f9aa7be97c
* Add MI 350 hardware information
* Refactor MI GPU YAML file and corresponding interface
* Add SoC file for gfx950 architecture
* Add analysis report configs for MI 350 containing existing metrics
* Add placeholder None valued metrics for previous architectures to make
baseline comparison work
* Enable testing on MI 350
* Analysis config metric changes
- SPI changes
- Update metric formula for default SPI pipe counter
- Use efficiently collected pipe wise SPI counters
- Add SPI Wave Occupancy
- Add Scheduler-Pipe Wave Utilization
- Update formula for VGPR Writes
- Add Scheduler-Pipe FIFO Full Rate
- CPC changes
- Add CPC SYNC FIFO Full Rate
- Add CPC CANE Stall Rate
- Add CPC ADC Utilization
- SQ changes
- Add VALU co-issue efficiency
- Add F6F4 datatype metrics
- Update formula for total FLOPs by adding F6F4 counters
- Add LDS STORE / LOAD / ATOMIC metrics
- Add LDS STORE / LOAD / ATOMIC bandwidth
- Add LDS FIFO and TA ADDR / CMD / DATA FIFO full rates
* Collect TCP_TCP_LATENCY_sum only for gfx950 (MI 350)
* Do not inject SQ_ACCUM_PREV_HIRES unnecesarily
* Do not hardcode memory and shader clock speeds
* Write num_hbm_channels to sysinfo.csv instead of hbm_bw while profiling
* Move generate sysinfo.csv to pre processing step of profiling
* Add warnings to use --specs-correction for missing sysinfo.csv values during analysis phase
* Update CHANGELOG
* Analysis phase warning to use --specs-correction when needed
7.5 KiB
7.5 KiB
| 1 | Dispatch_ID | GPU_ID | Queue_ID | PID | TID | Grid_Size | Workgroup_Size | LDS_Per_Workgroup | Scratch_Per_Workitem | Arch_VGPR | Accum_VGPR | SGPR | Wave_Size | Kernel_Name | Start_Timestamp | End_Timestamp | Correlation_ID | Kernel_ID | CPC_CPC_STAT_BUSY | CPC_CPC_STAT_IDLE | CPF_CPF_STAT_IDLE | CPF_CPF_STAT_STALL | SPI_CSQ_P0_OCCUPANCY | SPI_CSQ_P1_OCCUPANCY | SPI_CSQ_P2_OCCUPANCY | SPI_CSQ_P3_OCCUPANCY | SPI_RA_SGPR_SIMD_FULL_CSN | SQC_DCACHE_REQ_READ_16 | SQ_BUSY_CU_CYCLES | SQ_INSTS_LDS_LOAD | SQ_INSTS_LDS_STORE | SQ_INSTS_VALU_MUL_F64 | SQ_INST_LEVEL_SMEM | SQ_LDS_CMD_FIFO_FULL | TA_BUFFER_TOTAL_CYCLES_sum | TA_FLAT_READ_WAVEFRONTS_sum | TCC_ATOMIC[0] | TCC_ATOMIC[10] | TCC_ATOMIC[11] | TCC_ATOMIC[12] | TCC_ATOMIC[13] | TCC_ATOMIC[15] | TCC_ATOMIC[16] | TCC_ATOMIC[17] | TCC_ATOMIC[18] | TCC_ATOMIC[19] | TCC_ATOMIC[1] | TCC_ATOMIC[21] | TCC_ATOMIC[22] | TCC_ATOMIC[23] | TCC_ATOMIC[24] | TCC_ATOMIC[26] | TCC_ATOMIC[27] | TCC_ATOMIC[28] | TCC_ATOMIC[29] | TCC_ATOMIC[2] | TCC_ATOMIC[31] | TCC_ATOMIC[32] | TCC_ATOMIC[33] | TCC_ATOMIC[34] | TCC_ATOMIC[35] | TCC_ATOMIC[37] | TCC_ATOMIC[38] | TCC_ATOMIC[39] | TCC_ATOMIC[3] | TCC_ATOMIC[40] | TCC_ATOMIC[42] | TCC_ATOMIC[43] | TCC_ATOMIC[44] | TCC_ATOMIC[45] | TCC_ATOMIC[47] | TCC_ATOMIC[48] | TCC_ATOMIC[49] | TCC_ATOMIC[50] | TCC_ATOMIC[51] | TCC_ATOMIC[53] | TCC_ATOMIC[54] | TCC_ATOMIC[55] | TCC_ATOMIC[56] | TCC_ATOMIC[58] | TCC_ATOMIC[59] | TCC_ATOMIC[5] | TCC_ATOMIC[60] | TCC_ATOMIC[61] | TCC_ATOMIC[63] | TCC_ATOMIC[6] | TCC_ATOMIC[7] | TCC_ATOMIC[8] | TCC_BUBBLE[0] | TCC_BUBBLE[10] | TCC_BUBBLE[11] | TCC_BUBBLE[12] | TCC_BUBBLE[13] | TCC_BUBBLE[15] | TCC_BUBBLE[16] | TCC_BUBBLE[17] | TCC_BUBBLE[18] | TCC_BUBBLE[19] | TCC_BUBBLE[1] | TCC_BUBBLE[21] | TCC_BUBBLE[22] | TCC_BUBBLE[23] | TCC_BUBBLE[24] | TCC_BUBBLE[26] | TCC_BUBBLE[27] | TCC_BUBBLE[28] | TCC_BUBBLE[29] | TCC_BUBBLE[2] | TCC_BUBBLE[31] | TCC_BUBBLE[32] | TCC_BUBBLE[33] | TCC_BUBBLE[34] | TCC_BUBBLE[35] | TCC_BUBBLE[37] | TCC_BUBBLE[38] | TCC_BUBBLE[39] | TCC_BUBBLE[3] | TCC_BUBBLE[40] | TCC_BUBBLE[42] | TCC_BUBBLE[43] | TCC_BUBBLE[44] | TCC_BUBBLE[45] | TCC_BUBBLE[47] | TCC_BUBBLE[48] | TCC_BUBBLE[49] | TCC_BUBBLE[50] | TCC_BUBBLE[51] | TCC_BUBBLE[53] | TCC_BUBBLE[54] | TCC_BUBBLE[55] | TCC_BUBBLE[56] | TCC_BUBBLE[58] | TCC_BUBBLE[59] | TCC_BUBBLE[5] | TCC_BUBBLE[60] | TCC_BUBBLE[61] | TCC_BUBBLE[63] | TCC_BUBBLE[6] | TCC_BUBBLE[7] | TCC_BUBBLE[8] | TCC_EA0_ATOMIC[0] | TCC_EA0_ATOMIC[10] | TCC_EA0_ATOMIC[11] | TCC_EA0_ATOMIC[12] | TCC_EA0_ATOMIC[13] | TCC_EA0_ATOMIC[15] | TCC_EA0_ATOMIC[16] | TCC_EA0_ATOMIC[17] | TCC_EA0_ATOMIC[18] | TCC_EA0_ATOMIC[19] | TCC_EA0_ATOMIC[1] | TCC_EA0_ATOMIC[21] | TCC_EA0_ATOMIC[22] | TCC_EA0_ATOMIC[23] | TCC_EA0_ATOMIC[24] | TCC_EA0_ATOMIC[26] | TCC_EA0_ATOMIC[27] | TCC_EA0_ATOMIC[28] | TCC_EA0_ATOMIC[29] | TCC_EA0_ATOMIC[2] | TCC_EA0_ATOMIC[31] | TCC_EA0_ATOMIC[32] | TCC_EA0_ATOMIC[33] | TCC_EA0_ATOMIC[34] | TCC_EA0_ATOMIC[35] | TCC_EA0_ATOMIC[37] | TCC_EA0_ATOMIC[38] | TCC_EA0_ATOMIC[39] | TCC_EA0_ATOMIC[3] | TCC_EA0_ATOMIC[40] | TCC_EA0_ATOMIC[42] | TCC_EA0_ATOMIC[43] | TCC_EA0_ATOMIC[44] | TCC_EA0_ATOMIC[45] | TCC_EA0_ATOMIC[47] | TCC_EA0_ATOMIC[48] | TCC_EA0_ATOMIC[49] | TCC_EA0_ATOMIC[50] | TCC_EA0_ATOMIC[51] | TCC_EA0_ATOMIC[53] | TCC_EA0_ATOMIC[54] | TCC_EA0_ATOMIC[55] | TCC_EA0_ATOMIC[56] | TCC_EA0_ATOMIC[58] | TCC_EA0_ATOMIC[59] | TCC_EA0_ATOMIC[5] | TCC_EA0_ATOMIC[60] | TCC_EA0_ATOMIC[61] | TCC_EA0_ATOMIC[63] | TCC_EA0_ATOMIC[6] | TCC_EA0_ATOMIC[7] | TCC_EA0_ATOMIC[8] | TCC_HIT[0] | TCC_HIT[10] | TCC_HIT[11] | TCC_HIT[12] | TCC_HIT[13] | TCC_HIT[15] | TCC_HIT[16] | TCC_HIT[17] | TCC_HIT[18] | TCC_HIT[19] | TCC_HIT[1] | TCC_HIT[21] | TCC_HIT[22] | TCC_HIT[23] | TCC_HIT[24] | TCC_HIT[26] | TCC_HIT[27] | TCC_HIT[28] | TCC_HIT[29] | TCC_HIT[2] | TCC_HIT[31] | TCC_HIT[32] | TCC_HIT[33] | TCC_HIT[34] | TCC_HIT[35] | TCC_HIT[37] | TCC_HIT[38] | TCC_HIT[39] | TCC_HIT[3] | TCC_HIT[40] | TCC_HIT[42] | TCC_HIT[43] | TCC_HIT[44] | TCC_HIT[45] | TCC_HIT[47] | TCC_HIT[48] | TCC_HIT[49] | TCC_HIT[50] | TCC_HIT[51] | TCC_HIT[53] | TCC_HIT[54] | TCC_HIT[55] | TCC_HIT[56] | TCC_HIT[58] | TCC_HIT[59] | TCC_HIT[5] | TCC_HIT[60] | TCC_HIT[61] | TCC_HIT[63] | TCC_HIT[6] | TCC_HIT[7] | TCC_HIT[8] | TCP_TA_TCP_STATE_READ_sum | TCP_TCC_UC_READ_REQ_sum | TCP_TOTAL_ATOMIC_WITH_RET_sum | TCP_UTCL1_TRANSLATION_HIT_sum | TD_LOAD_WAVEFRONT_sum | TD_STORE_WAVEFRONT_sum |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | 0 | 0 | 1 | 15751 | 15751 | 1048576 | 256 | 0 | 0 | 4 | 0 | 16 | 64 | vecCopy(double*, double*, double*, int, int) | 1248225586306035 | 1248225586327794 | 1 | 13 | 77276.0 | 0.0 | 0.0 | 3438.0 | 26015654.