Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 146192 - gaim-1.5.0 segfaults after gcc-4.1.1 upgrade
Summary: gaim-1.5.0 segfaults after gcc-4.1.1 upgrade
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: Current packages (show other bugs)
Hardware: AMD64 Linux
: High normal (vote)
Assignee: Gentoo Net-im project
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-09-03 15:51 UTC by Jose Marino
Modified: 2006-10-13 08:31 UTC (History)
1 user (show)

See Also:
Package list:
Runtime testing required: ---


Attachments
gaim segfault backtrace (gaim.bt,7.89 KB, text/plain)
2006-09-03 15:53 UTC, Jose Marino
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Jose Marino 2006-09-03 15:51:26 UTC
Since the upgrade to gcc-4.1.1 and glibc-2.4-r3, gaim-1.5.0 segfaults when trying to open a gtk file dialog. It's reproducible 100% of the times. 
I get a segfault whenever I click on the "Send File" button or menu "Conversation/Save As" in a conversation window. Both of these actions open a gtk file chooser dialog.

This happens only with gaim, any other application that uses the gtk file chooser dialog works fine: such as gedit, glade, gtk-demo and gnome-system-log.

I've recompiled gcc, glibc, gtk+, glib, pango, gnome-vfs and gaim several times since the upgrade and still no solution.

This problem only occurs in the amd64 machine. I also did the gcc upgrade in a x86 laptop (a pentium-m) and gaim seems to works fine in there.

I'll post backtrack from gdb below.

# emerge --info
Portage 2.1-r2 (default-linux/amd64/2006.0, gcc-4.1.1, glibc-2.4-r3, 2.6.17-gentoo-r7 x86_64)
=================================================================
System uname: 2.6.17-gentoo-r7 x86_64 AMD Athlon(tm) 64 Processor 3500+
Gentoo Base System version 1.12.4
distcc 2.18.3 x86_64-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled]
app-admin/eselect-compiler: [Not Present]
dev-lang/python:     2.4.3-r1
dev-python/pycrypto: 2.0.1-r5
dev-util/ccache:     [Not Present]
dev-util/confcache:  [Not Present]
sys-apps/sandbox:    1.2.17
sys-devel/autoconf:  2.13, 2.59-r7
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2
sys-devel/binutils:  2.16.1-r3
sys-devel/gcc-config: 1.3.13-r3
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="amd64"
AUTOCLEAN="yes"
CBUILD="x86_64-pc-linux-gnu"
CFLAGS="-march=athlon64 -O2 -pipe -fprefetch-loop-arrays -ftracer"
CHOST="x86_64-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/share/X11/xkb /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/"
CONFIG_PROTECT_MASK="/etc/env.d /etc/gconf /etc/revdep-rebuild /etc/splash /etc/terminfo"
CXXFLAGS="-march=athlon64 -O2 -pipe -fprefetch-loop-arrays -ftracer"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig distlocks metadata-transfer sandbox sfperms splitdebug strict"
GENTOO_MIRRORS="http://ftp.ucsb.edu/pub/mirrors/linux/gentoo/"
LC_ALL="en_US.utf8"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude='/distfiles' --exclude='/local' --exclude='/packages'"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://localhost:3000/gentoo-portage"
USE="amd64 X a52 aac acpi alsa asf audiofile avi berkdb bitmap-fonts bonjour bzip2 cairo caps cddb cdparanoia cdr cli crypt cups curl dbus divx4linux dlloader dri dvd dvdread dvi emacs emboss encode esd ffmpeg fftw flac fontconfig fortran gdbm gif gimpprint glitz glut gnome gnutls gpm gs gstreamer gtk gtk2 gtkhtml hal howl idn imagemagick imlib ipv6 isdnlog java jpeg lcms ldap logrotate lzo lzw lzw-tiff mad md5sum mng mono mozcalendar moznocompose moznoirc moznomail moznoroaming mozsvg mozxmlterm mp3 mpeg mpi nautilus ncurses network nls nntp nomotif nptl nptlonly nsplugin nvidia offensive ogg opengl oss pam pcre pdf pdflib perl png pppd python qt4 quicktime readline reflection rtc samba sdl session sndfile speex spell spl ssl svg symlink t1lib tcpd tetex theora tiff truetype truetype-fonts type1-fonts ucs2 unicode usb v4l v4l2 vcd video_cards_nvidia vorbis xine xml xmms xorg xpm xscreensaver xv xvid xvmc zlib elibc_glibc input_devices_keyboard input_devices_mouse input_devices_evdev kernel_linux userland_GNU video_cards_vesa video_cards_v4l"
Unset:  CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LDFLAGS, LINGUAS, PORTAGE_RSYNC_EXTRA_OPTS
Comment 1 Jose Marino 2006-09-03 15:53:00 UTC
Created attachment 95888 [details]
gaim segfault backtrace

backtrace obtained following instructions from: http://gaim.sourceforge.net/gdb.php
Comment 2 Luke Bratch 2006-09-03 16:56:11 UTC
I'm getting exactly the same problem since gcc-4.1.1

Hopefully the fix won't be just to upgrade to gaim-2, as it's changed so much from 1.5.0.
Comment 3 Christoph Mende (RETIRED) gentoo-dev 2006-09-03 23:26:19 UTC
Don't have that problem here with gcc-4.1.1, glibc-2.4-r3 and gaim-1.5.0 on amd64
Comment 4 Jakub Moc (RETIRED) gentoo-dev 2006-09-04 00:38:06 UTC
(In reply to comment #3)
> Don't have that problem here with gcc-4.1.1, glibc-2.4-r3 and gaim-1.5.0 on
> amd64

gaim crashes pretty much randomly w/ whatever gcc/glibc version you use, don't think this is gcc upgrade-related at all.


