Check for fine grain support using memory allocation
This commit is contained in:
@@ -245,8 +245,6 @@ end:
|
||||
return dev;
|
||||
}
|
||||
|
||||
extern bool useFineGrainVramPcie;
|
||||
|
||||
NCCL_PARAM(NetGdrRead, "NET_GDR_READ", -2);
|
||||
NCCL_PARAM(NetGdrLevel, "NET_GDR_LEVEL", PATH_PHB);
|
||||
|
||||
@@ -257,7 +255,7 @@ static ncclResult_t netGetGdrSupport(int dev, int read, int* useGdr) {
|
||||
CUDACHECK(hipGetDevice(&cudaDev));
|
||||
NCCLCHECK(getNvmlDevice(cudaDev, &nvmlDev))
|
||||
|
||||
if (!useFineGrainVramPcie) {
|
||||
if (!hasFineGrainVramPcie()) {
|
||||
INFO(NCCL_INIT|NCCL_NET,"NET/%s : GPU Direct RDMA Disabled for GPU %d / Need Fine Grain VRAM over PCIe", ncclNetName(), cudaDev);
|
||||
return ncclSuccess;
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user