h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv

5 lines
160 KiB
CSV
Raw Normal View History

2026-08-25 22:32:48 +07:00
"ID","Process ID","Process Name","Host Name","Kernel Name","Context","Stream","Block Size","Grid Size","Device","CC","c2clink__enabled_mask","c2clink__present","derived__avg_thread_executed","derived__avg_thread_executed_true","derived__avg_thread_unexecuted_true","derived__derivative_avg_thread_executed_true","derived__l1tex__lsu_writeback_bytes_mem_lgds.sum.peak_sustained","derived__l1tex__lsu_writeback_bytes_mem_lgds.sum.per_second","derived__local_spilling_requests","derived__local_spilling_requests_pct","derived__lts__lts2xbar_bytes.sum.peak_sustained","derived__lts__lts2xbar_bytes.sum.per_second","derived__memory_l1_conflicts_shared_nway","derived__memory_l1_wavefronts_shared_excessive","derived__memory_l2_theoretical_sectors_global_excessive","derived__pct_occupancy_per_barrier_count","derived__pct_occupancy_per_block_size","derived__pct_occupancy_per_register_count","derived__pct_occupancy_per_shared_mem_size","derived__sm__sass_thread_inst_executed_op_dfma_pred_on_x2","derived__sm__sass_thread_inst_executed_op_ffma_pred_on_x2","derived__sm__sass_thread_inst_executed_op_hfma_pred_on_x4","derived__smsp__inst_executed_op_branch_pct","derived__smsp__sass_thread_inst_executed_op_dfma_pred_on_x2","derived__smsp__sass_thread_inst_executed_op_ffma_pred_on_x2","derived__smsp__sass_thread_inst_executed_op_hadd_pred_on_x2","derived__smsp__sass_thread_inst_executed_op_hfma_pred_on_x4","derived__smsp__sass_thread_inst_executed_op_hmul_pred_on_x2","derived_tempMetric0","derived_tempMetric1","derived_tempMetric2","derived_tempMetric3","derived_tempMetric4","derived_tempMetric5","device__attribute_architecture","device__attribute_async_engine_count","device__attribute_can_flush_remote_writes","device__attribute_can_map_host_memory","device__attribute_can_tex2d_gather","device__attribute_can_use_64_bit_stream_mem_ops","device__attribute_can_use_64_bit_stream_mem_ops_v1","device__attribute_can_use_host_pointer_for_registered_mem","device__attribute_can_use_stream_mem_ops_v1","device__attribute_can_use_stream_wait_value_nor","device__attribute_can_use_stream_wait_value_nor_v1","device__attribute_chip","device__attribute_clock_rate","device__attribute_cluster_launch","device__attribute_compute_capability_major","device__attribute_compute_capability_minor","device__attribute_compute_mode","device__attribute_compute_preemption_supported","device__attribute_concurrent_kernels","device__attribute_concurrent_managed_access","device__attribute_confidential_computing_mode","device__attribute_cooperative_launch","device__attribute_cooperative_multi_device_launch","device__attribute_deferred_mapping_cuda_array_supported","device__attribute_device_index","device__attribute_direct_managed_mem_access_from_host","device__attribute_display_name","device__attribute_dma_buf_supported","device__attribute_ecc_enabled","device__attribute_fb_bus_width","device__attribute_fbp_count","device__attribute_generic_compression_supported","device__attribute_global_l1_cache_supported","device__attribute_global_memory_bus_width","device__attribute_gpu_direct_rdma_flush_writes_options","device__attribute_gpu_direct_rdma_supported","device__attribute_gpu_direct_rdma_with_cuda_vmm_supported","device__attribute_gpu_direct_rdma_writes_ordering","device__attribute_gpu_overlap","device__attribute_gpu_pci_device_id","device__attribute_gpu_pci_ext_device_id","device__attribute_gpu_pci_ext_downstream_link_rate","device__attribute_gpu_pci_ext_downstream_link_width","device__attribute_gpu_pci_ext_gen","device__attribute_gpu_pci_ext_gpu_gen","device__attribute_gpu_pci_ext_gpu_link_rate","device__attribute_gpu_pci_ext_gpu_link_width","device__attribute_gpu_pci_revision_id","device__attribute_gpu_pci_sub_system_id","device__attribute_handle_type_fabric_supported","device__attribute_handle_type_posix_file_descriptor_supported","device__attribute_handle_type_win32_handle_supported","device__attribute_handle_type_win32_kmt_handle_supported","device__attribute_host_native_atomic_supported","device__attribute_host_numa_id","device__attribute_host_register_supported","device__attr
"","","","","","","","","","","","","","thread","thread","thread","thread","","","","%","Kbyte","Gbyte","","","byte","","%","%/register","%/Kbyte","thread","thread","thread","%","thread","thread","thread","thread","thread","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","request","%","request","%","","%","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","ms","ms","ms","ms","","","","","block","block","block","block","","","","","%","%","%","%","%","","%","inst","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","","","","","","%","%","%","%","%","%","%","%","%","%","%","%","","%","","%","","%","","%","%","%","%","%","%","%","%","%","%","%","%","%","cycle","%","","Mhz","%","%","%","%","%","%","%","%","Mbyte","%","Gbyte/s","byte","%","byte/s","byte","%","byte/s","Mbyte","%","Gbyte/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector","%","sector","%","sector","%","sector","%","sector/s","sector","%","sector/us","sector","%","sector","%","sector","%","sector","%","Gbyte","%","Gbyte/s","byte","%","byte/s","Gbyte","%","Gbyte/s","%","%","%","%","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector","%","sector","%","sector","%","sector/