Compiling ipu-tests v0.1.0 (/srv/home/gc-sdk/ipu-stack/crates/ipu-tests) Finished `release` profile [optimized] target(s) in 1.10s Running `target/release/cast_check --sdk /srv/home/gc-sdk/poplar_sdk-ubuntu_20_04-3.4.0+1507-69d9d03fd8/poplar-ubuntu_20_04-3.4.0+73-aa67dd6164 --kernel /tmp/cast-before.cpp --existing-layouts-only --output artifacts/fp8-cast/before` 2026-09-08T14:48:48.781508Z INFO opening IPU device path="/dev/ipu0" 2026-09-08T14:48:48.781552Z INFO IPU device attached path=/dev/ipu0 2026-09-08T14:48:48.781556Z INFO initializing IPU device 2026-09-08T14:48:48.789622Z INFO IPU device initialized 2026-09-08T14:48:48.789636Z INFO replaying device configuration register_writes=2003 2026-09-08T14:48:48.793047Z INFO loading application tile_count=1472 final_mark=23 2026-09-08T14:48:48.980171Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-08T14:48:48.983654Z INFO starting host exchange session startup_mark=23 2026-09-08T14:48:48.983661Z INFO attaching host exchange pages pages=2 2026-09-08T14:48:48.984375Z INFO host exchange session started 2026-09-08T14:48:48.984400Z INFO invoking streaming host exchange call call="run" command=0 phases=9 case=(1, 1, 0, "separate-standard", -4) cycles=750 case=(1, 1, 0, "separate-standard", 0) cycles=738 case=(1, 1, 0, "separate-standard", 3) cycles=738 case=(1, 1, 0, "shared-standard", -4) cycles=750 case=(1, 1, 0, "shared-standard", 0) cycles=738 case=(1, 1, 0, "shared-standard", 3) cycles=738 case=(1, 1, 0, "interleaved", -4) cycles=750 case=(1, 1, 0, "interleaved", 0) cycles=738 case=(1, 1, 0, "interleaved", 3) cycles=738 case=(1, 7, 0, "separate-standard", -4) cycles=882 case=(1, 7, 0, "separate-standard", 0) cycles=870 case=(1, 7, 0, "separate-standard", 3) cycles=870 case=(1, 7, 0, "shared-standard", -4) cycles=882 case=(1, 7, 0, "shared-standard", 0) cycles=870 case=(1, 7, 0, "shared-standard", 3) cycles=870 case=(1, 7, 0, "interleaved", -4) cycles=882 case=(1, 7, 0, "interleaved", 0) cycles=870 case=(1, 7, 0, "interleaved", 3) cycles=870 case=(1, 8, 0, "separate-standard", -4) cycles=486 case=(1, 8, 0, "separate-standard", 0) cycles=474 case=(1, 8, 0, "separate-standard", 3) cycles=474 case=(1, 8, 0, "shared-standard", -4) cycles=486 case=(1, 8, 0, "shared-standard", 0) cycles=474 case=(1, 8, 0, "shared-standard", 3) cycles=474 case=(1, 8, 0, "interleaved", -4) cycles=486 case=(1, 8, 0, "interleaved", 0) cycles=474 case=(1, 8, 0, "interleaved", 3) cycles=474 case=(1, 9, 0, "separate-standard", -4) cycles=750 case=(1, 9, 0, "separate-standard", 0) cycles=738 case=(1, 9, 0, "separate-standard", 3) cycles=738 case=(1, 9, 0, "shared-standard", -4) cycles=750 case=(1, 9, 0, "shared-standard", 0) cycles=738 case=(1, 9, 0, "shared-standard", 3) cycles=738 case=(1, 9, 0, "interleaved", -4) cycles=750 case=(1, 9, 0, "interleaved", 0) cycles=738 case=(1, 9, 0, "interleaved", 3) cycles=738 case=(1, 16, 0, "separate-standard", -4) cycles=486 case=(1, 16, 0, "separate-standard", 0) cycles=474 case=(1, 16, 0, "separate-standard", 3) cycles=474 case=(1, 16, 0, "shared-standard", -4) cycles=486 case=(1, 16, 0, "shared-standard", 0) cycles=474 case=(1, 16, 0, "shared-standard", 3) cycles=474 case=(1, 16, 0, "interleaved", -4) cycles=486 case=(1, 16, 0, "interleaved", 0) cycles=474 case=(1, 16, 0, "interleaved", 3) cycles=474 case=(1, 16, 1, "separate-standard", -4) cycles=774 case=(1, 16, 1, "separate-standard", 