squiid/llvm
1
0
Fork 0
Code Issues Pull requests Activity

Revert "Fix CUDA runtime wrapper for GPU mem alloc/free to async"

Browse source 
This reverts commit b4117fede2.
This broke one of the MLIR bot, a test is failing.
This commit is contained in:
Mehdi Amini 2022-04-12 06:50:27 +00:00
parent 1889170aeb
commit 6b7e6ea489
1 changed files with 2 additions and 21 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

23
mlir/lib/ExecutionEngine/CudaRuntimeWrappers.cpp
View file

@ -18,12 +18,6 @@
#include "cuda.h"
// We need to know the CUDA version to determine how to map some of the runtime
// calls below.
#if !defined(CUDA_VERSION)
#error "cuda.h did not define CUDA_VERSION"
#endif
#ifdef _WIN32
#define MLIR_CUDA_WRAPPERS_EXPORT __declspec(dllexport)
#else
@ -140,28 +134,15 @@ extern MLIR_CUDA_WRAPPERS_EXPORT "C" void mgpuEventRecord(CUevent event,
CUDA_REPORT_IF_ERROR(cuEventRecord(event, stream));
}
extern "C" void *mgpuMemAlloc(uint64_t sizeBytes, CUstream stream) {
extern "C" void *mgpuMemAlloc(uint64_t sizeBytes, CUstream /*stream*/) {
ScopedContext scopedContext;
CUdeviceptr ptr;
#if CUDA_VERSION >= 11020
// Use the async version that was available since CUDA 11.2.
CUDA_REPORT_IF_ERROR(cuMemAllocAsync(&ptr, sizeBytes, stream));
#else
CUDA_REPORT_IF_ERROR(cuMemAlloc(&ptr, sizeBytes));
(void)stream;
#endif
return reinterpret_cast<void *>(ptr);
}
extern "C" void mgpuMemFree(void *ptr, CUstream stream) {
#if CUDA_VERSION >= 11020
// Use the async version that was available since CUDA 11.2.
CUDA_REPORT_IF_ERROR(
cuMemFreeAsync(reinterpret_cast<CUdeviceptr>(ptr), stream));
#else
extern "C" void mgpuMemFree(void *ptr, CUstream /*stream*/) {
CUDA_REPORT_IF_ERROR(cuMemFree(reinterpret_cast<CUdeviceptr>(ptr)));
(void)stream;
#endif
}
extern "C" void mgpuMemcpy(void *dst, void *src, size_t sizeBytes,