2026-09-22T14:57:06.182234Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_runtime: close time.busy=288µs time.idle=6.07µs
2026-09-22T14:57:53.815994Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:construct_mid:gemm_candidates: close time.busy=47.6s time.idle=3.28µs
2026-09-22T14:59:12.038885Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:construct_mid:gemm_candidates: close time.busy=78.2s time.idle=3.58µs
2026-09-22T14:59:14.913503Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:construct_mid: close time.busy=129s time.idle=2.58µs
2026-09-22T14:59:14.967668Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: wrote planner memory profile path=artifacts/planner-memory-20260922/fp8-b2-n1-256k/memory/baseline-1172369-0000.json scope="baseline" peak_step=2 effective_peak_bytes=235008
2026-09-22T14:59:15.635548Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:group_exchanges: close time.busy=100µs time.idle=2.61µs
2026-09-22T14:59:15.637039Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:merge_copies: close time.busy=1.29ms time.idle=1.13µs
2026-09-22T14:59:15.657763Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:eliminate: eliminated padding clears using kernel access contracts removed=0
2026-09-22T14:59:15.658815Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:eliminate: close time.busy=21.8ms time.idle=725ns
2026-09-22T14:59:15.691775Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:place_provisional_storage: close time.busy=32.4ms time.idle=2.00µs
2026-09-22T14:59:16.072320Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=69984
2026-09-22T14:59:16.328686Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=4 transfers=3096 maximum_identical_destinations=1
2026-09-22T14:59:16.328708Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=4 initial_horizon=15897 selected_horizon=15868 endpoint_lower_bound=15360 lower_bound_gap=508 selected_kind="critical-neighborhood" neighborhood_improvements=1
2026-09-22T14:59:16.633960Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=69984 elapsed_ms=561 success=true
2026-09-22T14:59:16.739682Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=39366
2026-09-22T14:59:17.234848Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=6 transfers=69984 maximum_identical_destinations=1
2026-09-22T14:59:17.234873Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=6 initial_horizon=554 selected_horizon=554 endpoint_lower_bound=384 lower_bound_gap=170 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:59:17.285565Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=1 transfers=91056 maximum_identical_destinations=1
2026-09-22T14:59:17.285594Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=1 initial_horizon=8422 selected_horizon=7576 endpoint_lower_bound=7056 lower_bound_gap=520 selected_kind="matching-waves" neighborhood_improvements=0
2026-09-22T14:59:18.129137Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=39366 elapsed_ms=1389 success=true
2026-09-22T14:59:18.503896Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=16932
2026-09-22T14:59:19.608691Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=39366
2026-09-22T14:59:20.800623Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=16932 elapsed_ms=2296 success=true
2026-09-22T14:59:21.015257Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=365560
2026-09-22T14:59:21.019581Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=39366 elapsed_ms=1410 success=true
2026-09-22T14:59:21.644324Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: compared ordinary and paired exchange schedules phase=3 candidates=486 ordinary_horizon=6855 paired_horizon=7003 use_paired=false
2026-09-22T14:59:21.698316Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=3 transfers=39366 maximum_identical_destinations=1
2026-09-22T14:59:21.698332Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=3 initial_horizon=6855 selected_horizon=6855 endpoint_lower_bound=6400 lower_bound_gap=455 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:59:23.115418Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=16932
2026-09-22T14:59:25.515677Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=16932 elapsed_ms=2400 success=true
2026-09-22T14:59:26.675239Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: compared ordinary and paired exchange schedules phase=0 candidates=88 ordinary_horizon=29457 paired_horizon=29529 use_paired=false
2026-09-22T14:59:26.785853Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=0 transfers=16932 maximum_identical_destinations=1
