This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [patch 1/2] Re: Crash regression(?) printing Fortran strings in bt [Re: [V2 00/23] Fortran dynamic array support]
- From: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- To: Keven Boell <keven dot boell at linux dot intel dot com>
- Cc: Keven Boell <keven dot boell at intel dot com>, gdb-patches at sourceware dot org, sanimir dot agovic at intel dot com
- Date: Tue, 12 Aug 2014 23:55:32 +0200
- Subject: Re: [patch 1/2] Re: Crash regression(?) printing Fortran strings in bt [Re: [V2 00/23] Fortran dynamic array support]
- Authentication-results: sourceware.org; auth=none
- References: <1405070495-6948-1-git-send-email-keven dot boell at intel dot com> <20140729183023 dot GA17443 at host2 dot jankratochvil dot net> <20140729202710 dot GA20874 at host2 dot jankratochvil dot net> <53DB3FB3 dot 6050601 at linux dot intel dot com> <20140801210217 dot GA26015 at host2 dot jankratochvil dot net> <53E9BA7B dot 1080202 at linux dot intel dot com>
On Tue, 12 Aug 2014 08:55:55 +0200, Keven Boell wrote:
> On 01.08.2014 23:02, Jan Kratochvil wrote:
> > I have it reproducible on Fedora 20 i686 with plain
> > CFLAGS=-g ./configure;make;cd gdb/testsuite;make site.exp;runtest gdb.fortran/dynamic-other-frame.exp
> >
> > Besides that I have updated the testcase with
> > gdb_test_no_output "set print frame-arguments all"
> > so that there is no longer needed the patch:
> > [patch] Display Fortran strings in backtraces
> > https://sourceware.org/ml/gdb-patches/2014-07/msg00709.html
> >
> > The fix below has no regressions for me. Unfortunately I do not see why you
> > cannot reproduce it.
>
> I installed gcc 4.9 on my FC20 32bit system
This is unrelated to gcc 4.9. The problem occurs with standard Fedora 20 i686
compiler which is currently gcc-4.8.3-1.fc20.i686.
I have tried this evening to really reproduce it in 32-bit VM but it does not
happen there, despite all the involved packages are exactly the same build.
IMO it has to depend also on my x86_64 kernel where I run the i686 host in
mock chroot - but I did not try to reproduce it cleanly now.
Providing at least 32-bit build of GDB and its core:
http://people.redhat.com/jkratoch/gdb-bt.tar.xz
It is from the latest Intel vla.git clean commit:
364a416f43095996835d635561e22edceb53d380
> and tried to reproduce the issue like
> you mentioned above. Also this command sequence doesn't give me a crash. However
> I added the patch, which fixes the issue on your end, you mentioned in a different mail
> on the patch series and ran a make check with and without the patch. No regressions
> detected on my system. I've pushed the new series to Github:
> https://github.com/intel-gdb/vla/tree/vla-fortran
The difference of this new branch is only in:
https://github.com/intel-gdb/vla/commit/364a416f43095996835d635561e22edceb53d380
This is a part of:
[patch 2/2] Re: Crash regression(?) printing Fortran strings in bt [Re: [V2 00/23] Fortran dynamic array support]
https://sourceware.org/ml/gdb-patches/2014-08/msg00026.html
but that patch did not fix the bug, it was only trying to add some sanity
checking (which did not work, though).
The fix of the F20 i686 bug was this patch:
[patch 1/2] Re: Crash regression(?) printing Fortran strings in bt [Re: [V2 00/23] Fortran dynamic array support]
https://sourceware.org/ml/gdb-patches/2014-08/msg00025.html
No part of that patch is in the github branch, even not the testcase.
I have verified If I apply the testcase from [patch 1/2] (=the latter one)
GDB still crashes - that is the core file gdb-bt.tar.xz above.
Thanks,
Jan