2026-09-07T14:49:13.516854Z  INFO retained operator-plan finalist finalist=0 values=9 operations=6 estimated_cycles=247014 estimated_exchange_cycles=106200 exchange_row_bytes=23240 peak_standard_bytes=91336 peak_interleaved_bytes=93184 peak_total_bytes=164424 maximum_standard_allocation_bytes=57408 active_tile_counts={1440} padding_group_counts={} conversion_sources=[Some(OperationId(0)), Some(OperationId(2)), Some(OperationId(2))]
2026-09-07T14:49:13.520484Z  INFO retained operator-plan finalist finalist=1 values=9 operations=6 estimated_cycles=251306 estimated_exchange_cycles=93560 exchange_row_bytes=20396 peak_standard_bytes=109228 peak_interleaved_bytes=79360 peak_total_bytes=143084 maximum_standard_allocation_bytes=56064 active_tile_counts={1440, 1470} padding_group_counts={} conversion_sources=[Some(OperationId(0)), Some(OperationId(2)), Some(OperationId(2))]
2026-09-07T14:49:13.523688Z  INFO retained operator-plan finalist finalist=2 values=9 operations=6 estimated_cycles=252238 estimated_exchange_cycles=112600 exchange_row_bytes=24680 peak_standard_bytes=92776 peak_interleaved_bytes=110080 peak_total_bytes=161768 maximum_standard_allocation_bytes=55936 active_tile_counts={1440, 1458} padding_group_counts={} conversion_sources=[Some(OperationId(0)), Some(OperationId(2)), Some(OperationId(2))]
2026-09-07T14:49:13.526245Z  INFO retained operator-plan finalist finalist=3 values=9 operations=6 estimated_cycles=252486 estimated_exchange_cycles=102840 exchange_row_bytes=22484 peak_standard_bytes=90580 peak_interleaved_bytes=79360 peak_total_bytes=145172 maximum_standard_allocation_bytes=55936 active_tile_counts={1440} padding_group_counts={} conversion_sources=[Some(OperationId(0)), Some(OperationId(2)), Some(OperationId(2))]
2026-09-07T14:49:13.598575Z  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=9590 success=true
2026-09-07T14:49:17.029223Z  INFO screened tile mappings by exchange resource load candidates=96 baseline_cycles=39384 candidate_cycles=39384 baseline_pressure=36899 candidate_pressure=36899
2026-09-07T14:49:17.048738Z  INFO modelled expanded operator plan finalist=0 modelled_cycles=181216 estimated_cycles=217480 estimated_exchange_cycles=78648
2026-09-07T14:49:17.554926Z  INFO screened tile mappings by exchange resource load candidates=117 baseline_cycles=38496 candidate_cycles=38496 baseline_pressure=31154 candidate_pressure=31154
2026-09-07T14:49:17.566904Z  INFO modelled expanded operator plan finalist=1 modelled_cycles=195972 estimated_cycles=222948 estimated_exchange_cycles=68472
2026-09-07T14:49:18.438260Z  INFO screened tile mappings by exchange resource load candidates=102 baseline_cycles=42960 candidate_cycles=42960 baseline_pressure=34154 candidate_pressure=34154
2026-09-07T14:49:18.446719Z  INFO modelled expanded operator plan finalist=3 modelled_cycles=190130 estimated_cycles=216482 estimated_exchange_cycles=72312
2026-09-07T14:49:19.039261Z  INFO screened tile mappings by exchange resource load candidates=104 baseline_cycles=46464 candidate_cycles=46464 baseline_pressure=38020 candidate_pressure=38020
2026-09-07T14:49:19.051365Z  INFO modelled expanded operator plan finalist=2 modelled_cycles=183204 estimated_cycles=214020 estimated_exchange_cycles=80280
2026-09-07T14:49:20.863586Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: compared ordinary and paired exchange schedules phase=2 candidates=4164 ordinary_horizon=8966 paired_horizon=6795 use_paired=true
2026-09-07T14:49:20.900333Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: prepared large physical exchange phase phase=2 transfers=4842 maximum_identical_destinations=1
2026-09-07T14:49:20.900380Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: optimized physical exchange schedule phase=2 initial_horizon=6795 selected_horizon=6795 endpoint_lower_bound=5544 lower_bound_gap=1251 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-07T14:49:20.912191Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: finished large physical exchange phase phase=2 tile=1014 row_words=137 horizon=6795
2026-09-07T14:49:29.700761Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: prepared large physical exchange phase phase=4 transfers=74200 maximum_identical_destinations=1
2026-09-07T14:49:29.700887Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: optimized physical exchange schedule phase=4 initial_horizon=18482 selected_horizon=18482 endpoint_lower_bound=13680 lower_bound_gap=4802 selected_kind="full-duplex" neighborhood_improvements=0
2026-09-07T14:49:29.779288Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: finished large physical exchange phase phase=4 tile=716 row_words=406 horizon=18482
2026-09-07T14:49:34.794400Z  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=57928 maximum_identical_destinations=1
2026-09-07T14:49:34.794467Z  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=17325 selected_horizon=17035 endpoint_lower_bound=13248 lower_bound_gap=3787 selected_kind="critical-neighborhood" neighborhood_improvements=1
2026-09-07T14:49:34.834633Z  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=1074 row_words=270 horizon=17035
2026-09-07T14:49:34.865711Z  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=217480 analytical_exchange_cycles=78648 scheduled_exchange_cycles=52505 refined_cycles=191337
2026-09-07T14:49:35.039611Z  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
2026-09-07T14:49:35.039651Z  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=21440 success=true
