Files
rocm-systems/tests/workloads/ipblocks_SQ_SPI/MI200/pmc_perf.csv
T

5 lines
5.1 KiB
CSV
Raw Normal View History

Dispatch_ID,Kernel_Name,GPU_ID,Grid_Size,Workgroup_Size,LDS_Per_Workgroup,Scratch_Per_Workitem,Arch_VGPR,Accum_VGPR,SGPR,wave_size,obj,SQ_INSTS_VALU_ADD_F16,SQ_INSTS_VALU_MUL_F16,SQ_INSTS_VALU_FMA_F16,SQ_INSTS_VALU_TRANS_F16,SQ_INSTS_VALU_ADD_F32,SQ_INSTS_VALU_MUL_F32,SQ_INSTS_VALU_FMA_F32,SQ_INSTS_VALU_TRANS_F32,GRBM_SPI_BUSY,SPI_CSN_NUM_THREADGROUPS,SPI_CSN_WAVE,wave_size_1,obj_1,SQ_INSTS_SMEM_NORM,SQ_INSTS_MFMA,SQ_INSTS_VALU_MFMA_I8,SQ_INSTS_VALU_MFMA_F16,SQ_INSTS_VALU_MFMA_BF16,SQ_INSTS_VALU_MFMA_F32,SQ_INSTS_VALU_MFMA_F64,SQ_VALU_MFMA_BUSY_CYCLES,SPI_VWC_CSC_WR,SPI_RA_BULKY_CU_FULL_CSN,wave_size_2,obj_2,SQC_ICACHE_MISSES_DUPLICATE,SQC_DCACHE_INPUT_VALID_READYB,SQC_DCACHE_ATOMIC,SQC_DCACHE_REQ_READ_8,SQC_DCACHE_REQ,SQC_DCACHE_HITS,SQC_DCACHE_MISSES,SQC_DCACHE_MISSES_DUPLICATE,wave_size_3,obj_3,SQ_INSTS_VALU_CVT,SQ_INSTS_VMEM_WR,SQ_INSTS_VMEM_RD,SQ_INSTS_VMEM,SQ_INSTS_SALU,SQ_INSTS_VSKIPPED,SQ_INSTS,SQ_INSTS_VALU,GRBM_COUNT,GRBM_GUI_ACTIVE,SPI_CSN_WINDOW_VALID,SPI_CSN_BUSY,wave_size_4,obj_4,SQ_IFETCH,SQ_LDS_BANK_CONFLICT,SQ_LDS_ADDR_CONFLICT,SQ_LDS_UNALIGNED_STALL,SQ_WAVES,SQ_WAVES_EQ_64,SQ_WAVES_LT_64,SQ_WAVES_LT_48,SPI_RA_BAR_CU_FULL_CSN,SPI_RA_TGLIM_CU_FULL_CSN,wave_size_5,obj_5,SQC_DCACHE_REQ_READ_1,SQC_DCACHE_REQ_READ_2,SQC_DCACHE_REQ_READ_4,wave_size_6,obj_6,SQ_WAVES_LT_32,SQ_WAVES_LT_16,SQ_ITEMS,SQ_LDS_MEM_VIOLATIONS,SQ_LDS_ATOMIC_RETURN,SQ_LDS_IDX_ACTIVE,SQ_WAVES_RESTORED,SQ_WAVES_SAVED,SPI_RA_WVLIM_STALL_CSN,SPI_SWC_CSC_WR,wave_size_7,obj_7,SQC_TC_DATA_WRITE_REQ,SQC_TC_DATA_ATOMIC_REQ,SQC_TC_STALL,SQC_TC_REQ,SQC_DCACHE_REQ_READ_16,SQC_ICACHE_REQ,SQC_ICACHE_HITS,SQC_ICACHE_MISSES,wave_size_8,obj_8,SQ_ACTIVE_INST_EXP_GDS,SQ_ACTIVE_INST_MISC,SQ_ACTIVE_INST_FLAT,SQ_INST_CYCLES_VMEM_WR,SQ_INST_CYCLES_VMEM_RD,SQ_INST_CYCLES_SMEM,SQ_INST_CYCLES_SALU,SQ_THREAD_CYCLES_VALU,SPI_RA_SGPR_SIMD_FULL_CSN,SPI_RA_LDS_CU_FULL_CSN,wave_size_9,obj_9,SQ_INSTS_FLAT_LDS_ONLY,SQ_INSTS_VALU_MFMA_MOPS_I8,SQ_INSTS_VALU_MFMA_MOPS_F16,SQ_INSTS_VALU_MFMA_MOPS_BF16,SQ_INSTS_VALU_MFMA_MOPS_F32,SQ_INSTS_VALU_MFMA_MOPS_F64,SQC_TC_INST_REQ,SQC_TC_DATA_READ_REQ,wave_size_10,obj_10,SQ_ACTIVE_INST_ANY,SQ_CYCLES,SQ_BUSY_CYCLES,SQ_BUSY_CU_CYCLES,SQ_ACTIVE_INST_VMEM,SQ_ACTIVE_INST_LDS,SQ_ACTIVE_INST_VALU,SQ_ACTIVE_INST_SCA,SPI_RA_WAVE_SIMD_FULL_CSN,SPI_RA_VGPR_SIMD_FULL_CSN,wave_size_11,obj_11,SQ_INSTS_LDS,SQ_INSTS_GDS,SQ_INSTS_EXP_GDS,SQ_INSTS_BRANCH,SQ_INSTS_SENDMSG,SQ_WAVE_CYCLES,SQ_WAIT_ANY,SQ_WAIT_INST_ANY,SPI_RA_RES_STALL_CSN,SPI_RA_TMP_STALL_CSN,wave_size_12,obj_12,SQ_INSTS_VALU_ADD_F64,SQ_INSTS_VALU_MUL_F64,SQ_INSTS_VALU_FMA_F64,SQ_INSTS_VALU_TRANS_F64,SQ_INSTS_VALU_INT32,SQ_INSTS_VALU_INT64,SQ_INSTS_SMEM,SQ_INSTS_FLAT,SPI_RA_REQ_NO_ALLOC,SPI_RA_REQ_NO_ALLOC_CSN,Start_Timestamp,End_Timestamp
