Wenkai Du
40 lines
1.5 KiB
C++
40 lines
1.5 KiB
C++
/*************************************************************************
|
|
* Copyright (c) 2015-2022, NVIDIA CORPORATION. All rights reserved.
|
|
*
|
|
* See LICENSE.txt for license information
|
|
************************************************************************/
|
|
|
|
#include "enqueue.h"
|
|
#include "collectives.h"
|
|
#include "argcheck.h" // Need some checks here since we access comm
|
|
|
|
NCCL_API(ncclResult_t, ncclSend, const void* sendbuff, size_t count, ncclDataType_t datatype, int peer,
|
|
ncclComm_t comm, cudaStream_t stream);
|
|
ncclResult_t ncclSend(const void* sendbuff, size_t count, ncclDataType_t datatype, int peer,
|
|
ncclComm_t comm, cudaStream_t stream) {
|
|
NVTX3_FUNC_RANGE_IN(nccl_domain);
|
|
struct ncclInfo info = { ncclFuncSend, "Send",
|
|
NULL, (void*)sendbuff, count, datatype, ncclSum, peer, comm, stream, /* Args */
|
|
1, 1 };
|
|
ncclResult_t ret;
|
|
NCCLCHECK(ncclGroupStart());
|
|
ret = ncclEnqueueCheck(&info);
|
|
NCCLCHECK(ncclGroupEnd());
|
|
return ret;
|
|
}
|
|
|
|
NCCL_API(ncclResult_t, ncclRecv, void* recvbuff, size_t count, ncclDataType_t datatype, int peer,
|
|
ncclComm_t comm, cudaStream_t stream);
|
|
ncclResult_t ncclRecv(void* recvbuff, size_t count, ncclDataType_t datatype, int peer,
|
|
ncclComm_t comm, cudaStream_t stream) {
|
|
NVTX3_FUNC_RANGE_IN(nccl_domain);
|
|
struct ncclInfo info = { ncclFuncRecv, "Recv",
|
|
NULL, recvbuff, count, datatype, ncclSum, peer, comm, stream, /* Args */
|
|
1, 1 };
|
|
ncclResult_t ret;
|
|
NCCLCHECK(ncclGroupStart());
|
|
ret = ncclEnqueueCheck(&info);
|
|
NCCLCHECK(ncclGroupEnd());
|
|
return ret;
|
|
}
|