2026-09-06T11:28:00.045917Z INFO applied GEMM plan constraint source_operation=0 matching_plans=2 constraint=GemmPlanConstraint { source_operation: 0, orientation: Normal, row_partitions: 6, column_partitions: 16, inner_partitions: 15, result_row_partitions: 15, result_column_partitions: 1, output_column_block: 272, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:28:00.416467Z 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 } 2026-09-06T11:28:00.967616Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=439718 estimated_exchange_cycles=173240 exchange_row_bytes=38324 peak_standard_bytes=153012 peak_interleaved_bytes=169472 peak_total_bytes=256756 maximum_standard_allocation_bytes=114688 active_tile_counts={1440} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:28:00.971941Z 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=1218 success=true 2026-09-06T11:28:05.575183Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=451662 estimated_exchange_cycles=187800 2026-09-06T11:28:05.669910Z 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 2026-09-06T11:28:05.674273Z 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=4702 success=true 2026-09-06T11:28:05.674292Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=25440 exchange_phases=4 2026-09-06T11:28:05.674499Z 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 2026-09-06T11:28:05.677562Z 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=3 success=true 2026-09-06T11:28:05.677708Z 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_k80_c256_r120_r122" target=ipu21 2026-09-06T11:28:05.729852Z 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/fb650dfde5faf629fd01d105ea9de35d303e3e8ba54c93266ade235d8e9e2c52/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/fb650dfde5faf629fd01d105ea9de35d303e3e8ba54c93266ade235d8e9e2c52/kernel.json 2026-09-06T11:28:05.730184Z 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_k80_c272_r120_r122" target=ipu21 2026-09-06T11:28:05.782383Z 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/298787e028dd67b95ae16cc20d3648c049d46f5fa7415b24eb8ac0e172153b47/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/298787e028dd67b95ae16cc20d3648c049d46f5fa7415b24eb8ac0e172153b47/kernel.json 2026-09-06T11:28:05.782879Z 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=105 success=true 2026-09-06T11:28:05.785664Z 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 2026-09-06T11:28:05.905107Z 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=119 success=true 2026-09-06T11:30:11.989313Z 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=351314 maximum_identical_destinations=1 2026-09-06T11:30:11.989365Z 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=31256 selected_horizon=31256 endpoint_lower_bound=19040 lower_bound_gap=12216 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:30:12.576754Z 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=948 row_words=1405 horizon=31256 2026-09-06T11:31:12.949986Z 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=25737 maximum_identical_destinations=1 2026-09-06T11:31:12.950030Z 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=48366 selected_horizon=48366 endpoint_lower_bound=33408 lower_bound_gap=14958 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:31:13.124141Z 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=932 row_words=1216 horizon=48366 2026-09-06T11:31:16.926916Z 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 2026-09-06T11:31:16.926944Z 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 2026-09-06T11:31:16.969045Z 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 2026-09-06T11:31:16.983834Z 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=191078 success=true 2026-09-06T11:31:17.193575Z 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 2026-09-06T11:31:17.193608Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319128 profile_bytes=132 exchange_table_bytes=10452 host_code_bytes=1324 generated_code_bytes=2028 code_address=320456 standard_ranges=[(409600, 524288)] 2026-09-06T11:31:17.297345Z 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 2026-09-06T11:31:17.308410Z 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 2026-09-06T11:33:28.752495Z 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=351314 maximum_identical_destinations=1 2026-09-06T11:33:28.752536Z 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=31256 selected_horizon=31256 endpoint_lower_bound=19040 lower_bound_gap=12216 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:33:29.074855Z 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=948 row_words=1405 horizon=31256 2026-09-06T11:34:33.025240Z 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=25737 maximum_identical_destinations=1 2026-09-06T11:34:33.025273Z 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=48366 selected_horizon=48366 endpoint_lower_bound=33408 lower_bound_gap=14958 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:34:33.176929Z 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=932 row_words=1216 horizon=48366 2026-09-06T11:34:37.394805Z 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 2026-09-06T11:34:37.394834Z 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 2026-09-06T11:34:37.436285Z 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 2026-09-06T11:34:37.452417Z 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=200155 success=true 2026-09-06T11:34:39.150551Z 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 2026-09-06T11:36:56.988808Z 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=351314 maximum_identical_destinations=1 2026-09-06T11:36:56.988847Z 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=31256 selected_horizon=31256 endpoint_lower_bound=19040 lower_bound_gap=12216 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:36:57.304024Z 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=948 row_words=1405 horizon=31256 2026-09-06T11:37:59.494608Z 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=25737 maximum_identical_destinations=1 2026-09-06T11:37:59.494651Z 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=54910 selected_horizon=54910 endpoint_lower_bound=33408 lower_bound_gap=21502 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:37:59.651396Z 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=932 row_words=1214 horizon=54910 2026-09-06T11:38:04.215048Z 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 2026-09-06T11:38:04.215076Z 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 2026-09-06T11:38:04.259074Z 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 2026-09-06T11:38:04.296254Z 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=448685899776 score=171487494144 baseline_cycles=98615 candidate_cycles=105159 row_bytes=10452 row_capacity=16320 accepted=false 2026-09-06T11:38:04.340040Z 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=206887 success=true 2026-09-06T11:38:04.451323Z 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=9 success=true 2026-09-06T11:38:04.467880Z 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=16 success=true 2026-09-06T11:38:04.576541Z 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 2026-09-06T11:38:04.650756Z INFO writing application package tiles=1472 2026-09-06T11:38:04.752012Z INFO application package written 2026-09-06T11:38:04.754587Z INFO reading application package 2026-09-06T11:38:04.909537Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:38:04.913372Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:38:04.913384Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:38:04.913389Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:38:04.913418Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:38:04.913421Z INFO initializing IPU device 2026-09-06T11:38:04.921503Z INFO IPU device initialized 2026-09-06T11:38:04.921526Z INFO replaying device configuration register_writes=2003 2026-09-06T11:38:04.927742Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:38:05.229014Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:38:06.840161Z INFO starting host exchange session startup_mark=23 2026-09-06T11:38:06.840191Z INFO attaching host exchange pages pages=2 2026-09-06T11:38:06.840869Z INFO host exchange session started 2026-09-06T11:38:06.840921Z INFO invoking streaming host exchange call call="initialize" command=0 phases=22 2026-09-06T11:38:06.847631Z INFO invoking streaming host exchange call call="run" command=0 phases=17 profile=artifacts/layout-sweep/up-06/execution.ipuprofile tiles=1472 samples=25548 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=28032000 weightBytes=20282880 cycles=271446 minimumTileCycles=266022 deviceMicroseconds=180.964 effectiveGemmTflops=79.895 peakTflops=282.624 efficiencyPercent=28.27 maximumAbsoluteError=0.011719 package=artifacts/layout-sweep/up-06/model.ipuexe tiles=1472 hardwareTest=PASS