echristo accepted this revision.
echristo added a comment.
This revision is now accepted and ready to land.
A couple of inline comments, but this is fine for now.
We need a command line option to select CUDA version as well.
-eric
================
Comment at: lib/Driver/ToolChains.cpp:1635
@@ -1634,2 +1634,3 @@
CudaPathCandidates.push_back(D.SysRoot + "/usr/local/cuda");
+ CudaPathCandidates.push_back(D.SysRoot + "/usr/local/cuda-7.5");
CudaPathCandidates.push_back(D.SysRoot + "/usr/local/cuda-7.0");
----------------
Commit this part separately as an "add a newer version of cuda that we support".
================
Comment at: lib/Driver/ToolChains.cpp:4082
@@ +4081,3 @@
+ // than LLVM defaults to. Use PTX4.2 which is the PTX version that
+ // came with CUDA-7.0.
+ CC1Args.push_back("-target-feature");
----------------
Explain in a bit more detail here, I understand why, but it's oddly complicated
and annoying.
We should look into making this part of TargetInfo with the cuda version being
a feature that then turns on associated minimal cpu features etc.
http://reviews.llvm.org/D14556
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
