2026-09-22T14:53:10.019400Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_runtime: close time.busy=276µs time.idle=4.83µs
2026-09-22T14:54:22.874823Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:construct_mid:gemm_candidates: close time.busy=72.9s time.idle=2.95µs
2026-09-22T14:56:34.314884Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:construct_mid:gemm_candidates: close time.busy=131s time.idle=3.55µs
2026-09-22T14:56:38.767987Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:construct_mid: close time.busy=209s time.idle=3.05µs
2026-09-22T14:56:38.877362Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: wrote planner memory profile path=artifacts/planner-memory-20260922/fp8-b4-n1-384k/memory/baseline-1171362-0000.json scope="baseline" peak_step=2 effective_peak_bytes=320512
2026-09-22T14:56:40.332933Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:group_exchanges: close time.busy=133µs time.idle=3.53µs
2026-09-22T14:56:40.335074Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:merge_copies: close time.busy=1.83ms time.idle=1.51µs
2026-09-22T14:56:40.375817Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:eliminate: eliminated padding clears using kernel access contracts removed=0
2026-09-22T14:56:40.377517Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:eliminate: close time.busy=42.4ms time.idle=2.78µs
2026-09-22T14:56:40.430692Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:place_provisional_storage: close time.busy=52.5ms time.idle=3.59µs
2026-09-22T14:56:40.935109Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=1 transfers=27312 maximum_identical_destinations=1
2026-09-22T14:56:40.935144Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=1 initial_horizon=15233 selected_horizon=15233 endpoint_lower_bound=13176 lower_bound_gap=2057 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:56:40.949804Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=4 transfers=2304 maximum_identical_destinations=1
2026-09-22T14:56:40.949831Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=4 initial_horizon=21139 selected_horizon=21125 endpoint_lower_bound=20736 lower_bound_gap=389 selected_kind="critical-neighborhood" neighborhood_improvements=1
2026-09-22T14:56:42.469806Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=209304
2026-09-22T14:56:43.385603Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=19852
2026-09-22T14:56:44.563663Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=384480
2026-09-22T14:56:46.171008Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=209304 elapsed_ms=3701 success=true
2026-09-22T14:56:46.410109Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: compared ordinary and paired exchange schedules phase=3 candidates=41292 ordinary_horizon=17903 paired_horizon=16538 use_paired=true
2026-09-22T14:56:46.514029Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=3 transfers=43740 maximum_identical_destinations=1
2026-09-22T14:56:46.514052Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=3 initial_horizon=16538 selected_horizon=16538 endpoint_lower_bound=9216 lower_bound_gap=7322 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:56:49.900736Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=6 transfers=209304 maximum_identical_destinations=1
2026-09-22T14:56:49.900769Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=6 initial_horizon=1663 selected_horizon=1663 endpoint_lower_bound=1296 lower_bound_gap=367 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:56:50.608329Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=19852 elapsed_ms=7222 success=true
2026-09-22T14:56:50.689445Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=384480 elapsed_ms=6125 success=true
2026-09-22T14:56:55.278803Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=766680
2026-09-22T14:56:57.293029Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=19852
2026-09-22T14:56:57.446494Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=5 transfers=384480 maximum_identical_destinations=1
2026-09-22T14:56:57.446522Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=5 initial_horizon=19831 selected_horizon=19831 endpoint_lower_bound=18144 lower_bound_gap=1687 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:57:03.966871Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=19852 elapsed_ms=6673 success=true
