Justin Lebar f179364341 [CUDA] Conservatively mark inline asm as convergent. ...

Summary:
This is particularly important because a some convergent CUDA intrinsics
(e.g.  __shfl_down) are implemented in terms of inline asm.

Reviewers: tra

Subscribers: cfe-commits

Differential Revision: http://reviews.llvm.org/D20836

llvm-svn: 271336

2016-05-31 21:27:13 +00:00

81 KiB

Raw Blame History

View Raw