s 0.154 0.451 -0.162 391s -0.099 -0.066 0.162 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.893 0.827 391s -0.104 -0.392 391s 391s x1: [2, 3, 1, 1] 391s -0.977 -0.256 391s -0.441 0.758 391s -0.471 -0.079 391s 391s y: [2, 3, 1, 1] 391s -1.084 0.202 391s 0.234 -0.252 391s -0.485 0.080 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.623 0.607 391s -0.365 0.655 391s 391s x1: [2, 2, 1, 1] 391s -0.210 -0.461 391s 0.423 0.011 391s 391s y: [2, 2, 1, 1] 391s -0.410 -0.226 391s 0.270 -0.147 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.348 -0.844 391s -0.472 0.406 391s 391s x1: [2, 2, 1, 1] 391s -0.425 -0.166 391s -0.767 0.326 391s 391s y: [2, 2, 1, 1] 391s -0.008 0.133 391s -0.543 0.494 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.187 0.962 391s -0.222 -0.329 391s -0.975 0.852 391s 391s x1: [2, 2, 1, 1] 391s 0.504 -0.456 391s 0.429 -0.751 391s 391s y: [3, 2, 1, 1] 391s -0.344 0.038 -0.880 391s -0.643 0.152 -1.059 391s test-mul-mat0: iter:59/500 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.810 0.007 0.561 391s -0.282 0.163 -0.033 391s -0.777 0.707 -0.604 391s -0.529 -0.189 0.583 391s 391s x1: [3, 3, 1, 1] 391s -0.299 0.654 -0.881 391s -0.675 -0.711 0.726 391s 0.980 -0.136 -0.440 391s 391s y: [4, 3, 1, 1] 391s -0.248 0.220 1.227 -0.479 391s 0.949 0.050 -0.417 0.914 391s -1.042 -0.284 -0.593 -0.749 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.274 0.645 -0.795 391s -0.863 -0.795 -0.583 391s -0.673 0.212 0.978 391s 391s x1: [3, 4, 1, 1] 391s 0.249 -0.164 0.545 391s 0.920 -0.312 -0.597 391s 0.198 0.331 0.801 391s 0.900 -0.015 0.919 391s 391s y: [3, 4, 1, 1] 391s -0.471 -0.402 0.331 391s 0.525 -0.198 -1.269 391s -0.369 -0.901 0.720 391s -0.495 -1.300 0.291 391s testing: mul_mat, [2, 3, 1, 2] = [3, 2, 1, 2] * [3, 3, 1, 2] 391s x0: [3, 2, 1, 2] 391s -0.266 -0.409 -0.770 391s -0.993 -0.764 -0.565 391s 391s x1: [3, 3, 1, 2] 391s 0.046 -0.625 -0.055 391s 0.268 -0.917 0.341 391s 0.739 -0.106 -0.077 391s 391s y: [2, 3, 1, 2] 391s 0.286 0.463 391s 0.041 0.242 391s -0.094 -0.609 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.005 -0.266 0.152 391s -0.961 -0.493 0.379 391s 0.514 -0.578 -0.471 391s 391s x1: [3, 2, 1, 1] 391s 0.518 -0.971 0.776 391s -0.214 -0.889 0.116 391s 391s y: [3, 2, 1, 1] 391s 0.379 0.275 0.462 391s 0.253 0.687 0.349 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.319 -0.117 -0.437 391s -0.947 0.777 0.112 391s -0.741 0.001 0.775 391s -0.028 -0.341 0.089 391s 391s x1: [3, 3, 1, 1] 391s -0.099 -0.950 -0.465 391s 0.635 -0.359 -0.235 391s 0.642 -0.122 0.200 391s 391s y: [4, 3, 1, 1] 391s 0.283 -0.696 -0.288 0.286 391s 0.347 -0.906 -0.653 0.084 391s 0.132 -0.681 -0.321 0.042 391s testing: mul_mat, [2, 3, 1, 2] = [4, 2, 1, 2] * [4, 3, 1, 2] 391s x0: [4, 2, 1, 2] 391s -0.437 -0.433 0.423 0.792 391s 0.648 0.675 0.656 0.238 391s 391s x1: [4, 3, 1, 2] 391s -0.883 0.814 -0.397 -0.148 391s -0.678 0.025 -0.996 0.700 391s 0.554 -0.095 0.751 -0.912 391s 391s y: [2, 3, 1, 2] 391s -0.252 -0.318 391s 0.419 -0.909 391s -0.606 0.570 391s test-mul-mat0: iter:60/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.759 0.262 0.938 391s -0.532 -0.177 -0.073 391s -0.528 0.386 -0.424 391s 0.039 0.061 0.999 391s 391s x1: [3, 1, 1, 1] 391s 0.778 -0.727 -0.327 391s 391s y: [4, 1, 1, 1] 391s -1.088 -0.262 -0.552 -0.341 391s testing: mul_mat, [1, 4, 2, 1] = [3, 1, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 1, 2, 1] 391s 0.242 -0.387 0.049 391s 391s x1: [3, 4, 2, 1] 391s -0.306 -0.147 0.237 391s -0.047 0.497 -0.204 391s -0.764 -0.856 -0.944 391s 0.073 0.696 0.067 391s 391s y: [1, 4, 2, 1] 391s -0.005 391s -0.214 391s 0.101 391s -0.249 391s testing: mul_mat, [2, 1, 2, 1] = [3, 2, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.833 0.136 0.237 391s 0.863 -0.797 0.235 391s 391s x1: [3, 1, 2, 1] 391s -0.025 -0.474 -0.716 391s 391s y: [2, 1, 2, 1] 391s -0.214 0.188 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.940 -0.693 -0.267 391s 0.686 0.734 0.282 391s 391s x1: [3, 2, 1, 1] 391s -0.986 -0.780 -0.836 391s -0.216 0.447 -0.363 391s 391s y: [2, 2, 1, 1] 391s 1.691 -1.484 391s -0.010 0.078 391s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 3, 2, 1] 391s -0.892 0.924 391s -0.144 0.475 391s 0.677 0.691 391s 391s x1: [2, 3, 2, 1] 391s -0.740 0.017 391s 0.210 -0.717 391s -0.903 0.374 391s 391s y: [3, 3, 2, 1] 391s 0.675 0.114 -0.489 391s -0.850 -0.370 -0.353 391s 1.151 0.307 -0.353 391s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.040 -0.309 0.130 391s 0.284 -0.524 -0.344 391s 391s x1: [3, 2, 2, 1] 391s 0.651 0.961 -0.193 391s 0.385 -0.306 -0.911 391s 391s y: [2, 2, 2, 1] 391s -0.296 -0.253 391s -0.008 0.583 391s test-mul-mat0: iter:61/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.741 0.194 391s -0.149 0.548 391s 0.867 0.229 391s -0.412 0.151 391s 391s x1: [2, 4, 1, 1] 391s 0.375 -0.001 391s 0.894 0.336 391s 0.806 0.279 391s -0.970 0.895 391s 391s y: [4, 4, 1, 1] 391s 0.277 -0.056 0.325 -0.154 391s 0.727 0.051 0.852 -0.318 391s 0.651 0.033 0.763 -0.290 391s -0.545 0.635 -0.636 0.535 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.066 -0.192 391s -0.174 0.433 391s 391s x1: [2, 4, 1, 1] 391s 0.921 0.064 391s -0.720 -0.423 391s 0.648 -0.734 391s -0.090 -0.844 391s 391s y: [2, 4, 1, 1] 391s -0.073 -0.133 391s 0.128 -0.058 391s 0.098 -0.431 391s 0.168 -0.350 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.761 -0.776 391s 0.566 -0.866 391s 391s x1: [2, 2, 1, 1] 391s 0.051 -0.800 391s -0.567 -0.055 391s 391s y: [2, 2, 1, 1] 391s 0.582 0.721 391s 0.474 -0.274 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.195 -0.654 391s 0.228 0.149 391s 0.443 -0.493 391s 391s x1: [2, 2, 1, 1] 391s 0.147 0.307 391s 0.328 0.999 391s 391s y: [3, 2, 1, 1] 391s -0.172 0.079 -0.086 391s -0.589 0.223 -0.347 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.570 0.445 -0.829 391s 0.120 0.004 0.645 391s 391s x1: [3, 2, 1, 1] 391s -0.373 0.726 -0.845 391s -0.107 -0.364 -0.689 391s 391s y: [2, 2, 1, 1] 391s 0.811 -0.587 391s 0.348 -0.459 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.513 -0.091 391s -0.946 -0.666 391s 0.139 0.714 391s -0.514 -0.648 391s 391s x1: [2, 3, 1, 1] 391s 0.437 0.116 391s 0.181 0.676 391s 0.340 -0.252 391s 391s y: [4, 3, 1, 1] 391s -0.234 -0.490 0.143 -0.299 391s -0.154 -0.622 0.508 -0.531 391s -0.151 -0.153 -0.133 -0.011 391s test-mul-mat0: iter:62/500 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.839 -0.280 0.558 0.893 391s 391s x1: [4, 4, 1, 1] 391s -0.412 0.014 -0.800 0.224 391s 0.325 -0.181 -0.206 -0.555 391s -0.736 0.798 -0.384 0.909 391s 0.234 0.731 0.091 -0.090 391s 391s y: [1, 4, 1, 1] 391s -0.596 391s -0.287 391s -0.243 391s -0.038 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.155 -0.126 -0.3ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 80 -0.986 391s -0.819 0.208 0.028 -0.619 391s 391s x1: [4, 1, 1, 1] 391s 0.859 -0.956 -0.198 -0.807 391s 391s y: [2, 1, 1, 1] 391s 1.124 -0.409 391s testing: mul_mat, [2, 2, 1, 2] = [4, 2, 1, 2] * [4, 2, 1, 2] 391s x0: [4, 2, 1, 2] 391s -0.804 0.748 0.818 -0.002 391s 0.941 0.576 -0.847 -0.185 391s 391s x1: [4, 2, 1, 2] 391s -0.568 -0.647 0.200 0.226 391s -0.202 0.464 0.024 0.414 391s 391s y: [2, 2, 1, 2] 391s 0.136 -1.118 391s 0.528 -0.019 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.877 0.513 0.122 -0.820 391s 391s x1: [4, 2, 1, 1] 391s -0.424 0.063 -0.654 -0.960 391s -0.913 0.760 0.413 -0.655 391s 391s y: [1, 2, 1, 1] 391s 0.368 391s 0.177 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.374 391s 0.090 391s 391s x1: [1, 4, 1, 1] 391s 0.402 391s -0.103 391s -0.624 391s 0.149 391s 391s y: [2, 4, 1, 1] 391s 0.150 0.036 391s -0.039 -0.009 391s -0.233 -0.056 391s 0.056 0.013 391s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 391s x0: [2, 4, 1, 2] 391s -0.100 0.730 391s 0.893 -0.564 391s -0.112 -0.148 391s -0.724 -0.206 391s 391s x1: [2, 1, 1, 2] 391s -0.708 0.527 391s 391s y: [4, 1, 1, 2] 391s 0.456 -0.930 0.001 0.404 391s test-mul-mat0: iter:63/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.637 -0.716 -0.038 391s 0.331 0.184 -0.145 391s 391s x1: [3, 2, 1, 1] 391s 0.467 -0.263 0.290 391s -0.818 -0.889 -0.620 391s 391s y: [2, 2, 1, 1] 391s 0.475 0.064 391s 0.139 -0.344 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.203 -0.740 0.978 391s 0.507 0.486 0.426 391s 0.688 -0.775 0.996 391s -0.951 -0.634 0.463 391s 391s x1: [3, 2, 1, 1] 391s -0.782 0.460 -0.414 391s -0.346 -0.688 0.379 391s 391s y: [4, 2, 1, 1] 391s -0.904 -0.349 -1.307 0.260 391s 0.810 -0.348 0.673 0.942 391s testing: mul_mat, [2, 4, 1, 3] = [3, 2, 1, 3] * [3, 4, 1, 3] 391s x0: [3, 2, 1, 3] 391s 0.937 -0.752 0.881 391s 0.801 0.751 0.818 391s 391s x1: [3, 4, 1, 3] 391s -0.214 0.655 0.645 391s -0.104 -0.965 -0.881 391s -0.466 -0.681 0.081 391s 0.865 0.503 0.937 391s 391s y: [2, 4, 1, 3] 391s -0.125 0.848 391s -0.147 -1.528 391s 0.147 -0.819 391s 1.256 1.837 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.649 0.753 0.690 391s -0.943 0.985 -0.767 391s 0.808 -0.392 -0.512 391s 391s x1: [3, 2, 1, 1] 391s -0.480 -0.876 0.925 391s -0.908 0.844 -0.837 391s 391s y: [3, 2, 1, 1] 391s 0.289 -1.120 -0.517 391s 0.648 2.330 -0.637 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.158 0.609 0.733 391s 0.292 0.344 0.269 391s 0.734 -0.746 -0.654 391s 391s x1: [3, 3, 1, 1] 391s 0.491 0.985 -0.694 391s 0.376 -0.300 -0.354 391s 0.997 -0.310 -0.112 391s 391s y: [3, 3, 1, 1] 391s 0.169 0.296 0.080 391s -0.383 -0.089 0.731 391s -0.114 0.154 1.036 391s testing: mul_mat, [1, 3, 1, 3] = [3, 1, 1, 3] * [3, 3, 1, 3] 391s x0: [3, 1, 1, 3] 391s 0.582 -0.877 0.359 391s 391s x1: [3, 3, 1, 3] 391s 0.577 0.432 0.474 391s 0.928 0.488 0.282 391s 0.682 0.474 0.889 391s 391s y: [1, 3, 1, 3] 391s 0.127 391s 0.212 391s 0.300 391s test-mul-mat0: iter:64/500 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.243 391s -0.694 391s -0.606 391s 391s x1: [1, 4, 1, 1] 391s -0.777 391s -0.619 391s 0.004 391s 0.915 391s 391s y: [3, 4, 1, 1] 391s -0.189 0.539 0.471 391s -0.150 0.429 0.375 391s 0.001 -0.003 -0.002 391s 0.222 -0.634 -0.554 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.627 391s -0.878 391s 0.262 391s 391s x1: [1, 3, 1, 1] 391s 0.478 391s 0.388 391s -0.388 391s 391s y: [3, 3, 1, 1] 391s 0.300 -0.420 0.125 391s 0.243 -0.341 0.102 391s -0.243 0.341 -0.102 391s testing: mul_mat, [1, 3, 1, 3] = [1, 1, 1, 3] * [1, 3, 1, 3] 391s x0: [1, 1, 1, 3] 391s 0.695 391s 391s x1: [1, 3, 1, 3] 391s 0.476 391s -0.024 391s -0.725 391s 391s y: [1, 3, 1, 3] 391s 0.331 391s -0.017 391s -0.504 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.787 391s 391s x1: [1, 1, 1, 1] 391s -0.789 391s 391s y: [1, 1, 1, 1] 391s 0.621 391s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.217 391s -0.652 391s -0.970 391s 391s x1: [1, 1, 1, 1] 391s -0.566 391s 391s y: [3, 1, 1, 1] 391s -0.123 0.369 0.549 391s testing: mul_mat, [2, 1, 1, 3] = [3, 2, 1, 3] * [3, 1, 1, 3] 391s x0: [3, 2, 1, 3] 391s 0.146 0.978 0.232 391s -0.866 -0.797 0.612 391s 391s x1: [3, 1, 1, 3] 391s -0.541 -0.345 -0.576 391s 391s y: [2, 1, 1, 3] 391s -0.550 0.391 391s test-mul-mat0: iter:65/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.435 -0.569 0.691 -0.904 391s -0.774 0.481 0.342 -0.766 391s 0.705 0.553 0.459 -0.722 391s 0.795 0.143 0.505 -0.539 391s 391s x1: [4, 4, 1, 1] 391s 0.227 -0.334 0.622 0.270 391s 0.702 -0.938 -0.060 -0.140 391s 0.765 -0.914 -0.006 0.743 391s -0.711 -0.774 0.355 -0.291 391s 391s y: [4, 4, 1, 1] 391s 0.474 -0.331 0.066 0.301 391s 0.924 -0.907 0.050 0.469 391s 0.176 -1.602 -0.505 0.073 391s 0.640 0.522 -0.556 -0.339 391s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 2, 3, 1] 391s -0.149 -0.112 0.739 0.414 391s -0.420 -0.030 0.531 0.731 391s 391s x1: [4, 4, 3, 1] 391s -0.235 -0.224 -0.838 -0.173 391s 0.716 0.022 -0.408 0.802 391s -0.984 -0.665 -0.909 -0.758 391s 0.690 -0.200 -0.647 0.125 391s 391s y: [2, 4, 3, 1] 391s -0.630 -0.465 391s -0.079 0.069 391s -0.763 -0.603 391s -0.506 -0.535 391s testing: mul_mat, [2, 2, 3, 3] = [4, 2, 3, 3] * [4, 2, 3, 3] 391s x0: [4, 2, 3, 3] 391s -0.904 0.667 0.260 0.861 391s 0.948 -0.707 0.976 -0.414 391s 391s x1: [4, 2, 3, 3] 391s 0.294 -0.231 0.694 0.463 391s -0.602 0.124 0.676 0.249 391s 391s y: [2, 2, 3, 3] 391s 0.159 0.928 391s 1.018 -0.102 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.020 -0.325 -0.654 0.584 391s 0.955 -0.716 -0.674 0.379 391s 0.077 -0.370 -0.571 0.884 391s -0.601 -0.917 0.126 -0.737 391s 391s x1: [4, 2, 1, 1] 391s -0.359 -0.819 0.410 -0.792 391s 0.333 0.116 -0.429 -0.294 391s 391s y: [4, 2, 1, 1] 391s -0.457 -0.333 -0.659 1.601 391s 0.065 0.412 -0.032 -0.144 391s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.853 -0.638 -0.932 -0.929 391s -0.561 0.304 -0.054 0.057 391s 0.666 -0.916 -0.555 -0.723 391s -0.195 -0.286 -0.284 0.538 391s 391s x1: [4, 4, 3, 1] 391s -0.119 -0.087 -0.659 0.522 391s -0.010 0.344 -0.837 0.171 391s -0.247 -0.629 -0.496 0.870 391s -0.058 0.210 0.931 0.922 391s 391s y: [4, 4, 3, 1] 391s 0.083 0.106 -0.012 0.516 391s 0.394 0.166 0.019 0.234 391s -0.154 0.024 0.059 0.837 391s -1.908 0.099 -1.414 0.183 391s testing: mul_mat, [2, 4, 3, 3] = [4, 2, 3, 3] * [4, 4, 3, 3] 391s x0: [4, 2, 3, 3] 391s 0.233 0.024 -0.862 0.606 391s -0.531 -0.322 -0.264 0.225 391s 391s x1: [4, 4, 3, 3] 391s 0.944 -0.058 -0.495 -0.507 391s -0.145 -0.340 0.267 -0.799 391s -0.575 0.476 -0.087 0.721 391s -0.716 0.997 -0.129 0.530 391s 391s y: [2, 4, 3, 3] 391s 0.338 -0.466 391s -0.756 -0.064 391s 0.390 0.338 391s 0.290 0.213 391s test-mul-mat0: iter:66/500 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.222 -0.767 0.203 391s 0.093 0.139 -0.973 391s 0.963 0.206 0.983 391s -0.433 -0.958 -0.701 391s 391s x1: [3, 4, 1, 1] 391s -0.241 -0.366 -0.274 391s -0.170 -0.321 -0.790 391s -0.390 -0.400 -0.361 391s -0.925 0.325 0.811 391s 391s y: [4, 4, 1, 1] 391s 0.171 0.193 -0.577 0.646 391s 0.048 0.709 -1.007 0.934 391s 0.147 0.259 -0.813 0.805 391s -0.290 -0.830 -0.027 -0.479 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.654 -0.804 -0.490 391s 0.671 -0.994 0.886 391s 391s x1: [3, 4, 1, 1] 391s -0.961 0.502 0.054 391s -0.765 -0.567 -0.380 391s -0.006 0.067 0.347 391s 0.825 0.746 0.557 391s 391s y: [2, 4, 1, 1] 391s 0.198 -1.096 391s 1.142 -0.285 391s -0.220 0.237 391s -1.412 0.305 391s testing: mul_mat, [2, 2, 1, 3] = [3, 2, 1, 3] * [3, 2, 1, 3] 391s x0: [3, 2, 1, 3] 391s 0.628 -0.580 0.610 391s 0.374 0.977 -0.954 391s 391s x1: [3, 2, 1, 3] 391s -0.107 -0.761 0.089 391s 0.987 0.378 0.116 391s 391s y: [2, 2, 1, 3] 391s 0.429 -0.868 391s 0.471 0.627 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.134 0.596 -0.916 391s 391s x1: [3, 2, 1, 1] 391s -0.779 -0.620 0.745 391s -0.262 0.006 0.144 391s 391s y: [1, 2, 1, 1] 391s -1.157 391s -0.163 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.089 391s 0.480 391s -0.303 391s 391s x1: [1, 3, 1, 1] 391s 0.695 391s -0.331 391s 0.870 391s 391s y: [3, 3, 1, 1] 391s 0.062 0.334 -0.211 391s -0.030 -0.159 0.100 391s 0.078 0.418 -0.264 391s testing: mul_mat, [3, 1, 1, 3] = [3, 3, 1, 3] * [3, 1, 1, 3] 391s x0: [3, 3, 1, 3] 391s -0.947 -0.301 -0.556 391s -0.637 -0.416 0.321 391s 0.631 -0.990 0.016 391s 391s x1: [3, 1, 1, 3] 391s 0.066 0.526 -0.584 391s 391s y: [3, 1, 1, 3] 391s 0.103 -0.449 -0.488 391s test-mul-mat0: iter:67/500 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.762 0.151 -1.000 391s -0.669 -0.681 -0.057 391s 391s x1: [3, 4, 1, 1] 391s -0.768 -0.457 -0.870 391s -0.069 0.126 -0.859 391s 0.375 -0.552 0.156 391s -0.037 0.062 0.306 391s 391s y: [2, 4, 1, 1] 391s 0.215 0.875 391s 0.826 0.009 391s 0.046 0.116 391s -0.325 -0.035 391s testing: mul_mat, [3, 2, 4, 1] = [3, 3, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.672 0.249 0.440 391s -0.997 0.568 -0.616 391s -0.617 0.752 0.336 391s 391s x1: [3, 2, 4, 1] 391s -0.620 -0.817 -0.048 391s 0.956 -0.536 -0.800 391s 391s y: [3, 2, 4, 1] 391s -0.641 0.183 -0.247 391s 0.156 -0.765 -1.262 391s testing: mul_mat, [4, 3, 4, 1] = [3, 4, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 4, 4, 1] 391s 0.954 -0.459 -0.842 391s -0.164 -0.438 -0.008 391s 0.482 -0.651 -0.401 391s 0.386 -0.142 0.380 391s 391s x1: [3, 3, 4, 1] 391s 0.610 0.989 -0.647 391s -0.054 -0.672 -0.432 391s -0.519 -0.722 0.335 391s 391s y: [4, 3, 4, 1] 391s 0.673 -0.529 -0.091 -0.151 391s 0.621 0.307 0.584 -0.090 391s -0.446 0.399 0.085 0.029 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.632 -0.626 -0.576 391s -0.302 0.740 -0.126 391s 391s x1: [3, 4, 1, 1] 391s 0.444 -0.615 -0.528 391s -0.300 -0.378 0.408 391s 0.901 -0.239 -0.129 391s 0.037 0.153 0.807 391s 391s y: [2, 4, 1, 1] 391s 0.408 -0.523 391s 0.191 -0.240 391s -0.346 -0.433 391s -0.584 0.000 391s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.744 0.829 391s 0.269 0.484 391s 391s x1: [2, 3, 4, 1] 391s -0.896 0.632 391s 0.724 0.068 391s -0.773 0.432 391s 391s y: [2, 3, 4, 1] 391s -0.142 0.064 391s 0.595 0.228 391s -0.217 0.001 391s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.765 -0.889 391s -0.856 -0.532 391s 391s x1: [2, 2, 4, 1] 391s 0.014 0.432 391s 0.391 0.854 391s 391s y: [2, 2, 4, 1] 391s -0.374 -0.242 391s -0.461 -0.789 391s test-mul-mat0: iter:68/500 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.274 0.910 391s 0.269 -0.809 391s 0.058 0.205 391s -0.148 0.475 391s 391s x1: [2, 1, 1, 1] 391s 0.733 0.335 391s 391s y: [4, 1, 1, 1] 391s 0.506 -0.073 0.111 0.050 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.089 0.575 391s 0.558 0.262 391s -0.960 -0.880 391s -0.132 0.896 391s 391s x1: [2, 4, 1, 1] 391s -0.030 -0.004 391s 0.585 0.438 391s -0.271 0.022 391s -0.666 0.118 391s 391s y: [4, 4, 1, 1] 391s -0.005 -0.018 0.032 0.001 391s 0.304 0.442 -0.948 0.315 391s -0.012 -0.146 0.241 0.055 391s 0.009 -0.341 0.536 0.194 391s testing: mul_mat, [1, 4, 1, 2] = [2, 1, 1, 2] * [2, 4, 1, 2] 391s x0: [2, 1, 1, 2] 391s -0.102 -0.689 391s 391s x1: [2, 4, 1, 2] 391s -0.464 -0.609 391s 0.226 -0.731 391s 0.725 -0.984 391s 0.543 0.636 391s 391s y: [1, 4, 1, 2] 391s 0.467 391s 0.480 391s 0.604 391s -0.493 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.004 0.583 391s -0.656 0.044 391s 391s x1: [2, 1, 1, 1] 391s 0.429 -0.214 391s 391s y: [2, 1, 1, 1] 391s -0.123 -0.291 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.834 -0.818 391s 0.269 -0.188 391s -0.433 -0.797 391s 391s x1: [2, 2, 1, 1] 391s -0.536 -0.550 391s -0.061 -0.001 391s 391s y: [3, 2, 1, 1] 391s 0.897 -0.040 0.671 391s 0.051 -0.016 0.027 391s testing: mul_mat, [1, 2, 1, 2] = [3, 1, 1, 2] * [3, 2, 1, 2] 391s x0: [3, 1, 1, 2] 391s -0.546 -0.620 0.866 391s 391s x1: [3, 2, 1, 2] 391s 0.467 -0.454 -0.104 391s 0.957 0.132 0.065 391s 391s y: [1, 2, 1, 2] 391s -0.063 391s -0.548 391s test-mul-mat0: iter:69/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.307 0.789 -0.733 391s 391s x1: [3, 1, 1, 1] 391s -0.813 -0.259 0.607 391s 391s y: [1, 1, 1, 1] 391s -0.899 391s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 1, 2, 1] 391s -0.372 -0.341 -0.710 391s 391s x1: [3, 1, 2, 1] 391s 0.510 0.255 -0.187 391s 391s y: [1, 1, 2, 1] 391s -0.144 391s testing: mul_mat, [4, 1, 2, 3] = [3, 4, 2, 3] * [3, 1, 2, 3] 391s x0: [3, 4, 2, 3] 391s -0.809 -0.102 0.667 391s 0.403 -0.157 0.138 391s -0.969 0.503 0.428 391s -0.027 0.187 0.311 391s 391s x1: [3, 1, 2, 3] 391s -0.512 -0.861 -0.736 391s 391s y: [4, 1, 2, 3] 391s 0.011 -0.173 -0.252 -0.377 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.260 0.957 -0.823 391s -0.735 0.011 -0.463 391s 0.954 0.578 -0.963 391s 391s x1: [3, 4, 1, 1] 391s -0.451 -0.112 -0.087 391s 0.311 -0.215 0.433 391s -0.004 0.245 0.573 391s 0.704 -0.558 0.036 391s 391s y: [3, 4, 1, 1] 391s 0.081 0.371 -0.412 391s -0.643 -0.432 -0.245 391s -0.237 -0.260 -0.414 391s -0.746 -0.541 0.315 391s testing: mul_mat, [4, 3, 2, 1] = [3, 4, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.986 0.130 0.345 391s -0.135 0.767 -0.141 391s 0.813 -0.317 -0.322 391s -0.274 -0.859 0.382 391s 391s x1: [3, 3, 2, 1] 391s -0.702 -0.493 -0.265 391s 0.164 -0.939 0.701 391s -0.722 -0.374 0.911 391s 391s y: [4, 3, 2, 1] 391s -0.848 -0.246 -0.329 0.514 391s 0.282 -0.840 0.205 1.029 391s -0.446 -0.317 -0.761 0.866 391s testing: mul_mat, [3, 3, 2, 3] = [4, 3, 2, 3] * [4, 3, 2, 3] 391s x0: [4, 3, 2, 3] 391s 0.903 -0.677 -0.803 -0.291 391s 0.492 0.217 -0.215 -0.687 391s -0.483 0.840 0.803 -0.271 391s 391s x1: [4, 3, 2, 3] 391s 0.566 0.686 -0.191 -0.300 391s 0.229 -0.361 0.538 0.215 391s 0.504 0.351 0.941 -0.366 391s 391s y: [3, 3, 2, 3] 391s 0.287 0.674 0.230 391s -0.043 -0.229 -0.040 391s -0.432 0.373 0.906 391s test-mul-mat0: iter:70/500 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.933 0.951 0.655 391s 391s x1: [3, 2, 1, 1] 391s -0.319 -0.850 0.881 391s -0.492 -0.237 0.392 391s 391s y: [1, 2, 1, 1] 391s 0.065 391s 0.490 391s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 1, 2, 1] 391s -0.919 -0.736 0.405 391s 391s x1: [3, 1, 2, 1] 391s -0.750 -0.194 -0.648 391s 391s y: [1, 1, 2, 1] 391s 0.570 391s testing: mul_mat, [2, 1, 2, 3] = [3, 2, 2, 3] * [3, 1, 2, 3] 391s x0: [3, 2, 2, 3] 391s 0.560 0.325 0.754 391s 0.776 -0.284 -0.925 391s 391s x1: [3, 1, 2, 3] 391s -0.654 0.946 0.509 391s 391s y: [2, 1, 2, 3] 391s 0.324 -1.246 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.835 -0.890 -0.537 391s 0.497 -0.255 -0.974 391s 391s x1: [3, 2, 1, 1] 391s -0.573 -0.344 0.219 391s 0.908 0.238 -0.614 391s 391s y: [2, 2, 1, 1] 391s 0.666 -0.411 391s -0.639 0.989 391s testing: mul_mat, [2, 3, 2, 1] = [2, 2, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.936 -0.238 391s -0.457 -0.136 391s 391s x1: [2, 3, 2, 1] 391s 0.074 -0.041 391s 0.684 0.302 391s -0.134 0.593 391s 391s y: [2, 3, 2, 1] 391s -0.059 -0.028 391s -0.712 -0.354 391s -0.016 -0.019 391s testing: mul_mat, [1, 2, 2, 3] = [2, 1, 2, 3] * [2, 2, 2, 3] 391s x0: [2, 1, 2, 3] 391s 0.218 0.716 391s 391s x1: [2, 2, 2, 3] 391s 0.097 -0.242 391s -0.612 -0.649 391s 391s y: [1, 2, 2, 3] 391s -0.153 391s -0.598 391s test-mul-mat0: iter:71/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.649 0.573 391s 391s x1: [2, 4, 1, 1] 391s 0.325 -0.138 391s -0.026 0.368 391s 0.035 -0.262 391s 0.603 -0.216 391s 391s y: [1, 4, 1, 1] 391s 0.132 391s 0.194 391s -0.127 391s 0.268 391s testing: mul_mat, [1, 1, 4, 1] = [2, 1, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.072 0.233 391s 391s x1: [2, 1, 4, 1] 391s -0.077 -0.895 391s 391s y: [1, 1, 4, 1] 391s -0.203 391s testing: mul_mat, [3, 1, 4, 3] = [2, 3, 4, 3] * [2, 1, 4, 3] 391s x0: [2, 3, 4, 3] 391s -0.420 -0.596 391s -0.703 -0.131 391s 0.857 0.699 391s 391s x1: [2, 1, 4, 3] 391s -0.126 0.008 391s 391s y: [3, 1, 4, 3] 391s 0.048 0.088 -0.103 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.747 -0.258 391s -0.457 0.100 391s -0.438 0.298 391s 391s x1: [2, 3, 1, 1] 391s 0.729 -0.676 391s 0.767 -0.747 391s -0.833 -0.593 391s 391s y: [3, 3, 1, 1] 391s -0.371 -0.401 -0.521 391s -0.381 -0.426 -0.559 391s 0.775 0.322 0.188 391s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.754 0.104 0.488 391s 391s x1: [3, 2, 4, 1] 391s -0.133 -0.674 -0.853 391s 0.854 -0.803 -0.682 391s 391s y: [1, 2, 4, 1] 391s -0.586 391s 0.227 391s testing: mul_mat, [3, 3, 4, 3] = [1, 3, 4, 3] * [1, 3, 4, 3] 391s x0: [1, 3, 4, 3] 391s -0.571 391s -0.811 391s 0.905 391s 391s x1: [1, 3, 4, 3] 391s 0.075 391s -0.352 391s 0.640 391s 391s y: [3, 3, 4, 3] 391s -0.043 -0.061 0.068 391s 0.201 0.285 -0.318 391s -0.365 -0.519 0.579 391s test-mul-mat0: iter:72/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.266 0.799 0.691 0.498 391s -0.109 -0.942 -0.655 -0.801 391s -0.121 0.622 0.695 0.606 391s -0.493 -0.635 -0.924 -0.253 391s 391s x1: [4, 3, 1, 1] 391s -0.846 0.471 -0.500 0.557 391s -0.645 -0.549 0.935 0.309 391s -0.734 -0.925 0.696 -0.527 391s 391s y: [4, 3, 1, 1] 391s 0.532 -0.470 0.385 0.438 391s 0.534 -0.273 0.574 -0.276 391s -0.324 0.917 -0.321 0.440 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.517 0.904 -0.397 -0.585 391s 391s x1: [4, 4, 3, 1] 391s -0.475 -0.716 -0.099 0.996 391s -0.216 -0.542 -0.650 0.235 391s -0.607 0.659 0.501 -0.532 391s 0.355 0.974 0.379 -0.911 391s 391s y: [1, 4, 3, 1] 391s -0.944 391s -0.259 391s 1.022 391s 1.079 391s testing: mul_mat, [1, 1, 3, 1] = [4, 1, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 1, 3, 1] 391s 0.191 0.591 -0.296 -0.912 391s 391s x1: [4, 1, 3, 1] 391s -0.162 0.695 0.883 -0.454 391s 391s y: [1, 1, 3, 1] 391s 0.532 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.526 0.409 -0.952 -0.321 391s 0.822 -0.632 0.080 -0.649 391s -0.231 0.907 -0.161 0.223 391s 391s x1: [4, 1, 1, 1] 391s -0.046 -0.174 0.615 -0.016 391s 391s y: [3, 1, 1, 1] 391s -0.676 0.132 -0.250 391s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.409 0.466 0.159 391s 0.151 -0.854 0.114 391s -0.913 -0.171 -0.503 391s 0.369 0.174 0.157 391s 391s x1: [3, 4, 3, 1] 391s -0.197 -0.351 -0.585 391s -0.605 ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 0.352 -0.498 391s 0.225 0.532 0.073 391s 0.959 -0.685 -0.941 391s 391s y: [4, 4, 3, 1] 391s -0.337 0.203 0.534 -0.225 391s -0.163 -0.449 0.743 -0.240 391s 0.352 -0.412 -0.334 0.187 391s -0.077 0.623 -0.284 0.087 391s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.547 -0.749 -0.996 0.499 391s 0.588 -0.081 0.191 -0.229 391s 0.743 -0.065 -0.642 -0.347 391s 0.112 0.550 0.986 -0.874 391s 391s x1: [4, 3, 3, 1] 391s 0.365 -0.525 0.730 -0.476 391s 0.589 -0.773 0.682 -0.937 391s 0.572 -0.342 -0.337 0.176 391s 391s y: [4, 3, 3, 1] 391s -0.771 0.506 0.001 0.888 391s -0.890 0.754 0.376 1.132 391s 0.366 0.260 0.603 -0.610 391s test-mul-mat0: iter:73/500 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.038 -0.703 391s 391s x1: [2, 2, 1, 1] 391s 0.226 0.560 391s 0.639 0.790 391s 391s y: [1, 2, 1, 1] 391s -0.385 391s -0.531 391s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.242 -0.265 391s 0.640 0.557 391s 391s x1: [2, 1, 3, 1] 391s -0.478 0.964 391s 391s y: [2, 1, 3, 1] 391s -0.371 0.231 391s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.621 0.605 391s 0.709 -0.144 391s 391s x1: [2, 2, 3, 1] 391s -0.581 0.564 391s 0.316 -0.869 391s 391s y: [2, 2, 3, 1] 391s 0.701 -0.492 391s -0.721 0.349 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.484 -0.890 391s 0.978 0.917 391s -0.821 -0.626 391s 0.832 0.800 391s 391s x1: [2, 2, 1, 1] 391s 0.886 0.223 391s -0.923 0.559 391s 391s y: [4, 2, 1, 1] 391s -0.628 1.071 -0.867 0.915 391s -0.051 -0.391 0.408 -0.321 391s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 3, 3, 1] 391s 0.508 0.425 -0.282 -0.573 391s -0.428 -0.055 0.651 -0.014 391s -0.768 -0.968 0.178 -0.809 391s 391s x1: [4, 2, 3, 1] 391s 0.294 -0.294 0.146 0.709 391s 0.335 0.013 -0.912 -0.060 391s 391s y: [3, 2, 3, 1] 391s -0.423 -0.024 -0.489 391s 0.467 -0.737 -0.383 391s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.622 -0.630 0.667 391s -0.462 -0.026 0.371 391s 391s x1: [3, 4, 3, 1] 391s -0.684 -0.274 -0.778 391s 0.493 0.154 0.208 391s 0.684 0.669 0.134 391s -0.404 0.128 0.001 391s 391s y: [2, 4, 3, 1] 391s 0.079 0.034 391s -0.265 -0.155 391s -0.757 -0.284 391s 0.171 0.183 391s test-mul-mat0: iter:74/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.970 0.797 0.213 0.268 391s 0.711 -0.447 0.397 0.201 391s -0.864 -0.855 -0.988 -0.724 391s 0.442 0.222 -0.974 0.339 391s 391s x1: [4, 3, 1, 1] 391s -0.218 0.180 -0.208 0.982 391s -0.104 -0.180 -0.324 0.765 391s -0.944 0.054 -0.697 -0.574 391s 391s y: [4, 3, 1, 1] 391s 0.150 -0.121 -0.472 0.479 391s -0.109 0.031 0.010 0.489 391s -1.175 -1.087 1.874 0.078 391s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 4, 4, 1] 391s 0.030 -0.675 0.692 0.575 391s 0.830 0.814 0.375 -0.984 391s 0.768 -0.269 0.309 0.582 391s 0.616 0.507 -0.531 0.295 391s 391s x1: [4, 4, 4, 1] 391s 0.413 0.012 -0.879 -0.407 391s 0.805 -0.897 0.489 -0.375 391s -0.221 0.254 -0.319 0.832 391s 0.557 0.108 0.860 0.528 391s 391s y: [4, 4, 4, 1] 391s -0.838 0.424 -0.195 0.607 391s 0.752 0.490 0.792 -0.330 391s 0.080 -0.916 0.148 0.408 391s 0.842 0.353 0.972 0.097 391s testing: mul_mat, [4, 4, 4, 4] = [4, 4, 4, 4] * [4, 4, 4, 4] 391s x0: [4, 4, 4, 4] 391s 0.271 0.256 -0.850 0.715 391s -0.806 0.583 0.184 -0.461 391s -0.363 0.938 0.351 -0.963 391s 0.373 0.658 -0.665 0.091 391s 391s x1: [4, 4, 4, 4] 391s -0.758 -0.710 -0.508 -0.430 391s 0.165 -0.958 -0.333 -0.229 391s -0.100 -0.654 0.993 0.266 391s -0.507 -0.490 0.952 0.621 391s 391s y: [4, 4, 4, 4] 391s -0.263 0.302 -0.155 -0.452 391s -0.082 -0.648 -0.856 -0.369 391s -0.849 -0.241 -0.484 -1.104 391s -0.628 0.011 -0.540 -1.088 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.298 -0.397 0.156 -0.057 391s -0.807 -0.764 -0.308 -0.646 391s -0.228 -0.206 -0.388 0.142 391s -0.217 -0.859 0.169 -0.592 391s 391s x1: [4, 4, 1, 1] 391s 0.973 0.465 0.623 -0.787 391s 0.151 -0.225 0.438 0.937 391s 0.083 -0.831 0.918 -0.472 391s 0.540 -0.968 0.234 -0.214 391s 391s y: [4, 4, 1, 1] 391s -0.332 -0.824 -0.671 -0.040 391s 0.059 -0.691 -0.025 -0.320 391s 0.475 0.590 -0.272 1.130 391s 0.271 0.369 -0.046 0.880 391s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.699 0.414 0.714 -0.828 391s -0.883 -0.945 0.800 0.816 391s 391s x1: [4, 4, 4, 1] 391s 0.976 0.356 0.559 -0.249 391s -0.206 0.496 0.835 -0.037 391s 0.413 -0.638 -0.496 0.446 391s 0.ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 596 0.289 -0.991 -0.702 391s 391s y: [2, 4, 4, 1] 391s 0.070 -0.954 391s 0.976 0.351 391s -1.276 0.204 391s -0.423 -2.165 391s testing: mul_mat, [3, 4, 4, 4] = [2, 3, 4, 4] * [2, 4, 4, 4] 391s x0: [2, 3, 4, 4] 391s -0.231 -0.207 391s 0.828 0.398 391s 0.740 -0.950 391s 391s x1: [2, 4, 4, 4] 391s 0.944 0.352 391s -0.804 0.658 391s 0.152 -0.633 391s 0.474 -0.216 391s 391s y: [3, 4, 4, 4] 391s -0.291 0.922 0.364 391s 0.050 -0.404 -1.221 391s 0.096 -0.126 0.714 391s -0.065 0.306 0.556 391s test-mul-mat0: iter:75/500 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.256 -0.673 -0.023 391s -0.059 -0.159 0.299 391s 0.050 0.195 0.333 391s -0.854 0.372 -0.972 391s 391s x1: [3, 3, 1, 1] 391s 0.823 0.537 -0.410 391s -0.867 0.633 -0.574 391s 0.901 -0.761 -0.513 391s 391s y: [4, 3, 1, 1] 391s -0.142 -0.257 0.009 -0.104 391s -0.635 -0.221 -0.111 1.534 391s 0.755 -0.085 -0.274 -0.554 391s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 3, 2, 1] 391s -0.420 -0.521 0.909 391s -0.274 0.851 0.938 391s 0.905 0.229 0.040 391s 391s x1: [3, 4, 2, 1] 391s 0.179 0.378 0.102 391s -0.075 -0.036 0.064 391s 0.560 0.926 -0.386 391s 0.383 0.463 0.204 391s 391s y: [3, 4, 2, 1] 391s -0.179 0.369 0.253 391s 0.109 0.050 -0.074 391s -1.068 0.273 0.704 391s -0.216 0.480 0.461 391s testing: mul_mat, [3, 3, 2, 4] = [3, 3, 2, 4] * [3, 3, 2, 4] 391s x0: [3, 3, 2, 4] 391s 0.400 -0.776 0.471 391s -0.506 0.472 0.885 391s 0.314 0.426 0.643 391s 391s x1: [3, 3, 2, 4] 391s 0.289 -0.324 0.552 391s -0.295 -0.989 0.669 391s 0.006 -0.399 -0.888 391s 391s y: [3, 3, 2, 4] 391s 0.626 0.189 0.307 391s 0.965 0.274 -0.084 391s -0.106 -0.977 -0.739 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.477 0.807 -0.809 391s 391s x1: [3, 3, 1, 1] 391s 0.237 0.622 -0.772 391s -0.832 -0.126 0.329 391s 0.893 0.023 0.669 391s 391s y: [1, 3, 1, 1] 391s 1.014 391s 0.029 391s -0.948 391s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.243 391s -0.892 391s -0.873 391s 0.609 391s 391s x1: [1, 3, 2, 1] 391s -0.836 391s 0.838 391s 0.774 391s 391s y: [4, 3, 2, 1] 391s 0.203 0.745 0.730 -0.509 391s -0.204 -0.748 -0.732 0.511 391s -0.188 -0.690 -0.676 0.471 391s testing: mul_mat, [1, 1, 2, 4] = [4, 1, 2, 4] * [4, 1, 2, 4] 391s x0: [4, 1, 2, 4] 391s 0.766 -0.619 0.544 -0.481 391s 391s x1: [4, 1, 2, 4] 391s -0.042 -0.132 -0.475 -0.806 391s 391s y: [1, 1, 2, 4] 391s 0.178 391s test-mul-mat0: iter:76/500 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.884 -0.654 391s 0.065 0.191 391s 0.317 0.489 391s 391s x1: [2, 3, 1, 1] 391s -0.719 0.622 391s -0.659 0.577 391s 0.974 -0.388 391s 391s y: [3, 3, 1, 1] 391s -1.042 0.072 0.076 391s -0.959 0.067 0.073 391s 1.115 -0.011 0.119 391s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.709 -0.625 391s 0.149 -0.735 391s 391s x1: [2, 3, 3, 1] 391s 0.885 0.472 391s 0.039 -0.277 391s -0.080 0.841 391s 391s y: [2, 3, 3, 1] 391s -0.922 -0.215 391s 0.145 0.210 391s -0.469 -0.631 391s testing: mul_mat, [1, 2, 3, 2] = [2, 1, 3, 2] * [2, 2, 3, 2] 391s x0: [2, 1, 3, 2] 391s 0.626 0.752 391s 391s x1: [2, 2, 3, 2] 391s 0.224 -0.878 391s -0.475 0.947 391s 391s y: [1, 2, 3, 2] 391s -0.519 391s 0.415 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.250 0.623 391s 391s x1: [2, 1, 1, 1] 391s 0.399 0.020 391s 391s y: [1, 1, 1, 1] 391s -0.087 391s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.284 391s -0.990 391s 391s x1: [1, 2, 3, 1] 391s 0.161 391s -0.980 391s 391s y: [2, 2, 3, 1] 391s -0.046 -0.159 391s 0.278 0.971 391s testing: mul_mat, [4, 1, 3, 2] = [2, 4, 3, 2] * [2, 1, 3, 2] 391s x0: [2, 4, 3, 2] 391s 0.805 0.976 391s -0.648 -0.072 391s 0.767 -0.425 391s -0.445 0.851 391s 391s x1: [2, 1, 3, 2] 391s -0.786 0.113 391s 391s y: [4, 1, 3, 2] 391s -0.522 0.501 -0.651 0.446 391s test-mul-mat0: iter:77/500 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.685 -0.939 0.132 0.609 391s 0.979 -0.051 0.254 0.665 391s 0.321 -0.125 -0.722 -0.732 391s 391s x1: [4, 1, 1, 1] 391s -0.556 0.515 -0.646 0.079 391s 391s y: [3, 1, 1, 1] 391s -0.141 -0.682 0.166 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.465 -0.427 0.857 -0.112 391s 391s x1: [4, 3, 1, 1] 391s -0.653 -0.999 -0.629 -0.534 391s -0.925 -0.652 0.243 0.170 391s -0.804 -0.836 -0.624 -0.871 391s 391s y: [1, 3, 1, 1] 391s 0.251 391s 0.898 391s 0.294 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.188 0.692 0.059 -0.934 391s 0.356 -0.620 -0.059 0.634 391s 391s x1: [4, 1, 1, 1] 391s 0.652 -0.023 0.204 0.713 391s 391s y: [2, 1, 1, 1] 391s -0.792 0.686 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.562 0.257 0.504 -0.736 391s 0.162 0.691 0.792 0.361 391s 391s x1: [4, 2, 1, 1] 391s -0.353 0.288 0.635 0.019 391s 0.755 0.711 0.367 -0.002 391s 391s y: [2, 2, 1, 1] 391s 0.182 0.652 391s 0.793 0.902 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.238 0.886 391s -0.324 0.704 391s 391s x1: [2, 4, 1, 1] 391s -0.320 0.916 391s -0.662 0.884 391s 0.628 -0.552 391s -0.304 0.320 391s 391s y: [2, 4, 1, 1] 391s 0.888 0.748 391s 0.941 0.836 391s -0.639 -0.592 391s 0.356 0.324 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.253 -0.081 391s 0.657 0.745 391s 391s x1: [2, 2, 1, 1] 391s -0.690 -0.466 391s -0.008 0.946 391s 391s y: [2, 2, 1, 1] 391s 0.212 -0.800 391s -0.074 0.699 391s test-mul-mat0: iter:78/500 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.564 0.770 391s -0.732 -0.648 391s 391s x1: [2, 1, 1, 1] 391s -0.828 -0.590 391s 391s y: [2, 1, 1, 1] 391s 0.012 0.989 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.301 -0.937 391s 0.610 0.977 391s 391s x1: [2, 2, 1, 1] 391s -0.891 -0.848 391s 0.980 -0.444 391s 391s y: [2, 2, 1, 1] 391s 0.527 -1.373 391s 0.710 0.165 391s testing: mul_mat, [2, 2, 1, 2] = [2, 2, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 2, 1, 2] 391s -0.045 -0.202 391s -0.573 0.437 391s 391s x1: [2, 2, 1, 2] 391s 0.949 0.314 391s -0.713 -0.517 391s 391s y: [2, 2, 1, 2] 391s -0.106 -0.406 391s 0.136 0.182 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.993 -0.622 391s -0.771 -0.451 391s 391s x1: [2, 2, 1, 1] 391s 0.809 -0.955 391s 0.884 -0.923 391s 391s y: [2, 2, 1, 1] 391s 1.398 -0.194 391s 1.453 -0.266 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.438 0.942 391s -0.999 0.045 391s 391s x1: [2, 2, 1, 1] 391s 0.077 0.678 391s -0.388 -0.313 391s 391s y: [2, 2, 1, 1] 391s 0.605 -0.046 391s -0.125 0.373 391s testing: mul_mat, [2, 2, 1, 2] = [2, 2, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 2, 1, 2] 391s 0.862 0.082 391s -0.285 0.976 391s 391s x1: [2, 2, 1, 2] 391s -0.693 -0.825 391s 0.081 0.360 391s 391s y: [2, 2, 1, 2] 391s -0.665 -0.607 391s 0.099 0.328 391s test-mul-mat0: iter:79/500 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.099 0.353 0.701 391s 391s x1: [3, 3, 1, 1] 391s -0.598 0.906 0.580 391s 0.014 -0.407 0.235 391s 0.576 -0.406 0.177 391s 391s y: [1, 3, 1, 1] 391s 0.786 391s 0.020 391s -0.076 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.910 0.211 -0.719 391s -0.375 -0.707 -0.743 391s 0.149 0.421 -0.883 391s 391s x1: [3, 1, 1, 1] 391s -0.739 0.418 0.738 391s 391s y: [3, 1, 1, 1] 391s -1.115 -0.567 -0.586 391s testing: mul_mat, [3, 3, 1, 2] = [3, 3, 1, 2] * [3, 3, 1, 2] 391s x0: [3, 3, 1, 2] 391s 0.611 0.021 0.255 391s 0.521 -0.768 0.536 391s -0.853 -0.475 0.793 391s 391s x1: [3, 3, 1, 2] 391s 0.750 -0.055 0.970 391s -0.420 -0.152 -0.628 391s -0.514 -0.572 0.386 391s 391s y: [3, 3, 1, 2] 391s 0.705 0.953 0.154 391s -0.420 -0.439 -0.068 391s -0.227 0.378 1.016 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.137 -0.525 0.996 391s 0.463 -0.516 -0.284 391s -0.723 0.532 0.068 391s 391s x1: [3, 3, 1, 1] 391s -0.887 -0.833 0.637 391s 0.193 0.830 -0.402 391s -0.134 -0.330 -0.820 391s 391s y: [3, 3, 1, 1] 391s 1.193 -0.162 0.242 391s -0.863 -0.225 0.275 391s -0.625 0.341 -0.134 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.819 0.986 0.816 391s -0.099 -0.462 0.136 391s -0.776 0.417 -0.717 391s 391s x1: [3, 3, 1, 1] 391s -0.857 0.242 -0.139 391s 0.438 0.188 -0.230 391s -0.515 -0.920 -0.350 391s 391s y: [3, 3, 1, 1] 391s -0.578 -0.046 0.866 391s 0.357 -0.162 -0.097 391s -1.616 0.429 0.267 391s testing: mul_mat, [2, 3, 1, 2] = [3, 2, 1, 2] * [3, 3, 1, 2] 391s x0: [3, 2, 1, 2] 391s -0.510 0.998 0.782 391s -0.119 -0.691 -0.778 391s 391s x1: [3, 3, 1, 2] 391s -0.513 0.316 0.724 391s 0.350 -0.221 -1.000 391s 0.825 0.263 0.532 391s 391s y: [2, 3, 1, 2] 391s 1.142 -0.720 391s -1.181 0.889 391s 0.257 -0.694 391s test-mul-mat0: iter:80/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.986 0.237 391s -0.041 -0.576 391s 391s x1: [2, 4, 1, 1] 391s 0.550 0.645 391s 0.774 -0.187 391s 0.177 0.594 391s -0.207 0.776 391s 391s y: [2, 4, 1, 1] 391s -0.390 -0.394 391s -0.807 0.076 391s -0.034 -0.350 391s 0.388 -0.439 391s testing: mul_mat, [3, 2, 3, 1] = [2, 3, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.375 -0.772 391s -0.325 -0.730 391s -0.823 -0.775 391s 391s x1: [2, 2, 3, 1] 391s 0.467 -0.154 391s -0.087 -0.652 391s 391s y: [3, 2, 3, 1] 391s -0.056 -0.039 -0.265 391s 0.536 0.504 0.577 391s testing: mul_mat, [2, 3, 3, 2] = [2, 2, 3, 2] * [2, 3, 3, 2] 391s x0: [2, 2, 3, 2] 391s 0.930 -0.909 391s 0.792 -0.572 391s 391s x1: [2, 3, 3, 2] 391s 0.571 -0.092 391s 0.182 0.919 391s 0.751 -0.596 391s 391s y: [2, 3, 3, 2] 391s 0.615 0.505 391s -0.666 -0.382 391s 1.240 0.936 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.598 -0.507 391s 0.357 0.149 391s -0.630 -0.202 391s 391s x1: [2, 2, 1, 1] 391s -0.980 -0.442 391s 0.127 0.486 391s 391s y: [3, 2, 1, 1] 391s 0.810 -0.416 0.707 391s -0.322 0.118 -0.178 391s testing: mul_mat, [1, 2, 3, 1] = [3, 1, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.299 0.766 0.807 391s 391s x1: [3, 2, 3, 1] 391s -0.896 -0.508 0.628 391s 0.860 0.614 0.535 391s 391s y: [1, 2, 3, 1] 391s 0.385 391s 0.645 391s testing: mul_mat, [2, 3, 3, 2] = [1, 2, 3, 2] * [1, 3, 3, 2] 391s x0: [1, 2, 3, 2] 391s 0.439 391s 0.372 391s 391s x1: [1, 3, 3, 2] 391s -0.276 391s 0.759 391s 0.414 391s 391s y: [2, 3, 3, 2] 391s -0.121 -0.103 391s 0.333 0.282 391s 0.182 0.154 391s test-mul-mat0: iter:81/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.559 391s -0.125 391s 391s x1: [1, 1, 1, 1] 391s 0.010 391s 391s y: [2, 1, 1, 1] 391s -0.006 -0.001 391s testing: mul_mat, [2, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.856 391s -0.557 391s 391s x1: [1, 2, 4, 1] 391s 0.296 391s 0.159 391s 391s y: [2, 2, 4, 1] 391s 0.254 -0.165 391s 0.136 -0.089 391s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 3, 4, 1] 391s -0.571 391s 0.527 391s 0.738 391s 391s x1: [1, 2, 4, 1] 391s 0.553 391s -0.243 391s 391s y: [3, 2, 4, 1] 391s -0.315 0.291 0.408 391s 0.138 -0.128 -0.179 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.053 391s -0.034 391s 0.011 391s 391s x1: [1, 3, 1, 1] 391s -0.467 391s 0.504 391s 0.738 391s 391s y: [3, 3, 1, 1] 391s 0.025 0.016 -0.005 391s -0.026 -0.017 0.006 391s -0.039 -0.025 0.008 391s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.892 0.626 0.813 391s 391s x1: [3, 1, 4, 1] 391s -0.469 0.332 -0.554 391s 391s y: [1, 1, 4, 1] 391s -0.661 391s testing: mul_mat, [2, 3, 4, 1] = [1, 2, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 2, 4, 1] 391s -0.833 391s 0.640 391s 391s x1: [1, 3, 4, 1] 391s 0.131 391s 0.932 391s 0.788 391s 391s y: [2, 3, 4, 1] 391s -0.109 0.084 391s -0.776 0.596 391s -0.656 0.504 391s test-mul-mat0: iter:82/500 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.749 -0.479 391s 391s x1: [2, 3, 1, 1] 391s -0.154 -0.875 391s 0.922 -0.527 391s -0.681 -0.267 391s 391s y: [1, 3, 1, 1] 391s 0.535 391s -0.438 391s 0.638 391s testing: mul_mat, [3, 1, 4, 1] = [2, 3, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.516 0.864 391s -0.026 -0.440 391s -0.604 -0.498 391s 391s x1: [2, 1, 4, 1] 391s 0.682 0.148 391s 391s y: [3, 1, 4, 1] 391s 0.480 -0.083 -0.486 391s testing: mul_mat, [1, 3, 4, 2] = [2, 1, 4, 2] * [2, 3, 4, 2] 391s x0: [2, 1, 4, 2] 391s 0.737 0.914 391s 391s x1: [2, 3, 4, 2] 391s -0.515 -0.568 391s -0.309 0.049 391s 0.417 -0.611 391s 391s y: [1, 3, 4, 2] 391s -0.898 391s -0.183 391s -0.251 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.461 -0.551 391s 0.152 -0.165 391s 391s x1: [2, 1, 1, 1] 391s 0.445 0.532 391s 391s y: [2, 1, 1, 1] 391s -0.088 -0.020 391s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.152 -0.602 391s 0.435 -0.411 391s 391s x1: [2, 2, 4, 1] 391s -0.734 -0.600 391s -0.930 0.418 391s 391s y: [2, 2, 4, 1] 391s 0.250 -0.072 391s -0.393 -0.576 391s testing: mul_mat, [1, 2, 4, 2] = [2, 1, 4, 2] * [2, 2, 4, 2] 391s x0: [2, 1, 4, 2] 391s -0.275 -0.834 391s 391s x1: [2, 2, 4, 2] 391s 0.057 0.612 391s -0.499 0.879 391s 391s y: [1, 2, 4, 2] 391s -0.527 391s -0.596 391s test-mul-mat0: iter:83/500 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.529 -0.607 -0.037 -0.363 391s -0.200 -0.574 0.240 -0.353 391s -0.332 0.154 0.005 0.772 391s -0.936 -0.921 -0.439 0.122 391s 391s x1: [4, 2, 1, 1] 391s -0.288 -0.701 -0.526 -0.046 391s -0.670 -0.986 -0.816 -0.445 391s 391s y: [4, 2, 1, 1] 391s 0.614 0.350 -0.050 1.141 391s 1.144 0.661 -0.277 1.839 391s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.430 0.506 -0.782 -0.826 391s 0.741 -0.662 0.372 -0.768 391s 391s x1: [4, 4, 4, 1] 391s 0.304 0.431 -0.084 0.926 391s 0.628 -0.867 -0.362 0.927 391s -0.394 0.593 -0.743 -0.380 391s 0.777 -0.188 -0.142 -0.752 391s 391s y: [2, 4, 4, 1] 391s -0.611 -0.803 391s -1.191 0.193 391s 1.364 -0.669 391s 0.302 1.224 391s testing: mul_mat, [1, 2, 4, 3] = [4, 1, 4, 3] * [4, 2, 4, 3] 391s x0: [4, 1, 4, 3] 391s -0.755 0.194 0.565 -0.961 391s 391s x1: [4, 2, 4, 3] 391s -0.158 -0.476 0.462 -0.417 391s -0.138 -0.166 -0.185 0.942 391s 391s y: [1, 2, 4, 3] 391s 0.688 391s -0.938 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.867 -0.938 -0.823 0.690 391s -0.320 0.672 -0.348 0.135 391s 0.522 0.652 -0.661 0.610 391s -0.102 -0.397 -0.753 -0.546 391s 391s x1: [4, 1, 1, 1] 391s 0.682 0.479 0.368 0.075 391s 391s y: [4, 1, 1, 1] 391s -0.109 -0.014 0.471 -0.577 391s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.331 -0.689 -0.678 -0.420 391s -0.081 -0.250 -0.533 0.372 391s 391s x1: [4, 4, 4, 1] 391s 0.439 -0.086 0.246 -0.294 391s -0.324 0.467 0.731 0.966 391s -0.548 -0.589 0.648 0.931 391s 0.779 -0.277 -0.383 0.645 391s 391s y: [2, 4, 4, 1] 391s 0.160 -0.255 391s -1.330 -0.120 391s -0.605 0.192 391s 0.438 0.450 391s testing: mul_mat, [2, 4, 4, 3] = [2, 2, 4, 3] * [2, 4, 4, 3] 391s x0: [2, 2, 4, 3] 391s -0.164 0.258 391s 0.947 0.936 391s 391s x1: [2, 4, 4, 3] 391s 0.400 -0.263 391s 0.173 0.722 391s 0.204 0.754 391s 0.094 0.759 391s 391s y: [2, 4, 4, 3] 391s -0.134 0.133 391s 0.158 0.840 391s 0.161 0.899 391s 0.180 0.799 391s test-mul-mat0: iter:84/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.852 0.157 391s 0.691 0.340 391s 0.756 -0.172 391s 391s x1: [2, 2, 1, 1] 391s -0.047 -0.847 391s 0.268 -0.688 391s 391s y: [3, 2, 1, 1] 391s -0.093 -0.321 0.110 391s -0.336 -0.049 0.321 391s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.453 -0.027 391s 391s x1: [2, 3, 2, 1] 391s -0.521 0.613 391s 0.901 0.019 391s 0.709 -0.646 391s 391s y: [1, 3, 2, 1] 391s 0.220 391s -0.408 391s -0.304 391s testing: mul_mat, [2, 1, 2, 4] = [2, 2, 2, 4] * [2, 1, 2, 4] 391s x0: [2, 2, 2, 4] 391s 0.255 0.428 391s 0.177 -0.125 391s 391s x1: [2, 1, 2, 4] 391s 0.929 -0.140 391s 391s y: [2, 1, 2, 4] 391s 0.177 0.182 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.214 -0.593 391s 391s x1: [2, 2, 1, 1] 391s 0.207 -0.320 391s 0.139 0.692 391s 391s y: [1, 2, 1, 1] 391s 0.145 391s -0.440 391s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.005 391s 391s x1: [1, 2, 2, 1] 391s 0.573 391s 0.091 391s 391s y: [1, 2, 2, 1] 391s -0.003 391s -0.000 391s testing: mul_mat, [3, 1, 2, 4] = [1, 3, 2, 4] * [1, 1, 2, 4] 391s x0: [1, 3, 2, 4] 391s 0.414 391s 0.706 391s -0.284 391s 391s x1: [1, 1, 2, 4] 391s -0.272 391s 391s y: [3, 1, 2, 4] 391s -0.113 -0.192 0.077 391s test-mul-mat0: iter:85/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.652 391s 0.850 391s -0.641 391s 391s x1: [1, 3, 1, 1] 391s -0.151 391s -0.359 391s 0.134 391s 391s y: [3, 3, 1, 1] 391s 0.098 -0.128 0.097 391s 0.234 -0.305 0.230 391s -0.088 0.114 -0.086 391s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 2, 3, 1] 391s 0.240 391s -0.733 391s 391s x1: [1, 3, 3, 1] 391s 0.063 391s -0.656 391s 0.958 391s 391s y: [2, 3, 3, 1] 391s 0.015 -0.046 391s -0.157 0.480 391s 0.229 -0.702 391s testing: mul_mat, [1, 2, 3, 2] = [1, 1, 3, 2] * [1, 2, 3, 2] 391s x0: [1, 1, 3, 2] 391s -0.858 391s 391s x1: [1, 2, 3, 2] 391s 0.173 391s -0.092 391s 391s y: [1, 2, 3, 2] 391s -0.149 391s 0.079 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.078 391s 0.971 391s 391s x1: [1, 1, 1, 1] 391s -0.468 391s 391s y: [2, 1, 1, 1] 391s 0.036 -0.455 391s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.417 -0.946 391s 0.093 0.158 391s -0.522 0.255 391s 0.322 -0.279 391s 391s x1: [2, 1, 3, 1] 391s 0.064 0.987 391s 391s y: [4, 1, 3, 1] 391s -0.961 0.161 0.219 -0.254 391s testing: mul_mat, [1, 2, 3, 2] = [4, 1, 3, 2] * [4, 2, 3, 2] 391s x0: [4, 1, 3, 2] 391s 0.258 -0.918 0.189 0.977 391s 391s x1: [4, 2, 3, 2] 391s -0.486 -0.921 -0.053 0.846 391s 0.126 -0.134 0.416 0.709 391s 391s y: [1, 2, 3, 2] 391s 1.537 391s 0.927 391s test-mul-mat0: iter:86/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.435 0.467 0.166 0.204 391s 0.868 -0.327 0.401 0.594 391s 0.447 -0.427 -0.645 0.448 391s -0.387 0.182 -0.312 -0.428 391s 391s x1: [4, 3, 1, 1] 391s 0.245 0.852 0.950 -0.497 391s 0.934 0.139 -0.520 0.913 391s 0.188 -0.589 0.640 0.267 391s 391s y: [4, 3, 1, 1] 391s 0.348 0.019 -1.090 -0.024 391s -0.241 1.099 1.103 -0.565 391s -0.196 0.771 0.043 -0.494 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.561 0.468 -0.221 0.112 391s 391s x1: [4, 4, 1, 1] 391s -0.641 0.900 0.817 -0.789 391s 0.850 -0.680 -0.854 -0.011 391s -0.200 -0.941 -0.823 0.212 391s 0.699 0.444 0.273 -0.736 391s 391s y: [1, 4, 1, 1] 391s 0.511 391s -0.607 391s -0.123 391s -0.327 391s testing: mul_mat, [3, 1, 1, 3] = [4, 3, 1, 3] * [4, 1, 1, 3] 391s x0: [4, 3, 1, 3] 391s -0.520 -0.555 -0.236 -0.374 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 0.434 0.564 -0.316 0.610 391s -0.224 -0.617 0.054 -0.951 391s 391s x1: [4, 1, 1, 3] 391s -0.131 -0.627 -0.441 0.442 391s 391s y: [3, 1, 1, 3] 391s 0.355 -0.002 -0.028 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.944 0.670 0.980 0.513 391s 0.825 0.118 0.961 0.947 391s -0.049 -0.148 0.445 0.340 391s -0.629 0.423 0.437 0.829 391s 391s x1: [4, 3, 1, 1] 391s 0.880 0.812 -0.054 -0.344 391s -0.805 -1.000 -0.295 -0.157 391s 0.966 0.192 -0.041 -0.291 391s 391s y: [4, 3, 1, 1] 391s 1.145 0.445 -0.304 -0.519 391s -1.799 -1.215 0.002 -0.176 391s 0.850 0.504 -0.193 -0.786 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.994 0.197 0.792 0.124 391s 391s x1: [4, 4, 1, 1] 391s 0.075 -0.324 0.710 -0.113 391s 0.622 -0.634 0.082 0.622 391s 0.070 0.925 0.588 -0.737 391s -0.117 -0.703 0.055 -0.173 391s 391s y: [1, 4, 1, 1] 391s 0.410 391s -0.602 391s 0.487 391s -0.001 391s testing: mul_mat, [4, 4, 1, 3] = [1, 4, 1, 3] * [1, 4, 1, 3] 391s x0: [1, 4, 1, 3] 391s -0.431 391s -0.822 391s 0.428 391s 0.651 391s 391s x1: [1, 4, 1, 3] 391s -0.950 391s 0.215 391s 0.104 391s -0.990 391s 391s y: [4, 4, 1, 3] 391s 0.409 0.781 -0.407 -0.619 391s -0.093 -0.177 0.092 0.140 391s -0.045 -0.085 0.044 0.067 391s 0.426 0.814 -0.424 -0.645 391s test-mul-mat0: iter:87/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.457 391s 391s x1: [1, 4, 1, 1] 391s 0.599 391s -0.055 391s -0.467 391s -0.186 391s 391s y: [1, 4, 1, 1] 391s -0.273 391s 0.025 391s 0.213 391s 0.085 391s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 1, 4, 1] 391s 0.419 391s 391s x1: [1, 1, 4, 1] 391s 0.475 391s 391s y: [1, 1, 4, 1] 391s 0.199 391s testing: mul_mat, [4, 1, 4, 2] = [1, 4, 4, 2] * [1, 1, 4, 2] 391s x0: [1, 4, 4, 2] 391s -0.978 391s -0.924 391s -0.015 391s 0.308 391s 391s x1: [1, 1, 4, 2] 391s -0.665 391s 391s y: [4, 1, 4, 2] 391s 0.650 0.615 0.010 -0.205 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.377 391s 0.922 391s 0.865 391s -0.053 391s 391s x1: [1, 4, 1, 1] 391s -0.442 391s 0.307 391s -0.168 391s -0.559 391s 391s y: [4, 4, 1, 1] 391s 0.166 -0.407 -0.382 0.023 391s -0.116 0.283 0.266 -0.016 391s 0.063 -0.155 -0.145 0.009 391s 0.211 -0.516 -0.483 0.030 391s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.836 -0.142 -0.541 -0.842 391s 391s x1: [4, 1, 4, 1] 391s -0.782 0.664 0.815 0.988 391s 391s y: [1, 1, 4, 1] 391s -2.021 391s testing: mul_mat, [3, 4, 4, 2] = [1, 3, 4, 2] * [1, 4, 4, 2] 391s x0: [1, 3, 4, 2] 391s 0.823 391s -0.319 391s 0.661 391s 391s x1: [1, 4, 4, 2] 391s -0.401 391s 0.587 391s 0.161 391s 0.348 391s 391s y: [3, 4, 4, 2] 391s -0.330 0.128 -0.265 391s 0.483 -0.187 0.388 391s 0.133 -0.051 0.106 391s 0.286 -0.111 0.230 391s test-mul-mat0: iter:88/500 391s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.243 391s 0.246 391s -0.267 391s 391s x1: [1, 1, 1, 1] 391s 0.423 391s 391s y: [3, 1, 1, 1] 391s 0.103 0.104 -0.113 391s testing: mul_mat, [1, 3, 3, 1] = [1, 1, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 1, 3, 1] 391s 0.682 391s 391s x1: [1, 3, 3, 1] 391s -0.096 391s 0.142 391s 0.865 391s 391s y: [1, 3, 3, 1] 391s -0.066 391s 0.097 391s 0.590 391s testing: mul_mat, [1, 1, 3, 2] = [1, 1, 3, 2] * [1, 1, 3, 2] 391s x0: [1, 1, 3, 2] 391s -0.649 391s 391s x1: [1, 1, 3, 2] 391s 0.126 391s 391s y: [1, 1, 3, 2] 391s -0.082 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.286 391s 391s x1: [1, 1, 1, 1] 391s -0.464 391s 391s y: [1, 1, 1, 1] 391s 0.133 391s testing: mul_mat, [1, 1, 3, 1] = [1, 1, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 1, 3, 1] 391s 0.148 391s 391s x1: [1, 1, 3, 1] 391s 0.782 391s 391s y: [1, 1, 3, 1] 391s 0.116 391s testing: mul_mat, [2, 1, 3, 2] = [1, 2, 3, 2] * [1, 1, 3, 2] 391s x0: [1, 2, 3, 2] 391s -0.970 391s 0.072 391s 391s x1: [1, 1, 3, 2] 391s -0.466 391s 391s y: [2, 1, 3, 2] 391s 0.452 -0.033 391s test-mul-mat0: iter:89/500 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.280 0.386 391s -0.347 -0.017 391s 0.899 -0.157 391s -0.071 0.891 391s 391s x1: [2, 3, 1, 1] 391s -0.411 -0.039 391s 0.703 -0.793 391s 0.886 -0.149 391s 391s y: [4, 3, 1, 1] 391s 0.100 0.143 -0.363 -0.005 391s -0.503 -0.231 0.756 -0.756 391s -0.306 -0.305 0.820 -0.195 391s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.570 0.181 391s 0.790 -0.323 391s 0.435 0.583 391s 0.315 0.059 391s 391s x1: [2, 4, 4, 1] 391s -0.127 -0.999 391s 0.928 -0.335 391s 0.849 0.107 391s -0.746 0.793 391s 391s y: [4, 4, 4, 1] 391s -0.253 0.222 -0.638 -0.099 391s 0.468 0.841 0.208 0.272 391s 0.503 0.636 0.432 0.274 391s -0.282 -0.845 0.138 -0.188 391s testing: mul_mat, [2, 4, 4, 2] = [2, 2, 4, 2] * [2, 4, 4, 2] 391s x0: [2, 2, 4, 2] 391s 0.551 0.384 391s -0.481 ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s -0.595 391s 391s x1: [2, 4, 4, 2] 391s 0.668 -0.509 391s -0.959 0.103 391s -0.926 0.356 391s -0.838 0.855 391s 391s y: [2, 4, 4, 2] 391s 0.172 -0.019 391s -0.488 0.400 391s -0.373 0.233 391s -0.133 -0.105 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.436 0.025 391s 391s x1: [2, 2, 1, 1] 391s 0.209 0.631 391s 0.463 -0.626 391s 391s y: [1, 2, 1, 1] 391s 0.107 391s 0.186 391s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 3, 4, 1] 391s 0.025 391s 0.973 391s 0.220 391s 391s x1: [1, 2, 4, 1] 391s -0.729 391s -0.844 391s 391s y: [3, 2, 4, 1] 391s -0.018 -0.709 -0.161 391s -0.021 -0.821 -0.186 391s testing: mul_mat, [2, 1, 4, 2] = [3, 2, 4, 2] * [3, 1, 4, 2] 391s x0: [3, 2, 4, 2] 391s 0.740 0.798 -0.714 391s -0.586 -0.602 -0.443 391s 391s x1: [3, 1, 4, 2] 391s -0.517 -0.016 0.817 391s 391s y: [2, 1, 4, 2] 391s -0.978 -0.050 391s test-mul-mat0: iter:90/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.099 -0.437 -0.019 391s 0.689 -0.377 -0.702 391s 391s x1: [3, 3, 1, 1] 391s -0.631 0.449 -0.883 391s -0.159 -0.547 0.428 391s 0.257 -0.094 -0.211 391s 391s y: [2, 3, 1, 1] 391s -0.241 0.016 391s 0.215 -0.204 391s 0.071 0.361 391s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.683 0.494 0.395 391s 0.223 -0.717 0.614 391s 391s x1: [3, 2, 3, 1] 391s 0.161 0.953 0.989 391s 0.207 0.170 0.333 391s 391s y: [2, 2, 3, 1] 391s 0.751 -0.040 391s 0.074 0.129 391s testing: mul_mat, [1, 2, 3, 3] = [3, 1, 3, 3] * [3, 2, 3, 3] 391s x0: [3, 1, 3, 3] 391s -0.383 0.019 0.420 391s 391s x1: [3, 2, 3, 3] 391s -0.729 -0.506 0.260 391s -0.423 -0.448 0.260 391s 391s y: [1, 2, 3, 3] 391s 0.380 391s 0.263 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.609 -0.567 0.119 391s 391s x1: [3, 1, 1, 1] 391s 0.067 0.775 0.414 391s 391s y: [1, 1, 1, 1] 391s -0.431 391s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 2, 3, 1] 391s 0.929 391s -0.812 391s 391s x1: [1, 3, 3, 1] 391s 0.328 391s 0.708 391s 0.919 391s 391s y: [2, 3, 3, 1] 391s 0.305 -0.266 391s 0.658 -0.574 391s 0.854 -0.746 391s testing: mul_mat, [1, 1, 3, 3] = [2, 1, 3, 3] * [2, 1, 3, 3] 391s x0: [2, 1, 3, 3] 391s -0.829 0.737 391s 391s x1: [2, 1, 3, 3] 391s -0.871 -0.001 391s 391s y: [1, 1, 3, 3] 391s 0.720 391s test-mul-mat0: iter:91/500 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.059 391s -0.885 391s 391s x1: [1, 2, 1, 1] 391s 0.987 391s -0.109 391s 391s y: [2, 2, 1, 1] 391s -0.058 -0.873 391s 0.006 0.097 391s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.252 391s 391s x1: [1, 2, 1, 1] 391s 0.768 391s 0.603 391s 391s y: [1, 2, 1, 1] 391s -0.194 391s -0.152 391s testing: mul_mat, [2, 1, 1, 2] = [1, 2, 1, 2] * [1, 1, 1, 2] 391s x0: [1, 2, 1, 2] 391s -0.640 391s -0.572 391s 391s x1: [1, 1, 1, 2] 391s 0.774 391s 391s y: [2, 1, 1, 2] 391s -0.495 -0.443 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.425 391s 0.126 391s 0.744 391s 391s x1: [1, 2, 1, 1] 391s -0.952 391s 0.625 391s 391s y: [3, 2, 1, 1] 391s 0.404 -0.120 -0.708 391s -0.265 0.079 0.465 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.459 -0.655 -0.145 391s 0.609 0.914 -0.668 391s 0.927 -0.783 0.702 391s 0.326 0.933 -0.543 391s 391s x1: [3, 1, 1, 1] 391s -0.899 -0.836 0.333 391s 391s y: [4, 1, 1, 1] 391s 0.912 -1.534 0.056 -1.255 391s testing: mul_mat, [1, 3, 1, 2] = [4, 1, 1, 2] * [4, 3, 1, 2] 391s x0: [4, 1, 1, 2] 391s -0.538 -0.324 -0.105 -0.206 391s 391s x1: [4, 3, 1, 2] 391s 0.591 -0.551 -0.769 -0.412 391s -0.759 -0.409 0.016 0.464 391s 0.467 0.064 0.088 0.712 391s 391s y: [1, 3, 1, 2] 391s 0.026 391s 0.443 391s -0.427 391s test-mul-mat0: iter:92/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.620 0.353 -0.308 391s 0.835 -0.191 -0.568 391s 391s x1: [3, 1, 1, 1] 391s 0.989 -0.468 0.641 391s 391s y: [2, 1, 1, 1] 391s 0.251 0.551 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.261 -0.201 0.839 391s 391s x1: [3, 2, 1, 1] 391s -0.788 0.599 -0.368 391s 0.131 0.997 -0.242 391s 391s y: [1, 2, 1, 1] 391s -0.224 391s -0.438 391s testing: mul_mat, [1, 1, 1, 4] = [3, 1, 1, 4] * [3, 1, 1, 4] 391s x0: [3, 1, 1, 4] 391s -0.554 0.036 0.841 391s 391s x1: [3, 1, 1, 4] 391s 0.415 0.694 -0.366 391s 391s y: [1, 1, 1, 4] 391s -0.514 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.651 -0.131 -0.228 391s -0.765 -0.849 0.718 391s -0.642 0.089 0.881 391s -0.957 0.888 -0.540 391s 391s x1: [3, 1, 1, 1] 391s -0.797 0.550 -0.604 391s 391s y: [4, 1, 1, 1] 391s 0.584 -0.291 0.028 1.577 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.578 0.502ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 0.086 -0.391 391s 0.106 -0.273 0.520 -0.506 391s 0.785 0.477 -0.394 -0.991 391s 0.128 0.737 0.237 -0.673 391s 391s x1: [4, 3, 1, 1] 391s -0.478 0.863 -0.319 -0.032 391s -0.101 -0.478 -0.233 -0.374 391s -0.802 0.375 -0.214 -0.914 391s 391s y: [4, 3, 1, 1] 391s 0.142 -0.435 0.195 0.521 391s -0.172 0.187 0.155 -0.169 391s 0.064 0.165 0.540 0.738 391s testing: mul_mat, [4, 4, 1, 4] = [4, 4, 1, 4] * [4, 4, 1, 4] 391s x0: [4, 4, 1, 4] 391s 0.528 0.742 -0.398 -0.362 391s -0.934 -0.795 0.050 0.359 391s 0.767 0.589 -0.208 0.800 391s -0.322 0.288 -0.462 0.481 391s 391s x1: [4, 4, 1, 4] 391s -0.625 -0.531 -0.151 -0.762 391s 0.150 0.817 0.136 0.673 391s -0.416 0.763 0.871 0.959 391s -0.451 0.957 0.540 -0.873 391s 391s y: [4, 4, 1, 4] 391s -0.388 0.725 -1.371 -0.248 391s 0.388 -0.542 1.107 0.448 391s -0.349 0.170 0.716 0.413 391s 0.572 -0.626 -0.592 -0.248 391s test-mul-mat0: iter:93/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.034 -0.547 391s 0.353 -0.427 391s 391s x1: [2, 4, 1, 1] 391s 0.189 -0.417 391s 0.213 0.552 391s -0.863 0.689 391s -0.447 0.050 391s 391s y: [2, 4, 1, 1] 391s 0.235 0.245 391s -0.294 -0.161 391s -0.406 -0.599 391s -0.043 -0.179 391s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 3, 4, 1] 391s -0.538 0.280 391s 0.972 0.676 391s -0.168 -0.891 391s 391s x1: [2, 2, 4, 1] 391s 0.779 -0.692 391s -0.005 0.366 391s 391s y: [3, 2, 4, 1] 391s -0.612 0.290 0.486 391s 0.105 0.243 -0.325 391s testing: mul_mat, [4, 3, 4, 2] = [2, 4, 4, 2] * [2, 3, 4, 2] 391s x0: [2, 4, 4, 2] 391s -0.949 0.553 391s -0.902 0.724 391s 0.902 0.071 391s -0.846 -0.104 391s 391s x1: [2, 3, 4, 2] 391s -0.529 -0.339 391s 0.990 -0.960 391s 0.847 -0.919 391s 391s y: [4, 3, 4, 2] 391s 0.314 0.231 -0.501 0.482 391s -1.470 -1.588 0.825 -0.737 391s -1.312 -1.430 0.699 -0.621 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.309 -0.261 391s -0.080 -0.705 391s 0.993 0.408 391s -0.092 -0.330 391s 391s x1: [2, 4, 1, 1] 391s 0.067 -0.440 391s 0.778 0.466 391s 0.743 -0.530 391s -0.114 0.224 391s 391s y: [4, 4, 1, 1] 391s 0.094 0.305 -0.113 0.139 391s -0.362 -0.391 0.963 -0.226 391s -0.092 0.314 0.522 0.106 391s -0.023 -0.149 -0.022 -0.063 391s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 3, 4, 1] 391s 0.527 0.978 -0.060 0.640 391s -0.096 -0.243 -0.164 -0.554 391s 0.473 0.191 -0.003 -0.216 391s 391s x1: [4, 2, 4, 1] 391s 0.395 -0.626 -0.940 -0.118 391s 0.972 -0.548 0.853 0.718 391s 391s y: [3, 2, 4, 1] 391s -0.423 0.333 0.095 391s 0.384 -0.498 0.197 391s testing: mul_mat, [4, 4, 4, 2] = [3, 4, 4, 2] * [3, 4, 4, 2] 391s x0: [3, 4, 4, 2] 391s -0.002 0.301 0.278 391s 0.816 0.071 0.830 391s -0.814 -0.864 -0.927 391s -0.449 -0.014 0.686 391s 391s x1: [3, 4, 4, 2] 391s 0.097 -0.118 0.461 391s 0.821 -0.546 0.183 391s -0.123 0.744 0.703 391s -0.439 -0.840 0.689 391s 391s y: [4, 4, 4, 2] 391s 0.092 0.452 -0.403 0.274 391s -0.115 0.783 -0.366 -0.235 391s 0.419 0.536 -1.194 0.527 391s -0.060 0.154 0.445 0.681 391s test-mul-mat0: iter:94/500 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.655 -0.917 391s -0.411 0.628 391s 391s x1: [2, 3, 1, 1] 391s -0.104 0.125 391s -0.413 0.089 391s -0.722 0.317 391s 391s y: [2, 3, 1, 1] 391s -0.046 0.121 391s 0.188 0.226 391s 0.182 0.495 391s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.728 0.800 391s 391s x1: [2, 2, 2, 1] 391s 0.004 -0.989 391s -0.808 0.564 391s 391s y: [1, 2, 2, 1] 391s -0.788 391s -0.137 391s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.856 0.467 391s 391s x1: [2, 1, 2, 1] 391s 0.378 0.297 391s 391s y: [1, 1, 2, 1] 391s 0.462 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.372 0.603 391s -0.679 0.454 391s 391s x1: [2, 1, 1, 1] 391s -0.945 0.885 391s 391s y: [2, 1, 1, 1] 391s 0.183 1.043 391s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 4, 2, 1] 391s 0.812 0.484 391s -0.679 0.564 391s -0.710 -0.010 391s 0.612 -0.219 391s 391s x1: [2, 2, 2, 1] 391s -0.089 0.231 391s -0.542 -0.078 391s 391s y: [4, 2, 2, 1] 391s 0.039 0.191 0.061 -0.105 391s -0.478 0.324 0.386 -0.315 391s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.772 0.792 -0.129 0.806 391s -0.084 0.906 -0.544 -0.528 391s -0.099 0.682 -0.637 0.839 391s 391s x1: [4, 2, 2, 1] 391s 0.362 0.174 0.443 0.683 391s -0.223 -0.103 -0.406 -0.992 391s 391s y: [3, 2, 2, 1] 391s 0.352 -0.475 0.374 391s -0.657 0.669 -0.622 391s test-mul-mat0: iter:95/500 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.069 -0.430 -0.750 391s 0.542 -0.591 0.971 391s -0.915 0.611 -0.134 391s -0.928 -0.504 -0.463 391s 391s x1: [3, 3, 1, 1] 391s 0.553 0.791 0.624 391s 0.781 -0.293 -0.475 391s 0.573 -0.386 -0.793 391s 391s y: [4, 3, 1, 1] 391s -0.770 0.438 -0.106 -1.200 391s 0.536 0.134 -0.829 -0.357 391s 0.801 -0.231 -0.653 0.030 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.957 0.684 0.715 391s -0.429 0.891 -0.842 391s 0.640 -0.539 -0.592 391s 391s x1: [3, 4, 1, 1] 391s -0.044 0.975 -0.783 391s 0.194 -0.029 0.328 391s -0.191 0.458 0.535 391s -0.639 0.250 0.159 391s 391s y: [3, 4, 1, 1] 391s 0.066 1.547 -0.090 391s 0.401 -0.386 -0.054 391s 0.513 0.040 -0.686 391s -0.327 0.363 -0.637 391s testing: mul_mat, [3, 3, 1, 4] = [3, 3, 1, 4] * [3, 3, 1, 4] 391s x0: [3, 3, 1, 4] 391s -0.414 -0.613 -0.512 391s -0.119 0.339 -0.513 391s -0.630 0.263 0.814 391s 391s x1: [3, 3, 1, 4] 391s 0.181 -0.130 -0.621 391s 0.267 -0.519 0.245 391s 0.339 -0.023 0.782 391s 391s y: [3, 3, 1, 4] 391s 0.322 0.253 -0.654 391s 0.082 -0.333 -0.105 391s -0.526 -0.449 0.417 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.009 0.141 -0.808 391s 0.170 0.654 0.143 391s 391s x1: [3, 3, 1, 1] 391s 0.545 -0.117 0.388 391s 0.360 0.552 -0.968 391s -0.963 -0.121 0.307 391s 391s y: [2, 3, 1, 1] 391s -0.326 0.072 391s 0.863 0.283 391s -0.273 -0.199 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.141 0.809 391s -0.407 -0.509 391s 391s x1: [2, 3, 1, 1] 391s -0.027 -0.988 391s 0.204 0.375 391s 0.812 -0.806 391s 391s y: [2, 3, 1, 1] 391s -0.795 0.514 391s 0.274 -0.274 391s -0.767 0.080 391s testing: mul_mat, [4, 2, 1, 4] = [2, 4, 1, 4] * [2, 2, 1, 4] 391s x0: [2, 4, 1, 4] 391s -0.019 0.607 391s 0.595 -0.966 391s -0.170 0.329 391s 0.954 0.419 391s 391s x1: [2, 2, 1, 4] 391s 0.380 -0.133 391s 0.004 0.506 391s 391s y: [4, 2, 1, 4] 391s -0.088 0.354 -0.108 0.307 391s 0.307 -0.486 0.166 0.216 391s test-mul-mat0: iter:96/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.583 0.199 391s 391s x1: [2, 4, 1, 1] 391s -0.667 0.011 391s 0.784 -0.439 391s 0.579 -0.939 391s 0.382 0.956 391s 391s y: [1, 4, 1, 1] 391s -0.387 391s 0.370 391s 0.151 391s 0.413 391s testing: mul_mat, [2, 1, 2, 1] = [2, 2, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.716 0.364 391s 0.417 0.522 391s 391s x1: [2, 1, 2, 1] 391s -0.397 0.290 391s 391s y: [2, 1, 2, 1] 391s 0.390 -0.014 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.247 0.968 391s 0.532 0.308 391s 391s x1: [2, 2, 2, 1] 391s -0.661 -0.225 391s 0.259 0.098 391s 391s y: [2, 2, 2, 1] 391s -0.381 -0.421 391s 0.158 0.168 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.887 -0.631 391s 0.825 0.530 391s 391s x1: [2, 2, 1, 1] 391s -0.314 -0.259 391s -0.777 0.542 391s 391s y: [2, 2, 1, 1] 391s 0.442 -0.396 391s 0.347 -0.354 391s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.499 -0.117 391s 391s x1: [2, 2, 2, 1] 391s 0.347 0.153 391s -0.293 -0.348 391s 391s y: [1, 2, 2, 1] 391s -0.191 391s 0.187 391s testing: mul_mat, [2, 2, 2, 1] = [1, 2, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.225 391s 0.855 391s 391s x1: [1, 2, 2, 1] 391s -0.164 391s -0.147 391s 391s y: [2, 2, 2, 1] 391s -0.037 -0.140 391s -0.033 -0.126 391s test-mul-mat0: iter:97/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.255 391s 391s x1: [1, 4, 1, 1] 391s -0.955 391s -0.505 391s 0.908 391s -0.586 391s 391s y: [1, 4, 1, 1] 391s -0.244 391s -0.129 391s 0.232 391s -0.150 391s testing: mul_mat, [3, 1, 3, 1] = [1, 3, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 3, 3, 1] 391s 0.778 391s 0.335 391s -0.808 391s 391s x1: [1, 1, 3, 1] 391s 0.566 391s 391s y: [3, 1, 3, 1] 391s 0.440 0.190 -0.458 391s testing: mul_mat, [3, 3, 3, 2] = [1, 3, 3, 2] * [1, 3, 3, 2] 391s x0: [1, 3, 3, 2] 391s -0.364 391s 0.829 391s -0.845 391s 391s x1: [1, 3, 3, 2] 391s -0.619 391s 0.620 391s -0.721 391s 391s y: [3, 3, 3, 2] 391s 0.226 -0.513 0.523 391s -0.226 0.514 -0.524 391s 0.263 -0.598 0.609 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.598 391s 0.178 391s 0.432 391s 391s x1: [1, 3, 1, 1] 391s 0.625 391s -0.787 391s -0.580 391s 391s y: [3, 3, 1, 1] 391s 0.374 0.111 0.270 391s -0.471 -0.140 -0.340 391s -0.346 -0.103 -0.250 391s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.243 -0.234 0.601 391s -0.821 0.757 -0.372 391s 391s x1: [3, 1, 3, 1] 391s 0.477 0.843 0.772 391s 391s y: [2, 1, 3, 1] 391s 0.150 -0.041 391s testing: mul_mat, [2, 3, 3, 2] = [2, 2, 3, 2] * [2, 3, 3, 2] 391s x0: [2, 2, 3, 2] 391s 0.317 0.057 391s -0.115 -0.922 391s 391s x1: [2, 3, 3, 2] 391s 0.884 0.890 391s -0.781 0.361 391s 0.733 0.991 391s 391s y: [2, 3, 3, 2] 391s 0.331 -0.923 391s -ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 0.227 -0.243 391s 0.289 -0.998 391s test-mul-mat0: iter:98/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.263 391s 0.976 391s 391s x1: [1, 1, 1, 1] 391s -0.341 391s 391s y: [2, 1, 1, 1] 391s 0.090 -0.333 391s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 4, 2, 1] 391s 0.323 391s 0.034 391s -0.411 391s -0.451 391s 391s x1: [1, 2, 2, 1] 391s 0.236 391s 0.794 391s 391s y: [4, 2, 2, 1] 391s 0.076 0.008 -0.097 -0.106 391s 0.256 0.027 -0.326 -0.358 391s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 3, 2, 1] 391s 0.189 391s -0.276 391s -0.003 391s 391s x1: [1, 4, 2, 1] 391s 0.840 391s 0.627 391s -0.505 391s -0.666 391s 391s y: [3, 4, 2, 1] 391s 0.159 -0.232 -0.003 391s 0.119 -0.173 -0.002 391s -0.096 0.139 0.002 391s -0.126 0.184 0.002 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.398 391s 0.685 391s -0.048 391s -0.046 391s 391s x1: [1, 3, 1, 1] 391s -0.496 391s -0.192 391s 0.631 391s 391s y: [4, 3, 1, 1] 391s 0.197 -0.339 0.024 0.023 391s 0.076 -0.131 0.009 0.009 391s -0.251 0.432 -0.031 -0.029 391s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 2, 2, 1] 391s 0.543 0.824 -0.025 -0.677 391s -0.064 0.549 0.703 0.887 391s 391s x1: [4, 1, 2, 1] 391s 0.007 0.986 0.543 0.556 391s 391s y: [2, 1, 2, 1] 391s 0.426 1.415 391s testing: mul_mat, [4, 4, 2, 1] = [2, 4, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 4, 2, 1] 391s 0.991 0.750 391s 0.426 -0.086 391s 0.028 -0.413 391s 0.694 -0.203 391s 391s x1: [2, 4, 2, 1] 391s -0.830 0.291 391s -0.471 0.854 391s -0.758 0.483 391s -0.138 -0.772 391s 391s y: [4, 4, 2, 1] 391s -0.604 -0.379 -0.143 -0.635 391s 0.174 -0.274 -0.366 -0.501 391s -0.389 -0.364 -0.220 -0.624 391s -0.716 0.007 0.315 0.061 391s test-mul-mat0: iter:99/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.121 391s 0.640 391s 0.621 391s 0.555 391s 391s x1: [1, 1, 1, 1] 391s 0.222 391s 391s y: [4, 1, 1, 1] 391s -0.027 0.142 0.138 0.123 391s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 1, 2, 1] 391s 0.027 391s 391s x1: [1, 4, 2, 1] 391s 0.974 391s -0.263 391s 0.721 391s 0.852 391s 391s y: [1, 4, 2, 1] 391s 0.027 391s -0.007 391s 0.020 391s 0.023 391s testing: mul_mat, [3, 1, 2, 4] = [1, 3, 2, 4] * [1, 1, 2, 4] 391s x0: [1, 3, 2, 4] 391s 0.284 391s 0.952 391s 0.730 391s 391s x1: [1, 1, 2, 4] 391s 0.993 391s 391s y: [3, 1, 2, 4] 391s 0.282 0.945 0.725 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.505 391s 0.869 391s 0.405 391s -0.212 391s 391s x1: [1, 3, 1, 1] 391s 0.170 391s -0.305 391s -0.713 391s 391s y: [4, 3, 1, 1] 391s 0.086 0.148 0.069 -0.036 391s -0.154 -0.265 -0.124 0.064 391s -0.360 -0.619 -0.289 0.151 391s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.454 -0.494 -0.823 0.253 391s 0.246 0.664 0.359 -0.918 391s 0.605 0.698 0.608 -0.750 391s 391s x1: [4, 1, 2, 1] 391s 0.820 0.136 -0.211 0.625 391s 391s y: [3, 1, 2, 1] 391s -0.107 -0.357 -0.006 391s testing: mul_mat, [1, 4, 2, 4] = [3, 1, 2, 4] * [3, 4, 2, 4] 391s x0: [3, 1, 2, 4] 391s 0.003 -0.138 0.991 391s 391s x1: [3, 4, 2, 4] 391s -0.900 -0.769 -0.915 391s 0.835 -0.460 -0.235 391s -0.863 0.086 -0.989 391s 0.742 0.591 0.675 391s 391s y: [1, 4, 2, 4] 391s -0.804 391s -0.167 391s -0.995 391s 0.589 391s test-mul-mat0: iter:100/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.241 -0.852 0.192 391s 391s x1: [3, 1, 1, 1] 391s -0.470 0.897 -0.843 391s 391s y: [1, 1, 1, 1] 391s -0.814 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.048 -0.746 -0.447 391s 0.714 0.220 0.641 391s 391s x1: [3, 1, 1, 1] 391s 0.272 0.929 0.008 391s 391s y: [2, 1, 1, 1] 391s -0.710 0.405 391s testing: mul_mat, [4, 2, 1, 3] = [3, 4, 1, 3] * [3, 2, 1, 3] 391s x0: [3, 4, 1, 3] 391s -0.356 -0.589 0.746 391s -0.427 0.419 -0.252 391s 0.525 0.673 0.301 391s 0.240 -0.107 -0.058 391s 391s x1: [3, 2, 1, 3] 391s -0.986 0.545 0.732 391s -0.635 0.226 -0.046 391s 391s y: [4, 2, 1, 3] 391s 0.576 0.465 0.069 -0.337 391s 0.058 0.377 -0.195 -0.174 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.973 0.088 -0.327 391s 0.882 0.321 0.628 391s 0.655 0.995 -0.324 391s -0.924 -0.366 0.244 391s 391s x1: [3, 4, 1, 1] 391s -0.399 -0.216 -0.926 391s 0.902 -0.976 -0.033 391s -0.156 0.277 0.405 391s 0.517 -0.105 0.069 391s 391s y: [4, 4, 1, 1] 391s -0.104 -1.003 -0.176 0.222 391s 0.803 0.462 -0.369 -0.485 391s -0.260 0.206 0.042 0.142 391s 0.471 0.466 0.212 -0.423 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.294 -0.547 -0.857 0.575 391s 0.073 0.602 0.597 -0.975 391s 391s x1: [4, 3, 1, 1] 391s 0.735 0.362 -0.774 0.588 391s -0.379 -0.373 -0.810 0.405 391s -0.299 -0.908 0.429 0.668 391s 391s y: [2, 3, 1, 1] 391s 0.589 -0.764 391s 1.243 -1.131 391s 0.601 -0.963 391s testing: mul_mat, [2, 4, 1, 3] = [2, 2, 1, 3] * [2, 4, 1, 3] 391s x0: [2, 2, 1, 3] 391s -0.732 -0.074 391s 0.549 0.865 391s 391s x1: [2, 4, 1, 3] 391s 0.252 0.651 391s 0.113 -0.427 391s 0.646 0.747 391s 0.246 0.274 391s 391s y: [2, 4, 1, 3] 391s -0.232 0.701 391s -0.051 -0.308 391s -0.528 1.001 391s -0.200 0.372 391s test-mul-mat0: iter:101/500 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.763 0.525 391s -0.033 0.478 391s 0.235 -0.217 391s 391s x1: [2, 3, 1, 1] 391s -0.296 -0.980 391s -0.860 0.301 391s -0.574 0.482 391s 391s y: [3, 3, 1, 1] 391s -0.288 -0.459 0.143 391s 0.814 0.172 -0.268 391s 0.691 0.249 -0.240 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.896 -0.967 391s -0.127 0.960 391s 0.723 0.294 391s 0.771 -0.341 391s 391s x1: [2, 3, 1, 1] 391s 0.425 -0.981 391s 0.051 -0.026 391s -0.055 -0.084 391s 391s y: [4, 3, 1, 1] 391s 1.329 -0.996 0.019 0.663 391s 0.071 -0.032 0.029 0.048 391s 0.032 -0.073 -0.064 -0.014 391s testing: mul_mat, [2, 4, 1, 2] = [2, 2, 1, 2] * [2, 4, 1, 2] 391s x0: [2, 2, 1, 2] 391s -0.514 0.882 391s -0.434 0.431 391s 391s x1: [2, 4, 1, 2] 391s -0.012 -0.994 391s -0.916 0.692 391s -0.974 -0.776 391s -0.007 -0.548 391s 391s y: [2, 4, 1, 2] 391s -0.871 -0.423 391s 1.082 0.695 391s -0.183 0.088 391s -0.480 -0.233 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.214 0.064 391s 0.820 0.298 391s 391s x1: [2, 2, 1, 1] 391s 0.538 0.287 391s -0.445 -0.169 391s 391s y: [2, 2, 1, 1] 391s 0.133 0.526 391s -0.106 -0.415 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.228 -0.769 391s 0.586 -0.540 391s 391s x1: [2, 2, 1, 1] 391s 0.512 0.090 391s 0.522 -0.495 391s 391s y: [2, 2, 1, 1] 391s 0.048 0.251 391s 0.500 0.573 391s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 1, 1, 2] 391s 0.146 0.221 391s 391s x1: [2, 2, 1, 2] 391s 0.634 -0.602 391s -0.074 -0.535 391s 391s y: [1, 2, 1, 2] 391s -0.040 391s -0.129 391s test-mul-mat0: iter:102/500 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.112 -0.171 -0.408 0.459 391s 391s x1: [4, 4, 1, 1] 391s -0.485 0.747 0.410 0.579 391s 0.044 -0.079 -0.331 -0.434 391s 0.426 -0.790 0.794 0.012 391s -0.559 -0.746 -0.354 -0.161 391s 391s y: [1, 4, 1, 1] 391s -0.083 391s -0.046 391s -0.136 391s 0.135 391s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 1, 4, 1] 391s -0.527 -0.287 -0.965 -0.733 391s 391s x1: [4, 1, 4, 1] 391s 0.386 0.435 -0.395 -0.393 391s 391s y: [1, 1, 4, 1] 391s 0.340 391s testing: mul_mat, [1, 1, 4, 3] = [4, 1, 4, 3] * [4, 1, 4, 3] 391s x0: [4, 1, 4, 3] 391s 0.639 -0.130 0.211 -0.085 391s 391s x1: [4, 1, 4, 3] 391s 0.547 -0.862 0.332 -0.870 391s 391s y: [1, 1, 4, 3] 391s 0.605 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.931 0.731 0.128 -0.208 391s -0.741 -0.094 -0.149 -0.363 391s -0.566 0.824 -0.817 0.981 391s -0.564 0.522 -0.439 -0.375 391s 391s x1: [4, 1, 1, 1] 391s 0.681 -0.167 0.821 -0.720 391s 391s y: [4, 1, 1, 1] 391s 0.767 -0.351 -1.900 -0.562 391s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.481 0.055 0.944 -0.716 391s 391s x1: [4, 4, 4, 1] 391s 0.287 -0.691 0.618 0.480 391s 0.996 -0.234 -0.890 0.633 391s -0.603 -0.799 0.315 0.230 391s -0.978 0.595 0.956 0.953 391s 391s y: [1, 4, 4, 1] 391s 0.339 391s -0.827 391s -0.202 391s -0.217 391s testing: mul_mat, [4, 4, 4, 3] = [1, 4, 4, 3] * [1, 4, 4, 3] 391s x0: [1, 4, 4, 3] 391s 0.027 391s 0.814 391s -0.135 391s 0.357 391s 391s x1: [1, 4, 4, 3] 391s -0.090 391s 0.122 391s -0.761 391s 0.268 391s 391s y: [4, 4, 4, 3] 391s -0.002 -0.073 0.012 -0.032 391s 0.003 0.099 -0.017 0.044 391s -0.020 -0.620 0.103 -0.272 391s 0.007 0.218 -0.036 0.096 391s test-mul-mat0: iter:103/500 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.561 391s 0.479 391s 0.673 391s 0.586 391s 391s x1: [1, 3, 1, 1] 391s 0.505 391s 0.520 391s -0.633 391s 391s y: [4, 3, 1, 1] 391s 0.283 0.242 0.340 0.296 391s 0.292 0.249 0.350 0.305 391s -0.355 -0.303 -0.426 -0.371 391s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.223 391s 0.389 391s 0.803 391s -0.664 391s 391s x1: [1, 4, 2, 1] 391s 0.871 391s -0.518 391s -0.362 391s -0.922 391s 391s y: [4, 4, 2, 1] 391s -0.195 0.339 0.700 -0.578 391s 0.116 -0.201 -0.416 0.344 391s 0.081 -0.141 -0.291 0.240 391s 0.206 -0.359 -0.741 0.612 391s testing: mul_mat, [1, 4, 2, 2] = [1, 1, 2, 2] * [1, 4, 2, 2] 391s x0: [1, 1, 2, 2] 391s -0.741 391s 391s x1: [1, 4, 2, 2] 391s -0.076 391s -0.398 391s -0.495 391s -0.750 391s 391s y: [1, 4, 2, 2] 391s 0.056 391s 0.295 391s 0.367 391s 0.556 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.847 391s 391s x1: [1, 1, 1, 1] 391s -0.813 391s 391s y: [1, 1, 1, 1] 391s -0.689 391s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.620 391s 391s x1: [1, 1, 2, 1] 391s 0.576 391s 391s y: [1, 1, 2, 1] 391s -0.357 391s testing: mul_mat, [4, 1, 2, 2] = [1, 4, 2, 2] * [1, 1, 2, 2] 391s x0: [1, 4, 2, 2] 391s -0.226 391s -0.143 391s 0.117 391s -0.471 391s 391s x1: [1, 1, 2, 2] 391s -0.052 391s 391s y: [4, 1, 2, 2] 391s 0.012 0.007 -0.006 0.025 391s test-mul-mat0: iter:104/500 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.209 -0.085 -0.478 0.204 391s 391s x1: [4, 4, 1, 1] 391s 0.618 -0.982 -0.643 0.802 391s -0.602 0.958 -0.250 -0.454 391s -0.518 -0.700 0.198 0.256 391s 0.157 -0.685 0.784 -0.159 391s 391s y: [1, 4, 1, 1] 391s 0.684 391s -0.181 391s -0.091 391s -0.316 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.494 0.945 0.434 -0.052 391s 0.382 0.053 -0.034 0.740 391s -0.145 0.364 0.698 0.605 391s 391s x1: [4, 1, 1, 1] 391s -0.037 -0.037 -0.749 0.624 391s 391s y: [3, 1, 1, 1] 391s -0.374 0.471 -0.153 391s testing: mul_mat, [3, 3, 1, 4] = [4, 3, 1, 4] * [4, 3, 1, 4] 391s x0: [4, 3, 1, 4] 391s -0.989 0.342 -0.999 -0.721 391s -0.320 -0.268 0.600 -0.015 391s 0.203 0.776 -0.347 0.371 391s 391s x1: [4, 3, 1, 4] 391s 0.910 -0.820 0.905 0.108 391s 0.436 0.062 0.423 0.220 391s 0.903 0.686 -0.571 -0.182 391s 391s y: [3, 3, 1, 4] 391s -2.163 0.470 -0.726 391s -0.991 0.095 0.071 391s 0.043 -0.813 0.846 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.439 -0.433 0.304 0.244 391s 0.480 0.777 0.673 0.346 391s 0.248 0.228 -0.807 0.811 391s 391s x1: [4, 3, 1, 1] 391s -0.466 -0.541 0.799 0.042 391s 0.098 -0.375 -0.018 0.203 391s 0.456 0.989 -0.444 0.463 391s 391s y: [3, 3, 1, 1] 391s 0.692 -0.091 -0.850 391s 0.164 -0.186 0.118 391s -0.650 0.849 1.072 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.807 0.775 0.329 391s -0.688 -0.918 0.492 391s -0.866 -0.075 0.031 391s -0.637 0.255 -0.443 391s 391s x1: [3, 4, 1, 1] 391s -0.444 -0.424 0.661 391s 0.587 0.047 -0.569 391s -0.879 0.506 -0.769 391s 0.163 -0.396 -0.144 391s 391s y: [4, 4, 1, 1] 391s -0.469 1.020 0.437 -0.118 391s 0.322 -0.727 -0.529 -0.110 391s -0.570 -0.239 0.698 1.029 391s -0.223 0.181 -0.116 -0.141 391s testing: mul_mat, [1, 3, 1, 4] = [4, 1, 1, 4] * [4, 3, 1, 4] 391s x0: [4, 1, 1, 4] 391s 0.372 0.726 -0.424 0.118 391s 391s x1: [4, 3, 1, 4] 391s -0.204 -0.295 -0.250 -0.959 391s -0.949 -0.439 -0.404 -0.373 391s -0.778 -0.817 0.674 -0.347 391s 391s y: [1, 3, 1, 4] 391s -0.298 391s -0.545 391s -1.209 391s test-mul-mat0: iter:105/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.392 -0.703 391s 0.246 0.151 391s -0.569 0.771 391s 391s x1: [2, 2, 1, 1] 391s -0.223 -0.441 391s 0.099 0.947 391s 391s y: [3, 2, 1, 1] 391s 0.223 -0.121 -0.213 391s -0.627 0.167 0.673 391s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.847 -0.699 391s -0.329 -0.054 391s 391s x1: [2, 3, 3, 1] 391s -0.477 -0.843 391s -0.653 0.641 391s 0.838 -0.250 391s 391s y: [2, 3, 3, 1] 391s 0.185 0.203 391s -1.001 0.181 391s 0.884 -0.263 391s testing: mul_mat, [3, 2, 3, 2] = [2, 3, 3, 2] * [2, 2, 3, 2] 391s x0: [2, 3, 3, 2] 391s 0.992 -0.833 391s -0.463 0.480 391s 0.142 -0.315 391s 391s x1: [2, 2, 3, 2] 391s -0.989 0.133 391s -0.505 0.652 391s 391s y: [3, 2, 3, 2] 391s -1.091 0.521 -0.183 391s -1.045 0.547 -0.278 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.850 -0.712 391s -0.730 -0.532 391s 391s x1: [2, 3, 1, 1] 391s -0.216 0.622 391s -0.588 0.963 391s -0.234 0.657 391s 391s y: [2, 3, 1, 1] 391s -0.259 -0.173 391s -0.186 -0.083 391s -0.269 -0.178 391s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.885 0.584 391s 391s x1: [2, 2, 3, 1] 391s 0.581 0.496 391s -0.766 0.018 391s 391s y: [1, 2, 3, 1] 391s -0.225 391s 0.688 391s testing: mul_mat, [2, 2, 3, 2] = [1, 2, 3, 2] * [1, 2, 3, 2] 391s x0: [1, 2, 3, 2] 391s -0.399 391s -0.001 391s 391s x1: [1, 2, 3, 2] 391s 0.246 391s -0.532 391s 391s y: [2, 2, 3, 2] 391s -0.098 -0.000 391s 0.213 0.001 391s test-mul-mat0: iter:106/500 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.126 -0.990 0.250 0.979 391s 391s x1: [4, 2, 1, 1] 391s 0.158 -0.814 -0.306 0.626 391s 0.567 -0.649 -0.944 0.714 391s 391s y: [1, 2, 1, 1] 391s 1.322 391s 1.034 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.433 0.021 -0.902 -0.630 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s 391s x1: [4, 1, 1, 1] 391s -0.225 -0.585 -0.420 0.774 391s 391s y: [1, 1, 1, 1] 391s -0.219 391s testing: mul_mat, [4, 1, 1, 3] = [4, 4, 1, 3] * [4, 1, 1, 3] 391s x0: [4, 4, 1, 3] 391s -0.104 0.426 0.094 -0.390 391s 0.091 0.968 -0.380 -0.659 391s 0.947 0.395 -0.243 -0.474 391s 0.169 0.356 0.960 -0.809 391s 391s x1: [4, 1, 1, 3] 391s -0.111 0.918 -0.406 -0.221 391s 391s y: [4, 1, 1, 3] 391s 0.451 1.178 0.461 0.097 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.497 0.853 0.451 0.265 391s 0.280 0.186 0.422 -0.622 391s 391s x1: [4, 4, 1, 1] 391s -0.615 -0.482 0.109 0.308 391s -0.623 0.196 0.164 -0.511 391s -0.574 0.401 -0.987 -0.015 391s -0.287 -0.425 -0.561 0.451 391s 391s y: [2, 4, 1, 1] 391s 0.026 -0.408 391s 0.416 0.248 391s 0.178 -0.493 391s -0.353 -0.677 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.265 -0.154 391s 391s x1: [2, 4, 1, 1] 391s 0.517 -0.487 391s -0.339 0.573 391s 0.951 -0.602 391s 0.957 -0.531 391s 391s y: [1, 4, 1, 1] 391s 0.212 391s -0.178 391s 0.344 391s 0.335 391s testing: mul_mat, [3, 2, 1, 3] = [1, 3, 1, 3] * [1, 2, 1, 3] 391s x0: [1, 3, 1, 3] 391s -0.459 391s 0.923 391s -0.447 391s 391s x1: [1, 2, 1, 3] 391s -0.297 391s -0.572 391s 391s y: [3, 2, 1, 3] 391s 0.136 -0.274 0.133 391s 0.262 -0.528 0.256 391s test-mul-mat0: iter:107/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.483 0.972 391s -0.839 -0.526 391s -0.559 0.668 391s 391s x1: [2, 1, 1, 1] 391s -0.055 -0.980 391s 391s y: [3, 1, 1, 1] 391s -0.925 0.562 -0.624 391s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.058 -0.035 391s 0.551 0.416 391s 0.406 0.220 391s 0.155 -0.307 391s 391s x1: [2, 3, 4, 1] 391s 0.739 0.287 391s -0.628 -0.833 391s -0.377 0.501 391s 391s y: [4, 3, 4, 1] 391s -0.053 0.527 0.363 0.026 391s 0.066 -0.693 -0.438 0.159 391s 0.004 0.000 -0.043 -0.212 391s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 3, 4, 1] 391s -0.495 -0.089 391s -0.683 0.202 391s 0.508 -0.592 391s 391s x1: [2, 4, 4, 1] 391s -0.502 0.280 391s 0.542 0.904 391s -0.501 -0.303 391s -0.403 -0.909 391s 391s y: [3, 4, 4, 1] 391s 0.224 0.399 -0.421 391s -0.349 -0.188 -0.259 391s 0.275 0.281 -0.075 391s 0.280 0.092 0.333 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.452 -0.037 391s 391s x1: [2, 3, 1, 1] 391s -0.637 0.985 391s 0.044 -0.099 391s -0.053 -0.948 391s 391s y: [1, 3, 1, 1] 391s -0.324 391s 0.024 391s 0.011 391s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 4, 4, 1] 391s -0.481 391s -0.823 391s 0.513 391s -0.229 391s 391s x1: [1, 2, 4, 1] 391s 0.035 391s -0.347 391s 391s y: [4, 2, 4, 1] 391s -0.017 -0.029 0.018 -0.008 391s 0.167 0.285 -0.178 0.079 391s testing: mul_mat, [3, 1, 4, 1] = [4, 3, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 3, 4, 1] 391s -0.566 -0.444 -0.823 -0.053 391s -0.081 -0.188 -0.736 -0.750 391s 0.785 -0.635 -0.502 -0.006 391s 391s x1: [4, 1, 4, 1] 391s 0.397 0.701 0.082 -0.885 391s 391s y: [3, 1, 4, 1] 391s -0.557 0.439 -0.170 391s test-mul-mat0: iter:108/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.420 -0.234 -0.415 391s 391s x1: [3, 1, 1, 1] 391s -0.610 0.247 -0.113 391s 391s y: [1, 1, 1, 1] 391s 0.246 391s testing: mul_mat, [2, 1, 2, 1] = [3, 2, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.498 0.837 -0.813 391s 0.520 -0.076 -0.445 391s 391s x1: [3, 1, 2, 1] 391s -0.166 -0.207 0.024 391s 391s y: [2, 1, 2, 1] 391s -0.275 -0.082 391s testing: mul_mat, [3, 2, 2, 1] = [3, 3, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.540 -0.444 0.763 391s 0.280 0.804 -0.471 391s -0.223 0.641 -0.284 391s 391s x1: [3, 2, 2, 1] 391s 0.259 -0.542 -0.876 391s 0.582 0.094 -0.486 391s 391s y: [3, 2, 2, 1] 391s -0.288 0.049 -0.156 391s -0.098 0.467 0.069 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.330 -0.248 0.663 391s 0.107 -0.130 -0.485 391s 391s x1: [3, 3, 1, 1] 391s -0.308 0.508 -0.886 391s -0.726 -0.398 -0.371 391s 0.102 -0.417 0.575 391s 391s y: [2, 3, 1, 1] 391s -0.815 0.330 391s -0.387 0.154 391s 0.518 -0.214 391s testing: mul_mat, [2, 3, 2, 1] = [2, 2, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.706 0.438 391s -0.084 0.802 391s 391s x1: [2, 3, 2, 1] 391s -0.850 0.467 391s 0.045 -0.072 391s 0.108 0.761 391s 391s y: [2, 3, 2, 1] 391s 0.804 0.446 391s -0.063 -0.062 391s 0.257 0.601 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.409 0.471 391s 0.338 0.636 391s 391s x1: [2, 2, 2, 1] 391s -0.186 -0.013 391s 0.235 0.324 391s 391s y: [2, 2, 2, 1] 391s 0.070 -0.071 391s 0.057 0.285 391s test-mul-mat0: iter:109/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.012 391s 0.593 391s 0.300 391s 391s x1: [1, 3, 1, 1] 391s 0.353 391s 0.550 391s 0.791 391s 391s y: [3, 3, 1, 1] 391s -0.004 0.210 0.106 391s -0.006 0.326 0.165 391s -0.009 0.469 0.238 391s testing: mul_mat, [2, 3, 2, 1] = [1, 2, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.301 391s -0.464 391s 391s x1: [1, 3, 2, 1] 391s -0.300 391s 0.997 391s -0.040 391s 391s y: [2, 3, 2, 1] 391s -0.090 0.139 391s 0.300 -0.463 391s -0.012 0.019 391s testing: mul_mat, [3, 2, 2, 4] = [1, 3, 2, 4] * [1, 2, 2, 4] 391s x0: [1, 3, 2, 4] 391s -0.077 391s 0.723 391s 0.888 391s 391s x1: [1, 2, 2, 4] 391s -0.460 391s 0.505 391s 391s y: [3, 2, 2, 4] 391s 0.035 -0.333 -0.408 391s -0.039 0.365 0.448 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.685 391s -0.968 391s 0.942 391s 0.238 391s 391s x1: [1, 3, 1, 1] 391s 0.036 391s -0.620 391s 0.902 391s 391s y: [4, 3, 1, 1] 391s -0.025 -0.035 0.034 0.009 391s 0.425 0.600 -0.584 -0.148 391s -0.618 -0.873 0.849 0.215 391s testing: mul_mat, [4, 1, 2, 1] = [4, 4, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 4, 2, 1] 391s -0.814 0.416 -0.206 0.408 391s 0.298 0.259 0.414 0.829 391s -0.135 0.002 -0.734 0.902 391s -0.319 0.546 -0.134 0.788 391s 391s x1: [4, 1, 2, 1] 391s 0.756 0.830 -0.138 0.476 391s 391s y: [4, 1, 2, 1] 391s -0.047 0.777 0.430 0.605 391s testing: mul_mat, [2, 4, 2, 4] = [4, 2, 2, 4] * [4, 4, 2, 4] 391s x0: [4, 2, 2, 4] 391s -0.556 -0.396 -0.664 0.449 391s 0.754 -0.437 -0.287 -0.574 391s 391s x1: [4, 4, 2, 4] 391s -0.278 0.042 -0.957 -0.862 391s -0.699 0.044 0.684 0.095 391s -0.542 0.950 0.961 0.867 391s 0.778 0.863 0.655 0.509 391s 391s y: [2, 4, 2, 4] 391s 0.387 0.542 391s -0.041 -0.797 391s -0.325 -1.596 391s -0.982 -0.270 391s test-mul-mat0: iter:110/500 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.792 0.563 391s -0.563 -0.376 391s -0.371 0.355 391s 391s x1: [2, 3, 1, 1] 391s 0.092 0.179 391s -0.886 -0.255 391s 0.979 0.153 391s 391s y: [3, 3, 1, 1] 391s 0.028 -0.119 0.029 391s 0.558 0.595 0.238 391s -0.690 -0.609 -0.309 391s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.856 -0.153 391s 0.765 -0.233 391s 0.476 0.120 391s -0.530 -0.322 391s 391s x1: [2, 3, 4, 1] 391s 0.703 0.202 391s 0.118 -0.296 391s 0.937 0.744 391s 391s y: [4, 3, 4, 1] 391s -0.633 0.491 0.359 -0.437 391s -0.056 0.159 0.021 0.032 391s -0.916 0.544 0.535 -0.736 391s testing: mul_mat, [4, 4, 4, 3] = [2, 4, 4, 3] * [2, 4, 4, 3] 391s x0: [2, 4, 4, 3] 391s 0.847 0.948 391s -0.692 0.272 391s -0.006 -0.222 391s 0.419 -0.605 391s 391s x1: [2, 4, 4, 3] 391s -0.258 0.496 391s -0.389 -0.782 391s -0.384 0.081 391s -0.104 -0.146 391s 391s y: [4, 4, 4, 3] 391s 0.252 0.314 -0.109 -0.408 391s -1.070 0.056 0.176 0.310 391s -0.248 0.288 -0.016 -0.210 391s -0.226 0.032 0.033 0.045 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.434 -0.633 391s 391s x1: [2, 4, 1, 1] 391s 0.185 0.121 391s -0.935 -0.439 391s -0.244 0.800 391s -0.654 0.065 391s 391s y: [1, 4, 1, 1] 391s 0.004 391s -0.129 391s -0.612 391s -0.326 391s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 3, 4, 1] 391s 0.975 391s 0.824 391s -0.654 391s 391s x1: [1, 2, 4, 1] 391s -0.342 391s -0.877 391s 391s y: [3, 2, 4, 1] 391s -0.334 -0.282 0.224 391s -0.855 -0.722 0.573 391s testing: mul_mat, [4, 1, 4, 3] = [3, 4, 4, 3] * [3, 1, 4, 3] 391s x0: [3, 4, 4, 3] 391s 0.420 -0.280 0.936 391s 0.106 -0.787 0.556 391s -0.464 0.090 -0.707 391s -0.501 -0.433 0.124 391s 391s x1: [3, 1, 4, 3] 391s -0.110 -0.761 0.774 391s 391s y: [4, 1, 4, 3] 391s 0.892 1.018 -0.565 0.480 391s test-mul-mat0: iter:111/500 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.093 -0.529 -0.654 -0.424 391s -0.914 -0.711 -0.504 -0.587 391s 391s x1: [4, 3, 1, 1] 391s -0.855 0.586 0.746 0.154 391s 0.296 -0.482 -0.519 0.243 391s -0.906 -0.513 -0.095 -0.734 391s 391s y: [2, 3, 1, 1] 391s -0.782 -0.101 391s 0.463 0.191 391s 0.729 1.672 391s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.520 -0.547 -0.533 -0.231 391s -0.051 -0.119 0.116 0.343 391s 0.364 0.236 -0.779 -0.390 391s 0.462 0.269 -0.449 -0.252 391s 391s x1: [4, 2, 3, 1] 391s -0.653 -0.606 -0.517 -0.880 391s -0.122 0.246 -0.774 0.049 391s 391s y: [4, 2, 3, 1] 391s 1.149 -0.256 0.365 -0.012 391s 0.330 -0.096 0.598 0.346 391s testing: mul_mat, [1, 4, 3, 3] = [4, 1, 3, 3] * [4, 4, 3, 3] 391s x0: [4, 1, 3, 3] 391s -0.207 0.234 0.732 -0.926 391s 391s x1: [4, 4, 3, 3] 391s 0.218 -0.319 0.322 -0.884 391s -0.987 -0.038 0.080 -0.012 391s 0.917 0.015 0.874 0.006 391s -0.172 0.142 -0.425 0.093 391s 391s y: [1, 4, 3, 3] 391s 0.934 391s 0.265 391s 0.448 391s -0.328 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.711 -0.365 0.712 0.168 391s 391s x1: [4, 1, 1, 1] 391s 0.526 0.518 -0.056 -0.502 391s 391s y: [1, 1, 1, 1] 391s 0.061 391s testing: mul_mat, [2, 4, 3, 1] = [1, 2, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.376 391s -0.215 391s 391s x1: [1, 4, 3, 1] 391s -0.648 391s -0.860 391s -0.248 391s 0.099 391s 391s y: [2, 4, 3, 1] 391s 0.244 0.139 391s 0.324 0.185 391s 0.093 0.053 391s -0.037 -0.021 391s testing: mul_mat, [4, 1, 3, 3] = [2, 4, 3, 3] * [2, 1, 3, 3] 391s x0: [2, 4, 3, 3] 391s 0.672 -0.140 391s 0.172 0.196 391s -0.083 0.684 391s 0.060 0.645 391s 391s x1: [2, 1, 3, 3] 391s 0.978 -0.361 391s 391s y: [4, 1, 3, 3] 391s 0.708 0.097 -0.329 -0.174 391s test-mul-mat0: iter:112/500 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.587 -0.562 -0.181 391s -0.623 -0.932 -0.604 391s 391s x1: [3, 4, 1, 1] 391s -0.792 -0.179 -0.749 391s -0.607 -0.613 -0.606 391s -0.598 0.194 0.705 391s -0.624 -0.769 -0.793 391s 391s y: [2, 4, 1, 1] 391s -0.228 1.112 391s 0.098 1.316 391s -0.587 -0.234 391s 0.210 1.584 391s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.849 0.052 0.612 391s -0.618 0.259 -0.186 391s 0.968 0.697 0.634 391s 391s x1: [3, 2, 3, 1] 391s 0.520 -0.842 -0.205 391s -0.790 -0.608 0.659 391s 391s y: [3, 2, 3, 1] 391s -0.611 -0.502 -0.214 391s 1.043 0.208 -0.771 391s testing: mul_mat, [4, 3, 3, 2] = [3, 4, 3, 2] * [3, 3, 3, 2] 391s x0: [3, 4, 3, 2] 391s 0.470 -0.598 0.706 391s 0.331 0.861 0.189 391s -0.360 -0.743 -0.850 391s -0.253 -0.175 -0.261 391s 391s x1: [3, 3, 3, 2] 391s -0.063 0.478 -0.203 391s -0.827 0.628 0.848 391s 0.785 -0.990 0.108 391s 391s y: [4, 3, 3, 2] 391s -0.459 0.352 -0.160 -0.014 391s -0.166 0.428 -0.891 -0.122 391s 1.037 -0.573 0.362 -0.054 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.980 0.324 -0.659 391s 0.832 0.721 -0.234 391s 0.977 0.266 -0.560 391s -0.529 -0.709 0.168 391s 391s x1: [3, 4, 1, 1] 391s -0.463 -0.698 0.561 391s 0.364 -0.099 0.621 391s -0.678 0.415 0.312 391s -0.402 -0.255 0.073 391s 391s y: [4, 4, 1, 1] 391s -1.049 -1.020 -0.952 0.834 391s -0.084 0.087 -0.018 -0.018 391s -0.735 -0.337 -0.726 0.116 391s -0.525 -0.536 -0.502 0.406 391s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.336 -0.963 0.889 -0.045 391s -0.170 0.289 -0.226 0.373 391s -0.999 -0.768 -0.974 0.068 391s -0.713 -0.341 0.562 0.112 391s 391s x1: [4, 3, 3, 1] 391s -0.167 -0.039 -0.160 0.722 391s -0.997 -0.894 -0.741 -0.695 391s 0.667 0.623 0.207 0.288 391s 391s y: [4, 3, 3, 1] 391s -0.081 0.323 0.402 0.123 391s 0.569 -0.181 2.357 0.522 391s -0.653 0.127 -1.326 -0.540 391s testing: mul_mat, [2, 4, 3, 2] = [4, 2, 3, 2] * [4, 4, 3, 2] 391s x0: [4, 2, 3, 2] 391s 0.283 0.653 0.980 -0.385 391s -0.706 -0.906 0.784 0.715 391s 391s x1: [4, 4, 3, 2] 391s -0.134 -0.772 -0.982 0.884 391s 0.438 -0.855 -0.537 -0.965 391s -0.045 0.563 -0.140 -0.830 391s 0.188 -0.981 -0.950 0.644 391s 391s y: [2, 4, 3, 2] 391s -1.845 0.656 391s -0.590 -0.646 391s 0.537 -1.181 391s -1.767 0.470 391s test-mul-mat0: iter:113/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.862 0.819 0.567 -0.428 391s 0.359 0.714 0.871 -0.555 391s 0.941 0.192 0.223 -0.807 391s 391s x1: [4, 3, 1, 1] 391s 0.401 0.031 0.532 0.618 391s -0.848 -0.402 -0.514 -0.406 391s -0.851 -0.517 -0.681 0.994 391s 391s y: [3, 3, 1, 1] 391s -0.283 0.287 0.003 391s 0.285 -0.814 -0.662 391s -0.502 -1.820 -1.854 391s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.960 0.621 0.365 -0.187 391s 391s x1: [4, 3, 4, 1] 391s -0.775 -0.237 -0.719 0.164 391s -0.154 -0.655 -0.434 0.877 391s 0.878 -0.816 -0.971 -0.524 391s 391s y: [1, 3, 4, 1] 391s -1.184 391s -0.877 391s 0.079 391s testing: mul_mat, [1, 1, 4, 3] = [4, 1, 4, 3] * [4, 1, 4, 3] 391s x0: [4, 1, 4, 3] 391s -0.165 0.299 -0.764 0.251 391s 391s x1: [4, 1, 4, 3] 391s -0.453 -0.825 -0.106 0.216 391s 391s y: [1, 1, 4, 3] 391s -0.037 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.515 -0.224 0.835 0.730 391s -0.109 -0.264 -0.264 -0.965 391s 0.182 0.754 0.024 -0.469 391s 391s x1: [4, 1, 1, 1] 391s 0.794 -0.999 0.827 0.327 391s 391s y: [3, 1, 1, 1] 391s 1.563 -0.357 -0.741 391s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.535 -0.289 0.641 391s 0.517 0.454 0.226 391s -0.732 -0.918 -0.369 391s 391s x1: [3, 4, 4, 1] 391s -0.125 -0.428 0.217 391s 0.574 0.789 -0.104 391s -0.284 0.576 -0.518 391s -0.240 -0.631 0.065 391s 391s y: [3, 4, 4, 1] 391s 0.196 -0.210 0.405 391s 0.012 0.631 -1.106 391s -0.650 -0.002 -0.130 391s 0.096 -0.396 0.731 391s testing: mul_mat, [4, 3, 4, 3] = [3, 4, 4, 3] * [3, 3, 4, 3] 391s x0: [3, 4, 4, 3] 391s -0.496 -0.966 -0.399 391s -0.531 -0.822 0.109 391s -0.814 0.187 -0.615 391s 0.112 -0.083 -0.732 391s 391s x1: [3, 3, 4, 3] 391s -0.849 0.060 0.994 391s -0.217 -0.344 -0.277 391s 0.999 -0.263 0.828 391s 391s y: [4, 3, 4, 3] 391s -0.034 0.509 0.090 -0.827 391s 0.551 0.368 0.283 0.207 391s -0.571 -0.223 -1.371 -0.472 391s test-mul-mat0: iter:114/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.938 391s 0.822 391s 391s x1: [1, 4, 1, 1] 391s 0.886 391s 0.706 391s -0.432 391s 0.615 391s 391s y: [2, 4, 1, 1] 391s 0.831 0.728 391s 0.662 0.581 391s -0.405 -0.355 391s 0.577 0.506 391s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.560 391s -0.906 391s 0.587 391s 391s x1: [1, 2, 2, 1] 391s 0.842 391s 0.036 391s 391s y: [3, 2, 2, 1] 391s -0.472 -0.763 0.494 391s -0.020 -0.032 0.021 391s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.395 391s 391s x1: [1, 3, 2, 1] 391s 0.784 391s 0.809 391s -0.160 391s 391s y: [1, 3, 2, 1] 391s -0.310 391s -0.320 391s 0.063 391s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.316 391s 0.406 391s -0.593 391s 391s x1: [1, 1, 1, 1] 391s -0.966 391s 391s y: [3, 1, 1, 1] 391s 0.305 -0.392 0.572 391s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 4, 2, 1] 391s -0.053 0.453 -0.824 391s 0.013 0.179 0.274 391s 0.085 0.957 -0.619 391s -0.629 -0.403 0.985 391s 391s x1: [3, 1, 2, 1] 391s -0.609 0.974 -0.977 391s 391s y: [4, 1, 2, 1] 391s 1.279 -0.101 1.486 -0.972 391s testing: mul_mat, [3, 3, 2, 1] = [4, 3, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.176 -0.749 -0.931 0.042 391s -0.762 0.941 0.617 0.205 391s 0.653 0.029 -0.572 0.671 391s 391s x1: [4, 3, 2, 1] 391s 0.818 0.676 0.398 -0.027 391s 0.588 0.243 0.970 -0.465 391s -0.743 0.055 -0.094 0.710 391s 391s y: [3, 3, 2, 1] 391s -1.023 0.254 0.308 391s -1.208 0.284 -0.476 391s 0.207 0.706 0.046 391s test-mul-mat0: iter:115/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.252 -0.188 0.178 -0.007 391s -0.260 0.246 0.639 -0.623 391s 0.548 0.148 -0.679 -0.739 391s 391s x1: [4, 3, 1, 1] 391s 0.979 -0.749 -0.969 -0.197 391s -0.511 0.684 0.054 -0.570 391s -0.287 0.123 -0.953 0.141 391s 391s y: [3, 3, 1, 1] 391s 0.217 -0.935 1.230 391s -0.244 0.691 0.206 391s -0.266 -0.592 0.404 391s testing: mul_mat, [1, 3, 2, 1] = [4, 1, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 1, 2, 1] 391s -0.639 -0.787 0.789 0.509 391s 391s x1: [4, 3, 2, 1] 391s -0.205 -0.048 0.369 -0.307 391s 0.262 -0.285 -0.328 0.512 391s -0.254 0.475 -0.999 -0.570 391s 391s y: [1, 3, 2, 1] 391s 0.304 391s 0.059 391s -1.290 391s testing: mul_mat, [3, 1, 2, 4] = [4, 3, 2, 4] * [4, 1, 2, 4] 391s x0: [4, 3, 2, 4] 391s -0.346 0.772 -0.535 -0.085 391s -0.347 0.610 0.176 0.963 391s 0.284 0.869 -0.582 -0.390 391s 391s x1: [4, 1, 2, 4] 391s 0.419 -0.002 -0.252 -0.865 391s 391s y: [3, 1, 2, 4] 391s 0.062 -1.025 0.602 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.801 0.356 0.146 0.077 391s 391s x1: [4, 3, 1, 1] 391s 0.180 0.682 0.148 0.075 391s -0.953 0.013 -0.541 0.512 391s -0.522 -0.632 0.231 0.295 391s 391s y: [1, 3, 1, 1] 391s 0.125 391s 0.729 391s 0.250 391s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 1, 2, 1] 391s 0.616 391s 391s x1: [1, 4, 2, 1] 391s -0.071 391s 0.120 391s -0.583 391s -0.547 391s 391s y: [1, 4, 2, 1] 391s -0.044 391s 0.074 391s -0.360 391s -0.337 391s testing: mul_mat, [4, 1, 2, 4] = [1, 4, 2, 4] * [1, 1, 2, 4] 391s x0: [1, 4, 2, 4] 391s 0.877 391s -0.145 391s 0.745 391s -0.755 391s 391s x1: [1, 1, 2, 4] 391s 0.163 391s 391s y: [4, 1, 2, 4] 391s 0.143 -0.024 0.121 -0.123 391s test-mul-mat0: iter:116/500 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.702 -0.215 -0.222 0.822 391s 0.169 -0.628 -0.548 0.897 391s 391s x1: [4, 3, 1, 1] 391s -0.064 0.258 0.457 0.332 391s 0.393 -0.180 -0.343 -0.351 391s -0.402 -0.565 0.608 0.563 391s 391s y: [2, 3, 1, 1] 391s 0.161 -0.126 391s -0.450 0.053 391s 0.731 0.458 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.753 -0.681 0.592 0.210 391s 0.651 -0.015 -0.970 -0.692 391s 391s x1: [4, 2, 1, 1] 391s 0.261 0.336 -0.957 0.493 391s 0.487 -0.934 -0.420 0.382 391s 391s y: [2, 2, 1, 1] 391s -0.495 0.752 391s 0.834 0.473 391s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 391s x0: [4, 4, 1, 2] 391s 0.446 -0.365 -0.849 -0.974 391s -0.982 -0.515 0.778 -0.663 391s -0.923 -0.011 0.987 0.063 391s 0.018 -0.705 -0.303 0.646 391s 391s x1: [4, 2, 1, 2] 391s 0.634 -0.372 -0.257 0.242 391s -0.809 0.542 0.540 -0.024 391s 391s y: [4, 2, 1, 2] 391s 0.401 -0.792 -0.819 0.508 391s -0.993 0.952 1.272 -0.576 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.717 -0.026 0.590 -0.945 391s -0.309 0.097 0.814 0.181 391s 391s x1: [4, 4, 1, 1] 391s -0.563 0.236 0.014 -0.545 391s 0.721 -0.207 -0.209 0.799 391s 0.781 -0.221 -0.139 -0.201 391s 0.074 0.558 -0.554 -0.888 391s 391s y: [2, 4, 1, 1] 391s 0.921 0.110 391s -1.389 -0.268 391s -0.446 -0.412 391s 0.446 -0.580 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.160 0.952 391s 0.097 -0.104 391s 0.519 0.300 391s 391s x1: [2, 4, 1, 1] 391s -0.533 0.055 391s -0.276 -0.732 391s -0.860 0.712 391s -0.294 0.376 391s 391s y: [3, 4, 1, 1] 391s -0.033 -0.058 -0.260 391s -0.741 0.049 -0.363 391s 0.540 -0.158 -0.233 391s 0.311 -0.068 -0.040 391s testing: mul_mat, [2, 2, 1, 2] = [3, 2, 1, 2] * [3, 2, 1, 2] 391s x0: [3, 2, 1, 2] 391s 0.600 0.378 0.035 391s 0.980 -0.934 -0.899 391s 391s x1: [3, 2, 1, 2] 391s -0.270 0.757 -0.901 391s 0.804 0.315 -0.455 391s 391s y: [2, 2, 1, 2] 391s 0.093 -0.161 391s 0.585 0.903 391s test-mul-mat0: iter:117/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.456 -0.313 0.617 391s 391s x1: [3, 1, 1, 1] 391s -0.456 0.357 -0.117 391s 391s y: [1, 1, 1, 1] 391s -0.392 391s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.217 0.879 0.769 391s -0.116 -0.786 -0.056 391s 391s x1: [3, 1, 3, 1] 391s -0.999 -0.398 -0.572 391s 391s y: [2, 1, 3, 1] 391s -0.573 0.460 391s testing: mul_mat, [3, 2, 3, 2] = [3, 3, 3, 2] * [3, 2, 3, 2] 391s x0: [3, 3, 3, 2] 391s 0.097 -0.093 0.342 391s -0.210 0.302 0.402 391s 0.218 -0.520 0.341 391s 391s x1: [3, 2, 3, 2] 391s -0.892 0.177 0.606 391s -0.891 0.780 -0.966 391s 391s y: [3, 2, 3, 2] 391s 0.104 0.484 -0.080 391s -0.489 0.034 -0.929 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.337 0.816 -0.671 391s 0.974 0.849 0.197 391s 0.911 0.576 -0.118 391s 0.449 0.072 -0.888 391s 391s x1: [3, 3, 1, 1] 391s 0.713 -0.714 -0.222 391s 0.045 0.235 0.515 391s -0.705 0.278 0.349 391s 391s y: [4, 3, 1, 1] 391s -0.674 0.045 0.264 0.466 391s -0.170 0.345 0.115 -0.421 391s 0.230 -0.381 -0.523 -0.607 391s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.823 -0.052 -0.461 0.307 391s -0.582 0.184 0.276 0.462 391s 0.345 0.698 0.756 0.919 391s -0.586 -0.287 0.242 0.248 391s 391s x1: [4, 3, 3, 1] 391s 0.719 0.881 0.192 -0.995 391s 0.846 -0.138 -0.476 -0.755 391s 0.061 -0.762 -0.469 0.839 391s 391s y: [4, 3, 3, 1] 391s -1.032 -0.663 0.094 -0.874 391s -0.702 -0.998 -0.858 -0.758 391s 0.463 0.082 -0.095 0.277 391s testing: mul_mat, [1, 4, 3, 2] = [4, 1, 3, 2] * [4, 4, 3, 2] 391s x0: [4, 1, 3, 2] 391s 0.581 0.967 0.284 0.576 391s 391s x1: [4, 4, 3, 2] 391s 0.738 0.732 -0.819 0.714 391s -0.110 0.456 -0.864 -0.722 391s -0.622 -0.115 0.090 -0.048 391s 0.806 -0.814 0.703 0.145 391s 391s y: [1, 4, 3, 2] 391s 1.315 391s -0.285 391s -0.475 391s -0.036 391s test-mul-mat0: iter:118/500 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.319 -0.367 391s 0.044 -0.829 391s 0.957 -0.206 391s -0.018 0.589 391s 391s x1: [2, 3, 1, 1] 391s -0.683 0.548 391s 0.758 0.898 391s 0.515 0.042 391s 391s y: [4, 3, 1, 1] 391s -0.419 -0.484 -0.767 0.335 391s -0.088 -0.711 0.541 0.515 391s 0.149 -0.012 0.484 0.015 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.881 0.552 391s 0.764 -0.614 391s 0.372 -0.074 391s 0.543 0.689 391s 391s x1: [2, 4, 1, 1] 391s 0.538 0.533 391s 0.531 -0.660 391s 0.337 -0.777 391s -0.315 0.777 391s 391s y: [4, 4, 1, 1] 391s 0.768 0.083 0.160 0.659 391s 0.104 0.812 0.247 -0.166 391s -0.132 0.735 0.183 -0.352 391s 0.152 -0.718 -0.175 0.364 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.478 -0.067 391s -0.102 -0.541 391s 391s x1: [2, 4, 1, 1] 391s -0.527 0.301 391s 0.586 0.988 391s -0.657 