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

cuda : fix nkvo, offload and cuda graph node properties matching (#19165)

Browse Source 
* cuda : fix nkvo

* cont : more robust cuda graph node property matching

* cont : restore pre-leafs implementation

* cont : comments + static_assert
This commit is contained in:
Georgi Gerganov
2026-01-29 18:45:30 +02:00
committed by GitHub
parent 7b7ae857f6
commit 4fdbc1e4db
4 changed files with 59 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/fattn.cu
-5
View File
@@ -310,8 +310,6 @@ static best_fattn_kernel ggml_cuda_get_best_fattn_kernel(const int device, const
}
}
const bool V_is_K_view = V->view_src && (V->view_src == K || (V->view_src == K->view_src && V->view_offs == K->view_offs));
const int cc = ggml_cuda_info().devices[device].cc;
switch (K->ne[0]) {
@@ -334,9 +332,6 @@ static best_fattn_kernel ggml_cuda_get_best_fattn_kernel(const int device, const
if (!gqa_opt_applies) {
return BEST_FATTN_KERNEL_NONE;
}
if (!V_is_K_view) {
return BEST_FATTN_KERNEL_NONE;
}
break;
default:
return BEST_FATTN_KERNEL_NONE;