2026-09-07T14:03:49.007053Z INFO retained operator-plan finalist finalist=0 values=11 operations=8 estimated_cycles=116890 estimated_exchange_cycles=24120 exchange_row_bytes=4772 peak_standard_bytes=36004 peak_interleaved_bytes=37120 peak_total_bytes=69796 maximum_standard_allocation_bytes=21760 active_tile_counts={60, 64} padding_group_counts={} conversion_sources=[Some(OperationId(0)), Some(OperationId(0)), Some(OperationId(2)), Some(OperationId(2)), Some(OperationId(2))] 2026-09-07T14:03:49.007153Z INFO retained operator-plan finalist finalist=1 values=11 operations=8 estimated_cycles=127982 estimated_exchange_cycles=23480 exchange_row_bytes=4628 peak_standard_bytes=31764 peak_interleaved_bytes=37120 peak_total_bytes=65556 maximum_standard_allocation_bytes=21760 active_tile_counts={64} padding_group_counts={} conversion_sources=[Some(OperationId(0)), Some(OperationId(0)), Some(OperationId(2)), Some(OperationId(2)), Some(OperationId(2))] 2026-09-07T14:03:49.007179Z INFO retained operator-plan finalist finalist=2 values=11 operations=8 estimated_cycles=129016 estimated_exchange_cycles=22360 exchange_row_bytes=4376 peak_standard_bytes=34072 peak_interleaved_bytes=37120 peak_total_bytes=67352 maximum_standard_allocation_bytes=21760 active_tile_counts={64} padding_group_counts={} conversion_sources=[Some(OperationId(0)), Some(OperationId(0)), Some(OperationId(2)), Some(OperationId(2)), Some(OperationId(2))] 2026-09-07T14:03:49.007198Z INFO retained operator-plan finalist finalist=3 values=11 operations=8 estimated_cycles=129386 estimated_exchange_cycles=18680 exchange_row_bytes=3548 peak_standard_bytes=43484 peak_interleaved_bytes=37120 peak_total_bytes=60892 maximum_standard_allocation_bytes=17408 active_tile_counts={60, 64} padding_group_counts={} conversion_sources=[Some(OperationId(0)), Some(OperationId(0)), Some(OperationId(2)), Some(OperationId(2)), Some(OperationId(2))] 2026-09-07T14:03:49.013339Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="lower_mid"}: package build phase finished phase="lower_mid" elapsed_ms=346 success=true 2026-09-07T14:03:49.062028Z INFO screened tile mappings by exchange resource load candidates=11 baseline_cycles=9312 candidate_cycles=9312 baseline_pressure=8094 candidate_pressure=8094 2026-09-07T14:03:49.062421Z INFO modelled expanded operator plan finalist=1 modelled_cycles=145968 estimated_cycles=215744 estimated_exchange_cycles=81488 2026-09-07T14:03:49.069480Z INFO screened tile mappings by exchange resource load candidates=11 baseline_cycles=8832 candidate_cycles=8832 baseline_pressure=7486 candidate_pressure=7486 2026-09-07T14:03:49.069707Z INFO modelled expanded operator plan finalist=2 modelled_cycles=80814 estimated_cycles=175390 estimated_exchange_cycles=105808 2026-09-07T14:03:49.070406Z INFO screened tile mappings by exchange resource load candidates=21 baseline_cycles=11968 candidate_cycles=11968 baseline_pressure=8257 candidate_pressure=8257 2026-09-07T14:03:49.070622Z INFO modelled expanded operator plan finalist=0 modelled_cycles=75916 estimated_cycles=130132 estimated_exchange_cycles=68584 2026-09-07T14:03:49.072713Z INFO screened tile mappings by exchange resource load candidates=22 baseline_cycles=8256 candidate_cycles=8256 baseline_pressure=6315 candidate_pressure=6315 2026-09-07T14:03:49.072986Z INFO modelled expanded operator plan finalist=3 modelled_cycles=134260 estimated_cycles=187836 estimated_exchange_cycles=64232 2026-09-07T14:03:50.582901Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: prepared large physical exchange phase phase=2 transfers=8262 maximum_identical_destinations=1 2026-09-07T14:03:50.582963Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: optimized physical exchange schedule phase=2 initial_horizon=5166 selected_horizon=5166 endpoint_lower_bound=2880 lower_bound_gap=2286 selected_kind="full-duplex" neighborhood_improvements=0 2026-09-07T14:03:50.587857Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: finished large physical exchange phase phase=2 tile=7 row_words=1047 horizon=5166 2026-09-07T14:03:50.591136Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: scheduled operator-plan finalist finalist=0 mapped=false analytical_cycles=130132 analytical_exchange_cycles=68584 scheduled_exchange_cycles=15380 refined_cycles=76928 2026-09-07T14:03:50.597754Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: selected physically scheduled operator-plan finalist selected=0 mapped=false 2026-09-07T14:03:50.597785Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="select_finalist"}: package build phase finished phase="select_finalist" elapsed_ms=1584 success=true 2026-09-07T14:03:50.597806Z INFO ipu_codegen.package.build{tile_count=64 operations=3}: lowered graph for package construction logical_shards=1458 exchange_phases=4 2026-09-07T14:03:50.598280Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="compile_runtime"}: package build phase finished phase="compile_runtime" elapsed_ms=0 success=true 2026-09-07T14:03:50.598548Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="plan_kernels"}: package build phase finished phase="plan_kernels" elapsed_ms=0 success=true 2026-09-07T14:03:50.598817Z INFO ipu_codegen.package.build{tile_count=64 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_k96_c64_r48_r56" target=ipu21 2026-09-07T14:03:50.667650Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="compile_kernels"}: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/7077cc16e264b96747821dbb22a4b4adb8c61b085b0dae6673bd55a5e5e14230/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/7077cc16e264b96747821dbb22a4b4adb8c61b085b0dae6673bd55a5e5e14230/kernel.json 