0) cycles=762 case=(1, 16, 1, "separate-standard", 3) cycles=762 case=(1, 16, 1, "shared-standard", -4) cycles=774 case=(1, 16, 1, "shared-standard", 0) cycles=762 case=(1, 16, 1, "shared-standard", 3) cycles=762 case=(1, 16, 1, "interleaved", -4) cycles=774 case=(1, 16, 1, "interleaved", 0) cycles=762 case=(1, 16, 1, "interleaved", 3) cycles=762 case=(5, 48, 0, "separate-standard", -4) cycles=582 case=(5, 48, 0, "separate-standard", 0) cycles=570 case=(5, 48, 0, "separate-standard", 3) cycles=570 case=(5, 48, 0, "shared-standard", -4) cycles=582 case=(5, 48, 0, "shared-standard", 0) cycles=570 case=(5, 48, 0, "shared-standard", 3) cycles=570 case=(5, 48, 0, "interleaved", -4) cycles=582 case=(5, 48, 0, "interleaved", 0) cycles=570 case=(5, 48, 0, "interleaved", 3) cycles=570 case=(5, 48, 1, "separate-standard", -4) cycles=1062 case=(5, 48, 1, "separate-standard", 0) cycles=1050 case=(5, 48, 1, "separate-standard", 3) cycles=1050 case=(5, 48, 1, "shared-standard", -4) cycles=1062 case=(5, 48, 1, "shared-standard", 0) cycles=1050 case=(5, 48, 1, "shared-standard", 3) cycles=1050 case=(5, 48, 1, "interleaved", -4) cycles=1062 case=(5, 48, 1, "interleaved", 0) cycles=1050 case=(5, 48, 1, "interleaved", 3) cycles=1050 case=(33, 48, 0, "separate-standard", -4) cycles=1254 case=(33, 48, 0, "separate-standard", 0) cycles=1242 case=(33, 48, 0, "separate-standard", 3) cycles=1242 case=(33, 48, 0, "shared-standard", -4) cycles=1254 case=(33, 48, 0, "shared-standard", 0) cycles=1242 case=(33, 48, 0, "shared-standard", 3) cycles=1242 case=(33, 48, 0, "interleaved", -4) cycles=1254 case=(33, 48, 0, "interleaved", 0) cycles=1242 case=(33, 48, 0, "interleaved", 3) cycles=1242 case=(33, 48, 1, "separate-standard", -4) cycles=1782 case=(33, 48, 1, "separate-standard", 0) cycles=1770 case=(33, 48, 1, "separate-standard", 3) cycles=1770 case=(33, 48, 1, "shared-standard", -4) cycles=1782 case=(33, 48, 1, "shared-standard", 0) cycles=1770 case=(33, 48, 1, "shared-standard", 3) cycles=1770 case=(33, 48, 1, "interleaved", -4) cycles=1782 case=(33, 48, 1, "interleaved", 0) cycles=1770 case=(33, 48, 1, "interleaved", 3) cycles=1770 case=(1, 47, 0, "separate-standard", -4) cycles=876 case=(1, 47, 0, "separate-standard", 0) cycles=864 case=(1, 47, 0, "separate-standard", 3) cycles=864 case=(1, 47, 0, "shared-standard", -4) cycles=876 case=(1, 47, 0, "shared-standard", 0) cycles=864 case=(1, 47, 0, "shared-standard", 3) cycles=864 case=(1, 47, 0, "interleaved", -4) cycles=876 case=(1, 47, 0, "interleaved", 0) cycles=864 case=(1, 47, 0, "interleaved", 3) cycles=864 case=(1, 48, 0, "separate-standard", -4) cycles=486 case=(1, 48, 0, "separate-standard", 0) cycles=474 case=(1, 48, 0, "separate-standard", 3) cycles=474 case=(1, 48, 0, "shared-standard", -4) cycles=486 case=(1, 48, 0, "shared-standard", 0) cycles=474 case=(1, 48, 0, "shared-standard", 3) cycles=474 case=(1, 48, 0, "interleaved", -4) cycles=486 case=(1, 48, 0, "interleaved", 0) cycles=474 case=(1, 48, 0, "interleaved", 3) cycles=474 case=(1, 48, 1, "separate-standard", -4) cycles=1062 case=(1, 48, 1, "separate-standard", 0) cycles=1050 case=(1, 48, 1, "separate-standard", 3) cycles=1050 case=(1, 48, 1, "shared-standard", -4) cycles=1062 case=(1, 48, 1, "shared-standard", 0) cycles=1050 case=(1, 48, 1, "shared-standard", 3) cycles=1050 case=(1, 48, 1, "interleaved", -4) cycles=1062 case=(1, 48, 1, "interleaved", 0) cycles=1050 