AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Updating RCCL based on NCCL 2.3.7

Browse Source 
- Contains modifications to support AMD hardware
- Adds unit tests


[ROCm/rccl commit: 55a4b22ad7]
This commit is contained in:
Gilbert Lee
2019-05-16 16:16:18 +00:00
parent 3a7974d4c4
commit 60f91f645d
79 changed files with 2678 additions and 399 deletions
Download Patch File Download Diff File Expand all files Collapse all files
projects/rccl/src/include/enqueue.h
+3 -2
View File
@@ -1,5 +1,6 @@
/*************************************************************************
* Copyright (c) 2015-2018, NVIDIA CORPORATION. All rights reserved.
* Modifications Copyright (c) 2019 Advanced Micro Devices, Inc. All rights reserved.
*
* See LICENSE.txt for license information
************************************************************************/
@@ -11,11 +12,11 @@
#include "group.h"
typedef ncclResult_t(*ncclFunc_t)(const void* sendbuff, void* recvbuff, size_t count,
ncclDataType_t type, ncclRedOp_t op, int root, ncclComm_t comm, cudaStream_t stream);
ncclDataType_t type, ncclRedOp_t op, int root, ncclComm_t comm, hipStream_t stream);
ncclResult_t ncclEnqueueCheck(ncclFunc_t func, const char* primName, const void* sendbuff,
void* recvbuff, size_t count, ncclDataType_t type, ncclRedOp_t op, int root,
ncclComm_t comm, cudaStream_t stream);
ncclComm_t comm, hipStream_t stream);
ncclResult_t ncclCpuBarrierIn(ncclComm_t comm, int* isLast);
ncclResult_t ncclCpuBarrierLast(ncclComm_t comm);
ncclResult_t ncclCpuBarrierOut(ncclComm_t comm);