2026-09-07T16:21:41.761176Z  INFO retained operator-plan finalist finalist=0 values=8 operations=5 estimated_cycles=169834 estimated_exchange_cycles=78240 exchange_row_bytes=17080 peak_standard_bytes=213688 peak_interleaved_bytes=79360 peak_total_bytes=264376 maximum_standard_allocation_bytes=196608 active_tile_counts={1345, 1440} padding_group_counts={} conversion_sources=[Some(OperationId(2)), Some(OperationId(2))]
2026-09-07T16:21:41.769622Z  INFO retained operator-plan finalist finalist=1 values=8 operations=5 estimated_cycles=175368 estimated_exchange_cycles=84120 exchange_row_bytes=18272 peak_standard_bytes=86368 peak_interleaved_bytes=79360 peak_total_bytes=144224 maximum_standard_allocation_bytes=65536 active_tile_counts={1440, 1472} padding_group_counts={} conversion_sources=[Some(OperationId(2)), Some(OperationId(2))]
2026-09-07T16:21:41.773166Z  INFO retained operator-plan finalist finalist=2 values=8 operations=5 estimated_cycles=176364 estimated_exchange_cycles=81720 exchange_row_bytes=17732 peak_standard_bytes=116036 peak_interleaved_bytes=79360 peak_total_bytes=181060 maximum_standard_allocation_bytes=98304 active_tile_counts={1440, 1470} padding_group_counts={} conversion_sources=[Some(OperationId(2)), Some(OperationId(2))]
2026-09-07T16:21:41.776387Z  INFO retained operator-plan finalist finalist=3 values=8 operations=5 estimated_cycles=186884 estimated_exchange_cycles=92320 exchange_row_bytes=20248 peak_standard_bytes=216856 peak_interleaved_bytes=79360 peak_total_bytes=267544 maximum_standard_allocation_bytes=196608 active_tile_counts={1345, 1440} padding_group_counts={} conversion_sources=[Some(OperationId(2)), Some(OperationId(2))]
2026-09-07T16:21:41.894694Z  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=10304 success=true
2026-09-07T16:21:43.954903Z  INFO screened tile mappings by exchange resource load candidates=82 baseline_cycles=29728 candidate_cycles=29728 baseline_pressure=25015 candidate_pressure=25015
2026-09-07T16:21:43.965590Z  INFO modelled expanded operator plan finalist=0 modelled_cycles=121178 estimated_cycles=143034 estimated_exchange_cycles=53984
2026-09-07T16:21:44.435612Z  INFO screened tile mappings by exchange resource load candidates=72 baseline_cycles=35056 candidate_cycles=35056 baseline_pressure=24829 candidate_pressure=24829
2026-09-07T16:21:44.441705Z  INFO modelled expanded operator plan finalist=3 modelled_cycles=132760 estimated_cycles=254672 estimated_exchange_cycles=159368
2026-09-07T16:21:45.566950Z  INFO screened tile mappings by exchange resource load candidates=92 baseline_cycles=33000 candidate_cycles=33000 baseline_pressure=26982 candidate_pressure=26982
2026-09-07T16:21:45.579389Z  INFO modelled expanded operator plan finalist=1 modelled_cycles=125399 estimated_cycles=148271 estimated_exchange_cycles=58872
2026-09-07T16:21:45.934097Z  INFO screened tile mappings by exchange resource load candidates=116 baseline_cycles=32640 candidate_cycles=32640 baseline_pressure=25873 candidate_pressure=25873
2026-09-07T16:21:45.944742Z  INFO modelled expanded operator plan finalist=2 modelled_cycles=127014 estimated_cycles=148326 estimated_exchange_cycles=56952
2026-09-07T16:21:46.082785Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: prepared large physical exchange phase phase=0 transfers=5380 maximum_identical_destinations=1
2026-09-07T16:21:46.082850Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: optimized physical exchange schedule phase=0 initial_horizon=7141 selected_horizon=4841 endpoint_lower_bound=4672 lower_bound_gap=169 selected_kind="matching-waves" neighborhood_improvements=0
2026-09-07T16:21:46.086499Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: finished large physical exchange phase phase=0 tile=337 row_words=30 horizon=4841
2026-09-07T16:21:46.310354Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: compared ordinary and paired exchange schedules phase=1 candidates=2119 ordinary_horizon=9165 paired_horizon=8052 use_paired=true
2026-09-07T16:21:46.322717Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: prepared large physical exchange phase phase=1 transfers=2152 maximum_identical_destinations=1
2026-09-07T16:21:46.322762Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: optimized physical exchange schedule phase=1 initial_horizon=8052 selected_horizon=8052 endpoint_lower_bound=7920 lower_bound_gap=132 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-07T16:21:46.328104Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: finished large physical exchange phase phase=1 tile=111 row_words=61 horizon=8052
2026-09-07T16:21:54.955466Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: prepared large physical exchange phase phase=3 transfers=74200 maximum_identical_destinations=1
2026-09-07T16:21:54.955566Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: optimized physical exchange schedule phase=3 initial_horizon=18482 selected_horizon=18482 endpoint_lower_bound=13680 lower_bound_gap=4802 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-07T16:21:55.030658Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: finished large physical exchange phase phase=3 tile=716 row_words=406 horizon=18482
2026-09-07T16:21:55.082860Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: scheduled operator-plan finalist finalist=0 mapped=false analytical_cycles=143034 analytical_exchange_cycles=53984 scheduled_exchange_cycles=37363 refined_cycles=126413
2026-09-07T16:21:55.219602Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected physically scheduled operator-plan finalist selected=0 mapped=false
