2.18.1-1

Add support for IB SHARP to NVLS (NVLink SHARP algorithm).
Add NVLS+Tree algorithm.
Add support for memory management using cuMem* functions.
Use all NICs for Send/Receive operations on systems with more than
one NIC per GPU (#804).
Add ncclCommSplit primitive, with resource sharing option in config.
Fix alltoallv hang (#788)
Increase number of channels on H100 when we're not limited by NVLink.
Improve error reporting in case of IB failure, printing local and
remote ID (#779).
Add build option to allow compilation against RDMA includes instead
of dynamically loading IB verbs symbols (#802).
Fix context creation for progress thread (#803).
NET/IB: add option to use multiple QPs in round-robin mode.
Fix tree performance issue when NVB is disabled on HCM topologies.


[ROCm/rccl commit: d97a32fac8]

projects/rccl/src/include/p2p.h

@@ -9,4 +9,21 @@
 #ifndef NCCL_P2P_H_
 #define NCCL_P2P_H_
 
+#define NCCL_P2P_HANDLE_TYPE CU_MEM_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR
+
+typedef struct {
+  int data; // Currently only support an fd based descriptor
+} ncclCuDesc;
+
+typedef union {
+  // Legacy CUDA IPC
+  cudaIpcMemHandle_t devIpc;
+  // cuMem API support
+  ncclCuDesc cuDesc;
+} ncclIpcDesc;
+
+ncclResult_t ncclP2pAllocateShareableBuffer(size_t size, ncclIpcDesc *ipcDesc, void **ptr);
+ncclResult_t ncclP2pFreeShareableBuffer(ncclIpcDesc *ipcDesc);
+ncclResult_t ncclP2pImportShareableBuffer(struct ncclComm *comm, int tpPeer, size_t size, ncclIpcDesc *ipcDesc, void **devMemPtr);
+
 #endif