: unassigned at gcc dot gnu.org
Reporter: nico...@morey-chaisemartin.com
Target Milestone: ---
I found what seems to be a GCC issue while building/running the OpenMPI 2.1.0
testsuite using GCC 6.3.1 on i586 (Suse Tumbleweed)
I attached the tarball with the preprocessed source file and
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80640
--- Comment #2 from Nicolas Morey-Chaisemartin
---
Created attachment 41325
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41325&action=edit
Test case
Previous tarball was too big. I stripped all debug info from the lib and it
should work
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80640
--- Comment #4 from Nicolas Morey-Chaisemartin
---
I agree the volatile shoud fix thing> I'll have to see with the ompi guys to
fix that.
But shouldn't __atomic_thread_fence () have a side effect here and force the
memory to be reloaded ?
If it
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80640
--- Comment #6 from Nicolas Morey-Chaisemartin
---
Ok. So there's something wrong :)
I'll make a work around for SUSE while waiting for a fix in GCC
