2026-09-06T12:40:15.844417Z  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 }
2026-09-06T12:40:16.415737Z  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: Ipu21Standard, reduction_staging: Complete, local_weight_staging: Direct }
2026-09-06T12:40:16.971580Z  INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=259590 estimated_exchange_cycles=123960 exchange_row_bytes=27236 peak_standard_bytes=169572 peak_interleaved_bytes=60416 peak_total_bytes=229988 maximum_standard_allocation_bytes=114688 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2))]
2026-09-06T12:40:16.976899Z  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=1435 success=true
2026-09-06T12:40:20.608563Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=192930 estimated_exchange_cycles=70152
2026-09-06T12:40:20.701663Z  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
2026-09-06T12:40:20.704102Z  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=3727 success=true
2026-09-06T12:40:20.704131Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26264 exchange_phases=4
2026-09-06T12:40:20.704420Z  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
2026-09-06T12:40:20.705444Z  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=1 success=true
2026-09-06T12:40:20.705733Z  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
2026-09-06T12:40:20.708080Z  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
2026-09-06T12:40:20.857465Z  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=149 success=true
2026-09-06T12:40:21.899758Z  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
2026-09-06T12:40:21.899801Z  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=4492 selected_horizon=3807 endpoint_lower_bound=3408 lower_bound_gap=399 selected_kind="matching-waves" neighborhood_improvements=0
2026-09-06T12:40:21.921449Z  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=1230 row_words=75 horizon=3807
2026-09-06T12:40:26.006460Z  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=5562 maximum_identical_destinations=1
2026-09-06T12:40:26.006535Z  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=48919 selected_horizon=48919 endpoint_lower_bound=33408 lower_bound_gap=15511 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T12:40:26.056729Z  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=795 row_words=254 horizon=48919
2026-09-06T12:40:33.583417Z  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
2026-09-06T12:40:33.583476Z  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=8184 selected_horizon=8184 endpoint_lower_bound=4704 lower_bound_gap=3480 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T12:40:33.658845Z  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=764 row_words=280 horizon=8184
2026-09-06T12:40:33.702195Z  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=12844 success=true
2026-09-06T12:40:34.637816Z  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
2026-09-06T12:40:34.637850Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=317036 profile_bytes=132 exchange_table_bytes=2280 host_code_bytes=2416 generated_code_bytes=1544 code_address=319456 standard_ranges=[(409600, 524288)]
2026-09-06T12:40:34.811109Z  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=173 success=true
2026-09-06T12:40:34.834313Z  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
2026-09-06T12:40:35.082773Z  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
2026-09-06T12:40:35.092672Z  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
2026-09-06T12:40:35.092695Z  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=3807 selected_horizon=3807 endpoint_lower_bound=3408 lower_bound_gap=399 selected_kind="reused" neighborhood_improvements=0
2026-09-06T12:40:35.157753Z  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=1230 row_words=75 horizon=3807
2026-09-06T12:40:40.365894Z  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=5562 maximum_identical_destinations=1
2026-09-06T12:40:40.365964Z  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=48919 selected_horizon=48919 endpoint_lower_bound=33408 lower_bound_gap=15511 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T12:40:40.414034Z  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=795 row_words=254 horizon=48919
2026-09-06T12:40:48.686039Z  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
2026-09-06T12:40:48.686088Z  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=8184 selected_horizon=8184 endpoint_lower_bound=4704 lower_bound_gap=3480 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T12:40:48.775111Z  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=764 row_words=280 horizon=8184
2026-09-06T12:40:48.810664Z  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=13999 success=true
2026-09-06T12:40:51.232845Z  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
2026-09-06T12:40:51.445590Z  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
2026-09-06T12:40:51.455651Z  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
2026-09-06T12:40:51.455674Z  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=3807 selected_horizon=3807 endpoint_lower_bound=3408 lower_bound_gap=399 selected_kind="reused" neighborhood_improvements=0
2026-09-06T12:40:51.475893Z  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=1230 row_words=75 horizon=3807
2026-09-06T12:40:56.349009Z  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=5562 maximum_identical_destinations=1
2026-09-06T12:40:56.349057Z  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=40549 selected_horizon=40549 endpoint_lower_bound=33408 lower_bound_gap=7141 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T12:40:56.400342Z  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=795 row_words=265 horizon=40549
2026-09-06T12:41:03.524178Z  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
2026-09-06T12:41:03.524230Z  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=8184 selected_horizon=8184 endpoint_lower_bound=4704 lower_bound_gap=3480 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T12:41:03.584512Z  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=764 row_words=280 horizon=8184
2026-09-06T12:41:03.615103Z  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=170111729664 score=151709220864 baseline_cycles=67945 candidate_cycles=59575 row_bytes=2280 row_capacity=16320 accepted=true
2026-09-06T12:41:03.628241Z  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=14817 success=true
2026-09-06T12:41:03.945330Z  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
2026-09-06T12:41:03.955688Z  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=10 success=true
2026-09-06T12:41:04.049570Z  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=89 success=true
2026-09-06T12:41:04.140508Z  INFO writing application package tiles=1472
2026-09-06T12:41:04.186906Z  INFO application package written
2026-09-06T12:41:04.187456Z  INFO reading application package
2026-09-06T12:41:04.280818Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-06T12:41:04.281310Z  INFO waiting for hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/alternatives-followup/device.lock
2026-09-06T12:41:22.493358Z  INFO acquired hardware ownership path=/srv/home/gc-sdk/ipu-stack/artifacts/layout-sweep/alternatives-followup/device.lock
2026-09-06T12:41:22.493413Z  INFO opening IPU device path="/dev/ipu0"
2026-09-06T12:41:22.493484Z  INFO IPU device attached path=/dev/ipu0
2026-09-06T12:41:22.493491Z  INFO initializing IPU device
2026-09-06T12:41:22.501340Z  INFO IPU device initialized
2026-09-06T12:41:22.501370Z  INFO replaying device configuration register_writes=2003
2026-09-06T12:41:22.510753Z  INFO loading application tile_count=1472 final_mark=23
2026-09-06T12:41:22.775498Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-06T12:41:25.721926Z  INFO starting host exchange session startup_mark=23
2026-09-06T12:41:25.721960Z  INFO attaching host exchange pages pages=2
2026-09-06T12:41:25.722463Z  INFO host exchange session started
2026-09-06T12:41:25.722507Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=14
2026-09-06T12:41:25.727754Z  INFO invoking streaming host exchange call call="run" command=0 phases=59
profile=artifacts/layout-sweep/alternatives-followup/mapping-block-360-24-base/execution.ipuprofile tiles=1472 samples=25788
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=192750 minimumTileCycles=180216 deviceMicroseconds=128.500 effectiveGemmTflops=112.515 peakTflops=282.624 efficiencyPercent=39.81 maximumAbsoluteError=0.011719
package=artifacts/layout-sweep/alternatives-followup/mapping-block-360-24-base/model.ipuexe tiles=1472 hardwareTest=PASS