2026-09-22T14:39:33.755097Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_runtime: close time.busy=284µs time.idle=8.22µs 2026-09-22T14:40:28.379937Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:construct_mid:gemm_candidates: close time.busy=54.6s time.idle=2.99µs 2026-09-22T14:41:39.700932Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:construct_mid:gemm_candidates: close time.busy=71.3s time.idle=3.16µs 2026-09-22T14:41:43.362774Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:construct_mid: close time.busy=130s time.idle=2.22µs 2026-09-22T14:41:44.230519Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:group_exchanges: close time.busy=206µs time.idle=3.02µs 2026-09-22T14:41:44.231737Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:merge_copies: close time.busy=1.03ms time.idle=810ns 2026-09-22T14:41:44.256440Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:eliminate: eliminated padding clears using kernel access contracts removed=0 2026-09-22T14:41:44.257512Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:eliminate: close time.busy=25.8ms time.idle=670ns 2026-09-22T14:41:44.294608Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:place_provisional_storage: close time.busy=36.6ms time.idle=815ns 2026-09-22T14:41:44.657993Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=4 transfers=1728 maximum_identical_destinations=1 2026-09-22T14:41:44.658015Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=4 initial_horizon=21814 selected_horizon=21802 endpoint_lower_bound=21504 lower_bound_gap=298 selected_kind="critical-neighborhood" neighborhood_improvements=1 2026-09-22T14:41:44.707710Z 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 2026-09-22T14:41:44.707725Z 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 2026-09-22T14:41:45.542193Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=58320 2026-09-22T14:41:45.565762Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=209448 2026-09-22T14:41:47.162934Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=19852 2026-09-22T14:41:47.361649Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=58320 elapsed_ms=1819 success=true 2026-09-22T14:41:47.790280Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=209448 elapsed_ms=2224 success=true 2026-09-22T14:41:49.527942Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=58320 2026-09-22T14:41:50.155349Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=6 transfers=209448 maximum_identical_destinations=1 2026-09-22T14:41:50.155369Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=6 initial_horizon=1672 selected_horizon=1672 endpoint_lower_bound=1344 lower_bound_gap=328 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-22T14:41:52.141536Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=19852 elapsed_ms=4978 success=true 2026-09-22T14:41:52.144182Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=58320 elapsed_ms=2616 success=true 2026-09-22T14:41:52.962882Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=685728 2026-09-22T14:41:53.304785Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: compared ordinary and paired exchange schedules phase=3 candidates=816 ordinary_horizon=14660 paired_horizon=14886 use_paired=false 2026-09-22T14:41:53.404888Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=3 transfers=58320 maximum_identical_destinations=1 2026-09-22T14:41:53.404907Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=3 initial_horizon=14660 selected_horizon=14660 endpoint_lower_bound=13664 lower_bound_gap=996 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-22T14:41:56.403827Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=19852 2026-09-22T14:41:57.685577Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=766680 2026-09-22T14:42:01.124814Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=19852 elapsed_ms=4720 success=true 2026-09-22T14:42:02.885537Z 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 2026-09-22T14:42:03.046878Z 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 2026-09-22T14:42:03.046892Z 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 2026-09-22T14:42:05.613495Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=685728 elapsed_ms=12650 success=true 2026-09-22T14:42:17.434100Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: finished incremental instruction-alignment validation transfers=766680 elapsed_ms=19748 success=true 2026-09-22T14:42:23.833691Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: prepared large physical exchange phase phase=5 transfers=685728 