0.060 391s 0.307 -0.024 391s 391s y: [2, 4, 1, 1] 391s -0.272 -0.109 391s 0.214 -0.594 391s -0.318 0.035 391s 0.148 -0.018 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.330 0.461 391s -0.724 0.107 391s 391s x1: [2, 2, 1, 1] 391s -0.478 -0.564 391s 0.992 -0.947 391s 391s y: [2, 2, 1, 1] 391s -0.418 0.286 391s -0.109 -0.820 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.968 -0.272 391s -0.745 -0.428 391s 0.271 0.307 391s 0.720 -0.284 391s 391s x1: [2, 2, 1, 1] 391s 0.565 0.342 391s -0.340 0.329 391s 391s y: [4, 2, 1, 1] 391s -0.640 -0.567 0.258 0.309 391s 0.240 0.113 0.009 -0.339 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.346 0.804 0.952 0.399 391s 391s x1: [4, 2, 1, 1] 391s -0.084 -0.634 -0.977 0.892 391s -0.530 0.500 -0.175 0.368 391s 391s y: [1, 2, 1, 1] 391s -1.113 391s 0.199 391s test-mul-mat0: iter:119/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.177 -0.617 -0.200 391s 0.551 -0.045 -0.893 391s 391s x1: [3, 2, 1, 1] 391s -0.611 0.240 0.383 391s 0.049 -0.431 0.111 391s 391s y: [2, 2, 1, 1] 391s -0.117 -0.690 391s 0.235 -0.053 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.412 -0.688 0.749 391s 0.934 -0.884 0.701 391s 0.333 -0.304 -0.018 391s 0.008 -0.263 0.371 391s 391s x1: [3, 2, 1, 1] 391s -0.733 0.871 -0.527 391s -0.711 0.763 -0.056 391s 391s y: [4, 2, 1, 1] 391s -0.691 -1.823 -0.500 -0.431 391s -0.274 -1.377 -0.468 -0.227 391s testing: mul_mat, [4, 4, 1, 2] = [3, 4, 1, 2] * [3, 4, 1, 2] 391s x0: [3, 4, 1, 2] 391s -0.942 -0.050 0.969 391s 0.067 0.687 0.340 391s 0.315 0.807 0.760 391s -0.868 0.039 0.260 391s 391s x1: [3, 4, 1, 2] 391s -0.752 -0.879 -0.580 391s -0.183 0.232 0.500 391s 0.640 0.615 -0.699 391s 0.191 -0.430 -0.592 391s 391s y: [4, 4, 1, 2] 391s 0.190 -0.851 -1.388 0.467 391s 0.645 0.317 0.510 0.298 391s -1.311 0.227 0.166 -0.714 391s -0.732 -0.484 -0.737 -0.337 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.826 0.509 0.047 391s 391s x1: [3, 4, 1, 1] 391s 0.706 -0.340 -0.188 391s -0.008 0.386 0.065 391s 0.978 0.444 -0.985 391s 0.947 -0.489 0.702 391s 391s y: [1, 4, 1, 1] 391s 0.401 391s 0.193 391s 0.987 391s 0.566 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.202 391s 0.869 391s -0.308 391s -0.573 391s 391s x1: [1, 3, 1, 1] 391s 0.957 391s -0.452 391s -0.692 391s 391s y: [4, 3, 1, 1] 391s 0.194 0.832 -0.295 -0.549 391s -0.091 -0.393 0.139 0.259 391s -0.140 -0.601 0.213 0.397 391s testing: mul_mat, [4, 1, 1, 2] = [4, 4, 1, 2] * [4, 1, 1, 2] 391s x0: [4, 4, 1, 2] 391s -0.825 0.958 0.962 -0.534 391s 0.980 0.451 -0.043 -0.760 391s -0.993 0.255 0.192 0.746 391s 0.240 0.905 0.788 0.014 391s 391s x1: [4, 1, 1, 2] 391s 0.838 0.114 -0.908 -0.312 391s 391s y: [4, 1, 1, 2] 391s -1.290 1.149 -1.210 -0.416 391s test-mul-mat0: iter:120/500 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.934 391s 391s x1: [1, 3, 1, 1] 391s -0.223 391s -0.023 391s 0.395 391s 391s y: [1, 3, 1, 1] 391s -0.208 391s -0.022 391s 0.369 391s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 1, 4, 1] 391s 0.414 391s 391s x1: [1, 1, 4, 1] 391s -0.413 391s 391s y: [1, 1, 4, 1] 391s -0.171 391s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.010 391s -0.011 391s 391s x1: [1, 1, 4, 1] 391s 0.792 391s 391s y: [2, 1, 4, 1] 391s 0.008 -0.009 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.668 391s 0.127 391s -0.527 391s 391s x1: [1, 2, 1, 1] 391s -0.157 391s -0.650 391s 391s y: [3, 2, 1, 1] 391s 0.105 -0.020 0.083 391s 0.434 -0.082 0.342 391s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.671 0.135 -0.361 391s -0.875 0.344 -0.405 391s -0.645 0.284 0.866 391s 0.525 -0.159 -0.586 391s 391s x1: [3, 1, 4, 1] 391s 0.728 0.866 -0.593 391s 391s y: [4, 1, 4, 1] 391s -0.157 -0.098 -0.737 0.592 391s testing: mul_mat, [2, 3, 4, 1] = [4, 2, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.165 -0.678 0.570 0.197 391s -0.301 -0.637 0.067 -0.028 391s 391s x1: [4, 3, 4, 1] 391s -0.173 0.122 -0.597 0.618 391s -0.065 -0.995 0.216 0.889 391s -0.727 0.034 0.360 -0.043 391s 391s y: [2, 3, 4, 1] 391s -0.330 -0.083 391s 0.962 0.643 391s 0.054 0.222 391s test-mul-mat0: iter:121/500 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.924 0.210 0.155 0.596 391s 391s x1: [4, 2, 1, 1] 391s 0.279 0.519 0.079 -0.054 391s 0.402 0.895 -0.162 0.165 391s 391s y: [1, 2, 1, 1] 391s -0.168 391s -0.110 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.494 -0.152 0.306 -0.601 391s -0.255 -0.900 -0.763 -0.364 391s -0.940 -0.226 -0.152 -0.618 391s -0.118 -0.873 0.902 0.961 391s 391s x1: [4, 1, 1, 1] 391s 0.543 -0.567 0.286 -0.563 391s 391s y: [4, 1, 1, 1] 391s 0.244 0.359 -0.079 0.147 391s testing: mul_mat, [2, 4, 1, 2] = [4, 2, 1, 2] * [4, 4, 1, 2] 391s x0: [4, 2, 1, 2] 391s 0.482 0.846 -0.403 -0.282 391s 0.832 -0.724 0.110 0.414 391s 391s x1: [4, 4, 1, 2] 391s 0.073 0.304 0.856 0.910 391s -0.531 0.236 0.986 0.679 391s -0.609 0.582 0.223 -0.175 391s -0.133 0.659 0.074 0.374 391s 391s y: [2, 4, 1, 2] 391s -0.310 0.311 391s -0.646 -0.224 391s 0.158 -0.976 391s 0.359 -0.425 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.885 -0.595 -0.847 0.512 391s 391s x1: [4, 2, 1, 1] 391s 0.427 -0.082 0.868 0.436 391s -0.194 -0.592 -0.780 0.542 391s 391s y: [1, 2, 1, 1] 391s -0.085 391s 1.120 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.712 391s -0.982 391s 0.153 391s 391s x1: [1, 4, 1, 1] 391s -0.659 391s -0.720 391s -0.006 391s -0.814 391s 391s y: [3, 4, 1, 1] 391s -0.470 0.648 -0.101 391s -0.513 0.707 -0.110 391s -0.004 0.006 -0.001 391s -0.580 0.799 -0.124 391s testing: mul_mat, [4, 1, 1, 2] = [3, 4, 1, 2] * [3, 1, 1, 2] 391s x0: [3, 4, 1, 2] 391s 0.367 -0.607 0.428 391s 0.475 0.021 -0.550 391s 0.153 -0.770 -0.258 391s -0.206 0.200 -0.242 391s 391s x1: [3, 1, 1, 2] 391s -0.178 0.431 -0.348 391s 391s y: [4, 1, 1, 2] 391s -0.476 0.116 -0.270 0.207 391s test-mul-mat0: iter:122/500 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.550 -0.930 0.226 391s -0.250 0.498 0.676 391s 0.492 -0.744 -0.989 391s -0.361 0.167 0.858 391s 391s x1: [3, 2, 1, 1] 391s -0.164 0.682 0.896 391s -0.798 0.157 0.049 391s 391s y: [4, 2, 1, 1] 391s -0.522 0.987 -1.475 0.942 391s -0.574 0.311 -0.558 0.356 391s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 3, 2, 1] 391s -0.850 0.528 0.563 391s -0.211 -0.305 0.421 391s -0.084 0.053 -0.880 391s 391s x1: [3, 4, 2, 1] 391s -0.873 -0.642 -0.301 391s 0.801 -0.574 0.252 391s 0.492 0.647 -0.373 391s -0.384 0.519 -0.242 391s 391s y: [3, 4, 2, 1] 391s 0.233 0.253 0.304 391s -0.842 0.112 -0.319 391s -0.287 -0.458 0.321 391s 0.465 -0.179 0.272 391s testing: mul_mat, [2, 3, 2, 2] = [3, 2, 2, 2] * [3, 3, 2, 2] 391s x0: [3, 2, 2, 2] 391s 0.166 -0.146 -0.340 391s 0.623 0.627 -0.178 391s 391s x1: [3, 3, 2, 2] 391s -0.724 0.847 0.412 391s 0.930 -0.795 0.116 391s -0.420 -0.886 -0.110 391s 391s y: [2, 3, 2, 2] 391s -0.384 0.006 391s 0.231 0.061 391s 0.097 -0.797 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.424 -0.801 0.831 391s -0.884 0.743 -0.708 391s -0.391 0.431 0.944 391s 391s x1: [3, 2, 1, 1] 391s 0.524 -0.176 -0.478 391s -0.822 0.409 0.262 391s 391s y: [3, 2, 1, 1] 391s -0.478 -0.255 -0.732 391s 0.238 0.845 0.746 391s testing: mul_mat, [4, 3, 2, 1] = [3, 4, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.867 0.619 -0.687 391s 0.503 -0.118 -0.528 391s -0.727 -0.359 0.654 391s -0.558 0.362 -0.744 391s 391s x1: [3, 3, 2, 1] 391s 0.035 -0.447 -0.158 391s 0.233 0.443 0.179 391s -0.455 0.677 -0.566 391s 391s y: [4, 3, 2, 1] 391s -0.137 0.154 0.031 -0.064 391s 0.353 -0.030 -0.211 -0.103 391s 0.414 -0.010 -0.282 0.919 391s testing: mul_mat, [2, 3, 2, 2] = [4, 2, 2, 2] * [4, 3, 2, 2] 391s x0: [4, 2, 2, 2] 391s -0.646 0.158 -0.711 0.724 391s 0.734 0.900 0.557 -0.870 391s 391s x1: [4, 3, 2, 2] 391s -0.749 -0.573 -0.837 -0.602 391s -0.115 -0.596 -0.277 -0.249 391s 0.907 -0.004 0.194 0.526 391s 391s y: [2, 3, 2, 2] 391s 0.552 -1.008 391s -0.003 -0.559 391s -0.344 0.313 391s test-mul-mat0: iter:123/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.975 391s -0.775 391s -0.548 391s 391s x1: [1, 3, 1, 1] 391s 0.365 391s 0.714 391s -0.374 391s 391s y: [3, 3, 1, 1] 391s 0.355 -0.283 -0.200 391s 0.696 -0.554 -0.392 391s -0.365 0.290 0.205 391s testing: mul_mat, [2, 3, 2, 1] = [1, 2, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.258 391s -0.051 391s 391s x1: [1, 3, 2, 1] 391s 0.217 391s 0.544 391s 0.202 391s 391s y: [2, 3, 2, 1] 391s 0.056 -0.011 391s 0.140 -0.028 391s 0.052 -0.010 391s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.060 391s -0.082 391s -0.813 391s 0.031 391s 391s x1: [1, 2, 2, 1] 391s 0.891 391s 0.299 391s 391s y: [4, 2, 2, 1] 391s -0.053 -0.073 -0.724 0.027 391s -0.018 -0.025 -0.243 0.009 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.529 391s -0.541 391s 391s x1: [1, 4, 1, 1] 391s 0.359 391s -0.265 391s -0.754 391s -0.993 391s 391s y: [2, 4, 1, 1] 391s -0.190 -0.194 391s 0.140 0.143 391s 0.399 0.408 391s 0.526 0.537 391s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 4, 2, 1] 391s 0.445 0.995 391s 0.067 0.626 391s -0.747 -0.714 391s -0.606 0.293 391s 391s x1: [2, 1, 2, 1] 391s 0.926 -0.986 391s 391s y: [4, 1, 2, 1] 391s -0.568 -0.555 0.012 -0.850 391s testing: mul_mat, [4, 2, 2, 1] = [4, 4, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 4, 2, 1] 391s -0.630 -0.375 0.908 0.396 391s -0.661 0.077 -0.404 0.775 391s 0.783 -0.174 0.003 0.940 391s 0.996 0.355 -0.073 0.217 391s 391s x1: [4, 2, 2, 1] 391s 0.013 0.821 -0.608 -0.627 391s -0.444 -0.363 -0.620 0.266 391s 391s y: [4, 2, 2, 1] 391s -1.117 -0.186 -0.725 0.213 391s -0.042 0.723 -0.036 -0.468 391s test-mul-mat0: iter:124/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.061 0.298 -0.564 391s -0.665 0.074 -0.623 391s 0.552 0.241 -0.310 391s 0.325 -0.955 0.382 391s 391s x1: [3, 1, 1, 1] 391s 0.601 -0.656 -0.705 391s 391s y: [4, 1, 1, 1] 391s 0.165 -0.009 0.392 0.552 391s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 4, 3, 1] 391s -0.819 0.221 0.587 391s -0.531 -0.115 -0.177 391s 0.282 -0.933 -0.001 391s 0.695 0.593 0.469 391s 391s x1: [3, 4, 3, 1] 391s -0.637 0.957 -0.086 391s -0.406 -0.008 -0.055 391s 0.547 -0.857 -0.079 391s -0.712 0.664 0.236 391s 391s y: [4, 4, 3, 1] 391s 0.683 0.243 -1.072 0.084 391s 0.298 0.226 -0.106 -0.313 391s -0.684 -0.178 0.954 -0.165 391s 0.868 0.260 -0.820 0.010 391s testing: mul_mat, [3, 4, 3, 2] = [3, 3, 3, 2] * [3, 4, 3, 2] 391s x0: [3, 3, 3, 2] 391s -0.115 -0.832 0.964 391s -0.402 0.035 -0.980 391s -0.451 0.893 -0.347 391s 391s x1: [3, 4, 3, 2] 391s 0.743 0.490 0.633 391s -0.259 0.186 0.226 391s -0.789 -0.196 0.284 391s 0.449 -0.219 -0.858 391s 391s y: [3, 4, 3, 2] 391s 0.117 -0.901 -0.117 391s 0.093 -0.111 0.204 391s 0.528 0.032 0.083 391s -0.698 0.653 -0.100 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.337 0.913 0.878 391s -0.754 0.795 -0.348 391s 391s x1: [3, 3, 1, 1] 391s -0.035 0.494 0.079 391s 0.611 -0.573 -0.321 391s 0.255 -0.100 0.163 391s 391s y: [2, 3, 1, 1] 391s 0.508 0.392 391s -0.599 -0.805 391s 0.138 -0.328 391s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.073 -0.482 391s 0.355 0.300 391s -0.767 0.164 391s 0.826 -0.236 391s 391s x1: [2, 3, 3, 1] 391s -0.094 0.170 391s -0.485 -0.453 391s 0.955 -0.436 391s 391s y: [4, 3, 3, 1] 391s -0.075 0.018 0.100 -0.118 391s 0.254 -0.308 0.298 -0.294 391s 0.140 0.208 -0.804 0.892 391s testing: mul_mat, [2, 2, 3, 2] = [4, 2, 3, 2] * [4, 2, 3, 2] 391s x0: [4, 2, 3, 2] 391s 0.873 0.088 0.376 -0.086 391s -0.408 0.922 0.133 0.741 391s 391s x1: [4, 2, 3, 2] 391s -0.289 0.131 -0.489 -0.921 391s 0.631 0.187 0.057 -0.442 391s 391s y: [2, 2, 3, 2] 391s -0.345 -0.509 391s 0.627 -0.406 391s test-mul-mat0: iter:125/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.484 0.581 -0.497 391s 0.485 -0.829 -0.036 391s 391s x1: [3, 3, 1, 1] 391s -0.574 0.199 0.340 391s 0.509 0.199 0.743 391s -0.063 0.405 -0.634 391s 391s y: [2, 3, 1, 1] 391s -0.331 -0.455 391s -0.007 0.055 391s 0.520 -0.343 391s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.500 0.935 -0.698 391s 0.985 -0.894 0.266 391s -0.616 -0.391 -0.841 391s -0.078 0.860 -0.090 391s 391s x1: [3, 2, 4, 1] 391s -0.600 -0.497 -0.107 391s -0.462 0.251 -0.248 391s 391s y: [4, 2, 4, 1] 391s -0.089 -0.175 0.654 -0.371 391s 0.639 -0.745 0.396 0.274 391s testing: mul_mat, [1, 4, 4, 3] = [3, 1, 4, 3] * [3, 4, 4, 3] 391s x0: [3, 1, 4, 3] 391s -0.802 -0.071 0.527 391s 391s x1: [3, 4, 4, 3] 391s -0.060 -0.035 -0.282 391s 0.144 -0.179 0.614 391s 0.595 -0.638 -0.880 391s -0.966 -0.474 0.642 391s 391s y: [1, 4, 4, 3] 391s -0.098 391s 0.221 391s -0.896 391s 1.146 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.167 -0.846 0.559 391s 0.003 0.919 -0.384 391s 391s x1: [3, 1, 1, 1] 391s -0.322 0.181 -0.458 391s 391s y: [2, 1, 1, 1] 391s -0.463 0.341 391s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.676 0.595 391s 0.172 -0.269 391s -0.283 0.333 391s 391s x1: [2, 3, 4, 1] 391s -0.345 0.899 391s 0.012 0.239 391s 0.970 0.202 391s 391s y: [3, 3, 4, 1] 391s 0.302 -0.301 0.397 391s 0.150 -0.062 0.076 391s 0.777 0.113 -0.207 391s testing: mul_mat, [4, 2, 4, 3] = [3, 4, 4, 3] * [3, 2, 4, 3] 391s x0: [3, 4, 4, 3] 391s 0.359 0.206 -0.696 391s -0.763 0.697 0.702 391s -0.932 -0.614 -0.195 391s 0.382 0.621 -0.136 391s 391s x1: [3, 2, 4, 3] 391s 0.697 0.360 0.677 391s -0.059 0.869 0.846 391s 391s y: [4, 2, 4, 3] 391s -0.146 0.194 -1.002 0.398 391s -0.431 1.245 -0.643 0.402 391s test-mul-mat0: iter:126/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.906 0.888 391s 0.061 -0.208 391s 0.238 0.671 391s 391s x1: [2, 2, 1, 1] 391s 0.483 0.611 391s 0.662 -0.582 391s 391s y: [3, 2, 1, 1] 391s 0.980 -0.098 0.525 391s 0.083 0.161 -0.233 391s testing: mul_mat, [2, 3, 2, 1] = [2, 2, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.101 -0.329 391s -0.054 0.025 391s 391s x1: [2, 3, 2, 1] 391s -0.728 0.925 391s 0.572 -0.710 391s 0.823 -0.339 391s 391s y: [2, 3, 2, 1] 391s -0.378 0.063 391s 0.291 -0.049 391s 0.195 -0.053 391s testing: mul_mat, [3, 2, 2, 2] = [2, 3, 2, 2] * [2, 2, 2, 2] 391s x0: [2, 3, 2, 2] 391s 0.685 0.698 391s 0.747 -0.727 391s -0.481 0.585 391s 391s x1: [2, 2, 2, 2] 391s 0.242 0.162 391s -0.274 -0.078 391s 391s y: [3, 2, 2, 2] 391s 0.279 0.06ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 3 -0.021 391s -0.242 -0.148 0.086 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.420 0.293 391s -0.226 0.882 391s -0.703 -0.773 391s -0.147 -0.517 391s 391s x1: [2, 3, 1, 1] 391s -0.878 0.294 391s 0.189 -0.456 391s -0.105 0.076 391s 391s y: [4, 3, 1, 1] 391s 0.456 0.458 0.390 -0.023 391s -0.213 -0.445 0.219 0.208 391s 0.067 0.091 0.015 -0.024 391s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.650 0.244 0.824 -0.323 391s 0.766 -0.158 0.616 -0.090 391s -0.627 -0.077 0.219 0.101 391s 391s x1: [4, 2, 2, 1] 391s -0.587 -0.373 -0.501 -0.425 391s -0.474 -0.177 0.978 0.588 391s 391s y: [3, 2, 2, 1] 391s 0.016 -0.661 0.244 391s 0.881 0.215 0.584 391s testing: mul_mat, [2, 4, 2, 2] = [3, 2, 2, 2] * [3, 4, 2, 2] 391s x0: [3, 2, 2, 2] 391s -0.855 0.159 -0.340 391s -0.050 -0.826 0.138 391s 391s x1: [3, 4, 2, 2] 391s 0.363 -0.874 -0.016 391s 0.070 0.181 0.561 391s -0.100 0.531 0.326 391s 0.273 -0.225 -0.832 391s 391s y: [2, 4, 2, 2] 391s -0.444 0.702 391s -0.222 -0.076 391s 0.059 -0.389 391s 0.013 0.058 391s test-mul-mat0: iter:127/500 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.999 -0.571 391s 0.678 -0.573 391s -0.740 -0.848 391s 391s x1: [2, 3, 1, 1] 391s 0.070 -0.611 391s -0.868 0.215 391s 0.339 0.291 391s 391s y: [3, 3, 1, 1] 391s 0.419 0.397 0.466 391s -0.990 -0.712 0.460 391s 0.172 0.063 -0.498 391s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.632 0.374 391s 0.507 0.707 391s 391s x1: [2, 3, 3, 1] 391s -0.732 0.373 391s 0.179 -0.115 391s 0.893 0.745 391s 391s y: [2, 3, 3, 1] 391s 0.602 -0.108 391s -0.156 0.010 391s -0.286 0.980 391s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.785 -0.262 391s 391s x1: [2, 2, 3, 1] 391s -0.598 -0.274 391s 0.905 0.287 391s 391s y: [1, 2, 3, 1] 391s 0.542 391s -0.786 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.732 0.974 391s -0.982 0.112 391s -0.110 0.464 391s 391s x1: [2, 1, 1, 1] 391s -0.439 0.353 391s 391s y: [3, 1, 1, 1] 391s 0.665 0.471 0.212 391s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.697 0.432 -0.845 391s -0.922 0.265 -0.500 391s 0.271 -0.474 -0.439 391s -0.742 0.450 0.618 391s 391s x1: [3, 2, 3, 1] 391s -0.892 -0.535 -0.810 391s -0.490 0.192 -0.905 391s 391s y: [4, 2, 3, 1] 391s -0.168 1.086 0.367 -0.080 391s 0.506 0.955 0.173 -0.110 391s testing: mul_mat, [2, 3, 3, 1] = [4, 2, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 2, 3, 1] 391s -0.777 -0.648 -0.854 0.032 391s 0.180 -0.324 0.030 -0.975 391s 391s x1: [4, 3, 3, 1] 391s -0.128 -0.260 -0.129 0.026 391s 0.240 0.432 -0.355 0.848 391s -0.541 -0.547 -0.034 -0.322 391s 391s y: [2, 3, 3, 1] 391s 0.379 0.032 391s -0.136 -0.934 391s 0.793 0.392 391s test-mul-mat0: iter:128/500 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.013 0.289 0.171 -0.977 391s -0.463 0.082 0.514 0.971 391s 0.919 0.272 0.142 0.940 391s 0.209 -0.614 0.100 0.315 391s 391s x1: [4, 2, 1, 1] 391s 0.056 0.937 -0.950 0.279 391s 0.117 -0.598 0.955 0.263 391s 391s y: [4, 2, 1, 1] 391s -0.166 -0.166 0.434 -0.570 391s -0.268 0.643 0.328 0.570 391s testing: mul_mat, [3, 4, 2, 1] = [4, 3, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.590 0.960 -0.906 0.577 391s 0.804 -0.752 -0.206 -0.072 391s 0.520 -0.587 -0.994 -0.945 391s 391s x1: [4, 4, 2, 1] 391s 0.311 0.613 0.011 -0.211 391s -0.783 0.167 0.844 -0.846 391s 0.217 0.123 0.272 0.619 391s 0.079 -0.465 0.052 -0.934 391s 391s y: [3, 4, 2, 1] 391s 0.273 -0.197 -0.009 391s -0.630 -0.868 -0.546 391s 0.101 -0.019 -0.815 391s -1.078 0.470 1.145 391s testing: mul_mat, [3, 3, 2, 3] = [4, 3, 2, 3] * [4, 3, 2, 3] 391s x0: [4, 3, 2, 3] 391s -0.844 0.991 0.467 -0.045 391s 0.072 0.576 0.528 0.592 391s -0.537 0.947 0.430 0.816 391s 391s x1: [4, 3, 2, 3] 391s 0.380 -0.277 0.905 0.644 391s -0.878 0.076 0.521 -0.469 391s 0.036 0.615 -0.892 -0.160 391s 391s y: [3, 3, 2, 3] 391s -0.202 0.727 0.448 391s 1.081 -0.022 0.385 391s 0.169 -0.209 0.049 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.036 0.279 -0.974 -0.658 391s 391s x1: [4, 3, 1, 1] 391s 0.297 -0.727 -0.674 0.847 391s 0.584 0.743 -0.251 -0.995 391s -0.661 0.700 0.077 0.853 391s 391s y: [1, 3, 1, 1] 391s -0.093 391s 1.128 391s -0.464 391s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.905 391s -0.580 391s 391s x1: [1, 4, 2, 1] 391s 0.396 391s 0.460 391s 0.287 391s -0.124 391s 391s y: [2, 4, 2, 1] 391s 0.358 -0.230 391s 0.416 -0.267 391s 0.260 -0.167 391s -0.112 0.072 391s testing: mul_mat, [1, 1, 2, 3] = [2, 1, 2, 3] * [2, 1, 2, 3] 391s x0: [2, 1, 2, 3] 391s -0.211 0.467 391s 391s x1: [2, 1, 2, 3] 391s 0.898 -0.989 391s 391s y: [1, 1, 2, 3] 391s -0.652 391s test-mul-mat0: iter:129/500 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.234 0.444 0.597 391s 0.197 0.604 -0.614 391s -0.335 0.042 0.201 391s -0.526 0.876 0.476 391s 391s x1: [3, 3, 1, 1] 391s -0.186 -0.704 -0.054 391s -0.176 -0.938 0.170 391s -0.318 0.412 0.138 391s 391s y: [4, 3, 1, 1] 391s -0.301 -0.428 0.022 -0.544 391s -0.274 -0.705 0.054 -0.648 391s 0.339 0.101 0.151 0.593 391s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.686 0.709 0.423 391s -0.840 0.585 -0.101 391s 0.730 0.172 0.589 391s 391s x1: [3, 4, 2, 1] 391s 0.570 0.587 -0.311 391s -0.470 -0.515 0.318 391s -0.370 0.288 -0.635 391s 0.443 0.584 0.312 391s 391s y: [3, 4, 2, 1] 391s 0.676 -0.104 0.334 391s -0.554 0.061 -0.245 391s -0.318 0.543 -0.594 391s 0.851 -0.062 0.608 391s testing: mul_mat, [2, 3, 2, 1] = [3, 2, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.656 -0.118 0.759 391s 0.916 -0.462 -0.335 391s 391s x1: [3, 3, 2, 1] 391s -0.399 -0.825 0.065 391s 0.551 0.233 -0.315 391s 0.791 0.057 0.749 391s 391s y: [2, 3, 2, 1] 391s -0.115 -0.006 391s 0.095 0.503 391s 1.080 0.447 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.524 -0.566 0.828 391s -0.981 -0.443 -0.753 391s -0.291 -0.679 -0.731 391s -0.444 -0.441 0.998 391s 391s x1: [3, 2, 1, 1] 391s 0.740 -0.814 0.585 391s 0.291 0.419 -0.730 391s 391s y: [4, 2, 1, 1] 391s 0.557 -0.806 -0.090 0.614 391s -0.994 0.078 0.164 -1.043 391s testing: mul_mat, [3, 3, 2, 1] = [4, 3, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.859 -0.613 -0.710 -0.891 391s 0.745 -0.985 0.166 -0.144 391s -0.029 -0.031 -0.274 0.148 391s 391s x1: [4, 3, 2, 1] 391s 0.903 0.151 0.757 0.819 391s 0.689 -0.578 -0.375 0.172 391s -0.642 0.764 0.183 -0.122 391s 391s y: [3, 3, 2, 1] 391s -2.136 0.531 -0.117 391s -0.125 0.996 0.126 391s 0.063 -1.183 -0.073 391s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 4, 2, 1] 391s -0.393 -0.520 -0.135 391s 0.114 -0.466 0.337 391s -0.046 -0.922 0.559 391s 0.404 0.110 -0.618 391s 391s x1: [3, 4, 2, 1] 391s -0.326 -0.370 -0.669 391s 0.260 -0.551 0.517 391s -0.598 -0.410 0.262 391s 0.373 -0.023 0.277 391s 391s y: [4, 4, 2, 1] 391s 0.411 -0.090 -0.018 0.241 391s 0.115 0.461 0.785 -0.275 391s 0.412 0.212 0.552 -0.449 391s -0.172 0.147 0.159 -0.023 391s test-mul-mat0: iter:130/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.229 -0.074 0.910 391s -0.212 0.308 0.650 391s 391s x1: [3, 3, 1, 1] 391s -0.516 -0.475 -0.427 391s 0.091 0.639 0.527 391s -0.505 -0.881 -0.939 391s 391s y: [2, 3, 1, 1] 391s -0.471 -0.314 391s 0.454 0.520 391s -0.905 -0.774 391s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 4, 4, 1] 391s 0.402 0.040 -0.347 391s -0.810 -0.652 -0.697 391s 0.126 0.301 0.066 391s 0.189 0.174 0.284 391s 391s x1: [3, 2, 4, 1] 391s -0.063 -0.048 -0.237 391s -0.937 0.874 -0.781 391s 391s y: [4, 2, 4, 1] 391s 0.055 0.248 -0.038 -0.088 391s -0.071 0.735 0.093 -0.247 391s testing: mul_mat, [4, 4, 4, 2] = [3, 4, 4, 2] * [3, 4, 4, 2] 391s x0: [3, 4, 4, 2] 391s -0.205 0.704 -0.579 391s -0.496 0.865 0.213 391s -0.569 -0.653 -0.391 391s 0.139 -0.507 0.461 391s 391s x1: [3, 4, 4, 2] 391s 0.992 -0.765 -0.046 391s 0.459 -0.212 0.573 391s 0.451 0.018 -0.753 391s 0.496 0.418 0.093 391s 391s y: [4, 4, 4, 2] 391s -0.715 -1.164 -0.047 0.504 391s -0.574 -0.288 -0.347 0.435 391s 0.356 -0.369 0.027 -0.294 391s 0.139 0.135 -0.591 -0.100 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.980 0.723 0.856 391s 391s x1: [3, 4, 1, 1] 391s 0.847 0.154 0.091 391s 0.702 0.996 0.141 391s 0.403 0.306 -0.487 391s 0.933 0.163 0.377 391s 391s y: [1, 4, 1, 1] 391s -0.641 391s 0.153 391s -0.591 391s -0.473 391s testing: mul_mat, [1, 3, 4, 1] = [1, 1, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 1, 4, 1] 391s -0.219 391s 391s x1: [1, 3, 4, 1] 391s 0.582 391s 0.736 391s 0.388 391s 391s y: [1, 3, 4, 1] 391s -0.128 391s -0.161 391s -0.085 391s testing: mul_mat, [1, 1, 4, 2] = [1, 1, 4, 2] * [1, 1, 4, 2] 391s x0: [1, 1, 4, 2] 391s 0.213 391s 391s x1: [1, 1, 4, 2] 391s -0.818 391s 391s y: [1, 1, 4, 2] 391s -0.174 391s test-mul-mat0: iter:131/500 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.139 391s 0.240 391s 0.279 391s 0.952 391s 391s x1: [1, 3, 1, 1] 391s -0.996 391s -0.330 391s 0.813 391s 391s y: [4, 3, 1, 1] 391s 0.139 -0.239 -0.278 -0.948 391s 0.046 -0.079 -0.092 -0.314 391s -0.113 0.195 0.227 0.774 391s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.819 391s 391s x1: [1, 4, 2, 1] 391s -0.486 391s 0.214 391s -0.866 391s 0.490 391s 391s y: [1, 4, 2, 1] 391s 0.399 391s -0.175 391s 0.710 391s -0.401 391s testing: mul_mat, [4, 1, 2, 3] = [1, 4, 2, 3] * [1, 1, 2, 3] 391s x0: [1, 4, 2, 3] 391s 0.377 391s 0.387 391s -0.331 391s 0.820 391s 391s x1: [1, 1, 2, 3] 391s -0.494 391s 391s y: [4, 1, 2, 3] 391s -0.186 -0.191 0.164 -0.405 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.620 391s -0.753 391s 391s x1: [1, 4, 1, 1] 391s -0.309 391s 0.791 391s 0.175 391s -0.867 391s 391s y: [2, 4, 1, 1] 391s 0.192 0.233 391s -0.491 -0.596 391s -0.109 -0.132 391s 0.538 0.653 391s testing: mul_mat, [2, 1, 2, 1] = [2, 2, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.588 -0.476 391s 0.578 -0.241 391s 391s x1: [2, 1, 2, 1] 391s 0.268 0.766 391s 391s y: [2, 1, 2, 1] 391s -0.207 -0.030 391s testing: mul_mat, [1, 2, 2, 3] = [2, 1, 2, 3] * [2, 2, 2, 3] 391s x0: [2, 1, 2, 3] 391s -0.072 0.077 391s 391s x1: [2, 2, 2, 3] 391s -0.913 -0.128 391s 0.667 0.616 391s 391s y: [1, 2, 2, 3] 391s 0.056 391s -0.000 391s test-mul-mat0: iter:132/500 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.060 391s -0.024 391s 0.158 391s 391s x1: [1, 4, 1, 1] 391s -0.443 391s 0.774 391s -0.333 391s -0.423 391s 391s y: [3, 4, 1, 1] 391s 0.027 0.010 -0.070 391s -0.046 -0.018 0.123 391s 0.020 0.008 -0.053 391s 0.025 0.010 -0.067 391s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 2, 3, 1] 391s 0.452 391s 0.571 391s 391s x1: [1, 3, 3, 1] 391s -0.518 391s 0.142 391s 0.889 391s 391s y: [2, 3, 3, 1] 391s -0.234 -0.296 391s 0.064 0.081 391s 0.402 0.508 391s testing: mul_mat, [1, 2, 3, 4] = [1, 1, 3, 4] * [1, 2, 3, 4] 391s x0: [1, 1, 3, 4] 391s -0.992 391s 391s x1: [1, 2, 3, 4] 391s 0.197 391s 0.569 391s 391s y: [1, 2, 3, 4] 391s -0.195 391s -0.565 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.371 391s 0.629 391s 391s x1: [1, 1, 1, 1] 391s 0.787 391s 391s y: [2, 1, 1, 1] 391s -0.292 0.495 391s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.449 0.856 391s -0.951 0.711 391s 391s x1: [2, 1, 3, 1] 391s 0.850 -0.466 391s 391s y: [2, 1, 3, 1] 391s -0.781 -1.140 391s testing: mul_mat, [4, 2, 3, 4] = [2, 4, 3, 4] * [2, 2, 3, 4] 391s x0: [2, 4, 3, 4] 391s 0.250 -0.915 391s 0.027 0.748 391s 0.178 0.640 391s -0.761 0.300 391s 391s x1: [2, 2, 3, 4] 391s -0.080 -0.309 391s -0.334 0.236 391s 391s y: [4, 2, 3, 4] 391s 0.262 -0.233 -0.212 -0.032 391s -0.299 0.167 0.091 0.325 391s test-mul-mat0: iter:133/500 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.439 -0.610 -0.220 391s 391s x1: [3, 4, 1, 1] 391s -0.390 0.736 0.010 391s -0.487 0.525 -0.823 391s -0.676 0.065 -0.745 391s -0.415 -0.108 0.612 391s 391s y: [1, 4, 1, 1] 391s -0.280 391s 0.074 391s 0.420 391s 0.113 391s testing: mul_mat, [3, 1, 3, 1] = [3, 3, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.776 0.230 -0.273 391s -0.627 -0.671 0.337 391s -0.890 0.339 0.851 391s 391s x1: [3, 1, 3, 1] 391s 0.316 0.199 -0.401 391s 391s y: [3, 1, 3, 1] 391s -0.090 -0.467 -0.554 391s testing: mul_mat, [2, 3, 3, 4] = [3, 2, 3, 4] * [3, 3, 3, 4] 391s x0: [3, 2, 3, 4] 391s -0.722 -0.961 -0.617 391s 0.828 0.335 -0.483 391s 391s x1: [3, 3, 3, 4] 391s 0.022 0.189 0.485 391s -0.866 0.413 -0.284 391s -0.139 0.786 0.045 391s 391s y: [2, 3, 3, 4] 391s -0.497 -0.153 391s 0.405 -0.442 391s -0.682 0.127 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.259 -0.616 0.223 391s 391s x1: [3, 2, 1, 1] 391s 0.396 0.936 0.558 391s 0.494 0.306 -0.448 391s 391s y: [1, 2, 1, 1] 391s -0.350 391s -0.160 391s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 3, 3, 1] 391s -0.592 391s -0.373 391s 0.514 391s 391s x1: [1, 3, 3, 1] 391s 0.365 391s -0.748 391s 0.473 391s 391s y: [3, 3, 3, 1] 391s -0.216 -0.136 0.188 391s 0.442 0.279 -0.384 391s -0.280 -0.177 0.243 391s testing: mul_mat, [2, 1, 3, 4] = [3, 2, 3, 4] * [3, 1, 3, 4] 391s x0: [3, 2, 3, 4] 391s -0.564 -0.962 -0.848 391s 0.802 -0.946 0.002 391s 391s x1: [3, 1, 3, 4] 391s 0.117 0.856 0.985 391s 391s y: [2, 1, 3, 4] 391s -1.724 -0.713 391s test-mul-mat0: iter:134/500 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.743 391s 391s x1: [1, 1, 1, 1] 391s -0.467 391s 391s y: [1, 1, 1, 1] 391s -0.347 391s testing: mul_mat, [4, 1, 4, 1] = [1, 4, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 4, 4, 1] 391s -0.097 391s -0.539 391s 0.561 391s -0.409 391s 391s x1: [1, 1, 4, 1] 391s -0.242 391s 391s y: [4, 1, 4, 1] 391s 0.024 0.131 -0.136 0.099 391s testing: mul_mat, [4, 4, 4, 3] = [1, 4, 4, 3] * [1, 4, 4, 3] 391s x0: [1, 4, 4, 3] 391s -0.479 391s -0.317 391s 0.554 391s -0.142 391s 391s x1: [1, 4, 4, 3] 391s -0.956 391s 0.598 391s 0.932 391s -0.303 391s 391s y: [4, 4, 4, 3] 391s 0.458 0.303 -0.529 0.136 391s -0.286 -0.189 0.331 -0.085 391s -0.446 -0.295 0.516 -0.132 391s 0.145 0.0ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 96 -0.168 0.043 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.041 391s 0.437 391s -0.249 391s 391s x1: [1, 4, 1, 1] 391s -0.685 391s 0.644 391s -0.910 391s -0.866 391s 391s y: [3, 4, 1, 1] 391s -0.028 -0.299 0.171 391s 0.026 0.281 -0.161 391s -0.037 -0.397 0.227 391s -0.035 -0.378 0.216 391s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.572 0.711 0.829 391s 0.431 0.698 0.773 391s -0.917 -0.514 0.228 391s 391s x1: [3, 1, 4, 1] 391s -0.366 -0.993 0.574 391s 391s y: [3, 1, 4, 1] 391s -0.020 -0.407 0.977 391s testing: mul_mat, [3, 3, 4, 3] = [3, 3, 4, 3] * [3, 3, 4, 3] 391s x0: [3, 3, 4, 3] 391s -0.934 -0.293 0.499 391s -0.652 -0.985 0.987 391s -0.029 -0.985 0.226 391s 391s x1: [3, 3, 4, 3] 391s 0.318 0.166 -0.610 391s -0.454 -0.961 -0.523 391s 0.419 0.001 -0.301 391s 391s y: [3, 3, 4, 3] 391s -0.651 -0.973 -0.311 391s 0.445 0.727 0.842 391s -0.542 -0.572 -0.081 391s test-mul-mat0: iter:135/500 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.267 0.658 391s 0.564 0.346 391s 391s x1: [2, 1, 1, 1] 391s 0.819 -0.818 391s 391s y: [2, 1, 1, 1] 391s -0.319 0.179 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.737 -0.136 391s 0.022 0.341 391s -0.516 -0.764 391s -0.039 -0.966 391s 391s x1: [2, 2, 1, 1] 391s 0.846 0.475 391s 0.756 0.786 391s 391s y: [4, 2, 1, 1] 391s -0.688 0.181 -0.800 -0.492 391s -0.664 0.285 -0.991 -0.789 391s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 391s x0: [2, 4, 1, 2] 391s 0.561 -0.865 391s -0.487 0.317 391s 0.921 -0.786 391s 0.580 -0.215 391s 391s x1: [2, 4, 1, 2] 391s 0.363 0.393 391s -0.629 0.033 391s -0.313 0.352 391s -0.335 -0.140 391s 391s y: [4, 4, 1, 2] 391s -0.136 -0.052 0.025 0.126 391s -0.382 0.317 -0.606 -0.372 391s -0.480 0.264 -0.565 -0.257 391s -0.067 0.119 -0.198 -0.164 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.102 0.610 391s 0.611 0.316 391s 391s x1: [2, 4, 1, 1] 391s 0.674 0.868 391s 0.962 0.026 391s -0.467 -0.178 391s -0.846 -0.019 391s 391s y: [2, 4, 1, 1] 391s 0.598 0.687 391s 0.114 0.596 391s -0.156 -0.342 391s -0.097 -0.523 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.806 -0.259 391s 0.956 -0.389 391s 0.375 -0.306 391s 0.773 -0.989 391s 391s x1: [2, 2, 1, 1] 391s 0.900 -0.352 391s -0.123 -0.965 391s 391s y: [4, 2, 1, 1] 391s -0.635 0.997 0.445 1.044 391s 0.349 0.258 0.249 0.860 391s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 391s x0: [4, 4, 1, 2] 391s -0.414 0.844 0.566 -0.876 391s 0.809 -0.809 0.430 -0.448 391s 0.030 0.657 -0.459 -0.741 391s 0.463 0.800 -0.823 -0.299 391s 391s x1: [4, 2, 1, 2] 391s -0.917 -0.424 0.314 -0.083 391s 0.851 -0.012 -0.216 0.812 391s 391s y: [4, 2, 1, 2] 391s 0.272 -0.227 -0.388 -0.997 391s -1.196 0.241 -0.485 0.319 391s test-mul-mat0: iter:136/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.479 391s 391s x1: [1, 4, 1, 1] 391s 0.372 391s -0.988 391s -0.062 391s -0.062 391s 391s y: [1, 4, 1, 1] 391s 0.178 391s -0.474 391s -0.030 391s -0.030 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.969 391s 0.374 391s -0.733 391s 0.897 391s 391s x1: [1, 1, 1, 1] 391s 0.115 391s 391s y: [4, 1, 1, 1] 391s -0.111 0.043 -0.084 0.103 391s testing: mul_mat, [4, 4, 1, 4] = [1, 4, 1, 4] * [1, 4, 1, 4] 391s x0: [1, 4, 1, 4] 391s -0.793 391s 0.279 391s -0.345 391s -0.781 391s 391s x1: [1, 4, 1, 4] 391s -0.450 391s -0.181 391s -0.250 391s -0.490 391s 391s y: [4, 4, 1, 4] 391s 0.357 -0.126 0.155 0.352 391s 0.144 -0.051 0.063 0.142 391s 0.198 -0.070 0.086 0.195 391s 0.388 -0.137 0.169 0.383 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.676 391s -0.354 391s -0.709 391s 391s x1: [1, 4, 1, 1] 391s -0.626 391s 0.141 391s 0.496 391s 0.992 391s 391s y: [3, 4, 1, 1] 391s -0.423 0.221 0.444 391s 0.096 -0.050 -0.100 391s 0.335 -0.175 -0.351 391s 0.671 -0.351 -0.703 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.458 0.279 0.010 391s 391s x1: [3, 1, 1, 1] 391s 0.627 0.606 0.138 391s 391s y: [1, 1, 1, 1] 391s 0.458 391s testing: mul_mat, [2, 3, 1, 4] = [1, 2, 1, 4] * [1, 3, 1, 4] 391s x0: [1, 2, 1, 4] 391s 0.823 391s -0.322 391s 391s x1: [1, 3, 1, 4] 391s 0.665 391s -0.441 391s 0.665 391s 391s y: [2, 3, 1, 4] 391s 0.548 -0.214 391s -0.364 0.142 391s 0.547 -0.214 391s test-mul-mat0: iter:137/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.985 0.783 0.451 391s -0.209 0.040 -0.004 391s 391s x1: [3, 2, 1, 1] 391s -0.557 -0.823 0.108 391s 0.901 0.457 -0.882 391s 391s y: [2, 2, 1, 1] 391s -0.046 0.083 391s -0.927 -0.167 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.730 -0.381 0.483 391s 0.549 -0.304 0.673 391s 391s x1: [3, 2, 1, 1] 391s 0.263 0.005 0.770 391s 0.462 0.052 -0.962 391s 391s y: [2, 2, 1, 1] 391s 0.562 0.661 391s -0.148 -0.410 391s testing: mul_mat, [4, 2, 1, 3] = [3, 4, 1, 3] * [3, 2, 1, 3] 391s x0: [3, 4, 1, 3] 391s -0.117 -0.364 -0.499 391s 0.935 -0.327 0.787 391s 0.665 0.292 0.270 391s 0.213 0.988 -0.057 391s 391s x1: [3, 2, 1, 3] 391s -0.703 -0.551 -0.491 391s 0.336 0.829 -0.048 391s 391s y: [4, 2, 1, 3] 391s 0.527 -0.864 -0.761 -0.667 391s -0.317 0.005 0.452 0.894 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.385 -0.287 0.027 391s -0.090 0.168 -0.736 391s -0.255 -0.579 0.968 391s 391s x1: [3, 4, 1, 1] 391s 0.417 0.457 0.014 391s -0.313 -0.329 0.002 391s 0.630 -0.819 0.439 391s 0.083 0.026 0.706 391s 391s y: [3, 4, 1, 1] 391s 0.029 0.029 -0.357 391s -0.026 -0.028 0.272 391s 0.489 -0.518 0.739 391s 0.043 -0.523 0.647 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.654 0.166 -0.309 391s -0.836 0.608 -0.777 391s 0.978 0.197 -0.689 391s 0.016 -0.395 -0.822 391s 391s x1: [3, 3, 1, 1] 391s 0.116 0.238 -0.620 391s -0.755 0.096 0.428 391s 0.875 -0.112 0.150 391s 391s y: [4, 3, 1, 1] 391s 0.155 0.529 0.588 0.417 391s 0.378 0.357 -1.015 -0.402 391s -0.637 -0.916 0.730 -0.065 391s testing: mul_mat, [1, 3, 1, 3] = [4, 1, 1, 3] * [4, 3, 1, 3] 391s x0: [4, 1, 1, 3] 391s 0.933 -0.222 0.908 -0.603 391s 391s x1: [4, 3, 1, 3] 391s 0.608 0.221 -0.077 -0.679 391s -0.611 0.344 0.347 -0.347 391s 0.312 -0.536 0.891 0.693 391s 391s y: [1, 3, 1, 3] 391s 0.858 391s -0.122 391s 0.802 391s test-mul-mat0: iter:138/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.654 0.485 391s -0.397 -0.068 391s 0.610 0.199 391s -0.135 -0.612 391s 391s x1: [2, 4, 1, 1] 391s -0.010 0.027 391s -0.127 0.424 391s 0.232 0.024 391s -0.919 0.748 391s 391s y: [4, 4, 1, 1] 391s 0.020 0.002 -0.001 -0.015 391s 0.288 0.021 0.007 -0.242 391s -0.140 -0.094 0.146 -0.046 391s 0.964 0.314 -0.412 -0.334 391s testing: mul_mat, [1, 4, 3, 1] = [2, 1, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.117 0.770 391s 391s x1: [2, 4, 3, 1] 391s 0.107 0.262 391s 0.057 -0.292 391s -0.633 -0.287 391s 0.429 -0.945 391s 391s y: [1, 4, 3, 1] 391s 0.189 391s -0.232 391s -0.147 391s -0.777 391s testing: mul_mat, [1, 1, 3, 4] = [2, 1, 3, 4] * [2, 1, 3, 4] 391s x0: [2, 1, 3, 4] 391s -0.036 0.850 391s 391s x1: [2, 1, 3, 4] 391s 0.189 0.897 391s 391s y: [1, 1, 3, 4] 391s 0.755 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.710 0.589 391s -0.375 -0.460 391s -0.310 -0.205 391s -0.933 -0.365 391s 391s x1: [2, 1, 1, 1] 391s 0.854 -0.456 391s 391s y: [4, 1, 1, 1] 391s -0.875 -0.111 -0.171 -0.631 391s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 3, 3, 1] 391s 0.390 0.930 -0.045 -0.638 391s 0.954 -0.252 0.610 -0.066 391s 0.294 0.929 -0.200 0.803 391s 391s x1: [4, 2, 3, 1] 391s 0.025 -0.139 -0.129 0.407 391s 0.324 0.003 -0.947 -0.103 391s 391s y: [3, 2, 3, 1] 391s -0.373 -0.047 0.231 391s 0.238 -0.262 0.205 391s testing: mul_mat, [2, 4, 3, 4] = [3, 2, 3, 4] * [3, 4, 3, 4] 391s x0: [3, 2, 3, 4] 391s 0.455 0.053 -0.939 391s 0.191 -0.327 -0.410 391s 391s x1: [3, 4, 3, 4] 391s 0.690 0.540 0.939 391s -0.510 0.902 -0.127 391s -0.707 -0.684 -0.296 391s -0.443 0.749 -0.642 391s 391s y: [2, 4, 3, 4] 391s -0.538 -0.430 391s -0.065 -0.340 391s -0.080 0.210 391s 0.440 -0.066 391s test-mul-mat0: iter:139/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.928 -0.217 -0.205 -0.499 391s -0.431 -0.544 0.294 0.582 391s 0.977 -0.104 -0.717 -0.645 391s 391s x1: [4, 3, 1, 1] 391s -0.974 0.790 0.851 0.838 391s 0.419 0.073 0.868 -0.707 391s 0.412 0.676 -0.480 0.549 391s 391s y: [3, 3, 1, 1] 391s -1.668 0.728 -2.184 391s 0.547 -0.376 0.235 391s 0.060 -0.367 0.323 391s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 1, 4, 1] 391s -0.921 0.304 -0.546 -0.621 391s 391s x1: [4, 3, 4, 1] 391s 0.129 0.314 0.429 0.165 391s 0.950 -0.288 0.191 0.740 391s -0.437 0.029 0.160 0.636 391s 391s y: [1, 3, 4, 1] 391s -0.360 391s -1.525 391s -0.072 391s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.927 -0.179 -0.410 -0.670 391s 0.955 0.992 -0.736 0.822 391s 391s x1: [4, 1, 4, 1] 391s -0.397 0.953 -0.300 -0.320 391s 391s y: [2, 1, 4, 1] 391s -0.201 0.524 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.260 -0.013 0.934 0.373 391s 0.272 0.578 -0.435 0.337 391s -0.349 0.492 0.103 0.462 391s 391s x1: [4, 2, 1, 1] 391s -0.813 0.590 0.487 -0.858 391s 0.582 0.751 0.964 0.569 391s 391s y: [3, 2, 1, 1] 391s 0.339 -0.381 0.228 391s 0.952 0.365 0.529 391s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 2, 4, 1] 391s 0.705 0.794 -0.697 391s -0.903 -0.762 -0.500 391s 391s x1: [3, 4, 4, 1] 391s -0.267 0.897 0.998 391s -0.509 -0.180 -0.885 391s 0.722 -0.814 0.115 391s -0.121 -0.627 -0.295 391s 391s y: [2, 4, 4, 1] 391s -0.171 -0.941 391s 0.116 1.039 391s -0.218 -0.089 391s -0.378 0.734 391s testing: mul_mat, [1, 3, 4, 1] = [2, 1, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 1, 4, 1] 391s 0.686 -0.773 391s 391s x1: [2, 3, 4, 1] 391s 0.612 -0.556 391s -0.993 -0.578 391s -0.732 -0.115 391s 391s y: [1, 3, 4, 1] 391s 0.850 391s -0.234 391s -0.413 391s test-mul-mat0: iter:140/500 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.954 -0.582 0.275 391s -0.540 0.245 0.907 391s -0.854 0.472 -0.737 391s -0.079 -0.563 -0.773 391s 391s x1: [3, 4, 1, 1] 391s 0.898 0.650 0.836 391s -0.219 0.861 0.121 391s -0.836 -0.665 -0.152 391s 0.319 -0.800 -0.468 391s 391s y: [4, 4, 1, 1] 391s -1.005 0.434 -1.076 -1.084 391s -0.259 0.439 0.504 -0.561 391s 1.143 0.151 0.512 0.558 391s 0.032 -0.793 -0.305 0.787 391s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.415 0.506 -0.056 391s 0.381 -0.458 -0.047 391s -0.658 0.913 0.810 391s 391s x1: [3, 4, 3, 1] 391s 0.308 0.847 -0.803 391s -0.883 0.524 0.052 391s -0.985 0.174 -0.112 391s -0.204 0.036 -0.991 391s 391s y: [3, 4, 3, 1] 391s 0.346 -0.232 -0.080 391s 0.629 -0.579 1.102 391s 0.503 -0.450 0.717 391s 0.159 -0.047 -0.636 391s testing: mul_mat, [2, 3, 3, 2] = [3, 2, 3, 2] * [3, 3, 3, 2] 391s x0: [3, 2, 3, 2] 391s 0.964 -0.510 0.577 391s -0.009 -0.694 -0.069 391s 391s x1: [3, 3, 3, 2] 391s -0.133 0.568 -0.315 391s -0.078 -0.847 -0.808 391s 0.866 0.534 -0.266 391s 391s y: [2, 3, 3, 2] 391s -0.599 -0.371 391s -0.109 0.645 391s 0.409 -0.360 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.450 0.548 0.219 391s -0.137 -0.763 0.568 391s 391s x1: [3, 2, 1, 1] 391s 0.232 0.417 -0.797 391s 0.084 0.855 -0.202 391s 391s y: [2, 2, 1, 1] 391s -0.050 -0.802 391s 0.386 -0.778 391s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.268 0.425 391s 391s x1: [2, 3, 3, 1] 391s 0.563 -0.506 391s -0.430 -0.343 391s -0.483 -0.198 391s 391s y: [1, 3, 3, 1] 391s -0.366 391s -0.030 391s 0.046 391s testing: mul_mat, [4, 2, 3, 2] = [1, 4, 3, 2] * [1, 2, 3, 2] 391s x0: [1, 4, 3, 2] 391s 0.440 391s -0.222 391s -0.399 391s 0.656 391s 391s x1: [1, 2, 3, 2] 391s 0.137 391s 0.085 391s 391s y: [4, 2, 3, 2] 391s 0.060 -0.030 -0.055 0.090 391s 0.038 -0.019 -0.034 0.056 391s test-mul-mat0: iter:141/500 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.500 -0.637 -0.512 391s -0.970 0.334 0.047 391s -0.217 0.914 -0.257 391s 391s x1: [3, 4, 1, 1] 391s -0.745 0.855 0.947 391s 0.695 -0.367 -0.452 391s 0.350 -0.490 -0.872 391s 0.275 -0.606 -0.011 391s 391s y: [3, 4, 1, 1] 391s -1.402 1.053 0.700 391s 0.812 -0.817 -0.370 391s 0.933 -0.544 -0.300 391s 0.529 -0.470 -0.611 391s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.577 -0.351 0.072 391s -0.802 -0.700 -0.114 391s 391s x1: [3, 3, 3, 1] 391s -0.379 0.651 0.814 391s -0.237 -0.886 0.822 391s 0.704 0.871 0.424 391s 391s y: [2, 3, 3, 1] 391s -0.389 -0.245 391s 0.234 0.716 391s 0.130 -1.222 391s testing: mul_mat, [1, 2, 3, 4] = [3, 1, 3, 4] * [3, 2, 3, 4] 391s x0: [3, 1, 3, 4] 391s -0.994 -0.409 0.414 391s 391s x1: [3, 2, 3, 4] 391s 0.279 0.238 0.470 391s 0.557 -0.499 -0.494 391s 391s y: [1, 2, 3, 4] 391s -0.180 391s -0.554 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.413 -0.696 -0.891 391s 0.256 -0.380 0.442 391s 0.027 -0.629 0.296 391s 0.642 0.090 0.131 391s 391s x1: [3, 1, 1, 1] 391s -0.634 -0.090 0.684 391s 391s y: [4, 1, 1, 1] 391s -0.285 0.174 0.242 -0.325 391s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.506 0.972 0.505 0.852 391s 391s x1: [4, 3, 3, 1] 391s -0.241 -0.925 -0.119 -0.273 391s 0.088 -0.095 0.017 0.900 391s -0.619 0.358 -0.282 -0.552 391s 391s y: [1, 3, 3, 1] 391s -1.069 391s 0.639 391s 0.048 391s testing: mul_mat, [2, 4, 3, 4] = [1, 2, 3, 4] * [1, 4, 3, 4] 391s x0: [1, 2, 3, 4] 391s -0.086 391s 0.544 391s 391s x1: [1, 4, 3, 4] 391s -0.107 391s -0.786 391s -0.590 391s -0.630 391s 391s y: [2, 4, 3, 4] 391s 0.009 -0.058 391s 0.068 -0.428 391s 0.051 -0.321 391s 0.054 -0.343 391s test-mul-mat0: iter:142/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.670 0.946 391s -0.992 -0.500 391s -0.468 0.525 391s 391s x1: [2, 2, 1, 1] 391s 0.259 0.415 391s 0.328 -0.827 391s 391s y: [3, 2, 1, 1] 391s 0.219 -0.464 0.097ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 391s -1.002 0.088 -0.588 391s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.439 -0.016 391s -0.274 -0.233 391s 391s x1: [2, 3, 3, 1] 391s 0.569 0.161 391s 0.955 0.214 391s -0.975 -0.810 391s 391s y: [2, 3, 3, 1] 391s 0.247 -0.193 391s 0.416 -0.311 391s -0.415 0.456 391s testing: mul_mat, [2, 2, 3, 2] = [2, 2, 3, 2] * [2, 2, 3, 2] 391s x0: [2, 2, 3, 2] 391s -0.242 -0.387 391s 0.496 -0.645 391s 391s x1: [2, 2, 3, 2] 391s 0.166 -0.205 391s 0.174 -0.620 391s 391s y: [2, 2, 3, 2] 391s 0.039 0.215 391s 0.198 0.486 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.713 -0.674 391s -0.101 -0.606 391s -0.202 0.153 391s 391s x1: [2, 2, 1, 1] 391s -0.021 -0.622 391s -0.798 0.096 391s 391s y: [3, 2, 1, 1] 391s 0.405 0.379 -0.091 391s -0.633 0.023 0.176 391s testing: mul_mat, [1, 2, 3, 1] = [3, 1, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.369 0.801 -0.216 391s 391s x1: [3, 2, 3, 1] 391s -0.427 0.109 -0.327 391s 0.615 -0.002 -0.426 391s 391s y: [1, 2, 3, 1] 391s 0.316 391s -0.136 391s testing: mul_mat, [4, 3, 3, 2] = [1, 4, 3, 2] * [1, 3, 3, 2] 391s x0: [1, 4, 3, 2] 391s 0.100 391s -0.701 391s -0.540 391s 0.731 391s 391s x1: [1, 3, 3, 2] 391s -0.128 391s -0.810 391s -0.226 391s 391s y: [4, 3, 3, 2] 391s -0.013 0.090 0.069 -0.094 391s -0.081 0.568 0.438 -0.593 391s -0.023 0.159 0.122 -0.166 391s test-mul-mat0: iter:143/500 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.641 0.896 0.945 391s 0.340 -0.643 0.372 391s 0.619 -0.735 0.859 391s 391s x1: [3, 4, 1, 1] 391s 0.455 -0.825 -0.561 391s -0.445 -0.526 -0.101 391s -0.713 -0.427 -0.858 391s -0.149 0.568 -0.791 391s 391s y: [3, 4, 1, 1] 391s -1.561 0.476 0.407 391s -0.282 0.150 0.025 391s -0.736 -0.287 -0.864 391s -0.143 -0.710 -1.189 391s testing: mul_mat, [3, 3, 3, 1] = [3, 3, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 3, 3, 1] 391s 0.452 -0.269 -0.257 391s 0.647 -0.075 0.624 391s -0.033 0.572 0.761 391s 391s x1: [3, 3, 3, 1] 391s -0.804 -0.806 -0.119 391s 0.319 -0.354 -0.863 391s -0.543 0.344 -0.478 391s 391s y: [3, 3, 3, 1] 391s -0.116 -0.535 -0.524 391s 0.461 -0.306 -0.870 391s -0.215 -0.676 -0.149 391s testing: mul_mat, [4, 3, 3, 2] = [3, 4, 3, 2] * [3, 3, 3, 2] 391s x0: [3, 4, 3, 2] 391s -0.257 0.781 -0.765 391s -0.940 -0.684 -0.135 391s -0.197 -0.906 -0.055 391s 0.440 -0.998 0.926 391s 391s x1: [3, 3, 3, 2] 391s 0.244 0.450 0.893 391s -0.871 -0.099 -0.376 391s -0.128 -0.451 0.549 391s 391s y: [4, 3, 3, 2] 391s -0.395 -0.658 -0.504 0.486 391s 0.434 0.937 0.281 -0.633 391s -0.740 0.355 0.404 0.902 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.269 0.837 0.507 391s -0.553 -0.622 0.190 391s -0.744 -0.370 -0.949 391s 0.917 0.627 0.703 391s 391s x1: [3, 4, 1, 1] 391s 0.236 -0.260 0.385 391s -0.724 -0.433 -0.094 391s 0.371 -0.105 0.867 391s -0.947 0.709 -0.765 391s 391s y: [4, 4, 1, 1] 391s 0.042 0.104 -0.445 0.325 391s -0.606 0.652 0.789 -1.002 391s 0.451 0.025 -1.059 0.883 391s -0.050 -0.062 1.169 -0.962 391s testing: mul_mat, [3, 3, 3, 1] = [4, 3, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.812 -0.410 -0.638 -0.813 391s -0.096 -0.444 0.102 0.356 391s 0.506 -0.060 -0.759 0.776 391s 391s x1: [4, 3, 3, 1] 391s 0.911 0.976 0.768 -0.110 391s -0.655 0.143 -0.874 0.085 391s -0.472 -0.598 0.652 0.434 391s 391s y: [3, 3, 3, 1] 391s -1.541 -0.483 -0.267 391s 0.961 -0.059 0.390 391s -0.140 0.532 -0.360 391s testing: mul_mat, [1, 4, 3, 2] = [3, 1, 3, 2] * [3, 4, 3, 2] 391s x0: [3, 1, 3, 2] 391s -0.485 0.236 -0.782 391s 391s x1: [3, 4, 3, 2] 391s 0.425 -0.971 -0.623 391s 0.666 -0.784 0.733 391s 0.442 -0.375 -0.386 391s 0.201 0.511 -0.064 391s 391s y: [1, 4, 3, 2] 391s 0.052 391s -1.081 391s -0.001 391s 0.073 391s test-mul-mat0: iter:144/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.555 -0.835 -0.585 -0.370 391s -0.232 -0.915 -0.222 -0.147 391s -0.626 0.031 -0.074 0.059 391s -0.752 0.978 0.584 -0.434 391s 391s x1: [4, 3, 1, 1] 391s 0.931 0.928 -0.159 0.790 391s -0.067 -0.843 -0.491 -0.337 391s 0.298 0.025 0.899 0.516 391s 391s y: [4, 3, 1, 1] 391s -1.491 -1.146 -0.496 -0.228 391s 1.153 0.945 0.033 -0.914 391s -0.903 -0.367 -0.223 0.100 391s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.421 -0.778 -0.650 -0.426 391s -0.670 0.477 -0.480 -0.844 391s 391s x1: [4, 4, 4, 1] 391s -0.573 0.422 -0.503 -0.645 391s -0.737 -0.713 0.288 -0.580 391s -0.204 0.951 0.718 0.821 391s 0.849 0.234 -0.513 -0.706 391s 391s y: [2, 4, 4, 1] 391s 0.514 1.371 391s 0.925 0.505 391s -1.471 -0.448 391s 0.094 0.385 391s testing: mul_mat, [4, 2, 4, 4] = [4, 4, 4, 4] * [4, 2, 4, 4] 391s x0: [4, 4, 4, 4] 391s -0.200 0.394 0.239 -0.144 391s 0.304 ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s -0.730 -0.901 0.207 391s -0.671 -0.871 -0.520 -0.406 391s 0.425 0.315 -0.715 -0.078 391s 391s x1: [4, 2, 4, 4] 391s 0.177 -0.645 -0.907 0.506 391s 0.832 -0.387 0.662 -0.742 391s 391s y: [4, 2, 4, 4] 391s -0.579 1.447 0.709 0.481 391s -0.053 -0.215 -0.264 -0.184 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.408 -0.425 -0.870 -0.159 391s -0.308 0.425 0.383 0.833 391s -0.521 0.941 0.641 -0.434 391s -0.195 0.842 0.430 0.206 391s 391s x1: [4, 4, 1, 1] 391s 0.135 -0.635 -0.031 0.230 391s 0.844 0.909 -0.646 -0.807 391s 0.885 0.005 -0.632 -0.042 391s 0.066 -0.129 0.419 0.575 391s 391s y: [4, 4, 1, 1] 391s 0.316 -0.132 -0.788 -0.527 391s 0.648 -0.793 0.352 0.156 391s 0.915 -0.548 -0.844 -0.450 391s -0.373 0.564 -0.137 0.177 391s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.367 -0.425 0.724 0.893 391s -0.177 0.853 0.036 -0.153 391s 391s x1: [4, 4, 4, 1] 391s -0.199 0.796 0.050 -0.289 391s -0.850 0.243 -0.404 0.155 391s 0.611 0.554 -0.779 -0.518 391s -0.028 0.796 0.771 -0.620 391s 391s y: [2, 4, 4, 1] 391s -0.633 0.760 391s -0.570 0.320 391s -1.039 0.415 391s -0.344 0.806 391s testing: mul_mat, [3, 4, 4, 4] = [2, 3, 4, 4] * [2, 4, 4, 4] 391s x0: [2, 3, 4, 4] 391s -0.260 0.620 391s -0.918 0.848 391s -0.102 -0.214 391s 391s x1: [2, 4, 4, 4] 391s 0.788 -0.227 391s 0.923 0.512 391s 0.809 0.816 391s -0.641 0.787 391s 391s y: [3, 4, 4, 4] 391s -0.346 -0.916 -0.032 391s 0.077 -0.414 -0.204 391s 0.296 -0.051 -0.257 391s 0.655 1.256 -0.103 391s test-mul-mat0: iter:145/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.420 -0.973 -0.403 391s -0.725 -0.230 0.558 391s -0.112 0.576 -0.102 391s 0.169 -0.156 0.924 391s 391s x1: [3, 1, 1, 1] 391s -0.225 -0.896 -0.931 391s 391s y: [4, 1, 1, 1] 391s 1.342 -0.151 -0.396 -0.759 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.710 -0.488 -0.281 391s 0.814 -0.419 0.797 391s -0.606 -0.393 -0.605 391s 391s x1: [3, 4, 1, 1] 391s 0.103 -0.391 0.038 391s -0.411 -0.873 -0.681 391s -0.631 0.411 -0.414 391s 0.263 0.373 -0.723 391s 391s y: [3, 4, 1, 1] 391s 0.253 0.279 0.068 391s 0.326 -0.511 1.004 391s -0.532 -1.017 0.472 391s 0.208 -0.518 0.131 391s testing: mul_mat, [3, 3, 1, 2] = [3, 3, 1, 2] * [3, 3, 1, 2] 391s x0: [3, 3, 1, 2] 391s 0.317 -0.004 0.994 391s 0.027 0.508 -0.287 391s -0.158 -0.912 -0.490 391s 391s x1: [3, 3, 1, 2] 391s -0.331 0.377 0.953 391s 0.557 -0.047 -0.149 391s -0.273 0.797 -0.224 391s 391s y: [3, 3, 1, 2] 391s 0.841 -0.091 -0.758 391s 0.029 0.034 0.027 391s -0.313 0.462 -0.574 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.706 -0.853 0.983 391s 0.151 0.271 -0.108 391s 0.276 -0.950 0.743 391s 0.023 0.654 -0.506 391s 391s x1: [3, 3, 1, 1] 391s -0.811 -0.177 -0.515 391s -0.982 0.230 -0.701 391s -0.563 0.762 -0.568 391s 391s y: [4, 3, 1, 1] 391s -0.928 -0.114 -0.439 0.127 391s -1.578 -0.010 -1.010 0.483 391s -1.606 0.183 -1.301 0.773 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.590 -0.862 0.184 0.058 391s -0.249 0.352 0.364 -0.666 391s 391s x1: [4, 3, 1, 1] 391s -0.873 -0.562 0.399 -0.968 391s -0.489 -0.743 0.494 0.537 391s 0.967 0.683 -0.640 -0.548 391s 391s y: [2, 3, 1, 1] 391s -0.014 0.809 391s 0.473 -0.318 391s -0.167 0.132 391s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 391s x0: [2, 4, 1, 2] 391s -0.636 -0.538 391s -0.990 -0.941 391s -0.642 -0.686 391s -0.276 0.213 391s 391s x1: [2, 4, 1, 2] 391s -0.360 -0.919 391s -0.519 0.910 391s -0.869 -0.865 391s 0.894 0.023 391s 391s y: [4, 4, 1, 2] 391s 0.724 1.221 0.861 -0.097 391s -0.160 -0.342 -0.291 0.337 391s 1.018 1.674 1.151 0.055 391s -0.580 -0.906 -0.589 -0.241 391s test-mul-mat0: iter:146/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.710 -0.631 -0.777 391s -0.034 -0.267 -0.767 391s 391s x1: [3, 2, 1, 1] 391s -0.995 0.959 0.225 391s -0.607 0.108 0.542 391s 391s y: [2, 2, 1, 1] 391s -0.074 -0.395 391s -0.058 -0.423 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.438 -0.695 0.143 391s -0.837 0.036 0.344 391s 391s x1: [3, 2, 1, 1] 391s 0.324 0.457 -0.305 391s 0.384 0.772 0.908 391s 391s y: [2, 2, 1, 1] 391s -0.219 -0.360 391s -0.238 0.020 391s testing: mul_mat, [1, 2, 1, 3] = [3, 1, 1, 3] * [3, 2, 1, 3] 391s x0: [3, 1, 1, 3] 391s -0.079 0.186 -0.065 391s 391s x1: [3, 2, 1, 3] 391s -0.032 -0.752 0.456 391s 0.904 0.596 0.461 391s 391s y: [1, 2, 1, 3] 391s -0.167 391s 0.009 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.316 -0.720 0.101 391s -0.540 0.922 ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s 0.776 391s 0.166 0.029 -1.000 391s 391s x1: [3, 1, 1, 1] 391s -0.706 -0.564 0.004 391s 391s y: [3, 1, 1, 1] 391s 0.629 -0.136 -0.138 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.744 0.273 -0.509 391s 0.975 -0.162 0.404 391s 391s x1: [3, 3, 1, 1] 391s 0.464 0.810 0.261 391s -0.805 -0.932 0.488 391s 0.659 -0.442 0.040 391s 391s y: [2, 3, 1, 1] 391s 0.433 0.426 391s -1.102 -0.436 391s 0.349 0.730 391s testing: mul_mat, [4, 3, 1, 3] = [2, 4, 1, 3] * [2, 3, 1, 3] 391s x0: [2, 4, 1, 3] 391s 0.191 -0.054 391s 0.906 -0.365 391s -0.945 0.492 391s 0.748 -0.080 391s 391s x1: [2, 3, 1, 3] 391s -0.763 -0.849 391s 0.699 -0.327 391s 0.155 -0.039 391s 391s y: [4, 3, 1, 3] 391s -0.099 -0.382 0.303 -0.503 391s 0.151 0.752 -0.821 0.549 391s 0.032 0.155 -0.166 0.119 391s test-mul-mat0: iter:147/500 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.682 -0.740 -0.265 391s -0.398 -0.832 0.064 391s -0.986 0.481 0.631 391s 391s x1: [3, 2, 1, 1] 391s -0.696 -0.170 0.570 391s 0.494 -0.264 0.625 391s 391s y: [3, 2, 1, 1] 391s -0.500 0.455 0.965 391s 0.367 0.063 -0.220 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.818 -0.327 0.673 391s -0.770 0.846 0.977 391s 391s x1: [3, 3, 1, 1] 391s 0.576 -0.508 -0.129 391s -0.707 -0.536 0.467 391s -0.376 -0.437 0.701 391s 391s y: [2, 3, 1, 1] 391s 0.550 -1.000 391s -0.089 0.547 391s 0.308 0.604 391s testing: mul_mat, [4, 2, 1, 3] = [3, 4, 1, 3] * [3, 2, 1, 3] 391s x0: [3, 4, 1, 3] 391s 0.906 -0.518 -0.414 391s -0.531 -0.817 0.078 391s -0.712 -0.144 -0.248 391s -0.482 -0.297 -0.271 391s 391s x1: [3, 2, 1, 3] 391s 0.060 0.416 0.471 391s 0.796 0.041 -0.286 391s 391s y: [4, 2, 1, 3] 391s -0.357 -0.335 -0.219 -0.280 391s 0.818 -0.478 -0.502 -0.319 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.305 -0.765 0.257 391s -0.998 0.933 -0.186 391s 0.275 -0.848 0.173 391s 391s x1: [3, 4, 1, 1] 391s -0.683 -0.013 0.558 391s 0.069 0.505 -0.739 391s 0.797 0.082 -0.620 391s 0.998 -0.545 0.540 391s 391s y: [3, 4, 1, 1] 391s -0.055 0.566 -0.080 391s -0.556 0.539 -0.537 391s 0.021 -0.604 0.042 391s 0.861 -1.605 0.831 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.056 -0.440 0.642 391s 0.989 -0.750 -0.370 391s 0.226 0.023 0.021 391s 391s x1: [3, 3, 1, 1] 391s 0.949 0.138 0.210 391s -0.398 0.369 0.729 391s 0.840 -0.932 -0.569 391s 391s y: [3, 3, 1, 1] 391s 0.127 0.757 0.222 391s 0.284 -0.940 -0.066 391s 0.091 1.740 0.156 391s testing: mul_mat, [4, 3, 1, 3] = [3, 4, 1, 3] * [3, 3, 1, 3] 391s x0: [3, 4, 1, 3] 391s -0.247 -0.056 0.854 391s -0.227 -0.354 0.890 391s -0.287 -0.747 0.221 391s -0.149 0.116 0.215 391s 391s x1: [3, 3, 1, 3] 391s -0.903 -0.830 0.933 391s 0.402 -0.828 0.209 391s 0.637 -0.895 0.361 391s 391s y: [4, 3, 1, 3] 391s 1.066 1.329 1.085 0.239 391s 0.125 0.388 0.549 -0.111 391s 0.201 0.494 0.565 -0.121 391s test-mul-mat0: iter:148/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.101 -0.663 -0.058 391s 0.299 0.831 -0.985 391s -0.075 -0.203 -0.628 391s -0.955 0.666 -0.827 391s 391s x1: [3, 1, 1, 1] 391s -0.471 -0.040 -0.412 391s 391s y: [4, 1, 1, 1] 391s 0.003 0.232 0.302 0.763 391s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.658 -0.881 -0.768 391s 0.704 0.785 -0.596 391s 0.420 -0.786 0.046 391s 0.918 0.547 0.847 391s 391s x1: [3, 4, 2, 1] 391s 0.716 -0.572 -0.359 391s -0.501 0.334 -0.198 391s -0.966 0.013 0.675 391s -0.580 0.635 0.193 391s 391s y: [4, 4, 2, 1] 391s 1.251 0.269 0.734 0.041 391s -0.472 0.028 -0.482 -0.446 391s -1.165 -1.072 -0.385 -0.309 391s -1.090 -0.024 -0.734 -0.021 391s testing: mul_mat, [4, 4, 2, 2] = [3, 4, 2, 2] * [3, 4, 2, 2] 391s x0: [3, 4, 2, 2] 391s 0.459 -0.933 -0.412 391s -0.863 0.164 -0.738 391s 0.530 0.693 0.504 391s -0.795 0.242 -0.349 391s 391s x1: [3, 4, 2, 2] 391s 0.924 0.338 0.929 391s 0.768 -0.929 0.252 391s 0.628 0.730 0.371 391s 0.859 0.433 0.157 391s 391s y: [4, 4, 2, 2] 391s -0.274 -1.428 1.192 -0.977 391s 1.114 -1.002 -0.109 -0.924 391s -0.546 -0.696 1.026 -0.452 391s -0.075 -0.787 0.835 -0.633 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.714 0.484 -0.027 391s -0.471 -0.262 -0.675 391s 391s x1: [3, 4, 1, 1] 391s -0.735 -0.497 0.272 391s -0.760 -0.183 -0.173 391s 0.934 0.952 0.373 391s -0.801 0.278 -0.778 391s 391s y: [2, 4, 1, 1] 391s -0.773 0.292 391s -0.626 0.523 391s 1.118 -0.941 391s -0.416 0.829 391s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.021 0.669 391s -0.158 0.267 391s 0.333 0.748 391s 0.796 0.042 391s 391s x1: [2, 3, 2, 1] 391s -0.977 -0.053 391s -0.142 -0.187 391s 0.127 -0.367 391s 391s y: [4, 3, 2, 1] 391s -0.015 0.140 -0.365 -0.780 391s -0.122 -0.028 -0.187 -0.121 391s -0.248 -0.118 -0.232 0.085 391s testing: mul_mat, [2, 2, 2, 2] = [4, 2, 2, 2] * [4, 2, 2, 2] 391s x0: [4, 2, 2, 2] 391s 0.207 0.614 0.199 -0.912 391s -0.642 0.521 -0.812 -0.338 391s 391s x1: [4, 2, 2, 2] 391s -0.233 0.084 0.501 0.714 391s 0.942 -0.686 -0.159 -0.425 391s 391s y: [2, 2, 2, 2] 391s -0.549 -0.454 391s 0.130 -0.689 391s test-mul-mat0: iter:149/500 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.070 0.738 -0.465 391s -0.493 0.924 -0.520 391s 391s x1: [3, 4, 1, 1] 391s 0.329 -0.061 -0.315 391s 0.216 -0.578 0.286 391s -0.630 -0.606 -0.194 391s 0.633 -0.813 0.040 391s 391s y: [2, 4, 1, 1] 391s 0.124 -0.055 391s -0.544 -0.789 391s -0.401 -0.148 391s -0.574 -1.084 391s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 4, 3, 1] 391s -0.605 -0.610 -0.155 391s -0.417 0.430 0.519 391s 0.652 0.168 -0.946 391s -0.841 0.092 -0.466 391s 391s x1: [3, 2, 3, 1] 391s 0.774 -0.661 -0.698 391s 0.629 -0.539 -0.232 391s 391s y: [4, 2, 3, 1] 391s 0.043 -0.970 1.054 -0.386 391s -0.016 -0.615 0.539 -0.470 391s testing: mul_mat, [2, 4, 3, 2] = [3, 2, 3, 2] * [3, 4, 3, 2] 391s x0: [3, 2, 3, 2] 391s 0.122 -0.171 0.402 391s 0.142 -0.201 0.370 391s 391s x1: [3, 4, 3, 2] 391s 0.061 -0.465 0.681 391s 0.115 -0.306 -0.227 391s 0.649 -0.373 -0.796 391s -0.515 -0.811 -0.904 391s 391s y: [2, 4, 3, 2] 391s 0.361 0.354 391s -0.025 -0.006 391s -0.178 -0.128 391s -0.288 -0.244 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.779 0.779 -0.162 391s 0.421 0.439 0.213 391s 391s x1: [3, 2, 1, 1] 391s 0.577 -0.028 -0.989 391s 0.479 -0.641 -0.863 391s 391s y: [2, 2, 1, 1] 391s 0.588 0.020 391s 0.013 -0.264 391s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.027 0.115 391s 0.858 0.332 391s 391s x1: [2, 3, 3, 1] 391s -0.569 0.878 391s 0.614 -0.621 391s 0.709 0.448 391s 391s y: [2, 3, 3, 1] 391s 0.116 -0.197 391s -0.088 0.321 391s 0.032 0.757 391s testing: mul_mat, [1, 2, 3, 2] = [2, 1, 3, 2] * [2, 2, 3, 2] 391s x0: [2, 1, 3, 2] 391s 0.545 -0.881 391s 391s x1: [2, 2, 3, 2] 391s -0.019 -0.868 391s 0.686 0.360 391s 391s y: [1, 2, 3, 2] 391s 0.755 391s 0.057 391s test-mul-mat0: iter:150/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.193 -0.441 391s 0.620 -0.250 391s 0.661 -0.671 391s 391s x1: [2, 2, 1, 1] 391s -0.699 0.765 391s 0.271 -0.570 391s 391s y: [3, 2, 1, 1] 391s -0.472 -0.625 -0.976 391s 0.304 0.311 0.562 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.737 0.524 391s -0.537 -0.186 391s 391s x1: [2, 3, 1, 1] 391s 0.422 -0.905 391s 0.032 0.173 391s 0.641 0.151 391s 391s y: [2, 3, 1, 1] 391s -0.163 -0.058 391s 0.114 -0.049 391s 0.551 -0.372 391s testing: mul_mat, [2, 2, 1, 2] = [2, 2, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 2, 1, 2] 391s 0.136 0.553 391s 0.330 0.565 391s 391s x1: [2, 2, 1, 2] 391s -0.972 -0.171 391s 0.251 -0.627 391s 391s y: [2, 2, 1, 2] 391s -0.226 -0.418 391s -0.312 -0.271 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.303 0.651 391s 0.868 0.258 391s 0.522 0.977 391s -0.546 -0.886 391s 391s x1: [2, 2, 1, 1] 391s 0.273 0.785 391s -0.729 -0.305 391s 391s y: [4, 2, 1, 1] 391s 0.594 0.440 0.910 -0.845 391s -0.419 -0.711 -0.678 0.668 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.968 -0.388 0.914 0.885 391s -0.651 0.362 -0.066 -0.962 391s 391s x1: [4, 2, 1, 1] 391s -0.928 -0.995 -0.057 0.323 391s -0.622 0.272 0.973 0.213 391s 391s y: [2, 2, 1, 1] 391s 1.518 -0.063 391s 1.575 0.234 391s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 391s x0: [2, 4, 1, 2] 391s 0.879 -0.521 391s 0.266 -0.240 391s 0.701 0.884 391s 0.852 -0.171 391s 391s x1: [2, 4, 1, 2] 391s -0.123 0.158 391s 0.824 0.148 391s 0.853 0.704 391s -0.549 0.721 391s 391s y: [4, 4, 1, 2] 391s -0.191 -0.071 0.053 -0.132 391s 0.647 0.184 0.708 0.676 391s 0.383 0.058 1.221 0.606 391s -0.858 -0.319 0.253 -0.591 391s test-mul-mat0: iter:151/500 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.236 391s 391s x1: [1, 3, 1, 1] 391s -0.218 391s 0.864 391s 0.535 391s 391s y: [1, 3, 1, 1] 391s -0.051 391s 0.204 391s 0.126 391s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 1, 4, 1] 391s 0.151 391s 391s x1: [1, 1, 4, 1] 391s -0.981 391s 391s y: [1, 1, 4, 1] 391s -0.148 391s testing: mul_mat, [3, 1, 4, 1] = [1, 3, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 3, 4, 1] 391s -0.016 391s 0.437 391s -0.226 391s 391s x1: [1, 1, 4, 1] 391s -0.694 391s 391s y: [3, 1, 4, 1] 391s 0.011 -0.303 0.157 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.900 391s -0.978 391s 0.843 391s -0.178 391s 391s x1: [1, 3, 1, 1] 391s 0.586 391s -0.136 391s -0.997 391s 391s y: [4, 3, 1, 1] 391s -0.527 -0.573 0.494 -0.104 391s 0.123 0.133 -0.115 0.024 391s 0.897 0.974 -0.840 0.178 391s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.182 -0.589 0.748 0.986 391s 391s x1: [4, 1, 4, 1] 391s -0.403 0.301 0.973 0.052 391s 391s y: [1, 1, 4, 1] 391s 0.528 391s testing: mul_mat, [1, 4, 4, 1] = [1, 1, 4, 1] * [1, 4, 4, 1] 391s x0: [1, 1, 4, 1] 391s 0.870 391s 391s x1: [1, 4, 4, 1] 391s 0.169 391s 0.892 391s -0.987 391s 0.994 391s 391s y: [1, 4, 4, 1] 391s 0.147 391s 0.776 391s -0.858 391s 0.864 391s test-mul-mat0: iter:152/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.469 -0.924 -0.401 391s 0.330 -0.424 -0.552 391s 0.114 0.109 0.007 391s 391s x1: [3, 3, 1, 1] 391s 0.810 -0.450 -0.168 391s 0.007 0.389 -0.179 391s -0.824 0.281 -0.165 391s 391s y: [3, 3, 1, 1] 391s 0.104 0.551 0.042 391s -0.291 -0.065 0.042 391s 0.193 -0.300 -0.065 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.495 0.918 -0.393 391s -0.956 -0.250 0.613 391s 0.434 0.572 0.789 391s 391s x1: [3, 3, 1, 1] 391s -0.548 0.231 -0.252 391s -0.799 0.101 -0.525 391s 0.361 0.188 0.774 391s 391s y: [3, 3, 1, 1] 391s 0.040 0.312 -0.304 391s -0.096 0.417 -0.703 391s 0.046 0.082 0.874 391s testing: mul_mat, [3, 3, 1, 3] = [3, 3, 1, 3] * [3, 3, 1, 3] 391s x0: [3, 3, 1, 3] 391s -0.342 0.250 -0.491 391s 0.373 0.656 -0.532 391s -0.381 -0.861 0.026 391s 391s x1: [3, 3, 1, 3] 391s -0.285 -0.594 0.959 391s 0.246 -0.518 -0.442 391s -0.424 0.058 0.006 391s 391s y: [3, 3, 1, 3] 391s -0.521 -1.006 0.645 391s 0.004 -0.013 0.341 391s 0.157 -0.124 0.112 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.999 -0.294 0.401 391s 391s x1: [3, 3, 1, 1] 391s -0.321 0.349 0.661 391s 0.399 -0.993 -0.089 391s 0.908 0.380 -0.432 391s 391s y: [1, 3, 1, 1] 391s -0.158 391s 0.654 391s 0.622 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.266 391s 0.479 391s 0.102 391s 391s x1: [1, 3, 1, 1] 391s 0.193 391s 0.903 391s 0.433 391s 391s y: [3, 3, 1, 1] 391s 0.051 0.092 0.020 391s 0.240 0.432 0.092 391s 0.115 0.207 0.044 391s testing: mul_mat, [1, 1, 1, 3] = [3, 1, 1, 3] * [3, 1, 1, 3] 391s x0: [3, 1, 1, 3] 391s 0.367 0.793 0.704 391s 391s x1: [3, 1, 1, 3] 391s -0.973 -0.714 -0.530 391s 391s y: [1, 1, 1, 3] 391s -1.297 391s test-mul-mat0: iter:153/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.475 0.925 0.788 -0.411 391s -0.720 0.156 -0.618 0.984 391s -0.816 0.574 0.695 -0.876 391s 391s x1: [4, 3, 1, 1] 391s -0.123 -0.717 0.117 -0.690 391s -0.640 -0.617 0.789 0.462 391s -0.591 -0.925 0.933 0.695 391s 391s y: [3, 3, 1, 1] 391s -0.346 -0.774 0.374 391s -0.443 0.332 0.311 391s -0.687 0.389 -0.009 391s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.054 0.969 -0.780 -0.553 391s 391s x1: [4, 3, 4, 1] 391s -0.326 -0.213 -0.368 -0.850 391s 0.878 0.846 0.027 -0.839 391s -0.038 0.337 -0.479 0.345 391s 391s y: [1, 3, 4, 1] 391s 0.533 391s 1.310 391s 0.508 391s testing: mul_mat, [2, 1, 4, 2] = [4, 2, 4, 2] * [4, 1, 4, 2] 391s x0: [4, 2, 4, 2] 391s -0.473 0.394 0.102 -0.764 391s 0.919 -0.095 0.589 -0.306 391s 391s x1: [4, 1, 4, 2] 391s 0.616 -0.036 0.412 0.402 391s 391s y: [2, 1, 4, 2] 391s -0.570 0.689 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.200 0.800 -0.458 0.009 391s -0.081 0.708 -0.602 0.604 391s 391s x1: [4, 2, 1, 1] 391s -0.431 0.507 0.085 -0.740 391s 0.107 -0.354 0.322 0.046 391s 391s y: [2, 2, 1, 1] 391s 0.446 -0.104 391s -0.452 -0.426 391s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.700 0.405 391s -0.994 -0.544 391s 0.366 0.213 391s 391s x1: [2, 4, 4, 1] 391s 0.640 0.977 391s 0.309 0.978 391s -0.869 0.705 391s -0.550 0.847 391s 391s y: [3, 4, 4, 1] 391s 0.844 -1.168 0.442 391s 0.613 -0.840 0.321 391s -0.323 0.480 -0.168 391s -0.042 0.086 -0.021 391s testing: mul_mat, [2, 2, 4, 2] = [3, 2, 4, 2] * [3, 2, 4, 2] 391s x0: [3, 2, 4, 2] 391s 0.196 -0.182 0.989 391s -0.948 -0.749 0.143 391s 391s x1: [3, 2, 4, 2] 391s 0.905 -0.901 0.715 391s 0.163 -0.510 0.743 391s 391s y: [2, 2, 4, 2] 391s 1.049 -0.081 391s 0.860 0.334 391s test-mul-mat0: iter:154/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.774 -0.590 -0.865 391s -0.628 0.742 -0.560 391s 391s x1: [3, 2, 1, 1] 391s 0.255 0.196 0.341 391s 0.395 -0.628 0.131 391s 391s y: [2, 2, 1, 1] 391s -0.213 -0.206 391s 0.562 -0.787 391s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 4, 3, 1] 391s -0.233 0.763 0.661 391s 0.140 -0.106 -0.691 391s -0.086 0.304 -0.557 391s 0.285 0.046 -0.117 391s 391s x1: [3, 2, 3, 1] 391s 0.221 0.062 0.888 391s -0.475 0.424 0.873 391s 391s y: [4, 2, 3, 1] 391s 0.583 -0.590 -0.495 -0.038 391s 1.011 -0.715 -0.316 -0.218 391s testing: mul_mat, [1, 4, 3, 4] = [3, 1, 3, 4] * [3, 4, 3, 4] 391s x0: [3, 1, 3, 4] 391s 0.953 -0.168 0.108 391s 391s x1: [3, 4, 3, 4] 391s 0.713 -0.893 0.485 391s -0.844 0.392 -0.469 391s 0.039 0.899 -0.361 391s 0.811 -0.069 -0.829 391s 391s y: [1, 4, 3, 4] 391s 0.882 391s -0.921 391s -0.153 391s 0.695 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.315 -0.363 0.623 391s -0.004 0.235 -0.341 391s 391s x1: [3, 1, 1, 1] 391s 0.193 0.916 0.471 391s 391s y: [2, 1, 1, 1] 391s -0.100 0.054 391s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.830 -0.145 391s 0.384 0.577 391s 0.247 -0.229 391s -0.914 0.851 391s 391s x1: [2, 3, 3, 1] 391s -0.721 -0.644 391s 0.762 -0.678 391s 0.405 0.024 391s 391s y: [4, 3, 3, 1] 391s 0.692 -0.649 -0.031 0.112 391s -0.534 -0.099 0.344 -1.273 391s -0.340 0.170 0.095 -0.350 391s testing: mul_mat, [4, 2, 3, 4] = [4, 4, 3, 4] * [4, 2, 3, 4] 391s x0: [4, 4, 3, 4] 391s -0.842 -0.832 0.976 -0.243 391s -0.320 0.972 0.469 0.596 391s 0.165 -0.327 -0.805 0.566 391s 0.768 0.442 0.456 0.422 391s 391s x1: [4, 2, 3, 4] 391s -0.339 -0.649 0.020 0.187 391s 0.569 0.603 0.241 0.739 391s 391s y: [4, 2, 3, 4] 391s 0.799 -0.402 0.245 -0.459 391s -0.925 0.958 0.121 1.126 391s test-mul-mat0: iter:155/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.733 -0.551 0.879 -0.974 391s -0.778 -0.782 -0.168 0.193 391s -0.185 -0.989 0.999 0.716 391s 0.589 -0.464 0.784 -0.242 391s 391s x1: [4, 4, 1, 1] 391s -0.933 0.169 0.566 -0.884 391s -0.160 -0.755 -0.531 -0.460 391s 0.630 0.713 0.115 0.601 391s -0.710 0.947 -0.856 0.732 391s 391s y: [4, 4, 1, 1] 391s 1.949 0.329 -0.063 0.030 391s 0.515 0.716 -0.083 -0.048 391s -1.339 -0.951 -0.275 -0.015 391s -1.467 0.098 -1.135 -1.705 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 1.000 -0.985 -0.226 -0.782 391s 391s x1: [4, 4, 1, 1] 391s -0.898 0.900 0.598 -0.653 391s -0.631 -0.862 0.977 -0.918 391s 0.253 0.578 -0.629 0.200 391s 0.722 -0.896 -0.879 0.989 391s 391s y: [1, 4, 1, 1] 391s -1.408 391s 0.716 391s -0.330 391s 1.030 391s testing: mul_mat, [2, 1, 1, 2] = [4, 2, 1, 2] * [4, 1, 1, 2] 391s x0: [4, 2, 1, 2] 391s 0.205 -0.397 -0.567 -0.197 391s -0.050 -0.198 -0.058 -0.073 391s 391s x1: [4, 1, 1, 2] 391s -0.153 0.967 0.033 -0.142 391s 391s y: [2, 1, 1, 2] 391s -0.406 -0.175 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.936 -0.383 0.517 -0.618 391s 391s x1: [4, 2, 1, 1] 391s -0.784 0.911 0.516 -0.769 391s -0.315 0.733 0.077 -0.348 391s 391s y: [1, 2, 1, 1] 391s -0.341 391s -0.320 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.609 391s 391s x1: [1, 4, 1, 1] 391s 0.118 391s -0.950 391s 0.587 391s 0.722 391s 391s y: [1, 4, 1, 1] 391s -0.072 391s 0.579 391s -0.358 391s -0.440 391s testing: mul_mat, [1, 1, 1, 2] = [1, 1, 1, 2] * [1, 1, 1, 2] 391s x0: [1, 1, 1, 2] 391s 0.599 391s 391s x1: [1, 1, 1, 2] 391s -0.328 391s 391s y: [1, 1, 1, 2] 391s -0.196 391s test-mul-mat0: iter:156/500 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.954 -0.223 0.380 0.687 391s 0.854 -0.969 -0.546 0.790 391s -0.352 0.970 -0.828 0.767 391s -0.980 0.760 0.488 -0.497 391s 391s x1: [4, 1, 1, 1] 391s -0.670 -0.217 -0.562 -0.454 391s 391s y: [4, 1, 1, 1] 391s -1.116 -0.414 0.142 0.443 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.629 0.770 0.623 -0.164 391s 391s x1: [4, 4, 3, 1] 391s -0.850 -0.840 0.789 0.483 391s 0.759 0.958 0.010 0.862 391s 0.382 0.932 -0.808 -0.835 391s -0.630 -0.262 -0.140 -0.676 391s 391s y: [1, 4, 3, 1] 391s 0.300 391s 0.126 391s 0.111 391s 0.218 391s testing: mul_mat, [4, 1, 3, 2] = [4, 4, 3, 2] * [4, 1, 3, 2] 391s x0: [4, 4, 3, 2] 391s 0.590 -0.821 -0.659 0.318 391s 0.530 0.744 -0.995 -0.879 391s 0.480 0.440 0.437 0.310 391s 0.582 -0.514 0.095 0.811 391s 391s x1: [4, 1, 3, 2] 391s -0.313 -0.408 0.737 0.435 391s 391s y: [4, 1, 3, 2] 391s -0.196 -1.585 0.127 0.451 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.552 -0.213 0.621 -0.074 391s 0.822 0.113 0.444 0.336 391s -0.385 0.952 -0.363 0.123 391s 391s x1: [4, 4, 1, 1] 391s -0.942 -0.753 0.354 0.404 391s -0.632 -0.524 0.966 0.735 391s -0.003 0.325 -0.169 -0.990 391s 0.292 -0.143 -0.455 -0.839 391s 391s y: [3, 4, 1, 1] 391s 0.870 -0.567 -0.433 391s 1.006 0.097 -0.515 391s -0.099 -0.374 0.250 391s -0.352 -0.260 -0.187 391s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.743 0.465 0.504 391s 0.198 0.681 -0.826 391s 0.306 0.183 -0.647 391s 391s x1: [3, 4, 3, 1] 391s -0.141 0.656 0.181 391s 0.105 0.010 -0.414 391s 0.473 0.486 -0.448 391s 0.208 -0.517 0.877 391s 391s y: [3, 4, 3, 1] 391s 0.502 0.268 -0.041 391s -0.283 0.370 0.302 391s -0.351 0.795 0.524 391s 0.047 -1.036 -0.599 391s testing: mul_mat, [1, 3, 3, 2] = [3, 1, 3, 2] * [3, 3, 3, 2] 391s x0: [3, 1, 3, 2] 391s -0.872 -0.403 0.066 391s 391s x1: [3, 3, 3, 2] 391s -0.417 -0.852 -0.726 391s 0.337 -0.595 0.472 391s -0.357 0.871 0.153 391s 391s y: [1, 3, 3, 2] 391s 0.659 391s -0.023 391s -0.030 391s test-mul-mat0: iter:157/500 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.028 -0.684 391s 391s x1: [2, 2, 1, 1] 391s 0.096 -0.202 391s 0.786 -0.196 391s 391s y: [1, 2, 1, 1] 391s 0.141 391s 0.156 391s testing: mul_mat, [1, 1, 3, 1] = [2, 1, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 1, 3, 1] 391s 0.590 -0.308 391s 391s x1: [2, 1, 3, 1] 391s -0.068 0.339 391s 391s y: [1, 1, 3, 1] 391s -0.145 391s testing: mul_mat, [2, 1, 3, 3] = [2, 2, 3, 3] * [2, 1, 3, 3] 391s x0: [2, 2, 3, 3] 391s -0.077 0.530 391s -0.565 0.329 391s 391s x1: [2, 1, 3, 3] 391s 0.163 0.394 391s 391s y: [2, 1, 3, 3] 391s 0.196 0.037 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.663 0.976 391s -0.815 -0.927 391s -0.422 0.528 391s 391s x1: [2, 2, 1, 1] 391s 0.964 0.494 391s -0.257 -0.860 391s 391s y: [3, 2, 1, 1] 391s 1.121 -1.243 -0.147 391s -1.010 1.007 -0.345 391s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.809 0.331 0.235 391s 0.095 -0.448 -0.214 391s 391s x1: [3, 2, 3, 1] 391s 0.365 -0.255 -0.116 391s -0.224 -0.714 -0.081 391s 391s y: [2, 2, 3, 1] 391s 0.184 0.174 391s -0.437 0.316 391s testing: mul_mat, [3, 3, 3, 3] = [2, 3, 3, 3] * [2, 3, 3, 3] 391s x0: [2, 3, 3, 3] 391s -0.807 -0.556 391s 0.875 -0.301 391s 0.207 0.688 391s 391s x1: [2, 3, 3, 3] 391s 0.290 -0.681 391s 0.124 -0.014 391s -0.607 0.039 391s 391s y: [3, 3, 3, 3] 391s 0.144 0.459 -0.408 391s -0.092 0.113 0.016 391s 0.468 -0.543 -0.099 391s test-mul-mat0: iter:158/500 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.707 0.479 -0.662 391s -0.032 0.403 -0.359 391s -0.337 0.712 -0.152 391s -0.692 -0.118 -0.869 391s 391s x1: [3, 4, 1, 1] 391s 0.127 0.465 0.902 391s 0.660 -0.721 -0.612 391s 0.399 0.537 0.573 391s -0.173 -0.815 0.570 391s 391s y: [4, 4, 1, 1] 391s -0.464 -0.140 0.151 -0.927 391s -0.406 -0.092 -0.642 0.160 391s -0.404 -0.002 0.161 -0.837 391s -0.645 -0.527 -0.609 -0.279 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.050 0.346 0.255 391s 0.136 -0.084 0.997 391s 0.428 -0.606 -0.666 391s -0.604 0.797 -0.025 391s 391s x1: [3, 4, 1, 1] 391s -0.601 -0.177 -0.806 391s -0.757 0.668 0.484 391s 0.370 0.133 0.386 391s 0.029 0.413 0.774 391s 391s y: [4, 4, 1, 1] 391s -0.237 -0.870 0.386 0.242 391s 0.393 0.324 -1.051 0.978 391s 0.126 0.424 -0.179 -0.126 391s 0.339 0.741 -0.753 0.292 391s testing: mul_mat, [2, 4, 1, 4] = [3, 2, 1, 4] * [3, 4, 1, 4] 391s x0: [3, 2, 1, 4] 391s -0.837 0.071 -0.102 391s -0.745 -0.414 0.952 391s 391s x1: [3, 4, 1, 4] 391s 0.059 0.510 0.823 391s 0.367 -0.608 0.954 391s 0.766 0.215 -0.853 391s -0.991 -0.117 0.631 391s 391s y: [2, 4, 1, 4] 391s -0.097 0.528 391s -0.447 0.886 391s -0.539 -1.472 391s 0.757 1.388 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.482 0.761 0.812 391s 0.175 -0.587 0.880 391s -0.539 0.115 -0.840 391s -0.227 0.291 -0.543 391s 391s x1: [3, 2, 1, 1] 391s -0.168 0.027 0.413 391s -0.198 -0.645 -0.896 391s 391s y: [4, 2, 1, 1] 391s 0.437 0.318 -0.253 -0.178 391s -1.123 -0.444 0.785 0.344 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.836 -0.127 0.497 -0.482 391s 391s x1: [4, 3, 1, 1] 391s -0.957 0.428 0.856 0.427 391s -0.538 -0.599 0.984 0.029 391s 0.308 0.864 0.136 0.459 391s 391s y: [1, 3, 1, 1] 391s 0.966 391s 1.001 391s -0.521 391s testing: mul_mat, [4, 4, 1, 4] = [1, 4, 1, 4] * [1, 4, 1, 4] 391s x0: [1, 4, 1, 4] 391s -0.726 391s 0.256 391s -0.541 391s -0.741 391s 391s x1: [1, 4, 1, 4] 391s -0.023 391s 0.060 391s 0.036 391s -0.847 391s 391s y: [4, 4, 1, 4] 391s 0.017 -0.006 0.012 0.017 391s -0.043 0.015 -0.032 -0.044 391s -0.026 0.009 -0.020 -0.027 391s 0.615 -0.217 0.459 0.628 391s test-mul-mat0: iter:159/500 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.494 -0.165 0.800 0.761 391s 0.293 -0.9ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 41 -0.954 0.060 391s 391s x1: [4, 2, 1, 1] 391s 0.142 -0.171 0.873 -0.653 391s -0.425 -0.322 0.712 -0.223 391s 391s y: [2, 2, 1, 1] 391s 0.160 -0.669 391s 0.664 -0.514 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.121 -0.845 0.893 0.180 391s -0.799 -0.047 0.362 0.669 391s 391s x1: [4, 2, 3, 1] 391s -0.819 0.467 -0.715 -0.001 391s 0.052 -0.616 0.314 0.155 391s 391s y: [2, 2, 3, 1] 391s -1.132 0.372 391s 0.835 0.205 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.657 -0.736 0.863 -0.982 391s -0.205 0.637 0.366 0.122 391s 391s x1: [4, 2, 3, 1] 391s 0.348 -0.673 0.662 0.113 391s 0.648 -0.944 0.712 -0.231 391s 391s y: [2, 2, 3, 1] 391s 1.185 -0.244 391s 1.963 -0.501 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.806 0.831 0.129 -0.517 391s 0.203 -0.431 0.047 0.238 391s 0.447 -0.431 0.339 -0.123 391s 391s x1: [4, 2, 1, 1] 391s 0.567 0.818 0.538 -0.935 391s 0.528 -0.612 -0.669 0.185 391s 391s y: [3, 2, 1, 1] 391s 0.775 -0.435 0.198 391s -1.116 0.384 0.250 391s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.299 -0.082 0.290 391s -0.816 -0.757 -0.629 391s 391s x1: [3, 4, 3, 1] 391s -0.952 0.034 -0.635 391s -0.974 0.389 0.164 391s -0.916 0.539 0.635 391s -0.056 0.106 0.452 391s 391s y: [2, 4, 3, 1] 391s -0.471 1.150 391s -0.275 0.398 391s -0.134 -0.060 391s 0.106 -0.319 391s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 1, 3, 1] 391s 0.657 0.867 391s 391s x1: [2, 3, 3, 1] 391s 0.627 0.845 391s 0.806 -0.496 391s -0.015 -0.972 391s 391s y: [1, 3, 3, 1] 391s 1.144 391s 0.100 391s -0.852 391s test-mul-mat0: iter:160/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.330 -0.407 0.507 391s 0.358 -0.148 0.681 391s -0.356 0.230 0.070 391s -0.425 -0.108 0.758 391s 391s x1: [3, 1, 1, 1] 391s -0.323 0.524 -0.911 391s 391s y: [4, 1, 1, 1] 391s -0.782 -0.814 0.172 -0.610 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.246 -0.565 0.959 391s -0.230 -0.477 -0.519 391s 391s x1: [3, 4, 1, 1] 391s -0.310 0.473 -0.311 391s -0.904 -0.576 -0.654 391s 0.962 -0.274 0.600 391s 0.792 -0.456 0.973 391s 391s y: [2, 4, 1, 1] 391s -0.642 0.008 391s -0.524 0.822 391s 0.967 -0.402 391s 1.386 -0.469 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.198 -0.260 -0.324 391s -0.910 -0.503 0.366 391s 0.563 0.186 0.462 391s 391s x1: [3, 2, 1, 1] 391s -0.900 0.116 0.988 391s 0.458 0.968 0.670 391s 391s y: [3, 2, 1, 1] 391s -0.529 1.122 -0.029 391s -0.378 -0.658 0.747 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.539 -0.937 0.179 391s -0.544 -0.062 -0.074 391s 0.272 0.372 -0.170 391s 391s x1: [3, 3, 1, 1] 391s 0.987 0.532 0.424 391s -0.288 0.131 0.216 391s 0.256 0.105 -0.503 391s 391s y: [3, 3, 1, 1] 391s 0.110 -0.601 0.394 391s -0.240 0.132 -0.066 391s -0.050 -0.109 0.194 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.019 0.859 0.954 391s -0.565 -0.302 -0.213 391s 0.986 0.682 0.571 391s -0.449 0.074 -0.544 391s 391s x1: [3, 3, 1, 1] 391s 0.147 -0.404 -0.069 391s -0.655 0.336 0.607 391s -0.564 0.833 -0.027 391s 391s y: [4, 3, 1, 1] 391s -0.410 0.054 -0.170 -0.059 391s 0.855 0.139 -0.069 -0.011 391s 0.679 0.073 -0.003 0.330 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.823 0.496 0.442 -0.325 391s -0.531 0.656 0.495 0.877 391s 391s x1: [4, 3, 1, 1] 391s -0.674 -0.972 0.728 -0.611 391s -0.035 0.100 0.568 0.891 391s 0.930 -0.285 -0.513 -0.139 391s 391s y: [2, 3, 1, 1] 391s -0.516 -0.454 391s -0.018 1.147 391s 0.442 -1.057 391s test-mul-mat0: iter:161/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.265 0.348 -0.308 0.752 391s 391s x1: [4, 3, 1, 1] 391s -0.092 -0.191 -0.867 0.694 391s -0.619 -0.411 -0.980 -0.592 391s -0.682 -0.591 0.373 0.418 391s 391s y: [1, 3, 1, 1] 391s 0.698 391s -0.450 391s -0.187 391s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.595 -0.963 -0.458 -0.462 391s 0.731 -0.078 0.127 0.752 391s 0.331 0.445 -0.839 -0.296 391s 391s x1: [4, 1, 3, 1] 391s -0.790 0.753 0.574 -0.322 391s 391s y: [3, 1, 3, 1] 391s -0.369 -0.806 -0.313 391s testing: mul_mat, [3, 3, 3, 4] = [4, 3, 3, 4] * [4, 3, 3, 4] 391s x0: [4, 3, 3, 4] 391s 0.430 -0.032 -0.746 -0.241 391s -0.069 0.852 0.598 0.186 391s 0.675 -0.002 -0.072 0.134 391s 391s x1: [4, 3, 3, 4] 391s -0.588 -0.128 0.078 0.743 391s -0.683 0.240 -0.553 0.180 391s 0.378 -0.584 0.391 -0.791 391s 391s y: [3, 3, ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 3, 4] 391s -0.486 0.117 -0.302 391s 0.068 -0.046 -0.398 391s 0.080 -0.437 0.122 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.254 -0.853 -0.606 0.930 391s 0.590 0.965 0.190 -0.589 391s -0.096 -0.450 0.484 0.551 391s 391s x1: [4, 3, 1, 1] 391s -0.908 0.708 -0.333 -0.574 391s -0.814 0.428 -0.861 -0.283 391s 0.546 0.564 0.393 -0.344 391s 391s y: [3, 3, 1, 1] 391s -1.167 0.422 -0.709 391s -0.313 -0.064 -0.687 391s -0.900 1.143 -0.305 391s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.221 -0.696 0.480 391s 0.377 -0.167 0.031 391s 0.731 -0.328 -0.014 391s 0.877 0.949 -0.215 391s 391s x1: [3, 4, 3, 1] 391s -0.910 0.142 -0.051 391s -0.861 0.033 0.007 391s -0.770 -0.259 0.674 391s -0.343 -0.073 0.102 391s 391s y: [4, 4, 3, 1] 391s -0.324 -0.368 -0.711 -0.652 391s -0.210 -0.330 -0.640 -0.726 391s 0.334 -0.226 -0.487 -1.066 391s 0.024 -0.114 -0.229 -0.392 391s testing: mul_mat, [2, 3, 3, 4] = [4, 2, 3, 4] * [4, 3, 3, 4] 391s x0: [4, 2, 3, 4] 391s -0.445 -0.299 -0.502 -0.821 391s 0.585 -0.677 0.689 -0.491 391s 391s x1: [4, 3, 3, 4] 391s -0.453 -0.080 -0.085 -0.972 391s 0.952 0.900 -0.188 -0.406 391s 0.178 -0.509 0.753 -0.107 391s 391s y: [2, 3, 3, 4] 391s 1.066 0.207 391s -0.266 0.017 391s -0.217 1.020 391s test-mul-mat0: iter:162/500 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.232 0.880 391s 0.533 0.677 391s 391s x1: [2, 2, 1, 1] 391s -0.259 -0.373 391s 0.096 0.831 391s 391s y: [2, 2, 1, 1] 391s -0.388 -0.390 391s 0.754 0.614 391s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 1, 3, 1] 391s 0.082 -0.876 391s 391s x1: [2, 2, 3, 1] 391s -0.922 0.036 391s 0.340 -0.842 391s 391s y: [1, 2, 3, 1] 391s -0.107 391s 0.765 391s testing: mul_mat, [4, 1, 3, 2] = [2, 4, 3, 2] * [2, 1, 3, 2] 391s x0: [2, 4, 3, 2] 391s -0.617 -0.273 391s -0.403 -0.940 391s 0.041 0.800 391s 0.514 0.288 391s 391s x1: [2, 1, 3, 2] 391s 0.908 -0.898 391s 391s y: [4, 1, 3, 2] 391s -0.316 0.478 -0.681 0.208 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.764 0.664 391s -0.609 0.634 391s 0.495 0.567 391s 0.302 0.534 391s 391s x1: [2, 4, 1, 1] 391s 0.214 0.090 391s -0.266 0.175 391s 0.083 0.329 391s 0.614 0.557 391s 391s y: [4, 4, 1, 1] 391s 0.223 -0.073 0.157 0.113 391s -0.087 0.273 -0.033 0.013 391s 0.283 0.158 0.228 0.201 391s 0.840 -0.021 0.620 0.483 391s testing: mul_mat, [1, 2, 3, 1] = [4, 1, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.476 0.164 -0.142 -0.842 391s 391s x1: [4, 2, 3, 1] 391s 0.833 -0.127 0.204 0.546 391s 0.468 0.090 0.016 0.255 391s 391s y: [1, 2, 3, 1] 391s -0.906 391s -0.425 391s testing: mul_mat, [1, 4, 3, 2] = [1, 1, 3, 2] * [1, 4, 3, 2] 391s x0: [1, 1, 3, 2] 391s -0.134 391s 391s x1: [1, 4, 3, 2] 391s -0.448 391s -0.673 391s -0.854 391s 0.878 391s 391s y: [1, 4, 3, 2] 391s 0.060 391s 0.090 391s 0.115 391s -0.118 391s test-mul-mat0: iter:163/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.822 -0.107 0.367 391s 0.856 0.248 -0.470 391s 391s x1: [3, 1, 1, 1] 391s 0.580 -0.728 0.782 391s 391s y: [2, 1, 1, 1] 391s 0.841 -0.052 391s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.238 0.583 0.090 391s -0.354 -0.462 0.984 391s 391s x1: [3, 2, 2, 1] 391s 0.987 -0.594 -0.277 391s 0.981 -0.244 0.440 391s 391s y: [2, 2, 2, 1] 391s -0.606 -0.347 391s -0.336 0.198 391s testing: mul_mat, [4, 2, 2, 3] = [3, 4, 2, 3] * [3, 2, 2, 3] 391s x0: [3, 4, 2, 3] 391s 0.025 -0.701 -0.968 391s 0.618 -0.245 0.149 391s 0.350 0.176 -0.419 391s -0.889 0.161 0.194 391s 391s x1: [3, 2, 2, 3] 391s -0.230 -0.954 -0.577 391s -0.375 0.294 -0.046 391s 391s y: [4, 2, 2, 3] 391s 1.221 0.005 -0.007 -0.061 391s -0.171 -0.311 -0.060 0.372 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.867 0.898 0.055 391s 0.655 0.129 0.382 391s 0.454 0.982 -0.876 391s 391s x1: [3, 4, 1, 1] 391s 0.701 0.301 -0.259 391s -0.607 -0.573 -0.403 391s -0.611 0.679 -0.421 391s 0.758 0.161 -0.395 391s 391s y: [3, 4, 1, 1] 391s -0.352 0.398 0.841 391s -0.010 -0.625 -0.485 391s 1.116 -0.474 0.758 391s -0.535 0.366 0.848 391s testing: mul_mat, [1, 3, 2, 1] = [3, 1, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 1, 2, 1] 391s 0.604 -0.369 0.874 391s 391s x1: [3, 3, 2, 1] 391s 0.904 -0.554 0.499 391s 0.199 0.471 0.457 391s -0.884 0.036 -0.891 391s 391s y: [1, 3, 2, 1] 391s 1.187 391s 0.346 391s -1.327 391s testing: mul_mat, [3, 3, 2, 3] = [1, 3, 2, 3] * [1, 3, 2, 3] 391s x0: [1, 3, 2, 3] 391s -0.148 391s -0.204 391s -0.466 391s 391s x1: [1, 3, 2, 3] 391s 0.635 391s 0.414 391s 0.310 391s 391s y: [3, 3, 2, 3] 391s -0.094 -0.129 -0.296 391s -0.061 -0.084 -0.193 391s -0.046 -0.063 -0.144 391s test-mul-mat0: iter:164/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.339 -0.019 0.741 391s -0.990 -0.084 -0.408 391s -0.194 0.450 0.887 391s 0.628 -0.658 -0.213 391s 391s x1: [3, 1, 1, 1] 391s 0.904 -0.838 0.256 391s 391s y: [4, 1, 1, 1] 391s -0.101 -0.929 -0.326 1.065 391s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.067 0.488 0.823 391s 0.475 0.535 -0.466 391s -0.524 -0.612 -0.305 391s 391s x1: [3, 4, 4, 1] 391s 0.081 0.107 0.423 391s -0.181 0.201 0.694 391s 0.972 -0.395 -0.345 391s 0.920 0.443 -0.790 391s 391s y: [3, 4, 4, 1] 391s 0.406 -0.101 -0.237 391s 0.658 -0.302 -0.240 391s -0.412 0.411 -0.162 391s -0.373 1.043 -0.512 391s testing: mul_mat, [2, 3, 4, 2] = [3, 2, 4, 2] * [3, 3, 4, 2] 391s x0: [3, 2, 4, 2] 391s 0.548 -0.843 0.354 391s -0.103 -0.289 -0.948 391s 391s x1: [3, 3, 4, 2] 391s 0.638 -0.762 -0.169 391s -0.667 0.603 0.701 391s 0.531 -0.358 -0.868 391s 391s y: [2, 3, 4, 2] 391s 0.932 0.314 391s -0.625 -0.769 391s 0.286 0.871 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.511 0.874 0.417 391s 0.500 0.380 -0.162 391s 0.501 0.912 -0.386 391s -0.660 0.796 -0.835 391s 391s x1: [3, 2, 1, 1] 391s 0.389 0.261 0.386 391s 0.935 -0.475 0.436 391s 391s y: [4, 2, 1, 1] 391s 0.588 0.231 0.284 -0.372 391s 0.244 0.217 -0.133 -1.359 391s testing: mul_mat, [3, 3, 4, 1] = [4, 3, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 3, 4, 1] 391s 0.237 0.727 0.597 -0.116 391s -0.498 -0.130 0.151 -0.730 391s -0.677 -0.472 -0.471 0.144 391s 391s x1: [4, 3, 4, 1] 391s 0.816 0.226 0.067 0.669 391s -0.100 -0.576 -0.406 -0.509 391s -0.184 0.446 -0.268 -0.992 391s 391s y: [3, 3, 4, 1] 391s 0.320 -0.914 -0.595 391s -0.625 0.435 0.458 391s 0.235 0.718 -0.102 391s testing: mul_mat, [4, 4, 4, 2] = [3, 4, 4, 2] * [3, 4, 4, 2] 391s x0: [3, 4, 4, 2] 391s -0.877 -0.829 -0.579 391s 0.660 -0.430 0.702 391s 0.748 0.111 -0.357 391s -0.520 -0.142 -0.006 391s 391s x1: [3, 4, 4, 2] 391s 0.722 0.813 -0.836 391s -0.319 0.375 0.812 391s 0.604 -0.294 -0.062 391s -0.516 0.199 -0.687 391s 391s y: [4, 4, 4, 2] 391s -0.823 -0.460 0.929 -0.486 391s -0.501 0.198 -0.487 0.108 391s -0.250 0.482 0.441 -0.272 391s 0.686 -0.909 -0.119 0.244 391s test-mul-mat0: iter:165/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.446 0.410 391s -0.186 -0.554 391s 391s x1: [2, 4, 1, 1] 391s 0.980 -0.613 391s -0.678 -0.662 391s 0.590 0.255 391s -0.336 0.315 391s 391s y: [2, 4, 1, 1] 391s 0.185 0.158 391s -0.574 0.492 391s 0.368 -0.251 391s -0.021 -0.112 391s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.499 -0.400 391s -0.271 0.500 391s 391s x1: [2, 2, 3, 1] 391s -0.257 -0.801 391s 0.435 -0.143 391s 391s y: [2, 2, 3, 1] 391s 0.449 -0.331 391s -0.159 -0.190 391s testing: mul_mat, [3, 2, 3, 1] = [2, 3, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 3, 3, 1] 391s 0.915 -0.846 391s -0.840 0.984 391s -0.056 -0.912 391s 391s x1: [2, 2, 3, 1] 391s -0.208 0.224 391s 0.156 -0.798 391s 391s y: [3, 2, 3, 1] 391s -0.380 0.396 -0.193 391s 0.818 -0.917 0.719 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.009 -0.921 391s 391s x1: [2, 3, 1, 1] 391s -0.858 -0.267 391s 0.439 -0.656 391s -0.228 0.042 391s 391s y: [1, 3, 1, 1] 391s 0.238 391s 0.608 391s -0.040 391s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 3, 3, 1] 391s 0.019 391s -0.676 391s 0.725 391s 391s x1: [1, 2, 3, 1] 391s 0.091 391s 0.120 391s 391s y: [3, 2, 3, 1] 391s 0.002 -0.061 0.066 391s 0.002 -0.081 0.087 391s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.642 -0.511 0.481 391s 0.402 0.368 0.579 391s 391s x1: [3, 1, 3, 1] 391s 0.364 0.344 -0.361 391s 391s y: [2, 1, 3, 1] 391s -0.583 0.063 391s test-mul-mat0: iter:166/500 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.479 -0.654 -0.840 0.888 391s 0.827 0.739 -0.624 -0.467 391s -0.205 0.900 -0.381 0.487 391s -0.557 0.029 0.904 -0.510 391s 391s x1: [4, 2, 1, 1] 391s 0.722 0.664 0.616 -0.424 391s -0.415 -0.529 -0.838 -0.055 391s 391s y: [4, 2, 1, 1] 391s -1.674 0.902 0.008 0.389 391s 1.200 -0.186 -0.098 -0.513 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.152 -0.447 0.253 -0.413 391s -0.071 0.786 0.382 -0.171 391s 391s x1: [4, 4, 1, 1] 391s 0.306 -0.138 -0.971 0.684 391s 0.887 -0.069 0.406 0.550 391s -0.453 0.982 -0.865 0.018 391s -0.856 0.080 -0.311 -0.335 391s 391s y: [2, 4, 1, 1] 391s -0.513 -0.618 391s -0.228 -0.056 391s -0.596 0.471 391s 0.154 0.062 391s testing: mul_mat, [4, 2, 1, 4] = [4, 4, 1, 4] * [4, 2, 1, 4] 391s x0: [4, 4, 1, 4] 391s -0.795 -0.409 0.434 0.967 391s 0.331 0.069 0.758 0.755 391s -0.410 -0.520 0.585 0.661 391s 0.749 0.398 -0.132 -0.965 391s 391s x1: [4, 2, 1, 4] 391s -0.595 -0.131 0.272 0.434 391s -0.227 0.762 -0.260 0.635 391s 391s y: [4, 2, 1, 4] 391s 1.064 0.328 0.758 -0.952 391s 0.369 0.259 -0.035 -0.445 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.773 0.097 -0.915 0.795 391s 0.926 0.577 0.091 -0.833 391s -0.892 -0.223 -0.656 0.632 391s -0.351 -0.148 -0.708 -0.145 391s 391s x1: [4, 4, 1, 1] 391s 0.341 0.060 0.482 0.543 391s 0.175 -0.103 -0.665 0.770 391s 0.897 -0.212 -0.886 -0.706 391s -0.890 -0.507 -0.954 -0.415 391s 391s y: [4, 4, 1, 1] 391s 0.259 -0.058 -0.290 -0.549 391s 1.346 -0.600 0.790 0.313 391s 0.922 1.215 -0.619 0.446 391s -0.194 -0.857 1.270 1.124 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.602 0.981 0.212 0.984 391s 0.204 -0.133 -0.648 0.723 391s -0.984 -0.940 -0.133 0.027 391s 0.975 0.073 0.379 0.159 391s 391s x1: [4, 4, 1, 1] 391s 0.649 0.175 -0.970 -0.454 391s 0.510 0.800 -0.558 -0.701 391s 0.914 -0.263 -0.592 -0.593 391s -0.217 -0.007 -0.710 -0.445 391s 391s y: [4, 4, 1, 1] 391s -0.871 0.410 -0.687 0.206 391s -0.330 -0.147 -1.199 0.233 391s -1.517 0.177 -0.590 0.554 391s -0.464 0.095 0.302 -0.552 391s testing: mul_mat, [4, 4, 1, 4] = [4, 4, 1, 4] * [4, 4, 1, 4] 391s x0: [4, 4, 1, 4] 391s 0.759 0.043 -0.505 0.110 391s -0.087 0.018 -0.200 0.511 391s 0.911 0.053 -0.538 0.329 391s 0.469 -0.293 0.260 0.956 391s 391s x1: [4, 4, 1, 4] 391s 0.753 0.573 -0.330 0.553 391s -0.985 -0.032 0.467 -0.248 391s 0.377 0.874 0.535 -0.630 391s -0.836 -0.910 0.288 -0.438 391s 391s y: [4, 4, 1, 4] 391s 0.824 0.293 1.075 0.628 391s -1.012 -0.135 -1.231 -0.568 391s -0.016 -0.446 -0.106 -0.543 391s -0.867 -0.225 -1.108 -0.470 391s test-mul-mat0: iter:167/500 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.827 0.005 391s 391s x1: [2, 3, 1, 1] 391s -0.908 0.494 391s -0.129 0.909 391s 0.905 -0.726 391s 391s y: [1, 3, 1, 1] 391s -0.748 391s -0.102 391s 0.745 391s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.497 0.863 391s 0.277 -0.619 391s -0.320 0.011 391s 391s x1: [2, 1, 3, 1] 391s 0.720 0.583 391s 391s y: [3, 1, 3, 1] 391s 0.145 -0.162 -0.224 391s testing: mul_mat, [3, 3, 3, 2] = [2, 3, 3, 2] * [2, 3, 3, 2] 391s x0: [2, 3, 3, 2] 391s -0.351 -0.983 391s -0.901 -0.611 391s 0.302 0.621 391s 391s x1: [2, 3, 3, 2] 391s 0.180 -0.970 391s -0.582 -0.984 391s 0.857 0.423 391s 391s y: [3, 3, 3, 2] 391s 0.891 0.431 -0.548 391s 1.171 1.125 -0.787 391s -0.716 -1.030 0.521 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.640 -0.373 391s -0.496 -0.802 391s 0.055 -0.093 391s 391s x1: [2, 3, 1, 1] 391s 0.063 -0.084 391s -0.181 -0.932 391s -0.776 -0.921 391s 391s y: [3, 3, 1, 1] 391s 0.072 0.036 0.011 391s 0.232 0.838 0.076 391s -0.153 1.123 0.043 391s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 4, 3, 1] 391s -0.444 0.725 -0.682 391s 0.646 0.264 0.046 391s -0.874 -0.581 -0.573 391s -0.221 -0.771 0.226 391s 391s x1: [3, 2, 3, 1] 391s 0.729 -0.662 0.326 391s -0.782 -0.804 -0.564 391s 391s y: [4, 2, 3, 1] 391s -1.026 0.311 -0.439 0.423 391s 0.149 -0.743 1.473 0.665 391s testing: mul_mat, [1, 3, 3, 2] = [4, 1, 3, 2] * [4, 3, 3, 2] 391s x0: [4, 1, 3, 2] 391s 0.633 0.223 0.371 -0.937 391s 391s x1: [4, 3, 3, 2] 391s 0.440 -0.142 0.455 0.758 391s 0.904 0.882 -0.016 -0.322 391s -0.352 0.536 0.670 0.610 391s 391s y: [1, 3, 3, 2] 391s -0.295 391s 1.064 391s -0.426 391s test-mul-mat0: iter:168/500 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.672 -0.999 -0.930 -0.230 391s -0.454 -0.855 0.621 0.086 391s -0.768 -0.788 -0.759 -0.080 391s 391s x1: [4, 1, 1, 1] 391s -0.696 -0.948 -0.043 0.937 391s 391s y: [3, 1, 1, 1] 391s 1.240 1.181 1.239 391s testing: mul_mat, [4, 3, 2, 1] = [4, 4, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 4, 2, 1] 391s -0.731 0.289 0.215 -0.519 391s 0.530 -0.866 -0.599 0.818 391s -0.568 -0.297 0.761 -0.889 391s 0.052 -0.458 0.945 0.870 391s 391s x1: [4, 3, 2, 1] 391s 0.920 -0.712 -0.703 -0.698 391s 0.943 0.679 -0.650 -0.219 391s 0.834 -0.182 -0.721 0.415 391s 391s y: [4, 3, 2, 1] 391s -0.667 0.954 -0.225 -0.897 391s -0.519 0.122 -1.038 -1.067 391s -1.033 1.371 -1.338 -0.194 391s testing: mul_mat, [3, 4, 2, 3] = [4, 3, 2, 3] * [4, 4, 2, 3] 391s x0: [4, 3, 2, 3] 391s 0.037 -0.748 0.672 -0.492 391s -0.248 -0.907 0.546 -0.326 391s -0.047 0.383 -0.719 -0.395 391s 391s x1: [4, 4, 2, 3] 391s 0.883 -0.391 -0.763 0.413 391s -0.257 -0.362 0.230 0.174 391s 0.341 -0.009 0.285 -0.607 391s 0.533 0.230 -0.737 -0.647 391s 391s y: [3, 4, 2, 3] 391s -0.391 -0.415 0.194 391s 0.330 0.461 -0.361 391s 0.510 0.277 0.015 391s -0.329 -0.532 0.848 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.927 0.332 -0.503 0.810 391s 0.561 0.931 0.369 0.677 391s 0.839 -0.340 0.507 -0.882 391s -0.070 -0.910 -0.956 -0.344 391s 391s x1: [4, 3, 1, 1] 391s 0.532 -0.813 -0.498 0.368 391s 0.314 0.909 -0.138 -0.244 391s 0.950 -0.886 -0.130 -0.381 391s 391s y: [4, 3, 1, 1] 391s 0.773 -0.393 0.146 1.052 391s 0.466 0.807 0.099 -0.634 391s 0.343 -0.597 1.368 0.994 391s testing: mul_mat, [3, 4, 2, 1] = [4, 3, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.780 0.777 -0.804 -0.271 391s -0.458 0.595 0.405 -0.915 391s -0.150 0.363 0.826 -0.091 391s 391s x1: [4, 4, 2, 1] 391s 0.553 -0.898 -0.811 0.739 391s -0.396 0.557 0.054 -0.487 391s -0.581 0.810 -0.537 -0.467 391s -0.320 0.082 0.716 -0.393 391s 391s y: [3, 4, 2, 1] 391s -0.677 -1.793 -1.147 391s 0.830 0.980 0.350 391s 1.641 0.958 -0.020 391s -0.156 0.845 0.705 391s testing: mul_mat, [3, 4, 2, 3] = [3, 3, 2, 3] * [3, 4, 2, 3] 391s x0: [3, 3, 2, 3] 391s 0.649 0.831 -0.141 391s 0.162 -0.366 -0.064 391s 0.321 0.202 0.849 391s 391s x1: [3, 4, 2, 3] 391s 0.492 0.367 -0.584 391s 0.636 0.399 0.219 391s 0.882 0.619 0.760 391s -0.269 0.396 0.355 391s 391s y: [3, 4, 2, 3] 391s 0.707 -0.017 -0.263 391s 0.713 -0.057 0.471 391s 0.980 -0.132 1.054 391s 0.105 -0.211 0.296 391s test-mul-mat0: iter:169/500 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.748 -0.605 391s 391s x1: [2, 1, 1, 1] 391s -0.796 0.416 391s 391s y: [1, 1, 1, 1] 391s 0.343 391s testing: mul_mat, [4, 1, 4, 1] = [2, 4, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.010 -0.703 391s -0.177 0.657 391s -0.098 0.968 391s -0.383 0.807 391s 391s x1: [2, 1, 4, 1] 391s 0.893 0.621 391s 391s y: [4, 1, 4, 1] 391s -0.428 0.251 0.514 0.160 391s testing: mul_mat, [3, 4, 4, 2] = [2, 3, 4, 2] * [2, 4, 4, 2] 391s x0: [2, 3, 4, 2] 391s 0.910 -0.919 391s -0.161 -0.895 391s -0.509 -0.162 391s 391s x1: [2, 4, 4, 2] 391s -0.362 -0.589 391s 0.834 0.540 391s -0.621 -0.549 391s 0.347 -0.504 391s 391s y: [3, 4, 4, 2] 391s 0.212 0.586 0.279 391s 0.262 -0.618 -0.512 391s -0.061 0.591 0.405 391s 0.780 0.395 -0.095 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.634 0.054 391s 391s x1: [2, 3, 1, 1] 391s -1.000 -0.340 391s 0.377 0.732 391s -0.310 0.338 391s 391s y: [1, 3, 1, 1] 391s -0.652 391s 0.278 391s -0.179 391s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 4, 4, 1] 391s 0.584 391s 0.847 391s 0.345 391s 0.125 391s 391s x1: [1, 2, 4, 1] 391s -0.165 391s -0.905 391s 391s y: [4, 2, 4, 1] 391s -0.097 -0.140 -0.057 -0.021 391s -0.529 -0.767 -0.313 -0.113 391s testing: mul_mat, [2, 1, 4, 2] = [4, 2, 4, 2] * [4, 1, 4, 2] 391s x0: [4, 2, 4, 2] 391s 0.191 0.231 -0.469 0.399 391s -0.292 0.380 -0.340 0.222 391s 391s x1: [4, 1, 4, 2] 391s -0.980 0.700 0.135 -0.011 391s 391s y: [2, 1, 4, 2] 391s -0.093 0.503 391s test-mul-mat0: iter:170/500 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.631 -0.975 -0.982 -0.788 391s 0.351 -0.762 0.471 -0.036 391s 391s x1: [4, 2, 1, 1] 391s -0.695 -0.408 0.049 -0.187 391s -0.204 -0.480 0.104 0.147 391s 391s y: [2, 2, 1, 1] 391s 0.936 0.097 391s 0.379 0.338 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s -0.230 -0.304 0.775 -0.342 391s 391s x1: [4, 2, 4, 1] 391s 0.889 0.332 0.124 -0.329 391s 0.991 0.888 -0.697 0.344 391s 391s y: [1, 2, 4, 1] 391s -0.096 391s -1.155 391s testing: mul_mat, [4, 1, 4, 2] = [4, 4, 4, 2] * [4, 1, 4, 2] 391s x0: [4, 4, 4, 2] 391s 0.845 -0.257 0.655 -0.929 391s -0.830 0.275 0.277 -0.722 391s 0.031 -0.064 -0.967 -0.187 391s -0.621 -0.443 0.291 0.596 391s 391s x1: [4, 1, 4, 2] 391s -0.212 0.839 0.457 -0.905 391s 391s y: [4, 1, 4, 2] 391s 0.746 1.187 -0.334 -0.647 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.064 0.503 -0.474 -0.497 391s -0.024 0.746 -1.000 -0.447 391s 391s x1: [4, 4, 1, 1] 391s -0.988 -0.177 0.712 0.347 391s 0.643 -0.633 -0.139 0.707 391s 0.026 -0.187 0.439 0.105 391s 0.200 -0.965 0.954 -0.952 391s 391s y: [2, 4, 1, 1] 391s -0.536 -0.975 391s -0.645 -0.665 391s -0.356 -0.626 391s -0.479 -1.254 391s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 3, 4, 1] 391s -0.323 0.987 391s 0.765 -0.949 391s -0.849 -0.225 391s 391s x1: [2, 4, 4, 1] 391s 0.116 0.690 391s -0.932 -0.726 391s 0.801 0.310 391s -0.714 -0.376 391s 391s y: [3, 4, 4, 1] 391s 0.643 -0.565 -0.253 391s -0.416 -0.024 0.954 391s 0.048 0.319 -0.750 391s -0.141 -0.190 0.690 391s testing: mul_mat, [1, 2, 4, 2] = [3, 1, 4, 2] * [3, ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 2, 4, 2] 391s x0: [3, 1, 4, 2] 391s -0.612 0.341 0.945 391s 391s x1: [3, 2, 4, 2] 391s -0.622 -0.866 0.502 391s 0.038 -0.257 -0.007 391s 391s y: [1, 2, 4, 2] 391s 0.560 391s -0.118 391s test-mul-mat0: iter:171/500 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.699 -0.744 391s -0.103 -0.607 391s -0.863 0.980 391s 0.178 0.455 391s 391s x1: [2, 2, 1, 1] 391s -0.327 -0.751 391s -0.644 0.061 391s 391s y: [4, 2, 1, 1] 391s 0.787 0.489 -0.454 -0.400 391s 0.404 0.029 0.616 -0.086 391s testing: mul_mat, [2, 4, 2, 1] = [2, 2, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.124 -0.612 391s -0.103 0.480 391s 391s x1: [2, 4, 2, 1] 391s 0.059 -0.268 391s 0.337 -0.487 391s 0.189 0.284 391s 0.380 0.293 391s 391s y: [2, 4, 2, 1] 391s 0.172 -0.135 391s 0.340 -0.268 391s -0.150 0.117 391s -0.132 0.102 391s testing: mul_mat, [2, 2, 2, 2] = [2, 2, 2, 2] * [2, 2, 2, 2] 391s x0: [2, 2, 2, 2] 391s 0.319 -0.584 391s 0.139 0.878 391s 391s x1: [2, 2, 2, 2] 391s 0.384 -0.825 391s 0.842 0.364 391s 391s y: [2, 2, 2, 2] 391s 0.604 -0.671 391s 0.057 0.436 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.875 0.280 391s 0.448 0.148 391s 391s x1: [2, 2, 1, 1] 391s 0.516 0.920 391s 0.791 -0.186 391s 391s y: [2, 2, 1, 1] 391s 0.709 0.367 391s 0.640 0.327 391s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.228 -0.974 391s 0.507 -0.647 391s 0.668 -0.744 391s 0.774 0.994 391s 391s x1: [2, 2, 2, 1] 391s 0.367 0.597 391s 0.713 0.575 391s 391s y: [4, 2, 2, 1] 391s -0.665 -0.200 -0.199 0.877 391s -0.723 -0.011 0.049 1.124 391s testing: mul_mat, [1, 2, 2, 2] = [4, 1, 2, 2] * [4, 2, 2, 2] 391s x0: [4, 1, 2, 2] 391s -0.064 0.379 0.702 -0.292 391s 391s x1: [4, 2, 2, 2] 391s 0.998 -0.656 -0.376 0.446 391s 0.624 0.772 -0.187 0.221 391s 391s y: [1, 2, 2, 2] 391s -0.707 391s 0.057 391s test-mul-mat0: iter:172/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.821 0.997 -0.416 391s 0.556 -0.691 -0.482 391s 0.633 0.245 0.897 391s 0.335 0.952 -0.911 391s 391s x1: [3, 1, 1, 1] 391s 0.896 -0.912 -0.565 391s 391s y: [4, 1, 1, 1] 391s 0.062 1.401 -0.163 -0.054 391s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 4, 2, 1] 391s -0.071 -0.992 -0.902 391s -0.736 0.960 -0.813 391s -0.478 -0.895 -0.145 391s 0.882 -0.330 0.385 391s 391s x1: [3, 4, 2, 1] 391s -0.743 -0.855 -0.332 391s 0.360 -0.435 -0.470 391s -0.980 0.093 -0.553 391s 0.742 0.398 -0.808 391s 391s y: [4, 4, 2, 1] 391s 1.200 -0.004 1.169 -0.500 391s 0.830 -0.300 0.285 0.280 391s 0.476 1.261 0.466 -1.108 391s 0.282 0.493 -0.594 0.212 391s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.289 -0.622 -0.940 391s 0.938 0.290 -0.437 391s 0.687 0.942 0.521 391s 391s x1: [3, 4, 2, 1] 391s 0.749 -0.348 -0.043 391s 0.432 0.949 0.720 391s 0.633 -0.122 0.728 391s 0.731 0.143 0.688 391s 391s y: [3, 4, 2, 1] 391s 0.472 0.621 0.165 391s -1.143 0.366 1.566 391s -0.426 0.241 0.699 391s -0.524 0.427 0.995 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.658 0.334 -0.099 391s -0.477 -0.053 0.394 391s 391s x1: [3, 3, 1, 1] 391s -0.599 0.576 -0.550 391s 0.174 0.121 -0.088 391s 0.332 -0.432 0.137 391s 391s y: [2, 3, 1, 1] 391s -0.147 0.038 391s 0.164 -0.124 391s 0.060 -0.082 391s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 3, 2, 1] 391s -0.295 -0.719 391s -0.655 -0.206 391s 0.330 -0.496 391s 391s x1: [2, 3, 2, 1] 391s -0.114 -0.809 391s 0.957 -0.427 391s -0.867 0.478 391s 391s y: [3, 3, 2, 1] 391s 0.616 0.242 0.364 391s 0.025 -0.539 0.528 391s -0.088 0.470 -0.523 391s testing: mul_mat, [3, 2, 2, 1] = [3, 3, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 3, 2, 1] 391s -0.387 0.406 -0.889 391s -0.962 0.726 -0.929 391s -0.829 0.099 -0.571 391s 391s x1: [3, 2, 2, 1] 391s 0.629 0.366 -0.509 391s -0.424 -0.734 0.885 391s 391s y: [3, 2, 2, 1] 391s 0.358 0.134 -0.194 391s -0.921 -0.947 -0.226 391s test-mul-mat0: iter:173/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.502 -0.857 391s 0.464 -0.908 391s 0.463 0.190 391s 391s x1: [2, 1, 1, 1] 391s 0.045 -0.699 391s 391s y: [3, 1, 1, 1] 391s 0.622 0.656 -0.111 391s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 3, 2, 1] 391s -0.266 -0.410 391s -0.828 -0.853 391s -0.485 0.564 391s 391s x1: [2, 3, 2, 1] 391s -0.295 0.501 391s 0.360 -0.888 391s 0.228 -0.540 391s 391s y: [3, 3, 2, 1] 391s -0.127 -0.184 0.426 391s 0.268 0.459 -0.676 391s 0.161 0.272 -0.415 391s testing: mul_mat, [1, 3, 2, 4] = [2, 1, 2, 4] * [2, 3, 2, 4] 391s x0: [2, 1, 2, 4] 391s 0.962 0.197 391s 391s x1: [2, 3, 2, 4] 391s 0.474 0.376 391s -0.575 0.566 391s -0.161 0.615 391s 391s y: [1, 3, 2, 4] 391s 0.531 391s -0.442 391s -0.034 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.006 -0.054 391s -0.245 0.510 391s 391s x1: [2, 1, 1, 1] 391s -0.116 -0.215 391s 391s y: [2, 1, 1, 1] 391s 0.011 -0.082 391s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 4, 2, 1] 391s 0.331 -0.053 391s -0.818 0.810 391s -0.319 -0.618 391s -0.562 0.895 391s 391s x1: [2, 2, 2, 1] 391s -0.706 0.255 391s -0.378 -0.301 391s 391s y: [4, 2, 2, 1] 391s -0.247 0.784 0.067 0.625 391s -0.109 0.066 0.307 -0.057 391s testing: mul_mat, [4, 2, 2, 4] = [4, 4, 2, 4] * [4, 2, 2, 4] 391s x0: [4, 4, 2, 4] 391s 0.945 0.364 0.624 -0.391 391s -0.694 -0.928 0.359 0.055 391s 0.667 0.520 -0.686 0.600 391s 0.759 -0.523 0.240 0.706 391s 391s x1: [4, 2, 2, 4] 391s -0.803 0.966 -0.941 -0.047 391s -0.088 0.569 0.247 -0.833 391s 391s y: [4, 2, 2, 4] 391s -0.976 -0.680 0.584 -1.374 391s 0.604 -0.424 -0.432 -0.894 391s test-mul-mat0: iter:174/500 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.185 0.171 391s -0.014 -0.749 391s 0.621 0.891 391s -0.130 0.051 391s 391s x1: [2, 1, 1, 1] 391s -0.252 -0.710 391s 391s y: [4, 1, 1, 1] 391s -0.168 0.535 -0.789 -0.004 391s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 1, 4, 1] 391s 0.832 0.908 391s 391s x1: [2, 4, 4, 1] 391s -0.300 0.788 391s -0.410 0.002 391s -0.003 0.315 391s -0.517 0.177 391s 391s y: [1, 4, 4, 1] 391s 0.465 391s -0.340 391s 0.283 391s -0.269 391s testing: mul_mat, [2, 1, 4, 2] = [2, 2, 4, 2] * [2, 1, 4, 2] 391s x0: [2, 2, 4, 2] 391s 0.534 0.387 391s -0.922 -0.224 391s 391s x1: [2, 1, 4, 2] 391s -0.769 -0.961 391s 391s y: [2, 1, 4, 2] 391s -0.782 0.924 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.720 -0.058 391s 391s x1: [2, 2, 1, 1] 391s -0.553 0.524 391s 0.112 -0.260 391s 391s y: [1, 2, 1, 1] 391s -0.429 391s 0.096 391s testing: mul_mat, [2, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 2, 4, 1] 391s -0.498 391s 0.034 391s 391s x1: [1, 2, 4, 1] 391s 0.530 391s 0.476 391s 391s y: [2, 2, 4, 1] 391s -0.264 0.018 391s -0.237 0.016 391s testing: mul_mat, [3, 1, 4, 2] = [2, 3, 4, 2] * [2, 1, 4, 2] 391s x0: [2, 3, 4, 2] 391s 0.494 0.458 391s -0.785 0.641 391s -0.582 -0.707 391s 391s x1: [2, 1, 4, 2] 391s 0.337 -0.129 391s 391s y: [3, 1, 4, 2] 391s 0.107 -0.348 -0.105 391s test-mul-mat0: iter:175/500 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.384 0.373 391s 0.167 0.231 391s 391s x1: [2, 3, 1, 1] 391s 0.776 -0.459 391s -0.824 0.003 391s 0.032 -0.617 391s 391s y: [2, 3, 1, 1] 391s 0.127 0.024 391s -0.316 -0.137 391s -0.218 -0.137 391s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 4, 2, 1] 391s 0.508 0.560 391s 0.136 0.387 391s -0.310 0.071 391s 0.391 -0.345 391s 391s x1: [2, 2, 2, 1] 391s -0.423 -0.536 391s -0.465 0.423 391s 391s y: [4, 2, 2, 1] 391s -0.515 -0.265 0.093 0.019 391s 0.001 0.101 0.174 -0.328 391s testing: mul_mat, [1, 4, 2, 2] = [2, 1, 2, 2] * [2, 4, 2, 2] 391s x0: [2, 1, 2, 2] 391s 0.261 0.920 391s 391s x1: [2, 4, 2, 2] 391s -0.509 0.805 391s -0.269 0.864 391s -0.028 0.962 391s -0.559 0.437 391s 391s y: [1, 4, 2, 2] 391s 0.608 391s 0.725 391s 0.878 391s 0.256 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.402 -0.223 391s 0.311 0.745 391s 0.352 0.056 391s 391s x1: [2, 1, 1, 1] 391s 0.979 -0.986 391s 391s y: [3, 1, 1, 1] 391s -0.174 -0.429 0.289 391s testing: mul_mat, [3, 2, 2, 1] = [3, 3, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 3, 2, 1] 391s -0.399 0.652 -0.860 391s 0.992 0.215 -0.067 391s -0.828 -0.467 -0.242 391s 391s x1: [3, 2, 2, 1] 391s 0.647 0.249 -0.850 391s 0.171 0.492 0.018 391s 391s y: [3, 2, 2, 1] 391s 0.636 0.752 -0.447 391s 0.237 0.274 -0.376 391s testing: mul_mat, [1, 3, 2, 2] = [3, 1, 2, 2] * [3, 3, 2, 2] 391s x0: [3, 1, 2, 2] 391s -0.370 -0.357 0.275 391s 391s x1: [3, 3, 2, 2] 391s 0.140 0.253 -0.650 391s -0.688 -0.255 0.369 391s -0.059 -0.006 0.570 391s 391s y: [1, 3, 2, 2] 391s -0.321 391s 0.447 391s 0.181 391s test-mul-mat0: iter:176/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.659 0.412 391s 391s x1: [2, 4, 1, 1] 391s 0.982 0.209 391s -0.710 -0.274 391s -0.143 -0.924 391s -0.037 -0.492 391s 391s y: [1, 4, 1, 1] 391s -0.561 391s 0.355 391s -0.286 391s -0.179 391s testing: mul_mat, [1, 1, 3, 1] = [2, 1, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.575 0.864 391s 391s x1: [2, 1, 3, 1] 391s -0.601 -0.970 391s 391s y: [1, 1, 3, 1] 391s -0.493 391s testing: mul_mat, [3, 1, 3, 4] = [2, 3, 3, 4] * [2, 1, 3, 4] 391s x0: [2, 3, 3, 4] 391s 0.271 -0.361 391s 0.999 -0.153 391s -0.220 -0.176 391s 391s x1: [2, 1, 3, 4] 391s -0.715 -0.813 391s 391s y: [3, 1, 3, 4] 391s 0.100 -0.590 0.300 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.868 -0.116 391s -0.1ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 25 0.658 391s -0.241 -0.657 391s 391s x1: [2, 3, 1, 1] 391s -0.948 -0.868 391s -0.148 0.598 391s 0.816 -0.673 391s 391s y: [3, 3, 1, 1] 391s -0.722 -0.452 0.798 391s -0.198 0.412 -0.357 391s 0.787 -0.544 0.245 391s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.102 0.971 -0.868 391s 0.805 -0.082 0.200 391s 391s x1: [3, 2, 3, 1] 391s -0.129 -0.764 -0.834 391s -0.100 -0.725 0.073 391s 391s y: [2, 2, 3, 1] 391s -0.031 -0.208 391s -0.777 -0.007 391s testing: mul_mat, [1, 3, 3, 4] = [2, 1, 3, 4] * [2, 3, 3, 4] 391s x0: [2, 1, 3, 4] 391s 0.601 -0.051 391s 391s x1: [2, 3, 3, 4] 391s 0.185 0.995 391s 0.878 0.230 391s 0.861 -0.321 391s 391s y: [1, 3, 3, 4] 391s 0.060 391s 0.515 391s 0.534 391s test-mul-mat0: iter:177/500 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.395 -0.244 0.073 0.843 391s 391s x1: [4, 4, 1, 1] 391s -0.669 0.549 -0.353 0.932 391s -0.502 0.172 -0.413 -0.684 391s -0.091 0.465 -0.319 0.737 391s -0.258 0.225 0.203 0.259 391s 391s y: [1, 4, 1, 1] 391s 0.362 391s -0.847 391s 0.448 391s 0.076 391s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 1, 4, 1] 391s -0.058 0.112 0.740 0.623 391s 391s x1: [4, 1, 4, 1] 391s -0.849 -0.272 0.259 -0.645 391s 391s y: [1, 1, 4, 1] 391s -0.192 391s testing: mul_mat, [3, 1, 4, 3] = [4, 3, 4, 3] * [4, 1, 4, 3] 391s x0: [4, 3, 4, 3] 391s 0.660 0.068 0.327 -0.459 391s 0.844 0.646 -0.842 -0.933 391s -0.537 0.583 0.580 0.927 391s 391s x1: [4, 1, 4, 3] 391s -0.451 0.004 0.214 -0.866 391s 391s y: [3, 1, 4, 3] 391s 0.169 0.248 -0.434 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.676 -0.229 0.120 0.516 391s 0.014 0.722 -0.949 0.342 391s -0.233 -0.317 0.782 0.894 391s -0.614 -0.098 -0.802 -0.690 391s 391s x1: [4, 3, 1, 1] 391s -0.296 0.807 -0.956 0.403 391s -0.922 0.384 -0.583 -0.368 391s 0.727 -0.063 0.461 -0.914 391s 391s y: [4, 3, 1, 1] 391s 0.108 1.624 -0.574 0.591 391s 0.275 0.691 -0.692 1.250 391s -0.894 -0.786 -0.606 -0.179 391s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.936 0.368 -0.455 -0.964 391s -0.251 0.663 0.298 0.007 391s -0.449 -0.538 -0.053 0.047 391s 0.204 -0.261 -0.916 0.063 391s 391s x1: [4, 4, 4, 1] 391s 0.600 0.597 0.014 0.407 391s 0.640 -0.584 0.485 0.024 391s -0.167 -0.883 -0.249 0.770 391s 0.578 -0.163 -0.343 0.902 391s 391s y: [4, 4, 4, 1] 391s -0.741 0.251 -0.572 -0.020 391s -1.059 -0.403 0.002 -0.160 391s -0.798 -0.612 0.599 0.473 391s -1.315 -0.350 -0.112 0.532 391s testing: mul_mat, [4, 4, 4, 3] = [4, 4, 4, 3] * [4, 4, 4, 3] 391s x0: [4, 4, 4, 3] 391s -0.139 -0.944 0.920 0.210 391s 0.581 0.069 -0.056 0.876 391s 0.618 -0.045 -0.482 0.093 391s 0.233 -0.310 -0.180 -0.497 391s 391s x1: [4, 4, 4, 3] 391s -0.848 0.335 0.863 -0.308 391s 0.039 0.438 0.777 -0.960 391s 0.961 -0.054 0.077 0.491 391s 0.162 0.178 -0.291 -0.467 391s 391s y: [4, 4, 4, 3] 391s 0.531 -0.787 -0.983 -0.304 391s 0.095 -0.832 -0.460 0.210 391s 0.091 0.980 0.604 -0.018 391s -0.557 -0.286 0.189 0.267 391s test-mul-mat0: iter:178/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.597 -0.982 0.508 0.585 391s -0.756 0.107 0.028 0.764 391s 391s x1: [4, 1, 1, 1] 391s -0.827 0.929 -0.990 -0.336 391s 391s y: [2, 1, 1, 1] 391s -1.118 0.440 391s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 3, 4, 1] 391s -0.646 0.712 0.052 0.782 391s 0.108 -0.905 0.607 -0.847 391s -0.017 0.100 0.842 -0.144 391s 391s x1: [4, 2, 4, 1] 391s 0.894 -0.683 0.530 0.800 391s -0.923 0.428 0.396 0.043 391s 391s y: [3, 2, 4, 1] 391s -0.411 0.358 0.248 391s 0.955 -0.284 0.385 391s testing: mul_mat, [1, 3, 4, 2] = [4, 1, 4, 2] * [4, 3, 4, 2] 391s x0: [4, 1, 4, 2] 391s -0.357 0.278 -0.645 -0.446 391s 391s x1: [4, 3, 4, 2] 391s -0.102 0.118 -0.200 0.562 391s -0.588 -0.647 0.138 0.336 391s -0.033 -0.268 -0.083 -0.595 391s 391s y: [1, 3, 4, 2] 391s -0.053 391s -0.209 391s 0.256 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.770 0.512 -0.832 -0.318 391s -0.617 0.252 0.347 -0.112 391s 0.682 0.639 0.440 0.803 391s 391s x1: [4, 1, 1, 1] 391s 0.027 0.260 -0.378 0.473 391s 391s y: [3, 1, 1, 1] 391s 0.277 -0.135 0.398 391s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.302 -0.555 0.781 391s 0.825 -0.645 0.198 391s 391s x1: [3, 4, 4, 1] 391s 0.009 0.333 -0.972 391s -0.882 0.632 -0.876 391s 0.627 -0.902 -0.130 391s -0.303 0.524 -0.381 391s 391s y: [2, 4, 4, 1] 391s -0.947 -0.400 391s -0.769 -1.309 391s 0.210 1.073 391s -0.497 -0.663 391s testing: mul_mat, [3, 3, 4, 2] = [2, 3, 4, 2] * [2, 3, 4, 2] 391s x0: [2, ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 3, 4, 2] 391s 0.707 -0.761 391s 0.069 0.436 391s 0.627 0.575 391s 391s x1: [2, 3, 4, 2] 391s 0.926 -0.866 391s 0.361 0.752 391s -0.144 -0.258 391s 391s y: [3, 3, 4, 2] 391s 1.314 -0.314 0.083 391s -0.317 0.353 0.658 391s 0.094 -0.123 -0.239 391s test-mul-mat0: iter:179/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.270 -0.303 0.926 -0.082 391s 391s x1: [4, 3, 1, 1] 391s 0.413 0.229 -0.892 0.128 391s -0.647 0.215 -0.983 0.261 391s -0.483 0.044 0.876 -0.111 391s 391s y: [1, 3, 1, 1] 391s -1.017 391s -0.823 391s 0.938 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.664 0.180 -0.054 -0.266 391s -0.730 -0.968 0.353 -0.476 391s 391s x1: [4, 1, 1, 1] 391s -0.505 0.695 -0.848 -0.420 391s 391s y: [2, 1, 1, 1] 391s 0.618 -0.403 391s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 391s x0: [4, 4, 1, 2] 391s -0.107 0.205 -0.727 0.045 391s 0.785 0.516 0.381 -0.035 391s -0.538 -0.885 0.235 -0.506 391s 0.467 0.760 0.778 -0.767 391s 391s x1: [4, 2, 1, 2] 391s 0.284 -0.235 0.754 0.392 391s 0.893 -0.894 -0.393 0.910 391s 391s y: [4, 2, 1, 2] 391s -0.609 0.375 0.034 0.240 391s 0.048 0.058 -0.242 -1.266 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.464 0.707 0.687 0.499 391s 391s x1: [4, 4, 1, 1] 391s -0.171 -0.509 -0.857 -0.386 391s -0.993 0.524 0.579 -0.531 391s 0.638 -0.186 -0.037 0.106 391s -0.426 -0.259 0.338 -0.913 391s 391s y: [1, 4, 1, 1] 391s -1.221 391s 0.043 391s 0.192 391s -0.604 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.872 391s 0.131 391s 391s x1: [1, 4, 1, 1] 391s -0.305 391s -0.413 391s -0.512 391s -0.074 391s 391s y: [2, 4, 1, 1] 391s 0.266 -0.040 391s 0.360 -0.054 391s 0.447 -0.067 391s 0.065 -0.010 391s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 391s x0: [2, 4, 1, 2] 391s -0.628 -0.854 391s -0.433 0.312 391s -0.157 0.534 391s 0.896 -0.040 391s 391s x1: [2, 1, 1, 2] 391s 0.124 -0.654 391s 391s y: [4, 1, 1, 2] 391s 0.481 -0.258 -0.369 0.137 391s test-mul-mat0: iter:180/500 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.902 -0.690 391s 0.779 -0.261 391s 391s x1: [2, 3, 1, 1] 391s 0.201 0.429 391s -0.050 -0.298 391s 0.557 -0.919 391s 391s y: [2, 3, 1, 1] 391s -0.114 0.045 391s 0.160 0.039 391s 1.136 0.673 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.247 -0.014 391s -0.814 -0.269 391s 391s x1: [2, 2, 2, 1] 391s -0.318 -0.654 391s 0.582 -0.421 391s 391s y: [2, 2, 2, 1] 391s 0.088 0.434 391s -0.138 -0.360 391s testing: mul_mat, [4, 2, 2, 2] = [2, 4, 2, 2] * [2, 2, 2, 2] 391s x0: [2, 4, 2, 2] 391s -0.438 -0.928 391s 0.216 -0.325 391s 0.524 -0.614 391s 0.428 -0.490 391s 391s x1: [2, 2, 2, 2] 391s 0.728 0.114 391s 0.960 -0.071 391s 391s y: [4, 2, 2, 2] 391s -0.425 0.120 0.311 0.255 391s -0.355 0.231 0.547 0.446 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.468 -0.574 391s 0.013 -0.199 391s 0.355 0.536 391s -0.478 -0.231 391s 391s x1: [2, 4, 1, 1] 391s -0.683 -0.632 391s 0.435 0.841 391s -0.246 -0.137 391s -0.649 0.327 391s 391s y: [4, 4, 1, 1] 391s 0.043 0.118 -0.581 0.473 391s -0.279 -0.162 0.605 -0.403 391s -0.037 0.024 -0.160 0.149 391s -0.491 -0.073 -0.055 0.235 391s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 3, 2, 1] 391s 0.240 0.590 0.058 0.580 391s -0.932 0.394 -0.593 -0.167 391s 0.849 -0.128 -0.774 0.027 391s 391s x1: [4, 2, 2, 1] 391s -0.211 0.460 -0.258 -0.123 391s -0.625 -0.829 -0.035 -0.409 391s 391s y: [3, 2, 2, 1] 391s 0.135 0.552 -0.042 391s -0.878 0.344 -0.408 391s testing: mul_mat, [3, 4, 2, 2] = [3, 3, 2, 2] * [3, 4, 2, 2] 391s x0: [3, 3, 2, 2] 391s 0.536 -0.806 0.544 391s 0.650 -0.359 -0.416 391s 0.536 0.896 -0.230 391s 391s x1: [3, 4, 2, 2] 391s -0.668 -0.568 -0.602 391s 0.883 0.831 0.712 391s -0.131 0.070 0.780 391s -0.282 -0.339 0.174 391s 391s y: [3, 4, 2, 2] 391s -0.227 0.020 -0.729 391s 0.190 -0.020 1.055 391s 0.297 -0.435 -0.186 391s 0.218 -0.134 -0.495 391s test-mul-mat0: iter:181/500 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.512 0.934 -0.838 391s 391s x1: [3, 2, 1, 1] 391s -0.160 -0.962 -0.861 391s 0.852 0.925 0.793 391s 391s y: [1, 2, 1, 1] 391s -0.259 391s 0.636 391s testing: mul_mat, [4, 1, 3, 1] = [3, 4, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 4, 3, 1] 391s -0.492 0.627 0.000 391s 0.252 -0.896 0.745 391s -0.362 -0.523 -0.967 391s -0.393 0.620 -0.126 391s 391s x1: [3, 1, 3, 1] 391s 0.668 0.635 -0.048 391s 391s y: [4, 1, 3, 1] 391s 0.069 -0.436 -0.528 0.138 391s testing: mul_mat, [4, 4, 3, 3] = [3, 4, 3, 3] * [3, 4, 3, 3] 391s x0: [3, 4, 3, 3] 391s -0.691 0.150 0.395 391s 0.864 0.751 -0.827 ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 391s -0.611 0.013 0.450 391s 0.937 0.565 -0.550 391s 391s x1: [3, 4, 3, 3] 391s 0.493 -0.511 0.722 391s 0.526 0.096 0.342 391s -0.600 0.742 0.102 391s 0.126 -0.688 -0.346 391s 391s y: [4, 4, 3, 3] 391s -0.132 -0.555 0.017 -0.224 391s -0.214 0.244 -0.166 0.359 391s 0.566 -0.046 0.422 -0.199 391s -0.327 -0.122 -0.241 -0.080 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.745 0.964 0.921 391s -0.698 -0.621 -0.945 391s -0.699 -0.367 0.020 391s 391s x1: [3, 4, 1, 1] 391s -0.493 0.845 0.958 391s 0.865 -0.068 -0.839 391s 0.722 -0.754 0.633 391s 0.974 -0.770 0.059 391s 391s y: [3, 4, 1, 1] 391s 1.329 -1.086 0.054 391s -0.194 0.231 -0.597 391s 0.395 -0.634 -0.216 391s 0.038 -0.257 -0.398 391s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.493 1.000 0.577 391s 391s x1: [3, 3, 3, 1] 391s 0.042 0.110 0.912 391s -0.312 -0.115 -0.791 391s -0.470 -0.390 -0.547 391s 391s y: [1, 3, 3, 1] 391s 0.656 391s -0.725 391s -0.937 391s testing: mul_mat, [3, 3, 3, 3] = [1, 3, 3, 3] * [1, 3, 3, 3] 391s x0: [1, 3, 3, 3] 391s -0.626 391s 0.214 391s 0.782 391s 391s x1: [1, 3, 3, 3] 391s 0.753 391s -0.395 391s -0.850 391s 391s y: [3, 3, 3, 3] 391s -0.471 0.161 0.589 391s 0.247 -0.084 -0.309 391s 0.532 -0.181 -0.665 391s test-mul-mat0: iter:182/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.738 0.214 0.079 -0.244 391s 0.823 -0.595 -0.618 -0.249 391s 391s x1: [4, 1, 1, 1] 391s -0.830 -0.618 -0.671 -0.703 391s 391s y: [2, 1, 1, 1] 391s 0.600 0.274 391s testing: mul_mat, [2, 2, 4, 1] = [4, 2, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.380 -0.359 -0.316 -0.641 391s -0.294 0.324 -0.815 -0.203 391s 391s x1: [4, 2, 4, 1] 391s 0.812 0.346 -0.405 0.163 391s -0.090 -0.021 -0.935 -0.360 391s 391s y: [2, 2, 4, 1] 391s 0.208 0.171 391s 0.500 0.855 391s testing: mul_mat, [1, 2, 4, 2] = [4, 1, 4, 2] * [4, 2, 4, 2] 391s x0: [4, 1, 4, 2] 391s -0.264 -0.697 -0.970 0.587 391s 391s x1: [4, 2, 4, 2] 391s 0.036 0.983 -0.464 0.741 391s 0.307 -0.280 -0.461 0.110 391s 391s y: [1, 2, 4, 2] 391s 0.191 391s 0.626 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.136 -0.748 -0.304 0.816 391s -0.355 0.058 -0.608 -0.914 391s 0.992 -0.764 -0.626 0.442 391s -0.295 0.837 0.070 0.158 391s 391s x1: [4, 1, 1, 1] 391s 0.250 -0.122 -0.374 -0.965 391s 391s y: [4, 1, 1, 1] 391s -0.548 1.014 0.149 -0.354 391s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.444 -0.865 -0.140 0.885 391s 391s x1: [4, 4, 4, 1] 391s -0.887 -0.406 0.866 0.993 391s -0.389 0.565 -0.704 0.559 391s -0.586 0.335 -0.192 0.292 391s 0.961 -0.156 0.677 0.097 391s 391s y: [1, 4, 4, 1] 391s 0.714 391s -0.068 391s -0.265 391s 0.553 391s testing: mul_mat, [4, 4, 4, 2] = [1, 4, 4, 2] * [1, 4, 4, 2] 391s x0: [1, 4, 4, 2] 391s 0.859 391s 0.328 391s -0.210 391s -0.496 391s 391s x1: [1, 4, 4, 2] 391s -0.884 391s 0.369 391s 0.540 391s 0.276 391s 391s y: [4, 4, 4, 2] 391s -0.760 -0.290 0.185 0.439 391s 0.318 0.121 -0.078 -0.183 391s 0.464 0.177 -0.113 -0.268 391s 0.237 0.091 -0.058 -0.137 391s test-mul-mat0: iter:183/500 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.082 0.468 391s -0.095 0.604 391s 391s x1: [2, 1, 1, 1] 391s 0.955 -0.778 391s 391s y: [2, 1, 1, 1] 391s -0.286 -0.560 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.694 0.772 391s 0.729 0.346 391s 391s x1: [2, 2, 2, 1] 391s 0.486 0.278 391s 0.657 0.136 391s 391s y: [2, 2, 2, 1] 391s -0.123 0.450 391s -0.351 0.526 391s testing: mul_mat, [2, 2, 2, 3] = [2, 2, 2, 3] * [2, 2, 2, 3] 391s x0: [2, 2, 2, 3] 391s 0.369 0.610 391s 0.698 0.561 391s 391s x1: [2, 2, 2, 3] 391s -0.292 -0.162 391s 0.396 -0.447 391s 391s y: [2, 2, 2, 3] 391s -0.207 -0.295 391s -0.127 0.025 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.505 -0.895 391s 0.734 0.432 391s -0.445 -0.883 391s 391s x1: [2, 2, 1, 1] 391s 0.738 0.545 391s -0.770 -0.150 391s 391s y: [3, 2, 1, 1] 391s -0.861 0.777 -0.810 391s 0.524 -0.630 0.475 391s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.232 0.585 -0.484 391s 0.707 -0.929 -0.309 391s 391s x1: [3, 2, 2, 1] 391s -0.886 0.304 -0.614 391s -0.038 -0.428 0.209 391s 391s y: [2, 2, 2, 1] 391s 0.269 -0.719 391s -0.360 0.306 391s testing: mul_mat, [2, 3, 2, 3] = [2, 2, 2, 3] * [2, 3, 2, 3] 391s x0: [2, 2, 2, 3] 391s 0.537 -0.737 391s -0.892 0.784 391s 391s x1: [2, 3, 2, 3] 391s 0.066 -0.162 391s 0.978 0.171 391s -0.730 -0.905 391s 391s y: [2, 3, 2, 3] 391s 0.155 -0.186 391s 0.400 -0.739 391s 0.275 -0.058 391s test-mul-mat0: iter:184/500 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.306 0.516 0.480 -0.859 391s -0.326 0.112 -0.252 -0.137 391s -0.706 -0.287 -0.848 0.544 391s 391s x1: [4, 2, 1, 1] 391s -0.921 0.793 0.476 0.617 391s 0.901 0.739 0.401 -0.845 391s 391s y: [3, 2, 1, 1] 391s -0.174 0.184 0.353 391s 1.575 -0.196 -1.648 391s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.244 0.501 -0.160 0.871 391s 0.502 -0.751 -0.902 -0.749 391s -0.479 -0.381 0.713 -0.056 391s -0.316 0.901 -0.749 0.863 391s 391s x1: [4, 3, 4, 1] 391s -0.274 -0.116 0.114 0.025 391s 0.630 -0.890 0.033 0.458 391s 0.370 -0.273 0.537 0.163 391s 391s y: [4, 3, 4, 1] 391s 0.013 -0.172 0.255 -0.082 391s -0.206 0.612 0.034 -0.631 391s -0.170 -0.216 0.300 -0.623 391s testing: mul_mat, [3, 4, 4, 3] = [4, 3, 4, 3] * [4, 4, 4, 3] 391s x0: [4, 3, 4, 3] 391s 0.956 -0.864 0.744 0.831 391s 0.311 -0.373 0.859 0.893 391s -0.424 -0.951 0.105 -0.428 391s 391s x1: [4, 4, 4, 3] 391s 0.727 -0.283 0.556 0.441 391s -0.924 0.053 0.676 -0.946 391s -0.738 -0.146 -0.284 0.372 391s -0.716 0.241 0.597 -0.097 391s 391s y: [3, 4, 4, 3] 391s 1.721 1.204 -0.169 391s -1.213 -0.572 0.818 391s -0.481 -0.087 0.263 391s -0.530 0.113 0.179 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.231 0.225 0.086 -0.387 391s 0.041 0.479 -0.092 -0.202 391s -0.722 -0.278 0.913 0.625 391s 0.206 0.850 0.213 -0.561 391s 391s x1: [4, 3, 1, 1] 391s 0.072 -0.770 0.507 -0.669 391s 0.603 0.408 -0.797 0.406 391s 0.274 0.608 -0.015 -0.695 391s 391s y: [4, 3, 1, 1] 391s 0.145 -0.277 0.207 -0.157 391s 0.006 0.211 -1.022 0.073 391s 0.468 0.445 -0.815 0.961 391s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.729 0.541 -0.062 -0.762 391s 391s x1: [4, 4, 4, 1] 391s -0.462 0.511 0.511 -0.232 391s 0.018 0.842 -0.629 -0.574 391s -0.954 0.777 0.700 0.654 391s -0.238 -0.995 0.574 0.993 391s 391s y: [1, 4, 4, 1] 391s 0.085 391s 0.945 391s -0.817 391s -1.504 391s testing: mul_mat, [2, 4, 4, 3] = [1, 2, 4, 3] * [1, 4, 4, 3] 391s x0: [1, 2, 4, 3] 391s -0.851 391s 0.828 391s 391s x1: [1, 4, 4, 3] 391s 0.767 391s 0.712 391s 0.014 391s 0.121 391s 391s y: [2, 4, 4, 3] 391s -0.653 0.635 391s -0.606 0.590 391s -0.012 0.011 391s -0.103 0.100 391s test-mul-mat0: iter:185/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.050 -0.703 391s 391s x1: [2, 4, 1, 1] 391s -0.403 0.467 391s 0.660 -0.254 391s 0.296 0.158 391s 0.707 -0.619 391s 391s y: [1, 4, 1, 1] 391s -0.349 391s 0.211 391s -0.096 391s 0.470 391s testing: mul_mat, [3, 1, 2, 1] = [2, 3, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 3, 2, 1] 391s 0.993 -0.512 391s -0.367 -0.340 391s 0.733 -0.446 391s 391s x1: [2, 1, 2, 1] 391s -0.905 0.390 391s 391s y: [3, 1, 2, 1] 391s -1.099 0.199 -0.838 391s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.962 0.143 391s 0.239 0.617 391s -0.642 -0.181 391s 0.610 -0.154 391s 391s x1: [2, 3, 2, 1] 391s 0.748 -0.404 391s -0.185 -0.654 391s -0.936 -0.525 391s 391s y: [4, 3, 2, 1] 391s -0.778 -0.070 -0.408 0.519 391s 0.085 -0.448 0.237 -0.012 391s 0.826 -0.548 0.696 -0.490 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.534 0.358 391s 0.748 -0.793 391s 391s x1: [2, 4, 1, 1] 391s -0.553 -0.174 391s 0.122 0.043 391s 0.641 0.468 391s 0.107 -0.885 391s 391s y: [2, 4, 1, 1] 391s 0.233 -0.276 391s -0.050 0.057 391s -0.174 0.108 391s -0.374 0.781 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.142 0.742 391s -0.283 0.721 391s 391s x1: [2, 2, 2, 1] 391s -0.204 0.207 391s -0.755 0.939 391s 391s y: [2, 2, 2, 1] 391s 0.124 0.207 391s 0.589 0.890 391s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.375 0.870 391s 391s x1: [2, 2, 2, 1] 391s 0.476 0.254 391s 0.330 -0.402 391s 391s y: [1, 2, 2, 1] 391s 0.043 391s -0.474 391s test-mul-mat0: iter:186/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.574 0.087 0.145 0.639 391s -0.509 -0.769 -0.736 -0.639 391s 0.065 -0.753 -0.562 -0.305 391s 0.436 -0.239 -0.737 0.071 391s 391s x1: [4, 4, 1, 1] 391s 0.568 0.635 -0.815 0.865 391s -0.100 0.657 0.007 0.617 391s 0.399 -0.272 -0.660 0.501 391s 0.535 -0.155 -0.023 -0.210 391s 391s y: [4, 4, 1, 1] 391s 0.816 -0.731 -0.248 0.758 391s 0.395 -0.853 -0.694 -0.162 391s 0.429 0.171 0.449 0.761 391s 0.156 -0.002 0.229 0.273 391s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.402 -0.902 -0.662 -0.447 391s -0.776 0.728 -0.245 -0.213 391s 0.813 -0.607 -0.888 -0.839 391s 0.102 -0.618 -0.364 0.782 391s 391s x1: [4, 4, 3, 1] 391s -0.054 -0.872 0.971 -0.398 391s 0.135 0.588 -0.999 0.863 391s 0.929 0.50ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 2 0.398 -0.226 391s -0.522 -0.812 -0.052 -0.948 391s 391s y: [4, 4, 3, 1] 391s 0.343 -0.747 -0.043 -0.131 391s -0.309 0.385 -0.085 0.689 391s -0.988 -0.405 0.287 -0.537 391s 1.400 0.027 0.910 -0.274 391s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.145 0.545 -0.595 -0.647 391s -0.449 -0.480 0.055 -0.249 391s -0.698 0.344 -0.219 0.532 391s -0.525 -0.113 -0.738 -0.424 391s 391s x1: [4, 4, 3, 1] 391s -0.923 0.500 -0.029 0.797 391s 0.749 0.476 -0.613 -0.031 391s -0.165 -0.963 -0.891 0.100 391s -0.980 0.262 0.963 -0.899 391s 391s y: [4, 4, 3, 1] 391s -0.092 -0.025 1.247 0.113 391s 0.535 -0.591 -0.241 0.018 391s -0.035 0.462 0.033 0.811 391s 0.293 0.592 0.085 0.155 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.622 -0.150 -0.334 -0.043 391s -0.288 0.693 0.454 -0.897 391s 0.770 0.404 -0.430 0.998 391s 391s x1: [4, 4, 1, 1] 391s -0.831 0.021 -0.478 -0.643 391s 0.587 0.932 -0.897 0.665 391s 0.690 -0.334 0.218 0.898 391s -0.199 0.625 -0.038 -0.519 391s 391s y: [3, 4, 1, 1] 391s 0.701 0.613 -1.068 391s -0.234 -0.528 1.878 391s -0.490 -1.137 1.199 391s 0.065 0.938 -0.401 391s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.500 -0.404 0.911 391s -0.858 0.829 0.414 391s -0.565 0.727 0.821 391s 391s x1: [3, 4, 3, 1] 391s 0.824 0.543 -0.834 391s -0.155 -0.935 -0.476 391s -0.568 0.997 -0.374 391s -0.903 0.687 0.292 391s 391s y: [3, 4, 3, 1] 391s -1.391 -0.603 -0.755 391s 0.021 -0.839 -0.983 391s -0.459 1.159 0.739 391s 0.441 1.466 1.250 391s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.610 -0.583 0.866 391s -0.925 0.842 -0.728 391s 391s x1: [3, 3, 3, 1] 391s 0.452 0.792 0.855 391s 0.261 -0.709 0.996 391s 0.696 -0.113 0.825 391s 391s y: [2, 3, 3, 1] 391s 0.003 -0.374 391s 1.117 -1.564 391s 0.356 -1.340 391s test-mul-mat0: iter:187/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.152 -0.849 -0.293 0.245 391s -0.626 -0.450 -0.424 0.841 391s -0.959 0.108 -0.053 0.408 391s 0.725 0.375 -0.975 0.089 391s 391s x1: [4, 4, 1, 1] 391s -0.383 0.364 -0.590 0.793 391s 0.213 0.487 -0.983 0.563 391s -0.806 -0.593 0.638 -0.389 391s -0.751 0.504 -0.117 0.352 391s 391s y: [4, 4, 1, 1] 391s 0.000 0.993 0.761 0.504 391s 0.045 0.538 0.131 1.345 391s 0.099 0.174 0.516 -1.463 391s -0.421 0.589 0.924 -0.211 391s testing: mul_mat, [4, 4, 2, 1] = [4, 4, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 4, 2, 1] 391s -0.843 0.589 0.055 -0.978 391s 0.056 0.297 0.525 0.133 391s 0.788 -0.467 -0.905 0.632 391s 0.202 -0.793 0.627 -0.422 391s 391s x1: [4, 4, 2, 1] 391s 0.785 0.819 -0.698 0.272 391s 0.835 0.865 0.466 -0.758 391s 0.503 -0.922 -0.509 0.007 391s -0.039 0.844 -0.331 -0.887 391s 391s y: [4, 4, 2, 1] 391s -0.484 -0.044 1.041 -1.043 391s 0.572 0.447 -0.647 0.095 391s -1.002 -0.512 1.293 0.510 391s 1.379 -0.044 -0.686 -0.509 391s testing: mul_mat, [3, 4, 2, 4] = [4, 3, 2, 4] * [4, 4, 2, 4] 391s x0: [4, 3, 2, 4] 391s -0.243 0.217 0.111 -0.974 391s 0.034 -0.333 -0.674 -0.636 391s 0.701 -0.263 0.621 -0.165 391s 391s x1: [4, 4, 2, 4] 391s 0.800 0.521 0.343 -0.144 391s -0.182 -0.132 0.989 -0.394 391s 0.401 -0.916 -0.762 -0.398 391s -0.709 0.865 0.180 0.856 391s 391s y: [3, 4, 2, 4] 391s 0.097 -0.285 0.660 391s 0.509 -0.379 0.586 391s 0.006 1.085 0.114 391s -0.454 -0.977 -0.754 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.203 -0.610 -0.716 -0.842 391s 0.517 0.710 0.132 0.296 391s -0.128 -0.907 0.897 -0.543 391s 391s x1: [4, 3, 1, 1] 391s 0.923 -0.692 0.492 0.759 391s -0.378 -0.607 0.743 0.045 391s -0.949 0.896 -0.854 0.227 391s 391s y: [3, 3, 1, 1] 391s -0.757 0.275 0.539 391s -0.123 -0.515 1.241 391s 0.067 0.101 -1.581 391s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 4, 2, 1] 391s -0.852 0.549 0.980 391s -0.747 0.764 -0.568 391s 0.307 -0.379 -0.821 391s -0.693 0.624 -0.290 391s 391s x1: [3, 4, 2, 1] 391s -0.686 -0.926 -0.784 391s 0.811 -0.444 -0.269 391s 0.734 -0.136 -0.777 391s 0.492 0.486 -0.385 391s 391s y: [4, 4, 2, 1] 391s -0.692 0.251 0.784 0.125 391s -1.199 -0.792 0.639 -0.761 391s -1.461 -0.210 0.915 -0.368 391s -0.529 0.222 0.282 0.073 391s testing: mul_mat, [1, 3, 2, 4] = [4, 1, 2, 4] * [4, 3, 2, 4] 391s x0: [4, 1, 2, 4] 391s 0.664 0.079 -0.537 -0.324 391s 391s x1: [4, 3, 2, 4] 391s -0.531 0.212 0.518 0.786 391s 0.076 -0.361 0.196 0.225 391s -0.109 -0.497 0.531 -0.559 391s 391s y: [1, 3, 2, 4] 391s -0.868 391s -0.156 391s -0.215 391s test-mul-mat0: iter:188/500 391s testing: mul_mat, [3, 2, 1, 1] = [1,ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.851 391s -0.710 391s 0.101 391s 391s x1: [1, 2, 1, 1] 391s -0.374 391s -0.544 391s 391s y: [3, 2, 1, 1] 391s -0.318 0.265 -0.038 391s -0.463 0.386 -0.055 391s testing: mul_mat, [3, 3, 2, 1] = [1, 3, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.923 391s -0.964 391s -0.292 391s 391s x1: [1, 3, 2, 1] 391s -0.977 391s 0.007 391s -0.926 391s 391s y: [3, 3, 2, 1] 391s 0.902 0.942 0.285 391s -0.007 -0.007 -0.002 391s 0.854 0.893 0.270 391s testing: mul_mat, [3, 3, 2, 4] = [1, 3, 2, 4] * [1, 3, 2, 4] 391s x0: [1, 3, 2, 4] 391s 0.093 391s 0.360 391s -0.995 391s 391s x1: [1, 3, 2, 4] 391s -0.790 391s 0.208 391s 0.846 391s 391s y: [3, 3, 2, 4] 391s -0.073 -0.285 0.786 391s 0.019 0.075 -0.206 391s 0.078 0.305 -0.842 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.821 391s -0.116 391s 0.918 391s 0.271 391s 391s x1: [1, 3, 1, 1] 391s 0.418 391s 0.164 391s 0.559 391s 391s y: [4, 3, 1, 1] 391s -0.343 -0.049 0.384 0.113 391s -0.134 -0.019 0.150 0.044 391s -0.459 -0.065 0.514 0.152 391s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.622 -0.455 -0.778 0.866 391s -0.355 0.068 -0.599 -0.615 391s -0.788 0.588 -0.597 -0.492 391s 391s x1: [4, 1, 2, 1] 391s -0.756 0.924 0.202 -0.175 391s 391s y: [3, 1, 2, 1] 391s -0.258 0.318 1.105 391s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 391s x0: [3, 2, 2, 4] 391s -0.709 0.818 0.914 391s 0.153 0.933 -0.469 391s 391s x1: [3, 4, 2, 4] 391s 0.581 0.631 0.781 391s -0.078 -0.754 -0.513 391s -0.430 -0.376 0.132 391s -0.218 0.169 -0.800 391s 391s y: [2, 4, 2, 4] 391s 0.819 0.312 391s -1.031 -0.474 391s 0.118 -0.479 391s -0.439 0.499 391s test-mul-mat0: iter:189/500 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.780 0.429 0.476 391s 0.029 0.137 0.682 391s -0.652 0.497 0.610 391s 391s x1: [3, 4, 1, 1] 391s -0.145 -0.949 -0.437 391s 0.911 -0.623 0.926 391s -0.540 -0.994 0.686 391s 0.410 0.927 -0.400 391s 391s y: [3, 4, 1, 1] 391s -0.728 -0.433 -0.644 391s 0.884 0.572 -0.340 391s -0.520 0.316 0.276 391s 0.527 -0.134 -0.050 391s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.326 0.014 -0.967 391s 0.134 -0.542 0.982 391s 0.504 -0.723 0.857 391s 391s x1: [3, 3, 4, 1] 391s -0.116 -0.927 0.789 391s 0.275 0.325 0.672 391s 0.605 0.393 0.386 391s 391s y: [3, 3, 4, 1] 391s -0.738 1.262 1.287 391s -0.735 0.521 0.480 391s -0.565 0.247 0.352 391s testing: mul_mat, [2, 3, 4, 3] = [3, 2, 4, 3] * [3, 3, 4, 3] 391s x0: [3, 2, 4, 3] 391s -0.519 -0.413 0.218 391s -0.204 -0.621 -0.166 391s 391s x1: [3, 3, 4, 3] 391s 0.285 0.033 0.786 391s 0.141 0.368 0.889 391s -0.441 0.943 0.745 391s 391s y: [2, 3, 4, 3] 391s 0.010 -0.209 391s -0.032 -0.405 391s 0.002 -0.620 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.070 -0.746 0.554 391s 391s x1: [3, 2, 1, 1] 391s 0.726 -0.059 0.967 391s -0.764 0.541 0.030 391s 391s y: [1, 2, 1, 1] 391s 0.630 391s -0.441 391s testing: mul_mat, [2, 3, 4, 1] = [1, 2, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.274 391s -0.144 391s 391s x1: [1, 3, 4, 1] 391s 0.426 391s -0.129 391s 0.791 391s 391s y: [2, 3, 4, 1] 391s 0.117 -0.061 391s -0.035 0.019 391s 0.216 -0.114 391s testing: mul_mat, [4, 1, 4, 3] = [2, 4, 4, 3] * [2, 1, 4, 3] 391s x0: [2, 4, 4, 3] 391s -0.564 -0.898 391s 0.642 0.118 391s -0.935 0.216 391s 0.963 0.504 391s 391s x1: [2, 1, 4, 3] 391s 0.253 0.401 391s 391s y: [4, 1, 4, 3] 391s -0.503 0.210 -0.150 0.446 391s test-mul-mat0: iter:190/500 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.111 -0.991 391s -0.226 0.852 391s 0.461 0.977 391s 391s x1: [2, 3, 1, 1] 391s 0.794 -0.773 391s -0.279 -0.166 391s 0.570 0.768 391s 391s y: [3, 3, 1, 1] 391s 0.678 -0.838 -0.389 391s 0.195 -0.078 -0.291 391s -0.824 0.526 1.013 391s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.020 0.877 391s 0.636 -0.450 391s 0.645 -0.836 391s 0.439 0.654 391s 391s x1: [2, 3, 3, 1] 391s -0.151 0.599 391s -0.064 -0.077 391s 0.198 -0.966 391s 391s y: [4, 3, 3, 1] 391s 0.528 -0.365 -0.598 0.326 391s -0.066 -0.006 0.023 -0.078 391s -0.851 0.560 0.935 -0.545 391s testing: mul_mat, [2, 4, 3, 1] = [2, 2, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.582 -0.267 391s -0.605 -0.936 391s 391s x1: [2, 4, 3, 1] 391s -0.459 0.091 391s -0.459 0.549 391s 0.843 0.698 391s -0.649 -0.177 391s 391s y: [2, 4, 3, 1] 391s -0.292 0.193 391s -0.413 -0.236 391s 0.305 -1.164 391s -0.331 0.558 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.148 0.381 391s -0.356 -0.977 391s 391s x1: [2, 2, 1, 1] 391s -0.495 0.195 391s 0.927 -0.508 391s 391s y: [2, 2, 1, 1] 391s 0.001 -0.014 391s -0.057 0.166 391s testing: mul_mat, [3, 2, 3, 1ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ] = [2, 3, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.316 0.289 391s 0.507 -0.964 391s 0.845 -0.502 391s 391s x1: [2, 2, 3, 1] 391s 0.734 0.485 391s -0.987 0.739 391s 391s y: [3, 2, 3, 1] 391s -0.092 -0.096 0.377 391s 0.525 -1.212 -1.204 391s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.841 -0.586 0.637 391s -0.966 0.788 0.191 391s 391s x1: [3, 2, 3, 1] 391s -0.184 0.038 0.961 391s -0.445 -0.232 -0.321 391s 391s y: [2, 2, 3, 1] 391s 0.435 0.391 391s -0.443 0.186 391s test-mul-mat0: iter:191/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.315 0.776 -0.235 -0.506 391s -0.109 -0.926 0.857 0.953 391s -0.801 -0.742 -0.024 -0.121 391s 391s x1: [4, 3, 1, 1] 391s 0.400 0.799 0.540 -0.489 391s 0.389 -0.394 -0.091 0.014 391s -0.829 -0.250 0.048 -0.415 391s 391s y: [3, 3, 1, 1] 391s 0.614 -0.787 -0.868 391s -0.414 0.258 -0.018 391s 0.266 -0.033 0.898 391s testing: mul_mat, [2, 3, 4, 1] = [4, 2, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.425 0.929 0.889 0.849 391s -0.250 0.486 0.197 0.427 391s 391s x1: [4, 3, 4, 1] 391s 0.426 -0.323 0.710 -0.600 391s -0.607 -0.972 0.801 -0.809 391s 0.568 -0.689 0.580 -0.825 391s 391s y: [2, 3, 4, 1] 391s -0.359 -0.380 391s -0.619 -0.508 391s -1.067 -0.716 391s testing: mul_mat, [4, 2, 4, 3] = [4, 4, 4, 3] * [4, 2, 4, 3] 391s x0: [4, 4, 4, 3] 391s 0.056 -0.500 0.142 0.142 391s -0.778 0.981 -0.298 -0.313 391s -0.553 0.022 0.817 -0.705 391s 0.755 0.011 0.327 -0.304 391s 391s x1: [4, 2, 4, 3] 391s 0.006 -0.562 0.504 0.756 391s 0.924 -0.299 0.184 -0.039 391s 391s y: [4, 2, 4, 3] 391s 0.461 -0.944 -0.137 -0.066 391s 0.222 -1.054 -0.340 0.766 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.657 -0.710 -0.410 0.135 391s 0.841 -0.261 0.024 -0.895 391s -0.871 -0.116 0.592 0.711 391s 391s x1: [4, 4, 1, 1] 391s 0.147 0.865 0.181 -0.207 391s -0.441 -0.452 -0.948 0.417 391s 0.558 0.777 0.252 -0.839 391s 0.688 0.218 0.841 0.319 391s 391s y: [3, 4, 1, 1] 391s -0.813 0.087 -0.268 391s 1.055 -0.648 0.172 391s -1.135 1.022 -1.022 391s -0.908 0.256 0.101 391s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.404 0.439 -0.642 391s -0.079 0.151 0.972 391s 0.899 -0.033 0.058 391s 391s x1: [3, 4, 4, 1] 391s 0.001 -0.692 -0.142 391s -0.133 0.489 0.651 391s 0.426 -0.963 0.703 391s -0.158 0.595 0.481 391s 391s y: [3, 4, 4, 1] 391s -0.212 -0.243 0.016 391s -0.258 0.717 -0.098 391s -0.702 0.504 0.456 391s -0.111 0.569 -0.134 391s testing: mul_mat, [3, 3, 4, 3] = [3, 3, 4, 3] * [3, 3, 4, 3] 391s x0: [3, 3, 4, 3] 391s -0.118 -0.599 -0.714 391s 0.283 0.953 -0.466 391s 0.478 -0.039 0.201 391s 391s x1: [3, 3, 4, 3] 391s -0.233 -0.067 0.816 391s 0.825 0.801 0.743 391s 0.079 -0.109 -0.373 391s 391s y: [3, 3, 4, 3] 391s -0.515 -0.511 0.056 391s -1.108 0.651 0.512 391s 0.322 0.093 -0.033 391s test-mul-mat0: iter:192/500 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.263 391s -0.579 391s -0.602 391s -0.085 391s 391s x1: [1, 4, 1, 1] 391s -0.549 391s 0.386 391s 0.891 391s -0.308 391s 391s y: [4, 4, 1, 1] 391s 0.144 0.318 0.331 0.047 391s -0.102 -0.224 -0.233 -0.033 391s -0.234 -0.516 -0.536 -0.076 391s 0.081 0.178 0.185 0.026 391s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 3, 3, 1] 391s -0.934 391s 0.650 391s -0.512 391s 391s x1: [1, 4, 3, 1] 391s -0.267 391s -0.546 391s -0.842 391s 0.609 391s 391s y: [3, 4, 3, 1] 391s 0.250 -0.174 0.137 391s 0.510 -0.355 0.279 391s 0.787 -0.548 0.431 391s -0.569 0.396 -0.312 391s testing: mul_mat, [4, 3, 3, 4] = [1, 4, 3, 4] * [1, 3, 3, 4] 391s x0: [1, 4, 3, 4] 391s 0.367 391s 0.459 391s -0.848 391s -0.545 391s 391s x1: [1, 3, 3, 4] 391s 0.243 391s 0.228 391s 0.447 391s 391s y: [4, 3, 3, 4] 391s 0.089 0.111 -0.206 -0.132 391s 0.083 0.105 -0.193 -0.124 391s 0.164 0.205 -0.379 -0.244 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.971 391s 391s x1: [1, 4, 1, 1] 391s -0.969 391s 0.744 391s -0.550 391s 0.166 391s 391s y: [1, 4, 1, 1] 391s -0.941 391s 0.722 391s -0.534 391s 0.161 391s testing: mul_mat, [1, 1, 3, 1] = [1, 1, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 1, 3, 1] 391s -0.495 391s 391s x1: [1, 1, 3, 1] 391s -0.307 391s 391s y: [1, 1, 3, 1] 391s 0.152 391s testing: mul_mat, [2, 1, 3, 4] = [1, 2, 3, 4] * [1, 1, 3, 4] 391s x0: [1, 2, 3, 4] 391s 0.149 391s 0.429 391s 391s x1: [1, 1, 3, 4] 391s -0.801 391s 391s y: [2, 1, 3, 4] 391s -0.120 -0.344 391s test-mul-mat0: iter:193/500 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.340 -0.178 0.519 -0.764 391s -0.025 -0.839 -0.702 0.221 391s -0.225 0.054 0.333 0.955 391s 391s x1: [4, 1, 1, 1] 391s 0.264 0.898 -0.525 -0.587 391s 391s y: [3, 1, 1, 1] 391s 0.106 -0.522 -0.746 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.157 -0.597 -0.289 0.318 391s 391s x1: [4, 3, 1, 1] 391s 0.650 0.950 0.311 0.575 391s -0.234 0.815 0.774 0.626 391s -0.103 -0.774 0.785 0.140 391s 391s y: [1, 3, 1, 1] 391s -0.576 391s -0.474 391s 0.295 391s testing: mul_mat, [2, 1, 1, 3] = [4, 2, 1, 3] * [4, 1, 1, 3] 391s x0: [4, 2, 1, 3] 391s -0.552 0.897 0.116 0.759 391s 0.472 0.882 0.574 0.246 391s 391s x1: [4, 1, 1, 3] 391s -0.184 -0.962 -0.342 0.638 391s 391s y: [2, 1, 1, 3] 391s -0.317 -0.975 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.198 0.080 -0.411 0.882 391s 0.010 -0.416 0.055 -0.334 391s 0.814 0.060 -0.468 -0.184 391s 391s x1: [4, 2, 1, 1] 391s 0.081 -0.691 -0.801 -0.085 391s -0.198 -0.555 -0.918 0.251 391s 391s y: [3, 2, 1, 1] 391s 0.215 0.273 0.415 391s 0.515 0.095 0.189 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.976 0.237 0.797 391s -0.454 -0.942 0.579 391s 391s x1: [3, 4, 1, 1] 391s -0.981 -0.320 0.812 391s 0.473 0.511 -0.155 391s 0.586 0.980 0.435 391s -0.408 0.061 0.744 391s 391s y: [2, 4, 1, 1] 391s -0.387 1.217 391s 0.459 -0.786 391s 1.150 -0.937 391s 0.209 0.558 391s testing: mul_mat, [3, 3, 1, 3] = [2, 3, 1, 3] * [2, 3, 1, 3] 391s x0: [2, 3, 1, 3] 391s 0.282 -0.657 391s 0.893 0.637 391s -0.522 0.269 391s 391s x1: [2, 3, 1, 3] 391s 0.256 -0.194 391s 0.394 -0.664 391s 0.390 -0.546 391s 391s y: [3, 3, 1, 3] 391s 0.199 0.105 -0.186 391s 0.547 -0.071 -0.384 391s 0.469 0.000 -0.351 391s test-mul-mat0: iter:194/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.173 -0.174 -0.208 -0.288 391s -0.892 -0.315 0.190 -0.549 391s 391s x1: [4, 1, 1, 1] 391s 0.732 -0.280 -0.472 0.524 391s 391s y: [2, 1, 1, 1] 391s -0.131 -0.942 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.223 0.414 -0.741 0.908 391s -0.397 -0.289 -0.769 0.062 391s 391s x1: [4, 2, 3, 1] 391s -0.677 -0.541 -0.229 -0.494 391s -0.036 0.148 -0.514 0.049 391s 391s y: [2, 2, 3, 1] 391s -0.654 0.571 391s 0.479 0.370 391s testing: mul_mat, [3, 2, 3, 4] = [4, 3, 3, 4] * [4, 2, 3, 4] 391s x0: [4, 3, 3, 4] 391s -0.911 -0.347 0.129 0.126 391s 0.327 0.623 -0.013 -0.441 391s 0.897 0.002 0.159 0.689 391s 391s x1: [4, 2, 3, 4] 391s 0.812 0.539 0.029 0.137 391s 0.970 0.731 0.288 0.193 391s 391s y: [3, 2, 3, 4] 391s -0.906 0.541 0.828 391s -1.076 0.684 1.050 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.402 -0.482 -0.021 0.773 391s -0.011 0.125 -0.945 -0.652 391s -0.122 -0.940 0.784 0.865 391s 391s x1: [4, 3, 1, 1] 391s -0.718 -0.957 -0.817 0.928 391s 0.262 -0.643 0.082 -0.321 391s 0.551 -0.877 0.264 0.472 391s 391s y: [3, 3, 1, 1] 391s 0.908 0.056 1.150 391s 0.165 0.048 0.359 391s 1.003 -0.673 1.372 391s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.520 -0.886 0.531 391s 391s x1: [3, 4, 3, 1] 391s 0.891 0.967 -0.960 391s 0.038 -0.689 0.402 391s 0.320 -0.646 0.586 391s 0.248 0.615 0.943 391s 391s y: [1, 4, 3, 1] 391s -1.829 391s 0.805 391s 0.717 391s -0.174 391s testing: mul_mat, [3, 3, 3, 4] = [1, 3, 3, 4] * [1, 3, 3, 4] 391s x0: [1, 3, 3, 4] 391s -0.579 391s -0.183 391s -0.532 391s 391s x1: [1, 3, 3, 4] 391s -0.489 391s 0.584 391s 0.934 391s 391s y: [3, 3, 3, 4] 391s 0.283 0.090 0.260 391s -0.338 -0.107 -0.311 391s -0.541 -0.171 -0.497 391s test-mul-mat0: iter:195/500 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.492 0.907 391s 0.412 0.317 391s 0.694 0.611 391s 391s x1: [2, 4, 1, 1] 391s -0.689 -0.784 391s -0.006 -0.508 391s 0.061 -0.019 391s -0.625 -0.618 391s 391s y: [3, 4, 1, 1] 391s -1.050 -0.533 -0.958 391s -0.464 -0.164 -0.315 391s 0.013 0.019 0.031 391s -0.868 -0.453 -0.811 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.804 0.351 391s 0.610 -0.126 391s 391s x1: [2, 3, 1, 1] 391s -0.183 -0.746 391s -0.942 0.926 391s -0.986 0.296 391s 391s y: [2, 3, 1, 1] 391s -0.115 -0.018 391s 1.083 -0.691 391s 0.896 -0.639 391s testing: mul_mat, [4, 2, 1, 3] = [2, 4, 1, 3] * [2, 2, 1, 3] 391s x0: [2, 4, 1, 3] 391s -0.089 0.431 391s 0.640 -0.677 391s -0.252 0.334 391s 0.935 0.565 391s 391s x1: [2, 2, 1, 3] 391s -0.295 0.257 391s -0.239 -0.927 391s 391s y: [4, 2, 1, 3] 391s 0.137 -0.362 0.160 -0.131 391s -0.379 0.475 -0.249 -0.747 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.572 -0.793 391s -0.728 0.753 391s -0.811 0.350 391s 0.762 -0.800 391s 391s x1: [2, 4, 1, 1] 391s 0.706 -0.996 391s -0.930 0.056 391s 0.038 -0.991 391s -0.211 0.949 391s 391s y: [4, 4, 1, 1] 391s 0.386 -1.264 -0.921 1.335 ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 391s 0.487 0.719 0.775 -0.754 391s 0.764 -0.775 -0.378 0.822 391s -0.632 0.869 0.503 -0.921 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.798 -0.110 0.363 0.754 391s -0.143 0.749 -0.404 0.433 391s 391s x1: [4, 2, 1, 1] 391s 0.244 -0.048 0.488 0.270 391s -0.250 -0.872 -0.095 0.423 391s 391s y: [2, 2, 1, 1] 391s 0.581 -0.151 391s 0.181 -0.397 391s testing: mul_mat, [2, 4, 1, 3] = [2, 2, 1, 3] * [2, 4, 1, 3] 391s x0: [2, 2, 1, 3] 391s 0.399 0.326 391s 0.791 0.995 391s 391s x1: [2, 4, 1, 3] 391s 0.653 -0.208 391s 0.442 -0.559 391s -0.259 -0.118 391s -0.130 0.014 391s 391s y: [2, 4, 1, 3] 391s 0.192 0.309 391s -0.006 -0.206 391s -0.142 -0.322 391s -0.047 -0.089 391s test-mul-mat0: iter:196/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.323 -0.371 391s 0.294 0.186 391s 391s x1: [2, 4, 1, 1] 391s 0.312 -0.022 391s 0.572 -0.268 391s 0.046 -0.495 391s 0.641 -0.615 391s 391s y: [2, 4, 1, 1] 391s -0.093 0.088 391s -0.085 0.119 391s 0.169 -0.078 391s 0.021 0.074 391s testing: mul_mat, [4, 2, 2, 1] = [2, 4, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.658 0.544 391s -0.683 0.419 391s -0.373 0.276 391s 0.510 -0.219 391s 391s x1: [2, 2, 2, 1] 391s -0.330 0.257 391s -0.415 -0.539 391s 391s y: [4, 2, 2, 1] 391s 0.357 0.333 0.194 -0.225 391s -0.020 0.058 0.006 -0.094 391s testing: mul_mat, [1, 4, 2, 2] = [2, 1, 2, 2] * [2, 4, 2, 2] 391s x0: [2, 1, 2, 2] 391s 0.666 -0.866 391s 391s x1: [2, 4, 2, 2] 391s 0.190 0.133 391s 0.871 0.818 391s -0.573 0.057 391s -0.511 0.685 391s 391s y: [1, 4, 2, 2] 391s 0.011 391s -0.129 391s -0.430 391s -0.933 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.415 0.798 391s 0.033 -0.461 391s 0.824 -0.649 391s -0.430 0.839 391s 391s x1: [2, 1, 1, 1] 391s 0.804 -0.415 391s 391s y: [4, 1, 1, 1] 391s 0.002 0.218 0.931 -0.694 391s testing: mul_mat, [2, 2, 2, 1] = [4, 2, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 2, 2, 1] 391s 0.192 0.005 0.544 -0.590 391s -0.041 -0.005 0.532 -0.424 391s 391s x1: [4, 2, 2, 1] 391s -0.235 -0.086 -0.254 0.812 391s 0.235 -0.908 -0.588 -0.639 391s 391s y: [2, 2, 2, 1] 391s -0.663 -0.469 391s 0.098 -0.047 391s testing: mul_mat, [1, 4, 2, 2] = [2, 1, 2, 2] * [2, 4, 2, 2] 391s x0: [2, 1, 2, 2] 391s 0.526 0.234 391s 391s x1: [2, 4, 2, 2] 391s -0.041 -0.528 391s 0.479 0.051 391s -0.116 0.840 391s 0.433 0.091 391s 391s y: [1, 4, 2, 2] 391s -0.146 391s 0.264 391s 0.136 391s 0.249 391s test-mul-mat0: iter:197/500 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.733 391s 0.970 391s 391s x1: [1, 2, 1, 1] 391s 0.968 391s -0.112 391s 391s y: [2, 2, 1, 1] 391s 0.709 0.939 391s -0.082 -0.108 391s testing: mul_mat, [2, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.634 391s 0.439 391s 391s x1: [1, 2, 4, 1] 391s -0.242 391s 0.030 391s 391s y: [2, 2, 4, 1] 391s -0.153 -0.106 391s 0.019 0.013 391s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 4, 4, 1] 391s 0.732 391s 0.316 391s -0.168 391s 0.464 391s 391s x1: [1, 2, 4, 1] 391s -0.326 391s 0.720 391s 391s y: [4, 2, 4, 1] 391s -0.239 -0.103 0.055 -0.151 391s 0.527 0.227 -0.121 0.334 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.352 391s -0.817 391s 391s x1: [1, 4, 1, 1] 391s -0.799 391s -0.152 391s -0.035 391s 0.509 391s 391s y: [2, 4, 1, 1] 391s 0.281 0.653 391s 0.053 0.124 391s 0.012 0.028 391s -0.179 -0.416 391s testing: mul_mat, [4, 1, 4, 1] = [2, 4, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.011 -0.500 391s -0.090 -0.808 391s 0.698 0.024 391s -0.725 0.487 391s 391s x1: [2, 1, 4, 1] 391s 0.620 0.986 391s 391s y: [4, 1, 4, 1] 391s -0.499 -0.852 0.457 0.031 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.039 -0.589 0.585 -0.469 391s 391s x1: [4, 2, 4, 1] 391s 0.153 0.191 -0.532 0.653 391s 0.384 0.492 0.140 0.753 391s 391s y: [1, 2, 4, 1] 391s -0.724 391s -0.546 391s test-mul-mat0: iter:198/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.327 -0.621 -0.928 391s -0.742 -0.090 0.623 391s 391s x1: [3, 3, 1, 1] 391s 0.933 0.944 -0.608 391s 0.296 0.461 -0.230 391s -0.098 -0.529 0.341 391s 391s y: [2, 3, 1, 1] 391s -0.327 -1.156 391s -0.170 -0.404 391s 0.044 0.332 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.606 -0.080 -0.760 391s -0.421 -0.508 -0.025 391s -0.735 0.709 -0.091 391s 391s x1: [3, 2, 1, 1] 391s 0.596 -0.360 0.198 391s 0.463 0.693 -0.291 391s 391s y: [3, 2, 1, 1] 391s -0.482 -0.073 -0.711 391s -0.115 -0.540 0.177 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.409 -0.160 0.074 391s 0.502 -0.537 -0.330 391s -0.858 0.660 -0.867 391s 0.834 -0.631 -0.385 391s 391s x1: [3, 3, 1, 1] 391s -0.791 -0.899 -0.795 391s 0.671 -0.129 0.107 391s -0.858 -0.788 -0.232 391s 391s y: [4, 3, 1, 1] 391s 0.409 0.349 0.775 0.214 391s -0.246 0.370 -0.754 0.599 391s 0.460 0.070 0.417 -0.129 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.538 -0.326 0.265 391s -0.948 -0.180 0.659 391s 0.095 0.910 -0.017 391s 391s x1: [3, 4, 1, 1] 391s -0.772 0.289 -0.145 391s -0.193 0.781 0.830 391s 0.072 0.490 -0.261 391s 0.282 0.591 -0.055 391s 391s y: [3, 4, 1, 1] 391s 0.282 0.584 0.192 391s 0.069 0.590 0.678 391s -0.268 -0.329 0.457 391s -0.359 -0.410 0.565 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.440 0.255 0.778 391s 391s x1: [3, 3, 1, 1] 391s -0.233 -0.878 0.653 391s -0.092 0.782 0.786 391s -0.257 -0.849 -0.600 391s 391s y: [1, 3, 1, 1] 391s 0.181 391s 0.771 391s -0.797 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.825 391s 391s x1: [1, 3, 1, 1] 391s 0.221 391s 0.086 391s -0.150 391s 391s y: [1, 3, 1, 1] 391s 0.182 391s 0.071 391s -0.123 391s test-mul-mat0: iter:199/500 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.205 -0.489 0.121 391s 0.356 -0.089 -0.908 391s -0.204 -0.834 0.870 391s -0.983 0.252 -0.280 391s 391s x1: [3, 4, 1, 1] 391s -0.236 -0.178 -0.821 391s 0.438 0.641 -0.911 391s -0.297 0.301 0.072 391s 0.470 0.423 -0.275 391s 391s y: [4, 4, 1, 1] 391s -0.060 0.678 -0.517 0.417 391s -0.334 0.927 -1.417 -0.014 391s -0.199 -0.198 -0.128 0.348 391s -0.144 0.380 -0.688 -0.278 391s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.610 0.589 -0.135 391s -0.593 -0.159 0.585 391s 391s x1: [3, 4, 2, 1] 391s -0.273 0.078 -0.148 391s 0.029 0.847 0.937 391s 0.793 -0.331 -0.885 391s 0.232 -0.690 -0.796 391s 391s y: [2, 4, 2, 1] 391s 0.232 0.063 391s 0.354 0.396 391s -0.559 -0.935 391s -0.440 -0.493 391s testing: mul_mat, [1, 2, 2, 1] = [3, 1, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 1, 2, 1] 391s 0.629 -0.519 0.213 391s 391s x1: [3, 2, 2, 1] 391s 0.957 -0.565 -0.511 391s 0.188 -0.255 -0.307 391s 391s y: [1, 2, 2, 1] 391s 0.786 391s 0.185 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.666 -0.747 -0.981 391s -0.342 0.095 0.468 391s 0.155 0.318 0.275 391s 391s x1: [3, 1, 1, 1] 391s 0.614 0.493 0.073 391s 391s y: [3, 1, 1, 1] 391s -0.849 -0.129 0.272 391s testing: mul_mat, [1, 3, 2, 1] = [3, 1, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 1, 2, 1] 391s -0.534 -0.082 0.433 391s 391s x1: [3, 3, 2, 1] 391s -0.546 0.957 -0.537 391s 0.199 -0.350 0.586 391s 0.556 -0.381 -0.887 391s 391s y: [1, 3, 2, 1] 391s -0.019 391s 0.176 391s -0.650 391s testing: mul_mat, [3, 3, 2, 1] = [1, 3, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.904 391s 0.644 391s -0.276 391s 391s x1: [1, 3, 2, 1] 391s 0.053 391s -0.329 391s 0.906 391s 391s y: [3, 3, 2, 1] 391s -0.047 0.034 -0.014 391s 0.297 -0.212 0.091 391s -0.819 0.583 -0.250 391s test-mul-mat0: iter:200/500 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.403 391s 0.600 391s 0.998 391s 391s x1: [1, 2, 1, 1] 391s 0.766 391s 0.802 391s 391s y: [3, 2, 1, 1] 391s 0.308 0.460 0.764 391s 0.323 0.481 0.800 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.554 391s 0.264 391s 0.743 391s 391s x1: [1, 3, 1, 1] 391s 0.248 391s 0.013 391s -0.536 391s 391s y: [3, 3, 1, 1] 391s -0.137 0.065 0.184 391s -0.007 0.003 0.010 391s 0.297 -0.141 -0.398 391s testing: mul_mat, [2, 3, 1, 4] = [1, 2, 1, 4] * [1, 3, 1, 4] 391s x0: [1, 2, 1, 4] 391s 0.917 391s -0.239 391s 391s x1: [1, 3, 1, 4] 391s 0.034 391s -0.684 391s -0.586 391s 391s y: [2, 3, 1, 4] 391s 0.031 -0.008 391s -0.627 0.163 391s -0.538 0.140 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.247 391s 0.558 391s -0.798 391s -0.081 391s 391s x1: [1, 2, 1, 1] 391s 0.505 391s -0.455 391s 391s y: [4, 2, 1, 1] 391s -0.125 0.282 -0.403 -0.041 391s 0.112 -0.254 0.363 0.037 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.075 0.170 -0.900 -0.723 391s 0.979 0.628 -0.917 0.864 391s 391s x1: [4, 1, 1, 1] 391s -0.995 0.466 -0.338 0.039 391s 391s y: [2, 1, 1, 1] 391s 0.281 -0.338 391s testing: mul_mat, [3, 4, 1, 4] = [2, 3, 1, 4] * [2, 4, 1, 4] 391s x0: [2, 3, 1, 4] 391s 0.673 -0.134 391s -0.309 0.471 391s -0.777 0.374 391s 391s x1: [2, 4, 1, 4] 391s 0.720 0.673 391s -0.142 0.637 391s -0.565 0.194 391s -0.219 0.151 391s 391s y: [3, 4, 1, 4] 391s 0.394 0.095 -0.308 391s -0.181 0.344 0.349 391s -0.406 0.266 0.512 391s -0.168 0.139 0.227 391s test-mul-mat0: iter:201/500 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.901 0.674 391s 391s x1: [2, 3, 1, 1] 391s -0.380 0.739 391s -0.923 -0.707 391s -0.570 -0.700 391s 391s y: [1, 3, 1, 1ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ] 391s 0.156 391s -1.309 391s -0.986 391s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 3, 3, 1] 391s 0.863 -0.444 391s 0.795 -0.007 391s -0.422 0.921 391s 391s x1: [2, 1, 3, 1] 391s 0.322 0.098 391s 391s y: [3, 1, 3, 1] 391s 0.234 0.255 -0.046 391s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.697 0.352 391s -0.888 -0.847 391s 0.384 -0.459 391s 391s x1: [2, 3, 3, 1] 391s -0.431 -0.114 391s 0.094 0.729 391s -0.213 -0.233 391s 391s y: [3, 3, 3, 1] 391s 0.260 0.479 -0.113 391s 0.191 -0.700 -0.299 391s 0.066 0.386 0.025 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.120 -0.078 391s -0.750 -0.296 391s -0.152 -0.510 391s -0.813 -0.726 391s 391s x1: [2, 3, 1, 1] 391s 0.434 -0.884 391s -0.382 -0.229 391s -0.433 -0.844 391s 391s y: [4, 3, 1, 1] 391s 0.017 -0.064 0.385 0.289 391s 0.063 0.354 0.175 0.477 391s 0.118 0.575 0.496 0.965 391s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.543 -0.769 0.601 0.035 391s 0.740 0.943 -0.580 0.867 391s 0.669 0.063 0.328 0.678 391s -0.753 -0.125 0.361 0.701 391s 391s x1: [4, 2, 3, 1] 391s -0.184 -0.356 0.658 0.357 391s -0.547 0.759 -0.385 -0.461 391s 391s y: [4, 2, 3, 1] 391s 0.583 -0.544 0.312 0.671 391s -1.129 0.135 -0.756 -0.146 391s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.449 0.787 -0.375 0.935 391s -0.697 -0.275 -0.528 -0.656 391s 391s x1: [4, 4, 3, 1] 391s -0.488 0.651 0.500 0.940 391s -0.113 0.150 0.474 -0.535 391s -0.416 0.680 -0.460 -0.288 391s 0.440 0.362 -0.616 -0.060 391s 391s y: [2, 4, 3, 1] 391s 0.984 -0.719 391s -0.610 0.138 391s 0.252 0.535 391s 0.657 -0.041 391s test-mul-mat0: iter:202/500 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.135 391s 0.113 391s 391s x1: [1, 2, 1, 1] 391s 0.416 391s -0.190 391s 391s y: [2, 2, 1, 1] 391s -0.056 0.047 391s 0.026 -0.021 391s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 4, 4, 1] 391s 0.039 391s 0.286 391s 0.659 391s -0.196 391s 391s x1: [1, 2, 4, 1] 391s -0.879 391s 0.590 391s 391s y: [4, 2, 4, 1] 391s -0.035 -0.252 -0.579 0.172 391s 0.023 0.169 0.389 -0.116 391s testing: mul_mat, [2, 4, 4, 1] = [1, 2, 4, 1] * [1, 4, 4, 1] 391s x0: [1, 2, 4, 1] 391s -0.766 391s -0.469 391s 391s x1: [1, 4, 4, 1] 391s 0.454 391s -0.399 391s -0.576 391s 0.789 391s 391s y: [2, 4, 4, 1] 391s -0.348 -0.213 391s 0.306 0.187 391s 0.442 0.270 391s -0.604 -0.370 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.609 391s 0.054 391s 0.526 391s 391s x1: [1, 2, 1, 1] 391s 0.985 391s 0.869 391s 391s y: [3, 2, 1, 1] 391s 0.600 0.053 0.518 391s 0.529 0.047 0.457 391s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.810 -0.851 0.960 391s -0.308 -0.074 -0.497 391s 0.677 -0.792 0.106 391s 391s x1: [3, 1, 4, 1] 391s 0.063 0.655 0.949 391s 391s y: [3, 1, 4, 1] 391s 0.405 -0.540 -0.376 391s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 1, 4, 1] 391s -0.498 -0.896 -0.462 391s 391s x1: [3, 3, 4, 1] 391s 0.319 -0.620 -0.933 391s -0.575 -0.668 0.893 391s -0.379 -0.731 0.589 391s 391s y: [1, 3, 4, 1] 391s 0.827 391s 0.473 391s 0.572 391s test-mul-mat0: iter:203/500 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.454 0.622 0.645 -0.573 391s 391s x1: [4, 2, 1, 1] 391s 0.928 -0.708 -0.825 -0.047 391s -0.005 0.600 0.090 -0.881 391s 391s y: [1, 2, 1, 1] 391s -1.367 391s 0.939 391s testing: mul_mat, [1, 1, 3, 1] = [4, 1, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 1, 3, 1] 391s 0.517 0.834 -0.609 0.376 391s 391s x1: [4, 1, 3, 1] 391s 0.458 -0.406 -0.072 0.562 391s 391s y: [1, 1, 3, 1] 391s 0.153 391s testing: mul_mat, [1, 1, 3, 4] = [4, 1, 3, 4] * [4, 1, 3, 4] 391s x0: [4, 1, 3, 4] 391s -0.285 -0.832 -0.016 0.893 391s 391s x1: [4, 1, 3, 4] 391s -0.218 0.585 -0.091 -0.595 391s 391s y: [1, 1, 3, 4] 391s -0.955 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.279 -0.453 -0.245 0.859 391s 391s x1: [4, 1, 1, 1] 391s 0.885 0.999 0.896 0.965 391s 391s y: [1, 1, 1, 1] 391s 0.405 391s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 3, 3, 1] 391s -0.903 391s 0.884 391s -0.001 391s 391s x1: [1, 4, 3, 1] 391s -0.404 391s 0.398 391s -0.632 391s -0.296 391s 391s y: [3, 4, 3, 1] 391s 0.365 -0.357 0.000 391s -0.360 0.352 -0.000 391s 0.571 -0.559 0.001 391s 0.267 -0.262 0.000 391s testing: mul_mat, [2, 1, 3, 4] = [3, 2, 3, 4] * [3, 1, 3, 4] 391s x0: [3, 2, 3, 4] 391s 0.129 0.432 -0.358 391s 0.009 -0.884 0.217 391s 391s x1: [3, 1, 3, 4] 391s 0.878 0.133 -0.721 391s 391s y: [2, 1, 3, 4] 391s 0.429 -0.266 391s test-mul-mat0: iter:204/500 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s x0: [3, 2, 1, 1] 391s 0.768 0.526 0.258 391s 0.513 0.877 -0.650 391s 391s x1: [3, 4, 1, 1] 391s -0.410 -0.473 0.347 391s 0.603 0.658 0.814 391s -0.187 -0.924 0.428 391s -0.618 0.961 0.086 391s 391s y: [2, 4, 1, 1] 391s -0.473 -0.851 391s 1.019 0.357 391s -0.519 -1.185 391s 0.052 0.470 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.006 -0.441 0.959 391s 0.883 0.570 -0.451 391s -0.589 -0.083 -0.848 391s -0.932 -0.269 -0.035 391s 391s x1: [3, 2, 1, 1] 391s -0.372 -0.743 0.522 391s -0.670 -0.530 -0.218 391s 391s y: [4, 2, 1, 1] 391s 0.830 -0.987 -0.161 0.528 391s 0.029 -0.796 0.623 0.775 391s testing: mul_mat, [1, 4, 1, 4] = [3, 1, 1, 4] * [3, 4, 1, 4] 391s x0: [3, 1, 1, 4] 391s -0.845 -0.890 -0.379 391s 391s x1: [3, 4, 1, 4] 391s -0.855 -0.841 0.347 391s -0.895 0.245 -0.048 391s 0.873 -0.230 -0.790 391s 0.386 -0.352 -0.440 391s 391s y: [1, 4, 1, 4] 391s 1.341 391s 0.557 391s -0.234 391s 0.154 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.609 0.011 0.150 391s 391s x1: [3, 1, 1, 1] 391s 0.138 -0.371 -0.034 391s 391s y: [1, 1, 1, 1] 391s 0.075 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.289 391s 0.095 391s -0.044 391s 391s x1: [1, 3, 1, 1] 391s -0.538 391s 0.826 391s -0.158 391s 391s y: [3, 3, 1, 1] 391s -0.155 -0.051 0.024 391s 0.239 0.078 -0.036 391s -0.046 -0.015 0.007 391s testing: mul_mat, [4, 1, 1, 4] = [3, 4, 1, 4] * [3, 1, 1, 4] 391s x0: [3, 4, 1, 4] 391s -0.614 -0.792 -0.118 391s 0.070 0.273 -0.472 391s -0.355 0.517 0.026 391s -0.449 0.876 0.345 391s 391s x1: [3, 1, 1, 4] 391s 0.656 0.429 0.554 391s 391s y: [4, 1, 1, 4] 391s -0.808 -0.098 0.003 0.272 391s test-mul-mat0: iter:205/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.020 391s -0.692 391s 391s x1: [1, 1, 1, 1] 391s -0.896 391s 391s y: [2, 1, 1, 1] 391s 0.018 0.620 391s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 2, 3, 1] 391s 0.213 391s -0.050 391s 391s x1: [1, 2, 3, 1] 391s 0.449 391s -0.376 391s 391s y: [2, 2, 3, 1] 391s 0.096 -0.022 391s -0.080 0.019 391s testing: mul_mat, [1, 2, 3, 2] = [1, 1, 3, 2] * [1, 2, 3, 2] 391s x0: [1, 1, 3, 2] 391s -0.269 391s 391s x1: [1, 2, 3, 2] 391s 0.164 391s 0.594 391s 391s y: [1, 2, 3, 2] 391s -0.044 391s -0.160 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.947 391s 0.718 391s 391s x1: [1, 1, 1, 1] 391s -0.666 391s 391s y: [2, 1, 1, 1] 391s 0.630 -0.478 391s testing: mul_mat, [1, 1, 3, 1] = [2, 1, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.963 0.345 391s 391s x1: [2, 1, 3, 1] 391s 0.961 -0.856 391s 391s y: [1, 1, 3, 1] 391s -1.221 391s testing: mul_mat, [4, 2, 3, 2] = [1, 4, 3, 2] * [1, 2, 3, 2] 391s x0: [1, 4, 3, 2] 391s 0.392 391s 0.688 391s 0.227 391s 0.445 391s 391s x1: [1, 2, 3, 2] 391s -0.805 391s -0.640 391s 391s y: [4, 2, 3, 2] 391s -0.315 -0.554 -0.183 -0.358 391s -0.251 -0.440 -0.145 -0.285 391s test-mul-mat0: iter:206/500 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.267 0.356 391s 391s x1: [2, 2, 1, 1] 391s -0.481 0.022 391s -0.961 0.911 391s 391s y: [1, 2, 1, 1] 391s -0.120 391s 0.068 391s testing: mul_mat, [3, 1, 2, 1] = [2, 3, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 3, 2, 1] 391s -0.167 -0.030 391s 0.394 -0.242 391s -0.167 -0.430 391s 391s x1: [2, 1, 2, 1] 391s -0.883 -0.546 391s 391s y: [3, 1, 2, 1] 391s 0.164 -0.215 0.382 391s testing: mul_mat, [3, 3, 2, 3] = [2, 3, 2, 3] * [2, 3, 2, 3] 391s x0: [2, 3, 2, 3] 391s 0.709 -0.015 391s -0.234 0.426 391s -0.449 -0.910 391s 391s x1: [2, 3, 2, 3] 391s 0.189 -0.467 391s 0.923 -0.752 391s 0.136 0.591 391s 391s y: [3, 3, 2, 3] 391s 0.141 -0.243 0.340 391s 0.665 -0.536 0.271 391s 0.088 0.220 -0.599 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.163 0.403 391s 391s x1: [2, 3, 1, 1] 391s -0.872 0.476 391s -0.979 0.658 391s 0.324 0.362 391s 391s y: [1, 3, 1, 1] 391s 0.050 391s 0.106 391s 0.199 391s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.575 391s 0.183 391s 0.485 391s 391s x1: [1, 2, 2, 1] 391s 0.370 391s -0.613 391s 391s y: [3, 2, 2, 1] 391s -0.213 0.068 0.179 391s 0.353 -0.112 -0.297 391s testing: mul_mat, [2, 1, 2, 3] = [3, 2, 2, 3] * [3, 1, 2, 3] 391s x0: [3, 2, 2, 3] 391s -0.305 0.432 -0.934 391s -0.289 0.596 -0.002 391s 391s x1: [3, 1, 2, 3] 391s -0.557 -0.647 0.163 391s 391s y: [2, 1, 2, 3] 391s -0.261 -0.226 391s test-mul-mat0: iter:207/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.347 0.210 -0.216 391s 0.650 0.806 0.261 391s 391s x1: [3, 2, 1, 1] 391s 0.002 -0.256 0.399 391s -0.992 0.016 0.087 391s 391s y: [2, 2, 1, 1] 391s -0.141 -0.101 391s 0.329 -0.609 391s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.907 0.234 -0.792 391s -0.076 ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s -0.678 0.597 391s 0.577 0.532 -0.620 391s 0.227 0.338 0.641 391s 391s x1: [3, 2, 3, 1] 391s -0.444 0.003 -0.704 391s 0.391 -0.541 0.213 391s 391s y: [4, 2, 3, 1] 391s 0.155 -0.388 0.182 -0.552 391s 0.060 0.464 -0.194 0.042 391s testing: mul_mat, [3, 4, 3, 3] = [3, 3, 3, 3] * [3, 4, 3, 3] 391s x0: [3, 3, 3, 3] 391s -0.186 0.842 0.630 391s -0.601 -0.028 -0.540 391s -0.583 -0.918 0.483 391s 391s x1: [3, 4, 3, 3] 391s -0.009 -0.269 -0.223 391s 0.372 0.958 -0.885 391s 0.013 0.741 -0.544 391s -0.050 -0.085 0.256 391s 391s y: [3, 4, 3, 3] 391s -0.365 0.133 0.144 391s 0.181 0.227 -1.523 391s 0.279 0.265 -0.950 391s 0.099 -0.105 0.231 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.685 -0.880 0.335 391s -0.078 0.176 0.124 391s -0.979 -0.075 0.898 391s 391s x1: [3, 3, 1, 1] 391s 0.844 -0.366 -0.465 391s 0.211 -0.596 -0.518 391s -0.230 0.061 0.901 391s 391s y: [3, 3, 1, 1] 391s 0.744 -0.188 -1.216 391s 0.495 -0.186 -0.627 391s 0.091 0.140 1.030 391s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.546 0.844 -0.960 391s 0.793 -0.039 -0.745 391s 391s x1: [3, 3, 3, 1] 391s 0.415 -0.752 -0.588 391s 0.377 0.028 -0.979 391s -0.995 0.949 -0.612 391s 391s y: [2, 3, 3, 1] 391s 0.156 0.796 391s 1.169 1.027 391s 0.845 -0.370 391s testing: mul_mat, [4, 3, 3, 3] = [2, 4, 3, 3] * [2, 3, 3, 3] 391s x0: [2, 4, 3, 3] 391s 0.379 -0.683 391s 0.068 0.194 391s 0.783 0.893 391s 0.384 -0.978 391s 391s x1: [2, 3, 3, 3] 391s -0.545 0.609 391s 0.173 -0.315 391s -0.985 0.744 391s 391s y: [4, 3, 3, 3] 391s -0.622 0.081 0.118 -0.805 391s 0.281 -0.049 -0.146 0.375 391s -0.881 0.078 -0.106 -1.106 391s test-mul-mat0: iter:208/500 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.198 -0.086 -0.338 391s 391s x1: [3, 3, 1, 1] 391s 0.348 0.420 0.376 391s 0.982 0.511 0.064 391s 0.788 -0.898 0.355 391s 391s y: [1, 3, 1, 1] 391s -0.094 391s 0.129 391s 0.114 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.720 0.553 -0.924 391s -0.149 -0.922 0.104 391s 391s x1: [3, 1, 1, 1] 391s 0.611 0.429 0.854 391s 391s y: [2, 1, 1, 1] 391s -0.992 -0.398 391s testing: mul_mat, [3, 2, 1, 2] = [3, 3, 1, 2] * [3, 2, 1, 2] 391s x0: [3, 3, 1, 2] 391s -0.108 -0.044 0.004 391s -0.753 0.867 -0.798 391s 0.161 -0.470 0.813 391s 391s x1: [3, 2, 1, 2] 391s 0.034 0.450 0.275 391s -0.594 -0.920 0.839 391s 391s y: [3, 2, 1, 2] 391s -0.023 0.145 0.018 391s 0.108 -1.021 1.019 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.824 -0.284 -0.037 391s -0.564 0.391 0.259 391s 0.423 -0.573 -0.371 391s 391s x1: [3, 3, 1, 1] 391s -0.245 0.464 -0.757 391s 0.161 0.544 -0.918 391s 0.131 0.996 0.268 391s 391s y: [3, 3, 1, 1] 391s -0.306 0.124 -0.089 391s 0.012 -0.116 0.097 391s -0.185 0.385 -0.615 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.877 -0.941 0.307 391s 0.063 0.587 0.220 391s 391s x1: [3, 3, 1, 1] 391s -0.876 0.789 -0.558 391s -0.286 -0.828 -0.769 391s 0.584 -0.891 -0.905 391s 391s y: [2, 3, 1, 1] 391s -0.145 0.284 391s 0.794 -0.674 391s 0.049 -0.686 391s testing: mul_mat, [1, 3, 1, 2] = [2, 1, 1, 2] * [2, 3, 1, 2] 391s x0: [2, 1, 1, 2] 391s 0.742 0.352 391s 391s x1: [2, 3, 1, 2] 391s 0.130 0.388 391s -0.648 0.126 391s -0.344 -0.127 391s 391s y: [1, 3, 1, 2] 391s 0.233 391s -0.437 391s -0.300 391s test-mul-mat0: iter:209/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.183 -0.584 0.521 -0.511 391s 0.636 -0.349 0.878 0.988 391s 391s x1: [4, 1, 1, 1] 391s 0.052 0.811 -0.881 -0.644 391s 391s y: [2, 1, 1, 1] 391s -0.595 -1.658 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s -0.649 -0.159 -0.294 0.002 391s -0.281 -0.306 -0.221 0.253 391s 391s x1: [4, 2, 3, 1] 391s 0.777 -0.466 -0.139 -0.272 391s 0.626 0.157 -0.606 -0.891 391s 391s y: [2, 2, 3, 1] 391s -0.390 -0.114 391s -0.255 -0.315 391s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.802 0.157 -0.099 -0.853 391s -0.539 -0.200 -0.607 -0.340 391s 0.869 0.293 0.690 0.290 391s 391s x1: [4, 2, 3, 1] 391s -0.754 -0.800 0.069 -0.100 391s 0.030 0.421 0.969 0.380 391s 391s y: [3, 2, 3, 1] 391s 0.558 0.559 -0.871 391s -0.378 -0.818 0.928 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.885 0.302 0.713 -0.779 391s -0.981 0.736 -0.924 -0.949 391s 0.857 0.982 -0.496 -0.301 391s -0.884 0.155 -0.372 0.267 391s 391s x1: [4, 3, 1, 1] 391s 0.846 -0.914 0.192 0.715 391s 0.379 -0.118 0.005 -0.359 391s 0.952 -0.443 -0.422 0.405 391s 391s y: [4, 3, 1, 1] 391s 0.053 -2.359 -0.483 -0.770 391s 0.583 -0.123 0.314 -0.451 391s 0.093 -1.255 0.468 -0.645 391s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.499 -0.961 -0.211 -0.601 391s 0.652 0.911 -0.174 -0.571 391s 391s x1: [4, 4, 3, 1] 391s 0.744 0.231 0.114 0.831 391s -0.577 -0.171 0.210 0.305 391s 0.834 0.851 0.257 -0.609 391s -0.571 -0.339 0.482 -0.686 391s 391s y: [2, 4, 3, 1] 391s -0.373 0.202 391s -0.351 -0.743 391s -0.090 1.622 391s 0.351 -0.373 391s testing: mul_mat, [3, 4, 3, 1] = [2, 3, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 3, 3, 1] 391s 0.719 0.566 391s -0.806 0.817 391s 0.753 -0.378 391s 391s x1: [2, 4, 3, 1] 391s 0.992 -0.622 391s 0.859 0.619 391s 0.448 0.530 391s -0.416 -0.053 391s 391s y: [3, 4, 3, 1] 391s 0.361 -1.308 0.982 391s 0.968 -0.188 0.413 391s 0.622 0.072 0.136 391s -0.329 0.292 -0.293 391s test-mul-mat0: iter:210/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.250 -0.722 0.429 391s 0.067 -0.101 -0.012 391s 0.262 0.380 -0.835 391s 391s x1: [3, 3, 1, 1] 391s 0.070 -0.916 0.567 391s -0.480 -0.840 0.312 391s -0.856 0.057 -0.475 391s 391s y: [3, 3, 1, 1] 391s 0.922 0.090 -0.804 391s 0.621 0.049 -0.706 391s -0.459 -0.057 0.194 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.311 0.662 -0.286 391s 391s x1: [3, 3, 1, 1] 391s -0.096 0.391 0.925 391s 0.493 -0.416 -0.692 391s -0.059 0.747 -0.860 391s 391s y: [1, 3, 1, 1] 391s -0.035 391s 0.075 391s 0.722 391s testing: mul_mat, [3, 1, 1, 2] = [3, 3, 1, 2] * [3, 1, 1, 2] 391s x0: [3, 3, 1, 2] 391s 0.066 0.241 0.862 391s 0.344 -0.330 -0.071 391s -0.757 0.658 -0.809 391s 391s x1: [3, 1, 1, 2] 391s 0.395 0.229 0.234 391s 391s y: [3, 1, 1, 2] 391s 0.283 0.044 -0.338 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.628 -0.782 0.742 391s -0.499 -0.306 -0.920 391s 391s x1: [3, 3, 1, 1] 391s 0.345 -0.803 0.174 391s -0.344 0.858 0.888 391s -0.948 0.087 0.122 391s 391s y: [2, 3, 1, 1] 391s 0.974 -0.086 391s -0.228 -0.909 391s -0.573 0.334 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.893 -0.706 391s 391s x1: [2, 3, 1, 1] 391s -0.963 -0.588 391s 0.008 -0.719 391s -0.930 0.199 391s 391s y: [1, 3, 1, 1] 391s -0.444 391s 0.515 391s -0.971 391s testing: mul_mat, [3, 2, 1, 2] = [1, 3, 1, 2] * [1, 2, 1, 2] 391s x0: [1, 3, 1, 2] 391s -0.589 391s 0.861 391s 0.894 391s 391s x1: [1, 2, 1, 2] 391s -0.082 391s 0.641 391s 391s y: [3, 2, 1, 2] 391s 0.048 -0.071 -0.073 391s -0.377 0.552 0.573 391s test-mul-mat0: iter:211/500 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.482 0.479 0.321 -0.319 391s 391s x1: [4, 2, 1, 1] 391s -0.824 -0.533 0.097 -0.606 391s 0.161 -0.162 -0.527 0.198 391s 391s y: [1, 2, 1, 1] 391s -0.428 391s -0.232 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.858 -0.183 -0.448 -0.731 391s -0.322 -0.555 0.854 0.196 391s -0.802 0.328 0.765 0.037 391s 391s x1: [4, 1, 1, 1] 391s 0.383 0.214 -0.025 -0.699 391s 391s y: [3, 1, 1, 1] 391s 0.811 -0.400 -0.282 391s testing: mul_mat, [2, 3, 1, 4] = [4, 2, 1, 4] * [4, 3, 1, 4] 391s x0: [4, 2, 1, 4] 391s 0.741 0.576 0.693 -0.209 391s 0.649 -0.787 -0.713 0.576 391s 391s x1: [4, 3, 1, 4] 391s 0.924 0.941 0.504 0.021 391s -0.665 -0.336 0.859 -0.192 391s 0.863 0.110 -0.710 0.341 391s 391s y: [2, 3, 1, 4] 391s 1.572 -0.488 391s -0.051 -0.891 391s 0.139 1.177 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.237 0.457 -0.426 0.163 391s -0.374 -0.268 0.726 0.202 391s 0.788 0.458 0.196 0.882 391s 0.248 -0.956 0.177 -0.802 391s 391s x1: [4, 2, 1, 1] 391s 0.181 0.076 -0.189 -0.170 391s 0.289 0.097 -0.594 0.217 391s 391s y: [4, 2, 1, 1] 391s 0.131 -0.260 -0.010 0.075 391s 0.402 -0.522 0.348 -0.301 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.269 -0.536 -0.002 -0.425 391s 0.540 -0.999 -0.492 -0.276 391s 391s x1: [4, 4, 1, 1] 391s 0.035 0.414 -0.524 -0.361 391s -0.202 -0.504 0.820 0.874 391s 0.306 -0.351 0.163 -0.596 391s 0.055 -0.620 0.481 -0.708 391s 391s y: [2, 4, 1, 1] 391s -0.058 -0.038 391s -0.158 -0.250 391s 0.524 0.600 391s 0.647 0.608 391s testing: mul_mat, [3, 4, 1, 4] = [2, 3, 1, 4] * [2, 4, 1, 4] 391s x0: [2, 3, 1, 4] 391s 0.546 0.631 391s 0.853 0.794 391s 0.478 -0.474 391s 391s x1: [2, 4, 1, 4] 391s 0.704 0.751 391s 0.887 -0.093 391s 0.634 -0.915 391s 0.942 0.048 391s 391s y: [3, 4, 1, 4] 391s 0.859 1.198 -0.019 391s 0.425 0.682 0.468 391s -0.231 -0.186 0.736 391s 0.544 0.841 0.428 391s test-mul-mat0: iter:212/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.084 391s -0.966 391s 391s x1: [1, 4, 1, 1] 391s -0.142 391s 0.425 391s 0.606 391s -0.597 391s 391s y: [2, 4, 1, 1] 391s -0.012 0.138 391s 0.036 -0.410 391s 0.051 -0.585 391s -0.050 0.576 391s testing: mul_mat, [4, 2, 3, 1] = [1, 4, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 4, 3, 1] 391s -0.927 391s 0.092 391s -0.953 391s 0.047 391s 391s x1: [1, 2, 3, 1] 391s 0.072 391s 0.611 391s 391s y: [4, 2, 3, 1] 391s -0.067 0.007 -0.069 0.003 391s -0.566 0.056 -0.582 0.028 391s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.984 391s 0.289 391s -0.275 391s 0.967 391s 391s x1: [1, 4, 3, 1] 391s 0.024 391s 0.922 391s 0.766 391s 0.503 391s 391s y: [4, 4, 3, 1] 391s 0.024 0.007 -0.007 0.023 391s 0.908 0.266 -0.254 0.891 391s 0.754 0.221 -0.211 0.741 391s 0.495 0.145 -0.138 0.486 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.843 391s 391s x1: [1, 4, 1, 1] 391s -0.631 391s 0.438 391s 0.275 391s -0.845 391s 391s y: [1, 4, 1, 1] 391s -0.532 391s 0.369 391s 0.232 391s -0.712 391s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 2, 3, 1] 391s 0.298 391s -0.767 391s 391s x1: [1, 1, 3, 1] 391s 0.179 391s 391s y: [2, 1, 3, 1] 391s 0.053 -0.137 391s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.623 -0.048 391s -0.234 -0.764 391s 391s x1: [2, 1, 3, 1] 391s -0.039 -0.678 391s 391s y: [2, 1, 3, 1] 391s 0.057 0.528 391s test-mul-mat0: iter:213/500 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.802 -0.048 0.850 0.158 391s 391s x1: [4, 1, 1, 1] 391s 0.459 -0.576 0.611 0.691 391s 391s y: [1, 1, 1, 1] 391s 0.289 391s testing: mul_mat, [4, 1, 3, 1] = [4, 4, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.981 -0.586 -0.927 -0.974 391s -0.023 -0.228 0.875 -0.281 391s 0.085 0.334 0.144 -0.304 391s 0.025 0.178 -0.106 0.977 391s 391s x1: [4, 1, 3, 1] 391s 0.273 -0.056 -0.896 -0.116 391s 391s y: [4, 1, 3, 1] 391s 0.710 -0.746 -0.089 -0.022 391s testing: mul_mat, [2, 4, 3, 4] = [4, 2, 3, 4] * [4, 4, 3, 4] 391s x0: [4, 2, 3, 4] 391s -0.238 -0.790 0.486 -0.983 391s -0.534 0.925 -0.122 -0.460 391s 391s x1: [4, 4, 3, 4] 391s 0.529 0.056 -0.424 0.685 391s 0.190 0.310 -0.144 -0.416 391s 0.571 0.994 0.476 0.120 391s 0.139 0.433 -0.312 -0.302 391s 391s y: [2, 4, 3, 4] 391s -1.049 -0.494 391s 0.048 0.394 391s -0.808 0.501 391s -0.230 0.503 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.921 -0.819 -0.274 -0.921 391s -0.668 -0.887 -0.825 0.832 391s -0.994 0.245 -0.037 -0.433 391s -0.185 -0.316 -0.599 -0.528 391s 391s x1: [4, 2, 1, 1] 391s 0.500 -0.528 0.876 -0.007 391s -0.778 -0.482 0.425 -0.191 391s 391s y: [4, 2, 1, 1] 391s 0.659 -0.593 -0.657 -0.447 391s -0.262 0.437 0.722 0.143 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.793 -0.414 0.262 0.915 391s 391s x1: [4, 4, 3, 1] 391s 0.718 -0.487 -0.374 0.808 391s 0.740 0.300 0.309 -0.788 391s 0.176 -0.698 -0.566 0.694 391s 0.727 0.243 0.303 0.647 391s 391s y: [1, 4, 3, 1] 391s 0.274 391s -1.351 391s 0.636 391s -0.005 391s testing: mul_mat, [4, 4, 3, 4] = [1, 4, 3, 4] * [1, 4, 3, 4] 391s x0: [1, 4, 3, 4] 391s 0.601 391s -0.873 391s 0.514 391s 0.618 391s 391s x1: [1, 4, 3, 4] 391s -0.806 391s -0.386 391s -0.301 391s 0.254 391s 391s y: [4, 4, 3, 4] 391s -0.485 0.704 -0.415 -0.499 391s -0.232 0.337 -0.199 -0.239 391s -0.181 0.263 -0.155 -0.186 391s 0.153 -0.222 0.131 0.157 391s test-mul-mat0: iter:214/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.173 0.680 0.664 391s 0.913 0.731 0.322 391s 391s x1: [3, 3, 1, 1] 391s -0.137 -0.528 0.761 391s 0.512 0.493 0.518 391s 0.602 0.359 0.536 391s 391s y: [2, 3, 1, 1] 391s 0.171 -0.266 391s 0.590 0.995 391s 0.496 0.985 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.667 -0.402 0.527 391s -0.820 -0.256 -0.735 391s 0.564 0.352 0.128 391s 391s x1: [3, 2, 1, 1] 391s -0.956 0.267 0.632 391s -0.397 0.624 -0.962 391s 391s y: [3, 2, 1, 1] 391s 0.863 0.251 -0.365 391s -0.494 0.872 -0.127 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.847 -0.218 0.627 391s 0.578 -0.896 0.671 391s 391s x1: [3, 3, 1, 1] 391s -0.964 0.113 -0.360 391s 0.529 -0.369 -0.758 391s -0.112 -0.833 0.118 391s 391s y: [2, 3, 1, 1] 391s -1.066 -0.900 391s 0.053 0.127 391s 0.161 0.761 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.628 0.848 -0.018 391s 391s x1: [3, 2, 1, 1] 391s -0.155 0.736 -0.725 391s -0.531 0.774 -0.679 391s 391s y: [1, 2, 1, 1] 391s 0.735 391s 1.002 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.241 391s -0.418 391s 391s x1: [1, 3, 1, 1] 391s 0.116 391s -0.887 391s -0.454 391s 391s y: [2, 3, 1, 1] 391s 0.028 -0.048 391s -0.214 0.370 391s -0.109 0ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s .189 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.788 -0.622 391s -0.877 0.241 391s -0.001 -0.286 391s 391s x1: [2, 1, 1, 1] 391s 0.581 0.881 391s 391s y: [3, 1, 1, 1] 391s -1.005 -0.297 -0.253 391s test-mul-mat0: iter:215/500 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.863 -0.714 391s 0.116 0.638 391s -0.393 0.918 391s -0.990 -0.544 391s 391s x1: [2, 1, 1, 1] 391s -0.873 -0.988 391s 391s y: [4, 1, 1, 1] 391s -0.049 -0.731 -0.565 1.403 391s testing: mul_mat, [1, 4, 2, 1] = [2, 1, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.077 -0.806 391s 391s x1: [2, 4, 2, 1] 391s -0.100 0.125 391s -0.431 0.447 391s 0.593 0.810 391s -0.971 0.173 391s 391s y: [1, 4, 2, 1] 391s -0.108 391s -0.393 391s -0.607 391s -0.214 391s testing: mul_mat, [2, 1, 2, 3] = [2, 2, 2, 3] * [2, 1, 2, 3] 391s x0: [2, 2, 2, 3] 391s 0.820 0.432 391s 0.933 0.849 391s 391s x1: [2, 1, 2, 3] 391s -0.794 -0.562 391s 391s y: [2, 1, 2, 3] 391s -0.894 -1.218 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.096 0.191 391s -0.349 -0.472 391s 391s x1: [2, 2, 1, 1] 391s -0.028 -0.002 391s 0.257 -0.473 391s 391s y: [2, 2, 1, 1] 391s -0.003 0.011 391s -0.065 0.133 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.864 -0.381 391s 0.139 0.703 391s 391s x1: [2, 2, 2, 1] 391s -0.416 0.040 391s 0.133 -0.792 391s 391s y: [2, 2, 2, 1] 391s 0.344 -0.030 391s 0.187 -0.539 391s testing: mul_mat, [3, 2, 2, 3] = [2, 3, 2, 3] * [2, 2, 2, 3] 391s x0: [2, 3, 2, 3] 391s 0.490 0.641 391s 0.994 -0.532 391s -0.112 -0.417 391s 391s x1: [2, 2, 2, 3] 391s -0.877 -0.687 391s 0.568 0.648 391s 391s y: [3, 2, 2, 3] 391s -0.870 -0.506 0.384 391s 0.694 0.219 -0.334 391s test-mul-mat0: iter:216/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.413 -0.569 -0.495 -0.851 391s -0.128 -0.249 -0.418 0.604 391s 0.571 -0.495 -0.670 0.279 391s -0.931 0.524 0.570 0.397 391s 391s x1: [4, 3, 1, 1] 391s -0.260 0.612 0.133 -0.201 391s 0.362 -0.109 -0.124 -0.678 391s -0.634 -0.839 0.019 -0.856 391s 391s y: [4, 3, 1, 1] 391s -0.136 -0.296 -0.596 0.558 391s 0.551 -0.377 0.155 -0.735 391s 1.459 -0.236 -0.199 -0.179 391s testing: mul_mat, [3, 4, 4, 1] = [4, 3, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 3, 4, 1] 391s 0.573 -0.788 -0.053 0.961 391s -0.248 -0.062 0.144 0.693 391s -0.096 -0.859 0.180 -0.463 391s 391s x1: [4, 4, 4, 1] 391s 0.499 -0.987 -0.862 0.111 391s 0.145 -0.063 -0.527 -0.964 391s 0.813 -0.205 -0.598 0.974 391s 0.814 -0.455 0.668 -0.600 391s 391s y: [3, 4, 4, 1] 391s 1.216 -0.111 0.593 391s -0.766 -0.777 0.392 391s 1.595 0.399 -0.461 391s 0.213 -0.493 0.710 391s testing: mul_mat, [1, 3, 4, 3] = [4, 1, 4, 3] * [4, 3, 4, 3] 391s x0: [4, 1, 4, 3] 391s -0.576 0.777 -0.157 -0.466 391s 391s x1: [4, 3, 4, 3] 391s -0.254 -0.816 -0.478 0.051 391s 0.465 0.788 -0.640 0.168 391s 0.907 -0.982 -0.601 0.683 391s 391s y: [1, 3, 4, 3] 391s -0.436 391s 0.366 391s -1.509 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.418 -0.240 0.621 -0.093 391s 0.186 0.840 -0.245 0.448 391s -0.530 0.233 -0.826 0.959 391s 391s x1: [4, 1, 1, 1] 391s -0.077 -0.514 0.051 0.614 391s 391s y: [3, 1, 1, 1] 391s 0.131 -0.183 0.468 391s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 2, 4, 1] 391s 0.555 -0.056 0.905 391s -0.727 -0.488 -0.266 391s 391s x1: [3, 4, 4, 1] 391s -0.273 -0.538 -0.319 391s 0.594 0.080 -0.824 391s -0.376 -0.212 0.671 391s -0.565 0.547 -0.772 391s 391s y: [2, 4, 4, 1] 391s -0.410 0.546 391s -0.421 -0.252 391s 0.410 0.199 391s -1.043 0.349 391s testing: mul_mat, [4, 3, 4, 3] = [2, 4, 4, 3] * [2, 3, 4, 3] 391s x0: [2, 4, 4, 3] 391s -0.332 -0.423 391s -0.211 0.038 391s -0.529 0.716 391s 0.882 0.322 391s 391s x1: [2, 3, 4, 3] 391s 0.502 0.394 391s -0.433 0.731 391s -0.538 0.967 391s 391s y: [4, 3, 4, 3] 391s -0.333 -0.091 0.017 0.570 391s -0.166 0.119 0.753 -0.147 391s -0.231 0.151 0.978 -0.164 391s test-mul-mat0: iter:217/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.879 0.051 0.085 391s 391s x1: [3, 1, 1, 1] 391s -0.835 -0.849 0.527 391s 391s y: [1, 1, 1, 1] 391s -0.732 391s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.835 -0.787 -0.014 391s 391s x1: [3, 1, 4, 1] 391s 0.960 0.960 0.814 391s 391s y: [1, 1, 4, 1] 391s 0.035 391s testing: mul_mat, [2, 1, 4, 1] = [3, 2, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.779 0.905 0.667 391s -0.972 0.342 -0.950 391s 391s x1: [3, 1, 4, 1] 391s 0.731 -0.903 -0.096 391s 391s y: [2, 1, 4, 1] 391s -1.451 -0.928 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.332 -0.691 0.723 391s -0.620 -0.304 -0.641 391s 391s x1: [3, 2, 1, 1] 391s 0.870 -0.069 -0.341 391s 0.505 -0.553 -0.525 391s 391s y: [2, 2, 1, 1] 391s -0.488 -0.300 391s -0.165 0.191 391s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.205 -0.657 391s -0.379 0.548 391s 0.189 -0.539 391s 0.901 0.621 391s 391s x1: [2, 3, 4, 1] 391s -0.440 0.278 391s 0.395 0.529 391s -0.929 0.480 391s 391s y: [4, 3, 4, 1] 391s -0.273 0.319 -0.233 -0.224 391s -0.266 0.140 -0.211 0.685 391s -0.505 0.615 -0.434 -0.540 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.746 0.001 -0.533 -0.135 391s 391s x1: [4, 2, 4, 1] 391s 0.214 0.235 -0.342 0.557 391s -0.217 0.119 0.179 0.726 391s 391s y: [1, 2, 4, 1] 391s 0.267 391s -0.356 391s test-mul-mat0: iter:218/500 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.850 391s 391s x1: [1, 1, 1, 1] 391s 0.537 391s 391s y: [1, 1, 1, 1] 391s -0.456 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.518 391s 0.404 391s 0.274 391s 0.570 391s 391s x1: [1, 1, 1, 1] 391s 0.174 391s 391s y: [4, 1, 1, 1] 391s -0.090 0.070 0.048 0.099 391s testing: mul_mat, [2, 4, 1, 2] = [1, 2, 1, 2] * [1, 4, 1, 2] 391s x0: [1, 2, 1, 2] 391s 0.079 391s 0.960 391s 391s x1: [1, 4, 1, 2] 391s -0.152 391s -0.041 391s 0.316 391s 0.849 391s 391s y: [2, 4, 1, 2] 391s -0.012 -0.146 391s -0.003 -0.040 391s 0.025 0.303 391s 0.067 0.815 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.846 391s 0.773 391s 0.144 391s 391s x1: [1, 2, 1, 1] 391s -0.707 391s -0.102 391s 391s y: [3, 2, 1, 1] 391s -0.598 -0.546 -0.102 391s -0.086 -0.079 -0.015 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.378 0.147 0.061 391s -0.335 -0.787 -0.564 391s -0.191 0.477 -0.968 391s -0.798 -0.283 -0.090 391s 391s x1: [3, 1, 1, 1] 391s -0.902 -0.044 -0.485 391s 391s y: [4, 1, 1, 1] 391s -0.376 0.610 0.621 0.776 391s testing: mul_mat, [3, 3, 1, 2] = [4, 3, 1, 2] * [4, 3, 1, 2] 391s x0: [4, 3, 1, 2] 391s 0.186 0.622 0.483 -0.037 391s -0.485 -0.453 -0.240 0.126 391s 0.393 -0.698 0.241 0.258 391s 391s x1: [4, 3, 1, 2] 391s 0.861 -0.787 -0.062 0.480 391s -0.634 -0.983 0.439 -0.498 391s -0.211 0.732 0.400 -0.673 391s 391s y: [3, 3, 1, 2] 391s -0.377 0.014 0.996 391s -0.498 0.584 0.414 391s 0.634 -0.410 -0.671 391s test-mul-mat0: iter:219/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.393 0.337 0.882 0.766 391s 0.050 -0.917 0.610 -0.562 391s -0.586 0.439 0.138 -0.748 391s 0.919 0.957 0.799 -0.803 391s 391s x1: [4, 4, 1, 1] 391s 0.547 -0.722 0.152 -0.267 391s -0.208 -0.455 -0.645 0.339 391s -0.153 0.839 -0.901 -0.912 391s -0.199 0.225 0.346 0.203 391s 391s y: [4, 4, 1, 1] 391s -0.099 0.932 -0.418 0.147 391s -0.544 -0.177 -0.421 -1.415 391s -1.271 -0.813 1.016 0.674 391s 0.459 -0.119 0.111 0.146 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.612 0.093 0.118 -0.529 391s 391s x1: [4, 4, 1, 1] 391s 0.108 -0.467 -0.011 0.653 391s -0.112 -0.672 -0.500 -0.273 391s -0.573 -0.412 0.528 0.652 391s 0.934 -0.269 -0.494 0.327 391s 391s y: [1, 4, 1, 1] 391s -0.457 391s 0.091 391s 0.030 391s -0.828 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.138 -0.938 -0.349 -0.063 391s -0.741 0.759 0.470 0.248 391s 391s x1: [4, 1, 1, 1] 391s -0.297 0.556 -0.115 -0.857 391s 391s y: [2, 1, 1, 1] 391s -0.469 0.376 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.565 0.413 0.021 0.701 391s 0.345 -0.450 0.558 -0.218 391s -0.838 -0.145 -0.912 0.044 391s 0.762 -0.468 0.907 0.763 391s 391s x1: [4, 2, 1, 1] 391s 0.412 -0.642 0.576 0.913 391s 0.085 -0.996 -0.499 -0.387 391s 391s y: [4, 2, 1, 1] 391s 0.154 0.554 -0.738 1.834 391s -0.742 0.283 0.511 -0.217 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.331 -0.734 -0.931 -0.759 391s 0.214 0.219 -0.202 0.628 391s 391s x1: [4, 4, 1, 1] 391s 0.433 0.982 -0.978 0.192 391s 0.452 0.270 -0.396 0.810 391s -0.153 -0.483 -0.105 -0.149 391s 0.018 0.508 -0.493 0.452 391s 391s y: [2, 4, 1, 1] 391s -0.099 0.626 391s -0.594 0.744 391s 0.616 -0.211 391s -0.263 0.499 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.707 -0.375 391s 391s x1: [2, 4, 1, 1] 391s -0.114 -0.853 391s 0.329 0.668 391s 0.192 0.092 391s 0.101 0.173 391s 391s y: [1, 4, 1, 1] 391s 0.400 391s -0.483 391s -0.170 391s -0.136 391s test-mul-mat0: iter:220/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.244 -0.281 -0.205 0.642 391s -0.134 -0.876 0.310 -0.942 391s 0.216 -0.589 0.231 -0.670 391s -0.296 0.856 -0.285 0.601 391s 391s x1: [4, 3, 1, 1] 391s 0.414 0.330 -0.917 -0.568 391s -0.162 -0.410 0.884 0.691 391s 0.259 0.098 0.957 -0.522 391s 391s y: [4, 3, 1, 1] 391s -0.371 -0.093 0.064 0.079 391s 0.417 0.004 -0.052 -0.140 391s -0.622 0.668 0.569 -0.579 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s 0.076 -0.375 0.254 -0.110 391s 391s x1: [4, 4, 3, 1] 391s -0.710 0.947 0.015 0.620 391s -0.970 0.447 -0.542 -0.381 391s 0.331 -0.851 0.878 -0.571 391s -0.894 -0.643 0.596 -0.138 391s 391s y: [1, 4, 3, 1] 391s -0.473 391s -0.337 391s 0.629 391s 0.339 391s testing: mul_mat, [2, 1, 3, 3] = [4, 2, 3, 3] * [4, 1, 3, 3] 391s x0: [4, 2, 3, 3] 391s 0.585 -0.336 -0.105 0.019 391s 0.036 -0.180 0.203 0.119 391s 391s x1: [4, 1, 3, 3] 391s -0.426 -0.682 -0.454 -0.881 391s 391s y: [2, 1, 3, 3] 391s 0.011 -0.089 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.382 0.054 -0.931 0.001 391s -0.637 -0.974 0.820 0.716 391s -0.123 -0.306 0.451 -0.601 391s -0.960 -0.215 0.228 0.257 391s 391s x1: [4, 2, 1, 1] 391s -0.820 -0.064 0.564 0.210 391s -0.904 -0.888 -0.634 -0.869 391s 391s y: [4, 2, 1, 1] 391s -0.842 1.198 0.249 0.984 391s 0.197 0.299 0.619 0.691 391s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.980 -0.828 -0.585 -0.922 391s 0.465 0.119 -0.145 -0.183 391s 391s x1: [4, 4, 3, 1] 391s -0.385 -0.599 0.790 0.268 391s 0.586 -0.663 0.448 -0.478 391s 0.901 -0.342 -0.382 -0.987 391s 0.023 -0.251 0.711 -0.595 391s 391s y: [2, 4, 3, 1] 391s -0.590 -0.414 391s 1.302 0.216 391s 2.299 0.614 391s 0.363 -0.013 391s testing: mul_mat, [1, 4, 3, 3] = [2, 1, 3, 3] * [2, 4, 3, 3] 391s x0: [2, 1, 3, 3] 391s 0.182 -0.976 391s 391s x1: [2, 4, 3, 3] 391s -0.892 -0.651 391s -0.472 0.077 391s 0.884 0.827 391s -0.366 0.864 391s 391s y: [1, 4, 3, 3] 391s 0.473 391s -0.161 391s -0.646 391s -0.909 391s test-mul-mat0: iter:221/500 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.750 391s -0.164 391s -0.453 391s 391s x1: [1, 4, 1, 1] 391s -0.330 391s -0.083 391s 0.305 391s 0.623 391s 391s y: [3, 4, 1, 1] 391s 0.247 0.054 0.149 391s 0.062 0.014 0.038 391s -0.229 -0.050 -0.138 391s -0.467 -0.102 -0.282 391s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 1, 2, 1] 391s 0.315 391s 391s x1: [1, 3, 2, 1] 391s 0.399 391s -0.982 391s -0.428 391s 391s y: [1, 3, 2, 1] 391s 0.126 391s -0.309 391s -0.135 391s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.068 391s 391s x1: [1, 1, 2, 1] 391s -0.906 391s 391s y: [1, 1, 2, 1] 391s 0.062 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.106 391s 391s x1: [1, 1, 1, 1] 391s -0.218 391s 391s y: [1, 1, 1, 1] 391s 0.023 391s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.319 391s 0.204 391s 391s x1: [1, 1, 2, 1] 391s -0.369 391s 391s y: [2, 1, 2, 1] 391s -0.118 -0.075 391s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.552 0.122 391s -0.449 -0.540 391s -0.353 -0.842 391s -0.534 0.505 391s 391s x1: [2, 1, 2, 1] 391s -0.879 -0.002 391s 391s y: [4, 1, 2, 1] 391s 0.485 0.396 0.313 0.468 391s test-mul-mat0: iter:222/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.124 391s -0.027 391s 391s x1: [1, 1, 1, 1] 391s 0.769 391s 391s y: [2, 1, 1, 1] 391s -0.096 -0.021 391s testing: mul_mat, [2, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.640 391s 0.391 391s 391s x1: [1, 2, 4, 1] 391s -0.252 391s -0.135 391s 391s y: [2, 2, 4, 1] 391s -0.161 -0.098 391s -0.087 -0.053 391s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 4, 4, 1] 391s 0.887 391s 0.563 391s -0.443 391s -0.237 391s 391s x1: [1, 2, 4, 1] 391s 0.261 391s -0.697 391s 391s y: [4, 2, 4, 1] 391s 0.232 0.147 -0.116 -0.062 391s -0.618 -0.392 0.309 0.165 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.009 391s -0.319 391s 0.243 391s 0.496 391s 391s x1: [1, 4, 1, 1] 391s 0.145 391s 0.785 391s 0.969 391s 0.420 391s 391s y: [4, 4, 1, 1] 391s -0.001 -0.046 0.035 0.072 391s -0.007 -0.251 0.191 0.390 391s -0.009 -0.309 0.235 0.481 391s -0.004 -0.134 0.102 0.208 391s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.487 -0.749 -0.753 0.983 391s 0.828 -0.803 0.396 -0.784 391s 391s x1: [4, 1, 4, 1] 391s -0.915 0.627 -0.947 -0.584 391s 391s y: [2, 1, 4, 1] 391s 0.115 -1.177 391s testing: mul_mat, [2, 4, 4, 1] = [2, 2, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 2, 4, 1] 391s -0.457 0.981 391s -0.512 0.544 391s 391s x1: [2, 4, 4, 1] 391s -0.312 -0.009 391s 0.019 -0.065 391s -0.614 0.001 391s 0.152 0.202 391s 391s y: [2, 4, 4, 1] 391s 0.133 0.155 391s -0.072 -0.045 391s 0.281 0.315 391s 0.129 0.032 391s test-mul-mat0: iter:223/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.036 -0.319 391s -0.333 0.829 391s 0.285 -0.809 391s 391s x1: [2, 1, 1, 1] 391s -0.041 -0.184 391s 391s y: [3, 1, 1, 1] 391s 0.060 -0.139 0.137 391s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.915 -0.260 391s 391s x1: [2, 3, 2, 1] 391s 0.382 -0.121 391s -0.265 0.869 391s -0.140 -0.721 391s 391s y: [1, 3, 2, 1] 391s -0.318 391s 0.016 391s 0.316 391s testing: mul_mat, [3, 1, 2, 3] = [2, 3, 2, 3] * [2, 1, 2, 3] 391s x0: [2, 3, 2, 3] 391s -0.150 -0.328 391s -0.256 0.585 391s -0.458 0.604 391s 391s x1: [2, 1, 2, 3] 391s 0.268 -0.538 391s 391s y: [3, 1, 2, 3] 391s 0.136 -0.383 -0.448 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.540 0.913 391s 391s x1: [2, 3, 1, 1] 391s -0.303 0.016 391s -0.769 -0.655 391s -0.374 0.788 391s 391s y: [1, 3, 1, 1] 391s -0.149 391s -1.013 391s 0.518 391s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 1, 2, 1] 391s 0.522 391s 391s x1: [1, 2, 2, 1] 391s 0.573 391s 0.194 391s 391s y: [1, 2, 2, 1] 391s 0.300 391s 0.101 391s testing: mul_mat, [4, 1, 2, 3] = [1, 4, 2, 3] * [1, 1, 2, 3] 391s x0: [1, 4, 2, 3] 391s -0.638 391s -0.669 391s -0.142 391s 0.929 391s 391s x1: [1, 1, 2, 3] 391s -0.880 391s 391s y: [4, 1, 2, 3] 391s 0.561 0.589 0.125 -0.817 391s test-mul-mat0: iter:224/500 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.058 0.034 391s 0.698 0.210 391s 0.683 0.794 391s 391s x1: [2, 3, 1, 1] 391s -0.674 -0.243 391s -0.775 -0.312 391s 0.088 0.083 391s 391s y: [3, 3, 1, 1] 391s -0.047 -0.521 -0.653 391s -0.056 -0.606 -0.776 391s 0.008 0.079 0.126 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.932 -0.763 391s 391s x1: [2, 3, 1, 1] 391s 0.592 -0.324 391s -0.483 0.763 391s 0.585 -0.220 391s 391s y: [1, 3, 1, 1] 391s -0.305 391s -0.132 391s -0.377 391s testing: mul_mat, [1, 1, 1, 3] = [2, 1, 1, 3] * [2, 1, 1, 3] 391s x0: [2, 1, 1, 3] 391s 0.405 0.218 391s 391s x1: [2, 1, 1, 3] 391s 0.418 -0.696 391s 391s y: [1, 1, 1, 3] 391s 0.018 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.691 0.518 391s 0.561 0.244 391s -0.480 -0.686 391s 391s x1: [2, 1, 1, 1] 391s 0.250 -0.473 391s 391s y: [3, 1, 1, 1] 391s -0.418 0.025 0.204 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.573 -0.861 0.656 391s 0.847 0.877 -0.350 391s 391s x1: [3, 2, 1, 1] 391s 0.111 0.920 -0.169 391s -0.126 0.505 0.611 391s 391s y: [2, 2, 1, 1] 391s -0.839 0.959 391s -0.106 0.122 391s testing: mul_mat, [1, 3, 1, 3] = [2, 1, 1, 3] * [2, 3, 1, 3] 391s x0: [2, 1, 1, 3] 391s 0.070 -0.948 391s 391s x1: [2, 3, 1, 3] 391s 0.888 0.157 391s 0.841 0.293 391s -0.625 0.359 391s 391s y: [1, 3, 1, 3] 391s -0.087 391s -0.219 391s -0.384 391s test-mul-mat0: iter:225/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.912 391s -0.677 391s 0.869 391s 0.206 391s 391s x1: [1, 1, 1, 1] 391s 0.166 391s 391s y: [4, 1, 1, 1] 391s 0.151 -0.112 0.144 0.034 391s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 2, 2, 1] 391s -0.231 391s -0.915 391s 391s x1: [1, 4, 2, 1] 391s -0.302 391s 0.227 391s -0.338 391s -0.483 391s 391s y: [2, 4, 2, 1] 391s 0.070 0.276 391s -0.052 -0.208 391s 0.078 0.310 391s 0.112 0.442 391s testing: mul_mat, [2, 2, 2, 4] = [1, 2, 2, 4] * [1, 2, 2, 4] 391s x0: [1, 2, 2, 4] 391s -0.146 391s -0.284 391s 391s x1: [1, 2, 2, 4] 391s 0.306 391s -0.440 391s 391s y: [2, 2, 2, 4] 391s -0.045 -0.087 391s 0.064 0.125 391s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.338 391s 391s x1: [1, 2, 1, 1] 391s 0.773 391s 0.536 391s 391s y: [1, 2, 1, 1] 391s 0.261 391s 0.181 391s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.086 391s 0.832 391s 391s x1: [1, 1, 2, 1] 391s -0.444 391s 391s y: [2, 1, 2, 1] 391s -0.038 -0.369 391s testing: mul_mat, [4, 1, 2, 4] = [2, 4, 2, 4] * [2, 1, 2, 4] 391s x0: [2, 4, 2, 4] 391s -0.507 -0.054 391s 0.410 0.705 391s 0.884 0.604 391s -0.733 0.502 391s 391s x1: [2, 1, 2, 4] 391s 0.554 -0.566 391s 391s y: [4, 1, 2, 4] 391s -0.250 -0.171 0.148 -0.690 391s test-mul-mat0: iter:226/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.504 391s 391s x1: [1, 4, 1, 1] 391s -0.544 391s -0.404 391s -0.355 391s -0.537 391s 391s y: [1, 4, 1, 1] 391s -0.274 391s -0.204 391s -0.179 391s -0.271 391s testing: mul_mat, [4, 1, 3, 1] = [1, 4, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.343 391s -0.948 391s 0.716 391s -0.486 391s 391s x1: [1, 1, 3, 1] 391s 0.910 391s 391s y: [4, 1, 3, 1] 391s 0.312 -0.863 0.652 -0.442 391s testing: mul_mat, [1, 4, 3, 4] = [1, 1, 3, 4] * [1, 4, 3, 4] 391s x0: [1, 1, 3, 4] 391s 0.640 391s 391s x1: [1, 4, 3, 4] 391s -0.123 391s -0.005 391s 0.480 391s -0.768 391s 391s y: [1, 4, 3, 4] 391s -0.079 391s -0.003 391s 0.307 391s -0.492 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.271 391s -0.808 391s 391s x1: [1, 1, 1, 1] 391s -0.849 391s 391s y: [2, 1, 1, 1] 391s 0.230 0.686 391s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 4, 3, 1] 391s 0.599 -0.479 391s -0.080 -0.727 391s -0.970 0.649 391s 0.319 -0.081 391s 391s x1: [2, 1, 3, 1] 391s -0.907 0.873 391s 391s y: [4, 1, 3, 1] 391s -0.962 -0.561 1.447 -0.360 391s testing: mul_mat, [1, 2, 3, 4] = [4, 1, 3, 4] * [4, 2, 3, 4] 391s x0: [4, 1, 3, 4] 391s 0.676 0.841 -0.210 -0.176 391s 391s x1: [4, 2, 3, 4] 391s -0.567 0.864 -0.540 -0.440 391s -0.505 -0.016 -0.489 -0.906 391s 391s y: [1, 2, 3, 4] 391s 0.535 391s -0.093 391s test-mul-mat0: iter:227/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.426 0.791 391s 0.633 -0.546 391s -0.064 -0.158 391s 391s x1: [2, 2, 1, 1] 391s -0.533 -0.280 391s 0.054 0.464 391s 391s y: [3, 2, 1, 1] 391s 0.005 -0.184 0.079 391s 0.344 -0.219 -0.077 391s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 2, 4, 1] 391s -0.278 -0.867 391s -0.109 0.176 391s 391s x1: [2, 3, 4, 1] 391s -0.825 -0.130 391s -0.125 -0.988 391s -0.405 0.918 391s 391s y: [2, 3, 4, 1] 391s 0.342 0.067 391s 0.891 -0.160 391s -0.684 0.206 391s testing: mul_mat, [4, 2, 4, 1] = [2, 4, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.517 -0.072 391s -0.993 0.249 391s -0.721 0.945 391s 0.857 0.643 391s 391s x1: [2, 2, 4, 1] 391s -0.163 -0.943 391s -0.166 0.995 391s 391s y: [4, 2, 4, 1] 391s -0.016 -0.073 -0.774 -0.747 391s -0.158 0.412 1.060 0.498 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.559 0.393 391s -0.409 -0.951 391s -0.173 -0.912 391s 0.406 -0.810 391s 391s x1: [2, 4, 1, 1] 391s 0.172 -0.800 391s 0.422 0.452 391s -0.855 0.279 391s 0.095 -0.375 391s 391s y: [4, 4, 1, 1] 391s -0.218 0.690 0.700 0.718 391s 0.414 -0.602 -0.485 -0.194 391s -0.368 0.085 -0.106 -0.573 391s -0.094 0.318 0.326 0.342 391s testing: mul_mat, [4, 2, 4, 1] = [4, 4, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.754 0.063 -0.383 -0.663 391s -0.818 -0.274 -0.810 -0.944 391s 0.492 0.214 0.528 -0.832 391s -0.517 0.199 -0.319 0.937 391s 391s x1: [4, 2, 4, 1] 391s -0.552 -0.459 -0.227 -0.568 391s 0.204 -0.762 -0.119 0.234 391s 391s y: [4, 2, 4, 1] 391s 0.850 1.297 -0.017 -0.265 391s -0.312 -0.083 -0.321 0.000 391s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 4, 4, 1] 391s 0.949 0.173 0.580 0.859 391s -0.759 -0.699 -0.442 0.779 391s 0.957 0.115 0.545 -0.337 391s -0.666 0.247 -0.784 0.574 391s 391s x1: [4, 4, 4, 1] 391s -0.915 -0.428 0.499 -0.117 391s 0.190 0.607 -0.701 0.997 391s 0.243 -0.604 0.142 -0.065 391s -0.339 -0.776 0.358 0.858 391s 391s y: [4, 4, 4, 1] 391s -0.753 0.682 -0.614 0.046 391s 0.735 0.519 -0.466 1.145 391s 0.152 0.124 0.262 -0.460 391s 0.489 1.310 -0.508 0.246 391s test-mul-mat0: iter:228/500 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.034 391s 391s x1: [1, 3, 1, 1] 391s 0.992 391s 0.828 391s 0.575 391s 391s y: [1, 3, 1, 1] 391s -0.034 391s -0.028 391s -0.019 391s testing: mul_mat, [1, 1, 3, 1] = [1, 1, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 1, 3, 1] 391s -0.697 391s 391s x1: [1, 1, 3, 1] 391s -0.960 391s 391s y: [1, 1, 3, 1] 391s 0.669 391s testing: mul_mat, [4, 1, 3, 4] = [1, 4, 3, 4] * [1, 1, 3, 4] 391s x0: [1, 4, 3, 4] 391s -0.659 391s -0.766 391s -0.087 391s 0.884 391s 391s x1: [1, 1, 3, 4] 391s 0.611 391s 391s y: [4, 1, 3, 4] 391s -0.403 -0.468 -0.053 0.541 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.327 391s 0.578 391s 391s x1: [1, 4, 1, 1] 391s 0.153 391s 0.507 391s -0.715 391s 0.445 391s 391s y: [2, 4, 1, 1] 391s 0.050 0.089 391s 0.166 0.293 391s -0.234 -0.413 391s 0.145 0.257 391s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 4, 3, 1] 391s 0.815 -0.418 391s 0.776 -0.019 391s 0.132 -0.674 391s -0.868 0.685 391s 391s x1: [2, 1, 3, 1] 391s -0.611 0.289 391s 391s y: [4, 1, 3, 1] 391s -0.619 -0.480 -0.275 0.728 391s testing: mul_mat, [2, 2, 3, 4] = [4, 2, 3, 4] * [4, 2, 3, 4] 391s x0: [4, 2, 3, 4] 391s 0.255 0.642 0.445 0.160 391s 0.424 0.358 0.288 -0.289 391s 391s x1: [4, 2, 3, 4] 391s 0.863 -0.532 -0.778 -0.307 391s 0.850 0.661 -0.783 0.665 391s 391s y: [2, 2, 3, 4] 391s -0.518 0.041 391s 0.399 0.180 391s test-mul-mat0: iter:229/500 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.934 -0.386 391s 0.326 0.611 391s -0.180 -0.395 391s 0.390 0.372 391s 391s x1: [2, 2, 1, 1] 391s -0.545 0.792 391s 0.682 -0.908 391s 391s y: [4, 2, 1, 1] 391s -0.815 0.306 -0.215 0.083 391s 0.988 -0.332 0.236 -0.072 391s testing: mul_mat, [4, 4, 2, 1] = [2, 4, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.592 0.829 391s 0.846 -0.910 391s 0.921 -0.177 391s -0.976 -0.465 391s 391s x1: [2, 4, 2, 1] 391s -0.657 0.045 391s -0.550 -0.186 391s -0.527 0.324 391s 0.914 -0.584 391s 391s y: [4, 4, 2, 1] 391s 0.426 -0.597 -0.613 0.620 391s 0.172 -0.296 -0.473 0.623 391s 0.581 -0.741 -0.543 0.364 391s -1.025 1.305 0.945 -0.620 391s testing: mul_mat, [4, 4, 2, 4] = [2, 4, 2, 4] * [2, 4, 2, 4] 391s x0: [2, 4, 2, 4] 391s 0.260 0.212 391s 0.716 0.011 391s -0.586 0.162 391s -0.407 0.129 391s 391s x1: [2, 4, 2, 4] 391s 0.177 0.911 391s -0.458 -0.499 391s 0.825 -0.042 391s -0.180 0.991 391s 391s y: [4, 4, 2, 4] 391s 0.239 0.137 0.043 0.045 391s -0.225 -0.333 0.188 0.122 391s 0.206 0.590 -0.490 -0.341 391s 0.163 -0.119 0.266 0.201 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.688 0.879 391s 0.277 -0.235 391s -0.442 0.015 391s -0.228 0.890 391s 391s x1: [2, 4, 1, 1] 391s -0.401 -0.700 391s -0.239 0.934 391s -0.837 0.191 391s -0.372 -0.625 391s 391s y: [4, 4, 1, 1] 391s -0.891 0.054 0.166 -0.532 391s 0.656 -0.285 0.120 0.886 391s -0.408 -0.276 0.373 0.361 391s -0.805 0.044 0.155 -0.471 391s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.416 0.212 -0.100 0.671 391s -0.774 -0.399 0.878 0.757 391s 0.288 0.639 -0.802 -0.037 391s 391s x1: [4, 2, 2, 1] 391s 0.095 0.180 0.350 -0.805 391s -0.596 0.654 0.965 0.302 391s 391s y: [3, 2, 2, 1] 391s -0.577 -0.448 -0.108 391s 0.493 1.277 -0.539 391s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 391s x0: [3, 2, 2, 4] 391s 0.375 -0.840 -0.091 391s -0.206 0.687 -0.537 391s 391s x1: [3, 4, 2, 4] 391s 0.448 -0.675 -0.284 391s -0.215 -0.025 -0.221 391s -0.118 0.559 0.005 391s -0.830 -0.229 0.606 391s 391s y: [2, 4, 2, 4] 391s 0.761 -0.403 391s -0.040 0.146 391s -0.514 0.406 391s -0.174 -0.312 391s test-mul-mat0: iter:230/500 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.326 -0.396 0.770 391s -0.445 -0.044 -0.392 391s 391s x1: [3, 4, 1, 1] 391s -0.893 0.448 0.458 391s 0.397 0.096 -0.100 391s 0.135 0.476 0.969 391s 0.752 0.231 0.832 391s 391s y: [2, 4, 1, 1] 391s 0.467 0.197 391s -0.244 -0.141 391s 0.513 -0.461 391s 0.303 -0.671 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.865 -0.504 0.567 391s 0.864 -0.085 0.674 391s 0.312 -0.627 0.071 391s 391s x1: [3, 2, 1, 1] 391s 0.739 0.595 0.015 391s -0.458 -0.488 -0.227 391s 391s y: [3, 2, 1, 1] 391s 0.348 0.599 -0.141 391s -0.279 -0.508 0.147 391s testing: mul_mat, [1, 3, 1, 4] = [3, 1, 1, 4] * [3, 3, 1, 4] 391s x0: [3, 1, 1, 4] 391s 0.188 -0.975 -0.655 391s 391s x1: [3, 3, 1, 4] 391s -0.592 0.273 -0.795 391s 0.884 0.242 0.957 391s 0.115 0.074 -0.514 391s 391s y: [1, 3, 1, 4] 391s 0.143 391s -0.697 391s 0.286 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.915 0.204 -0.317 391s 0.142 -0.013 -0.642 391s -0.638 -0.832 0.348 391s -0.615 0.639 -0.416 391s 391s x1: [3, 1, 1, 1] 391s -0.973 0.308 0.995 391s 391s y: [4, 1, 1, 1] 391s -1.143 -0.781 0.710 0.382 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.069 -0.810 0.120 -0.293 391s -0.805 -0.242 -0.461 -0.825 391s 0.346 -0.846 -0.464 -0.210 391s -0.760 0.332 -0.893 0.065 391s 391s x1: [4, 3, 1, 1] 391s -0.568 -0.701 -0.558 0.620 391s -0.676 -0.213 -0.403 -0.973 391s 0.388 0.917 -0.635 -0.148 391s 391s y: [4, 3, 1, 1] 391s 0.358 0.372 0.524 0.738 391s 0.456 1.584 0.338 0.741 391s -0.802 -0.120 -0.315 0.567 391s testing: mul_mat, [1, 4, 1, 4] = [4, 1, 1, 4] * [4, 4, 1, 4] 391s x0: [4, 1, 1, 4] 391s -0.696 -0.343 -0.339 0.068 391s 391s x1: [4, 4, 1, 4] 391s 0.523 0.373 0.496 0.822 391s 0.816 0.117 -0.854 -0.397 391s 0.714 -0.826 0.991 0.630 391s -0.462 -0.157 -0.050 0.469 391s 391s y: [1, 4, 1, 4] 391s -0.605 391s -0.345 391s -0.506 391s 0.424 391s test-mul-mat0: iter:231/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.284 -0.135 -0.803 -0.555 391s -0.130 -0.430 0.738 -0.649 391s -0.362 -0.889 -0.225 0.442 391s 391s x1: [4, 3, 1, 1] 391s -0.215 -0.484 -0.086 0.670 391s -0.455 0.806 0.624 -0.251 391s 0.493 -0.978 -0.377 0.798 391s 391s y: [3, 3, 1, 1] 391s -0.298 -0.262 0.824 391s -0.599 0.336 -0.804 391s 0.132 -0.439 1.129 391s testing: mul_mat, [2, 3, 4, 1] = [4, 2, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.778 -0.533 0.640 -0.202 391s -0.921 -0.109 -0.613 0.429 391s 391s x1: [4, 3, 4, 1] 391s 0.861 0.854 -0.789 -0.958 391s 0.450 0.504 -0.172 0.966 391s -0.582 -0.502 -0.489 -0.776 391s 391s y: [2, 3, 4, 1] 391s -0.096 -0.813 391s -0.224 0.051 391s -0.341 0.558 391s testing: mul_mat, [1, 2, 4, 2] = [4, 1, 4, 2] * [4, 2ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s , 4, 2] 391s x0: [4, 1, 4, 2] 391s 0.960 0.992 -0.161 0.349 391s 391s x1: [4, 2, 4, 2] 391s 0.773 -0.980 -0.197 0.852 391s -0.088 0.189 0.281 0.405 391s 391s y: [1, 2, 4, 2] 391s 0.099 391s 0.199 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.343 -0.229 -0.579 -0.149 391s 0.758 0.063 -0.353 0.785 391s 0.333 -0.738 0.904 0.998 391s 0.508 0.957 -0.443 -0.665 391s 391s x1: [4, 1, 1, 1] 391s -0.597 0.102 0.748 -0.240 391s 391s y: [4, 1, 1, 1] 391s -0.216 -0.898 0.163 -0.377 391s testing: mul_mat, [4, 4, 4, 1] = [4, 4, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.616 -0.168 0.049 0.352 391s -0.581 -0.257 0.132 0.633 391s 0.698 0.517 -0.820 0.062 391s 0.850 0.833 -0.956 0.686 391s 391s x1: [4, 4, 4, 1] 391s -0.223 -0.606 -0.215 0.364 391s 0.773 0.894 0.803 -0.816 391s -0.843 -0.799 -0.413 0.260 391s 0.948 0.347 -0.310 -0.395 391s 391s y: [4, 4, 4, 1] 391s 0.356 0.487 -0.270 -0.239 391s -0.874 -1.089 0.293 0.075 391s 0.724 0.804 -0.646 -0.809 391s -0.796 -0.931 1.071 1.121 391s testing: mul_mat, [2, 4, 4, 2] = [4, 2, 4, 2] * [4, 4, 4, 2] 391s x0: [4, 2, 4, 2] 391s 0.324 0.954 -0.524 0.725 391s -0.855 0.815 0.925 0.878 391s 391s x1: [4, 4, 4, 2] 391s -0.697 0.931 -0.689 0.103 391s 0.847 0.768 -0.049 -0.105 391s 0.405 -0.157 0.294 -0.325 391s 0.798 0.380 -0.094 0.917 391s 391s y: [2, 4, 4, 2] 391s 1.098 0.808 391s 0.956 -0.237 391s -0.408 -0.487 391s 1.334 0.346 391s test-mul-mat0: iter:232/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.238 391s -0.909 391s 391s x1: [1, 1, 1, 1] 391s 0.430 391s 391s y: [2, 1, 1, 1] 391s -0.102 -0.391 391s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.997 391s 0.764 391s 391s x1: [1, 2, 3, 1] 391s 0.388 391s -0.810 391s 391s y: [2, 2, 3, 1] 391s -0.387 0.297 391s 0.808 -0.619 391s testing: mul_mat, [4, 2, 3, 4] = [1, 4, 3, 4] * [1, 2, 3, 4] 391s x0: [1, 4, 3, 4] 391s 0.292 391s -0.566 391s 0.431 391s -0.388 391s 391s x1: [1, 2, 3, 4] 391s -0.218 391s 0.420 391s 391s y: [4, 2, 3, 4] 391s -0.064 0.123 -0.094 0.085 391s 0.123 -0.238 0.181 -0.163 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.630 391s -0.418 391s 0.391 391s 391s x1: [1, 4, 1, 1] 391s -0.794 391s 0.398 391s -0.969 391s -0.837 391s 391s y: [3, 4, 1, 1] 391s -0.500 0.332 -0.310 391s 0.251 -0.166 0.156 391s -0.611 0.405 -0.379 391s -0.528 0.350 -0.327 391s testing: mul_mat, [1, 1, 3, 1] = [3, 1, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.258 -0.937 -0.432 391s 391s x1: [3, 1, 3, 1] 391s 0.888 -0.411 0.714 391s 391s y: [1, 1, 3, 1] 391s 0.306 391s testing: mul_mat, [2, 3, 3, 4] = [1, 2, 3, 4] * [1, 3, 3, 4] 391s x0: [1, 2, 3, 4] 391s -0.339 391s 0.596 391s 391s x1: [1, 3, 3, 4] 391s -0.816 391s 0.801 391s 0.357 391s 391s y: [2, 3, 3, 4] 391s 0.276 -0.486 391s -0.271 0.477 391s -0.121 0.213 391s test-mul-mat0: iter:233/500 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.739 0.046 391s 391s x1: [2, 3, 1, 1] 391s 0.385 -0.682 391s -0.967 -0.954 391s 0.914 0.817 391s 391s y: [1, 3, 1, 1] 391s -0.316 391s 0.671 391s -0.638 391s testing: mul_mat, [4, 1, 4, 1] = [2, 4, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.425 0.058 391s 0.995 0.730 391s -0.843 0.848 391s 0.444 0.534 391s 391s x1: [2, 1, 4, 1] 391s 0.494 0.950 391s 391s y: [4, 1, 4, 1] 391s -0.154 1.185 0.389 0.726 391s testing: mul_mat, [3, 4, 4, 3] = [2, 3, 4, 3] * [2, 4, 4, 3] 391s x0: [2, 3, 4, 3] 391s -0.383 -0.363 391s -0.594 -0.922 391s -0.892 0.753 391s 391s x1: [2, 4, 4, 3] 391s 0.489 -0.598 391s 0.716 0.646 391s -0.750 0.160 391s 0.180 0.571 391s 391s y: [3, 4, 4, 3] 391s 0.030 0.260 -0.886 391s -0.508 -1.021 -0.152 391s 0.229 0.299 0.789 391s -0.276 -0.633 0.269 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.690 -0.254 391s 0.551 -0.322 391s 0.472 0.402 391s -0.832 -0.548 391s 391s x1: [2, 3, 1, 1] 391s 0.156 0.958 391s -0.148 -0.508 391s -0.510 0.466 391s 391s y: [4, 3, 1, 1] 391s -0.351 -0.222 0.459 -0.655 391s 0.231 0.082 -0.274 0.402 391s 0.234 -0.431 -0.053 0.169 391s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 3, 4, 1] 391s -0.101 0.955 0.685 -0.773 391s 0.860 0.469 -0.903 -0.222 391s -0.407 0.921 0.102 -0.832 391s 391s x1: [4, 2, 4, 1] 391s 0.981 -0.188 -0.235 0.419 391s 0.241 0.056 0.609 0.329 391s 391s y: [3, 2, 4, 1] 391s -0.764 0.875 -0.946 391s 0.192 -0.389 -0.258 391s testing: mul_mat, [2, 4, 4, 3] = [3, 2, 4, 3] * [3, 4, 4, 3] 391s x0: [3, 2, 4, 3] 391s -0.430 -0.418 -0.600 391s -0.318 -0.401 0.205 391s 391s x1: [3, 4, 4, 3] 391s -0.456 -0.014 0.066 391s 0.959 0.223 0.154 391s 0.155 -0.403 -0.075 391s -0.188 -0.489 -0.169 391s 391s y: [2, 4, 4, 3] 391s 0.162 0.164 391s -0.598 -0.363 391s 0.147 0.097 391s 0.387 0.222 391s test-mul-mat0: iter:234/500 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.345 -0.436 391s -0.675 0.495 391s 0.102 -0.703 391s 391s x1: [2, 4, 1, 1] 391s -0.647 -0.414 391s -0.227 0.894 391s -0.072 -0.393 391s 0.265 -0.836 391s 391s y: [3, 4, 1, 1] 391s -0.043 0.232 0.225 391s -0.468 0.596 -0.652 391s 0.146 -0.146 0.269 391s 0.455 -0.593 0.615 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.861 0.338 391s 0.573 -0.700 391s 391s x1: [2, 3, 1, 1] 391s -0.244 0.339 391s -0.549 0.808 391s 0.857 0.389 391s 391s y: [2, 3, 1, 1] 391s -0.096 -0.377 391s -0.200 -0.880 391s 0.869 0.219 391s testing: mul_mat, [2, 2, 1, 3] = [2, 2, 1, 3] * [2, 2, 1, 3] 391s x0: [2, 2, 1, 3] 391s -0.877 -0.312 391s -0.554 -0.552 391s 391s x1: [2, 2, 1, 3] 391s 0.476 -0.679 391s 0.830 -0.797 391s 391s y: [2, 2, 1, 3] 391s -0.205 0.112 391s -0.479 -0.020 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.096 -0.266 391s 0.397 0.600 391s 0.692 -0.533 391s 391s x1: [2, 2, 1, 1] 391s -0.416 0.798 391s -0.396 0.922 391s 391s y: [3, 2, 1, 1] 391s -0.172 0.313 -0.713 391s -0.207 0.396 -0.765 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.987 0.025 -0.257 391s -0.705 -0.567 0.208 391s 391s x1: [3, 2, 1, 1] 391s -0.083 -0.611 -0.984 391s -0.005 -0.894 -0.828 391s 391s y: [2, 2, 1, 1] 391s 0.156 0.200 391s 0.185 0.339 391s testing: mul_mat, [3, 3, 1, 3] = [2, 3, 1, 3] * [2, 3, 1, 3] 391s x0: [2, 3, 1, 3] 391s -0.425 -0.319 391s -0.763 -0.591 391s 0.765 -0.897 391s 391s x1: [2, 3, 1, 3] 391s 0.980 0.487 391s -0.853 -0.133 391s -0.318 -0.107 391s 391s y: [3, 3, 1, 3] 391s -0.572 -1.036 0.313 391s 0.405 0.730 -0.533 391s 0.169 0.306 -0.147 391s test-mul-mat0: iter:235/500 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.418 -0.662 0.279 -0.902 391s -0.253 0.382 -0.234 0.451 391s 391s x1: [4, 2, 1, 1] 391s -0.929 -0.872 -0.109 -0.343 391s 0.881 -0.256 -0.157 -0.899 391s 391s y: [2, 2, 1, 1] 391s 0.468 -0.227 391s 1.304 -0.689 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s -0.687 0.392 0.000 -0.855 391s 391s x1: [4, 2, 4, 1] 391s -0.490 0.867 -0.102 -0.154 391s -0.052 0.832 0.329 -0.104 391s 391s y: [1, 2, 4, 1] 391s 0.809 391s 0.452 391s testing: mul_mat, [1, 1, 4, 3] = [4, 1, 4, 3] * [4, 1, 4, 3] 391s x0: [4, 1, 4, 3] 391s -0.662 0.500 0.892 0.738 391s 391s x1: [4, 1, 4, 3] 391s 0.872 0.555 0.290 -0.249 391s 391s y: [1, 1, 4, 3] 391s -0.224 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.476 0.107 -0.779 0.673 391s 0.962 0.851 -0.044 0.561 391s 0.011 0.655 0.026 -0.711 391s 391s x1: [4, 1, 1, 1] 391s -0.028 0.828 -0.189 0.726 391s 391s y: [3, 1, 1, 1] 391s 0.737 1.094 0.021 391s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 2, 4, 1] 391s 0.461 -0.125 -0.418 391s -0.841 0.728 -0.954 391s 391s x1: [3, 4, 4, 1] 391s -0.138 0.375 -0.040 391s 0.114 -0.697 0.129 391s -0.920 0.342 -0.418 391s -0.201 0.361 -0.341 391s 391s y: [2, 4, 4, 1] 391s -0.093 0.428 391s 0.086 -0.727 391s -0.292 1.421 391s 0.005 0.757 391s testing: mul_mat, [1, 3, 4, 3] = [2, 1, 4, 3] * [2, 3, 4, 3] 391s x0: [2, 1, 4, 3] 391s 0.637 0.318 391s 391s x1: [2, 3, 4, 3] 391s 0.836 0.330 391s 0.120 -0.965 391s 0.534 0.963 391s 391s y: [1, 3, 4, 3] 391s 0.638 391s -0.230 391s 0.647 391s test-mul-mat0: iter:236/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.731 0.109 391s 391s x1: [2, 4, 1, 1] 391s -0.130 -0.181 391s -0.578 -0.493 391s -0.864 0.999 391s 0.817 -0.949 391s 391s y: [1, 4, 1, 1] 391s 0.075 391s 0.369 391s 0.740 391s -0.701 391s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.160 -0.340 391s 391s x1: [2, 1, 2, 1] 391s 0.187 -0.946 391s 391s y: [1, 1, 2, 1] 391s 0.292 391s testing: mul_mat, [2, 1, 2, 1] = [2, 2, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.650 0.025 391s 0.176 0.326 391s 391s x1: [2, 1, 2, 1] 391s 0.277 0.530 391s 391s y: [2, 1, 2, 1] 391s -0.167 0.221 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.444 -0.447 391s 0.193 -0.620 391s -0.211 -0.156 391s 391s x1: [2, 2, 1, 1] 391s -0.456 -0.297 391s 0.393 -0.457 391s 391s y: [3, 2, 1, 1] 391s -0.070 0.096 0.142 391s 0.379 0.359 -0.011 391s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.413 -0.876 0.847 391s 0.725 0.851 -0.681 391s 391s x1: [3, 2, 2, 1] 391s 0.337 -0.165 0.219 391s -0.823 0.495 0.039 391s 391s y: [2, 2, 2, 1] 391s 0.470 -0.045 391s -0.741 -0.202 391s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.609 -0.997 391s 391s x1: [2, 3, 2, 1] 391s 0.946 -0.872 391s -0.668 -0.501 391s -0.493 0.176 391s 391s y: [1, 3, 2, 1] 391s 1.446 391s 0.092 391s -0.475 391s test-mul-mat0: iter:237/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.317 -0.914 0.998 0.153 391s 0.428 -0.607 0.167 -0.734 391s 0.827 0.402 0.875 0.830 391s 391s x1: [4, 3, 1, 1] 391s 0.605 0.262 -0.876 0.924 391s -0.630 -0.484 -0.053 -0.128 391s 0.451 -0.510 -0.183 0.579 391s 391s y: [3, 3, 1, 1] 391s -0.780 -0.725 0.607 391s 0.170 0.109 -0.868 391s 0.516 0.048 0.488 391s testing: mul_mat, [2, 3, 3, 1] = [4, 2, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 2, 3, 1] 391s -0.537 0.681 0.224 0.558 391s -0.404 -0.476 0.600 -0.618 391s 391s x1: [4, 3, 3, 1] 391s -0.915 0.872 -0.618 0.566 391s 0.282 -0.341 0.171 -0.456 391s -0.217 0.095 -0.086 0.300 391s 391s y: [2, 3, 3, 1] 391s 1.263 -0.766 391s -0.600 0.433 391s 0.329 -0.194 391s testing: mul_mat, [2, 2, 3, 4] = [4, 2, 3, 4] * [4, 2, 3, 4] 391s x0: [4, 2, 3, 4] 391s -0.919 -0.167 -0.954 0.800 391s -0.370 0.253 0.773 0.238 391s 391s x1: [4, 2, 3, 4] 391s 0.552 0.586 0.809 0.335 391s -0.122 -0.054 0.776 0.341 391s 391s y: [2, 2, 3, 4] 391s -1.109 0.649 391s -0.346 0.712 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.408 0.781 -0.174 0.695 391s 0.340 -0.206 0.504 0.580 391s -0.939 0.730 0.747 -0.831 391s -0.777 -0.907 -0.340 0.143 391s 391s x1: [4, 2, 1, 1] 391s 0.761 -0.339 -0.207 0.787 391s 0.331 -0.849 -0.475 0.461 391s 391s y: [4, 2, 1, 1] 391s 0.008 0.681 -1.771 -0.101 391s -0.395 0.316 -1.668 0.741 391s testing: mul_mat, [2, 4, 3, 1] = [4, 2, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 2, 3, 1] 391s -0.286 -0.919 -0.002 0.652 391s 0.287 0.431 -0.051 -0.545 391s 391s x1: [4, 4, 3, 1] 391s -0.849 0.853 -0.122 -0.902 391s 0.045 0.978 0.859 0.705 391s -0.230 0.646 0.036 -0.079 391s -0.829 -0.503 0.186 -0.236 391s 391s y: [2, 4, 3, 1] 391s -1.128 0.621 391s -0.453 0.006 391s -0.579 0.253 391s 0.545 -0.335 391s testing: mul_mat, [4, 4, 3, 4] = [2, 4, 3, 4] * [2, 4, 3, 4] 391s x0: [2, 4, 3, 4] 391s 0.088 -0.927 391s 0.743 0.863 391s -0.511 -0.694 391s 0.258 -0.995 391s 391s x1: [2, 4, 3, 4] 391s -0.224 -0.129 391s 0.783 0.569 391s -0.402 0.452 391s 0.609 0.312 391s 391s y: [4, 4, 3, 4] 391s 0.100 -0.278 0.204 0.071 391s -0.459 1.073 -0.794 -0.364 391s -0.454 0.092 -0.108 -0.554 391s -0.235 0.721 -0.527 -0.153 391s test-mul-mat0: iter:238/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.419 0.435 0.938 0.909 391s -0.666 0.873 -0.125 0.975 391s -0.687 -0.842 -0.386 0.318 391s 391s x1: [4, 3, 1, 1] 391s 0.950 -0.985 -0.840 -0.443 391s 0.238 -0.649 0.492 0.568 391s -0.475 -0.255 -0.154 -0.408 391s 391s y: [3, 3, 1, 1] 391s -1.220 -1.821 0.359 391s 0.794 -0.233 0.374 391s -0.824 -0.284 0.470 391s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.197 0.049 0.383 0.781 391s -0.976 -0.497 0.857 -0.433 391s -0.853 -0.321 0.957 -0.617 391s 0.036 -0.234 -0.945 0.330 391s 391s x1: [4, 3, 3, 1] 391s 0.341 -0.006 -0.791 -0.923 391s 0.723 -0.392 -0.972 0.738 391s -0.232 -0.416 -0.024 0.119 391s 391s y: [4, 3, 3, 1] 391s -0.957 -0.609 -0.477 0.457 391s 0.326 -1.663 -1.877 1.280 391s 0.018 0.360 0.234 0.151 391s testing: mul_mat, [2, 4, 3, 2] = [4, 2, 3, 2] * [4, 4, 3, 2] 391s x0: [4, 2, 3, 2] 391s -0.698 -0.599 0.087 -0.139 391s 0.028 -0.915 -0.705 -0.007 391s 391s x1: [4, 4, 3, 2] 391s -0.071 -0.537 -0.546 -0.031 391s -0.560 0.281 0.603 -0.785 391s 0.199 0.825 0.998 -0.714 391s 0.357 -0.605 -0.431 0.938 391s 391s y: [2, 4, 3, 2] 391s 0.328 0.874 391s 0.384 -0.692 391s -0.446 -1.447 391s -0.054 0.861 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.023 -0.085 -0.349 -0.213 391s -0.785 0.933 0.987 -0.215 391s -0.924 -0.821 -0.424 0.674 391s 0.400 -0.297 0.664 -0.551 391s 391s x1: [4, 2, 1, 1] 391s -0.477 -0.623 -0.748 0.021 391s 0.741 0.587 0.471 0.500 391s 391s y: [4, 2, 1, 1] 391s 0.299 -0.949 1.284 -0.514 391s -0.304 0.322 -1.030 0.160 391s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 3, 3, 1] 391s 0.603 -0.065 -0.746 -0.640 391s -0.995 -0.373 0.824 0.807 391s 0.804 0.925 0.709 0.167 391s 391s x1: [4, 4, 3, 1] 391s -0.227 -0.174 0.386 0.715 391s 0.171 0.102 -0.762 0.548 391s 0.353 0.259 0.289 -0.060 391s -0.270 -0.211 0.461 0.753 391s 391s y: [3, 4, 3, 1] 391s -0.872 1.186 0.050 391s 0.314 -0.394 -0.217 391s 0.019 -0.258 0.719 391s -0.975 1.335 0.041 391s testing: mul_mat, [2, 4, 3, 2] = [3, 2, 3, 2] * [3, 4, 3, 2] 391s x0: [3, 2, 3, 2] 391s -0.736 -0.106 -0.434 391s -0.018 -0.978 -0.315 391s 391s x1: [3, 4, 3, 2] 391s -0.511 0.648 0.830 391s -0.802 -0.992 0.636 391s 0.366 -0.078 -0.802 391s -0.182 0.777 0.231 391s 391s y: [2, 4, 3, 2] 391s -0.053 -0.886 391s 0.418 0.785 391s 0.088 0.322 391s -0.048 -0.829 391s test-mul-mat0: iter:239/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.582 0.245 0.929 391s 0.568 0.258 -0.445 391s 0.478 -0.912 -0.896 391s 391s x1: [3, 3, 1, 1] 391s -0.882 0.962 0.753 391s 0.566 0.585 0.429 391s -0.120 -0.839 -0.353 391s 391s y: [3, 3, 1, 1] 391s 0.422 -0.589 -1.974 391s 0.871 0.282 -0.647 391s -0.603 -0.127 1.025 391s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.959 0.127 0.331 391s 391s x1: [3, 3, 3, 1] 391s -0.135 0.954 -0.277 391s -0.483 -0.198 -0.763 391s -0.808 -0.784 -0.564 391s 391s y: [1, 3, 3, 1] 391s -0.099 391s -0.740 391s -1.061 391s testing: mul_mat, [2, 1, 3, 3] = [3, 2, 3, 3] * [3, 1, 3, 3] 391s x0: [3, 2, 3, 3] 391s 0.328 0.236 0.522 391s -0.139 0.336 -0.749 391s 391s x1: [3, 1, 3, 3] 391s 0.533 -0.900 -0.272 391s 391s y: [2, 1, 3, 3] 391s -0.180 -0.173 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.532 -0.515 0.050 391s -0.436 -0.140 -0.659 391s 0.730 0.531 -0.212 391s -0.594 0.125 0.608 391s 391s x1: [3, 2, 1, 1] 391s -0.827 -0.941 0.773 391s 0.378 -0.820 0.504 391s 391s y: [4, 2, 1, 1] 391s 0.083 -0.016 -1.267 0.843 391s 0.649 -0.382 -0.266 -0.021 391s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.657 0.675 0.388 0.563 391s 0.226 0.068 -0.098 0.089 391s -0.480 -0.823 -0.130 0.601 391s -0.169 -0.818 0.213 0.953 391s 391s x1: [4, 3, 3, 1] 391s -0.267 0.786 0.014 -0.510 391s 0.319 0.077 0.936 0.334 391s -0.479 0.174 -0.552 -0.452 391s 391s y: [4, 3, 3, 1] 391s 0.073 -0.054 -0.826 -1.080 391s 0.813 0.016 -0.138 0.400 391s -0.666 -0.083 -0.113 -0.610 391s testing: mul_mat, [3, 4, 3, 3] = [4, 3, 3, 3] * [4, 4, 3, 3] 391s x0: [4, 3, 3, 3] 391s -0.137 -0.440 -0.840 -0.867 391s -0.347 0.028 -0.297 -0.395 391s -0.743 -0.083 0.506 -0.250 391s 391s x1: [4, 4, 3, 3] 391s -0.403 0.608 -0.645 0.071 391s 0.111 -0.276 -0.668 -0.477 391s -0.676 -0.834 -0.144 0.645 391s 0.892 0.189 -0.378 -0.155 391s 391s y: [3, 4, 3, 3] 391s 0.268 0.321 -0.095 391s 1.082 0.341 -0.279 391s 0.022 -0.001 0.338 391s 0.247 -0.131 -0.831 391s test-mul-mat0: iter:240/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.471 0.535 0.662 0.201 391s -0.532 0.064 -0.958 -0.787 391s 391s x1: [4, 1, 1, 1] 391s -0.141 0.244 0.610 0.887 391s 391s y: [2, 1, 1, 1] 391s 0.779 -1.192 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.000 -0.463 0.679 0.162 391s 0.056 -0.646 -0.233 -0.806 391s 391s x1: [4, 2, 1, 1] 391s -0.249 0.728 -0.145 -0.472 391s 0.849 -0.359 -0.093 0.503 391s 391s y: [2, 2, 1, 1] 391s -0.512 -0.070 391s 0.185 -0.104 391s testing: mul_mat, [4, 2, 1, 4] = [4, 4, 1, 4] * [4, 2, 1, 4] 391s x0: [4, 4, 1, 4] 391s -0.366 -0.289 -0.818 0.836 391s 0.796 0.230 -0.300 -0.918 391s -0.254 -0.372 -0.589 -0.526 391s 0.307 -0.533 0.058 -0.485 391s 391s x1: [4, 2, 1, 4] 391s -0.053 0.626 0.439 -0.443 391s 0.513 0.202 0.086 0.048 391s 391s y: [4, 2, 1, 4] 391s -0.891 0.376 -0.245 -0.110 391s -0.276 0.385 -0.281 0.031 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.901 -0.952 0.814 -0.150 391s -0.659 0.112 0.981 -0.195 391s 391s x1: [4, 4, 1, 1] 391s -0.353 -0.399 0.226 0.792 391s 0.689 0.488 0.684 0.406 391s -0.840 -0.950 -0.570 0.299 391s 0.473 0.969 -0.191 0.179 391s 391s y: [2, 4, 1, 1] 391s 0.763 0.255 391s -0.590 0.192 391s 1.153 -0.171 391s -1.530 -0.425 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.851 -0.585 391s 0.009 0.387 391s 0.227 0.882 391s 0.833 -0.155 391s 391s x1: [2, 4, 1, 1] 391s -0.021 -0.492 391s 0.468 0.710 391s -0.281 -0.486 391s -0.643 0.320 391s 391s y: [4, 4, 1, 1] 391s 0.306 -0.191 -0.439 0.058 391s -0.813 0.279 0.733 0.280 391s 0.523 -0.191 -0.493 -0.159 391s 0.360 0.118 0.136 -0.585 391s testing: mul_mat, [2, 2, 1, 4] = [4, 2, 1, 4] * [4, 2, 1, 4] 391s x0: [4, 2, 1, 4] 391s 0.187 0.779 0.607 0.136 391s -0.987 -0.863 -0.345 -0.686 391s 391s x1: [4, 2, 1, 4] 391s -0.314 0.356 -0.186 0.496 391s -0.466 -0.050 0.594 -0.125 391s 391s y: [2, 2, 1, 4] 391s 0.173 -0.273 391s 0.218 0.383 391s test-mul-mat0: iter:241/500 391s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.220 391s -0.138 391s -0.590 391s 391s x1: [1, 1, 1, 1] 391s 0.518 391s 391s y: [3, 1, 1, 1] 391s 0.114 -0.071 -0.305 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.501 391s -0.774 391s 391s x1: [1, 3, 1, 1] 391s 0.090 391s 0.700 391s -0.516 391s 391s y: [2, 3, 1, 1] 391s 0.045 -0.070 391s 0.351 -0.542 391s -0.258 0.399 391s testing: mul_mat, [3, 2, 1, 2] = [1, 3, 1, 2] * [1, 2, 1, 2] 391s x0: [1, 3, 1, 2] 391s 0.936 391s -0.502 391s -0.792 391s 391s x1: [1, 2, 1, 2] 391s 0.742 391s -0.222 391s 391s y: [3, 2, 1, 2] 391s 0.695 -0.373 -0.588 391s -0.208 0.112 0.176 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.182 391s 0.560 391s 391s x1: [1, 3, 1, 1] 391s 0.723 391s 0.796 391s -0.739 391s 391s y: [2, 3, 1, 1] 391s -0.131 0.404 391s -0.145 0.445 391s 0.134 -0.414 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.187 0.686 391s -0.728 -0.951 391s 391s x1: [2, 1, 1, 1] 391s 0.669 0.977 391s 391s y: [2, 1, 1, 1] 391s 0.795 -1.417 391s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 1, 1, 2] 391s 0.387 -0.894 391s 391s x1: [2, 2, 1, 2] 391s -0.318 -0.225 391s 0.749 0.166 391s 391s y: [1, 2, 1, 2] 391s 0.078 391s 0.141 391s test-mul-mat0: iter:242/500 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.258 391s 0.735 391s 0.179 391s 391s x1: [1, 2, 1, 1] 391s -0.519 391s -0.071 391s 391s y: [3, 2, 1, 1] 391s 0.134 -0.381 -0.093 391s 0.018 -0.052 -0.013 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.408 391s 0.428 391s -0.547 391s 391s x1: [1, 3, 1, 1] 391s -0.698 391s 0.405 391s 0.156 391s 391s y: [3, 3, 1, 1] 391s 0.285 -0.299 0.382 391s -0.165 0.173 -0.221 391s -0.064 0.067 -0.085 391s testing: mul_mat, [2, 3, 1, 4] = [1, 2, 1, 4] * [1, 3, 1, 4] 391s x0: [1, 2, 1, 4] 391s 0.378 391s 0.282 391s 391s x1: [1, 3, 1, 4] 391s -0.359 391s -0.890 391s 0.229 391s 391s y: [2, 3, 1, 4] 391s -0.136 -0.101 391s -0.336 -0.251 391s 0.086 0.064 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.998 391s -0.347 391s -0.458 391s -0.234 391s 391s x1: [1, 2, 1, 1] 391s 0.696 391s 0.248 391s 391s y: [4, 2, 1, 1] 391s 0.695 -0.242 -0.319 -0.163 391s 0.248 -0.086 -0.114 -0.058 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.552 0.356 0.916 -0.952 391s -0.724 0.477 -0.975 0.585 391s 391s x1: [4, 1, 1, 1] 391s 0.245 0.970 0.220 0.886 391s 391s y: [2, 1, 1, 1] 391s -0.432 0.589 391s testing: mul_mat, [4, 4, 1, 4] = [2, 4, 1, 4] * [2, 4, 1, 4] 391s x0: [2, 4, 1, 4] 391s -0.085 0.001 391s -0.041 -0.782 391s -0.636 0.916 391s 0.939 -0.444 391s 391s x1: [2, 4, 1, 4] 391s -0.919 -0.275 391s -0.449 0.458 391s -0.993 0.024 391s -0.958 0.854 391s 391s y: [4, 4, 1, 4] 391s 0.078 0.253 0.333 -0.741 391s 0.038 -0.340 0.705 -0.625 391s 0.084 0.022 0.653 -0.943 391s 0.082 -0.628 1.392 -1.279 391s test-mul-mat0: iter:243/500 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.337 0.436 -0.765 0.011 391s -0.205 -0.660 -0.277 -0.739 391s 391s x1: [4, 2, 1, 1] 391s 0.619 -0.118 -0.456 0.897 391s -0.527 0.903 0.623 0.735 391s 391s y: [2, 2, 1, 1] 391s 0.516 -0.586 391s -0.252 -1.204 391s testing: mul_mat, [4, 2, 2, 1] = [4, 4, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 4, 2, 1] 391s 0.945 -0.486 0.808 0.568 391s -0.751 -0.473 -0.095 0.685 391s -0.238 0.916 -0.519 0.102 391s -0.360 -0.258 -0.591 -0.530 391s 391s x1: [4, 2, 2, 1] 391s 0.307 0.831 0.310 0.054 391s -0.642 0.990 0.166 -0.613 391s 391s y: [4, 2, 2, 1] 391s 0.168 -0.616 0.533 -0.537 391s -1.302 -0.423 0.912 0.202 391s testing: mul_mat, [3, 4, 2, 1] = [4, 3, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.545 -0.292 0.293 -0.580 391s 0.179 -0.752 -0.698 -0.869 391s -0.695 -0.177 0.824 -0.230 391s 391s x1: [4, 4, 2, 1] 391s -0.463 0.666 0.060 -0.214 391s -0.807 0.965 -0.529 -0.045 391s 0.882 -0.048 -0.943 -0.479 391s 0.694 -0.534 -0.008 0.745 391s 391s y: [3, 4, 2, 1] 391s 0.200 -0.439 0.303 391s 0.029 -0.461 -0.035 391s -0.465 1.269 -1.272 391s -0.657 -0.116 -0.566 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.046 0.869 0.683 0.318 391s 0.324 0.148 -0.282 -0.925 391s -0.746 0.025 0.724 -0.018 391s 391s x1: [4, 3, 1, 1] 391s 0.874 -0.526 0.386 0.308 391s 0.869 0.308 -0.591 -0.676 391s -0.985 0.702 -0.256 0.194 391s 391s y: [3, 3, 1, 1] 391s -0.055 -0.189 -0.391 391s -0.311 1.120 -1.056 391s 0.451 -0.323 0.563 391s testing: mul_mat, [1, 4, 2, 1] = [3, 1, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 1, 2, 1] 391s -0.130 -0.286 -0.469 391s 391s x1: [3, 4, 2, 1] 391s 0.988 -0.711 0.620 391s -0.715 -0.140 0.621 391s -0.841 0.335 0.007 391s 0.467 0.204 -0.686 391s 391s y: [1, 4, 2, 1] 391s -0.216 391s -0.159 391s 0.010 391s 0.203 391s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 1, 2, 1] 391s 0.273 391s 391s x1: [1, 3, 2, 1] 391s -0.824 391s 0.078 391s 0.132 391s 391s y: [1, 3, 2, 1] 391s -0.225 391s 0.021 391s 0.036 391s test-mul-mat0: iter:244/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.216 -0.229 391s 0.655 0.655 391s 0.486 -0.814 391s 0.842 -0.725 391s 391s x1: [2, 4, 1, 1] 391s -0.994 -0.596 391s 0.540 -0.721 391s 0.927 0.067 391s 0.597 -0.225 391s 391s y: [4, 4, 1, 1] 391s 0.351 -1.042 0.003 -0.405 391s 0.049 -0.119 0.850 0.978 391s -0.215 0.651 0.396 0.732 391s -0.077 0.244 0.473 0.666 391s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 3, 4, 1] 391s -0.592 -0.027 391s 0.704 0.699 391s -0.443 -0.197 391s 391s x1: [2, 4, 4, 1] 391s 0.698 -0.981 391s 0.353 -0.170 391s -0.520 0.292 391s 0.584 0.100 391s 391s y: [3, 4, 4, 1] 391s -0.387 -0.195 -0.116 391s -0.204 0.130 -0.123 391s 0.300 -0.162 0.173 391s -0.348 0.480 -0.278 391s testing: mul_mat, [2, 3, 4, 3] = [2, 2, 4, 3] * [2, 3, 4, 3] 391s x0: [2, 2, 4, 3] 391s -0.317 0.834 391s -0.398 0.395 391s 391s x1: [2, 3, 4, 3] 391s 0.675 0.486 391s -0.113 -0.120 391s -0.459 0.030 391s 391s y: [2, 3, 4, 3] 391s 0.191 -0.077 391s -0.064 -0.003 391s 0.170 0.195 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.749 0.405 391s 0.023 0.042 391s 0.702 0.232 391s 0.275 0.371 391s 391s x1: [2, 2, 1, 1] 391s -0.735 0.927 391s -0.035 -0.413 391s 391s y: [4, 2, 1, 1] 391s -0.176 0.022 -0.302 0.142 391s -0.193 -0.018 -0.120 -0.163 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.952 -0.825 -0.289 0.607 391s 391s x1: [4, 2, 4, 1] 391s -0.473 0.872 0.171 -0.387 391s 0.907 0.655 0.979 0.827 391s 391s y: [1, 2, 4, 1] 391s -1.454 391s 0.543 391s testing: mul_mat, [1, 4, 4, 3] = [1, 1, 4, 3] * [1, 4, 4, 3] 391s x0: [1, 1, 4, 3] 391s 0.964 391s 391s x1: [1, 4, 4, 3] 391s -0.647 391s 0.061 391s -0.530 391s 0.335 391s 391s y: [1, 4, 4, 3] 391s -0.624 391s 0.058 391s -0.511 391s 0.323 391s test-mul-mat0: iter:245/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.187 0.707 0.380 391s -0.827 -0.217 -0.228 391s 391s x1: [3, 1, 1, 1] 391s -0.594 0.832 -0.841 391s 391s y: [2, 1, 1, 1] 391s 0.380 0.502 391s testing: mul_mat, [1, 2, 3, 1] = [3, 1, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.333 0.388 0.922 391s 391s x1: [3, 2, 3, 1] 391s 0.476 -0.808 0.647 391s -0.105 -0.355 0.611 391s 391s y: [1, 2, 3, 1] 391s 0.125 391s 0.461 391s testing: mul_mat, [3, 1, 3, 4] = [3, 3, 3, 4] * [3, 1, 3, 4] 391s x0: [3, 3, 3, 4] 391s 0.238 -0.473 0.956 391s 0.678 -0.919 -0.513 391s -0.312 0.509 -0.524 391s 391s x1: [3, 1, 3, 4] 391s 0.485 -0.962 0.592 391s 391s y: [3, 1, 3, 4] 391s 1.137 0.910 -0.951 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.823 -0.785 -0.701 391s 391s x1: [3, 3, 1, 1] 391s 0.551 0.712 -0.831 391s 0.917 0.622 0.786 391s 0.323 -0.581 0.732 391s 391s y: [1, 3, 1, 1] 391s -0.431 391s -1.795 391s -0.322 391s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 2, 3, 1] 391s 0.875 391s 0.572 391s 391s x1: [1, 3, 3, 1] 391s -0.649 391s 0.098 391s -0.699 391s 391s y: [2, 3, 3, 1] 391s -0.568 -0.371 391s 0.086 0.056 391s -0.612 -0.399 391s testing: mul_mat, [1, 1, 3, 4] = [2, 1, 3, 4] * [2, 1, 3, 4] 391s x0: [2, 1, 3, 4] 391s 0.170 0.049 391s 391s x1: [2, 1, 3, 4] 391s 0.800 0.819 391s 391s y: [1, 1, 3, 4] 391s 0.176 391s test-mul-mat0: iter:246/500 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.621 391s -0.652 391s 391s x1: [1, 3, 1, 1] 391s 0.285 391s 0.330 391s 0.451 391s 391s y: [2, 3, 1, 1] 391s -0.177 -0.186 391s -0.205 -0.215 391s -0.280 -0.294 391s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.672 391s -0.937 391s 391s x1: [1, 2, 3, 1] 391s -0.614 391s 0.154 391s 391s y: [2, 2, 3, 1] 391s 0.412 0.575 391s -0.103 -0.144 391s testing: mul_mat, [2, 2, 3, 4] = [1, 2, 3, 4] * [1, 2, 3, 4] 391s x0: [1, 2, 3, 4] 391s 0.330 391s 0.904 391s 391s x1: [1, 2, 3, 4] 391s -0.456 391s 0.511 391s 391s y: [2, 2, 3, 4] 391s -0.150 -0.412 391s 0.169 0.462 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.173 391s 0.843 391s 0.907 391s 0.292 391s 391s x1: [1, 2, 1, 1] 391s -0.585 391s -0.159 391s 391s y: [4, 2, 1, 1] 391s -0.101 -0.493 -0.531 -0.171 391s -0.027 -0.134 -0.144 -0.046 391s testing: mul_mat, [4, 1, 3, 1] = [4, 4, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.187 -0.390 -0.347 -0.508 391s 0.985 -0.122 0.651 0.031 391s 0.357 -0.937 -0.035 -0.759 391s -0.423 -0.180 0.068 -0.665 391s 391s x1: [4, 1, 3, 1] 391s 0.173 0.811 0.152 0.187 391s 391s y: [4, 1, 3, 1] 391s -0.497 0.176 -0.845 -0.333 391s testing: mul_mat, [1, 4, 3, 4] = [4, 1, 3, 4] * [4, 4, 3, 4] 391s x0: [4, 1, 3, 4] 391s 0.610 -0.210 0.514 0.247 391s 391s x1: [4, 4, 3, 4] 391s -0.770 0.694 0.510 -0.086 391s -0.611 0.714 -0.718 -0.172 391s -0.473 0.199 0.695 -0.218 391s 0.612 -0.804 0.849 -0.990 391s 391s y: [1, 4, 3, 4] 391s -0.375 391s -0.935 391s -0.027 391s 0.735 391s test-mul-mat0: iter:247/500 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.091 0.296 -0.127 -0.622 391s 0.650 -0.694 -0.985 -0.035 391s -0.183 0.070 -0.166 0.235 391s 391s x1: [4, 2, 1, 1] 391s 0.283 -0.170 -0.326 0.297 391s -0.308 -0.606 -0.046 -0.811 391s 391s y: [3, 2, 1, 1] 391s -0.220 0.613 0.060 391s 0.359 0.295 -0.169 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.846 0.248 0.404 -0.200 391s 0.437 0.642 0.709 -0.267 391s -0.486 -0.913 -0.616 -0.180 391s 391s x1: [4, 3, 1, 1] 391s 0.427 -0.841 0.864 0.953 391s -0.377 0.919 -0.750 0.267 391s 0.726 -0.664 -0.451 -0.444 391s 391s y: [3, 3, 1, 1] 391s 0.311 0.006 -0.143 391s -0.447 -0.177 -0.242 391s 0.356 -0.310 0.612 391s testing: mul_mat, [1, 3, 1, 2] = [4, 1, 1, 2] * [4, 3, 1, 2] 391s x0: [4, 1, 1, 2] 391s 0.136 -0.645 0.708 0.223 391s 391s x1: [4, 3, 1, 2] 391s -0.898 0.349 0.637 0.172 391s -0.817 -0.128 -0.401 -0.658 391s -0.264 -0.448 -0.035 -0.345 391s 391s y: [1, 3, 1, 2] 391s 0.142 391s -0.460 391s 0.151 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.388 -0.571 0.588 0.216 391s 0.774 -0.550 -0.411 0.931 391s 0.648 0.808 0.747 0.288 391s 0.577 -0.190 0.737 -0.356 391s 391s x1: [4, 1, 1, 1] 391s -0.834 -0.503 -0.730 0.038 391s 391s y: [4, 1, 1, 1] 391s 0.190 -0.034 -1.481 -0.937 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.569 0.191 -0.875 -0.981 391s 391s x1: [4, 4, 1, 1] 391s 0.553 -0.518 0.780 0.908 391s -0.810 0.003 -0.354 -0.282 391s -0.672 -0.266 -0.116 -0.175 391s 0.003 0.921 0.921 0.615 391s 391s y: [1, 4, 1, 1] 391s -1.358 391s 0.126 391s -0.160 391s -1.231 391s testing: mul_mat, [2, 4, 1, 2] = [1, 2, 1, 2] * [1, 4, 1, 2] 391s x0: [1, 2, 1, 2] 391s 0.448 391s -0.834 391s 391s x1: [1, 4, 1, 2] 391s -0.000 391s -0.065 391s 0.607 391s 0.588 391s 391s y: [2, 4, 1, 2] 391s -0.000 0.000 391s -0.029 0.054 391s 0.272 -0.506 391s 0.264 -0.491 391s test-mul-mat0: iter:248/500 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.543 -0.173 0.305 -0.928 391s 391s x1: [4, 1, 1, 1] 391s -0.979 -0.441 0.454 -0.095 391s 391s y: [1, 1, 1, 1] 391s 0.834 391s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.332 0.116 -0.108 0.563 391s 391s x1: [4, 1, 4, 1] 391s -0.478 -0.126 0.263 -0.353 391s 391s y: [1, 1, 4, 1] 391s -0.400 391s testing: mul_mat, [3, 1, 4, 3] = [4, 3, 4, 3] * [4, 1, 4, 3] 391s x0: [4, 3, 4, 3] 391s -0.040 -0.370 -0.391 0.933 391s -0.549 0.220 -0.588 -0.981 391s -0.523 0.703 0.406 -0.099 391s 391s x1: [4, 1, 4, 3] 391s 0.082 0.814 -0.422 0.345 391s 391s y: [3, 1, 4, 3] 391s 0.183 0.043 0.324 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.348 0.512 0.849 0.369 391s 0.488 0.315 0.876 0.447 391s 0.513 -0.676 -0.660 -0.010 391s 0.247 -0.185 -0.282 -0.482 391s 391s x1: [4, 3, 1, 1] 391s -0.164 0.901 -0.453 -0.243 391s 0.900 0.278 -0.579 0.719 391s 0.732 -0.269 -0.117 0.982 391s 391s y: [4, 3, 1, 1] 391s 0.045 -0.302 -0.392 0.037 391s -0.397 0.341 0.649 -0.012 391s -0.130 0.609 0.625 -0.209 391s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.061 0.129 0.732 0.408 391s 0.358 -0.975 -0.707 0.575 391s 391s x1: [4, 4, 4, 1] 391s -0.223 0.910 0.354 -0.322 391s -0.542 -0.889 -0.422 0.736 391s -0.468 -0.703 0.468 0.263 391s 0.180 0.449 0.886 0.833 391s 391s y: [2, 4, 4, 1] 391s 0.259 -1.402 391s -0.091 1.393 391s 0.388 0.338 391s 1.036 -0.521 391s testing: mul_mat, [4, 4, 4, 3] = [2, 4, 4, 3] * [2, 4, 4, 3] 391s x0: [2, 4, 4, 3] 391s -0.428 -0.481 391s 0.372 -0.774 391s 0.443 0.901 391s -0.032 0.776 391s 391s x1: [2, 4, 4, 3] 391s 0.040 0.295 391s 0.269 -0.227 391s 0.589 -0.323 391s -0.652 -0.690 391s 391s y: [4, 4, 4, 3] 391s -0.159 -0.214 0.284 0.228 391s -0.006 0.276 -0.086 -0.185 391s -0.097 0.469 -0.030 -0.270 391s 0.612 0.292 -0.911 -0.515 391s test-mul-mat0: iter:249/500 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.266 -0.944 391s -0.127 -0.650 391s -0.494 -0.106 391s 391s x1: [2, 4, 1, 1] 391s 0.510 0.336 391s 0.726 -0.584 391s -0.168 -0.244 391s 0.290 0.705 391s 391s y: [3, 4, 1, 1] 391s -0.453 -0.283 -0.288 391s 0.358 0.287 -0.297 391s 0.275 0.180 0.109 391s -0.743 -0.495 -0.218 391s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.651 -0.405 391s 0.042 0.638 391s 391s x1: [2, 3, 3, 1] 391s 0.769 0.535 391s 0.026 0.671 391s 0.334 0.196 391s 391s y: [2, 3, 3, 1] 391s -0.718 0.374 391s -0.288 0.429 391s -0.297 0.139 391s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.795 0.490 391s -0.479 -0.862 391s 391s x1: [2, 2, 3, 1] 391s -0.509 -0.602 391s 0.124 -0.838 391s 391s y: [2, 2, 3, 1] 391s 0.109 0.762 391s -0.509 0.663 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.481 0.766 391s 0.710 -0.395 391s 391s x1: [2, 2, 1, 1] 391s -0.503 -0.495 391s 0.384 0.219 391s 391s y: [2, 2, 1, 1] 391s -0.621 -0.162 391s 0.352 0.186 391s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.234 -0.098 391s 391s x1: [2, 2, 3, 1] 391s 0.872 0.498 391s -0.076 -0.863 391s 391s y: [1, 2, 3, 1] 391s -0.253 391s 0.102 391s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 3, 3, 1] 391s -0.228 391s -0.727 391s -0.196 391s 391s x1: [1, 2, 3, 1] 391s 0.378 391s -0.101 391s 391s y: [3, 2, 3, 1] 391s -0.086 -0.275 -0.074 391s 0.023 0.073 0.020 391s test-mul-mat0: iter:250/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.639 -0.865 391s 0.124 0.260 391s 0.543 -0.001 391s -0.336 -0.665 391s 391s x1: [2, 4, 1, 1] 391s 0.058 -0.046 391s 0.727 -0.292 391s 0.527 -0.507 391s 0.610 -0.093 391s 391s y: [4, 4, 1, 1] 391s 0.002 -0.005 0.032 0.011 391s -0.212 0.015 0.395 -0.050 391s 0.101 -0.066 0.287 0.160 391s -0.309 0.052 0.332 -0.143 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.614 -0.313 391s 391s x1: [2, 4, 1, 1] 391s -0.907 0.436 391s -0.392 -0.103 391s 0.974 0.486 391s 0.573 -0.990 391s 391s y: [1, 4, 1, 1] 391s 0.420 391s 0.273 391s -0.750 391s -0.041 391s testing: mul_mat, [4, 1, 1, 3] = [2, 4, 1, 3] * [2, 1, 1, 3] 391s x0: [2, 4, 1, 3] 391s 0.771 -0.993 391s 0.564 -0.323 391s 0.753 0.925 391s -0.188 -0.123 391s 391s x1: [2, 1, 1, 3] 391s -0.813 0.289 391s 391s y: [4, 1, 1, 3] 391s -0.914 -0.552 -0.344 0.117 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.024 -0.193 391s 391s x1: [2, 4, 1, 1] 391s -0.739 -0.885 391s 0.741 0.507 391s -0.641 0.255 391s 0.460 -0.871 391s 391s y: [1, 4, 1, 1] 391s 0.153 391s -0.080 391s -0.065 391s 0.179 391s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.891 391s 391s x1: [1, 2, 1, 1] 391s 0.014 391s -0.051 391s 391s y: [1, 2, 1, 1] 391s 0.013 391s -0.045 391s testing: mul_mat, [2, 1, 1, 3] = [1, 2, 1, 3] * [1, 1, 1, 3] 391s x0: [1, 2, 1, 3] 391s -0.334 391s 0.737 391s 391s x1: [1, 1, 1, 3] 391s -0.866 391s 391s y: [2, 1, 1, 3] 391s 0.289 -0.639 391s test-mul-mat0: iter:251/500 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.022 -0.973 -0.103 391s -0.087 -0.839 -0.128 391s 391s x1: [3, 4, 1, 1] 391s 0.868 0.751 0.087 391s 0.983 0.492 -0.406 391s -0.658 -0.253 -0.946 391s -0.528 -0.992 0.078 391s 391s y: [2, 4, 1, 1] 391s -0.720 -0.716 391s -0.415 -0.446 391s 0.329 0.390 391s 0.945 0.868 391s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.723 -0.184 -0.270 391s 0.731 0.895 -0.992 391s -0.247 0.922 -0.095 391s 0.666 -0.917 0.778 391s 391s x1: [3, 2, 3, 1] 391s -0.319 -0.856 0.538 391s -0.582 0.411 0.504 391s 391s y: [4, 2, 3, 1] 391s -0.219 -1.533 -0.761 0.992 391s -0.632 -0.557 0.475 -0.373 391s testing: mul_mat, [1, 4, 3, 2] = [3, 1, 3, 2] * [3, 4, 3, 2] 391s x0: [3, 1, 3, 2] 391s -0.655 0.030 0.197 391s 391s x1: [3, 4, 3, 2] 391s -0.304 0.972 0.720 391s 0.602 0.638 0.803 391s 0.379 0.985 -0.971 391s -0.305 -0.249 0.666 391s 391s y: [1, 4, 3, 2] 391s 0.370 391s -0.217 391s -0.410 391s 0.323 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.962 0.391 0.096 391s 391s x1: [3, 1, 1, 1] 391s -0.336 0.488 0.861 391s 391s y: [1, 1, 1, 1] 391s -0.049 391s testing: mul_mat, [1, 3, 3, 1] = [1, 1, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 1, 3, 1] 391s -0.202 391s 391s x1: [1, 3, 3, 1] 391s 0.819 391s 0.119 391s 0.434 391s 391s y: [1, 3, 3, 1] 391s -0.165 391s -0.024 391s -0.088 391s testing: mul_mat, [2, 1, 3, 2] = [1, 2, 3, 2] * [1, 1, 3, 2] 391s x0: [1, 2, 3, 2] 391s 0.306 391s 0.343 391s 391s x1: [1, 1, 3, 2] 391s -0.431 391s 391s y: [2, 1, 3, 2] 391s -0.132 -0.148 391s test-mul-mat0: iter:252/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.437 -0.424 0.097 0.906 391s 391s x1: [4, 3, 1, 1] 391s -0.362 0.095 0.339 0.767 391s 0.189 -0.863 0.181 -0.189 391s -0.293 -0.857 -0.834 0.862 391s 391s y: [1, 3, 1, 1] 391s 0.846 391s 0.129 391s 1.192 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.910 0.439 0.578 0.363 391s 391s x1: [4, 1, 1, 1] 391s 0.884 0.553 0.914 -0.320 391s 391s y: [1, 1, 1, 1] 391s -0.149 391s testing: mul_mat, [3, 1, 1, 4] = [4, 3, 1, 4] * [4, 1, 1, 4] 391s x0: [4, 3, 1, 4] 391s 0.590 0.505 0.595 -0.313 391s 0.411 0.479 -0.760 0.325 391s -0.840 -0.890 0.415 0.598 391s 391s x1: [4, 1, 1, 4] 391s -0.984 -0.595 -0.931 0.681 391s 391s y: [3, 1, 1, 4] 391s -1.648 0.240 1.377 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.165 0.682 -0.383 0.850 391s -0.143 -0.233 -0.648 -0.384 391s 391s x1: [4, 3, 1, 1] 391s 0.750 -0.042 -0.412 0.821 391s -0.662 -0.384 0.858 -0.831 391s -0.699 -0.330 0.157 0.784 391s 391s y: [2, 3, 1, 1] 391s 0.951 -0.146 391s -1.405 -0.053 391s 0.265 -0.226 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.023 0.373 391s -0.864 0.564 391s 391s x1: [2, 4, 1, 1] 391s -0.654 0.091 391s -0.274 0.756 391s 0.931 -0.476 391s 0.943 -0.614 391s 391s y: [2, 4, 1, 1] 391s 0.049 0.616 391s 0.288 0.663 391s -0.199 -1.073 391s -0.251 -1.161 391s testing: mul_mat, [4, 2, 1, 4] = [2, 4, 1, 4] * [2, 2, 1, 4] 391s x0: [2, 4, 1, 4] 391s -0.645 0.045 391s 0.290 0.842 391s 0.911 0.604 391s 0.081 -0.012 391s 391s x1: [2, 2, 1, 4] 391s 0.958 0.711 391s -0.819 0.815 391s 391s y: [4, 2, 1, 4] 391s -0.586 0.876 1.302 0.069 391s 0.566 0.449 -0.254 -0.076 391s test-mul-mat0: iter:253/500 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.994 391s -0.625 391s 391s x1: [1, 2, 1, 1] 391s 0.765 391s -0.644 391s 391s y: [2, 2, 1, 1] 391s 0.760 -0.478 391s -0.640 0.403 391s testing: mul_mat, [2, 2, 2, 1] = [1, 2, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 2, 2, 1] 391s -0.131 391s 0.594 391s 391s x1: [1, 2, 2, 1] 391s -0.645 391s -0.641 391s 391s y: [2, 2, 2, 1] 391s 0.085 -0.383 391s 0.084 -0.381 391s testing: mul_mat, [4, 2, 2, 3] = [1, 4, 2, 3] * [1, 2, 2, 3] 391s x0: [1, 4, 2, 3] 391s -0.479 391s 0.222 391s 0.559 391s 0.923 391s 391s x1: [1, 2, 2, 3] 391s 0.786 391s 0.443 391s 391s y: [4, 2, 2, 3] 391s -0.376 0.174 0.439 0.725 391s -0.212 0.098 0.247 0.408 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.327 391s 0.938 391s 391s x1: [1, 4, 1, 1] 391s 0.994 391s -0.645 391s -0.972 391s 0.131 391s 391s y: [2, 4, 1, 1] 391s 0.326 0.933 391s -0.211 -0.605 391s -0.318 -0.911 391s 0.043 0.123 391s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.412 0.276 391s 391s x1: [2, 1, 2, 1] 391s 0.041 -0.451 391s 391s y: [1, 1, 2, 1] 391s -0.108 391s testing: mul_mat, [4, 2, 2, 3] = [1, 4, 2, 3] * [1, 2, 2, 3] 391s x0: [1, 4, 2, 3] 391s -0.743 391s -0.153 391s -0.256 391s -0.388 391s 391s x1: [1, 2, 2, 3] 391s 0.944 391s 0.856 391s 391s y: [4, 2, 2, 3] 391s -0.702 -0.144 -0.241 -0.367 391s -0.637 -0.131 -0.219 -0.333 391s test-mul-mat0: iter:254/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.274 0.566 0.123 391s 391s x1: [3, 1, 1, 1] 391s -0.745 -0.573 -0.178 391s 391s y: [1, 1, 1, 1] 391s -0.551 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.254 -0.573 0.105 391s -0.435 0.562 0.394 391s 391s x1: [3, 1, 1, 1] 391s -0.850 0.441 -0.745 391s 391s y: [2, 1, 1, 1] 391s -0.548 0.325 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.183 0.400 -0.247 391s -0.321 -0.345 0.074 391s -0.249 -0.089 0.501 391s 391s x1: [3, 2, 1, 1] 391s 0.286 0.974 -0.801 391s 0.710 -0.670 -0.139 391s 391s y: [3, 2, 1, 1] 391s 0.640 -0.487 -0.560 391s -0.104 -0.007 -0.187 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.227 -0.211 -0.236 391s -0.617 0.776 0.976 391s -0.602 0.684 0.394 391s 391s x1: [3, 3, 1, 1] 391s 0.572 0.422 -0.225 391s 0.138 -0.455 -0.076 391s -0.421 -0.200 0.277 391s 391s y: [3, 3, 1, 1] 391s 0.094 -0.245 -0.144 391s 0.145 -0.512 -0.424 391s -0.119 0.375 0.226 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.803 -0.387 0.279 391s 391s x1: [3, 3, 1, 1] 391s 0.094 -0.163 0.284 391s -0.723 -0.763 -0.962 391s -0.044 -0.108 0.112 391s 391s y: [1, 3, 1, 1] 391s 0.218 391s -0.553 391s 0.038 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.688 391s -0.004 391s -0.429 391s 391s x1: [1, 3, 1, 1] 391s 0.225 391s 0.387 391s -0.583 391s 391s y: [3, 3, 1, 1] 391s -0.155 -0.001 -0.097 391s -0.267 -0.002 -0.166 391s 0.401 0.002 0.250 391s test-mul-mat0: iter:255/500 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.555 -0.439 391s 0.591 -0.277 391s -0.202 0.629 391s 391s x1: [2, 3, 1, 1] 391s 0.228 -0.413 391s 0.748 -0.088 391s 0.351 0.724 391s 391s y: [3, 3, 1, 1] 391s 0.055 0.249 -0.305 391s -0.376 0.466 -0.206 391s -0.513 0.007 0.384 391s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 3, 3, 1] 391s 0.446 -0.638 391s 0.047 -0.203 391s -0.914 0.354 391s 391s x1: [2, 3, 3, 1] 391s 0.679 0.691 391s -0.260 0.386 391s 0.494 0.353 391s 391s y: [3, 3, 3, 1] 391s -0.138 -0.108 -0.376 391s -0.3ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s 62 -0.091 0.374 391s -0.005 -0.048 -0.326 391s testing: mul_mat, [3, 3, 3, 4] = [2, 3, 3, 4] * [2, 3, 3, 4] 391s x0: [2, 3, 3, 4] 391s -0.812 0.200 391s -0.194 -0.256 391s 0.006 0.868 391s 391s x1: [2, 3, 3, 4] 391s 0.695 0.348 391s 0.407 -0.224 391s 0.838 -0.540 391s 391s y: [3, 3, 3, 4] 391s -0.495 -0.224 0.307 391s -0.376 -0.022 -0.192 391s -0.789 -0.024 -0.464 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.948 -0.815 391s -0.452 -0.978 391s 391s x1: [2, 3, 1, 1] 391s -0.667 -0.721 391s 0.299 0.690 391s 0.972 0.345 391s 391s y: [2, 3, 1, 1] 391s -0.045 1.007 391s -0.278 -0.810 391s 0.641 -0.777 391s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.638 -0.302 391s 391s x1: [2, 2, 3, 1] 391s 0.475 -0.311 391s -0.641 0.903 391s 391s y: [1, 2, 3, 1] 391s -0.209 391s 0.136 391s testing: mul_mat, [1, 2, 3, 4] = [1, 1, 3, 4] * [1, 2, 3, 4] 391s x0: [1, 1, 3, 4] 391s 0.703 391s 391s x1: [1, 2, 3, 4] 391s 0.476 391s 0.621 391s 391s y: [1, 2, 3, 4] 391s 0.335 391s 0.436 391s test-mul-mat0: iter:256/500 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.872 -0.670 391s 0.200 -0.339 391s -0.968 -0.042 391s 0.229 -0.987 391s 391s x1: [2, 1, 1, 1] 391s -0.003 -0.040 391s 391s y: [4, 1, 1, 1] 391s 0.029 0.013 0.005 0.039 391s testing: mul_mat, [2, 4, 3, 1] = [2, 2, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.609 -0.538 391s -0.664 -0.432 391s 391s x1: [2, 4, 3, 1] 391s 0.359 -0.430 391s 0.393 0.263 391s -0.728 0.471 391s 0.328 -0.760 391s 391s y: [2, 4, 3, 1] 391s 0.450 -0.052 391s 0.098 -0.375 391s -0.696 0.280 391s 0.608 0.111 391s testing: mul_mat, [2, 2, 3, 3] = [2, 2, 3, 3] * [2, 2, 3, 3] 391s x0: [2, 2, 3, 3] 391s 0.582 -0.312 391s -0.288 0.447 391s 391s x1: [2, 2, 3, 3] 391s -0.483 -0.226 391s -0.950 -0.301 391s 391s y: [2, 2, 3, 3] 391s -0.210 0.038 391s -0.459 0.139 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.187 0.423 391s 0.289 -0.504 391s 391s x1: [2, 2, 1, 1] 391s 0.697 0.911 391s 0.758 0.421 391s 391s y: [2, 2, 1, 1] 391s 0.515 -0.257 391s 0.320 0.007 391s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 1, 3, 1] 391s 0.088 -0.588 391s 391s x1: [2, 2, 3, 1] 391s 0.763 0.248 391s -0.746 -0.465 391s 391s y: [1, 2, 3, 1] 391s -0.078 391s 0.207 391s testing: mul_mat, [4, 2, 3, 3] = [1, 4, 3, 3] * [1, 2, 3, 3] 391s x0: [1, 4, 3, 3] 391s -0.121 391s 0.260 391s 0.866 391s 0.082 391s 391s x1: [1, 2, 3, 3] 391s 0.007 391s -0.866 391s 391s y: [4, 2, 3, 3] 391s -0.001 0.002 0.006 0.001 391s 0.104 -0.225 -0.750 -0.071 391s test-mul-mat0: iter:257/500 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.117 -0.976 391s -0.397 -0.493 391s 391s x1: [2, 3, 1, 1] 391s 0.465 -0.351 391s -0.325 0.334 391s -0.139 0.920 391s 391s y: [2, 3, 1, 1] 391s 0.397 -0.012 391s -0.364 -0.036 391s -0.915 -0.398 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.793 -0.650 391s -0.004 -0.322 391s 0.016 0.570 391s 391s x1: [2, 2, 1, 1] 391s -0.827 -0.287 391s -0.545 -0.136 391s 391s y: [3, 2, 1, 1] 391s 0.843 0.096 -0.177 391s 0.521 0.046 -0.087 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.443 0.429 391s 391s x1: [2, 3, 1, 1] 391s -0.972 -0.057 391s -0.989 -0.539 391s 0.986 -0.102 391s 391s y: [1, 3, 1, 1] 391s 0.406 391s 0.206 391s -0.480 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.864 0.184 391s -0.933 -0.951 391s -0.975 -0.627 391s 0.770 0.691 391s 391s x1: [2, 1, 1, 1] 391s -0.471 0.206 391s 391s y: [4, 1, 1, 1] 391s -0.369 0.243 0.330 -0.220 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.547 0.650 0.059 0.396 391s 391s x1: [4, 2, 1, 1] 391s -0.778 0.086 -0.854 0.086 391s 0.084 -0.647 0.436 -0.920 391s 391s y: [1, 2, 1, 1] 391s 0.465 391s -0.805 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.939 391s 391s x1: [1, 4, 1, 1] 391s 0.661 391s 0.520 391s 0.382 391s -0.441 391s 391s y: [1, 4, 1, 1] 391s 0.621 391s 0.488 391s 0.359 391s -0.414 391s test-mul-mat0: iter:258/500 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.767 0.477 0.290 391s -0.414 -0.087 0.370 391s 0.616 0.366 0.020 391s -0.325 -0.239 -0.319 391s 391s x1: [3, 3, 1, 1] 391s -0.023 0.212 -0.881 391s -0.253 -0.604 -0.832 391s 0.120 -0.913 -0.610 391s 391s y: [4, 3, 1, 1] 391s -0.137 -0.335 0.046 0.238 391s -0.335 -0.150 -0.393 0.492 391s -0.704 -0.196 -0.272 0.374 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.164 0.608 -0.263 391s -0.077 0.999 -0.057 391s 0.156 0.475 -0.767 391s 0.742 -0.612 0.602 391s 391s x1: [3, 4, 1, 1] 391s -0.805 -0.857 0.240 391s -0.221 -0.918 0.228 391s 0.893 0.228 0.322 391s -0.130 -0.560 0.440 391s 391s y: [4, 4, 1, 1] 391s -0.452 -0.808 -0.717 0.072 391s -0.582 -0.912 -0.646 0.535 391s -0.092 0.141 0.001 0.717 391s -0.435 -0.574 -0.624 0.511 391s testing: mul_mat, [2, 4, 1, 3] = [3, 2, 1, 3] * [3, 4, 1, 3] 391s x0: [3, 2, 1, 3] 391s -0.313 -0.071 0.369 391s 0.695 0.670 0.140 391s 391s x1: [3, 4, 1, 3] 391s 0.358 0.753 -0.378 391s -0.966 -0.485 0.303 391s -0.771 -0.342 -0.457 391s 0.008 -0.259 0.771 391s 391s y: [2, 4, 1, 3] 391s -0.305 0.701 391s 0.449 -0.954 391s 0.097 -0.829 391s 0.301 -0.060 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.727 -0.333 -0.307 391s -0.578 -0.269 0.494 391s 391s x1: [3, 2, 1, 1] 391s -0.057 -0.217 -0.338 391s -0.758 0.849 -0.335 391s 391s y: [2, 2, 1, 1] 391s 0.217 -0.075 391s 0.371 0.045 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.070 0.002 391s -0.764 0.827 391s -0.926 -0.732 391s -0.023 0.910 391s 391s x1: [2, 3, 1, 1] 391s 0.418 0.622 391s -0.137 0.113 391s 0.292 -0.997 391s 391s y: [4, 3, 1, 1] 391s -0.028 0.196 -0.842 0.556 391s 0.010 0.198 0.044 0.106 391s -0.023 -1.047 0.459 -0.914 391s testing: mul_mat, [2, 2, 1, 3] = [4, 2, 1, 3] * [4, 2, 1, 3] 391s x0: [4, 2, 1, 3] 391s 0.676 -0.701 -0.511 0.207 391s -0.779 0.503 -0.791 -0.784 391s 391s x1: [4, 2, 1, 3] 391s 0.704 0.727 -0.043 -0.353 391s -0.491 0.620 -0.111 -0.642 391s 391s y: [2, 2, 1, 3] 391s -0.085 0.128 391s -0.842 1.285 391s test-mul-mat0: iter:259/500 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.382 -0.468 391s -0.532 0.591 391s 0.739 -0.317 391s -0.243 0.593 391s 391s x1: [2, 2, 1, 1] 391s 0.353 0.758 391s 0.083 0.028 391s 391s y: [4, 2, 1, 1] 391s -0.220 0.260 0.020 0.364 391s 0.018 -0.027 0.052 -0.003 391s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.474 0.071 391s 0.739 -0.414 391s 0.277 -0.869 391s 0.174 -0.439 391s 391s x1: [2, 4, 4, 1] 391s 0.408 0.543 391s -0.664 -0.508 391s -0.386 0.112 391s -0.794 -0.608 391s 391s y: [4, 4, 4, 1] 391s 0.232 0.077 -0.359 -0.167 391s -0.350 -0.280 0.258 0.108 391s -0.175 -0.332 -0.204 -0.116 391s -0.419 -0.335 0.308 0.129 391s testing: mul_mat, [1, 4, 4, 2] = [2, 1, 4, 2] * [2, 4, 4, 2] 391s x0: [2, 1, 4, 2] 391s 0.595 0.792 391s 391s x1: [2, 4, 4, 2] 391s 0.123 0.539 391s -0.339 -0.599 391s 0.670 0.835 391s -0.038 0.372 391s 391s y: [1, 4, 4, 2] 391s 0.501 391s -0.676 391s 1.060 391s 0.272 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.621 0.222 391s -0.920 0.921 391s -0.535 0.173 391s 391s x1: [2, 1, 1, 1] 391s 0.518 -0.826 391s 391s y: [3, 1, 1, 1] 391s -0.505 -1.237 -0.420 391s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.595 -0.172 -0.322 391s 391s x1: [3, 2, 4, 1] 391s -0.900 0.496 -0.841 391s -0.864 -0.655 -0.913 391s 391s y: [1, 2, 4, 1] 391s -0.350 391s -0.107 391s testing: mul_mat, [1, 3, 4, 2] = [1, 1, 4, 2] * [1, 3, 4, 2] 391s x0: [1, 1, 4, 2] 391s 0.255 391s 391s x1: [1, 3, 4, 2] 391s -0.215 391s -0.273 391s -0.114 391s 391s y: [1, 3, 4, 2] 391s -0.055 391s -0.070 391s -0.029 391s test-mul-mat0: iter:260/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.612 391s -0.847 391s -0.950 391s 391s x1: [1, 3, 1, 1] 391s 0.506 391s 0.375 391s -0.310 391s 391s y: [3, 3, 1, 1] 391s -0.310 -0.429 -0.481 391s -0.229 -0.318 -0.356 391s 0.190 0.262 0.294 391s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.417 391s 391s x1: [1, 3, 2, 1] 391s 0.099 391s -0.811 391s -0.832 391s 391s y: [1, 3, 2, 1] 391s -0.041 391s 0.338 391s 0.347 391s testing: mul_mat, [4, 1, 2, 1] = [1, 4, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.860 391s 0.549 391s 0.239 391s -0.551 391s 391s x1: [1, 1, 2, 1] 391s -0.116 391s 391s y: [4, 1, 2, 1] 391s 0.099 -0.063 -0.028 0.064 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.068 391s 391s x1: [1, 4, 1, 1] 391s -0.941 391s 0.269 391s 0.557 391s -0.668 391s 391s y: [1, 4, 1, 1] 391s 0.064 391s -0.018 391s -0.038 391s 0.045 391s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.175 391s 0.007 391s 391s x1: [1, 1, 2, 1] 391s 0.022 391s 391s y: [2, 1, 2, 1] 391s 0.004 0.000 391s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.081 0.182 391s 391s x1: [2, 1, 2, 1] 391s 0.175 0.445 391s 391s y: [1, 1, 2, 1] 391s 0.067 391s test-mul-mat0: iter:261/500 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.007 -0.181 391s 0.056 -0.725 391s -0.624 0.388 391s 0.050 0.309 391s 391s x1: [2, 1, 1, 1] 391s 0.464 -0.846 391s 391s y: [4, 1, 1, 1] 391s 0.156 0.639 -0.618 -0.238 391s testing: mul_mat, [2, 4, 2, 1] = [2, 2, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.623 0.477 391s 0.386 0.841 391s 391s x1: [2, 4, 2, 1] 391s -0.589 -0.993 391s -0.371 0.586 391s 0.014 0.049 391s -0.050 -0.812 391s 391s y: [2, 4, 2, 1] 391s -0.107 -1.063 391s 0.510 0.350 391s 0.015 0.047 391s -0.356 -0.702 391s testing: mul_mat, [3, 2, 2, 3] = [2, 3, 2, 3] * [2, 2, 2, 3] 391s x0: [2, 3, 2, 3] 391s -0.727 -0.626 391s -0.562 -0.344 391s 0.223 -0.112 391s 391s x1: [2, 2, 2, 3] 391s 0.439 0.124 391s -0.174 -0.173 391s 391s y: [3, 2, 2, 3] 391s -0.397 -0.289 0.084 391s 0.234 0.157 -0.019 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.810 0.387 391s 391s x1: [2, 3, 1, 1] 391s -0.458 -0.272 391s -0.836 -0.743 391s 0.650 -0.315 391s 391s y: [1, 3, 1, 1] 391s -0.476 391s -0.964 391s 0.404 391s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.434 391s 0.888 391s 0.366 391s 391s x1: [1, 2, 2, 1] 391s -0.278 391s -0.207 391s 391s y: [3, 2, 2, 1] 391s 0.121 -0.247 -0.102 391s 0.090 -0.184 -0.076 391s testing: mul_mat, [4, 1, 2, 3] = [3, 4, 2, 3] * [3, 1, 2, 3] 391s x0: [3, 4, 2, 3] 391s 0.975 -0.497 -0.877 391s 0.352 -0.243 -0.081 391s 0.207 0.826 0.811 391s 0.260 -0.609 -0.432 391s 391s x1: [3, 1, 2, 3] 391s 0.911 0.107 -0.146 391s 391s y: [4, 1, 2, 3] 391s 0.964 0.307 0.158 0.235 391s test-mul-mat0: iter:262/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.368 0.220 391s -0.542 -0.445 391s 0.715 0.630 391s 391s x1: [2, 2, 1, 1] 391s -0.598 -0.184 391s -0.385 -0.796 391s 391s y: [3, 2, 1, 1] 391s 0.180 0.406 -0.544 391s -0.034 0.563 -0.776 391s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.682 -0.479 391s 0.531 -0.764 391s -0.764 0.161 391s 391s x1: [2, 3, 4, 1] 391s -0.715 -0.790 391s -0.375 -0.588 391s 0.336 0.857 391s 391s y: [3, 3, 4, 1] 391s -0.109 0.224 0.419 391s 0.026 0.250 0.192 391s -0.181 -0.476 -0.120 391s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.541 0.532 391s 0.549 0.638 391s 391s x1: [2, 3, 4, 1] 391s 0.849 -0.120 391s 0.977 -0.583 391s 0.930 0.277 391s 391s y: [2, 3, 4, 1] 391s 0.396 0.389 391s 0.218 0.164 391s 0.651 0.687 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.438 -0.744 391s 0.920 -0.830 391s 391s x1: [2, 2, 1, 1] 391s -0.752 0.515 391s -0.405 -0.383 391s 391s y: [2, 2, 1, 1] 391s -0.053 -1.119 391s 0.463 -0.055 391s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.382 0.058 391s 0.060 -0.137 391s 391s x1: [2, 2, 4, 1] 391s 0.067 -0.255 391s 0.953 0.099 391s 391s y: [2, 2, 4, 1] 391s 0.011 0.039 391s 0.370 0.044 391s testing: mul_mat, [4, 2, 4, 1] = [2, 4, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.497 -0.450 391s -0.828 -0.042 391s -0.110 0.731 391s 0.808 -0.078 391s 391s x1: [2, 2, 4, 1] 391s -0.247 -0.992 391s -0.126 -0.931 391s 391s y: [4, 2, 4, 1] 391s 0.569 0.246 -0.697 -0.123 391s 0.482 0.144 -0.667 -0.030 391s test-mul-mat0: iter:263/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.533 -0.200 -0.649 391s -0.050 0.724 0.492 391s 0.442 0.522 -0.298 391s 391s x1: [3, 1, 1, 1] 391s -0.624 -0.106 0.249 391s 391s y: [3, 1, 1, 1] 391s 0.192 0.077 -0.406 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.894 0.115 0.043 391s 0.460 -0.108 -0.539 391s -0.377 0.218 -0.163 391s 0.517 -0.533 -0.200 391s 391s x1: [3, 3, 1, 1] 391s -0.418 0.225 -0.155 391s -0.352 0.846 0.215 391s 0.403 0.254 0.511 391s 391s y: [4, 3, 1, 1] 391s 0.393 -0.133 0.232 -0.305 391s 0.422 -0.369 0.282 -0.676 391s -0.309 -0.117 -0.180 -0.028 391s testing: mul_mat, [2, 4, 1, 2] = [3, 2, 1, 2] * [3, 4, 1, 2] 391s x0: [3, 2, 1, 2] 391s 0.027 -0.859 0.868 391s -0.456 -0.392 -0.332 391s 391s x1: [3, 4, 1, 2] 391s 0.985 0.267 0.152 391s -0.065 -0.009 -0.356 391s -0.623 -0.487 0.346 391s -0.041 0.738 -0.809 391s 391s y: [2, 4, 1, 2] 391s -0.071 -0.604 391s -0.303 0.151 391s 0.702 0.360 391s -1.337 -0.002 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.298 0.406 0.417 391s 391s x1: [3, 2, 1, 1] 391s 0.841 0.380 0.810 391s -0.200 0.118 -1.000 391s 391s y: [1, 2, 1, 1] 391s 0.241 391s -0.310 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.647 391s 0.512 391s -0.114 391s 0.276 391s 391s x1: [1, 3, 1, 1] 391s 0.539 391s -0.678 391s -0.057 391s 391s y: [4, 3, 1, 1] 391s -0.349 0.276 -0.061 0.149 391s 0.439 -0.347 0.077 -0.187 391s 0.037 -0.029 0.006 -0.016 391s testing: mul_mat, [1, 1, 1, 2] = [4, 1, 1, 2] * [4, 1, 1, 2] 391s x0: [4, 1, 1, 2] 391s -0.102 -0.550 -0.232 0.362 391s 391s x1: [4, 1, 1, 2] 391s -0.457 0.913 0.417 -0.590 391s 391s y: [1, 1, 1, 2] 391s -0.766 391s test-mul-mat0: iter:264/500 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.989 0.253 391s 0.370 -0.124 391s -0.471 0.912 391s -0.211 0.946 391s 391s x1: [2, 2, 1, 1] 391s -0.422 -0.085 391s 0.506 -0.100 391s 391s y: [4, 2, 1, 1] 391s -0.439 -0.145 0.121 0.008 391s 0.476 0.200 -0.330 -0.202 391s testing: mul_mat, [4, 4, 2, 1] = [2, 4, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.981 -0.416 391s -0.522 0.526 391s 0.484 -0.664 391s 0.515 -0.263 391s 391s x1: [2, 4, 2, 1] 391s -0.677 0.247 391s 0.971 0.401 391s -0.767 0.872 391s -0.700 -0.316 391s 391s y: [4, 4, 2, 1] 391s 0.561 0.483 -0.492 -0.414 391s -1.119 -0.296 0.203 0.395 391s 0.389 0.859 -0.950 -0.624 391s 0.818 0.199 -0.129 -0.278 391s testing: mul_mat, [3, 4, 2, 4] = [2, 3, 2, 4] * [2, 4, 2, 4] 391s x0: [2, 3, 2, 4] 391s -0.600 -0.605 391s 0.181 0.756 391s -0.290 0.094 391s 391s x1: [2, 4, 2, 4] 391s 0.182 0.342 391s -0.340 0.054 391s 0.642 0.344 391s 0.694 0.304 391s 391s y: [3, 4, 2, 4] 391s -0.316 0.292 -0.021 391s 0.171 -0.021 0.104 391s -0.593 0.376 -0.154 391s -0.600 0.355 -0.173 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.525 0.658 391s 391s x1: [2, 3, 1, 1] 391s 0.095 0.655 391s 0.893 -0.254 391s -0.282 0.711 391s 391s y: [1, 3, 1, 1] 391s 0.481 391s 0.302 391s 0.319 391s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 4, 2, 1] 391s 0.430 391s -0.485 391s -0.695 391s -0.727 391s 391s x1: [1, 2, 2, 1] 391s -0.959 391s 0.668 391s 391s y: [4, 2, 2, 1] 391s -0.413 0.466 0.666 0.698 391s 0.287 -0.324 -0.464 -0.486 391s testing: mul_mat, [4, 1, 2, 4] = [4, 4, 2, 4] * [4, 1, 2, 4] 391s x0: [4, 4, 2, 4] 391s 0.606 0.665 -0.738 -0.805 391s 0.580 0.044 -0.377 0.082 391s 0.482 -0.238 -0.229 0.658 391s -0.205 -0.155 -0.803 0.425 391s 391s x1: [4, 1, 2, 4] 391s 0.704 0.107 0.034 -0.262 391s 391s y: [4, 1, 2, 4] 391s 0.683 0.379 0.134 -0.300 391s test-mul-mat0: iter:265/500 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.664 -0.527 391s 0.662 0.346 391s 391s x1: [2, 2, 1, 1] 391s -0.506 0.638 391s 0.790 -0.251 391s 391s y: [2, 2, 1, 1] 391s -0.000 -0.114 391s -0.392 0.436 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.027 0.652 391s -0.677 0.465 391s 391s x1: [2, 2, 1, 1] 391s 0.446 -0.650 391s 0.185 -0.239 391s 391s y: [2, 2, 1, 1] 391s -0.412 -0.604 391s -0.151 -0.237 391s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 1, 1, 2] 391s -0.373 -0.589 391s 391s x1: [2, 2, 1, 2] 391s -0.368 -0.788 391s 0.080 -0.665 391s 391s y: [1, 2, 1, 2] 391s 0.602 391s 0.362 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.654 -0.037 391s 0.302 -0.693 391s -0.039 0.573 391s -0.873 0.313 391s 391s x1: [2, 1, 1, 1] 391s -0.095 0.080 391s 391s y: [4, 1, 1, 1] 391s -0.065 -0.084 0.050 0.108 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.065 0.111 -0.961 0.481 391s -0.048 -0.949 -0.323 0.513 391s -0.842 0.856 -0.733 0.172 391s 391s x1: [4, 2, 1, 1] 391s 0.492 -0.666 0.993 -0.480 391s 0.986 -0.684 0.985 -0.383 391s 391s y: [3, 2, 1, 1] 391s -1.228 0.042 -1.795 391s -1.143 0.088 -2.203 391s testing: mul_mat, [2, 4, 1, 2] = [3, 2, 1, 2] * [3, 4, 1, 2] 391s x0: [3, 2, 1, 2] 391s 0.617 0.258 -0.344 391s 0.143 0.098 -0.570 391s 391s x1: [3, 4, 1, 2] 391s -0.439 -0.863 -0.174 391s 0.863 0.098 -0.047 391s -0.174 0.405 -0.474 391s -0.861 -0.102 -0.141 391s 391s y: [2, 4, 1, 2] 391s -0.433 -0.048 391s 0.574 0.160 391s 0.161 0.285 391s -0.509 -0.053 391s test-mul-mat0: iter:266/500 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.482 -0.684 0.180 0.060 391s -0.219 -0.829 -0.324 0.924 391s 391s x1: [4, 2, 1, 1] 391s -0.806 0.594 -0.380 0.612 391s -0.561 0.068 0.015 0.901 391s 391s y: [2, 2, 1, 1] 391s -0.826 0.373 391s -0.261 0.894 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s -0.297 0.923 0.087 -0.126 391s -0.400 0.011 -0.698 -0.626 391s 391s x1: [4, 2, 3, 1] 391s 0.429 0.774 -0.420 0.858 391s 0.433 0.373 0.114 -0.045 391s 391s y: [2, 2, 3, 1] 391s 0.442 -0.407 391s 0.231 -0.220 391s testing: mul_mat, [2, 2, 3, 4] = [4, 2, 3, 4] * [4, 2, 3, 4] 391s x0: [4, 2, 3, 4] 391s 0.665 0.962 -0.388 0.418 391s 0.231 -0.159 -0.203 -0.472 391s 391s x1: [4, 2, 3, 4] 391s -0.606 0.743 0.294 -0.616 391s 0.266 -0.843 -0.310 0.969 391s 391s y: [2, 2, 3, 4] 391s -0.060 -0.027 391s -0.108 -0.199 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.389 0.543 0.197 0.269 391s 391s x1: [4, 2, 1, 1] 391s -0.575 -0.250 -0.388 0.809 391s -0.391 -0.374 0.478 0.796 391s 391s y: [1, 2, 1, 1] 391s 0.229 391s 0.258 391s testing: mul_mat, [2, 4, 3, 1] = [1, 2, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.464 391s -0.721 391s 391s x1: [1, 4, 3, 1] 391s -0.026 391s -0.920 391s -0.763 391s -0.638 391s 391s y: [2, 4, 3, 1] 391s 0.012 0.018 391s 0.427 0.663 391s 0.354 0.550 391s 0.296 0.460 391s testing: mul_mat, [3, 1, 3, 4] = [2, 3, 3, 4] * [2, 1, 3, 4] 391s x0: [2, 3, 3, 4] 391s 0.312 0.157 391s -0.644 0.561 391s 0.141 0.222 391s 391s x1: [2, 1, 3, 4] 391s -0.404 0.817 391s 391s y: [3, 1, 3, 4] 391s 0.002 0.718 0.125 391s test-mul-mat0: iter:267/500 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.640 -0.142 391s 0.901 0.083 391s -0.636 -0.564 391s 0.171 0.571 391s 391s x1: [2, 1, 1, 1] 391s -0.165 0.254 391s 391s y: [4, 1, 1, 1] 391s 0.070 -0.128 -0.039 0.117 391s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.551 -0.838 391s 0.847 0.646 391s -0.162 -0.155 391s 0.887 -0.181 391s 391s x1: [2, 4, 4, 1] 391s -0.863 -0.245 391s -0.312 0.779 391s 0.912 0.197 391s -0.325 -0.002 391s 391s y: [4, 4, 4, 1] 391s 0.681 -0.890 0.178 -0.721 391s -0.481 0.239 -0.070 -0.418 391s -0.668 0.900 -0.179 0.773 391s 0.181 -0.276 0.053 -0.287 391s testing: mul_mat, [1, 4, 4, 3] = [2, 1, 4, 3] * [2, 4, 4, 3] 391s x0: [2, 1, 4, 3] 391s -0.089 -0.850 391s 391s x1: [2, 4, 4, 3] 391s -0.526 0.834 391s -0.136 0.312 391s -0.320 -0.249 391s -0.869 0.099 391s 391s y: [1, 4, 4, 3] 391s -0.662 391s -0.253 391s 0.240 391s -0.007 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.976 0.962 391s -0.165 0.745 391s 0.936 -0.914 391s 391s x1: [2, 1, 1, 1] 391s 0.909 0.169 391s 391s y: [3, 1, 1, 1] 391s -0.724 -0.024 0.696 391s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.103 0.056 0.136 391s 391s x1: [3, 2, 4, 1] 391s -0.478 0.909 0.021 391s -0.094 0.869 -0.102 391s 391s y: [1, 2, 4, 1] 391s 0.004 391s 0.025 391s testing: mul_mat, [2, 3, 4, 3] = [1, 2, 4, 3] * [1, 3, 4, 3] 391s x0: [1, 2, 4, 3] 391s -0.514 391s 0.256 391s 391s x1: [1, 3, 4, 3] 391s 0.525 391s 0.867 391s 0.702 391s 391s y: [2, 3, 4, 3] 391s -0.270 0.135 391s -0.446 0.222 391s -0.361 0.180 391s test-mul-mat0: iter:268/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.501 0.558 0.823 391s -0.804 -0.400 0.168 391s 0.500 0.670 -0.007 391s 391s x1: [3, 1, 1, 1] 391s 0.002 -0.755 0.492 391s 391s y: [3, 1, 1, 1] 391s -0.016 0.383 -0.508 391s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.280 -0.430 -0.064 391s -0.219 -0.872 -0.241 391s -0.022 -0.272 0.927 391s 391s x1: [3, 3, 2, 1] 391s -0.510 -0.011 -0.502 391s -0.731 0.613 -0.180 391s 0.903 -0.658 0.180 391s 391s y: [3, 3, 2, 1] 391s -0.106 0.242 -0.451 391s -0.457 -0.331 -0.318 391s 0.525 0.333 0.326 391s testing: mul_mat, [4, 3, 2, 3] = [3, 4, 2, 3] * [3, 3, 2, 3] 391s x0: [3, 4, 2, 3] 391s 0.478 0.997 -0.058 391s 0.435 -0.190 0.357 391s 0.629 -0.237 0.114 391s 0.242 -0.169 -0.900 391s 391s x1: [3, 3, 2, 3] 391s 0.139 0.341 0.418 391s -0.885 0.229 0.240 391s -0.902 -0.940 0.128 391s 391s y: [4, 3, 2, 3] 391s 0.382 0.145 0.054 -0.401 391s -0.208 -0.342 -0.584 -0.469 391s -1.376 -0.168 -0.330 -0.174 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.793 0.987 -0.343 391s -0.454 0.904 0.165 391s -0.056 0.146 -0.847 391s -0.174 0.542 -0.362 391s 391s x1: [3, 4, 1, 1] 391s 0.900 -0.821 0.131 391s -0.001 -0.049 0.402 391s 0.807 0.009 0.935 391s 0.028 -0.130 0.695 391s 391s y: [4, 4, 1, 1] 391s -1.569 -1.129 -0.281 -0.649 391s -0.185 0.022 -0.347 -0.172 391s -0.951 -0.203 -0.836 -0.474 391s -0.389 -0.016 -0.609 -0.327 391s testing: mul_mat, [4, 3, 2, 1] = [4, 4, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 4, 2, 1] 391s -0.081 -0.195 0.143 -0.346 391s -0.349 0.667 0.265 0.624 391s 0.029 0.775 0.714 0.101 391s -0.916 0.983 0.557 0.042 391s 391s x1: [4, 3, 2, 1] 391s 0.692 0.178 0.476 0.489 391s 0.731 -0.587 0.389 0.910 391s 0.545 -0.612 -0.139 -0.054 391s 391s y: [4, 3, 2, 1] 391s -0.192 0.309 0.548 -0.172 391s -0.204 0.025 -0.064 -0.991 391s 0.074 -0.668 -0.563 -1.180 391s testing: mul_mat, [4, 4, 2, 3] = [4, 4, 2, 3] * [4, 4, 2, 3] 391s x0: [4, 4, 2, 3] 391s 0.620 0.254 0.585 0.930 391s 0.611 -0.988 0.054 -0.393 391s -0.213 0.613 -0.421 -0.867 391s -0.759 0.725 -0.494 -0.463 391s 391s x1: [4, 4, 2, 3] 391s -0.609 0.850 -0.182 0.196 391s 0.517 -0.406 0.179 -0.340 391s 0.858 -0.107 -0.783 -0.487 391s -0.483 0.318 0.555 0.002 391s 391s y: [4, 4, 2, 3] 391s -0.085 -1.298 0.557 1.077 391s 0.006 0.860 -0.140 -0.618 391s -0.407 0.779 0.504 -0.117 391s 0.107 -0.580 0.063 0.323 391s test-mul-mat0: iter:269/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.313 0.321 391s 0.881 -0.250 391s -0.912 -0.498 391s -0.073 0.257 391s 391s x1: [2, 4, 1, 1] 391s -0.280 0.879 391s -0.316 0.793 391s -0.645 0.698 391s -0.905 0.196 391s 391s y: [4, 4, 1, 1] 391s 0.370 -0.467 -0.182 0.247 391s 0.354 -0.477 -0.107 0.227 391s 0.426 -0.743 0.241 0.227 391s 0.346 -0.847 0.727 0.117 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.577 -0.844 391s 0.895 -0.942 391s 0.783 -0.386 391s 0.936 -0.533 391s 391s x1: [2, 4, 1, 1] 391s -0.918 0.185 391s 0.999 -0.430 391s 0.023 0.768 391s 0.307 0.561 391s 391s y: [4, 4, 1, 1] 391s 0.374 -0.996 -0.790 -0.958 391s -0.214 1.300 0.948 1.165 391s -0.661 -0.703 -0.278 -0.388 391s -0.650 -0.253 0.024 -0.011 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.940 -0.103 391s 0.754 -0.134 391s 391s x1: [2, 4, 1, 1] 391s -0.592 -0.709 391s -0.835 -0.497 391s 0.487 0.370 391s 0.190 -0.191 391s 391s y: [2, 4, 1, 1] 391s -0.484 -0.352 391s -0.734 -0.563 391s 0.420 0.318 391s 0.198 0.169 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.075 0.712 391s 0.922 -0.365 391s 391s x1: [2, 2, 1, 1] 391s -0.845 0.835 391s -0.949 -0.846 391s 391s y: [2, 2, 1, 1] 391s 0.532 -1.084 391s -0.673 -0.566 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.426 0.041 391s -0.942 -0.534 391s -0.871 -0.579 391s 391s x1: [2, 2, 1, 1] 391s 0.549 -0.865 391s -0.209 0.443 391s 391s y: [3, 2, 1, 1] 391s -0.269 -0.055 0.023 391s 0.107 -0.040 -0.075 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.902 0.614 0.581 391s 391s x1: [3, 2, 1, 1] 391s 0.206 -0.032 0.908 391s -0.424 -0.842 -0.283 391s 391s y: [1, 2, 1, 1] 391s 0.323 391s -0.299 391s test-mul-mat0: iter:270/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.164 0.862 391s -0.908 -0.106 391s 0.992 0.133 391s 0.359 -0.588 391s 391s x1: [2, 4, 1, 1] 391s 0.922 0.822 391s -0.909 0.844 391s 0.534 -0.274 391s 0.393 0.669 391s 391s y: [4, 4, 1, 1] 391s 0.557 -0.925 1.024 -0.152 391s 0.877 0.736 -0.790 -0.823 391s -0.324 -0.456 0.493 0.353 391s 0.513 -0.428 0.478 -0.252 391s testing: mul_mat, [2, 4, 4, 1] = [2, 2, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.353 0.377 391s 0.481 -0.609 391s 391s x1: [2, 4, 4, 1] 391s -0.661 -0.672 391s -0.680 -0.085 391s -0.514 0.037 391s -0.257 -0.416 391s 391s y: [2, 4, 4, 1] 391s -0.487 0.092 391s -0.272 -0.275 391s -0.168 -0.270 391s -0.248 0.130 391s testing: mul_mat, [3, 2, 4, 2] = [2, 3, 4, 2] * [2, 2, 4, 2] 391s x0: [2, 3, 4, 2] 391s 0.505 0.033 391s -0.547 0.487 391s 0.277 -0.443 391s 391s x1: [2, 2, 4, 2] 391s -0.152 -0.785 391s 0.567 0.615 391s 391s y: [3, 2, 4, 2] 391s -0.103 -0.299 0.306 391s 0.307 -0.011 -0.115 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.688 -0.547 391s -0.022 -0.372 391s 0.284 -0.553 391s 391s x1: [2, 3, 1, 1] 391s -0.994 -0.430 391s 0.501 -0.637 391s 0.296 0.265 391s 391s y: [3, 3, 1, 1] 391s 0.919 0.182 -0.044 391s 0.004 0.226 0.495 391s -0.349 -0.105 -0.063 391s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.964 0.964 0.843 391s 391s x1: [3, 2, 4, 1] 391s 0.932 0.782 -0.357 391s 0.864 -0.576 -0.883 391s 391s y: [1, 2, 4, 1] 391s 1.352 391s -0.466 391s testing: mul_mat, [3, 3, 4, 2] = [1, 3, 4, 2] * [1, 3, 4, 2] 391s x0: [1, 3, 4, 2] 391s -0.010 391s -0.636 391s -0.484 391s 391s x1: [1, 3, 4, 2] 391s -0.434 391s -0.359 391s -0.218 391s 391s y: [3, 3, 4, 2] 391s 0.004 0.276 0.210 391s 0.004 0.228 0.174 391s 0.002 0.139 0.106 391s test-mul-mat0: iter:271/500 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.148 -0.650 0.258 -0.187 391s -0.211 0.775 0.336 0.693 391s -0.200 0.073 -0.151 -0.449 391s -0.320 -0.387 -0.251 0.180 391s 391s x1: [4, 2, 1, 1] 391s -0.111 0.021 -0.172 0.879 391s 0.385 0.344 0.082 -0.453 391s 391s y: [4, 2, 1, 1] 391s -0.206 0.591 -0.345 0.229 391s -0.175 -0.101 0.139 -0.358 391s testing: mul_mat, [1, 4, 2, 1] = [4, 1, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 1, 2, 1] 391s 0.800 -0.819 -0.615 -0.824 391s 391s x1: [4, 4, 2, 1] 391s -0.363 -0.618 0.131 -0.115 391s 0.413 -0.625 0.774 -0.566 391s 0.203 0.653 0.818 -0.453 391s -0.264 -0.634 0.457 0.587 391s 391s y: [1, 4, 2, 1] 391s 0.230 391s 0.833 391s -0.502 391s -0.457 391s testing: mul_mat, [3, 1, 2, 4] = [4, 3, 2, 4] * [4, 1, 2, 4] 391s x0: [4, 3, 2, 4] 391s -0.487 -0.878 0.697 -0.556 391s 0.840 0.794 -0.225 0.063 391s -0.620 -0.157 0.484 -0.744 391s 391s x1: [4, 1, 2, 4] 391s -0.363 0.925 -0.656 0.003 391s 391s y: [3, 1, 2, 4] 391s -1.094 0.577 -0.239 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.600 -0.296 -0.146 -0.825 391s 391s x1: [4, 3, 1, 1] 391s 0.249 0.731 -0.796 -0.586 391s -0.333 -0.422 -0.013 -0.065 391s -0.247 -0.311 -0.513 -0.298 391s 391s y: [1, 3, 1, 1] 391s 0.234 391s 0.380 391s 0.562 391s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.918 391s -0.672 391s -0.223 391s -0.275 391s 391s x1: [1, 4, 2, 1] 391s -0.683 391s -0.605 391s 0.917 391s -0.238 391s 391s y: [4, 4, 2, 1] 391s 0.627 0.459 0.152 0.188 391s 0.555 0.406 0.135 0.166 391s -0.842 -0.616 -0.204 -0.252 391s 0.219 0.160 0.053 0.066 391s testing: mul_mat, [3, 1, 2, 4] = [4, 3, 2, 4] * [4, 1, 2, 4] 391s x0: [4, 3, 2, 4] 391s 0.205 -0.558 -0.618 0.024 391s -0.875 -0.760 0.959 -0.467 391s 0.906 -0.550 0.395 -0.485 391s 391s x1: [4, 1, 2, 4] 391s -0.561 0.122 -0.515 0.427 391s 391s y: [3, 1, 2, 4] 391s 0.146 -0.296 -0.985 391s test-mul-mat0: iter:272/500 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.532 391s -0.112 391s -0.188 391s 0.738 391s 391s x1: [1, 4, 1, 1] 391s -0.550 391s -0.517 391s 0.031 391s 0.728 391s 391s y: [4, 4, 1, 1] 391s 0.292 0.061 0.104 -0.405 391s 0.275 0.058 0.097 -0.381 391s -0.017 -0.003 -0.006 0.023 391s -0.387 -0.081 -0.137 0.537 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.992 391s -0.436 391s 391s x1: [1, 4, 1, 1] 391s 0.598 391s -0.081 391s -0.992 391s -0.696 391s 391s y: [2, 4, 1, 1] 391s 0.593 -0.261 391s -0.080 0.035 391s -0.985 0.433 391s -0.691 0.303 391s testing: mul_mat, [4, 2, 1, 2] = [1, 4, 1, 2] * [1, 2, 1, 2] 391s x0: [1, 4, 1, 2] 391s 0.947 391s -0.338 391s -0.173 391s -0.545 391s 391s x1: [1, 2, 1, 2] 391s 0.473 391s -0.324 391s 391s y: [4, 2, 1, 2] 391s 0.447 -0.160 -0.082 -0.258 391s -0.307 0.110 0.056 0.177 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.099 391s -0.410 391s -0.769 391s 391s x1: [1, 4, 1, 1] 391s 0.147 391s -0.269 391s 0.173 391s 0.630 391s 391s y: [3, 4, 1, 1] 391s -0.014 -0.060 -0.113 391s 0.027 0.110 0.207 391s -0.017 -0.071 -0.133 391s -0.062 -0.259 -0.485 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.680 0.968 0.003 391s 0.976 -0.745 -0.278 391s 0.691 0.410 -0.587 391s 0.937 -0.356 -0.050 391s 391s x1: [3, 1, 1, 1] 391s 0.789 0.401 0.968 391s 391s y: [4, 1, 1, 1] 391s -0.145 0.203 0.141 0.549 391s testing: mul_mat, [2, 3, 1, 2] = [4, 2, 1, 2] * [4, 3, 1, 2] 391s x0: [4, 2, 1, 2] 391s -0.009 0.476 -0.890 0.097 391s 0.697 0.713 -0.574 0.350 391s 391s x1: [4, 3, 1, 2] 391s 0.384 0.240 0.405 -0.447 391s -0.445 -0.585 0.276 0.702 391s 0.146 -0.551 0.332 -0.091 391s 391s y: [2, 3, 1, 2] 391s -0.293 0.050 391s -0.451 -0.639 391s -0.568 -0.513 391s test-mul-mat0: iter:273/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.392 0.409 -0.313 391s 0.081 -0.601 -0.616 391s 391s x1: [3, 2, 1, 1] 391s 0.441 0.864 -0.815 391s -0.450 -0.610 0.828 391s 391s y: [2, 2, 1, 1] 391s 0.435 0.019 391s -0.332 -0.180 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.141 0.750 0.059 391s 0.784 0.357 -0.548 391s 391s x1: [3, 2, 1, 1] 391s -0.444 -0.888 -0.506 391s -0.018 0.209 0.844 391s 391s y: [2, 2, 1, 1] 391s -0.633 -0.387 391s 0.209 -0.402 391s testing: mul_mat, [4, 2, 1, 4] = [3, 4, 1, 4] * [3, 2, 1, 4] 391s x0: [3, 4, 1, 4] 391s 0.159 0.538 0.236 391s -0.057 -0.105 0.687 391s -0.899 0.192 -0.560 391s 0.983 -0.650 -0.365 391s 391s x1: [3, 2, 1, 4] 391s 0.158 -0.703 -0.247 391s 0.882 0.158 -0.805 391s 391s y: [4, 2, 1, 4] 391s -0.411 -0.105 -0.139 0.702 391s 0.036 -0.620 -0.311 1.059 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.976 -0.378 -0.588 391s -0.295 -0.217 -0.446 391s -0.655 -0.161 -0.358 391s 0.564 -0.360 -0.608 391s 391s x1: [3, 4, 1, 1] 391s 0.486 -0.797 0.981 391s -0.691 -0.365 -0.597 391s -0.418 0.448 0.917 391s -0.201 0.624 0.731 391s 391s y: [4, 4, 1, 1] 391s 0.200 -0.408 -0.541 -0.035 391s -0.186 0.549 0.725 0.105 391s -1.116 -0.383 -0.128 -0.955 391s -0.862 -0.402 -0.231 -0.783 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.747 -0.365 -0.043 -0.744 391s -0.097 -0.388 0.417 -0.398 391s 0.809 0.853 0.961 -0.848 391s 0.877 0.340 0.740 -0.190 391s 391s x1: [4, 3, 1, 1] 391s -0.097 -0.628 -0.935 -0.368 391s 0.967 -0.398 -0.881 -0.830 391s -0.417 -0.572 -0.195 -0.014 391s 391s y: [4, 3, 1, 1] 391s 0.615 0.010 -1.201 -0.920 391s 0.079 0.024 0.298 0.219 391s 0.539 0.187 -1.001 -0.702 391s testing: mul_mat, [2, 4, 1, 4] = [4, 2, 1, 4] * [4, 4, 1, 4] 391s x0: [4, 2, 1, 4] 391s -0.851 -0.988 0.417 0.220 391s 0.881 -0.191 0.556 -0.253 391s 391s x1: [4, 4, 1, 4] 391s 0.243 -0.456 0.713 0.615 391s -0.391 -0.655 0.582 0.211 391s -0.536 0.751 0.794 -0.108 391s -0.444 -0.220 0.902 -0.ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 191 391s 391s y: [2, 4, 1, 4] 391s 0.676 0.542 391s 1.269 0.051 391s 0.021 -0.148 391s 0.929 0.201 391s test-mul-mat0: iter:274/500 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.750 -0.044 391s 0.359 -0.421 391s 391s x1: [2, 1, 1, 1] 391s -0.742 -0.231 391s 391s y: [2, 1, 1, 1] 391s 0.567 -0.169 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.702 -0.222 391s 0.297 -0.903 391s 0.169 0.212 391s 0.160 -0.267 391s 391s x1: [2, 2, 1, 1] 391s -0.080 -0.991 391s -0.045 0.347 391s 391s y: [4, 2, 1, 1] 391s 0.164 0.872 -0.223 0.252 391s -0.109 -0.327 0.066 -0.100 391s testing: mul_mat, [1, 4, 1, 4] = [2, 1, 1, 4] * [2, 4, 1, 4] 391s x0: [2, 1, 1, 4] 391s -0.220 0.823 391s 391s x1: [2, 4, 1, 4] 391s -0.267 -0.937 391s -0.507 0.326 391s -0.479 0.184 391s -0.933 -0.140 391s 391s y: [1, 4, 1, 4] 391s -0.713 391s 0.379 391s 0.257 391s 0.090 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.934 -0.610 391s 391s x1: [2, 1, 1, 1] 391s -0.053 0.996 391s 391s y: [1, 1, 1, 1] 391s -0.557 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.493 391s 0.031 391s 391s x1: [1, 2, 1, 1] 391s 0.232 391s -0.456 391s 391s y: [2, 2, 1, 1] 391s 0.115 0.007 391s -0.225 -0.014 391s testing: mul_mat, [3, 1, 1, 4] = [2, 3, 1, 4] * [2, 1, 1, 4] 391s x0: [2, 3, 1, 4] 391s 0.536 -0.515 391s 0.330 -0.890 391s 0.100 -0.077 391s 391s x1: [2, 1, 1, 4] 391s 0.292 0.102 391s 391s y: [3, 1, 1, 4] 391s 0.104 0.006 0.021 391s test-mul-mat0: iter:275/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.897 0.703 -0.440 391s 391s x1: [3, 1, 1, 1] 391s -0.461 -0.433 -0.398 391s 391s y: [1, 1, 1, 1] 391s -0.543 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.438 -0.197 -0.993 391s 391s x1: [3, 1, 1, 1] 391s -0.993 -0.374 -0.484 391s 391s y: [1, 1, 1, 1] 391s 0.119 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.335 -0.761 0.406 391s 391s x1: [3, 1, 1, 1] 391s 0.756 -0.964 0.928 391s 391s y: [1, 1, 1, 1] 391s 1.364 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.119 0.213 0.901 391s -0.623 -0.229 0.213 391s 391s x1: [3, 1, 1, 1] 391s 0.178 0.914 -0.659 391s 391s y: [2, 1, 1, 1] 391s -0.421 -0.460 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.628 -0.546 391s -0.857 0.582 391s -0.539 0.264 391s 391s x1: [2, 3, 1, 1] 391s 0.310 -0.532 391s 0.815 0.385 391s -0.635 0.518 391s 391s y: [3, 3, 1, 1] 391s 0.096 -0.575 -0.307 391s -0.722 -0.474 -0.337 391s 0.116 0.845 0.479 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.291 0.814 -0.457 391s 391s x1: [3, 2, 1, 1] 391s -0.539 0.338 0.300 391s -0.611 0.303 -0.365 391s 391s y: [1, 2, 1, 1] 391s 0.294 391s 0.591 391s test-mul-mat0: iter:276/500 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.109 -0.839 0.624 391s -0.164 -0.467 0.768 391s 0.298 -0.013 0.349 391s -0.438 0.449 -0.313 391s 391s x1: [3, 2, 1, 1] 391s 0.079 -0.805 0.328 391s -0.708 0.506 0.796 391s 391s y: [4, 2, 1, 1] 391s 0.871 0.614 0.148 -0.498 391s 0.150 0.491 0.061 0.288 391s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.610 -0.258 0.520 391s 391s x1: [3, 4, 3, 1] 391s 0.862 0.838 0.990 391s -0.503 -0.534 -0.301 391s -0.689 0.009 0.748 391s -0.823 -0.567 0.175 391s 391s y: [1, 4, 3, 1] 391s -0.228 391s 0.288 391s 0.807 391s 0.740 391s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.982 -0.074 0.747 391s 0.702 -0.645 -0.170 391s 391s x1: [3, 1, 3, 1] 391s 0.371 0.937 0.495 391s 391s y: [2, 1, 3, 1] 391s 0.665 -0.429 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.973 -0.159 0.482 391s -0.465 -0.204 -0.807 391s 391s x1: [3, 2, 1, 1] 391s -0.859 0.074 0.086 391s 0.512 0.011 -0.419 391s 391s y: [2, 2, 1, 1] 391s -0.806 0.315 391s 0.294 0.098 391s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.821 0.097 391s -0.484 0.386 391s -0.151 0.163 391s 0.190 0.632 391s 391s x1: [2, 3, 3, 1] 391s -0.222 -0.592 391s 0.940 -0.521 391s -0.237 -0.230 391s 391s y: [4, 3, 3, 1] 391s 0.125 -0.121 -0.063 -0.416 391s -0.822 -0.656 -0.227 -0.151 391s 0.172 0.026 -0.002 -0.190 391s testing: mul_mat, [1, 2, 3, 1] = [4, 1, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.233 0.956 0.687 -0.129 391s 391s x1: [4, 2, 3, 1] 391s -0.019 -0.758 0.691 -0.823 391s 0.347 0.121 0.514 0.526 391s 391s y: [1, 2, 3, 1] 391s -0.140 391s 0.320 391s test-mul-mat0: iter:277/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.387 0.630 0.909 391s -0.133 -0.975 -0.517 391s 391s x1: [3, 1, 1, 1] 391s -0.805 0.020 0.654 391s 391s y: [2, 1, 1, 1] 391s 0.296 -0.251 391s testing: mul_mat, [3, 2, 4, 1] = [3, 3, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.058 -0.478 0.694 391s 0.926 -0.452 -0.824 391s 0.795 0.328 -0.552 391s 391s x1: [3, 2, 4, 1] 391s 0.869 -0.219 -0.729 391s 0.156 -0.334 0.039 391s 391s y: [3, 2, 4, 1] 391s -0.350 1.504 1.021 391s 0.196 0.264 -0.007 391s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.314 0.824 0.401 391s -0.439 -0.902 0.092 391s 391s x1: [3, 3, 4, 1] 391s -0.456 0.720 -0.187 391s -0.008 -0.329 -0.193 391s -0.522 -0.266 0.154 391s 391s y: [2, 3, 4, 1] 391s 0.661 -0.466 391s -0.346 0.283 391s 0.007 0.483 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.013 0.104 -0.183 391s 0.005 0.781 -0.256 391s -0.311 0.168 -0.340 391s 0.574 0.101 -0.586 391s 391s x1: [3, 2, 1, 1] 391s 0.602 -0.880 -0.908 391s 0.041 -0.673 0.182 391s 391s y: [4, 2, 1, 1] 391s 0.083 -0.452 -0.026 0.787 391s -0.103 -0.572 -0.187 -0.151 391s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 1, 4, 1] 391s -0.886 -0.082 -0.999 0.762 391s 391s x1: [4, 3, 4, 1] 391s -0.413 -0.042 0.844 -0.782 391s -0.114 -0.954 -0.947 0.118 391s 0.022 0.028 -0.978 -0.304 391s 391s y: [1, 3, 4, 1] 391s -1.070 391s 1.216 391s 0.724 391s testing: mul_mat, [2, 4, 4, 1] = [1, 2, 4, 1] * [1, 4, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.079 391s 0.686 391s 391s x1: [1, 4, 4, 1] 391s -0.508 391s 0.583 391s 0.155 391s -0.682 391s 391s y: [2, 4, 4, 1] 391s -0.040 -0.348 391s 0.046 0.400 391s 0.012 0.106 391s -0.054 -0.468 391s test-mul-mat0: iter:278/500 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.002 -0.825 391s 391s x1: [2, 1, 1, 1] 391s 0.415 0.020 391s 391s y: [1, 1, 1, 1] 391s -0.017 391s testing: mul_mat, [4, 1, 4, 1] = [2, 4, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.036 -0.608 391s 0.030 -0.040 391s -0.075 -0.891 391s -0.354 0.433 391s 391s x1: [2, 1, 4, 1] 391s 0.401 0.385 391s 391s y: [4, 1, 4, 1] 391s -0.220 -0.003 -0.373 0.025 391s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 1, 4, 1] 391s 0.967 0.152 391s 391s x1: [2, 4, 4, 1] 391s -0.235 0.235 391s -0.123 0.690 391s 0.344 0.523 391s 0.123 -0.389 391s 391s y: [1, 4, 4, 1] 391s -0.191 391s -0.013 391s 0.412 391s 0.060 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.663 -0.018 391s -0.435 -0.770 391s -0.322 0.503 391s 0.776 -0.149 391s 391s x1: [2, 1, 1, 1] 391s -0.834 0.423 391s 391s y: [4, 1, 1, 1] 391s -0.560 0.038 0.481 -0.710 391s testing: mul_mat, [2, 2, 4, 1] = [4, 2, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.695 0.171 0.746 -0.916 391s 0.418 0.854 -0.578 0.165 391s 391s x1: [4, 2, 4, 1] 391s -0.134 -0.770 0.073 0.157 391s -0.446 0.549 -0.671 0.252 391s 391s y: [2, 2, 4, 1] 391s -0.127 -0.729 391s -0.328 0.712 391s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.976 -0.866 391s 0.675 0.501 391s 0.422 -0.943 391s -0.099 0.144 391s 391s x1: [2, 4, 4, 1] 391s 0.548 -0.480 391s -0.158 0.295 391s -0.057 -0.074 391s -0.541 -0.921 391s 391s y: [4, 4, 4, 1] 391s 0.950 0.130 0.684 -0.123 391s -0.409 0.041 -0.345 0.058 391s 0.009 -0.076 0.046 -0.005 391s 0.270 -0.826 0.641 -0.079 391s test-mul-mat0: iter:279/500 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.643 0.931 -0.505 391s 0.664 0.196 -0.756 391s -0.706 -0.394 -0.176 391s -0.118 0.150 0.785 391s 391s x1: [3, 4, 1, 1] 391s 0.989 0.477 0.914 391s -0.429 -0.674 0.491 391s 0.378 -0.120 -0.660 391s 0.234 0.497 -0.870 391s 391s y: [4, 4, 1, 1] 391s 0.619 0.059 -1.047 0.672 391s -1.152 -0.789 0.482 0.335 391s 0.464 0.727 -0.104 -0.581 391s 1.053 0.911 -0.208 -0.636 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.772 -0.065 0.611 391s 0.269 0.065 0.059 391s 391s x1: [3, 4, 1, 1] 391s 0.949 0.669 -0.580 391s -0.562 0.090 0.190 391s -0.573 -0.433 0.104 391s -0.002 -0.108 -0.405 391s 391s y: [2, 4, 1, 1] 391s 0.335 0.265 391s -0.324 -0.134 391s -0.351 -0.176 391s -0.242 -0.032 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.202 -0.378 0.752 391s 391s x1: [3, 2, 1, 1] 391s -0.088 -0.004 0.554 391s -0.424 -0.808 0.798 391s 391s y: [1, 2, 1, 1] 391s 0.436 391s 0.991 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.173 -0.983 0.121 391s -0.861 -0.434 0.564 391s -0.293 0.274 -0.834 391s 391s x1: [3, 1, 1, 1] 391s 0.948 -0.593 0.702 391s 391s y: [3, 1, 1, 1] 391s 0.504 -0.163 -1.025 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.836 -0.096 0.566 391s -0.304 0.962 0.702 391s 391s x1: [3, 3, 1, 1] 391s 0.716 0.940 0.938 391s -0.349 0.550 0.208 391s 0.716 -0.390 -0.880 391s 391s y: [2, 3, 1, 1] 391s -0.157 1.346 391s 0.357 0.781 391s -1.060 -1.212 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.160 0.726 391s 0.974 0.247 391s 0.835 -0.045 391s 391s x1: [2, 3, 1, 1] 391s -0.416 0.319 391s 0.650 -0.009 391s 0.020 0.267 391s 391s y: [3, 3, 1, 1] 391s 0.165 -0.326 -0.361 391s 0.097 0.631 0.543 391s 0.197 0.086 0.005 391s test-mul-mat0: iter:280/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.220 391s -0.441 391s 0.039 391s 0.389 391s 391s x1: [1, 1, 1, 1] 391s 0.352 391s 391s y: [4, 1, 1, 1] 391s -0.077 -0.155 0.014 0.137 391s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.740 391s -0.992 391s 0.090 391s 0.280 391s 391s x1: [1, 4, 2, 1] 391s -0.321 391s 0.751 391s 0.553 391s 0.375 391s 391s y: [4, 4, 2, 1] 391s 0.238 0.318 -0.029 -0.090 391s -0.556 -0.745 0.068 0.210 391s -0.410 -0.549 0.050 0.155 391s -0.278 -0.372 0.034 0.105 391s testing: mul_mat, [3, 4, 2, 4] = [1, 3, 2, 4] * [1, 4, 2, 4] 391s x0: [1, 3, 2, 4] 391s -0.477 391s -0.539 391s -0.582 391s 391s x1: [1, 4, 2, 4] 391s -0.258 391s -0.834 391s -0.503 391s 0.698 391s 391s y: [3, 4, 2, 4] 391s 0.123 0.139 0.150 391s 0.398 0.450 0.486 391s 0.240 0.271 0.293 391s -0.333 -0.376 -0.406 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.329 391s 391s x1: [1, 3, 1, 1] 391s -0.859 391s -0.356 391s -0.092 391s 391s y: [1, 3, 1, 1] 391s 0.283 391s 0.117 391s 0.030 391s testing: mul_mat, [3, 1, 2, 1] = [1, 3, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.867 391s -0.683 391s 0.627 391s 391s x1: [1, 1, 2, 1] 391s 0.672 391s 391s y: [3, 1, 2, 1] 391s -0.583 -0.459 0.421 391s testing: mul_mat, [3, 1, 2, 4] = [3, 3, 2, 4] * [3, 1, 2, 4] 391s x0: [3, 3, 2, 4] 391s -0.637 0.326 -0.189 391s -0.469 0.951 0.937 391s 0.707 0.029 0.866 391s 391s x1: [3, 1, 2, 4] 391s 0.485 0.987 -0.163 391s 391s y: [3, 1, 2, 4] 391s 0.043 0.558 0.231 391s test-mul-mat0: iter:281/500 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.238 0.820 0.485 0.525 391s 0.032 -0.750 -0.649 -0.909 391s -0.241 0.946 0.208 0.093 391s -0.328 -0.084 -0.041 0.857 391s 391s x1: [4, 1, 1, 1] 391s -0.390 -0.691 -0.067 -0.862 391s 391s y: [4, 1, 1, 1] 391s -1.144 1.333 -0.653 -0.550 391s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 3, 3, 1] 391s 0.993 -0.547 0.464 0.448 391s -0.620 0.536 -0.152 0.514 391s -0.300 -0.079 0.566 -0.176 391s 391s x1: [4, 4, 3, 1] 391s 0.166 0.385 -0.690 0.896 391s -0.762 0.581 -0.551 -0.764 391s 0.691 0.408 -0.154 -1.000 391s -0.660 -0.016 0.577 0.579 391s 391s y: [3, 4, 3, 1] 391s 0.035 0.668 -0.628 391s -1.673 0.476 0.005 391s -0.056 -0.701 -0.150 391s -0.119 0.611 0.423 391s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.359 -0.717 0.909 -0.327 391s 391s x1: [4, 3, 3, 1] 391s -0.824 -0.085 0.665 -0.124 391s 0.836 0.230 0.700 -0.314 391s -0.987 -0.968 0.777 0.910 391s 391s y: [1, 3, 3, 1] 391s 1.001 391s 0.273 391s 1.456 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.326 -0.770 0.703 -0.701 391s -0.147 0.453 -0.882 -0.297 391s 391s x1: [4, 1, 1, 1] 391s -0.986 -0.091 -0.485 0.774 391s 391s y: [2, 1, 1, 1] 391s -0.492 0.301 391s testing: mul_mat, [2, 4, 3, 1] = [2, 2, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.269 -0.480 391s 0.800 -0.151 391s 391s x1: [2, 4, 3, 1] 391s -0.817 -0.940 391s 0.368 0.123 391s 0.657 0.263 391s 0.764 0.941 391s 391s y: [2, 4, 3, 1] 391s 0.672 -0.512 391s -0.158 0.276 391s -0.304 0.486 391s -0.658 0.470 391s testing: mul_mat, [3, 2, 3, 1] = [2, 3, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.361 0.140 391s 0.024 0.371 391s 0.905 -0.279 391s 391s x1: [2, 2, 3, 1] 391s 0.919 0.758 391s 0.026 -0.643 391s 391s y: [3, 2, 3, 1] 391s -0.226 0.303 0.621 391s -0.099 -0.238 0.203 391s test-mul-mat0: iter:282/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.322 -0.240 -0.666 0.786 391s -0.489 0.765 -0.160 -0.583 391s 0.014 0.750 -0.757 -0.980 391s 0.685 -0.574 -0.216 0.638 391s 391s x1: [4, 4, 1, 1] 391s -0.235 0.952 0.760 0.273 391s -0.854 0.513 -0.608 0.497 391s 0.455 0.031 -0.479 0.360 391s -0.829 0.892 -0.919 -0.959 391s 391s y: [4, 4, 1, 1] 391s -0.596 0.564 -0.132 -0.698 391s 0.397 0.618 0.346 -0.432 391s 0.741 -0.332 0.039 0.627 391s -0.623 1.793 2.293 -1.494 391s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.354 0.217 -0.668 0.737 391s 0.877 -0.187 0.342 -0.812 391s 391s x1: [4, 4, 4, 1] 391s -0.814 -0.943 0.783 0.700 391s -0.551 0.280 0.155 0.480 391s 0.801 -0.486 0.651 0.693 391s -0.405 0.692 -0.244 0.917 391s 391s y: [2, 4, 4, 1] 391s 0.075 -0.837 391s 0.506 -0.873 391s -0.313 0.454 391s 1.132 -1.312 391s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 3, 4, 1] 391s -0.907 0.992 -0.641 -0.708 391s -0.191 0.532 -0.967 0.817 391s -0.379 0.433 0.473 0.644 391s 391s x1: [4, 2, 4, 1] 391s -0.249 0.374 -0.143 -0.372 391s -0.813 -0.801 -0.183 0.173 391s 391s y: [3, 2, 4, 1] 391s 0.951 0.082 -0.050 391s -0.062 0.048 -0.014 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.708 0.429 -0.317 -0.442 391s -0.643 0.397 -0.530 -0.024 391s -0.024 0.278 0.398 -0.009 391s 391s x1: [4, 3, 1, 1] 391s 0.866 -0.870 -0.769 -0.685 391s -0.235 0.534 0.841 0.156 391s -0.047 0.980 -0.655 0.608 391s 391s y: [3, 3, 1, 1] 391s 0.785 -0.478 -0.563 391s -0.272 -0.087 0.488 391s 0.326 0.751 0.008 391s testing: mul_mat, [4, 4, 4, 1] = [3, 4, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.499 -0.621 -0.875 391s 0.189 0.226 0.054 391s -0.203 -0.155 -0.316 391s -0.966 -0.748 0.999 391s 391s x1: [3, 4, 4, 1] 391s -0.923 0.796 0.778 391s 0.073 0.657 -0.651 391s -0.061 0.787 -0.420 391s 0.254 -0.448 -0.886 391s 391s y: [4, 4, 4, 1] 391s -0.715 0.047 -0.182 1.074 391s 0.125 0.127 0.089 -1.213 391s -0.091 0.144 0.023 -0.949 391s 0.927 -0.101 0.298 -0.795 391s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.433 -0.175 0.569 0.912 391s 0.625 0.401 0.374 0.340 391s 0.801 0.889 0.349 0.860 391s -0.316 0.510 0.536 -0.205 391s 391s x1: [4, 3, 4, 1] 391s -0.848 -0.643 -0.654 -0.734 391s -0.178 0.413 -0.953 0.308 391s 0.349 0.203 -0.118 0.230 391s 391s y: [4, 3, 4, 1] 391s -0.562 -1.283 -2.111 -0.260 391s -0.257 -0.198 0.156 -0.307 391s -0.044 0.334 0.617 -0.118 391s test-mul-mat0: iter:283/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.146 0.274 -0.020 -0.332 391s 0.313 0.196 0.181 0.703 391s 0.511 0.248 -0.648 -0.560 391s 391s x1: [4, 3, 1, 1] 391s 0.686 0.462 -0.456 0.754 391s 0.031 0.464 0.735 0.941 391s -0.757 -0.490 -0.553 -0.971 391s 391s y: [3, 3, 1, 1] 391s -0.215 0.753 0.338 391s -0.205 0.895 -0.872 391s 0.310 -1.116 0.394 391s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 4, 4, 1] 391s 0.006 -0.473 -0.311 -0.525 391s 0.059 -0.501 0.139 -0.228 391s -0.989 0.314 0.734 0.876 391s 0.929 -0.870 -0.375 -0.673 391s 391s x1: [4, 3, 4, 1] 391s 0.435 -0.115 0.783 -0.573 391s -0.872 -0.057 -0.887 0.590 391s 0.486 0.867 -0.378 -0.050 391s 391s y: [4, 3, 4, 1] 391s 0.114 0.323 -0.393 0.596 391s -0.012 -0.280 0.710 -0.825 391s -0.263 -0.447 -0.530 -0.127 391s testing: mul_mat, [3, 4, 4, 3] = [4, 3, 4, 3] * [4, 4, 4, 3] 391s x0: [4, 3, 4, 3] 391s -0.989 -0.880 -0.796 -0.862 391s 0.779 -0.126 0.909 -0.603 391s 0.978 -0.226 -0.402 0.067 391s 391s x1: [4, 4, 4, 3] 391s 0.042 0.573 -0.390 0.664 391s 0.810 -0.658 -0.703 0.361 391s 0.807 -0.194 -0.159 0.403 391s 0.940 -0.353 0.953 -0.424 391s 391s y: [3, 4, 4, 3] 391s -0.807 -0.795 0.112 391s 0.027 -0.143 1.247 391s -0.848 0.265 0.924 391s -1.012 1.898 0.587 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.030 -0.922 -0.367 -0.929 391s -0.114 -0.211 -0.806 0.268 391s -0.696 0.400 0.354 0.299 391s 391s x1: [4, 3, 1, 1] 391s 0.092 0.260 -0.503 -0.537 391s 0.412 0.480 0.450 0.401 391s 0.641 0.524 0.815 -0.294 391s 391s y: [3, 3, 1, 1] 391s 0.446 0.196 -0.299 391s -0.968 -0.404 0.185 391s -0.489 -0.920 -0.036 391s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.142 -0.936 0.889 391s -0.239 -0.996 0.571 391s 391s x1: [3, 4, 4, 1] 391s -0.076 -0.807 -0.808 391s 0.416 -0.869 0.507 391s -0.492 0.391 -0.995 391s -0.029 -0.197 0.485 391s 391s y: [2, 4, 4, 1] 391s 0.048 0.361 391s 1.205 1.056 391s -1.181 -0.840 391s 0.619 0.480 391s testing: mul_mat, [4, 3, 4, 3] = [2, 4, 4, 3] * [2, 3, 4, 3] 391s x0: [2, 4, 4, 3] 391s -0.034 -0.021 391s 0.872 0.402 391s -0.062 -0.136 391s 0.890 0.327 391s 391s x1: [2, 3, 4, 3] 391s 0.231 -0.189 391s -0.443 -0.255 391s -0.172 -0.731 391s 391s y: [4, 3, 4, 3] 391s -0.004 0.125 0.011 0.144 391s 0.020 -0.488 0.062 -0.477 391s 0.021 -0.444 0.110 -0.392 391s test-mul-mat0: iter:284/500 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.513 -0.088 0.905 -0.930 391s -0.355 0.046 0.909 -0.601 391s 0.623 -0.768 0.896 0.333 391s 391s x1: [4, 4, 1, 1] 391s 0.866 0.300 -0.136 0.364 391s -0.627 -0.301 -0.153 -0.508 391s 0.161 -0.996 0.164 -0.650 391s 0.151 -0.072 0.973 0.369 391s 391s y: [3, 4, 1, 1] 391s -0.044 -0.636 0.308 391s 0.039 0.374 -0.466 391s 0.923 0.437 0.796 391s 0.621 0.606 1.144 391s testing: mul_mat, [1, 3, 2, 1] = [4, 1, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 1, 2, 1] 391s -0.741 0.890 0.952 -0.509 391s 391s x1: [4, 3, 2, 1] 391s 0.641 0.720 0.199 -0.059 391s -0.416 -0.437 0.314 0.283 391s 0.409 0.806 -0.556 0.413 391s 391s y: [1, 3, 2, 1] 391s 0.386 391s 0.074 391s -0.326 391s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 3, 2, 1] 391s 0.869 0.161 -0.389 -0.566 391s 0.059 0.378 -0.633 -0.563 391s 0.657 0.213 0.720 -0.993 391s 391s x1: [4, 1, 2, 1] 391s -0.516 0.072 -0.910 0.047 391s 391s y: [3, 1, 2, 1] 391s -0.109 0.546 -1.026 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.945 0.195 -0.187 0.158 391s 391s x1: [4, 3, 1, 1] 391s 0.977 -0.182 -0.654 0.350 391s 0.274 -0.463 0.167 0.144 391s 0.699 0.778 0.577 -0.243 391s 391s y: [1, 3, 1, 1] 391s 1.065 391s 0.161 391s 0.665 391s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.746 391s 391s x1: [1, 4, 2, 1] 391s -0.085 391s -0.180 391s 0.862 391s -0.035 391s 391s y: [1, 4, 2, 1] 391s 0.063 391s 0.135 391s -0.643 391s 0.026 391s testing: mul_mat, [4, 1, 2, 1] = [1, 4, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.299 391s -0.157 391s 0.034 391s -0.948 391s 391s x1: [1, 1, 2, 1] 391s -0.276 391s 391s y: [4, 1, 2, 1] 391s 0.082 0.043 -0.009 0.262 391s test-mul-mat0: iter:285/500 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.863 0.785 0.803 0.102 391s -0.130 -0.741 -0.931 -0.289 391s 0.096 -0.678 0.440 0.821 391s 391s x1: [4, 1, 1, 1] 391s -0.848 -0.216 0.223 -0.035 391s 391s y: [3, 1, 1, 1] 391s 0.737 0.073 0.134 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.967 -0.059 -0.807 0.255 391s 0.906 -0.865 0.393 0.691 391s 0.938 -0.505 -0.439 -0.803 391s 391s x1: [4, 3, 1, 1] 391s -0.653 -0.871 -0.478 0.191 391s 0.163 -0.426 0.309 -0.265 391s 0.792 -0.430 -0.995 -0.211 391s 391s y: [3, 3, 1, 1] 391s 1.117 0.107 -0.115 391s -0.449 0.454 0.446 391s 0.008 0.553 1.567 391s testing: mul_mat, [3, 3, 1, 2] = [4, 3, 1, 2] * [4, 3, 1, 2] 391s x0: [4, 3, 1, 2] 391s 0.424 0.329 0.978 0.918 391s 0.890 -0.825 -0.518 0.161 391s 0.468 0.367 0.873 -0.418 391s 391s x1: [4, 3, 1, 2] 391s -0.437 0.272 0.293 -0.114 391s -0.288 0.114 0.233 -0.160 391s 0.636 -0.576 -0.997 -0.790 391s 391s y: [3, 3, 1, 2] 391s 0.086 -0.783 0.199 391s -0.004 -0.497 0.177 391s -1.621 1.431 -0.454 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.174 -0.154 0.137 -0.425 391s -0.027 -0.081 0.818 0.729 391s -0.663 -0.206 -0.725 0.994 391s -0.806 0.559 0.564 0.634 391s 391s x1: [4, 3, 1, 1] 391s -0.228 -0.028 -0.047 -0.323 391s 0.003 -0.408 -0.283 -0.574 391s 0.921 -0.305 -0.655 0.811 391s 391s y: [4, 3, 1, 1] 391s 0.175 -0.265 -0.130 -0.063 391s 0.268 -0.617 -0.283 -0.754 391s -0.547 0.054 0.734 -0.769 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.700 -0.689 -0.516 0.756 391s 0.690 -0.381 0.870 -0.697 391s 391s x1: [4, 4, 1, 1] 391s 0.925 0.215 -0.594 -0.103 391s -0.831 0.083 0.900 -0.239 391s 0.800 -0.674 -0.318 -0.506 391s -0.330 -0.507 0.364 0.497 391s 391s y: [2, 4, 1, 1] 391s 0.728 0.111 391s -1.284 0.344 391s 0.806 0.885 391s 0.306 -0.065 391s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 391s x0: [2, 4, 1, 2] 391s -0.706 0.088 391s 0.592 -0.900 391s -0.365 0.358 391s 0.657 -0.221 391s 391s x1: [2, 4, 1, 2] 391s -0.855 0.320 391s -0.123 0.874 391s 0.315 -0.489 391s 0.799 -0.470 391s 391s y: [4, 4, 1, 2] 391s 0.632 -0.794 0.427 -0.633 391s 0.164 -0.859 0.358 -0.274 391s -0.265 0.626 -0.290 0.315 391s -0.606 0.896 -0.460 0.629 391s test-mul-mat0: iter:286/500 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.597 391s 391s x1: [1, 3, 1, 1] 391s -0.900 391s -0.102 391s 0.871 391s 391s y: [1, 3, 1, 1] 391s -0.537 391s -0.061 391s 0.520 391s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.198 391s 0.964 391s 391s x1: [1, 1, 4, 1] 391s 0.871 391s 391s y: [2, 1, 4, 1] 391s 0.173 0.840 391s testing: mul_mat, [1, 2, 4, 2] = [1, 1, 4, 2] * [1, 2, 4, 2] 391s x0: [1, 1, 4, 2] 391s -0.394 391s 391s x1: [1, 2, 4, 2] 391s 0.773 391s -0.655 391s 391s y: [1, 2, 4, 2] 391s -0.305 391s 0.258 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.002 391s 391s x1: [1, 1, 1, 1] 391s 0.864 391s 391s y: [1, 1, 1, 1] 391s -0.002 391s testing: mul_mat, [3, 1, 4, 1] = [1, 3, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 3, 4, 1] 391s -0.066 391s 0.668 391s 0.599 391s 391s x1: [1, 1, 4, 1] 391s -0.141 391s 391s y: [3, 1, 4, 1] 391s 0.009 -0.094 -0.085 391s testing: mul_mat, [2, 1, 4, 2] = [3, 2, 4, 2] * [3, 1, 4, 2] 391s x0: [3, 2, 4, 2] 391s -0.386 -0.835 0.236 391s 0.923 -0.909 0.626 391s 391s x1: [3, 1, 4, 2] 391s -0.514 0.184 0.713 391s 391s y: [2, 1, 4, 2] 391s 0.213 -0.195 391s test-mul-mat0: iter:287/500 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.891 -0.169 -0.562 391s -0.519 -0.169 0.077 391s 0.429 0.250 0.055 391s -0.953 0.015 0.728 391s 391s x1: [3, 3, 1, 1] 391s -0.298 0.744 -0.007 391s 0.887 -0.837 0.226 391s -0.423 0.829 -0.933 391s 391s y: [4, 3, 1, 1] 391s 0.144 0.028 0.057 0.290 391s -0.775 -0.301 0.184 -0.692 391s 0.760 0.007 -0.025 -0.264 391s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.701 0.839 -0.309 391s -0.083 -0.585 -0.981 391s -0.388 -0.580 -0.403 391s 391s x1: [3, 4, 4, 1] 391s 0.088 -0.539 -0.589 391s -0.834 0.302 -0.717 391s 0.589 -0.750 0.381 391s -0.709 0.994 -0.626 391s 391s y: [3, 4, 4, 1] 391s -0.333 0.885 0.515 391s 1.060 0.595 0.437 391s -1.160 0.016 0.053 391s 1.525 0.092 -0.049 391s testing: mul_mat, [4, 3, 4, 3] = [3, 4, 4, 3] * [3, 3, 4, 3] 391s x0: [3, 4, 4, 3] 391s 0.807 0.259 0.804 391s -0.953 0.544 0.474 391s -0.139 0.825 0.521 391s -0.745 -0.445 -0.579 391s 391s x1: [3, 3, 4, 3] 391s -0.186 -0.684 -0.018 391s 0.411 0.678 0.455 391s 0.601 0.519 0.068 391s 391s y: [4, 3, 4, 3] 391s -0.342 -0.203 -0.548 0.454 391s 0.873 0.193 0.740 -0.872 391s 0.675 -0.258 0.380 -0.719 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.562 -0.915 -0.587 391s -0.252 -0.915 0.084 391s 391s x1: [3, 4, 1, 1] 391s 0.326 0.604 -0.644 391s 0.554 -0.428 -0.728 391s -0.260 -0.288 0.390 391s -0.015 0.506 0.108 391s 391s y: [2, 4, 1, 1] 391s 0.009 -0.689 391s 1.130 0.191 391s -0.111 0.362 391s -0.535 -0.451 391s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.723 0.017 391s -0.396 -0.340 391s 0.576 -0.795 391s 0.807 -0.736 391s 391s x1: [2, 3, 4, 1] 391s -0.602 0.246 391s 0.811 -0.635 391s -0.542 -0.496 391s 391s y: [4, 3, 4, 1] 391s -0.431 0.155 -0.543 -0.668 391s 0.575 -0.105 0.972 1.123 391s -0.400 0.383 0.082 -0.072 391s testing: mul_mat, [1, 2, 4, 3] = [4, 1, 4, 3] * [4, 2, 4, 3] 391s x0: [4, 1, 4, 3] 391s 0.983 -0.582 0.198 0.785 391s 391s x1: [4, 2, 4, 3] 391s 0.968 -0.341 -0.730 -0.015 391s 0.319 -0.525 0.249 0.791 391s 391s y: [1, 2, 4, 3] 391s 0.995 391s 1.289 391s test-mul-mat0: iter:288/500 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.748 391s -0.609 391s -0.874 391s 391s x1: [1, 2, 1, 1] 391s -0.301 391s 0.461 391s 391s y: [3, 2, 1, 1] 391s -0.225 0.184 0.263 391s 0.345 -0.281 -0.403 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.973 391s 391s x1: [1, 3, 1, 1] 391s -0.455 391s 0.190 391s 0.140 391s 391s y: [1, 3, 1, 1] 391s -0.443 391s 0.185 391s 0.137 391s testing: mul_mat, [1, 1, 1, 2] = [1, 1, 1, 2] * [1, 1, 1, 2] 391s x0: [1, 1, 1, 2] 391s 0.160 391s 391s x1: [1, 1, 1, 2] 391s -0.427 391s 391s y: [1, 1, 1, 2] 391s -0.069 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.933 391s 391s x1: [1, 1, 1, 1] 391s 0.148 391s 391s y: [1, 1, 1, 1] 391s 0.138 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.626 391s 391s x1: [1, 1, 1, 1] 391s -0.143 391s 391s y: [1, 1, 1, 1] 391s -0.090 391s testing: mul_mat, [4, 1, 1, 2] = [1, 4, 1, 2] * [1, 1, 1, 2] 391s x0: [1, 4, 1, 2] 391s 0.288 391s 0.045 391s -0.183 391s -0.424 391s 391s x1: [1, 1, 1, 2] 391s -0.793 391s 391s y: [4, 1, 1, 2] 391s -0.228 -0.036 0.145 0.336 391s test-mul-mat0: iter:289/500 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.736 0.802 -0.213 0.425 391s -0.553 0.935 0.191 -0.620 391s -0.208 0.356 -0.993 -0.001 391s 0.478 -0.342 -0.713 -0.477 391s 391s x1: [4, 1, 1, 1] 391s 0.567 -0.727 0.692 0.428 391s 391s y: [4, 1, 1, 1] 391s -0.966 -1.127 -1.064 -0.178 391s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 3, 3, 1] 391s 0.470 -0.747 -0.723 0.216 391s 0.038 0.838 -0.683 0.828 391s -0.601 -0.103 0.277 0.246 391s 391s x1: [4, 4, 3, 1] 391s 0.475 -0.137 0.753 -0.802 391s -0.862 -0.555 -0.377 0.160 391s 0.527 0.174 -0.272 0.800 391s -0.134 -0.844 -0.953 0.130 391s 391s y: [3, 4, 3, 1] 391s -0.392 -1.275 -0.260 391s 0.317 -0.108 0.511 391s 0.487 1.015 -0.214 391s 1.285 0.046 -0.064 391s testing: mul_mat, [3, 3, 3, 1] = [4, 3, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.921 0.114 0.548 -0.414 391s -0.447 -0.239 0.900 -0.855 391s 0.043 0.028 -0.282 0.310 391s 391s x1: [4, 3, 3, 1] 391s 0.905 0.689 -0.587 -0.696 391s -0.414 0.690 0.550 -0.415 391s 0.891 -0.493 0.439 0.214 391s 391s y: [3, 3, 3, 1] 391s -0.788 -0.502 0.008 391s 0.933 0.870 -0.282 391s -0.725 -0.069 -0.033 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.082 0.303 0.078 -0.182 391s -0.914 -0.979 -0.817 -0.127 391s -0.566 0.642 -0.207 -0.528 391s 391s x1: [4, 3, 1, 1] 391s 0.705 0.255 0.491 -0.252 391s -0.716 -0.791 -0.942 0.993 391s 0.108 0.406 0.828 -0.530 391s 391s y: [3, 3, 1, 1] 391s 0.219 -1.265 -0.204 391s -0.553 2.072 -0.431 391s 0.293 -1.106 0.308 391s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.125 0.879 -0.514 391s 0.948 -0.513 -0.808 391s 391s x1: [3, 4, 3, 1] 391s -0.786 0.642 0.266 391s 0.158 -0.002 -0.388 391s -0.136 -0.746 -0.896 391s 0.612 -0.463 -0.687 391s 391s y: [2, 4, 3, 1] 391s 0.527 -1.290 391s 0.178 0.464 391s -0.179 0.978 391s -0.130 1.373 391s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.600 0.632 391s -0.442 0.932 391s 0.358 -0.137 391s 391s x1: [2, 3, 3, 1] 391s 0.411 0.379 391s 0.662 -0.164 391s -0.983 -0.284 391s 391s y: [3, 3, 3, 1] 391s -0.007 0.171 0.096 391s -0.501 -0.445 0.260 391s 0.410 0.169 -0.313 391s test-mul-mat0: iter:290/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.250 391s 0.209 391s 391s x1: [1, 1, 1, 1] 391s 0.452 391s 391s y: [2, 1, 1, 1] 391s -0.113 0.095 391s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 4, 4, 1] 391s 0.811 391s -0.731 391s -0.095 391s -0.443 391s 391s x1: [1, 2, 4, 1] 391s 0.779 391s -0.697 391s 391s y: [4, 2, 4, 1] 391s 0.631 -0.569 -0.074 -0.345 391s -0.565 0.509 0.066 0.309 391s testing: mul_mat, [3, 4, 4, 3] = [1, 3, 4, 3] * [1, 4, 4, 3] 391s x0: [1, 3, 4, 3] 391s 0.872 391s 0.992 391s 0.447 391s 391s x1: [1, 4, 4, 3] 391s 0.748 391s 0.418 391s -0.733 391s -0.529 391s 391s y: [3, 4, 4, 3] 391s 0.652 0.742 0.334 391s 0.364 0.414 0.187 391s -0.639 -0.727 -0.327 391s -0.461 -0.524 -0.236 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.720 391s -0.770 391s 0.495 391s 391s x1: [1, 3, 1, 1] 391s 0.827 391s -0.516 391s -0.498 391s 391s y: [3, 3, 1, 1] 391s 0.595 -0.637 0.409 391s -0.372 0.397 -0.255 391s -0.358 0.383 -0.246 391s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 4, 4, 1] 391s 0.094 -0.634 0.305 391s 0.513 0.915 -0.737 391s 0.972 0.549 0.295 391s 0.386 0.210 0.131 391s 391s x1: [3, 1, 4, 1] 391s 0.237 0.014 -0.516 391s 391s y: [4, 1, 4, 1] 391s -0.144 0.515 0.086 0.027 391s testing: mul_mat, [3, 3, 4, 3] = [4, 3, 4, 3] * [4, 3, 4, 3] 391s x0: [4, 3, 4, 3] 391s -0.371 -0.522 -0.234 -0.063 391s 0.700 0.874 -0.654 -0.728 391s 0.569 0.441 -0.751 -0.549 391s 391s x1: [4, 3, 4, 3] 391s -0.070 0.853 0.335 -0.545 391s -0.142 0.703 0.531 0.250 391s 0.490 0.772 -0.931 -0.865 391s 391s y: [3, 3, 4, 3] 391s -0.464 0.874 0.385 391s -0.454 -0.015 -0.307 391s -0.313 2.256 1.793 391s test-mul-mat0: iter:291/500 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.941 391s 0.722 391s 0.535 391s 391s x1: [1, 4, 1, 1] 391s -0.504 391s 0.982 391s -0.843 391s 0.058 391s 391s y: [3, 4, 1, 1] 391s 0.475 -0.364 -0.270 391s -0.924 0.709 0.525 391s 0.793 -0.608 -0.451 391s -0.055 0.042 0.031 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.166 391s -0.573 391s 391s x1: [1, 3, 1, 1] 391s -0.136 391s 0.523 391s 0.169 391s 391s y: [2, 3, 1, 1] 391s -0.023 0.078 391s 0.087 -0.299 391s 0.028 -0.097 391s testing: mul_mat, [4, 2, 1, 3] = [1, 4, 1, 3] * [1, 2, 1, 3] 391s x0: [1, 4, 1, 3] 391s -0.115 391s 0.644 391s -0.660 391s -0.669 391s 391s x1: [1, 2, 1, 3] 391s 0.206 391s 0.275 391s 391s y: [4, 2, 1, 3] 391s -0.024 0.132 -0.136 -0.138 391s -0.032 0.177 -0.182 -0.184 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.210 391s 0.638 391s -0.235 391s -0.621 391s 391s x1: [1, 4, 1, 1] 391s -0.284 391s 0.520 391s 0.675 391s -0.226 391s 391s y: [4, 4, 1, 1] 391s -0.060 -0.181 0.067 0.176 391s 0.109 0.332 -0.122 -0.323 391s 0.142 0.431 -0.159 -0.419 391s -0.047 -0.144 0.053 0.140 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.898 0.163 0.264 0.948 391s 0.887 -0.067 0.982 -0.117 391s -0.952 0.241 0.073 0.438 391s 0.597 0.503 -0.120 -0.590 391s 391s x1: [4, 1, 1, 1] 391s 0.389 0.931 -0.194 -0.405 391s 391s y: [4, 1, 1, 1] 391s -0.632 0.140 -0.337 0.963 391s testing: mul_mat, [4, 4, 1, 3] = [4, 4, 1, 3] * [4, 4, 1, 3] 391s x0: [4, 4, 1, 3] 391s 0.613 -0.068 -0.308 -0.398 391s 0.828 -0.506 -0.421 0.901 391s 0.177 -0.035 -0.832 -0.883 391s 0.298 0.108 -0.672 -0.445 391s 391s x1: [4, 4, 1, 3] 391s -0.326 -0.846 0.930 -0.651 391s -0.072 0.172 0.559 -0.434 391s 0.936 0.938 0.955 0.867 391s -0.256 -0.450 0.073 -0.154 391s 391s y: [4, 4, 1, 3] 391s -0.169 -0.821 -0.227 -0.524 391s -0.055 -0.772 -0.101 -0.186 391s -0.129 0.679 -1.428 -0.647 391s -0.088 -0.154 0.046 -0.106 391s test-mul-mat0: iter:292/500 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.064 -0.932 391s -0.653 -0.560 391s -0.140 -0.344 391s 0.644 -0.172 391s 391s x1: [2, 2, 1, 1] 391s -0.271 -0.876 391s 0.914 -0.602 391s 391s y: [4, 2, 1, 1] 391s 0.799 0.667 0.339 -0.024 391s 0.619 -0.260 0.079 0.692 391s testing: mul_mat, [4, 4, 3, 1] = [2, 4, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.913 0.938 391s -0.912 -0.257 391s 0.582 -0.083 391s 0.006 0.362 391s 391s x1: [2, 4, 3, 1] 391s -0.737 0.780 391s 0.291 -0.165 391s 0.024 -0.554 391s 0.985 -0.564 391s 391s y: [4, 4, 3, 1] 391s 1.404 0.472 -0.494 0.278 391s -0.421 -0.223 0.183 -0.058 391s -0.541 0.121 0.060 -0.200 391s -1.428 -0.753 0.620 -0.198 391s testing: mul_mat, [4, 4, 3, 4] = [2, 4, 3, 4] * [2, 4, 3, 4] 391s x0: [2, 4, 3, 4] 391s -0.141 0.187 391s 0.441 0.614 391s 0.766 0.822 391s 0.569 -0.554 391s 391s x1: [2, 4, 3, 4] 391s -0.597 0.670 391s 0.031 0.563 391s 0.411 -0.472 391s -0.208 0.498 391s 391s y: [4, 4, 3, 4] 391s 0.210 0.148 0.093 -0.711 391s 0.101 0.359 0.486 -0.295 391s -0.147 -0.109 -0.073 0.495 391s 0.123 0.214 0.249 -0.395 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.746 0.280 391s -0.656 0.051 391s -0.719 0.103 391s 391s x1: [2, 4, 1, 1] 391s 0.367 -0.899 391s -0.479 0.882 391s 0.712 -0.643 391s -0.293 0.749 391s 391s y: [3, 4, 1, 1] 391s 0.022 -0.286 -0.357 391s -0.111 0.359 0.436 391s 0.351 -0.500 -0.579 391s -0.009 0.230 0.288 391s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.334 0.636 0.768 391s 0.887 0.487 0.066 391s 0.402 0.787 0.031 391s -0.959 0.980 -0.129 391s 391s x1: [3, 2, 3, 1] 391s -0.109 0.948 -0.293 391s -0.537 0.551 0.095 391s 391s y: [4, 2, 3, 1] 391s 0.341 0.345 0.693 1.072 391s 0.244 -0.202 0.221 1.044 391s testing: mul_mat, [4, 3, 3, 4] = [4, 4, 3, 4] * [4, 3, 3, 4] 391s x0: [4, 4, 3, 4] 391s -0.056 -0.377 -0.830 -0.293 391s 0.243 0.742 0.676 -0.702 391s 0.059 0.340 0.741 -0.381 391s 0.968 0.021 -0.040 -0.481 391s 391s x1: [4, 3, 3, 4] 391s -0.187 0.218 -0.789 -0.420 391s -0.716 0.242 0.452 -0.759 391s -0.779 0.030 -0.339 -0.249 391s 391s y: [4, 3, 3, 4] 391s 0.706 -0.123 -0.362 0.057 391s -0.204 0.844 0.664 -0.340 391s 0.387 -0.222 -0.192 -0.620 391s test-mul-mat0: iter:293/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.898 -0.354 0.184 -0.986 391s 0.842 0.098 0.575 -0.074 391s 391s x1: [4, 1, 1, 1] 391s 0.265 0.012 -0.370 0.050 391s 391s y: [2, 1, 1, 1] 391s 0.116 0.008 391s testing: mul_mat, [4, 2, 2, 1] = [4, 4, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 4, 2, 1] 391s -0.453 0.088 0.243 0.177 391s -0.862 0.387 0.075 -0.216 391s -0.429 0.089 -0.375 0.668 391s -0.336 0.551 -0.331 -0.611 391s 391s x1: [4, 2, 2, 1] 391s 0.001 0.726 0.299 -0.213 391s 0.557 -0.703 0.690 -0.876 391s 391s y: [4, 2, 2, 1] 391s 0.098 0.349 -0.190 0.431 391s -0.301 -0.512 -1.145 -0.268 391s testing: mul_mat, [3, 4, 2, 4] = [4, 3, 2, 4] * [4, 4, 2, 4] 391s x0: [4, 3, 2, 4] 391s 0.097 0.836 0.193 -0.264 391s -0.794 -0.780 0.836 0.277 391s -0.919 -0.010 -0.580 0.903 391s 391s x1: [4, 4, 2, 4] 391s -0.152 -0.140 -0.509 -0.015 391s -0.753 0.566 0.769 -0.182 391s -0.345 -0.605 -0.514 0.319 391s 0.946 0.155 0.708 -0.204 391s 391s y: [3, 4, 2, 4] 391s -0.226 -0.200 0.423 391s 0.597 0.750 0.076 391s -0.723 0.405 0.909 391s 0.412 -0.337 -1.465 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.956 0.437 -0.016 0.840 391s -0.835 0.913 0.858 -0.176 391s -0.757 -0.034 -0.614 0.764 391s 391s x1: [4, 3, 1, 1] 391s -0.522 -0.771 0.524 0.495 391s -0.895 -0.437 -0.551 -0.863 391s -0.254 0.785 0.273 -0.357 391s 391s y: [3, 3, 1, 1] 391s -0.429 0.094 0.478 391s -1.763 0.028 0.371 391s -0.205 1.226 -0.274 391s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.430 -0.196 0.783 391s 0.981 -0.063 -0.753 391s 0.174 -0.445 0.337 391s 0.270 -0.972 -0.649 391s 391s x1: [3, 4, 2, 1] 391s 0.285 0.224 -0.155 391s 0.691 0.309 -0.431 391s -0.831 0.538 -0.908 391s 0.664 0.643 -0.344 391s 391s y: [4, 4, 2, 1] 391s -0.042 0.382 -0.102 -0.040 391s -0.101 0.983 -0.163 0.166 391s -1.173 -0.165 -0.690 -0.159 391s -0.109 0.871 -0.287 -0.223 391s testing: mul_mat, [2, 3, 2, 4] = [4, 2, 2, 4] * [4, 3, 2, 4] 391s x0: [4, 2, 2, 4] 391s 0.019 0.853 0.323 -0.422 391s 0.288 0.303 -0.621 -0.369 391s 391s x1: [4, 3, 2, 4] 391s 0.670 -0.638 -0.244 0.492 391s -0.192 -0.121 -0.720 -0.762 391s -0.141 0.454 0.508 0.664 391s 391s y: [2, 3, 2, 4] 391s -0.819 -0.030 391s -0.018 0.636 391s 0.268 -0.463 391s test-mul-mat0: iter:294/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.196 -0.091 -0.228 -0.306 391s -0.317 0.719 -0.382 0.605 391s 0.181 -0.207 -0.460 -0.527 391s -0.762 -0.224 0.448 -0.744 391s 391s x1: [4, 3, 1, 1] 391s 0.168 -0.886 -0.616 0.225 391s -0.725 -0.524 0.882 0.610 391s 0.851 -0.077 -0.711 0.095 391s 391s y: [4, 3, 1, 1] 391s 0.119 -0.319 0.379 -0.373 391s -0.198 -0.115 -0.750 0.612 391s -0.026 0.003 0.448 -1.020 391s testing: mul_mat, [4, 4, 2, 1] = [4, 4, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 4, 2, 1] 391s -0.249 -0.813 0.430 0.399 391s -0.789 -0.226 0.930 0.845 391s 0.970 0.384 0.768 -0.740 391s 0.865 -0.505 0.460 0.119 391s 391s x1: [4, 4, 2, 1] 391s 0.635 -0.515 0.424 0.749 391s -0.131 -0.351 -0.976 0.345 391s -0.469 0.634 0.196 0.454 391s 0.923 -0.710 -0.519 -0.273 391s 391s y: [4, 4, 2, 1] 391s 0.742 0.642 0.189 1.093 391s 0.037 -0.434 -1.266 -0.344 391s -0.133 0.792 -0.397 -0.582 391s 0.015 -1.282 0.426 0.887 391s testing: mul_mat, [3, 4, 2, 4] = [4, 3, 2, 4] * [4, 4, 2, 4] 391s x0: [4, 3, 2, 4] 391s 0.618 0.541 0.273 -0.822 391s -0.362 0.829 -0.409 0.872 391s -0.910 -0.494 0.761 0.788 391s 391s x1: [4, 4, 2, 4] 391s -0.718 -0.061 0.336 -0.507 391s 0.714 0.266 -0.662 0.684 391s -0.349 -0.895 0.944 -0.484 391s -0.400 0.404 0.635 -0.483 391s 391s y: [3, 4, 2, 4] 391s 0.032 -0.370 0.539 391s -0.158 0.829 -0.746 391s -0.044 -1.423 1.097 391s 0.542 -0.202 0.267 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.836 -0.091 0.239 -0.548 391s -0.411 -0.755 0.826 0.632 391s -0.430 0.539 0.846 0.262 391s 391s x1: [4, 3, 1, 1] 391s -0.956 0.017 0.623 -0.170 391s -0.471 0.121 0.034 -0.268 391s -0.195 0.952 -0.076 -0.092 391s 391s y: [3, 3, 1, 1] 391s -0.558 0.787 0.903 391s -0.250 -0.039 0.227 391s -0.217 -0.760 0.508 391s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 3, 2, 1] 391s -0.838 -0.386 0.686 391s -0.367 -0.736 -0.194 391s -0.053 -0.791 0.202 391s 391s x1: [3, 4, 2, 1] 391s 0.072 -0.458 -0.007 391s 0.244 -0.024 0.516 391s 0.288 0.993 0.139 391s -0.882 -0.478 -0.740 391s 391s y: [3, 4, 2, 1] 391s 0.112 0.312 0.357 391s 0.159 -0.172 0.110 391s -0.530 -0.863 -0.772 391s 0.417 0.819 0.276 391s testing: mul_mat, [1, 3, 2, 4] = [3, 1, 2, 4] * [3, 3, 2, 4] 391s x0: [3, 1, 2, 4] 391s 0.945 -0.811 0.013 391s 391s x1: [3, 3, 2, 4] 391s 0.486 0.551 0.182 391s -0.410 0.728 0.155 391s 0.581 0.742 0.717 391s 391s y: [1, 3, 2, 4] 391s 0.015 391s -0.976 391s -0.043 391s test-mul-mat0: iter:295/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.630 391s 391s x1: [1, 4, 1, 1] 391s -0.858 391s 0.718 391s -0.383 391s -0.913 391s 391s y: [1, 4, 1, 1] 391s -0.541 391s 0.452 391s -0.241 391s -0.575 391s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.793 391s -0.119 391s 391s x1: [1, 1, 3, 1] 391s 0.059 391s 391s y: [2, 1, 3, 1] 391s -0.046 -0.007 391s testing: mul_mat, [4, 2, 3, 3] = [1, 4, 3, 3] * [1, 2, 3, 3] 391s x0: [1, 4, 3, 3] 391s -0.767 391s -0.590 391s 0.132 391s 0.170 391s 391s x1: [1, 2, 3, 3] 391s -0.406 391s 0.203 391s 391s y: [4, 2, 3, 3] 391s 0.312 0.240 -0.054 -0.069 391s -0.156 -0.120 0.027 0.034 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.642 391s 0.099 391s -0.825 391s 391s x1: [1, 4, 1, 1] 391s 0.364 391s -0.813 391s 0.375 391s 0.552 391s 391s y: [3, 4, 1, 1] 391s 0.234 0.036 -0.301 391s -0.522 -0.080 0.670 391s 0.241 0.037 -0.309 391s 0.354 0.055 -0.455 391s testing: mul_mat, [1, 1, 3, 1] = [3, 1, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.820 -0.119 0.386 391s 391s x1: [3, 1, 3, 1] 391s 0.111 0.260 -0.330 391s 391s y: [1, 1, 3, 1] 391s -0.067 391s testing: mul_mat, [4, 3, 3, 3] = [1, 4, 3, 3] * [1, 3, 3, 3] 391s x0: [1, 4, 3, 3] 391s 0.413 391s -0.556 391s -0.748 391s -0.398 391s 391s x1: [1, 3, 3, 3] 391s 0.189 391s 0.801 391s 0.245 391s 391s y: [4, 3, 3, 3] 391s 0.078 -0.105 -0.141 -0.075 391s 0.331 -0.445 -0.599 -0.319 391s 0.101 -0.136 -0.183 -0.097 391s test-mul-mat0: iter:296/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.483 -0.463 391s -0.683 0.380 391s 0.552 -0.737 391s 391s x1: [2, 2, 1, 1] 391s -0.758 0.347 391s 0.748 0.027 391s 391s y: [3, 2, 1, 1] 391s -0.527 0.649 -0.674 391s 0.349 -0.500 0.393 391s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.890 0.171 391s -0.830 0.490 391s -0.277 -0.566 391s -0.466 -0.376 391s 391s x1: [2, 3, 4, 1] 391s 0.044 0.901 391s 0.982 -0.205 391s 0.886 0.293 391s 391s y: [4, 3, 4, 1] 391s 0.115 0.405 -0.522 -0.359 391s -0.909 -0.915 -0.156 -0.381 391s -0.739 -0.591 -0.412 -0.524 391s testing: mul_mat, [2, 4, 4, 4] = [2, 2, 4, 4] * [2, 4, 4, 4] 391s x0: [2, 2, 4, 4] 391s -0.386 0.511 391s 0.874 0.160 391s 391s x1: [2, 4, 4, 4] 391s -0.651 0.456 391s -0.594 0.071 391s 0.890 -0.060 391s 0.695 -0.694 391s 391s y: [2, 4, 4, 4] 391s 0.485 -0.496 391s 0.266 -0.507 391s -0.374 0.768 391s -0.623 0.496 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.746 0.722 391s -0.987 0.110 391s 391s x1: [2, 2, 1, 1] 391s -0.220 -0.358 391s 0.608 -0.111 391s 391s y: [2, 2, 1, 1] 391s -0.422 0.177 391s 0.374 -0.612 391s testing: mul_mat, [1, 2, 4, 1] = [2, 1, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.361 -0.905 391s 391s x1: [2, 2, 4, 1] 391s -0.089 0.593 391s -0.028 -0.540 391s 391s y: [1, 2, 4, 1] 391s -0.504 391s 0.499 391s testing: mul_mat, [4, 2, 4, 4] = [1, 4, 4, 4] * [1, 2, 4, 4] 391s x0: [1, 4, 4, 4] 391s 0.797 391s -0.400 391s -0.502 391s 0.828 391s 391s x1: [1, 2, 4, 4] 391s 0.686 391s 0.453 391s 391s y: [4, 2, 4, 4] 391s 0.547 -0.275 -0.345 0.568 391s 0.361 -0.181 -0.228 0.375 391s test-mul-mat0: iter:297/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.086 391s 0.546 391s 0.747 391s -0.375 391s 391s x1: [1, 1, 1, 1] 391s -0.532 391s 391s y: [4, 1, 1, 1] 391s 0.046 -0.290 -0.397 0.199 391s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 3, 3, 1] 391s -0.522 391s 0.669 391s 0.102 391s 391s x1: [1, 4, 3, 1] 391s 0.885 391s 0.618 391s -0.226 391s -0.327 391s 391s y: [3, 4, 3, 1] 391s -0.462 0.592 0.090 391s -0.323 0.413 0.063 391s 0.118 -0.151 -0.023 391s 0.171 -0.219 -0.033 391s testing: mul_mat, [1, 3, 3, 2] = [1, 1, 3, 2] * [1, 3, 3, 2] 391s x0: [1, 1, 3, 2] 391s 0.416 391s 391s x1: [1, 3, 3, 2] 391s -0.257 391s 0.462 391s 0.854 391s 391s y: [1, 3, 3, 2] 391s -0.107 391s 0.192 391s 0.355 391s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.327 391s -0.217 391s 0.464 391s 391s x1: [1, 1, 1, 1] 391s -0.401 391s 391s y: [3, 1, 1, 1] 391s 0.131 0.087 -0.186 391s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.863 0.005 -0.831 391s 0.968 0.452 -0.963 391s 391s x1: [3, 1, 3, 1] 391s -0.176 0.526 -0.074 391s 391s y: [2, 1, 3, 1] 391s 0.217 0.139 391s testing: mul_mat, [1, 3, 3, 2] = [2, 1, 3, 2] * [2, 3, 3, 2] 391s x0: [2, 1, 3, 2] 391s -0.280 0.621 391s 391s x1: [2, 3, 3, 2] 391s -0.432 -0.428 391s 0.370 0.392 391s -0.902 -0.704 391s 391s y: [1, 3, 3, 2] 391s -0.145 391s 0.140 391s -0.185 391s test-mul-mat0: iter:298/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.031 -0.294 -0.975 391s 0.085 0.260 0.283 391s 0.773 0.183 0.933 391s -0.441 0.403 -0.347 391s 391s x1: [3, 1, 1, 1] 391s -0.460 -0.358 -0.941 391s 391s y: [4, 1, 1, 1] 391s 1.037 -0.399 -1.300 0.385 391s testing: mul_mat, [3, 4, 4, 1] = [3, 3, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.948 -0.744 -0.119 391s -0.984 0.215 -0.233 391s 0.367 0.948 0.836 391s 391s x1: [3, 4, 4, 1] 391s -0.820 0.487 -0.113 391s -0.870 0.395 0.474 391s -0.454 0.097 0.419 391s -0.121 -0.240 -0.242 391s 391s y: [3, 4, 4, 1] 391s -1.126 0.938 0.066 391s -1.175 0.830 0.452 391s -0.552 0.370 0.275 391s 0.093 0.123 -0.474 391s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 1, 4, 1] 391s -0.842 -0.217 -0.738 391s 391s x1: [3, 3, 4, 1] 391s -0.855 0.920 -0.446 391s 0.981 -0.428 0.793 391s -0.712 0.161 -0.495 391s 391s y: [1, 3, 4, 1] 391s 0.849 391s -1.318 391s 0.929 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.143 -0.656 0.194 391s -0.383 0.222 0.149 391s 391s x1: [3, 1, 1, 1] 391s -0.347 -0.011 -0.423 391s 391s y: [2, 1, 1, 1] 391s -0.025 0.068 391s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.990 0.542 391s -0.701 0.212 391s 391s x1: [2, 3, 4, 1] 391s -0.034 -0.677 391s -0.441 -0.145 391s 0.373 -0.477 391s 391s y: [2, 3, 4, 1] 391s -0.400 -0.120 391s -0.515 0.279 391s 0.111 -0.363 391s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.339 -0.970 391s -0.643 -0.464 391s -0.223 -0.407 391s 391s x1: [2, 2, 4, 1] 391s 0.962 -0.405 391s -0.761 0.097 391s 391s y: [3, 2, 4, 1] 391s 0.720 -0.430 -0.050 391s -0.353 0.444 0.130 391s test-mul-mat0: iter:299/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.270 -0.837 0.196 0.756 391s -0.923 -0.872 -0.340 -0.812 391s 0.394 0.585 0.602 0.261 391s 0.666 -0.998 -0.075 -0.108 391s 391s x1: [4, 4, 1, 1] 391s 0.665 0.226 0.684 0.004 391s 0.584 -0.539 0.034 -0.881 391s 0.053 -0.688 0.688 -0.165 391s 0.363 0.016 -0.198 -0.535 391s 391s y: [4, 4, 1, 1] 391s -0.231 -1.047 0.807 0.166 391s -0.365 0.635 -0.295 1.019 391s 0.572 0.450 -0.010 0.688 391s -0.554 0.153 -0.107 0.298 391s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.935 0.707 -0.842 0.919 391s 0.238 0.161 0.489 -0.629 391s 0.425 -0.275 0.506 -0.386 391s -0.147 -0.672 0.733 0.458 391s 391s x1: [4, 4, 3, 1] 391s 0.686 0.930 -0.524 -0.853 391s -0.037 -0.405 0.200 0.275 391s -0.717 -0.964 -0.362 0.299 391s -0.162 0.103 -0.714 0.568 391s 391s y: [4, 4, 3, 1] 391s 0.956 0.594 0.100 -1.500 391s -0.236 -0.149 0.091 0.550 391s -0.773 -0.691 -0.338 0.625 391s 1.045 -0.729 -0.678 -0.309 391s testing: mul_mat, [1, 4, 3, 4] = [4, 1, 3, 4] * [4, 4, 3, 4] 391s x0: [4, 1, 3, 4] 391s -0.789 0.030 0.427 -0.440 391s 391s x1: [4, 4, 3, 4] 391s 0.444 -0.642 -0.458 -0.384 391s -0.503 -0.789 0.783 0.418 391s -0.078 -0.951 0.407 -0.757 391s -0.321 0.945 0.500 0.018 391s 391s y: [1, 4, 3, 4] 391s -0.396 391s 0.523 391s 0.539 391s 0.487 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.557 0.240 0.862 -0.476 391s 0.408 -0.346 -0.307 -0.924 391s 0.536 0.421 -0.279 -0.199 391s -0.639 -0.535 -0.502 0.148 391s 391s x1: [4, 1, 1, 1] 391s -0.399 -0.756 -0.900 0.449 391s 391s y: [4, 1, 1, 1] 391s -0.949 -0.039 -0.370 1.178 391s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.117 -0.310 0.359 -0.101 391s 0.294 -0.765 -0.901 0.777 391s 0.609 -0.553 -0.002 -0.926 391s 0.318 -0.194 0.283 -0.250 391s 391s x1: [4, 4, 3, 1] 391s -0.414 -0.133 0.050 -0.976 391s -0.467 -0.459 -0.909 -0.768 391s 0.763 -0.889 -0.166 -0.992 391s -0.789 -0.717 -0.426 -0.347 391s 391s y: [4, 4, 3, 1] 391s 0.206 -0.822 0.725 0.152 391s -0.052 0.436 0.682 -0.125 391s 0.227 0.284 1.876 0.617 391s 0.197 0.432 0.238 -0.146 391s testing: mul_mat, [1, 4, 3, 4] = [4, 1, 3, 4] * [4, 4, 3, 4] 391s x0: [4, 1, 3, 4] 391s 0.751 -0.612 -0.239 -0.475 391s 391s x1: [4, 4, 3, 4] 391s 0.001 0.100 0.999 0.956 391s -0.602 -0.523 0.302 0.857 391s -0.830 0.321 -0.946 0.732 391s 0.703 0.096 -0.823 0.379 391s 391s y: [1, 4, 3, 4] 391s -0.754 391s -0.612 391s -0.941 391s 0.486 391s test-mul-mat0: iter:300/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.468 0.785 391s -0.862 0.166 391s 0.120 -0.483 391s 391s x1: [2, 1, 1, 1] 391s -0.306 -0.046 391s 391s y: [3, 1, 1, 1] 391s -0.179 0.256 -0.015 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.315 0.618 391s 391s x1: [2, 3, 1, 1] 391s -0.774 -0.794 391s 0.284 -0.953 391s 0.816 -0.512 391s 391s y: [1, 3, 1, 1] 391s -0.735 391s -0.500 391s -0.059 391s testing: mul_mat, [1, 1, 1, 4] = [2, 1, 1, 4] * [2, 1, 1, 4] 391s x0: [2, 1, 1, 4] 391s 0.099 0.481 391s 391s x1: [2, 1, 1, 4] 391s -0.722 0.530 391s 391s y: [1, 1, 1, 4] 391s 0.184 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.549 -0.343 391s 0.093 -0.622 391s 0.036 0.083 391s 391s x1: [2, 1, 1, 1] 391s 0.331 -0.424 391s 391s y: [3, 1, 1, 1] 391s -0.036 0.295 -0.023 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.186 0.103 0.911 391s -0.199 0.359 -0.542 391s 0.265 -0.133 -0.651 391s 391s x1: [3, 2, 1, 1] 391s -0.527 -0.134 -0.539 391s 0.789 -0.516 -0.261 391s 391s y: [3, 2, 1, 1] 391s -0.407 0.348 0.229 391s -0.437 -0.201 0.447 391s testing: mul_mat, [2, 3, 1, 4] = [3, 2, 1, 4] * [3, 3, 1, 4] 391s x0: [3, 2, 1, 4] 391s -0.249 -0.783 0.116 391s 0.848 -0.419 -0.771 391s 391s x1: [3, 3, 1, 4] 391s -0.498 0.160 -0.476 391s -0.952 0.598 -0.833 391s 0.379 -0.825 -0.023 391s 391s y: [2, 3, 1, 4] 391s -0.057 -0.122 391s -0.328 -0.416 391s 0.549 0.685 391s test-mul-mat0: iter:301/500 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.608 0.891 0.712 -0.340 391s 0.068 -0.444 -0.263 0.342 391s 391s x1: [4, 3, 1, 1] 391s -0.491 0.393 0.238 0.260 391s 0.241 0.455 0.841 -0.643 391s 0.685 -0.091 0.119 0.009 391s 391s y: [2, 3, 1, 1] 391s 0.730 -0.181 391s 1.076 -0.627 391s -0.416 0.059 391s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.570 -0.590 0.193 0.549 391s 0.420 -0.871 0.941 0.310 391s 0.841 -0.399 -0.622 -0.604 391s 391s x1: [4, 2, 2, 1] 391s -0.268 0.726 -0.927 0.109 391s 0.537 0.719 -0.966 -0.327 391s 391s y: [3, 2, 2, 1] 391s -0.394 -1.583 -0.004 391s -1.095 -1.412 0.962 391s testing: mul_mat, [2, 3, 2, 1] = [4, 2, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 2, 2, 1] 391s 0.382 -0.155 -0.621 -0.343 391s 0.797 -0.318 0.113 0.581 391s 391s x1: [4, 3, 2, 1] 391s -0.544 0.784 0.574 -0.101 391s 0.857 -0.701 0.183 -0.712 391s -0.291 -0.624 0.837 -0.871 391s 391s y: [2, 3, 2, 1] 391s -0.651 -0.676 391s 0.567 0.513 391s -0.235 -0.444 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.436 0.521 0.517 0.370 391s -0.811 0.139 -0.744 0.744 391s 0.031 0.600 -0.837 -0.545 391s -0.782 -0.634 -0.801 0.382 391s 391s x1: [4, 2, 1, 1] 391s 0.562 0.393 -0.925 0.340 391s -0.167 -0.580 0.719 -0.350 391s 391s y: [4, 2, 1, 1] 391s 0.098 0.539 0.842 0.181 391s -0.132 -0.739 -0.765 -0.211 391s testing: mul_mat, [3, 4, 2, 1] = [4, 3, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.633 -0.485 -0.261 0.382 391s 0.223 0.692 0.502 0.031 391s 0.431 0.099 0.600 -0.249 391s 391s x1: [4, 4, 2, 1] 391s 0.182 0.010 -0.423 -0.093 391s -0.725 -0.852 -0.531 0.668 391s -0.777 0.810 -0.499 -0.357 391s 0.529 0.151 0.805 -0.035 391s 391s y: [3, 4, 2, 1] 391s -0.045 -0.168 -0.151 391s 1.266 -0.997 -0.882 391s 0.093 0.126 -0.465 391s -0.631 0.625 0.734 391s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.410 -0.396 -0.146 391s 0.603 -0.848 0.406 391s -0.417 -0.921 0.215 391s -0.101 -0.986 0.872 391s 391s x1: [3, 4, 2, 1] 391s 0.180 0.410 -0.298 391s 0.327 0.200 0.018 391s -0.517 0.568 -0.759 391s 0.914 -0.917 0.933 391s 391s y: [4, 4, 2, 1] 391s -0.045 -0.360 -0.517 -0.682 391s 0.052 0.034 -0.317 -0.215 391s -0.326 -1.102 -0.470 -1.170 391s 0.601 1.708 0.663 1.625 391s test-mul-mat0: iter:302/500 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.485 -0.889 -0.624 391s 391s x1: [3, 4, 1, 1] 391s -0.307 -0.950 -0.350 391s -0.897 0.653 0.233 391s 0.002 -0.743 0.385 391s 0.081 -0.729 -0.761 391s 391s y: [1, 4, 1, 1] 391s 0.914 391s -1.161 391s 0.422 391s 1.163 391s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 1, 2, 1] 391s 0.750 -0.575 -0.289 391s 391s x1: [3, 1, 2, 1] 391s -0.732 -0.776 -0.820 391s 391s y: [1, 1, 2, 1] 391s 0.134 391s testing: mul_mat, [3, 1, 2, 2] = [3, 3, 2, 2] * [3, 1, 2, 2] 391s x0: [3, 3, 2, 2] 391s -0.030 0.359 0.821 391s 0.081 0.735 -0.911 391s 0.304 0.915 0.222 391s 391s x1: [3, 1, 2, 2] 391s -0.894 0.463 -0.314 391s 391s y: [3, 1, 2, 2] 391s -0.065 0.555 0.082 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.479 -0.308 -0.836 391s 391s x1: [3, 3, 1, 1] 391s 0.399 0.562 0.891 391s -0.379 0.135 -0.198 391s -0.336 -0.541 -0.862 391s 391s y: [1, 3, 1, 1] 391s -0.726 391s -0.058 391s 0.726 391s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.686 391s 0.828 391s -0.870 391s -0.479 391s 391s x1: [1, 3, 2, 1] 391s -0.395 391s -0.292 391s 0.597 391s 391s y: [4, 3, 2, 1] 391s 0.271 -0.327 0.343 0.189 391s 0.200 -0.242 0.254 0.140 391s -0.409 0.494 -0.519 -0.285 391s testing: mul_mat, [4, 1, 2, 2] = [4, 4, 2, 2] * [4, 1, 2, 2] 391s x0: [4, 4, 2, 2] 391s 0.159 -0.390 -0.810 -0.468 391s -0.207 -0.940 -0.167 -0.898 391s 0.084 -0.689 0.149 -0.261 391s 0.997 -0.562 0.809 -0.512 391s 391s x1: [4, 1, 2, 2] 391s -0.070 0.957 -0.071 -0.671 391s 391s y: [4, 1, 2, 2] 391s -0.012 -0.269 -0.500 -0.322 391s test-mul-mat0: iter:303/500 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.159 391s 391s x1: [1, 3, 1, 1] 391s 0.049 391s -0.692 391s 0.630 391s 391s y: [1, 3, 1, 1] 391s 0.008 391s -0.110 391s 0.100 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.888 391s 0.785 391s -0.451 391s -0.905 391s 391s x1: [1, 1, 1, 1] 391s -0.662 391s 391s y: [4, 1, 1, 1] 391s 0.588 -0.520 0.298 0.599 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.232 391s -0.001 391s -0.327 391s 0.767 391s 391s x1: [1, 4, 1, 1] 391s -0.699 391s 0.559 391s 0.928 391s -0.166 391s 391s y: [4, 4, 1, 1] 391s -0.162 0.001 0.229 -0.536 391s 0.129 -0.001 -0.183 0.428 391s 0.215 -0.001 -0.303 0.711 391s -0.038 0.000 0.054 -0.127 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.044 391s 391s x1: [1, 4, 1, 1] 391s 0.828 391s -0.517 391s 0.985 391s -0.482 391s 391s y: [1, 4, 1, 1] 391s 0.037 391s -0.023 391s 0.043 391s -0.021 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.057 391s 391s x1: [1, 1, 1, 1] 391s 0.861 391s 391s y: [1, 1, 1, 1] 391s -0.049 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.991 391s 391s x1: [1, 1, 1, 1] 391s -0.532 391s 391s y: [1, 1, 1, 1] 391s -0.528 391s test-mul-mat0: iter:304/500 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.323 -0.930 0.498 391s 0.906 0.302 -0.503 391s -0.421 0.068 -0.674 391s 0.062 0.053 -0.156 391s 391s x1: [3, 2, 1, 1] 391s -0.422 -0.430 -0.952 391s -0.636 0.119 -0.858 391s 391s y: [4, 2, 1, 1] 391s 0.062 -0.033 0.790 0.100 391s -0.332 -0.110 0.854 0.101 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.376 0.932 -0.409 391s 391s x1: [3, 4, 1, 1] 391s -0.081 -0.903 -0.295 391s -0.202 0.040 0.173 391s 0.408 0.031 0.949 391s 0.649 -0.946 0.884 391s 391s y: [1, 4, 1, 1] 391s -0.690 391s 0.042 391s -0.513 391s -1.487 391s testing: mul_mat, [2, 1, 1, 2] = [3, 2, 1, 2] * [3, 1, 1, 2] 391s x0: [3, 2, 1, 2] 391s 0.446 0.251 -0.095 391s -0.485 0.576 0.967 391s 391s x1: [3, 1, 1, 2] 391s 0.742 -0.926 0.255 391s 391s y: [2, 1, 1, 2] 391s 0.075 -0.647 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.869 0.559 -0.425 391s -0.032 -0.844 -0.718 391s -0.587 -0.863 0.568 391s -0.937 -0.186 -0.098 391s 391s x1: [3, 2, 1, 1] 391s -0.296 -0.702 0.092 391s 0.735 -0.754 -0.259 391s 391s y: [4, 2, 1, 1] 391s -0.175 0.536 0.832 0.398 391s -0.950 0.799 0.071 -0.524 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.763 0.914 -0.822 -0.386 391s 0.437 -0.517 0.704 0.146 391s 391s x1: [4, 3, 1, 1] 391s 0.728 -0.182 0.807 -0.758 391s -0.606 0.774 0.810 -0.186 391s -0.340 0.475 0.939 -0.656 391s 391s y: [2, 3, 1, 1] 391s -1.093 0.869 391s 0.576 -0.122 391s 0.174 0.172 391s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 391s x0: [2, 4, 1, 2] 391s 0.114 -0.664 391s 0.991 -0.470 391s -0.649 0.290 391s 0.053 0.772 391s 391s x1: [2, 4, 1, 2] 391s -0.883 -0.323 391s -0.295 -0.728 391s -0.186 0.519 391s -0.152 0.096 391s 391s y: [4, 4, 1, 2] 391s 0.114 -0.723 0.480 -0.296 391s 0.450 0.050 -0.020 -0.578 391s -0.366 -0.428 0.271 0.391 391s -0.081 -0.196 0.127 0.066 391s test-mul-mat0: iter:305/500 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.979 0.742 391s 0.727 -0.988 391s -0.908 -0.220 391s 391s x1: [2, 4, 1, 1] 391s 0.384 -0.024 391s -0.592 0.134 391s 0.800 -0.687 391s -0.310 -0.134 391s 391s y: [3, 4, 1, 1] 391s -0.395 0.304 -0.343 391s 0.679 -0.562 0.508 391s -1.292 1.260 -0.574 391s 0.204 -0.093 0.311 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.213 -0.769 391s 0.166 0.478 391s 391s x1: [2, 3, 1, 1] 391s -0.653 -0.378 391s -0.930 -0.880 391s 0.066 0.844 391s 391s y: [2, 3, 1, 1] 391s 0.151 -0.289 391s 0.478 -0.575 391s -0.634 0.414 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.399 -0.849 391s -0.303 -0.468 391s 391s x1: [2, 2, 1, 1] 391s 0.296 -0.032 391s -0.986 0.176 391s 391s y: [2, 2, 1, 1] 391s 0.145 -0.075 391s -0.542 0.217 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.243 0.350 391s -0.442 0.255 391s 391s x1: [2, 2, 1, 1] 391s 0.951 0.010 391s 0.222 -0.183 391s 391s y: [2, 2, 1, 1] 391s 0.235 -0.417 391s -0.010 -0.145 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.722 -0.957 391s -0.661 0.792 391s 391s x1: [2, 2, 1, 1] 391s -0.349 -0.871 391s 0.602 0.273 391s 391s y: [2, 2, 1, 1] 391s 0.581 -0.458 391s 0.174 -0.182 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.926 0.044 391s 391s x1: [2, 2, 1, 1] 391s 0.552 -0.725 391s -0.042 0.030 391s 391s y: [1, 2, 1, 1] 391s 0.479 391s -0.037 391s test-mul-mat0: iter:306/500 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.440 -0.776 391s -0.107 -0.305 391s -0.126 0.023 391s 391s x1: [2, 3, 1, 1] 391s -0.089 -0.025 391s -0.151 0.921 391s -0.803 0.665 391s 391s y: [3, 3, 1, 1] 391s -0.019 0.017 0.011 391s -0.782 -0.265 0.040 391s -0.870 -0.117 0.116 391s testing: mul_mat, [1, 3, 4, 1] = [2, 1, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 1, 4, 1] 391s 0.165 0.493 391s 391s x1: [2, 3, 4, 1] 391s -0.703 -0.853 391s 0.223 -0.981 391s -0.514 0.266 391s 391s y: [1, 3, 4, 1] 391s -0.537 391s -0.447 391s 0.046 391s testing: mul_mat, [1, 1, 4, 1] = [2, 1, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.657 0.452 391s 391s x1: [2, 1, 4, 1] 391s -0.575 -0.824 391s 391s y: [1, 1, 4, 1] 391s 0.005 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.975 0.325 391s 0.832 0.392 391s 0.212 0.050 391s 0.190 0.693 391s 391s x1: [2, 1, 1, 1] 391s -0.700 -0.438 391s 391s y: [4, 1, 1, 1] 391s 0.541 -0.754 -0.170 -0.437 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.064 -0.733 -0.501 -0.727 391s 391s x1: [4, 2, 4, 1] 391s -0.406 0.208 -0.792 0.020 391s 0.384 0.209 -0.289 0.371 391s 391s y: [1, 2, 4, 1] 391s 0.204 391s -0.254 391s testing: mul_mat, [3, 4, 4, 1] = [1, 3, 4, 1] * [1, 4, 4, 1] 391s x0: [1, 3, 4, 1] 391s -0.518 391s -0.664 391s -0.901 391s 391s x1: [1, 4, 4, 1] 391s 0.137 391s 0.644 391s -0.566 391s 0.143 391s 391s y: [3, 4, 4, 1] 391s -0.071 -0.091 -0.123 391s -0.333 -0.427 -0.580 391s 0.293 0.376 0.510 391s -0.074 -0.095 -0.128 391s test-mul-mat0: iter:307/500 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.177 391s 391s x1: [1, 3, 1, 1] 391s -0.594 391s -0.803 391s -0.213 391s 391s y: [1, 3, 1, 1] 391s -0.105 391s -0.142 391s -0.038 391s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.348 391s -0.665 391s 391s x1: [1, 1, 2, 1] 391s 0.741 391s 391s y: [2, 1, 2, 1] 391s 0.258 -0.493 391s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 1, 2, 1] 391s 0.421 391s 391s x1: [1, 2, 2, 1] 391s -0.682 391s -0.617 391s 391s y: [1, 2, 2, 1] 391s -0.287 391s -0.260 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.505 391s 0.310 391s -0.335 391s 0.435 391s 391s x1: [1, 1, 1, 1] 391s 0.577 391s 391s y: [4, 1, 1, 1] 391s 0.291 0.179 -0.193 0.251 391s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 3, 2, 1] 391s 0.376 -0.157 0.178 0.495 391s 0.012 -0.668 -0.348 0.035 391s -0.493 0.854 -0.944 0.379 391s 391s x1: [4, 1, 2, 1] 391s -0.510 -0.417 -0.115 -0.538 391s 391s y: [3, 1, 2, 1] 391s -0.413 0.293 -0.200 391s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.950 0.760 0.849 391s -0.282 -0.561 -0.751 391s 391s x1: [3, 4, 2, 1] 391s 0.274 0.593 -0.874 391s 0.253 -0.539 0.961 391s -0.981 0.836 -0.027 391s -0.474 -0.321 0.305 391s 391s y: [2, 4, 2, 1] 391s -0.552 0.246 391s 0.164 -0.491 391s 1.545 -0.172 391s 0.466 0.085 391s test-mul-mat0: iter:308/500 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.671 0.627 391s 0.562 -0.856 391s 0.274 -0.692 391s -0.241 -0.630 391s 391s x1: [2, 2, 1, 1] 391s -0.582 -0.387 391s 0.892 -0.725 391s 391s y: [4, 2, 1, 1] 391s 0.148 0.004 0.108 0.384 391s -1.053 1.122 0.746 0.242 391s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.581 0.183 391s -0.852 0.278 391s -0.866 -0.229 391s 0.849 -0.472 391s 391s x1: [2, 4, 4, 1] 391s -0.615 -0.386 391s -0.580 0.103 391s -0.625 -0.141 391s -0.048 -0.377 391s 391s y: [4, 4, 4, 1] 391s 0.286 0.417 0.620 -0.340 391s 0.356 0.523 0.479 -0.541 391s 0.337 0.494 0.574 -0.464 391s -0.041 -0.063 0.128 0.137 391s testing: mul_mat, [4, 4, 4, 4] = [2, 4, 4, 4] * [2, 4, 4, 4] 391s x0: [2, 4, 4, 4] 391s -0.271 0.434 391s 0.102 0.837 391s -0.958 0.898 391s -0.137 0.050 391s 391s x1: [2, 4, 4, 4] 391s 0.164 0.611 391s 0.719 0.299 391s -0.618 0.569 391s 0.827 -0.848 391s 391s y: [4, 4, 4, 4] 391s 0.220 0.528 0.391 0.008 391s -0.065 0.323 -0.420 -0.084 391s 0.414 0.413 1.102 0.113 391s -0.592 -0.626 -1.553 -0.156 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.783 -0.013 391s 0.998 -0.202 391s -0.760 0.544 391s 0.913 0.515 391s 391s x1: [2, 4, 1, 1] 391s -0.936 -0.062 391s -0.867 0.094 391s -0.426 -0.305 391s -0.930 -0.293 391s 391s y: [4, 4, 1, 1] 391s -0.733 -0.922 0.678 -0.887 391s -0.680 -0.884 0.710 -0.742 391s -0.330 -0.363 0.158 -0.546 391s -0.725 -0.869 0.548 -1.000 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s -0.331 -0.223 0.838 -0.040 391s 391s x1: [4, 2, 4, 1] 391s -0.476 0.993 0.870 -0.272 391s -0.465 -0.708 0.511 0.096 391s 391s y: [1, 2, 4, 1] 391s 0.676 391s 0.736 391s testing: mul_mat, [4, 4, 4, 4] = [1, 4, 4, 4] * [1, 4, 4, 4] 391s x0: [1, 4, 4, 4] 391s 0.801 391s 0.727 391s 0.209 391s -0.603 391s 391s x1: [1, 4, 4, 4] 391s -0.699 391s -0.071 391s 0.618 391s -0.402 391s 391s y: [4, 4, 4, 4] 391s -0.560 -0.508 -0.146 0.421 391s -0.057 -0.051 -0.015 0.043 391s 0.495 0.450 0.129 -0.373 391s -0.322 -0.292 -0.084 0.242 391s test-mul-mat0: iter:309/500 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.054 0.839 0.459 391s 0.536 0.762 0.292 391s 0.775 -0.075 0.539 391s 391s x1: [3, 4, 1, 1] 391s 0.026 0.634 0.486 391s 0.168 0.521 0.051 391s 0.619 -0.811 -0.244 391s -0.625 0.335 0.782 391s 391s y: [3, 4, 1, 1] 391s 0.756 0.639 0.234 391s 0.470 0.502 0.118 391s -0.759 -0.357 0.409 391s 0.606 0.149 -0.087 391s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.882 0.497 0.734 391s 391s x1: [3, 3, 3, 1] 391s 0.868 -0.831 -0.830 391s 0.617 0.221 -0.606 391s 0.419 -0.162 0.895 391s 391s y: [1, 3, 3, 1] 391s -0.257 391s 0.209 391s 0.947 391s testing: mul_mat, [2, 1, 3, 4] = [3, 2, 3, 4] * [3, 1, 3, 4] 391s x0: [3, 2, 3, 4] 391s -0.572 0.713 0.941 391s -0.495 -0.914 0.137 391s 391s x1: [3, 1, 3, 4] 391s -0.214 0.725 -0.806 391s 391s y: [2, 1, 3, 4] 391s -0.119 -0.667 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.913 0.440 -0.409 391s 391s x1: [3, 2, 1, 1] 391s -0.764 -0.677 -0.150 391s 0.497 -0.321 0.376 391s 391s y: [1, 2, 1, 1] 391s -0.934 391s 0.159 391s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 3, 3, 1] 391s 0.720 391s 0.622 391s -0.809 391s 391s x1: [1, 3, 3, 1] 391s -0.261 391s -0.094 391s -0.810 391s 391s y: [3, 3, 3, 1] 391s -0.188 -0.162 0.211 391s -0.067 -0.058 0.076 391s -0.583 -0.504 0.656 391s testing: mul_mat, [2, 1, 3, 4] = [3, 2, 3, 4] * [3, 1, 3, 4] 391s x0: [3, 2, 3, 4] 391s 0.616 0.076 0.416 391s 0.914 -0.773 0.952 391s 391s x1: [3, 1, 3, 4] 391s 0.695 0.837 -0.427 391s 391s y: [2, 1, 3, 4] 391s 0.314 -0.418 391s test-mul-mat0: iter:310/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.350 0.356 0.966 0.485 391s 391s x1: [4, 3, 1, 1] 391s 0.153 -0.631 -0.307 -0.184 391s 0.418 -0.402 0.198 -0.929 391s 0.407 0.626 -0.525 -0.979 391s 391s y: [1, 3, 1, 1] 391s -0.557 391s -0.257 391s -0.617 391s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.570 -0.234 -0.550 -0.233 391s -0.163 0.857 -0.607 0.312 391s 391s x1: [4, 1, 4, 1] 391s 0.031 0.643 0.956 0.270 391s 391s y: [2, 1, 4, 1] 391s -0.722 0.050 391s testing: mul_mat, [1, 2, 4, 4] = [4, 1, 4, 4] * [4, 2, 4, 4] 391s x0: [4, 1, 4, 4] 391s 0.732 0.596 0.943 -0.435 391s 391s x1: [4, 2, 4, 4] 391s -0.859 0.107 -0.712 -0.022 391s -0.036 -0.319 -0.711 -0.159 391s 391s y: [1, 2, 4, 4] 391s -1.226 391s -0.817 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.421 0.288 -0.367 0.014 391s 0.326 0.122 0.581 0.905 391s 0.380 0.582 0.982 -0.792 391s 0.919 0.354 0.184 0.428 391s 391s x1: [4, 1, 1, 1] 391s -0.397 -0.209 -0.973 -0.514 391s 391s y: [4, 1, 1, 1] 391s 0.457 -1.186 -0.821 -0.838 391s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 1, 4, 1] 391s -0.622 -0.197 -0.169 0.887 391s 391s x1: [4, 4, 4, 1] 391s -0.566 -0.782 -0.536 0.414 391s 0.121 -0.924 0.813 -0.097 391s 0.995 0.227 0.506 -0.213 391s 0.254 0.992 -0.544 0.833 391s 391s y: [1, 4, 4, 1] 391s 0.963 391s -0.117 391s -0.938 391s 0.478 391s testing: mul_mat, [3, 4, 4, 4] = [1, 3, 4, 4] * [1, 4, 4, 4] 391s x0: [1, 3, 4, 4] 391s 0.266 391s -0.756 391s -0.094 391s 391s x1: [1, 4, 4, 4] 391s -0.822 391s 0.445 391s -0.079 391s -0.212 391s 391s y: [3, 4, 4, 4] 391s -0.218 0.622 0.077 391s 0.118 -0.337 -0.042 391s -0.021 0.060 0.007 391s -0.056 0.160 0.020 391s test-mul-mat0: iter:311/500 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.767 -0.313 391s 0.898 -0.896 391s 391s x1: [2, 3, 1, 1] 391s -0.487 -0.709 391s -0.921 -0.030 391s -0.304 0.246 391s 391s y: [2, 3, 1, 1] 391s -0.151 0.199 391s -0.697 -0.800 391s -0.311 -0.494 391s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 2, 4, 1] 391s -0.069 -0.171 391s -0.430 0.445 391s 391s x1: [2, 2, 4, 1] 391s -0.629 0.701 391s 0.647 0.536 391s 391s y: [2, 2, 4, 1] 391s -0.077 0.582 391s -0.136 -0.040 391s testing: mul_mat, [1, 2, 4, 3] = [2, 1, 4, 3] * [2, 2, 4, 3] 391s x0: [2, 1, 4, 3] 391s -0.301 0.319 391s 391s x1: [2, 2, 4, 3] 391s 0.333 -0.512 391s 0.962 0.411 391s 391s y: [1, 2, 4, 3] 391s -0.264 391s -0.158 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.619 -0.521 391s -0.142 0.557 391s 0.320 -0.360 391s 391s x1: [2, 1, 1, 1] 391s -0.224 0.962 391s 391s y: [3, 1, 1, 1] 391s -0.640 0.568 -0.418 391s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 1, 4, 1] 391s -0.836 0.645 0.723 391s 391s x1: [3, 2, 4, 1] 391s -0.996 -0.527 -0.607 391s -0.910 0.022 -0.838 391s 391s y: [1, 2, 4, 1] 391s 0.054 391s 0.169 391s testing: mul_mat, [3, 3, 4, 3] = [1, 3, 4, 3] * [1, 3, 4, 3] 391s x0: [1, 3, 4, 3] 391s 0.400 391s -0.062 391s -0.942 391s 391s x1: [1, 3, 4, 3] 391s -0.945 391s 0.213 391s -0.447 391s 391s y: [3, 3, 4, 3] 391s -0.378 0.058 0.891 391s 0.085 -0.013 -0.201 391s -0.179 0.028 0.422 391s test-mul-mat0: iter:312/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.008 -0.638 391s 391s x1: [2, 4, 1, 1] 391s 0.680 -0.807 391s 0.275 0.010 391s 0.336 0.767 391s 0.332 -0.490 391s 391s y: [1, 4, 1, 1] 391s 0.520 391s -0.004 391s -0.487 391s 0.315 391s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.990 0.272 391s -0.335 -0.317 391s 0.576 -0.276 391s 0.414 0.995 391s 391s x1: [2, 1, 3, 1] 391s 0.362 -0.206 391s 391s y: [4, 1, 3, 1] 391s -0.414 -0.056 0.265 -0.055 391s testing: mul_mat, [4, 4, 3, 3] = [2, 4, 3, 3] * [2, 4, 3, 3] 391s x0: [2, 4, 3, 3] 391s -0.223 0.376 391s -0.389 0.582 391s 0.596 0.387 391s -0.684 -0.789 391s 391s x1: [2, 4, 3, 3] 391s -0.079 0.437 391s 0.104 0.758 391s 0.047 0.473 391s 0.303 0.057 391s 391s y: [4, 4, 3, 3] 391s 0.182 0.285 0.122 -0.291 391s 0.262 0.401 0.355 -0.670 391s 0.167 0.257 0.211 -0.406 391s -0.046 -0.085 0.202 -0.252 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.726 -0.322 391s 0.145 0.353 391s 0.755 0.069 391s 0.150 0.498 391s 391s x1: [2, 4, 1, 1] 391s -0.152 0.416 391s -0.930 0.300 391s 0.082 0.926 391s -0.324 -0.375 391s 391s y: [4, 4, 1, 1] 391s -0.244 0.125 -0.086 0.184 391s -0.772 -0.029 -0.681 0.010 391s -0.239 0.339 0.126 0.473 391s -0.114 -0.179 -0.270 -0.235 391s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.252 -0.536 -0.187 0.725 391s 0.395 -0.271 0.424 0.712 391s 0.231 0.291 -0.730 -0.097 391s -0.395 -0.765 -0.375 -0.694 391s 391s x1: [4, 2, 3, 1] 391s -0.314 0.771 -0.916 -0.868 391s -0.965 0.812 -0.698 0.442 391s 391s y: [4, 2, 3, 1] 391s -0.950 -1.339 0.905 0.480 391s -0.227 -0.582 0.480 -0.284 391s testing: mul_mat, [4, 4, 3, 3] = [4, 4, 3, 3] * [4, 4, 3, 3] 391s x0: [4, 4, 3, 3] 391s -0.621 0.257 -0.734 0.247 391s -0.501 -0.646 -0.425 -0.995 391s 0.129 -0.255 -0.277 -0.573 391s -0.655 -0.722 -0.191 0.254 391s 391s x1: [4, 4, 3, 3] 391s 0.937 0.487 -0.236 0.076 391s 0.780 -0.339 -0.943 -0.144 391s 0.457 0.557 0.342 0.749 391s 0.302 -0.619 0.294 -0.559 391s 391s y: [4, 4, 3, 3] 391s -0.265 -0.759 0.019 -0.900 391s 0.085 0.372 0.531 -0.122 391s -0.206 -1.480 -0.607 -0.576 391s -0.701 0.680 0.435 0.051 391s test-mul-mat0: iter:313/500 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.902 0.166 391s 0.573 0.522 391s 391s x1: [2, 1, 1, 1] 391s -0.244 0.548 391s 391s y: [2, 1, 1, 1] 391s -0.129 0.146 391s testing: mul_mat, [3, 2, 3, 1] = [2, 3, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 3, 3, 1] 391s 0.248 0.364 391s -0.319 -0.599 391s 0.862 -0.136 391s 391s x1: [2, 2, 3, 1] 391s 0.147 0.061 391s -0.943 -0.177 391s 391s y: [3, 2, 3, 1] 391s 0.059 -0.083 0.118 391s -0.299 0.406 -0.789 391s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.179 0.550 391s 0.373 0.351 391s 391s x1: [2, 3, 3, 1] 391s -0.968 -0.250 391s -0.420 -0.145 391s -0.960 -0.280 391s 391s y: [2, 3, 3, 1] 391s -0.311 -0.449 391s -0.155 -0.208 391s -0.326 -0.456 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.298 -0.482 391s 0.150 -0.118 391s 391s x1: [2, 2, 1, 1] 391s 0.986 0.424 391s -0.422 -0.159 391s 391s y: [2, 2, 1, 1] 391s 0.090 0.098 391s -0.049 -0.045 391s testing: mul_mat, [4, 2, 3, 1] = [2, 4, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.557 0.389 391s -0.595 -0.741 391s -0.215 0.468 391s 0.513 -0.938 391s 391s x1: [2, 2, 3, 1] 391s -0.739 -0.983 391s 0.054 0.745 391s 391s y: [4, 2, 3, 1] 391s 0.029 1.168 -0.301 0.542 391s 0.260 -0.584 0.337 -0.671 391s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.582 0.675 0.270 0.310 391s 0.425 -0.375 -0.303 -0.204 391s 0.967 -0.604 0.096 -0.964 391s 0.034 0.839 0.451 -0.292 391s 391s x1: [4, 2, 3, 1] 391s 0.362 -0.836 -0.356 -0.699 391s 0.257 -0.819 -0.387 0.700 391s 391s y: [4, 2, 3, 1] 391s -0.666 0.718 1.495 -0.645 391s -0.291 0.391 0.032 -1.058 391s test-mul-mat0: iter:314/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.186 0.298 -0.573 0.329 391s 391s x1: [4, 3, 1, 1] 391s 0.501 -0.663 -0.214 -0.470 391s 0.593 -0.172 -0.559 -0.276 391s -0.191 -0.750 0.438 -0.784 391s 391s y: [1, 3, 1, 1] 391s -0.136 391s 0.288 391s -0.768 391s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 2, 2, 1] 391s 0.872 0.546 -0.604 0.384 391s -0.238 0.897 0.721 0.548 391s 391s x1: [4, 1, 2, 1] 391s -0.780 -0.902 -0.832 -0.763 391s 391s y: [2, 1, 2, 1] 391s -0.963 -1.642 391s testing: mul_mat, [3, 2, 2, 2] = [4, 3, 2, 2] * [4, 2, 2, 2] 391s x0: [4, 3, 2, 2] 391s -0.326 -0.220 0.072 -0.336 391s 0.699 -0.260 -0.177 -0.145 391s 0.628 0.299 0.060 -0.107 391s 391s x1: [4, 2, 2, 2] 391s 0.401 0.144 -0.338 -0.301 391s 0.571 0.991 -0.081 0.668 391s 391s y: [3, 2, 2, 2] 391s -0.086 0.346 0.307 391s -0.634 0.059 0.578 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.655 0.981 -0.343 0.824 391s 391s x1: [4, 3, 1, 1] 391s 0.718 0.171 -0.435 -0.887 391s -0.519 -0.157 -0.997 0.031 391s 0.657 -0.426 0.937 -0.518 391s 391s y: [1, 3, 1, 1] 391s -0.885 391s 0.553 391s -1.596 391s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 2, 2, 1] 391s -0.350 391s 0.004 391s 391s x1: [1, 4, 2, 1] 391s -0.856 391s 0.788 391s -0.957 391s 0.130 391s 391s y: [2, 4, 2, 1] 391s 0.299 -0.004 391s -0.276 0.003 391s 0.335 -0.004 391s -0.045 0.001 391s testing: mul_mat, [4, 1, 2, 2] = [2, 4, 2, 2] * [2, 1, 2, 2] 391s x0: [2, 4, 2, 2] 391s -0.148 -0.296 391s 0.479 0.846 391s 0.501 0.314 391s 0.333 0.361 391s 391s x1: [2, 1, 2, 2] 391s 0.787 -0.211 391s 391s y: [4, 1, 2, 2] 391s -0.054 0.198 0.328 0.186 391s test-mul-mat0: iter:315/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.442 391s -0.301 391s 391s x1: [1, 1, 1, 1] 391s 0.029 391s 391s y: [2, 1, 1, 1] 391s 0.013 -0.009 391s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.987 391s 0.631 391s 0.259 391s 391s x1: [1, 2, 2, 1] 391s 0.040 391s -0.409 391s 391s y: [3, 2, 2, 1] 391s -0.039 0.025 0.010 391s 0.403 -0.258 -0.106 391s testing: mul_mat, [4, 3, 2, 2] = [1, 4, 2, 2] * [1, 3, 2, 2] 391s x0: [1, 4, 2, 2] 391s -0.607 391s -0.285 391s 0.987 391s 0.422 391s 391s x1: [1, 3, 2, 2] 391s 0.546 391s 0.820 391s -0.458 391s 391s y: [4, 3, 2, 2] 391s -0.332 -0.156 0.539 0.231 391s -0.498 -0.234 0.809 0.346 391s 0.278 0.130 -0.452 -0.193 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.973 391s 391s x1: [1, 4, 1, 1] 391s -0.119 391s -0.199 391s -0.665 391s -0.299 391s 391s y: [1, 4, 1, 1] 391s -0.116 391s -0.194 391s -0.647 391s -0.291 391s testing: mul_mat, [2, 1, 2, 1] = [1, 2, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.634 391s -0.714 391s 391s x1: [1, 1, 2, 1] 391s -0.892 391s 391s y: [2, 1, 2, 1] 391s -0.566 0.637 391s testing: mul_mat, [1, 1, 2, 2] = [2, 1, 2, 2] * [2, 1, 2, 2] 391s x0: [2, 1, 2, 2] 391s 0.082 -0.525 391s 391s x1: [2, 1, 2, 2] 391s -0.267 0.040 391s 391s y: [1, 1, 2, 2] 391s -0.043 391s test-mul-mat0: iter:316/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.722 391s 0.558 391s 0.966 391s 391s x1: [1, 3, 1, 1] 391s -0.776 391s -0.051 391s -0.047 391s 391s y: [3, 3, 1, 1] 391s -0.560 -0.433 -0.750 391s -0.037 -0.029 -0.050 391s -0.034 -0.026 -0.045 391s testing: mul_mat, [2, 3, 4, 1] = [1, 2, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.688 391s -0.205 391s 391s x1: [1, 3, 4, 1] 391s -0.992 391s -0.309 391s -0.190 391s 391s y: [2, 3, 4, 1] 391s -0.683 0.203 391s -0.212 0.063 391s -0.131 0.039 391s testing: mul_mat, [3, 2, 4, 2] = [1, 3, 4, 2] * [1, 2, 4, 2] 391s x0: [1, 3, 4, 2] 391s 0.967 391s 0.308 391s 0.651 391s 391s x1: [1, 2, 4, 2] 391s 0.226 391s -0.892 391s 391s y: [3, 2, 4, 2] 391s 0.219 0.070 0.147 391s -0.862 -0.275 -0.580 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.486 391s -0.200 391s 0.046 391s 391s x1: [1, 3, 1, 1] 391s 0.801 391s 0.304 391s -0.804 391s 391s y: [3, 3, 1, 1] 391s -0.389 -0.160 0.037 391s -0.148 -0.061 0.014 391s 0.390 0.161 -0.037 391s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.741 0.285 -0.408 391s 0.178 0.578 -0.067 391s 0.343 0.107 -0.186 391s -0.931 0.175 -0.607 391s 391s x1: [3, 1, 4, 1] 391s -0.519 -0.891 -0.303 391s 391s y: [4, 1, 4, 1] 391s 0.253 -0.587 -0.217 0.510 391s testing: mul_mat, [2, 3, 4, 2] = [4, 2, 4, 2] * [4, 3, 4, 2] 391s x0: [4, 2, 4, 2] 391s -0.283 0.071 -0.565 -0.438 391s -0.741 -0.814 0.154 0.090 391s 391s x1: [4, 3, 4, 2] 391s 0.084 -0.374 0.632 -0.450 391s 0.520 -0.548 0.379 -0.055 391s 0.759 0.298 0.465 0.793 391s 391s y: [2, 3, 4, 2] 391s -0.210 0.299 391s -0.376 0.114 391s -0.803 -0.662 391s test-mul-mat0: iter:317/500 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.727 -0.010 0.143 -0.505 391s 0.816 -0.287 -0.216 -0.857 391s 391s x1: [4, 4, 1, 1] 391s -0.716 0.224 -0.475 -0.713 391s 0.382 0.915 -0.045 -0.924 391s -0.340 0.979 -0.537 0.515 391s -0.942 -0.059 0.348 0.475 391s 391s y: [2, 4, 1, 1] 391s 0.810 0.064 391s 0.173 0.849 391s -0.099 -0.884 391s 0.496 -1.234 391s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.886 -0.602 0.662 -0.173 391s 0.182 0.805 -0.631 -0.422 391s -0.153 -0.814 0.348 -0.486 391s 391s x1: [4, 2, 3, 1] 391s 0.542 0.397 0.042 0.817 391s -0.230 0.667 -0.899 0.994 391s 391s y: [3, 2, 3, 1] 391s -0.833 0.047 -0.788 391s -0.965 0.644 -1.304 391s testing: mul_mat, [2, 3, 3, 3] = [4, 2, 3, 3] * [4, 3, 3, 3] 391s x0: [4, 2, 3, 3] 391s -0.190 0.673 -0.535 0.502 391s 0.113 -0.098 -0.567 0.607 391s 391s x1: [4, 3, 3, 3] 391s -0.109 0.221 0.550 0.739 391s 0.407 -0.102 -0.747 0.694 391s -0.760 -0.042 -0.631 -0.042 391s 391s y: [2, 3, 3, 3] 391s 0.246 0.103 391s 0.602 0.901 391s 0.433 0.251 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.223 0.211 -0.161 0.985 391s 0.602 0.804 0.323 0.381 391s -0.533 0.358 0.197 0.423 391s 391s x1: [4, 2, 1, 1] 391s -0.291 -0.808 -0.869 0.859 391s -0.317 0.841 0.951 0.105 391s 391s y: [3, 2, 1, 1] 391s 0.880 -0.778 0.058 391s 0.198 0.832 0.701 391s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.655 0.536 0.628 391s 391s x1: [3, 4, 3, 1] 391s 0.459 0.653 -0.498 391s -0.239 0.250 0.335 391s -0.675 0.780 -0.236 391s 0.210 -0.511 -0.044 391s 391s y: [1, 4, 3, 1] 391s 0.337 391s 0.188 391s -0.172 391s -0.164 391s testing: mul_mat, [3, 3, 3, 3] = [1, 3, 3, 3] * [1, 3, 3, 3] 391s x0: [1, 3, 3, 3] 391s -0.107 391s 0.984 391s -0.121 391s 391s x1: [1, 3, 3, 3] 391s 0.386 391s -0.328 391s -0.034 391s 391s y: [3, 3, 3, 3] 391s -0.041 0.380 -0.047 391s 0.035 -0.323 0.040 391s 0.004 -0.034 0.004 391s test-mul-mat0: iter:318/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.518 0.238 -0.198 391s 0.921 -0.606 -0.937 391s -0.642 0.011 -0.255 391s 0.955 -0.439 -0.547 391s 391s x1: [3, 1, 1, 1] 391s -0.722 -0.138 0.392 391s 391s y: [4, 1, 1, 1] 391s -0.484 -0.949 0.362 -0.843 391s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.716 -0.058 0.479 391s 0.813 0.420 -0.329 391s 0.296 -0.224 -0.625 391s 0.057 -0.006 0.472 391s 391s x1: [3, 4, 3, 1] 391s 0.821 0.899 -0.788 391s -0.474 0.531 0.551 391s 0.400 0.212 0.335 391s -0.902 -0.522 0.191 391s 391s y: [4, 4, 3, 1] 391s 0.157 1.305 0.534 -0.331 391s -0.106 -0.344 -0.604 0.230 391s 0.435 0.304 -0.138 0.180 391s -0.524 -1.016 -0.270 0.042 391s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.852 -0.262 0.137 391s 391s x1: [3, 4, 3, 1] 391s -0.125 0.300 0.641 391s 0.250 -0.642 -0.365 391s -0.278 0.733 0.364 391s -0.472 -0.772 0.749 391s 391s y: [1, 4, 3, 1] 391s -0.098 391s 0.331 391s -0.379 391s -0.097 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.879 0.382 0.190 391s -0.770 -0.441 0.766 391s 0.373 -0.240 -0.928 391s 0.957 -0.275 -0.313 391s 391s x1: [3, 1, 1, 1] 391s -0.221 -0.619 0.553 391s 391s y: [4, 1, 1, 1] 391s -0.325 0.866 -0.447 -0.214 391s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.903 0.524 0.886 0.107 391s 391s x1: [4, 3, 3, 1] 391s 0.647 -0.280 0.824 -0.222 391s 0.725 0.597 -0.370 -0.537 391s -0.266 0.010 0.428 0.620 391s 391s y: [1, 3, 3, 1] 391s -0.024 391s -0.727 391s 0.692 391s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 3, 3, 1] 391s -0.981 391s 0.126 391s -0.803 391s 391s x1: [1, 4, 3, 1] 391s -0.284 391s -0.507 391s -0.193 391s -0.989 391s 391s y: [3, 4, 3, 1] 391s 0.278 -0.036 0.228 391s 0.498 -0.064 0.408 391s 0.189 -0.024 0.155 391s 0.970 -0.124 0.794 391s test-mul-mat0: iter:319/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.761 -0.658 391s -0.921 0.773 391s 0.351 -0.294 391s 391s x1: [2, 2, 1, 1] 391s 0.917 -0.097 391s 0.045 0.849 391s 391s y: [3, 2, 1, 1] 391s 0.761 -0.919 0.351 391s -0.525 0.615 -0.234 391s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.005 0.273 391s 0.616 -0.961 391s 0.122 -0.112 391s 391s x1: [2, 3, 3, 1] 391s 0.638 0.638 391s 0.878 -0.452 391s -0.033 0.667 391s 391s y: [3, 3, 3, 1] 391s 0.171 -0.219 0.006 391s -0.128 0.975 0.158 391s 0.182 -0.661 -0.079 391s testing: mul_mat, [2, 3, 3, 1] = [2, 2, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.492 -0.103 391s 0.225 -0.919 391s 391s x1: [2, 3, 3, 1] 391s -0.051 -0.595 391s -0.990 -0.665 391s -0.124 -0.010 391s 391s y: [2, 3, 3, 1] 391s 0.087 0.535 391s 0.555 0.388 391s 0.062 -0.019 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.333 0.793 391s 0.408 -0.651 391s 391s x1: [2, 2, 1, 1] 391s -0.998 0.988 391s 0.677 -0.663 391s 391s y: [2, 2, 1, 1] 391s 1.116 -1.050 391s -0.751 0.707 391s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.432 0.956 391s 0.739 -0.275 391s 391s x1: [2, 2, 3, 1] 391s -0.550 0.467 391s 0.387 -0.796 391s 391s y: [2, 2, 3, 1] 391s 0.209 -0.535 391s -0.594 0.505 391s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.425 -0.020 391s 391s x1: [2, 2, 3, 1] 391s -0.884 -0.662 391s -0.586 0.524 391s 391s y: [1, 2, 3, 1] 391s 0.389 391s 0.239 391s test-mul-mat0: iter:320/500 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.863 0.077 391s 391s x1: [2, 1, 1, 1] 391s -0.349 0.006 391s 391s y: [1, 1, 1, 1] 391s 0.302 391s testing: mul_mat, [3, 1, 4, 1] = [2, 3, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.252 0.678 391s -0.847 0.430 391s 0.411 -0.998 391s 391s x1: [2, 1, 4, 1] 391s 0.638 0.611 391s 391s y: [3, 1, 4, 1] 391s 0.576 -0.278 -0.348 391s testing: mul_mat, [2, 3, 4, 4] = [2, 2, 4, 4] * [2, 3, 4, 4] 391s x0: [2, 2, 4, 4] 391s 0.645 -0.548 391s 0.200 0.706 391s 391s x1: [2, 3, 4, 4] 391s -0.076 -0.396 391s 0.086 -0.837 391s 0.883 0.166 391s 391s y: [2, 3, 4, 4] 391s 0.168 -0.294 391s 0.514 -0.574 391s 0.478 0.293 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.459 -0.367 391s 0.370 -0.147 391s 391s x1: [2, 2, 1, 1] 391s 0.128 0.079 391s 0.366 0.832 391s 391s y: [2, 2, 1, 1] 391s -0.088 0.036 391s -0.473 0.013 391s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 3, 4, 1] 391s -0.459 -0.044 391s -0.324 -0.819 391s 0.045 -0.827 391s 391s x1: [2, 2, 4, 1] 391s 0.305 -0.473 391s 0.946 -0.244 391s 391s y: [3, 2, 4, 1] 391s -0.119 0.289 0.405 391s -0.423 -0.107 0.244 391s testing: mul_mat, [1, 2, 4, 4] = [3, 1, 4, 4] * [3, 2, 4, 4] 391s x0: [3, 1, 4, 4] 391s -0.752 0.163 0.580 391s 391s x1: [3, 2, 4, 4] 391s -0.748 -0.379 0.222 391s 0.345 0.707 0.088 391s 391s y: [1, 2, 4, 4] 391s 0.630 391s -0.093 391s test-mul-mat0: iter:321/500 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.507 0.710 0.514 -0.755 391s 0.415 0.387 -0.437 -0.004 391s 391s x1: [4, 3, 1, 1] 391s -0.171 0.358 0.205 -0.577 391s 0.939 -0.281 0.085 -0.200 391s 0.625 -0.813 0.909 0.694 391s 391s y: [2, 3, 1, 1] 391s 0.882 -0.019 391s -0.481 0.244 391s -0.951 -0.456 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.397 -0.863 -0.335 0.808 391s -0.440 -0.397 -0.472 0.645 391s 391s x1: [4, 2, 1, 1] 391s -0.698 -0.647 -0.330 0.636 391s 0.041 0.201 -0.954 0.307 391s 391s y: [2, 2, 1, 1] 391s 1.461 1.130 391s 0.378 0.551 391s testing: mul_mat, [4, 2, 1, 2] = [4, 4, 1, 2] * [4, 2, 1, 2] 391s x0: [4, 4, 1, 2] 391s 0.541 -0.993 0.839 0.588 391s 0.314 0.145 -0.809 0.451 391s 0.810 0.999 -0.989 -0.587 391s -0.473 0.655 0.817 -0.320 391s 391s x1: [4, 2, 1, 2] 391s 0.403 -0.847 0.832 0.312 391s 0.847 -0.297 0.805 0.557 391s 391s y: [4, 2, 1, 2] 391s 1.942 -0.529 -1.526 -0.166 391s 1.757 -0.177 -0.735 -0.116 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.031 0.053 -0.890 0.134 391s -0.322 0.084 0.094 0.010 391s -0.251 0.733 0.330 0.892 391s 391s x1: [4, 4, 1, 1] 391s -0.431 0.501 0.044 0.883 391s -0.353 0.235 0.334 -0.543 391s 0.235 0.344 -0.130 0.762 391s -0.000 -0.314 -0.558 -0.513 391s 391s y: [3, 4, 1, 1] 391s 0.119 0.194 1.279 391s -0.347 0.160 -0.114 391s 0.229 -0.051 0.831 391s 0.411 -0.084 -0.871 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.743 -0.490 0.522 391s -0.977 -0.227 -0.793 391s -0.042 0.154 -0.728 391s -0.815 0.188 0.255 391s 391s x1: [3, 4, 1, 1] 391s -0.203 -0.533 -0.539 391s -0.702 0.511 -0.655 391s -0.055 -0.254 0.678 391s 0.401 0.981 0.023 391s 391s y: [4, 4, 1, 1] 391s -0.171 0.747 0.319 -0.072 391s -1.114 1.090 0.585 0.501 391s 0.437 -0.426 -0.530 0.170 391s -0.171 -0.633 0.117 -0.137 391s testing: mul_mat, [4, 3, 1, 2] = [4, 4, 1, 2] * [4, 3, 1, 2] 391s x0: [4, 4, 1, 2] 391s -0.838 0.977 -0.744 -0.615 391s 0.705 0.671 -0.863 -0.437 391s -0.584 -0.314 0.720 0.673 391s -0.631 0.344 -0.266 -0.070 391s 391s x1: [4, 3, 1, 2] 391s -0.683 0.366 -0.415 0.451 391s -0.624 -0.522 -0.752 -0.157 391s -0.061 -0.454 -0.646 0.284 391s 391s y: [4, 3, 1, 2] 391s 0.962 -0.075 0.289 0.636 391s 0.669 -0.072 -0.119 0.425 391s -0.087 0.086 -0.096 0.034 391s test-mul-mat0: iter:322/500 391s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.455 391s 0.834 391s -0.719 391s 391s x1: [1, 1, 1, 1] 391s 0.578 391s 391s y: [3, 1, 1, 1] 391s 0.263 0.482 -0.415 391s testing: mul_mat, [1, 3, 4, 1] = [1, 1, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 1, 4, 1] 391s 0.021 391s 391s x1: [1, 3, 4, 1] 391s 0.175 391s -0.432 391s -0.334 391s 391s y: [1, 3, 4, 1] 391s 0.004 391s -0.009 391s -0.007 391s testing: mul_mat, [2, 1, 4, 4] = [1, 2, 4, 4] * [1, 1, 4, 4] 391s x0: [1, 2, 4, 4] 391s 0.975 391s -0.977 391s 391s x1: [1, 1, 4, 4] 391s -0.598 391s 391s y: [2, 1, 4, 4] 391s -0.582 0.584 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.283 391s 0.118 391s 0.957 391s 391s x1: [1, 2, 1, 1] 391s 0.496 391s 0.248 391s 391s y: [3, 2, 1, 1] 391s -0.140 0.058 0.474 391s -0.070 0.029 0.237 391s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.132 -0.546 -0.730 391s 391s x1: [3, 1, 4, 1] 391s 0.056 -0.683 0.022 391s 391s y: [1, 1, 4, 1] 391s 0.364 391s testing: mul_mat, [1, 3, 4, 4] = [1, 1, 4, 4] * [1, 3, 4, 4] 391s x0: [1, 1, 4, 4] 391s 0.634 391s 391s x1: [1, 3, 4, 4] 391s -0.729 391s 0.782 391s -0.168 391s 391s y: [1, 3, 4, 4] 391s -0.462 391s 0.496 391s -0.106 391s test-mul-mat0: iter:323/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.421 -0.837 -0.266 391s -0.107 -0.250 -0.181 391s 0.310 -0.616 0.336 391s 391s x1: [3, 1, 1, 1] 391s -0.567 -0.321 0.108 391s 391s y: [3, 1, 1, 1] 391s 0.001 0.121 0.058 391s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.191 -0.707 -0.108 391s 0.755 0.514 0.136 391s -0.601 -0.085 0.179 391s 391s x1: [3, 3, 4, 1] 391s 0.127 0.335 0.895 391s -0.653 0.691 0.800 391s -0.892 0.595 0.826 391s 391s y: [3, 3, 4, 1] 391s -0.358 0.390 0.055 391s -0.450 -0.029 0.477 391s -0.339 -0.255 0.634 391s testing: mul_mat, [2, 3, 4, 4] = [3, 2, 4, 4] * [3, 3, 4, 4] 391s x0: [3, 2, 4, 4] 391s 0.499 0.641 -0.379 391s 0.563 0.357 0.599 391s 391s x1: [3, 3, 4, 4] 391s -0.495 0.630 -0.258 391s 0.685 -0.870 -0.418 391s -0.870 0.163 -0.975 391s 391s y: [2, 3, 4, 4] 391s 0.255 -0.208 391s -0.058 -0.176 391s 0.040 -1.016 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.351 0.855 -0.039 391s -0.492 -0.673 -0.020 391s 391s x1: [3, 2, 1, 1] 391s -0.696 0.735 0.421 391s 0.683 -0.280 -0.220 391s 391s y: [2, 2, 1, 1] 391s 0.856 -0.161 391s -0.471 -0.143 391s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.445 -0.822 391s 0.217 -0.391 391s 0.629 0.317 391s -0.118 -0.835 391s 391s x1: [2, 3, 4, 1] 391s -0.565 0.987 391s -0.142 0.171 391s -0.490 -0.008 391s 391s y: [4, 3, 4, 1] 391s -0.559 -0.509 -0.043 -0.757 391s -0.077 -0.098 -0.035 -0.126 391s 0.225 -0.103 -0.311 0.064 391s testing: mul_mat, [1, 2, 4, 4] = [4, 1, 4, 4] * [4, 2, 4, 4] 391s x0: [4, 1, 4, 4] 391s 0.317 0.542 0.796 -0.290 391s 391s x1: [4, 2, 4, 4] 391s 0.053 0.896 -0.987 0.232 391s -0.495 0.329 0.397 0.972 391s 391s y: [1, 2, 4, 4] 391s -0.351 391s 0.056 391s test-mul-mat0: iter:324/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.568 -0.302 -0.226 0.868 391s 391s x1: [4, 3, 1, 1] 391s 0.223 0.438 -0.671 0.983 391s -0.677 0.009 -0.309 0.003 391s -0.590 -0.151 -0.597 0.333 391s 391s y: [1, 3, 1, 1] 391s 1.000 391s -0.315 391s 0.134 391s testing: mul_mat, [4, 1, 2, 1] = [4, 4, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 4, 2, 1] 391s 0.440 -0.474 0.179 0.689 391s 0.790 -0.598 0.127 -0.881 391s -0.615 0.450 0.128 0.077 391s -0.547 0.538 0.925 -0.143 391s 391s x1: [4, 1, 2, 1] 391s -0.785 -0.378 0.239 -0.636 391s 391s y: [4, 1, 2, 1] 391s -0.562 0.197 0.295 0.538 391s testing: mul_mat, [1, 4, 2, 4] = [4, 1, 2, 4] * [4, 4, 2, 4] 391s x0: [4, 1, 2, 4] 391s 0.544 -0.303 -0.548 -0.806 391s 391s x1: [4, 4, 2, 4] 391s 0.750 -0.938 -0.798 0.540 391s -0.536 0.328 0.659 -0.151 391s -0.222 -0.213 0.925 0.231 391s -0.675 0.851 -0.912 0.195 391s 391s y: [1, 4, 2, 4] 391s 0.694 391s -0.630 391s -0.750 391s -0.283 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.207 -0.375 -0.059 0.522 391s 0.119 0.093 0.805 0.266 391s -0.624 -0.871 0.628 0.950 391s -0.924 -0.806 0.216 0.602 391s 391s x1: [4, 1, 1, 1] 391s 0.734 0.695 -0.343 0.713 391s 391s y: [4, 1, 1, 1] 391s -0.020 0.065 -0.602 -0.884 391s testing: mul_mat, [2, 4, 2, 1] = [4, 2, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 2, 2, 1] 391s 0.758 -0.065 -0.429 0.105 391s 0.353 -0.190 0.964 -0.748 391s 391s x1: [4, 4, 2, 1] 391s 0.747 0.186 0.493 -0.701 391s 0.992 0.052 0.218 0.029 391s -0.705 0.681 -0.237 0.990 391s -0.662 -0.524 -0.819 0.131 391s 391s y: [2, 4, 2, 1] 391s 0.269 1.228 391s 0.658 0.529 391s -0.373 -1.348 391s -0.102 -1.022 391s testing: mul_mat, [1, 4, 2, 4] = [2, 1, 2, 4] * [2, 4, 2, 4] 391s x0: [2, 1, 2, 4] 391s 0.349 -0.437 391s 391s x1: [2, 4, 2, 4] 391s -0.072 0.016 391s 0.482 -0.736 391s 0.429 -0.926 391s 0.404 0.470 391s 391s y: [1, 4, 2, 4] 391s -0.032 391s 0.490 391s 0.555 391s -0.065 391s test-mul-mat0: iter:325/500 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.397 0.363 391s 0.637 0.166 391s -0.211 0.900 391s 391s x1: [2, 4, 1, 1] 391s 0.525 0.952 391s 0.856 -0.995 391s 0.724 -0.551 391s -0.307 0.700 391s 391s y: [3, 4, 1, 1] 391s 0.138 0.493 0.746 391s -0.701 0.380 -1.076 391s -0.487 0.369 -0.649 391s 0.376 -0.079 0.695 391s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.291 0.199 391s 0.700 -0.543 391s 0.988 0.601 391s 391s x1: [2, 3, 4, 1] 391s 0.852 0.856 391s 0.144 0.438 391s -0.795 0.698 391s 391s y: [3, 3, 4, 1] 391s 0.418 0.132 1.356 391s 0.129 -0.137 0.405 391s -0.092 -0.936 -0.366 391s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.785 0.806 391s 0.061 -0.365 391s -0.265 0.935 391s 391s x1: [2, 3, 4, 1] 391s -0.824 -0.076 391s 0.010 0.952 391s 0.611 -0.154 391s 391s y: [3, 3, 4, 1] 391s -0.708 -0.023 0.147 391s 0.775 -0.347 0.887 391s 0.356 0.094 -0.306 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.725 -0.641 391s -0.265 0.469 391s -0.368 -0.433 391s 391s x1: [2, 3, 1, 1] 391s -0.627 0.534 391s -0.938 -0.029 391s 0.940 -0.071 391s 391s y: [3, 3, 1, 1] 391s -0.797 0.417 -0.000 391s -0.661 0.235 0.358 391s 0.727 -0.283 -0.315 391s testing: mul_mat, [3, 2, 4, 1] = [3, 3, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.117 0.476 -0.353 391s -0.848 0.622 -0.250 391s -0.570 -0.003 -0.491 391s 391s x1: [3, 2, 4, 1] 391s 0.170 0.128 0.512 391s 0.933 -0.602 0.304 391s 391s y: [3, 2, 4, 1] 391s -0.100 -0.193 -0.349 391s -0.285 -1.242 -0.679 391s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.172 0.204 0.542 391s -0.909 0.581 -0.420 391s 0.653 0.910 0.288 391s 391s x1: [3, 3, 4, 1] 391s 0.100 0.124 0.575 391s 0.609 0.704 -0.278 391s 0.422 -0.005 -0.561 391s 391s y: [3, 3, 4, 1] 391s 0.320 -0.261 0.344 391s -0.112 -0.028 0.958 391s -0.377 -0.152 0.109 391s test-mul-mat0: iter:326/500 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.401 -0.973 0.131 -0.193 391s 391s x1: [4, 1, 1, 1] 391s -0.405 -0.815 -0.213 0.393 391s 391s y: [1, 1, 1, 1] 391s 0.528 391s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 1, 2, 1] 391s -0.302 -0.684 -0.705 -0.072 391s 391s x1: [4, 1, 2, 1] 391s 0.546 0.655 -0.994 0.156 391s 391s y: [1, 1, 2, 1] 391s 0.077 391s testing: mul_mat, [4, 1, 2, 3] = [4, 4, 2, 3] * [4, 1, 2, 3] 391s x0: [4, 4, 2, 3] 391s -0.208 -0.956 0.010 0.137 391s -0.146 0.034 0.113 0.459 391s 0.054 0.301 0.145 0.670 391s 0.674 0.618 0.679 0.246 391s 391s x1: [4, 1, 2, 3] 391s 0.976 -0.396 -0.980 -0.812 391s 391s y: [4, 1, 2, 3] 391s 0.054 -0.639 -0.752 -0.451 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.349 0.453 -0.904 -0.944 391s 0.488 0.330 -0.201 0.944 391s 391s x1: [4, 4, 1, 1] 391s -0.310 0.251 -0.722 -0.230 391s -0.138 -0.900 -0.286 0.540 391s -0.203 -0.805 -0.095 -0.951 391s -0.541 -0.552 0.568 0.455 391s 391s y: [2, 4, 1, 1] 391s 1.092 -0.140 391s -0.610 0.203 391s 0.690 -1.243 391s -1.004 -0.131 391s testing: mul_mat, [3, 4, 2, 1] = [2, 3, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 3, 2, 1] 391s -0.975 0.593 391s 0.902 0.357 391s -0.036 0.307 391s 391s x1: [2, 4, 2, 1] 391s 0.902 -0.604 391s 0.140 0.074 391s 0.959 -0.199 391s 0.765 0.210 391s 391s y: [3, 4, 2, 1] 391s -1.238 0.598 -0.218 391s -0.092 0.153 0.018 391s -1.053 0.794 -0.096 391s -0.621 0.765 0.037 391s testing: mul_mat, [1, 2, 2, 3] = [3, 1, 2, 3] * [3, 2, 2, 3] 391s x0: [3, 1, 2, 3] 391s -0.860 -0.696 -0.721 391s 391s x1: [3, 2, 2, 3] 391s 0.631 -0.116 0.461 391s -0.973 -0.976 -0.465 391s 391s y: [1, 2, 2, 3] 391s -0.794 391s 1.851 391s test-mul-mat0: iter:327/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.242 391s 391s x1: [1, 4, 1, 1] 391s 0.532 391s 0.135 391s 0.313 391s 0.660 391s 391s y: [1, 4, 1, 1] 391s -0.129 391s -0.033 391s -0.076 391s -0.160 391s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.017 391s 0.291 391s 391s x1: [1, 1, 4, 1] 391s 0.814 391s 391s y: [2, 1, 4, 1] 391s 0.014 0.237 391s testing: mul_mat, [2, 2, 4, 3] = [1, 2, 4, 3] * [1, 2, 4, 3] 391s x0: [1, 2, 4, 3] 391s -0.552 391s 0.681 391s 391s x1: [1, 2, 4, 3] 391s -0.777 391s -0.950 391s 391s y: [2, 2, 4, 3] 391s 0.429 -0.529 391s 0.524 -0.647 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.293 391s 0.721 391s 0.113 391s 391s x1: [1, 2, 1, 1] 391s -0.062 391s 0.853 391s 391s y: [3, 2, 1, 1] 391s 0.018 -0.045 -0.007 391s -0.250 0.615 0.096 391s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.208 -0.488 -0.812 391s 391s x1: [3, 1, 4, 1] 391s 0.767 -0.831 -0.206 391s 391s y: [1, 1, 4, 1] 391s 0.733 391s testing: mul_mat, [1, 3, 4, 3] = [1, 1, 4, 3] * [1, 3, 4, 3] 391s x0: [1, 1, 4, 3] 391s 0.642 391s 391s x1: [1, 3, 4, 3] 391s 0.173 391s 0.751 391s -0.791 391s 391s y: [1, 3, 4, 3] 391s 0.111 391s 0.482 391s -0.508 391s test-mul-mat0: iter:328/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.276 -0.845 -0.035 391s -0.521 0.977 -0.040 391s 391s x1: [3, 2, 1, 1] 391s 0.202 -0.729 0.578 391s 0.909 0.825 -0.457 391s 391s y: [2, 2, 1, 1] 391s 0.652 -0.841 391s -0.429 0.351 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.556 -0.953 0.523 391s 391s x1: [3, 2, 1, 1] 391s 0.154 0.121 -0.398 391s 0.469 -0.620 -0.387 391s 391s y: [1, 2, 1, 1] 391s -0.238 391s 0.650 391s testing: mul_mat, [2, 1, 1, 4] = [3, 2, 1, 4] * [3, 1, 1, 4] 391s x0: [3, 2, 1, 4] 391s -0.841 0.097 0.987 391s -0.685 -0.938 -0.534 391s 391s x1: [3, 1, 1, 4] 391s 0.510 -0.461 0.528 391s 391s y: [2, 1, 1, 4] 391s 0.047 -0.199 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.556 -0.064 0.501 391s -0.126 0.881 -0.537 391s 0.415 0.050 0.582 391s -0.241 0.958 -0.411 391s 391s x1: [3, 2, 1, 1] 391s 0.832 -0.331 0.529 391s -0.282 -0.715 0.776 391s 391s y: [4, 2, 1, 1] 391s -0.177 -0.681 0.637 -0.735 391s 0.591 -1.011 0.299 -0.936 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.774 -0.105 -0.771 -0.489 391s 391s x1: [4, 3, 1, 1] 391s -0.745 0.819 0.593 0.604 391s -0.897 -0.880 -0.538 -0.074 391s -0.825 0.679 0.557 0.366 391s 391s y: [1, 3, 1, 1] 391s -0.262 391s 1.238 391s -0.041 391s testing: mul_mat, [3, 4, 1, 4] = [1, 3, 1, 4] * [1, 4, 1, 4] 391s x0: [1, 3, 1, 4] 391s 0.085 391s 0.943 391s -0.243 391s 391s x1: [1, 4, 1, 4] 391s -0.329 391s 0.657 391s -0.045 391s 0.545 391s 391s y: [3, 4, 1, 4] 391s -0.028 -0.310 0.080 391s 0.056 0.620 -0.160 391s -0.004 -0.043 0.011 391s 0.046 0.513 -0.132 391s test-mul-mat0: iter:329/500 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.831 0.889 391s 391s x1: [2, 1, 1, 1] 391s -0.883 0.408 391s 391s y: [1, 1, 1, 1] 391s -0.371 391s testing: mul_mat, [1, 1, 4, 1] = [2, 1, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.216 -0.612 391s 391s x1: [2, 1, 4, 1] 391s -0.350 -0.405 391s 391s y: [1, 1, 4, 1] 391s 0.324 391s testing: mul_mat, [3, 1, 4, 1] = [2, 3, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.357 0.235 391s 0.185 0.188 391s 0.124 0.835 391s 391s x1: [2, 1, 4, 1] 391s 0.425 -0.229 391s 391s y: [3, 1, 4, 1] 391s 0.098 0.036 -0.138 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.214 0.684 391s 391s x1: [2, 3, 1, 1] 391s -0.712 -0.382 391s -0.378 0.152 391s 0.858 -0.551 391s 391s y: [1, 3, 1, 1] 391s -0.414 391s 0.023 391s -0.193 391s testing: mul_mat, [4, 2, 4, 1] = [1, 4, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 4, 4, 1] 391s 0.688 391s -0.306 391s -0.952 391s 0.559 391s 391s x1: [1, 2, 4, 1] 391s 0.826 391s -0.598 391s 391s y: [4, 2, 4, 1] 391s 0.569 -0.253 -0.787 0.462 391s -0.411 0.183 0.569 -0.334 391s testing: mul_mat, [3, 1, 4, 1] = [4, 3, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 3, 4, 1] 391s 0.449 0.614 -0.103 -0.280 391s 0.024 0.996 -0.146 -0.630 391s -0.429 0.788 0.395 0.289 391s 391s x1: [4, 1, 4, 1] 391s 0.047 -0.700 -0.414 0.688 391s 391s y: [3, 1, 4, 1] 391s -0.559 -1.069 -0.536 391s test-mul-mat0: iter:330/500 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.461 0.522 0.557 -0.394 391s 0.352 -0.225 -0.529 0.552 391s 391s x1: [4, 3, 1, 1] 391s -0.640 0.610 0.120 0.353 391s 0.394 0.611 -0.766 -0.421 391s 0.566 0.698 0.213 -0.454 391s 391s y: [2, 3, 1, 1] 391s -0.049 -0.231 391s 0.240 0.174 391s 0.923 -0.321 391s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.607 -0.910 0.117 0.168 391s -0.439 -0.330 -0.549 -0.939 391s -0.071 -0.423 -0.950 -0.009 391s 391s x1: [4, 2, 3, 1] 391s 0.283 0.499 -0.740 -0.874 391s 0.990 -0.939 -0.006 0.457 391s 391s y: [3, 2, 3, 1] 391s -0.859 0.938 0.480 391s 0.329 -0.550 0.328 391s testing: mul_mat, [1, 3, 3, 2] = [4, 1, 3, 2] * [4, 3, 3, 2] 391s x0: [4, 1, 3, 2] 391s -0.637 -0.510 0.719 0.200 391s 391s x1: [4, 3, 3, 2] 391s 0.114 0.847 -0.681 -0.957 391s -0.576 -0.631 0.034 0.995 391s 0.876 0.252 -0.033 0.200 391s 391s y: [1, 3, 3, 2] 391s -1.186 391s 0.913 391s -0.671 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.616 0.167 0.469 0.698 391s 0.447 0.839 0.775 -0.601 391s -0.361 0.439 0.465 -0.786 391s -0.126 -0.293 -0.882 -0.368 391s 391s x1: [4, 1, 1, 1] 391s -0.222 0.057 -0.626 0.470 391s 391s y: [4, 1, 1, 1] 391s 0.182 -0.819 -0.555 0.390 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s 0.675 -0.747 -0.887 0.281 391s 391s x1: [4, 4, 3, 1] 391s 0.794 -0.588 0.696 -0.620 391s -0.237 -0.729 0.998 0.166 391s -0.947 -0.465 0.944 0.110 391s -0.091 0.414 0.386 0.293 391s 391s y: [1, 4, 3, 1] 391s 0.183 391s -0.454 391s -1.099 391s -0.630 391s testing: mul_mat, [2, 4, 3, 2] = [1, 2, 3, 2] * [1, 4, 3, 2] 391s x0: [1, 2, 3, 2] 391s -0.628 391s 0.443 391s 391s x1: [1, 4, 3, 2] 391s 0.452 391s -0.819 391s -0.367 391s -0.141 391s 391s y: [2, 4, 3, 2] 391s -0.284 0.200 391s 0.515 -0.363 391s 0.231 -0.163 391s 0.089 -0.063 391s test-mul-mat0: iter:331/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.085 -0.923 0.631 -0.642 391s 391s x1: [4, 3, 1, 1] 391s 0.121 0.594 -0.668 -0.633 391s -0.618 0.664 0.121 0.394 391s 0.245 0.549 -0.340 -0.061 391s 391s y: [1, 3, 1, 1] 391s -0.573 391s -0.737 391s -0.703 391s testing: mul_mat, [3, 1, 4, 1] = [4, 3, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 3, 4, 1] 391s -0.055 -0.236 0.913 -0.934 391s -0.842 0.158 0.615 -0.182 391s -0.903 -0.125 0.733 -0.826 391s 391s x1: [4, 1, 4, 1] 391s -0.890 0.743 0.589 -0.300 391s 391s y: [3, 1, 4, 1] 391s 0.691 1.284 1.390 391s testing: mul_mat, [3, 3, 4, 2] = [4, 3, 4, 2] * [4, 3, 4, 2] 391s x0: [4, 3, 4, 2] 391s -0.980 0.137 0.466 0.634 391s 0.667 -0.192 0.639 0.976 391s -0.533 0.231 0.351 -0.909 391s 391s x1: [4, 3, 4, 2] 391s -0.247 0.894 -0.380 0.660 391s 0.181 -0.332 0.185 0.158 391s -0.827 0.497 0.052 -0.743 391s 391s y: [3, 3, 4, 2] 391s 0.606 0.065 -0.395 391s -0.037 0.457 -0.252 391s 0.432 -1.338 1.248 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.403 0.930 0.165 -0.003 391s 0.207 0.185 0.108 0.487 391s 391s x1: [4, 3, 1, 1] 391s 0.412 -0.662 -0.945 0.049 391s -0.706 -0.925 0.287 0.782 391s -0.321 0.786 -0.027 -0.186 391s 391s y: [2, 3, 1, 1] 391s -0.606 -0.115 391s -1.100 0.094 391s 0.598 -0.015 391s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.466 -0.197 391s 391s x1: [2, 4, 4, 1] 391s 0.315 0.169 391s 0.443 -0.944 391s 0.779 -0.566 391s 0.082 -0.076 391s 391s y: [1, 4, 4, 1] 391s -0.180 391s -0.021 391s -0.251 391s -0.023 391s testing: mul_mat, [4, 2, 4, 2] = [1, 4, 4, 2] * [1, 2, 4, 2] 391s x0: [1, 4, 4, 2] 391s -0.413 391s -0.160 391s 0.448 391s 0.678 391s 391s x1: [1, 2, 4, 2] 391s -0.824 391s 0.298 391s 391s y: [4, 2, 4, 2] 391s 0.340 0.132 -0.369 -0.559 391s -0.123 -0.048 0.134 0.202 391s test-mul-mat0: iter:332/500 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.586 0.288 391s 0.471 0.357 391s 391s x1: [2, 2, 1, 1] 391s -0.552 0.146 391s 0.310 0.949 391s 391s y: [2, 2, 1, 1] 391s -0.281 -0.207 391s 0.455 0.484 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.888 0.188 391s -0.010 0.252 391s 0.188 0.419 391s -0.063 -0.728 391s 391s x1: [2, 2, 1, 1] 391s -0.197 0.913 391s -0.714 0.879 391s 391s y: [4, 2, 1, 1] 391s 0.346 0.232 0.346 -0.652 391s 0.800 0.228 0.235 -0.595 391s testing: mul_mat, [1, 4, 1, 4] = [2, 1, 1, 4] * [2, 4, 1, 4] 391s x0: [2, 1, 1, 4] 391s -0.583 0.475 391s 391s x1: [2, 4, 1, 4] 391s -0.644 0.607 391s -0.235 -0.228 391s -0.137 -0.903 391s -0.364 0.798 391s 391s y: [1, 4, 1, 4] 391s 0.664 391s 0.029 391s -0.349 391s 0.591 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.661 -0.450 391s 0.771 0.080 391s 391s x1: [2, 1, 1, 1] 391s 0.068 0.858 391s 391s y: [2, 1, 1, 1] 391s -0.432 0.121 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.671 0.183 391s -0.736 -0.750 391s -0.130 -0.054 391s -0.401 0.040 391s 391s x1: [2, 2, 1, 1] 391s 0.041 0.674 391s 0.734 -0.403 391s 391s y: [4, 2, 1, 1] 391s 0.151 -0.536 -0.042 0.010 391s 0.419 -0.238 -0.073 -0.311 391s testing: mul_mat, [4, 2, 1, 4] = [4, 4, 1, 4] * [4, 2, 1, 4] 391s x0: [4, 4, 1, 4] 391s -0.380 0.732 -0.174 -0.214 391s -0.100 -0.604 -0.575 -0.691 391s 0.162 -0.597 -0.684 -0.307 391s 0.450 -0.905 0.485 -0.774 391s 391s x1: [4, 2, 1, 4] 391s 0.902 0.648 -0.497 -0.682 391s 0.123 -0.171 -0.168 0.234 391s 391s y: [4, 2, 1, 4] 391s 0.365 0.275 0.308 0.106 391s -0.193 0.025 0.164 -0.053 391s test-mul-mat0: iter:333/500 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.763 0.717 0.522 0.439 391s 0.108 -0.105 0.470 -0.383 391s 391s x1: [4, 4, 1, 1] 391s -0.273 -0.224 -0.033 0.689 391s -0.352 -0.045 0.437 -0.171 391s 0.634 0.411 -0.647 0.890 391s -0.214 -0.151 0.280 -0.594 391s 391s y: [2, 4, 1, 1] 391s 0.333 -0.285 391s 0.389 0.238 391s -0.137 -0.620 391s -0.059 0.352 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.638 -0.728 -0.579 -0.467 391s 0.742 0.037 0.484 -0.029 391s 391s x1: [4, 2, 3, 1] 391s -0.049 0.229 0.406 0.141 391s -0.552 0.871 0.868 0.224 391s 391s y: [2, 2, 3, 1] 391s -0.498 0.164 391s -1.593 0.036 391s testing: mul_mat, [4, 2, 3, 4] = [4, 4, 3, 4] * [4, 2, 3, 4] 391s x0: [4, 4, 3, 4] 391s 0.221 -0.913 -0.155 0.117 391s 0.433 -0.022 -0.978 -0.378 391s -0.901 0.878 0.758 0.655 391s 0.249 -0.635 0.408 0.196 391s 391s x1: [4, 2, 3, 4] 391s -0.363 -0.828 -0.715 -0.766 391s -0.713 0.118 0.547 0.925 391s 391s y: [4, 2, 3, 4] 391s 0.697 0.851 -1.444 -0.007 391s -0.242 -1.195 1.767 0.152 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.319 0.670 0.175 -0.657 391s 391s x1: [4, 4, 1, 1] 391s 0.213 0.933 -0.194 0.976 391s -0.756 0.537 -0.053 -0.907 391s 0.093 -0.906 -0.507 0.178 391s -0.175 0.172 -0.572 -0.243 391s 391s y: [1, 4, 1, 1] 391s 0.018 391s 0.706 391s -0.784 391s 0.119 391s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.614 391s 0.257 391s -0.339 391s -0.439 391s 391s x1: [1, 4, 3, 1] 391s 0.200 391s 0.231 391s -0.102 391s -0.033 391s 391s y: [4, 4, 3, 1] 391s 0.123 0.051 -0.068 -0.088 391s 0.142 0.059 -0.078 -0.101 391s -0.063 -0.026 0.035 0.045 391s -0.020 -0.009 0.011 0.015 391s testing: mul_mat, [4, 1, 3, 4] = [4, 4, 3, 4] * [4, 1, 3, 4] 391s x0: [4, 4, 3, 4] 391s 0.124 -0.263 -0.032 -0.735 391s 0.015 0.529 -0.945 0.068 391s 0.933 0.205 -0.027 -0.828 391s 0.287 0.115 0.423 0.155 391s 391s x1: [4, 1, 3, 4] 391s -0.762 -0.707 -0.321 0.908 391s 391s y: [4, 1, 3, 4] 391s -0.566 -0.020 -1.599 -0.295 391s test-mul-mat0: iter:334/500 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.193 0.395 0.690 391s 0.506 0.923 -0.784 391s 391s x1: [3, 4, 1, 1] 391s -0.353 -0.821 0.623 391s 0.670 0.575 0.283 391s 0.541 -0.781 -0.569 391s 0.252 -0.560 -0.418 391s 391s y: [2, 4, 1, 1] 391s 0.038 -1.425 391s 0.551 0.648 391s -0.596 -0.001 391s -0.461 -0.061 391s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 3, 3, 1] 391s 0.709 -0.885 0.220 391s -0.851 -0.303 0.318 391s 0.342 -0.909 0.008 391s 391s x1: [3, 2, 3, 1] 391s 0.798 0.874 -0.944 391s -0.928 -0.188 -0.953 391s 391s y: [3, 2, 3, 1] 391s -0.416 -1.244 -0.528 391s -0.702 0.543 -0.155 391s testing: mul_mat, [4, 3, 3, 2] = [3, 4, 3, 2] * [3, 3, 3, 2] 391s x0: [3, 4, 3, 2] 391s 0.653 -0.304 -0.488 391s 0.392 0.114 0.839 391s 0.248 0.273 0.851 391s -0.095 0.143 -0.743 391s 391s x1: [3, 3, 3, 2] 391s 0.742 -0.316 0.801 391s 0.710 -0.607 0.916 391s -0.070 -0.457 -0.387 391s 391s y: [4, 3, 3, 2] 391s 0.189 0.927 0.779 -0.711 391s 0.201 0.978 0.790 -0.835 391s 0.282 -0.404 -0.472 0.229 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.194 -0.621 0.925 391s 391s x1: [3, 4, 1, 1] 391s -0.483 0.529 0.680 391s 0.877 -0.524 -0.736 391s -0.773 0.462 -0.599 391s 0.870 -0.948 0.701 391s 391s y: [1, 4, 1, 1] 391s 0.207 391s -0.185 391s -0.991 391s 1.406 391s testing: mul_mat, [1, 3, 3, 1] = [1, 1, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 1, 3, 1] 391s 0.792 391s 391s x1: [1, 3, 3, 1] 391s 0.936 391s 0.962 391s -0.490 391s 391s y: [1, 3, 3, 1] 391s 0.742 391s 0.762 391s -0.388 391s testing: mul_mat, [2, 1, 3, 2] = [1, 2, 3, 2] * [1, 1, 3, 2] 391s x0: [1, 2, 3, 2] 391s -0.212 391s -0.764 391s 391s x1: [1, 1, 3, 2] 391s 0.578 391s 391s y: [2, 1, 3, 2] 391s -0.123 -0.441 391s test-mul-mat0: iter:335/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.948 -0.970 -0.654 391s -0.939 0.613 0.134 391s -0.703 -0.343 -0.677 391s 391s x1: [3, 3, 1, 1] 391s 0.124 0.789 0.554 391s 0.894 -0.695 0.346 391s -0.021 0.920 -0.076 391s 391s y: [3, 3, 1, 1] 391s -1.245 0.442 -0.733 391s -0.400 -1.218 -0.624 391s -0.822 0.574 -0.249 391s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.010 0.967 -0.126 391s -0.143 -0.666 -0.352 391s 0.828 -0.509 0.976 391s 391s x1: [3, 3, 4, 1] 391s -0.597 0.753 0.213 391s -0.903 -0.956 -0.519 391s 0.717 -0.729 0.713 391s 391s y: [3, 3, 4, 1] 391s 0.695 -0.491 -0.670 391s -0.868 0.948 -0.767 391s -0.787 0.132 1.660 391s testing: mul_mat, [3, 3, 4, 4] = [3, 3, 4, 4] * [3, 3, 4, 4] 391s x0: [3, 3, 4, 4] 391s 0.612 0.766 -0.100 391s 0.090 -0.241 -0.111 391s 0.124 -0.757 0.102 391s 391s x1: [3, 3, 4, 4] 391s -0.812 0.243 0.157 391s -0.836 0.935 -0.889 391s -0.185 -0.531 0.447 391s 391s y: [3, 3, 4, 4] 391s -0.326 -0.149 -0.268 391s 0.294 -0.202 -0.902 391s -0.565 0.062 0.425 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.043 0.140 0.149 391s 0.341 -0.312 -0.313 391s -0.095 -0.232 -0.285 391s 391s x1: [3, 3, 1, 1] 391s 0.133 1.000 -0.130 391s 0.332 0.474 -0.132 391s 0.072 -0.964 0.101 391s 391s y: [3, 3, 1, 1] 391s 0.126 -0.226 -0.207 391s 0.061 0.007 -0.104 391s -0.117 0.294 0.188 391s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.538 -0.076 -0.442 391s 0.596 0.179 0.912 391s 391s x1: [3, 3, 4, 1] 391s 0.172 0.538 -0.298 391s 0.118 0.196 0.386 391s -0.632 -0.681 0.423 391s 391s y: [2, 3, 4, 1] 391s -0.002 -0.073 391s -0.249 0.458 391s 0.204 -0.112 391s testing: mul_mat, [1, 3, 4, 4] = [2, 1, 4, 4] * [2, 3, 4, 4] 391s x0: [2, 1, 4, 4] 391s 0.279 0.853 391s 391s x1: [2, 3, 4, 4] 391s 0.528 0.377 391s 0.969 0.643 391s 0.472 0.246 391s 391s y: [1, 3, 4, 4] 391s 0.469 391s 0.818 391s 0.341 391s test-mul-mat0: iter:336/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.858 -0.644 0.706 -0.305 391s 0.529 0.602 -0.308 0.889 391s 391s x1: [4, 1, 1, 1] 391s 0.076 -0.327 0.189 0.605 391s 391s y: [2, 1, 1, 1] 391s 0.095 0.323 391s testing: mul_mat, [1, 2, 3, 1] = [4, 1, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.907 -0.659 -0.687 0.695 391s 391s x1: [4, 2, 3, 1] 391s -0.342 0.215 -0.736 -0.648 391s 0.189 -0.117 -0.997 0.781 391s 391s y: [1, 2, 3, 1] 391s 0.224 391s 1.134 391s testing: mul_mat, [1, 1, 3, 4] = [4, 1, 3, 4] * [4, 1, 3, 4] 391s x0: [4, 1, 3, 4] 391s 0.416 0.204 -0.808 0.441 391s 391s x1: [4, 1, 3, 4] 391s -0.292 0.218 -0.379 0.004 391s 391s y: [1, 1, 3, 4] 391s 0.231 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.368 -0.134 -0.845 0.846 391s 0.758 -0.040 0.011 -0.289 391s -0.706 0.618 -0.705 0.816 391s 0.641 -0.022 0.542 -0.650 391s 391s x1: [4, 1, 1, 1] 391s -0.482 -0.094 0.871 0.549 391s 391s y: [4, 1, 1, 1] 391s -0.436 -0.511 0.116 -0.192 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s 0.413 -0.649 -0.592 0.118 391s 391s x1: [4, 4, 3, 1] 391s -0.527 0.920 0.216 -0.464 391s -0.766 -0.514 -0.441 -0.646 391s -0.177 0.207 -0.129 0.730 391s 0.078 -0.580 -0.436 -0.554 391s 391s y: [1, 4, 3, 1] 391s -0.998 391s 0.202 391s -0.046 391s 0.602 391s testing: mul_mat, [2, 4, 3, 4] = [1, 2, 3, 4] * [1, 4, 3, 4] 391s x0: [1, 2, 3, 4] 391s -0.080 391s 0.697 391s 391s x1: [1, 4, 3, 4] 391s 0.308 391s 0.217 391s -0.525 391s 0.436 391s 391s y: [2, 4, 3, 4] 391s -0.025 0.214 391s -0.017 0.151 391s 0.042 -0.366 391s -0.035 0.304 391s test-mul-mat0: iter:337/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.398 -0.872 0.316 391s -0.090 0.635 0.046 391s 391s x1: [3, 2, 1, 1] 391s -0.981 -0.181 0.485 391s -0.060 -0.483 -0.025 391s 391s y: [2, 2, 1, 1] 391s 0.701 -0.004 391s 0.437 -0.303 391s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.700 -0.853 -0.926 391s -0.183 0.121 0.319 391s 0.419 0.250 -0.365 391s 391s x1: [3, 2, 3, 1] 391s 0.835 0.212 0.662 391s -0.316 -0.926 0.638 391s 391s y: [3, 2, 3, 1] 391s -1.378 0.084 0.161 391s 0.420 0.149 -0.597 391s testing: mul_mat, [1, 3, 3, 4] = [3, 1, 3, 4] * [3, 3, 3, 4] 391s x0: [3, 1, 3, 4] 391s 0.407 0.164 -0.446 391s 391s x1: [3, 3, 3, 4] 391s -0.586 0.477 0.292 391s 0.548 0.777 0.438 391s 0.622 -0.406 -0.440 391s 391s y: [1, 3, 3, 4] 391s -0.291 391s 0.155 391s 0.383 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.199 0.963 -0.961 391s -0.769 -0.359 0.232 391s 0.001 0.199 0.397 391s 0.172 -0.759 0.789 391s 391s x1: [3, 1, 1, 1] 391s 0.724 -0.651 0.484 391s 391s y: [4, 1, 1, 1] 391s -1.236 -0.210 0.063 1.000 391s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.267 0.360 0.595 -0.269 391s 391s x1: [4, 3, 3, 1] 391s 0.860 0.120 0.157 -0.151 391s -0.526 0.885 -0.282 -0.762 391s -0.107 0.998 0.409 0.063 391s 391s y: [1, 3, 3, 1] 391s -0.053 391s 0.496 391s 0.614 391s testing: mul_mat, [1, 4, 3, 4] = [1, 1, 3, 4] * [1, 4, 3, 4] 391s x0: [1, 1, 3, 4] 391s -0.209 391s 391s x1: [1, 4, 3, 4] 391s -0.691 391s 0.929 391s -0.505 391s 0.501 391s 391s y: [1, 4, 3, 4] 391s 0.145 391s -0.195 391s 0.106 391s -0.105 391s test-mul-mat0: iter:338/500 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.417 0.694 -0.344 391s -0.408 0.715 0.446 391s -0.297 -0.090 -0.786 391s 0.645 -0.950 0.618 391s 391s x1: [3, 3, 1, 1] 391s 0.001 0.547 -0.494 391s 0.659 -0.349 0.975 391s 0.522 0.941 0.488 391s 391s y: [4, 3, 1, 1] 391s 0.549 0.170 0.339 -0.824 391s -0.853 -0.083 -0.931 1.359 391s 0.268 0.678 -0.623 -0.256 391s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.334 0.891 -0.020 391s 0.179 0.114 -0.450 391s 0.351 -0.011 -0.906 391s 391s x1: [3, 4, 3, 1] 391s 0.982 -0.544 -0.472 391s 0.835 -0.644 0.512 391s 0.511 0.677 0.064 391s -0.488 0.223 0.570 391s 391s y: [3, 4, 3, 1] 391s -0.803 0.326 0.778 391s -0.863 -0.155 -0.163 391s 0.431 0.140 0.114 391s 0.350 -0.318 -0.689 391s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.785 -0.032 -0.400 391s 0.952 -0.333 0.725 391s 391s x1: [3, 3, 3, 1] 391s -0.833 0.698 0.125 391s -0.917 -0.635 0.016 391s 0.063 0.543 -0.870 391s 391s y: [2, 3, 3, 1] 391s -0.726 -0.935 391s -0.706 -0.650 391s 0.380 -0.752 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.024 0.562 0.982 391s 0.557 -0.001 0.251 391s 391s x1: [3, 2, 1, 1] 391s 0.251 -0.668 0.038 391s 0.940 -0.160 -0.645 391s 391s y: [2, 2, 1, 1] 391s -0.344 0.150 391s -0.746 0.362 391s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.570 -0.869 391s 391s x1: [2, 3, 3, 1] 391s -0.216 -0.049 391s 0.851 -0.264 391s 0.617 0.577 391s 391s y: [1, 3, 3, 1] 391s 0.166 391s -0.256 391s -0.852 391s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.024 391s 0.778 391s 391s x1: [1, 2, 3, 1] 391s -0.753 391s -0.956 391s 391s y: [2, 2, 3, 1] 391s 0.018 -0.586 391s 0.022 -0.744 391s test-mul-mat0: iter:339/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.018 0.255 0.135 391s 0.229 0.685 0.267 391s 0.415 -0.045 0.753 391s -0.472 0.202 0.797 391s 391s x1: [3, 1, 1, 1] 391s -0.138 0.985 0.156 391s 391s y: [4, 1, 1, 1] 391s 0.275 0.685 0.016 0.389 391s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.798 -0.849 -0.007 391s -0.674 0.353 -0.210 391s 391s x1: [3, 4, 2, 1] 391s 0.203 -0.553 0.823 391s 0.129 -0.523 -0.201 391s -0.093 0.243 0.393 391s -0.739 -0.220 -0.365 391s 391s y: [2, 4, 2, 1] 391s 0.626 -0.504 391s 0.549 -0.229 391s -0.284 0.066 391s -0.400 0.497 391s testing: mul_mat, [3, 2, 2, 4] = [3, 3, 2, 4] * [3, 2, 2, 4] 391s x0: [3, 3, 2, 4] 391s 0.577 -0.945 0.458 391s 0.176 -0.626 -0.537 391s 0.565 -0.777 0.268 391s 391s x1: [3, 2, 2, 4] 391s 0.565 0.521 -0.194 391s 0.826 -0.699 0.441 391s 391s y: [3, 2, 2, 4] 391s -0.255 -0.122 -0.138 391s 1.339 0.346 1.128 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.546 -0.085 0.875 391s 391s x1: [3, 3, 1, 1] 391s -0.130 0.513 0.485 391s -0.723 -0.827 -0.301 391s -0.618 -0.311 -0.225 391s 391s y: [1, 3, 1, 1] 391s 0.452 391s 0.202 391s 0.168 391s testing: mul_mat, [1, 3, 2, 1] = [1, 1, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.586 391s 391s x1: [1, 3, 2, 1] 391s 0.518 391s 0.116 391s 0.081 391s 391s y: [1, 3, 2, 1] 391s -0.303 391s -0.068 391s -0.047 391s testing: mul_mat, [4, 1, 2, 4] = [1, 4, 2, 4] * [1, 1, 2, 4] 391s x0: [1, 4, 2, 4] 391s 0.454 391s -0.457 391s 0.889 391s 0.730 391s 391s x1: [1, 1, 2, 4] 391s 0.932 391s 391s y: [4, 1, 2, 4] 391s 0.423 -0.426 0.829 0.681 391s test-mul-mat0: iter:340/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.495 0.857 0.759 0.352 391s 391s x1: [4, 3, 1, 1] 391s -0.842 -0.547 0.181 0.521 391s 0.408 0.039 0.799 -0.762 391s -0.585 0.194 -0.851 0.536 391s 391s y: [1, 3, 1, 1] 391s -0.565 391s 0.574 391s -0.580 391s testing: mul_mat, [2, 1, 3, 1] = [4, 2, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 2, 3, 1] 391s -0.133 0.383 0.243 -0.951 391s -0.096 -0.349 0.088 -0.297 391s 391s x1: [4, 1, 3, 1] 391s -0.890 -0.977 -0.302 -0.317 391s 391s y: [2, 1, 3, 1] 391s -0.028 0.494 391s testing: mul_mat, [4, 2, 3, 3] = [4, 4, 3, 3] * [4, 2, 3, 3] 391s x0: [4, 4, 3, 3] 391s -0.184 -0.515 0.021 0.601 391s 0.693 -0.475 -0.287 0.492 391s 0.154 -0.680 -0.587 0.802 391s -0.654 -0.657 0.476 -0.261 391s 391s x1: [4, 2, 3, 3] 391s 0.528 0.488 -0.786 0.977 391s -0.421 0.440 -0.232 0.447 391s 391s y: [4, 2, 3, 3] 391s 0.222 0.839 0.994 -1.294 391s 0.114 -0.214 0.131 -0.241 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.877 -0.891 -0.310 0.298 391s 0.099 -0.291 0.251 0.297 391s 391s x1: [4, 4, 1, 1] 391s 0.382 -0.394 0.721 0.165 391s -0.783 -0.428 -0.341 0.276 391s -0.718 -0.024 -0.479 -0.806 391s 0.704 0.995 0.932 -0.397 391s 391s y: [2, 4, 1, 1] 391s -0.159 0.382 391s 1.256 0.043 391s 0.559 -0.424 391s -1.911 -0.103 391s testing: mul_mat, [3, 4, 3, 1] = [2, 3, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.813 0.438 391s 0.729 0.027 391s 0.377 -0.326 391s 391s x1: [2, 4, 3, 1] 391s -0.096 0.901 391s 0.124 0.099 391s 0.893 -0.806 391s -0.519 -0.501 391s 391s y: [3, 4, 3, 1] 391s 0.473 -0.045 -0.330 391s -0.058 0.093 0.014 391s -1.079 0.629 0.599 391s 0.202 -0.392 -0.032 391s testing: mul_mat, [3, 2, 3, 3] = [3, 3, 3, 3] * [3, 2, 3, 3] 391s x0: [3, 3, 3, 3] 391s -0.874 -0.709 -0.148 391s -0.219 -0.193 -0.292 391s 0.656 0.565 0.500 391s 391s x1: [3, 2, 3, 3] 391s 0.875 -0.973 0.166 391s 0.156 -0.461 -0.870 391s 391s y: [3, 2, 3, 3] 391s -0.100 -0.052 0.108 391s 0.319 0.309 -0.593 391s test-mul-mat0: iter:341/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.744 -0.423 0.667 391s 0.619 -0.958 -0.992 391s 391s x1: [3, 2, 1, 1] 391s -0.590 0.247 -0.868 391s 0.663 -0.983 0.371 391s 391s y: [2, 2, 1, 1] 391s -0.244 0.259 391s 0.170 0.983 391s testing: mul_mat, [1, 2, 3, 1] = [3, 1, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.953 -0.053 -0.953 391s 391s x1: [3, 2, 3, 1] 391s 0.671 -0.381 0.551 391s 0.913 -0.114 -0.938 391s 391s y: [1, 2, 3, 1] 391s -1.145 391s 0.030 391s testing: mul_mat, [4, 1, 3, 3] = [3, 4, 3, 3] * [3, 1, 3, 3] 391s x0: [3, 4, 3, 3] 391s -0.295 0.475 -0.733 391s -0.356 -0.589 0.810 391s -0.745 0.967 -0.095 391s 0.778 -0.590 -0.938 391s 391s x1: [3, 1, 3, 3] 391s 0.939 -0.063 0.543 391s 391s y: [4, 1, 3, 3] 391s -0.705 0.144 -0.813 0.258 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.366 0.186 0.697 391s 0.252 -0.443 -0.450 391s -0.438 0.999 -0.520 391s 391s x1: [3, 4, 1, 1] 391s -0.077 0.199 0.604 391s -0.290 0.612 0.636 391s 0.799 0.355 -0.661 391s 0.740 0.465 0.941 391s 391s y: [3, 4, 1, 1] 391s 0.430 -0.380 -0.082 391s 0.451 -0.631 0.407 391s -0.103 0.342 0.348 391s 1.013 -0.443 -0.349 391s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.058 0.545 0.191 391s 0.296 -0.130 0.676 391s 391s x1: [3, 3, 3, 1] 391s 0.927 0.895 -0.870 391s -0.318 0.665 0.140 391s -0.613 0.063 0.418 391s 391s y: [2, 3, 3, 1] 391s 0.268 -0.430 391s 0.407 -0.086 391s 0.150 0.093 391s testing: mul_mat, [3, 3, 3, 3] = [2, 3, 3, 3] * [2, 3, 3, 3] 391s x0: [2, 3, 3, 3] 391s 0.699 -0.491 391s 0.224 -0.675 391s -0.938 -0.657 391s 391s x1: [2, 3, 3, 3] 391s 0.088 0.282 391s -0.089 -0.353 391s 0.976 0.849 391s 391s y: [3, 3, 3, 3] 391s -0.077 -0.171 -0.268 391s 0.111 0.218 0.315 391s 0.266 -0.354 -1.473 391s test-mul-mat0: iter:342/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.464 -0.656 0.632 391s 391s x1: [3, 1, 1, 1] 391s 0.860 -0.489 0.803 391s 391s y: [1, 1, 1, 1] 391s 0.429 391s testing: mul_mat, [2, 1, 4, 1] = [3, 2, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 2, 4, 1] 391s 0.009 -0.201 -0.644 391s 0.660 -0.753 0.965 391s 391s x1: [3, 1, 4, 1] 391s 0.070 0.210 0.891 391s 391s y: [2, 1, 4, 1] 391s -0.616 0.748 391s testing: mul_mat, [2, 2, 4, 2] = [3, 2, 4, 2] * [3, 2, 4, 2] 391s x0: [3, 2, 4, 2] 391s -0.954 0.576 0.377 391s 0.269 -0.013 0.993 391s 391s x1: [3, 2, 4, 2] 391s -0.722 -0.435 -0.372 391s -0.265 0.958 -0.075 391s 391s y: [2, 2, 4, 2] 391s 0.298 -0.557 391s 0.777 -0.158 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.503 -0.668 0.644 391s -0.028 0.581 -0.798 391s 0.095 -0.352 -0.586 391s -0.529 0.739 -0.306 391s 391s x1: [3, 2, 1, 1] 391s 0.296 -0.846 -0.313 391s 0.172 -0.348 0.816 391s 391s y: [4, 2, 1, 1] 391s 0.215 -0.251 0.510 -0.686 391s 0.672 -0.859 -0.339 -0.598 391s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.823 0.331 -0.258 -0.950 391s 0.547 -0.096 -0.656 -0.220 391s -0.194 0.331 -0.602 0.838 391s 0.963 0.764 -0.007 0.867 391s 391s x1: [4, 3, 4, 1] 391s 0.748 -0.406 -0.889 -0.445 391s 0.416 -0.917 0.131 -0.021 391s 0.076 -0.449 -0.989 -0.709 391s 391s y: [4, 3, 4, 1] 391s -0.098 1.130 -0.118 0.031 391s -0.660 0.234 -0.481 -0.319 391s 0.717 0.889 -0.162 -0.878 391s testing: mul_mat, [4, 4, 4, 2] = [4, 4, 4, 2] * [4, 4, 4, 2] 391s x0: [4, 4, 4, 2] 391s -0.693 0.517 -0.325 -0.188 391s 0.340 -0.410 -0.331 0.586 391s 0.327 -0.326 -0.032 0.205 391s 0.796 -0.996 -0.684 0.402 391s 391s x1: [4, 4, 4, 2] 391s -0.744 -0.791 0.753 0.260 391s -0.966 -0.366 0.006 -0.798 391s 0.912 0.744 -0.579 0.970 391s -0.678 -0.492 0.744 -0.002 391s 391s y: [4, 4, 4, 2] 391s -0.188 -0.026 0.043 -0.216 391s 0.628 -0.648 -0.360 -0.729 391s -0.241 0.765 0.273 0.770 391s -0.026 -0.277 -0.086 -0.560 391s test-mul-mat0: iter:343/500 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.543 -0.600 391s 391s x1: [2, 3, 1, 1] 391s -0.761 0.729 391s -0.473 -0.786 391s 0.254 -0.805 391s 391s y: [1, 3, 1, 1] 391s -0.025 391s 0.728 391s 0.345 391s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 4, 2, 1] 391s 0.142 0.324 391s 0.104 0.895 391s 0.220 0.765 391s 0.369 0.346 391s 391s x1: [2, 1, 2, 1] 391s 0.523 -0.080 391s 391s y: [4, 1, 2, 1] 391s 0.048 -0.017 0.054 0.165 391s testing: mul_mat, [3, 4, 2, 4] = [2, 3, 2, 4] * [2, 4, 2, 4] 391s x0: [2, 3, 2, 4] 391s -0.904 -0.144 391s -0.876 -0.741 391s -0.589 0.976 391s 391s x1: [2, 4, 2, 4] 391s -0.069 0.872 391s -0.409 0.459 391s -0.914 0.845 391s 0.654 -0.760 391s 391s y: [3, 4, 2, 4] 391s -0.064 -0.586 0.891 391s 0.303 0.018 0.688 391s 0.704 0.174 1.363 391s -0.482 -0.010 -1.127 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.913 0.780 391s 391s x1: [2, 3, 1, 1] 391s -0.251 0.929 391s 0.947 -0.222 391s -0.871 0.594 391s 391s y: [1, 3, 1, 1] 391s 0.954 391s -1.038 391s 1.260 391s testing: mul_mat, [2, 2, 2, 1] = [1, 2, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.424 391s 0.578 391s 391s x1: [1, 2, 2, 1] 391s 0.054 391s -0.084 391s 391s y: [2, 2, 2, 1] 391s 0.023 0.031 391s -0.036 -0.049 391s testing: mul_mat, [3, 1, 2, 4] = [2, 3, 2, 4] * [2, 1, 2, 4] 391s x0: [2, 3, 2, 4] 391s -0.181 -0.095 391s 0.858 0.638 391s -0.937 0.117 391s 391s x1: [2, 1, 2, 4] 391s 0.679 0.352 391s 391s y: [3, 1, 2, 4] 391s -0.156 0.807 -0.595 391s test-mul-mat0: iter:344/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.947 -0.291 0.521 391s -0.935 -0.369 -0.742 391s 391s x1: [3, 2, 1, 1] 391s 0.629 -0.962 -0.156 391s -0.524 -0.353 -0.355 391s 391s y: [2, 2, 1, 1] 391s 0.794 -0.117 391s -0.579 0.884 391s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.466 0.141 -0.769 391s -0.400 0.772 -0.511 391s 0.688 0.964 -0.399 391s 0.621 -0.092 0.571 391s 391s x1: [3, 2, 4, 1] 391s 0.089 -0.808 -0.888 391s 0.933 -0.056 -0.030 391s 391s y: [4, 2, 4, 1] 391s 0.528 -0.206 -0.364 -0.378 391s -0.419 -0.401 0.600 0.567 391s testing: mul_mat, [3, 4, 4, 4] = [3, 3, 4, 4] * [3, 4, 4, 4] 391s x0: [3, 3, 4, 4] 391s -0.151 -0.698 -0.949 391s 0.283 0.145 -0.889 391s -0.960 0.722 -0.191 391s 391s x1: [3, 4, 4, 4] 391s 0.314 0.255 0.699 391s -0.012 0.643 -0.614 391s 0.606 0.418 0.013 391s -0.989 0.056 -0.772 391s 391s y: [3, 4, 4, 4] 391s -0.888 -0.495 -0.251 391s 0.136 0.636 0.593 391s -0.396 0.220 -0.283 391s 0.843 0.414 1.138 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.584 -0.893 -0.763 391s -0.292 0.303 0.405 391s -0.964 0.085 0.115 391s 391s x1: [3, 3, 1, 1] 391s -0.945 0.803 0.021 391s 0.846 -0.652 -0.160 391s -0.317 -0.431 -0.489 391s 391s y: [3, 3, 1, 1] 391s -0.181 0.528 0.982 391s 0.210 -0.510 -0.889 391s 0.943 -0.237 0.212 391s testing: mul_mat, [4, 3, 4, 1] = [3, 4, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.376 0.303 0.695 391s 0.140 -0.391 -0.524 391s -0.456 -0.390 0.582 391s -0.410 0.269 0.622 391s 391s x1: [3, 3, 4, 1] 391s -0.412 -0.408 -0.798 391s 0.861 0.731 -0.202 391s -0.466 -0.037 0.769 391s 391s y: [4, 3, 4, 1] 391s -0.524 0.519 -0.118 -0.437 391s -0.242 -0.059 -0.795 -0.282 391s 0.699 -0.453 0.675 0.659 391s testing: mul_mat, [4, 3, 4, 4] = [4, 4, 4, 4] * [4, 3, 4, 4] 391s x0: [4, 4, 4, 4] 391s 0.856 -0.694 -0.671 -0.250 391s -0.697 0.344 0.021 0.216 391s -0.379 0.899 0.795 0.458 391s -0.817 0.912 0.789 -0.473 391s 391s x1: [4, 3, 4, 4] 391s 0.064 0.166 -0.180 -0.283 391s 0.968 0.630 -0.777 -0.453 391s -0.330 -0.824 0.102 0.801 391s 391s y: [4, 3, 4, 4] 391s 0.132 -0.053 -0.148 0.091 391s 1.027 -0.572 -0.626 -0.615 391s 0.021 0.121 -0.168 -0.781 391s test-mul-mat0: iter:345/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.316 0.435 0.550 391s -0.547 0.817 0.259 391s 391s x1: [3, 3, 1, 1] 391s 0.376 0.053 0.155 391s 0.529 0.112 -0.319 391s -0.980 -0.862 0.804 391s 391s y: [2, 3, 1, 1] 391s -0.011 -0.123 391s -0.294 -0.280 391s 0.377 0.040 391s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.169 -0.413 0.014 391s -0.778 0.742 -0.456 391s 391s x1: [3, 2, 2, 1] 391s 0.934 0.810 0.717 391s 0.251 0.892 -0.994 391s 391s y: [2, 2, 2, 1] 391s -0.166 -0.453 391s -0.340 0.920 391s testing: mul_mat, [3, 2, 2, 1] = [3, 3, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.353 -0.163 0.061 391s 0.164 0.350 -0.301 391s -0.667 0.937 0.713 391s 391s x1: [3, 2, 2, 1] 391s -0.844 0.661 -0.623 391s -0.391 0.478 0.636 391s 391s y: [3, 2, 2, 1] 391s -0.444 0.281 0.738 391s -0.178 -0.088 1.162 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.934 0.287 -0.549 391s -0.628 0.209 -0.441 391s 0.098 -0.842 -0.143 391s 391s x1: [3, 3, 1, 1] 391s 0.518 0.988 -0.975 391s -0.872 0.466 0.661 391s -0.329 -0.246 0.013 391s 391s y: [3, 3, 1, 1] 391s 1.303 0.311 -0.641 391s -1.043 0.354 -0.572 391s -0.385 0.149 0.173 391s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 3, 2, 1] 391s -0.189 -0.256 -0.969 391s 0.992 -0.636 0.573 391s 0.374 -0.528 -0.370 391s 391s x1: [3, 3, 2, 1] 391s -0.216 -0.504 -0.430 391s 0.339 -0.825 -0.174 391s 0.229 -0.725 0.646 391s 391s y: [3, 3, 2, 1] 391s 0.587 -0.141 0.345 391s 0.315 0.762 0.627 391s -0.484 1.058 0.229 391s testing: mul_mat, [4, 3, 2, 1] = [3, 4, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.914 -0.819 0.700 391s -0.952 0.644 -0.732 391s -0.843 0.663 0.612 391s -0.603 -0.191 -0.417 391s 391s x1: [3, 3, 2, 1] 391s 0.835 0.079 -0.472 391s -0.574 0.681 0.822 391s -0.464 -0.728 0.140 391s 391s y: [4, 3, 2, 1] 391s 0.369 -0.399 -0.940 -0.322 391s -0.508 0.384 1.439 -0.126 391s 0.270 -0.129 -0.005 0.361 391s test-mul-mat0: iter:346/500 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.910 391s -0.174 391s -0.742 391s 391s x1: [1, 2, 1, 1] 391s 0.996 391s -0.222 391s 391s y: [3, 2, 1, 1] 391s 0.906 -0.173 -0.739 391s -0.202 0.039 0.165 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.816 391s 391s x1: [1, 3, 1, 1] 391s -0.693 391s 0.007 391s 0.902 391s 391s y: [1, 3, 1, 1] 391s -0.566 391s 0.006 391s 0.736 391s testing: mul_mat, [2, 1, 1, 4] = [1, 2, 1, 4] * [1, 1, 1, 4] 391s x0: [1, 2, 1, 4] 391s -0.880 391s 0.296 391s 391s x1: [1, 1, 1, 4] 391s 0.602 391s 391s y: [2, 1, 1, 4] 391s -0.530 0.179 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.289 391s -0.778 391s -0.202 391s 0.863 391s 391s x1: [1, 2, 1, 1] 391s 0.771 391s -0.890 391s 391s y: [4, 2, 1, 1] 391s 0.223 -0.600 -0.156 0.666 391s -0.258 0.692 0.180 -0.768 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.778 -0.876 0.726 0.911 391s -0.565 -0.596 -0.358 -0.018 391s 0.409 0.225 -0.346 -0.227 391s 391s x1: [4, 1, 1, 1] 391s -0.782 0.576 0.964 -0.872 391s 391s y: [3, 1, 1, 1] 391s 0.009 -0.231 -0.327 391s testing: mul_mat, [4, 4, 1, 4] = [3, 4, 1, 4] * [3, 4, 1, 4] 391s x0: [3, 4, 1, 4] 391s 0.929 -0.119 -0.980 391s 0.862 0.459 -0.705 391s 0.539 0.995 -0.718 391s -0.650 0.626 0.564 391s 391s x1: [3, 4, 1, 4] 391s 0.207 0.852 0.266 391s 0.437 0.786 -0.597 391s 0.740 0.557 -0.487 391s -0.765 -0.154 -0.265 391s 391s y: [4, 4, 1, 4] 391s -0.169 0.382 0.769 0.548 391s 0.898 1.159 1.447 -0.130 391s 1.099 1.237 1.303 -0.408 391s -0.433 -0.543 -0.375 0.251 391s test-mul-mat0: iter:347/500 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.462 -0.776 0.609 391s 391s x1: [3, 4, 1, 1] 391s -0.218 -0.785 0.685 391s -0.528 -0.714 -0.557 391s 0.105 0.275 -0.832 391s 0.761 -0.807 0.755 391s 391s y: [1, 4, 1, 1] 391s 0.926 391s -0.028 391s -0.671 391s 1.437 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.305 -0.479 0.210 391s 0.216 -0.280 0.843 391s 0.183 0.060 -0.077 391s 0.636 -0.719 0.705 391s 391s x1: [3, 1, 1, 1] 391s -0.274 -0.681 0.283 391s 391s y: [4, 1, 1, 1] 391s 0.470 0.370 -0.113 0.515 391s testing: mul_mat, [4, 4, 1, 3] = [3, 4, 1, 3] * [3, 4, 1, 3] 391s x0: [3, 4, 1, 3] 391s 0.006 0.424 -0.739 391s -0.389 0.757 0.260 391s 0.678 -0.687 0.868 391s 0.471 0.188 0.758 391s 391s x1: [3, 4, 1, 3] 391s 0.851 0.966 -0.823 391s -0.863 -0.591 0.283 391s 0.412 -0.423 0.043 391s 0.605 -0.668 -0.001 391s 391s y: [4, 4, 1, 3] 391s 1.023 0.186 -0.800 -0.041 391s -0.465 -0.038 0.066 -0.303 391s -0.209 -0.469 0.608 0.147 391s -0.279 -0.741 0.868 0.159 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.282 0.299 0.074 391s 0.748 -0.132 0.445 391s 0.037 0.003 0.406 391s 391s x1: [3, 4, 1, 1] 391s -0.551 0.466 -0.255 391s -0.684 -0.063 0.933 391s -0.926 -0.516 -0.309 391s -0.756 0.318 -0.765 391s 391s y: [3, 4, 1, 1] 391s -0.035 -0.587 -0.123 391s -0.143 -0.088 0.354 391s -0.438 -0.762 -0.161 391s -0.175 -0.948 -0.338 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.953 -0.016 0.468 391s -0.489 -0.556 0.135 391s 0.665 0.739 0.983 391s 391s x1: [3, 3, 1, 1] 391s -0.217 0.616 -0.342 391s 0.711 -0.207 0.333 391s -0.100 -0.419 0.766 391s 391s y: [3, 3, 1, 1] 391s 0.036 -0.283 -0.025 391s -0.518 -0.187 0.647 391s 0.461 0.385 0.376 391s testing: mul_mat, [3, 3, 1, 3] = [3, 3, 1, 3] * [3, 3, 1, 3] 391s x0: [3, 3, 1, 3] 391s -0.081 0.706 0.775 391s -0.180 -0.810 0.308 391s -0.430 -0.903 -0.338 391s 391s x1: [3, 3, 1, 3] 391s -0.213 0.370 0.527 391s -0.931 0.118 -0.436 391s 0.368 0.986 0.567 391s 391s y: [3, 3, 1, 3] 391s 0.687 -0.099 -0.421 391s -0.179 -0.063 0.441 391s 1.106 -0.690 -1.240 391s test-mul-mat0: iter:348/500 391s testing: mul_mat, [3, 4, 1, 1] = [2, 3, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.493 0.068 391s 0.763 0.078 391s 0.793 -0.035 391s 391s x1: [2, 4, 1, 1] 391s -0.133 -0.725 391s -0.225 -0.427 391s -0.535 -0.127 391s -0.651 0.773 391s 391s y: [3, 4, 1, 1] 391s -0.115 -0.158 -0.080 391s -0.140 -0.205 -0.163 391s -0.272 -0.418 -0.420 391s -0.268 -0.436 -0.543 391s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 1, 3, 1] 391s 0.091 -0.915 391s 391s x1: [2, 3, 3, 1] 391s 0.890 0.565 391s 0.798 0.637 391s 0.176 0.596 391s 391s y: [1, 3, 3, 1] 391s -0.435 391s -0.510 391s -0.529 391s testing: mul_mat, [4, 1, 3, 4] = [2, 4, 3, 4] * [2, 1, 3, 4] 391s x0: [2, 4, 3, 4] 391s 0.629 -0.815 391s -0.782 0.069 391s -0.956 0.214 391s -0.998 -0.029 391s 391s x1: [2, 1, 3, 4] 391s -0.067 -0.073 391s 391s y: [4, 1, 3, 4] 391s 0.018 0.047 0.048 0.069 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.691 0.718 391s 391s x1: [2, 4, 1, 1] 391s 0.766 0.620 391s 0.262 0.567 391s -0.191 -0.597 391s 0.109 -0.817 391s 391s y: [1, 4, 1, 1] 391s -0.085 391s 0.226 391s -0.296 391s -0.662 391s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.831 391s -0.321 391s 391s x1: [1, 2, 3, 1] 391s 0.957 391s -0.382 391s 391s y: [2, 2, 3, 1] 391s -0.795 -0.307 391s 0.317 0.123 391s testing: mul_mat, [3, 1, 3, 4] = [2, 3, 3, 4] * [2, 1, 3, 4] 391s x0: [2, 3, 3, 4] 391s 0.172 -0.133 391s -0.667 -0.327 391s -0.055 0.063 391s 391s x1: [2, 1, 3, 4] 391s -0.684 0.689 391s 391s y: [3, 1, 3, 4] 391s -0.209 0.231 0.081 391s test-mul-mat0: iter:349/500 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.677 -0.047 0.334 0.573 391s 0.012 0.147 0.005 -0.801 391s 391s x1: [4, 2, 1, 1] 391s -0.771 -0.865 -0.907 0.555 391s 0.051 -0.315 0.662 0.133 391s 391s y: [2, 2, 1, 1] 391s -0.467 -0.585 391s 0.347 -0.150 391s testing: mul_mat, [3, 2, 3, 1] = [4, 3, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.809 0.303 0.665 0.338 391s -0.693 0.865 -0.202 -0.387 391s 0.326 0.936 -0.238 0.186 391s 391s x1: [4, 2, 3, 1] 391s 0.460 -0.694 0.461 0.138 391s -0.851 0.860 0.367 -0.104 391s 391s y: [3, 2, 3, 1] 391s -0.229 -1.065 -0.584 391s 1.158 1.300 0.422 391s testing: mul_mat, [4, 3, 3, 2] = [4, 4, 3, 2] * [4, 3, 3, 2] 391s x0: [4, 4, 3, 2] 391s 0.534 0.707 -0.895 0.918 391s 0.894 -0.938 0.717 -0.188 391s -0.665 0.446 0.872 -0.002 391s -0.526 -0.789 -0.224 -0.637 391s 391s x1: [4, 3, 3, 2] 391s -0.744 0.697 0.362 0.581 391s 0.633 -0.876 -0.232 -0.064 391s 0.782 -0.714 0.847 0.732 391s 391s y: [4, 3, 3, 2] 391s 0.304 -1.168 1.121 -0.610 391s -0.131 1.233 -1.014 0.451 391s -0.173 1.839 -0.101 -0.504 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.841 -0.018 -0.642 0.178 391s -0.976 0.920 -0.217 -0.765 391s 0.374 -0.355 -0.409 -0.701 391s 391s x1: [4, 4, 1, 1] 391s -0.524 0.395 0.870 0.968 391s 0.233 -0.903 0.628 -0.753 391s 0.445 0.931 -0.353 -0.310 391s 0.240 -0.295 0.893 0.567 391s 391s y: [3, 4, 1, 1] 391s 0.047 -0.054 -1.371 391s -0.717 -0.618 0.679 391s -0.219 0.736 0.198 391s -0.670 -1.133 -0.568 391s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.187 -0.711 0.322 391s 0.909 0.728 -0.410 391s 391s x1: [3, 4, 3, 1] 391s -0.961 -0.999 -0.225 391s 0.435 0.871 -0.257 391s -0.332 0.967 -0.630 391s -0.085 0.412 -0.699 391s 391s y: [2, 4, 3, 1] 391s 0.817 -1.509 391s -0.783 1.135 391s -0.829 0.660 391s -0.502 0.509 391s testing: mul_mat, [2, 3, 3, 2] = [2, 2, 3, 2] * [2, 3, 3, 2] 391s x0: [2, 2, 3, 2] 391s -0.055 0.053 391s -0.636 -0.955 391s 391s x1: [2, 3, 3, 2] 391s 0.422 -0.797 391s 0.259 0.524 391s -0.125 0.329 391s 391s y: [2, 3, 3, 2] 391s -0.065 0.492 391s 0.013 -0.665 391s 0.024 -0.235 391s test-mul-mat0: iter:350/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.760 391s 0.294 391s 0.371 391s 0.293 391s 391s x1: [1, 1, 1, 1] 391s -0.651 391s 391s y: [4, 1, 1, 1] 391s 0.495 -0.191 -0.242 -0.191 391s testing: mul_mat, [1, 4, 3, 1] = [1, 1, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 1, 3, 1] 391s 0.086 391s 391s x1: [1, 4, 3, 1] 391s 0.339 391s 0.467 391s 0.689 391s -0.599 391s 391s y: [1, 4, 3, 1] 391s 0.029 391s 0.040 391s 0.059 391s -0.051 391s testing: mul_mat, [2, 1, 3, 4] = [1, 2, 3, 4] * [1, 1, 3, 4] 391s x0: [1, 2, 3, 4] 391s -0.025 391s -0.516 391s 391s x1: [1, 1, 3, 4] 391s -0.159 391s 391s y: [2, 1, 3, 4] 391s 0.004 0.082 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.995 391s -0.133 391s -0.986 391s 0.646 391s 391s x1: [1, 2, 1, 1] 391s 0.657 391s 0.263 391s 391s y: [4, 2, 1, 1] 391s -0.654 -0.088 -0.648 0.425 391s -0.262 -0.035 -0.259 0.170 391s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.908 -0.650 -0.351 0.316 391s 0.868 0.282 -0.208 -0.336 391s 0.608 -0.404 -0.745 0.912 391s 391s x1: [4, 1, 3, 1] 391s 0.842 -0.502 -0.374 -0.716 391s 391s y: [3, 1, 3, 1] 391s -0.534 0.908 0.341 391s testing: mul_mat, [1, 4, 3, 4] = [3, 1, 3, 4] * [3, 4, 3, 4] 391s x0: [3, 1, 3, 4] 391s -0.018 -0.514 0.701 391s 391s x1: [3, 4, 3, 4] 391s -0.530 0.188 0.645 391s -0.275 -0.496 -0.691 391s -0.363 0.082 0.796 391s -0.446 -0.472 0.296 391s 391s y: [1, 4, 3, 4] 391s 0.366 391s -0.225 391s 0.523 391s 0.459 391s test-mul-mat0: iter:351/500 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.618 0.077 0.632 391s -0.821 0.710 -0.274 391s 0.226 0.486 0.748 391s 0.255 0.749 -0.529 391s 391s x1: [3, 4, 1, 1] 391s 0.857 0.249 0.994 391s 0.891 -0.300 -0.147 391s 0.230 -0.925 -0.723 391s -0.474 -0.222 -0.169 391s 391s y: [4, 4, 1, 1] 391s 0.117 -0.799 1.059 -0.121 391s -0.667 -0.905 -0.054 0.080 391s -0.671 -0.647 -0.939 -0.251 391s 0.169 0.277 -0.342 -0.198 391s testing: mul_mat, [4, 4, 4, 1] = [3, 4, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 4, 4, 1] 391s 0.176 0.220 -0.535 391s 0.211 0.646 0.052 391s 0.973 0.298 0.126 391s -0.517 -0.796 -0.052 391s 391s x1: [3, 4, 4, 1] 391s -0.856 0.809 0.465 391s 0.526 0.063 0.844 391s 0.384 -0.687 0.839 391s 0.275 0.013 -0.308 391s 391s y: [4, 4, 4, 1] 391s -0.221 0.366 -0.534 -0.225 391s -0.345 0.196 0.637 -0.366 391s -0.533 -0.319 0.274 0.305 391s 0.216 0.050 0.232 -0.136 391s testing: mul_mat, [2, 4, 4, 2] = [3, 2, 4, 2] * [3, 4, 4, 2] 391s x0: [3, 2, 4, 2] 391s -0.023 0.129 -0.381 391s -0.689 0.132 -0.924 391s 391s x1: [3, 4, 4, 2] 391s 0.968 0.951 0.633 391s -0.749 -0.723 -0.338 391s -0.445 0.422 0.431 391s 0.967 -0.645 -0.029 391s 391s y: [2, 4, 4, 2] 391s -0.141 -1.127 391s 0.053 0.733 391s -0.100 -0.036 391s -0.094 -0.725 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.086 0.587 0.864 391s 0.977 -0.596 0.808 391s 0.901 -0.836 0.556 391s 391s x1: [3, 2, 1, 1] 391s 0.306 -0.287 0.035 391s 0.741 -0.837 -0.659 391s 391s y: [3, 2, 1, 1] 391s -0.112 0.498 0.535 391s -0.997 0.690 1.001 391s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 1, 4, 1] 391s -0.767 -0.510 0.034 391s 391s x1: [3, 3, 4, 1] 391s -0.240 0.523 -0.920 391s 0.732 -0.615 -0.206 391s 0.202 0.609 0.502 391s 391s y: [1, 3, 4, 1] 391s -0.114 391s -0.254 391s -0.449 391s testing: mul_mat, [1, 3, 4, 2] = [1, 1, 4, 2] * [1, 3, 4, 2] 391s x0: [1, 1, 4, 2] 391s 0.619 391s 391s x1: [1, 3, 4, 2] 391s -0.468 391s -0.887 391s -0.970 391s 391s y: [1, 3, 4, 2] 391s -0.290 391s -0.549 391s -0.600 391s test-mul-mat0: iter:352/500 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.374 0.250 0.582 391s 0.102 0.189 -0.446 391s -0.279 0.400 -0.670 391s 391s x1: [3, 4, 1, 1] 391s -0.427 0.513 0.799 391s -0.815 0.930 -0.273 391s -0.544 0.647 -0.848 391s 0.465 -0.606 -0.544 391s 391s y: [3, 4, 1, 1] 391s 0.433 -0.303 -0.211 391s -0.231 0.215 0.782 391s -0.535 0.446 0.979 391s -0.294 0.175 -0.007 391s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 2, 4, 1] 391s 0.562 -0.114 -0.612 391s 0.371 -0.811 -0.566 391s 391s x1: [3, 3, 4, 1] 391s -0.319 0.721 0.352 391s 0.174 -0.845 -0.596 391s 0.236 0.790 -0.753 391s 391s y: [2, 3, 4, 1] 391s -0.477 -0.902 391s 0.559 1.087 391s 0.503 -0.127 391s testing: mul_mat, [4, 2, 4, 3] = [3, 4, 4, 3] * [3, 2, 4, 3] 391s x0: [3, 4, 4, 3] 391s -0.950 -0.419 -0.268 391s 0.951 -0.569 0.112 391s 0.872 0.918 0.259 391s 0.391 -0.445 -0.176 391s 391s x1: [3, 2, 4, 3] 391s 0.596 0.599 -0.876 391s 0.333 0.925 0.219 391s 391s y: [4, 2, 4, 3] 391s -0.583 0.127 0.842 0.120 391s -0.763 -0.185 1.197 -0.320 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.966 0.556 -0.500 391s -0.241 -0.225 0.656 391s -0.006 -0.652 -0.570 391s 0.417 0.549 -0.630 391s 391s x1: [3, 4, 1, 1] 391s -0.694 0.435 -0.454 391s 0.024 0.936 0.577 391s -0.234 0.689 0.924 391s -0.213 0.218 -0.199 391s 391s y: [4, 4, 1, 1] 391s -0.202 -0.228 -0.021 0.235 391s 0.254 0.162 -0.939 0.160 391s -0.305 0.507 -0.975 -0.301 391s 0.015 -0.128 -0.027 0.156 391s testing: mul_mat, [3, 3, 4, 1] = [4, 3, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 3, 4, 1] 391s -0.414 0.252 -0.307 -0.366 391s -0.790 -0.170 0.798 -0.577 391s 0.253 -0.760 -0.223 0.743 391s 391s x1: [4, 3, 4, 1] 391s 0.319 0.102 -0.178 0.271 391s -0.041 0.825 0.577 -0.606 391s -0.629 -0.399 -0.670 0.948 391s 391s y: [3, 3, 4, 1] 391s -0.151 -0.568 0.244 391s 0.270 0.703 -1.216 391s 0.019 -0.517 0.998 391s testing: mul_mat, [2, 4, 4, 3] = [3, 2, 4, 3] * [3, 4, 4, 3] 391s x0: [3, 2, 4, 3] 391s 0.068 -0.610 0.428 391s 0.433 -0.280 -0.832 391s 391s x1: [3, 4, 4, 3] 391s 0.278 -0.335 -0.635 391s 0.538 0.032 -0.870 391s 0.549 -0.324 -0.347 391s 0.330 0.470 -0.910 391s 391s y: [2, 4, 4, 3] 391s -0.048 0.742 391s -0.355 0.948 391s 0.086 0.617 391s -0.653 0.769 391s test-mul-mat0: iter:353/500 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.718 0.003 -0.473 0.574 391s 391s x1: [4, 4, 1, 1] 391s -0.615 0.894 0.189 -0.069 391s 0.463 -0.003 -0.721 0.854 391s 0.131 -0.298 0.510 -0.609 391s -0.133 -0.214 -0.769 0.823 391s 391s y: [1, 4, 1, 1] 391s -0.568 391s 1.164 391s -0.498 391s 0.740 391s testing: mul_mat, [4, 1, 2, 1] = [4, 4, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 4, 2, 1] 391s 0.478 -0.992 0.395 -0.333 391s -0.061 -0.141 0.664 0.218 391s -0.287 -0.206 0.920 -0.777 391s 0.185 -0.213 0.009 0.416 391s 391s x1: [4, 1, 2, 1] 391s 0.834 0.550 -0.693 -0.031 391s 391s y: [4, 1, 2, 1] 391s -0.410 -0.595 -0.966 0.018 391s testing: mul_mat, [4, 4, 2, 3] = [4, 4, 2, 3] * [4, 4, 2, 3] 391s x0: [4, 4, 2, 3] 391s 0.644 0.163 -0.844 -0.748 391s -0.530 -0.590 -0.935 0.722 391s -0.181 0.284 -0.037 -0.012 391s -0.221 0.371 -0.862 -0.382 391s 391s x1: [4, 4, 2, 3] 391s -0.866 0.396 -0.015 0.073 391s -0.746 -0.351 -0.709 -0.033 391s 0.443 -0.789 0.190 -0.372 391s -0.002 -0.800 -0.956 0.608 391s 391s y: [4, 4, 2, 3] 391s -0.535 0.292 0.269 0.323 391s 0.085 1.241 0.062 0.658 391s 0.274 -0.216 -0.307 -0.412 391s 0.221 1.806 -0.199 0.296 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.879 0.629 0.198 0.811 391s 391s x1: [4, 4, 1, 1] 391s -0.267 0.486 -0.630 0.623 391s 0.751 0.062 0.320 0.197 391s -0.426 -0.395 0.193 0.045 391s -0.906 -0.429 0.408 0.405 391s 391s y: [1, 4, 1, 1] 391s 0.920 391s -0.399 391s 0.201 391s 0.936 391s testing: mul_mat, [1, 4, 2, 1] = [1, 1, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.147 391s 391s x1: [1, 4, 2, 1] 391s 0.256 391s 0.960 391s -0.833 391s 0.678 391s 391s y: [1, 4, 2, 1] 391s -0.038 391s -0.141 391s 0.122 391s -0.100 391s testing: mul_mat, [2, 1, 2, 3] = [1, 2, 2, 3] * [1, 1, 2, 3] 391s x0: [1, 2, 2, 3] 391s -0.141 391s 0.071 391s 391s x1: [1, 1, 2, 3] 391s -0.027 391s 391s y: [2, 1, 2, 3] 391s 0.004 -0.002 391s test-mul-mat0: iter:354/500 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.525 -0.373 391s -0.053 0.328 391s -0.850 0.920 391s -0.449 0.296 391s 391s x1: [2, 1, 1, 1] 391s 0.132 0.838 391s 391s y: [4, 1, 1, 1] 391s -0.382 0.268 0.659 0.188 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.366 -0.091 391s -0.433 -0.376 391s -0.659 -0.235 391s 0.080 0.584 391s 391s x1: [2, 4, 1, 1] 391s -0.355 -0.165 391s 0.762 -0.434 391s 0.694 -0.168 391s 0.680 0.598 391s 391s y: [4, 4, 1, 1] 391s -0.115 0.216 0.273 -0.125 391s 0.318 -0.167 -0.400 -0.192 391s 0.269 -0.237 -0.418 -0.042 391s 0.195 -0.519 -0.588 0.404 391s testing: mul_mat, [1, 4, 1, 2] = [2, 1, 1, 2] * [2, 4, 1, 2] 391s x0: [2, 1, 1, 2] 391s -0.469 -0.766 391s 391s x1: [2, 4, 1, 2] 391s -0.780 0.497 391s -0.006 0.352 391s 0.335 -0.547 391s 0.827 0.962 391s 391s y: [1, 4, 1, 2] 391s -0.015 391s -0.266 391s 0.262 391s -1.124 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.259 -0.798 391s -0.435 0.082 391s 391s x1: [2, 1, 1, 1] 391s 0.832 0.998 391s 391s y: [2, 1, 1, 1] 391s -1.013 -0.280 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.660 -0.887 391s -0.098 -0.887 391s 391s x1: [2, 2, 1, 1] 391s 0.331 0.282 391s -0.334 0.550 391s 391s y: [2, 2, 1, 1] 391s -0.032 -0.282 391s -0.708 -0.455 391s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 1, 1, 2] 391s -0.872 -0.236 391s 391s x1: [2, 2, 1, 2] 391s -0.271 -0.925 391s 0.514 0.883 391s 391s y: [1, 2, 1, 2] 391s 0.455 391s -0.656 391s test-mul-mat0: iter:355/500 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.543 0.841 0.572 0.971 391s 0.724 0.759 -0.194 0.312 391s 391s x1: [4, 3, 1, 1] 391s 0.121 -0.065 0.329 0.323 391s -0.983 -0.340 -0.395 -0.316 391s -0.790 0.383 -0.656 0.112 391s 391s y: [2, 3, 1, 1] 391s 0.381 0.075 391s -0.286 -0.992 391s 0.485 -0.119 391s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.981 -0.595 -0.477 0.741 391s 0.210 0.835 0.095 -0.184 391s 0.275 0.213 0.754 0.077 391s -0.903 -0.764 0.074 0.745 391s 391s x1: [4, 2, 3, 1] 391s 0.354 0.606 0.440 -0.881 391s -0.062 0.802 -0.116 -0.519 391s 391s y: [4, 2, 3, 1] 391s -0.876 0.784 0.491 -1.408 391s -0.867 0.741 0.027 -0.952 391s testing: mul_mat, [4, 4, 3, 4] = [4, 4, 3, 4] * [4, 4, 3, 4] 391s x0: [4, 4, 3, 4] 391s -0.743 0.777 0.403 0.184 391s -0.787 -0.646 -0.465 0.679 391s 0.405 0.732 -0.292 0.132 391s 0.114 -0.141 0.696 -0.968 391s 391s x1: [4, 4, 3, 4] 391s 0.750 -0.792 -0.981 -0.296 391s -0.434 -0.689 0.291 -0.230 391s 0.516 0.854 0.932 0.131 391s -0.149 0.893 -0.274 -0.748 391s 391s y: [4, 4, 3, 4] 391s -1.623 0.176 -0.028 -0.199 391s -0.137 0.495 -0.796 0.473 391s 0.679 -1.303 0.579 0.460 391s 0.556 -0.839 0.575 0.391 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.113 -0.979 0.662 0.442 391s 391s x1: [4, 4, 1, 1] 391s 0.136 -0.856 0.470 0.334 391s -0.450 0.874 0.378 -0.725 391s 0.117 -0.257 -0.303 -0.833 391s 0.847 -0.845 -0.946 0.803 391s 391s y: [1, 4, 1, 1] 391s 1.311 391s -0.977 391s -0.304 391s 0.652 391s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 4, 3, 1] 391s -0.285 391s 0.556 391s 0.585 391s -0.907 391s 391s x1: [1, 4, 3, 1] 391s 0.935 391s -0.213 391s 0.043 391s 0.844 391s 391s y: [4, 4, 3, 1] 391s -0.267 0.520 0.547 -0.848 391s 0.061 -0.118 -0.125 0.193 391s -0.012 0.024 0.025 -0.039 391s -0.241 0.469 0.493 -0.765 391s testing: mul_mat, [3, 1, 3, 4] = [4, 3, 3, 4] * [4, 1, 3, 4] 391s x0: [4, 3, 3, 4] 391s -0.377 -0.277 0.525 0.197 391s 0.824 0.325 0.662 -0.318 391s -0.426 0.776 0.508 0.922 391s 391s x1: [4, 1, 3, 4] 391s -0.514 -0.060 -0.963 -0.493 391s 391s y: [3, 1, 3, 4] 391s -0.393 -0.924 -0.771 391s test-mul-mat0: iter:356/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.280 391s 391s x1: [1, 4, 1, 1] 391s 0.245 391s 0.852 391s -0.611 391s 0.630 391s 391s y: [1, 4, 1, 1] 391s 0.069 391s 0.238 391s -0.171 391s 0.176 391s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.664 391s 391s x1: [1, 1, 2, 1] 391s 0.030 391s 391s y: [1, 1, 2, 1] 391s -0.020 391s testing: mul_mat, [1, 1, 2, 3] = [1, 1, 2, 3] * [1, 1, 2, 3] 391s x0: [1, 1, 2, 3] 391s -0.728 391s 391s x1: [1, 1, 2, 3] 391s 0.679 391s 391s y: [1, 1, 2, 3] 391s -0.495 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.722 391s 391s x1: [1, 1, 1, 1] 391s 0.528 391s 391s y: [1, 1, 1, 1] 391s -0.381 391s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.734 391s 391s x1: [1, 1, 2, 1] 391s -0.394 391s 391s y: [1, 1, 2, 1] 391s 0.289 391s testing: mul_mat, [1, 1, 2, 3] = [1, 1, 2, 3] * [1, 1, 2, 3] 391s x0: [1, 1, 2, 3] 391s 0.232 391s 391s x1: [1, 1, 2, 3] 391s -0.221 391s 391s y: [1, 1, 2, 3] 391s -0.051 391s test-mul-mat0: iter:357/500 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.295 0.361 0.471 391s 391s x1: [3, 3, 1, 1] 391s 0.296 0.403 0.401 391s 0.071 -0.866 -0.844 391s -0.401 -0.230 -0.566 391s 391s y: [1, 3, 1, 1] 391s 0.422 391s -0.690 391s -0.468 391s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.235 0.293 0.663 391s 0.237 -0.796 0.548 391s -0.577 -0.285 -0.427 391s 0.866 -0.920 0.869 391s 391s x1: [3, 1, 2, 1] 391s 0.065 -0.859 -0.874 391s 391s y: [4, 1, 2, 1] 391s -0.816 0.221 0.580 0.088 391s testing: mul_mat, [2, 4, 2, 3] = [3, 2, 2, 3] * [3, 4, 2, 3] 391s x0: [3, 2, 2, 3] 391s -0.387 -0.090 0.670 391s -0.455 0.933 -0.147 391s 391s x1: [3, 4, 2, 3] 391s -0.467 0.574 -0.659 391s -0.002 -0.815 -0.598 391s 0.894 0.421 0.695 391s 0.557 -0.342 0.899 391s 391s y: [2, 4, 2, 3] 391s -0.313 0.845 391s -0.327 -0.671 391s 0.082 -0.116 391s 0.417 -0.705 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.530 -0.389 0.505 391s 0.064 0.553 -0.128 391s 0.202 0.336 0.323 391s 391s x1: [3, 2, 1, 1] 391s 0.614 -0.025 -0.136 391s 0.865 -0.054 -0.576 391s 391s y: [3, 2, 1, 1] 391s 0.267 0.043 0.072 391s 0.188 0.099 -0.030 391s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.321 -0.148 0.463 391s 0.498 -0.438 -0.885 391s -0.395 0.807 0.143 391s 391s x1: [3, 3, 2, 1] 391s 0.104 -0.656 0.952 391s -0.516 0.693 0.905 391s -0.858 -0.662 0.422 391s 391s y: [3, 3, 2, 1] 391s 0.571 -0.503 -0.434 391s 0.150 -1.362 0.893 391s 0.017 -0.512 -0.134 391s testing: mul_mat, [2, 3, 2, 3] = [3, 2, 2, 3] * [3, 3, 2, 3] 391s x0: [3, 2, 2, 3] 391s 0.693 0.576 0.873 391s 0.905 0.140 0.973 391s 391s x1: [3, 3, 2, 3] 391s -0.302 0.363 0.744 391s 0.931 -0.040 0.594 391s 0.926 -0.640 0.804 391s 391s y: [2, 3, 2, 3] 391s 0.650 0.501 391s 1.140 1.415 391s 0.975 1.531 391s test-mul-mat0: iter:358/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.142 391s -0.637 391s 391s x1: [1, 4, 1, 1] 391s -0.611 391s 0.097 391s 0.611 391s 0.920 391s 391s y: [2, 4, 1, 1] 391s -0.087 0.389 391s 0.014 -0.062 391s 0.087 -0.389 391s 0.131 -0.586 391s testing: mul_mat, [3, 2, 3, 1] = [1, 3, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 3, 3, 1] 391s 0.704 391s -0.490 391s -0.721 391s 391s x1: [1, 2, 3, 1] 391s -0.527 391s -0.512 391s 391s y: [3, 2, 3, 1] 391s -0.371 0.258 0.380 391s -0.360 0.251 0.369 391s testing: mul_mat, [1, 3, 3, 3] = [1, 1, 3, 3] * [1, 3, 3, 3] 391s x0: [1, 1, 3, 3] 391s -0.898 391s 391s x1: [1, 3, 3, 3] 391s -0.955 391s 0.515 391s 0.478 391s 391s y: [1, 3, 3, 3] 391s 0.858 391s -0.462 391s -0.429 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.687 391s 0.564 391s 391s x1: [1, 1, 1, 1] 391s 0.040 391s 391s y: [2, 1, 1, 1] 391s 0.027 0.023 391s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.728 0.984 391s 0.608 0.985 391s 391s x1: [2, 1, 3, 1] 391s -0.261 0.441 391s 391s y: [2, 1, 3, 1] 391s 0.624 0.276 391s testing: mul_mat, [2, 2, 3, 3] = [2, 2, 3, 3] * [2, 2, 3, 3] 391s x0: [2, 2, 3, 3] 391s -0.064 0.819 391s 0.377 0.951 391s 391s x1: [2, 2, 3, 3] 391s 0.081 -0.223 391s -0.410 0.064 391s 391s y: [2, 2, 3, 3] 391s -0.188 -0.182 391s 0.079 -0.094 391s test-mul-mat0: iter:359/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.234 0.701 391s -0.693 -0.635 391s 0.669 0.884 391s 391s x1: [2, 1, 1, 1] 391s 0.432 0.713 391s 391s y: [3, 1, 1, 1] 391s 0.398 -0.752 0.919 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.627 -0.889 391s 391s x1: [2, 3, 1, 1] 391s -0.034 0.826 391s 0.990 -0.093 391s -0.215 -0.527 391s 391s y: [1, 3, 1, 1] 391s -0.756 391s 0.703 391s 0.333 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.405 -0.859 391s 0.440 0.898 391s 391s x1: [2, 1, 1, 1] 391s -0.269 0.416 391s 391s y: [2, 1, 1, 1] 391s -0.466 0.255 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.589 0.396 391s -0.370 -0.971 391s -0.671 -0.359 391s -0.666 0.762 391s 391s x1: [2, 2, 1, 1] 391s -0.183 -0.106 391s 0.016 0.198 391s 391s y: [4, 2, 1, 1] 391s 0.066 0.170 0.160 0.041 391s 0.069 -0.198 -0.082 0.140 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.937 0.186 -0.699 0.510 391s 0.099 -0.597 -0.689 -0.864 391s -0.002 0.494 0.521 0.729 391s 0.353 -0.374 0.318 -0.860 391s 391s x1: [4, 2, 1, 1] 391s 0.699 -0.475 -0.273 0.525 391s -0.486 0.635 -0.690 -0.012 391s 391s y: [4, 2, 1, 1] 391s 1.024 0.087 0.005 -0.114 391s 0.139 0.059 -0.054 -0.617 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.075 -0.188 0.162 -0.324 391s 0.318 0.641 -0.053 -0.227 391s 0.015 0.629 -0.367 0.763 391s 0.328 0.123 0.298 -0.261 391s 391s x1: [4, 4, 1, 1] 391s -0.535 -0.937 0.536 -0.506 391s -0.296 0.298 -0.807 0.228 391s -0.975 0.718 0.743 0.660 391s -0.972 -0.270 0.927 0.965 391s 391s y: [4, 4, 1, 1] 391s 0.467 -0.684 -1.180 0.001 391s -0.238 0.088 0.654 -0.361 391s -0.156 -0.039 0.668 -0.183 391s -0.039 -0.750 0.212 -0.328 391s test-mul-mat0: iter:360/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.987 0.472 -0.709 0.628 391s 0.101 0.414 -0.210 -0.952 391s -0.953 -0.912 -0.276 -0.180 391s 391s x1: [4, 3, 1, 1] 391s -0.549 -0.329 0.632 -0.831 391s -0.587 0.292 -0.803 0.144 391s 0.219 -0.838 -0.027 -0.856 391s 391s y: [3, 3, 1, 1] 391s -1.668 0.466 0.799 391s 0.218 0.093 0.488 391s -0.698 0.496 0.717 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.133 -0.121 0.087 -0.894 391s 0.023 -0.433 -0.907 -0.505 391s -0.141 0.722 0.597 -0.727 391s -0.488 0.645 -0.680 -0.401 391s 391s x1: [4, 3, 1, 1] 391s 0.851 0.463 0.129 -0.892 391s -0.095 -0.264 -0.441 0.576 391s -0.632 -0.272 0.990 0.660 391s 391s y: [4, 3, 1, 1] 391s 0.866 0.153 0.939 0.153 391s -0.534 0.221 -0.859 -0.055 391s -0.555 -1.127 0.003 -0.804 391s testing: mul_mat, [2, 4, 1, 4] = [4, 2, 1, 4] * [4, 4, 1, 4] 391s x0: [4, 2, 1, 4] 391s -0.420 -0.489 -0.074 -0.811 391s -0.199 0.519 -0.889 -0.537 391s 391s x1: [4, 4, 1, 4] 391s -0.631 0.140 -0.550 0.832 391s -0.731 -0.442 -0.263 0.004 391s 0.117 -0.687 0.372 0.845 391s -0.698 0.032 -0.230 0.436 391s 391s y: [2, 4, 1, 4] 391s -0.437 0.241 391s 0.539 0.148 391s -0.426 -1.164 391s -0.059 0.126 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.195 -0.928 0.090 -0.194 391s 0.842 0.056 -0.465 0.116 391s 0.859 -0.739 -0.711 0.612 391s -0.042 0.794 0.320 0.746 391s 391s x1: [4, 2, 1, 1] 391s -0.227 0.200 0.417 0.574 391s -0.281 0.528 -0.963 -0.324 391s 391s y: [4, 2, 1, 1] 391s -0.303 -0.308 -0.288 0.730 391s -0.569 0.203 -0.145 -0.118 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.710 -0.051 0.219 0.346 391s 391s x1: [4, 4, 1, 1] 391s -0.849 -0.564 -0.607 -0.587 391s -0.290 0.622 0.185 0.910 391s 0.039 -0.241 -0.371 -0.432 391s -0.204 0.306 -0.569 0.991 391s 391s y: [1, 4, 1, 1] 391s 0.295 391s 0.529 391s -0.246 391s 0.347 391s testing: mul_mat, [2, 4, 1, 4] = [1, 2, 1, 4] * [1, 4, 1, 4] 391s x0: [1, 2, 1, 4] 391s -0.576 391s 0.352 391s 391s x1: [1, 4, 1, 4] 391s 0.210 391s 0.013 391s -0.564 391s 0.746 391s 391s y: [2, 4, 1, 4] 391s -0.121 0.074 391s -0.008 0.005 391s 0.325 -0.199 391s -0.429 0.262 391s test-mul-mat0: iter:361/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.873 391s 0.739 391s 391s x1: [1, 1, 1, 1] 391s -0.274 391s 391s y: [2, 1, 1, 1] 391s -0.239 -0.203 391s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 3, 4, 1] 391s 0.349 391s 0.160 391s 0.010 391s 391s x1: [1, 2, 4, 1] 391s -0.578 391s 0.619 391s 391s y: [3, 2, 4, 1] 391s -0.202 -0.092 -0.006 391s 0.216 0.099 0.006 391s testing: mul_mat, [2, 3, 4, 1] = [1, 2, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 2, 4, 1] 391s -0.681 391s 0.865 391s 391s x1: [1, 3, 4, 1] 391s 0.700 391s 0.094 391s 0.540 391s 391s y: [2, 3, 4, 1] 391s -0.476 0.605 391s -0.064 0.082 391s -0.368 0.467 391s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.614 391s 391s x1: [1, 2, 1, 1] 391s 0.209 391s 0.075 391s 391s y: [1, 2, 1, 1] 391s -0.128 391s -0.046 391s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 1, 4, 1] 391s -0.683 391s 391s x1: [1, 1, 4, 1] 391s 0.362 391s 391s y: [1, 1, 4, 1] 391s -0.247 391s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 2, 4, 1] 391s -0.864 391s 0.210 391s 391s x1: [1, 1, 4, 1] 391s 0.608 391s 391s y: [2, 1, 4, 1] 391s -0.526 0.128 391s test-mul-mat0: iter:362/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.884 0.059 -0.725 391s -0.626 0.762 -0.192 391s 391s x1: [3, 1, 1, 1] 391s 0.520 0.497 0.697 391s 391s y: [2, 1, 1, 1] 391s -0.016 -0.081 391s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 1, 4, 1] 391s -0.798 -0.565 -0.422 391s 391s x1: [3, 2, 4, 1] 391s -0.309 -0.836 -0.564 391s -0.452 0.772 0.798 391s 391s y: [1, 2, 4, 1] 391s 0.957 391s -0.413 391s testing: mul_mat, [3, 1, 4, 4] = [3, 3, 4, 4] * [3, 1, 4, 4] 391s x0: [3, 3, 4, 4] 391s -0.842 -0.609 -0.316 391s 0.166 0.252 0.686 391s 0.559 0.379 0.049 391s 391s x1: [3, 1, 4, 4] 391s -0.576 0.446 -0.499 391s 391s y: [3, 1, 4, 4] 391s 0.371 -0.326 -0.177 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.276 0.347 -0.157 391s 0.355 -0.771 0.689 391s 391s x1: [3, 3, 1, 1] 391s -0.816 -0.674 0.213 391s -0.730 -0.328 -0.732 391s 0.650 0.985 0.149 391s 391s y: [2, 3, 1, 1] 391s -0.493 0.376 391s -0.201 -0.511 391s 0.498 -0.426 391s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.969 -0.802 391s 0.228 -0.929 391s -0.879 0.809 391s 391s x1: [2, 3, 4, 1] 391s -0.113 0.581 391s -0.993 -0.301 391s 0.326 0.773 391s 391s y: [3, 3, 4, 1] 391s -0.576 -0.566 0.570 391s -0.720 0.053 0.629 391s -0.304 -0.644 0.339 391s testing: mul_mat, [1, 2, 4, 4] = [3, 1, 4, 4] * [3, 2, 4, 4] 391s x0: [3, 1, 4, 4] 391s 0.585 0.220 -0.946 391s 391s x1: [3, 2, 4, 4] 391s -0.542 -0.552 -0.919 391s 0.268 0.141 0.954 391s 391s y: [1, 2, 4, 4] 391s 0.430 391s -0.714 391s test-mul-mat0: iter:363/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.774 391s -0.203 391s -0.595 391s 0.195 391s 391s x1: [1, 1, 1, 1] 391s -0.838 391s 391s y: [4, 1, 1, 1] 391s 0.649 0.170 0.499 -0.163 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.225 391s -0.801 391s 391s x1: [1, 4, 1, 1] 391s -0.671 391s -0.981 391s -0.625 391s 0.237 391s 391s y: [2, 4, 1, 1] 391s 0.151 0.537 391s 0.220 0.785 391s 0.140 0.501 391s -0.053 -0.190 391s testing: mul_mat, [1, 2, 1, 2] = [1, 1, 1, 2] * [1, 2, 1, 2] 391s x0: [1, 1, 1, 2] 391s -0.211 391s 391s x1: [1, 2, 1, 2] 391s 0.581 391s -0.890 391s 391s y: [1, 2, 1, 2] 391s -0.122 391s 0.188 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.382 391s 391s x1: [1, 1, 1, 1] 391s 0.138 391s 391s y: [1, 1, 1, 1] 391s -0.053 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.813 391s -0.285 391s 391s x1: [1, 1, 1, 1] 391s -0.349 391s 391s y: [2, 1, 1, 1] 391s -0.284 0.099 391s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 391s x0: [2, 4, 1, 2] 391s 0.407 -0.269 391s 0.536 0.206 391s -0.123 0.831 391s 0.812 0.106 391s 391s x1: [2, 1, 1, 2] 391s 0.328 0.918 391s 391s y: [4, 1, 1, 2] 391s -0.114 0.365 0.723 0.364 391s test-mul-mat0: iter:364/500 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.659 -0.854 -0.327 391s 0.547 0.977 0.779 391s 391s x1: [3, 4, 1, 1] 391s -0.015 0.651 0.833 391s 0.768 0.464 -0.452 391s 0.096 0.383 -0.073 391s -0.594 0.463 -0.666 391s 391s y: [2, 4, 1, 1] 391s -0.819 1.276 391s -0.755 0.522 391s -0.367 0.370 391s 0.213 -0.390 391s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 4, 4, 1] 391s 0.846 -0.337 -0.907 391s 0.393 -0.360 0.872 391s 0.383 0.667 0.531 391s -0.984 0.326 0.181 391s 391s x1: [3, 2, 4, 1] 391s 0.990 0.027 0.659 391s -0.367 0.659 0.650 391s 391s y: [4, 2, 4, 1] 391s 0.230 0.954 0.748 -0.846 391s -1.122 0.186 0.644 0.694 391s testing: mul_mat, [4, 4, 4, 1] = [3, 4, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.899 -0.513 0.780 391s -0.013 -0.077 0.033 391s -0.416 -0.042 0.253 391s -0.272 0.467 0.210 391s 391s x1: [3, 4, 4, 1] 391s -0.775 -0.285 -0.500 391s -0.317 0.667 0.733 391s -0.603 0.691 0.008 391s -0.707 0.193 0.788 391s 391s y: [4, 4, 4, 1] 391s 0.453 0.015 0.208 -0.027 391s 0.515 -0.023 0.289 0.552 391s 0.194 -0.046 0.223 0.489 391s 1.151 0.020 0.485 0.448 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.367 -0.737 -0.504 391s 391s x1: [3, 4, 1, 1] 391s -0.088 0.186 -0.633 391s 0.162 -0.835 0.905 391s -0.365 -0.757 -0.742 391s 0.447 0.832 0.210 391s 391s y: [1, 4, 1, 1] 391s 0.150 391s 0.218 391s 0.798 391s -0.555 391s testing: mul_mat, [4, 3, 4, 1] = [1, 4, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 4, 4, 1] 391s 0.945 391s -0.797 391s 0.045 391s 0.131 391s 391s x1: [1, 3, 4, 1] 391s -0.319 391s -0.697 391s 0.187 391s 391s y: [4, 3, 4, 1] 391s -0.302 0.254 -0.014 -0.042 391s -0.659 0.555 -0.031 -0.092 391s 0.177 -0.149 0.008 0.025 391s testing: mul_mat, [4, 1, 4, 1] = [4, 4, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 4, 4, 1] 391s 0.697 -0.481 -0.261 -0.951 391s 0.222 -0.995 0.033 -0.831 391s 0.529 -0.255 0.430 -0.110 391s 0.904 -0.941 -0.317 0.561 391s 391s x1: [4, 1, 4, 1] 391s -0.793 0.512 0.419 -0.490 391s 391s y: [4, 1, 4, 1] 391s -0.443 -0.265 -0.316 -1.607 391s test-mul-mat0: iter:365/500 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.943 -0.882 -0.119 391s 391s x1: [3, 4, 1, 1] 391s 0.353 -0.742 0.993 391s 0.019 -0.722 0.948 391s -0.266 -0.533 -0.208 391s -0.990 -0.756 0.036 391s 391s y: [1, 4, 1, 1] 391s 0.203 391s 0.506 391s 0.746 391s 1.596 391s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.985 -0.278 0.303 391s -0.727 -0.285 -0.678 391s -0.449 -0.337 0.056 391s 391s x1: [3, 1, 4, 1] 391s -0.015 0.753 0.499 391s 391s y: [3, 1, 4, 1] 391s -0.043 -0.542 -0.219 391s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.174 -0.733 -0.944 391s -0.386 -0.285 0.678 391s 0.895 0.468 0.664 391s 391s x1: [3, 3, 4, 1] 391s 0.053 -0.152 0.068 391s -0.891 0.866 -0.478 391s 0.175 0.129 0.014 391s 391s y: [3, 3, 4, 1] 391s 0.038 0.069 0.021 391s -0.029 -0.227 -0.709 391s -0.138 -0.095 0.226 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.779 0.387 -0.022 391s 0.963 0.120 0.811 391s 391s x1: [3, 3, 1, 1] 391s 0.135 -0.968 0.709 391s -0.202 -0.799 0.781 391s -0.103 0.795 -0.293 391s 391s y: [2, 3, 1, 1] 391s -0.286 0.588 391s -0.484 0.343 391s 0.234 -0.242 391s testing: mul_mat, [3, 3, 4, 1] = [2, 3, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.690 -0.190 391s 0.964 -0.059 391s 0.871 0.682 391s 391s x1: [2, 3, 4, 1] 391s 0.508 -0.518 391s -0.341 0.348 391s 0.527 -0.221 391s 391s y: [3, 3, 4, 1] 391s 0.449 0.520 0.089 391s -0.301 -0.349 -0.060 391s 0.406 0.521 0.308 391s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 1, 4, 1] 391s -0.876 0.505 -0.471 391s 391s x1: [3, 2, 4, 1] 391s 0.240 -0.168 0.329 391s -0.366 -0.258 0.329 391s 391s y: [1, 2, 4, 1] 391s -0.450 391s 0.036 391s test-mul-mat0: iter:366/500 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.015 -0.341 -0.144 391s 391s x1: [3, 2, 1, 1] 391s 0.470 -0.250 0.517 391s 0.030 -0.593 -0.472 391s 391s y: [1, 2, 1, 1] 391s 0.018 391s 0.271 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.953 -0.941 0.003 391s -0.518 -0.075 0.225 391s 0.046 0.031 -0.721 391s 391s x1: [3, 1, 1, 1] 391s -0.204 0.946 0.935 391s 391s y: [3, 1, 1, 1] 391s -1.081 0.245 -0.655 391s testing: mul_mat, [1, 3, 1, 4] = [3, 1, 1, 4] * [3, 3, 1, 4] 391s x0: [3, 1, 1, 4] 391s -0.614 0.738 0.803 391s 391s x1: [3, 3, 1, 4] 391s -0.060 0.771 -0.925 391s -0.837 0.628 -0.697 391s -0.321 -0.526 0.773 391s 391s y: [1, 3, 1, 4] 391s -0.137 391s 0.418 391s 0.430 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.625 0.183 0.787 391s 0.343 -0.822 -0.957 391s 0.442 -0.183 -0.531 391s 391s x1: [3, 1, 1, 1] 391s -0.852 -0.452 0.433 391s 391s y: [3, 1, 1, 1] 391s -0.274 -0.335 -0.524 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.590 -0.698 0.538 391s -0.364 0.390 -0.370 391s -0.111 -0.918 0.318 391s -0.343 0.885 -0.168 391s 391s x1: [3, 3, 1, 1] 391s 0.332 0.680 -0.803 391s 0.039 -0.056 -0.773 391s 0.425 -0.318 -0.970 391s 391s y: [4, 3, 1, 1] 391s -1.102 0.442 -0.916 0.623 391s -0.400 0.250 -0.199 0.067 391s -0.551 0.080 -0.063 -0.263 391s testing: mul_mat, [2, 3, 1, 4] = [4, 2, 1, 4] * [4, 3, 1, 4] 391s x0: [4, 2, 1, 4] 391s -0.683 -0.762 -0.357 -0.995 391s 0.388 0.466 0.190 0.981 391s 391s x1: [4, 3, 1, 4] 391s -0.745 -0.339 -0.726 0.437 391s -0.162 0.091 0.225 -0.118 391s 0.560 -0.443 -0.439 0.757 391s 391s y: [2, 3, 1, 4] 391s 0.592 -0.156 391s 0.079 -0.094 391s -0.641 0.670 391s test-mul-mat0: iter:367/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.118 0.635 -0.329 391s -0.890 0.806 0.668 391s 391s x1: [3, 1, 1, 1] 391s -0.538 0.035 0.687 391s 391s y: [2, 1, 1, 1] 391s -0.267 0.966 391s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.056 -0.655 0.371 391s -0.491 0.408 -0.357 391s 391s x1: [3, 2, 2, 1] 391s -0.715 0.636 -0.682 391s -0.266 -0.650 0.504 391s 391s y: [2, 2, 2, 1] 391s -0.629 0.853 391s 0.627 -0.314 391s testing: mul_mat, [4, 2, 2, 2] = [3, 4, 2, 2] * [3, 2, 2, 2] 391s x0: [3, 4, 2, 2] 391s 0.419 0.080 -0.815 391s 0.993 0.880 0.780 391s 0.686 -0.685 -0.954 391s 0.488 -0.444 0.979 391s 391s x1: [3, 2, 2, 2] 391s 0.693 -0.565 -0.734 391s 0.803 -0.759 0.934 391s 391s y: [4, 2, 2, 2] 391s 0.843 -0.382 1.562 -0.130 391s -0.486 0.857 0.179 1.644 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.125 -0.050 0.610 391s 0.132 -0.395 -0.821 391s 391s x1: [3, 4, 1, 1] 391s 0.754 0.865 -0.421 391s 0.284 -0.169 -0.605 391s 0.028 -0.138 0.581 391s 0.331 0.329 -0.353 391s 391s y: [2, 4, 1, 1] 391s -0.395 0.103 391s -0.396 0.601 391s 0.358 -0.419 391s -0.273 0.203 391s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.854 -0.479 391s -0.874 0.041 391s -0.988 0.839 391s -0.023 -0.898 391s 391s x1: [2, 3, 2, 1] 391s -0.885 0.773 391s 0.724 0.671 391s 0.425 0.825 391s 391s y: [4, 3, 2, 1] 391s 0.386 0.804 1.522 -0.674 391s -0.940 -0.605 -0.152 -0.620 391s -0.758 -0.338 0.273 -0.750 391s testing: mul_mat, [1, 2, 2, 2] = [4, 1, 2, 2] * [4, 2, 2, 2] 391s x0: [4, 1, 2, 2] 391s 0.205 -0.309 0.585 0.640 391s 391s x1: [4, 2, 2, 2] 391s 0.492 0.100 -0.237 0.265 391s -0.175 -0.565 -0.311 -0.351 391s 391s y: [1, 2, 2, 2] 391s 0.100 391s -0.268 391s test-mul-mat0: iter:368/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.502 -0.011 -0.034 391s -0.842 0.431 -0.607 391s -0.081 -0.364 0.084 391s 391s x1: [3, 1, 1, 1] 391s -0.021 0.759 0.194 391s 391s y: [3, 1, 1, 1] 391s -0.004 0.227 -0.258 391s testing: mul_mat, [4, 3, 3, 1] = [3, 4, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.614 -0.806 -0.625 391s -0.882 -0.530 -0.161 391s -0.143 -0.973 -0.975 391s -0.164 0.481 0.556 391s 391s x1: [3, 3, 3, 1] 391s -0.496 -0.724 -0.536 391s -0.261 0.557 0.186 391s 0.979 0.454 0.531 391s 391s y: [4, 3, 3, 1] 391s 0.614 0.907 1.298 -0.565 391s -0.726 -0.095 -0.686 0.414 391s -0.096 -1.189 -1.099 0.352 391s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.134 0.581 0.689 391s -0.068 0.483 0.964 391s -0.570 -0.835 0.981 391s 0.573 -0.473 0.779 391s 391s x1: [3, 4, 3, 1] 391s 0.019 0.601 0.156 391s 0.044 -0.563 -0.362 391s -0.401 0.046 -0.188 391s 0.798 0.902 -0.725 391s 391s y: [4, 4, 3, 1] 391s 0.459 0.440 -0.359 -0.152 391s -0.571 -0.625 0.090 0.009 391s -0.157 -0.132 0.006 -0.398 391s 0.131 -0.318 -1.920 -0.535 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.294 -0.807 0.532 391s -0.562 -0.841 -0.568 391s -0.115 0.532 0.219 391s -0.129 0.439 0.440 391s 391s x1: [3, 4, 1, 1] 391s 0.432 -0.723 0.137 391s 0.413 0.849 0.664 391s 0.192 0.345 -0.834 391s 0.086 0.672 0.156 391s 391s y: [4, 4, 1, 1] 391s 0.783 0.288 -0.404 -0.313 391s -0.211 -1.323 0.549 0.611 391s -0.666 0.076 -0.022 -0.240 391s -0.434 -0.702 0.381 0.352 391s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.752 0.683 -0.208 -0.230 391s 391s x1: [4, 3, 3, 1] 391s -0.790 -0.866 -0.071 0.677 391s 0.980 0.901 0.109 -0.743 391s 0.039 -0.478 -0.894 -0.298 391s 391s y: [1, 3, 3, 1] 391s -0.139 391s 0.028 391s -0.101 391s testing: mul_mat, [2, 4, 3, 1] = [1, 2, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 2, 3, 1] 391s 0.602 391s -0.643 391s 391s x1: [1, 4, 3, 1] 391s -0.808 391s -0.220 391s -0.152 391s 0.985 391s 391s y: [2, 4, 3, 1] 391s -0.487 0.520 391s -0.133 0.141 391s -0.092 0.098 391s 0.593 -0.633 391s test-mul-mat0: iter:369/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.831 0.704 0.707 391s 0.334 0.640 -0.405 391s -0.069 0.059 -0.265 391s 391s x1: [3, 3, 1, 1] 391s 0.256 -0.766 -0.756 391s -0.504 -0.403 0.528 391s -0.141 -0.017 0.719 391s 391s y: [3, 3, 1, 1] 391s -0.860 -0.098 0.137 391s -0.329 -0.640 -0.129 391s 0.379 -0.349 -0.182 391s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.807 0.724 -0.262 391s 391s x1: [3, 3, 3, 1] 391s 0.841 0.401 -0.662 391s -0.802 0.596 0.079 391s 0.982 0.544 -0.930 391s 391s y: [1, 3, 3, 1] 391s -0.215 391s 1.057 391s -0.155 391s testing: mul_mat, [2, 1, 3, 3] = [3, 2, 3, 3] * [3, 1, 3, 3] 391s x0: [3, 2, 3, 3] 391s 0.930 -0.689 -0.289 391s -0.855 -0.424 -0.064 391s 391s x1: [3, 1, 3, 3] 391s -0.786 -0.735 -0.775 391s 391s y: [2, 1, 3, 3] 391s 0.000 1.034 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.153 -0.906 0.970 391s -0.475 -0.704 0.876 391s -0.501 -0.837 -0.049 391s 391s x1: [3, 2, 1, 1] 391s 0.446 0.249 0.202 391s -0.367 -0.077 -0.102 391s 391s y: [3, 2, 1, 1] 391s 0.039 -0.210 -0.441 391s -0.086 0.139 0.253 391s testing: mul_mat, [3, 3, 3, 1] = [3, 3, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.924 -0.087 0.738 391s -0.837 0.797 -0.062 391s -0.765 -0.835 0.449 391s 391s x1: [3, 3, 3, 1] 391s -0.163 0.634 0.931 391s -0.175 0.142 0.284 391s 0.664 -0.970 0.291 391s 391s y: [3, 3, 3, 1] 391s 0.782 0.584 0.013 391s 0.359 0.242 0.144 391s -0.314 -1.347 0.432 391s testing: mul_mat, [4, 3, 3, 3] = [3, 4, 3, 3] * [3, 3, 3, 3] 391s x0: [3, 4, 3, 3] 391s 0.040 0.701 -0.467 391s 0.661 -0.849 -0.036 391s -0.575 0.962 -0.985 391s -0.315 -0.611 -0.962 391s 391s x1: [3, 3, 3, 3] 391s -0.534 -0.803 -0.422 391s 0.749 -0.728 -0.259 391s 0.984 0.185 -0.462 391s 391s y: [4, 3, 3, 3] 391s -0.388 0.344 -0.051 1.065 391s -0.360 1.122 -0.876 0.458 391s 0.385 0.510 0.068 0.021 391s test-mul-mat0: iter:370/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.421 -0.202 391s -0.673 0.877 391s 391s x1: [2, 4, 1, 1] 391s -0.174 -0.412 391s 0.854 0.935 391s -0.047 -0.590 391s 0.399 0.533 391s 391s y: [2, 4, 1, 1] 391s 0.157 -0.244 391s -0.549 0.245 391s 0.139 -0.485 391s -0.276 0.199 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.130 0.205 391s 391s x1: [2, 2, 1, 1] 391s -0.122 0.923 391s -0.623 0.607 391s 391s y: [1, 2, 1, 1] 391s 0.174 391s 0.044 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.094 0.323 391s 0.112 0.182 391s 0.565 -0.230 391s 391s x1: [2, 1, 1, 1] 391s 0.179 -0.855 391s 391s y: [3, 1, 1, 1] 391s -0.293 -0.135 0.298 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.532 -0.071 391s 391s x1: [2, 3, 1, 1] 391s 0.118 0.268 391s 0.596 0.706 391s 0.122 0.530 391s 391s y: [1, 3, 1, 1] 391s 0.044 391s 0.267 391s 0.027 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.011 391s 0.125 391s 0.385 391s 391s x1: [1, 2, 1, 1] 391s -0.808 391s -0.202 391s 391s y: [3, 2, 1, 1] 391s 0.009 -0.101 -0.311 391s 0.002 -0.025 -0.078 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.584 -0.108 -0.320 391s 391s x1: [3, 1, 1, 1] 391s 0.867 0.049 0.762 391s 391s y: [1, 1, 1, 1] 391s 0.257 391s test-mul-mat0: iter:371/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.597 -0.667 0.846 391s -0.474 -0.475 -0.355 391s -0.966 0.514 0.770 391s 391s x1: [3, 3, 1, 1] 391s -0.640 0.701 -0.150 391s 0.925 -0.530 0.968 391s 0.193 -0.934 0.674 391s 391s y: [3, 3, 1, 1] 391s -0.975 0.024 0.863 391s 1.725 -0.531 -0.421 391s 1.308 0.113 -0.147 391s testing: mul_mat, [4, 3, 3, 1] = [3, 4, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 4, 3, 1] 391s -0.482 0.700 0.165 391s 0.937 0.081 0.983 391s 0.118 0.937 -0.614 391s 0.191 0.473 0.553 391s 391s x1: [3, 3, 3, 1] 391s 0.419 0.381 -0.181 391s 0.181 -0.144 -0.597 391s -0.927 0.536 0.166 391s 391s y: [4, 3, 3, 1] 391s 0.035 0.245 0.518 0.160 391s -0.286 -0.429 0.253 -0.364 391s 0.849 -0.661 0.291 0.169 391s testing: mul_mat, [4, 4, 3, 4] = [3, 4, 3, 4] * [3, 4, 3, 4] 391s x0: [3, 4, 3, 4] 391s 0.275 0.343 -0.753 391s 0.852 0.871 0.091 391s -0.994 0.524 -0.940 391s 0.887 -0.429 -0.551 391s 391s x1: [3, 4, 3, 4] 391s 0.952 -0.906 0.620 391s -0.661 0.285 0.093 391s 0.891 0.586 -0.310 391s -0.720 0.358 -0.887 391s 391s y: [4, 4, 3, 4] 391s -0.516 0.079 -2.003 0.892 391s -0.154 -0.306 0.720 -0.760 391s 0.679 1.242 -0.288 0.709 391s 0.593 -0.383 1.738 -0.304 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.102 0.916 -0.600 391s -0.182 -0.398 0.442 391s 0.843 -0.975 -0.161 391s 391s x1: [3, 4, 1, 1] 391s -0.523 -0.912 0.716 391s -0.785 -0.791 -0.928 391s 0.357 -0.035 -0.639 391s -0.385 -0.764 -0.354 391s 391s y: [3, 4, 1, 1] 391s -1.212 0.775 0.333 391s -0.089 0.048 0.259 391s 0.315 -0.334 0.439 391s -0.449 0.218 0.478 391s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.848 -0.759 0.355 391s -0.447 -0.790 -0.216 391s 391s x1: [3, 3, 3, 1] 391s -0.941 -0.198 -0.457 391s -0.380 0.968 0.693 391s 0.440 0.440 -0.103 391s 391s y: [2, 3, 3, 1] 391s 0.787 0.676 391s -0.167 -0.745 391s -0.743 -0.522 391s testing: mul_mat, [3, 3, 3, 4] = [2, 3, 3, 4] * [2, 3, 3, 4] 391s x0: [2, 3, 3, 4] 391s 0.834 0.257 391s 0.758 0.657 391s 0.699 0.192 391s 391s x1: [2, 3, 3, 4] 391s 0.808 -0.273 391s 0.719 0.913 391s 0.664 0.410 391s 391s y: [3, 3, 3, 4] 391s 0.604 0.433 0.513 391s 0.834 1.145 0.678 391s 0.659 0.773 0.543 391s test-mul-mat0: iter:372/500 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.609 391s 391s x1: [1, 1, 1, 1] 391s -0.720 391s 391s y: [1, 1, 1, 1] 391s -0.439 391s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.876 391s 391s x1: [1, 1, 2, 1] 391s -0.735 391s 391s y: [1, 1, 2, 1] 391s 0.644 391s testing: mul_mat, [2, 1, 2, 2] = [1, 2, 2, 2] * [1, 1, 2, 2] 391s x0: [1, 2, 2, 2] 391s 0.280 391s 0.577 391s 391s x1: [1, 1, 2, 2] 391s 0.144 391s 391s y: [2, 1, 2, 2] 391s 0.040 0.083 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.741 391s -0.833 391s 0.640 391s 0.955 391s 391s x1: [1, 2, 1, 1] 391s -0.492 391s -0.628 391s 391s y: [4, 2, 1, 1] 391s 0.365 0.410 -0.315 -0.470 391s 0.465 0.523 -0.402 -0.600 391s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 3, 2, 1] 391s 0.213 0.799 -0.433 0.561 391s 0.889 0.154 -0.566 -0.477 391s 0.463 0.315 -0.108 0.046 391s 391s x1: [4, 1, 2, 1] 391s -0.455 0.883 0.251 -0.175 391s 391s y: [3, 1, 2, 1] 391s 0.402 -0.326 0.032 391s testing: mul_mat, [2, 4, 2, 2] = [3, 2, 2, 2] * [3, 4, 2, 2] 391s x0: [3, 2, 2, 2] 391s 0.651 0.776 -0.526 391s -0.248 0.990 0.139 391s 391s x1: [3, 4, 2, 2] 391s -0.444 0.820 0.506 391s 0.416 -0.091 0.169 391s 0.735 -0.878 0.058 391s 0.198 0.921 -0.788 391s 391s y: [2, 4, 2, 2] 391s 0.081 0.993 391s 0.111 -0.169 391s -0.234 -1.043 391s 1.259 0.753 391s test-mul-mat0: iter:373/500 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.716 391s 0.204 391s -0.415 391s 391s x1: [1, 4, 1, 1] 391s 0.563 391s 0.359 391s -0.492 391s 0.214 391s 391s y: [3, 4, 1, 1] 391s -0.403 0.115 -0.233 391s -0.257 0.073 -0.149 391s 0.352 -0.100 0.204 391s -0.153 0.044 -0.089 391s testing: mul_mat, [3, 3, 4, 1] = [1, 3, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 3, 4, 1] 391s -0.047 391s 0.402 391s -0.954 391s 391s x1: [1, 3, 4, 1] 391s 0.424 391s -0.322 391s -0.868 391s 391s y: [3, 3, 4, 1] 391s -0.020 0.171 -0.405 391s 0.015 -0.130 0.308 391s 0.041 -0.349 0.829 391s testing: mul_mat, [1, 3, 4, 1] = [1, 1, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 1, 4, 1] 391s -0.488 391s 391s x1: [1, 3, 4, 1] 391s -0.041 391s 0.112 391s 0.926 391s 391s y: [1, 3, 4, 1] 391s 0.020 391s -0.055 391s -0.452 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.685 391s 0.126 391s -0.785 391s -0.959 391s 391s x1: [1, 1, 1, 1] 391s 0.724 391s 391s y: [4, 1, 1, 1] 391s 0.496 0.091 -0.568 -0.694 391s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.047 0.479 0.676 0.943 391s 0.165 -0.770 0.088 -0.592 391s 391s x1: [4, 1, 4, 1] 391s 0.097 0.430 -0.319 0.432 391s 391s y: [2, 1, 4, 1] 391s 0.402 -0.599 391s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 3, 4, 1] 391s -0.141 0.456 391s 0.996 0.995 391s 0.570 -0.162 391s 391s x1: [2, 4, 4, 1] 391s 0.914 0.368 391s -0.345 0.590 391s -0.544 -0.403 391s 0.999 0.268 391s 391s y: [3, 4, 4, 1] 391s 0.039 1.277 0.461 391s 0.318 0.243 -0.293 391s -0.107 -0.943 -0.245 391s -0.019 1.261 0.526 391s test-mul-mat0: iter:374/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.118 391s 0.240 391s -0.923 391s 391s x1: [1, 3, 1, 1] 391s 0.762 391s -0.115 391s 0.670 391s 391s y: [3, 3, 1, 1] 391s -0.090 0.183 -0.704 391s 0.014 -0.028 0.106 391s -0.079 0.161 -0.618 391s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 2, 3, 1] 391s 0.396 391s 0.860 391s 391s x1: [1, 3, 3, 1] 391s -0.123 391s -0.922 391s -0.759 391s 391s y: [2, 3, 3, 1] 391s -0.049 -0.106 391s -0.365 -0.794 391s -0.301 -0.653 391s testing: mul_mat, [1, 2, 3, 1] = [1, 1, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 1, 3, 1] 391s -0.615 391s 391s x1: [1, 2, 3, 1] 391s 0.136 391s 0.019 391s 391s y: [1, 2, 3, 1] 391s -0.084 391s -0.012 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.368 391s 0.295 391s 391s x1: [1, 1, 1, 1] 391s -0.945 391s 391s y: [2, 1, 1, 1] 391s -0.348 -0.279 391s testing: mul_mat, [4, 1, 3, 1] = [2, 4, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 4, 3, 1] 391s -0.513 0.294 391s 0.898 0.128 391s 0.268 -0.899 391s 0.121 -0.813 391s 391s x1: [2, 1, 3, 1] 391s -0.847 -0.755 391s 391s y: [4, 1, 3, 1] 391s 0.213 -0.857 0.452 0.511 391s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.866 -0.720 0.882 -0.529 391s -0.624 -0.269 -0.725 0.630 391s -0.047 0.915 0.786 0.067 391s 0.948 -0.572 -0.474 0.934 391s 391s x1: [4, 2, 3, 1] 391s 0.967 -0.851 0.151 -0.639 391s 0.358 -0.356 -0.805 0.844 391s 391s y: [4, 2, 3, 1] 391s 0.247 -0.886 -0.749 0.735 391s -1.209 0.988 -0.918 1.713 391s test-mul-mat0: iter:375/500 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.315 -0.341 -0.665 -0.541 391s -0.567 -0.102 -0.036 -0.877 391s 0.889 0.351 -0.913 -0.334 391s -0.737 -0.478 0.072 -0.758 391s 391s x1: [4, 2, 1, 1] 391s 0.205 0.845 -0.920 0.514 391s 0.141 -0.855 0.865 -0.167 391s 391s y: [4, 2, 1, 1] 391s -0.019 -0.620 1.147 -1.010 391s -0.237 0.122 -0.908 0.493 391s testing: mul_mat, [2, 4, 2, 1] = [4, 2, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 2, 2, 1] 391s 0.456 0.055 -0.121 -0.612 391s -0.955 0.460 0.399 -0.451 391s 391s x1: [4, 4, 2, 1] 391s -0.109 -0.926 0.124 0.785 391s -0.264 -0.068 -0.010 -0.419 391s 0.011 -0.497 0.722 0.157 391s -0.632 -0.445 0.209 0.053 391s 391s y: [2, 4, 2, 1] 391s -0.597 -0.627 391s 0.134 0.406 391s -0.206 -0.022 391s -0.371 0.458 391s testing: mul_mat, [4, 2, 2, 3] = [4, 4, 2, 3] * [4, 2, 2, 3] 391s x0: [4, 4, 2, 3] 391s 0.431 0.078 -0.766 -0.120 391s -0.397 0.047 -0.115 0.044 391s 0.596 -0.346 0.455 0.096 391s 0.732 0.348 0.195 -0.440 391s 391s x1: [4, 2, 2, 3] 391s 0.482 -0.402 0.751 0.958 391s -0.632 -0.918 0.785 -0.775 391s 391s y: [4, 2, 2, 3] 391s -0.514 -0.255 0.860 -0.063 391s -0.853 0.084 0.224 -0.288 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.666 -0.258 -0.608 -0.911 391s -0.833 0.566 -0.349 -0.232 391s 391s x1: [4, 4, 1, 1] 391s -0.543 0.783 -0.623 -0.922 391s -0.741 0.453 -0.005 0.840 391s -0.863 -0.061 0.873 0.622 391s -0.530 0.559 0.570 0.059 391s 391s y: [2, 4, 1, 1] 391s 1.378 1.327 391s -0.386 0.681 391s -0.507 0.236 391s -0.192 0.546 391s testing: mul_mat, [3, 4, 2, 1] = [2, 3, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 3, 2, 1] 391s -0.674 0.896 391s 0.325 -0.616 391s 0.098 -0.168 391s 391s x1: [2, 4, 2, 1] 391s -0.093 -0.173 391s -0.073 -0.616 391s 0.784 -0.263 391s -0.158 0.567 391s 391s y: [3, 4, 2, 1] 391s -0.092 0.076 0.020 391s -0.502 0.355 0.096 391s -0.764 0.417 0.121 391s 0.615 -0.401 -0.111 391s testing: mul_mat, [1, 2, 2, 3] = [3, 1, 2, 3] * [3, 2, 2, 3] 391s x0: [3, 1, 2, 3] 391s 0.098 -0.027 0.829 391s 391s x1: [3, 2, 2, 3] 391s 0.884 0.710 0.132 391s -0.289 -0.363 0.516 391s 391s y: [1, 2, 2, 3] 391s 0.177 391s 0.409 391s test-mul-mat0: iter:376/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.998 -0.714 0.059 391s 0.596 0.757 -0.843 391s 391s x1: [3, 1, 1, 1] 391s -0.436 0.036 -0.668 391s 391s y: [2, 1, 1, 1] 391s -0.500 0.331 391s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.674 -0.020 -0.348 391s 0.362 0.312 0.107 391s 0.360 0.598 -0.834 391s 391s x1: [3, 2, 3, 1] 391s -0.432 0.586 0.318 391s -0.971 -0.307 -0.632 391s 391s y: [3, 2, 3, 1] 391s 0.168 0.060 -0.070 391s 0.881 -0.515 -0.005 391s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.783 0.116 -0.539 391s 391s x1: [3, 3, 3, 1] 391s 0.641 0.569 0.677 391s 0.728 0.895 -0.344 391s -0.620 0.257 0.968 391s 391s y: [1, 3, 3, 1] 391s 0.203 391s 0.859 391s -0.978 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.546 0.247 -0.734 391s -0.673 -0.351 0.773 391s -0.328 0.123 -0.398 391s -0.974 0.108 0.891 391s 391s x1: [3, 1, 1, 1] 391s 0.532 0.837 -0.239 391s 391s y: [4, 1, 1, 1] 391s 0.091 -0.837 0.024 -0.641 391s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.102 -0.748 -0.464 -0.744 391s 0.642 -0.606 0.149 0.945 391s 0.045 0.061 -0.534 -0.330 391s -0.495 0.440 0.440 -0.964 391s 391s x1: [4, 3, 3, 1] 391s -0.477 -0.278 -0.503 -0.803 391s -0.191 -0.249 0.980 0.925 391s 0.212 0.404 -0.391 0.349 391s 391s y: [4, 3, 3, 1] 391s 0.990 -0.971 0.495 0.666 391s -0.975 1.048 -0.852 -0.475 391s -0.359 0.163 0.128 -0.435 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.258 0.925 0.888 0.167 391s 391s x1: [4, 4, 3, 1] 391s 0.892 0.100 -0.180 -0.376 391s -0.741 0.300 0.543 -0.997 391s -0.296 0.371 -0.171 0.230 391s -0.083 0.488 -0.142 -0.895 391s 391s y: [1, 4, 3, 1] 391s -0.360 391s 0.785 391s 0.306 391s 0.197 391s test-mul-mat0: iter:377/500 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.653 391s 391s x1: [1, 1, 1, 1] 391s -0.506 391s 391s y: [1, 1, 1, 1] 391s 0.330 391s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.632 391s -0.170 391s 391s x1: [1, 1, 3, 1] 391s 0.298 391s 391s y: [2, 1, 3, 1] 391s -0.189 -0.051 391s testing: mul_mat, [1, 2, 3, 2] = [1, 1, 3, 2] * [1, 2, 3, 2] 391s x0: [1, 1, 3, 2] 391s -0.600 391s 391s x1: [1, 2, 3, 2] 391s -0.557 391s 0.247 391s 391s y: [1, 2, 3, 2] 391s 0.334 391s -0.148 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.695 391s 391s x1: [1, 1, 1, 1] 391s 0.859 391s 391s y: [1, 1, 1, 1] 391s 0.597 391s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.016 391s 0.733 391s 391s x1: [1, 1, 3, 1] 391s -0.726 391s 391s y: [2, 1, 3, 1] 391s 0.012 -0.533 391s testing: mul_mat, [2, 1, 3, 2] = [2, 2, 3, 2] * [2, 1, 3, 2] 391s x0: [2, 2, 3, 2] 391s 0.299 -0.474 391s -0.878 -0.867 391s 391s x1: [2, 1, 3, 2] 391s 0.775 -0.999 391s 391s y: [2, 1, 3, 2] 391s 0.705 0.186 391s test-mul-mat0: iter:378/500 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.741 0.105 0.118 -0.098 391s 391s x1: [4, 4, 1, 1] 391s 0.458 0.157 0.640 -0.243 391s 0.279 -0.834 -0.110 0.798 391s -0.449 -0.796 -0.988 0.209 391s 0.673 0.393 -0.491 -0.875 391s 391s y: [1, 4, 1, 1] 391s -0.223 391s -0.386 391s 0.112 391s -0.429 391s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.202 0.563 0.140 0.688 391s 0.362 0.691 0.892 0.373 391s -0.100 0.565 -0.234 0.409 391s 391s x1: [4, 1, 3, 1] 391s 0.859 0.730 -0.194 0.524 391s 391s y: [3, 1, 3, 1] 391s 0.571 0.838 0.587 391s testing: mul_mat, [3, 3, 3, 2] = [4, 3, 3, 2] * [4, 3, 3, 2] 391s x0: [4, 3, 3, 2] 391s -0.340 0.659 0.335 0.391 391s 0.547 -0.109 -0.911 0.890 391s -0.703 0.475 0.381 -0.285 391s 391s x1: [4, 3, 3, 2] 391s -0.032 0.140 0.275 0.868 391s -0.295 -0.959 0.277 -0.605 391s 0.315 -0.055 0.190 0.707 391s 391s y: [3, 3, 3, 2] 391s 0.535 0.489 -0.053 391s -0.676 -0.848 0.029 391s 0.197 0.634 -0.376 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.633 -0.007 -0.518 0.076 391s -0.138 -0.587 0.271 -0.936 391s -0.874 -0.188 -0.519 -0.774 391s -0.847 -0.579 0.723 -0.884 391s 391s x1: [4, 3, 1, 1] 391s 0.657 0.940 -0.709 0.269 391s -0.243 0.500 -0.845 0.847 391s -0.023 0.158 -0.409 -0.773 391s 391s y: [4, 3, 1, 1] 391s 0.798 -1.086 -0.590 -1.851 391s 0.345 -1.282 -0.098 -1.444 391s 0.138 0.523 0.801 0.316 391s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 3, 3, 1] 391s 0.564 -0.234 0.239 -0.213 391s 0.109 0.848 0.067 -0.650 391s -0.362 -0.026 0.033 0.319 391s 391s x1: [4, 4, 3, 1] 391s 0.468 0.467 0.773 0.408 391s 0.758 0.041 -0.835 0.258 391s 0.196 -0.988 -0.765 -0.646 391s -0.397 -0.538 0.315 -0.764 391s 391s y: [3, 4, 3, 1] 391s 0.252 0.234 -0.026 391s 0.164 -0.106 -0.221 391s 0.297 -0.448 -0.277 391s 0.139 0.018 -0.076 391s testing: mul_mat, [1, 4, 3, 2] = [3, 1, 3, 2] * [3, 4, 3, 2] 391s x0: [3, 1, 3, 2] 391s -0.576 0.214 0.773 391s 391s x1: [3, 4, 3, 2] 391s 0.809 0.602 -0.364 391s -0.158 -0.610 -0.014 391s -0.839 -0.159 -0.912 391s 0.232 -0.271 0.349 391s 391s y: [1, 4, 3, 2] 391s -0.618 391s -0.050 391s -0.256 391s 0.078 391s test-mul-mat0: iter:379/500 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.142 -0.315 -0.880 391s -0.057 -0.333 0.674 391s -0.434 0.975 -0.226 391s 391s x1: [3, 4, 1, 1] 391s -0.001 -0.961 -0.911 391s -0.837 -0.950 -0.770 391s -0.717 0.755 0.639 391s -0.293 -0.031 0.411 391s 391s y: [3, 4, 1, 1] 391s 1.104 -0.293 -0.730 391s 1.095 -0.154 -0.388 391s -0.698 0.219 0.903 391s -0.311 0.304 0.004 391s testing: mul_mat, [2, 3, 2, 1] = [3, 2, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 2, 2, 1] 391s 0.966 0.203 0.193 391s -0.176 0.888 0.313 391s 391s x1: [3, 3, 2, 1] 391s 0.683 0.424 -0.238 391s 0.969 -0.337 0.346 391s -0.328 0.153 0.834 391s 391s y: [2, 3, 2, 1] 391s 0.700 0.182 391s 0.935 -0.361 391s -0.126 0.454 391s testing: mul_mat, [2, 2, 2, 3] = [3, 2, 2, 3] * [3, 2, 2, 3] 391s x0: [3, 2, 2, 3] 391s -0.255 -0.918 -0.553 391s 0.073 0.042 -0.928 391s 391s x1: [3, 2, 2, 3] 391s -0.985 0.953 -0.478 391s 0.985 -0.384 0.868 391s 391s y: [2, 2, 2, 3] 391s -0.360 0.412 391s -0.379 -0.750 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.188 0.723 -0.758 391s -0.765 0.830 0.181 391s 0.071 0.923 0.990 391s 391s x1: [3, 2, 1, 1] 391s 0.609 -0.316 -0.813 391s 0.459 -0.804 -0.371 391s 391s y: [3, 2, 1, 1] 391s 0.273 -0.876 -1.054 391s -0.386 -1.086 -1.077 391s testing: mul_mat, [4, 3, 2, 1] = [3, 4, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 4, 2, 1] 391s -0.498 0.023 -0.266 391s -0.212 -0.189 0.280 391s 0.740 0.037 -0.966 391s 0ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s .314 -0.147 -0.276 391s 391s x1: [3, 3, 2, 1] 391s 0.356 -0.379 0.800 391s -0.176 -0.836 0.307 391s 0.168 0.796 0.127 391s 391s y: [4, 3, 2, 1] 391s -0.399 0.220 -0.524 -0.053 391s -0.013 0.281 -0.459 -0.017 391s -0.099 -0.150 0.031 -0.099 391s testing: mul_mat, [1, 3, 2, 3] = [4, 1, 2, 3] * [4, 3, 2, 3] 391s x0: [4, 1, 2, 3] 391s 0.808 0.147 -0.175 0.770 391s 391s x1: [4, 3, 2, 3] 391s -0.321 -0.590 0.961 -0.919 391s -0.284 -0.880 -0.058 0.218 391s -0.092 -0.317 -0.468 0.931 391s 391s y: [1, 3, 2, 3] 391s -1.221 391s -0.182 391s 0.678 391s test-mul-mat0: iter:380/500 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.413 0.015 0.917 391s 0.637 -0.277 0.268 391s -0.835 0.423 -0.928 391s 0.265 0.575 -0.974 391s 391s x1: [3, 3, 1, 1] 391s -0.014 0.322 -0.914 391s -0.206 -0.531 -0.089 391s -0.437 -0.818 -0.274 391s 391s y: [4, 3, 1, 1] 391s -0.840 -0.343 0.997 1.072 391s -0.174 -0.008 0.030 -0.273 391s -0.444 -0.125 0.273 -0.318 391s testing: mul_mat, [2, 4, 2, 1] = [3, 2, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.440 -0.196 0.234 391s 0.825 -0.622 0.260 391s 391s x1: [3, 4, 2, 1] 391s 0.192 -0.686 0.235 391s -0.836 -0.141 -0.480 391s -0.480 0.371 0.920 391s 0.506 -0.307 -0.994 391s 391s y: [2, 4, 2, 1] 391s 0.105 0.646 391s 0.283 -0.726 391s 0.353 -0.388 391s -0.395 0.350 391s testing: mul_mat, [4, 2, 2, 4] = [3, 4, 2, 4] * [3, 2, 2, 4] 391s x0: [3, 4, 2, 4] 391s -0.372 -0.612 -0.316 391s 0.264 -0.953 0.405 391s 0.159 -0.278 -0.182 391s 0.549 -0.422 0.782 391s 391s x1: [3, 2, 2, 4] 391s 0.700 -0.078 -0.065 391s 0.206 0.615 -0.059 391s 391s y: [4, 2, 2, 4] 391s -0.192 0.233 0.145 0.367 391s -0.435 -0.555 -0.127 -0.192 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.549 -0.117 0.887 391s 391s x1: [3, 4, 1, 1] 391s -0.460 0.252 -0.879 391s -0.650 0.383 -0.517 391s -0.229 -0.074 0.036 391s -0.243 -0.983 0.447 391s 391s y: [1, 4, 1, 1] 391s -1.062 391s -0.860 391s -0.085 391s 0.379 391s testing: mul_mat, [3, 3, 2, 1] = [1, 3, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.335 391s 0.904 391s -0.077 391s 391s x1: [1, 3, 2, 1] 391s -0.312 391s -0.212 391s 0.325 391s 391s y: [3, 3, 2, 1] 391s 0.104 -0.282 0.024 391s 0.071 -0.192 0.016 391s -0.109 0.294 -0.025 391s testing: mul_mat, [1, 1, 2, 4] = [3, 1, 2, 4] * [3, 1, 2, 4] 391s x0: [3, 1, 2, 4] 391s -0.694 -0.869 0.345 391s 391s x1: [3, 1, 2, 4] 391s 0.396 0.875 0.255 391s 391s y: [1, 1, 2, 4] 391s -0.948 391s test-mul-mat0: iter:381/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.903 391s 0.701 391s -0.160 391s 391s x1: [1, 3, 1, 1] 391s -0.213 391s 0.967 391s -0.644 391s 391s y: [3, 3, 1, 1] 391s 0.192 -0.149 0.034 391s -0.873 0.678 -0.155 391s 0.581 -0.451 0.103 391s testing: mul_mat, [2, 3, 2, 1] = [1, 2, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.996 391s 0.592 391s 391s x1: [1, 3, 2, 1] 391s 0.936 391s 0.761 391s 0.981 391s 391s y: [2, 3, 2, 1] 391s 0.933 0.554 391s 0.758 0.450 391s 0.978 0.581 391s testing: mul_mat, [2, 2, 2, 3] = [1, 2, 2, 3] * [1, 2, 2, 3] 391s x0: [1, 2, 2, 3] 391s -0.634 391s -0.613 391s 391s x1: [1, 2, 2, 3] 391s -0.630 391s 0.252 391s 391s y: [2, 2, 2, 3] 391s 0.400 0.386 391s -0.160 -0.155 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.644 391s -0.021 391s -0.999 391s -0.987 391s 391s x1: [1, 2, 1, 1] 391s -0.948 391s 0.431 391s 391s y: [4, 2, 1, 1] 391s -0.610 0.020 0.948 0.935 391s 0.277 -0.009 -0.430 -0.425 391s testing: mul_mat, [3, 1, 2, 1] = [4, 3, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.839 -0.473 0.987 -0.714 391s 0.712 -0.900 -0.416 0.005 391s -0.766 -0.592 -0.717 0.918 391s 391s x1: [4, 1, 2, 1] 391s -0.769 0.723 -0.865 -0.849 391s 391s y: [3, 1, 2, 1] 391s 0.056 -0.843 0.001 391s testing: mul_mat, [4, 4, 2, 3] = [3, 4, 2, 3] * [3, 4, 2, 3] 391s x0: [3, 4, 2, 3] 391s -0.000 -0.184 -0.965 391s 0.092 -0.707 0.248 391s -0.470 0.414 0.499 391s 0.024 -0.258 -0.627 391s 391s x1: [3, 4, 2, 3] 391s -0.589 0.868 0.673 391s -0.057 -0.549 -0.035 391s 0.359 -0.388 -0.322 391s 0.593 0.139 -0.223 391s 391s y: [4, 4, 2, 3] 391s -0.809 -0.501 0.973 -0.660 391s 0.134 0.374 -0.218 0.162 391s 0.382 0.227 -0.490 0.311 391s 0.189 -0.099 -0.332 0.118 391s test-mul-mat0: iter:382/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.489 -0.380 -0.423 391s 0.766 -0.508 0.611 391s 0.047 0.260 0.751 391s 391s x1: [3, 3, 1, 1] 391s 0.650 -0.910 -0.079 391s -0.619 -0.415 -0.771 391s -0.106 0.102 -0.355 391s 391s y: [3, 3, 1, 1] 391s 0.062 0.913 -0.266 391s 0.786 -0.734 -0.715 391s 0.163 -0.350 -0.245 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.202 -0.104 0.605 391s -0.112 0.817 0.986 391s 0.472 -0.953 -0.120 391s -0.426 -0.308 -0.564 391s 391s x1: [3, 3, 1, 1] 391s -0.118 0.948 0.812 391s 0.566 0.971 -0.464 391s 0.132 -0.423 0.422 391s 391s y: [4, 3, 1, 1] 391s 0.417 1.588 -1.057 -0.700 391s -0.496 0.272 -0.603 -0.278 391s 0.273 0.056 0.415 -0.164 391s testing: mul_mat, [1, 4, 1, 3] = [3, 1, 1, 3] * [3, 4, 1, 3] 391s x0: [3, 1, 1, 3] 391s 0.127 0.774 -0.649 391s 391s x1: [3, 4, 1, 3] 391s 0.086 0.312 0.013 391s -0.148 0.803 -0.376 391s 0.899 0.063 -0.626 391s -0.219 0.011 -0.814 391s 391s y: [1, 4, 1, 3] 391s 0.244 391s 0.847 391s 0.569 391s 0.509 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.682 -0.556 -0.795 391s 391s x1: [3, 1, 1, 1] 391s 0.733 0.325 0.403 391s 391s y: [1, 1, 1, 1] 391s -0.002 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.793 391s 0.134 391s -0.065 391s 0.803 391s 391s x1: [1, 3, 1, 1] 391s -0.073 391s 0.302 391s 0.229 391s 391s y: [4, 3, 1, 1] 391s 0.058 -0.010 0.005 -0.059 391s -0.240 0.040 -0.020 0.243 391s -0.182 0.031 -0.015 0.184 391s testing: mul_mat, [4, 1, 1, 3] = [4, 4, 1, 3] * [4, 1, 1, 3] 391s x0: [4, 4, 1, 3] 391s 0.829 0.072 -0.202 -0.838 391s 0.747 -0.646 -0.067 0.325 391s -0.049 -0.272 0.681 -0.693 391s 0.562 -0.246 -0.900 0.296 391s 391s x1: [4, 1, 1, 3] 391s 0.474 0.311 -0.289 -0.173 391s 391s y: [4, 1, 1, 3] 391s 0.619 0.117 -0.185 0.399 391s test-mul-mat0: iter:383/500 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.009 0.194 -0.808 -0.538 391s -0.813 0.545 -0.135 -0.140 391s -0.002 0.431 -0.142 -0.242 391s 0.293 0.116 -0.372 0.926 391s 391s x1: [4, 3, 1, 1] 391s -0.769 0.938 -0.066 -0.367 391s 0.715 -0.628 -0.811 0.067 391s 0.791 0.438 -0.752 -0.043 391s 391s y: [4, 3, 1, 1] 391s 0.441 1.197 0.504 -0.433 391s 0.490 -0.824 -0.174 0.501 391s 0.709 -0.297 0.304 0.522 391s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.853 0.484 0.318 -0.918 391s 0.098 -0.892 0.743 0.570 391s 0.657 -0.302 0.159 0.834 391s 0.578 0.690 0.330 -0.385 391s 391s x1: [4, 4, 3, 1] 391s -0.056 0.319 -0.843 -0.118 391s -0.747 -0.211 -0.403 -0.375 391s -0.021 0.664 -0.584 -0.583 391s 0.912 0.373 -0.945 -0.097 391s 391s y: [4, 4, 3, 1] 391s -0.053 -0.984 -0.366 -0.045 391s -0.523 -0.399 -0.804 -0.566 391s 0.653 -1.360 -0.793 0.478 391s 0.747 -1.001 0.255 0.510 391s testing: mul_mat, [3, 4, 3, 2] = [4, 3, 3, 2] * [4, 4, 3, 2] 391s x0: [4, 3, 3, 2] 391s -0.342 -0.813 0.269 0.988 391s 0.724 0.106 0.148 0.944 391s -0.620 -0.253 0.940 0.548 391s 391s x1: [4, 4, 3, 2] 391s -0.403 0.873 -0.645 -0.950 391s 0.120 0.143 -0.706 0.590 391s 0.516 -0.300 0.337 -0.623 391s 0.837 -0.391 0.091 -0.542 391s 391s y: [3, 4, 3, 2] 391s -1.683 -1.191 -1.098 391s 0.235 0.555 -0.451 391s -0.457 -0.197 -0.269 391s -0.479 0.066 -0.631 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.427 0.913 0.894 0.883 391s 0.147 0.163 0.882 -0.598 391s 391s x1: [4, 3, 1, 1] 391s -0.600 -0.747 0.672 -0.224 391s 0.105 0.258 -0.499 0.518 391s -0.847 0.079 0.718 0.130 391s 391s y: [2, 3, 1, 1] 391s -0.535 0.516 391s 0.292 -0.693 391s 0.467 0.443 391s testing: mul_mat, [1, 4, 3, 1] = [2, 1, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 1, 3, 1] 391s 0.100 -0.014 391s 391s x1: [2, 4, 3, 1] 391s -0.184 0.773 391s 0.939 -0.295 391s -0.444 0.612 391s -0.305 -0.822 391s 391s y: [1, 4, 3, 1] 391s -0.030 391s 0.098 391s -0.053 391s -0.019 391s testing: mul_mat, [2, 2, 3, 2] = [1, 2, 3, 2] * [1, 2, 3, 2] 391s x0: [1, 2, 3, 2] 391s 0.307 391s 0.756 391s 391s x1: [1, 2, 3, 2] 391s 0.799 391s -0.359 391s 391s y: [2, 2, 3, 2] 391s 0.245 0.603 391s -0.110 -0.271 391s test-mul-mat0: iter:384/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.163 391s 0.145 391s 391s x1: [1, 1, 1, 1] 391s -0.496 391s 391s y: [2, 1, 1, 1] 391s -0.081 -0.072 391s testing: mul_mat, [1, 2, 4, 1] = [1, 1, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 1, 4, 1] 391s -0.358 391s 391s x1: [1, 2, 4, 1] 391s -0.237 391s 0.667 391s 391s y: [1, 2, 4, 1] 391s 0.085 391s -0.239 391s testing: mul_mat, [3, 1, 4, 4] = [1, 3, 4, 4] * [1, 1, 4, 4] 391s x0: [1, 3, 4, 4] 391s 0.596 391s -0.788 391s -0.906 391s 391s x1: [1, 1, 4, 4] 391s 0.934 391s 391s y: [3, 1, 4, 4] 391s 0.556 -0.735 -0.846 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.858 391s 391s x1: [1, 3, 1, 1] 391s -0.521 391s 0.945 391s 0.145 391s 391s y: [1, 3, 1, 1] 391s 0.447 391s -0.811 391s -0.124 391s testing: mul_mat, [4, 1, 4, 1] = [1, 4, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 4, 4, 1] 391s 0.600 391s -0.707 391s -0.584 391s 0.388 391s 391s x1: [1, 1, 4, 1] 391s -0.253 391s 391s y: [4, 1, 4, 1] 391s -0.152 0.179 0.148 -0.098 391s testing: mul_mat, [3, 1, 4, 4] = [4, 3, 4, 4] * [4, 1, 4, 4] 391s x0: [4, 3, 4, 4] 391s -0.704 0.81ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s 2 -0.376 -0.032 391s -0.171 0.093 -0.051 -0.307 391s -0.702 -0.676 0.717 0.571 391s 391s x1: [4, 1, 4, 4] 391s 0.943 0.849 -0.501 0.208 391s 391s y: [3, 1, 4, 4] 391s 0.208 -0.121 -1.475 391s test-mul-mat0: iter:385/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.738 0.460 0.109 391s -0.823 0.513 0.359 391s 391s x1: [3, 1, 1, 1] 391s -0.380 -0.304 0.457 391s 391s y: [2, 1, 1, 1] 391s 0.190 0.321 391s testing: mul_mat, [2, 2, 4, 1] = [3, 2, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.979 0.987 0.050 391s -0.801 0.500 -0.591 391s 391s x1: [3, 2, 4, 1] 391s 0.111 0.084 -0.085 391s -0.045 0.474 -0.387 391s 391s y: [2, 2, 4, 1] 391s -0.029 0.004 391s 0.493 0.501 391s testing: mul_mat, [2, 2, 4, 4] = [3, 2, 4, 4] * [3, 2, 4, 4] 391s x0: [3, 2, 4, 4] 391s 0.479 -0.422 0.365 391s -0.747 0.161 -0.822 391s 391s x1: [3, 2, 4, 4] 391s 0.094 -0.189 0.857 391s -0.243 -0.430 -0.616 391s 391s y: [2, 2, 4, 4] 391s 0.437 -0.805 391s -0.159 0.618 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s -1.000 0.029 0.085 391s 0.499 -0.333 0.675 391s -0.281 0.946 -0.355 391s 391s x1: [3, 2, 1, 1] 391s 0.765 0.940 0.309 391s 0.361 0.774 -0.493 391s 391s y: [3, 2, 1, 1] 391s -0.712 0.277 0.565 391s -0.381 -0.411 0.806 391s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 2, 4, 1] 391s 0.033 -0.699 -0.743 391s 0.912 -0.499 0.339 391s 391s x1: [3, 3, 4, 1] 391s 0.190 0.869 0.079 391s 0.062 -0.677 0.450 391s 0.703 -0.645 0.421 391s 391s y: [2, 3, 4, 1] 391s -0.660 -0.234 391s 0.141 0.547 391s 0.161 1.106 391s testing: mul_mat, [4, 3, 4, 4] = [2, 4, 4, 4] * [2, 3, 4, 4] 391s x0: [2, 4, 4, 4] 391s -0.566 0.267 391s -0.278 -0.244 391s -0.441 -0.064 391s 0.751 -0.787 391s 391s x1: [2, 3, 4, 4] 391s -0.183 0.968 391s -0.099 -0.753 391s 0.298 -0.543 391s 391s y: [4, 3, 4, 4] 391s 0.362 -0.185 0.018 -0.899 391s -0.145 0.211 0.092 0.518 391s -0.313 0.050 -0.096 0.650 391s test-mul-mat0: iter:386/500 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.431 391s 0.104 391s 391s x1: [1, 1, 1, 1] 391s 0.769 391s 391s y: [2, 1, 1, 1] 391s -0.331 0.080 391s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 3, 4, 1] 391s 0.375 391s -0.430 391s -0.613 391s 391s x1: [1, 2, 4, 1] 391s 0.998 391s -0.215 391s 391s y: [3, 2, 4, 1] 391s 0.375 -0.430 -0.612 391s -0.081 0.092 0.132 391s testing: mul_mat, [2, 3, 4, 2] = [1, 2, 4, 2] * [1, 3, 4, 2] 391s x0: [1, 2, 4, 2] 391s 0.589 391s 0.404 391s 391s x1: [1, 3, 4, 2] 391s 0.373 391s -0.130 391s 0.170 391s 391s y: [2, 3, 4, 2] 391s 0.220 0.151 391s -0.076 -0.052 391s 0.100 0.069 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.315 391s 0.345 391s 0.205 391s 391s x1: [1, 2, 1, 1] 391s 0.875 391s -0.062 391s 391s y: [3, 2, 1, 1] 391s -0.276 0.302 0.180 391s 0.020 -0.021 -0.013 391s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.372 -0.017 0.612 391s 0.485 0.935 -0.426 391s -0.166 -0.526 -0.071 391s 391s x1: [3, 1, 4, 1] 391s -0.654 -0.292 0.357 391s 391s y: [3, 1, 4, 1] 391s 0.467 -0.742 0.237 391s testing: mul_mat, [2, 3, 4, 2] = [3, 2, 4, 2] * [3, 3, 4, 2] 391s x0: [3, 2, 4, 2] 391s 0.260 0.099 0.703 391s -0.246 0.914 -0.613 391s 391s x1: [3, 3, 4, 2] 391s 0.149 0.710 -0.960 391s -0.921 -0.425 -0.991 391s 0.383 0.315 -0.084 391s 391s y: [2, 3, 4, 2] 391s -0.566 1.200 391s -0.979 0.446 391s 0.072 0.245 391s test-mul-mat0: iter:387/500 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.966 391s 0.990 391s 0.784 391s -0.809 391s 391s x1: [1, 2, 1, 1] 391s -0.893 391s 0.869 391s 391s y: [4, 2, 1, 1] 391s -0.863 -0.884 -0.700 0.723 391s 0.839 0.860 0.681 -0.703 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.307 391s 391s x1: [1, 4, 1, 1] 391s -0.153 391s -0.914 391s 0.892 391s -0.035 391s 391s y: [1, 4, 1, 1] 391s 0.047 391s 0.281 391s -0.274 391s 0.011 391s testing: mul_mat, [2, 1, 1, 2] = [1, 2, 1, 2] * [1, 1, 1, 2] 391s x0: [1, 2, 1, 2] 391s 0.326 391s -0.180 391s 391s x1: [1, 1, 1, 2] 391s -0.004 391s 391s y: [2, 1, 1, 2] 391s -0.001 0.001 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.752 391s -0.763 391s 391s x1: [1, 2, 1, 1] 391s 0.705 391s 0.416 391s 391s y: [2, 2, 1, 1] 391s 0.530 -0.538 391s 0.313 -0.318 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.833 -0.614 391s -0.787 0.940 391s 391s x1: [2, 1, 1, 1] 391s 0.590 0.084 391s 391s y: [2, 1, 1, 1] 391s 0.440 -0.385 391s testing: mul_mat, [3, 2, 1, 2] = [2, 3, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 3, 1, 2] 391s 0.884 -0.120 391s -0.741 -0.840 391s -0.319 -0.439 391s 391s x1: [2, 2, 1, 2] 391s -0.918 0.248 391s 0.906 -0.928 391s 391s y: [3, 2, 1, 2] 391s -0.841 0.472 0.184 391s 0.912 0.109 0.119 391s test-mul-mat0: iter:388/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.250 391s 0.816 391s 391s x1: [1, 4, 1, 1] 391s 0.521 391s -0.891 391s 0.310 391s 0.734 391s 391s y: [2, 4, 1, 1] 391s 0.130 0.426 391s -0.222 -0.727 391s 0.077 0.253 391s 0.183 0.600 391s testing: mul_mat, [2, 2, 2, 1] = [1, 2, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 2, 2, 1] 391s -0.193 391s 0.892 391s 391s x1: [1, 2, 2, 1] 391s 0.744 391s 0.156 391s 391s y: [2, 2, 2, 1] 391s -0.144 0.663 391s -0.030 0.139 391s testing: mul_mat, [1, 2, 2, 4] = [1, 1, 2, 4] * [1, 2, 2, 4] 391s x0: [1, 1, 2, 4] 391s -0.931 391s 391s x1: [1, 2, 2, 4] 391s -0.849 391s 0.922 391s 391s y: [1, 2, 2, 4] 391s 0.790 391s -0.858 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.820 391s 0.223 391s -0.032 391s -0.129 391s 391s x1: [1, 1, 1, 1] 391s 0.332 391s 391s y: [4, 1, 1, 1] 391s -0.272 0.074 -0.011 -0.043 391s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 1, 2, 1] 391s -0.048 -0.590 -0.103 0.365 391s 391s x1: [4, 1, 2, 1] 391s 0.374 -0.111 -0.558 0.685 391s 391s y: [1, 1, 2, 1] 391s 0.354 391s testing: mul_mat, [1, 4, 2, 4] = [1, 1, 2, 4] * [1, 4, 2, 4] 391s x0: [1, 1, 2, 4] 391s 0.599 391s 391s x1: [1, 4, 2, 4] 391s 0.726 391s -0.032 391s 0.750 391s 0.045 391s 391s y: [1, 4, 2, 4] 391s 0.435 391s -0.019 391s 0.449 391s 0.027 391s test-mul-mat0: iter:389/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.867 -0.582 0.029 0.983 391s 0.852 -0.492 0.581 -0.668 391s -0.559 0.487 -0.667 -0.801 391s 391s x1: [4, 3, 1, 1] 391s -0.359 -0.434 0.964 -0.438 391s 0.043 0.503 0.458 -0.074 391s -0.342 0.305 0.221 -0.291 391s 391s y: [3, 3, 1, 1] 391s 0.161 0.760 -0.303 391s -0.389 0.104 -0.025 391s -0.160 -0.120 0.425 391s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.418 -0.435 0.357 0.532 391s 0.479 0.071 -0.011 0.974 391s 0.117 -0.113 0.956 0.495 391s -0.016 -0.032 0.453 0.047 391s 391s x1: [4, 3, 4, 1] 391s 0.030 0.580 0.965 0.436 391s 0.042 -0.391 -0.923 0.608 391s -0.428 -0.361 -0.349 -0.924 391s 391s y: [4, 3, 4, 1] 391s 0.312 0.470 1.076 0.439 391s 0.146 0.595 -0.532 -0.378 391s -0.281 -1.127 -0.800 -0.184 391s testing: mul_mat, [3, 4, 4, 4] = [4, 3, 4, 4] * [4, 4, 4, 4] 391s x0: [4, 3, 4, 4] 391s 0.410 0.694 -0.258 0.495 391s -0.162 0.405 -0.200 -0.326 391s -0.890 -0.379 -0.315 -0.018 391s 391s x1: [4, 4, 4, 4] 391s 0.362 -0.663 -0.692 0.621 391s -0.533 0.119 -0.004 0.484 391s 0.887 0.867 0.485 0.623 391s -0.810 -0.467 -0.226 -0.032 391s 391s y: [3, 4, 4, 4] 391s 0.175 -0.391 0.136 391s 0.105 -0.023 0.422 391s 1.148 -0.093 -1.282 391s -0.613 -0.002 0.969 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.067 -0.944 0.300 0.723 391s -0.811 0.032 -0.497 -0.092 391s 391s x1: [4, 3, 1, 1] 391s 0.127 -0.892 0.843 0.267 391s -0.777 -0.225 0.465 0.568 391s 0.049 -0.553 0.746 0.097 391s 391s y: [2, 3, 1, 1] 391s 1.280 -0.575 391s 0.815 0.340 391s 0.813 -0.437 391s testing: mul_mat, [2, 4, 4, 1] = [2, 2, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.285 -0.251 391s 0.788 0.992 391s 391s x1: [2, 4, 4, 1] 391s -0.060 0.195 391s -0.610 -0.075 391s -0.004 -0.293 391s 0.852 -0.531 391s 391s y: [2, 4, 4, 1] 391s -0.066 0.146 391s -0.155 -0.556 391s 0.072 -0.294 391s 0.376 0.144 391s testing: mul_mat, [4, 2, 4, 4] = [2, 4, 4, 4] * [2, 2, 4, 4] 391s x0: [2, 4, 4, 4] 391s 0.814 -0.377 391s -0.376 -0.703 391s -0.589 -0.961 391s -0.008 -0.889 391s 391s x1: [2, 2, 4, 4] 391s -0.301 -0.221 391s -0.900 0.196 391s 391s y: [4, 2, 4, 4] 391s -0.162 0.268 0.390 0.199 391s -0.806 0.201 0.342 -0.167 391s test-mul-mat0: iter:390/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.809 0.204 -0.024 391s 391s x1: [3, 1, 1, 1] 391s 0.417 0.205 -0.163 391s 391s y: [1, 1, 1, 1] 391s 0.383 391s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.589 -0.104 -0.455 391s 0.285 0.205 -0.702 391s -0.869 -0.983 -0.736 391s -0.997 0.187 0.681 391s 391s x1: [3, 1, 4, 1] 391s -0.740 0.349 -0.035 391s 391s y: [4, 1, 4, 1] 391s 0.415 -0.115 0.326 0.779 391s testing: mul_mat, [4, 4, 4, 3] = [3, 4, 4, 3] * [3, 4, 4, 3] 391s x0: [3, 4, 4, 3] 391s 0.665 0.655 -0.936 391s -0.656 -0.970 -0.651 391s 0.612 0.768 -0.899 391s 0.362 -0.641 -0.054 391s 391s x1: [3, 4, 4, 3] 391s 0.474 -0.926 0.088 391s 0.003 0.253 -0.376 391s -0.587 -0.755 0.505 391s 0.623 -0.075 -0.753 391s 391s y: [4, 4, 4, 3] 391s -0.373 0.530 -0.500 0.761 391s 0.520 -0.002 0.534 -0.141 391s -1.358 0.789 -1.393 0.245 391s 1.070 0.154 1.000 0.314 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.491 0.419 -0.646 391s 391s x1: [3, 4, 1, 1] 391s -0.158 0.019 0.886 391s 0.462 0.425 -0.658 391s -0.994 0.927 -0.867 391s 0.272 -0.011 0.702 391s 391s y: [1, 4, 1, 1] 391s -0.642 391s 0.830 391s 0.460 391s -0.325 391s testing: mul_mat, [1, 3, 4, 1] = [1, 1, 4, 1] * [1, 3, 4, 1] 391s x0: [1, 1, 4, 1] 391s 0.029 391s 391s x1: [1, 3, 4, 1] 391s 0.086 391s 0.192 391s -0.216 391s 391s y: [1, 3, 4, 1] 391s 0.002 391s 0.006 391s -0.006 391s testing: mul_mat, [3, 1, 4, 3] = [1, 3, 4, 3] * [1, 1, 4, 3] 391s x0: [1, 3, 4, 3] 391s -0.747 391s -0.034 391s 0.603 391s 391s x1: [1, 1, 4, 3] 391s -0.184 391s 391s y: [3, 1, 4, 3] 391s 0.137 0.006 -0.111 391s test-mul-mat0: iter:391/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.333 391s -0.710 391s 0.908 391s 0.470 391s 391s x1: [1, 1, 1, 1] 391s 0.386 391s 391s y: [4, 1, 1, 1] 391s 0.129 -0.274 0.351 0.181 391s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.680 391s -0.986 391s 391s x1: [1, 4, 2, 1] 391s -0.876 391s 0.616 391s 0.655 391s -0.226 391s 391s y: [2, 4, 2, 1] 391s -0.596 0.864 391s 0.419 -0.607 391s 0.445 -0.645 391s -0.154 0.223 391s testing: mul_mat, [4, 2, 2, 2] = [1, 4, 2, 2] * [1, 2, 2, 2] 391s x0: [1, 4, 2, 2] 391s -0.289 391s 0.244 391s -0.850 391s -0.903 391s 391s x1: [1, 2, 2, 2] 391s 0.898 391s 0.140 391s 391s y: [4, 2, 2, 2] 391s -0.259 0.219 -0.763 -0.811 391s -0.040 0.034 -0.119 -0.126 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.316 391s 0.436 391s 391s x1: [1, 4, 1, 1] 391s 0.524 391s -0.712 391s 0.545 391s 0.538 391s 391s y: [2, 4, 1, 1] 391s 0.166 0.228 391s -0.225 -0.310 391s 0.172 0.237 391s 0.170 0.234 391s testing: mul_mat, [3, 1, 2, 1] = [2, 3, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 3, 2, 1] 391s -0.323 0.388 391s 0.655 -0.100 391s 0.897 -0.953 391s 391s x1: [2, 1, 2, 1] 391s 0.241 0.591 391s 391s y: [3, 1, 2, 1] 391s 0.151 0.099 -0.347 391s testing: mul_mat, [1, 2, 2, 2] = [3, 1, 2, 2] * [3, 2, 2, 2] 391s x0: [3, 1, 2, 2] 391s -0.713 0.084 0.430 391s 391s x1: [3, 2, 2, 2] 391s -0.619 0.279 0.092 391s 0.665 -0.227 0.012 391s 391s y: [1, 2, 2, 2] 391s 0.504 391s -0.488 391s test-mul-mat0: iter:392/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.974 0.361 391s 0.180 -0.287 391s -0.325 0.134 391s 391s x1: [2, 1, 1, 1] 391s -0.295 -0.281 391s 391s y: [3, 1, 1, 1] 391s -0.389 0.028 0.058 391s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.076 0.234 391s -0.879 0.499 391s 0.252 0.855 391s -0.631 -0.430 391s 391s x1: [2, 3, 2, 1] 391s -0.871 0.645 391s -0.015 -0.068 391s 0.514 -0.116 391s 391s y: [4, 3, 2, 1] 391s 0.217 1.087 0.331 0.272 391s -0.015 -0.021 -0.062 0.039 391s -0.066 -0.510 0.030 -0.274 391s testing: mul_mat, [1, 4, 2, 1] = [2, 1, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.458 -0.374 391s 391s x1: [2, 4, 2, 1] 391s 0.324 -0.826 391s 0.891 -0.031 391s 0.159 -0.178 391s -0.517 -0.958 391s 391s y: [1, 4, 2, 1] 391s 0.458 391s 0.419 391s 0.139 391s 0.122 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.930 0.835 391s -0.071 0.118 391s 391s x1: [2, 1, 1, 1] 391s -0.005 0.640 391s 391s y: [2, 1, 1, 1] 391s 0.539 0.076 391s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.645 -0.750 391s 391s x1: [2, 2, 2, 1] 391s 0.578 -0.125 391s -0.120 -0.097 391s 391s y: [1, 2, 2, 1] 391s -0.279 391s 0.151 391s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.232 391s 0.006 391s 0.848 391s 391s x1: [1, 2, 2, 1] 391s -0.394 391s 0.054 391s 391s y: [3, 2, 2, 1] 391s 0.092 -0.002 -0.334 391s -0.013 0.000 0.046 391s test-mul-mat0: iter:393/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.004 0.103 391s -0.180 0.352 391s 391s x1: [2, 4, 1, 1] 391s -0.356 -0.029 391s 0.577 -0.778 391s 0.847 -0.543 391s 0.125 -0.104 391s 391s y: [2, 4, 1, 1] 391s -0.002 0.054 391s -0.082 -0.377 391s -0.059 -0.344 391s -0.011 -0.059 391s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.092 -0.021 391s 0.946 -0.956 391s 391s x1: [2, 2, 3, 1] 391s 0.353 -0.546 391s -0.212 0.959 391s 391s y: [2, 2, 3, 1] 391s -0.021 0.856 391s -0.000 -1.117 391s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.380 -0.302 391s 0.477 0.127 391s 391s x1: [2, 2, 3, 1] 391s 0.758 -0.302 391s -0.067 -0.139 391s 391s y: [2, 2, 3, 1] 391s 0.379 0.323 391s 0.017 -0.049 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.007 0.927 391s 0.714 -0.424 391s -0.110 -0.592 391s 391s x1: [2, 2, 1, 1] 391s 0.094 -0.280 391s 0.056 0.956 391s 391s y: [3, 2, 1, 1] 391s -0.260 0.186 0.155 391s 0.886 -0.366 -0.572 391s testing: mul_mat, [3, 2, 3, 1] = [3, 3, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 3, 3, 1] 391s 0.370 -0.574 0.419 391s 0.577 0.534 0.248 391s 0.465 0.656 -0.454 391s 391s x1: [3, 2, 3, 1] 391s -0.788 -0.209 -0.619 391s 0.285 -0.035 0.862 391s 391s y: [3, 2, 3, 1] 391s -0.431 -0.720 -0.223 391s 0.487 0.359 -0.282 391s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.820 -0.324 0.288 391s 391s x1: [3, 3, 3, 1] 391s -0.074 -0.769 0.696 391s 0.671 0.601 0.273 391s 0.136 -0.973 -0.194 391s 391s y: [1, 3, 3, 1] 391s 0.389 391s 0.434 391s 0.371 391s test-mul-mat0: iter:394/500 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.775 -0.111 0.746 0.063 391s 0.635 -0.347 -0.953 -0.948 391s 0.160 -0.697 0.604 0.693 391s 391s x1: [4, 4, 1, 1] 391s 0.980 0.117 0.683 -0.228 391s -0.005 0.977 0.122 0.488 391s 0.606 -0.686 0.560 -0.567 391s 0.135 0.499 0.699 0.069 391s 391s y: [3, 4, 1, 1] 391s 1.242 0.147 0.330 391s 0.010 -0.921 -0.270 391s 0.928 0.627 0.520 391s 0.575 -0.820 0.144 391s testing: mul_mat, [2, 3, 4, 1] = [4, 2, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.528 -0.214 0.989 0.314 391s -0.029 -0.403 -0.191 0.203 391s 391s x1: [4, 3, 4, 1] 391s 0.366 0.949 0.673 -0.517 391s 0.632 -0.554 0.478 0.609 391s 0.567 -0.035 0.215 0.881 391s 391s y: [2, 3, 4, 1] 391s 0.107 -0.626 391s 0.448 0.237 391s 0.197 0.135 391s testing: mul_mat, [2, 2, 4, 2] = [4, 2, 4, 2] * [4, 2, 4, 2] 391s x0: [4, 2, 4, 2] 391s 0.746 -0.121 -0.587 0.670 391s 0.243 0.210 0.983 0.165 391s 391s x1: [4, 2, 4, 2] 391s -0.523 -0.821 0.223 0.448 391s -0.224 -0.968 -0.349 -0.965 391s 391s y: [2, 2, 4, 2] 391s -0.122 -0.006 391s -0.492 -0.759 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.224 -0.093 0.693 -0.398 391s 0.670 0.865 -0.190 0.703 391s 391s x1: [4, 2, 1, 1] 391s 0.567 -0.572 -0.973 0.580 391s -0.610 -0.427 -0.096 -0.325 391s 391s y: [2, 2, 1, 1] 391s -0.980 0.478 391s 0.239 -0.988 391s testing: mul_mat, [2, 4, 4, 1] = [2, 2, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 2, 4, 1] 391s -0.585 -0.727 391s 0.370 -0.135 391s 391s x1: [2, 4, 4, 1] 391s -0.148 0.428 391s 0.357 -0.663 391s 0.966 -0.550 391s 0.912 0.213 391s 391s y: [2, 4, 4, 1] 391s -0.225 -0.112 391s 0.273 0.222 391s -0.165 0.432 391s -0.688 0.309 391s testing: mul_mat, [2, 2, 4, 2] = [2, 2, 4, 2] * [2, 2, 4, 2] 391s x0: [2, 2, 4, 2] 391s -0.236 0.484 391s -0.313 0.974 391s 391s x1: [2, 2, 4, 2] 391s -0.927 -0.075 391s -0.203 -0.935 391s 391s y: [2, 2, 4, 2] 391s 0.182 0.217 391s -0.404 -0.848 391s test-mul-mat0: iter:395/500 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.593 -0.521 -0.413 391s 0.859 -0.912 0.727 391s 0.745 -0.197 -0.172 391s 0.269 0.900 0.721 391s 391s x1: [3, 2, 1, 1] 391s 0.725 -0.927 -0.343 391s -0.087 0.903 -0.709 391s 391s y: [4, 2, 1, 1] 391s 0.194 1.219 0.781 -0.886 391s -0.126 -1.415 -0.121 0.278 391s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.986 0.150 0.321 391s 0.367 -0.926 -0.541 391s 391s x1: [3, 4, 3, 1] 391s -0.998 0.805 0.162 391s 0.904 0.677 0.225 391s 0.900 -0.687 0.975 391s 0.381 -0.076 0.138 391s 391s y: [2, 4, 3, 1] 391s -0.812 -1.199 391s 1.066 -0.417 391s 1.098 0.439 391s 0.408 0.135 391s testing: mul_mat, [4, 2, 3, 4] = [3, 4, 3, 4] * [3, 2, 3, 4] 391s x0: [3, 4, 3, 4] 391s -0.295 -0.026 0.382 391s 0.675 0.400 0.098 391s -0.764 -0.056 -0.557 391s 0.459 0.862 0.907 391s 391s x1: [3, 2, 3, 4] 391s 0.676 0.356 0.013 391s -0.086 0.536 -0.631 391s 391s y: [4, 2, 3, 4] 391s -0.203 0.600 -0.544 0.629 391s -0.229 0.094 0.388 -0.150 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.696 -0.017 0.722 391s 0.592 -0.065 0.517 391s 391s x1: [3, 4, 1, 1] 391s 0.202 0.131 0.885 391s 0.847 -0.564 0.187 391s 0.774 0.481 0.505 391s -0.044 0.544 -0.453 391s 391s y: [2, 4, 1, 1] 391s 0.496 0.569 391s -0.445 0.635 391s -0.182 0.688 391s -0.306 -0.296 391s testing: mul_mat, [1, 3, 3, 1] = [2, 1, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.571 -0.251 391s 391s x1: [2, 3, 3, 1] 391s -0.452 0.284 391s 0.492 -0.898 391s 0.206 -0.876 391s 391s y: [1, 3, 3, 1] 391s 0.187 391s -0.055 391s 0.102 391s testing: mul_mat, [4, 2, 3, 4] = [1, 4, 3, 4] * [1, 2, 3, 4] 391s x0: [1, 4, 3, 4] 391s -0.546 391s 0.879 391s -0.391 391s -0.573 391s 391s x1: [1, 2, 3, 4] 391s -0.722 391s -0.112 391s 391s y: [4, 2, 3, 4] 391s 0.395 -0.635 0.283 0.414 391s 0.061 -0.098 0.044 0.064 391s test-mul-mat0: iter:396/500 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.880 0.703 -0.010 391s -0.780 0.274 0.841 391s -0.762 0.520 -0.763 391s 391s x1: [3, 4, 1, 1] 391s -0.992 -0.251 -0.913 391s 0.810 0.240 0.410 391s -0.749 0.755 0.862 391s -0.162 -0.276 -0.805 391s 391s y: [3, 4, 1, 1] 391s 0.706 -0.063 1.322 391s -0.548 -0.221 -0.805 391s 1.181 1.516 0.305 391s -0.044 -0.627 0.594 391s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 1, 3, 1] 391s 0.198 0.063 -0.710 391s 391s x1: [3, 3, 3, 1] 391s 0.933 -0.666 0.322 391s 0.168 -0.183 -0.638 391s -0.720 -0.848 0.012 391s 391s y: [1, 3, 3, 1] 391s -0.086 391s 0.475 391s -0.204 391s testing: mul_mat, [4, 1, 3, 1] = [3, 4, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.303 -0.036 -0.499 391s 0.922 0.530 0.722 391s -0.338 0.799 0.934 391s 0.621 0.342 0.986 391s 391s x1: [3, 1, 3, 1] 391s -0.421 -0.634 -0.015 391s 391s y: [4, 1, 3, 1] 391s -0.097 -0.735 -0.378 -0.493 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.580 -0.861 0.761 391s 0.380 -0.592 -0.015 391s 391s x1: [3, 4, 1, 1] 391s -0.804 -0.059 0.817 391s -0.871 -0.437 0.159 391s -0.884 0.381 0.563 391s -0.608 0.331 0.365 391s 391s y: [2, 4, 1, 1] 391s 0.205 -0.283 391s -0.008 -0.074 391s -0.413 -0.570 391s -0.361 -0.433 391s testing: mul_mat, [4, 3, 3, 1] = [2, 4, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 4, 3, 1] 391s 0.985 0.845 391s -0.519 -0.679 391s -0.958 -0.103 391s 0.316 0.225 391s 391s x1: [2, 3, 3, 1] 391s -0.897 0.408 391s 0.915 -0.237 391s -0.537 0.438 391s 391s y: [4, 3, 3, 1] 391s -0.538 0.188 0.816 -0.192 391s 0.701 -0.314 -0.851 0.236 391s -0.159 -0.018 0.469 -0.071 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.414 0.857 0.584 0.055 391s -0.193 0.532 -0.478 -0.597 391s 391s x1: [4, 2, 3, 1] 391s 0.523 -0.181 -0.576 0.058 391s 0.423 0.341 -0.291 0.408 391s 391s y: [2, 2, 3, 1] 391s -0.272 0.044 391s 0.320 -0.004 391s test-mul-mat0: iter:397/500 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.374 391s 391s x1: [1, 1, 1, 1] 391s 0.349 391s 391s y: [1, 1, 1, 1] 391s -0.130 391s testing: mul_mat, [1, 1, 2, 1] = [1, 1, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.705 391s 391s x1: [1, 1, 2, 1] 391s -0.237 391s 391s y: [1, 1, 2, 1] 391s 0.167 391s testing: mul_mat, [3, 1, 2, 2] = [1, 3, 2, 2] * [1, 1, 2, 2] 391s x0: [1, 3, 2, 2] 391s -0.431 391s -0.764 391s -0.717 391s 391s x1: [1, 1, 2, 2] 391s 0.005 391s 391s y: [3, 1, 2, 2] 391s -0.002 -0.004 -0.004 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.847 391s 0.322 391s -0.956 391s 391s x1: [1, 3, 1, 1] 391s 0.852 391s 0.334 391s 0.830 391s 391s y: [3, 3, 1, 1] 391s -0.721 0.274 -0.815 391s -0.283 0.108 -0.320 391s -0.703 0.267 -0.794 391s testing: mul_mat, [3, 1, 2, 1] = [3, 3, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.437 0.387 0.623 391s -0.548 0.915 -0.802 391s 0.940 -0.491 -0.686 391s 391s x1: [3, 1, 2, 1] 391s -0.898 0.948 -0.193 391s 391s y: [3, 1, 2, 1] 391s -0.146 1.514 -1.177 391s testing: mul_mat, [2, 3, 2, 2] = [3, 2, 2, 2] * [3, 3, 2, 2] 391s x0: [3, 2, 2, 2] 391s 0.175 -0.149 -0.149 391s -0.910 -0.503 -0.229 391s 391s x1: [3, 3, 2, 2] 391s -0.790 0.059 0.968 391s -0.637 -0.619 -0.989 391s -0.535 -0.671 -0.182 391s 391s y: [2, 3, 2, 2] 391s -0.292 0.468 391s 0.128 1.117 391s 0.033 0.866 391s test-mul-mat0: iter:398/500 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.628 -0.830 -0.915 0.470 391s -0.914 0.811 -0.513 -0.459 391s 391s x1: [4, 3, 1, 1] 391s 0.476 0.005 0.974 -0.348 391s 0.095 -0.176 0.149 0.946 391s 0.596 -0.835 -0.005 -0.261 391s 391s y: [2, 3, 1, 1] 391s -0.760 -0.771 391s 0.513 -0.740 391s 0.949 -1.099 391s testing: mul_mat, [2, 2, 2, 1] = [4, 2, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 2, 2, 1] 391s 0.243 0.017 0.392 -0.762 391s 0.756 0.033 0.866 0.927 391s 391s x1: [4, 2, 2, 1] 391s -0.104 -0.234 -0.914 -0.238 391s 0.184 0.952 -0.318 0.950 391s 391s y: [2, 2, 2, 1] 391s -0.206 -1.097 391s -0.788 0.775 391s testing: mul_mat, [2, 2, 2, 3] = [4, 2, 2, 3] * [4, 2, 2, 3] 391s x0: [4, 2, 2, 3] 391s -0.222 -0.487 0.426 0.413 391s 0.124 0.421 0.384 0.748 391s 391s x1: [4, 2, 2, 3] 391s -0.360 -0.412 0.773 -0.408 391s 0.270 0.722 0.789 0.893 391s 391s y: [2, 2, 2, 3] 391s 0.441 -0.227 391s 0.294 1.308 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.384 0.692 -0.610 -0.355 391s 391s x1: [4, 2, 1, 1] 391s 0.880 -0.360 -0.189 0.420 391s 0.255 -0.651 0.951 -0.878 391s 391s y: [1, 2, 1, 1] 391s 0.054 391s -0.621 391s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 3, 2, 1] 391s 0.632 391s 0.787 391s 0.535 391s 391s x1: [1, 4, 2, 1] 391s 0.811 391s -0.807 391s -0.581 391s 0.355 391s 391s y: [3, 4, 2, 1] 391s 0.513 0.638 0.434 391s -0.510 -0.635 -0.432 391s -0.367 -0.457 -0.311 391s 0.225 0.280 0.190 391s testing: mul_mat, [1, 1, 2, 3] = [3, 1, 2, 3] * [3, 1, 2, 3] 391s x0: [3, 1, 2, 3] 391s 0.048 0.829 0.894 391s 391s x1: [3, 1, 2, 3] 391s -0.123 -0.827 0.311 391s 391s y: [1, 1, 2, 3] 391s -0.413 391s test-mul-mat0: iter:399/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.207 -0.350 0.566 391s 0.626 0.227 -0.387 391s 391s x1: [3, 1, 1, 1] 391s 0.992 -0.019 -0.507 391s 391s y: [2, 1, 1, 1] 391s -0.075 0.813 391s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.826 0.587 0.558 391s 0.800 -0.187 -0.829 391s 0.255 0.934 -0.846 391s -0.451 -0.086 0.137 391s 391s x1: [3, 2, 4, 1] 391s 0.455 0.121 0.982 391s 0.293 -0.021 -0.017 391s 391s y: [4, 2, 4, 1] 391s 0.243 -0.472 -0.602 -0.081 391s -0.264 0.253 0.070 -0.133 391s testing: mul_mat, [4, 4, 4, 4] = [3, 4, 4, 4] * [3, 4, 4, 4] 391s x0: [3, 4, 4, 4] 391s 0.482 -0.937 0.937 391s 0.523 -0.117 0.738 391s -0.098 -0.534 -0.616 391s -0.741 -0.531 -0.592 391s 391s x1: [3, 4, 4, 4] 391s 0.568 -0.032 0.057 391s 0.279 0.020 -0.036 391s -0.547 0.200 -0.144 391s -0.929 0.499 0.715 391s 391s y: [4, 4, 4, 4] 391s 0.357 0.343 -0.074 -0.438 391s 0.082 0.117 -0.016 -0.196 391s -0.586 -0.416 0.036 0.385 391s -0.246 -0.017 -0.616 0.000 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.276 0.701 0.647 391s 0.020 0.265 0.494 391s 391s x1: [3, 4, 1, 1] 391s -0.198 0.217 0.329 391s -0.546 0.522 0.885 391s 0.771 0.708 0.249 391s -0.947 0.551 -0.295 391s 391s y: [2, 4, 1, 1] 391s 0.310 0.216 391s 0.788 0.565 391s 0.870 0.326 391s -0.066 -0.019 391s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.308 -0.521 391s -0.168 0.233 391s 0.739 -0.686 391s 0.957 -0.243 391s 391s x1: [2, 3, 4, 1] 391s -0.643 -0.721 391s -0.914 -0.306 391s 0.004 -0.248 391s 391s y: [4, 3, 4, 1] 391s 0.574 -0.060 0.020 -0.440 391s 0.441 0.083 -0.466 -0.801 391s 0.128 -0.058 0.173 0.064 391s testing: mul_mat, [3, 2, 4, 4] = [4, 3, 4, 4] * [4, 2, 4, 4] 391s x0: [4, 3, 4, 4] 391s 0.601 0.559 -0.083 -0.871 391s 0.980 -0.091 -0.342 -0.905 391s 0.225 0.602 -0.394 -0.889 391s 391s x1: [4, 2, 4, 4] 391s 0.050 0.549 0.012 0.528 391s -0.218 0.326 -0.715 -0.899 391s 391s y: [3, 2, 4, 4] 391s -0.124 -0.484 -0.132 391s 0.893 0.815 1.228 391s test-mul-mat0: iter:400/500 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.595 391s 0.317 391s -0.969 391s 391s x1: [1, 3, 1, 1] 391s -0.271 391s 0.259 391s -0.567 391s 391s y: [3, 3, 1, 1] 391s -0.161 -0.086 0.262 391s 0.154 0.082 -0.251 391s -0.337 -0.180 0.549 391s testing: mul_mat, [4, 3, 3, 1] = [1, 4, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.697 391s 0.547 391s 0.354 391s 0.032 391s 391s x1: [1, 3, 3, 1] 391s 0.397 391s 0.128 391s 0.301 391s 391s y: [4, 3, 3, 1] 391s 0.277 0.217 0.140 0.013 391s 0.089 0.070 0.045 0.004 391s 0.210 0.165 0.106 0.010 391s testing: mul_mat, [4, 4, 3, 3] = [1, 4, 3, 3] * [1, 4, 3, 3] 391s x0: [1, 4, 3, 3] 391s 0.112 391s 0.966 391s 0.690 391s -0.483 391s 391s x1: [1, 4, 3, 3] 391s 0.155 391s 0.413 391s -0.384 391s -0.587 391s 391s y: [4, 4, 3, 3] 391s 0.017 0.149 0.107 -0.075 391s 0.046 0.399 0.285 -0.200 391s -0.043 -0.371 -0.265 0.186 391s -0.066 -0.567 -0.405 0.284 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.372 391s 391s x1: [1, 4, 1, 1] 391s 0.300 391s 0.059 391s 0.981 391s -0.095 391s 391s y: [1, 4, 1, 1] 391s 0.111 391s 0.022 391s 0.364 391s -0.035 391s testing: mul_mat, [2, 1, 3, 1] = [1, 2, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 2, 3, 1] 391s 0.365 391s -0.975 391s 391s x1: [1, 1, 3, 1] 391s -0.529 391s 391s y: [2, 1, 3, 1] 391s -0.193 0.515 391s testing: mul_mat, [1, 1, 3, 3] = [2, 1, 3, 3] * [2, 1, 3, 3] 391s x0: [2, 1, 3, 3] 391s -0.521 0.680 391s 391s x1: [2, 1, 3, 3] 391s 0.242 0.414 391s 391s y: [1, 1, 3, 3] 391s 0.155 391s test-mul-mat0: iter:401/500 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.282 -0.080 0.071 0.525 391s -0.025 0.598 0.697 0.056 391s -0.574 0.852 0.088 -0.462 391s 0.250 -0.466 -0.542 0.676 391s 391s x1: [4, 2, 1, 1] 391s 0.028 -0.395 0.974 0.104 391s -0.068 0.579 -0.761 0.240 391s 391s y: [4, 2, 1, 1] 391s 0.148 0.447 -0.315 -0.266 391s 0.045 -0.169 0.354 0.288 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.123 0.126 0.084 -0.430 391s 391s x1: [4, 4, 3, 1] 391s 0.547 0.669 -0.941 0.383 391s -0.548 0.727 -0.589 0.056 391s 0.701 0.514 0.989 0.279 391s 0.754 0.229 0.369 -0.529 391s 391s y: [1, 4, 3, 1] 391s -0.227 391s 0.085 391s -0.059 391s 0.194 391s testing: mul_mat, [4, 1, 3, 4] = [4, 4, 3, 4] * [4, 1, 3, 4] 391s x0: [4, 4, 3, 4] 391s 0.977 0.836 -0.982 -0.916 391s -0.179 -0.718 0.738 0.920 391s 0.119 0.941 0.930 -0.671 391s -0.847 0.322 0.260 0.501 391s 391s x1: [4, 1, 3, 4] 391s 0.383 0.721 0.137 0.914 391s 391s y: [4, 1, 3, 4] 391s 0.005 0.355 0.238 0.402 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1]ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.649 0.114 0.127 -0.056 391s -0.956 -0.537 -0.093 -0.546 391s 0.556 0.900 -0.434 -0.830 391s -0.820 -0.900 0.770 0.280 391s 391s x1: [4, 4, 1, 1] 391s -0.601 -0.664 0.611 -0.554 391s 0.250 -0.622 0.802 0.928 391s 0.003 0.035 0.928 -0.319 391s -0.125 0.931 -0.075 0.806 391s 391s y: [4, 4, 1, 1] 391s 0.422 1.176 -0.737 1.406 391s -0.184 -0.485 -1.540 1.233 391s 0.138 0.066 -0.105 0.592 391s 0.133 -0.814 0.132 -0.568 391s testing: mul_mat, [3, 4, 3, 1] = [4, 3, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.471 -0.988 -0.599 0.734 391s 0.989 0.381 0.339 0.751 391s 0.336 -0.033 -0.631 0.941 391s 391s x1: [4, 4, 3, 1] 391s -0.935 -0.384 -0.470 -0.557 391s -0.582 -0.543 0.446 0.453 391s -0.614 -0.873 -0.672 -0.683 391s 0.052 -0.866 -0.514 0.403 391s 391s y: [3, 4, 3, 1] 391s 0.693 -1.649 -0.529 391s 0.875 -0.292 -0.033 391s 1.053 -1.681 -0.397 391s 1.435 -0.150 0.749 391s testing: mul_mat, [4, 4, 3, 4] = [3, 4, 3, 4] * [3, 4, 3, 4] 391s x0: [3, 4, 3, 4] 391s 0.613 -0.013 -0.545 391s 0.155 0.279 -0.421 391s 0.868 0.564 -0.572 391s 0.506 0.190 0.909 391s 391s x1: [3, 4, 3, 4] 391s 0.300 0.440 0.102 391s 0.670 0.174 -0.147 391s 0.611 0.521 0.081 391s 0.739 0.011 -0.772 391s 391s y: [4, 4, 3, 4] 391s 0.123 0.126 0.450 0.328 391s 0.488 0.214 0.764 0.238 391s 0.324 0.206 0.778 0.481 391s 0.874 0.442 1.089 -0.325 391s test-mul-mat0: iter:402/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.809 0.613 391s -0.756 0.594 391s 391s x1: [2, 4, 1, 1] 391s 0.125 0.496 391s -0.211 -0.669 391s 0.174 -0.715 391s 0.404 0.289 391s 391s y: [2, 4, 1, 1] 391s 0.405 0.201 391s -0.581 -0.239 391s -0.298 -0.556 391s 0.504 -0.133 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.407 0.369 391s 0.091 0.450 391s -0.360 0.678 391s 391s x1: [2, 2, 1, 1] 391s 0.104 -0.869 391s 0.845 -0.515 391s 391s y: [3, 2, 1, 1] 391s -0.363 -0.381 -0.627 391s -0.533 -0.155 -0.654 391s testing: mul_mat, [1, 3, 1, 2] = [2, 1, 1, 2] * [2, 3, 1, 2] 391s x0: [2, 1, 1, 2] 391s -0.770 -0.273 391s 391s x1: [2, 3, 1, 2] 391s 0.223 0.746 391s -0.340 -0.216 391s 0.371 -0.232 391s 391s y: [1, 3, 1, 2] 391s -0.375 391s 0.321 391s -0.222 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.444 0.930 391s -0.707 0.246 391s 0.479 -0.927 391s 391s x1: [2, 1, 1, 1] 391s -0.660 -0.838 391s 391s y: [3, 1, 1, 1] 391s -1.072 0.260 0.461 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.325 0.055 -0.922 391s 0.530 0.443 0.551 391s 391s x1: [3, 2, 1, 1] 391s 0.298 -0.663 0.522 391s 0.938 -0.985 -0.254 391s 391s y: [2, 2, 1, 1] 391s -0.615 0.152 391s -0.125 -0.079 391s testing: mul_mat, [1, 3, 1, 2] = [2, 1, 1, 2] * [2, 3, 1, 2] 391s x0: [2, 1, 1, 2] 391s 0.412 0.760 391s 391s x1: [2, 3, 1, 2] 391s 0.707 0.416 391s -0.623 0.145 391s 0.039 -0.392 391s 391s y: [1, 3, 1, 2] 391s 0.608 391s -0.146 391s -0.282 391s test-mul-mat0: iter:403/500 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.981 -0.837 -0.901 391s 0.853 0.120 -0.255 391s -0.934 -0.761 0.026 391s 0.723 0.651 -0.214 391s 391s x1: [3, 4, 1, 1] 391s -0.177 0.011 -0.899 391s -0.492 0.686 0.631 391s 0.563 0.129 0.709 391s 0.114 -0.164 0.124 391s 391s y: [4, 4, 1, 1] 391s 0.628 0.079 0.133 0.071 391s -1.626 -0.499 -0.046 -0.044 391s -0.194 0.315 -0.605 0.340 391s 0.137 0.046 0.022 -0.051 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.130 0.414 -0.797 391s 0.966 -0.461 0.694 391s 391s x1: [3, 4, 1, 1] 391s 0.310 -0.691 -0.381 391s -0.804 0.305 0.974 391s 0.019 -0.684 -0.925 391s 0.527 -0.999 0.705 391s 391s y: [2, 4, 1, 1] 391s 0.058 0.354 391s -0.754 -0.241 391s 0.456 -0.308 391s -0.907 1.459 391s testing: mul_mat, [4, 2, 1, 3] = [3, 4, 1, 3] * [3, 2, 1, 3] 391s x0: [3, 4, 1, 3] 391s 0.641 0.019 0.642 391s 0.642 -0.276 -0.269 391s -0.228 -0.861 -0.066 391s -0.262 -0.323 -0.371 391s 391s x1: [3, 2, 1, 3] 391s 0.947 -0.298 0.793 391s 0.800 0.821 -0.462 391s 391s y: [4, 2, 1, 3] 391s 1.110 0.477 -0.011 -0.446 391s 0.232 0.412 -0.859 -0.303 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.369 -0.843 -0.387 391s 0.187 -0.232 0.353 391s 0.893 0.415 0.464 391s -0.427 0.148 0.402 391s 391s x1: [3, 4, 1, 1] 391s -0.575 -0.629 -0.329 391s 0.359 0.109 0.348 391s 0.988 0.794 -0.273 391s -0.591 -0.721 0.927 391s 391s y: [4, 4, 1, 1] 391s 0.869 -0.077 -0.927 0.020 391s -0.359 0.165 0.528 0.003 391s -0.928 -0.096 1.084 -0.414 391s 0.466 0.384 -0.396 0.518 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.706 0.280 -0.584 0.205 391s -0.511 -0.426 -0.351 0.603 391s 391s x1: [4, 3, 1, 1] 391s 0.405 0.421 0.210 0.099 391s 0.020 0.743 0.524 0.391 391s -0.586 -0.117 -0.500 0.762 391s 391s y: [2, 3, 1, 1] 391s -0.270 -0.400 391s -0.032 -0.275 391s 0.829 0.984 391s testing: mul_mat, [2, 4, 1, 3] = [2, 2, 1, 3] * [2, 4, 1, 3] 391s x0: [2, 2, 1, 3] 391s 0.609 0.357 391s 0.574 -0.743 391s 391s x1: [2, 4, 1, 3] 391s -0.459 0.778 391s 0.760 0.309 391s 0.193 -0.092 391s 0.921 -0.455 391s 391s y: [2, 4, 1, 3] 391s -0.001 -0.841 391s 0.573 0.207 391s 0.084 0.179 391s 0.398 0.867 391s test-mul-mat0: iter:404/500 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.392 391s -0.839 391s 391s x1: [1, 3, 1, 1] 391s -0.755 391s 0.794 391s 0.885 391s 391s y: [2, 3, 1, 1] 391s 0.296 0.634 391s -0.311 -0.666 391s -0.347 -0.743 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.325 391s 0.520 391s 391s x1: [1, 2, 1, 1] 391s 0.868 391s -0.186 391s 391s y: [2, 2, 1, 1] 391s 0.282 0.451 391s -0.060 -0.097 391s testing: mul_mat, [1, 2, 1, 2] = [1, 1, 1, 2] * [1, 2, 1, 2] 391s x0: [1, 1, 1, 2] 391s -0.906 391s 391s x1: [1, 2, 1, 2] 391s 0.302 391s 0.653 391s 391s y: [1, 2, 1, 2] 391s -0.273 391s -0.591 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.440 391s 391s x1: [1, 1, 1, 1] 391s 0.519 391s 391s y: [1, 1, 1, 1] 391s -0.228 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.280 391s 391s x1: [1, 1, 1, 1] 391s -0.682 391s 391s y: [1, 1, 1, 1] 391s -0.191 391s testing: mul_mat, [2, 1, 1, 2] = [1, 2, 1, 2] * [1, 1, 1, 2] 391s x0: [1, 2, 1, 2] 391s -0.545 391s 0.473 391s 391s x1: [1, 1, 1, 2] 391s -0.575 391s 391s y: [2, 1, 1, 2] 391s 0.313 -0.272 391s test-mul-mat0: iter:405/500 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.875 -0.297 391s -0.795 0.645 391s -0.995 0.857 391s -0.835 -0.752 391s 391s x1: [2, 1, 1, 1] 391s 0.311 0.853 391s 391s y: [4, 1, 1, 1] 391s -0.525 0.302 0.422 -0.901 391s testing: mul_mat, [2, 4, 3, 1] = [2, 2, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.029 0.287 391s 0.896 -0.114 391s 391s x1: [2, 4, 3, 1] 391s 0.119 -0.741 391s 0.848 -0.362 391s 0.132 -0.592 391s -0.045 -0.439 391s 391s y: [2, 4, 3, 1] 391s -0.209 0.191 391s -0.080 0.801 391s -0.166 0.185 391s -0.127 0.010 391s testing: mul_mat, [3, 2, 3, 4] = [2, 3, 3, 4] * [2, 2, 3, 4] 391s x0: [2, 3, 3, 4] 391s -0.906 0.409 391s -0.052 -0.143 391s 0.381 0.239 391s 391s x1: [2, 2, 3, 4] 391s 0.598 0.403 391s -0.913 -0.022 391s 391s y: [3, 2, 3, 4] 391s -0.376 -0.089 0.324 391s 0.818 0.051 -0.353 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.489 0.185 391s -0.901 0.579 391s 0.530 -0.024 391s -0.504 -0.421 391s 391s x1: [2, 3, 1, 1] 391s 0.143 0.181 391s -0.340 -0.452 391s 0.988 -0.325 391s 391s y: [4, 3, 1, 1] 391s -0.036 -0.024 0.072 -0.148 391s 0.082 0.045 -0.170 0.362 391s -0.543 -1.078 0.532 -0.361 391s testing: mul_mat, [4, 2, 3, 1] = [4, 4, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.439 0.370 -0.317 0.279 391s -0.606 0.847 0.113 0.501 391s 0.268 0.877 -0.866 0.957 391s -0.861 0.588 -0.253 -0.333 391s 391s x1: [4, 2, 3, 1] 391s 0.738 -0.782 -0.164 0.236 391s -0.455 -0.936 -0.833 0.367 391s 391s y: [4, 2, 3, 1] 391s -0.496 -1.010 -0.120 -1.132 391s 0.220 -0.428 0.131 -0.071 391s testing: mul_mat, [1, 4, 3, 4] = [4, 1, 3, 4] * [4, 4, 3, 4] 391s x0: [4, 1, 3, 4] 391s -0.999 -0.855 -0.493 -0.452 391s 391s x1: [4, 4, 3, 4] 391s 0.327 -0.611 -0.508 0.016 391s -0.635 0.197 -0.895 0.650 391s 0.592 -0.421 -0.082 0.018 391s -0.048 -0.731 0.004 0.719 391s 391s y: [1, 4, 3, 4] 391s 0.440 391s 0.613 391s -0.199 391s 0.346 391s test-mul-mat0: iter:406/500 391s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.624 391s 391s x1: [1, 2, 1, 1] 391s -0.737 391s 0.198 391s 391s y: [1, 2, 1, 1] 391s 0.460 391s -0.123 391s testing: mul_mat, [4, 1, 2, 1] = [1, 4, 2, 1] * [1, 1, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.344 391s -0.879 391s 0.863 391s -0.610 391s 391s x1: [1, 1, 2, 1] 391s -0.078 391s 391s y: [4, 1, 2, 1] 391s 0.027 0.068 -0.067 0.047 391s testing: mul_mat, [4, 4, 2, 3] = [1, 4, 2, 3] * [1, 4, 2, 3] 391s x0: [1, 4, 2, 3] 391s -0.162 391s 0.064 391s 0.232 391s 0.106 391s 391s x1: [1, 4, 2, 3] 391s 0.218 391s -0.894 391s 0.692 391s -0.881 391s 391s y: [4, 4, 2, 3] 391s -0.035 0.014 0.050 0.023 391s 0.144 -0.058 -0.207 -0.095 391s -0.112 0.045 0.160 0.073 391s 0.142 -0.057 -0.204 -0.093 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.702 391s -0.220 391s -0.925 391s -0.233 391s 391s x1: [1, 4, 1, 1] 391s -0.711 391s -0.740 391s 0.028 391s 0.474 391s 391s y: [4, 4, 1, 1] 391s -0.500 0.157 0.658 0.166 391s -0.519 0.163 0.684 0.173 391s 0.020 -0.006 -0.026 -0.007 391s 0.333 -0.104 -0.438 -0.111 391s tesggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ting: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 1, 2, 1] 391s -0.189 -0.473 -0.253 -0.682 391s 391s x1: [4, 1, 2, 1] 391s -0.988 0.180 -0.497 0.952 391s 391s y: [1, 1, 2, 1] 391s -0.423 391s testing: mul_mat, [4, 4, 2, 3] = [1, 4, 2, 3] * [1, 4, 2, 3] 391s x0: [1, 4, 2, 3] 391s -0.573 391s -0.782 391s 0.548 391s 0.417 391s 391s x1: [1, 4, 2, 3] 391s 0.605 391s -0.755 391s -0.002 391s -0.145 391s 391s y: [4, 4, 2, 3] 391s -0.346 -0.473 0.332 0.252 391s 0.432 0.590 -0.414 -0.315 391s 0.001 0.001 -0.001 -0.001 391s 0.083 0.114 -0.080 -0.061 391s test-mul-mat0: iter:407/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.128 391s -0.971 391s 0.834 391s -0.324 391s 391s x1: [1, 1, 1, 1] 391s 0.601 391s 391s y: [4, 1, 1, 1] 391s 0.077 -0.584 0.502 -0.195 391s testing: mul_mat, [2, 4, 4, 1] = [1, 2, 4, 1] * [1, 4, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.627 391s 0.825 391s 391s x1: [1, 4, 4, 1] 391s 0.446 391s -0.615 391s 0.659 391s -0.532 391s 391s y: [2, 4, 4, 1] 391s 0.279 0.368 391s -0.386 -0.507 391s 0.413 0.544 391s -0.334 -0.439 391s testing: mul_mat, [1, 2, 4, 3] = [1, 1, 4, 3] * [1, 2, 4, 3] 391s x0: [1, 1, 4, 3] 391s 0.858 391s 391s x1: [1, 2, 4, 3] 391s -0.213 391s 0.583 391s 391s y: [1, 2, 4, 3] 391s -0.183 391s 0.500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.490 391s 0.635 391s -0.058 391s -0.551 391s 391s x1: [1, 1, 1, 1] 391s -0.094 391s 391s y: [4, 1, 1, 1] 391s -0.046 -0.059 0.005 0.052 391s testing: mul_mat, [1, 1, 4, 1] = [4, 1, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.711 -0.684 0.903 -0.549 391s 391s x1: [4, 1, 4, 1] 391s -0.291 0.154 -0.684 0.119 391s 391s y: [1, 1, 4, 1] 391s -0.995 391s testing: mul_mat, [4, 4, 4, 3] = [1, 4, 4, 3] * [1, 4, 4, 3] 391s x0: [1, 4, 4, 3] 391s 0.003 391s 0.903 391s 0.372 391s -0.500 391s 391s x1: [1, 4, 4, 3] 391s 0.255 391s -0.083 391s 0.589 391s -0.625 391s 391s y: [4, 4, 4, 3] 391s 0.001 0.230 0.095 -0.128 391s -0.000 -0.075 -0.031 0.042 391s 0.002 0.532 0.219 -0.295 391s -0.002 -0.564 -0.232 0.313 391s test-mul-mat0: iter:408/500 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.123 391s 0.307 391s 0.110 391s 0.879 391s 391s x1: [1, 1, 1, 1] 391s -0.899 391s 391s y: [4, 1, 1, 1] 391s 0.111 -0.276 -0.099 -0.790 391s testing: mul_mat, [3, 4, 4, 1] = [1, 3, 4, 1] * [1, 4, 4, 1] 391s x0: [1, 3, 4, 1] 391s -0.446 391s -0.904 391s -0.212 391s 391s x1: [1, 4, 4, 1] 391s -0.351 391s -0.235 391s -0.130 391s -0.528 391s 391s y: [3, 4, 4, 1] 391s 0.156 0.317 0.074 391s 0.105 0.213 0.050 391s 0.058 0.118 0.028 391s 0.236 0.478 0.112 391s testing: mul_mat, [3, 3, 4, 2] = [1, 3, 4, 2] * [1, 3, 4, 2] 391s x0: [1, 3, 4, 2] 391s -0.202 391s -0.412 391s -0.019 391s 391s x1: [1, 3, 4, 2] 391s 0.731 391s -0.560 391s 0.438 391s 391s y: [3, 3, 4, 2] 391s -0.148 -0.301 -0.014 391s 0.113 0.231 0.011 391s -0.088 -0.180 -0.008 391s testing: mul_mat, [3, 3, 1, 1] = [1, 3, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.542 391s -0.551 391s -0.233 391s 391s x1: [1, 3, 1, 1] 391s 0.150 391s -0.288 391s 0.546 391s 391s y: [3, 3, 1, 1] 391s -0.081 -0.083 -0.035 391s 0.156 0.159 0.067 391s -0.295 -0.301 -0.127 391s testing: mul_mat, [2, 1, 4, 1] = [3, 2, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.354 0.107 0.383 391s -0.465 0.163 0.972 391s 391s x1: [3, 1, 4, 1] 391s 0.257 0.037 0.748 391s 391s y: [2, 1, 4, 1] 391s 0.199 0.614 391s testing: mul_mat, [3, 3, 4, 2] = [2, 3, 4, 2] * [2, 3, 4, 2] 391s x0: [2, 3, 4, 2] 391s 0.807 0.215 391s -0.195 -0.809 391s 0.759 -0.203 391s 391s x1: [2, 3, 4, 2] 391s -0.130 0.574 391s -0.033 -0.655 391s -0.752 -0.191 391s 391s y: [3, 3, 4, 2] 391s 0.019 -0.439 -0.215 391s -0.168 0.537 0.108 391s -0.648 0.301 -0.532 391s test-mul-mat0: iter:409/500 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.134 -0.378 0.868 -0.851 391s 391s x1: [4, 3, 1, 1] 391s 0.928 -0.998 0.344 0.110 391s -0.946 -0.183 0.249 -0.286 391s -0.182 -0.316 0.182 0.625 391s 391s y: [1, 3, 1, 1] 391s 0.458 391s 0.656 391s -0.230 391s testing: mul_mat, [2, 1, 4, 1] = [4, 2, 4, 1] * [4, 1, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.618 -0.823 -0.754 -0.132 391s -0.109 0.064 0.552 -0.927 391s 391s x1: [4, 1, 4, 1] 391s 0.562 0.305 -0.857 0.240 391s 391s y: [2, 1, 4, 1] 391s 0.711 -0.738 391s testing: mul_mat, [4, 2, 4, 2] = [4, 4, 4, 2] * [4, 2, 4, 2] 391s x0: [4, 4, 4, 2] 391s -0.023 -0.504 0.064 0.655 391s 0.633 -0.538 0.911 -0.978 391s 0.779 -0.168 -0.463 0.617 391s 0.649 0.293 0.215 0.792 391s 391s x1: [4, 2, 4, 2] 391s 0.674 -0.813 0.637 -0.435 391s 0.251 0.189 -0.361 0.941 391s 391s y: [4, 2, 4, 2] 391s 0.150 1.870 0.099 -0.008 391s 0.492 -1.192 0.911 0.886 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.886 -0.263 -0.139 0.489 391s 391s x1: [4, 4, 1, 1] 391s -0.566 -0.955 -0.207 -0.349 391s -0.554 0.621 -0.631 -0.175 391s 0.384 0.294 0.924 -0.953 391s -0.140 -0.819 -0.100 0.585 391s 391s y: [1, 4, 1, 1] 391s 0.611 391s 0.329 391s -1.013 391s 0.640 391s testing: mul_mat, [3, 4, 4, 1] = [1, 3, 4, 1] * [1, 4, 4, 1] 391s x0: [1, 3, 4, 1] 391s 0.232 391s -0.203 391s 0.093 391s 391s x1: [1, 4, 4, 1] 391s 0.794 391s 0.665 391s 0.149 391s 0.388 391s 391s y: [3, 4, 4, 1] 391s 0.184 -0.161 0.074 391s 0.154 -0.135 0.062 391s 0.035 -0.030 0.014 391s 0.090 -0.079 0.036 391s testing: mul_mat, [3, 1, 4, 2] = [3, 3, 4, 2] * [3, 1, 4, 2] 391s x0: [3, 3, 4, 2] 391s -0.668 0.169 -0.132 391s 0.323 0.181 -0.471 391s 0.345 0.621 0.690 391s 391s x1: [3, 1, 4, 2] 391s 0.698 0.347 -0.930 391s 391s y: [3, 1, 4, 2] 391s -0.285 0.727 -0.186 391s test-mul-mat0: iter:410/500 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.583 -0.659 391s 0.119 0.698 391s 391s x1: [2, 3, 1, 1] 391s -0.003 0.007 391s 0.632 0.589 391s 0.773 -0.209 391s 391s y: [2, 3, 1, 1] 391s -0.003 0.005 391s -0.756 0.486 391s -0.313 -0.054 391s testing: mul_mat, [2, 2, 4, 1] = [2, 2, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 2, 4, 1] 391s -0.856 0.691 391s 0.767 0.140 391s 391s x1: [2, 2, 4, 1] 391s -0.881 -0.272 391s 0.468 0.121 391s 391s y: [2, 2, 4, 1] 391s 0.566 -0.713 391s -0.318 0.376 391s testing: mul_mat, [2, 2, 4, 4] = [2, 2, 4, 4] * [2, 2, 4, 4] 391s x0: [2, 2, 4, 4] 391s 0.640 -0.987 391s -0.549 0.226 391s 391s x1: [2, 2, 4, 4] 391s -0.945 0.771 391s 0.617 0.590 391s 391s y: [2, 2, 4, 4] 391s -1.365 0.693 391s -0.187 -0.206 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.027 0.702 391s 0.585 -0.075 391s 0.415 -0.818 391s 391s x1: [2, 2, 1, 1] 391s -0.655 -0.016 391s -0.086 -0.347 391s 391s y: [3, 2, 1, 1] 391s 0.007 -0.382 -0.259 391s -0.241 -0.024 0.248 391s testing: mul_mat, [4, 2, 4, 1] = [3, 4, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.461 0.641 0.833 391s -0.285 -0.393 -0.914 391s -0.576 -0.376 0.900 391s -0.759 0.505 0.100 391s 391s x1: [3, 2, 4, 1] 391s -0.617 0.865 0.226 391s 0.461 -0.606 0.267 391s 391s y: [4, 2, 4, 1] 391s 1.027 -0.371 0.234 0.928 391s -0.378 -0.137 0.202 -0.630 391s testing: mul_mat, [3, 3, 4, 4] = [4, 3, 4, 4] * [4, 3, 4, 4] 391s x0: [4, 3, 4, 4] 391s 0.368 -0.426 -0.303 -0.655 391s -0.414 0.685 -0.908 -0.801 391s -0.301 -0.283 0.929 -0.451 391s 391s x1: [4, 3, 4, 4] 391s -0.511 -0.940 0.144 -0.047 391s -0.886 0.787 0.549 -0.187 391s 0.083 -0.378 0.141 0.945 391s 391s y: [3, 3, 4, 4] 391s 0.199 -0.525 0.575 391s -0.705 0.556 0.638 391s -0.471 -1.179 -0.213 391s test-mul-mat0: iter:411/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.168 -0.284 391s 0.089 -0.401 391s 0.312 0.190 391s 391s x1: [2, 2, 1, 1] 391s 0.404 0.187 391s 0.377 -0.520 391s 391s y: [3, 2, 1, 1] 391s 0.015 -0.039 0.162 391s 0.211 0.242 0.019 391s testing: mul_mat, [3, 3, 3, 1] = [2, 3, 3, 1] * [2, 3, 3, 1] 391s x0: [2, 3, 3, 1] 391s -0.246 -0.548 391s 0.251 -0.869 391s -0.067 -0.504 391s 391s x1: [2, 3, 3, 1] 391s 0.729 0.663 391s -0.547 0.489 391s -0.621 0.819 391s 391s y: [3, 3, 3, 1] 391s -0.543 -0.394 -0.384 391s -0.134 -0.562 -0.210 391s -0.296 -0.867 -0.371 391s testing: mul_mat, [2, 3, 3, 3] = [2, 2, 3, 3] * [2, 3, 3, 3] 391s x0: [2, 2, 3, 3] 391s 0.051 0.227 391s -0.380 -0.295 391s 391s x1: [2, 3, 3, 3] 391s 0.388 -0.955 391s -0.891 -0.204 391s -0.896 0.488 391s 391s y: [2, 3, 3, 3] 391s -0.197 0.134 391s -0.092 0.399 391s 0.065 0.196 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.160 0.084 391s -0.948 0.159 391s 391s x1: [2, 2, 1, 1] 391s -0.445 -0.650 391s 0.285 -0.573 391s 391s y: [2, 2, 1, 1] 391s 0.017 0.318 391s -0.094 -0.362 391s testing: mul_mat, [4, 2, 3, 1] = [2, 4, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 4, 3, 1] 391s 0.778 -0.242 391s -0.482 -0.338 391s 0.975 0.821 391s 0.490 0.702 391s 391s x1: [2, 2, 3, 1] 391s -0.272 -0.165 391s -0.501 0.215 391s 391s y: [4, 2, 3, 1] 391s -0.172 0.187 -0.401 -0.249 391s -0.442 0.169 -0.312 -0.095 391s testing: mul_mat, [4, 2, 3, 3] = [4, 4, 3, 3] * [4, 2, 3, 3] 391s x0: [4, 4, 3, 3] 391s -0.634 -0.844 0.172 -0.630 391s -0.718 0.893 0.944 -0.508 391s 0.746 -0.911 0.394 0.278 391s 0.103 -0.569 -0.561 0.906 391s 391s x1: [4, 2, 3, 3] 391s 0.827 -0.222 0.889 0.641 391s -0.137 0.219 -0.844 -0.360 391s 391s y: [4, 2, 3, 3] 391s -0.588 -0.278 1.347 0.293 391s -0.016 -0.320 -0.734 0.009 391s test-mul-mat0: iter:412/500 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.112 391s 391s x1: [1, 1, 1, 1] 391s 0.148 391s 391s y: [1, 1, 1, 1] 391s 0.017 391s testing: mul_mat, [1, 1, 3, 1] = [1, 1, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 1, 3, 1] 391s -0.855 391s 391s x1: [1, 1, 3, 1] 391s -0.167 391s 391s y: [1, 1, 3, 1] 391s 0.143 391s testing: mul_mat, [1, 1, 3, 2] = [1, 1, 3, 2] * [1, 1, 3, 2] 391s x0: [1, 1, 3, 2] 391s 0.562 391s 391s x1: [1, 1, 3, 2] 391s -0.249 391s 391s y: [1, 1, 3, 2] 391s -0.140 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.210 391s -0.026 391s 391s x1: [1, 1, 1, 1] 391s 0.351 391s 391s y: [2, 1, 1, 1] 391s -0.074 -0.009 391s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 2, 3, 1] 391s -0.380 0.018 391s 0.542 -0.172 391s 391s x1: [2, 1, 3, 1] 391s 0.543 -0.361 391s 391s y: [2, 1, 3, 1] 391s -0.213 0.356 391s testing: mul_mat, [4, 2, 3, 2] = [2, 4, 3, 2] * [2, 2, 3, 2] 391s x0: [2, 4, 3, 2] 391s 0.316 -0.576 391s -0.306 -0.115 391s -0.804 0.397 391s -0.063 0.458 391s 391s x1: [2, 2, 3, 2] 391s -0.540 0.034 391s -0.748 -0.898 391s 391s y: [4, 2, 3, 2] 391s -0.190 0.162 0.447 0.049 391s 0.281 0.332 0.245 -0.365 391s test-mul-mat0: iter:413/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.702 0.020 391s -0.295 0.493 391s 391s x1: [2, 4, 1, 1] 391s -0.700 0.055 391s 0.099 -0.127 391s 0.451 0.232 391s 0.300 0.578 391s 391s y: [2, 4, 1, 1] 391s 0.493 0.233 391s -0.072 -0.092 391s -0.312 -0.019 391s -0.199 0.196 391s testing: mul_mat, [2, 2, 3, 1] = [2, 2, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.914 0.207 391s -0.549 0.474 391s 391s x1: [2, 2, 3, 1] 391s 0.399 -0.787 391s 0.572 0.176 391s 391s y: [2, 2, 3, 1] 391s 0.203 -0.592 391s 0.560 -0.231 391s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.268 -0.224 391s 391s x1: [2, 2, 3, 1] 391s 0.557 -0.307 391s 0.222 -0.423 391s 391s y: [1, 2, 3, 1] 391s -0.080 391s 0.035 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.689 -0.227 391s -0.065 0.532 391s 0.908 -0.803 391s -0.782 0.845 391s 391s x1: [2, 1, 1, 1] 391s -0.784 -0.805 391s 391s y: [4, 1, 1, 1] 391s -0.358 -0.377 -0.066 -0.067 391s testing: mul_mat, [1, 2, 3, 1] = [4, 1, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 1, 3, 1] 391s 0.655 0.494 0.425 0.187 391s 391s x1: [4, 2, 3, 1] 391s -0.544 0.755 -0.462 0.678 391s -0.668 0.936 -0.607 -0.692 391s 391s y: [1, 2, 3, 1] 391s -0.054 391s -0.362 391s testing: mul_mat, [1, 4, 3, 1] = [1, 1, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 1, 3, 1] 391s 0.086 391s 391s x1: [1, 4, 3, 1] 391s 0.714 391s -0.914 391s 0.290 391s -0.606 391s 391s y: [1, 4, 3, 1] 391s 0.061 391s -0.078 391s 0.025 391s -0.052 391s test-mul-mat0: iter:414/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.341 -0.181 391s 391s x1: [2, 4, 1, 1] 391s -0.574 -1.000 391s 0.674 -0.882 391s 0.270 0.316 391s 0.564 0.077 391s 391s y: [1, 4, 1, 1] 391s -0.015 391s 0.389 391s 0.035 391s 0.178 391s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 4, 2, 1] 391s 0.737 -0.104 391s 0.587 -0.565 391s 0.582 -0.917 391s -0.496 0.789 391s 391s x1: [2, 1, 2, 1] 391s -0.650 0.427 391s 391s y: [4, 1, 2, 1] 391s -0.523 -0.622 -0.770 0.659 391s testing: mul_mat, [2, 4, 2, 2] = [2, 2, 2, 2] * [2, 4, 2, 2] 391s x0: [2, 2, 2, 2] 391s -0.353 0.325 391s 0.913 0.846 391s 391s x1: [2, 4, 2, 2] 391s -0.712 -0.837 391s 0.636 0.558 391s 0.479 0.200 391s -0.365 -0.884 391s 391s y: [2, 4, 2, 2] 391s -0.021 -1.358 391s -0.043 1.053 391s -0.104 0.607 391s -0.158 -1.082 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.782 -0.830 391s -0.346 0.999 391s 391s x1: [2, 2, 1, 1] 391s -0.986 0.502 391s -0.563 -0.746 391s 391s y: [2, 2, 1, 1] 391s 0.355 0.843 391s 1.059 -0.550 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.650 -0.355 391s -0.299 -0.963 391s 391s x1: [2, 2, 2, 1] 391s -0.854 0.649 391s -0.731 0.157 391s 391s y: [2, 2, 2, 1] 391s -0.786 -0.370 391s -0.531 0.067 391s testing: mul_mat, [4, 2, 2, 2] = [2, 4, 2, 2] * [2, 2, 2, 2] 391s x0: [2, 4, 2, 2] 391s 0.597 0.119 391s -0.622 0.758 391s 0.142 0.051 391s 0.401 0.820 391s 391s x1: [2, 2, 2, 2] 391s -0.153 0.690 391s 0.371 -0.340 391s 391s y: [4, 2, 2, 2] 391s -0.009 0.618 0.013 0.505 391s 0.181 -0.488 0.036 -0.130 391s test-mul-mat0: iter:415/500 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.882 -0.732 391s -0.225 -0.090 391s 391s x1: [2, 2, 1, 1] 391s -0.678 -0.794 391s 0.946 -0.276 391s 391s y: [2, 2, 1, 1] 391s -0.017 0.224 391s 1.036 -0.189 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.094 -0.079 391s 0.272 -0.481 391s 0.276 -0.916 391s 391s x1: [2, 2, 1, 1] 391s -0.342 0.945 391s 0.309 0.708 391s 391s y: [3, 2, 1, 1] 391s -0.107 -0.548 -0.960 391s -0.027 -0.257 -0.563 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.421 0.050 391s 0.305 -0.295 391s 0.372 0.511 391s -0.349 -0.904 391s 391s x1: [2, 3, 1, 1] 391s -0.763 -0.056 391s 0.066 -0.605 391s 0.573 -0.513 391s 391s y: [4, 3, 1, 1] 391s 0.318 -0.216 -0.313 0.317 391s -0.058 0.198 -0.285 0.524 391s -0.267 0.326 -ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 0.049 0.264 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.844 0.650 391s 0.847 0.429 391s 391s x1: [2, 4, 1, 1] 391s -0.195 -0.467 391s -0.636 0.579 391s 0.443 0.022 391s 0.524 -0.248 391s 391s y: [2, 4, 1, 1] 391s -0.139 -0.366 391s 0.913 -0.290 391s -0.360 0.384 391s -0.603 0.338 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.422 0.146 391s -0.004 -0.978 391s 0.443 -0.885 391s 0.065 -0.549 391s 391s x1: [2, 2, 1, 1] 391s -0.635 -0.074 391s 0.513 -0.398 391s 391s y: [4, 2, 1, 1] 391s 0.257 0.074 -0.216 -0.001 391s -0.274 0.388 0.579 0.252 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.355 0.698 0.455 -0.079 391s -0.031 -0.974 -0.681 -0.855 391s 0.870 0.954 -0.725 0.546 391s -0.893 0.349 -0.032 0.918 391s 391s x1: [4, 2, 1, 1] 391s -0.273 0.487 0.961 0.379 391s 0.583 -0.234 0.912 0.947 391s 391s y: [4, 2, 1, 1] 391s 0.651 -1.444 -0.263 0.730 391s 0.384 -1.221 0.140 0.238 391s test-mul-mat0: iter:416/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.476 0.715 0.797 391s -0.758 0.627 0.745 391s 391s x1: [3, 1, 1, 1] 391s -0.273 0.514 -0.663 391s 391s y: [2, 1, 1, 1] 391s -0.291 0.036 391s testing: mul_mat, [4, 2, 3, 1] = [3, 4, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.265 0.377 0.789 391s -0.220 0.714 -0.997 391s -0.744 0.429 0.801 391s -0.503 0.057 0.545 391s 391s x1: [3, 2, 3, 1] 391s -0.413 -0.018 -0.287 391s -0.544 0.090 -0.588 391s 391s y: [4, 2, 3, 1] 391s -0.343 0.364 0.070 0.050 391s -0.575 0.770 -0.028 -0.043 391s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.825 0.101 -0.337 391s 0.603 -0.454 0.822 391s -0.949 -0.987 0.152 391s -0.069 0.442 -0.983 391s 391s x1: [3, 4, 3, 1] 391s -0.827 -0.678 -0.065 391s 0.973 -0.181 0.992 391s 0.519 0.903 -0.969 391s 0.778 0.444 0.159 391s 391s y: [4, 4, 3, 1] 391s -0.730 -0.244 1.444 -0.179 391s 0.450 1.484 -0.594 -1.121 391s 0.847 -0.894 -1.531 1.316 391s 0.633 0.398 -1.152 -0.013 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.773 -0.947 -0.696 391s -0.445 -0.530 -0.544 391s 0.015 0.233 -0.980 391s -0.350 0.884 0.335 391s 391s x1: [3, 4, 1, 1] 391s -0.660 0.616 -0.308 391s 0.492 -0.453 -0.866 391s 0.509 -0.175 0.870 391s -0.789 0.095 0.217 391s 391s y: [4, 4, 1, 1] 391s -0.880 0.135 0.435 0.672 391s 1.412 0.492 0.750 -0.862 391s -0.046 -0.607 -0.885 -0.041 391s -0.851 0.183 -0.202 0.433 391s testing: mul_mat, [3, 3, 3, 1] = [4, 3, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.945 -0.284 -0.114 0.268 391s 0.265 -0.025 0.511 0.141 391s 0.576 -0.451 0.392 0.452 391s 391s x1: [4, 3, 3, 1] 391s -0.688 0.982 0.468 -0.170 391s 0.547 -0.853 0.170 0.163 391s -0.161 -0.339 0.711 -0.027 391s 391s y: [3, 3, 3, 1] 391s 0.273 0.009 -0.733 391s -0.250 0.275 0.840 391s 0.161 0.325 0.326 391s testing: mul_mat, [1, 4, 3, 1] = [3, 1, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.512 -0.235 0.997 391s 391s x1: [3, 4, 3, 1] 391s -0.874 -0.773 -0.735 391s 0.518 0.494 0.406 391s -0.030 -0.239 0.951 391s 0.037 0.041 -0.450 391s 391s y: [1, 4, 3, 1] 391s -0.104 391s 0.024 391s 1.020 391s -0.477 391s test-mul-mat0: iter:417/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.941 -0.938 -0.536 391s -0.657 -0.563 -0.536 391s 391s x1: [3, 2, 1, 1] 391s 0.999 -0.203 -0.402 391s -0.706 0.116 -0.559 391s 391s y: [2, 2, 1, 1] 391s -0.534 -0.327 391s 0.855 0.698 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.120 0.994 0.430 391s 0.666 0.290 -0.959 391s -0.352 0.643 0.010 391s 0.695 -0.087 0.009 391s 391s x1: [3, 2, 1, 1] 391s -0.147 0.077 0.982 391s -0.142 0.123 -0.530 391s 391s y: [4, 2, 1, 1] 391s 0.481 -1.017 0.111 -0.100 391s -0.123 0.449 0.124 -0.115 391s testing: mul_mat, [3, 4, 1, 2] = [3, 3, 1, 2] * [3, 4, 1, 2] 391s x0: [3, 3, 1, 2] 391s -0.916 -0.797 0.825 391s 0.779 0.116 -0.166 391s -0.729 -0.373 -0.862 391s 391s x1: [3, 4, 1, 2] 391s -0.508 0.511 0.304 391s 0.608 0.952 0.894 391s 0.667 -0.987 -0.641 391s -0.990 -0.550 -0.177 391s 391s y: [3, 4, 1, 2] 391s 0.309 -0.387 -0.082 391s -0.577 0.436 -1.569 391s -0.354 0.511 0.434 391s 1.199 -0.806 1.080 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.575 0.124 0.069 391s 391s x1: [3, 3, 1, 1] 391s -0.582 -0.430 -0.981 391s -0.862 0.220 0.295 391s 0.481 -0.011 -0.436 391s 391s y: [1, 3, 1, 1] 391s 0.214 391s 0.544 391s -0.308 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.556 391s 391s x1: [1, 3, 1, 1] 391s -0.492 391s 0.328 391s -0.107 391s 391s y: [1, 3, 1, 1] 391s 0.273 391s -0.182 391s 0.059 391s testing: mul_mat, [4, 1, 1, 2] = [1, 4, 1, 2] * [1, 1, 1, 2] 391s x0: [1, 4, 1, 2] 391s 0.865 391s -0.563 391s 0.651 391s 0.897 391s 391s x1: [1, 1, 1, 2] 391s 0.727 391s 391s y: [4, 1, 1, 2] 391s 0.629 -0.410 0.473 0.653 391s test-mul-mat0: iter:418/500 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.828 -0.510 -0.495 391s 391s x1: [3, 2, 1, 1] 391s 0.521 -0.908 0.076 391s -0.184 0.574 -0.935 391s 391s y: [1, 2, 1, 1] 391s 0.857 391s 0.018 391s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.719 -0.001 0.281 391s -0.737 0.242 0.569 391s 391s x1: [3, 1, 3, 1] 391s -0.103 -0.002 0.833 391s 391s y: [2, 1, 3, 1] 391s 0.308 0.549 391s testing: mul_mat, [3, 2, 3, 2] = [3, 3, 3, 2] * [3, 2, 3, 2] 391s x0: [3, 3, 3, 2] 391s -0.163 -0.789 0.371 391s -0.708 -0.647 -0.878 391s 0.129 -0.224 0.872 391s 391s x1: [3, 2, 3, 2] 391s 0.249 -0.986 0.294 391s -0.854 0.012 0.126 391s 391s y: [3, 2, 3, 2] 391s 0.846 0.204 0.509 391s 0.176 0.486 -0.003 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.660 0.315 -0.157 391s 0.147 -0.159 0.783 391s 0.521 -0.786 -0.632 391s 391s x1: [3, 3, 1, 1] 391s 0.252 0.771 -0.126 391s -0.310 0.965 -0.690 391s -0.958 -0.873 -0.417 391s 391s y: [3, 3, 1, 1] 391s 0.429 -0.184 -0.395 391s 0.208 -0.739 -0.484 391s -0.841 -0.330 0.451 391s testing: mul_mat, [4, 3, 3, 1] = [3, 4, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.944 -0.146 -0.294 391s 0.194 0.950 -0.735 391s 0.481 0.190 0.930 391s 0.179 -0.916 -0.978 391s 391s x1: [3, 3, 3, 1] 391s 0.235 0.660 -0.531 391s -0.989 0.230 -0.244 391s 0.075 0.845 0.316 391s 391s y: [4, 3, 3, 1] 391s 0.282 1.063 -0.255 -0.043 391s -0.896 0.206 -0.658 -0.150 391s -0.146 0.585 0.491 -1.070 391s testing: mul_mat, [3, 3, 3, 2] = [4, 3, 3, 2] * [4, 3, 3, 2] 391s x0: [4, 3, 3, 2] 391s 0.304 -0.618 -0.021 0.239 391s 0.683 0.018 0.903 0.376 391s -0.517 0.367 -0.378 0.124 391s 391s x1: [4, 3, 3, 2] 391s 0.419 0.557 -0.418 -0.875 391s 0.823 -0.488 -0.853 0.877 391s -0.295 0.825 -0.028 -0.065 391s 391s y: [3, 3, 3, 2] 391s -0.417 -0.411 0.037 391s 0.779 0.114 -0.173 391s -0.615 -0.237 0.458 391s test-mul-mat0: iter:419/500 391s testing: mul_mat, [1, 1, 1, 1] = [1, 1, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.214 391s 391s x1: [1, 1, 1, 1] 391s 0.563 391s 391s y: [1, 1, 1, 1] 391s 0.121 391s testing: mul_mat, [1, 1, 4, 1] = [1, 1, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 1, 4, 1] 391s -0.736 391s 391s x1: [1, 1, 4, 1] 391s 0.338 391s 391s y: [1, 1, 4, 1] 391s -0.249 391s testing: mul_mat, [2, 1, 4, 1] = [1, 2, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.745 391s 0.568 391s 391s x1: [1, 1, 4, 1] 391s 0.513 391s 391s y: [2, 1, 4, 1] 391s 0.383 0.291 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.839 391s -0.727 391s 391s x1: [1, 2, 1, 1] 391s 0.070 391s -0.303 391s 391s y: [2, 2, 1, 1] 391s -0.058 -0.051 391s 0.254 0.220 391s testing: mul_mat, [2, 1, 4, 1] = [2, 2, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 2, 4, 1] 391s 0.702 -0.281 391s 0.345 0.295 391s 391s x1: [2, 1, 4, 1] 391s 0.282 0.724 391s 391s y: [2, 1, 4, 1] 391s -0.006 0.311 391s testing: mul_mat, [1, 2, 4, 1] = [2, 1, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.205 -0.296 391s 391s x1: [2, 2, 4, 1] 391s -0.846 -0.751 391s 0.137 -0.509 391s 391s y: [1, 2, 4, 1] 391s 0.396 391s 0.123 391s test-mul-mat0: iter:420/500 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.722 -0.199 391s -0.098 -0.092 391s 0.070 -0.901 391s -0.536 -0.223 391s 391s x1: [2, 3, 1, 1] 391s -0.892 0.731 391s -0.462 -0.643 391s -0.132 0.029 391s 391s y: [4, 3, 1, 1] 391s 0.500 0.020 -0.721 0.316 391s 0.461 0.104 0.547 0.391 391s 0.090 0.010 -0.035 0.065 391s testing: mul_mat, [2, 4, 2, 1] = [2, 2, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.905 -0.571 391s -0.602 -0.227 391s 391s x1: [2, 4, 2, 1] 391s 0.338 -0.461 391s -0.814 0.921 391s 0.334 -0.111 391s 0.362 0.670 391s 391s y: [2, 4, 2, 1] 391s 0.569 -0.098 391s -1.263 0.281 391s 0.366 -0.176 391s -0.055 -0.370 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.425 -0.705 391s 0.182 -0.213 391s 391s x1: [2, 2, 2, 1] 391s -0.932 0.959 391s 0.330 -0.833 391s 391s y: [2, 2, 2, 1] 391s -1.072 -0.373 391s 0.727 0.237 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.054 0.345 391s 0.352 -0.875 391s 0.315 -0.227 391s 391s x1: [2, 2, 1, 1] 391s 0.272 -0.005 391s -0.095 0.447 391s 391s y: [3, 2, 1, 1] 391s -0.016 0.100 0.087 391s 0.159 -0.424 -0.132 391s testing: mul_mat, [2, 2, 2, 1] = [3, 2, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 2, 2, 1] 391s -0.187 -0.537 -0.669 391s 0.942 -0.222 0.951 391s 391s x1: [3, 2, 2, 1] 391s -0.489 0.176 0.032 391s -0ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s .421 0.134 -0.638 391s 391s y: [2, 2, 2, 1] 391s -0.024 -0.470 391s 0.434 -1.034 391s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.606 0.781 391s 391s x1: [2, 3, 2, 1] 391s 0.600 -0.598 391s 0.147 -0.055 391s -0.473 0.919 391s 391s y: [1, 3, 2, 1] 391s -0.832 391s -0.132 391s 1.005 391s test-mul-mat0: iter:421/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.748 -0.582 -0.931 391s -0.094 0.467 0.901 391s -0.493 0.869 0.048 391s 391s x1: [3, 1, 1, 1] 391s 0.171 -0.497 -0.185 391s 391s y: [3, 1, 1, 1] 391s 0.334 -0.415 -0.525 391s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.434 -0.963 -0.108 391s 391s x1: [3, 3, 4, 1] 391s 0.452 -0.604 -0.033 391s -0.093 -0.902 0.919 391s -0.058 -0.065 0.232 391s 391s y: [1, 3, 4, 1] 391s 0.781 391s 0.729 391s 0.012 391s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.898 0.977 0.834 391s -0.406 0.079 0.495 391s -0.971 0.116 -0.613 391s 0.186 -0.517 -0.708 391s 391s x1: [3, 1, 4, 1] 391s -0.992 -0.961 -0.646 391s 391s y: [4, 1, 4, 1] 391s -0.588 0.007 1.248 0.770 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.800 0.805 -0.141 391s -0.199 -0.121 0.430 391s 391s x1: [3, 4, 1, 1] 391s -0.743 0.322 -0.327 391s 0.863 0.567 0.132 391s 0.094 0.402 0.976 391s -0.310 -0.445 0.337 391s 391s y: [2, 4, 1, 1] 391s -0.289 -0.031 391s 1.128 -0.184 391s 0.261 0.352 391s -0.654 0.260 391s testing: mul_mat, [2, 3, 4, 1] = [2, 2, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 2, 4, 1] 391s -0.922 0.639 391s 0.600 -0.043 391s 391s x1: [2, 3, 4, 1] 391s 0.453 -0.054 391s -0.829 -0.586 391s -0.781 0.394 391s 391s y: [2, 3, 4, 1] 391s -0.452 0.274 391s 0.390 -0.472 391s 0.972 -0.486 391s testing: mul_mat, [1, 2, 4, 1] = [2, 1, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 1, 4, 1] 391s -0.177 0.753 391s 391s x1: [2, 2, 4, 1] 391s 0.644 0.022 391s -0.906 -0.603 391s 391s y: [1, 2, 4, 1] 391s -0.098 391s -0.294 391s test-mul-mat0: iter:422/500 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.624 0.185 391s -0.634 -0.603 391s 0.625 -0.310 391s 0.746 -0.656 391s 391s x1: [2, 3, 1, 1] 391s 0.106 -0.358 391s -0.931 -0.872 391s -0.263 -0.533 391s 391s y: [4, 3, 1, 1] 391s 0.000 0.148 0.178 0.314 391s -0.741 1.116 -0.311 -0.122 391s -0.263 0.489 0.001 0.153 391s testing: mul_mat, [1, 4, 4, 1] = [2, 1, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 1, 4, 1] 391s 0.772 -0.317 391s 391s x1: [2, 4, 4, 1] 391s 0.812 -0.070 391s 0.140 -0.465 391s 0.753 -0.106 391s 0.718 0.645 391s 391s y: [1, 4, 4, 1] 391s 0.650 391s 0.256 391s 0.615 391s 0.350 391s testing: mul_mat, [1, 1, 4, 4] = [2, 1, 4, 4] * [2, 1, 4, 4] 391s x0: [2, 1, 4, 4] 391s -0.841 -0.502 391s 391s x1: [2, 1, 4, 4] 391s 0.772 0.766 391s 391s y: [1, 1, 4, 4] 391s -1.034 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.436 0.010 391s -0.082 0.636 391s 391s x1: [2, 1, 1, 1] 391s -0.025 -0.986 391s 391s y: [2, 1, 1, 1] 391s -0.021 -0.625 391s testing: mul_mat, [3, 2, 4, 1] = [2, 3, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.596 -0.378 391s 0.287 0.764 391s -0.247 0.315 391s 391s x1: [2, 2, 4, 1] 391s -0.824 0.165 391s -0.146 -0.689 391s 391s y: [3, 2, 4, 1] 391s -0.554 -0.110 0.256 391s 0.173 -0.569 -0.181 391s testing: mul_mat, [4, 2, 4, 4] = [3, 4, 4, 4] * [3, 2, 4, 4] 391s x0: [3, 4, 4, 4] 391s -0.432 0.111 0.372 391s 0.853 0.147 -0.832 391s 0.923 -0.419 -0.080 391s -0.219 0.626 0.580 391s 391s x1: [3, 2, 4, 4] 391s -0.124 -0.334 -0.263 391s 0.231 -0.828 0.731 391s 391s y: [4, 2, 4, 4] 391s -0.081 0.064 0.047 -0.334 391s 0.080 -0.533 0.502 -0.145 391s test-mul-mat0: iter:423/500 391s testing: mul_mat, [3, 1, 1, 1] = [1, 3, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.430 391s 0.796 391s -0.779 391s 391s x1: [1, 1, 1, 1] 391s -0.130 391s 391s y: [3, 1, 1, 1] 391s 0.056 -0.103 0.101 391s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.477 391s 0.997 391s 0.600 391s 0.440 391s 391s x1: [1, 3, 2, 1] 391s -0.897 391s -0.347 391s 0.662 391s 391s y: [4, 3, 2, 1] 391s 0.428 -0.895 -0.538 -0.395 391s 0.165 -0.346 -0.208 -0.153 391s -0.316 0.661 0.398 0.292 391s testing: mul_mat, [4, 4, 2, 2] = [1, 4, 2, 2] * [1, 4, 2, 2] 391s x0: [1, 4, 2, 2] 391s 0.933 391s -0.598 391s -0.470 391s 0.569 391s 391s x1: [1, 4, 2, 2] 391s -0.264 391s 0.470 391s -0.087 391s -0.564 391s 391s y: [4, 4, 2, 2] 391s -0.246 0.158 0.124 -0.150 391s 0.439 -0.281 -0.221 0.268 391s -0.081 0.052 0.041 -0.049 391s -0.526 0.337 0.265 -0.321 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.997 391s 0.798 391s 0.323 391s 391s x1: [1, 4, 1, 1] 391s 0.048 391s -0.869 391s -0.351 391s 0.185 391s 391s y: [3, 4, 1, 1] 391s -0.048 0.038 0.015 391s 0.866 -0.694 -0.281 391s 0.350 -0.280 -0.113 391s -ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s 0.185 0.148 0.060 391s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 1, 2, 1] 391s -0.507 -0.018 0.549 391s 391s x1: [3, 1, 2, 1] 391s -0.406 0.281 0.841 391s 391s y: [1, 1, 2, 1] 391s 0.662 391s testing: mul_mat, [3, 3, 2, 2] = [1, 3, 2, 2] * [1, 3, 2, 2] 391s x0: [1, 3, 2, 2] 391s -0.153 391s -0.395 391s 0.187 391s 391s x1: [1, 3, 2, 2] 391s -0.845 391s 0.946 391s -0.547 391s 391s y: [3, 3, 2, 2] 391s 0.129 0.333 -0.158 391s -0.144 -0.373 0.177 391s 0.083 0.216 -0.102 391s test-mul-mat0: iter:424/500 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.514 -0.019 0.348 391s -0.880 -0.832 0.993 391s -0.952 0.950 0.920 391s 0.818 0.487 -0.592 391s 391s x1: [3, 4, 1, 1] 391s 0.797 0.306 0.198 391s 0.744 0.760 -0.114 391s -0.776 0.171 0.262 391s -0.162 -0.723 0.797 391s 391s y: [4, 4, 1, 1] 391s 0.473 -0.760 -0.286 0.685 391s 0.328 -1.400 -0.091 1.046 391s -0.311 0.801 1.142 -0.707 391s 0.208 1.535 0.201 -0.956 391s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.856 -0.756 -0.856 391s -0.904 -0.459 -0.133 391s 391s x1: [3, 4, 4, 1] 391s -0.831 -0.957 0.389 391s 0.501 -0.852 -0.568 391s 0.298 0.454 0.629 391s 0.042 0.214 -0.485 391s 391s y: [2, 4, 4, 1] 391s 1.102 1.140 391s 0.702 0.014 391s -1.137 -0.561 391s 0.218 -0.072 391s testing: mul_mat, [3, 2, 4, 3] = [3, 3, 4, 3] * [3, 2, 4, 3] 391s x0: [3, 3, 4, 3] 391s -0.992 0.482 -0.317 391s -0.528 -0.681 0.675 391s -0.791 -0.087 -0.472 391s 391s x1: [3, 2, 4, 3] 391s 0.899 0.952 -0.513 391s 0.914 0.492 -0.446 391s 391s y: [3, 2, 4, 3] 391s -0.270 -1.469 -0.551 391s -0.528 -1.119 -0.556 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.875 0.766 0.622 391s 0.577 0.595 -0.848 391s 0.424 0.345 -0.698 391s 391s x1: [3, 3, 1, 1] 391s -0.741 -0.481 -0.813 391s -0.018 0.487 -0.040 391s -0.828 0.823 -0.822 391s 391s y: [3, 3, 1, 1] 391s -1.523 -0.025 0.086 391s 0.332 0.313 0.188 391s -0.606 0.709 0.507 391s testing: mul_mat, [1, 3, 4, 1] = [3, 1, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.111 -0.953 0.137 391s 391s x1: [3, 3, 4, 1] 391s -0.954 0.487 -0.730 391s 0.845 -0.249 0.752 391s -0.572 -0.733 -0.665 391s 391s y: [1, 3, 4, 1] 391s -0.670 391s 0.434 391s 0.544 391s testing: mul_mat, [1, 3, 4, 3] = [1, 1, 4, 3] * [1, 3, 4, 3] 391s x0: [1, 1, 4, 3] 391s 0.935 391s 391s x1: [1, 3, 4, 3] 391s -0.750 391s 0.204 391s 0.363 391s 391s y: [1, 3, 4, 3] 391s -0.701 391s 0.191 391s 0.339 391s test-mul-mat0: iter:425/500 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.371 391s 391s x1: [1, 4, 1, 1] 391s 0.583 391s -0.578 391s -0.844 391s 0.440 391s 391s y: [1, 4, 1, 1] 391s -0.216 391s 0.214 391s 0.313 391s -0.163 391s testing: mul_mat, [3, 1, 4, 1] = [1, 3, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 3, 4, 1] 391s 0.233 391s 0.808 391s 0.099 391s 391s x1: [1, 1, 4, 1] 391s -0.923 391s 391s y: [3, 1, 4, 1] 391s -0.215 -0.746 -0.092 391s testing: mul_mat, [4, 3, 4, 3] = [1, 4, 4, 3] * [1, 3, 4, 3] 391s x0: [1, 4, 4, 3] 391s -0.411 391s 0.328 391s -0.506 391s -0.750 391s 391s x1: [1, 3, 4, 3] 391s -0.277 391s -0.654 391s 0.435 391s 391s y: [4, 3, 4, 3] 391s 0.114 -0.091 0.140 0.208 391s 0.269 -0.215 0.331 0.490 391s -0.179 0.143 -0.220 -0.326 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.090 391s -0.425 391s 0.277 391s 391s x1: [1, 4, 1, 1] 391s -0.786 391s -0.844 391s 0.547 391s 0.095 391s 391s y: [3, 4, 1, 1] 391s -0.071 0.334 -0.217 391s -0.076 0.359 -0.234 391s 0.049 -0.233 0.151 391s 0.009 -0.041 0.026 391s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 3, 4, 1] 391s 0.228 -0.705 0.277 391s 0.452 0.164 0.999 391s -0.625 -0.649 -0.201 391s 391s x1: [3, 1, 4, 1] 391s -0.527 -0.001 0.556 391s 391s y: [3, 1, 4, 1] 391s 0.034 0.318 0.218 391s testing: mul_mat, [2, 3, 4, 3] = [3, 2, 4, 3] * [3, 3, 4, 3] 391s x0: [3, 2, 4, 3] 391s 0.674 0.300 0.906 391s 0.802 0.217 -0.188 391s 391s x1: [3, 3, 4, 3] 391s -0.275 0.730 -0.379 391s 0.524 -0.871 -0.166 391s -0.522 -0.924 -0.857 391s 391s y: [2, 3, 4, 3] 391s -0.310 0.009 391s -0.058 0.263 391s -1.405 -0.458 391s test-mul-mat0: iter:426/500 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.879 391s 0.959 391s 0.122 391s 391s x1: [1, 4, 1, 1] 391s 0.887 391s 0.279 391s -0.659 391s -0.240 391s 391s y: [3, 4, 1, 1] 391s -0.779 0.851 0.108 391s -0.246 0.268 0.034 391s 0.579 -0.632 -0.081 391s 0.211 -0.230 -0.029 391s testing: mul_mat, [3, 3, 2, 1] = [1, 3, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 3, 2, 1] 391s 0.795 391s 0.374 391s 0.789 391s 391s x1: [1, 3, 2, 1] 391s -0.509 391s -0.599 391s -0.718 391s 391s y: [3, 3, 2, 1] 391s -0.405 -0.190 -0.402 391s -0.477 -0.224 -0.473 391s -0.571 -0.269 -0.567 391s testing: mul_mat, [2, 3, 2, 3] = [1, 2, 2, 3] * [1, 3, 2, 3] 391s x0: [1, 2, 2, 3] 391s -0.002 391s -0.519 391s 391s x1: [1, 3, 2, 3] 391s 0.661 391s 0.256 391s 0.080 391s 391s y: [2, 3, 2, 3] 391s -0.001 -0.343 391s -0.000 -0.133 391s -0.000 -0.042 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.191 391s -0.759 391s -0.826 391s 391s x1: [1, 2, 1, 1] 391s 0.637 391s 0.678 391s 391s y: [3, 2, 1, 1] 391s -0.122 -0.483 -0.526 391s -0.130 -0.515 -0.560 391s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 4, 2, 1] 391s -0.252 0.348 0.666 391s 0.149 -0.904 -0.455 391s -0.653 -0.656 0.879 391s 0.345 0.829 -0.736 391s 391s x1: [3, 1, 2, 1] 391s 0.154 -0.768 -0.517 391s 391s y: [4, 1, 2, 1] 391s -0.650 0.952 -0.050 -0.203 391s testing: mul_mat, [1, 3, 2, 3] = [4, 1, 2, 3] * [4, 3, 2, 3] 391s x0: [4, 1, 2, 3] 391s 0.123 0.773 -0.581 0.422 391s 391s x1: [4, 3, 2, 3] 391s -0.882 0.876 0.806 -0.932 391s 0.986 0.878 0.156 -0.266 391s 0.027 0.502 -0.920 -0.625 391s 391s y: [1, 3, 2, 3] 391s -0.293 391s 0.597 391s 0.663 391s test-mul-mat0: iter:427/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.423 0.023 391s 391s x1: [2, 4, 1, 1] 391s -0.772 0.609 391s -0.976 0.351 391s 0.382 -0.557 391s -0.227 -0.990 391s 391s y: [1, 4, 1, 1] 391s 0.340 391s 0.421 391s -0.174 391s 0.073 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.591 -0.409 391s 0.280 0.672 391s -0.581 -0.247 391s 391s x1: [2, 1, 1, 1] 391s 0.456 -0.853 391s 391s y: [3, 1, 1, 1] 391s 0.619 -0.446 -0.054 391s testing: mul_mat, [2, 3, 1, 2] = [2, 2, 1, 2] * [2, 3, 1, 2] 391s x0: [2, 2, 1, 2] 391s -0.871 -0.306 391s -0.954 -0.428 391s 391s x1: [2, 3, 1, 2] 391s -0.151 0.787 391s -0.974 -0.611 391s -0.276 0.916 391s 391s y: [2, 3, 1, 2] 391s -0.109 -0.193 391s 1.035 1.191 391s -0.040 -0.129 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.218 -0.626 391s -0.042 0.201 391s 391s x1: [2, 2, 1, 1] 391s 0.079 0.112 391s -0.809 0.678 391s 391s y: [2, 2, 1, 1] 391s -0.087 0.019 391s -0.248 0.170 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.712 -0.232 391s 0.653 -0.704 391s 0.241 0.064 391s -0.041 -0.889 391s 391s x1: [2, 2, 1, 1] 391s 0.711 0.224 391s -0.012 0.737 391s 391s y: [4, 2, 1, 1] 391s 0.454 0.307 0.185 -0.228 391s -0.179 -0.527 0.044 -0.655 391s testing: mul_mat, [2, 2, 1, 2] = [4, 2, 1, 2] * [4, 2, 1, 2] 391s x0: [4, 2, 1, 2] 391s -0.509 -0.169 -0.776 0.191 391s -0.513 0.479 -0.556 0.877 391s 391s x1: [4, 2, 1, 2] 391s -0.040 0.192 -0.195 0.005 391s 0.764 -0.727 0.061 0.420 391s 391s y: [2, 2, 1, 2] 391s 0.140 0.226 391s -0.234 -0.406 391s test-mul-mat0: iter:428/500 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.342 391s -0.036 391s 0.355 391s 0.948 391s 391s x1: [1, 4, 1, 1] 391s 0.007 391s 0.590 391s -0.850 391s 0.146 391s 391s y: [4, 4, 1, 1] 391s -0.002 -0.000 0.002 0.006 391s -0.202 -0.021 0.209 0.560 391s 0.290 0.030 -0.301 -0.806 391s -0.050 -0.005 0.052 0.139 391s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 2, 2, 1] 391s -0.671 391s 0.923 391s 391s x1: [1, 4, 2, 1] 391s 0.741 391s -0.038 391s -0.561 391s -0.772 391s 391s y: [2, 4, 2, 1] 391s -0.497 0.684 391s 0.025 -0.035 391s 0.376 -0.518 391s 0.518 -0.713 391s testing: mul_mat, [2, 2, 2, 2] = [1, 2, 2, 2] * [1, 2, 2, 2] 391s x0: [1, 2, 2, 2] 391s 0.796 391s -0.359 391s 391s x1: [1, 2, 2, 2] 391s 0.904 391s -0.526 391s 391s y: [2, 2, 2, 2] 391s 0.720 -0.325 391s -0.419 0.189 391s testing: mul_mat, [3, 2, 1, 1] = [1, 3, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.681 391s 0.876 391s -0.127 391s 391s x1: [1, 2, 1, 1] 391s 0.880 391s 0.648 391s 391s y: [3, 2, 1, 1] 391s -0.599 0.771 -0.112 391s -0.441 0.568 -0.082 391s testing: mul_mat, [1, 1, 2, 1] = [3, 1, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 1, 2, 1] 391s 0.631 -0.062 0.080 391s 391s x1: [3, 1, 2, 1] 391s -0.763 -0.672 0.111 391s 391s y: [1, 1, 2, 1] 391s -0.431 391s testing: mul_mat, [1, 3, 2, 2] = [1, 1, 2, 2] * [1, 3, 2, 2] 391s x0: [1, 1, 2, 2] 391s -0.134 391s 391s x1: [1, 3, 2, 2] 391s -0.349 391s -0.428 391s 0.010 391s 391s y: [1, 3, 2, 2] 391s 0.047 391s 0.058 391s -0.001 391s test-mul-mat0: iter:429/500 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.600 0.485 0.171 -0.182 391s -0.984 -0.968 0.683 0.626 391s 391s x1: [4, 4, 1, 1] 391s -0.236 0.727 -0.371 0.370 391s 0.358 0.567 -0.550 -0.282 391s 0.349 -0.590 0.369 0.921 391s 0.420 0.143 -0.711 0.072 391s 391s y: [2, 4, 1, 1] 391s 0.080 -0.493 391s 0.447 -1.453 391s -0.182 1.057 391s 0.187 -0.992 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.237 -0.421 0.900 0.213 391s 391s x1: [4, 2, 1, 1] 391s 0.212 -0.575 -0.890 -0.371 391s 0.495 0.704 0.393 0.221 391s 391s y: [1, 2, 1, 1] 391s -0.689 391s -0.012 391s testing: mul_mat, [3, 1, 1, 2] = [4ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-forward.dot -o test-grad0-forward.dot.png && open test-grad0-forward.dot.png 391s ggml_graph_dump_dot: dot -Tpng test-grad0-backward.dot -o test-grad0-backward.dot.png && open test-grad0-backward.dot.png 391s , 3, 1, 2] * [4, 1, 1, 2] 391s x0: [4, 3, 1, 2] 391s 1.000 0.060 0.601 0.171 391s 0.877 0.617 0.203 0.561 391s 0.243 -0.585 0.985 0.353 391s 391s x1: [4, 1, 1, 2] 391s 0.297 0.250 0.623 -0.334 391s 391s y: [3, 1, 1, 2] 391s 0.629 0.354 0.422 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.426 0.669 0.657 0.841 391s 391s x1: [4, 3, 1, 1] 391s -0.963 0.244 0.712 0.080 391s 0.053 0.172 0.196 0.052 391s -0.769 -0.204 -0.777 -0.891 391s 391s y: [1, 3, 1, 1] 391s 0.288 391s 0.310 391s -1.723 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.322 391s 0.836 391s 391s x1: [1, 4, 1, 1] 391s 0.655 391s 0.010 391s -0.115 391s 0.135 391s 391s y: [2, 4, 1, 1] 391s 0.211 0.547 391s 0.003 0.008 391s -0.037 -0.097 391s 0.043 0.113 391s testing: mul_mat, [4, 1, 1, 2] = [2, 4, 1, 2] * [2, 1, 1, 2] 391s x0: [2, 4, 1, 2] 391s 0.695 0.002 391s 0.291 -0.947 391s -0.028 0.991 391s -0.061 -0.826 391s 391s x1: [2, 1, 1, 2] 391s 0.459 -0.478 391s 391s y: [4, 1, 1, 2] 391s 0.317 0.586 -0.487 0.367 391s test-mul-mat0: iter:430/500 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.261 0.926 0.401 391s 391s x1: [3, 2, 1, 1] 391s 0.125 0.418 -0.027 391s 0.193 -0.261 -0.191 391s 391s y: [1, 2, 1, 1] 391s 0.409 391s -0.268 391s testing: mul_mat, [3, 1, 4, 1] = [3, 3, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.140 -0.211 -0.028 391s 0.553 -0.028 0.087 391s -0.914 -0.189 -0.788 391s 391s x1: [3, 1, 4, 1] 391s -0.804 0.621 -0.308 391s 391s y: [3, 1, 4, 1] 391s -0.010 -0.488 0.859 391s testing: mul_mat, [4, 3, 4, 3] = [3, 4, 4, 3] * [3, 3, 4, 3] 391s x0: [3, 4, 4, 3] 391s -0.932 -0.447 -0.680 391s 0.294 -0.641 0.063 391s -0.742 0.763 0.624 391s 0.844 -0.143 -0.264 391s 391s x1: [3, 3, 4, 3] 391s 0.132 0.343 -0.441 391s 0.344 -0.154 -0.657 391s -0.251 0.089 -0.064 391s 391s y: [4, 3, 4, 3] 391s 0.023 -0.209 -0.111 0.179 391s 0.195 0.158 -0.782 0.486 391s 0.238 -0.135 0.214 -0.208 391s testing: mul_mat, [4, 4, 1, 1] = [3, 4, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.995 -0.480 -0.574 391s -0.758 -0.355 0.825 391s -0.233 0.533 -0.121 391s 0.988 0.001 0.229 391s 391s x1: [3, 4, 1, 1] 391s 0.386 -0.298 0.630 391s -0.337 0.816 0.308 391s 0.979 0.097 -0.675 391s -0.937 0.569 0.329 391s 391s y: [4, 4, 1, 1] 391s -0.602 0.333 -0.325 0.525 391s -0.232 0.220 0.476 -0.263 391s -0.633 -1.333 -0.095 0.813 391s 0.470 0.780 0.482 -0.850 391s testing: mul_mat, [3, 3, 4, 1] = [4, 3, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 3, 4, 1] 391s 0.891 -0.055 -0.356 -0.230 391s -0.082 0.699 -0.988 -0.850 391s 0.856 -0.379 -0.614 0.030 391s 391s x1: [4, 3, 4, 1] 391s -0.897 0.448 -0.218 -0.238 391s 0.603 0.410 -0.852 -0.695 391s 0.041 -0.189 -0.879 -0.652 391s 391s y: [3, 3, 4, 1] 391s -0.692 0.804 -0.811 391s 0.977 1.670 0.863 391s 0.509 1.287 0.627 391s testing: mul_mat, [2, 4, 4, 3] = [3, 2, 4, 3] * [3, 4, 4, 3] 391s x0: [3, 2, 4, 3] 391s 0.420 0.304 0.151 391s -0.322 -0.888 -0.352 391s 391s x1: [3, 4, 4, 3] 391s -0.615 0.311 0.515 391s 0.201 0.708 -0.976 391s -0.862 0.040 0.285 391s -0.586 0.982 -0.850 391s 391s y: [2, 4, 4, 3] 391s -0.086 -0.260 391s 0.152 -0.351 391s -0.307 0.141 391s -0.076 -0.385 391s test-mul-mat0: iter:431/500 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.042 -0.146 -0.560 391s 391s x1: [3, 1, 1, 1] 391s 0.879 0.136 0.131 391s 391s y: [1, 1, 1, 1] 391s -0.056 391s testing: mul_mat, [2, 1, 4, 1] = [3, 2, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.677 0.768 -0.248 391s 0.316 -0.079 0.154 391s 391s x1: [3, 1, 4, 1] 391s 0.028 0.244 0.067 391s 391s y: [2, 1, 4, 1] 391s 0.151 0.000 391s testing: mul_mat, [3, 2, 4, 2] = [3, 3, 4, 2] * [3, 2, 4, 2] 391s x0: [3, 3, 4, 2] 391s -0.659 -0.420 0.859 391s 0.459 0.407 0.865 391s 0.487 0.445 0.958 391s 391s x1: [3, 2, 4, 2] 391s -0.076 0.670 0.469 391s -0.793 0.392 0.302 391s 391s y: [3, 2, 4, 2] 391s 0.172 0.643 0.710 391s 0.617 0.057 0.077 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.089 0.099 -0.354 391s -0.986 -0.386 0.057 391s 0.146 0.084 -0.070 391s 391s x1: [3, 3, 1, 1] 391s 0.992 0.175 0.257 391s 0.788 -0.628 -0.447 391s -0.663 0.649 0.732 391s 391s y: [3, 3, 1, 1] 391s -0.162 -1.031 0.142 391s 0.026 -0.560 0.094 391s -0.136 0.445 -0.094 391s testing: mul_mat, [3, 3, 4, 1] = [3, 3, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 3, 4, 1] 391s -0.301 0.650 -0.406 391s 0.270 -0.801 -0.697 391s 0.338 -0.141 0.637 391s 391s x1: [3, 3, 4, 1] 391s 0.490 -0.961 0.026 391s -0.789 0.420 -0.170 391s -0.783 0.064 0.831 391s 391s y: [3, 3, 4, 1] 391s -0.783 0.884 0.318 391s 0.579 -0.431 -0.434 391s -0.060 -0.841 0.255 391s testing: mul_mat, [3, 3, 4, 2] = [3, 3, 4, 2] * [3, 3, 4, 2] 391s x0: [3, 3, 4, 2] 391s 0.795 0.175 0.411 391s 0.980 0.723 -0.168 391s 0.768 -0.938 -0.820 391s 391s x1: [3, 3, 4, 2] 391s 0.280 -0.452 -0.690 391s -0.083 0.874 0.089 391s 0.885 -0.731 0.992 391s 391s y: [3, 3, 4, 2] 391s -0.140 0.063 1.205 391s 0.123 0.535 -0.956 391s 0.983 0.173 0.551 391s test-mul-mat0: iter:432/500 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.621 391s -0.093 391s 0.225 391s 391s x1: [1, 4, 1, 1] 391s 0.574 391s -0.396 391s 0.721 391s 0.407 391s 391s y: [3, 4, 1, 1] 391s -0.356 -0.053 0.129 391s 0.246 0.037 -0.089 391s -0.447 -0.067 0.162 391s -0.252 -0.038 0.091 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.396 391s 0.321 391s 391s x1: [1, 3, 1, 1] 391s 0.494 391s -0.747 391s 0.549 391s 391s y: [2, 3, 1, 1] 391s -0.196 0.159 391s 0.296 -0.240 391s -0.218 0.176 391s testing: mul_mat, [4, 2, 1, 2] = [1, 4, 1, 2] * [1, 2, 1, 2] 391s x0: [1, 4, 1, 2] 391s -0.820 391s 0.311 391s 0.706 391s 0.096 391s 391s x1: [1, 2, 1, 2] 391s -0.572 391s -0.981 391s 391s y: [4, 2, 1, 2] 391s 0.469 -0.178 -0.404 -0.055 391s 0.805 -0.305 -0.692 -0.094 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.862 391s -0.678 391s 0.180 391s 0.774 391s 391s x1: [1, 4, 1, 1] 391s -0.399 391s -0.800 391s 0.534 391s -0.398 391s 391s y: [4, 4, 1, 1] 391s 0.344 0.271 -0.072 -0.309 391s 0.689 0.542 -0.144 -0.620 391s -0.460 -0.362 0.096 0.413 391s 0.343 0.270 -0.072 -0.309 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.452 -0.338 0.442 -0.913 391s -0.766 -0.037 -0.009 -0.177 391s -0.056 -0.188 -0.566 0.680 391s 0.870 0.123 0.147 0.958 391s 391s x1: [4, 1, 1, 1] 391s -0.298 -0.912 -0.001 -0.804 391s 391s y: [4, 1, 1, 1] 391s 1.177 0.405 -0.357 -1.142 391s testing: mul_mat, [3, 4, 1, 2] = [4, 3, 1, 2] * [4, 4, 1, 2] 391s x0: [4, 3, 1, 2] 391s 0.015 -0.053 -0.941 0.259 391s 0.200 0.016 -0.393 0.197 391s -0.122 -0.918 0.988 -0.910 391s 391s x1: [4, 4, 1, 2] 391s -0.856 -0.720 -0.842 0.678 391s -0.118 0.932 0.816 0.204 391s 0.112 0.591 0.906 0.200 391s -0.410 -0.898 0.540 0.138 391s 391s y: [3, 4, 1, 2] 391s 0.993 0.282 -0.684 391s -0.766 -0.290 -0.220 391s -0.830 -0.285 0.158 391s -0.431 -0.281 1.282 391s test-mul-mat0: iter:433/500 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.015 -0.594 0.806 391s 0.350 -0.168 0.089 391s 391s x1: [3, 4, 1, 1] 391s 0.726 -0.004 0.367 391s -0.259 -0.804 -0.755 391s 0.688 0.211 -0.673 391s 0.747 0.818 -0.685 391s 391s y: [2, 4, 1, 1] 391s 0.309 0.288 391s -0.134 -0.023 391s -0.658 0.145 391s -1.027 0.063 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.896 0.094 -0.304 391s 0.371 -0.536 -0.577 391s 391s x1: [3, 2, 1, 1] 391s -0.308 -0.745 0.820 391s 0.854 0.799 0.439 391s 391s y: [2, 2, 1, 1] 391s -0.595 -0.188 391s 0.707 -0.364 391s testing: mul_mat, [2, 2, 1, 2] = [3, 2, 1, 2] * [3, 2, 1, 2] 391s x0: [3, 2, 1, 2] 391s 0.125 0.412 0.958 391s 0.956 -0.499 -0.350 391s 391s x1: [3, 2, 1, 2] 391s -0.633 0.831 0.164 391s -0.437 -0.924 -0.148 391s 391s y: [2, 2, 1, 2] 391s 0.421 -1.077 391s -0.577 0.095 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.149 -0.387 -0.792 391s 391s x1: [3, 2, 1, 1] 391s 0.906 -0.147 0.026 391s 0.278 0.318 0.449 391s 391s y: [1, 2, 1, 1] 391s 0.171 391s -0.437 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.935 391s 0.575 391s 391s x1: [1, 3, 1, 1] 391s 0.225 391s 0.465 391s 0.982 391s 391s y: [2, 3, 1, 1] 391s -0.211 0.129 391s -0.435 0.267 391s -0.919 0.564 391s testing: mul_mat, [3, 1, 1, 2] = [2, 3, 1, 2] * [2, 1, 1, 2] 391s x0: [2, 3, 1, 2] 391s -0.860 -0.211 391s 0.428 0.639 391s 0.584 0.905 391s 391s x1: [2, 1, 1, 2] 391s -0.653 0.671 391s 391s y: [3, 1, 1, 2] 391s 0.420 0.149 0.226 391s test-mul-mat0: iter:434/500 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.758 -0.815 0.912 -0.576 391s 391s x1: [4, 2, 1, 1] 391s -0.522 -0.016 -0.289 -0.315 391s -0.791 -0.824 -0.333 0.838 391s 391s y: [1, 2, 1, 1] 391s 0.327 391s 0.485 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.353 -0.276 -0.518 -0.354 391s -0.627 -0.854 -0.731 0.928 391s 391s x1: [4, 1, 1, 1] 391s 0.657 -0.006 -0.436 0.085 391s 391s y: [2, 1, 1, 1] 391s 0.429 -0.010 391s testing: mul_mat, [2, 2, 1, 4] = [4, 2, 1, 4] * [4, 2, 1, 4] 391s x0: [4, 2, 1, 4] 391s -0.760 0.421 -0.974 0.479 391s -0.097 0.754 0.403 -0.405 391s 391s x1: [4, 2, 1, 4] 391s -0.336 -0.081 -0.239 -0.501 391s 0.017 0.239 0.483 0.729 391s 391s y: [2, 2, 1, 4] 391s 0.214 0.078 391s -0.034 0.078 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.121 0.106 0.807 -0.567 391s 0.243 0.901 0.590 0.222 391s 0.817 0.926 -0.435 0.786 391s -0.282 -0.979 -0.114 -0.173 391s 391s x1: [4, 2, 1, 1] 391s 0.257 -0.976 0.593 -0.840 391s 0.778 -0.004 -0.245 0.438 391s 391s y: [4, 2, 1, 1] 391s 0.820 -0.653 -1.612 0.960 391s -0.540 0.138 1.083 -0.264 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.789 -0.636 0.668 0.919 391s 0.742 -0.888 -0.615 0.258 391s 391s x1: [4, 4, 1, 1] 391s -0.864 -0.225 -0.299 0.972 391s -0.226 -0.185 0.229 -0.202 391s -0.592 0.389 -0.424 0.404 391s -0.855 -0.986 -0.606 0.024 391s 391s y: [2, 4, 1, 1] 391s 1.518 -0.007 391s 0.263 -0.197 391s 0.308 -0.420 391s 0.919 0.619 391s testing: mul_mat, [3, 4, 1, 4] = [2, 3, 1, 4] * [2, 4, 1, 4] 391s x0: [2, 3, 1, 4] 391s -0.573 -0.054 391s -0.341 -0.032 391s 0.878 0.245 391s 391s x1: [2, 4, 1, 4] 391s -0.049 -0.195 391s 0.691 -0.828 391s -0.409 -0.481 391s -0.692 0.366 391s 391s y: [3, 4, 1, 4] 391s 0.039 0.023 -0.091 391s -0.351 -0.209 0.405 391s 0.261 0.155 -0.477 391s 0.376 0.224 -0.518 391s test-mul-mat0: iter:435/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.799 -0.978 -0.471 391s -0.042 -0.655 0.172 391s 391s x1: [3, 3, 1, 1] 391s -0.209 0.326 0.514 391s 0.218 0.985 0.392 391s -0.836 -0.047 -0.363 391s 391s y: [2, 3, 1, 1] 391s -0.394 -0.116 391s -1.322 -0.587 391s 0.886 0.004 391s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.154 -0.552 -0.109 391s -0.893 0.085 0.220 391s 391s x1: [3, 2, 3, 1] 391s -0.385 -0.817 0.661 391s 0.316 0.549 -0.610 391s 391s y: [2, 2, 3, 1] 391s 0.320 0.420 391s -0.188 -0.370 391s testing: mul_mat, [1, 2, 3, 2] = [3, 1, 3, 2] * [3, 2, 3, 2] 391s x0: [3, 1, 3, 2] 391s -0.298 0.869 0.032 391s 391s x1: [3, 2, 3, 2] 391s 0.282 -0.116 -0.457 391s 0.944 0.889 0.858 391s 391s y: [1, 2, 3, 2] 391s -0.199 391s 0.519 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.633 0.809 0.894 391s 0.561 -0.739 -0.526 391s -0.513 0.772 0.814 391s 0.730 0.945 -0.074 391s 391s x1: [3, 1, 1, 1] 391s -0.003 -0.183 -0.802 391s 391s y: [4, 1, 1, 1] 391s -0.867 0.555 -0.793 -0.115 391s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 1, 3, 1] 391s -0.377 -0.604 -0.227 0.859 391s 391s x1: [4, 3, 3, 1] 391s 0.346 -0.415 -0.587 0.115 391s 0.412 -0.197 -0.328 -0.597 391s 0.180 0.236 0.865 -0.349 391s 391s y: [1, 3, 3, 1] 391s 0.352 391s -0.474 391s -0.706 391s testing: mul_mat, [1, 4, 3, 2] = [1, 1, 3, 2] * [1, 4, 3, 2] 391s x0: [1, 1, 3, 2] 391s -0.847 391s 391s x1: [1, 4, 3, 2] 391s -0.224 391s -0.281 391s 0.387 391s -0.927 391s 391s y: [1, 4, 3, 2] 391s 0.190 391s 0.238 391s -0.328 391s 0.786 391s test-mul-mat0: iter:436/500 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.681 0.692 0.711 0.140 391s 391s x1: [4, 1, 1, 1] 391s -0.757 -0.895 -0.401 0.223 391s 391s y: [1, 1, 1, 1] 391s -1.389 391s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 2, 2, 1] 391s 0.872 -0.731 0.888 -0.975 391s -0.375 -0.978 -0.855 -0.814 391s 391s x1: [4, 1, 2, 1] 391s -0.002 -0.908 0.389 -0.746 391s 391s y: [2, 1, 2, 1] 391s 1.734 1.163 391s testing: mul_mat, [1, 2, 2, 3] = [4, 1, 2, 3] * [4, 2, 2, 3] 391s x0: [4, 1, 2, 3] 391s 0.225 -0.024 -0.209 -0.723 391s 391s x1: [4, 2, 2, 3] 391s 0.932 0.612 -0.711 0.624 391s 0.322 0.429 -0.379 0.415 391s 391s y: [1, 2, 2, 3] 391s -0.107 391s -0.158 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.016 0.329 -0.594 0.993 391s 391s x1: [4, 1, 1, 1] 391s 0.152 -0.137 -0.946 -0.653 391s 391s y: [1, 1, 1, 1] 391s -0.129 391s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.464 391s -0.292 391s -0.253 391s 391s x1: [1, 4, 2, 1] 391s -0.880 391s -0.316 391s 0.169 391s 0.836 391s 391s y: [3, 4, 2, 1] 391s 0.409 0.257 0.223 391s 0.147 0.093 0.080 391s -0.079 -0.049 -0.043 391s -0.388 -0.244 -0.212 391s testing: mul_mat, [2, 1, 2, 3] = [3, 2, 2, 3] * [3, 1, 2, 3] 391s x0: [3, 2, 2, 3] 391s -0.861 -0.963 -0.664 391s -0.189 -0.787 -0.737 391s 391s x1: [3, 1, 2, 3] 391s -0.713 0.498 0.197 391s 391s y: [2, 1, 2, 3] 391s 0.003 -0.403 391s test-mul-mat0: iter:437/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.568 0.489 0.360 391s -0.647 -0.484 -0.665 391s 0.644 0.029 -0.841 391s 391s x1: [3, 1, 1, 1] 391s 0.311 -0.864 0.086 391s 391s y: [3, 1, 1, 1] 391s -0.215 0.160 0.103 391s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 3, 2, 1] 391s -0.200 -0.593 -0.283 391s -0.633 0.896 -0.923 391s -0.280 -0.588 -0.588 391s 391s x1: [3, 3, 2, 1] 391s 0.789 -0.790 0.846 391s -0.884 -0.535 0.391 391s -0.254 -0.522 0.963 391s 391s y: [3, 3, 2, 1] 391s 0.070 -1.988 -0.254 391s 0.383 -0.282 0.331 391s 0.088 -1.197 -0.188 391s testing: mul_mat, [2, 3, 2, 4] = [3, 2, 2, 4] * [3, 3, 2, 4] 391s x0: [3, 2, 2, 4] 391s 0.709 -0.749 -0.622 391s 0.015 -0.924 0.278 391s 391s x1: [3, 3, 2, 4] 391s 0.016 -0.405 0.771 391s -0.469 0.895 -0.658 391s 0.028 -0.645 0.783 391s 391s y: [2, 3, 2, 4] 391s -0.165 0.589 391s -0.593 -1.017 391s 0.015 0.814 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.783 0.612 0.353 391s -0.697 0.327 0.864 391s 391s x1: [3, 2, 1, 1] 391s 0.630 0.120 0.259 391s 0.390 0.006 0.168 391s 391s y: [2, 2, 1, 1] 391s 0.658 -0.176 391s 0.369 -0.125 391s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.014 0.478 391s 0.809 0.444 391s 0.815 -0.290 391s 0.589 -0.264 391s 391s x1: [2, 3, 2, 1] 391s -0.466 -0.889 391s -0.634 -0.880 391s 0.885 -0.054 391s 391s y: [4, 3, 2, 1] 391s -0.418 -0.772 -0.121 -0.040 391s -0.411 -0.903 -0.261 -0.142 391s -0.039 0.691 0.736 0.536 391s testing: mul_mat, [2, 2, 2, 4] = [4, 2, 2, 4] * [4, 2, 2, 4] 391s x0: [4, 2, 2, 4] 391s -0.712 0.179 -0.020 -0.121 391s -0.915 0.112 0.906 0.804 391s 391s x1: [4, 2, 2, 4] 391s 0.560 0.605 0.360 0.671 391s 0.971 0.480 0.555 -0.083 391s 391s y: [2, 2, 2, 4] 391s -0.379 0.421 391s -0.607 -0.398 391s test-mul-mat0: iter:438/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.212 0.180 -0.951 391s -0.566 -0.532 -0.936 391s -0.747 0.663 -0.595 391s 391s x1: [3, 1, 1, 1] 391s -0.650 -0.040 0.294 391s 391s y: [3, 1, 1, 1] 391s -0.424 0.115 0.285 391s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.925 0.476 -0.795 391s -0.282 0.980 0.841 391s 391s x1: [3, 3, 4, 1] 391s -0.776 0.804 0.178 391s 0.731 -0.411 0.090 391s -0.802 -0.879 -0.729 391s 391s y: [2, 3, 4, 1] 391s 0.959 1.156 391s -0.943 -0.533 391s 0.903 -1.249 391s testing: mul_mat, [4, 2, 4, 4] = [3, 4, 4, 4] * [3, 2, 4, 4] 391s x0: [3, 4, 4, 4] 391s 0.376 -0.719 -0.660 391s -0.289 -0.106 0.581 391s -0.880 0.876 0.711 391s -0.786 0.913 -0.650 391s 391s x1: [3, 2, 4, 4] 391s -0.409 0.868 -0.928 391s 0.436 0.746 0.427 391s 391s y: [4, 2, 4, 4] 391s -0.165 -0.513 0.460 1.717 391s -0.653 0.043 0.572 0.061 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.385 0.018 -0.087 391s -0.658 0.485 -0.216 391s -0.057 -0.027 0.595 391s 391s x1: [3, 4, 1, 1] 391s 0.581 0.299 0.776 391s 0.516 -0.671 -0.482 391s 0.979 -0.079 0.831 391s 0.417 -0.866 0.615 391s 391s y: [3, 4, 1, 1] 391s -0.286 -0.405 0.421 391s -0.169 -0.560 -0.298 391s -0.451 -0.862 0.441 391s -0.230 -0.827 0.366 391s testing: mul_mat, [2, 3, 4, 1] = [3, 2, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 2, 4, 1] 391s -0.104 0.294 0.848 391s 0.414 -0.876 0.249 391s 391s x1: [3, 3, 4, 1] 391s 0.730 0.542 -0.204 391s 0.155 -0.794 -0.930 391s 0.646 0.136 -0.820 391s 391s y: [2, 3, 4, 1] 391s -0.089 -0.224 391s -1.039 0.527 391s -0.723 -0.056 391s testing: mul_mat, [3, 3, 4, 4] = [2, 3, 4, 4] * [2, 3, 4, 4] 391s x0: [2, 3, 4, 4] 391s -0.464 -0.889 391s 0.756 -0.787 391s -0.659 0.469 391s 391s x1: [2, 3, 4, 4] 391s 0.045 -0.537 391s -0.446 -0.910 391s 0.713 -0.168 391s 391s y: [3, 3, 4, 4] 391s 0.457 0.457 -0.282 391s 1.016 0.379 -0.133 391s -0.181 0.671 -0.549 391s test-mul-mat0: iter:439/500 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.662 0.068 -0.623 0.193 391s 391s x1: [4, 1, 1, 1] 391s -0.613 -0.858 0.200 0.018 391s 391s y: [1, 1, 1, 1] 391s 0.226 391s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 1, 2, 1] 391s -0.933 -0.768 0.420 -0.890 391s 391s x1: [4, 1, 2, 1] 391s 0.034 0.329 -0.606 -0.340 391s 391s y: [1, 1, 2, 1] 391s -0.236 391s testing: mul_mat, [1, 1, 2, 4] = [4, 1, 2, 4] * [4, 1, 2, 4] 391s x0: [4, 1, 2, 4] 391s -0.137 -0.203 0.775 0.930 391s 391s x1: [4, 1, 2, 4] 391s 0.215 0.257 -0.319 -0.872 391s 391s y: [1, 1, 2, 4] 391s -1.141 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.193 -0.569 -0.718 0.978 391s 391s x1: [4, 1, 1, 1] 391s 0.457 0.944 -0.209 -0.476 391s 391s y: [1, 1, 1, 1] 391s -0.940 391s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.622 391s 0.403 391s 0.349 391s 391s x1: [1, 4, 2, 1] 391s 0.555 391s -0.624 391s 0.336 391s 0.225 391s 391s y: [3, 4, 2, 1] 391s -0.345 0.224 0.194 391s 0.388 -0.252 -0.218 391s -0.209 0.136 0.118 391s -0.140 0.091 0.079 391s testing: mul_mat, [4, 1, 2, 4] = [3, 4, 2, 4] * [3, 1, 2, 4] 391s x0: [3, 4, 2, 4] 391s 0.333 0.074 -0.322 391s 0.134 -0.791 -0.801 391s 0.084 0.306 0.597 391s -0.589 0.005 0.316 391s 391s x1: [3, 1, 2, 4] 391s -0.105 0.931 -0.531 391s 391s y: [4, 1, 2, 4] 391s 0.205 -0.325 -0.041 -0.101 391s test-mul-mat0: iter:440/500 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.934 -0.863 0.299 -0.884 391s -0.180 -0.360 0.209 -0.336 391s 391s x1: [4, 4, 1, 1] 391s 0.728 -0.661 0.008 0.346 391s -0.978 0.478 -0.016 -0.327 391s -0.840 -0.138 -0.360 -0.098 391s -0.803 0.001 -0.120 -0.232 391s 391s y: [2, 4, 1, 1] 391s 0.947 -0.008 391s -1.042 0.111 391s -0.686 0.159 391s -0.583 0.197 391s testing: mul_mat, [1, 2, 4, 1] = [4, 1, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 1, 4, 1] 391s -0.409 -0.812 0.760 -0.223 391s 391s x1: [4, 2, 4, 1] 391s -0.332 -0.920 0.930 0.155 391s 0.808 -0.815 -0.117 -0.852 391s 391s y: [1, 2, 4, 1] 391s 1.555 391s 0.433 391s testing: mul_mat, [4, 1, 4, 4] = [4, 4, 4, 4] * [4, 1, 4, 4] 391s x0: [4, 4, 4, 4] 391s 0.732 0.453 0.330 0.667 391s -0.386 -0.067 -0.322 0.177 391s 0.435 0.423 -0.823 -0.210 391s -0.504 0.799 -0.244 -0.328 391s 391s x1: [4, 1, 4, 4] 391s -0.140 -0.112 0.818 0.900 391s 391s y: [4, 1, 4, 4] 391s 0.718 -0.042 -0.971 -0.514 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.519 0.834 -0.642 0.414 391s 0.935 0.834 -0.859 0.896 391s 391s x1: [4, 4, 1, 1] 391s -0.688 -0.333 -0.150 -0.741 391s -0.566 0.692 0.645 0.585 391s 0.114 0.724 0.612 -0.074 391s 0.297 0.584 0.021 0.599 391s 391s y: [2, 4, 1, 1] 391s -0.846 -1.457 391s 0.111 0.018 391s 0.239 0.118 391s 0.876 1.284 391s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 3, 4, 1] 391s -0.938 -0.608 391s -0.690 0.981 391s 0.290 -0.812 391s 391s x1: [2, 4, 4, 1] 391s -0.095 -0.640 391s 0.680 0.359 391s -0.769 -0.670 391s 0.671 -0.102 391s 391s y: [3, 4, 4, 1] 391s 0.478 -0.562 0.492 391s -0.856 -0.117 -0.095 391s 1.129 -0.127 0.322 391s -0.567 -0.563 0.277 391s testing: mul_mat, [3, 2, 4, 4] = [3, 3, 4, 4] * [3, 2, 4, 4] 391s x0: [3, 3, 4, 4] 391s 0.053 0.185 -0.107 391s 0.701 -0.469 -0.356 391s 0.048 0.364 0.263 391s 391s x1: [3, 2, 4, 4] 391s -0.992 -0.813 0.098 391s -0.293 -0.430 -0.951 391s 391s y: [3, 2, 4, 4] 391s -0.214 -0.349 -0.318 391s 0.007 0.335 -0.420 391s test-mul-mat0: iter:441/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.206 0.292 391s -0.755 0.201 391s -0.947 0.805 391s 391s x1: [2, 1, 1, 1] 391s -0.806 -0.308 391s 391s y: [3, 1, 1, 1] 391s 0.076 0.547 0.516 391s testing: mul_mat, [1, 3, 4, 1] = [2, 1, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 1, 4, 1] 391s 0.375 -0.738 391s 391s x1: [2, 3, 4, 1] 391s -0.528 -0.251 391s 0.659 -0.565 391s 0.371 -0.113 391s 391s y: [1, 3, 4, 1] 391s -0.012 391s 0.664 391s 0.223 391s testing: mul_mat, [2, 1, 4, 1] = [2, 2, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 2, 4, 1] 391s -0.944 0.749 391s 0.025 -0.771 391s 391s x1: [2, 1, 4, 1] 391s -0.043 0.483 391s 391s y: [2, 1, 4, 1] 391s 0.402 -0.374 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.590 0.819 391s 391s x1: [2, 2, 1, 1] 391s 0.563 -0.410 391s -0.473 -0.138 391s 391s y: [1, 2, 1, 1] 391s -0.003 391s -0.392 391s testing: mul_mat, [3, 2, 4, 1] = [1, 3, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 3, 4, 1] 391s 0.096 391s -0.105 391s -0.196 391s 391s x1: [1, 2, 4, 1] 391s 0.121 391s -0.975 391s 391s y: [3, 2, 4, 1] 391s 0.012 -0.013 -0.024 391s -0.094 0.103 0.191 391s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.827 -0.973 0.596 391s -0.842 -0.208 0.901 391s -0.783 -0.384 0.488 391s 0.520 -0.078 0.105 391s 391s x1: [3, 1, 4, 1] 391s 0.790 0.279 -0.346 391s 391s y: [4, 1, 4, 1] 391s -1.131 -1.035 -0.894 0.353 391s test-mul-mat0: iter:442/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.380 -0.759 -0.581 0.417 391s 0.564 -0.912 0.568 -0.518 391s -0.721 0.096 0.437 -0.218 391s 0.280 0.083 -0.050 0.989 391s 391s x1: [4, 4, 1, 1] 391s -0.832 -0.291 0.519 -0.953 391s -0.260 -0.837 0.961 0.399 391s 0.605 0.573 -0.769 0.204 391s 0.402 -0.417 -0.169 -0.976 391s 391s y: [4, 4, 1, 1] 391s -0.795 0.584 1.007 -1.225 391s 0.144 0.955 0.440 0.204 391s 0.327 -0.724 -0.762 0.458 391s 0.160 1.017 -0.191 -0.879 391s testing: mul_mat, [4, 4, 3, 1] = [4, 4, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 4, 3, 1] 391s 0.474 0.350 0.698 -0.755 391s 0.595 -0.023 -0.888 0.277 391s 0.038 0.259 -0.958 -0.951 391s -0.606 0.410 -0.944 -0.741 391s 391s x1: [4, 4, 3, 1] 391s -0.398 -0.003 -0.271 -0.234 391s -0.042 -0.872 -0.629 -0.469 391s -0.641 0.575 0.934 -0.058 391s -0.594 0.957 -0.507 0.786 391s 391s y: [4, 4, 3, 1] 391s -0.202 -0.060 0.466 0.669 391s -0.410 0.424 0.821 0.609 391s 0.593 -1.239 -0.715 -0.214 391s -0.894 0.293 -0.037 0.649 391s testing: mul_mat, [4, 4, 3, 4] = [4, 4, 3, 4] * [4, 4, 3, 4] 391s x0: [4, 4, 3, 4] 391s 0.649 0.507 0.782 0.914 391s -0.463 0.968 -0.388 0.608 391s 0.256 -0.186 0.626 -0.261 391s 0.221 -0.004 -0.897 0.613 391s 391s x1: [4, 4, 3, 4] 391s 0.223 -0.124 -0.812 0.742 391s 0.479 -0.685 0.691 0.376 391s -0.073 0.028 0.059 0.809 391s 0.951 0.441 -0.836 -0.776 391s 391s y: [4, 4, 3, 4] 391s 0.125 0.542 -0.622 1.233 391s 0.847 -0.923 0.585 -0.281 391s 0.752 0.529 -0.198 0.426 391s -0.522 -0.161 -0.159 0.482 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.357 -0.255 -0.352 0.193 391s 0.222 -0.984 0.573 -0.041 391s -0.969 0.659 -0.460 0.491 391s 391s x1: [4, 4, 1, 1] 391s 0.973 -0.715 -0.769 -0.317 391s -0.775 -0.066 -0.192 0.760 391s 0.359 0.066 -0.761 -0.234 391s -0.721 0.680 0.434 0.928 391s 391s y: [3, 4, 1, 1] 391s 0.740 0.492 -1.215 391s -0.045 -0.248 1.168 391s 0.334 -0.413 -0.069 391s -0.405 -0.619 1.402 391s testing: mul_mat, [3, 4, 3, 1] = [3, 3, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 3, 3, 1] 391s -0.559 -0.305 0.188 391s -0.525 -0.975 -0.659 391s 0.998 -0.980 -0.678 391s 391s x1: [3, 4, 3, 1] 391s 0.984 0.026 0.464 391s -0.731 0.257 -0.853 391s -0.505 -0.809 -0.044 391s -0.745 0.550 -0.978 391s 391s y: [3, 4, 3, 1] 391s -0.471 -0.848 0.642 391s 0.170 0.695 -0.403 391s 0.521 1.083 0.319 391s 0.065 0.499 -0.620 391s testing: mul_mat, [1, 3, 3, 4] = [3, 1, 3, 4] * [3, 3, 3, 4] 391s x0: [3, 1, 3, 4] 391s -0.328 -0.791 0.388 391s 391s x1: [3, 3, 3, 4] 391s -0.515 -0.137 0.827 391s 0.528 0.303 -0.698 391s 0.526 0.998 -0.673 391s 391s y: [1, 3, 3, 4] 391s 0.598 391s -0.684 391s -1.224 391s test-mul-mat0: iter:443/500 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.554 -0.132 -0.233 391s 0.534 -0.865 0.051 391s 391s x1: [3, 1, 1, 1] 391s -0.710 -0.213 -0.652 391s 391s y: [2, 1, 1, 1] 391s -0.213 -0.228 391s testing: mul_mat, [2, 2, 3, 1] = [3, 2, 3, 1] * [3, 2, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.457 0.012 0.701 391s -0.756 0.360 -0.167 391s 391s x1: [3, 2, 3, 1] 391s 0.579 -0.055 0.775 391s 0.424 0.015 -0.780 391s 391s y: [2, 2, 3, 1] 391s 0.808 -0.587 391s -0.353 -0.184 391s testing: mul_mat, [3, 2, 3, 3] = [3, 3, 3, 3] * [3, 2, 3, 3] 391s x0: [3, 3, 3, 3] 391s -0.072 -0.242 0.660 391s -0.851 0.148 -0.281 391s -0.985 -0.258 0.755 391s 391s x1: [3, 2, 3, 3] 391s 0.204 0.977 0.669 391s 0.533 -0.495 -0.730 391s 391s y: [3, 2, 3, 3] 391s 0.191 -0.217 0.052 391s -0.400 -0.322 -0.948 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.552 -0.601 -0.078 391s 0.521 -0.160 -0.918 391s -0.014 -0.850 -0.579 391s 391s x1: [3, 3, 1, 1] 391s 0.430 -0.840 -0.106 391s -0.335 -0.940 -0.160 391s 0.781 0.921 0.408 391s 391s y: [3, 3, 1, 1] 391s 0.276 0.457 0.769 391s 0.763 0.123 0.897 391s -1.017 -0.115 -1.030 391s testing: mul_mat, [4, 3, 3, 1] = [3, 4, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 4, 3, 1] 391s 0.291 -0.095 0.050 391s 0.902 -0.241 -0.198 391s 0.410 0.827 -0.479 391s -0.951 0.803 0.294 391s 391s x1: [3, 3, 3, 1] 391s -0.242 0.003 0.350 391s 0.778 -0.102 -0.709 391s -0.025 -0.282 -0.756 391s 391s y: [4, 3, 3, 1] 391s -0.053 -0.288 -0.264 0.335 391s 0.200 0.867 0.574 -1.031 391s -0.018 0.195 0.119 -0.425 391s testing: mul_mat, [1, 3, 3, 3] = [4, 1, 3, 3] * [4, 3, 3, 3] 391s x0: [4, 1, 3, 3] 391s -0.109 0.644 -0.743 0.649 391s 391s x1: [4, 3, 3, 3] 391s -0.693 -0.928 0.744 0.357 391s -0.126 -0.735 -0.350 -0.260 391s -0.833 -0.738 0.100 0.611 391s 391s y: [1, 3, 3, 3] 391s -0.844 391s -0.369 391s -0.063 391s test-mul-mat0: iter:444/500 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.290 0.264 0.439 391s 391s x1: [3, 3, 1, 1] 391s 0.286 0.224 -0.879 391s -0.419 -0.543 0.485 391s 0.721 -0.853 -0.192 391s 391s y: [1, 3, 1, 1] 391s -0.244 391s -0.052 391s -0.101 391s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.073 0.578 0.671 391s -0.558 -0.467 -0.443 391s 391s x1: [3, 1, 3, 1] 391s 0.675 0.924 0.573 391s 391s y: [2, 1, 3, 1] 391s 0.869 -1.061 391s testing: mul_mat, [3, 2, 3, 3] = [3, 3, 3, 3] * [3, 2, 3, 3] 391s x0: [3, 3, 3, 3] 391s 0.209 0.252 0.737 391s 0.680 -0.786 -0.408 391s 0.845 -0.765 -0.070 391s 391s x1: [3, 2, 3, 3] 391s 0.059 0.482 0.133 391s -0.265 0.406 -0.294 391s 391s y: [3, 2, 3, 3] 391s 0.232 -0.393 -0.328 391s -0.170 -0.380 -0.514 391s testing: mul_mat, [1, 3, 1, 1] = [3, 1, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.153 0.231 0.700 391s 391s x1: [3, 3, 1, 1] 391s -0.027 0.580 0.304 391s 0.115 -0.008 -0.908 391s 0.641 0.395 -0.632 391s 391s y: [1, 3, 1, 1] 391s 0.351 391s -0.654 391s -0.449 391s testing: mul_mat, [3, 3, 3, 1] = [1, 3, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 3, 3, 1] 391s -0.061 391s 0.965 391s -0.375 391s 391s x1: [1, 3, 3, 1] 391s -0.398 391s -0.582 391s -0.877 391s 391s y: [3, 3, 3, 1] 391s 0.024 -0.384 0.149 391s 0.036 -0.561 0.218 391s 0.054 -0.846 0.329 391s testing: mul_mat, [1, 1, 3, 3] = [3, 1, 3, 3] * [3, 1, 3, 3] 391s x0: [3, 1, 3, 3] 391s -0.660 -0.464 0.932 391s 391s x1: [3, 1, 3, 3] 391s 0.362 0.405 -0.054 391s 391s y: [1, 1, 3, 3] 391s -0.477 391s test-mul-mat0: iter:445/500 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.497 391s 0.310 391s 391s x1: [1, 2, 1, 1] 391s -0.199 391s 0.369 391s 391s y: [2, 2, 1, 1] 391s 0.099 -0.062 391s -0.184 0.115 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.868 391s -0.635 391s 391s x1: [1, 2, 1, 1] 391s -0.597 391s -0.440 391s 391s y: [2, 2, 1, 1] 391s -0.518 0.379 391s -0.383 0.280 391s testing: mul_mat, [3, 2, 1, 2] = [1, 3, 1, 2] * [1, 2, 1, 2] 391s x0: [1, 3, 1, 2] 391s -0.634 391s -0.631 391s 0.600 391s 391s x1: [1, 2, 1, 2] 391s -0.947 391s 0.353 391s 391s y: [3, 2, 1, 2] 391s 0.601 0.598 -0.569 391s -0.224 -0.223 0.212 391s testing: mul_mat, [4, 3, 1, 1] = [1, 4, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.318 391s 0.365 391s 0.406 391s 0.143 391s 391s x1: [1, 3, 1, 1] 391s 0.666 391s -0.222 391s -0.034 391s 391s y: [4, 3, 1, 1] 391s -0.212 0.243 0.271 0.095 391s 0.071 -0.081 -0.090 -0.032 391s 0.011 -0.012 -0.014 -0.005 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.553 -0.071 0.627 0.372 391s -0.122 0.007 -0.203 0.680 391s 391s x1: [4, 1, 1, 1] 391s -0.802 -0.000 0.944 0.568 391s 391s y: [2, 1, 1, 1] 391s 1.246 0.292 391s testing: mul_mat, [4, 4, 1, 2] = [2, 4, 1, 2] * [2, 4, 1, 2] 391s x0: [2, 4, 1, 2] 391s 0.425 0.437 391s 0.869 0.050 391s -0.475 0.817 391s 0.369 -0.685 391s 391s x1: [2, 4, 1, 2] 391s -0.850 -0.862 391s -0.867 -0.294 391s -0.496 -0.498 391s -0.694 -0.461 391s 391s y: [4, 4, 1, 2] 391s -0.738 -0.782 -0.300 0.277 391s -0.497 -0.768 0.171 -0.118 391s -0.428 -0.456 -0.171 0.158 391s -0.497 -0.626 -0.047 0.060 391s test-mul-mat0: iter:446/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.276 0.580 391s -0.319 0.582 391s 0.698 -0.036 391s 391s x1: [2, 1, 1, 1] 391s 0.937 0.376 391s 391s y: [3, 1, 1, 1] 391s 0.476 -0.080 0.641 391s testing: mul_mat, [4, 3, 2, 1] = [2, 4, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 4, 2, 1] 391s 0.553 -0.407 391s -0.564 -0.522 391s -0.074 -0.809 391s -0.891 0.763 391s 391s x1: [2, 3, 2, 1] 391s 0.252 0.224 391s -0.611 0.121 391s 0.748 0.759 391s 391s y: [4, 3, 2, 1] 391s 0.048 -0.259 -0.200 -0.054 391s -0.387 0.281 -0.053 0.636 391s 0.105 -0.818 -0.669 -0.088 391s testing: mul_mat, [1, 4, 2, 2] = [2, 1, 2, 2] * [2, 4, 2, 2] 391s x0: [2, 1, 2, 2] 391s 0.531 -0.509 391s 391s x1: [2, 4, 2, 2] 391s -0.583 0.174 391s -0.844 0.640 391s 0.564 0.277 391s 0.389 0.322 391s 391s y: [1, 4, 2, 2] 391s -0.399 391s -0.774 391s 0.158 391s 0.042 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.410 -0.170 391s -0.128 -0.458 391s 0.547 0.295 391s -0.520 -0.022 391s 391s x1: [2, 1, 1, 1] 391s 0.294 0.350 391s 391s y: [4, 1, 1, 1] 391s -0.180 -0.198 0.264 -0.160 391s testing: mul_mat, [1, 2, 2, 1] = [4, 1, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 1, 2, 1] 391s 0.015 0.626 0.313 -0.692 391s 391s x1: [4, 2, 2, 1] 391s 0.646 0.873 0.575 -0.204 391s 0.590 0.701 0.268 -0.976 391s 391s y: [1, 2, 2, 1] 391s 0.877 391s 1.205 391s testing: mul_mat, [3, 4, 2, 2] = [1, 3, 2, 2] * [1, 4, 2, 2] 391s x0: [1, 3, 2, 2] 391s -0.486 391s -0.040 391s -0.598 391s 391s x1: [1, 4, 2, 2] 391s 0.891 391s 0.378 391s -0.322 391s -0.567 391s 391s y: [3, 4, 2, 2] 391s -0.433 -0.036 -0.533 391s -0.184 -0.015 -0.226 391s 0.157 0.013 0.193 391s 0.276 0.023 0.339 391s test-mul-mat0: iter:447/500 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.952 0.957 -0.419 -0.912 391s 391s x1: [4, 1, 1, 1] 391s 0.070 0.628 0.406 -0.274 391s 391s y: [1, 1, 1, 1] 391s 0.747 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.284 -0.630 -0.532 0.686 391s 0.503 -0.590 0.334 0.650 391s 391s x1: [4, 1, 1, 1] 391s -0.112 -0.277 -0.752 0.856 391s 391s y: [2, 1, 1, 1] 391s 1.129 0.413 391s testing: mul_mat, [2, 2, 1, 3] = [4, 2, 1, 3] * [4, 2, 1, 3] 391s x0: [4, 2, 1, 3] 391s 0.409 -0.142 0.282 0.819 391s -0.809 -0.067 0.797 -0.514 391s 391s x1: [4, 2, 1, 3] 391s 0.978 -0.706 -0.894 -0.479 391s 0.598 0.461 -0.087 -0.207 391s 391s y: [2, 2, 1, 3] 391s -0.144 -1.209 391s -0.015 -0.478 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.388 0.545 -0.416 0.084 391s 0.735 0.532 0.851 0.082 391s 0.596 -0.918 -0.834 0.044 391s 391s x1: [4, 2, 1, 1] 391s -0.860 0.566 0.375 0.550 391s -0.493 -0.453 0.330 0.916 391s 391s y: [3, 2, 1, 1] 391s 0.533 0.033 -1.321 391s -0.116 -0.248 -0.113 391s testing: mul_mat, [3, 4, 1, 1] = [3, 3, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.505 0.825 -0.563 391s 0.976 0.892 0.627 391s 0.325 -0.270 -0.674 391s 391s x1: [3, 4, 1, 1] 391s 0.778 -0.704 -0.658 391s -0.928 -0.782 0.768 391s 0.918 -0.195 0.903 391s 0.404 -0.055 0.469 391s 391s y: [3, 4, 1, 1] 391s -0.603 -0.281 0.886 391s -0.609 -1.121 -0.609 391s -1.133 1.289 -0.257 391s -0.513 0.640 -0.170 391s testing: mul_mat, [3, 3, 1, 3] = [3, 3, 1, 3] * [3, 3, 1, 3] 391s x0: [3, 3, 1, 3] 391s 0.114 0.095 0.661 391s -0.403 0.757 -0.233 391s 0.983 -0.609 -0.034 391s 391s x1: [3, 3, 1, 3] 391s 0.982 0.160 -0.592 391s -0.434 0.010 -0.425 391s 0.650 -0.908 0.618 391s 391s y: [3, 3, 1, 3] 391s -0.264 -0.137 0.888 391s -0.330 0.282 -0.419 391s 0.396 -1.093 1.170 391s test-mul-mat0: iter:448/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.041 0.021 391s -0.824 -0.993 391s -0.667 0.035 391s 391s x1: [2, 2, 1, 1] 391s -0.735 0.652 391s -0.351 0.360 391s 391s y: [3, 2, 1, 1] 391s -0.016 -0.041 0.514 391s -0.007 -0.069 0.247 391s testing: mul_mat, [3, 3, 2, 1] = [2, 3, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 3, 2, 1] 391s 0.535 0.500 391s -0.591 -0.456 391s -0.349 0.464 391s 391s x1: [2, 3, 2, 1] 391s -0.409 0.072 391s -0.717 0.136 391s 0.312 -0.862 391s 391s y: [3, 3, 2, 1] 391s -0.183 0.209 0.176 391s -0.315 0.361 0.313 391s -0.264 0.209 -0.509 391s testing: mul_mat, [2, 3, 2, 4] = [2, 2, 2, 4] * [2, 3, 2, 4] 391s x0: [2, 2, 2, 4] 391s -0.378 -0.035 391s -0.962 -0.282 391s 391s x1: [2, 3, 2, 4] 391s -0.099 0.297 391s -0.300 -0.093 391s 0.630 0.735 391s 391s y: [2, 3, 2, 4] 391s 0.027 0.012 391s 0.117 0.315 391s -0.264 -0.813 391s testing: mul_mat, [1, 2, 1, 1] = [2, 1, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.226 -0.234 391s 391s x1: [2, 2, 1, 1] 391s -0.924 -0.494 391s -0.052 0.557 391s 391s y: [1, 2, 1, 1] 391s -0.093 391s -0.142 391s testing: mul_mat, [1, 2, 2, 1] = [1, 1, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 1, 2, 1] 391s -0.155 391s 391s x1: [1, 2, 2, 1] 391s 0.707 391s -0.821 391s 391s y: [1, 2, 2, 1] 391s -0.110 391s 0.127 391s testing: mul_mat, [3, 1, 2, 4] = [1, 3, 2, 4] * [1, 1, 2, 4] 391s x0: [1, 3, 2, 4] 391s -0.443 391s -0.134 391s 0.743 391s 391s x1: [1, 1, 2, 4] 391s -0.615 391s 391s y: [3, 1, 2, 4] 391s 0.272 0.083 -0.457 391s test-mul-mat0: iter:449/500 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.759 -0.115 0.809 0.095 391s -0.575 -0.488 -0.976 -0.530 391s -0.837 -0.490 0.795 0.040 391s 0.763 -0.173 -0.903 -0.423 391s 391s x1: [4, 1, 1, 1] 391s 0.256 -0.530 -0.502 0.813 391s 391s y: [4, 1, 1, 1] 391s -0.462 0.171 -0.321 0.397 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s 0.003 -0.946 0.187 0.191 391s 391s x1: [4, 4, 3, 1] 391s 0.920 0.486 -0.763 -0.347 391s 0.260 0.419 0.459 0.106 391s 0.696 -0.669 -0.639 -0.835 391s -0.171 -0.826 0.501 0.070 391s 391s y: [1, 4, 3, 1] 391s -0.667 391s -0.290 391s 0.356 391s 0.888 391s testing: mul_mat, [4, 1, 3, 3] = [4, 4, 3, 3] * [4, 1, 3, 3] 391s x0: [4, 4, 3, 3] 391s 0.982 0.721 -0.903 0.735 391s 0.244 -0.798 -0.956 -0.466 391s 0.817 -0.590 -0.337 -0.998 391s -0.857 0.793 -0.760 0.062 391s 391s x1: [4, 1, 3, 3] 391s 0.391 -0.172 0.927 -0.333 391s 391s y: [4, 1, 3, 3] 391s -0.822 -0.498 0.441 -1.196 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.095 -0.138 0.984 0.453 391s -0.392 0.969 -0.272 0.993 391s 391s x1: [4, 4, 1, 1] 391s -0.433 0.113 -0.586 -0.217 391s -0.828 -0.304 -0.395 0.033 391s 0.128 0.198 -0.331 0.873 391s -0.435 -0.942 -0.725 -0.201 391s 391s y: [2, 4, 1, 1] 391s -0.649 0.223 391s -0.253 0.170 391s 0.031 1.098 391s -0.633 -0.745 391s testing: mul_mat, [1, 4, 3, 1] = [2, 1, 3, 1] * [2, 4, 3, 1] 391s x0: [2, 1, 3, 1] 391s 0.265 0.719 391s 391s x1: [2, 4, 3, 1] 391s 0.487 -0.820 391s -0.107 0.400 391s -0.027 -0.541 391s 0.968 -0.915 391s 391s y: [1, 4, 3, 1] 391s -0.460 391s 0.260 391s -0.396 391s -0.401 391s testing: mul_mat, [3, 2, 3, 3] = [1, 3, 3, 3] * [1, 2, 3, 3] 391s x0: [1, 3, 3, 3] 391s 0.932 391s -0.595 391s -0.453 391s 391s x1: [1, 2, 3, 3] 391s 0.480 391s 0.352 391s 391s y: [3, 2, 3, 3] 391s 0.447 -0.285 -0.217 391s 0.328 -0.209 -0.159 391s test-mul-mat0: iter:450/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.593 -0.045 -0.361 0.655 391s 0.356 -0.564 -0.988 -0.289 391s -0.811 0.361 0.577 0.857 391s -0.729 0.071 0.457 0.586 391s 391s x1: [4, 4, 1, 1] 391s 0.600 0.315 -0.959 0.935 391s 0.470 0.114 0.536 0.344 391s -0.304 0.129 -0.611 -0.220 391s -0.798 -0.214 -0.741 -0.846 391s 391s y: [4, 4, 1, 1] 391s 1.301 0.713 -0.124 -0.306 391s 0.305 -0.526 0.264 0.112 391s -0.110 0.486 -0.248 -0.177 391s -0.751 0.813 -0.581 -0.267 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.722 0.208 -0.276 -0.842 391s 0.220 0.436 -0.652 -0.419 391s 0.012 -0.795 -0.149 -0.917 391s 391s x1: [4, 4, 1, 1] 391s 0.112 0.393 0.056 -0.775 391s -0.071 -0.601 -0.079 -0.943 391s -0.211 0.701 -0.741 0.575 391s 0.960 -0.587 0.083 0.554 391s 391s y: [3, 4, 1, 1] 391s 0.799 0.484 0.391 391s 0.638 0.169 1.353 391s -0.286 0.501 -0.977 391s 0.082 -0.331 -0.042 391s testing: mul_mat, [1, 3, 1, 3] = [4, 1, 1, 3] * [4, 3, 1, 3] 391s x0: [4, 1, 1, 3] 391s -0.527 0.238 -0.707 -0.486 391s 391s x1: [4, 3, 1, 3] 391s 0.662 -0.563 0.195 0.055 391s 0.493 0.420 0.984 0.892 391s -0.659 -0.959 -0.319 -0.958 391s 391s y: [1, 3, 1, 3] 391s -0.647 391s -1.289 391s 0.810 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.356 -0.343 0.361 0.210 391s 391s x1: [4, 1, 1, 1] 391s 0.609 -0.426 -0.341 -0.584 391s 391s y: [1, 1, 1, 1] 391s -0.317 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.520 391s 0.779 391s 0.057 391s 391s x1: [1, 4, 1, 1] 391s 0.911 391s 0.133 391s -0.374 391s 0.248 391s 391s y: [3, 4, 1, 1] 391s -0.474 0.710 0.052 391s -0.069 0.104 0.008 391s 0.194 -0.291 -0.021 391s -0.129 0.193 0.014 391s testing: mul_mat, [1, 1, 1, 3] = [3, 1, 1, 3] * [3, 1, 1, 3] 391s x0: [3, 1, 1, 3] 391s -0.303 -0.615 -0.330 391s 391s x1: [3, 1, 1, 3] 391s 0.826 0.731 0.530 391s 391s y: [1, 1, 1, 3] 391s -0.875 391s test-mul-mat0: iter:451/500 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.269 -0.918 -0.880 391s 391s x1: [3, 4, 1, 1] 391s -0.022 0.533 0.730 391s -0.774 -0.899 -0.790 391s -0.994 0.158 -0.964 391s 0.737 -0.313 0.100 391s 391s y: [1, 4, 1, 1] 391s -1.126 391s 1.729 391s 0.972 391s 0.001 391s testing: mul_mat, [4, 1, 2, 1] = [3, 4, 2, 1] * [3, 1, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.154 -0.342 -0.624 391s -0.807 -0.448 0.443 391s -0.407 0.656 -0.578 391s -0.874 0.386 -0.352 391s 391s x1: [3, 1, 2, 1] 391s 0.588 0.239 -0.183 391s 391s y: [4, 1, 2, 1] 391s 0.123 -0.663 0.023 -0.357 391s testing: mul_mat, [1, 4, 2, 4] = [3, 1, 2, 4] * [3, 4, 2, 4] 391s x0: [3, 1, 2, 4] 391s 0.302 0.413 0.640 391s 391s x1: [3, 4, 2, 4] 391s -0.609 -0.947 0.088 391s -0.636 0.963 -0.504 391s -0.669 0.117 0.154 391s -0.293 0.309 0.706 391s 391s y: [1, 4, 2, 4] 391s -0.519 391s -0.118 391s -0.056 391s 0.491 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.837 -0.269 0.047 391s 0.633 0.139 -0.629 391s 391s x1: [3, 1, 1, 1] 391s 0.219 0.444 0.750 391s 391s y: [2, 1, 1, 1] 391s 0.098 -0.272 391s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.944 0.624 391s 391s x1: [2, 3, 2, 1] 391s -0.993 0.412 391s -0.893 -0.597 391s 0.860 -0.603 391s 391s y: [1, 3, 2, 1] 391s -0.681 391s -1.216 391s 0.436 391s testing: mul_mat, [3, 2, 2, 4] = [1, 3, 2, 4] * [1, 2, 2, 4] 391s x0: [1, 3, 2, 4] 391s -0.616 391s -0.557 391s 0.731 391s 391s x1: [1, 2, 2, 4] 391s 0.614 391s -0.972 391s 391s y: [3, 2, 2, 4] 391s -0.378 -0.342 0.449 391s 0.599 0.542 -0.711 391s test-mul-mat0: iter:452/500 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.919 0.762 -0.718 391s 391s x1: [3, 2, 1, 1] 391s 0.458 -0.685 -0.969 391s 0.843 -0.242 0.763 391s 391s y: [1, 2, 1, 1] 391s -0.247 391s -1.506 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.324 0.153 -0.787 391s 391s x1: [3, 1, 1, 1] 391s 0.118 0.731 0.942 391s 391s y: [1, 1, 1, 1] 391s -0.591 391s testing: mul_mat, [1, 1, 1, 1] = [3, 1, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.123 -0.450 -0.084 391s 391s x1: [3, 1, 1, 1] 391s -0.469 -0.608 0.883 391s 391s y: [1, 1, 1, 1] 391s 0.257 391s testing: mul_mat, [2, 1, 1, 1] = [3, 2, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 2, 1, 1] 391s 0.530 0.656 0.526 391s 0.211 -0.011 0.687 391s 391s x1: [3, 1, 1, 1] 391s 0.490 -0.947 0.478 391s 391s y: [2, 1, 1, 1] 391s -0.110 0.443 391s testing: mul_mat, [3, 3, 1, 1] = [2, 3, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.483 0.581 391s -0.057 -0.733 391s 0.477 -0.370 391s 391s x1: [2, 3, 1, 1] 391s -0.169 -0.716 391s 0.450 0.254 391s -0.635 0.212 391s 391s y: [3, 3, 1, 1] 391s -0.498 0.534 0.184 391s 0.365 -0.212 0.121 391s -0.184 -0.119 -0.381 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.807 0.389 -0.719 391s 391s x1: [3, 2, 1, 1] 391s 0.795 -0.202 0.022 391s 0.678 0.277 0.900 391s 391s y: [1, 2, 1, 1] 391s -0.736 391s -1.086 391s test-mul-mat0: iter:453/500 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.691 391s 0.631 391s 391s x1: [1, 2, 1, 1] 391s -0.996 391s -0.200 391s 391s y: [2, 2, 1, 1] 391s -0.688 -0.629 391s -0.138 -0.126 391s testing: mul_mat, [1, 2, 3, 1] = [1, 1, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 1, 3, 1] 391s 0.295 391s 391s x1: [1, 2, 3, 1] 391s -0.308 391s 0.141 391s 391s y: [1, 2, 3, 1] 391s -0.091 391s 0.042 391s testing: mul_mat, [4, 1, 3, 1] = [1, 4, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.550 391s 0.594 391s 0.637 391s 0.449 391s 391s x1: [1, 1, 3, 1] 391s 0.562 391s 391s y: [4, 1, 3, 1] 391s 0.309 0.334 0.358 0.253 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.482 391s -0.978 391s 391s x1: [1, 4, 1, 1] 391s 0.190 391s -0.286 391s -0.210 391s -0.119 391s 391s y: [2, 4, 1, 1] 391s 0.092 -0.186 391s -0.138 0.280 391s -0.101 0.206 391s -0.057 0.116 391s testing: mul_mat, [2, 1, 3, 1] = [2, 2, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 2, 3, 1] 391s 0.201 -0.925 391s 0.231 0.117 391s 391s x1: [2, 1, 3, 1] 391s 0.230 -0.461 391s 391s y: [2, 1, 3, 1] 391s 0.473 -0.001 391s testing: mul_mat, [4, 2, 3, 1] = [2, 4, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 4, 3, 1] 391s 0.751 0.236 391s 0.567 -0.963 391s -0.510 -0.969 391s -0.170 -0.532 391s 391s x1: [2, 2, 3, 1] 391s -0.363 -0.441 391s 0.100 -0.449 391s 391s y: [4, 2, 3, 1] 391s -0.377 0.219 0.613 0.296 391s -0.031 0.489 0.384 0.222 391s test-mul-mat0: iter:454/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.459 -0.953 391s 0.786 0.372 391s 0.284 0.823 391s 391s x1: [2, 1, 1, 1] 391s -0.210 -0.520 391s 391s y: [3, 1, 1, 1] 391s 0.591 -0.358 -0.487 391s testing: mul_mat, [4, 3, 4, 1] = [2, 4, 4, 1] * [2, 3, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.108 -0.031 391s 0.805 -0.520 391s -0.747 0.628 391s 0.883 -0.996 391s 391s x1: [2, 3, 4, 1] 391s 0.403 0.751 391s 0.936 -0.448 391s 0.811 0.674 391s 391s y: [4, 3, 4, 1] 391s 0.020 -0.066 0.171 -0.393 391s 0.115 0.987 -0.981 1.273 391s 0.067 0.302 -0.183 0.045 391s testing: mul_mat, [1, 4, 4, 4] = [2, 1, 4, 4] * [2, 4, 4, 4] 391s x0: [2, 1, 4, 4] 391s -0.403 -0.690 391s 391s x1: [2, 4, 4, 4] 391s 0.463 0.721 391s -0.158 0.715 391s 0.350 -0.276 391s 0.719 -0.086 391s 391s y: [1, 4, 4, 4] 391s -0.684 391s -0.429 391s 0.049 391s -0.231 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.832 -0.181 391s -0.680 -0.048 391s 0.860 0.931 391s 391s x1: [2, 1, 1, 1] 391s -0.391 0.547 391s 391s y: [3, 1, 1, 1] 391s -0.424 0.239 0.173 391s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.024 0.198 -0.918 391s 391s x1: [3, 2, 4, 1] 391s -0.901 0.209 0.961 391s 0.401 0.522 -0.663 391s 391s y: [1, 2, 4, 1] 391s -0.862 391s 0.721 391s testing: mul_mat, [4, 3, 4, 4] = [1, 4, 4, 4] * [1, 3, 4, 4] 391s x0: [1, 4, 4, 4] 391s -0.564 391s -0.711 391s -0.572 391s -0.159 391s 391s x1: [1, 3, 4, 4] 391s 0.426 391s 0.790 391s 0.319 391s 391s y: [4, 3, 4, 4] 391s -0.240 -0.303 -0.244 -0.068 391s -0.445 -0.562 -0.452 -0.126 391s -0.180 -0.227 -0.183 -0.051 391s test-mul-mat0: iter:455/500 391s testing: mul_mat, [2, 2, 1, 1] = [3, 2, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.174 0.907 -0.687 391s -0.113 0.547 0.282 391s 391s x1: [3, 2, 1, 1] 391s 0.509 0.891 0.478 391s 0.413 -0.858 -0.494 391s 391s y: [2, 2, 1, 1] 391s 0.392 0.565 391s -0.511 -0.656 391s testing: mul_mat, [4, 2, 2, 1] = [3, 4, 2, 1] * [3, 2, 2, 1] 391s x0: [3, 4, 2, 1] 391s -0.792 0.941 -0.090 391s 0.020 0.153 -0.580 391s -0.090 -0.369 0.833 391s 0.052 0.137 0.506 391s 391s x1: [3, 2, 2, 1] 391s -0.908 -0.079 0.966 391s -0.116 0.433 0.773 391s 391s y: [4, 2, 2, 1] 391s 0.559 -0.590 0.915 0.431 391s 0.430 -0.384 0.494 0.445 391s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 391s x0: [3, 2, 2, 4] 391s 0.451 -0.546 -0.514 391s -0.345 -0.105 -0.452 391s 391s x1: [3, 4, 2, 4] 391s 0.021 0.926 -0.692 391s -0.589 0.068 -0.643 391s -0.935 0.276 -0.703 391s -0.025 -0.705 0.451 391s 391s y: [2, 4, 2, 4] 391s -0.141 0.208 391s 0.028 0.487 391s -0.212 0.611 391s 0.142 -0.121 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.729 -0.258 -0.992 391s 0.454 0.788 -0.315 391s -0.978 -0.910 -0.124 391s 0.983 0.482 0.122 391s 391s x1: [3, 2, 1, 1] 391s 0.057 -0.401 0.055 391s -0.588 0.863 0.996 391s 391s y: [4, 2, 1, 1] 391s 0.007 -0.307 0.302 -0.130 391s -0.782 0.099 -0.333 -0.041 391s testing: mul_mat, [3, 3, 2, 1] = [4, 3, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 3, 2, 1] 391s 0.128 0.813 0.640 -0.413 391s -0.222 0.654 0.661 -0.298 391s -0.381 0.741 0.288 0.042 391s 391s x1: [4, 3, 2, 1] 391s 0.827 -0.993 0.547 -0.053 391s 0.040 0.754 -0.933 0.483 391s 0.492 -0.461 -0.678 -0.952 391s 391s y: [3, 3, 2, 1] 391s -0.329 -0.456 -0.896 391s -0.179 -0.277 0.295 391s -0.352 -0.576 -0.764 391s testing: mul_mat, [2, 4, 2, 4] = [3, 2, 2, 4] * [3, 4, 2, 4] 391s x0: [3, 2, 2, 4] 391s -0.847 -0.779 0.809 391s -0.197 -0.705 -0.960 391s 391s x1: [3, 4, 2, 4] 391s 0.599 0.266 0.341 391s -0.577 -0.696 -0.362 391s 0.034 0.432 0.416 391s 0.653 0.244 0.070 391s 391s y: [2, 4, 2, 4] 391s -0.439 -0.633 391s 0.738 0.953 391s -0.029 -0.711 391s -0.687 -0.369 391s test-mul-mat0: iter:456/500 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.097 391s 0.748 391s 0.942 391s 0.477 391s 391s x1: [1, 2, 1, 1] 391s -0.514 391s -0.137 391s 391s y: [4, 2, 1, 1] 391s -0.050 -0.384 -0.484 -0.245 391s -0.013 -0.102 -0.129 -0.065 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.512 391s 391s x1: [1, 4, 1, 1] 391s 0.411 391s -0.895 391s 0.872 391s 0.227 391s 391s y: [1, 4, 1, 1] 391s -0.211 391s 0.458 391s -0.447 391s -0.116 391s testing: mul_mat, [1, 1, 1, 4] = [1, 1, 1, 4] * [1, 1, 1, 4] 391s x0: [1, 1, 1, 4] 391s -0.336 391s 391s x1: [1, 1, 1, 4] 391s 0.338 391s 391s y: [1, 1, 1, 4] 391s -0.113 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.772 391s -0.291 391s 391s x1: [1, 1, 1, 1] 391s -0.080 391s 391s y: [2, 1, 1, 1] 391s -0.061 0.023 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.868 0.006 391s 391s x1: [2, 1, 1, 1] 391s 0.150 -0.118 391s 391s y: [1, 1, 1, 1] 391s 0.130 391s testing: mul_mat, [1, 2, 1, 4] = [1, 1, 1, 4] * [1, 2, 1, 4] 391s x0: [1, 1, 1, 4] 391s 0.964 391s 391s x1: [1, 2, 1, 4] 391s 0.517 391s -0.268 391s 391s y: [1, 2, 1, 4] 391s 0.499 391s -0.259 391s test-mul-mat0: iter:457/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.521 -0.035 0.799 391s 0.485 -0.518 -0.469 391s -0.226 0.097 0.010 391s 391s x1: [3, 3, 1, 1] 391s 0.467 -0.740 -0.695 391s 0.934 0.825 0.225 391s -0.219 0.596 0.934 391s 391s y: [3, 3, 1, 1] 391s -0.773 0.936 -0.184 391s -0.336 -0.080 -0.129 391s 0.839 -0.854 0.117 391s testing: mul_mat, [4, 3, 4, 1] = [3, 4, 4, 1] * [3, 3, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.286 -0.220 -0.148 391s -0.284 0.979 0.681 391s -0.815 0.439 -0.186 391s 0.588 -0.278 0.173 391s 391s x1: [3, 3, 4, 1] 391s -0.994 0.188 -0.098 391s 0.342 0.152 -0.980 391s 0.749 0.604 -0.524 391s 391s y: [4, 3, 4, 1] 391s 0.257 0.399 0.911 -0.654 391s 0.014 -0.615 -0.030 -0.010 391s -0.270 0.023 -0.248 0.182 391s testing: mul_mat, [3, 4, 4, 2] = [3, 3, 4, 2] * [3, 4, 4, 2] 391s x0: [3, 3, 4, 2] 391s 0.719 0.626 -0.584 391s 0.461 -0.189 -0.826 391s 0.195 0.822 -0.045 391s 391s x1: [3, 4, 4, 2] 391s 0.136 -0.328 0.782 391s 0.596 -0.084 0.204 391s -0.109 -0.386 -0.751 391s -0.790 -0.332 -0.741 391s 391s y: [3, 4, 4, 2] 391s -0.564 -0.521 -0.278 391s 0.257 0.122 0.038 391s 0.119 0.644 -0.305 391s -0.342 0.311 -0.393 391s testing: mul_mat, [4, 3, 1, 1] = [3, 4, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.988 0.155 -0.380 391s 0.554 -0.989 -0.396 391s 0.875 -0.433 0.396 391s -0.628 0.443 0.420 391s 391s x1: [3, 3, 1, 1] 391s 0.628 0.442 0.344 391s -0.934 -0.856 0.939 391s 0.535 0.777 -0.006 391s 391s y: [4, 3, 1, 1] 391s -0.683 -0.225 0.494 -0.054 391s 0.433 -0.042 -0.075 0.602 391s -0.406 -0.470 0.130 0.006 391s testing: mul_mat, [4, 3, 4, 1] = [4, 4, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 4, 4, 1] 391s 0.709 0.088 -0.566 0.576 391s -0.870 -0.337 0.732 -0.019 391s -0.664 0.307 -0.257 -0.804 391s 0.503 -0.387 0.601 0.295 391s 391s x1: [4, 3, 4, 1] 391s -0.891 0.429 0.874 0.893 391s 0.715 0.175 0.777 0.211 391s 0.341 0.405 -0.347 -0.315 391s 391s y: [4, 3, 4, 1] 391s -0.574 1.254 -0.219 0.174 391s 0.204 -0.117 -0.791 0.821 391s 0.292 -0.682 0.240 -0.287 391s testing: mul_mat, [1, 4, 4, 2] = [4, 1, 4, 2] * [4, 4, 4, 2] 391s x0: [4, 1, 4, 2] 391s -0.416 0.617 -0.610 0.259 391s 391s x1: [4, 4, 4, 2] 391s 0.307 0.493 -0.310 -0.643 391s -0.082 0.238 0.840 -0.926 391s -0.471 -0.076 -0.193 -0.200 391s -0.796 -0.165 -0.815 0.991 391s 391s y: [1, 4, 4, 2] 391s 0.199 391s -0.571 391s 0.215 391s 0.984 391s test-mul-mat0: iter:458/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.171 -0.474 391s 391s x1: [2, 4, 1, 1] 391s 0.169 -0.888 391s 0.273 -0.242 391s 0.021 -0.333 391s 0.243 0.738 391s 391s y: [1, 4, 1, 1] 391s 0.450 391s 0.161 391s 0.162 391s -0.309 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.893 0.383 391s -0.600 0.221 391s 0.160 -0.749 391s 391s x1: [2, 1, 1, 1] 391s -0.734 -0.053 391s 391s y: [3, 1, 1, 1] 391s 0.636 0.429 -0.078 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.648 0.663 391s -0.951 0.585 391s 391s x1: [2, 3, 1, 1] 391s 0.798 0.604 391s 0.826 0.740 391s 0.817 0.550 391s 391s y: [2, 3, 1, 1] 391s 0.918 -0.405 391s 1.026 -0.352 391s 0.894 -0.455 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.525 -0.627 391s -0.167 -0.996 391s -0.080 0.446 391s 391s x1: [2, 2, 1, 1] 391s -0.168 -0.839 391s -0.142 0.590 391s 391s y: [3, 2, 1, 1] 391s 0.438 0.864 -0.361 391s -0.445 -0.564 0.275 391s testing: mul_mat, [3, 2, 1, 1] = [3, 3, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.494 0.098 -0.085 391s 0.962 0.788 0.338 391s 0.764 0.504 0.769 391s 391s x1: [3, 2, 1, 1] 391s -0.361 -0.049 0.311 391s -0.255 -0.666 0.711 391s 391s y: [3, 2, 1, 1] 391s -0.210 -0.281 -0.062 391s -0.251 -0.529 0.017 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.743 -0.739 0.212 391s -0.918 0.527 -0.161 391s -0.096 0.543 -0.712 391s 391s x1: [3, 3, 1, 1] 391s -0.437 0.001 0.818 391s -0.852 0.834 0.979 391s 0.006 0.424 0.161 391s 391s y: [3, 3, 1, 1] 391s 0.497 0.270 -0.540 391s 0.225 1.064 -0.163 391s -0.283 0.192 0.115 391s test-mul-mat0: iter:459/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.088 -0.229 0.897 -0.751 391s 0.302 -0.846 -0.670 -0.794 391s -0.584 0.858 0.748 0.628 391s 391s x1: [4, 3, 1, 1] 391s -0.488 -0.982 -0.861 -0.700 391s 0.522 0.054 0.638 0.291 391s 0.617 -0.361 0.109 0.765 391s 391s y: [3, 3, 1, 1] 391s -0.064 1.816 -1.641 391s 0.387 -0.546 0.401 391s -0.341 -0.189 -0.109 391s testing: mul_mat, [1, 3, 2, 1] = [4, 1, 2, 1] * [4, 3, 2, 1] 391s x0: [4, 1, 2, 1] 391s 0.897 0.069 0.273 0.754 391s 391s x1: [4, 3, 2, 1] 391s -0.304 -0.964 0.175 0.246 391s -0.709 -0.785 0.758 -0.691 391s -0.646 -0.942 0.831 0.408 391s 391s y: [1, 3, 2, 1] 391s -0.105 391s -1.005 391s -0.109 391s testing: mul_mat, [3, 1, 2, 3] = [4, 3, 2, 3] * [4, 1, 2, 3] 391s x0: [4, 3, 2, 3] 391s 0.185 0.170 -0.819 -0.914 391s 0.621 0.035 -0.838 0.318 391s -0.820 -0.547 -0.227 0.634 391s 391s x1: [4, 1, 2, 3] 391s -0.924 0.697 0.145 -0.709 391s 391s y: [3, 1, 2, 3] 391s 0.477 -0.895 -0.106 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.716 0.583 0.292 -0.954 391s 391s x1: [4, 3, 1, 1] 391s -0.059 0.937 0.217 -0.047 391s 0.074 0.703 -0.541 0.534 391s -0.580 0.000 0.953 0.236 391s 391s y: [1, 3, 1, 1] 391s 0.696 391s -0.310 391s 0.468 391s testing: mul_mat, [4, 4, 2, 1] = [1, 4, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.023 391s 0.769 391s -0.248 391s -0.305 391s 391s x1: [1, 4, 2, 1] 391s 0.755 391s -0.995 391s -0.346 391s 0.115 391s 391s y: [4, 4, 2, 1] 391s -0.017 0.581 -0.187 -0.230 391s 0.023 -0.765 0.247 0.303 391s 0.008 -0.266 0.086 0.106 391s -0.003 0.088 -0.028 -0.035 391s testing: mul_mat, [1, 1, 2, 3] = [4, 1, 2, 3] * [4, 1, 2, 3] 391s x0: [4, 1, 2, 3] 391s 0.745 -0.802 -0.874 0.439 391s 391s x1: [4, 1, 2, 3] 391s -0.353 0.706 -0.427 0.106 391s 391s y: [1, 1, 2, 3] 391s -0.411 391s test-mul-mat0: iter:460/500 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.655 391s 0.078 391s -0.130 391s 0.784 391s 391s x1: [1, 2, 1, 1] 391s -0.400 391s -0.119 391s 391s y: [4, 2, 1, 1] 391s 0.262 -0.031 0.052 -0.313 391s 0.078 -0.009 0.015 -0.093 391s testing: mul_mat, [1, 4, 1, 1] = [1, 1, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.061 391s 391s x1: [1, 4, 1, 1] 391s 0.909 391s 0.866 391s 0.870 391s -0.555 391s 391s y: [1, 4, 1, 1] 391s 0.055 391s 0.053 391s 0.053 391s -0.034 391s testing: mul_mat, [3, 1, 1, 4] = [1, 3, 1, 4] * [1, 1, 1, 4] 391s x0: [1, 3, 1, 4] 391s -0.311 391s 0.641 391s 0.394 391s 391s x1: [1, 1, 1, 4] 391s -0.778 391s 391s y: [3, 1, 1, 4] 391s 0.242 -0.499 -0.307 391s testing: mul_mat, [2, 3, 1, 1] = [1, 2, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.681 391s 0.757 391s 391s x1: [1, 3, 1, 1] 391s -0.830 391s 0.974 391s -0.321 391s 391s y: [2, 3, 1, 1] 391s 0.565 -0.628 391s -0.663 0.737 391s 0.219 -0.243 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.651 0.034 391s 0.973 -0.668 391s -0.889 0.086 391s 0.087 0.392 391s 391s x1: [2, 1, 1, 1] 391s 0.785 -0.897 391s 391s y: [4, 1, 1, 1] 391s 0.481 1.363 -0.775 -0.283 391s testing: mul_mat, [2, 2, 1, 4] = [4, 2, 1, 4] * [4, 2, 1, 4] 391s x0: [4, 2, 1, 4] 391s -0.201 0.495 0.070 0.979 391s -0.711 -0.776 -0.668 0.617 391s 391s x1: [4, 2, 1, 4] 391s 0.525 0.876 -0.919 0.166 391s 0.271 0.425 -0.219 -0.163 391s 391s y: [2, 2, 1, 4] 391s 0.427 -0.337 391s -0.019 -0.477 391s test-mul-mat0: iter:461/500 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.520 -0.219 -0.610 -0.045 391s 0.156 -0.416 -0.397 0.019 391s 0.300 -0.430 -0.962 -0.809 391s -0.171 -0.676 -0.657 -0.588 391s 391s x1: [4, 1, 1, 1] 391s -0.448 0.303 0.982 -0.860 391s 391s y: [4, 1, 1, 1] 391s -0.859 -0.602 -0.512 -0.267 391s testing: mul_mat, [1, 4, 3, 1] = [4, 1, 3, 1] * [4, 4, 3, 1] 391s x0: [4, 1, 3, 1] 391s 0.151 -0.120 0.325 0.082 391s 391s x1: [4, 4, 3, 1] 391s -0.793 -0.156 -0.287 0.213 391s 0.569 0.812 -0.493 0.788 391s 0.136 -0.129 -0.660 -0.562 391s -0.147 -0.520 0.373 -0.627 391s 391s y: [1, 4, 3, 1] 391s -0.177 391s -0.108 391s -0.225 391s 0.110 391s testing: mul_mat, [1, 1, 3, 2] = [4, 1, 3, 2] * [4, 1, 3, 2] 391s x0: [4, 1, 3, 2] 391s -0.779 -0.639 0.925 0.135 391s 391s x1: [4, 1, 3, 2] 391s -0.344 0.594 -0.986 0.112 391s 391s y: [1, 1, 3, 2] 391s -1.010 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.337 0.613 0.429 0.698 391s 391s x1: [4, 1, 1, 1] 391s -0.648 -0.430 -0.232 -0.503 391s 391s y: [1, 1, 1, 1] 391s -0.933 391s testing: mul_mat, [3, 4, 3, 1] = [1, 3, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 3, 3, 1] 391s -0.392 391s -0.530 391s 0.667 391s 391s x1: [1, 4, 3, 1] 391s 0.538 391s -0.436 391s 0.648 391s -0.547 391s 391s y: [3, 4, 3, 1] 391s -0.211 -0.285 0.359 391s 0.171 0.231 -0.291 391s -0.254 -0.343 0.432 391s 0.214 0.290 -0.365 391s testing: mul_mat, [1, 1, 3, 2] = [3, 1, 3, 2] * [3, 1, 3, 2] 391s x0: [3, 1, 3, 2] 391s 0.023 0.959 -0.982 391s 391s x1: [3, 1, 3, 2] 391s -0.814 -0.335 -0.910 391s 391s y: [1, 1, 3, 2] 391s 0.554 391s test-mul-mat0: iter:462/500 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.284 0.294 -0.963 391s -0.325 0.312 0.668 391s 0.105 -0.003 0.599 391s -0.385 0.490 0.364 391s 391s x1: [3, 2, 1, 1] 391s 0.311 0.233 0.046 391s 0.533 0.200 -0.287 391s 391s y: [4, 2, 1, 1] 391s -0.065 0.003 0.060 0.011 391s 0.184 -0.303 -0.117 -0.212 391s testing: mul_mat, [4, 4, 2, 1] = [3, 4, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 4, 2, 1] 391s 0.933 -0.002 0.239 391s -0.452 -0.513 -0.397 391s -0.233 0.792 0.918 391s 0.256 0.003 0.865 391s 391s x1: [3, 4, 2, 1] 391s -0.781 0.305 0.315 391s -0.511 0.211 0.947 391s -0.368 -0.456 -0.125 391s -0.923 0.003 0.636 391s 391s y: [4, 4, 2, 1] 391s -0.654 0.072 0.712 0.073 391s -0.251 -0.253 1.155 0.689 391s -0.372 0.450 -0.390 -0.204 391s -0.710 0.163 0.801 0.314 391s testing: mul_mat, [2, 4, 2, 3] = [3, 2, 2, 3] * [3, 4, 2, 3] 391s x0: [3, 2, 2, 3] 391s -0.245 0.269 -0.100 391s -0.698 -0.042 0.997 391s 391s x1: [3, 4, 2, 3] 391s -0.792 -0.623 0.143 391s -0.820 0.205 0.144 391s 0.819 0.138 -0.859 391s 0.058 0.685 -0.372 391s 391s y: [2, 4, 2, 3] 391s 0.012 0.721 391s 0.242 0.707 391s -0.078 -1.434 391s 0.208 -0.440 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.921 -0.622 0.996 391s 391s x1: [3, 2, 1, 1] 391s -0.739 -0.808 0.695 391s -0.050 0.405 0.100 391s 391s y: [1, 2, 1, 1] 391s 1.876 391s -0.106 391s testing: mul_mat, [4, 3, 2, 1] = [1, 4, 2, 1] * [1, 3, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.419 391s -0.843 391s -0.499 391s -0.183 391s 391s x1: [1, 3, 2, 1] 391s -0.478 391s 0.147 391s -0.551 391s 391s y: [4, 3, 2, 1] 391s 0.200 0.403 0.239 0.088 391s -0.061 -0.124 -0.073 -0.027 391s 0.231 0.464 0.275 0.101 391s testing: mul_mat, [2, 1, 2, 3] = [4, 2, 2, 3] * [4, 1, 2, 3] 391s x0: [4, 2, 2, 3] 391s 0.408 -0.293 0.885 0.479 391s 0.204 0.762 -0.537 0.184 391s 391s x1: [4, 1, 2, 3] 391s 0.717 0.300 -0.597 0.981 391s 391s y: [2, 1, 2, 3] 391s 0.147 0.876 391s test-mul-mat0: iter:463/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.058 0.873 0.376 0.825 391s 0.871 0.729 -0.590 -0.550 391s -0.024 -0.370 -0.587 -0.802 391s -0.621 -0.056 -0.524 -0.922 391s 391s x1: [4, 4, 1, 1] 391s -0.723 0.698 0.428 -0.140 391s 0.332 -0.869 0.615 0.874 391s 0.221 -0.299 0.824 0.515 391s -0.506 0.535 0.694 -0.250 391s 391s y: [4, 4, 1, 1] 391s 0.613 -0.295 -0.379 0.315 391s 0.213 -1.188 -0.748 -1.286 391s 0.487 -0.795 -0.791 -1.028 391s 0.493 -0.322 -0.393 0.151 391s testing: mul_mat, [1, 4, 1, 1] = [4, 1, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.255 0.816 -0.360 -0.526 391s 391s x1: [4, 4, 1, 1] 391s -0.628 0.336 0.410 -0.497 391s -0.049 0.284 0.723 0.652 391s 0.108 0.239 -0.854 -0.357 391s -0.067 -0.104 0.369 0.990 391s 391s y: [1, 4, 1, 1] 391s 0.548 391s -0.359 391s 0.662 391s -0.721 391s testing: mul_mat, [1, 1, 1, 2] = [4, 1, 1, 2] * [4, 1, 1, 2] 391s x0: [4, 1, 1, 2] 391s 0.313 -0.180 0.460 -0.277 391s 391s x1: [4, 1, 1, 2] 391s -0.774 0.090 0.450 -0.144 391s 391s y: [1, 1, 1, 2] 391s -0.012 391s testing: mul_mat, [3, 1, 1, 1] = [4, 3, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.217 0.208 0.024 0.250 391s 0.103 0.872 -0.487 0.603 391s 0.127 0.872 -0.654 0.796 391s 391s x1: [4, 1, 1, 1] 391s -0.938 0.115 -0.715 -0.792 391s 391s y: [3, 1, 1, 1] 391s -0.396 -0.126 -0.182 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.258 0.500 -0.500 391s 391s x1: [3, 4, 1, 1] 391s -0.894 0.106 0.444 391s 0.342 0.554 0.568 391s -0.345 -0.626 0.028 391s 0.378 0.696 0.438 391s 391s y: [1, 4, 1, 1] 391s 0.061 391s -0.096 391s -0.238 391s 0.031 391s testing: mul_mat, [1, 3, 1, 2] = [1, 1, 1, 2] * [1, 3, 1, 2] 391s x0: [1, 1, 1, 2] 391s -0.316 391s 391s x1: [1, 3, 1, 2] 391s 0.026 391s 0.708 391s -0.742 391s 391s y: [1, 3, 1, 2] 391s -0.008 391s -0.224 391s 0.235 391s test-mul-mat0: iter:464/500 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.174 0.459 -0.588 0.829 391s 0.832 0.440 0.207 0.528 391s -0.122 -0.408 -0.730 -0.622 391s 0.092 0.297 -0.914 0.350 391s 391s x1: [4, 1, 1, 1] 391s -0.201 0.399 0.832 0.905 391s 391s y: [4, 1, 1, 1] 391s 0.409 0.658 -1.308 -0.345 391s testing: mul_mat, [3, 4, 4, 1] = [4, 3, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 3, 4, 1] 391s 0.590 0.321 -0.422 0.457 391s 0.605 0.166 -0.482 0.465 391s -0.254 -0.394 -0.196 -0.839 391s 391s x1: [4, 4, 4, 1] 391s 0.541 -0.103 -0.266 -0.008 391s 0.581 0.990 0.468 -0.223 391s 0.593 0.194 0.575 -0.008 391s 0.027 0.480 0.835 -0.799 391s 391s y: [3, 4, 4, 1] 391s 0.395 0.435 -0.038 391s 0.361 0.186 -0.443 391s 0.165 0.110 -0.333 391s -0.548 -0.678 0.311 391s testing: mul_mat, [4, 3, 4, 3] = [4, 4, 4, 3] * [4, 3, 4, 3] 391s x0: [4, 4, 4, 3] 391s 0.499 -0.984 -0.091 0.844 391s -0.704 0.607 0.296 -0.205 391s -0.969 0.373 0.659 -0.029 391s -0.689 0.400 -0.357 0.993 391s 391s x1: [4, 3, 4, 3] 391s -0.731 0.418 0.536 0.356 391s 0.573 -0.066 0.301 -0.966 391s -0.885 0.134 0.073 -0.512 391s 391s y: [4, 3, 4, 3] 391s -0.525 0.854 1.208 0.834 391s -0.492 -0.156 -0.353 -1.489 391s -1.012 0.831 0.971 0.129 391s testing: mul_mat, [3, 4, 1, 1] = [4, 3, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 3, 1, 1] 391s 0.202 -0.977 0.556 0.112 391s -0.457 0.235 0.313 -0.505 391s -0.043 0.941 0.355 0.017 391s 391s x1: [4, 4, 1, 1] 391s 0.062 0.563 0.876 -0.882 391s -0.232 -0.788 0.088 0.130 391s 0.598 -0.446 -0.623 -0.676 391s -0.932 0.139 -0.965 0.237 391s 391s y: [3, 4, 1, 1] 391s -0.150 0.824 0.823 391s 0.786 -0.117 -0.698 391s 0.134 -0.231 -0.678 391s -0.833 0.037 -0.168 391s testing: mul_mat, [2, 4, 4, 1] = [3, 2, 4, 1] * [3, 4, 4, 1] 391s x0: [3, 2, 4, 1] 391s 0.207 0.589 0.229 391s -0.826 -0.657 -0.489 391s 391s x1: [3, 4, 4, 1] 391s -0.811 0.784 -0.921 391s 0.752 0.660 -0.803 391s -0.480 0.873 0.285 391s 0.650 0.470 0.839 391s 391s y: [2, 4, 4, 1] 391s 0.083 0.605 391s 0.360 -0.663 391s 0.480 -0.317 391s 0.603 -1.257 391s testing: mul_mat, [3, 3, 4, 3] = [2, 3, 4, 3] * [2, 3, 4, 3] 391s x0: [2, 3, 4, 3] 391s 0.952 0.990 391s 0.896 0.442 391s 0.796 0.700 391s 391s x1: [2, 3, 4, 3] 391s -0.584 0.022 391s 0.035 0.173 391s -0.950 -0.823 391s 391s y: [3, 3, 4, 3] 391s -0.533 -0.513 -0.449 391s 0.205 0.108 0.149 391s -1.719 -1.215 -1.332 391s test-mul-mat0: iter:465/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.301 391s -0.189 391s 391s x1: [1, 4, 1, 1] 391s 0.075 391s -0.255 391s 0.534 391s -0.344 391s 391s y: [2, 4, 1, 1] 391s -0.023 -0.014 391s 0.077 0.048 391s -0.161 -0.101 391s 0.104 0.065 391s testing: mul_mat, [2, 2, 4, 1] = [1, 2, 4, 1] * [1, 2, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.705 391s -0.986 391s 391s x1: [1, 2, 4, 1] 391s 0.022 391s -0.242 391s 391s y: [2, 2, 4, 1] 391s 0.016 -0.022 391s -0.171 0.239 391s testing: mul_mat, [4, 2, 4, 2] = [1, 4, 4, 2] * [1, 2, 4, 2] 391s x0: [1, 4, 4, 2] 391s 0.671 391s -0.044 391s 0.044 391s -0.948 391s 391s x1: [1, 2, 4, 2] 391s -0.665 391s 0.441 391s 391s y: [4, 2, 4, 2] 391s -0.446 0.029 -0.029 0.631 391s 0.296 -0.019 0.019 -0.418 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.831 391s 0.797 391s 391s x1: [1, 4, 1, 1] 391s -0.674 391s 0.713 391s -0.926 391s 0.055 391s 391s y: [2, 4, 1, 1] 391s -0.560 -0.537 391s 0.593 0.568 391s -0.770 -0.738 391s 0.046 0.044 391s testing: mul_mat, [3, 1, 4, 1] = [2, 3, 4, 1] * [2, 1, 4, 1] 391s x0: [2, 3, 4, 1] 391s 0.355 -0.916 391s -0.523 -0.838 391s 0.077 -0.408 391s 391s x1: [2, 1, 4, 1] 391s 0.997 0.099 391s 391s y: [3, 1, 4, 1] 391s 0.262 -0.604 0.036 391s testing: mul_mat, [3, 2, 4, 2] = [3, 3, 4, 2] * [3, 2, 4, 2] 391s x0: [3, 3, 4, 2] 391s -0.789 -0.210 -0.170 391s -0.042 -0.472 0.577 391s 0.386 0.599 0.677 391s 391s x1: [3, 2, 4, 2] 391s -0.466 0.987 -0.744 391s -0.853 -0.426 -0.675 391s 391s y: [3, 2, 4, 2] 391s 0.287 -0.876 -0.093 391s 0.878 -0.152 -1.041 391s test-mul-mat0: iter:466/500 391s testing: mul_mat, [2, 3, 1, 1] = [3, 2, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.938 0.814 0.353 391s 0.268 0.022 0.658 391s 391s x1: [3, 3, 1, 1] 391s 0.484 -0.099 -0.021 391s 0.212 0.321 0.896 391s 0.418 0.018 0.111 391s 391s y: [2, 3, 1, 1] 391s -0.542 0.113 391s 0.378 0.653 391s -0.339 0.185 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.856 -0.546 0.450 391s -0.590 0.214 0.252 391s 0.060 0.198 -0.264 391s 0.961 -0.823 0.948 391s 391s x1: [3, 2, 1, 1] 391s -0.670 -0.054 -0.584 391s -0.364 -0.623 -0.844 391s 391s y: [4, 2, 1, 1] 391s 0.340 0.236 0.103 -1.153 391s 0.271 -0.132 0.077 -0.637 391s testing: mul_mat, [2, 4, 1, 2] = [3, 2, 1, 2] * [3, 4, 1, 2] 391s x0: [3, 2, 1, 2] 391s -0.044 -0.640 0.819 391s -0.083 -0.463 0.768 391s 391s x1: [3, 4, 1, 2] 391s 0.282 -0.927 0.366 391s -0.700 0.184 0.910 391s -0.638 -0.003 0.263 391s 0.630 -0.981 -0.079 391s 391s y: [2, 4, 1, 2] 391s 0.881 0.687 391s 0.659 0.672 391s 0.246 0.256 391s 0.536 0.342 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.774 -0.939 0.260 391s -0.433 0.429 -0.503 391s 0.049 -0.398 0.171 391s 0.116 -0.991 -0.657 391s 391s x1: [3, 2, 1, 1] 391s -0.139 -0.210 -0.692 391s -0.509 -0.191 0.229 391s 391s y: [4, 2, 1, 1] 391s -0.091 0.318 -0.042 0.646 391s -0.155 0.023 0.090 -0.020 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.262 0.646 0.923 0.563 391s -0.583 -0.125 -0.038 -0.718 391s -0.577 0.846 -0.727 -0.387 391s -0.929 -0.788 0.353 -0.939 391s 391s x1: [4, 3, 1, 1] 391s 0.454 0.531 -0.837 -0.629 391s -0.932 0.931 -0.429 -0.322 391s 0.065 0.070 -0.528 -0.891 391s 391s y: [4, 3, 1, 1] 391s -0.666 0.153 1.040 -0.544 391s -0.221 0.675 1.761 0.283 391s -0.928 0.614 0.751 0.535 391s testing: mul_mat, [1, 4, 1, 2] = [4, 1, 1, 2] * [4, 4, 1, 2] 391s x0: [4, 1, 1, 2] 391s 0.481 0.488 -0.280 0.367 391s 391s x1: [4, 4, 1, 2] 391s 0.452 -0.044 0.515 -0.017 391s 0.119 0.886 0.052 0.050 391s -0.544 0.730 -0.886 0.526 391s -0.799 -0.776 0.458 0.464 391s 391s y: [1, 4, 1, 2] 391s 0.046 391s 0.493 391s 0.536 391s -0.722 391s test-mul-mat0: iter:467/500 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.192 -0.698 -0.845 391s 0.098 0.354 -0.201 391s 0.065 0.859 -0.601 391s 391s x1: [3, 3, 1, 1] 391s -0.193 0.935 -0.253 391s -0.312 0.470 0.033 391s 0.444 -0.150 0.531 391s 391s y: [3, 3, 1, 1] 391s -0.476 0.363 0.943 391s -0.416 0.129 0.364 391s -0.259 -0.116 -0.419 391s testing: mul_mat, [2, 3, 3, 1] = [3, 2, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 2, 3, 1] 391s -0.761 0.707 0.499 391s -0.216 0.054 -0.381 391s 391s x1: [3, 3, 3, 1] 391s -0.454 0.347 0.119 391s 0.912 0.887 -0.112 391s 0.735 -0.843 -0.228 391s 391s y: [2, 3, 3, 1] 391s 0.651 0.072 391s -0.123 -0.106 391s -1.269 -0.117 391s testing: mul_mat, [1, 2, 3, 2] = [3, 1, 3, 2] * [3, 2, 3, 2] 391s x0: [3, 1, 3, 2] 391s 0.683 0.614 -0.386 391s 391s x1: [3, 2, 3, 2] 391s 0.986 -0.676 0.445 391s -0.990 0.208 0.636 391s 391s y: [1, 2, 3, 2] 391s 0.087 391s -0.794 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.311 0.896 -0.573 391s -0.225 0.942 0.417 391s 0.093 0.734 0.840 391s -0.793 -0.774 0.042 391s 391s x1: [3, 1, 1, 1] 391s 0.070 0.158 0.977 391s 391s y: [4, 1, 1, 1] 