]> granicus.if.org Git - clang/commit
projects / clang / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: a2b5d62) | patch
[OpenMP] NVPTX: Set default/minimum compute capability to sm_35
authorGeorge Rokos <grokos@us.ibm.com>
Thu, 7 Dec 2017 20:27:31 +0000 (20:27 +0000)
committerGeorge Rokos <grokos@us.ibm.com>
Thu, 7 Dec 2017 20:27:31 +0000 (20:27 +0000)
commite9a55b7c48d5fa09d405ac413b11e1ea850ce2bb
treef60c6584b707a4607055577d99416e8c250c93e6tree | snapshot
parenta2b5d62923c59e0f7af9ba5f2635d5fb2cf40129commit | diff
[OpenMP] NVPTX: Set default/minimum compute capability to sm_35

The current implementation of the nvptx runtime (to be upstreamed shortly) uses the atomicMax operation on 64-bit integers.
This is only supported in compute capabilities 3.5 and later. I've changed the clang default to sm_35.

Differential Revision: https://reviews.llvm.org/D40977

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@320082 91177308-0d34-0410-b5e6-96231b3b80d8
CMakeLists.txt diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.