2026-09-07T14:03:50.668198Z INFO ipu_codegen.package.build{tile_count=64 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_k32_c64_r128_r136" target=ipu21 2026-09-07T14:03:50.728669Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="compile_kernels"}: kernel artifact created object=/srv/home/.cache/ipu-stack/kernels/9094bdf5f6721326ee2075c6394df731ebae2f81d9b2aa5e4e24eb8134f47c55/kernel.o metadata=/srv/home/.cache/ipu-stack/kernels/9094bdf5f6721326ee2075c6394df731ebae2f81d9b2aa5e4e24eb8134f47c55/kernel.json 2026-09-07T14:03:50.729546Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="compile_kernels"}: package build phase finished phase="compile_kernels" elapsed_ms=130 success=true 2026-09-07T14:03:50.730065Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="link_runtime"}: package build phase finished phase="link_runtime" elapsed_ms=0 success=true 2026-09-07T14:03:50.763451Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="size_tile_code"}: package build phase finished phase="size_tile_code" elapsed_ms=3 success=true 2026-09-07T14:03:50.763545Z INFO ipu_codegen.package.build{tile_count=64 operations=3}: allocated package support memory linked_end=318356 profile_bytes=112 exchange_table_bytes=4376 host_code_bytes=640 generated_code_bytes=2584 code_address=319064 standard_ranges=[(409600, 524288)] 2026-09-07T14:03:50.767193Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="place_storage"}: package build phase finished phase="place_storage" elapsed_ms=3 success=true 2026-09-07T14:03:50.768386Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: reused exchange optimization after validating relocated rows phase=0 2026-09-07T14:03:50.769390Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: reused exchange optimization after validating relocated rows phase=3 2026-09-07T14:03:50.770187Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: reused exchange optimization after validating relocated rows phase=1 2026-09-07T14:03:52.016615Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: reused exchange optimization after validating relocated rows phase=2 2026-09-07T14:03:52.021895Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: prepared large physical exchange phase phase=2 transfers=8262 maximum_identical_destinations=1 2026-09-07T14:03:52.021927Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: optimized physical exchange schedule phase=2 initial_horizon=5166 selected_horizon=5166 endpoint_lower_bound=2880 lower_bound_gap=2286 selected_kind="reused" neighborhood_improvements=0 2026-09-07T14:03:52.029683Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: finished large physical exchange phase phase=2 tile=7 row_words=1047 horizon=5166 2026-09-07T14:03:52.032922Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="lower_exchanges"}: package build phase finished phase="lower_exchanges" elapsed_ms=1265 success=true 2026-09-07T14:03:52.058173Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="optimize_exchange_placement"}: package build phase finished phase="optimize_exchange_placement" elapsed_ms=25 success=true 2026-09-07T14:03:52.058356Z INFO ipu_codegen.package.build{tile_count=64 operations=3}: costed final placed program final_cycles=76928 final_exchange=15380 2026-09-07T14:03:52.076506Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="prepare_tile_code"}: package build phase finished phase="prepare_tile_code" elapsed_ms=4 success=true 2026-09-07T14:03:52.080890Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="emit_tile_code"}: package build phase finished phase="emit_tile_code" elapsed_ms=4 success=true 2026-09-07T14:03:52.187013Z INFO ipu_codegen.package.build{tile_count=64 operations=3}:ipu_codegen.package.phase{phase="build_tile_images"}: package build phase finished phase="build_tile_images" elapsed_ms=104 success=true 2026-09-07T14:03:52.193355Z INFO writing application package tiles=1472 2026-09-07T14:03:52.208323Z INFO application package written 2026-09-07T14:03:52.209031Z INFO reading application package 2026-09-07T14:03:52.234190Z INFO application package read tiles=1472 compiler=0.1.0 2026-09-07T14:03:52.234530Z INFO waiting for hardware ownership path=artifacts/layout-sweep/device.lock 2026-09-07T14:03:52.234545Z INFO acquired hardware ownership path=artifacts/layout-sweep/device.lock 2026-09-07T14:03:52.234552Z INFO opening IPU device path="/dev/ipu0" 2026-09-07T14:03:52.234592Z INFO IPU device attached path=/dev/ipu0 2026-09-07T14:03:52.234597Z INFO initializing IPU device 2026-09-07T14:03:52.242448Z INFO IPU device initialized 2026-09-07T14:03:52.242458Z INFO replaying device configuration register_writes=2003 2026-09-07T14:03:52.246866Z INFO loading application tile_count=1472 final_mark=23 2026-09-07T14:03:52.448737Z INFO application loaded and initial synchronization completed tile_count=1472 2026-09-07T14:03:52.545019Z INFO starting host exchange session startup_mark=23 2026-09-07T14:03:52.545067Z INFO attaching host exchange pages pages=2 2026-09-07T14:03:52.545739Z INFO host exchange session started 2026-09-07T14:03:52.545791Z INFO invoking streaming host exchange call call="initialize" command=0 phases=4 2026-09-07T14:03:52.546342Z INFO invoking streaming host exchange call call="run" command=0 phases=13 referenceMaximumAbsoluteError=0.025726 numericalTest=PASS package=artifacts/fp8/mlp-uneven/model.ipuexe tiles=1472 hardwareTest=PASS