2026-09-06T11:43:29.068106Z  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 }
2026-09-06T11:43:29.497579Z  INFO applied GEMM plan constraint source_operation=2 matching_plans=2 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 9, column_partitions: 18, inner_partitions: 9, result_row_partitions: 9, result_column_partitions: 1, output_column_block: 64, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct }
2026-09-06T11:43:29.818888Z  INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=316399 estimated_exchange_cycles=122680 exchange_row_bytes=26948 peak_standard_bytes=141636 peak_interleaved_bytes=122880 peak_total_bytes=263492 maximum_standard_allocation_bytes=114688 active_tile_counts={1458, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))]
2026-09-06T11:43:29.822168Z  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=1042 success=true
2026-09-06T11:43:33.073808Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=256217 estimated_exchange_cycles=70184
2026-09-06T11:43:33.099130Z  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=144
2026-09-06T11:43:33.100110Z  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=3277 success=true
2026-09-06T11:43:33.100125Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26210 exchange_phases=4
2026-09-06T11:43:33.100281Z  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
2026-09-06T11:43:33.100655Z  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
2026-09-06T11:43:33.100861Z  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_k480_c64_r80_r82" target=ipu21
2026-09-06T11:43:33.152757Z  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/6cb82f292b3d18645c8127708cc25b3b6036b333a5d932875b0ab1ce67901b92/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/6cb82f292b3d18645c8127708cc25b3b6036b333a5d932875b0ab1ce67901b92/kernel.json
2026-09-06T11:43:33.153085Z  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
2026-09-06T11:43:33.157911Z  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
2026-09-06T11:43:33.267904Z  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=109 success=true
2026-09-06T11:43:33.604513Z  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
2026-09-06T11:43:33.604547Z  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
2026-09-06T11:43:33.614980Z  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
2026-09-06T11:43:41.910392Z  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=4932 ordinary_horizon=43710 paired_horizon=42209 use_paired=true
2026-09-06T11:43:42.000586Z  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=7104 maximum_identical_destinations=1
2026-09-06T11:43:42.000612Z  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=42209 selected_horizon=42209 endpoint_lower_bound=31920 lower_bound_gap=10289 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:43:42.043612Z  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=246 row_words=305 horizon=42209
2026-09-06T11:43:46.041611Z  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=50400 maximum_identical_destinations=1
2026-09-06T11:43:46.041648Z  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=5028 selected_horizon=5028 endpoint_lower_bound=2880 lower_bound_gap=2148 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:43:46.087309Z  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=580 row_words=221 horizon=5028
2026-09-06T11:43:46.102582Z  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=12834 success=true
2026-09-06T11:43:46.709928Z  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
2026-09-06T11:43:46.709964Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319176 profile_bytes=180 exchange_table_bytes=2080 host_code_bytes=2908 generated_code_bytes=1820 code_address=322088 standard_ranges=[(409600, 524288)]
2026-09-06T11:43:46.812761Z  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=102 success=true
2026-09-06T11:43:46.820715Z  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
2026-09-06T11:43:46.911990Z  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
2026-09-06T11:43:46.917902Z  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
2026-09-06T11:43:46.917915Z  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
2026-09-06T11:43:46.928089Z  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
2026-09-06T11:43:55.794306Z  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=4932 ordinary_horizon=43710 paired_horizon=42209 use_paired=true
2026-09-06T11:43:55.878933Z  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=7104 maximum_identical_destinations=1
2026-09-06T11:43:55.878957Z  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=42209 selected_horizon=42209 endpoint_lower_bound=31920 lower_bound_gap=10289 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:43:55.925616Z  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=246 row_words=305 horizon=42209
2026-09-06T11:44:00.267880Z  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=50400 maximum_identical_destinations=1
2026-09-06T11:44:00.267932Z  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=5028 selected_horizon=5028 endpoint_lower_bound=2880 lower_bound_gap=2148 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:44:00.328647Z  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=580 row_words=221 horizon=5028
2026-09-06T11:44:00.356040Z  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=13543 success=true
2026-09-06T11:44:01.543059Z  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
2026-09-06T11:44:01.633605Z  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
2026-09-06T11:44:01.639774Z  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
2026-09-06T11:44:01.639787Z  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
2026-09-06T11:44:01.650152Z  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
2026-09-06T11:44:10.833747Z  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=4932 ordinary_horizon=43710 paired_horizon=42209 use_paired=true
2026-09-06T11:44:10.944088Z  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=7104 maximum_identical_destinations=1
2026-09-06T11:44:10.944128Z  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=42209 selected_horizon=42209 endpoint_lower_bound=31920 lower_bound_gap=10289 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:44:10.990953Z  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=246 row_words=306 horizon=42209
2026-09-06T11:44:15.321261Z  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=50400 maximum_identical_destinations=1
2026-09-06T11:44:15.321297Z  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=5028 selected_horizon=5028 endpoint_lower_bound=2880 lower_bound_gap=2148 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:44:15.368394Z  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=580 row_words=221 horizon=5028
2026-09-06T11:44:15.393361Z  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=394151854080 score=382591303680 baseline_cycles=58064 candidate_cycles=58064 row_bytes=2080 row_capacity=16320 accepted=false
2026-09-06T11:44:15.404050Z  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=15047 success=true
2026-09-06T11:44:15.712860Z  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=10 success=true
2026-09-06T11:44:15.724127Z  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
2026-09-06T11:44:15.829947Z  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=101 success=true
2026-09-06T11:44:15.887687Z  INFO writing application package tiles=1472
2026-09-06T11:44:15.945677Z  INFO application package written
2026-09-06T11:44:15.948230Z  INFO reading application package
2026-09-06T11:44:16.044148Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-06T11:44:16.046080Z  INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:44:16.046093Z  INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock
2026-09-06T11:44:16.046098Z  INFO opening IPU device path="/dev/ipu0"
2026-09-06T11:44:16.046128Z  INFO IPU device attached path=/dev/ipu0
2026-09-06T11:44:16.046130Z  INFO initializing IPU device
2026-09-06T11:44:16.053943Z  INFO IPU device initialized
2026-09-06T11:44:16.053972Z  INFO replaying device configuration register_writes=2003
2026-09-06T11:44:16.062715Z  INFO loading application tile_count=1472 final_mark=23
2026-09-06T11:44:16.346789Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-06T11:44:18.961005Z  INFO starting host exchange session startup_mark=23
2026-09-06T11:44:18.961039Z  INFO attaching host exchange pages pages=2
2026-09-06T11:44:18.961741Z  INFO host exchange session started
2026-09-06T11:44:18.961795Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=30
2026-09-06T11:44:18.970378Z  INFO invoking streaming host exchange call call="run" command=0 phases=59
profile=artifacts/layout-sweep/down-16/execution.ipuprofile tiles=1472 samples=24044
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=1458 executionTiles=1472 inputBytes=154736640 weightBytes=19869696 cycles=212880 minimumTileCycles=199944 deviceMicroseconds=141.920 effectiveGemmTflops=101.875 peakTflops=282.624 efficiencyPercent=36.05 maximumAbsoluteError=0.009766
package=artifacts/layout-sweep/down-16/model.ipuexe tiles=1472 hardwareTest=PASS