2026-09-06T11:43:29.099666Z 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 } 2026-09-06T11:43:29.538890Z INFO applied GEMM plan constraint source_operation=2 matching_plans=2 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 12, column_partitions: 12, inner_partitions: 10, result_row_partitions: 10, result_column_partitions: 1, output_column_block: 96, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:43:29.886453Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=330469 estimated_exchange_cycles=122680 exchange_row_bytes=26948 peak_standard_bytes=141636 peak_interleaved_bytes=165888 peak_total_bytes=284996 maximum_standard_allocation_bytes=114688 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:43:29.889578Z 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=1096 success=true 2026-09-06T11:43:36.319443Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=287009 estimated_exchange_cycles=87144 2026-09-06T11:43:36.349357Z 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=132 2026-09-06T11:43:36.350846Z 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=6461 success=true 2026-09-06T11:43:36.350865Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=25916 exchange_phases=4 2026-09-06T11:43:36.351061Z 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 2026-09-06T11:43:36.351622Z 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 2026-09-06T11:43:36.351950Z 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_k432_c96_r60_r62" target=ipu21 2026-09-06T11:43:36.407440Z 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/80469e264162c2de94b7dc5c4f761756ab7dccee2b9c86cfb1d8510c1fa56a56/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/80469e264162c2de94b7dc5c4f761756ab7dccee2b9c86cfb1d8510c1fa56a56/kernel.json 2026-09-06T11:43:36.407762Z 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=56 success=true 2026-09-06T11:43:36.411902Z 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 2026-09-06T11:43:36.524053Z 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 2026-09-06T11:43:36.859415Z 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 2026-09-06T11:43:36.859441Z 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 2026-09-06T11:43:36.870295Z 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 2026-09-06T11:43:42.696355Z 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=5254 ordinary_horizon=42732 paired_horizon=38219 use_paired=true 2026-09-06T11:43:42.765147Z 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=7863 maximum_identical_destinations=1 2026-09-06T11:43:42.765172Z 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=38219 selected_horizon=38219 endpoint_lower_bound=33696 lower_bound_gap=4523 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:43:42.801504Z 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=1149 row_words=251 horizon=38219 2026-09-06T11:43:51.092038Z 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=80064 maximum_identical_destinations=1 2026-09-06T11:43:51.092073Z 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=5575 selected_horizon=5575 endpoint_lower_bound=3456 lower_bound_gap=2119 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:43:51.166216Z 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=105 row_words=326 horizon=5575 2026-09-06T11:43:51.190736Z 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=14666 success=true 2026-09-06T11:43:51.799692Z 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 2026-09-06T11:43:51.799727Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319152 profile_bytes=164 exchange_table_bytes=2172 host_code_bytes=3276 generated_code_bytes=1720 code_address=322432 standard_ranges=[(409600, 524288)] 2026-09-06T11:43:51.900493Z 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=100 success=true 2026-09-06T11:43:51.908551Z 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 2026-09-06T11:43:51.999029Z 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 2026-09-06T11:43:52.006548Z 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 2026-09-06T11:43:52.006560Z 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 2026-09-06T11:43:52.016642Z 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 2026-09-06T11:43:58.138699Z 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=5254 ordinary_horizon=42732 paired_horizon=38219 use_paired=true 2026-09-06T11:43:58.207997Z 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=7863 maximum_identical_destinations=1 2026-09-06T11:43:58.208028Z 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=38219 selected_horizon=38219 endpoint_lower_bound=33696 lower_bound_gap=4523 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:43:58.246010Z 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=1149 row_words=251 horizon=38219 2026-09-06T11:44:07.262111Z 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=80064 maximum_identical_destinations=1 2026-09-06T11:44:07.262154Z 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=5575 selected_horizon=5575 endpoint_lower_bound=3456 lower_bound_gap=2119 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:44:07.336716Z 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=105 row_words=326 horizon=5575 2026-09-06T11:44:07.368527Z 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=15468 success=true 2026-09-06T11:44:08.578581Z 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 2026-09-06T11:44:08.671069Z 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 2026-09-06T11:44:08.677955Z 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 2026-09-06T11:44:08.677979Z 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 2026-09-06T11:44:08.692579Z 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 2026-09-06T11:44:14.998960Z 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=5254 ordinary_horizon=38612 paired_horizon=35894 use_paired=true 2026-09-06T11:44:15.076552Z 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=7863 maximum_identical_destinations=1 2026-09-06T11:44:15.076586Z 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=35894 selected_horizon=35894 endpoint_lower_bound=33696 lower_bound_gap=2198 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:44:15.114648Z 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=1149 row_words=253 horizon=35894 2026-09-06T11:44:24.146828Z 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=80064 maximum_identical_destinations=1 2026-09-06T11:44:24.146855Z 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=5575 selected_horizon=5575 endpoint_lower_bound=3456 lower_bound_gap=2119 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:44:24.194326Z 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=105 row_words=326 horizon=5575 2026-09-06T11:44:24.219937Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: evaluated exchange placement finalist offset=12288 baseline_score=1335900045312 score=530448777216 baseline_cycles=54621 candidate_cycles=52296 row_bytes=2172 row_capacity=16320 accepted=true 2026-09-06T11:44:24.236614Z 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=16868 success=true 2026-09-06T11:44:24.444291Z 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 2026-09-06T11:44:24.453867Z 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 2026-09-06T11:44:24.561484Z 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=103 success=true 2026-09-06T11:44:24.625085Z INFO writing application package tiles=1472 2026-09-06T11:44:24.675870Z INFO application package written 2026-09-06T11:44:24.677671Z INFO reading application package 2026-09-06T11:44:24.773116Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:44:24.775142Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:44:24.775155Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:44:24.775160Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:44:24.775195Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:44:24.775198Z INFO initializing IPU device 2026-09-06T11:44:24.783413Z INFO IPU device initialized 2026-09-06T11:44:24.783420Z INFO replaying device configuration register_writes=2003 2026-09-06T11:44:24.792660Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:44:25.089536Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:44:27.759081Z INFO starting host exchange session startup_mark=23 2026-09-06T11:44:27.759112Z INFO attaching host exchange pages pages=2 2026-09-06T11:44:27.759681Z INFO host exchange session started 2026-09-06T11:44:27.759721Z INFO invoking streaming host exchange call call="initialize" command=0 phases=42 2026-09-06T11:44:27.769568Z INFO invoking streaming host exchange call call="run" command=0 phases=59 profile=artifacts/layout-sweep/down-17/execution.ipuprofile tiles=1472 samples=22566 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=213636 minimumTileCycles=200196 deviceMicroseconds=142.424 effectiveGemmTflops=101.515 peakTflops=282.624 efficiencyPercent=35.92 maximumAbsoluteError=0.009766 package=artifacts/layout-sweep/down-17/model.ipuexe tiles=1472 hardwareTest=PASS