Since a emerge -u world last week Rhythmbox 0.11.2-r2 and Totem 2.21 crash. All the other applications work. I think it happenend after the upgrade of glibc to 2.7. Because it is Rhythmbox and Totem, I suspected a gstreamer problem, but vven stranger is the fact that totem plays mp3 fine but crashes on loading of a video and rhythmbox crashes randomly even when not playing anything. Besides Banshee and "gst-launch-0.10 filesrc location=./Downloads/Bittorrents/The.Office.S04E07/the.office.s04e07.hdtv.xvid-xor.avi ! decodebin name=debin ! { queue ! xvimagesink } { debin. ! queue ! alsasink }" Play mp3 and video just fine. I did a emerge -e system and emerge -e world. I've rebuilded the kernel and the modules not included in the kernel (alsa, nvidia, etc). I've tried to switch from compiz-fusion to metacity. All to no avail, both applications stay unstable , the rest of the system and other applications performing the same things (Banshee, mplayer, vlc, gst-launch) work. I like rhythmbox, so Banshee is only a temporary solution to me. Reproducible: Always Portage 2.1.3.19 (default-linux/amd64/2007.0, gcc-4.2.2, glibc-2.7-r0, 2.6.22-suspend2-r2 x86_64) ================================================================= System uname: 2.6.22-suspend2-r2 x86_64 AMD Athlon(tm) 64 Processor 3000+ Timestamp of tree: Mon, 12 Nov 2007 12:00:01 +0000 app-shells/bash: 3.2_p17-r1 dev-java/java-config: 1.3.7, 2.1.2-r1 dev-lang/python: 2.3.6-r2, 2.4.4-r4, 2.5.1-r3 dev-python/pycrypto: 2.0.1-r6 sys-apps/baselayout: 1.12.10-r5 sys-apps/sandbox: 1.2.18.1-r2 sys-devel/autoconf: 2.13, 2.61-r1 sys-devel/automake: 1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2, 1.10 sys-devel/binutils: 2.18-r1 sys-devel/gcc-config: 1.4.0-r4 sys-devel/libtool: 1.5.24 virtual/os-headers: 2.6.23-r1 ACCEPT_KEYWORDS="amd64 ~amd64" CBUILD="x86_64-pc-linux-gnu" CFLAGS="-march=k8 -O2 -pipe -msse3" CHOST="x86_64-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/kde/3.5/env /usr/kde/3.5/share/config /usr/kde/3.5/shutdown /usr/share/config" CONFIG_PROTECT_MASK="/etc/env.d /etc/env.d/java/ /etc/fonts/fonts.conf /etc/gconf /etc/revdep-rebuild /etc/splash /etc/terminfo /etc/udev/rules.d" CXXFLAGS="-march=k8 -O2 -pipe -msse3" DISTDIR="/usr/portage/distfiles" FEATURES="distlocks metadata-transfer sfperms strict unmerge-orphans userfetch" GENTOO_MIRRORS="http://distfiles.gentoo.org http://distro.ibiblio.org/pub/linux/distributions/gentoo" LANG="en_US.UTF-8" PKGDIR="/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 --filter=H_**/files/digest-*" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/portage/local/layman/musicbrainz /usr/portage/local/layman/sunrise /usr/portage/local/layman/xeffects /usr/local/portage /usr/local/misc_overlays /usr/local/portage-aeskulap" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="X a52 aac accesibility accessibility acl acpi aiglx alsa amd64 audiofile avahi avi beagle berkdb bitmap-fonts bluetooth bzip2 cairo cdparanoia cdr cli cracklib crypt cups daap dbus dpms dri dv dvd dvdr dvdread eds emul-linux-x86 encode evo exif fam ffmpeg firefox foomaticdb fortran galago gb gcj gdbm gif glitz glut gnome gnutls gpm gstreamer gtk gtk2 gtkhtml guile hal iconv ieee1394 imagemagick imap imlib ipv6 isdnlog java javascript joystick jpeg keyring libnotify mad midi mikmod mime mmx mng mono mozsvg mp3 mp4 mpeg mudflap ncurses networkmanager nls nptl nptlonly nsplugin ogg openal opengl openmp pam pcmcia pcre pdf perl pidgin png postscript pppd python quicktime readline reflection samba scanner sdl session sndfile sockets sox speex spell spl sse sse2 ssl svg tcpd theora threads tiff tracker truetype truetype-fonts type1-fonts unicode usb v4l vcd vorbis wifi wmf wxwindows xinerama xml xml2 xorg xosd xpm xrandr xv xvid zlib" ALSA_CARDS="intel8x0 usb-audio" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mulaw multi null plug rate route share shm softvol" ELIBC="glibc" INPUT_DEVICES="keyboard mouse" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" USERLAND="GNU" VIDEO_CARDS="nv nvidia vesa" Unset: CPPFLAGS, CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LC_ALL, LDFLAGS, LINGUAS, MAKEOPTS, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS
I forget one thing: rhythmbox sometimes, not always, gives a error on crash. Xlib: unexpected async reply (sequence 0x4dc6!)
