vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llamafile : disable sgemm for batch-size 1 (#9330)

Browse source
This commit is contained in:
Eve 2024-09-07 19:02:26 +00:00 • committed by GitHub
parent 1b9ae5189c
commit e536426ded
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 4 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

4
ggml/src/llamafile/sgemm.cpp
View file

@ -1006,6 +1006,10 @@ bool llamafile_sgemm(int64_t m, int64_t n, int64_t k, const void *A, int64_t lda
assert(nth > 0); assert(nth > 0);
assert(ith < nth); assert(ith < nth);
// only enable sgemm for prompt processing
if (n < 2)
return false;
if (Ctype != GGML_TYPE_F32) if (Ctype != GGML_TYPE_F32)
return false; return false;