diff options
| author | Andrew Godfrey <AndrewGodfrey@users.noreply.github.com> | 2023-11-17 00:01:15 -0800 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-11-17 10:01:15 +0200 |
| commit | b83e149ec6264d078e6a47412e7347bf5c2bfcc9 (patch) | |
| tree | 395a0764606ee5550a3ee9c2f6b2da9879f4bbec | |
| parent | 4f447a48339977073a1af4f33ae873465ff64994 (diff) | |
cuda : get_row_rounding F32 (#4095)
* Fix #4017
* Update ggml-cuda.cu
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
* Update ggml-cuda.cu
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
---------
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
| -rw-r--r-- | ggml-cuda.cu | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/ggml-cuda.cu b/ggml-cuda.cu index 9aa61fe4..874ad9ac 100644 --- a/ggml-cuda.cu +++ b/ggml-cuda.cu @@ -6356,6 +6356,7 @@ static int64_t get_row_rounding(ggml_type type) { case GGML_TYPE_Q8_0: return max_compute_capability >= CC_RDNA2 ? 128 : 64; case GGML_TYPE_F16: + case GGML_TYPE_F32: return 1; case GGML_TYPE_Q2_K: return max_compute_capability >= CC_RDNA2 ? 128 : 32; @@ -6378,6 +6379,7 @@ static int64_t get_row_rounding(ggml_type type) { case GGML_TYPE_Q8_0: return 64; case GGML_TYPE_F16: + case GGML_TYPE_F32: return 1; case GGML_TYPE_Q2_K: case GGML_TYPE_Q3_K: |