maximum_identical_destinations=1 2026-09-22T14:42:23.833716Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: optimized physical exchange schedule phase=5 initial_horizon=28630 selected_horizon=28630 endpoint_lower_bound=25536 lower_bound_gap=3094 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-22T14:42:40.356109Z 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 2026-09-22T14:42:40.356154Z 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 2026-09-22T14:42:42.475173Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:schedule_provisional_exchanges: close time.busy=58.2s time.idle=4.45µs 2026-09-22T14:42:42.515832Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:plan_kernels: close time.busy=40.5ms time.idle=2.53µs 2026-09-22T14:42:42.516156Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/rearrange_f8.cpp name="rearrange_row_major_to_amp_f8_o0_r61_p61_c384_p384_codelet" target="ipu21" 2026-09-22T14:42:42.828066Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/71906c1d87dca9066125cc0426d3674eae4c0c629f8629f1cdee2d5f66d9851f/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/71906c1d87dca9066125cc0426d3674eae4c0c629f8629f1cdee2d5f66d9851f/kernel.json 2026-09-22T14:42:42.828482Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/worker_call.S name="rearrange_row_major_to_amp_f8_o0_r61_p61_c384_p384_wrapper" target="ipu21" 2026-09-22T14:42:42.874816Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/c56b50fc1e1abf7f8e2aa623daf50b12b87ecc313346be4b424ce1f48df30853/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/c56b50fc1e1abf7f8e2aa623daf50b12b87ecc313346be4b424ce1f48df30853/kernel.json 2026-09-22T14:42:42.875145Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/rearrange_f8.cpp name="rearrange_row_major_to_amp_f8_o0_r60_p60_c384_p384_codelet" target="ipu21" 2026-09-22T14:42:43.179899Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/d8a417c2ef33aa6f64dca8af4db5fea9bf154ff54f914e90f6c515018bce2036/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/d8a417c2ef33aa6f64dca8af4db5fea9bf154ff54f914e90f6c515018bce2036/kernel.json 2026-09-22T14:42:43.180271Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/worker_call.S name="rearrange_row_major_to_amp_f8_o0_r60_p60_c384_p384_wrapper" target="ipu21" 2026-09-22T14:42:43.222955Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/b379f8d71da9c0ad54915e36883d7576287c6c4eeda7f0ccc052a42cdf983b5e/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/b379f8d71da9c0ad54915e36883d7576287c6c4eeda7f0ccc052a42cdf983b5e/kernel.json 2026-09-22T14:42:43.223950Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/gemm_f16_amp.S name="gemm_f8_interleaved_packed0_k384_c432_r61" target="ipu21" 2026-09-22T14:42:43.266926Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/79532092f6f84f50104e100312ebdf809562f6923dd45bd02d7468a1b4e8173e/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/79532092f6f84f50104e100312ebdf809562f6923dd45bd02d7468a1b4e8173e/kernel.json 2026-09-22T14:42:43.267449Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/gemm_f16_amp.S name="gemm_f8_interleaved_packed0_k384_c432_r60" target="ipu21" 2026-09-22T14:42:43.318949Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/ad8ca024738cc4e68f7436e7b16429a706d3a0515f60cf4f630350e6ff25410b/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/ad8ca024738cc4e68f7436e7b16429a706d3a0515f60cf4f630350e6ff25410b/kernel.json 2026-09-22T14:42:43.319109Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/reduce_add_f16.S name="reduce_add_f16" target="ipu21" 2026-09-22T14:42:43.362160Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/5047decebfdc7369d23b3f92f29d5c28e4afe04a40f0190c04eee25ab0797856/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/5047decebfdc7369d23b3f92f29d5c28e4afe04a40f0190c04eee25ab0797856/kernel.json 2026-09-22T14:42:43.362761Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/rearrange_amp_left_f16.S name="rearrange_row_major_to_amp_f16_o0_r122_p122_c224_p224" target="ipu21" 2026-09-22T14:42:43.407028Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/1b4322431a5052ce07925bdd90f1b0950c9aaa87bd270b49173c548ff2827c80/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/1b4322431a5052ce07925bdd90f1b0950c9aaa87bd270b49173c548ff2827c80/kernel.json 2026-09-22T14:42:43.407426Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/rearrange_amp_left_f16.S name="rearrange_row_major_to_amp_f16_o0_r122_p122_c48_p224" target="ipu21" 