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

cuda : fix "V is K view" check for non-unified KV cache (#19145)

Browse Source
This commit is contained in:
Georgi Gerganov
2026-01-28 09:15:27 +02:00
committed by GitHub
parent 2eee6c866c
commit 631cbfcc7a
3 changed files with 5 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/fattn-common.cuh
+1 -1
View File
@@ -789,7 +789,7 @@ void launch_fattn(
const ggml_tensor * K = dst->src[1];
const ggml_tensor * V = dst->src[2];
const bool V_is_K_view = V->view_src && V->view_offs == 0 && (V->view_src == K || V->view_src == K->view_src);
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 ggml_tensor * mask = dst->src[3];
const ggml_tensor * sinks = dst->src[4];