2026-09-06T11:58:07.372825Z INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=264707 estimated_exchange_cycles=106840 exchange_row_bytes=23384 peak_standard_bytes=224088 peak_interleaved_bytes=53248 peak_total_bytes=272216 maximum_standard_allocation_bytes=180224 active_tile_counts={1458, 1470} padding_group_counts={} conversion_sources=[Some(OperationId(2))] 2026-09-06T11:58:07.692862Z 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=12884 success=true 2026-09-06T11:58:09.263571Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=232689 estimated_exchange_cycles=81992 2026-09-06T11:58:09.327679Z 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=36 2026-09-06T11:58:09.329818Z 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=1636 success=true 2026-09-06T11:58:09.329843Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26664 exchange_phases=4 2026-09-06T11:58:09.330034Z 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 2026-09-06T11:58:09.330815Z 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 2026-09-06T11:58:09.331051Z 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=0 success=true 2026-09-06T11:58:09.332943Z 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 2026-09-06T11:58:09.470271Z 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=137 success=true 2026-09-06T11:58:09.955735Z 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=12105 maximum_identical_destinations=1 2026-09-06T11:58:09.955783Z 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=6557 selected_horizon=5174 endpoint_lower_bound=4800 lower_bound_gap=374 selected_kind="matching-waves" neighborhood_improvements=0 2026-09-06T11:58:09.966626Z 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=619 row_words=66 horizon=5174 2026-09-06T11:58:12.271229Z 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=3062 ordinary_horizon=28839 paired_horizon=31187 use_paired=false 2026-09-06T11:58:12.314924Z 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=4200 maximum_identical_destinations=1 2026-09-06T11:58:12.314953Z 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=29705 selected_horizon=28839 endpoint_lower_bound=24640 lower_bound_gap=4199 selected_kind="critical-neighborhood" neighborhood_improvements=1 2026-09-06T11:58:12.335330Z 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=1102 row_words=118 horizon=28839 2026-09-06T11:58:47.346774Z 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=155880 maximum_identical_destinations=1 2026-09-06T11:58:47.346807Z 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=15700 selected_horizon=15700 endpoint_lower_bound=8640 lower_bound_gap=7060 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:58:47.461447Z 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=447 row_words=633 horizon=15700 2026-09-06T11:58:47.494997Z 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=38024 success=true 2026-09-06T11:58:48.049977Z 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 2026-09-06T11:58:48.050013Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319128 profile_bytes=100 exchange_table_bytes=3140 host_code_bytes=3356 generated_code_bytes=1640 code_address=322488 standard_ranges=[(409600, 524288)] 2026-09-06T11:58:48.166666Z 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=116 success=true 2026-09-06T11:58:48.172824Z 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 2026-09-06T11:58:48.260009Z 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 2026-09-06T11:58:48.265065Z 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=12105 maximum_identical_destinations=1 2026-09-06T11:58:48.265097Z 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=5174 selected_horizon=5174 endpoint_lower_bound=4800 lower_bound_gap=374 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:58:48.274898Z 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=619 row_words=66 horizon=5174 2026-09-06T11:58:48.551258Z 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=2 2026-09-06T11:58:48.564296Z 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=4200 maximum_identical_destinations=1 2026-09-06T11:58:48.564311Z 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=28839 selected_horizon=28839 endpoint_lower_bound=24640 lower_bound_gap=4199 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:58:48.582520Z 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=1102 row_words=118 horizon=28839 2026-09-06T11:59:26.264939Z 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=155880 maximum_identical_destinations=1 2026-09-06T11:59:26.264984Z 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=15700 selected_horizon=15700 endpoint_lower_bound=8640 lower_bound_gap=7060 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T11:59:26.470319Z 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=447 row_words=633 horizon=15700 2026-09-06T11:59:26.517999Z 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=38351 success=true 2026-09-06T11:59:27.962069Z 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 2026-09-06T11:59:28.046501Z 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 2026-09-06T11:59:28.051668Z 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=12105 maximum_identical_destinations=1 2026-09-06T11:59:28.051682Z 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=5174 selected_horizon=5174 endpoint_lower_bound=4800 lower_bound_gap=374 selected_kind="reused" neighborhood_improvements=0 2026-09-06T11:59:28.064458Z 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=619 row_words=66 horizon=5174 2026-09-06T11:59:30.244094Z 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=3062 ordinary_horizon=28735 paired_horizon=28683 use_paired=true 2026-09-06T11:59:30.290551Z 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=4200 maximum_identical_destinations=1 2026-09-06T11:59:30.290575Z 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=29293 selected_horizon=28683 endpoint_lower_bound=24640 lower_bound_gap=4043 selected_kind="critical-neighborhood" neighborhood_improvements=2 2026-09-06T11:59:30.314194Z 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=905 row_words=165 horizon=28683 2026-09-06T12:00:11.544139Z 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=155880 maximum_identical_destinations=1 2026-09-06T12:00:11.544224Z 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=15700 selected_horizon=15700 endpoint_lower_bound=8640 lower_bound_gap=7060 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-06T12:00:11.764060Z 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=447 row_words=633 horizon=15700 2026-09-06T12:00:11.849079Z 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=127958384640 score=105171517440 baseline_cycles=53767 candidate_cycles=53611 row_bytes=3288 row_capacity=16320 accepted=true 2026-09-06T12:00:11.880353Z 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=45362 success=true 2026-09-06T12:00:12.336417Z 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=7 success=true 2026-09-06T12:00:12.352668Z 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=16 success=true 2026-09-06T12:00:12.471889Z 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=112 success=true 2026-09-06T12:00:12.577779Z INFO writing application package tiles=1472 2026-09-06T12:00:12.625135Z INFO application package written 2026-09-06T12:00:12.626027Z INFO reading application package 2026-09-06T12:00:12.750750Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-06T12:00:12.751519Z INFO waiting for hardware ownership path=artifacts/layout-sweep/device.lock 2026-09-06T12:00:12.751536Z INFO acquired hardware ownership path=artifacts/layout-sweep/device.lock 2026-09-06T12:00:12.751541Z INFO opening IPU device path="/dev/ipu0" 2026-09-06T12:00:12.751585Z INFO IPU device attached path=/dev/ipu0 2026-09-06T12:00:12.751588Z INFO initializing IPU device 2026-09-06T12:00:12.759643Z INFO IPU device initialized 2026-09-06T12:00:12.759677Z INFO replaying device configuration register_writes=2003 2026-09-06T12:00:12.774168Z INFO loading application tile_count=1472 final_mark=23 2026-09-06T12:00:13.058990Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-06T12:00:16.743167Z INFO starting host exchange session startup_mark=23 2026-09-06T12:00:16.743255Z INFO attaching host exchange pages pages=2 2026-09-06T12:00:16.743986Z INFO host exchange session started 2026-09-06T12:00:16.744102Z INFO invoking streaming host exchange call call="initialize" command=0 phases=10 2026-09-06T12:00:16.754412Z INFO invoking streaming host exchange call call="run" command=0 phases=93 profile=artifacts/layout-sweep/instruction-cost/execution.ipuprofile tiles=1472 samples=24048 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=257544000 weightBytes=20282880 cycles=198786 minimumTileCycles=168960 deviceMicroseconds=132.524 effectiveGemmTflops=109.098 peakTflops=282.624 efficiencyPercent=38.60 maximumAbsoluteError=0.011719 package=artifacts/layout-sweep/instruction-cost/model.ipuexe tiles=1472 hardwareTest=PASS