2026-09-06T11:45:21.261891Z  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 }
2026-09-06T11:45:21.662398Z  INFO applied GEMM plan constraint source_operation=2 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 8, column_partitions: 36, inner_partitions: 5, result_row_partitions: 5, result_column_partitions: 1, output_column_block: 32, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct }
2026-09-06T11:45:21.945846Z  INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=396621 estimated_exchange_cycles=183320 exchange_row_bytes=40592 peak_standard_bytes=204432 peak_interleaved_bytes=165888 peak_total_bytes=315024 maximum_standard_allocation_bytes=163840 active_tile_counts={1440} padding_group_counts={} conversion_sources=[Some(OperationId(2))]
2026-09-06T11:45:21.948386Z  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=976 success=true
2026-09-06T11:45:29.138717Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=289978 estimated_exchange_cycles=93896
2026-09-06T11:45:29.240789Z  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=252
2026-09-06T11:45:29.242842Z  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=7294 success=true
2026-09-06T11:45:29.242874Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26217 exchange_phases=4
2026-09-06T11:45:29.243134Z  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
2026-09-06T11:45:29.243981Z  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
2026-09-06T11:45:29.244359Z  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
2026-09-06T11:45:29.247441Z  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
2026-09-06T11:45:29.372569Z  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=125 success=true
2026-09-06T11:45:31.631143Z  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
2026-09-06T11:45:31.631190Z  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
2026-09-06T11:45:31.661370Z  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
2026-09-06T11:46:29.796480Z  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=6456 ordinary_horizon=69740 paired_horizon=60798 use_paired=true
2026-09-06T11:46:30.055811Z  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=11211 maximum_identical_destinations=1
2026-09-06T11:46:30.055839Z  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=60798 selected_horizon=60798 endpoint_lower_bound=44064 lower_bound_gap=16734 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:46:30.191233Z  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=688 row_words=796 horizon=60798
2026-09-06T11:46:30.601794Z  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=12960 maximum_identical_destinations=1
2026-09-06T11:46:30.601818Z  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=1771 selected_horizon=1732 endpoint_lower_bound=1408 lower_bound_gap=324 selected_kind="critical-neighborhood" neighborhood_improvements=1
2026-09-06T11:46:30.609016Z  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=507 row_words=62 horizon=1732
2026-09-06T11:46:30.612158Z  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=61239 success=true
2026-09-06T11:46:30.859525Z  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
2026-09-06T11:46:30.859549Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319104 profile_bytes=276 exchange_table_bytes=4024 host_code_bytes=2012 generated_code_bytes=2268 code_address=321120 standard_ranges=[(409600, 524288)]
2026-09-06T11:46:30.954799Z  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=95 success=true
2026-09-06T11:46:30.970057Z  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
2026-09-06T11:46:33.159050Z  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
2026-09-06T11:46:33.159086Z  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
2026-09-06T11:46:33.191732Z  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
2026-09-06T11:47:24.517411Z  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=6456 ordinary_horizon=69740 paired_horizon=60798 use_paired=true
2026-09-06T11:47:24.781825Z  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=11211 maximum_identical_destinations=1
2026-09-06T11:47:24.781856Z  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=60798 selected_horizon=60798 endpoint_lower_bound=44064 lower_bound_gap=16734 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:47:24.916884Z  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=688 row_words=796 horizon=60798
2026-09-06T11:47:25.059989Z  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
2026-09-06T11:47:25.067130Z  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=12960 maximum_identical_destinations=1
2026-09-06T11:47:25.067152Z  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=1732 selected_horizon=1732 endpoint_lower_bound=1408 lower_bound_gap=324 selected_kind="reused" neighborhood_improvements=0
2026-09-06T11:47:25.074624Z  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=507 row_words=62 horizon=1732
2026-09-06T11:47:25.077140Z  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=54122 success=true
2026-09-06T11:47:26.667911Z  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
2026-09-06T11:47:28.765004Z  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
2026-09-06T11:47:28.765035Z  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
2026-09-06T11:47:28.795024Z  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
2026-09-06T11:48:20.411461Z  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=5757 ordinary_horizon=61456 paired_horizon=54415 use_paired=true
2026-09-06T11:48:20.582271Z  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=11211 maximum_identical_destinations=1
2026-09-06T11:48:20.582296Z  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=54415 selected_horizon=54415 endpoint_lower_bound=44064 lower_bound_gap=10351 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:48:20.668629Z  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=688 row_words=795 horizon=54415
2026-09-06T11:48:20.799356Z  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
2026-09-06T11:48:20.805144Z  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=12960 maximum_identical_destinations=1
2026-09-06T11:48:20.805158Z  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=1732 selected_horizon=1732 endpoint_lower_bound=1408 lower_bound_gap=324 selected_kind="reused" neighborhood_improvements=0
2026-09-06T11:48:20.818026Z  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=507 row_words=62 horizon=1732
2026-09-06T11:48:20.831621Z  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=301300383744 score=236601409536 baseline_cycles=81211 candidate_cycles=74828 row_bytes=4024 row_capacity=16320 accepted=true
2026-09-06T11:48:20.843270Z  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=55766 success=true
2026-09-06T11:48:21.049142Z  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
2026-09-06T11:48:21.061413Z  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=12 success=true
2026-09-06T11:48:21.167005Z  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=100 success=true
2026-09-06T11:48:21.226440Z  INFO writing application package tiles=1472
2026-09-06T11:48:21.281853Z  INFO application package written
2026-09-06T11:48:21.284253Z  INFO reading application package
2026-09-06T11:48:21.390594Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-06T11:48:21.393239Z  INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:48:21.393253Z  INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:48:21.393258Z  INFO opening IPU device path="/dev/ipu0"
2026-09-06T11:48:21.393287Z  INFO IPU device attached path=/dev/ipu0
2026-09-06T11:48:21.393290Z  INFO initializing IPU device
2026-09-06T11:48:21.401322Z  INFO IPU device initialized
2026-09-06T11:48:21.401341Z  INFO replaying device configuration register_writes=2003
2026-09-06T11:48:21.408853Z  INFO loading application tile_count=1472 final_mark=23
2026-09-06T11:48:21.625488Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-06T11:48:23.446577Z  INFO starting host exchange session startup_mark=23
2026-09-06T11:48:23.446608Z  INFO attaching host exchange pages pages=2
2026-09-06T11:48:23.447373Z  INFO host exchange session started
2026-09-06T11:48:23.447443Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=28
2026-09-06T11:48:23.455894Z  INFO invoking streaming host exchange call call="run" command=0 phases=33
profile=artifacts/layout-sweep/joint-rows-8/execution.ipuprofile tiles=1472 samples=29234
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=248928 minimumTileCycles=231924 deviceMicroseconds=165.952 effectiveGemmTflops=87.122 peakTflops=282.624 efficiencyPercent=30.83 maximumAbsoluteError=0.011719
package=artifacts/layout-sweep/joint-rows-8/model.ipuexe tiles=1472 hardwareTest=PASS