http://www.gentoo.org/proj/en/qa/backtraces.xml http://www.gentoo.org/doc/en/bugzilla-howto.xml
backtrace rhythmbox: process 2563: arguments to dbus_message_new_signal() were incorrect, assertion "path != NULL" failed in file dbus-message.c line 1158. This is normally a bug in some application using the D-Bus library. D-Bus not built with -rdynamic so unable to print a backtrace Program received signal SIGABRT, Aborted. ---Type <return> to continue, or q <return> to quit--- 0x00002b9f10684185 in raise () from /lib/libc.so.6 (gdb) bt #0 0x00002b9f10684185 in raise () from /lib/libc.so.6 #1 0x00002b9f10685630 in abort () from /lib/libc.so.6 #2 0x00002b9f0b815235 in ?? () from /usr/lib/libdbus-1.so.3 #3 0x00002b9f0b811945 in ?? () from /usr/lib/libdbus-1.so.3 #4 0x00002b9f0b807743 in dbus_message_new_signal () from /usr/lib/libdbus-1.so.3 #5 0x00002b9f09196b60 in ?? () from /usr/lib/libdbus-glib-1.so.2 #6 0x00002b9f0f1afc9e in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0 #7 0x00002b9f0f1c3920 in ?? () from /usr/lib/libgobject-2.0.so.0 #8 0x00002b9f0f1c5356 in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0 #9 0x00002b9f0f1c5743 in g_signal_emit () from /usr/lib/libgobject-2.0.so.0 #10 0x000000000042a16c in rb_shell_window_state_cb () #11 0x00002b9f0d3bb0ef in ?? () from /usr/lib/libgtk-x11-2.0.so.0 #12 0x00002b9f0f1afc9e in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0 #13 0x00002b9f0f1c3920 in ?? () from /usr/lib/libgobject-2.0.so.0 #14 0x00002b9f0f1c5085 in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0 #15 0x00002b9f0f1c5743 in g_signal_emit () from /usr/lib/libgobject-2.0.so.0 #16 0x00002b9f0d4be4f5 in ?? () from /usr/lib/libgtk-x11-2.0.so.0 #17 0x00002b9f0d3b5434 in gtk_main_do_event () from /usr/lib/libgtk-x11-2.0.so.0 #18 0x00002b9f0d8651dc in ?? () from /usr/lib/libgdk-x11-2.0.so.0 ---Type <return> to continue, or q <return> to quit--- #19 0x00002b9f0ff71bd3 in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0 #20 0x00002b9f0ff74ead in ?? () from /usr/lib/libglib-2.0.so.0 #21 0x00002b9f0ff75189 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0 #22 0x00002b9f0d3b5772 in gtk_main () from /usr/lib/libgtk-x11-2.0.so.0 #23 0x00000000004236ae in main ()
totem got a some new stuff in plparser which is also used by rhythmbox. rhythmbox-0.11.3 might work better in this case. try to locally bump rhythmbox and tell us if it fixes the problem. And as pointed out by jakub, you need to recompile some more stuff before the backtrace can actually be useful. Thanks in advance.
