happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Multi-stream batching fix

This commit is contained in:
Pavel Rykov 2026-03-30 23:36:10 +03:00
parent 1c6a5e61a6
commit 759db688ad
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/llama-graph.cpp
View File

@ -2222,7 +2222,7 @@ ggml_tensor * llm_graph_context::build_attn_sparse(
cb(kqv, "kqv", il); cb(kqv, "kqv", il);
ggml_tensor * cur = ggml_permute(ctx0, kqv, 0, 2, 1, 3); ggml_tensor * cur = ggml_permute(ctx0, kqv, 0, 2, 1, 3);
cur = ggml_cont_2d(ctx0, cur, cur->ne[0]*cur->ne[1], cur->ne[2]); cur = ggml_cont_2d(ctx0, cur, cur->ne[0]*cur->ne[1], cur->ne[2]*cur->ne[3]);
cb(cur, "kqv_out", il); cb(cur, "kqv_out", il);
if (wo) { if (wo) {