2026-09-22T14:42:43.450015Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/f670ee8a9b1591729bf827be67423ff4110b9ef98f36946367bff7a8a0d1b045/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/f670ee8a9b1591729bf827be67423ff4110b9ef98f36946367bff7a8a0d1b045/kernel.json 2026-09-22T14:42:43.450408Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/rearrange_amp_left_f16.S name="rearrange_row_major_to_amp_f16_o0_r121_p121_c224_p224" target="ipu21" 2026-09-22T14:42:43.493297Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/a84da0bd442f5184696a00622283e5122fe00f4f5a2c7151d722f5e304841ac2/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/a84da0bd442f5184696a00622283e5122fe00f4f5a2c7151d722f5e304841ac2/kernel.json 2026-09-22T14:42:43.493649Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/rearrange_amp_left_f16.S name="rearrange_row_major_to_amp_f16_o0_r121_p121_c48_p224" target="ipu21" 2026-09-22T14:42:43.537764Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/6c28492222398ec7f19776eca66d2db86be2e9012085709ef4d044449a40cfea/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/6c28492222398ec7f19776eca66d2db86be2e9012085709ef4d044449a40cfea/kernel.json 2026-09-22T14:42:43.538778Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/gemm_f16_amp.S name="gemm_f8_interleaved_packed0_k224_c384_r122" target="ipu21" 2026-09-22T14:42:43.583932Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/eed73cc4f89a19faeac950a5035ea23d338cc858d89313772c9a562a4a65942c/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/eed73cc4f89a19faeac950a5035ea23d338cc858d89313772c9a562a4a65942c/kernel.json 2026-09-22T14:42:43.584449Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: compiling kernel source=/srv/home/gc-sdk/ipu-stack/crates/ipu-tests/../../device/gemm_f16_amp.S name="gemm_f8_interleaved_packed0_k224_c384_r121" target="ipu21" 2026-09-22T14:42:43.629703Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/5d71e49ae09640ee2917aea5912ff40a5aeea79eaf25b705c3b7b798edad4fd4/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/5d71e49ae09640ee2917aea5912ff40a5aeea79eaf25b705c3b7b798edad4fd4/kernel.json 2026-09-22T14:42:43.630050Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:compile_kernels: close time.busy=1.11s time.idle=5.04µs 2026-09-22T14:42:43.630865Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:link_runtime: close time.busy=515µs time.idle=3.46µs 2026-09-22T14:42:44.055325Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: exchange table and repeat patch storage exchange_table_bytes=28444 maximum_uncompressed_repeat_patch_bytes=0 maximum_elided_arithmetic_patch_bytes=0 2026-09-22T14:42:44.608359Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:size_tile_code: close time.busy=16.4ms time.idle=4.43µs 2026-09-22T14:42:44.608383Z 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=1804 exchange_table_bytes=28444 executable_ranges=[(377648, 524288)] 2026-09-22T14:42:44.608417Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=319744 profile_bytes=192 exchange_table_bytes=28444 host_code_bytes=748 host_data_bytes=528 generated_code_bytes=1804 code_address=377648 available_ranges=[(327680, 360448), (362736, 376832), (425984, 949168)] 2026-09-22T14:42:44.688577Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:place_storage: close time.busy=67.0ms time.idle=2.83µs 2026-09-22T14:42:44.857771Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=4 2026-09-22T14:42:44.869090Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=4 transfers=1728 maximum_identical_destinations=1 2026-09-22T14:42:44.869109Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=4 initial_horizon=21802 selected_horizon=21802 endpoint_lower_bound=21504 lower_bound_gap=298 selected_kind="reused" neighborhood_improvements=0 2026-09-22T14:42:44.903468Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=1 2026-09-22T14:42:44.911910Z 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 2026-09-22T14:42:44.911921Z 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 2026-09-22T14:42:46.890400Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=3 2026-09-22T14:42:46.934041Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=3 transfers=58320 maximum_identical_destinations=1 2026-09-22T14:42:46.934061Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=3 initial_horizon=14660 selected_horizon=14660 endpoint_lower_bound=13664 lower_bound_gap=996 selected_kind="reused" neighborhood_improvements=0 2026-09-22T14:42:51.080716Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=209448 2026-09-22T14:42:51.431484Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=0 2026-09-22T14:42:51.580532Z 