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 1005075.0 | 0.0 | 0.0 | 0.0 | 170074.0 | 0.0 | 0.0 | 16384.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 2135.0 | 1024.0 | 1041.0 | 1094.0 | 1024.0 | 1024.0 | 2118.0 | 2118.0 | 2048.0 | 1024.0 | 2118.0 | 1094.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1094.0 | 1024.0 | 2048.0 | 1024.0 | 2118.0 | 2118.0 | 2048.0 | 1024.0 | 1094.0 | 1024.0 | 1024.0 | 1041.0 | 1024.0 | 1024.0 | 1024.0 | 1094.0 | 1024.0 | 1024.0 | 2118.0 | 2153.0 | 2083.0 | 1024.0 | 1094.0 | 1059.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1094.0 | 1094.0 | 1059.0 | 1024.0 | 1024.0 | 1024.0 | 1041.0 | 32768.0 | 0.0 | 0.0 | 493416.0 | 32768.0 | 16384.0 |
| 3 | 1 | 0 | 1 | 15751 | 15751 | 1048576 | 256 | 0 | 0 | 4 | 0 | 16 | 64 | vecCopy(double*, double*, double*, int, int) | 1248225587128604 | 1248225587145724 | 2 | 13 | 66157.0 | 0.0 | 0.0 | 3477.0 | 18725576.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 751413.0 | 0.0 | 0.0 | 0.0 | 109530.0 | 0.0 | 0.0 | 16384.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 2118.0 | 1024.0 | 1024.0 | 1094.0 | 1024.0 | 1024.0 | 2135.0 | 2118.0 | 2048.0 | 1041.0 | 2118.0 | 1094.0 | 1024.0 | 1024.0 | 1041.0 | 1024.0 | 1041.0 | 1094.0 | 1024.0 | 2048.0 | 1024.0 | 2118.0 | 2118.0 | 2048.0 | 1024.0 | 1094.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1094.0 | 1024.0 | 1024.0 | 2118.0 | 2153.0 | 2083.0 | 1024.0 | 1094.0 | 1059.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1094.0 | 1094.0 | 1059.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 32768.0 | 0.0 | 0.0 | 524288.0 | 32768.0 | 16384.0 |
| 4 | 2 | 0 | 1 | 15751 | 15751 | 1048576 | 256 | 0 | 0 | 4 | 0 | 16 | 64 | vecCopy(double*, double*, double*, int, int) | 1248225587833139 | 1248225587849859 | 3 | 13 | 62396.0 | 0.0 | 0.0 | 376.0 | 18327092.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 740907.0 | 0.0 | 0.0 | 0.0 | 108626.0 | 0.0 | 0.0 | 16384.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 2118.0 | 1024.0 | 1024.0 | 1094.0 | 1024.0 | 1024.0 | 2118.0 | 2118.0 | 2048.0 | 1024.0 | 2118.0 | 1094.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1094.0 | 1024.0 | 2048.0 | 1024.0 | 2118.0 | 2118.0 | 2048.0 | 1024.0 | 1094.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 1094.0 | 1024.0 | 1024.0 | 2135.0 | 2153.0 | 2083.0 | 1041.0 | 1094.0 | 1059.0 | 1024.0 | 1041.0 | 1024.0 | 1041.0 | 1094.0 | 1094.0 | 1059.0 | 1024.0 | 1024.0 | 1024.0 | 1024.0 | 32768.0 | 0.0 | 0.0 | 524288.0 | 32768.0 | 16384.0 |