2026-09-06T11:38:57.247537Z  INFO applied GEMM plan constraint source_operation=0 matching_plans=2 constraint=GemmPlanConstraint { source_operation: 0, orientation: Normal, row_partitions: 4, column_partitions: 92, inner_partitions: 4, result_row_partitions: 1, result_column_partitions: 1, output_column_block: 48, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct }
2026-09-06T11:38:57.708784Z  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 }
2026-09-06T11:38:58.248125Z  INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=454409 estimated_exchange_cycles=148760 exchange_row_bytes=32816 peak_standard_bytes=147504 peak_interleaved_bytes=88064 peak_total_bytes=235568 maximum_standard_allocation_bytes=114688 active_tile_counts={368, 1440} padding_group_counts={} conversion_sources=[Some(OperationId(2))]
2026-09-06T11:38:58.253338Z  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=1293 success=true
2026-09-06T11:39:01.756464Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=393513 estimated_exchange_cycles=80904
2026-09-06T11:39:01.832837Z  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
2026-09-06T11:39:01.834182Z  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=3580 success=true
2026-09-06T11:39:01.834216Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=20192 exchange_phases=4
2026-09-06T11:39:01.834410Z  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
2026-09-06T11:39:01.834845Z  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
2026-09-06T11:39:01.835140Z  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
2026-09-06T11:39:01.836897Z  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
2026-09-06T11:39:01.927816Z  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=90 success=true
2026-09-06T11:39:01.990192Z  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=2655 maximum_identical_destinations=1
2026-09-06T11:39:01.990241Z  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=17620 selected_horizon=17620 endpoint_lower_bound=17472 lower_bound_gap=148 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:39:01.991969Z  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=157 row_words=27 horizon=17620
2026-09-06T11:39:02.428634Z  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=788 ordinary_horizon=41976 paired_horizon=41793 use_paired=true
2026-09-06T11:39:02.440823Z  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=1520 maximum_identical_destinations=1
2026-09-06T11:39:02.440836Z  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=41793 selected_horizon=41793 endpoint_lower_bound=33408 lower_bound_gap=8385 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:39:02.447325Z  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=323 row_words=56 horizon=41793
2026-09-06T11:39:08.479464Z  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
2026-09-06T11:39:08.479509Z  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
2026-09-06T11:39:08.539723Z  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
2026-09-06T11:39:08.559406Z  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=6631 success=true
2026-09-06T11:39:09.171829Z  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=5 success=true
2026-09-06T11:39:09.171865Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=317272 profile_bytes=112 exchange_table_bytes=1352 host_code_bytes=2412 generated_code_bytes=1292 code_address=319688 standard_ranges=[(409600, 524288)]
2026-09-06T11:39:09.249942Z  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=78 success=true
2026-09-06T11:39:09.257721Z  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
2026-09-06T11:39:09.275408Z  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
2026-09-06T11:39:09.276520Z  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=2655 maximum_identical_destinations=1
2026-09-06T11:39:09.276532Z  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=17620 selected_horizon=17620 endpoint_lower_bound=17472 lower_bound_gap=148 selected_kind="reused" neighborhood_improvements=0
2026-09-06T11:39:09.278269Z  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=157 row_words=27 horizon=17620
2026-09-06T11:39:09.400056Z  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
2026-09-06T11:39:09.404176Z  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=1520 maximum_identical_destinations=1
2026-09-06T11:39:09.404185Z  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=41793 selected_horizon=41793 endpoint_lower_bound=33408 lower_bound_gap=8385 selected_kind="reused" neighborhood_improvements=0
2026-09-06T11:39:09.410457Z  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=323 row_words=56 horizon=41793
2026-09-06T11:39:15.885989Z  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
2026-09-06T11:39:15.886023Z  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
2026-09-06T11:39:15.959375Z  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
2026-09-06T11:39:15.988426Z  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=6738 success=true
2026-09-06T11:39:17.109411Z  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
2026-09-06T11:39:17.127619Z  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
2026-09-06T11:39:17.128725Z  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=2655 maximum_identical_destinations=1
2026-09-06T11:39:17.128737Z  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=17620 selected_horizon=17620 endpoint_lower_bound=17472 lower_bound_gap=148 selected_kind="reused" neighborhood_improvements=0
2026-09-06T11:39:17.130579Z  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=157 row_words=27 horizon=17620
2026-09-06T11:39:17.681664Z  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=788 ordinary_horizon=47191 paired_horizon=44522 use_paired=true
2026-09-06T11:39:17.694669Z  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=1520 maximum_identical_destinations=1
2026-09-06T11:39:17.694688Z  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=44522 selected_horizon=44522 endpoint_lower_bound=33408 lower_bound_gap=11114 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:39:17.701307Z  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=323 row_words=56 horizon=44522
2026-09-06T11:39:24.181657Z  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
2026-09-06T11:39:24.181707Z  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
2026-09-06T11:39:24.242263Z  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
2026-09-06T11:39:24.270677Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: evaluated exchange placement finalist offset=16384 baseline_score=448685899776 score=421462573056 baseline_cycles=74452 candidate_cycles=77181 row_bytes=1352 row_capacity=16320 accepted=false
2026-09-06T11:39:24.280691Z  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=8292 success=true
2026-09-06T11:39:24.586221Z  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=8 success=true
2026-09-06T11:39:24.595135Z  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=8 success=true
2026-09-06T11:39:24.680522Z  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=82 success=true
2026-09-06T11:39:24.742448Z  INFO writing application package tiles=1472
2026-09-06T11:39:24.792793Z  INFO application package written
2026-09-06T11:39:24.794411Z  INFO reading application package
2026-09-06T11:39:24.875158Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-06T11:39:24.877238Z  INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:39:24.877250Z  INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:39:24.877255Z  INFO opening IPU device path="/dev/ipu0"
2026-09-06T11:39:24.877287Z  INFO IPU device attached path=/dev/ipu0
2026-09-06T11:39:24.877289Z  INFO initializing IPU device
2026-09-06T11:39:24.885455Z  INFO IPU device initialized
2026-09-06T11:39:24.885459Z  INFO replaying device configuration register_writes=2003
2026-09-06T11:39:24.893899Z  INFO loading application tile_count=1472 final_mark=23
2026-09-06T11:39:25.160563Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-06T11:39:27.796247Z  INFO starting host exchange session startup_mark=23
2026-09-06T11:39:27.796282Z  INFO attaching host exchange pages pages=2
2026-09-06T11:39:27.796712Z  INFO host exchange session started
2026-09-06T11:39:27.796765Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=14
2026-09-06T11:39:27.804606Z  INFO invoking streaming host exchange call call="run" command=0 phases=59
profile=artifacts/layout-sweep/up-20/execution.ipuprofile tiles=1472 samples=20826
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=279258 minimumTileCycles=262914 deviceMicroseconds=186.172 effectiveGemmTflops=77.660 peakTflops=282.624 efficiencyPercent=27.48 maximumAbsoluteError=0.011719
package=artifacts/layout-sweep/up-20/model.ipuexe tiles=1472 hardwareTest=PASS