0,"vecCopy(double*, double*, double*, int, int) [clone .kd]",2,1048576,256,0,0,8,0,16,64,0x7f60b2444ec0,0,0,0,0,0,0,0,0,17624,4096,16384,64,0x7fd9612e8ec0,131072,0,0,0,0,0,0,0,16384,0,64,0x7f21ea858ec0,1199,119879,0,0,65536,63800,56,1680,64,0x7f9ac19c8ec0,0,16384,16384,32768,49152,0,393216,163840,27512,27512,218393,134670,64,0x7ff9f9104ec0,65536,0,0,0,16384,16384,0,0,0,0,64,0x7f6a15b9cec0,32768,32768,0,64,0x7f11cbd5cec0,0,0,1048576,0,0,0,0,0,0,32768,64,0x7f03bc60cec0,0,0,0,224,0,65536,64280,56,64,0x7fce08ef4ec0,0,32768,32768,16384,16384,65536,49152,10485760,0,0,64,0x7f6157014ec0,0,0,0,0,0,0,168,56,64,0x7f0330014ec0,344064,219936,127406,1615032,0,0,163840,114688,488525,0,64,0x7f162504cec0,0,0,0,16384,16384,10847134,9773211,729859,10336,0,64,0x7f2513d74ec0,0,0,0,0,114688,0,65536,32768,13652,18752,1411654780403076,1411654780422756
1,"vecCopy(double*, double*, double*, int, int) [clone .kd]",2,1048576,256,0,0,8,0,16,64,0x7f60b2444ec0,0,0,0,0,0,0,0,0,29838,4096,16384,64,0x7fd9612e8ec0,131072,0,0,0,0,0,0,0,16384,0,64,0x7f21ea858ec0,0,186099,0,0,65536,63800,56,1680,64,0x7f9ac19c8ec0,0,16384,16384,32768,49152,0,393216,163840,41431,41431,331448,231545,64,0x7ff9f9104ec0,65536,0,0,0,16384,16384,0,0,0,0,64,0x7f6a15b9cec0,32768,32768,0,64,0x7f11cbd5cec0,0,0,1048576,0,0,0,0,0,0,32768,64,0x7f03bc60cec0,0,0,0,56,0,65536,65536,0,64,0x7fce08ef4ec0,0,32768,32768,16384,16384,65536,49152,10485760,0,0,64,0x7f6157014ec0,0,0,0,0,0,0,0,56,64,0x7f0330014ec0,344064,331416,222248,2781574,0,0,163840,114688,986828,0,64,0x7f162504cec0,0,0,0,16384,16384,19326373,18172977,809332,20575,0,64,0x7f2513d74ec0,0,0,0,0,114688,0,65536,32768,33945,27855,1411654780594116,1411654780610756
2,"vecCopy(double*, double*, double*, int, int) [clone .kd]",2,1048576,256,0,0,8,0,16,64,0x7f60b2444ec0,0,0,0,0,0,0,0,0,29277,4096,16384,64,0x7fd9612e8ec0,131072,0,0,0,0,0,0,0,16384,0,64,0x7f21ea858ec0,0,189505,0,0,65536,63800,56,1680,64,0x7f9ac19c8ec0,0,16384,16384,32768,49152,0,393216,163840,40657,40657,325256,228731,64,0x7ff9f9104ec0,65536,0,0,0,16384,16384,0,0,0,0,64,0x7f6a15b9cec0,32768,32768,0,64,0x7f11cbd5cec0,0,0,1048576,0,0,0,0,0,0,32768,64,0x7f03bc60cec0,0,0,0,56,0,65536,65536,0,64,0x7fce08ef4ec0,0,32768,32768,16384,16384,65536,49152,10485760,0,0,64,0x7f6157014ec0,0,0,0,0,0,0,0,56,64,0x7f0330014ec0,344064,340400,230274,2899401,0,0,163840,114688,1078763,0,64,0x7f162504cec0,0,0,0,16384,16384,19821355,18639520,837771,21844,0,64,0x7f2513d74ec0,0,0,0,0,114688,0,65536,32768,34504,27978,1411654780632356,1411654780648036