case=(1, 48, 1, "interleaved", 3) cycles=1050 case=(1, 49, 0, "separate-standard", -4) cycles=774 case=(1, 49, 0, "separate-standard", 0) cycles=762 case=(1, 49, 0, "separate-standard", 3) cycles=762 case=(1, 49, 0, "shared-standard", -4) cycles=774 case=(1, 49, 0, "shared-standard", 0) cycles=762 case=(1, 49, 0, "shared-standard", 3) cycles=762 case=(1, 49, 0, "interleaved", -4) cycles=774 case=(1, 49, 0, "interleaved", 0) cycles=762 case=(1, 49, 0, "interleaved", 3) cycles=762 case=(1, 128, 0, "separate-standard", -4) cycles=534 case=(1, 128, 0, "separate-standard", 0) cycles=522 case=(1, 128, 0, "separate-standard", 3) cycles=522 case=(1, 128, 0, "shared-standard", -4) cycles=534 case=(1, 128, 0, "shared-standard", 0) cycles=522 case=(1, 128, 0, "shared-standard", 3) cycles=522 case=(1, 128, 0, "interleaved", -4) cycles=534 case=(1, 128, 0, "interleaved", 0) cycles=522 case=(1, 128, 0, "interleaved", 3) cycles=522 case=(1, 128, 1, "separate-standard", -4) cycles=534 case=(1, 128, 1, "separate-standard", 0) cycles=522 case=(1, 128, 1, "separate-standard", 3) cycles=522 case=(1, 128, 1, "shared-standard", -4) cycles=534 case=(1, 128, 1, "shared-standard", 0) cycles=522 case=(1, 128, 1, "shared-standard", 3) cycles=522 case=(1, 128, 1, "interleaved", -4) cycles=534 case=(1, 128, 1, "interleaved", 0) cycles=522 case=(1, 128, 1, "interleaved", 3) cycles=522 case=(2, 128, 0, "separate-standard", -4) cycles=606 case=(2, 128, 0, "separate-standard", 0) cycles=594 case=(2, 128, 0, "separate-standard", 3) cycles=594 case=(2, 128, 0, "shared-standard", -4) cycles=606 case=(2, 128, 0, "shared-standard", 0) cycles=594 case=(2, 128, 0, "shared-standard", 3) cycles=594 case=(2, 128, 0, "interleaved", -4) cycles=606 case=(2, 128, 0, "interleaved", 0) cycles=594 case=(2, 128, 0, "interleaved", 3) cycles=594 case=(2, 128, 1, "separate-standard", -4) cycles=1704 case=(2, 128, 1, "separate-standard", 0) cycles=1692 case=(2, 128, 1, "separate-standard", 3) cycles=1692 case=(2, 128, 1, "shared-standard", -4) cycles=1704 case=(2, 128, 1, "shared-standard", 0) cycles=1692 case=(2, 128, 1, "shared-standard", 3) cycles=1692 case=(2, 128, 1, "interleaved", -4) cycles=1704 case=(2, 128, 1, "interleaved", 0) cycles=1692 case=(2, 128, 1, "interleaved", 3) cycles=1692 case=(5, 64, 0, "separate-standard", -4) cycles=630 case=(5, 64, 0, "separate-standard", 0) cycles=618 case=(5, 64, 0, "separate-standard", 3) cycles=618 case=(5, 64, 0, "shared-standard", -4) cycles=630 case=(5, 64, 0, "shared-standard", 0) cycles=618 case=(5, 64, 0, "shared-standard", 3) cycles=618 case=(5, 64, 0, "interleaved", -4) cycles=630 case=(5, 64, 0, "interleaved", 0) cycles=618 case=(5, 64, 0, "interleaved", 3) cycles=618 case=(5, 64, 1, "separate-standard", -4) cycles=1128 case=(5, 64, 1, "separate-standard", 0) cycles=1116 case=(5, 64, 1, "separate-standard", 3) cycles=1116 case=(5, 64, 1, "shared-standard", -4) cycles=1128 case=(5, 64, 1, "shared-standard", 0) cycles=1116 case=(5, 64, 1, "shared-standard", 3) cycles=1116 case=(5, 64, 1, "interleaved", -4) cycles=1128 case=(5, 64, 1, "interleaved", 0) cycles=1116 case=(5, 64, 1, "interleaved", 3) cycles=1116 case=(6, 64, 0, "separate-standard", -4) cycles=654 case=(6, 64, 0, "separate-standard", 0) cycles=642 case=(6, 64, 0, "separate-standard", 3) cycles=642 case=(6, 64, 0, "shared-standard", -4) cycles=654 case=(6, 64, 0, "shared-standard", 0) cycles=642 