Here you have three backtraces in rhythmbox, hope it will help. I downgrade totem to 2.18.3 and then totem works without problems, going back to 2.20.x and totem crashes. I run revdep-rebuild after switching totem. [New LWP 29095] Program received signal SIGABRT, Aborted. [Switching to LWP 29095] 0x00002b6bdff98185 in raise () from /lib/libc.so.6 (gdb) bt #0 0x00002b6bdff98185 in raise () from /lib/libc.so.6 #1 0x00002b6bdff99771 in abort () from /lib/libc.so.6 #2 0x00002b6bd61e1bba in free () from /usr/lib/libffi.so.4 #3 0x00002b6bdf892e32 in magazine_cache_push_magazine () from /usr/lib/libglib-2.0.so.0 #4 0x00002b6bdf89308c in private_thread_memory_cleanup () from /usr/lib/libglib-2.0.so.0 #5 0x00002b6bda1eda4c in ?? () from /lib/libpthread.so.0 #6 0x00002b6bda1ee075 in ?? () from /lib/libpthread.so.0 #7 0x00002b6be002cc7d in clone () from /lib/libc.so.6 Second Backtrace: Program received signal SIGSEGV, Segmentation fault. 0x00002b41cf7ade26 in free () from /usr/lib/libffi.so.4 (gdb) bt #0 0x00002b41cf7ade26 in free () from /usr/lib/libffi.so.4 Cannot access memory at address 0x7fffdbf190c8 Thirth Backtrace: [New LWP 29591] Program received signal SIGABRT, Aborted. [Switching to LWP 29591] 0x00002b0083651185 in raise () from /lib/libc.so.6 (gdb) bt #0 0x00002b0083651185 in raise () from /lib/libc.so.6 #1 0x00002b0083652771 in abort () from /lib/libc.so.6 #2 0x00002b007989abba in free () from /usr/lib/libffi.so.4 #3 0x00002b0082f4be32 in magazine_cache_push_magazine () from /usr/lib/libglib-2.0.so.0 #4 0x00002b0082f4c08c in private_thread_memory_cleanup () from /usr/lib/libglib-2.0.so.0 #5 0x00002b007d8a6a4c in ?? () from /lib/libpthread.so.0 #6 0x00002b007d8a7075 in ?? () from /lib/libpthread.so.0 #7 0x00002b00836e5c7d in clone () from /lib/libc.so.6
Created attachment 136385 [details] Backtrace Rhythmbox 0.11.2-r2 Finally Pissed of enough to recompile a lot of my system to show you a complete backtrace. As you can see in the attachment called "rhythmbox 0.11.2-r2 backtrace.txt". Hopefully this one will leave no question open where the bug is....
The error is consistent, but time it shows up is random....
Maybe you (whoever is assigned to this bug) would like to look to this thread on the forums http://forums.gentoo.org/viewtopic.php?p=4528355#4528355. I'm not the only one and a bunch of Gnome/GTK+ applications is crashing after a update. Totem Epiphany Rhythmbox Gedit Till now...
A couple things : 1) that last trace you attached shows nothing ... it could be pure bad luck. 2) totem and rhythmbox have more in common than they'll let you believe. Could you run gstreamer-properties and select a safe video output plugin, it should be called "X Window System (without Xv)". Try using your apps with that plugin. Thanks
Well, ok then.... I recompiled the complete stack with FEATURES="nostrip", totem will not run with gdb and gets stuck. Rhythmbox runs with gdb and always (over 10 times) generated this output, and never crashed on the same point. But okay bad luck is bad luck. By the way I've returned to totem 2.18.3 and this is running without problems, I've revdep-rebuild to update the dependencies and rhythmbox is rebuild with totem 2.18.3 and still crashes... But okay, if this is unresolvable fine... I will see what I do...