391s -0.440 0.541 0.943 -0.137 391s testing: mul_mat, [3, 3, 3, 1] = [4, 3, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.591 -0.156 -0.112 -0.998 391s 0.314 -0.333 0.054 -0.401 391s 0.767 0.379 -0.538 0.522 391s 391s x1: [4, 3, 3, 1] 391s 0.557 0.911 0.633 0.254 391s -0.321 0.907 -0.993 -0.073 391s -0.080 -0.566 0.353 0.612 391s 391s y: [3, 3, 3, 1] 391s -0.795 -0.195 0.564 391s 0.232 -0.428 0.594 391s -0.515 -0.063 -0.146 391s testing: mul_mat, [2, 4, 3, 2] = [3, 2, 3, 2] * [3, 4, 3, 2] 391s x0: [3, 2, 3, 2] 391s -0.283 0.671 -0.769 391s 0.966 0.509 -0.888 391s 391s x1: [3, 4, 3, 2] 391s 0.043 -0.265 0.974 391s 0.505 -0.263 -0.427 391s 0.028 0.542 0.407 391s -0.615 -0.757 -0.318 391s 391s y: [2, 4, 3, 2] 391s -0.939 -0.957 391s 0.009 0.733 391s 0.043 -0.059 391s -0.090 -0.697 391s test-mul-mat0: iter:468/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.588 391s 0.451 391s 391s x1: [1, 4, 1, 1] 391s -0.036 391s 0.402 391s 0.738 391s 0.075 391s 391s y: [2, 4, 1, 1] 391s 0.021 -0.016 391s -0.236 0.181 391s -0.434 0.333 391s -0.044 0.034 391s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.009 391s 0.583 391s -0.568 391s 0.992 391s 391s x1: [1, 2, 2, 1] 391s -0.526 391s -0.068 391s 391s y: [4, 2, 2, 1] 391s 0.005 -0.307 0.299 -0.522 391s 0.001 -0.040 0.039 -0.068 391s testing: mul_mat, [4, 4, 2, 2] = [1, 4, 2, 2] * [1, 4, 2, 2] 391s x0: [1, 4, 2, 2] 391s 0.912 391s 0.235 391s -0.434 391s -0.182 391s 391s x1: [1, 4, 2, 2] 391s -0.893 391s -0.698 391s 0.593 391s -0.979 391s 391s y: [4, 4, 2, 2] 391s -0.814 -0.210 0.388 0.163 391s -0.637 -0.164 0.303 0.127 391s 0.541 0.140 -0.258 -0.108 391s -0.893 -0.230 0.425 0.178 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.262 391s -0.223 391s 391s x1: [1, 4, 1, 1] 391s -0.914 391s 0.936 391s -0.622 391s -0.644 391s 391s y: [2, 4, 1, 1] 391s -0.240 0.204 391s 0.246 -0.209 391s -0.163 0.139 391s -0.169 0.144 391s testing: mul_mat, [4, 1, 2, 1] = [2, 4, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 4, 2, 1] 391s -0.376 0.839 391s -0.396 -0.834 391s 0.601 0.354 391s -0.464 -0.654 391s 391s x1: [2, 1, 2, 1] 391s -0.729 -0.254 391s 391s y: [4, 1, 2, 1] 391s 0.060 0.501 -0.528 0.504 391s testing: mul_mat, [2, 2, 2, 2] = [4, 2, 2, 2] * [4, 2, 2, 2] 391s x0: [4, 2, 2, 2] 391s 0.865 0.172 -0.844 -0.432 391s -0.282 -0.415 0.967 -0.628 391s 391s x1: [4, 2, 2, 2] 391s -0.003 -0.243 -0.657 0.083 391s -0.307 -0.279 0.439 -0.204 391s 391s y: [2, 2, 2, 2] 391s 0.474 -0.586 391s -0.595 0.755 391s test-mul-mat0: iter:469/500 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.868 -0.855 -0.441 391s 391s x1: [3, 4, 1, 1] 391s -0.185 0.803 -0.841 391s 0.850 0.100 -0.090 391s 0.793 0.092 -0.336 391s 0.248 0.381 0.958 391s 391s y: [1, 4, 1, 1] 391s -0.155 391s -0.783 391s -0.619 391s -0.963 391s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 1, 4, 1] 391s -0.678 -0.196 -0.254 391s 391s x1: [3, 1, 4, 1] 391s -0.270 -0.611 0.198 391s 391s y: [1, 1, 4, 1] 391s 0.253 391s testing: mul_mat, [1, 1, 4, 1] = [3, 1, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 1, 4, 1] 391s 0.117 0.311 0.257 391s 391s x1: [3, 1, 4, 1] 391s 0.698 -0.506 0.692 391s 391s y: [1, 1, 4, 1] 391s 0.102 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.499 -0.754 -0.718 391s -0.445 0.722 -0.337 391s 0.892 0.635 0.945 391s -0.005 -0.503 0.050 391s 391s x1: [3, 1, 1, 1] 391s 0.876 0.965 -0.082 391s 391s y: [4, 1, 1, 1] 391s -1.106 0.335 1.317 -0.494 391s testing: mul_mat, [1, 3, 4, 1] = [4, 1, 4, 1] * [4, 3, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.622 -0.886 -0.382 0.425 391s 391s x1: [4, 3, 4, 1] 391s 0.344 0.869 0.224 -0.539 391s 0.179 -0.520 -0.335 0.507 391s 0.668 0.190 -0.362 0.602 391s 391s y: [1, 3, 4, 1] 391s -0.871 391s 0.915 391s 0.640 391s testing: mul_mat, [2, 4, 4, 1] = [1, 2, 4, 1] * [1, 4, 4, 1] 391s x0: [1, 2, 4, 1] 391s 0.920 391s -0.376 391s 391s x1: [1, 4, 4, 1] 391s -0.239 391s -0.385 391s -0.739 391s 0.722 391s 391s y: [2, 4, 4, 1] 391s -0.220 0.090 391s -0.354 0.145 391s -0.679 0.278 391s 0.664 -0.272 391s test-mul-mat0: iter:470/500 391s testing: mul_mat, [3, 1, 1, 1] = [3, 3, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 3, 1, 1] 391s 0.910 -0.784 0.672 391s -0.599 -0.910 -0.377 391s -0.275 0.957 0.792 391s 391s x1: [3, 1, 1, 1] 391s 0.593 0.648 -0.506 391s 391s y: [3, 1, 1, 1] 391s -0.309 -0.754 0.056 391s testing: mul_mat, [3, 3, 3, 1] = [3, 3, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 3, 3, 1] 391s 0.898 -0.021 -0.962 391s -0.089 -0.487 0.915 391s 0.226 0.246 0.989 391s 391s x1: [3, 3, 3, 1] 391s 0.013 0.534 0.876 391s -0.685 -0.267 -0.926 391s -0.765 0.357 -0.421 391s 391s y: [3, 3, 3, 1] 391s -0.842 0.540 1.001 391s 0.281 -0.656 -1.136 391s -0.290 -0.490 -0.501 391s testing: mul_mat, [1, 3, 3, 2] = [3, 1, 3, 2] * [3, 3, 3, 2] 391s x0: [3, 1, 3, 2] 391s -0.234 -0.065 0.944 391s 391s x1: [3, 3, 3, 2] 391s 0.384 -0.186 0.171 391s 0.631 -0.288 -0.850 391s 0.669 0.623 -0.337 391s 391s y: [1, 3, 3, 2] 391s 0.084 391s -0.931 391s -0.515 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.721 -0.619 -0.997 391s -0.994 0.785 -0.600 391s -0.580 0.633 -0.935 391s 0.574 0.131 0.947 391s 391s x1: [3, 1, 1, 1] 391s 0.421 0.796 0.940 391s 391s y: [4, 1, 1, 1] 391s -1.126 -0.358 -0.619 1.236 391s testing: mul_mat, [4, 3, 3, 1] = [4, 4, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 4, 3, 1] 391s -0.942 -0.622 0.678 -0.804 391s -0.745 0.372 -0.836 -0.906 391s 0.333 -0.970 -0.557 -0.823 391s 0.316 -0.487 0.924 -0.352 391s 391s x1: [4, 3, 3, 1] 391s 0.090 -0.055 0.258 0.852 391s 0.155 -0.249 0.858 -0.675 391s 0.137 0.258 -0.877 -0.961 391s 391s y: [4, 3, 3, 1] 391s -0.561 -1.075 -0.763 -0.007 391s 1.134 -0.314 0.371 1.202 391s -0.111 1.597 1.075 -0.554 391s testing: mul_mat, [3, 4, 3, 2] = [4, 3, 3, 2] * [4, 4, 3, 2] 391s x0: [4, 3, 3, 2] 391s 0.593 0.244 -0.290 0.411 391s 0.090 -0.507 0.337 0.145 391s -0.303 -0.145 -0.412 -0.238 391s 391s x1: [4, 4, 3, 2] 391s -0.756 0.278 -0.715 -0.182 391s -0.598 0.719 0.221 0.713 391s 0.502 -0.602 -0.531 0.484 391s 0.505 0.420 0.524 -0.133 391s 391s y: [3, 4, 3, 2] 391s -0.248 -0.476 0.527 391s 0.050 -0.241 -0.183 391s 0.504 0.242 0.039 391s 0.196 -0.011 -0.399 391s test-mul-mat0: iter:471/500 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.072 -0.808 -0.172 391s 0.148 0.693 0.463 391s 0.256 0.236 -0.147 391s 0.584 -0.890 -0.359 391s 391s x1: [3, 1, 1, 1] 391s 0.467 -0.191 -0.882 391s 391s y: [4, 1, 1, 1] 391s 0.340 -0.472 0.204 0.760 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.677 0.073 0.721 391s 391s x1: [3, 4, 1, 1] 391s -0.260 0.907 0.916 391s 0.419 0.872 0.171 391s -0.523 -0.284 0.743 391s -0.499 0.848 -0.222 391s 391s y: [1, 4, 1, 1] 391s 0.903 391s -0.096 391s 0.870 391s 0.240 391s testing: mul_mat, [4, 1, 1, 1] = [3, 4, 1, 1] * [3, 1, 1, 1] 391s x0: [3, 4, 1, 1] 391s 0.383 -0.157 0.353 391s 0.076 -0.693 -0.391 391s -0.687 0.160 -0.807 391s -0.577 0.801 -0.067 391s 391s x1: [3, 1, 1, 1] 391s 0.132 0.191 -0.984 391s 391s y: [4, 1, 1, 1] 391s -0.327 0.262 0.734 0.143 391s testing: mul_mat, [2, 4, 1, 1] = [3, 2, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 2, 1, 1] 391s -0.518 -0.779 0.322 391s 0.306 0.674 -0.574 391s 391s x1: [3, 4, 1, 1] 391s -0.670 0.716 -0.648 391s -0.798 -0.113 -0.171 391s -0.082 -0.369 0.330 391s -0.234 0.409 0.585 391s 391s y: [2, 4, 1, 1] 391s -0.419 0.649 391s 0.446 -0.222 391s 0.436 -0.464 391s -0.009 -0.132 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.632 -0.132 391s -0.419 0.433 391s 391s x1: [2, 3, 1, 1] 391s 0.057 -0.834 391s 0.779 -0.867 391s -0.528 -0.612 391s 391s y: [2, 3, 1, 1] 391s 0.146 -0.385 391s 0.606 -0.701 391s -0.253 -0.044 391s testing: mul_mat, [4, 2, 1, 1] = [2, 4, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.964 -0.974 391s -0.305 0.085 391s -0.682 -0.924 391s -0.333 0.611 391s 391s x1: [2, 2, 1, 1] 391s 0.514 0.198 391s 0.149 0.866 391s 391s y: [4, 2, 1, 1] 391s -0.689 -0.140 -0.534 -0.050 391s -0.988 0.028 -0.902 0.479 391s test-mul-mat0: iter:472/500 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.123 -0.012 0.867 391s -0.490 0.568 -0.264 391s 0.942 0.083 0.934 391s 0.092 -0.051 0.334 391s 391s x1: [3, 2, 1, 1] 391s -0.769 -0.761 -0.179 391s 0.288 -0.595 -0.400 391s 391s y: [4, 2, 1, 1] 391s -0.051 -0.008 -0.954 -0.091 391s -0.375 -0.374 -0.151 -0.077 391s testing: mul_mat, [4, 4, 3, 1] = [3, 4, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 4, 3, 1] 391s -0.944 0.910 -0.457 391s -0.515 -0.892 0.851 391s 0.145 -0.455 0.979 391s 0.093 0.387 -0.292 391s 391s x1: [3, 4, 3, 1] 391s 0.128 0.644 0.653 391s 0.795 0.670 -0.263 391s 0.871 0.280 -0.088 391s -0.572 -0.802 0.307 391s 391s y: [4, 4, 3, 1] 391s 0.167 -0.085 0.365 0.070 391s -0.021 -1.230 -0.447 0.409 391s -0.526 -0.774 -0.088 0.215 391s -0.331 1.272 0.584 -0.453 391s testing: mul_mat, [2, 4, 3, 1] = [3, 2, 3, 1] * [3, 4, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.216 0.183 0.772 391s -0.930 0.470 0.144 391s 391s x1: [3, 4, 3, 1] 391s 0.742 0.031 0.545 391s 0.722 -0.876 -0.069 391s -0.570 -0.413 -0.963 391s -0.146 -0.713 0.372 391s 391s y: [2, 4, 3, 1] 391s 0.586 -0.598 391s -0.058 -1.093 391s -0.941 0.197 391s 0.125 -0.146 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s 0.234 -0.899 0.376 391s 391s x1: [3, 2, 1, 1] 391s 0.056 -0.152 -0.813 391s 0.314 -0.830 0.629 391s 391s y: [1, 2, 1, 1] 391s -0.156 391s 1.056 391s testing: mul_mat, [2, 3, 3, 1] = [1, 2, 3, 1] * [1, 3, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.218 391s -0.951 391s 391s x1: [1, 3, 3, 1] 391s -0.981 391s -0.303 391s 0.110 391s 391s y: [2, 3, 3, 1] 391s 0.214 0.933 391s 0.066 0.288 391s -0.024 -0.105 391s testing: mul_mat, [3, 1, 3, 1] = [2, 3, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 3, 3, 1] 391s 0.017 -0.355 391s -0.425 0.148 391s 0.530 -0.237 391s 391s x1: [2, 1, 3, 1] 391s 0.093 -0.802 391s 391s y: [3, 1, 3, 1] 391s 0.286 -0.158 0.239 391s test-mul-mat0: iter:473/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.647 0.028 391s 0.707 -0.002 391s 391s x1: [2, 4, 1, 1] 391s -0.691 -0.013 391s -0.695 -0.493 391s 0.192 -0.183 391s 0.336 -0.547 391s 391s y: [2, 4, 1, 1] 391s 0.447 -0.488 391s 0.436 -0.490 391s -0.129 0.136 391s -0.233 0.239 391s testing: mul_mat, [4, 2, 4, 1] = [2, 4, 4, 1] * [2, 2, 4, 1] 391s x0: [2, 4, 4, 1] 391s 0.858 -0.714 391s 0.309 -0.836 391s -0.208 -0.499 391s -0.019 -0.872 391s 391s x1: [2, 2, 4, 1] 391s -0.825 -0.530 391s 0.745 -0.302 391s 391s y: [4, 2, 4, 1] 391s -0.330 0.188 0.436 0.478 391s 0.855 0.482 -0.004 0.249 391s testing: mul_mat, [2, 4, 4, 2] = [2, 2, 4, 2] * [2, 4, 4, 2] 391s x0: [2, 2, 4, 2] 391s 0.172 0.725 391s -0.601 -0.665 391s 391s x1: [2, 4, 4, 2] 391s 0.681 0.932 391s -0.216 -0.527 391s -0.568 0.765 391s -0.398 0.385 391s 391s y: [2, 4, 4, 2] 391s 0.792 -1.029 391s -0.419 0.480 391s 0.457 -0.167 391s 0.211 -0.017 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.948 0.340 391s 0.892 0.933 391s 0.387 -0.216 391s 391s x1: [2, 2, 1, 1] 391s 0.015 -0.500 391s 0.367 0.427 391s 391s y: [3, 2, 1, 1] 391s -0.156 -0.453 0.114 391s 0.493 0.725 0.050 391s testing: mul_mat, [1, 2, 4, 1] = [3, 1, 4, 1] * [3, 2, 4, 1] 391s x0: [3, 1, 4, 1] 391s -0.782 0.805 0.659 391s 391s x1: [3, 2, 4, 1] 391s 0.328 0.925 -0.101 391s -0.884 0.429 0.299 391s 391s y: [1, 2, 4, 1] 391s 0.421 391s 1.234 391s testing: mul_mat, [4, 3, 4, 2] = [1, 4, 4, 2] * [1, 3, 4, 2] 391s x0: [1, 4, 4, 2] 391s 0.606 391s 0.213 391s -0.219 391s -0.392 391s 391s x1: [1, 3, 4, 2] 391s 0.834 391s 0.129 391s -0.360 391s 391s y: [4, 3, 4, 2] 391s 0.506 0.178 -0.183 -0.327 391s 0.078 0.028 -0.028 -0.051 391s -0.218 -0.077 0.079 0.141 391s test-mul-mat0: iter:474/500 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.433 0.874 0.734 0.194 391s -0.968 0.413 0.048 0.256 391s -0.962 -0.091 0.817 0.827 391s -0.417 0.375 -0.664 0.808 391s 391s x1: [4, 2, 1, 1] 391s 0.200 0.165 -0.823 0.263 391s -0.837 0.502 0.281 0.508 391s 391s y: [4, 2, 1, 1] 391s -0.322 -0.098 -0.663 0.738 391s 0.380 1.161 1.409 0.762 391s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.172 -0.666 0.787 0.407 391s 391s x1: [4, 4, 4, 1] 391s -0.629 -0.549 -0.753 -0.851 391s 0.174 -0.010 0.350 -0.661 391s 0.166 -0.387 -0.498 -0.332 391s 0.894 -0.990 0.399 0.327 391s 391s y: [1, 4, 4, 1] 391s -0.682 391s 0.043 391s -0.241 391s 1.261 391s testing: mul_mat, [3, 1, 4, 3] = [4, 3, 4, 3] * [4, 1, 4, 3] 391s x0: [4, 3, 4, 3] 391s 0.098 -0.434 -0.683 -0.233 391s -0.621 -0.444 0.432 0.672 391s 0.135 0.163 0.555 -0.474 391s 391s x1: [4, 1, 4, 3] 391s 0.612 0.832 0.813 0.809 391s 391s y: [3, 1, 4, 3] 391s -1.045 0.146 0.286 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.914 -0.768 -0.877 0.531 391s -0.764 -0.858 0.439 0.967 391s 0.949 0.529 0.105 -0.732 391s 0.227 0.128 -0.385 -0.256 391s 391s x1: [4, 3, 1, 1] 391s -0.844 0.346 0.760 -0.357 391s 0.080 -0.951 0.743 -0.706 391s 0.044 -0.119 0.179 -0.167 391s 391s y: [4, 3, 1, 1] 391s -1.893 0.336 -0.277 -0.348 391s -0.223 0.399 0.166 -0.209 391s -0.114 -0.014 0.120 -0.031 391s testing: mul_mat, [1, 4, 4, 1] = [4, 1, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 1, 4, 1] 391s 0.363 -0.639 0.333 0.444 391s 391s x1: [4, 4, 4, 1] 391s -0.454 -0.684 -0.100 0.892 391s -0.924 0.543 -0.028 -0.875 391s 0.286 0.267 0.169 -0.833 391s -0.554 -0.997 -0.431 -0.640 391s 391s y: [1, 4, 4, 1] 391s 0.635 391s -1.080 391s -0.380 391s 0.008 391s testing: mul_mat, [3, 4, 4, 3] = [1, 3, 4, 3] * [1, 4, 4, 3] 391s x0: [1, 3, 4, 3] 391s -0.089 391s 0.771 391s 0.046 391s 391s x1: [1, 4, 4, 3] 391s -0.089 391s -0.802 391s 0.338 391s -0.425 391s 391s y: [3, 4, 4, 3] 391s 0.008 -0.068 -0.004 391s 0.071 -0.618 -0.037 391s -0.030 0.260 0.016 391s 0.038 -0.328 -0.020 391s test-mul-mat0: iter:475/500 391s testing: mul_mat, [1, 2, 1, 1] = [3, 1, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.913 -0.332 -0.679 391s 391s x1: [3, 2, 1, 1] 391s -0.222 -0.581 -0.862 391s -0.266 -0.307 -0.485 391s 391s y: [1, 2, 1, 1] 391s 0.981 391s 0.675 391s testing: mul_mat, [2, 1, 3, 1] = [3, 2, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 2, 3, 1] 391s 0.961 -0.878 0.011 391s 0.254 -0.257 -0.691 391s 391s x1: [3, 1, 3, 1] 391s 0.544 -0.296 -0.044 391s 391s y: [2, 1, 3, 1] 391s 0.782 0.245 391s testing: mul_mat, [2, 2, 3, 3] = [3, 2, 3, 3] * [3, 2, 3, 3] 391s x0: [3, 2, 3, 3] 391s 0.279 -0.843 0.176 391s -0.205 -0.118 -0.415 391s 391s x1: [3, 2, 3, 3] 391s 0.743 -0.700 -0.735 391s 0.287 0.004 0.221 391s 391s y: [2, 2, 3, 3] 391s 0.668 0.234 391s 0.116 -0.151 391s testing: mul_mat, [4, 2, 1, 1] = [3, 4, 1, 1] * [3, 2, 1, 1] 391s x0: [3, 4, 1, 1] 391s -0.910 -0.128 0.446 391s 0.722 -0.186 0.941 391s 0.532 -0.067 0.659 391s -0.781 -0.724 0.183 391s 391s x1: [3, 2, 1, 1] 391s 0.739 0.585 -0.710 391s -0.108 0.138 0.366 391s 391s y: [4, 2, 1, 1] 391s -1.064 -0.245 -0.114 -1.131 391s 0.244 0.241 0.174 0.052 391s testing: mul_mat, [1, 3, 3, 1] = [4, 1, 3, 1] * [4, 3, 3, 1] 391s x0: [4, 1, 3, 1] 391s 0.482 0.484 -0.530 0.619 391s 391s x1: [4, 3, 3, 1] 391s 0.956 0.965 -0.890 0.745 391s 0.511 0.451 -0.433 -0.467 391s -0.967 -0.863 0.306 0.291 391s 391s y: [1, 3, 3, 1] 391s 1.861 391s 0.405 391s -0.866 391s testing: mul_mat, [3, 4, 3, 3] = [1, 3, 3, 3] * [1, 4, 3, 3] 391s x0: [1, 3, 3, 3] 391s 0.193 391s 0.283 391s 0.450 391s 391s x1: [1, 4, 3, 3] 391s -0.837 391s 0.830 391s -0.045 391s 0.596 391s 391s y: [3, 4, 3, 3] 391s -0.162 -0.237 -0.377 391s 0.161 0.235 0.373 391s -0.009 -0.013 -0.020 391s 0.115 0.168 0.268 391s test-mul-mat0: iter:476/500 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.830 0.712 0.383 -0.467 391s -0.576 0.681 0.197 -0.027 391s 391s x1: [4, 2, 1, 1] 391s 0.774 0.525 0.051 -0.498 391s 0.460 -0.714 0.638 0.218 391s 391s y: [2, 2, 1, 1] 391s -0.017 -0.065 391s -0.747 -0.631 391s testing: mul_mat, [4, 2, 1, 1] = [4, 4, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.565 -0.668 -0.863 -0.324 391s 0.933 0.912 -0.799 -0.016 391s -0.587 0.661 0.270 -0.949 391s -0.121 -0.984 -0.779 -0.409 391s 391s x1: [4, 2, 1, 1] 391s -0.571 0.923 -0.147 0.310 391s -0.948 -0.673 -0.965 0.870 391s 391s y: [4, 2, 1, 1] 391s -0.268 0.421 0.611 -0.852 391s 1.537 -0.742 -0.974 1.172 391s testing: mul_mat, [4, 4, 1, 3] = [4, 4, 1, 3] * [4, 4, 1, 3] 391s x0: [4, 4, 1, 3] 391s 0.252 0.947 -0.333 0.420 391s -0.009 0.853 0.248 0.305 391s 0.090 0.956 0.635 -0.451 391s 0.254 0.364 -0.239 0.567 391s 391s x1: [4, 4, 1, 3] 391s 0.400 -0.246 0.015 0.081 391s 0.951 0.987 0.510 0.874 391s -0.159 -0.180 0.926 0.168 391s -0.144 0.796 -0.276 0.290 391s 391s y: [4, 4, 1, 3] 391s -0.103 -0.185 -0.226 0.054 391s 1.372 1.226 0.959 0.975 391s -0.448 0.129 0.326 -0.232 391s 0.931 0.700 0.441 0.484 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.580 -0.938 -0.678 0.712 391s 0.250 0.921 -0.902 -0.694 391s -0.932 -0.950 -0.680 0.570 391s 0.257 -0.491 0.484 0.193 391s 391s x1: [4, 4, 1, 1] 391s -0.914 0.481 -0.520 0.048 391s -0.512 -0.885 0.973 0.805 391s 0.745 -0.344 -0.847 -0.338 391s 0.226 -0.604 -0.169 0.743 391s 391s y: [4, 4, 1, 1] 391s 0.466 0.650 0.776 -0.714 391s 1.040 -2.379 1.115 0.929 391s 0.223 0.868 0.015 -0.114 391s 1.078 -0.863 0.901 0.416 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.409 0.529 0.867 0.652 391s 0.816 0.330 -0.962 0.965 391s 391s x1: [4, 4, 1, 1] 391s -0.214 -0.382 0.681 -0.099 391s -0.409 0.486 -0.354 0.247 391s 0.639 0.308 -0.527 -0.965 391s -0.861 -0.784 0.341 -0.441 391s 391s y: [2, 4, 1, 1] 391s 0.236 -1.052 391s -0.056 0.405 391s -0.661 0.200 391s -0.759 -1.715 391s testing: mul_mat, [1, 4, 1, 3] = [2, 1, 1, 3] * [2, 4, 1, 3] 391s x0: [2, 1, 1, 3] 391s -0.153 0.791 391s 391s x1: [2, 4, 1, 3] 391s -0.642 0.136 391s 0.676 -0.337 391s -0.293 -0.131 391s 0.449 0.325 391s 391s y: [1, 4, 1, 3] 391s 0.206 391s -0.370 391s -0.059 391s 0.189 391s test-mul-mat0: iter:477/500 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s 0.705 -0.472 391s 0.782 -0.299 391s 0.691 0.457 391s 391s x1: [2, 2, 1, 1] 391s 0.287 0.196 391s 0.355 -0.388 391s 391s y: [3, 2, 1, 1] 391s 0.110 0.166 0.288 391s 0.434 0.394 0.068 391s testing: mul_mat, [1, 3, 1, 1] = [2, 1, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.502 -0.319 391s 391s x1: [2, 3, 1, 1] 391s 0.006 0.327 391s 0.168 0.449 391s 0.179 -0.781 391s 391s y: [1, 3, 1, 1] 391s -0.107 391s -0.228 391s 0.159 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.956 -0.770 391s 391s x1: [2, 1, 1, 1] 391s 0.270 0.345 391s 391s y: [1, 1, 1, 1] 391s -0.007 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.504 -0.068 391s 391s x1: [2, 1, 1, 1] 391s 0.916 -0.394 391s 391s y: [1, 1, 1, 1] 391s -0.435 391s testing: mul_mat, [2, 2, 1, 1] = [1, 2, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 2, 1, 1] 391s -0.266 391s -0.605 391s 391s x1: [1, 2, 1, 1] 391s 0.379 391s 0.783 391s 391s y: [2, 2, 1, 1] 391s -0.101 -0.229 391s -0.208 -0.474 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.656 0.141 391s 391s x1: [2, 1, 1, 1] 391s 0.874 -0.561 391s 391s y: [1, 1, 1, 1] 391s 0.495 391s test-mul-mat0: iter:478/500 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.653 0.273 0.193 0.569 391s 391s x1: [4, 2, 1, 1] 391s -0.719 0.595 0.121 -0.499 391s -0.953 0.619 0.181 0.317 391s 391s y: [1, 2, 1, 1] 391s -0.568 391s -0.238 391s testing: mul_mat, [3, 1, 3, 1] = [4, 3, 3, 1] * [4, 1, 3, 1] 391s x0: [4, 3, 3, 1] 391s -0.641 0.781 0.583 -0.495 391s 0.722 0.864 -0.899 -0.157 391s -0.635 -0.852 -0.539 0.546 391s 391s x1: [4, 1, 3, 1] 391s 0.879 -0.334 -0.935 -0.189 391s 391s y: [3, 1, 3, 1] 391s -1.276 1.216 0.127 391s testing: mul_mat, [3, 3, 3, 2] = [4, 3, 3, 2] * [4, 3, 3, 2] 391s x0: [4, 3, 3, 2] 391s 0.367 0.730 0.267 0.161 391s 0.416 0.338 -0.491 0.594 391s -0.538 0.519 0.188 0.820 391s 391s x1: [4, 3, 3, 2] 391s 0.098 -0.710 -0.564 0.463 391s -0.563 -0.102 0.009 0.902 391s -0.678 -0.792 -0.360 -0.059 391s 391s y: [3, 3, 3, 2] 391s -0.558 0.353 -0.147 391s -0.133 0.262 0.991 391s -0.932 -0.408 -0.162 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.890 0.211 -0.200 0.279 391s 391s x1: [4, 3, 1, 1] 391s -0.869 -0.376 0.489 0.431 391s -0.150 0.789 0.501 -0.496 391s 0.795 0.793 -0.830 -0.819 391s 391s y: [1, 3, 1, 1] 391s 0.717 391s 0.062 391s -0.602 391s testing: mul_mat, [4, 4, 3, 1] = [1, 4, 3, 1] * [1, 4, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.382 391s 0.145 391s 0.783 391s -0.129 391s 391s x1: [1, 4, 3, 1] 391s -0.364 391s -0.521 391s 0.532 391s -0.665 391s 391s y: [4, 4, 3, 1] 391s -0.139 -0.053 -0.285 0.047 391s -0.199 -0.076 -0.408 0.067 391s 0.203 0.077 0.417 -0.069 391s -0.254 -0.097 -0.521 0.086 391s testing: mul_mat, [1, 1, 3, 2] = [4, 1, 3, 2] * [4, 1, 3, 2] 391s x0: [4, 1, 3, 2] 391s -0.928 0.536 -0.196 -0.473 391s 391s x1: [4, 1, 3, 2] 391s -0.364 0.041 0.417 0.847 391s 391s y: [1, 1, 3, 2] 391s -0.123 391s test-mul-mat0: iter:479/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.477 0.469 391s 0.193 0.534 391s 0.756 0.416 391s 0.027 -0.636 391s 391s x1: [2, 4, 1, 1] 391s -0.238 -0.920 391s -0.250 -0.166 391s 0.616 0.554 391s 0.360 0.022 391s 391s y: [4, 4, 1, 1] 391s -0.318 -0.537 -0.563 0.579 391s 0.041 -0.137 -0.258 0.099 391s -0.033 0.414 0.696 -0.336 391s -0.162 0.081 0.282 -0.004 391s testing: mul_mat, [3, 4, 4, 1] = [2, 3, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 3, 4, 1] 391s -0.318 0.569 391s -0.901 0.199 391s -0.283 -0.058 391s 391s x1: [2, 4, 4, 1] 391s 0.868 0.338 391s -0.133 0.147 391s -0.955 -0.483 391s 0.149 -0.157 391s 391s y: [3, 4, 4, 1] 391s -0.083 -0.715 -0.265 391s 0.126 0.149 0.029 391s 0.029 0.765 0.298 391s -0.137 -0.165 -0.033 391s testing: mul_mat, [1, 3, 4, 3] = [2, 1, 4, 3] * [2, 3, 4, 3] 391s x0: [2, 1, 4, 3] 391s -0.242 0.546 391s 391s x1: [2, 3, 4, 3] 391s -0.614 0.277 391s -0.545 0.102 391s 0.092 -0.123 391s 391s y: [1, 3, 4, 3] 391s 0.300 391s 0.188 391s -0.089 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.167 0.425 391s -0.333 0.308 391s 0.762 -0.232 391s 0.437 0.749 391s 391s x1: [2, 1, 1, 1] 391s -0.541 -0.293 391s 391s y: [4, 1, 1, 1] 391s -0.215 0.090 -0.344 -0.455 391s testing: mul_mat, [2, 2, 4, 1] = [4, 2, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 2, 4, 1] 391s 0.048 -0.794 -0.653 -0.993 391s -0.146 -0.632 0.778 0.285 391s 391s x1: [4, 2, 4, 1] 391s -0.538 0.859 0.520 0.066 391s -0.956 0.320 0.494 0.571 391s 391s y: [2, 2, 4, 1] 391s -1.112 -0.041 391s -1.189 0.484 391s testing: mul_mat, [2, 4, 4, 3] = [2, 2, 4, 3] * [2, 4, 4, 3] 391s x0: [2, 2, 4, 3] 391s 0.872 -0.968 391s -0.231 0.634 391s 391s x1: [2, 4, 4, 3] 391s 0.454 -0.060 391s -0.020 0.801 391s -0.282 -0.013 391s 0.087 0.723 391s 391s y: [2, 4, 4, 3] 391s 0.453 -0.143 391s -0.793 0.513 391s -0.233 0.057 391s -0.624 0.439 391s test-mul-mat0: iter:480/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.296 391s 0.143 391s 391s x1: [1, 4, 1, 1] 391s 0.884 391s 0.439 391s 0.941 391s 0.364 391s 391s y: [2, 4, 1, 1] 391s 0.262 0.126 391s 0.130 0.063 391s 0.279 0.134 391s 0.108 0.052 391s testing: mul_mat, [1, 2, 3, 1] = [1, 1, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 1, 3, 1] 391s 0.335 391s 391s x1: [1, 2, 3, 1] 391s -0.910 391s 0.297 391s 391s y: [1, 2, 3, 1] 391s -0.305 391s 0.100 391s testing: mul_mat, [2, 1, 3, 4] = [1, 2, 3, 4] * [1, 1, 3, 4] 391s x0: [1, 2, 3, 4] 391s -0.633 391s -0.636 391s 391s x1: [1, 1, 3, 4] 391s 0.762 391s 391s y: [2, 1, 3, 4] 391s -0.482 -0.485 391s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.292 391s 391s x1: [1, 2, 1, 1] 391s -0.837 391s -0.930 391s 391s y: [1, 2, 1, 1] 391s 0.245 391s 0.272 391s testing: mul_mat, [4, 1, 3, 1] = [1, 4, 3, 1] * [1, 1, 3, 1] 391s x0: [1, 4, 3, 1] 391s 0.984 391s -0.723 391s -0.513 391s 0.711 391s 391s x1: [1, 1, 3, 1] 391s 0.812 391s 391s y: [4, 1, 3, 1] 391s 0.798 -0.587 -0.416 0.577 391s testing: mul_mat, [1, 1, 3, 4] = [4, 1, 3, 4] * [4, 1, 3, 4] 391s x0: [4, 1, 3, 4] 391s -0.133 0.859 0.268 -0.842 391s 391s x1: [4, 1, 3, 4] 391s 0.756 -0.683 -0.737 -0.406 391s 391s y: [1, 1, 3, 4] 391s -0.543 391s test-mul-mat0: iter:481/500 391s testing: mul_mat, [2, 3, 1, 1] = [4, 2, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.546 0.631 0.721 -0.663 391s -0.351 -0.933 0.043 -0.028 391s 391s x1: [4, 3, 1, 1] 391s -0.374 -0.186 -0.493 0.462 391s -0.024 0.332 0.305 0.201 391s 0.495 -0.853 -0.705 0.517 391s 391s y: [2, 3, 1, 1] 391s -0.983 0.270 391s 0.283 -0.293 391s -1.119 0.576 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.034 -0.994 0.658 0.541 391s 0.468 -0.366 -0.127 -0.226 391s 391s x1: [4, 2, 1, 1] 391s 0.920 0.999 -0.244 -0.784 391s 0.803 -0.248 -0.447 0.380 391s 391s y: [2, 2, 1, 1] 391s -1.546 0.273 391s 0.186 0.438 391s testing: mul_mat, [2, 2, 1, 2] = [4, 2, 1, 2] * [4, 2, 1, 2] 391s x0: [4, 2, 1, 2] 391s 0.103 -0.141 -0.355 0.657 391s -0.761 0.489 -0.309 -0.755 391s 391s x1: [4, 2, 1, 2] 391s 0.835 -0.632 -0.079 -0.870 391s 0.885 0.179 0.676 0.516 391s 391s y: [2, 2, 1, 2] 391s -0.368 -0.264 391s 0.165 -1.185 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.546 -0.823 0.548 -0.557 391s -0.508 0.340 -0.929 -0.593 391s 391s x1: [4, 2, 1, 1] 391s 0.409 0.747 0.359 -0.413 391s 0.423 -0.125 0.487 0.436 391s 391s y: [2, 2, 1, 1] 391s -0.412 -0.042 391s -0.104 -0.968 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.134 0.092 391s -0.872 -0.248 391s 391s x1: [2, 4, 1, 1] 391s -0.284 -0.900 391s -0.354 -0.795 391s -0.209 -0.109 391s 0.353 0.023 391s 391s y: [2, 4, 1, 1] 391s -0.121 0.471 391s -0.121 0.506 391s -0.038 0.209 391s 0.050 -0.313 391s testing: mul_mat, [4, 2, 1, 2] = [2, 4, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 4, 1, 2] 391s -0.125 -0.632 391s 0.028 0.398 391s 0.850 0.123 391s 0.052 -0.188 391s 391s x1: [2, 2, 1, 2] 391s 0.601 -0.500 391s -0.501 -0.040 391s 391s y: [4, 2, 1, 2] 391s 0.240 -0.182 0.450 0.125 391s 0.088 -0.030 -0.431 -0.019 391s test-mul-mat0: iter:482/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.090 0.176 -0.403 -0.693 391s -0.701 0.409 -0.888 -0.862 391s -0.679 -0.437 -0.818 -0.977 391s 391s x1: [4, 3, 1, 1] 391s 0.038 0.283 -0.545 0.538 391s 0.781 0.415 -0.374 0.703 391s 0.250 0.201 0.060 -0.876 391s 391s y: [3, 3, 1, 1] 391s -0.107 0.108 -0.229 391s -0.333 -0.652 -1.092 391s 0.595 0.609 0.549 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.124 -0.278 -0.746 0.936 391s 391s x1: [4, 3, 1, 1] 391s 0.905 0.578 -0.251 0.851 391s -0.891 0.603 -0.111 0.392 391s -0.942 -0.572 0.173 0.473 391s 391s y: [1, 3, 1, 1] 391s 0.711 391s 0.393 391s 0.589 391s testing: mul_mat, [3, 1, 1, 4] = [4, 3, 1, 4] * [4, 1, 1, 4] 391s x0: [4, 3, 1, 4] 391s 0.584 -0.671 0.184 -0.527 391s 0.721 0.242 -0.099 -0.106 391s -0.285 0.955 0.771 0.438 391s 391s x1: [4, 1, 1, 4] 391s -0.153 -0.517 -0.154 -0.977 391s 391s y: [3, 1, 1, 4] 391s 0.744 -0.118 -0.997 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.168 0.049 -0.665 0.499 391s 391s x1: [4, 3, 1, 1] 391s -0.345 -0.882 -0.370 0.285 391s -0.622 0.601 -0.626 0.447 391s -0.600 -0.895 0.349 -0.068 391s 391s y: [1, 3, 1, 1] 391s 0.403 391s 0.773 391s -0.208 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s -0.965 391s -0.425 391s 0.896 391s 391s x1: [1, 4, 1, 1] 391s 0.639 391s -0.665 391s 0.410 391s 0.129 391s 391s y: [3, 4, 1, 1] 391s -0.616 -0.272 0.572 391s 0.641 0.283 -0.596 391s -0.395 -0.174 0.367 391s -0.125 -0.055 0.116 391s testing: mul_mat, [3, 1, 1, 4] = [3, 3, 1, 4] * [3, 1, 1, 4] 391s x0: [3, 3, 1, 4] 391s -0.483 0.866 -0.302 391s 0.823 -0.245 0.804 391s -0.134 -0.078 0.256 391s 391s x1: [3, 1, 1, 4] 391s -0.901 -0.771 -0.732 391s 391s y: [3, 1, 1, 4] 391s -0.011 -1.142 -0.007 391s test-mul-mat0: iter:483/500 391s testing: mul_mat, [2, 3, 1, 1] = [2, 2, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.030 -0.385 391s 0.983 -0.654 391s 391s x1: [2, 3, 1, 1] 391s -0.007 0.251 391s 0.074 -0.400 391s 0.823 0.897 391s 391s y: [2, 3, 1, 1] 391s -0.096 -0.171 391s 0.152 0.335 391s -0.370 0.223 391s testing: mul_mat, [1, 2, 3, 1] = [2, 1, 3, 1] * [2, 2, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.120 -0.013 391s 391s x1: [2, 2, 3, 1] 391s 0.601 -0.829 391s -0.702 0.780 391s 391s y: [1, 2, 3, 1] 391s -0.061 391s 0.074 391s testing: mul_mat, [3, 1, 3, 4] = [2, 3, 3, 4] * [2, 1, 3, 4] 391s x0: [2, 3, 3, 4] 391s -0.173 -0.571 391s 0.457 0.934 391s -0.788 0.723 391s 391s x1: [2, 1, 3, 4] 391s -0.411 -0.224 391s 391s y: [3, 1, 3, 4] 391s 0.199 -0.397 0.162 391s testing: mul_mat, [4, 3, 1, 1] = [2, 4, 1, 1] * [2, 3, 1, 1] 391s x0: [2, 4, 1, 1] 391s 0.022 -0.342 391s 0.664 0.587 391s 0.685 0.543 391s 0.085 -0.758 391s 391s x1: [2, 3, 1, 1] 391s -0.946 0.438 391s 0.878 -0.578 391s 0.457 0.923 391s 391s y: [4, 3, 1, 1] 391s -0.171 -0.371 -0.410 -0.413 391s 0.217 0.244 0.287 0.513 391s -0.306 0.845 0.815 -0.661 391s testing: mul_mat, [2, 2, 3, 1] = [4, 2, 3, 1] * [4, 2, 3, 1] 391s x0: [4, 2, 3, 1] 391s 0.219 -0.836 -0.137 0.256 391s 0.319 -0.195 -0.594 0.813 391s 391s x1: [4, 2, 3, 1] 391s -0.549 -0.050 -0.040 -0.776 391s -0.253 0.812 0.735 0.364 391s 391s y: [2, 2, 3, 1] 391s -0.272 -0.773 391s -0.741 -0.379 391s testing: mul_mat, [2, 4, 3, 4] = [2, 2, 3, 4] * [2, 4, 3, 4] 391s x0: [2, 2, 3, 4] 391s -0.616 0.902 391s 0.691 0.114 391s 391s x1: [2, 4, 3, 4] 391s -0.617 0.516 391s -0.273 -0.774 391s 0.050 -0.039 391s 0.732 -0.732 391s 391s y: [2, 4, 3, 4] 391s 0.846 -0.368 391s -0.530 -0.277 391s -0.066 0.030 391s -1.111 0.423 391s test-mul-mat0: iter:484/500 391s testing: mul_mat, [1, 4, 1, 1] = [2, 1, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.751 -0.854 391s 391s x1: [2, 4, 1, 1] 391s 0.151 -0.508 391s -0.137 -0.223 391s -0.646 -0.363 391s 0.387 0.058 391s 391s y: [1, 4, 1, 1] 391s 0.320 391s 0.293 391s 0.795 391s -0.340 391s testing: mul_mat, [1, 1, 2, 1] = [2, 1, 2, 1] * [2, 1, 2, 1] 391s x0: [2, 1, 2, 1] 391s -0.343 -0.406 391s 391s x1: [2, 1, 2, 1] 391s 0.329 0.014 391s 391s y: [1, 1, 2, 1] 391s -0.118 391s testing: mul_mat, [2, 1, 2, 3] = [2, 2, 2, 3] * [2, 1, 2, 3] 391s x0: [2, 2, 2, 3] 391s -0.052 -0.055 391s 0.770 0.585 391s 391s x1: [2, 1, 2, 3] 391s 0.788 -0.463 391s 391s y: [2, 1, 2, 3] 391s -0.015 0.336 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.301 0.326 391s -0.630 -0.754 391s 391s x1: [2, 2, 1, 1] 391s -0.970 -0.707 391s 0.133 -0.046 391s 391s y: [2, 2, 1, 1] 391s -0.522 1.145 391s 0.025 -0.049 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.267 0.952 391s 0.876 0.248 391s 391s x1: [2, 2, 2, 1] 391s -0.860 -0.089 391s -0.423 -0.031 391s 391s y: [2, 2, 2, 1] 391s -0.315 -0.775 391s -0.143 -0.378 391s testing: mul_mat, [1, 2, 2, 3] = [2, 1, 2, 3] * [2, 2, 2, 3] 391s x0: [2, 1, 2, 3] 391s -0.891 -0.887 391s 391s x1: [2, 2, 2, 3] 391s 0.949 -0.125 391s 0.192 0.689 391s 391s y: [1, 2, 2, 3] 391s -0.735 391s -0.782 391s test-mul-mat0: iter:485/500 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.946 0.332 -0.840 0.284 391s -0.263 0.758 -0.800 -0.107 391s 391s x1: [4, 1, 1, 1] 391s 0.415 0.820 0.458 0.085 391s 391s y: [2, 1, 1, 1] 391s -0.481 0.137 391s testing: mul_mat, [3, 2, 4, 1] = [4, 3, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 3, 4, 1] 391s 0.939 -0.014 0.581 -0.267 391s 0.725 -0.288 0.267 0.340 391s 0.720 0.231 -0.177 0.829 391s 391s x1: [4, 2, 4, 1] 391s -0.377 -0.008 0.002 -0.264 391s -0.555 -0.206 -0.261 0.131 391s 391s y: [3, 2, 4, 1] 391s -0.282 -0.360 -0.492 391s -0.706 -0.368 -0.293 391s testing: mul_mat, [4, 3, 4, 3] = [4, 4, 4, 3] * [4, 3, 4, 3] 391s x0: [4, 4, 4, 3] 391s -0.556 -0.406 0.874 -0.326 391s 0.073 -0.432 -0.360 -0.579 391s -0.053 0.482 0.751 -0.360 391s -0.565 -0.562 -0.490 -0.311 391s 391s x1: [4, 3, 4, 3] 391s 0.717 0.342 0.390 0.842 391s -0.636 0.857 0.836 0.331 391s -0.632 0.457 -0.558 0.815 391s 391s y: [4, 3, 4, 3] 391s -0.471 -0.723 0.117 -1.050 391s 0.628 -0.909 0.955 -0.635 391s -0.588 -0.514 -0.459 0.120 391s testing: mul_mat, [2, 4, 1, 1] = [4, 2, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.528 0.890 0.437 0.555 391s -0.749 0.054 -0.572 0.657 391s 391s x1: [4, 4, 1, 1] 391s -0.961 -0.091 -0.463 -0.748 391s 0.417 -0.364 -0.918 0.083 391s 0.007 -0.382 -0.629 0.793 391s 0.525 -0.961 -0.500 -0.615 391s 391s y: [2, 4, 1, 1] 391s -1.206 0.488 391s -0.459 0.247 391s -0.171 0.854 391s -1.137 -0.564 391s testing: mul_mat, [4, 4, 4, 1] = [2, 4, 4, 1] * [2, 4, 4, 1] 391s x0: [2, 4, 4, 1] 391s -0.775 0.744 391s -0.902 0.489 391s 0.847 -0.357 391s -0.834 0.640 391s 391s x1: [2, 4, 4, 1] 391s -0.256 -0.438 391s -0.627 0.606 391s -0.267 0.941 391s 0.645 0.642 391s 391s y: [4, 4, 4, 1] 391s -0.127 0.017 -0.061 -0.067 391s 0.936 0.862 -0.747 0.911 391s 0.906 0.701 -0.562 0.825 391s -0.023 -0.268 0.318 -0.127 391s testing: mul_mat, [1, 2, 4, 3] = [4, 1, 4, 3] * [4, 2, 4, 3] 391s x0: [4, 1, 4, 3] 391s -0.872 -0.073 -0.515 0.123 391s 391s x1: [4, 2, 4, 3] 391s -0.159 -0.341 -0.955 -0.416 391s -0.852 -0.312 -0.775 -0.305 391s 391s y: [1, 2, 4, 3] 391s 0.605 391s 1.128 391s test-mul-mat0: iter:486/500 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.611 -0.212 -0.118 0.239 391s -0.773 0.578 0.016 -0.849 391s 0.397 -0.537 -0.182 -0.652 391s -0.989 0.060 -0.962 0.896 391s 391s x1: [4, 1, 1, 1] 391s 0.046 -0.495 0.088 -0.913 391s 391s y: [4, 1, 1, 1] 391s -0.152 0.455 0.863 -0.978 391s testing: mul_mat, [1, 4, 2, 1] = [4, 1, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 1, 2, 1] 391s -0.290 -0.130 -0.943 0.407 391s 391s x1: [4, 4, 2, 1] 391s -0.316 -0.991 -0.526 0.715 391s 0.266 -0.303 0.255 -0.706 391s 0.721 -0.865 0.340 -0.773 391s 0.223 0.428 -0.460 0.612 391s 391s y: [1, 4, 2, 1] 391s 1.007 391s -0.566 391s -0.733 391s 0.563 391s testing: mul_mat, [2, 1, 2, 1] = [4, 2, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 2, 2, 1] 391s 0.778 -0.683 0.813 -0.629 391s -0.287 -0.858 0.783 0.458 391s 391s x1: [4, 1, 2, 1] 391s 0.692 0.356 0.483 0.119 391s 391s y: [2, 1, 2, 1] 391s 0.613 -0.070 391s testing: mul_mat, [3, 2, 1, 1] = [4, 3, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.750 0.586 0.364 0.735 391s 0.217 -0.466 -0.149 0.564 391s 0.244 0.190 0.003 0.145 391s 391s x1: [4, 2, 1, 1] 391s 0.088 -0.163 -0.595 0.439 391s 0.966 -0.335 0.131 0.321 391s 391s y: [3, 2, 1, 1] 391s -0.055 0.431 0.052 391s -0.636 0.527 0.219 391s testing: mul_mat, [3, 4, 2, 1] = [3, 3, 2, 1] * [3, 4, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.150 -0.115 -0.597 391s -0.499 -0.934 0.088 391s -0.883 0.262 -0.111 391s 391s x1: [3, 4, 2, 1] 391s 0.518 0.022 0.627 391s 0.772 0.262 -0.654 391s 0.181 -0.707 0.775 391s 0.461 0.381 -0.388 391s 391s y: [3, 4, 2, 1] 391s -0.300 -0.224 -0.521 391s 0.476 -0.687 -0.541 391s -0.355 0.638 -0.431 391s 0.257 -0.620 -0.265 391s testing: mul_mat, [3, 3, 2, 1] = [3, 3, 2, 1] * [3, 3, 2, 1] 391s x0: [3, 3, 2, 1] 391s 0.217 0.391 -0.253 391s -0.481 -0.131 0.511 391s 0.203 -0.562 0.448 391s 391s x1: [3, 3, 2, 1] 391s -0.316 -0.158 -0.800 391s 0.216 -0.207 -0.901 391s 0.896 -0.007 0.215 391s 391s y: [3, 3, 2, 1] 391s 0.072 -0.236 -0.334 391s 0.193 -0.537 -0.244 391s 0.137 -0.320 0.282 391s test-mul-mat0: iter:487/500 391s testing: mul_mat, [3, 1, 1, 1] = [2, 3, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.130 -0.184 391s 0.168 0.684 391s -0.752 0.482 391s 391s x1: [2, 1, 1, 1] 391s -0.059 0.450 391s 391s y: [3, 1, 1, 1] 391s -0.075 0.298 0.261 391s testing: mul_mat, [1, 3, 2, 1] = [2, 1, 2, 1] * [2, 3, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.056 -0.087 391s 391s x1: [2, 3, 2, 1] 391s -0.099 -0.233 391s -0.315 -0.708 391s 0.636 0.123 391s 391s y: [1, 3, 2, 1] 391s 0.015 391s 0.044 391s 0.025 391s testing: mul_mat, [1, 1, 2, 3] = [2, 1, 2, 3] * [2, 1, 2, 3] 391s x0: [2, 1, 2, 3] 391s 0.626 -0.218 391s 391s x1: [2, 1, 2, 3] 391s -0.717 0.678 391s 391s y: [1, 1, 2, 3] 391s -0.597 391s testing: mul_mat, [4, 1, 1, 1] = [2, 4, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.441 0.566 391s 0.888 0.277 391s -0.109 0.710 391s -0.158 -0.575 391s 391s x1: [2, 1, 1, 1] 391s 0.646 0.571 391s 391s y: [4, 1, 1, 1] 391s 0.038 0.732 0.335 -0.431 391s testing: mul_mat, [3, 2, 2, 1] = [4, 3, 2, 1] * [4, 2, 2, 1] 391s x0: [4, 3, 2, 1] 391s -0.082 -0.436 0.123 0.965 391s 0.368 -0.061 -0.172 -0.606 391s 0.019 0.685 -0.424 0.853 391s 391s x1: [4, 2, 2, 1] 391s -0.397 -0.972 0.400 -0.200 391s 0.162 0.544 -0.319 -0.456 391s 391s y: [3, 2, 2, 1] 391s 0.313 -0.035 -1.014 391s -0.730 0.358 0.122 391s testing: mul_mat, [1, 4, 2, 3] = [3, 1, 2, 3] * [3, 4, 2, 3] 391s x0: [3, 1, 2, 3] 391s -0.424 -0.812 0.078 391s 391s x1: [3, 4, 2, 3] 391s -0.899 -0.587 0.935 391s -0.422 0.326 -0.415 391s 0.986 -0.756 0.953 391s 0.004 -0.193 -0.108 391s 391s y: [1, 4, 2, 3] 391s 0.931 391s -0.118 391s 0.270 391s 0.146 391s test-mul-mat0: iter:488/500 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.049 -0.683 391s 0.473 -0.126 391s 391s x1: [2, 4, 1, 1] 391s 0.017 -0.221 391s -0.993 0.426 391s 0.846 0.883 391s -0.527 -0.871 391s 391s y: [2, 4, 1, 1] 391s 0.152 0.036 391s -0.340 -0.523 391s -0.562 0.289 391s 0.570 -0.139 391s testing: mul_mat, [2, 2, 2, 1] = [2, 2, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 2, 2, 1] 391s 0.041 0.752 391s 0.763 -0.292 391s 391s x1: [2, 2, 2, 1] 391s 0.675 0.010 391s -0.938 0.323 391s 391s y: [2, 2, 2, 1] 391s 0.036 0.512 391s 0.205 -0.810 391s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.501 0.965 391s 391s x1: [2, 2, 2, 1] 391s 0.216 0.017 391s 0.744 0.223 391s 391s y: [1, 2, 2, 1] 391s 0.124 391s 0.588 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.099 -0.438 391s 0.034 -0.554 391s 391s x1: [2, 1, 1, 1] 391s -0.161 0.710 391s 391s y: [2, 1, 1, 1] 391s -0.295 -0.399 391s testing: mul_mat, [1, 2, 2, 1] = [2, 1, 2, 1] * [2, 2, 2, 1] 391s x0: [2, 1, 2, 1] 391s 0.618 0.768 391s 391s x1: [2, 2, 2, 1] 391s 0.869 -0.993 391s -0.080 -0.090 391s 391s y: [1, 2, 2, 1] 391s -0.225 391s -0.118 391s testing: mul_mat, [3, 2, 2, 1] = [1, 3, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 3, 2, 1] 391s 0.993 391s -0.181 391s -0.013 391s 391s x1: [1, 2, 2, 1] 391s 0.511 391s -0.620 391s 391s y: [3, 2, 2, 1] 391s 0.508 -0.092 -0.006 391s -0.616 0.112 0.008 391s test-mul-mat0: iter:489/500 391s testing: mul_mat, [2, 4, 1, 1] = [1, 2, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.065 391s -0.623 391s 391s x1: [1, 4, 1, 1] 391s 0.598 391s -0.492 391s -0.856 391s 0.467 391s 391s y: [2, 4, 1, 1] 391s 0.039 -0.372 391s -0.032 0.307 391s -0.055 0.533 391s 0.030 -0.291 391s testing: mul_mat, [2, 2, 3, 1] = [1, 2, 3, 1] * [1, 2, 3, 1] 391s x0: [1, 2, 3, 1] 391s -0.153 391s -0.753 391s 391s x1: [1, 2, 3, 1] 391s -0.725 391s 0.748 391s 391s y: [2, 2, 3, 1] 391s 0.111 0.546 391s -0.114 -0.563 391s testing: mul_mat, [1, 2, 3, 4] = [1, 1, 3, 4] * [1, 2, 3, 4] 391s x0: [1, 1, 3, 4] 391s 0.265 391s 391s x1: [1, 2, 3, 4] 391s -0.286 391s 0.318 391s 391s y: [1, 2, 3, 4] 391s -0.076 391s 0.084 391s testing: mul_mat, [2, 1, 1, 1] = [1, 2, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 2, 1, 1] 391s 0.740 391s -0.823 391s 391s x1: [1, 1, 1, 1] 391s -0.431 391s 391s y: [2, 1, 1, 1] 391s -0.319 0.354 391s testing: mul_mat, [1, 1, 3, 1] = [2, 1, 3, 1] * [2, 1, 3, 1] 391s x0: [2, 1, 3, 1] 391s -0.073 0.829 391s 391s x1: [2, 1, 3, 1] 391s -0.245 0.032 391s 391s y: [1, 1, 3, 1] 391s 0.045 391s testing: mul_mat, [3, 2, 3, 4] = [1, 3, 3, 4] * [1, 2, 3, 4] 391s x0: [1, 3, 3, 4] 391s -0.086 391s -0.542 391s 0.091 391s 391s x1: [1, 2, 3, 4] 391s -0.198 391s -0.362 391s 391s y: [3, 2, 3, 4] 391s 0.017 0.107 -0.018 391s 0.031 0.196 -0.033 391s test-mul-mat0: iter:490/500 391s testing: mul_mat, [2, 1, 1, 1] = [2, 2, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.758 0.463 391s 0.574 0.666 391s 391s x1: [2, 1, 1, 1] 391s -0.436 -0.726 391s 391s y: [2, 1, 1, 1] 391s -0.006 -0.734 391s testing: mul_mat, [2, 2, 1, 1] = [2, 2, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 2, 1, 1] 391s 0.461 0.172 391s -0.757 0.570 391s 391s x1: [2, 2, 1, 1] 391s -0.970 -0.498 391s 0.744 -0.981 391s 391s y: [2, 2, 1, 1] 391s -0.533 0.451 391s 0.174 -1.123 391s testing: mul_mat, [1, 2, 1, 2] = [2, 1, 1, 2] * [2, 2, 1, 2] 391s x0: [2, 1, 1, 2] 391s 0.327 -0.553 391s 391s x1: [2, 2, 1, 2] 391s -0.479 0.525 391s -0.257 0.857 391s 391s y: [1, 2, 1, 2] 391s -0.447 391s -0.558 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s -0.732 -0.406 391s 391s x1: [2, 1, 1, 1] 391s 0.993 0.281 391s 391s y: [1, 1, 1, 1] 391s -0.841 391s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 1, 1, 1] 391s -0.205 391s 391s x1: [1, 2, 1, 1] 391s 0.129 391s 0.731 391s 391s y: [1, 2, 1, 1] 391s -0.026 391s -0.150 391s testing: mul_mat, [1, 1, 1, 2] = [1, 1, 1, 2] * [1, 1, 1, 2] 391s x0: [1, 1, 1, 2] 391s 0.780 391s 391s x1: [1, 1, 1, 2] 391s 0.761 391s 391s y: [1, 1, 1, 2] 391s 0.593 391s test-mul-mat0: iter:491/500 391s testing: mul_mat, [4, 4, 1, 1] = [2, 4, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 4, 1, 1] 391s -0.237 -0.204 391s 0.016 0.085 391s -0.211 -0.703 391s 0.972 0.057 391s 391s x1: [2, 4, 1, 1] 391s 0.473 0.104 391s -0.687 0.329 391s -0.458 -0.059 391s 0.435 -0.651 391s 391s y: [4, 4, 1, 1] 391s -0.133 0.017 -0.173 0.465 391s 0.096 0.017 -0.087 -0.649 391s 0.121 -0.012 0.138 -0.449 391s 0.030 -0.049 0.366 0.386 391s testing: mul_mat, [2, 4, 2, 1] = [2, 2, 2, 1] * [2, 4, 2, 1] 391s x0: [2, 2, 2, 1] 391s -0.918 -0.708 391s 0.257 0.518 391s 391s x1: [2, 4, 2, 1] 391s -0.109 0.101 391s -0.212 0.059 391s 0.895 -0.451 391s 0.730 -0.566 391s 391s y: [2, 4, 2, 1] 391s 0.029 0.024 391s 0.152 -0.024 391s -0.503 -0.004 391s -0.269 -0.106 391s testing: mul_mat, [1, 2, 2, 4] = [2, 1, 2, 4] * [2, 2, 2, 4] 391s x0: [2, 1, 2, 4] 391s -0.143 -0.485 391s 391s x1: [2, 2, 2, 4] 391s -0.180 0.366 391s -0.774 0.116 391s 391s y: [1, 2, 2, 4] 391s -0.152 391s 0.054 391s testing: mul_mat, [1, 1, 1, 1] = [2, 1, 1, 1] * [2, 1, 1, 1] 391s x0: [2, 1, 1, 1] 391s 0.019 0.031 391s 391s x1: [2, 1, 1, 1] 391s -0.241 0.512 391s 391s y: [1, 1, 1, 1] 391s 0.011 391s testing: mul_mat, [4, 2, 2, 1] = [1, 4, 2, 1] * [1, 2, 2, 1] 391s x0: [1, 4, 2, 1] 391s -0.615 391s 0.509 391s 0.395 391s 0.784 391s 391s x1: [1, 2, 2, 1] 391s -0.937 391s 0.571 391s 391s y: [4, 2, 2, 1] 391s 0.576 -0.477 -0.370 -0.734 391s -0.351 0.291 0.226 0.448 391s testing: mul_mat, [2, 1, 2, 4] = [4, 2, 2, 4] * [4, 1, 2, 4] 391s x0: [4, 2, 2, 4] 391s -0.192 0.114 0.242 0.216 391s 0.803 -0.541 0.909 -0.535 391s 391s x1: [4, 1, 2, 4] 391s -0.607 -0.179 0.311 -0.394 391s 391s y: [2, 1, 2, 4] 391s 0.086 0.103 391s test-mul-mat0: iter:492/500 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.601 391s 0.369 391s 0.551 391s -0.455 391s 391s x1: [1, 2, 1, 1] 391s 0.517 391s -0.520 391s 391s y: [4, 2, 1, 1] 391s 0.311 0.191 0.285 -0.235 391s -0.312 -0.192 -0.286 0.236 391s testing: mul_mat, [2, 4, 2, 1] = [1, 2, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 2, 2, 1] 391s 0.353 391s -0.961 391s 391s x1: [1, 4, 2, 1] 391s -0.806 391s 0.716 391s 0.960 391s 0.253 391s 391s y: [2, 4, 2, 1] 391s -0.285 0.774 391s 0.253 -0.688 391s 0.339 -0.922 391s 0.089 -0.243 391s testing: mul_mat, [4, 2, 2, 2] = [1, 4, 2, 2] * [1, 2, 2, 2] 391s x0: [1, 4, 2, 2] 391s -0.592 391s -0.435 391s -0.593 391s -0.112 391s 391s x1: [1, 2, 2, 2] 391s 0.453 391s 0.919 391s 391s y: [4, 2, 2, 2] 391s -0.268 -0.197 -0.269 -0.051 391s -0.544 -0.400 -0.545 -0.103 391s testing: mul_mat, [4, 4, 1, 1] = [1, 4, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.585 391s 0.239 391s -0.556 391s 0.628 391s 391s x1: [1, 4, 1, 1] 391s 0.747 391s -0.980 391s -0.455 391s 0.786 391s 391s y: [4, 4, 1, 1] 391s 0.437 0.179 -0.415 0.469 391s -0.573 -0.235 0.545 -0.616 391s -0.266 -0.109 0.253 -0.286 391s 0.460 0.188 -0.437 0.494 391s testing: mul_mat, [1, 1, 2, 1] = [4, 1, 2, 1] * [4, 1, 2, 1] 391s x0: [4, 1, 2, 1] 391s -0.805 0.164 -0.894 -0.548 391s 391s x1: [4, 1, 2, 1] 391s 0.783 0.636 -0.198 -0.102 391s 391s y: [1, 1, 2, 1] 391s -0.295 391s testing: mul_mat, [4, 4, 2, 2] = [1, 4, 2, 2] * [1, 4, 2, 2] 391s x0: [1, 4, 2, 2] 391s 0.129 391s -0.358 391s -0.949 391s -0.565 391s 391s x1: [1, 4, 2, 2] 391s 0.172 391s -0.394 391s 0.674 391s -0.055 391s 391s y: [4, 4, 2, 2] 391s 0.022 -0.061 -0.163 -0.097 391s -0.051 0.141 0.374 0.222 391s 0.087 -0.241 -0.640 -0.381 391s -0.007 0.020 0.052 0.031 391s test-mul-mat0: iter:493/500 391s testing: mul_mat, [2, 2, 1, 1] = [4, 2, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 2, 1, 1] 391s -0.011 -0.424 -0.176 0.632 391s -0.373 0.259 0.911 -0.224 391s 391s x1: [4, 2, 1, 1] 391s 0.996 -0.268 0.019 0.486 391s -0.339 0.465 0.420 -0.783 391s 391s y: [2, 2, 1, 1] 391s 0.407 -0.533 391s -0.762 0.805 391s testing: mul_mat, [4, 2, 4, 1] = [4, 4, 4, 1] * [4, 2, 4, 1] 391s x0: [4, 4, 4, 1] 391s -0.446 0.077 -0.107 -0.498 391s -0.969 -0.352 -0.464 0.575 391s -0.208 0.602 0.508 -0.252 391s -0.335 -0.827 0.723 0.213 391s 391s x1: [4, 2, 4, 1] 391s 0.889 0.354 0.032 -0.030 391s 0.038 0.947 -0.046 0.754 391s 391s y: [4, 2, 4, 1] 391s -0.357 -1.018 0.053 -0.574 391s -0.314 0.085 0.349 -0.668 391s testing: mul_mat, [3, 4, 4, 1] = [4, 3, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 3, 4, 1] 391s 0.987 0.465 0.360 -0.199 391s 0.598 0.774 -0.323 -0.885 391s -0.798 0.505 0.517 -0.908 391s 391s x1: [4, 4, 4, 1] 391s -0.151 0.452 -0.979 0.476 391s -0.527 -0.712 -0.105 0.096 391s -0.936 0.062 -0.616 -0.987 391s -0.048 -0.115 0.462 0.004 391s 391s y: [3, 4, 4, 1] 391s -0.386 0.155 -0.590 391s -0.908 -0.917 -0.080 391s -0.920 0.560 1.355 391s 0.064 -0.271 0.216 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.433 0.895 0.279 -0.482 391s 391s x1: [4, 3, 1, 1] 391s -0.536 -0.637 -0.802 -0.553 391s 0.225 0.374 0.178 -0.007 391s -0.826 0.772 0.709 0.041 391s 391s y: [1, 3, 1, 1] 391s -0.759 391s 0.486 391s 0.511 391s testing: mul_mat, [3, 4, 4, 1] = [1, 3, 4, 1] * [1, 4, 4, 1] 391s x0: [1, 3, 4, 1] 391s 0.587 391s -0.828 391s 0.105 391s 391s x1: [1, 4, 4, 1] 391s 0.549 391s 0.364 391s 0.376 391s 0.881 391s 391s y: [3, 4, 4, 1] 391s 0.323 -0.455 0.058 391s 0.214 -0.301 0.038 391s 0.221 -0.311 0.040 391s 0.517 -0.729 0.093 391s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 4, 4, 1] 391s -0.096 0.782 0.671 391s -0.165 -0.434 -0.602 391s -0.165 0.632 0.120 391s -0.955 0.954 -0.050 391s 391s x1: [3, 1, 4, 1] 391s -0.389 -0.175 0.764 391s 391s y: [4, 1, 4, 1] 391s 0.413 -0.320 0.045 0.167 391s test-mul-mat0: iter:494/500 391s testing: mul_mat, [4, 2, 1, 1] = [1, 4, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 4, 1, 1] 391s -0.504 391s -0.088 391s 0.691 391s 0.705 391s 391s x1: [1, 2, 1, 1] 391s -0.895 391s -0.054 391s 391s y: [4, 2, 1, 1] 391s 0.451 0.079 -0.618 -0.631 391s 0.027 0.005 -0.037 -0.038 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.564 391s 0.345 391s -0.336 391s 391s x1: [1, 4, 1, 1] 391s 0.105 391s -0.802 391s -0.783 391s -0.798 391s 391s y: [3, 4, 1, 1] 391s 0.059 0.036 -0.035 391s -0.452 -0.276 0.270 391s -0.442 -0.270 0.263 391s -0.450 -0.275 0.268 391s testing: mul_mat, [1, 3, 1, 1] = [1, 1, 1, 1] * [1, 3, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.291 391s 391s x1: [1, 3, 1, 1] 391s 0.477 391s -0.881 391s 0.742 391s 391s y: [1, 3, 1, 1] 391s 0.139 391s -0.256 391s 0.216 391s testing: mul_mat, [4, 1, 1, 1] = [1, 4, 1, 1] * [1, 1, 1, 1] 391s x0: [1, 4, 1, 1] 391s 0.398 391s -0.869 391s 0.402 391s 0.773 391s 391s x1: [1, 1, 1, 1] 391s 0.939 391s 391s y: [4, 1, 1, 1] 391s 0.374 -0.816 0.378 0.726 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.131 0.189 -0.331 0.627 391s 391s x1: [4, 1, 1, 1] 391s -0.377 -0.307 0.026 -0.758 391s 391s y: [1, 1, 1, 1] 391s -0.591 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.549 391s 0.408 391s -0.902 391s 391s x1: [1, 4, 1, 1] 391s -0.900 391s -0.640 391s 0.332 391s 0.205 391s 391s y: [3, 4, 1, 1] 391s -0.494 -0.367 0.812 391s -0.351 -0.261 0.578 391s 0.182 0.135 -0.300 391s 0.113 0.084 -0.185 391s test-mul-mat0: iter:495/500 391s testing: mul_mat, [4, 1, 1, 1] = [4, 4, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 4, 1, 1] 391s 0.248 0.560 -0.427 -0.350 391s 0.333 0.197 0.343 -0.641 391s 0.439 -0.423 0.490 -0.372 391s 0.246 0.117 -0.272 0.606 391s 391s x1: [4, 1, 1, 1] 391s -0.129 -0.497 0.651 -0.838 391s 391s y: [4, 1, 1, 1] 391s -0.295 0.620 0.784 -0.775 391s testing: mul_mat, [2, 4, 4, 1] = [4, 2, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 2, 4, 1] 391s -0.189 0.055 -0.713 0.357 391s -0.040 0.760 0.603 0.829 391s 391s x1: [4, 4, 4, 1] 391s -0.551 0.934 -0.836 0.998 391s 0.341 -0.738 0.111 -0.906 391s 0.024 0.701 -0.035 0.527 391s 0.352 0.127 0.970 -0.400 391s 391s y: [2, 4, 4, 1] 391s 1.108 1.055 391s -0.508 -1.259 391s 0.247 0.947 391s -0.894 0.335 391s testing: mul_mat, [3, 2, 4, 2] = [4, 3, 4, 2] * [4, 2, 4, 2] 391s x0: [4, 3, 4, 2] 391s 0.402 0.806 0.619 -0.421 391s 0.662 0.456 -0.390 0.268 391s -0.774 -0.998 0.866 -0.224 391s 391s x1: [4, 2, 4, 2] 391s -0.759 0.500 -0.269 0.200 391s 0.260 -0.666 0.029 -0.980 391s 391s y: [3, 2, 4, 2] 391s -0.153 -0.116 -0.190 391s -0.002 -0.405 0.709 391s testing: mul_mat, [4, 3, 1, 1] = [4, 4, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.956 -0.632 -0.161 -0.415 391s 0.367 0.880 0.972 0.683 391s 0.790 0.262 -0.603 -0.549 391s -0.235 0.330 -0.890 -0.806 391s 391s x1: [4, 3, 1, 1] 391s -0.722 0.149 0.068 0.229 391s 0.722 -0.593 -0.538 0.012 391s -0.419 0.945 -0.030 0.579 391s 391s y: [4, 3, 1, 1] 391s 0.490 0.088 -0.697 -0.026 391s -0.234 -0.772 0.733 0.103 391s -0.432 1.044 -0.383 -0.030 391s testing: mul_mat, [3, 4, 4, 1] = [4, 3, 4, 1] * [4, 4, 4, 1] 391s x0: [4, 3, 4, 1] 391s 0.729 -0.907 0.449 -0.233 391s 0.464 -0.625 0.581 -0.340 391s 0.719 -0.161 -0.934 0.875 391s 391s x1: [4, 4, 4, 1] 391s 0.109 -0.761 -0.528 -0.742 391s 0.307 0.701 0.980 0.713 391s -0.837 -0.008 -0.706 -0.892 391s 0.962 0.874 -0.419 -0.994 391s 391s y: [3, 4, 4, 1] 391s 0.705 0.472 0.043 391s -0.138 0.031 -0.183 391s -0.712 -0.490 -0.723 391s -0.048 -0.005 0.071 391s testing: mul_mat, [1, 4, 4, 2] = [3, 1, 4, 2] * [3, 4, 4, 2] 391s x0: [3, 1, 4, 2] 391s -0.286 0.212 0.611 391s 391s x1: [3, 4, 4, 2] 391s -0.523 0.161 0.108 391s -0.515 0.607 0.079 391s 0.215 -0.225 -0.731 391s -0.681 0.449 -0.197 391s 391s y: [1, 4, 4, 2] 391s 0.250 391s 0.324 391s -0.556 391s 0.169 391s test-mul-mat0: iter:496/500 391s testing: mul_mat, [3, 3, 1, 1] = [4, 3, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 3, 1, 1] 391s -0.336 -0.698 -0.726 -0.180 391s -0.261 0.918 -0.675 0.124 391s -0.423 0.693 -0.584 0.629 391s 391s x1: [4, 3, 1, 1] 391s 0.209 -0.734 0.065 0.923 391s 0.478 -0.323 -0.674 -0.230 391s 0.864 -0.279 0.143 -0.169 391s 391s y: [3, 3, 1, 1] 391s 0.228 -0.658 -0.055 391s 0.595 0.005 -0.177 391s -0.169 -0.599 -0.749 391s testing: mul_mat, [1, 3, 1, 1] = [4, 1, 1, 1] * [4, 3, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.428 -0.408 0.407 -0.429 391s 391s x1: [4, 3, 1, 1] 391s 0.899 0.978 -0.532 0.228 391s -0.085 -0.014 -0.563 0.181 391s -0.948 -0.640 -0.342 -0.272 391s 391s y: [1, 3, 1, 1] 391s -0.329 391s -0.338 391s -0.168 391s testing: mul_mat, [1, 1, 1, 1] = [4, 1, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 1, 1, 1] 391s -0.945 -0.536 0.895 -0.620 391s 391s x1: [4, 1, 1, 1] 391s 0.423 0.702 0.235 0.725 391s 391s y: [1, 1, 1, 1] 391s -1.015 391s testing: mul_mat, [2, 1, 1, 1] = [4, 2, 1, 1] * [4, 1, 1, 1] 391s x0: [4, 2, 1, 1] 391s 0.971 -0.431 -0.103 0.636 391s 0.982 0.199 0.555 -0.922 391s 391s x1: [4, 1, 1, 1] 391s 0.588 -0.528 -0.928 -0.996 391s 391s y: [2, 1, 1, 1] 391s 0.260 0.876 391s testing: mul_mat, [2, 4, 1, 1] = [2, 2, 1, 1] * [2, 4, 1, 1] 391s x0: [2, 2, 1, 1] 391s -0.650 0.971 391s 0.792 -0.925 391s 391s x1: [2, 4, 1, 1] 391s 0.607 0.996 391s -0.264 -0.665 391s -0.318 -0.836 391s -0.073 -0.910 391s 391s y: [2, 4, 1, 1] 391s 0.572 -0.441 391s -0.474 0.406 391s -0.605 0.522 391s -0.836 0.784 391s testing: mul_mat, [3, 2, 1, 1] = [2, 3, 1, 1] * [2, 2, 1, 1] 391s x0: [2, 3, 1, 1] 391s -0.873 -0.869 391s -0.863 -0.762 391s -0.522 -0.551 391s 391s x1: [2, 2, 1, 1] 391s 0.769 -0.974 391s -0.461 0.665 391s 391s y: [3, 2, 1, 1] 391s 0.175 0.079 0.135 391s -0.175 -0.108 -0.125 391s test-mul-mat0: iter:497/500 391s testing: mul_mat, [4, 4, 1, 1] = [4, 4, 1, 1] * [4, 4, 1, 1] 391s x0: [4, 4, 1, 1] 391s -0.280 -0.806 0.279 0.125 391s -0.680 0.416 0.603 -0.549 391s 0.654 -0.949 -0.436 -0.539 391s 0.699 0.573 -0.177 -0.017 391s 391s x1: [4, 4, 1, 1] 391s -0.638 0.284 0.087 0.969 391s 0.280 0.823 -0.696 0.963 391s 0.988 0.231 -0.948 -0.277 391s 0.580 0.845 -0.306 0.655 391s 391s y: [4, 4, 1, 1] 391s 0.095 0.072 -1.247 -0.315 391s -0.816 -0.796 -0.814 0.775 391s -0.761 -0.994 0.989 0.995 391s -0.847 -0.587 -0.642 0.933 391s testing: mul_mat, [4, 4, 2, 1] = [4, 4, 2, 1] * [4, 4, 2, 1] 391s x0: [4, 4, 2, 1] 391s -0.666 0.809 0.256 -0.440 391s -0.719 0.582 -0.161 0.476 391s 0.489 -0.038 0.366 0.050 391s -0.301 0.975 -0.195 0.715 391s 391s x1: [4, 4, 2, 1] 391s -0.339 -0.208 -0.737 -0.516 391s 0.096 -0.774 -0.528 -0.674 391s -0.722 0.195 0.907 -0.877 391s 0.889 0.562 0.737 -0.391 391s 391s y: [4, 4, 2, 1] 391s 0.096 -0.004 -0.453 -0.326 391s -0.529 -0.755 -0.151 -1.162 391s 1.256 0.069 -0.073 -0.396 391s 0.223 -0.618 0.663 -0.143 391s testing: mul_mat, [2, 4, 2, 2] = [4, 2, 2, 2] * [4, 4, 2, 2] 391s x0: [4, 2, 2, 2] 391s 0.635 -0.614 0.462 0.491 391s -0.966 0.180 -0.234 0.365 391s 391s x1: [4, 4, 2, 2] 391s -0.703 0.562 -0.514 -0.422 391s 0.144 0.325 -0.946 -0.367 391s -0.714 0.420 0.683 -0.014 391s 0.394 -0.512 -0.299 0.103 391s 391s y: [2, 4, 2, 2] 391s -1.236 0.746 391s -0.725 0.007 391s -0.402 0.599 391s 0.477 -0.365 391s testing: mul_mat, [1, 2, 1, 1] = [4, 1, 1, 1] * [4, 2, 1, 1] 391s x0: [4, 1, 1, 1] 391s 0.092 -0.687 -0.757 0.465 391s 391s x1: [4, 2, 1, 1] 391s -0.492 0.960 -0.167 -0.458 391s 0.140 0.599 0.907 0.024 391s 391s y: [1, 2, 1, 1] 391s -0.792 391s -1.075 391s testing: mul_mat, [3, 4, 2, 1] = [1, 3, 2, 1] * [1, 4, 2, 1] 391s x0: [1, 3, 2, 1] 391s -0.967 391s 0.793 391s -0.474 391s 391s x1: [1, 4, 2, 1] 391s 0.853 391s 0.373 391s 0.804 391s -0.276 391s 391s y: [3, 4, 2, 1] 391s -0.824 0.676 -0.404 391s -0.361 0.296 -0.177 391s -0.777 0.637 -0.381 391s 0.267 -0.219 0.131 391s testing: mul_mat, [2, 1, 2, 2] = [3, 2, 2, 2] * [3, 1, 2, 2] 391s x0: [3, 2, 2, 2] 391s 0.195 -0.582 -0.340 391s 0.287 0.048 -0.778 391s 391s x1: [3, 1, 2, 2] 391s -0.095 -0.242 0.576 391s 391s y: [2, 1, 2, 2] 391s -0.074 -0.487 391s test-mul-mat0: iter:498/500 391s testing: mul_mat, [1, 4, 1, 1] = [3, 1, 1, 1] * [3, 4, 1, 1] 391s x0: [3, 1, 1, 1] 391s -0.414 -0.033 -0.655 391s 391s x1: [3, 4, 1, 1] 391s 0.186 -0.942 -0.542 391s -0.619 0.345 -0.124 391s 0.429 -0.995 0.098 391s 0.201 0.671 -0.293 391s 391s y: [1, 4, 1, 1] 391s 0.309 391s 0.326 391s -0.209 391s 0.087 391s testing: mul_mat, [4, 1, 3, 1] = [3, 4, 3, 1] * [3, 1, 3, 1] 391s x0: [3, 4, 3, 1] 391s -0.623 0.326 -0.265 391s 0.391 -0.169 0.920 391s 0.450 0.289 -0.699 391s -0.205 -0.836 0.730 391s 391s x1: [3, 1, 3, 1] 391s 0.294 0.754 0.717 391s 391s y: [4, 1, 3, 1] 391s -0.128 0.647 -0.152 -0.167 391s testing: mul_mat, [3, 4, 3, 3] = [3, 3, 3, 3] * [3, 4, 3, 3] 391s x0: [3, 3, 3, 3] 391s -0.168 0.395 -0.083 391s -0.996 -0.326 0.380 391s -0.553 0.716 -0.592 391s 391s x1: [3, 4, 3, 3] 391s 0.687 -0.947 0.872 391s 0.988 -0.152 -0.964 391s 0.718 0.648 0.299 391s -0.351 -0.881 0.269 391s 391s y: [3, 4, 3, 3] 391s -0.562 -0.045 -1.574 391s -0.147 -1.301 -0.085 391s 0.110 -0.813 -0.110 391s -0.311 0.738 -0.596 391s testing: mul_mat, [3, 3, 1, 1] = [3, 3, 1, 1] * [3, 3, 1, 1] 391s x0: [3, 3, 1, 1] 391s -0.494 -0.352 0.064 391s -0.305 0.945 0.394 391s -0.511 0.853 0.905 391s 391s x1: [3, 3, 1, 1] 391s -0.262 -0.188 -0.620 391s 0.001 0.626 0.736 391s -0.519 0.963 -0.735 391s 391s y: [3, 3, 1, 1] 391s 0.156 -0.343 -0.588 391s -0.174 0.882 1.199 391s -0.130 0.779 0.422 391s testing: mul_mat, [1, 3, 3, 1] = [3, 1, 3, 1] * [3, 3, 3, 1] 391s x0: [3, 1, 3, 1] 391s -0.603 0.406 0.957 391s 391s x1: [3, 3, 3, 1] 391s 0.469 0.451 -0.420 391s -0.207 -0.842 0.891 391s -0.644 -0.571 0.747 391s 391s y: [1, 3, 3, 1] 391s -0.502 391s 0.636 391s 0.872 391s testing: mul_mat, [1, 3, 3, 3] = [1, 1, 3, 3] * [1, 3, 3, 3] 391s x0: [1, 1, 3, 3] 391s 0.154 391s 391s x1: [1, 3, 3, 3] 391s 0.015 391s 0.445 391s -0.825 391s 391s y: [1, 3, 3, 3] 391s 0.002 391s 0.069 391s -0.127 391s test-mul-mat0: iter:499/500 391s testing: mul_mat, [1, 2, 1, 1] = [1, 1, 1, 1] * [1, 2, 1, 1] 391s x0: [1, 1, 1, 1] 391s 0.097 391s 391s x1: [1, 2, 1, 1] 391s -0.115 391s -0.391 391s 391s y: [1, 2, 1, 1] 391s -0.011 391s -0.038 391s testing: mul_mat, [4, 1, 4, 1] = [1, 4, 4, 1] * [1, 1, 4, 1] 391s x0: [1, 4, 4, 1] 391s -0.776 391s 0.796 391s 0.573 391s -0.148 391s 391s x1: [1, 1, 4, 1] 391s 0.623 391s 391s y: [4, 1, 4, 1] 391s -0.483 0.496 0.357 -0.092 391s testing: mul_mat, [4, 4, 4, 2] = [1, 4, 4, 2] * [1, 4, 4, 2] 391s x0: [1, 4, 4, 2] 391s -0.434 391s 0.259 391s 0.574 391s 0.023 391s 391s x1: [1, 4, 4, 2] 391s -0.122 391s 0.144 391s 0.287 391s 0.732 391s 391s y: [4, 4, 4, 2] 391s 0.053 -0.032 -0.070 -0.003 391s -0.062 0.037 0.082 0.003 391s -0.124 0.074 0.165 0.007 391s -0.318 0.190 0.420 0.017 391s testing: mul_mat, [3, 4, 1, 1] = [1, 3, 1, 1] * [1, 4, 1, 1] 391s x0: [1, 3, 1, 1] 391s 0.799 391s -0.840 391s -0.539 391s 391s x1: [1, 4, 1, 1] 391s -0.308 391s 0.534 391s -0.057 391s -0.431 391s 391s y: [3, 4, 1, 1] 391s -0.246 0.258 0.166 391s 0.426 -0.449 -0.288 391s -0.045 0.048 0.031 391s -0.344 0.362 0.232 391s testing: mul_mat, [4, 1, 4, 1] = [3, 4, 4, 1] * [3, 1, 4, 1] 391s x0: [3, 4, 4, 1] 391s 0.616 0.929 -0.493 391s -0.978 -0.453 -0.035 391s -0.949 0.044 -0.946 391s 0.691 0.362 -0.959 391s 391s x1: [3, 1, 4, 1] 391s 0.778 0.458 -0.860 391s 391s y: [4, 1, 4, 1] 391s 1.328 -0.937 0.095 1.527 391s testing: mul_mat, [4, 3, 4, 2] = [4, 4, 4, 2] * [4, 3, 4, 2] 391s x0: [4, 4, 4, 2] 391s -0.098 0.579 -0.110 0.879 391s -0.433 0.923 -0.015 0.200 391s 0.544 -0.770 0.666 0.494 391s -0.418 0.165 0.131 -0.611 391s 391s x1: [4, 3, 4, 2] 391s 0.187 -0.115 0.229 0.655 391s -0.346 -0.844 -0.215 0.320 391s 0.784 -0.766 0.291 -0.746 391s 391s y: [4, 3, 4, 2] 391s 0.466 -0.060 0.667 -0.468 391s -0.150 -0.562 0.476 -0.218 391s -1.208 -1.200 0.841 0.040 393s compression: 0.000000 393s convert time: 0.000000 ms / method = 0 393s -0.040188 0.785085 393s 0.405617 0.183192 393s 0.016901 0.564737 393s 0.001560 0.452003 393s -0.111647 -0.158779 393s 0.602449 0.452541 393s 0.464777 0.596264 393s 0.031770 0.389379 393s 0.522225 -0.094984 393s 0.246030 0.173205 393s 0.322603 -0.063522 393s 0.171129 0.240454 393s 0.435216 0.577889 393s 0.286599 -0.102425 393s -0.152230 0.570174 393s -0.116195 0.552043 393s main: elapsed ticks: 1833983 393s main: elapsed us: 1836413 / 1836.413000 ms 393s 135.421090 393s Testing 2 devices 393s 393s Backend 1/2: BLAS 393s Device description: BLAS 393s Device memory: 0 MB (0 MB free) 393s 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=1): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=2): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=3): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=4): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=5): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=6): OK 393s test_dataset(shuffle=no, ndata_shard=2, ndata_batch=2): OK 393s test_dataset(shuffle=no, ndata_shard=2, ndata_batch=4): OK 393s test_dataset(shuffle=no, ndata_shard=2, ndata_batch=6): OK 393s test_dataset(shuffle=no, ndata_shard=3, ndata_batch=3): OK 393s test_dataset(shuffle=no, ndata_shard=3, ndata_batch=6): OK 393s test_dataset(shuffle=no, ndata_shard=4, ndata_batch=4): OK 393s test_dataset(shuffle=no, ndata_shard=5, ndata_batch=5): OK 393s test_dataset(shuffle=no, ndata_shard=6, ndata_batch=6): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=1): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=2): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=3): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=4): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=5): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=6): OK 393s test_dataset(shuffle=yes, ndata_shard=2, ndata_batch=2): OK 393s test_dataset(shuffle=yes, ndata_shard=2, ndata_batch=4): OK 393s test_dataset(shuffle=yes, ndata_shard=2, ndata_batch=6): OK 393s test_dataset(shuffle=yes, ndata_shard=3, ndata_batch=3): OK 393s test_dataset(shuffle=yes, ndata_shard=3, ndata_batch=6): OK 393s test_dataset(shuffle=yes, ndata_shard=4, ndata_batch=4): OK 393s test_dataset(shuffle=yes, ndata_shard=5, ndata_batch=5): OK 393s test_dataset(shuffle=yes, ndata_shard=6, ndata_batch=6): OK 393s test_grad(): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=results_initial): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=weights_after_forward): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=results_after_forward): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=weights_after_forward_backward): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=result_after_forward_backward): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=results_initial): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=weights_after_forward): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=results_after_forward): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=weights_after_forward_backward): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=result_after_forward_backward): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=results_initial): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=weights_after_forward): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=results_after_forward): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=weights_after_forward_backward): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=result_after_forward_backward): OK 393s test_epoch_vs_fit(): OK 393s test_idata_split(high_level=no, epoch=1, subtest=weights): OK 393s test_idata_split(high_level=no, epoch=1, subtest=results_backward): OK 393s test_idata_split(high_level=no, epoch=1, subtest=results_forward): OK 393s test_idata_split(high_level=no, epoch=2, subtest=weights): OK 393s test_idata_split(high_level=no, epoch=2, subtest=results_backward): OK 393s test_idata_split(high_level=no, epoch=2, subtest=results_forward): OK 393s test_idata_split(high_level=no, epoch=3, subtest=weights): OK 393s test_idata_split(high_level=no, epoch=3, subtest=results_backward): OK 393s test_idata_split(high_level=no, epoch=3, subtest=results_forward): OK 393s test_idata_split(high_level=no, epoch=4, subtest=weights): OK 393s test_idata_split(high_level=no, epoch=4, subtest=results_backward): OK 393s test_idata_split(high_level=no, epoch=4, subtest=results_forward): OK 393s test_idata_split(high_level=yes, epoch=1, subtest=weights): OK 393s test_idata_split(high_level=yes, epoch=1, subtest=results_backward): OK 393s test_idata_split(high_level=yes, epoch=1, subtest=results_forward): OK 393s test_idata_split(high_level=yes, epoch=2, subtest=weights): OK 393s test_idata_split(high_level=yes, epoch=2, subtest=results_backward): OK 393s test_idata_split(high_level=yes, epoch=2, subtest=results_forward): OK 393s test_idata_split(high_level=yes, epoch=3, subtest=weights): OK 393s test_idata_split(high_level=yes, epoch=3, subtest=results_backward): OK 393s test_idata_split(high_level=yes, epoch=3, subtest=results_forward): OK 393s test_idata_split(high_level=yes, epoch=4, subtest=weights): OK 393s test_idata_split(high_level=yes, epoch=4, subtest=results_backward): OK 393s test_idata_split(high_level=yes, epoch=4, subtest=results_forward): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=1, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=1, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=1, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=2, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=2, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=2, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=3, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=3, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=3, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=4, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=4, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=4, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=1, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=1, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=1, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=2, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=2, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=2, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=3, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=3, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=3, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=4, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=4, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=4, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=1, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=1, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=1, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=2, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=2, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=2, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=3, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=3, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=3, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=4, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=4, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=4, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=1, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=1, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=1, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=2, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=2, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=2, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=3, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=3, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=3, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=4, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=4, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=4, subtest=results): OK 393s test_regression(subtest=weights): OK 393s 118/118 tests passed 393s Backend BLAS: OK 393s 393s Backend 2/2: CPU 393s Device description: CPU 393s Device memory: 3897 MB (3897 MB free) 393s 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=1): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=2): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=3): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=4): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=5): OK 393s test_dataset(shuffle=no, ndata_shard=1, ndata_batch=6): OK 393s test_dataset(shuffle=no, ndata_shard=2, ndata_batch=2): OK 393s test_dataset(shuffle=no, ndata_shard=2, ndata_batch=4): OK 393s test_dataset(shuffle=no, ndata_shard=2, ndata_batch=6): OK 393s test_dataset(shuffle=no, ndata_shard=3, ndata_batch=3): OK 393s test_dataset(shuffle=no, ndata_shard=3, ndata_batch=6): OK 393s test_dataset(shuffle=no, ndata_shard=4, ndata_batch=4): OK 393s test_dataset(shuffle=no, ndata_shard=5, ndata_batch=5): OK 393s test_dataset(shuffle=no, ndata_shard=6, ndata_batch=6): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=1): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=2): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=3): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=4): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=5): OK 393s test_dataset(shuffle=yes, ndata_shard=1, ndata_batch=6): OK 393s test_dataset(shuffle=yes, ndata_shard=2, ndata_batch=2): OK 393s test_dataset(shuffle=yes, ndata_shard=2, ndata_batch=4): OK 393s test_dataset(shuffle=yes, ndata_shard=2, ndata_batch=6): OK 393s test_dataset(shuffle=yes, ndata_shard=3, ndata_batch=3): OK 393s test_dataset(shuffle=yes, ndata_shard=3, ndata_batch=6): OK 393s test_dataset(shuffle=yes, ndata_shard=4, ndata_batch=4): OK 393s test_dataset(shuffle=yes, ndata_shard=5, ndata_batch=5): OK 393s test_dataset(shuffle=yes, ndata_shard=6, ndata_batch=6): OK 393s test_grad(): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=results_initial): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=weights_after_forward): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=results_after_forward): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=weights_after_forward_backward): OK 393s test_forward_backward(high_level=no, shuffle=no, subtest=result_after_forward_backward): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=results_initial): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=weights_after_forward): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=results_after_forward): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=weights_after_forward_backward): OK 393s test_forward_backward(high_level=yes, shuffle=no, subtest=result_after_forward_backward): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=results_initial): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=weights_after_forward): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=results_after_forward): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=weights_after_forward_backward): OK 393s test_forward_backward(high_level=yes, shuffle=yes, subtest=result_after_forward_backward): OK 393s test_epoch_vs_fit(): OK 393s test_idata_split(high_level=no, epoch=1, subtest=weights): OK 393s test_idata_split(high_level=no, epoch=1, subtest=results_backward): OK 393s test_idata_split(high_level=no, epoch=1, subtest=results_forward): OK 393s test_idata_split(high_level=no, epoch=2, subtest=weights): OK 393s test_idata_split(high_level=no, epoch=2, subtest=results_backward): OK 393s test_idata_split(high_level=no, epoch=2, subtest=results_forward): OK 393s test_idata_split(high_level=no, epoch=3, subtest=weights): OK 393s test_idata_split(high_level=no, epoch=3, subtest=results_backward): OK 393s test_idata_split(high_level=no, epoch=3, subtest=results_forward): OK 393s test_idata_split(high_level=no, epoch=4, subtest=weights): OK 393s test_idata_split(high_level=no, epoch=4, subtest=results_backward): OK 393s test_idata_split(high_level=no, epoch=4, subtest=results_forward): OK 393s test_idata_split(high_level=yes, epoch=1, subtest=weights): OK 393s test_idata_split(high_level=yes, epoch=1, subtest=results_backward): OK 393s test_idata_split(high_level=yes, epoch=1, subtest=results_forward): OK 393s test_idata_split(high_level=yes, epoch=2, subtest=weights): OK 393s test_idata_split(high_level=yes, epoch=2, subtest=results_backward): OK 393s test_idata_split(high_level=yes, epoch=2, subtest=results_forward): OK 393s test_idata_split(high_level=yes, epoch=3, subtest=weights): OK 393s test_idata_split(high_level=yes, epoch=3, subtest=results_backward): OK 393s test_idata_split(high_level=yes, epoch=3, subtest=results_forward): OK 393s test_idata_split(high_level=yes, epoch=4, subtest=weights): OK 393s test_idata_split(high_level=yes, epoch=4, subtest=results_backward): OK 393s test_idata_split(high_level=yes, epoch=4, subtest=results_forward): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=1, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=1, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=1, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=2, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=2, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=2, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=3, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=3, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=3, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=4, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=4, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=sum, epoch=4, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=1, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=1, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=1, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=2, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=2, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=2, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=3, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=3, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=3, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=4, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=4, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=2, loss_type=mean, epoch=4, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=1, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=1, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=1, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=2, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=2, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=2, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=3, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=3, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=3, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=4, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=4, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=sum, epoch=4, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=1, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=1, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=1, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=2, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=2, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=2, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=3, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=3, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=3, subtest=results): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=4, subtest=grads): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=4, subtest=weights): OK 393s test_gradient_accumulation(high_level=no, nbatch_physical=1, loss_type=mean, epoch=4, subtest=results): OK 393s test_regression(subtest=weights): OK 393s 118/118 tests passed 393s Backend CPU: OK 393s 393s 2/2 backends passed 393s OK 393s test_pad_reflect_1d: using CPU backend 410s Testing f32 410s Testing f16 410s Testing q4_0 410s Testing q4_1 410s Testing q5_0 410s Testing q5_1 410s Testing q8_0 410s Testing q8_1 410s Testing q2_K 410s Testing q3_K 410s Testing q4_K 410s Testing q5_K 410s Testing q6_K 410s Testing q8_K 410s Testing iq2_xxs 410s Testing iq2_xs 410s Testing iq3_xxs 410s Testing iq1_s 410s Testing iq4_nl 410s Testing iq3_s 410s Testing iq2_s 410s Testing iq4_xs 410s Testing i8 410s Testing i16 410s Testing i32 410s Testing i64 410s Testing f64 410s Testing iq1_m 410s Testing bf16 410s Testing tq1_0 410s Testing tq2_0 410s f16 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 6.63 GB/s 410s quantized throughput : 3.32 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 6.10 GB/s 410s quantized throughput : 3.05 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 6.10 GB/s 410s quantized throughput : 3.05 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 6.10 GB/s 410s quantized throughput : 3.05 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 11.74 GB/s 410s quantized throughput : 5.87 GB/s 410s 410s q4_0 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.63 GB/s 410s quantized throughput : 0.51 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.63 GB/s 410s quantized throughput : 0.51 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 11.74 GB/s 410s quantized throughput : 1.65 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.32 GB/s 410s quantized throughput : 0.47 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 30.52 GB/s 410s quantized throughput : 4.29 GB/s 410s 410s q4_1 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.06 GB/s 410s quantized throughput : 0.32 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.06 GB/s 410s quantized throughput : 0.32 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 5.26 GB/s 410s quantized throughput : 0.82 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.38 GB/s 410s quantized throughput : 0.37 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 25.43 GB/s 410s quantized throughput : 3.97 GB/s 410s 410s q5_0 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.77 GB/s 410s quantized throughput : 0.48 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.77 GB/s 410s quantized throughput : 0.48 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 7.63 GB/s 410s quantized throughput : 1.31 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.32 GB/s 410s quantized throughput : 0.57 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 21.80 GB/s 410s quantized throughput : 3.75 GB/s 410s 410s q5_1 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 1.41 GB/s 410s quantized throughput : 0.26 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 1.50 GB/s 410s quantized throughput : 0.28 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 8.48 GB/s 410s quantized throughput : 1.59 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.38 GB/s 410s quantized throughput : 0.45 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 19.07 GB/s 410s quantized throughput : 3.58 GB/s 410s 410s q8_0 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 1.66 GB/s 410s quantized throughput : 0.44 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.25 GB/s 410s quantized throughput : 0.86 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 11.74 GB/s 410s quantized throughput : 3.12 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.32 GB/s 410s quantized throughput : 0.88 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 30.52 GB/s 410s quantized throughput : 8.11 GB/s 410s 410s q2_K 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.05 GB/s 410s quantized throughput : 0.00 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.05 GB/s 410s quantized throughput : 0.00 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 8.98 GB/s 410s quantized throughput : 0.74 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.32 GB/s 410s quantized throughput : 0.27 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 10.90 GB/s 410s quantized throughput : 0.89 GB/s 410s 410s q3_K 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.55 GB/s 410s quantized throughput : 0.06 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.55 GB/s 410s quantized throughput : 0.06 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 10.90 GB/s 410s quantized throughput : 1.17 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.39 GB/s 410s quantized throughput : 0.36 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 19.07 GB/s 410s quantized throughput : 2.05 GB/s 410s 410s q4_K 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.04 GB/s 410s quantized throughput : 0.01 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.04 GB/s 410s quantized throughput : 0.01 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 10.17 GB/s 410s quantized throughput : 1.43 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.32 GB/s 410s quantized throughput : 0.47 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 30.52 GB/s 410s quantized throughput : 4.29 GB/s 410s 410s q5_K 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.05 GB/s 410s quantized throughput : 0.01 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.05 GB/s 410s quantized throughput : 0.01 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 8.03 GB/s 410s quantized throughput : 1.38 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.39 GB/s 410s quantized throughput : 0.58 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 21.80 GB/s 410s quantized throughput : 3.75 GB/s 410s 410s q6_K 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.07 GB/s 410s quantized throughput : 0.02 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.08 GB/s 410s quantized throughput : 0.02 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.93 GB/s 410s quantized throughput : 0.60 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.88 GB/s 410s quantized throughput : 0.59 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 19.07 GB/s 410s quantized throughput : 3.91 GB/s 410s 410s iq4_nl 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.19 GB/s 410s quantized throughput : 0.03 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.19 GB/s 410s quantized throughput : 0.03 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 4.49 GB/s 410s quantized throughput : 0.63 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.32 GB/s 410s quantized throughput : 0.47 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 25.43 GB/s 410s quantized throughput : 3.58 GB/s 410s 410s iq4_xs 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.02 GB/s 410s quantized throughput : 0.00 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.02 GB/s 410s quantized throughput : 0.00 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 4.24 GB/s 410s quantized throughput : 0.56 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.25 GB/s 410s quantized throughput : 0.43 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 30.52 GB/s 410s quantized throughput : 4.05 GB/s 410s 410s bf16 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.28 GB/s 410s quantized throughput : 1.14 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.28 GB/s 410s quantized throughput : 1.14 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 5.09 GB/s 410s quantized throughput : 2.54 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.28 GB/s 410s quantized throughput : 1.14 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 2.18 GB/s 410s quantized throughput : 1.09 GB/s 410s 410s tq1_0 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 1.01 GB/s 410s quantized throughput : 0.05 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 1.02 GB/s 410s quantized throughput : 0.05 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 7.63 GB/s 410s quantized throughput : 0.40 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.32 GB/s 410s quantized throughput : 0.17 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 38.15 GB/s 410s quantized throughput : 2.01 GB/s 410s 410s tq2_0 410s quantize_row_q_reference 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 1.01 GB/s 410s quantized throughput : 0.07 GB/s 410s 410s quantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 0.95 GB/s 410s quantized throughput : 0.06 GB/s 410s 410s dequantize_row_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 9.54 GB/s 410s quantized throughput : 0.61 GB/s 410s 410s quantize_row_q_dot 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 3.47 GB/s 410s quantized throughput : 0.22 GB/s 410s 410s vec_dot_q 410s 4096 values (0.02 MB) 410s min cycles/32 vals : 0.00 410s avg cycles/32 vals : 0.00 410s float32 throughput : 50.86 GB/s 410s quantized throughput : 3.28 GB/s 410s 411s ggml_roll(3(1), 7(0), 4(-1), 2(0), contiguous): PASSED 411s ggml_roll(37(-4), 42(3), 59(-7), 2(1), contiguous): PASSED 411s ggml_roll(37(-4), 42(3), 59(-7), 2(1), permuted): PASSED 411s 0.8439 -0.9970 0.6497 0.9733 0.9981 0.9999 1.0000 -0.5366 -0.0776 0.7602 0.2296 0.0621 0.0167 0.0045 0.0000 0.0000 0.4242 0.9880 -0.1558 0.8946 0.9923 0.9994 1.0000 -0.9056 0.1547 0.9878 0.4470 0.1240 0.0333 0.0089 0.0000 0.0000 411s ----------------------------------- 411s 0.8439 -0.9970 0.6497 0.9733 0.9981 0.9999 1.0000 -0.5366 -0.0776 0.7602 0.2296 0.0621 0.0167 0.0045 0.0000 0.0000 0.4242 0.9880 -0.1558 0.8946 0.9923 0.9994 1.0000 -0.9056 0.1547 0.9878 0.4470 0.1240 0.0333 0.0089 0.0000 0.0000 413s main: elapsed ticks: 1978106 413s 2951358968546669887488.000000 413s ggml_print_objects: objects in context 0xbec4fe287930: 413s - ggml_object: type = 0, offset = 32, size = 384, next = 0xfc6ca1fef1e0 413s - ggml_object: type = 0, offset = 448, size = 736, next = 0xfc6ca1fef4e0 413s - ggml_object: type = 0, offset = 1216, size = 24336, next = (nil) 413s ggml_print_objects: --- end --- 413s autopkgtest [19:41:03]: test libggml-cpu: -----------------------] 414s libggml-cpu PASS 414s autopkgtest [19:41:04]: test libggml-cpu: - - - - - - - - - - results - - - - - - - - - - 415s autopkgtest [19:41:05]: test libggml-hip: preparing testbed 438s Creating nova instance adt-questing-arm64-ggml-20250707-193409-juju-7f2275-prod-proposed-migration-environment-2-954b271b-a1cd-49e8-988e-09a11edc8a51 from image adt/ubuntu-questing-arm64-server-20250707.img (UUID 5a1f42c1-3f6f-4084-8e72-afd42fd719a8)... 496s autopkgtest [19:42:26]: testbed dpkg architecture: arm64 496s autopkgtest [19:42:26]: testbed apt version: 3.1.3 497s autopkgtest [19:42:27]: @@@@@@@@@@@@@@@@@@@@ test bed setup 497s autopkgtest [19:42:27]: testbed release detected to be: questing 498s autopkgtest [19:42:28]: updating testbed package index (apt update) 498s Get:1 http://ftpmaster.internal/ubuntu questing-proposed InRelease [249 kB] 498s Hit:2 http://ftpmaster.internal/ubuntu questing InRelease 498s Hit:3 http://ftpmaster.internal/ubuntu questing-updates InRelease 498s Hit:4 http://ftpmaster.internal/ubuntu questing-security InRelease 498s Get:5 http://ftpmaster.internal/ubuntu questing-proposed/multiverse Sources [15.8 kB] 498s Get:6 http://ftpmaster.internal/ubuntu questing-proposed/main Sources [27.5 kB] 498s Get:7 http://ftpmaster.internal/ubuntu questing-proposed/universe Sources [177 kB] 498s Get:8 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 Packages [38.5 kB] 498s Get:9 http://ftpmaster.internal/ubuntu questing-proposed/universe arm64 Packages [154 kB] 499s Get:10 http://ftpmaster.internal/ubuntu questing-proposed/multiverse arm64 Packages [15.7 kB] 499s Fetched 677 kB in 1s (808 kB/s) 500s Reading package lists... 500s autopkgtest [19:42:30]: upgrading testbed (apt dist-upgrade and autopurge) 501s Reading package lists... 501s Building dependency tree... 501s Reading state information... 501s Calculating upgrade... 502s The following packages will be upgraded: 502s apparmor libapparmor1 libc-bin libc-dev-bin libc6 libc6-dev libnss-systemd 502s libpam-systemd libsystemd-shared libsystemd0 libudev1 libxml2-16 locales 502s systemd systemd-cryptsetup systemd-resolved systemd-sysv systemd-timesyncd 502s udev 502s 19 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. 502s Need to get 19.1 MB of archives. 502s After this operation, 34.8 kB disk space will be freed. 502s Get:1 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc6-dev arm64 2.41-9ubuntu1 [1749 kB] 502s Get:2 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc-dev-bin arm64 2.41-9ubuntu1 [22.7 kB] 502s Get:3 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc6 arm64 2.41-9ubuntu1 [2919 kB] 502s Get:4 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 libc-bin arm64 2.41-9ubuntu1 [599 kB] 502s Get:5 http://ftpmaster.internal/ubuntu questing/main arm64 libsystemd0 arm64 257.7-1ubuntu1 [522 kB] 503s Get:6 http://ftpmaster.internal/ubuntu questing/main arm64 libnss-systemd arm64 257.7-1ubuntu1 [170 kB] 503s Get:7 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-sysv arm64 257.7-1ubuntu1 [11.9 kB] 503s Get:8 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-resolved arm64 257.7-1ubuntu1 [324 kB] 503s Get:9 http://ftpmaster.internal/ubuntu questing/main arm64 libpam-systemd arm64 257.7-1ubuntu1 [254 kB] 503s Get:10 http://ftpmaster.internal/ubuntu questing/main arm64 libsystemd-shared arm64 257.7-1ubuntu1 [2239 kB] 503s Get:11 http://ftpmaster.internal/ubuntu questing/main arm64 systemd arm64 257.7-1ubuntu1 [2939 kB] 503s Get:12 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-timesyncd arm64 257.7-1ubuntu1 [41.0 kB] 503s Get:13 http://ftpmaster.internal/ubuntu questing/main arm64 systemd-cryptsetup arm64 257.7-1ubuntu1 [121 kB] 503s Get:14 http://ftpmaster.internal/ubuntu questing/main arm64 udev arm64 257.7-1ubuntu1 [1407 kB] 503s Get:15 http://ftpmaster.internal/ubuntu questing/main arm64 libudev1 arm64 257.7-1ubuntu1 [203 kB] 503s Get:16 http://ftpmaster.internal/ubuntu questing/main arm64 libapparmor1 arm64 4.1.1-0ubuntu3 [56.2 kB] 503s Get:17 http://ftpmaster.internal/ubuntu questing/main arm64 libxml2-16 arm64 2.14.4+dfsg-0exp1 [595 kB] 503s Get:18 http://ftpmaster.internal/ubuntu questing-proposed/main arm64 locales all 2.41-9ubuntu1 [4253 kB] 503s Get:19 http://ftpmaster.internal/ubuntu questing/main arm64 apparmor arm64 4.1.1-0ubuntu3 [666 kB] 503s Preconfiguring packages ... 504s Fetched 19.1 MB in 1s (14.8 MB/s) 504s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 504s Preparing to unpack .../libc6-dev_2.41-9ubuntu1_arm64.deb ... 504s Unpacking libc6-dev:arm64 (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 504s Preparing to unpack .../libc-dev-bin_2.41-9ubuntu1_arm64.deb ... 504s Unpacking libc-dev-bin (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 504s Preparing to unpack .../libc6_2.41-9ubuntu1_arm64.deb ... 504s Unpacking libc6:arm64 (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 504s Setting up libc6:arm64 (2.41-9ubuntu1) ... 505s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 505s Preparing to unpack .../libc-bin_2.41-9ubuntu1_arm64.deb ... 505s Unpacking libc-bin (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 505s Setting up libc-bin (2.41-9ubuntu1) ... 505s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 505s Preparing to unpack .../libsystemd0_257.7-1ubuntu1_arm64.deb ... 505s Unpacking libsystemd0:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 505s Setting up libsystemd0:arm64 (257.7-1ubuntu1) ... 505s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 505s Preparing to unpack .../libnss-systemd_257.7-1ubuntu1_arm64.deb ... 505s Unpacking libnss-systemd:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 505s Preparing to unpack .../systemd-sysv_257.7-1ubuntu1_arm64.deb ... 505s Unpacking systemd-sysv (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 505s Preparing to unpack .../systemd-resolved_257.7-1ubuntu1_arm64.deb ... 505s Unpacking systemd-resolved (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 505s Preparing to unpack .../libpam-systemd_257.7-1ubuntu1_arm64.deb ... 505s Unpacking libpam-systemd:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 505s Preparing to unpack .../libsystemd-shared_257.7-1ubuntu1_arm64.deb ... 505s Unpacking libsystemd-shared:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 505s Setting up libsystemd-shared:arm64 (257.7-1ubuntu1) ... 505s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127317 files and directories currently installed.) 505s Preparing to unpack .../systemd_257.7-1ubuntu1_arm64.deb ... 506s Unpacking systemd (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 506s Preparing to unpack .../systemd-timesyncd_257.7-1ubuntu1_arm64.deb ... 506s Unpacking systemd-timesyncd (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 506s Preparing to unpack .../systemd-cryptsetup_257.7-1ubuntu1_arm64.deb ... 506s Unpacking systemd-cryptsetup (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 506s Preparing to unpack .../udev_257.7-1ubuntu1_arm64.deb ... 506s Unpacking udev (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 506s Preparing to unpack .../libudev1_257.7-1ubuntu1_arm64.deb ... 506s Unpacking libudev1:arm64 (257.7-1ubuntu1) over (257.6-1ubuntu1) ... 506s Setting up libudev1:arm64 (257.7-1ubuntu1) ... 506s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127318 files and directories currently installed.) 506s Preparing to unpack .../libapparmor1_4.1.1-0ubuntu3_arm64.deb ... 506s Unpacking libapparmor1:arm64 (4.1.1-0ubuntu3) over (4.1.1-0ubuntu2) ... 506s Preparing to unpack .../libxml2-16_2.14.4+dfsg-0exp1_arm64.deb ... 506s Unpacking libxml2-16:arm64 (2.14.4+dfsg-0exp1) over (2.14.3+dfsg-0exp3) ... 506s Preparing to unpack .../locales_2.41-9ubuntu1_all.deb ... 506s Unpacking locales (2.41-9ubuntu1) over (2.41-6ubuntu2) ... 507s Preparing to unpack .../apparmor_4.1.1-0ubuntu3_arm64.deb ... 507s Unpacking apparmor (4.1.1-0ubuntu3) over (4.1.1-0ubuntu2) ... 508s Setting up libapparmor1:arm64 (4.1.1-0ubuntu3) ... 508s Setting up libxml2-16:arm64 (2.14.4+dfsg-0exp1) ... 508s Setting up systemd (257.7-1ubuntu1) ... 508s /usr/lib/tmpfiles.d/legacy.conf:14: Duplicate line for path "/run/lock", ignoring. 508s /usr/lib/tmpfiles.d/legacy.conf:14: Duplicate line for path "/run/lock", ignoring. 509s Setting up locales (2.41-9ubuntu1) ... 509s Generating locales (this might take a while)... 512s en_US.UTF-8... done 512s Generation complete. 512s Setting up apparmor (4.1.1-0ubuntu3) ... 512s Installing new version of config file /etc/apparmor.d/lsusb ... 512s Installing new version of config file /etc/apparmor.d/mbsync ... 513s Reloading AppArmor profiles 514s Setting up systemd-cryptsetup (257.7-1ubuntu1) ... 514s Setting up systemd-timesyncd (257.7-1ubuntu1) ... 515s systemd-time-wait-sync.service is a disabled or a static unit not running, not starting it. 515s Setting up udev (257.7-1ubuntu1) ... 516s Creating group 'clock' with GID 988. 516s Setting up libc-dev-bin (2.41-9ubuntu1) ... 516s Setting up systemd-resolved (257.7-1ubuntu1) ... 517s Setting up systemd-sysv (257.7-1ubuntu1) ... 517s Setting up libnss-systemd:arm64 (257.7-1ubuntu1) ... 517s Setting up libpam-systemd:arm64 (257.7-1ubuntu1) ... 517s Setting up libc6-dev:arm64 (2.41-9ubuntu1) ... 517s Processing triggers for libc-bin (2.41-9ubuntu1) ... 517s Processing triggers for man-db (2.13.1-1) ... 519s Processing triggers for dbus (1.16.2-2ubuntu1) ... 519s Processing triggers for shared-mime-info (2.4-5build2) ... 519s Processing triggers for procps (2:4.0.4-8ubuntu2) ... 519s Processing triggers for initramfs-tools (0.149ubuntu1) ... 519s update-initramfs: Generating /boot/initrd.img-6.15.0-3-generic 519s W: No lz4 in /usr/bin:/sbin:/bin, using gzip 521s W: Possible missing firmware /lib/firmware/apple/dfrmtfw-*.bin for built-in driver apple_z2 539s System running in EFI mode, skipping. 540s Reading package lists... 540s Building dependency tree... 540s Reading state information... 540s Solving dependencies... 541s 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. 541s autopkgtest [19:43:11]: rebooting testbed after setup commands that affected boot 574s Reading package lists... 574s Building dependency tree... 574s Reading state information... 574s Solving dependencies... 575s The following NEW packages will be installed: 575s libamd-comgr2 libamdhip64-5 libfmt10 libggml libggml-hip libgomp1 575s libhipblas0 libhsa-runtime64-1 libhsakmt1 libllvm17t64 librocblas0 575s librocsolver0 pkg-rocm-tools 575s 0 upgraded, 13 newly installed, 0 to remove and 0 not upgraded. 575s Need to get 849 MB of archives. 575s After this operation, 5118 MB of additional disk space will be used. 575s Get:1 http://ftpmaster.internal/ubuntu questing/universe arm64 libllvm17t64 arm64 1:17.0.6-22build1 [24.8 MB] 587s Get:2 http://ftpmaster.internal/ubuntu questing/universe arm64 libamd-comgr2 arm64 6.0+git20231212.4510c28+dfsg-3build3 [14.0 MB] 590s Get:3 http://ftpmaster.internal/ubuntu questing/universe arm64 libhsakmt1 arm64 5.7.0-1build1 [63.6 kB] 590s Get:4 http://ftpmaster.internal/ubuntu questing/universe arm64 libhsa-runtime64-1 arm64 5.7.1-2build1 [286 kB] 590s Get:5 http://ftpmaster.internal/ubuntu questing/universe arm64 libamdhip64-5 arm64 5.7.1-6 [9707 kB] 591s Get:6 http://ftpmaster.internal/ubuntu questing/universe arm64 libfmt10 arm64 10.1.1+ds1-4 [75.6 kB] 591s Get:7 http://ftpmaster.internal/ubuntu questing/main arm64 libgomp1 arm64 15.1.0-8ubuntu1 [147 kB] 591s Get:8 http://ftpmaster.internal/ubuntu questing/universe arm64 librocblas0 arm64 5.5.1+dfsg-7 [181 MB] 600s Get:9 http://ftpmaster.internal/ubuntu questing/universe arm64 librocsolver0 arm64 5.5.1-7 [479 MB] 629s Get:10 http://ftpmaster.internal/ubuntu questing/universe arm64 libhipblas0 arm64 5.5.1-4build1 [68.4 kB] 629s Get:11 http://ftpmaster.internal/ubuntu questing/universe arm64 libggml-hip arm64 0.0~git20250618.9e4bee1-1 [139 MB] 636s Get:12 http://ftpmaster.internal/ubuntu questing/universe arm64 libggml arm64 0.0~git20250618.9e4bee1-1 [282 kB] 636s Get:13 http://ftpmaster.internal/ubuntu questing/universe arm64 pkg-rocm-tools all 0.8.2 [7130 B] 637s Fetched 849 MB in 1min 2s (13.7 MB/s) 637s Selecting previously unselected package libllvm17t64:arm64. 637s (Reading database ... (Reading database ... 5% (Reading database ... 10% (Reading database ... 15% (Reading database ... 20% (Reading database ... 25% (Reading database ... 30% (Reading database ... 35% (Reading database ... 40% (Reading database ... 45% (Reading database ... 50% (Reading database ... 55% (Reading database ... 60% (Reading database ... 65% (Reading database ... 70% (Reading database ... 75% (Reading database ... 80% (Reading database ... 85% (Reading database ... 90% (Reading database ... 95% (Reading database ... 100% (Reading database ... 127320 files and directories currently installed.) 637s Preparing to unpack .../00-libllvm17t64_1%3a17.0.6-22build1_arm64.deb ... 637s Unpacking libllvm17t64:arm64 (1:17.0.6-22build1) ... 638s Selecting previously unselected package libamd-comgr2:arm64. 638s Preparing to unpack .../01-libamd-comgr2_6.0+git20231212.4510c28+dfsg-3build3_arm64.deb ... 638s Unpacking libamd-comgr2:arm64 (6.0+git20231212.4510c28+dfsg-3build3) ... 638s Selecting previously unselected package libhsakmt1:arm64. 638s Preparing to unpack .../02-libhsakmt1_5.7.0-1build1_arm64.deb ... 638s Unpacking libhsakmt1:arm64 (5.7.0-1build1) ... 638s Selecting previously unselected package libhsa-runtime64-1. 638s Preparing to unpack .../03-libhsa-runtime64-1_5.7.1-2build1_arm64.deb ... 638s Unpacking libhsa-runtime64-1 (5.7.1-2build1) ... 638s Selecting previously unselected package libamdhip64-5:arm64. 638s Preparing to unpack .../04-libamdhip64-5_5.7.1-6_arm64.deb ... 638s Unpacking libamdhip64-5:arm64 (5.7.1-6) ... 638s Selecting previously unselected package libfmt10:arm64. 638s Preparing to unpack .../05-libfmt10_10.1.1+ds1-4_arm64.deb ... 638s Unpacking libfmt10:arm64 (10.1.1+ds1-4) ... 638s Selecting previously unselected package libgomp1:arm64. 638s Preparing to unpack .../06-libgomp1_15.1.0-8ubuntu1_arm64.deb ... 638s Unpacking libgomp1:arm64 (15.1.0-8ubuntu1) ... 638s Selecting previously unselected package librocblas0. 638s Preparing to unpack .../07-librocblas0_5.5.1+dfsg-7_arm64.deb ... 638s Unpacking librocblas0 (5.5.1+dfsg-7) ... 647s Selecting previously unselected package librocsolver0. 647s Preparing to unpack .../08-librocsolver0_5.5.1-7_arm64.deb ... 647s Unpacking librocsolver0 (5.5.1-7) ... 653s Selecting previously unselected package libhipblas0. 654s Preparing to unpack .../09-libhipblas0_5.5.1-4build1_arm64.deb ... 654s Unpacking libhipblas0 (5.5.1-4build1) ... 654s Selecting previously unselected package libggml-hip:arm64. 654s Preparing to unpack .../10-libggml-hip_0.0~git20250618.9e4bee1-1_arm64.deb ... 654s Unpacking libggml-hip:arm64 (0.0~git20250618.9e4bee1-1) ... 658s Selecting previously unselected package libggml:arm64. 658s Preparing to unpack .../11-libggml_0.0~git20250618.9e4bee1-1_arm64.deb ... 658s Unpacking libggml:arm64 (0.0~git20250618.9e4bee1-1) ... 658s Selecting previously unselected package pkg-rocm-tools. 658s Preparing to unpack .../12-pkg-rocm-tools_0.8.2_all.deb ... 658s Unpacking pkg-rocm-tools (0.8.2) ... 658s Setting up libgomp1:arm64 (15.1.0-8ubuntu1) ... 658s Setting up libllvm17t64:arm64 (1:17.0.6-22build1) ... 658s Setting up libamd-comgr2:arm64 (6.0+git20231212.4510c28+dfsg-3build3) ... 658s Setting up pkg-rocm-tools (0.8.2) ... 658s Setting up libfmt10:arm64 (10.1.1+ds1-4) ... 658s Setting up libhsakmt1:arm64 (5.7.0-1build1) ... 658s Setting up libhsa-runtime64-1 (5.7.1-2build1) ... 658s Setting up libamdhip64-5:arm64 (5.7.1-6) ... 658s Setting up librocblas0 (5.5.1+dfsg-7) ... 658s Setting up librocsolver0 (5.5.1-7) ... 658s Setting up libhipblas0 (5.5.1-4build1) ... 658s Setting up libggml:arm64 (0.0~git20250618.9e4bee1-1) ... 658s Setting up libggml-hip:arm64 (0.0~git20250618.9e4bee1-1) ... 658s Processing triggers for libc-bin (2.41-9ubuntu1) ... 658s Processing triggers for man-db (2.13.1-1) ... 663s autopkgtest [19:45:13]: test libggml-hip: rocm-test-launcher /usr/libexec/ggml/runtests 663s autopkgtest [19:45:13]: test libggml-hip: [----------------------- 663s /dev/kfd not present, system either lacks AMD GPU or 'amdgpu' driver is not loaded. 663s Skipping tests. 664s autopkgtest [19:45:14]: test libggml-hip: -----------------------] 664s libggml-hip SKIP exit status 77 and marked as skippable 664s autopkgtest [19:45:14]: test libggml-hip: - - - - - - - - - - results - - - - - - - - - - 665s autopkgtest [19:45:15]: @@@@@@@@@@@@@@@@@@@@ summary 665s libggml-cpu PASS 665s libggml-cpu PASS 665s libggml-hip SKIP exit status 77 and marked as skippable