2026-09-06T11:40:06.735743Z INFO applied GEMM plan constraint source_operation=0 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 0, orientation: Normal, row_partitions: 4, column_partitions: 92, inner_partitions: 4, result_row_partitions: 4, result_column_partitions: 1, output_column_block: 48, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:40:07.174989Z INFO applied GEMM plan constraint source_operation=2 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 13, column_partitions: 36, inner_partitions: 3, result_row_partitions: 3, result_column_partitions: 1, output_column_block: 32, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:40:07.376374Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=409699 estimated_exchange_cycles=192280 exchange_row_bytes=42608 peak_standard_bytes=222832 peak_interleaved_bytes=184320 peak_total_bytes=407152 maximum_standard_allocation_bytes=180224 active_tile_counts={1404, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:40:07.378522Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_mid"}: package build phase finished phase="lower_mid" elapsed_ms=931 success=true 2026-09-06T11:40:15.164571Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=284068 estimated_exchange_cycles=89472 2026-09-06T11:40:15.219110Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: eliminated finite GEMM padding clears using load-time SRAM initialization removed=432 2026-09-06T11:40:15.221054Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: package build phase finished phase="select_finalist" elapsed_ms=7842 success=true 2026-09-06T11:40:15.221071Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26804 exchange_phases=4 2026-09-06T11:40:15.221273Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="compile_runtime"}: package build phase finished phase="compile_runtime" elapsed_ms=0 success=true 2026-09-06T11:40:15.221848Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="plan_kernels"}: package build phase finished phase="plan_kernels" elapsed_ms=0 success=true 2026-09-06T11:40:15.222126Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: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_f16_interleaved_k1440_c32_r56_r58" target=ipu21 2026-09-06T11:40:15.277721Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="compile_kernels"}: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/68a7e89846fdb25e44b6f6b0ad41154ec2b5c37d8563bb27c9d696cc29eb74ef/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/68a7e89846fdb25e44b6f6b0ad41154ec2b5c37d8563bb27c9d696cc29eb74ef/kernel.json 2026-09-06T11:40:15.278047Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="compile_kernels"}: package build phase finished phase="compile_kernels" elapsed_ms=56 success=true 2026-09-06T11:40:15.281981Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="link_runtime"}: package build phase finished phase="link_runtime" elapsed_ms=0 success=true 2026-09-06T11:40:15.442383Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="plan_exchange_storage"}: package build phase finished phase="plan_exchange_storage" elapsed_ms=160 success=true 2026-09-06T11:40:15.699178Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: prepared large physical exchange phase phase=1 transfers=15064 maximum_identical_destinations=1 2026-09-06T11:40:15.699216Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: optimized physical exchange schedule phase=1 initial_horizon=4180 selected_horizon=3779 endpoint_lower_bound=3408 lower_bound_gap=371 selected_kind="matching-waves" neighborhood_improvements=0 2026-09-06T11:40:15.705938Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: finished large physical exchange phase phase=1 tile=690 row_words=75 horizon=3779 2026-09-06T11:40:47.540730Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: compared ordinary and paired exchange schedules phase=2 candidates=4038 ordinary_horizon=64925 paired_horizon=70240 use_paired=false 2026-09-06T11:40:47.708994Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: prepared large physical exchange phase phase=2 transfers=7911 maximum_identical_destinations=1 2026-09-06T11:40:47.709024Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: optimized physical exchange schedule phase=2 initial_horizon=64925 selected_horizon=64925 endpoint_lower_bound=64800 lower_bound_gap=125 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:40:47.794751Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: finished large physical exchange phase phase=2 tile=891 row_words=604 horizon=64925 2026-09-06T11:40:48.179783Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: prepared large physical exchange phase phase=3 transfers=8208 maximum_identical_destinations=1 2026-09-06T11:40:48.179814Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: optimized physical exchange schedule phase=3 initial_horizon=1055 selected_horizon=1044 endpoint_lower_bound=864 lower_bound_gap=180 selected_kind="critical-neighborhood" neighborhood_improvements=1 2026-09-06T11:40:48.186023Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: finished large physical exchange phase phase=3 tile=1278 row_words=38 horizon=1044 2026-09-06T11:40:48.188565Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: package build phase finished phase="lower_exchanges_provisional" elapsed_ms=32746 success=true 2026-09-06T11:40:48.793524Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="size_tile_code"}: package build phase finished phase="size_tile_code" elapsed_ms=7 success=true 2026-09-06T11:40:48.793559Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319176 profile_bytes=420 exchange_table_bytes=2728 host_code_bytes=3412 generated_code_bytes=3220 code_address=322592 standard_ranges=[(409600, 524288)] 2026-09-06T11:40:48.899643Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="place_storage"}: package build phase finished phase="place_storage" elapsed_ms=106 success=true 2026-09-06T11:40:48.907620Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: reused exchange optimization after validating relocated rows phase=0 2026-09-06T11:40:48.995967Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: reused exchange optimization after validating relocated rows phase=1 2026-09-06T11:40:49.001879Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: prepared large physical exchange phase phase=1 transfers=15064 maximum_identical_destinations=1 2026-09-06T11:40:49.001891Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: optimized physical exchange schedule phase=1 initial_horizon=3779 selected_horizon=3779 endpoint_lower_bound=3408 lower_bound_gap=371 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:40:49.011932Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: finished large physical exchange phase phase=1 tile=690 row_words=75 horizon=3779 2026-09-06T11:41:13.122542Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: compared ordinary