I've added a test with a relative -o path, but there's still some of
uncovered ground here. For example, I can't verify that the default case
emits the .gcno file to the CWD rather than the directory of the source
file (which was a bug I did have at first) since it's an absolute path by
the time is reaches FileCheck, and afaik there's no way to tell FileCheck
what the CWD is.

On Wed, Oct 8, 2014 at 11:15 PM, Dan Albert <[email protected]> wrote:

> Add an additional test with a relative -o path.
>
> http://reviews.llvm.org/D5697
>
> Files:
>   lib/Driver/Tools.cpp
>   test/Driver/clang_f_opts.c
>
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to