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

check for sycl blas, better performance

Browse source
This commit is contained in:
jianyuzh 2024-01-23 13:34:05 +08:00
parent 67e6b3cb7d
commit 533c647d0e
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama.cpp
View file

@ -6703,7 +6703,7 @@ static int llama_decode_internal(
}
const bool fully_offloaded = model.n_gpu_layers >= (int) hparams.n_layer + 1;
if (ggml_cpu_has_cublas() && fully_offloaded) {
if ((ggml_cpu_has_cublas() || ggml_cpu_has_sycl()) && fully_offloaded) {
n_threads = 1;
}