2026-09-06T11:20:46.518194Z  INFO applied GEMM plan constraint source_operation=0 matching_plans=2 constraint=GemmPlanConstraint { source_operation: 0, orientation: Normal, row_partitions: 6, column_partitions: 39, inner_partitions: 6, result_row_partitions: 6, result_column_partitions: 1, output_column_block: 112, weight_memory_class: Ipu21Interleaved, reduction_staging: Complete, local_weight_staging: Direct }
2026-09-06T11:20:46.918916Z  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 }
2026-09-06T11:20:47.461691Z  INFO retained operator-plan finalist finalist=0 values=7 operations=4 estimated_cycles=351859 estimated_exchange_cycles=140760 exchange_row_bytes=31016 peak_standard_bytes=145704 peak_interleaved_bytes=113664 peak_total_bytes=206120 maximum_standard_allocation_bytes=114688 active_tile_counts={1404, 1440} padding_group_counts={} conversion_sources=[Some(OperationId(2))]
2026-09-06T11:20:47.466191Z  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=1181 success=true
2026-09-06T11:20:51.944761Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected analytical operator plan estimated_cycles=286035 estimated_exchange_cycles=83952
2026-09-06T11:20:52.034040Z  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
2026-09-06T11:20:52.036305Z  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=4569 success=true
2026-09-06T11:20:52.036335Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=25434 exchange_phases=4
2026-09-06T11:20:52.036570Z  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
2026-09-06T11:20:52.037421Z  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
2026-09-06T11:20:52.037534Z  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_k192_c96_r120_r122" target=ipu21
2026-09-06T11:20:52.095012Z  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/8ec742f091450e58717c3388bc9d61fbd3fc216f55f3e43c2845a850ebb54f3e/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/8ec742f091450e58717c3388bc9d61fbd3fc216f55f3e43c2845a850ebb54f3e/kernel.json
2026-09-06T11:20:52.095330Z  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_k192_c112_r120_r122" target=ipu21
2026-09-06T11:20:52.147076Z  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/6a42c0a2490eae8092f8316f338eb15d6253851461364500a3daea081c3a002f/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/6a42c0a2490eae8092f8316f338eb15d6253851461364500a3daea081c3a002f/kernel.json
2026-09-06T11:20:52.147642Z  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=110 success=true
2026-09-06T11:20:52.150489Z  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
2026-09-06T11:20:52.297908Z  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=147 success=true
2026-09-06T11:20:55.400443Z  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=54876 maximum_identical_destinations=1
2026-09-06T11:20:55.400542Z  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=7914 selected_horizon=7914 endpoint_lower_bound=6160 lower_bound_gap=1754 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:20:55.451106Z  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=228 row_words=237 horizon=7914
2026-09-06T11:21:25.180520Z  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=8780 ordinary_horizon=46689 paired_horizon=44589 use_paired=true
2026-09-06T11:21:25.418323Z  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=11211 maximum_identical_destinations=1
2026-09-06T11:21:25.418369Z  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=44589 selected_horizon=44589 endpoint_lower_bound=22752 lower_bound_gap=21837 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:21:25.546359Z  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=826 row_words=626 horizon=44589
2026-09-06T11:21:32.981185Z  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
2026-09-06T11:21:32.981274Z  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
2026-09-06T11:21:33.045266Z  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
2026-09-06T11:21:33.072929Z  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=40774 success=true
2026-09-06T11:21:33.462585Z  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=7 success=true
2026-09-06T11:21:33.462646Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319128 profile_bytes=132 exchange_table_bytes=3912 host_code_bytes=1764 generated_code_bytes=1472 code_address=320896 standard_ranges=[(409600, 524288)]
2026-09-06T11:21:33.588443Z  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=125 success=true
2026-09-06T11:21:33.601037Z  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
2026-09-06T11:21:34.113474Z  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
2026-09-06T11:21:34.137966Z  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=54876 maximum_identical_destinations=1
