2026-09-06T12:42:49.874004Z  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 }
2026-09-06T12:42:50.341599Z  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: 5, result_column_partitions: 3, output_column_block: 48, weight_memory_class: Ipu21Standard, reduction_staging: Complete, local_weight_staging: Direct }
2026-09-06T12:42:50.958231Z  INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=258236 estimated_exchange_cycles=122840 exchange_row_bytes=26984 peak_standard_bytes=169320 peak_interleaved_bytes=60416 peak_total_bytes=229736 maximum_standard_allocation_bytes=114688 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))]
2026-09-06T12:42:50.963841Z  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=1298 success=true
2026-09-06T12:42:54.655547Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=186526 estimated_exchange_cycles=65184
2026-09-06T12:42:54.782368Z  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
2026-09-06T12:42:54.785963Z  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=3822 success=true
2026-09-06T12:42:54.785998Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26696 exchange_phases=4
2026-09-06T12:42:54.786321Z  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
2026-09-06T12:42:54.787672Z  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
2026-09-06T12:42:54.788041Z  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
2026-09-06T12:42:54.791188Z  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
2026-09-06T12:42:54.921428Z  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=130 success=true
2026-09-06T12:42:55.313238Z  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
2026-09-06T12:42:55.313281Z  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=3755 endpoint_lower_bound=3408 lower_bound_gap=347 selected_kind="matching-waves" neighborhood_improvements=0
2026-09-06T12:42:55.333529Z  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=3755
2026-09-06T12:43:03.914258Z  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=41192 paired_horizon=34545 use_paired=true
2026-09-06T12:43:04.015319Z  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
2026-09-06T12:43:04.015355Z  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=34545 selected_horizon=34545 endpoint_lower_bound=23544 lower_bound_gap=11001 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T12:43:04.071415Z  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=1337 row_words=320 horizon=34545
2026-09-06T12:43:04.971119Z  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=24760 maximum_identical_destinations=1
2026-09-06T12:43:04.971163Z  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=6230 selected_horizon=5860 endpoint_lower_bound=4560 lower_bound_gap=1300 selected_kind="matching-waves" neighborhood_improvements=0
2026-09-06T12:43:04.997391Z  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=964 row_words=132 horizon=5860
2026-09-06T12:43:05.009221Z  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=10087 success=true
2026-09-06T12:43:05.752164Z  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=6 success=true
2026-09-06T12:43:05.752241Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=317036 profile_bytes=132 exchange_table_bytes=1940 host_code_bytes=2400 generated_code_bytes=1504 code_address=319440 standard_ranges=[(409600, 524288)]
2026-09-06T12:43:05.866446Z  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=114 success=true
2026-09-06T12:43:05.875060Z  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
2026-09-06T12:43:05.979579Z  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
2026-09-06T12:43:05.986668Z  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
2026-09-06T12:43:05.986687Z  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=3755 selected_horizon=3755 endpoint_lower_bound=3408 lower_bound_gap=347 selected_kind="reused" neighborhood_improvements=0
2026-09-06T12:43:05.997300Z  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=3755
2026-09-06T12:43:15.485567Z  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=41192 paired_horizon=34545 use_paired=true
2026-09-06T12:43:15.597753Z  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
2026-09-06T12:43:15.597791Z  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=34545 selected_horizon=34545 endpoint_lower_bound=23544 lower_bound_gap=11001 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T12:43:15.657844Z  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=1337 row_words=320 horizon=34545
2026-09-06T12:43:15.904000Z  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
2026-09-06T12:43:15.918113Z  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=24760 maximum_identical_destinations=1
2026-09-06T12:43:15.918136Z  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=5860 selected_horizon=5860 endpoint_lower_bound=4560 lower_bound_gap=1300 selected_kind="reused" neighborhood_improvements=0
2026-09-06T12:43:15.951256Z  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=964 row_words=132 horizon=5860
2026-09-06T12:43:15.959925Z  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=10093 success=true
2026-09-06T12:43:17.894977Z  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
2026-09-06T12:43:17.999933Z  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
2026-09-06T12:43:18.007723Z  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
2026-09-06T12:43:18.007752Z  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=3755 selected_horizon=3755 endpoint_lower_bound=3408 lower_bound_gap=347 selected_kind="reused" neighborhood_improvements=0
2026-09-06T12:43:18.018872Z  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=3755
2026-09-06T12:43:27.605029Z  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=41192 paired_horizon=30716 use_paired=true
2026-09-06T12:43:27.698939Z  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
2026-09-06T12:43:27.698992Z  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=30716 selected_horizon=30716 endpoint_lower_bound=23544 lower_bound_gap=7172 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T12:43:27.747866Z  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=1337 row_words=318 horizon=30716
2026-09-06T12:43:27.999232Z  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
2026-09-06T12:43:28.013532Z  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=24760 maximum_identical_destinations=1
2026-09-06T12:43:28.013560Z  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=5860 selected_horizon=5860 endpoint_lower_bound=4560 lower_bound_gap=1300 selected_kind="reused" neighborhood_improvements=0
2026-09-06T12:43:28.039907Z  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=964 row_words=132 horizon=5860
2026-09-06T12:43:28.057296Z  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=51208 candidate_cycles=47379 row_bytes=1932 row_capacity=16320 accepted=true
2026-09-06T12:43:28.069032Z  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=12109 success=true
2026-09-06T12:43:28.406812Z  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=17 success=true
2026-09-06T12:43:28.419889Z  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=13 success=true
2026-09-06T12:43:28.522685Z  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=97 success=true
2026-09-06T12:43:28.615931Z  INFO writing application package tiles=1472
2026-09-06T12:43:28.658461Z  INFO application package written
2026-09-06T12:43:28.659402Z  INFO reading application package
2026-09-06T12:43:28.768135Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-06T12:43:28.768689Z  INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/alternatives-confirm/device.lock
2026-09-06T12:43:28.768704Z  INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/alternatives-confirm/device.lock
2026-09-06T12:43:28.768710Z  INFO opening IPU device path="/dev/ipu0"
2026-09-06T12:43:28.768757Z  INFO IPU device attached path=/dev/ipu0
2026-09-06T12:43:28.768760Z  INFO initializing IPU device
2026-09-06T12:43:28.776594Z  INFO IPU device initialized
2026-09-06T12:43:28.776599Z  INFO replaying device configuration register_writes=2003
2026-09-06T12:43:28.786263Z  INFO loading application tile_count=1472 final_mark=23
2026-09-06T12:43:29.068932Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-06T12:43:31.588883Z  INFO starting host exchange session startup_mark=23
2026-09-06T12:43:31.588933Z  INFO attaching host exchange pages pages=2
2026-09-06T12:43:31.589449Z  INFO host exchange session started
2026-09-06T12:43:31.589676Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=14
2026-09-06T12:43:31.598604Z  INFO invoking streaming host exchange call call="run" command=0 phases=59
profile=artifacts/layout-sweep/alternatives-confirm/winner-repeat-1/execution.ipuprofile tiles=1472 samples=24596
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=178962 minimumTileCycles=168396 deviceMicroseconds=119.308 effectiveGemmTflops=121.183 peakTflops=282.624 efficiencyPercent=42.88 maximumAbsoluteError=0.011719
package=artifacts/layout-sweep/alternatives-confirm/winner-repeat-1/model.ipuexe tiles=1472 hardwareTest=PASS