Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

rocm enable for v2.0.1

Browse Source
This commit is contained in:
Hui Liu
2023-10-27 18:50:28 -07:00
parent 2e7e9d3b2d a408254c2f
commit 782b73f2bb
447 changed files with 13518 additions and 8719 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/common/device_helpers.hip.h
View File

@@ -38,6 +38,8 @@
#include "xgboost/logging.h"
#include "xgboost/span.h"
#include "cuda_to_hip.h"
#ifdef XGBOOST_USE_RCCL
#include "rccl.h"
#endif // XGBOOST_USE_RCCL
@@ -107,7 +109,7 @@ inline ncclResult_t ThrowOnNcclError(ncclResult_t code, const char *file, int li
if (code == ncclUnhandledCudaError) {
// nccl usually preserves the last error so we can get more details.
auto err = hipPeekAtLastError();
ss << " CUDA error: " << thrust::system_error(err, thrust::cuda_category()).what() << "\n";
ss << " CUDA error: " << thrust::system_error(err, thrust::hip_category()).what() << "\n";
} else if (code == ncclSystemError) {
ss << " This might be caused by a network configuration issue. Please consider specifying "
"the network interface for RCCL via environment variables listed in its reference: "