ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-04-30 11:21:56 +00:00
Code Issues Packages Projects Releases Wiki Activity

Prevent FlashMLA-1 from running on CUDA (#268)

Browse Source 
as it is not supported.

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
This commit is contained in:
Kawrakow
2025-03-19 13:03:59 +01:00
committed by GitHub
parent 8e549b4234
commit c3b75c531c
1 changed files with 3 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
ggml/src/ggml-cuda.cu
View File

@@ -3579,6 +3579,9 @@ GGML_CALL static bool ggml_backend_cuda_supports_op(ggml_backend_t backend, cons
return (op->src[1]->type == GGML_TYPE_F16 && op->src[2]->type == GGML_TYPE_F16) || return (op->src[1]->type == GGML_TYPE_F16 && op->src[2]->type == GGML_TYPE_F16) ||
(op->src[1]->type == GGML_TYPE_Q8_0 && op->src[2]->type == GGML_TYPE_Q8_0); (op->src[1]->type == GGML_TYPE_Q8_0 && op->src[2]->type == GGML_TYPE_Q8_0);
} }
if (op->src[1]->ne[0] > 256) {
return false;
}
if (op->src[0]->ne[0] == 64 && op->src[1]->type == GGML_TYPE_F16) { if (op->src[0]->ne[0] == 64 && op->src[1]->type == GGML_TYPE_F16) {
return true; return true;
} }