On Fri, Feb 14, 2020 at 6:38 AM Christoph Junghans <junghans(a)votca.org>
wrote:
On Fri, Feb 14, 2020, 00:16 Petr Pisar <ppisar(a)redhat.com> wrote:
>
> On Thu, Feb 13, 2020 at 07:48:22PM -0700, Christoph Junghans wrote:
> > when running "mock -r fedora-rawhide-ppc64le --no-clean
> > gromacs-2019.5-2.fc32.1.src.rpm"
> > I am getting the following error:
> > Signal 4 (ILL) caught by ps (3.3.15).
> > /usr/bin/ps:ps/display.c:66: please report this bug
> >
> > I tried a couple of different src.rpm with the same result.
> > This is mock 1.4.21 on f31.
> >
> > Has anybody else seen this?
> >
> Not exactly, but if you check Koji build history for gromacs package
> <
https://koji.fedoraproject.org/koji/packageinfo?packageID=6961>. You
will
> find out that the build failed during F32 mass rebuild
> <
https://koji.fedoraproject.org/koji/taskinfo?taskID=41318028> on
ppc64le
and
> aarch64. The ppc64le failure is:
That was exactly the reason why I was looking into building this ppc64le
;-) But
unfortunately the Signal 4 shows up before it even gets to the
CMake part.
>
> -- Performing Test CXX_mvsx_COMPILE_WORKS - Failed
> -- Flag was accepted, but it did not build test source (this could be
due to
either the compiler or binutils)
> -- Performing Test CXX_maltivec_mabi_altivec_FLAG_ACCEPTED
> -- Performing Test CXX_maltivec_mabi_altivec_FLAG_ACCEPTED - Success
> -- Performing Test CXX_maltivec_mabi_altivec_COMPILE_WORKS
> -- Performing Test CXX_maltivec_mabi_altivec_COMPILE_WORKS - Failed
> -- Flag was accepted, but it did not build test source (this could be
due to
either the compiler or binutils)
> -- Performing Test CXX_qarch_auto_qaltivec_FLAG_ACCEPTED
> -- Performing Test CXX_qarch_auto_qaltivec_FLAG_ACCEPTED - Failed
> -- Performing Test CXX_COMPILE_WORKS_WITHOUT_SPECIAL_FLAGS
> -- Performing Test CXX_COMPILE_WORKS_WITHOUT_SPECIAL_FLAGS - Failed
> -- Could not find any flag to build test source (this could be due to
either
the compiler or binutils)
> CMake Error at cmake/gmxManageSimd.cmake:51 (message):
> Cannot find IBM VSX compiler flag. Use a newer compiler, or disable
SIMD
> support (slower).
> Call Stack (most recent call first):
> cmake/gmxManageSimd.cmake:265
(gmx_give_fatal_error_when_simd_support_not_found)
> CMakeLists.txt:719 (gmx_manage_simd)
> -- Configuring incomplete, errors occurred!
> See also
"/builddir/build/BUILD/gromacs-2019.5/serial/CMakeFiles/CMakeOutput.log".
> See also
"/builddir/build/BUILD/gromacs-2019.5/serial/CMakeFiles/CMakeError.log".
claims the
> build started to fail with these changes
> <
https://koschei.fedoraproject.org/build/7747640>. It's probably a
triggered by
> an upgrade of GCC to 10 version.
>
> SIGILL means the processor met an instruction it does not understand.
>
> In your case procps-ng package seems to be miscompiled (or run on an
> incopatible hardware). I recommend getting a shell in the mock enviroment
> (mock --shell) and investigeting whether the ps program works there).
Ok, I will try that and report back.
It is indeed broken in shell as well:
<mock-chroot> sh-5.0# ps
Signal 4 (ILL) caught by ps (3.3.15).
/usr/bin/ps:ps/display.c:66: please report this bug
Segmentation fault (core dumped)
>
> My gut feeling is that GCC 10 started to omit new instructions and your
> hardware is not compatible with them. Can you reproduce it on one of
these
> machines
> <
https://fedoraproject.org/wiki/Test_Machine_Resources_For_Package_Maintai...
?
Using these machines might be the easiest way the track down the gromacs
issue,
thanks for the reminder that we have these resources.
Christoph
>
> -- Petr
> _______________________________________________
> devel mailing list -- devel(a)lists.fedoraproject.org
> To unsubscribe send an email to devel-leave(a)lists.fedoraproject.org
> Fedora Code of Conduct:
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
--
Christoph Junghans
Web:
http://www.compphys.de
--
Christoph Junghans
Web:
http://www.compphys.de