2026-09-07T14:49:35.039669Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: lowered graph for package construction logical_shards=29268 exchange_phases=5
2026-09-07T14:49:35.040005Z  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
2026-09-07T14:49:35.041892Z  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=1 success=true
2026-09-07T14:49:35.042188Z  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_k96_c128_r182_r184" target=ipu21
2026-09-07T14:49:35.101727Z  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/c22220b5084f7c59bc46660049c9e2a0c84694631acf7780c9f257dec3be3b50/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/c22220b5084f7c59bc46660049c9e2a0c84694631acf7780c9f257dec3be3b50/kernel.json
2026-09-07T14:49:35.102930Z  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=61 success=true
2026-09-07T14:49:35.103946Z  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
2026-09-07T14:49:35.423684Z  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
2026-09-07T14:49:35.423741Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: allocated package support memory linked_end=318376 profile_bytes=96 exchange_table_bytes=3052 host_code_bytes=1116 generated_code_bytes=1088 code_address=319560 standard_ranges=[(409600, 524288)]
2026-09-07T14:49:35.468808Z  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=45 success=true
2026-09-07T14:49:35.504857Z  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
2026-09-07T14:49:36.055762Z  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
2026-09-07T14:49:36.099092Z  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=57928 maximum_identical_destinations=1
2026-09-07T14:49:36.099144Z  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=17035 selected_horizon=17035 endpoint_lower_bound=13248 lower_bound_gap=3787 selected_kind="reused" neighborhood_improvements=0
2026-09-07T14:49:36.166388Z  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=1074 row_words=270 horizon=17035
2026-09-07T14:49:36.329778Z  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
2026-09-07T14:49:36.342033Z  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=4842 maximum_identical_destinations=1
2026-09-07T14:49:36.342071Z  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=6795 selected_horizon=6795 endpoint_lower_bound=5544 lower_bound_gap=1251 selected_kind="reused" neighborhood_improvements=0
2026-09-07T14:49:36.353163Z  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=1014 row_words=137 horizon=6795
2026-09-07T14:49:38.512178Z  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=4
2026-09-07T14:49:38.580461Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: prepared large physical exchange phase phase=4 transfers=74200 maximum_identical_destinations=1
2026-09-07T14:49:38.580505Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: optimized physical exchange schedule phase=4 initial_horizon=18482 selected_horizon=18482 endpoint_lower_bound=13680 lower_bound_gap=4802 selected_kind="reused" neighborhood_improvements=0
2026-09-07T14:49:38.656453Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: finished large physical exchange phase phase=4 tile=716 row_words=406 horizon=18482
2026-09-07T14:49:38.697189Z  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
2026-09-07T14:49:38.699044Z  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=3230 success=true
2026-09-07T14:49:39.074930Z  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=375 success=true
2026-09-07T14:49:39.084605Z  INFO ipu_codegen.package.build{tile_count=1472 operations=3}: costed final placed program final_cycles=191337 final_exchange=52505
2026-09-07T14:49:39.248400Z  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=13 success=true
2026-09-07T14:49:39.263051Z  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=14 success=true
2026-09-07T14:49:39.400598Z  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=132 success=true
2026-09-07T14:49:39.476399Z  INFO writing application package tiles=1472
2026-09-07T14:49:39.525933Z  INFO application package written
2026-09-07T14:49:39.527587Z  INFO reading application package
2026-09-07T14:49:39.640451Z  INFO application package read tiles=1472 compiler=0.1.0
2026-09-07T14:49:39.642368Z  INFO waiting for hardware ownership path=artifacts/layout-sweep/device.lock
2026-09-07T14:49:39.642390Z  INFO acquired hardware ownership path=artifacts/layout-sweep/device.lock
2026-09-07T14:49:39.642397Z  INFO opening IPU device path="/dev/ipu0"
2026-09-07T14:49:39.642438Z  INFO IPU device attached path=/dev/ipu0
2026-09-07T14:49:39.642443Z  INFO initializing IPU device
2026-09-07T14:49:39.650745Z  INFO IPU device initialized
2026-09-07T14:49:39.650791Z  INFO replaying device configuration register_writes=2003
2026-09-07T14:49:39.658083Z  INFO loading application tile_count=1472 final_mark=23
2026-09-07T14:49:39.906081Z  INFO application loaded and initial synchronization completed tile_count=1472
2026-09-07T14:49:42.279960Z  INFO starting host exchange session startup_mark=23
2026-09-07T14:49:42.280029Z  INFO attaching host exchange pages pages=2
2026-09-07T14:49:42.280705Z  INFO host exchange session started
2026-09-07T14:49:42.280735Z  INFO invoking streaming host exchange call call="initialize" command=0 phases=8
2026-09-07T14:49:42.284172Z  INFO invoking streaming host exchange call call="run" command=0 phases=25
profile=artifacts/fp8-fast/mlp/profile.ipuprof tiles=1472 samples=23324
referenceMaximumAbsoluteError=0.026241 numericalTest=PASS
package=artifacts/fp8-fast/mlp/model.ipuexe tiles=1472 hardwareTest=PASS