rsmith marked an inline comment as done. rsmith added inline comments.
================ Comment at: test/Misc/pragma-attribute-supported-attributes-list.test:27-32 +// CHECK-NEXT: CUDAConstant (SubjectMatchRule_variable) +// CHECK-NEXT: CUDADevice (SubjectMatchRule_function, SubjectMatchRule_variable) +// CHECK-NEXT: CUDAGlobal (SubjectMatchRule_function) +// CHECK-NEXT: CUDAHost (SubjectMatchRule_function) +// CHECK-NEXT: CUDALaunchBounds (SubjectMatchRule_objc_method, SubjectMatchRule_hasType_functionType) +// CHECK-NEXT: CUDAShared (SubjectMatchRule_variable) ---------------- tra wrote: > I don't see much practical use of this pragma for CUDA, but I also don't have > any specific objections. > LGTM. > > Theoretically we could use it to apply `__host__ __device__` attribute to > some portable headers-only library so we could use it on device side. In > practice, though, there usually will be few things that would have to remain > host-only (anything involving file-io, for example) and we would need to be > more selective in applying the attributes or have a way to remove them from a > subset of objects later on. > > > > > The pragma is already flexible enough to allow selectively applying attributes to a subset of the declarations in its scope: https://clang.llvm.org/docs/LanguageExtensions.html#subject-match-rules Repository: rC Clang https://reviews.llvm.org/D51507 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
