Files
rocm-systems/tests/workloads/TCC/mi200/prev_analysis/201.csv
T
colramos-amd 62d130b458 Initial commit
2022-11-04 14:49:36 -05:00

1.0 KiB

1MetricValueUnitPeakPoP
2VALU FLOPsGflop22630.4
3VALU IOPsGiop22630.4
4MFMA FLOPs (BF16)Gflop90521.6
5MFMA FLOPs (F16)Gflop181043.2
6MFMA FLOPs (F32)Gflop45260.8
7MFMA FLOPs (F64)Gflop45260.8
8MFMA IOPs (Int8)Giop181043.2
9Active CUsCus104
10SALU UtilPct100
11VALU UtilPct100
12MFMA UtilPct100
13VALU Active Threads/WaveThreads64
14IPC - IssueInstr/cycle5
15LDS BWGb/sec22630.4
16LDS Bank ConflictConflicts/access32
17Instr Cache Hit RatePct100
18Instr Cache BWGb/s6092.8
19Scalar L1D Cache Hit RatePct100
20Scalar L1D Cache BWGb/s6092.8
21Vector L1D Cache Hit RatePct100
22Vector L1D Cache BWGb/s11315.199999999999
23L2 Cache Hit Rate3.901475373514127Pct1003.901475373514127
24L2-Fabric Read BW1158.2842873117752Gb/s1638.470.69606245799409
25L2-Fabric Write BW6.619265013283814Gb/s1638.40.4040078743459359
26L2-Fabric Read Latency537.3816507825869Cycles
27L2-Fabric Write Latency391.49073331318203Cycles
28Wave OccupancyWavefronts3328
29Instr Fetch BWGb/s3046.4
30Instr Fetch LatencyCycles