Do not use _Pragma in code blocks with CUDA
The current CUDA compiler is crashing when using _Pragma in code blocks, which were being used with clang on apple. Turn these off when compiling with CUDA.
This commit is contained in:
parent
fbddba74a1
commit
bfb26cf71b
@ -31,7 +31,7 @@ VTKM_THIRDPARTY_POST_INCLUDE
|
||||
// pragmas. However, not all compiler support pragmas in code blocks (although
|
||||
// fortunately clang does). Thus, only use these pragmas in the instance where
|
||||
// we need it and know we can use it.
|
||||
#if defined(VTKM_CLANG) && (__apple_build_version__ >= 7000072)
|
||||
#if defined(VTKM_CLANG) && (__apple_build_version__ >= 7000072) && !defined(VTKM_CUDA)
|
||||
|
||||
#define VTKM_STATIC_ASSERT(condition) \
|
||||
VTKM_THIRDPARTY_PRE_INCLUDE \
|
||||
|
Loading…
Reference in New Issue
Block a user