case=(6, 64, 0, "shared-standard", 3) cycles=642 case=(6, 64, 0, "interleaved", -4) cycles=654 case=(6, 64, 0, "interleaved", 0) cycles=642 case=(6, 64, 0, "interleaved", 3) cycles=642 case=(6, 64, 1, "separate-standard", -4) cycles=1128 case=(6, 64, 1, "separate-standard", 0) cycles=1116 case=(6, 64, 1, "separate-standard", 3) cycles=1116 case=(6, 64, 1, "shared-standard", -4) cycles=1128 case=(6, 64, 1, "shared-standard", 0) cycles=1116 case=(6, 64, 1, "shared-standard", 3) cycles=1116 case=(6, 64, 1, "interleaved", -4) cycles=1128 case=(6, 64, 1, "interleaved", 0) cycles=1116 case=(6, 64, 1, "interleaved", 3) cycles=1116 case=(7, 64, 0, "separate-standard", -4) cycles=702 case=(7, 64, 0, "separate-standard", 0) cycles=690 case=(7, 64, 0, "separate-standard", 3) cycles=690 case=(7, 64, 0, "shared-standard", -4) cycles=702 case=(7, 64, 0, "shared-standard", 0) cycles=690 case=(7, 64, 0, "shared-standard", 3) cycles=690 case=(7, 64, 0, "interleaved", -4) cycles=702 case=(7, 64, 0, "interleaved", 0) cycles=690 case=(7, 64, 0, "interleaved", 3) cycles=690 case=(7, 64, 1, "separate-standard", -4) cycles=1296 case=(7, 64, 1, "separate-standard", 0) cycles=1284 case=(7, 64, 1, "separate-standard", 3) cycles=1284 case=(7, 64, 1, "shared-standard", -4) cycles=1296 case=(7, 64, 1, "shared-standard", 0) cycles=1284 case=(7, 64, 1, "shared-standard", 3) cycles=1284 case=(7, 64, 1, "interleaved", -4) cycles=1296 case=(7, 64, 1, "interleaved", 0) cycles=1284 case=(7, 64, 1, "interleaved", 3) cycles=1284 case=(32, 192, 0, "separate-standard", -4) cycles=3534 case=(32, 192, 0, "separate-standard", 0) cycles=3522 case=(32, 192, 0, "separate-standard", 3) cycles=3522 case=(32, 192, 0, "shared-standard", -4) cycles=3534 case=(32, 192, 0, "shared-standard", 0) cycles=3522 case=(32, 192, 0, "shared-standard", 3) cycles=3522 case=(32, 192, 0, "interleaved", -4) cycles=3534 case=(32, 192, 0, "interleaved", 0) cycles=3522 case=(32, 192, 0, "interleaved", 3) cycles=3522 case=(32, 192, 1, "separate-standard", -4) cycles=3336 case=(32, 192, 1, "separate-standard", 0) cycles=3324 case=(32, 192, 1, "separate-standard", 3) cycles=3324 case=(32, 192, 1, "shared-standard", -4) cycles=3336 case=(32, 192, 1, "shared-standard", 0) cycles=3324 case=(32, 192, 1, "shared-standard", 3) cycles=3324 case=(32, 192, 1, "interleaved", -4) cycles=3336 case=(32, 192, 1, "interleaved", 0) cycles=3324 case=(32, 192, 1, "interleaved", 3) cycles=3324 case=(65, 128, 0, "separate-standard", -4) cycles=4638 case=(65, 128, 0, "separate-standard", 0) cycles=4626 case=(65, 128, 0, "separate-standard", 3) cycles=4626 case=(65, 128, 0, "shared-standard", -4) cycles=4638 case=(65, 128, 0, "shared-standard", 0) cycles=4626 case=(65, 128, 0, "shared-standard", 3) cycles=4626 case=(65, 128, 0, "interleaved", -4) cycles=4638 case=(65, 128, 0, "interleaved", 0) cycles=4626 case=(65, 128, 0, "interleaved", 3) cycles=4626 case=(65, 128, 1, "separate-standard", -4) cycles=5064 case=(65, 128, 1, "separate-standard", 0) cycles=5052 case=(65, 128, 1, "separate-standard", 3) cycles=5052 case=(65, 128, 1, "shared-standard", -4) cycles=5064 case=(65, 128, 1, "shared-standard", 0) cycles=5052 case=(65, 128, 1, "shared-standard", 3) cycles=5052 case=(65, 128, 1, "interleaved", -4) cycles=5064 case=(65, 128, 1, "interleaved", 0) cycles=5052 case=(65, 128, 1, "interleaved", 3) cycles=5052 cases=252 checkedBytes=333072 bitwiseTest=PASS