[Openmp-commits] [PATCH] D40384: [OMPT] Add annotations to testcases that are expected to fail when using certain compilers

[Jonas Hahnfeld via Phabricator via Openmp-commits]

Hahnfeld added inline comments.


================
Comment at: runtime/test/ompt/misc/control_tool.c:3
 // REQUIRES: ompt
+// XFAIL: gcc-4, gcc-5, gcc-6, gcc-7
 #include "callback.h"
----------------
protze.joachim wrote:
> Hahnfeld wrote:
> > sconvent wrote:
> > > Hahnfeld wrote:
> > > > This works on LLVM's buildbot, so this can't be entirely correct: http://lab.llvm.org:8011/builders/openmp-ompt-gcc-x86_64-linux-debian/builds/14/steps/test-libomp/logs/stdio
> > > > 
> > > > Why did you want to deactivate this?
> > > Which gcc version does the buildbot use? For me the testcase fails for versions up to 7.2 (I haven't tested more recent versions).
> > > The problem is that in those versions, gcc has problems with labels inside of OpenMP regions. So when using the "print_current_address" macro (which defines a label) in a parallel region, the compilation fails.
> > > Other testcases with the same problem also work in the buildbot so I believe this is a matter of the gcc version.
> > > 
> > GCC 7.2.0, probably the system default in Debian: http://lab.llvm.org:8011/builders/openmp-ompt-gcc-x86_64-linux-debian/builds/16/steps/configure-openmp/logs/stdio
> > 
> > Maybe the distribution carries some downstream patches?!?
> The context is this (fixed) issue: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81687
Yes, so the fix has been backported to GCC 5, 6, 7. This has happened after 7.2.0 has been released, but probably Debian tracks a more recent commit (which most distributions do unfortunately). So although I don't like we might need to turn this into an UNSUPPORTED because a passing XFAIL is treated as error :-(


https://reviews.llvm.org/D40384