| Summary: | gaim-2.0.0_beta3-r1 segfaults when mouse leaves chat window | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Jamie Anderson <gentoo.mail> |
| Component: | Current packages | Assignee: | Gentoo Net-im project <net-im> |
| Status: | RESOLVED WORKSFORME | ||
| Severity: | major | CC: | asheldon, chris, free.perception, lostlogic, ullum, world.root |
| Priority: | High | ||
| Version: | 2006.0 | ||
| Hardware: | x86 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: | backtrace and debug output from gaim | ||
|
Description
Jamie Anderson
2006-07-31 08:10:51 UTC
Created attachment 93156 [details] backtrace and debug output from gaim I'm having the same problem. I recently recompiled all of my system with gcc-4.1.1 and this is when this started happening - I had the same behaviour with gaim 1.5.0 and this is why I tried out the beta-release. Recompiling just gaim with a lower version of gcc did not remedy the problem, so I am not sure, whether this is actually a bug in gaim - it might be in the dependencies. Or gaim might be incompatible with a new version of one of it's dependencies. The backtrace & debug output is in the attachment. I noticed it says there is no symbol table info available. I did compile gaim with USE="debug" and removed all C[XX]FLAGS except -march=pentium-m from my make.conf before emerging, so I guess this has got to be in one of the libraries? Please tell me if you need me to recompile any of them with special settings, or if you need any other kind of info. This is the output from emerge --info: Portage 2.1-r1 (default-linux/x86/2006.0, gcc-4.1.1/vanilla, glibc-2.4-r3, 2.6.18-rc1-mm2 i686) ================================================================= System uname: 2.6.18-rc1-mm2 i686 Intel(R) Pentium(R) M processor 1.73GHz Gentoo Base System version 1.6.15 ccache version 2.3 [enabled] app-admin/eselect-compiler: 2.0.0_rc2-r1 dev-lang/python: 2.3.5, 2.4.3-r1 dev-python/pycrypto: 2.0.1-r5 dev-util/ccache: 2.3 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: 2.17 sys-devel/gcc-config: 2.0.0_rc1 sys-devel/libtool: 1.5.22 virtual/os-headers: 2.6.11-r2 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CBUILD="i686-pc-linux-gnu" CFLAGS="-march=pentium-m" CHOST="i686-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/kde/3.5/env /usr/kde/3.5/share/config /usr/kde/3.5/shutdown /usr/share/X11/xkb /usr/share/config /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/eselect/compiler /etc/gconf /etc/revdep-rebuild /etc/terminfo" CXXFLAGS="-march=pentium-m" DISTDIR="/usr/portage/distfiles" FEATURES="autoconfig ccache distlocks metadata-transfer sandbox sfperms strict" GENTOO_MIRRORS="http://gentoo.inode.at/ ftp://gentoo.inode.at/source/ http://gd.tuwien.ac.at/opsys/linux/gentoo/ ftp://gd.tuwien.ac.at/opsys/linux/gentoo/" LINGUAS="en de fi es ru en_GB en_US de_DE de_AT es_ES fi_FI ru_RU" 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 /usr/portage/local/layman/portage-xgl" SYNC="rsync://rsync.de.gentoo.org/gentoo-portage" USE="x86 X a52 aac acpi aim alsa apache2 avi bash-completion bcmath berkdb bitmap-fonts blas bluetooth bzip2 cdparanoia cdr cjk cli crypt cups dbx dhcp directfb dlloader doc dri dvd dvdr dvdread eds emacs encode examples exif fbcon ffmpg firefox flac flash foomaticdb ftp gdbm gif ginac gmp gnutls gphoto2 gpm graphviz gstreamer gtk gtk2 icc icq ieee1394 imap imlib ipv6 irc isdnlog ithreads jabber java javascript jpeg jpeg2k libg++ libwww mad mikmod mime ming mmx mng motif mozilla mp3 mpeg msn mule mysql mysqli ncurses nls nptl nptlonly ocaml offensive ogg opengl oscar pam pcmcia pcre pdf pdflib perl php png posix pppd python qt qt3 qt4 quicktime readline reflection samba scanner sdl session simplexml socks5 speex spell spl sql sse sse2 ssl svg tcpd theora tiff truetype truetype-fonts type1-fonts udev unicode usb vcd vorbis wifi win32codecs wmf x264 xemacs xine xinerama xml xmms xorg xpm xscreensaver xsl xv xvid zlib elibc_glibc input_devices_keyboard input_devices_mouse input_devices_synaptics kernel_linux linguas_en linguas_de linguas_fi linguas_es linguas_ru linguas_en_GB linguas_en_US linguas_de_DE linguas_de_AT linguas_es_ES linguas_fi_FI linguas_ru_RU userland_GNU video_cards_fbdev video_cards_fglrx video_cards_vesa video_cards_radeon" Unset: CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, LDFLAGS, PORTAGE_RSYNC_EXTRA_OPTS Someone seems to have posted something that looks like the same bug to me on gaim's sf page: http://sourceforge.net/tracker/index.php?func=detail&aid=1508790&group_id=235&atid=100235 There is a backtrace that looks more meaningful than mine there as well. I am also experiencing a similar problem, I've tried both gaim-2.0.0_beta3 and gaim-1.5.0 and also experimented with different versions of gtk+ (including the masked 2.10.1 version, 2.8.20-r1 and 2.8.19) to no avail. It doesn't seem to be actually when the mouse leaves the conversation window, but whenever the mouse hovers over text in the conversation window, or in an away message window. The fault appears in gtk_text_insert for me. I can confirm the hovering thing. It doesn't always seem to happen immediately though, especially if I move the mouse slowly. Moving the cursor across the line, that seperates the text of an old session from the rest (this text is inserted by a history-plugin I think) always seems to crash it immediately though. I have also tried the SVN version to no avail - same problem. Another (almost definitely related) bug on sourceforge is: http://sourceforge.net/tracker/index.php?func=detail&aid=1535181&group_id=235&atid=100235 This looks to me like a gcc-4.1.1 issue, either in gtk+ or gaim. Is everyone seeing this problem indeed using gcc-4.1.1 and did it indeed start close to the time of the upgrade to 4.1.1? Can anyone tell what library or libraries were upgraded on their system between the gcc-4.1.1 installation and the start of this gaim bug? Thanks. In my case, it was on a brand-new install built using the stage 1/3 method, so everything was built and linked with gcc-4.1.1 and binutils-2.17.1. I have noticed some problems with other software (namely Evolution), but nothing as severe or consistant as what Gaim is doing. However, I guess we should all hope it is not a Gaim issue, since I have read the other bug reports and the Gaim devs are quite clear that they have no intention of troubleshooting the problem, let alone fixing it. (In reply to comment #5) > Another (almost definitely related) bug on sourceforge is: > http://sourceforge.net/tracker/index.php?func=detail&aid=1535181&group_id=235&atid=100235 > > This looks to me like a gcc-4.1.1 issue, either in gtk+ or gaim. > > Is everyone seeing this problem indeed using gcc-4.1.1 and did it indeed start > close to the time of the upgrade to 4.1.1? > > Can anyone tell what library or libraries were upgraded on their system between > the gcc-4.1.1 installation and the start of this gaim bug? > > Thanks. > The Gaim devs will troubleshoot it if it can be proven to be a bug in gaim as opposed to the system. Due to Gentoo's nature as a build-it-yourself distro, they will not even approach a problem until a Gentoo developer can demonstrate that the problem exists in Gaim itself. Very similar issue happens for me. I found that gaim craches only with particular gnome themes, e.g. for "Myst" I have following error in console just before crash happens: <QUOTE>** CRITICAL **: mist_draw_border: assertion `shadow_type != GTK_SHADOW_NONE' failed</QUOTE> Crash doesn't happen with "ClearLooks" theme at all. Also "ClearLooks" theme make inkscape don't crush in "About" dialog. But inkscape still crushes in other places with error like <QUOTE>Gtk-CRITICAL **: gtk_style_detach: assertion `style->attach_count > 0' failed</QUOTE> May be this info will help other to avoid crashes and motivate somebody to look into gnome-theme code. PS. I have gnome-themes-2.15.91.1 installed on my box. I use openbox, not gnome, so no help there, interesting the variety of situations which cause this crash for different people. recent either gtk+ or xorg upgrade has solved this for me. reopen this bug if it's still occuring for others. First of all -- do NOT USE -O3 OR -march=pentium-m OR -fomit-frame-pointer WHEN EMERGING gtk+ OR gaim. It seems like both of them hate it. Re-emering with really safe CFLAGS and CXXFLAGS (-O2 -march=i686) did the trick. Secondly: this has to be fixed anyways since these are my standard CFLAGS, too :P (In reply to comment #11) > First of all -- do NOT USE -O3 OR -march=pentium-m OR -fomit-frame-pointer WHEN > EMERGING gtk+ OR gaim. It seems like both of them hate it. Re-emering with > really safe CFLAGS and CXXFLAGS (-O2 -march=i686) did the trick. > > Secondly: this has to be fixed anyways since these are my standard CFLAGS, too > :P > Ok, I tried experimenting a bit on this and found a more or less convinient solution: set the CFLAGS and CXXFLAGS to "-O2 -march=pentium-m -pipe", then re-emerge gtk+ (stable version) and gaim (Beta 3.1) w/ "emerge gtk+ gaim" and the problems are not orcurring anymore. Though, it's weird that GCC 4.1.1 has such problems with -O3 -- it worked very fine with GCC 3.4.6. Next time try to remove the ricer flags before reporting any bug and re-emerge all dependencies and the programm itself. Thanks. I suspect this is a duplicate of bug #133469. Please try workaround described there: http://bugs.gentoo.org/show_bug.cgi?id=133469#c43 Basically, appending "-fno-inline-functions" to CFLAGS and re-emerging gtk+ should solve the problem (for me it does) :-) |