2026-09-22T14:57:05.882142Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: compared ordinary and paired exchange schedules phase=0 candidates=7412 ordinary_horizon=50443 paired_horizon=50750 use_paired=false
2026-09-22T14:57:06.066719Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=0 transfers=19852 maximum_identical_destinations=1
2026-09-22T14:57:06.066748Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=0 initial_horizon=50443 selected_horizon=50443 endpoint_lower_bound=47328 lower_bound_gap=3115 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:57:17.044947Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=766680 elapsed_ms=21766 success=true
2026-09-22T14:57:34.611453Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=2 transfers=766680 maximum_identical_destinations=1
2026-09-22T14:57:34.611495Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=2 initial_horizon=6715 selected_horizon=6715 endpoint_lower_bound=4392 lower_bound_gap=2323 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:57:36.086578Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: close time.busy=55.7s time.idle=9.05µs
2026-09-22T14:57:36.139479Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:plan_kernels: close time.busy=52.8ms time.idle=3.45µs
2026-09-22T14:57:36.141647Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: close time.busy=2.14ms time.idle=1.39µs
2026-09-22T14:57:36.141848Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:link_runtime: close time.busy=135µs time.idle=735ns
2026-09-22T14:57:36.500372Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: exchange table and repeat patch storage exchange_table_bytes=23956 maximum_uncompressed_repeat_patch_bytes=0 maximum_elided_arithmetic_patch_bytes=0
2026-09-22T14:57:37.039901Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:size_tile_code: close time.busy=60.6ms time.idle=4.09µs
2026-09-22T14:57:37.039929Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: placing generated tile programs linked_end=319744 host_code_base=376832 host_code_bytes=748 generated_code_bytes=1232 exchange_table_bytes=23956 executable_ranges=[(377648, 524288)]
2026-09-22T14:57:37.039960Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319744 profile_bytes=96 exchange_table_bytes=23956 host_code_bytes=748 host_data_bytes=528 generated_code_bytes=1232 code_address=377648 available_ranges=[(327680, 360448), (362736, 376832), (425984, 949168)]
2026-09-22T14:57:37.111720Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:place_storage: close time.busy=59.4ms time.idle=3.36µs
2026-09-22T14:57:37.416997Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=4
2026-09-22T14:57:37.429132Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=4 transfers=2304 maximum_identical_destinations=1
2026-09-22T14:57:37.429150Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=4 initial_horizon=21125 selected_horizon=21125 endpoint_lower_bound=20736 lower_bound_gap=389 selected_kind="reused" neighborhood_improvements=0
2026-09-22T14:57:37.461975Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=1
2026-09-22T14:57:37.470296Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=1 transfers=27312 maximum_identical_destinations=1
2026-09-22T14:57:37.470310Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=1 initial_horizon=15233 selected_horizon=15233 endpoint_lower_bound=13176 lower_bound_gap=2057 selected_kind="reused" neighborhood_improvements=0
2026-09-22T14:57:38.627050Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=3
2026-09-22T14:57:38.669298Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=3 transfers=43740 maximum_identical_destinations=1
2026-09-22T14:57:38.669320Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=3 initial_horizon=16538 selected_horizon=16538 endpoint_lower_bound=9216 lower_bound_gap=7322 selected_kind="reused" neighborhood_improvements=0
2026-09-22T14:57:43.635302Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=209304
2026-09-22T14:57:45.322646Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=5
2026-09-22T14:57:45.457832Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=5 transfers=384480 maximum_identical_destinations=1
2026-09-22T14:57:45.457867Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=5 initial_horizon=19831 selected_horizon=19831 endpoint_lower_bound=18144 lower_bound_gap=1687 selected_kind="reused" neighborhood_improvements=0
2026-09-22T14:57:47.155182Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: finished incremental instruction-alignment validation transfers=209304 elapsed_ms=3519 success=true