and paired exchange schedules phase=2 candidates=4038 ordinary_horizon=64925 paired_horizon=70240 use_paired=false 2026-09-06T11:41:13.229525Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: prepared large physical exchange phase phase=2 transfers=7911 maximum_identical_destinations=1 2026-09-06T11:41:13.229540Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: optimized physical exchange schedule phase=2 initial_horizon=64925 selected_horizon=64925 endpoint_lower_bound=64800 lower_bound_gap=125 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:41:13.282406Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: finished large physical exchange phase phase=2 tile=891 row_words=604 horizon=64925 2026-09-06T11:41:13.335388Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: reused exchange optimization after validating relocated rows phase=3 2026-09-06T11:41:13.337623Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: prepared large physical exchange phase phase=3 transfers=8208 maximum_identical_destinations=1 2026-09-06T11:41:13.337631Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: optimized physical exchange schedule phase=3 initial_horizon=1044 selected_horizon=1044 endpoint_lower_bound=864 lower_bound_gap=180 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:41:13.341607Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: finished large physical exchange phase phase=3 tile=1278 row_words=38 horizon=1044 2026-09-06T11:41:13.343037Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: package build phase finished phase="lower_exchanges" elapsed_ms=24443 success=true 2026-09-06T11:41:14.492364Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: reused exchange optimization after validating relocated rows phase=0 2026-09-06T11:41:14.587562Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: reused exchange optimization after validating relocated rows phase=1 2026-09-06T11:41:14.593677Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: prepared large physical exchange phase phase=1 transfers=15064 maximum_identical_destinations=1 2026-09-06T11:41:14.593689Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: optimized physical exchange schedule phase=1 initial_horizon=3779 selected_horizon=3779 endpoint_lower_bound=3408 lower_bound_gap=371 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:41:14.604170Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: finished large physical exchange phase phase=1 tile=690 row_words=75 horizon=3779 2026-09-06T11:41:41.502300Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: compared ordinary and paired exchange schedules phase=2 candidates=4038 ordinary_horizon=64925 paired_horizon=70240 use_paired=false 2026-09-06T11:41:41.677532Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: prepared large physical exchange phase phase=2 transfers=7911 maximum_identical_destinations=1 2026-09-06T11:41:41.677554Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: optimized physical exchange schedule phase=2 initial_horizon=64925 selected_horizon=64925 endpoint_lower_bound=64800 lower_bound_gap=125 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:41:41.764941Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: finished large physical exchange phase phase=2 tile=891 row_words=604 horizon=64925 2026-09-06T11:41:41.847721Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: reused exchange optimization after validating relocated rows phase=3 2026-09-06T11:41:41.851310Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: prepared large physical exchange phase phase=3 transfers=8208 maximum_identical_destinations=1 2026-09-06T11:41:41.851320Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: optimized physical exchange schedule phase=3 initial_horizon=1044 selected_horizon=1044 endpoint_lower_bound=864 lower_bound_gap=180 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:41:41.857694Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: finished large physical exchange phase phase=3 tile=1278 row_words=38 horizon=1044 2026-09-06T11:41:41.867540Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: evaluated exchange placement finalist offset=4096 baseline_score=1470637670400 score=540862709760 baseline_cycles=76796 candidate_cycles=76796 row_bytes=2728 row_capacity=16320 accepted=false 2026-09-06T11:41:41.876103Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: package build phase finished phase="optimize_exchange_placement" elapsed_ms=28533 success=true 2026-09-06T11:41:42.188125Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="prepare_tile_code"}: package build phase finished phase="prepare_tile_code" elapsed_ms=11 success=true 2026-09-06T11:41:42.199902Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="emit_tile_code"}: package build phase finished phase="emit_tile_code" elapsed_ms=11 success=true 2026-09-06T11:41:42.306318Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="build_tile_images"}: package build phase finished phase="build_tile_images" elapsed_ms=101 success=true 2026-09-06T11:41:42.374730Z INFO writing application package tiles=1472 2026-09-06T11:41:42.477420Z INFO application package written 2026-09-06T11:41:42.478920Z INFO reading application package 2026-09-06T11:41:42.586009Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:41:42.588480Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:41:42.588499Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:41:42.588512Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:41:42.588540Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:41:42.588543Z INFO initializing IPU device 2026-09-06T11:41:42.596480Z INFO IPU device initialized 2026-09-06T11:41:42.596485Z INFO replaying device configuration register_writes=2003 2026-09-06T11:41:42.605581Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:41:42.955017Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:41:45.858770Z INFO starting host exchange session startup_mark=23 2026-09-06T11:41:45.858804Z INFO attaching host exchange pages pages=2 2026-09-06T11:41:45.859381Z INFO host exchange session started 2026-09-06T11:41:45.859474Z INFO invoking streaming host exchange call call="initialize" command=0 phases=46 2026-09-06T11:41:45.870620Z INFO invoking streaming host exchange call call="run" command=0 phases=59 profile=artifacts/layout-sweep/down-01/execution.ipuprofile tiles=1472 samples=29636 workload=siglip-mlp-f16-b1-t729-d1152-h4304-n1 benchmark=siglip-mlp-f16 batch=1 tokens=729 rows=729 dimension=1152 hiddenDimension=4304 blocks=1 biases=false activeTiles=1404 executionTiles=1472 inputBytes=154736640 weightBytes=19869696 cycles=255402 minimumTileCycles=246222 deviceMicroseconds=170.268 effectiveGemmTflops=84.914 peakTflops=282.624 efficiencyPercent=30.04 maximumAbsoluteError=0.017578 package=artifacts/layout-sweep/down-01/model.ipuexe tiles=1472 hardwareTest=PASS