r207808 - do not hide clang stderr output during a test

Reid Kleckner rnk at google.com
Thu May 1 17:04:31 PDT 2014


On Thu, May 1, 2014 at 4:56 PM, David Blaikie <dblaikie at gmail.com> wrote:

> On Thu, May 1, 2014 at 4:49 PM, Lubos Lunak <l.lunak at centrum.cz> wrote:
> > On Friday 02 of May 2014, David Blaikie wrote:
> >> On Thu, May 1, 2014 at 3:40 PM, Lubos Lunak <l.lunak at centrum.cz> wrote:
> >> > Author: llunak
> >> > Date: Thu May  1 17:40:00 2014
> >> > New Revision: 207808
> >> >
> >> > URL: http://llvm.org/viewvc/llvm-project?rev=207808&view=rev
> >> > Log:
> >> > do not hide clang stderr output during a test
> >> >
> >> > I don't know why this fails on the buildbot.
> >>
> >> What's the buildbot failure look like?
> >
> >  Until I did this commit, "Exit Code: 1".
> >
> >> This is probably not a change to the test that we should keep in
> >> perpetuity - it'd be nice to understand/fix the issue, or XFAIL or
> >> certain platforms, etc, if that's the right answer.
> >
> > "clang: error: unable to execute command: Executable "hexagon-as" doesn't
> > exist!"
> > (
> http://lab.llvm.org:8011/builders/clang-hexagon-elf/builds/11260/steps/check-all/logs/Clang::rewrite-includes-messages.c
> )
> >
> >  I suppose I should have used -fsyntax-only for the compile. But r207810
> has
> > already fixed the problem somehow.
>
> If you could add -fsyntax-only to this, that'd be nice - no need to
> run extra backend work when it's not needed.
>
> Also, since this isn't a driver test, it should be using %clang_cc1 not
> %clang.


Both are already done in r207810. :)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20140501/7d271423/attachment.html>


More information about the cfe-commits mailing list