2026-09-22T14:59:26.785869Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=0 initial_horizon=29457 selected_horizon=29457 endpoint_lower_bound=28008 lower_bound_gap=1449 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:59:28.156220Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=365560 elapsed_ms=7140 success=true
2026-09-22T14:59:30.287106Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=909792
2026-09-22T14:59:34.818484Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=2 transfers=365560 maximum_identical_destinations=1
2026-09-22T14:59:34.818511Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=2 initial_horizon=3211 selected_horizon=3211 endpoint_lower_bound=2352 lower_bound_gap=859 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:59:50.704408Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=909792 elapsed_ms=20417 success=true
2026-09-22T15:00:10.027036Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=5 transfers=909792 maximum_identical_destinations=1
2026-09-22T15:00:10.027068Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=5 initial_horizon=20277 selected_horizon=20277 endpoint_lower_bound=14976 lower_bound_gap=5301 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T15:00:11.783662Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: close time.busy=56.1s time.idle=4.54µs
2026-09-22T15:00:11.830769Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:plan_kernels: close time.busy=47.1ms time.idle=1.90µs
2026-09-22T15:00:11.832244Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: close time.busy=1.45ms time.idle=2.46µs
2026-09-22T15:00:11.832502Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:link_runtime: close time.busy=171µs time.idle=746ns
2026-09-22T15:00:12.171014Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: exchange table and repeat patch storage exchange_table_bytes=19668 maximum_uncompressed_repeat_patch_bytes=0 maximum_elided_arithmetic_patch_bytes=0
2026-09-22T15:00:12.691864Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:size_tile_code: close time.busy=42.0ms time.idle=6.56µs
2026-09-22T15:00:12.691923Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: placing generated tile programs linked_end=318984 host_code_base=376832 host_code_bytes=572 generated_code_bytes=1140 exchange_table_bytes=19668 executable_ranges=[(377472, 524288)]
2026-09-22T15:00:12.691979Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=318984 profile_bytes=96 exchange_table_bytes=19668 host_code_bytes=572 host_data_bytes=392 generated_code_bytes=1140 code_address=377472 available_ranges=[(327680, 360448), (362600, 376832), (425984, 949168)]
2026-09-22T15:00:12.767804Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:place_storage: close time.busy=57.0ms time.idle=3.08µs
2026-09-22T15:00:13.009260Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=4
2026-09-22T15:00:13.020862Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=4 transfers=3096 maximum_identical_destinations=1
2026-09-22T15:00:13.020879Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=4 initial_horizon=15868 selected_horizon=15868 endpoint_lower_bound=15360 lower_bound_gap=508 selected_kind="reused" neighborhood_improvements=0
2026-09-22T15:00:13.466187Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=1
2026-09-22T15:00:13.493900Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=1 transfers=91056 maximum_identical_destinations=1
2026-09-22T15:00:13.493920Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=1 initial_horizon=7576 selected_horizon=7576 endpoint_lower_bound=7056 lower_bound_gap=520 selected_kind="reused" neighborhood_improvements=0
2026-09-22T15:00:13.516495Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=6
2026-09-22T15:00:13.530156Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=6 transfers=69984 maximum_identical_destinations=1
2026-09-22T15:00:13.530181Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=6 initial_horizon=554 selected_horizon=554 endpoint_lower_bound=384 lower_bound_gap=170 selected_kind="reused" neighborhood_improvements=0
2026-09-22T15:00:13.783959Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=3
2026-09-22T15:00:13.802552Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=3 transfers=39366 maximum_identical_destinations=1
2026-09-22T15:00:13.802566Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=3 initial_horizon=6855 selected_horizon=6855 endpoint_lower_bound=6400 lower_bound_gap=455 selected_kind="reused" neighborhood_improvements=0
