2026-09-06T11:46:04.289718Z INFO applied GEMM plan constraint source_operation=0 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 0, orientation: Normal, row_partitions: 8, column_partitions: 90, inner_partitions: 2, result_row_partitions: 2, result_column_partitions: 1, output_column_block: 48, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:46:04.787258Z INFO applied GEMM plan constraint source_operation=2 matching_plans=2 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 16, column_partitions: 23, inner_partitions: 4, result_row_partitions: 4, result_column_partitions: 1, output_column_block: 64, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:46:04.976625Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=447385 estimated_exchange_cycles=192920 exchange_row_bytes=42752 peak_standard_bytes=157440 peak_interleaved_bytes=278528 peak_total_bytes=435968 maximum_standard_allocation_bytes=114688 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:46:04.979839Z 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=1010 success=true 2026-09-06T11:46:14.987251Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=340812 estimated_exchange_cycles=109992 2026-09-06T11:46:15.019951Z 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=345 2026-09-06T11:46:15.021528Z 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=10041 success=true 2026-09-06T11:46:15.021554Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26514 exchange_phases=4 2026-09-06T11:46:15.021741Z 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 2026-09-06T11:46:15.022358Z 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 2026-09-06T11:46:15.022803Z 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=0 success=true 2026-09-06T11:46:15.025033Z 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 2026-09-06T11:46:15.142309Z 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=117 success=true 2026-09-06T11:46:15.257288Z 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=5380 maximum_identical_destinations=1 2026-09-06T11:46:15.257315Z 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=1288 selected_horizon=1256 endpoint_lower_bound=1152 lower_bound_gap=104 selected_kind="critical-neighborhood" neighborhood_improvements=1 2026-09-06T11:46:15.260694Z 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=1193 row_words=31 horizon=1256 2026-09-06T11:46:15.911649Z 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=24 ordinary_horizon=63746 paired_horizon=63749 use_paired=false 2026-09-06T11:46:15.933305Z 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=2317 maximum_identical_destinations=1 2026-09-06T11:46:15.933319Z 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=64140 selected_horizon=63746 endpoint_lower_bound=59840 lower_bound_gap=3906 selected_kind="critical-neighborhood" neighborhood_improvements=1 2026-09-06T11:46:15.942982Z 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=1428 row_words=169 horizon=63746 2026-09-06T11:46:16.568866Z 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=18432 maximum_identical_destinations=1 2026-09-06T11:46:16.568892Z 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=2183 selected_horizon=2183 endpoint_lower_bound=1536 lower_bound_gap=647 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:46:16.584552Z 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=966 row_words=98 horizon=2183 2026-09-06T11:46:16.589895Z 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=1447 success=true 2026-09-06T11:46:17.170064Z 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 2026-09-06T11:46:17.170097Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=321032 profile_bytes=328 exchange_table_bytes=984 host_code_bytes=4132 generated_code_bytes=2664 code_address=409600 standard_ranges=[(425984, 524288)] 2026-09-06T11:46:17.270461Z 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=100 success=true 2026-09-06T11:46:17.283780Z 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 2026-09-06T11:46:17.315050Z 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 2026-09-06T11:46:17.316995Z 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=5380 maximum_identical_destinations=1 2026-09-06T11:46:17.317006Z 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=1256 selected_horizon=1256 endpoint_lower_bound=1152 lower_bound_gap=104 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:46:17.320327Z 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=1193 row_words=31 horizon=1256 2026-09-06T11:46:17.457602Z 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 2026-09-06T11:46:17.466826Z 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=2317 maximum_identical_destinations=1 2026-09-06T11:46:17.466840Z 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=63746 selected_horizon=63746 endpoint_lower_bound=59840 lower_bound_gap=3906 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:46:17.475815Z 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=1428 row_words=169 horizon=63746 2026-09-06T11:46:18.258711Z 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=18432 maximum_identical_destinations=1 2026-09-06T11:46:18.258745Z 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=2183 selected_horizon=2183 endpoint_lower_bound=1536 lower_bound_gap=647 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:46:18.274366Z 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=966 row_words=98 horizon=2183 2026-09-06T11:46:18.280112Z 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=1009 success=true 2026-09-06T11:46:19.354830Z 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 2026-09-06T11:46:19.386848Z 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 2026-09-06T11:46:19.388892Z 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=5380 maximum_identical_destinations=1 2026-09-06T11:46:19.388904Z 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=1256 selected_horizon=1256 endpoint_lower_bound=1152 lower_bound_gap=104 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:46:19.392544Z 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=1193 row_words=31 horizon=1256 2026-09-06T11:46:20.072917Z 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=21 ordinary_horizon=63746 paired_horizon=63746 use_paired=false 2026-09-06T11:46:20.087356Z 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=2317 maximum_identical_destinations=1 2026-09-06T11:46:20.087371Z 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=64140 selected_horizon=63746 endpoint_lower_bound=59840 lower_bound_gap=3906 selected_kind="critical-neighborhood" neighborhood_improvements=1 2026-09-06T11:46:20.093449Z 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=1428 row_words=169 horizon=63746 2026-09-06T11:46:20.550585Z 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=18432 maximum_identical_destinations=1 2026-09-06T11:46:20.550606Z 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=2183 selected_horizon=2183 endpoint_lower_bound=1536 lower_bound_gap=647 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:46:20.560179Z 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=966 row_words=98 horizon=2183 2026-09-06T11:46:20.567524Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: evaluated exchange placement finalist offset=24576 baseline_score=1775048916992 score=683956699136 baseline_cycles=81134 candidate_cycles=81134 row_bytes=984 row_capacity=16320 accepted=false 2026-09-06T11:46:20.571480Z 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=2291 success=true 2026-09-06T11:46:20.757452Z 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=14 success=true 2026-09-06T11:46:20.768507Z 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 2026-09-06T11:46:20.904871Z 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=131 success=true 2026-09-06T11:46:20.958006Z INFO writing application package tiles=1472 2026-09-06T11:46:21.011554Z INFO application package written 2026-09-06T11:46:21.013545Z INFO reading application package 2026-09-06T11:46:21.104432Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:46:21.106870Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:46:23.027984Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:46:23.028037Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:46:23.028111Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:46:23.028118Z INFO initializing IPU device 2026-09-06T11:46:23.036396Z INFO IPU device initialized 2026-09-06T11:46:23.036440Z INFO replaying device configuration register_writes=2003 2026-09-06T11:46:23.046240Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:46:23.234991Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:46:24.820666Z INFO starting host exchange session startup_mark=23 2026-09-06T11:46:24.820696Z INFO attaching host exchange pages pages=2 2026-09-06T11:46:24.821372Z INFO host exchange session started 2026-09-06T11:46:24.821413Z INFO invoking streaming host exchange call call="initialize" command=0 phases=68 2026-09-06T11:46:24.847676Z INFO invoking streaming host exchange call call="run" command=0 phases=59 profile=artifacts/layout-sweep/joint-result-rows-8/execution.ipuprofile tiles=1472 samples=25976 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=1472 executionTiles=1472 inputBytes=151372800 weightBytes=19943424 cycles=294528 minimumTileCycles=272232 deviceMicroseconds=196.352 effectiveGemmTflops=73.634 peakTflops=282.624 efficiencyPercent=26.05 maximumAbsoluteError=0.013672 package=artifacts/layout-sweep/joint-result-rows-8/model.ipuexe tiles=1472 hardwareTest=PASS