Enchanged rocm-triton-prof.py; regex, config-yaml file

[ravil-mobile]

No description provided.

[guacamoleo]

Looks like this still has the data type being printed.

Also, if I run without any args, I get an unintelligible output. It would be better than all commandline errors print the help menu.

python3 rocm-triton-prof.py
running original program...
Traceback (most recent call last):
  File "/home/dtanner/repos/rocm_triton/python/perf-kernels/fa_perf_4/rocm-triton-prof.py", line 74, in run_external_binary
    print(f"CURR.CMD: {' '.join(cmd)}")
TypeError: can only join an iterable

Timing info in `nsec`:
count       163.000000
mean     613441.061350
std       16550.151525
min      581941.000000
25%      595713.500000
50%      616461.000000
75%      625823.000000
max      650620.000000
dtype: float64


NON-FLOP related data:
   Counter Name         Max        Min          Mean     Median
0    GRBM_COUNT  10047431.0  7028708.0  7.173176e+06  7158677.0
1   TCC_HIT_sum   5252411.0  5094965.0  5.131720e+06  5129603.0
2  TCC_MISS_sum   6001565.0  5873660.0  5.941301e+06  5942057.0

FLOP related data:
                    Counter Name     Raw Data          FLOP  Relative FLOP, %
0          SQ_INSTS_VALU_ADD_F16          0.0  0.000000e+00          0.000000
1          SQ_INSTS_VALU_MUL_F16          0.0  0.000000e+00          0.000000
2          SQ_INSTS_VALU_FMA_F16     262144.0  3.355443e+07          0.023327
3        SQ_INSTS_VALU_TRANS_F16          0.0  0.000000e+00          0.000000
4          SQ_INSTS_VALU_ADD_F32    9003008.0  5.761925e+08          0.400571
5          SQ_INSTS_VALU_MUL_F32    8691712.0  5.562696e+08          0.386721
6          SQ_INSTS_VALU_FMA_F32    4536448.0  5.806653e+08          0.403681
7        SQ_INSTS_VALU_TRANS_F32    4608128.0  2.949202e+08          0.205030
8          SQ_INSTS_VALU_ADD_F64          0.0  0.000000e+00          0.000000
9          SQ_INSTS_VALU_MUL_F64          0.0  0.000000e+00          0.000000
10         SQ_INSTS_VALU_FMA_F64          0.0  0.000000e+00          0.000000
11       SQ_INSTS_VALU_TRANS_F64          0.0  0.000000e+00          0.000000
12   SQ_INSTS_VALU_MFMA_MOPS_F16  276955136.0  1.418010e+11         98.580670
13  SQ_INSTS_VALU_MFMA_MOPS_BF16          0.0  0.000000e+00          0.000000
14   SQ_INSTS_VALU_MFMA_MOPS_F32          0.0  0.000000e+00          0.000000
15   SQ_INSTS_VALU_MFMA_MOPS_F64          0.0  0.000000e+00          0.000000

Performance info in TFLOP/s:
count    163.000000
mean     234.655881
std        6.381366
min      221.085475
25%      229.845550
50%      233.336142
75%      241.462803
max      247.177346
dtype: float64