Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

HIP: enable mmf for RDNA3 (#17879)

Browse Source 
* enable mmf for RDNA3

* disable mmf for some shape

* move some mmvf to mmf

* more mmfv to mmf

* 3 is good in mmvf

---------

Co-authored-by: zhang hui <you@example.com>
This commit is contained in:
yulo
2025-12-12 11:34:33 +01:00
committed by GitHub
co-authored by zhang hui
parent a81a569577
commit c33a58bced
4 changed files with 84 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/mmvf.cu
+4 -1
View File
@@ -765,7 +765,10 @@ bool ggml_cuda_should_use_mmvf(enum ggml_type type, int cc, const int64_t * src0
return ne11 <= 8;
} else if (GGML_CUDA_CC_IS_AMD(cc)) {
if (fp16_mma_hardware_available(cc)) {
if (GGML_CUDA_CC_IS_RDNA3(cc) || GGML_CUDA_CC_IS_RDNA4(cc)) {
if (GGML_CUDA_CC_IS_RDNA3(cc)) {
return ne11 <= 3;
}
if (GGML_CUDA_CC_IS_RDNA4(cc)) {
return ne11 <= 5;
}
return ne11 <= 2;