2026-09-06T11:37:59.864572Z INFO applied GEMM plan constraint source_operation=0 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 0, orientation: Normal, row_partitions: 5, column_partitions: 135, inner_partitions: 2, result_row_partitions: 2, result_column_partitions: 1, output_column_block: 32, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:38:00.316328Z INFO applied GEMM plan constraint source_operation=2 matching_plans=1 constraint=GemmPlanConstraint { source_operation: 2, orientation: Normal, row_partitions: 4, column_partitions: 24, inner_partitions: 15, result_row_partitions: 15, result_column_partitions: 1, output_column_block: 48, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct } 2026-09-06T11:38:00.941427Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=311951 estimated_exchange_cycles=124440 exchange_row_bytes=27344 peak_standard_bytes=207568 peak_interleaved_bytes=101376 peak_total_bytes=308944 maximum_standard_allocation_bytes=180224 active_tile_counts={1350, 1440} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:38:00.947341Z 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=1377 success=true 2026-09-06T11:38:05.079437Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=247784 estimated_exchange_cycles=70536 2026-09-06T11:38:05.166019Z 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=72 2026-09-06T11:38:05.168034Z 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=4220 success=true 2026-09-06T11:38:05.168050Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26136 exchange_phases=4 2026-09-06T11:38:05.168269Z 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 2026-09-06T11:38:05.168871Z 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 2026-09-06T11:38:05.169097Z 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_k576_c16_r146_r146" target=ipu21 2026-09-06T11:38:05.221836Z 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/b4a5425246ddcd3ae1443c81d82e642abfb55582affeec1c1242f38c16653a93/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/b4a5425246ddcd3ae1443c81d82e642abfb55582affeec1c1242f38c16653a93/kernel.json 2026-09-06T11:38:05.222169Z 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_k576_c32_r146_r146" target=ipu21 2026-09-06T11:38:05.277223Z 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/cc1ae96453c079cd1a50714b1df27c49bd784f5762d6d3b27b37d1e18a5c0c60/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/cc1ae96453c079cd1a50714b1df27c49bd784f5762d6d3b27b37d1e18a5c0c60/kernel.json 2026-09-06T11:38:05.277378Z 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=108 success=true 2026-09-06T11:38:05.279410Z 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 2026-09-06T11:38:05.469404Z 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=189 success=true 2026-09-06T11:38:05.605290Z 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=4842 maximum_identical_destinations=1 2026-09-06T11:38:05.605325Z 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=1393 selected_horizon=1362 endpoint_lower_bound=1248 lower_bound_gap=114 selected_kind="critical-neighborhood" neighborhood_improvements=1 2026-09-06T11:38:05.611073Z 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=735 row_words=23 horizon=1362 2026-09-06T11:38:10.236011Z 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=2995 ordinary_horizon=46437 paired_horizon=36323 use_paired=true 2026-09-06T11:38:10.291841Z 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=4217 maximum_identical_destinations=1 2026-09-06T11:38:10.291862Z 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=36323 selected_horizon=36323 endpoint_lower_bound=25488 lower_bound_gap=10835 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:10.322589Z 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=906 row_words=223 horizon=36323 2026-09-06T11:38:16.449635Z 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=63720 maximum_identical_destinations=1 2026-09-06T11:38:16.449687Z 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=7991 selected_horizon=7991 endpoint_lower_bound=4704 lower_bound_gap=3287 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:16.510676Z 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=1058 row_words=294 horizon=7991 2026-09-06T11:38:16.533999Z 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=11064 success=true 2026-09-06T11:38:17.344251Z 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=4 success=true 2026-09-06T11:38:17.344286Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=317272 profile_bytes=132 exchange_table_bytes=1944 host_code_bytes=3556 generated_code_bytes=1728 code_address=320832 standard_ranges=[(409600, 524288)] 2026-09-06T11:38:17.449689Z 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=105 success=true 2026-09-06T11:38:17.459763Z 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 2026-09-06T11:38:17.488223Z 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 2026-09-06T11:38:17.490086Z 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=4842 maximum_identical_destinations=1 2026-09-06T11:38:17.490098Z 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=1362 selected_horizon=1362 endpoint_lower_bound=1248 