2026-09-06T12:37:31.420921Z 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: Batched(2), local_weight_staging: Direct } 2026-09-06T12:37:31.985906Z 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: Ipu21Standard, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T12:37:32.538257Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=263508 estimated_exchange_cycles=124560 exchange_row_bytes=27240 peak_standard_bytes=169576 peak_interleaved_bytes=60416 peak_total_bytes=229992 maximum_standard_allocation_bytes=114688 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T12:37:32.543592Z 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=1419 success=true 2026-09-06T12:37:36.255839Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=203778 estimated_exchange_cycles=77376 2026-09-06T12:37:36.541887Z 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 2026-09-06T12:37:36.546164Z 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=4002 success=true 2026-09-06T12:37:36.546230Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26264 exchange_phases=5 2026-09-06T12:37:36.546542Z 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 2026-09-06T12:37:36.548328Z 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=1 success=true 2026-09-06T12:37:36.548703Z 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 2026-09-06T12:37:36.552395Z 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 2026-09-06T12:37:37.228048Z 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=675 success=true 2026-09-06T12:37:37.543095Z 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=10222 maximum_identical_destinations=1 2026-09-06T12:37:37.543133Z 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=3498 selected_horizon=3498 endpoint_lower_bound=3312 lower_bound_gap=186 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:37:37.551899Z 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=1158 row_words=46 horizon=3498 2026-09-06T12:37:37.649029Z 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=4842 maximum_identical_destinations=1 2026-09-06T12:37:37.649068Z 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=3455 selected_horizon=3455 endpoint_lower_bound=3360 lower_bound_gap=95 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:37:37.652372Z 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=1091 row_words=39 horizon=3455 2026-09-06T12:37:45.893361Z 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=3 candidates=4028 ordinary_horizon=41592 paired_horizon=32939 use_paired=true 2026-09-06T12:37:45.991944Z 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=5562 maximum_identical_destinations=1 2026-09-06T12:37:45.991985Z 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=32939 selected_horizon=32939 endpoint_lower_bound=23544 lower_bound_gap=9395 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:37:46.039608Z 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=1182 row_words=303 horizon=32939 2026-09-06T12:37:53.776306Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: prepared large physical exchange phase phase=4 transfers=63720 maximum_identical_destinations=1 2026-09-06T12:37:53.776376Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: optimized physical exchange schedule phase=4 initial_horizon=7991 selected_horizon=7991 endpoint_lower_bound=4704 lower_bound_gap=3287 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:37:53.850436Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges_provisional"}: finished large physical exchange phase phase=4 tile=1058 row_words=294 horizon=7991 2026-09-06T12:37:53.895778Z 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=16667 success=true 2026-09-06T12:37:54.593178Z 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=7 success=true 2026-09-06T12:37:54.593253Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=317036 profile_bytes=140 exchange_table_bytes=2432 host_code_bytes=2424 generated_code_bytes=1640 code_address=319464 standard_ranges=[(409600, 524288)] 2026-09-06T12:37:54.721001Z 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=127 success=true 2026-09-06T12:37:54.730550Z 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 2026-09-06T12:37:55.055997Z 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=10222 maximum_identical_destinations=1 2026-09-06T12:37:55.056051Z 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=3498 selected_horizon=3498 endpoint_lower_bound=3312 lower_bound_gap=186 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:37:55.068709Z 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=1158 row_words=46 horizon=3498 2026-09-06T12:37:55.159828Z 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 2026-09-06T12:37:55.162043Z 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=4842 maximum_identical_destinations=1 2026-09-06T12:37:55.162060Z 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=3455 selected_horizon=3455 endpoint_lower_bound=3360 lower_bound_gap=95 selected_kind="reused" neighborhood_improvements=0 2026-09-06T12:37:55.165503Z 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=1091 row_words=39 horizon=3455 2026-09-06T12:38:07.134419Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: compared ordinary and paired exchange schedules phase=3 candidates=4028 ordinary_horizon=41592 paired_horizon=32939 use_paired=true 2026-09-06T12:38:07.229842Z 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=5562 maximum_identical_destinations=1 2026-09-06T12:38:07.229879Z 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=32939 selected_horizon=32939 endpoint_lower_bound=23544 