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 2026-09-22T14:42:51.580561Z 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 2026-09-22T14:42:54.531464Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: finished incremental instruction-alignment validation transfers=209448 elapsed_ms=3450 success=true 2026-09-22T14:42:57.953603Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=6 transfers=209448 maximum_identical_destinations=1 2026-09-22T14:42:57.953635Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=6 initial_horizon=1672 selected_horizon=1672 endpoint_lower_bound=1344 lower_bound_gap=328 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-22T14:43:12.927951Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: reused exchange optimization after validating relocated rows phase=2 2026-09-22T14:43:13.223450Z 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 2026-09-22T14:43:13.223483Z 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 2026-09-22T14:43:21.032285Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: retrying exchange schedule with incremental instruction-alignment validation transfers=685728 2026-09-22T14:43:37.751919Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: finished incremental instruction-alignment validation transfers=685728 elapsed_ms=16719 success=true 2026-09-22T14:43:52.197973Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: prepared large physical exchange phase phase=5 transfers=685728 maximum_identical_destinations=1 2026-09-22T14:43:52.197995Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: optimized physical exchange schedule phase=5 initial_horizon=28426 selected_horizon=28426 endpoint_lower_bound=25536 lower_bound_gap=2890 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-22T14:43:53.233407Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:lower_exchanges: close time.busy=68.5s time.idle=4.86µs 2026-09-22T14:43:53.237411Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: costed final placed program final_cycles=761840 final_exchange=143151 2026-09-22T14:43:53.758924Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:prepare_tile_code: close time.busy=32.6ms time.idle=4.15µs 2026-09-22T14:43:53.780555Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:emit_tile_code: close time.busy=21.6ms time.idle=1.57µs 2026-09-22T14:43:53.937421Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}:build_tile_images: close time.busy=151ms time.idle=1.83µs 2026-09-22T14:43:54.042164Z INFO ipu_codegen.package.build{tile_count=1472 operations=3}: close time.busy=260s time.idle=4.35µs 2026-09-22T14:43:54.045847Z INFO writing application package tiles=1472 2026-09-22T14:43:54.295982Z INFO application package written 2026-09-22T14:43:54.298020Z INFO reading application package 2026-09-22T14:43:54.549307Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-22T14:43:54.558236Z INFO waiting for hardware ownership path=/tmp/ipu-stack-device.lock 2026-09-22T14:43:54.558245Z INFO acquired hardware ownership path=/tmp/ipu-stack-device.lock 2026-09-22T14:43:54.558250Z INFO opening IPU device path="/dev/ipu0" 2026-09-22T14:43:54.558280Z INFO IPU device attached path=/dev/ipu0 2026-09-22T14:43:54.558282Z INFO initializing IPU device 2026-09-22T14:43:54.566742Z INFO IPU device initialized 2026-09-22T14:43:54.566764Z INFO replaying device configuration register_writes=2003 2026-09-22T14:43:54.573179Z INFO loading application tile_count=1472 final_mark=23 2026-09-22T14:43:54.767705Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-22T14:43:55.648216Z INFO prepared reference inputs elapsed_ms=877 2026-09-22T14:43:56.124883Z INFO configured host reference BLAS threads=4 referencePrecision=device 2026-09-22T14:43:57.550847Z INFO evaluated host reference elapsed_ms=1902 2026-09-22T14:43:57.553654Z INFO starting host exchange session startup_mark=23 2026-09-22T14:43:57.553664Z INFO attaching host exchange pages pages=2 2026-09-22T14:43:57.554337Z INFO host exchange session started 2026-09-22T14:43:57.554438Z INFO invoking host exchange call call="initialize" command=0 phases=4 2026-09-22T14:43:57.563019Z INFO invoking host exchange call call="run" command=0 phases=15 referenceInference=1 numericalTest=PASS profile=artifacts/packed-reduction/mlp.capnp tiles=1472 samples=28825 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=614700 minimumTileCycles=607116 deviceMicroseconds=409.800 effectiveGemmTflops=141.124 peakTflops=282.624 efficiencyPercent=49.93 maximumAbsoluteError=0.006470 package=artifacts/packed-reduction/mlp.ipuexe tiles=1472 hardwareTest=PASS