[llvm] [Offload] Use newer CUDA API functions when dynamically loaded (PR #93057)
via llvm-commits
llvm-commits at lists.llvm.org
Wed May 22 08:53:50 PDT 2024
github-actions[bot] wrote:
<!--LLVM CODE FORMAT COMMENT: {clang-format}-->
:warning: C/C++ code formatter, clang-format found issues in your code. :warning:
<details>
<summary>
You can test this locally with the following command:
</summary>
``````````bash
git-clang-format --diff 831d1435193e73026e03e006a5b86591f3d202b3 6a889513fa82de4a3a0ba5aace4f59e38ab5db69 -- offload/plugins-nextgen/cuda/dynamic_cuda/cuda.cpp offload/plugins-nextgen/cuda/dynamic_cuda/cuda.h
``````````
</details>
<details>
<summary>
View the diff from clang-format here.
</summary>
``````````diff
diff --git a/offload/plugins-nextgen/cuda/dynamic_cuda/cuda.h b/offload/plugins-nextgen/cuda/dynamic_cuda/cuda.h
index 5bae3539b7..d156831381 100644
--- a/offload/plugins-nextgen/cuda/dynamic_cuda/cuda.h
+++ b/offload/plugins-nextgen/cuda/dynamic_cuda/cuda.h
@@ -16,14 +16,14 @@
#include <cstddef>
#include <cstdint>
-#define cuDeviceTotalMem cuDeviceTotalMem_v2
-#define cuModuleGetGlobal cuModuleGetGlobal_v2
-#define cuMemGetInfo cuMemGetInfo_v2
-#define cuMemAlloc cuMemAlloc_v2
-#define cuMemFree cuMemFree_v2
-#define cuMemAllocHost cuMemAllocHost_v2
-#define cuDevicePrimaryCtxRelease cuDevicePrimaryCtxRelease_v2
-#define cuDevicePrimaryCtxSetFlags cuDevicePrimaryCtxSetFlags_v2
+#define cuDeviceTotalMem cuDeviceTotalMem_v2
+#define cuModuleGetGlobal cuModuleGetGlobal_v2
+#define cuMemGetInfo cuMemGetInfo_v2
+#define cuMemAlloc cuMemAlloc_v2
+#define cuMemFree cuMemFree_v2
+#define cuMemAllocHost cuMemAllocHost_v2
+#define cuDevicePrimaryCtxRelease cuDevicePrimaryCtxRelease_v2
+#define cuDevicePrimaryCtxSetFlags cuDevicePrimaryCtxSetFlags_v2
typedef int CUdevice;
typedef uintptr_t CUdeviceptr;
``````````
</details>
https://github.com/llvm/llvm-project/pull/93057
More information about the llvm-commits
mailing list