2026-09-22T15:00:17.298052Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=0
2026-09-22T15:00:17.348598Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=0 transfers=16932 maximum_identical_destinations=1
2026-09-22T15:00:17.348616Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=0 initial_horizon=29457 selected_horizon=29457 endpoint_lower_bound=28008 lower_bound_gap=1449 selected_kind="reused" neighborhood_improvements=0
2026-09-22T15:00:20.203409Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=2
2026-09-22T15:00:20.278315Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=2 transfers=365560 maximum_identical_destinations=1
2026-09-22T15:00:20.278335Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=2 initial_horizon=3211 selected_horizon=3211 endpoint_lower_bound=2352 lower_bound_gap=859 selected_kind="reused" neighborhood_improvements=0
2026-09-22T15:00:30.832212Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=5
2026-09-22T15:00:31.037623Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=5 transfers=909792 maximum_identical_destinations=1
2026-09-22T15:00:31.037652Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=5 initial_horizon=20277 selected_horizon=20277 endpoint_lower_bound=14976 lower_bound_gap=5301 selected_kind="reused" neighborhood_improvements=0
2026-09-22T15:00:32.683047Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: close time.busy=19.9s time.idle=5.90µs
2026-09-22T15:00:32.686731Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: costed final placed program final_cycles=450078 final_exchange=87998
2026-09-22T15:00:33.180826Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:prepare_tile_code: close time.busy=104ms time.idle=5.19µs
2026-09-22T15:00:33.205880Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:emit_tile_code: close time.busy=25.0ms time.idle=2.21µs
2026-09-22T15:00:33.357728Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:build_tile_images: close time.busy=146ms time.idle=1.99µs
2026-09-22T15:00:33.539402Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: wrote exact tile placement profile path=artifacts/planner-memory-20260922/fp8-b2-n1-256k/memory/placement-1172369.html
2026-09-22T15:00:33.610103Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: close time.busy=207s time.idle=4.89µs
2026-09-22T15:00:33.614486Z  INFO writing application package tiles=1472
2026-09-22T15:00:33.748586Z  INFO application package written
2026-09-22T15:00:33.749710Z  INFO reading application package
2026-09-22T15:00:33.971159Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-22T15:00:33.978770Z  INFO waiting for hardware ownership path=/tmp/ipu-stack-device.lock
2026-09-22T15:00:33.978784Z  INFO acquired hardware ownership path=/tmp/ipu-stack-device.lock
2026-09-22T15:00:33.978789Z  INFO opening IPU device path="/dev/ipu0"
2026-09-22T15:00:33.978819Z  INFO IPU device attached path=/dev/ipu0
2026-09-22T15:00:33.978822Z  INFO initializing IPU device
2026-09-22T15:00:33.989191Z  INFO IPU device initialized
2026-09-22T15:00:33.989281Z  INFO replaying device configuration register_writes=2003
2026-09-22T15:00:33.995854Z  INFO loading application tile_count=1472 final_mark=23
2026-09-22T15:00:34.144598Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-22T15:00:34.962043Z  INFO prepared reference inputs elapsed_ms=815
2026-09-22T15:00:35.185710Z  INFO configured host reference BLAS threads=4
referencePrecision=device
2026-09-22T15:00:36.053881Z  INFO evaluated host reference elapsed_ms=1091
2026-09-22T15:00:36.056586Z  INFO starting host exchange session startup_mark=23
2026-09-22T15:00:36.056602Z  INFO attaching host exchange pages pages=2
2026-09-22T15:00:36.057273Z  INFO host exchange session started
2026-09-22T15:00:36.057321Z  INFO invoking host exchange call call="initialize" command=0 phases=4
2026-09-22T15:00:36.065079Z  INFO invoking host exchange call call="run" command=0 phases=11
referenceInference=1 numericalTest=PASS
profile=artifacts/planner-memory-20260922/fp8-b2-n1-256k/profile.capnp tiles=1472 samples=30741
workload=siglip-mlp-f16-b2-t729-d1152-h4304-n1 benchmark=siglip-mlp-f16 batch=2 tokens=729 rows=1458 dimension=1152 hiddenDimension=4304 blocks=1 biases=false activeTiles=1458 executionTiles=1472 inputBytes=1679616 weightBytes=9934848 cycles=340494 minimumTileCycles=333816 deviceMicroseconds=226.996 effectiveGemmTflops=127.387 peakTflops=282.624 efficiencyPercent=45.07 maximumAbsoluteError=0.005493
package=artifacts/planner-memory-20260922/fp8-b2-n1-256k/model.ipuexe tiles=1472 hardwareTest=PASS