ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-05-09 20:13:14 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

CUDA: fix FA logic for PTX 7.0 and CC >= 7.5 (llama/12222)

Browse Source
This commit is contained in:
Johannes Gäßler 2025-03-06 18:45:09 +01:00 committed by Georgi Gerganov
parent b9eab73fa2
commit e3c85e75bd
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/src/ggml-cuda/fattn.cu
View File

@ -310,7 +310,7 @@ void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst
}
// The MMA implementation needs Turing or newer, use the old WMMA code for Volta:
if (cc == GGML_CUDA_CC_VOLTA) {
if (fp16_mma_available(cc) && !new_mma_available(cc)) {
ggml_cuda_flash_attn_ext_wmma_f16(ctx, dst);
return;
}