lower_bound_gap=114 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:38:17.493428Z 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=735 row_words=23 horizon=1362 2026-09-06T11:38:22.535390Z 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=2995 ordinary_horizon=46437 paired_horizon=36323 use_paired=true 2026-09-06T11:38:22.591275Z 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=4217 maximum_identical_destinations=1 2026-09-06T11:38:22.591296Z 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=36323 selected_horizon=36323 endpoint_lower_bound=25488 lower_bound_gap=10835 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:22.620841Z 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=906 row_words=223 horizon=36323 2026-09-06T11:38:28.771814Z 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=63720 maximum_identical_destinations=1 2026-09-06T11:38:28.771856Z 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=7991 selected_horizon=7991 endpoint_lower_bound=4704 lower_bound_gap=3287 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:28.808452Z 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=1058 row_words=294 horizon=7991 2026-09-06T11:38:28.824050Z 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=11374 success=true 2026-09-06T11:38:29.946340Z 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 2026-09-06T11:38:30.007487Z 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 2026-09-06T11:38:30.010571Z 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=4842 maximum_identical_destinations=1 2026-09-06T11:38:30.010593Z 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=1362 selected_horizon=1362 endpoint_lower_bound=1248 lower_bound_gap=114 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:38:30.015859Z 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=735 row_words=23 horizon=1362 2026-09-06T11:38:33.493092Z 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=2995 ordinary_horizon=40491 paired_horizon=30364 use_paired=true 2026-09-06T11:38:33.528291Z 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=4217 maximum_identical_destinations=1 2026-09-06T11:38:33.528308Z 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=30364 selected_horizon=30364 endpoint_lower_bound=25488 lower_bound_gap=4876 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:33.545720Z 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=906 row_words=221 horizon=30364 2026-09-06T11:38:37.727519Z 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=63720 maximum_identical_destinations=1 2026-09-06T11:38:37.727561Z 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=7991 selected_horizon=7991 endpoint_lower_bound=4704 lower_bound_gap=3287 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:38:37.782901Z 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=1058 row_words=294 horizon=7991 2026-09-06T11:38:37.819675Z 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=448685899776 score=173497614336 baseline_cycles=55013 candidate_cycles=49054 row_bytes=1936 row_capacity=16320 accepted=true 2026-09-06T11:38:37.833921Z 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=9009 success=true 2026-09-06T11:38:38.114225Z 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=9 success=true 2026-09-06T11:38:38.122671Z 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=8 success=true 2026-09-06T11:38:38.180727Z 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=54 success=true 2026-09-06T11:38:38.232957Z INFO writing application package tiles=1472 2026-09-06T11:38:38.273576Z INFO application package written 2026-09-06T11:38:38.275065Z INFO reading application package 2026-09-06T11:38:38.342657Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T11:38:38.344696Z INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:38:38.344705Z INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/device.lock 2026-09-06T11:38:38.344709Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T11:38:38.344732Z INFO IPU device attached path=/dev/ipu0 2026-09-06T11:38:38.344734Z INFO initializing IPU device 2026-09-06T11:38:38.352879Z INFO IPU device initialized 2026-09-06T11:38:38.352882Z INFO replaying device configuration register_writes=2003 2026-09-06T11:38:38.360450Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T11:38:38.623955Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T11:38:41.939347Z INFO starting host exchange session startup_mark=23 2026-09-06T11:38:41.939376Z INFO attaching host exchange pages pages=2 2026-09-06T11:38:41.940100Z INFO host exchange session started 2026-09-06T11:38:41.940144Z INFO invoking streaming host exchange call call="initialize" command=0 phases=18 2026-09-06T11:38:41.947143Z INFO invoking streaming host exchange call call="run" command=0 phases=91 profile=artifacts/layout-sweep/up-15/execution.ipuprofile tiles=1472 samples=25559 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=227059200 weightBytes=19869696 cycles=188076 minimumTileCycles=172794 deviceMicroseconds=125.384 effectiveGemmTflops=115.311 peakTflops=282.624 efficiencyPercent=40.80 maximumAbsoluteError=0.011719 package=artifacts/layout-sweep/up-15/model.ipuexe tiles=1472 hardwareTest=PASS