lower_bound_gap=9395 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:38:07.278574Z 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=1182 row_words=303 horizon=32939 2026-09-06T12:38:15.570425Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: prepared large physical exchange phase phase=4 transfers=63720 maximum_identical_destinations=1 2026-09-06T12:38:15.570467Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: optimized physical exchange schedule phase=4 initial_horizon=7991 selected_horizon=7991 endpoint_lower_bound=4704 lower_bound_gap=3287 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:38:15.661184Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: finished large physical exchange phase phase=4 tile=1058 row_words=294 horizon=7991 2026-09-06T12:38:15.695160Z 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=20974 success=true 2026-09-06T12:38:17.576251Z 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 2026-09-06T12:38:17.643578Z 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 2026-09-06T12:38:17.648436Z 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=10222 maximum_identical_destinations=1 2026-09-06T12:38:17.648452Z 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=3498 selected_horizon=3498 endpoint_lower_bound=3312 lower_bound_gap=186 selected_kind="reused" neighborhood_improvements=0 2026-09-06T12:38:17.655963Z 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=1158 row_words=46 horizon=3498 2026-09-06T12:38:17.687288Z 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=2 2026-09-06T12:38:17.689501Z 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=4842 maximum_identical_destinations=1 2026-09-06T12:38:17.689514Z 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=3455 selected_horizon=3455 endpoint_lower_bound=3360 lower_bound_gap=95 selected_kind="reused" neighborhood_improvements=0 2026-09-06T12:38:17.692621Z 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=1091 row_words=39 horizon=3455 2026-09-06T12:38:26.617394Z 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=3 candidates=4028 ordinary_horizon=41592 paired_horizon=31382 use_paired=true 2026-09-06T12:38:26.716619Z 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=5562 maximum_identical_destinations=1 2026-09-06T12:38:26.716652Z 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=31382 selected_horizon=31382 endpoint_lower_bound=23544 lower_bound_gap=7838 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:38:26.792029Z 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=1182 row_words=295 horizon=31382 2026-09-06T12:38:33.707309Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: prepared large physical exchange phase phase=4 transfers=63720 maximum_identical_destinations=1 2026-09-06T12:38:33.707346Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: optimized physical exchange schedule phase=4 initial_horizon=7991 selected_horizon=7991 endpoint_lower_bound=4704 lower_bound_gap=3287 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:38:33.770101Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: finished large physical exchange phase phase=4 tile=1058 row_words=294 horizon=7991 2026-09-06T12:38:33.811956Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: evaluated exchange placement finalist offset=28672 baseline_score=170111729664 score=151709220864 baseline_cycles=54931 candidate_cycles=53374 row_bytes=2432 row_capacity=16320 accepted=true 2026-09-06T12:38:33.825452Z 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=18130 success=true 2026-09-06T12:38:34.193143Z 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=13 success=true 2026-09-06T12:38:34.204748Z 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 2026-09-06T12:38:34.298485Z 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=88 success=true 2026-09-06T12:38:34.369093Z INFO writing application package tiles=1472 2026-09-06T12:38:34.404675Z INFO application package written 2026-09-06T12:38:34.405419Z INFO reading application package 2026-09-06T12:38:34.506120Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T12:38:34.506475Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/alternatives-v1/device.lock 2026-09-06T12:38:52.248962Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/alternatives-v1/device.lock 2026-09-06T12:38:52.249023Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T12:38:52.249115Z INFO IPU device attached path=/dev/ipu0 2026-09-06T12:38:52.249122Z INFO initializing IPU device 2026-09-06T12:38:52.257238Z INFO IPU device initialized 2026-09-06T12:38:52.257264Z INFO replaying device configuration register_writes=2003 2026-09-06T12:38:52.266699Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T12:38:52.545160Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T12:38:55.191057Z INFO starting host exchange session startup_mark=23 2026-09-06T12:38:55.191090Z INFO attaching host exchange pages pages=2 2026-09-06T12:38:55.191488Z INFO host exchange session started 2026-09-06T12:38:55.191532Z INFO invoking streaming host exchange call call="initialize" command=0 phases=14 2026-09-06T12:38:55.197362Z INFO invoking streaming host exchange call call="run" command=0 phases=59 profile=artifacts/layout-sweep/alternatives-v1/up-batch-2/execution.ipuprofile tiles=1472 samples=28732 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=188868 minimumTileCycles=176412 deviceMicroseconds=125.912 effectiveGemmTflops=114.827 peakTflops=282.624 efficiencyPercent=40.63 maximumAbsoluteError=0.011719 package=artifacts/layout-sweep/alternatives-v1/up-batch-2/model.ipuexe tiles=1472 hardwareTest=PASS