2026-09-06T11:21:34.137994Z  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=7914 selected_horizon=7914 endpoint_lower_bound=6160 lower_bound_gap=1754 selected_kind="reused" neighborhood_improvements=0
2026-09-06T11:21:34.188510Z  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=228 row_words=237 horizon=7914
2026-09-06T11:22:04.729431Z  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=8780 ordinary_horizon=46689 paired_horizon=44589 use_paired=true
2026-09-06T11:22:04.896943Z  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=11211 maximum_identical_destinations=1
2026-09-06T11:22:04.897000Z  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=44589 selected_horizon=44589 endpoint_lower_bound=22752 lower_bound_gap=21837 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:22:04.977335Z  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=826 row_words=626 horizon=44589
2026-09-06T11:22:12.079351Z  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
2026-09-06T11:22:12.079415Z  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
2026-09-06T11:22:12.149093Z  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
2026-09-06T11:22:12.177519Z  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=38589 success=true
2026-09-06T11:22:13.873631Z  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
2026-09-06T11:22:14.411761Z  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
2026-09-06T11:22:14.437651Z  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=54876 maximum_identical_destinations=1
2026-09-06T11:22:14.437682Z  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=7914 selected_horizon=7914 endpoint_lower_bound=6160 lower_bound_gap=1754 selected_kind="reused" neighborhood_improvements=0
2026-09-06T11:22:14.493542Z  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=228 row_words=237 horizon=7914
2026-09-06T11:22:50.805240Z  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=8780 ordinary_horizon=39963 paired_horizon=36704 use_paired=true
2026-09-06T11:22:50.930648Z  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=11211 maximum_identical_destinations=1
2026-09-06T11:22:50.930669Z  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=36704 selected_horizon=36704 endpoint_lower_bound=22752 lower_bound_gap=13952 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-06T11:22:51.002410Z  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=826 row_words=615 horizon=36704
2026-09-06T11:22:55.112423Z  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
2026-09-06T11:22:55.112477Z  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
2026-09-06T11:22:55.168960Z  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
2026-09-06T11:22:55.195186Z  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=172365152256 baseline_cycles=71367 candidate_cycles=63482 row_bytes=3912 row_capacity=16320 accepted=true
2026-09-06T11:22:55.208980Z  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=43031 success=true
2026-09-06T11:22:55.326823Z  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
2026-09-06T11:22:55.338643Z  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
2026-09-06T11:22:55.442163Z  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=98 success=true
2026-09-06T11:22:55.500546Z  INFO writing application package tiles=1472
2026-09-06T11:22:55.563832Z  INFO application package written
2026-09-06T11:22:55.566542Z  INFO reading application package
2026-09-06T11:22:55.668923Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-06T11:22:55.671558Z  INFO opening IPU device path="/dev/ipu0"
2026-09-06T11:22:55.671593Z  INFO IPU device attached path=/dev/ipu0
2026-09-06T11:22:55.671597Z  INFO initializing IPU device
2026-09-06T11:22:55.679429Z  INFO IPU device initialized
2026-09-06T11:22:55.679483Z  INFO replaying device configuration register_writes=2003
2026-09-06T11:22:55.685193Z  INFO loading application tile_count=1472 final_mark=23
2026-09-06T11:22:55.895338Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-06T11:22:56.952677Z  INFO starting host exchange session startup_mark=23
2026-09-06T11:22:56.952717Z  INFO attaching host exchange pages pages=2
2026-09-06T11:22:56.953378Z  INFO host exchange session started
2026-09-06T11:22:56.953443Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=22
2026-09-06T11:22:56.960003Z  INFO invoking streaming host exchange call call="run" command=0 phases=31
profile=artifacts/layout-sweep/up-03/execution.ipuprofile tiles=1472 samples=25454
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=65594880 weightBytes=19869696 cycles=220170 minimumTileCycles=205224 deviceMicroseconds=146.780 effectiveGemmTflops=98.502 peakTflops=282.624 efficiencyPercent=34.85 maximumAbsoluteError=0.009766
package=artifacts/layout-sweep/up-03/model.ipuexe tiles=1472 hardwareTest=PASS