Enable local sendrecv over network if GDR is available on all GPUs (#324)

[ROCm/rccl commit: c018edf0f2]
2021-03-05 19:59:41 -08:00
parent 8d90821062
commit bcf4ecb0e3
14 changed files with 163 additions and 44 deletions
@@ -82,8 +82,9 @@ struct ncclTopoLinkList {

 #define NCCL_TOPO_UNDEF (-1)

-#define RCCL_TOPO_CR8G 1
+#define RCCL_TOPO_CR8G      1
 #define RCCL_TOPO_4P2H_ROME 2
+#define RCCL_TOPO_GDR_ALL   4

 struct ncclTopoNode {
  int type;
@@ -132,6 +133,7 @@ struct ncclTopoSystem {
  float maxWidth;
  float totalWidth;
  int type;
+  int nRanks;
 };

 ncclResult_t ncclTopoGetNode(struct ncclTopoSystem* system, struct ncclTopoNode** node, int type, uint64_t id);