Skip to content

Commit

Permalink
fix config for f16
Browse files Browse the repository at this point in the history
  • Loading branch information
zhanglx13 committed Oct 1, 2024
1 parent b8a82a4 commit db1424e
Showing 1 changed file with 3 additions and 4 deletions.
7 changes: 3 additions & 4 deletions python/perf-kernels/tools/tune_gemm/config_fp16.yaml
Original file line number Diff line number Diff line change
@@ -1,7 +1,4 @@

- {'M': 9728, 'N': 8192, 'K': 8256, 'rowMajorA': 'T', 'rowMajorB': 'N', 'BLOCK_SIZE_M': 256, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 64, 'GROUP_SIZE_M': 4, 'SPLIT_K': 1, 'num_warps': 8, 'num_stages': 0, 'waves_per_eu': 0, 'matrix_instr_nonkdim': 16, 'kpack': 2}
- {'M': 9728, 'N': 8192, 'K': 8320, 'rowMajorA': 'T', 'rowMajorB': 'N', 'BLOCK_SIZE_M': 256, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 64, 'GROUP_SIZE_M': 4, 'SPLIT_K': 1, 'num_warps': 8, 'num_stages': 0, 'waves_per_eu': 0, 'matrix_instr_nonkdim': 16, 'kpack': 2}
- {'M': 9728, 'N': 8192, 'K': 8384, 'rowMajorA': 'T', 'rowMajorB': 'N', 'BLOCK_SIZE_M': 256, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 64, 'GROUP_SIZE_M': 4, 'SPLIT_K': 1, 'num_warps': 8, 'num_stages': 0, 'waves_per_eu': 0, 'matrix_instr_nonkdim': 16, 'kpack': 2}

## 2 workgroups / CU
#- {'M': 4864, 'N': 8192, 'K': 4096, 'rowMajorA': 'T', 'rowMajorB': 'N', 'BLOCK_SIZE_M': 256, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 64, 'GROUP_SIZE_M': 4, 'SPLIT_K': 1, 'num_warps': 8, 'num_stages': 0, 'waves_per_eu': 0, 'matrix_instr_nonkdim': 16, 'kpack': 2}
Expand All @@ -22,4 +19,6 @@
- {'M': 9728, 'N': 8192, 'K': 4288, 'rowMajorA': 'T', 'rowMajorB': 'N', 'BLOCK_SIZE_M': 256, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 64, 'GROUP_SIZE_M': 4, 'SPLIT_K': 1, 'num_warps': 8, 'num_stages': 0, 'waves_per_eu': 0, 'matrix_instr_nonkdim': 16, 'kpack': 2}



- {'M': 9728, 'N': 8192, 'K': 8256, 'rowMajorA': 'T', 'rowMajorB': 'N', 'BLOCK_SIZE_M': 256, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 64, 'GROUP_SIZE_M': 4, 'SPLIT_K': 1, 'num_warps': 8, 'num_stages': 0, 'waves_per_eu': 0, 'matrix_instr_nonkdim': 16, 'kpack': 2}
- {'M': 9728, 'N': 8192, 'K': 8320, 'rowMajorA': 'T', 'rowMajorB': 'N', 'BLOCK_SIZE_M': 256, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 64, 'GRO UP_SIZE_M': 4, 'SPLIT_K': 1, 'num_warps': 8, 'num_stages': 0, 'waves_per_eu': 0, 'matrix_instr_nonkdim': 16, 'kpack': 2}
- {'M': 9728, 'N': 8192, 'K': 8384, 'rowMajorA': 'T', 'rowMajorB': 'N', 'BLOCK_SIZE_M': 256, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 64, 'GROUP_SIZE_M': 4, 'SPLIT_K': 1, 'num_warps': 8, 'num_stages': 0, 'waves_per_eu': 0, 'matrix_instr_nonkdim': 16, 'kpack': 2}

0 comments on commit db1424e

Please sign in to comment.