2026-09-06T11:40:08.110798Z 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 } 2026-09-06T11:40:08.549241Z INFO applied GEMM plan constraint source_operation=2 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 10, column_partitions: 36, inner_partitions: 4, result_row_partitions: 4, result_column_partitions: 1, output_column_block: 32, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:40:08.788196Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=378447 estimated_exchange_cycles=175160 exchange_row_bytes=38756 peak_standard_bytes=202596 peak_interleaved_bytes=139264 peak_total_bytes=341860 maximum_standard_allocation_bytes=163840 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:40:08.791764Z 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=967 success=true 2026-09-06T11:40:15.271059Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=266049 estimated_exchange_cycles=82592 2026-09-06T11:40:15.334586Z 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=324 2026-09-06T11:40:15.336486Z 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=6544 success=true 2026-09-06T11:40:15.336503Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26264 exchange_phases=4 2026-09-06T11:40:15.336664Z 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 2026-09-06T11:40:15.337357Z 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 2026-09-06T11:40:15.337628Z 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_k1088_c32_r72_r74" target=ipu21 2026-09-06T11:40:15.385184Z 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/d516aa725165230dea453fb17292e26c43f5c9bacb8d834f1b416e39c03b8cf8/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/d516aa725165230dea453fb17292e26c43f5c9bacb8d834f1b416e39c03b8cf8/kernel.json 2026-09-06T11:40:15.385544Z 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=48 success=true 2026-09-06T11:40:15.388508Z 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 2026-09-06T11:40:15.510145Z 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=121 success=true 2026-09-06T11:40:15.918655Z 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 2026-09-06T11:40:15.918697Z 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 2026-09-06T11:40:15.936214Z 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 2026-09-06T11:40:41.277873Z 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=4348 ordinary_horizon=70209 paired_horizon=60630 use_paired=true 2026-09-06T11:40:41.439018Z 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 2026-09-06T11:40:41.439041Z 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=60630 selected_horizon=60630 endpoint_lower_bound=52768 lower_bound_gap=7862 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:40:41.524019Z 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=580 horizon=60630 2026-09-06T11:40:41.712220Z 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=9792 maximum_identical_destinations=1 2026-09-06T11:40:41.712244Z 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=1355 selected_horizon=1342 endpoint_lower_bound=1056 lower_bound_gap=286 selected_kind="critical-neighborhood" neighborhood_improvements=1 2026-09-06T11:40:41.716906Z 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=556 row_words=47 horizon=1342 2026-09-06T11:40:41.718978Z 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=26208 success=true 2026-09-06T11:40:42.091012Z 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 2026-09-06T11:40:42.091037Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319152 profile_bytes=332 exchange_table_bytes=2628 host_code_bytes=3036 generated_code_bytes=2692 code_address=322192 standard_ranges=[(409600, 524288)] 2026-09-06T11:40:42.185847Z 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=94 success=true 2026-09-06T11:40:42.193934Z 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 2026-09-06T11:40:42.282514Z 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 2026-09-06T11:40:42.288667Z 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 2026-09-06T11:40:42.288684Z 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 2026-09-06T11:40:42.300422Z 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 2026-09-06T11:41:08.711434Z 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=4348 ordinary_horizon=70209 paired_horizon=60630 use_paired=true 2026-09-06T11:41:08.820190Z 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 2026-09-06T11:41:08.820231Z 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=60630 selected_horizon=60630 endpoint_lower_bound=52768 lower_bound_gap=7862 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:41:08.872902Z 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=580 horizon=60630 2026-09-06T11:41:08.937800Z 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 2026-09-06T11:41:08.940894Z 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=9792 maximum_identical_destinations=1 2026-09-06T11:41:08.940904Z 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=1342 selected_horizon=1342 endpoint_lower_bound=1056 lower_bound_gap=286 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:41:08.947506Z 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=556 row_words=47 horizon=1342 2026-09-06T11:41:08.950224Z 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=26764 success=true 2026-09-06T11:41:10.119996Z 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 2026-09-06T11:41:10.213502Z 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 2026-09-06T11:41:10.219704Z 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 2026-09-06T11:41:10.219716Z 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 2026-09-06T11:41:10.230261Z 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 2026-09-06T11:41:37.443129Z 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=4348 ordinary_horizon=66383 paired_horizon=58894 use_paired=true 2026-09-06T11:41:37.559616Z 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 2026-09-06T11:41:37.559641Z 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=58894 selected_horizon=58894 endpoint_lower_bound=52768 lower_bound_gap=6126 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:41:37.611767Z 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=580 horizon=58894 2026-09-06T11:41:37.680014Z 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 2026-09-06T11:41:37.683186Z 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=9792 maximum_identical_destinations=1 2026-09-06T11:41:37.683198Z 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=1342 selected_horizon=1342 endpoint_lower_bound=1056 lower_bound_gap=286 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:41:37.688229Z 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=556 row_words=47 horizon=1342 2026-09-06T11:41:37.695289Z 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=1126630162432 score=249115443200 baseline_cycles=72799 candidate_cycles=71063 row_bytes=2628 row_capacity=16320 accepted=true 2026-09-06T11:41:37.702924Z 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=28752 success=true 2026-09-06T11:41:37.982919Z 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=34 success=true 2026-09-06T11:41:37.992715Z 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 2026-09-06T11:41:38.064420Z 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=67 success=true 2026-09-06T11:41:38.118063Z INFO writing application package tiles=1472 2026-09-06T11:41:38.189959Z INFO application package written 2026-09-06T11:41:38.191830Z INFO reading application package 2026-09-06T11:41:38.269866Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:41:38.271623Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:41:38.271637Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:41:38.271645Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:41:38.271670Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:41:38.271672Z INFO initializing IPU device 2026-09-06T11:41:38.279456Z INFO IPU device initialized 2026-09-06T11:41:38.279459Z INFO replaying device configuration register_writes=2003 2026-09-06T11:41:38.286400Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:41:38.566404Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:41:40.204445Z INFO starting host exchange session startup_mark=23 2026-09-06T11:41:40.204476Z INFO attaching host exchange pages pages=2 2026-09-06T11:41:40.204931Z INFO host exchange session started 2026-09-06T11:41:40.204981Z INFO invoking streaming host exchange call call="initialize" command=0 phases=34 2026-09-06T11:41:40.214406Z INFO invoking streaming host exchange call call="run" command=0 phases=59 profile=artifacts/layout-sweep/down-02/execution.ipuprofile tiles=1472 samples=29380 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=19943424 cycles=235098 minimumTileCycles=226944 deviceMicroseconds=156.732 effectiveGemmTflops=92.247 peakTflops=282.624 efficiencyPercent=32.64 maximumAbsoluteError=0.017578 package=artifacts/layout-sweep/down-02/model.ipuexe tiles=1472 hardwareTest=PASS