Implement NIC identification and remapping (#420)

* Add 1H16P GPU model

* Implement NIC identification and remapping

* Revert "Sort IB devices based on device name (#413)"

This reverts commit 2d0ed8dff6.

* Fix permute and check order

* Correction on IB speed reporting

* Revert "Allow user to link layer with RCCL_IB_HCA_SKIP_LINK_LAYER (#361)"

This reverts commit caf5c9992a.
This commit is contained in:
Wenkai Du
2021-08-24 09:42:04 -07:00
zatwierdzone przez GitHub
rodzic 5f15ed6e3e
commit 5c8380ff5b
11 zmienionych plików z 623 dodań i 47 usunięć
+2 -1
Wyświetl plik
@@ -22,8 +22,9 @@ THE SOFTWARE.
#ifndef RCCL_ROME_MODELS_H_
#define RCCL_ROME_MODELS_H_
ncclResult_t parseGraph(const char* str, struct ncclTopoSystem* system, struct ncclTopoGraph* graph, int* gpu_map);
ncclResult_t parseGraph(const char* str, struct ncclTopoSystem* system, struct ncclTopoGraph* graph, int* gpu_map, int* net_map);
ncclResult_t parseRome4P2H(struct ncclTopoSystem* system, struct ncclTopoGraph* graph);
ncclResult_t parseChordalRing(struct ncclTopoSystem* system, struct ncclTopoGraph* graph);
ncclResult_t parse1H16P(struct ncclTopoSystem* system, struct ncclTopoGraph* graph);
#endif