2026-09-06T11:37:55.628887Z  INFO applied GEMM plan constraint source_operation=0 matching_plans=2 constraint=GemmPlanConstraint { source_operation: 0, orientation: Normal, row_partitions: 8, column_partitions: 45, inner_partitions: 4, result_row_partitions: 4, result_column_partitions: 1, output_column_block: 96, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct }
2026-09-06T11:37:56.025725Z  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 }
2026-09-06T11:37:56.571509Z  INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=345851 estimated_exchange_cycles=141560 exchange_row_bytes=31196 peak_standard_bytes=145884 peak_interleaved_bytes=138240 peak_total_bytes=234972 maximum_standard_allocation_bytes=114688 active_tile_counts={1440} padding_group_counts={} conversion_sources=[Some(OperationId(2))]
2026-09-06T11:37:56.575754Z  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=1234 success=true
2026-09-06T11:38:01.978684Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=280780 estimated_exchange_cycles=84936
2026-09-06T11:38:02.067417Z  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
2026-09-06T11:38:02.069342Z  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=5493 success=true
2026-09-06T11:38:02.069362Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26181 exchange_phases=4
2026-09-06T11:38:02.069571Z  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
2026-09-06T11:38:02.070250Z  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
2026-09-06T11:38:02.070930Z  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
2026-09-06T11:38:02.072702Z  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
2026-09-06T11:38:02.185829Z  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=113 success=true
2026-09-06T11:38:04.170057Z  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=33356 maximum_identical_destinations=1
2026-09-06T11:38:04.170088Z  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=4732 selected_horizon=4732 endpoint_lower_bound=4032 lower_bound_gap=700 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:38:04.198862Z  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=707 row_words=155 horizon=4732
2026-09-06T11:38:30.319530Z  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=7784 ordinary_horizon=57134 paired_horizon=47741 use_paired=true
2026-09-06T11:38:30.493089Z  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=10135 maximum_identical_destinations=1
2026-09-06T11:38:30.493111Z  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=47741 selected_horizon=47741 endpoint_lower_bound=23040 lower_bound_gap=24701 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:38:30.594012Z  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=430 row_words=541 horizon=47741
2026-09-06T11:38:36.631486Z  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
2026-09-06T11:38:36.631521Z  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
2026-09-06T11:38:36.693497Z  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
2026-09-06T11:38:36.714156Z  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=34528 success=true
2026-09-06T11:38:37.099779Z  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
2026-09-06T11:38:37.099809Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319128 profile_bytes=136 exchange_table_bytes=3384 host_code_bytes=2028 generated_code_bytes=1584 code_address=321160 standard_ranges=[(409600, 524288)]
2026-09-06T11:38:37.199718Z  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
2026-09-06T11:38:37.214762Z  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
2026-09-06T11:38:39.330351Z  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=33356 maximum_identical_destinations=1
2026-09-06T11:38:39.330389Z  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=4732 selected_horizon=4732 endpoint_lower_bound=4032 lower_bound_gap=700 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:38:39.359453Z  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=707 row_words=155 horizon=4732
2026-09-06T11:39:07.305610Z  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=7784 ordinary_horizon=57134 paired_horizon=47741 use_paired=true
2026-09-06T11:39:07.479181Z  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=10135 maximum_identical_destinations=1
2026-09-06T11:39:07.479231Z  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=47741 selected_horizon=47741 endpoint_lower_bound=23040 lower_bound_gap=24701 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:39:07.578599Z  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=430 row_words=541 horizon=47741
2026-09-06T11:39:13.212297Z  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
2026-09-06T11:39:13.212326Z  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
2026-09-06T11:39:13.252241Z  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
2026-09-06T11:39:13.269695Z  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=36069 success=true
2026-09-06T11:39:14.611694Z  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
2026-09-06T11:39:17.110230Z  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=33356 maximum_identical_destinations=1
2026-09-06T11:39:17.110261Z  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=4732 selected_horizon=4732 endpoint_lower_bound=4032 lower_bound_gap=700 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:39:17.142013Z  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=707 row_words=155 horizon=4732
2026-09-06T11:39:45.071218Z  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=7784 ordinary_horizon=55131 paired_horizon=47741 use_paired=true
2026-09-06T11:39:45.185902Z  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=10135 maximum_identical_destinations=1
2026-09-06T11:39:45.185919Z  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=47741 selected_horizon=47741 endpoint_lower_bound=23040 lower_bound_gap=24701 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:39:45.247526Z  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=794 row_words=510 horizon=47741
2026-09-06T11:39:51.919000Z  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
2026-09-06T11:39:51.919037Z  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
2026-09-06T11:39:51.985680Z  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
2026-09-06T11:39:52.019357Z  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=172542099456 baseline_cycles=74413 candidate_cycles=74413 row_bytes=3392 row_capacity=16320 accepted=false
2026-09-06T11:39:52.034076Z  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=38764 success=true
2026-09-06T11:39:52.234083Z  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
2026-09-06T11:39:52.245785Z  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
2026-09-06T11:39:52.350394Z  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=99 success=true
2026-09-06T11:39:52.401587Z  INFO writing application package tiles=1472
2026-09-06T11:39:52.494050Z  INFO application package written
2026-09-06T11:39:52.495795Z  INFO reading application package
2026-09-06T11:39:52.596009Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-06T11:39:52.598556Z  INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:39:52.598569Z  INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:39:52.598573Z  INFO opening IPU device path="/dev/ipu0"
2026-09-06T11:39:52.598602Z  INFO IPU device attached path=/dev/ipu0
2026-09-06T11:39:52.598605Z  INFO initializing IPU device
2026-09-06T11:39:52.606480Z  INFO IPU device initialized
2026-09-06T11:39:52.606486Z  INFO replaying device configuration register_writes=2003
2026-09-06T11:39:52.613657Z  INFO loading application tile_count=1472 final_mark=23
2026-09-06T11:39:52.884127Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-06T11:39:54.513629Z  INFO starting host exchange session startup_mark=23
2026-09-06T11:39:54.513665Z  INFO attaching host exchange pages pages=2
2026-09-06T11:39:54.514421Z  INFO host exchange session started
2026-09-06T11:39:54.514464Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=28
2026-09-06T11:39:54.522589Z  INFO invoking streaming host exchange call call="run" command=0 phases=33
profile=artifacts/layout-sweep/up-14/execution.ipuprofile tiles=1472 samples=25718
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=75686400 weightBytes=19869696 cycles=226974 minimumTileCycles=216690 deviceMicroseconds=151.316 effectiveGemmTflops=95.549 peakTflops=282.624 efficiencyPercent=33.81 maximumAbsoluteError=0.011719
package=artifacts/layout-sweep/up-14/model.ipuexe tiles=1472 hardwareTest=PASS