furyhawk/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-07-21 10:15:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

disable NCCL path when virtual devices are used

Browse Source
This commit is contained in:
Anav Prasad
2026-07-08 13:56:10 +00:00
parent a8eca04c3f
commit 2976e33c04
1 changed files with 9 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/ggml-cuda.cu
+9
View File
@@ -1164,6 +1164,15 @@ static void ggml_backend_cuda_comm_init_internal(ggml_backend_cuda_comm_context
static void ggml_backend_cuda_comm_init_nccl(ggml_backend_cuda_comm_context * ret) {
#ifdef GGML_USE_NCCL
// Disabling NCCL path when CUDA virtual devices are in use since NCCL requires one distinct physical GPU per rank.
const ggml_cuda_device_info & info = ggml_cuda_info();
if (info.device_count > info.physical_device_count) {
GGML_LOG_WARN("NCCL disabled: virtual devices in use; "
"falling back to internal AllReduce\n");
ggml_backend_cuda_comm_init_internal(ret);
return;
}
const size_t n = ret->dev_ids.size();
ret->comms.resize(n);
ncclResult_t rc = ncclCommInitAll(ret->comms.data(), (int) n, ret->dev_ids.data());