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

ggml : check compute for fusion

This commit is contained in:
Georgi Gerganov 2026-01-16 17:52:21 +02:00
parent 9696c3e748
commit c12ec4a395
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
2 changed files with 7 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
ggml/src/ggml-impl.h
View File

@ -611,6 +611,9 @@ static inline bool ggml_can_fuse_ext(const struct ggml_cgraph * cgraph, const in
if (node->op != ops[i]) {
return false;
}
if ((node->flags & GGML_TENSOR_FLAG_COMPUTE) == 0) {
return false;
}
if (i < num_ops - 1 && !ggml_node_has_n_uses(cgraph, node_idxs[i], 1)) {
return false;
}

4
ggml/src/ggml.c
View File

@ -7259,6 +7259,10 @@ bool ggml_can_fuse_subgraph_ext(const struct ggml_cgraph * cgraph,
return false;
}
if ((node->flags & GGML_TENSOR_FLAG_COMPUTE) == 0) {
return false;
}
if (ggml_node_list_find_tensor(cgraph, outputs, num_outputs, node) != -1) {
continue;
}