ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-01-31 08:25:20 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

ggml : add NVPL BLAS support (ggml/8329) (llama/8425)

Browse Source 
* ggml : add NVPL BLAS support

* ggml : replace `<BLASLIB>_ENABLE_CBLAS` with `GGML_BLAS_USE_<BLASLIB>`

---------

Co-authored-by: ntukanov <ntukanov@nvidia.com>
This commit is contained in:
Nicholai Tukanov 2024-07-11 11:49:15 -05:00 committed by Georgi Gerganov
parent 84493d7f3e
commit 5a8ae474f0
1 changed files with 9 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
ggml/src/ggml-blas.cpp
View File

@ -8,11 +8,12 @@
# include <Accelerate/Accelerate.h>
#elif defined(GGML_BLAS_USE_MKL)
# include <mkl.h>
#elif defined(GGML_BLAS_USE_BLIS)
# include <blis.h>
#elif defined(GGML_BLAS_USE_NVPL)
# include <nvpl_blas.h>
#else
# include <cblas.h>
# ifdef BLIS_ENABLE_CBLAS
# include <blis.h>
# endif
#endif
struct ggml_backend_blas_context {
@ -140,10 +141,14 @@ static void ggml_backend_blas_mul_mat(ggml_backend_blas_context * ctx, struct gg
openblas_set_num_threads(ctx->n_threads);
#endif
#if defined(BLIS_ENABLE_CBLAS)
#if defined(GGML_BLAS_USE_BLIS)
bli_thread_set_num_threads(ctx->n_threads);
#endif
#if defined(GGML_BLAS_USE_NVPL)
nvpl_blas_set_num_threads(ctx->n_threads);
#endif
for (int64_t i13 = 0; i13 < ne13; i13++) {
for (int64_t i12 = 0; i12 < ne12; i12++) {
const int64_t i03 = i13/r3;