2026-09-07T13:52:31.139753Z  INFO retained operator-plan finalist finalist=0 values=4 operations=2 estimated_cycles=3506 estimated_exchange_cycles=2000 exchange_row_bytes=188 peak_standard_bytes=18108 peak_interleaved_bytes=33024 peak_total_bytes=49852 maximum_standard_allocation_bytes=16384 active_tile_counts={64} padding_group_counts={} conversion_sources=[Some(OperationId(0))]
2026-09-07T13:52:31.139852Z  INFO retained operator-plan finalist finalist=1 values=4 operations=2 estimated_cycles=3506 estimated_exchange_cycles=2000 exchange_row_bytes=188 peak_standard_bytes=16572 peak_interleaved_bytes=33280 peak_total_bytes=49852 maximum_standard_allocation_bytes=16384 active_tile_counts={64} padding_group_counts={} conversion_sources=[Some(OperationId(0))]
2026-09-07T13:52:31.139883Z  INFO retained operator-plan finalist finalist=2 values=4 operations=2 estimated_cycles=3626 estimated_exchange_cycles=2000 exchange_row_bytes=188 peak_standard_bytes=19644 peak_interleaved_bytes=33024 peak_total_bytes=50364 maximum_standard_allocation_bytes=16384 active_tile_counts={64} padding_group_counts={} conversion_sources=[Some(OperationId(0))]
2026-09-07T13:52:31.139893Z  INFO retained operator-plan finalist finalist=3 values=4 operations=2 estimated_cycles=3626 estimated_exchange_cycles=2000 exchange_row_bytes=188 peak_standard_bytes=18620 peak_interleaved_bytes=33792 peak_total_bytes=50364 maximum_standard_allocation_bytes=16384 active_tile_counts={64} padding_group_counts={} conversion_sources=[Some(OperationId(0))]
2026-09-07T13:52:31.141390Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="lower_mid"}: package build phase finished phase="lower_mid" elapsed_ms=16 success=true
2026-09-07T13:52:31.146605Z  INFO screened tile mappings by exchange resource load candidates=11 baseline_cycles=448 candidate_cycles=384 baseline_pressure=320 candidate_pressure=320
2026-09-07T13:52:31.146696Z  INFO modelled expanded operator plan finalist=0 modelled_cycles=3026 estimated_cycles=3282 estimated_exchange_cycles=1840
2026-09-07T13:52:31.146694Z  INFO screened tile mappings by exchange resource load candidates=8 baseline_cycles=320 candidate_cycles=320 baseline_pressure=320 candidate_pressure=320
2026-09-07T13:52:31.146762Z  INFO modelled expanded operator plan finalist=3 modelled_cycles=2954 estimated_cycles=3274 estimated_exchange_cycles=1840
2026-09-07T13:52:31.147517Z  INFO screened tile mappings by exchange resource load candidates=11 baseline_cycles=448 candidate_cycles=384 baseline_pressure=320 candidate_pressure=320
2026-09-07T13:52:31.147598Z  INFO modelled expanded operator plan finalist=1 modelled_cycles=3026 estimated_cycles=3282 estimated_exchange_cycles=1840
2026-09-07T13:52:31.148093Z  INFO screened tile mappings by exchange resource load candidates=8 baseline_cycles=320 candidate_cycles=320 baseline_pressure=320 candidate_pressure=320
2026-09-07T13:52:31.148161Z  INFO modelled expanded operator plan finalist=2 modelled_cycles=2954 estimated_cycles=3274 estimated_exchange_cycles=1840
2026-09-07T13:52:31.149305Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="select_finalist"}: scheduled operator-plan finalist finalist=2 mapped=false analytical_cycles=3274 analytical_exchange_cycles=1840 scheduled_exchange_cycles=1840 refined_cycles=3274
2026-09-07T13:52:31.150319Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="select_finalist"}: selected physically scheduled operator-plan finalist selected=2 mapped=false
2026-09-07T13:52:31.150336Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="select_finalist"}: package build phase finished phase="select_finalist" elapsed_ms=8 success=true
2026-09-07T13:52:31.150345Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}: lowered graph for package construction logical_shards=584 exchange_phases=2
2026-09-07T13:52:31.150610Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="compile_runtime"}: package build phase finished phase="compile_runtime" elapsed_ms=0 success=true
2026-09-07T13:52:31.150675Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="plan_kernels"}: package build phase finished phase="plan_kernels" elapsed_ms=0 success=true
2026-09-07T13:52:31.150914Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="compile_kernels"}: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/gemm_f16_amp.S name="gemm_f8_k64_c16_r16_r16" target=ipu21
2026-09-07T13:52:31.213462Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="compile_kernels"}: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/9f9f3e499e1fc9533f8318063c506198703f41566658b114071334a597a7abd3/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/9f9f3e499e1fc9533f8318063c506198703f41566658b114071334a597a7abd3/kernel.json
2026-09-07T13:52:31.213801Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="compile_kernels"}: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/cast_f8.cpp name="cast_f16_f8_codelet" target=ipu21
2026-09-07T13:52:31.552487Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="compile_kernels"}: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/24bbe6e37a5b0bfd3fa4963c541af221465c53d25913b718e9a484e949f26d33/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/24bbe6e37a5b0bfd3fa4963c541af221465c53d25913b718e9a484e949f26d33/kernel.json
