Sylvain Jeaugey
d97a32fac8
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.
30 líneas
912 B
C
30 líneas
912 B
C
/*************************************************************************
|
|
* Copyright (c) 2015-2020, NVIDIA CORPORATION. All rights reserved.
|
|
*
|
|
* See LICENSE.txt for license information
|
|
************************************************************************/
|
|
|
|
#include <stdlib.h>
|
|
|
|
#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
|