Comment 5 Steven Oliver 2006-09-08 08:21:05 UTC
I have a Pentium 4 M (x86) and Gaim 1.5 crashes for me too. Again though it's just started this after I upgraded my GCC to 4.1.1. I've never had Gaim to do this to me before now.
Comment 6 Robert Buchholz (RETIRED) gentoo-dev 2006-09-13 06:26:41 UTC
I am having the very same weird bug behaviour with net-im/gaim-2.0.0_beta3-r2 as well as 1.5.0, too. I tried compiling with gcc-4.1.1 and 3.4.6 and with -O0 cflags. This also happened with app-editors/bluefish, so I figured it might have to do with compiling gtk+ with gcc-4.

Now I just finished merging x11-libs/gtk+-2.8.19 with gcc-3.4.6 and that seems to fix the problem for me! So there seems to be a bug in gtk+ triggered by gcc-4 or my use flags were too aggressive. Nevertheless, using the *same* useflags it works perfectly when cooked with gcc-3.

Nevertheless, this bug is not limited to x86 as implied by the report-form.
Comment 7 Robert Buchholz (RETIRED) gentoo-dev 2006-09-13 06:29:02 UTC
"... not limited to AMD64 as implied by the report-form, but also happens on x86."
is what I meant.
Comment 8 Robert Buchholz (RETIRED) gentoo-dev 2006-09-13 06:41:08 UTC
At least for me, this bug seems to be a dupe of bug #133469.
Comment 9 Rance Smith 2006-09-13 07:42:29 UTC
I also did what robert had just posted I changed my cflags to o2 and recompiled gtk+ and it fixed the problem.  Link to forum discussion: http://forums.gentoo.org/viewtopic-p-3574872.html#3574872
link to reference bug:
http://bugs.gentoo.org/show_bug.cgi?id=133469
Would be nice if I could set the cflags somewhere for this package.  if you know how please let me know.
Comment 10 Roger 2006-09-14 20:47:47 UTC
Ditto. Haven't searched bugs.gentoo.org for my problem and went out on a binge to solve this one myself - since I was using the experimental releases.

re-emerging x11-libs/gtk+ seems to have solved this.  This also occurred following gcc-4 upgrade.

revdep-rebuild seems to be missing allot of stuff eh?
Comment 11 Robert Buchholz (RETIRED) gentoo-dev 2006-09-15 03:00:36 UTC
(In reply to comment #10)
> re-emerging x11-libs/gtk+ seems to have solved this.  This also occurred
> following gcc-4 upgrade.
> 
> revdep-rebuild seems to be missing allot of stuff eh?

revdep-rebuild can only detect linking errors, but this is a failure is because of the CFLAGS that are set. If you synced before rebuilding, you might have rebuilt a fixed version of gtk+, since as decided in bug #133469, the CFLAGS are now filtered.

Jose, can you confirm that that bugfix also fixed your bug so it can be closed?
Comment 12 Jose Marino 2006-09-15 16:25:22 UTC
Sadly no, this didn't solve my problem.

I want to point out that my original post refered to gaim crashing whenever I click on "Send File" button or the equivalent from the menu. Some people seemed to talk about gaim crashing in other ways, that apparently get fixed when gtk+ is compiled with -O2 or lower as in bug #133469.

In the original post of my emerge --info you can see that I was already using -O2. So that was not my problem. Still I figured I should give it a shot and compiled gtk+ and gaim with -O1 and also with no optimizations at all. No luck, gaim still crashes when I click on the "Send File" button.

Also tried removing my ~/.gaim directory. Got the same crash. I haven't had any other issue with any other program using gtk+ so far.

Maybe I'm the only one with this issue. There must be something wrong somewhere in my computer. If anybody has any suggestions on things to try please let me know. The backtrace from gdb doesn't really tell me anything and I'm out of ideas.
Comment 13 Robert Buchholz (RETIRED) gentoo-dev 2006-09-15 18:28:29 UTC
Sorry for mixing things up here.

Sounds pretty weird, especially if it doesn't happen in other gtk+ apps.
Did you try running gaim as another user?
You could also try the latest 2.0 beta if that bug was fixed.
Comment 14 Jose Marino 2006-09-15 18:35:24 UTC
Thanks for your reply Robert. I was just writing another comment about the beta when you posted your reply.

So I got curious and decided to try gaim-2.0.0_beta3-r2. And I can confirm that the problem goes away. Gaim doesn't crash anymore with the beta (as of yet :P). 

I think you've got a good point there, it may be a good idea to try gaim-1.5 in a brand new account. I'll let you know how that goes.
Comment 15 Jose Marino 2006-09-16 12:08:05 UTC
Just a quick update. I created a new user and tried gaim in there with no luck. It still crashed.

I'm now happily running the beta version which works fine.
Comment 16 Kathryn Kulick (RETIRED) gentoo-dev 2006-10-13 08:31:55 UTC
Due to the comments, I'm taking it that the issue has been resolved and am closing the bug. Please feel free to re-open if necessary.