2026-09-07T13:52:31.552890Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="compile_kernels"}: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/worker_call.S name="cast_f16_f8_wrapper" target=ipu21
2026-09-07T13:52:31.606889Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="compile_kernels"}: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/ac48960bfa12ac5985380ceee568fd51f345f63b4e1fed23f00949c7888b7117/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/ac48960bfa12ac5985380ceee568fd51f345f63b4e1fed23f00949c7888b7117/kernel.json
2026-09-07T13:52:31.607251Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="compile_kernels"}: package build phase finished phase="compile_kernels" elapsed_ms=456 success=true
2026-09-07T13:52:31.607539Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="link_runtime"}: package build phase finished phase="link_runtime" elapsed_ms=0 success=true
2026-09-07T13:52:31.634218Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="size_tile_code"}: package build phase finished phase="size_tile_code" elapsed_ms=1 success=true
2026-09-07T13:52:31.634259Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}: allocated package support memory linked_end=314164 profile_bytes=32 exchange_table_bytes=56 host_code_bytes=368 generated_code_bytes=372 code_address=314600 standard_ranges=[(409600, 524288)]
2026-09-07T13:52:31.635017Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="place_storage"}: package build phase finished phase="place_storage" elapsed_ms=0 success=true
2026-09-07T13:52:31.635481Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="lower_exchanges"}: reused exchange optimization after validating relocated rows phase=0
2026-09-07T13:52:31.635746Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="lower_exchanges"}: reused exchange optimization after validating relocated rows phase=1
2026-09-07T13:52:31.635974Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="lower_exchanges"}: package build phase finished phase="lower_exchanges" elapsed_ms=0 success=true
2026-09-07T13:52:31.636097Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: package build phase finished phase="optimize_exchange_placement" elapsed_ms=0 success=true
2026-09-07T13:52:31.636150Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}: costed final placed program final_cycles=3274 final_exchange=1840
2026-09-07T13:52:31.651037Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="prepare_tile_code"}: package build phase finished phase="prepare_tile_code" elapsed_ms=2 success=true
2026-09-07T13:52:31.654069Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="emit_tile_code"}: package build phase finished phase="emit_tile_code" elapsed_ms=3 success=true
2026-09-07T13:52:31.722011Z  INFO ipu_codegen.package.build{tile_count=64 operations=1}:ipu_codegen.package.phase{phase="build_tile_images"}: package build phase finished phase="build_tile_images" elapsed_ms=67 success=true
2026-09-07T13:52:31.727530Z  INFO writing application package tiles=1472
2026-09-07T13:52:31.734495Z  INFO application package written
2026-09-07T13:52:31.735019Z  INFO reading application package
2026-09-07T13:52:31.748616Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-07T13:52:31.748922Z  INFO waiting for hardware ownership path=artifacts/layout-sweep/device.lock
2026-09-07T13:52:31.748935Z  INFO acquired hardware ownership path=artifacts/layout-sweep/device.lock
2026-09-07T13:52:31.748949Z  INFO opening IPU device path="/dev/ipu0"
2026-09-07T13:52:31.748996Z  INFO IPU device attached path=/dev/ipu0
2026-09-07T13:52:31.749003Z  INFO initializing IPU device
2026-09-07T13:52:31.757533Z  INFO IPU device initialized
2026-09-07T13:52:31.757579Z  INFO replaying device configuration register_writes=2003
2026-09-07T13:52:31.761731Z  INFO loading application tile_count=1472 final_mark=23
2026-09-07T13:52:31.971719Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-07T13:52:31.989692Z  INFO starting host exchange session startup_mark=23
2026-09-07T13:52:31.989736Z  INFO attaching host exchange pages pages=2
2026-09-07T13:52:31.990544Z  INFO host exchange session started
2026-09-07T13:52:31.990591Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=2
2026-09-07T13:52:31.990819Z  INFO invoking streaming host exchange call call="run" command=0 phases=9
Error: numerical comparison failed for 7565/8192 logical shard values (maximum absolute error 1.4360352): [(0, -0.056854248, -0.0032958984, 0.05355835), (1, 0.74902344, 0.3996582, 0.34936523), (2, 0.030899048, -0.31469727, 0.3455963), (3, 0.096191406, 0.18823242, 0.092041016), (4, -0.07989502, 0.03488159, 0.11477661), (5, -0.49267578, -0.13781738, 0.3548584), (6, 0.22216797, 0.068359375, 0.1538086), (7, -0.15075684, -0.42871094, 0.2779541), (9, -0.22265625, -0.51416016, 0.2915039), (10, -0.3059082, -0.2265625, 0.0793457), (11, 0.037506104, 0.24768066, 0.21017456), (12, -0.30639648, -0.124572754, 0.18182373), (14, 0.29174805, 0.49316406, 0.20141602), (15, -0.3828125, -0.05908203, 0.32373047), (64, 0.014060974, -0.20581055, 0.21987152), (65, 0.35009766, 0.11621094, 0.23388672)]