2026-09-06T11:42:01.423349Z 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 } 2026-09-06T11:42:01.876550Z INFO applied GEMM plan constraint source_operation=2 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 3, column_partitions: 36, inner_partitions: 13, result_row_partitions: 13, result_column_partitions: 1, output_column_block: 32, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:42:02.542641Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=342005 estimated_exchange_cycles=154200 exchange_row_bytes=34040 peak_standard_bytes=214264 peak_interleaved_bytes=81920 peak_total_bytes=268536 maximum_standard_allocation_bytes=180224 active_tile_counts={1404, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:42:02.548873Z 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=1417 success=true 2026-09-06T11:42:06.113510Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=250841 estimated_exchange_cycles=77032 2026-09-06T11:42:06.262680Z 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=72 2026-09-06T11:42:06.264641Z 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=3715 success=true 2026-09-06T11:42:06.264660Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26084 exchange_phases=4 2026-09-06T11:42:06.264864Z 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 2026-09-06T11:42:06.265535Z 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 2026-09-06T11:42:06.265822Z 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_k336_c32_r242_r244" target=ipu21 2026-09-06T11:42:06.320437Z 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/5e01f836edd1a43c75be7d567637f5862e51033dce354118a6cad71517802216/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/5e01f836edd1a43c75be7d567637f5862e51033dce354118a6cad71517802216/kernel.json 2026-09-06T11:42:06.320763Z 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=55 success=true 2026-09-06T11:42:06.323997Z 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 2026-09-06T11:42:06.439463Z 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=115 success=true 2026-09-06T11:42:06.788722Z 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 2026-09-06T11:42:06.788762Z 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 2026-09-06T11:42:06.799228Z 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 2026-09-06T11:42:14.747149Z 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=4249 ordinary_horizon=54501 paired_horizon=44191 use_paired=true 2026-09-06T11:42:14.866352Z 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=5778 maximum_identical_destinations=1 2026-09-06T11:42:14.866376Z 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=46063 selected_horizon=44191 endpoint_lower_bound=30912 lower_bound_gap=13279 selected_kind="critical-neighborhood" neighborhood_improvements=3 2026-09-06T11:42:14.928044Z 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=1112 row_words=393 horizon=44191 2026-09-06T11:42:18.085873Z 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=35568 maximum_identical_destinations=1 2026-09-06T11:42:18.085923Z 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=6113 selected_horizon=5268 endpoint_lower_bound=4096 lower_bound_gap=1172 selected_kind="critical-neighborhood" neighborhood_improvements=2 2026-09-06T11:42:18.121293Z 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=905 row_words=163 horizon=5268 2026-09-06T11:42:18.134650Z 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=11695 success=true 2026-09-06T11:42:18.760583Z 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 2026-09-06T11:42:18.760620Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319176 profile_bytes=148 exchange_table_bytes=2264 host_code_bytes=2396 generated_code_bytes=1664 code_address=321576 standard_ranges=[(409600, 524288)] 2026-09-06T11:42:18.865042Z 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=104 success=true 2026-09-06T11:42:18.873016Z 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 2026-09-06T11:42:18.938051Z 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 2026-09-06T11:42:18.942173Z 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 2026-09-06T11:42:18.942185Z 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 2026-09-06T11:42:18.948597Z 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 2026-09-06T11:42:20.223439Z 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 2026-09-06T11:42:20.261452Z 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=5778 maximum_identical_destinations=1 2026-09-06T11:42:20.261470Z 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=44191 selected_horizon=44191 endpoint_lower_bound=30912 lower_bound_gap=13279 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:42:20.318192Z 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=1112 row_words=393 horizon=44191 2026-09-06T11:42:20.599871Z 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 2026-09-06T11:42:20.615165Z 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=35568 maximum_identical_destinations=1 2026-09-06T11:42:20.615179Z 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=5268 selected_horizon=5268 endpoint_lower_bound=4096 lower_bound_gap=1172 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:42:20.649310Z 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=905 row_words=163 horizon=5268 2026-09-06T11:42:20.658981Z 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=1793 success=true 2026-09-06T11:42:22.191260Z 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 2026-09-06T11:42:22.283387Z 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 2026-09-06T11:42:22.289563Z 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 2026-09-06T11:42:22.289576Z 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 2026-09-06T11:42:22.300016Z 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 2026-09-06T11:42:28.917389Z 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=3179 ordinary_horizon=55927 paired_horizon=56249 use_paired=false 2026-09-06T11:42:29.028981Z 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=5778 maximum_identical_destinations=1 2026-09-06T11:42:29.029012Z 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=55927 selected_horizon=55927 endpoint_lower_bound=46368 lower_bound_gap=9559 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:42:29.082810Z 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=720 row_words=336 horizon=55927 2026-09-06T11:42:29.369356Z 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 2026-09-06T11:42:29.385378Z 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=35568 maximum_identical_destinations=1 2026-09-06T11:42:29.385395Z 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=5268 selected_horizon=5268 endpoint_lower_bound=4096 lower_bound_gap=1172 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:42:29.419270Z 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=905 row_words=163 horizon=5268 2026-09-06T11:42:29.437411Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: evaluated exchange placement finalist offset=8192 baseline_score=345526960128 score=139917066240 baseline_cycles=60286 candidate_cycles=72022 row_bytes=2084 row_capacity=16320 accepted=false 2026-09-06T11:42:29.446293Z 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=8787 success=true 2026-09-06T11:42:29.759330Z 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=10 success=true 2026-09-06T11:42:29.771071Z 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 2026-09-06T11:42:29.876281Z 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 2026-09-06T11:42:29.934363Z INFO writing application package tiles=1472 2026-09-06T11:42:30.008606Z INFO application package written 2026-09-06T11:42:30.009921Z INFO reading application package 2026-09-06T11:42:30.114048Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:42:30.116521Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:42:30.116537Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:42:30.116542Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:42:30.116570Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:42:30.116573Z INFO initializing IPU device 2026-09-06T11:42:30.124641Z INFO IPU device initialized 2026-09-06T11:42:30.124647Z INFO replaying device configuration register_writes=2003 2026-09-06T11:42:30.133355Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:42:30.466624Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:42:33.020563Z INFO starting host exchange session startup_mark=23 2026-09-06T11:42:33.020598Z INFO attaching host exchange pages pages=2 2026-09-06T11:42:33.021041Z INFO host exchange session started 2026-09-06T11:42:33.021089Z INFO invoking streaming host exchange call call="initialize" command=0 phases=14 2026-09-06T11:42:33.026923Z INFO invoking streaming host exchange call call="run" command=0 phases=59 profile=artifacts/layout-sweep/down-11/execution.ipuprofile tiles=1472 samples=28919 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=19980288 cycles=206226 minimumTileCycles=191850 deviceMicroseconds=137.484 effectiveGemmTflops=105.162 peakTflops=282.624 efficiencyPercent=37.21 maximumAbsoluteError=0.009766 package=artifacts/layout-sweep/down-11/model.ipuexe tiles=1472 hardwareTest=PASS