2026-09-22T14:57:50.815620Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=6 transfers=209304 maximum_identical_destinations=1
2026-09-22T14:57:50.815664Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=6 initial_horizon=1663 selected_horizon=1663 endpoint_lower_bound=1296 lower_bound_gap=367 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-22T14:57:57.101866Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=0
2026-09-22T14:57:57.232310Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=0 transfers=19852 maximum_identical_destinations=1
2026-09-22T14:57:57.232334Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=0 initial_horizon=50443 selected_horizon=50443 endpoint_lower_bound=47328 lower_bound_gap=3115 selected_kind="reused" neighborhood_improvements=0
2026-09-22T14:57:57.570983Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=2
2026-09-22T14:57:57.852275Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=2 transfers=766680 maximum_identical_destinations=1
2026-09-22T14:57:57.852304Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=2 initial_horizon=6715 selected_horizon=6715 endpoint_lower_bound=4392 lower_bound_gap=2323 selected_kind="reused" neighborhood_improvements=0
2026-09-22T14:57:59.502177Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: close time.busy=22.4s time.idle=4.74µs
2026-09-22T14:57:59.522879Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: costed final placed program final_cycles=749633 final_exchange=135748
2026-09-22T14:58:00.074688Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:prepare_tile_code: close time.busy=114ms time.idle=6.34µs
2026-09-22T14:58:00.104064Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:emit_tile_code: close time.busy=29.3ms time.idle=2.79µs
2026-09-22T14:58:00.311437Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:build_tile_images: close time.busy=200ms time.idle=3.18µs
2026-09-22T14:58:00.569919Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: wrote exact tile placement profile path=artifacts/planner-memory-20260922/fp8-b4-n1-384k/memory/placement-1171362.html
2026-09-22T14:58:00.669333Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: close time.busy=291s time.idle=4.98µs
2026-09-22T14:58:00.675029Z  INFO writing application package tiles=1472
2026-09-22T14:58:00.847844Z  INFO application package written
2026-09-22T14:58:00.849062Z  INFO reading application package
2026-09-22T14:58:01.085413Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-22T14:58:01.093500Z  INFO waiting for hardware ownership path=/tmp/ipu-stack-device.lock
2026-09-22T14:58:01.803588Z  INFO acquired hardware ownership path=/tmp/ipu-stack-device.lock
2026-09-22T14:58:01.803621Z  INFO opening IPU device path="/dev/ipu0"
2026-09-22T14:58:01.803680Z  INFO IPU device attached path=/dev/ipu0
2026-09-22T14:58:01.803685Z  INFO initializing IPU device
2026-09-22T14:58:01.811647Z  INFO IPU device initialized
2026-09-22T14:58:01.811653Z  INFO replaying device configuration register_writes=2003
2026-09-22T14:58:01.817907Z  INFO loading application tile_count=1472 final_mark=23
2026-09-22T14:58:02.008581Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-22T14:58:02.915148Z  INFO prepared reference inputs elapsed_ms=903
2026-09-22T14:58:03.391438Z  INFO configured host reference BLAS threads=4
referencePrecision=device
2026-09-22T14:58:04.783759Z  INFO evaluated host reference elapsed_ms=1868
2026-09-22T14:58:04.785997Z  INFO starting host exchange session startup_mark=23
2026-09-22T14:58:04.786011Z  INFO attaching host exchange pages pages=2
2026-09-22T14:58:04.786798Z  INFO host exchange session started
2026-09-22T14:58:04.786862Z  INFO invoking host exchange call call="initialize" command=0 phases=4
2026-09-22T14:58:04.790994Z  INFO invoking host exchange call call="run" command=0 phases=15
referenceInference=1 numericalTest=PASS
profile=artifacts/planner-memory-20260922/fp8-b4-n1-384k/profile.capnp tiles=1472 samples=27117
workload=siglip-mlp-f16-b4-t729-d1152-h4304-n1 benchmark=siglip-mlp-f16 batch=4 tokens=729 rows=2916 dimension=1152 hiddenDimension=4304 blocks=1 biases=false activeTiles=1458 executionTiles=1472 inputBytes=3359232 weightBytes=9934848 cycles=577326 minimumTileCycles=569988 deviceMicroseconds=384.884 effectiveGemmTflops=150.260 peakTflops=282.624 efficiencyPercent=53.17 maximumAbsoluteError=0.006104
package=artifacts/planner-memory-20260922/fp8-b4-n1-384k/model.ipuexe tiles=1472 hardwareTest=PASS