root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 10:54:36 +00:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

CUDA: fix peer access logic (#3231)

Browse Source
This commit is contained in:
Johannes Gäßler 2023-09-17 23:35:20 +02:00 committed by GitHub
parent 111163e246
commit ee66942d7e
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 8 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
ggml-cuda.cu
View File

@ -6286,12 +6286,14 @@ void ggml_cuda_set_peer_access(const int n_tokens) {
continue;
}
int canAccessPeer;
CUDA_CHECK(cudaDeviceCanAccessPeer(&canAccessPeer, id, id_other));
if (enable_peer_access) {
CUDA_CHECK(cudaDeviceEnablePeerAccess(id_other, 0));
} else {
CUDA_CHECK(cudaDeviceDisablePeerAccess(id_other));
int can_access_peer;
CUDA_CHECK(cudaDeviceCanAccessPeer(&can_access_peer, id, id_other));
if (can_access_peer) {
if (enable_peer_access) {
CUDA_CHECK(cudaDeviceEnablePeerAccess(id_other, 0));
} else {
CUDA_CHECK(cudaDeviceDisablePeerAccess(id_other));
}
}
}
}