2026-09-06T11:37:19.688956Z INFO applied GEMM plan constraint source_operation=0 matching_plans=2 constraint=GemmPlanConstraint { source_operation: 0, orientation: Normal, row_partitions: 6, column_partitions: 54, inner_partitions: 4, result_row_partitions: 4, result_column_partitions: 1, output_column_block: 80, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:37:20.078407Z INFO applied GEMM plan constraint source_operation=2 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 4, column_partitions: 24, inner_partitions: 15, result_row_partitions: 15, result_column_partitions: 1, output_column_block: 48, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:37:20.617988Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=344453 estimated_exchange_cycles=136760 exchange_row_bytes=30116 peak_standard_bytes=144804 peak_interleaved_bytes=119808 peak_total_bytes=231844 maximum_standard_allocation_bytes=114688 active_tile_counts={1296, 1440} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:37:20.622364Z 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=1227 success=true 2026-09-06T11:37:25.035711Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=280376 estimated_exchange_cycles=80488 2026-09-06T11:37:25.119568Z 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 2026-09-06T11:37:25.121347Z 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=4498 success=true 2026-09-06T11:37:25.121365Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=24624 exchange_phases=4 2026-09-06T11:37:25.121551Z 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 2026-09-06T11:37:25.122101Z 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 2026-09-06T11:37:25.122327Z 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_k288_c64_r120_r122" target=ipu21 2026-09-06T11:37:25.175586Z 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/35c61145c66a9be0ebba82ad6d42e1a1d03bbfe99f9ebab3e50f9ac304018ab8/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/35c61145c66a9be0ebba82ad6d42e1a1d03bbfe99f9ebab3e50f9ac304018ab8/kernel.json 2026-09-06T11:37:25.175793Z 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_k288_c80_r120_r122" target=ipu21 2026-09-06T11:37:25.225065Z 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/3144630b871643a5ef771ff0fb9609aaf7943809ee16d9836c161254b3704288/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/3144630b871643a5ef771ff0fb9609aaf7943809ee16d9836c161254b3704288/kernel.json 2026-09-06T11:37:25.225454Z 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=103 success=true 2026-09-06T11:37:25.228571Z 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 2026-09-06T11:37:25.343458Z 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=114 success=true 2026-09-06T11:37:26.233952Z 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=23672 maximum_identical_destinations=1 2026-09-06T11:37:26.233993Z 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=4838 selected_horizon=4705 endpoint_lower_bound=4080 lower_bound_gap=625 selected_kind="critical-neighborhood" neighborhood_improvements=1 2026-09-06T11:37:26.251359Z 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=413 row_words=118 horizon=4705 2026-09-06T11:37:40.165476Z 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=6072 ordinary_horizon=45566 paired_horizon=37896 use_paired=true 2026-09-06T11:37:40.300743Z 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=7983 maximum_identical_destinations=1 2026-09-06T11:37:40.300767Z 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=37896 selected_horizon=37896 endpoint_lower_bound=22608 lower_bound_gap=15288 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:37:40.372604Z 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=507 row_words=403 horizon=37896 2026-09-06T11:37:46.797707Z 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=63720 maximum_identical_destinations=1 2026-09-06T11:37:46.797742Z 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=7991 selected_horizon=7991 endpoint_lower_bound=4704 lower_bound_gap=3287 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:37:46.859812Z 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=1058 row_words=294 horizon=7991 2026-09-06T11:37:46.880215Z 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=21536 success=true 2026-09-06T11:37:47.303072Z 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=4 success=true 2026-09-06T11:37:47.303105Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319128 profile_bytes=132 exchange_table_bytes=2848 host_code_bytes=2196 generated_code_bytes=1460 code_address=321328 standard_ranges=[(409600, 524288)] 2026-09-06T11:37:47.402462Z 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=99 success=true 2026-09-06T11:37:47.412694Z 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 2026-09-06T11:37:47.567592Z 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 2026-09-06T11:37:47.577165Z 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=23672 maximum_identical_destinations=1 2026-09-06T11:37:47.577182Z 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=4705 selected_horizon=4705 endpoint_lower_bound=4080 