2026-09-07T16:21:55.219666Z  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=13324 success=true
2026-09-07T16:21:55.219686Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=26073 exchange_phases=4
2026-09-07T16:21:55.220105Z  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
2026-09-07T16:21:55.224483Z  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=4 success=true
2026-09-07T16:21:55.225014Z  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_f8_interleaved_k256_c16_r730_r730" target=ipu21
2026-09-07T16:21:55.292932Z  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/d715656e7a2021d47c39107a208918dce1375da213a4ea60acb26faaac989978/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/d715656e7a2021d47c39107a208918dce1375da213a4ea60acb26faaac989978/kernel.json
2026-09-07T16:21:55.293937Z  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=69 success=true
2026-09-07T16:21:55.295076Z  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
2026-09-07T16:21:56.205570Z  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=9 success=true
2026-09-07T16:21:56.205652Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=317644 profile_bytes=88 exchange_table_bytes=1940 host_code_bytes=3544 generated_code_bytes=1508 code_address=321256 standard_ranges=[(409600, 524288)]
2026-09-07T16:21:56.253674Z  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=47 success=true
2026-09-07T16:21:56.273577Z  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=2
2026-09-07T16:21:56.307139Z  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
2026-09-07T16:21:56.310148Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: prepared large physical exchange phase phase=0 transfers=5380 maximum_identical_destinations=1
2026-09-07T16:21:56.310181Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: optimized physical exchange schedule phase=0 initial_horizon=4841 selected_horizon=4841 endpoint_lower_bound=4672 lower_bound_gap=169 selected_kind="reused" neighborhood_improvements=0
2026-09-07T16:21:56.313999Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: finished large physical exchange phase phase=0 tile=337 row_words=30 horizon=4841
2026-09-07T16:21:56.388318Z  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
2026-09-07T16:21:56.392718Z  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=2152 maximum_identical_destinations=1
2026-09-07T16:21:56.392782Z  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=8052 selected_horizon=8052 endpoint_lower_bound=7920 lower_bound_gap=132 selected_kind="reused" neighborhood_improvements=0
2026-09-07T16:21:56.403540Z  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=111 row_words=61 horizon=8052
2026-09-07T16:21:58.626513Z  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=3
2026-09-07T16:21:58.660281Z  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=74200 maximum_identical_destinations=1
2026-09-07T16:21:58.660314Z  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=18482 selected_horizon=18482 endpoint_lower_bound=13680 lower_bound_gap=4802 selected_kind="reused" neighborhood_improvements=0
2026-09-07T16:21:58.711653Z  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=716 row_words=406 horizon=18482
2026-09-07T16:21:58.735047Z  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=2481 success=true
2026-09-07T16:21:58.980159Z  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=245 success=true
2026-09-07T16:21:58.983432Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: costed final placed program final_cycles=126413 final_exchange=37363
2026-09-07T16:21:59.452119Z  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=8 success=true
2026-09-07T16:21:59.464412Z  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=12 success=true
2026-09-07T16:21:59.595980Z  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=125 success=true
2026-09-07T16:21:59.694914Z  INFO writing application package tiles=1472
2026-09-07T16:21:59.746172Z  INFO application package written
2026-09-07T16:21:59.746896Z  INFO reading application package
2026-09-07T16:21:59.851648Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-07T16:21:59.852015Z  INFO waiting for hardware ownership path=artifacts/layout-sweep/device.lock
2026-09-07T16:21:59.852030Z  INFO acquired hardware ownership path=artifacts/layout-sweep/device.lock
2026-09-07T16:21:59.852037Z  INFO opening IPU device path="/dev/ipu0"
2026-09-07T16:21:59.852078Z  INFO IPU device attached path=/dev/ipu0
2026-09-07T16:21:59.852084Z  INFO initializing IPU device
2026-09-07T16:21:59.860196Z  INFO IPU device initialized
2026-09-07T16:21:59.860230Z  INFO replaying device configuration register_writes=2003
2026-09-07T16:21:59.872249Z  INFO loading application tile_count=1472 final_mark=23
2026-09-07T16:22:00.117610Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-07T16:22:08.824885Z  INFO starting host exchange session startup_mark=23
2026-09-07T16:22:08.824961Z  INFO attaching host exchange pages pages=2
2026-09-07T16:22:08.825939Z  INFO host exchange session started
2026-09-07T16:22:08.825995Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=10
2026-09-07T16:22:08.832356Z  INFO invoking streaming host exchange call call="run" command=0 phases=99
profile=artifacts/fp8-before-fanout/mlp/profile.ipuprof tiles=1472 samples=21118
referenceMaximumAbsoluteError=0.006378 numericalTest=PASS
package=artifacts/fp8-before-fanout/mlp/model.ipuexe tiles=1472 hardwareTest=PASS