2026-09-06T11:40:12.399343Z  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 }
2026-09-06T11:40:12.844163Z  INFO applied GEMM plan constraint source_operation=2 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 10, column_partitions: 24, inner_partitions: 6, result_row_partitions: 6, result_column_partitions: 1, output_column_block: 48, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct }
2026-09-06T11:40:13.095068Z  INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=341668 estimated_exchange_cycles=142840 exchange_row_bytes=31484 peak_standard_bytes=146172 peak_interleaved_bytes=138240 peak_total_bytes=284412 maximum_standard_allocation_bytes=114688 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))]
2026-09-06T11:40:13.098480Z  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=987 success=true
2026-09-06T11:40:19.220732Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=256877 estimated_exchange_cycles=72024
2026-09-06T11:40:19.266993Z  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=216
2026-09-06T11:40:19.268649Z  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=6170 success=true
2026-09-06T11:40:19.268665Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26120 exchange_phases=4
2026-09-06T11:40:19.268850Z  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
2026-09-06T11:40:19.269406Z  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
2026-09-06T11:40:19.269678Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="compile_kernels"}: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/gemm_f16_amp.S name="gemm_f16_interleaved_k720_c48_r72_r74" target=ipu21
2026-09-06T11:40:19.321979Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="compile_kernels"}: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/6630acafd188c60706d9d6bd7c43b545567809a2e21419e172d0b2878a7eb73a/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/6630acafd188c60706d9d6bd7c43b545567809a2e21419e172d0b2878a7eb73a/kernel.json
2026-09-06T11:40:19.322354Z  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=52 success=true
2026-09-06T11:40:19.325329Z  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
2026-09-06T11:40:19.438143Z  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=112 success=true
2026-09-06T11:40:19.778295Z  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
2026-09-06T11:40:19.778322Z  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
2026-09-06T11:40:19.789058Z  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
2026-09-06T11:40:32.487913Z  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=4695 ordinary_horizon=56533 paired_horizon=47490 use_paired=true
2026-09-06T11:40:32.605476Z  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=7445 maximum_identical_destinations=1
2026-09-06T11:40:32.605499Z  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=47490 selected_horizon=47490 endpoint_lower_bound=40680 lower_bound_gap=6810 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:40:32.666269Z  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=247 row_words=452 horizon=47490
2026-09-06T11:40:33.741634Z  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=23328 maximum_identical_destinations=1
2026-09-06T11:40:33.741668Z  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=2756 selected_horizon=2756 endpoint_lower_bound=1680 lower_bound_gap=1076 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:40:33.760577Z  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=275 row_words=101 horizon=2756
2026-09-06T11:40:33.768030Z  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=14329 success=true
2026-09-06T11:40:34.374919Z  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=8 success=true
2026-09-06T11:40:34.374954Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319152 profile_bytes=240 exchange_table_bytes=2244 host_code_bytes=3036 generated_code_bytes=2160 code_address=322192 standard_ranges=[(409600, 524288)]
2026-09-06T11:40:34.474936Z  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
2026-09-06T11:40:34.483276Z  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
2026-09-06T11:40:34.574145Z  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
2026-09-06T11:40:34.580153Z  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
2026-09-06T11:40:34.580166Z  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
2026-09-06T11:40:34.590493Z  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
2026-09-06T11:40:49.562564Z  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=4695 ordinary_horizon=56533 paired_horizon=47490 use_paired=true
2026-09-06T11:40:49.683586Z  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=7445 maximum_identical_destinations=1
2026-09-06T11:40:49.683611Z  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=47490 selected_horizon=47490 endpoint_lower_bound=40680 lower_bound_gap=6810 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:40:49.744842Z  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=247 row_words=452 horizon=47490
2026-09-06T11:40:50.926301Z  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=23328 maximum_identical_destinations=1
2026-09-06T11:40:50.926332Z  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=2756 selected_horizon=2756 endpoint_lower_bound=1680 lower_bound_gap=1076 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:40:50.945116Z  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=275 row_words=101 horizon=2756
2026-09-06T11:40:50.952736Z  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=16477 success=true
2026-09-06T11:40:52.202683Z  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
2026-09-06T11:40:52.294182Z  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
2026-09-06T11:40:52.300295Z  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
2026-09-06T11:40:52.300307Z  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
2026-09-06T11:40:52.310518Z  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
2026-09-06T11:41:06.958995Z  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=4695 ordinary_horizon=48074 paired_horizon=43689 use_paired=true
2026-09-06T11:41:07.079193Z  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=7445 maximum_identical_destinations=1
2026-09-06T11:41:07.079239Z  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=43689 selected_horizon=43689 endpoint_lower_bound=40680 lower_bound_gap=3009 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:41:07.140185Z  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=247 row_words=450 horizon=43689
2026-09-06T11:41:08.388922Z  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=23328 maximum_identical_destinations=1
2026-09-06T11:41:08.388957Z  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=2756 selected_horizon=2756 endpoint_lower_bound=1680 lower_bound_gap=1076 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:41:08.407987Z  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=275 row_words=101 horizon=2756
2026-09-06T11:41:08.423130Z  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=1113250037760 score=248796610560 baseline_cycles=61073 candidate_cycles=57272 row_bytes=2236 row_capacity=16320 accepted=true
2026-09-06T11:41:08.435645Z  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=17482 success=true
2026-09-06T11:41:08.752232Z  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=16 success=true
2026-09-06T11:41:08.761313Z  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=9 success=true
2026-09-06T11:41:08.829164Z  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=64 success=true
2026-09-06T11:41:08.875840Z  INFO writing application package tiles=1472
2026-09-06T11:41:08.933382Z  INFO application package written
2026-09-06T11:41:08.934298Z  INFO reading application package
2026-09-06T11:41:08.995675Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-06T11:41:08.996535Z  INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:41:08.996543Z  INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:41:08.996546Z  INFO opening IPU device path="/dev/ipu0"
2026-09-06T11:41:08.996566Z  INFO IPU device attached path=/dev/ipu0
2026-09-06T11:41:08.996568Z  INFO initializing IPU device
2026-09-06T11:41:09.004846Z  INFO IPU device initialized
2026-09-06T11:41:09.004851Z  INFO replaying device configuration register_writes=2003
2026-09-06T11:41:09.013972Z  INFO loading application tile_count=1472 final_mark=23
2026-09-06T11:41:09.306526Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-06T11:41:11.411029Z  INFO starting host exchange session startup_mark=23
2026-09-06T11:41:11.411059Z  INFO attaching host exchange pages pages=2
2026-09-06T11:41:11.411461Z  INFO host exchange session started
2026-09-06T11:41:11.411539Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=34
2026-09-06T11:41:11.420943Z  INFO invoking streaming host exchange call call="run" command=0 phases=59
profile=artifacts/layout-sweep/down-03/execution.ipuprofile tiles=1472 samples=25954
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=215862 minimumTileCycles=207324 deviceMicroseconds=143.908 effectiveGemmTflops=100.468 peakTflops=282.624 efficiencyPercent=35.55 maximumAbsoluteError=0.011719
package=artifacts/layout-sweep/down-03/model.ipuexe tiles=1472 hardwareTest=PASS