2026-09-06T11:43:59.805624Z 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 } 2026-09-06T11:44:00.227127Z INFO applied GEMM plan constraint source_operation=2 matching_plans=2 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 16, column_partitions: 9, inner_partitions: 10, result_row_partitions: 10, result_column_partitions: 1, output_column_block: 128, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:44:00.539529Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=353095 estimated_exchange_cycles=131320 exchange_row_bytes=28892 peak_standard_bytes=143580 peak_interleaved_bytes=221184 peak_total_bytes=314588 maximum_standard_allocation_bytes=114688 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:44:00.543374Z 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=910 success=true 2026-09-06T11:44:06.278362Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=325505 estimated_exchange_cycles=110568 2026-09-06T11:44:06.291671Z 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=135 2026-09-06T11:44:06.292491Z 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=5749 success=true 2026-09-06T11:44:06.292502Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=25994 exchange_phases=4 2026-09-06T11:44:06.292630Z 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 2026-09-06T11:44:06.292943Z 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 2026-09-06T11:44:06.293116Z 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_k432_c128_r44_r46" target=ipu21 2026-09-06T11:44:06.345552Z 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/a79cbd4262fdbe84c18a2b4cef130d05313f5a44fa85b369121b2f020b99931f/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/a79cbd4262fdbe84c18a2b4cef130d05313f5a44fa85b369121b2f020b99931f/kernel.json 2026-09-06T11:44:06.346118Z 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=53 success=true 2026-09-06T11:44:06.351023Z 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 2026-09-06T11:44:06.463158Z 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=112 success=true 2026-09-06T11:44:06.798177Z 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 2026-09-06T11:44:06.798222Z 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 2026-09-06T11:44:06.808708Z 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 2026-09-06T11:44:07.110362Z 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=30 ordinary_horizon=37815 paired_horizon=37815 use_paired=false 2026-09-06T11:44:07.123105Z 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=2279 maximum_identical_destinations=1 2026-09-06T11:44:07.123122Z 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=37815 selected_horizon=37815 endpoint_lower_bound=37584 lower_bound_gap=231 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:44:07.129563Z 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=1402 row_words=87 horizon=37815 2026-09-06T11:44:22.217539Z 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=116208 maximum_identical_destinations=1 2026-09-06T11:44:22.217581Z 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=8233 selected_horizon=8233 endpoint_lower_bound=3840 lower_bound_gap=4393 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:44:22.333569Z 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=1041 row_words=480 horizon=8233 2026-09-06T11:44:22.370522Z 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=15907 success=true 2026-09-06T11:44:23.037883Z 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=5 success=true 2026-09-06T11:44:23.037921Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319176 profile_bytes=164 exchange_table_bytes=2344 host_code_bytes=3684 generated_code_bytes=1760 code_address=322864 standard_ranges=[(409600, 524288)] 2026-09-06T11:44:23.141287Z 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=103 success=true 2026-09-06T11:44:23.149302Z 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 2026-09-06T11:44:23.237959Z 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 2026-09-06T11:44:23.243870Z 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 2026-09-06T11:44:23.243882Z 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 2026-09-06T11:44:23.253926Z 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 2026-09-06T11:44:23.329865Z 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=2 2026-09-06T11:44:23.334832Z 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=2279 maximum_identical_destinations=1 2026-09-06T11:44:23.334843Z 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=37815 selected_horizon=37815 endpoint_lower_bound=37584 lower_bound_gap=231 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:44:23.340795Z 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=1402 row_words=87 horizon=37815 2026-09-06T11:44:43.870845Z 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=116208 maximum_identical_destinations=1 2026-09-06T11:44:43.870880Z 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=8233 selected_horizon=8233 endpoint_lower_bound=3840 lower_bound_gap=4393 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:44:43.990370Z 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=1041 row_words=480 horizon=8233 2026-09-06T11:44:44.025490Z 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=20884 success=true 2026-09-06T11:44:45.029961Z 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 2026-09-06T11:44:45.121001Z 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 2026-09-06T11:44:45.127336Z 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 2026-09-06T11:44:45.127347Z 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 2026-09-06T11:44:45.137592Z 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 2026-09-06T11:44:45.494277Z 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=30 ordinary_horizon=37815 paired_horizon=37815 use_paired=false 2026-09-06T11:44:45.508096Z 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=2279 maximum_identical_destinations=1 2026-09-06T11:44:45.508112Z 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=37815 selected_horizon=37815 endpoint_lower_bound=37584 lower_bound_gap=231 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:44:45.514680Z 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=1402 row_words=87 horizon=37815 2026-09-06T11:45:06.739712Z 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=116208 maximum_identical_destinations=1 2026-09-06T11:45:06.739758Z 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=8233 selected_horizon=8233 endpoint_lower_bound=3840 lower_bound_gap=4393 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:45:06.858967Z 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=1041 row_words=480 horizon=8233 2026-09-06T11:45:06.908454Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: evaluated exchange placement finalist offset=20480 baseline_score=1781200060416 score=680446918656 baseline_cycles=56875 candidate_cycles=56875 row_bytes=2344 row_capacity=16320 accepted=false 2026-09-06T11:45:06.926070Z 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=22900 success=true 2026-09-06T11:45:07.236159Z 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=8 success=true 2026-09-06T11:45:07.247188Z 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=10 success=true 2026-09-06T11:45:07.351249Z 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=100 success=true 2026-09-06T11:45:07.415337Z INFO writing application package tiles=1472 2026-09-06T11:45:07.458836Z INFO application package written 2026-09-06T11:45:07.460630Z INFO reading application package 2026-09-06T11:45:07.556898Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:45:07.559431Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:45:07.559445Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:45:07.559450Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:45:07.559478Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:45:07.559480Z INFO initializing IPU device 2026-09-06T11:45:07.567544Z INFO IPU device initialized 2026-09-06T11:45:07.567555Z INFO replaying device configuration register_writes=2003 2026-09-06T11:45:07.577027Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:45:07.881946Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:45:10.547433Z INFO starting host exchange session startup_mark=23 2026-09-06T11:45:10.547461Z INFO attaching host exchange pages pages=2 2026-09-06T11:45:10.547914Z INFO host exchange session started 2026-09-06T11:45:10.547960Z INFO invoking streaming host exchange call call="initialize" command=0 phases=54 2026-09-06T11:45:10.560446Z INFO invoking streaming host exchange call call="run" command=0 phases=59 profile=artifacts/layout-sweep/down-18/execution.ipuprofile tiles=1472 samples=21053 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=1440 executionTiles=1472 inputBytes=154736640 weightBytes=19869696 cycles=224802 minimumTileCycles=210096 deviceMicroseconds=149.868 effectiveGemmTflops=96.472 peakTflops=282.624 efficiencyPercent=34.13 maximumAbsoluteError=0.011719 package=artifacts/layout-sweep/down-18/model.ipuexe tiles=1472 hardwareTest=PASS