lower_bound_gap=625 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:37:47.595552Z 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=413 row_words=118 horizon=4705 2026-09-06T11:38:00.308622Z 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=6072 ordinary_horizon=45566 paired_horizon=37896 use_paired=true 2026-09-06T11:38:00.407040Z 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=7983 maximum_identical_destinations=1 2026-09-06T11:38:00.407070Z 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=37896 selected_horizon=37896 endpoint_lower_bound=22608 lower_bound_gap=15288 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:00.468078Z 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=507 row_words=403 horizon=37896 2026-09-06T11:38:04.939619Z 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=63720 maximum_identical_destinations=1 2026-09-06T11:38:04.939648Z 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=7991 selected_horizon=7991 endpoint_lower_bound=4704 lower_bound_gap=3287 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:04.978366Z 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=1058 row_words=294 horizon=7991 2026-09-06T11:38:04.994833Z 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=17592 success=true 2026-09-06T11:38:06.216386Z 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 2026-09-06T11:38:06.375330Z 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 2026-09-06T11:38:06.385243Z 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=23672 maximum_identical_destinations=1 2026-09-06T11:38:06.385262Z 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=4705 selected_horizon=4705 endpoint_lower_bound=4080 lower_bound_gap=625 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:38:06.403303Z 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=413 row_words=118 horizon=4705 2026-09-06T11:38:24.185926Z 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=6072 ordinary_horizon=41282 paired_horizon=35352 use_paired=true 2026-09-06T11:38:24.313627Z 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=7983 maximum_identical_destinations=1 2026-09-06T11:38:24.313659Z 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=35352 selected_horizon=35352 endpoint_lower_bound=22608 lower_bound_gap=12744 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:24.385493Z 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=504 row_words=382 horizon=35352 2026-09-06T11:38:30.935619Z 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=63720 maximum_identical_destinations=1 2026-09-06T11:38:30.935652Z 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=7991 selected_horizon=7991 endpoint_lower_bound=4704 lower_bound_gap=3287 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:31.001456Z 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=1058 row_words=294 horizon=7991 2026-09-06T11:38:31.033001Z 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=400669507584 score=171607818240 baseline_cycles=62234 candidate_cycles=59690 row_bytes=2800 row_capacity=16320 accepted=true 2026-09-06T11:38:31.045671Z 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=26050 success=true 2026-09-06T11:38:31.260325Z 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 2026-09-06T11:38:31.270975Z 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 2026-09-06T11:38:31.374496Z 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=98 success=true 2026-09-06T11:38:31.427087Z INFO writing application package tiles=1472 2026-09-06T11:38:31.473585Z INFO application package written 2026-09-06T11:38:31.475780Z INFO reading application package 2026-09-06T11:38:31.570375Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:38:31.572579Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:38:31.572591Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:38:31.572600Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:38:31.572629Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:38:31.572632Z INFO initializing IPU device 2026-09-06T11:38:31.580573Z INFO IPU device initialized 2026-09-06T11:38:31.580581Z INFO replaying device configuration register_writes=2003 2026-09-06T11:38:31.587869Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:38:31.862157Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:38:33.750811Z INFO starting host exchange session startup_mark=23 2026-09-06T11:38:33.750843Z INFO attaching host exchange pages pages=2 2026-09-06T11:38:33.751539Z INFO host exchange session started 2026-09-06T11:38:33.751599Z INFO invoking streaming host exchange call call="initialize" command=0 phases=24 2026-09-06T11:38:33.757860Z INFO invoking streaming host exchange call call="run" command=0 phases=43 profile=artifacts/layout-sweep/up-13/execution.ipuprofile tiles=1472 samples=24941 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=90823680 weightBytes=19869696 cycles=211002 minimumTileCycles=195978 deviceMicroseconds=140.668 effectiveGemmTflops=102.782 peakTflops=282.624 efficiencyPercent=36.37 maximumAbsoluteError=0.011719 package=artifacts/layout-sweep/up-13/model.ipuexe tiles=1472 hardwareTest=PASS