2026-09-06T12:37:31.413313Z 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 } 2026-09-06T12:37:31.983013Z 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: 1, result_column_partitions: 3, output_column_block: 48, weight_memory_class: Ipu21Standard, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T12:37:32.532628Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=310502 estimated_exchange_cycles=163640 exchange_row_bytes=36164 peak_standard_bytes=178500 peak_interleaved_bytes=60416 peak_total_bytes=238916 maximum_standard_allocation_bytes=114688 active_tile_counts={288, 1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T12:37:32.537956Z 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=1422 success=true 2026-09-06T12:37:36.096752Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=211862 estimated_exchange_cycles=78432 2026-09-06T12:37:36.176418Z 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 2026-09-06T12:37:36.177844Z 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=3639 success=true 2026-09-06T12:37:36.177875Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=21440 exchange_phases=4 2026-09-06T12:37:36.178089Z 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 2026-09-06T12:37:36.178819Z 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 2026-09-06T12:37:36.179102Z 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 2026-09-06T12:37:36.180831Z 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 2026-09-06T12:37:36.510697Z 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=329 success=true 2026-09-06T12:37:37.533183Z 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 2026-09-06T12:37:37.533234Z 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 2026-09-06T12:37:37.542409Z 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 2026-09-06T12:37:45.740261Z 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=4028 ordinary_horizon=41592 paired_horizon=32939 use_paired=true 2026-09-06T12:37:45.848612Z 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=5562 maximum_identical_destinations=1 2026-09-06T12:37:45.848655Z 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=32939 selected_horizon=32939 endpoint_lower_bound=23544 lower_bound_gap=9395 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:37:45.902943Z 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=1182 row_words=303 horizon=32939 2026-09-06T12:37:46.038428Z 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=4316 maximum_identical_destinations=1 2026-09-06T12:37:46.038472Z 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=23660 selected_horizon=22188 endpoint_lower_bound=22080 lower_bound_gap=108 selected_kind="matching-waves" neighborhood_improvements=0 2026-09-06T12:37:46.041684Z 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=229 row_words=47 horizon=22188 2026-09-06T12:37:46.044085Z 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=9533 success=true 2026-09-06T12:37:46.678054Z 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=14 success=true 2026-09-06T12:37:46.678119Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=317036 profile_bytes=120 exchange_table_bytes=1456 host_code_bytes=2592 generated_code_bytes=1412 code_address=319632 standard_ranges=[(409600, 524288)] 2026-09-06T12:37:46.777464Z 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 2026-09-06T12:37:46.790386Z 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 2026-09-06T12:37:46.959441Z 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 2026-09-06T12:37:46.970543Z 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 2026-09-06T12:37:46.970579Z 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 2026-09-06T12:37:46.991436Z 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 2026-09-06T12:37:57.278011Z 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=4028 ordinary_horizon=41592 paired_horizon=32939 use_paired=true 2026-09-06T12:37:57.374062Z 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=5562 maximum_identical_destinations=1 2026-09-06T12:37:57.374106Z 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=32939 selected_horizon=32939 endpoint_lower_bound=23544 lower_bound_gap=9395 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:37:57.422683Z 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=1182 row_words=303 horizon=32939 2026-09-06T12:37:57.487079Z 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 2026-09-06T12:37:57.489843Z 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=4316 maximum_identical_destinations=1 2026-09-06T12:37:57.489869Z 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=22188 selected_horizon=22188 endpoint_lower_bound=22080 lower_bound_gap=108 selected_kind="reused" neighborhood_improvements=0 2026-09-06T12:37:57.493131Z 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=229 row_words=47 horizon=22188 2026-09-06T12:37:57.494752Z 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=10717 success=true 2026-09-06T12:37:59.232644Z 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 2026-09-06T12:37:59.418339Z 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 2026-09-06T12:37:59.426123Z 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 2026-09-06T12:37:59.426147Z 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 2026-09-06T12:37:59.437050Z 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 2026-09-06T12:38:10.839608Z 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=4028 ordinary_horizon=41592 paired_horizon=31382 use_paired=true 2026-09-06T12:38:10.943847Z 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=5562 maximum_identical_destinations=1 2026-09-06T12:38:10.943919Z 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=31382 selected_horizon=31382 endpoint_lower_bound=23544 lower_bound_gap=7838 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:38:10.998993Z 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=1182 row_words=295 horizon=31382 2026-09-06T12:38:11.068355Z 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 2026-09-06T12:38:11.071077Z 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=4316 maximum_identical_destinations=1 2026-09-06T12:38:11.071107Z 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=22188 selected_horizon=22188 endpoint_lower_bound=22080 lower_bound_gap=108 selected_kind="reused" neighborhood_improvements=0 2026-09-06T12:38:11.074504Z 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=229 row_words=47 horizon=22188 2026-09-06T12:38:11.082474Z 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=65954 candidate_cycles=64397 row_bytes=1432 row_capacity=16320 accepted=true 2026-09-06T12:38:11.093827Z 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=13599 success=true 2026-09-06T12:38:11.471058Z 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=21 success=true 2026-09-06T12:38:11.487996Z 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 2026-09-06T12:38:11.636968Z 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=141 success=true 2026-09-06T12:38:11.727195Z INFO writing application package tiles=1472 2026-09-06T12:38:11.769394Z INFO application package written 2026-09-06T12:38:11.770636Z INFO reading application package 2026-09-06T12:38:11.883923Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T12:38:11.884293Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/alternatives-v1/device.lock 2026-09-06T12:38:13.571082Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/alternatives-v1/device.lock 2026-09-06T12:38:13.571158Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T12:38:13.571285Z INFO IPU device attached path=/dev/ipu0 2026-09-06T12:38:13.571298Z INFO initializing IPU device 2026-09-06T12:38:13.579248Z INFO IPU device initialized 2026-09-06T12:38:13.579299Z INFO replaying device configuration register_writes=2003 2026-09-06T12:38:13.589149Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T12:38:13.857456Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T12:38:17.523441Z INFO starting host exchange session startup_mark=23 2026-09-06T12:38:17.523473Z INFO attaching host exchange pages pages=2 2026-09-06T12:38:17.524015Z INFO host exchange session started 2026-09-06T12:38:17.524075Z INFO invoking streaming host exchange call call="initialize" command=0 phases=14 2026-09-06T12:38:17.530442Z INFO invoking streaming host exchange call call="run" command=0 phases=61 profile=artifacts/layout-sweep/alternatives-v1/down-result-1x3/execution.ipuprofile tiles=1472 samples=22000 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=288 executionTiles=1472 inputBytes=154736640 weightBytes=19869696 cycles=201810 minimumTileCycles=167274 deviceMicroseconds=134.540 effectiveGemmTflops=107.463 peakTflops=282.624 efficiencyPercent=38.02 maximumAbsoluteError=0.011719 package=artifacts/layout-sweep/alternatives-v1/down-result-1x3/model.ipuexe tiles=1472 hardwareTest=PASS