ns","sector","%","sector","%","sector","%","sector","%","byte","%","byte/s","","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","%","%","%","%","%","%","%","%","%","%","%","%","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","%","%","cycle","%","%","%","%","%","%","%","%","%","%","%","%","%","","block","block","block","","","","","cluster","block","","","","","","","","","","","","%","%","block","block","block","block","block","","","","","","Mbyte","","","","","register/thread","register/thread","Kbyte","Kbyte/block","Kbyte/block","Kbyte/block","Kbyte/block","byte/block","SM","","","thread","","","","","","","","","%","sector","%","","","","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","%","%","%","%","%","%","%","%","%","%","%","%","sector","%","sector/ns","sector","sector","sector","sector","%","","Mhz","%","","Mhz","%","","Mhz","%","","Ghz","request","request","request","request","request","request","request","request","%","%","%","sector","%","sector/cycle","sector/cycle","sector","%","sector","%","sector","%","sector/ns","sector","%","sector/us","%","sector/cycle","sector/cycle","sector","sector/s","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/ns","%","sector/cycle","sector/cycle","sector","sector/us","sector","sector","%","sector/cycle","sector/cycle","%","%","sector","%","sector/ns","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/s","%","sector/cycle","sector/cycle","sector","sector/s","sector","sector","%","sector/cycle","sector/cycle","sector","s
"0","73","python3.12","127.0.0.1","void at::vectorized_elementwise_kernel<4, at::BinaryFunctor<float, float, float, at::MulFunctor<float>>, std::array<char *, 3>>(int, T2, T3)","1","7","(128, 1, 1)","(1, 1, 1)","0","12.1","0","1","3802","2078","3874.000000","1694","6144","34334763.95","0","0.000000","1.024000","3.326180","0","0","0","692","5270.000000","10876.000000","1.959000","192.000000","12288.000000","12288.000000","0.075665","0","0.000000","0","0.128805","0","37","489","0","0","nan","0","432","1","0","1","1","1","0","1","0","1","0","443","2418000","1","12","1","0","1","1","1","No-CC","1","1","1","0","0","NVIDIA GB10","0","0","256","4","1","1","256","1","0","0","100","1","772935902","11794","32000","16","0","4","2500","16","161","4318","0","1","0","0","1","0","1","443","1","1","1","25165824","24","24","24","1","1","134217728","1024","1024","64","24","2418000","2147483647","65535","65535","4","1","8533000","18874368","2147483647","65536","65536","255","49152","101376","102400","1024","1536","48","12","2048","32768","32768","65536","32768","2048","32768","131072","16384","16384","16384","2046","32768","32768","2048","32768","268435456","32768","131072","32768","32768","65536","32768","2048","32768","65000","2097120","131072","32768","32768","131072","16384","32768","16384","8192","16384","8192","2046","32768","32768","4","8533000","1","1","0","0","0","0","48","6","4","1","0","1","1","1","0","15","1","0","1024","12","64","1","1","512","0","1","512","32","65536","130661769216","1","1","1","32","4.000000","0.006289","30.000000","0.047170","672","0.528302","15900.000000","2.132511","15903.000000","2.132913","15897.000000","2.132108","63600.000000","8.530043","0.264263","0.861043","0.014001","0.264263","0.146132","0.448022","0.014001","0.146132","0.251575","0.918239","0.056604","0.251575","0","0","0","0","0.264263","0.918239","0.056604","0.264263","0.007456","0.007456","0.007456","0.007456","0","0","0","0","0.000000","0.000000","0.000000","0.000000","0","0","0","0","0.026468","1.270440","0.000000","0.026468","53.623188","69","0.018082","489","46.625000","2238.000000","0.000000","2238.000000","15900.000000","2.132511","15903.000000","2.132913","15897.000000","2.132108","763200.000000","102.360515","0","0","0","0","0","0.000008","0.000393","0.000000","0.000008","0.000037","0.001769","0.000000","0.000037","0.009565","0.459119","0.000000","0.009565","70","0.009172","0","0","0","0.000000","0","0.000000","0","0","0","0","0","0","0","0","0.000786","0.037736","0.000000","0.000786","2.000000","0.000262","48","0.268240","0.019130","0.918239","0.000000","0.019130","0.012972","0.031447","0.012579","0.012972","0.000032","0.000131","0.004292","0","0","0","0","0","0","0.000000","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.000000","0.000000","0.000000","1.000000","0.000131","0","0","0","0","0","0","0.000000","0.000262","0.008584","0","0","0","0.000000","0.000000","0.000000","0.000262","0.012579","0.000000","0.000262","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.000000","0.000000","0.000000","2.000000","0.000262","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","2","0.000262","0","0","1","0.000131","0","0.000000","0","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","2","0.000262","0","0","1","0.000131","0","0.000000","0","0.000000","0","0","0","0","0","0","0","0","0","0","0.000000","0","0.000000","0","0","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","2.000000","0.000000","2.000000","0","0","0","1.000000","0","1.000000","0.000000","0.000000","0","0.000000","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.056604","0.056604","0.056604","0.056604","19.302949","0.056604","313.136729","0.918239","19.302949","0.056604","19.302949","0.056604","0","128","1","1","128","0","0","0","0","8","PolicySpread","0","1","0","CachePreferNone","1","1","1","1","1","","0","0","24.000000","24.000000","12.
