2026-09-06T11:42:33.949651Z 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 } 2026-09-06T11:42:34.390321Z INFO applied GEMM plan constraint source_operation=2 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 6, column_partitions: 24, inner_partitions: 10, result_row_partitions: 10, result_column_partitions: 1, output_column_block: 48, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:42:34.776875Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=315602 estimated_exchange_cycles=128120 exchange_row_bytes=28172 peak_standard_bytes=142860 peak_interleaved_bytes=101888 peak_total_bytes=244748 maximum_standard_allocation_bytes=114688 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:42:34.782923Z 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=1123 success=true 2026-09-06T11:42:39.134617Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=245771 estimated_exchange_cycles=68664 2026-09-06T11:42:39.196670Z 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=120 2026-09-06T11:42:39.198346Z 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=4415 success=true 2026-09-06T11:42:39.198367Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26216 exchange_phases=4 2026-09-06T11:42:39.198571Z 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 2026-09-06T11:42:39.199129Z 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 2026-09-06T11:42:39.199427Z 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_c48_r120_r122" target=ipu21 2026-09-06T11:42:39.252313Z 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/f35c7162059138bcc6df51a9c1fce43c4f76c0f0aa5e3883ebf1810b73fa159b/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/f35c7162059138bcc6df51a9c1fce43c4f76c0f0aa5e3883ebf1810b73fa159b/kernel.json 2026-09-06T11:42:39.252641Z 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=53 success=true 2026-09-06T11:42:39.256839Z 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 2026-09-06T11:42:39.427985Z 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=171 success=true 2026-09-06T11:42:39.823197Z 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 2026-09-06T11:42:39.823248Z 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 2026-09-06T11:42:39.833888Z 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 2026-09-06T11:42:49.830917Z 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=4427 ordinary_horizon=55281 paired_horizon=42100 use_paired=true 2026-09-06T11:42:49.943623Z 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=6369 maximum_identical_destinations=1 2026-09-06T11:42:49.943662Z 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=42100 selected_horizon=42100 endpoint_lower_bound=30024 lower_bound_gap=12076 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:42:50.003812Z 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=1151 row_words=330 horizon=42100 2026-09-06T11:42:52.850105Z 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=41760 maximum_identical_destinations=1 2026-09-06T11:42:52.850141Z 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=5019 selected_horizon=4999 endpoint_lower_bound=3024 lower_bound_gap=1975 selected_kind="matching-waves" neighborhood_improvements=0 2026-09-06T11:42:52.886082Z 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=223 row_words=178 horizon=4999 2026-09-06T11:42:52.899049Z 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=13471 success=true 2026-09-06T11:42:53.513566Z 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 2026-09-06T11:42:53.513605Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319152 profile_bytes=168 exchange_table_bytes=2084 host_code_bytes=2636 generated_code_bytes=1728 code_address=321792 standard_ranges=[(409600, 524288)] 2026-09-06T11:42:53.620129Z 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=106 success=true 2026-09-06T11:42:53.627882Z 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 2026-09-06T11:42:53.717176Z 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 2026-09-06T11:42:53.723236Z 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 2026-09-06T11:42:53.723248Z 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 2026-09-06T11:42:53.733261Z 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 2026-09-06T11:43:04.292826Z 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=4427 ordinary_horizon=55281 paired_horizon=42100 use_paired=true 2026-09-06T11:43:04.385099Z 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=6369 maximum_identical_destinations=1 2026-09-06T11:43:04.385124Z 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=42100 selected_horizon=42100 endpoint_lower_bound=30024 lower_bound_gap=12076 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:43:04.432425Z 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=1151 row_words=330 horizon=42100 2026-09-06T11:43:04.738563Z 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 2026-09-06T11:43:04.756459Z 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=41760 maximum_identical_destinations=1 2026-09-06T11:43:04.756473Z 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=4999 selected_horizon=4999 endpoint_lower_bound=3024 lower_bound_gap=1975 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:43:04.791278Z 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=223 row_words=178 horizon=4999 2026-09-06T11:43:04.802745Z 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=11182 success=true 2026-09-06T11:43:06.111334Z 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 2026-09-06T11:43:06.252470Z 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 2026-09-06T11:43:06.261435Z 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 2026-09-06T11:43:06.261455Z 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 2026-09-06T11:43:06.278234Z 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 2026-09-06T11:43:16.777376Z 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=4427 ordinary_horizon=45435 paired_horizon=34910 use_paired=true 2026-09-06T11:43:16.866662Z 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=6369 maximum_identical_destinations=1 2026-09-06T11:43:16.866688Z 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=34910 selected_horizon=34910 endpoint_lower_bound=30024 lower_bound_gap=4886 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:43:16.911493Z 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=1151 row_words=326 horizon=34910 2026-09-06T11:43:17.217686Z 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 2026-09-06T11:43:17.236378Z 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=41760 maximum_identical_destinations=1 2026-09-06T11:43:17.236393Z 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=4999 selected_horizon=4999 endpoint_lower_bound=3024 lower_bound_gap=1975 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:43:17.272016Z 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=223 row_words=178 horizon=4999 2026-09-06T11:43:17.291557Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: evaluated exchange placement finalist offset=20480 baseline_score=667950022656 score=267289362432 baseline_cycles=57926 candidate_cycles=50736 row_bytes=2068 row_capacity=16320 accepted=true 2026-09-06T11:43:17.301399Z 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=12498 success=true 2026-09-06T11:43:17.613262Z 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=11 success=true 2026-09-06T11:43:17.624364Z 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 2026-09-06T11:43:17.730497Z 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 2026-09-06T11:43:17.790907Z INFO writing application package tiles=1472 2026-09-06T11:43:17.844817Z INFO application package written 2026-09-06T11:43:17.847296Z INFO reading application package 2026-09-06T11:43:17.947531Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:43:17.949866Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:43:17.949883Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:43:17.949888Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:43:17.949921Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:43:17.949923Z INFO initializing IPU device 2026-09-06T11:43:17.958289Z INFO IPU device initialized 2026-09-06T11:43:17.958294Z INFO replaying device configuration register_writes=2003 2026-09-06T11:43:17.967070Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:43:18.249860Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:43:20.930281Z INFO starting host exchange session startup_mark=23 2026-09-06T11:43:20.930305Z INFO attaching host exchange pages pages=2 2026-09-06T11:43:20.930839Z INFO host exchange session started 2026-09-06T11:43:20.930949Z INFO invoking streaming host exchange call call="initialize" command=0 phases=22 2026-09-06T11:43:20.939222Z INFO invoking streaming host exchange call call="run" command=0 phases=59 profile=artifacts/layout-sweep/down-13/execution.ipuprofile tiles=1472 samples=25862 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=196896 minimumTileCycles=185490 deviceMicroseconds=131.264 effectiveGemmTflops=110.145 peakTflops=282.624 efficiencyPercent=38.97 maximumAbsoluteError=0.009766 package=artifacts/layout-sweep/down-13/model.ipuexe tiles=1472 hardwareTest=PASS