"1","73","python3.12","127.0.0.1","cutlass3x_sm120_bstensorop_s16864gemm_block_scaled_ue4m3xe2m1_ue4m3xe2m1_f32_bf16_bf16_128x128x256_1x1x1_0_tnn_align32_o_vs16_bias_bf16_relu_stream_k","1","7","(384, 1, 1)","(42, 296, 1)","0","12.1","0","1","56167","52394","354166.000000","7598","6144","1620772988.98","696768","100.000000","1.024000","466.239645","294","44556288","0","625","316.000000","4225.000000","4.975000","192.000000","12288.000000","12288.000000","0.033180","0","3.443262","0","0.000000","0","42996614","1295861238","576","696768","1","1","432","1","0","1","1","1","0","1","0","1","0","443","2418000","1","12","1","0","1","1","1","No-CC","1","1","1","0","0","NVIDIA GB10","0","0","256","4","1","1","256","1","0","0","100","1","772935902","11794","32000","16","0","4","2500","16","161","4318","0","1","0","0","1","0","1","443","1","1","1","25165824","24","24","24","1","1","134217728","1024","1024","64","24","2418000","2147483647","65535","65535","4","1","8533000","18874368","2147483647","65536","65536","255","49152","101376","102400","1024","1536","48","12","2048","32768","32768","65536","32768","2048","32768","131072","16384","16384","16384","2046","32768","32768","2048","32768","268435456","32768","131072","32768","32768","65536","32768","2048","32768","65000","2097120","131072","32768","32768","131072","16384","32768","16384","8192","16384","8192","2046","32768","32768","4","8533000","1","1","0","0","0","0","48","6","4","1","0","1","1","1","0","15","1","0","1024","12","64","1","1","512","0","1","512","32","65536","130661769216","1","1","1","32","2798544.000000","0.591359","9028.000000","0.001908","10712","0.001132","118309837.750000","2.148848","118334635.000000","2.149298","118285340.000000","2.148403","473239351.000000","8.595390","23.974686","23.993901","23.961532","23.974686","12.559269","12.628962","12.489907","12.559269","23.598833","23.616927","23.583289","23.598833","0","0","0","0","23.974686","23.993901","23.961532","23.974686","55.057344","55.057344","55.057344","55.057344","12384","12384","12384","12384","12384.000000","12384.000000","12384.000000","12384.000000","12432","12432","12432","12432","0.070323","0.070326","0.070310","0.070323","99.721630","1991232","0.070128","1259175041","118776167.583333","118805817.000000","118739425.000000","5701256044.000000","118309837.750000","2.148848","118334635.000000","2.149298","118285340.000000","2.148403","5678872212.000000","103.144682","44813290","0","44566514","0","250754","5.324639","5.324688","5.324632","5.324639","1.793456","1.793457","1.793455","1.793456","12.737439","12.738040","12.736932","12.737439","719808603","12.675203","0","0","713157733","12.558087","3440663","0.060587","0","0","0","0","0","0","0","0","11.787449","11.787449","11.787449","11.787449","697152.000000","0.012276","48","12.662289","7.999717","7.999717","7.999717","7.999717","3.560743","3.563408","3.559783","3.560743","406.757376","0.223833","7.387886","0","0","0","0","0","0","406.683648","0.223792","7.386547","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","12708864.000000","0.223792","230.829587","2304.000000","0.000041","0","0","0","0","0","0","25.664425","14.122756","466.139902","0","0","0","25.664422","14.122755","466.139847","14.122756","14.122756","14.122756","14.122756","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","802013184.000000","14.122755","14.566870","96.000000","0.000002","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960","0.000017","0","0","0","0.000000","696192","0.012259","576","0.000010","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960","0.000017","0","0","0","0.000000","696192","0.012259","576","0.000010","0","0","0","0","0","0","0","0","0","0","99.979413","0","90.000000","0","0","100.000000","97.916667","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960.000000","864.000000","96.000000","0","0","0","0.000000","0","0.000000","696192.000000","696192.000000","0","2304.000000","2256.000000","48.000000","0","0","0","0","0","0","