Created attachment 136612 [details] different Backtrace Ok, I switched of xv in gstreamer-properties and did several runs of rhythmbox. Mostly the same trace as I allready posted. But this one is a lot different. Please give feedback and I will try other things
The backtrace still doesn't speak a lot. At best it tells that there might be a problem in glib/gtk/librsvg. But I'm not sure this is even correct. No idea of what's going on here.
Created attachment 136615 [details] Traces of my problem These are various traces from my system. Emerge info is included, Totem, gedit, and epiphany too.
Your epiphany problem is related to gecko. That should probably get it's own bug. Your gedit crash is a failure to malloc. Smells like bad RAM? The totem crash isn't extremely helpful. Nothing in the backtrace should be crashing (ie, poll, pthread_cond_wait, and so on. All stopped in the kernel). The two rhythmbox crashes are either double-frees or bad RAM. Have you run memcheck recently?
Ok... I've run memcheck to be sure but that went fine... (I mean no errors). So I've shrunk my root partition and installed Fedora 8, which runs rhythmbox and totem just fine. You can close the bug if you want.
Daniel, If you refer to the traces from Mark (chaseguard) I also have the same problems as Marc. You mention bad ram, but (as noted in the forum thread for this) I ran memtest86 from the LiveCD for three complete cycles and no failure is noted. In addition, no other applications besides these Gnome ones are effected. Perhaps there is some unique problem related to the AMD 64 architecture, but, it seems unlikely considering that many people must use similar systems. I have not "switched" distos (yet) and would like the bug resolved if possible. Also note, there is another Gentoo user besides Marc and myself with the same affliction.
I've not switched distro's but I've spent a lot of time recompiling my system to give backtraces and I've done a lot of other things to help and rule out things. 1. I did a emerge --empty system and world. 2. Go back to previous versions of rhythmbox and totem (revdep-rebuilding everytime) and totem 2.18.3 works, which rules out gstreamer problems (besides I run gstreamer command-lines to test without any problems and Banshee (also gstreamer) never crashes. All of this is mentioned in my reports. 3. I use a lot of applications and only gedit, rhythmbox and totem are unstable (and gdmflexiserver, but that is another bug), epiphany after reading the thread I mentioned an behold it crashes too (I use firefox-bin) But my impression is that I'm not taking serious on this bug. so be it. I've spend many hours on this and now installed Fedora 8 under a little as 3 hours, a fraction of the time I spend on this bug, so I can use my computer a couple of days/weeks without getting frustrated. I've the feeling this only get resolved with updates and time, so I update and wait.
Marc, Please don't flame me for "switched" as I never meant anything derogatory. If this is not resolved I too may "switch" (I do not even know how to approach that problem yet). Further, I am piggy-backing on your bug and you and I have both taken remarkably similar approaches to resolve or identify the problem. My only concern is your suggestion that the "bug" be marked closed considering that the forum thread suggests there are four of us experiencing a similar problem. No, I do not believe in "bad luck" or the coincidental failure of RAM memory (which has tested OK) among unrelated users. I cannot understand why epiphany (which has never had any problem before (even though I do not use it consistently) has a Gecko problem that appeared at the same time as these other issues. My preference is that this problem gets fixed without changing distros (I have used Gentoo for over six years).
listen guys, nobody is taking this lightly. It just look as strange to us as it looks to you. Now, looking at your emerge --info, it appears that both of you have old python versions around, could you try to clean that up (python-updater, yadi yada) and see if it makes any difference. You also both have overlays, even though I sure you already checked that, could you please verify nothing in there would interact in the gnome stack ? Thanks
Sorry for the bug closed line in my previous remark, It was very selfish and egocentric. I wil remove my overlay, remove the old python builds and run a emerge --empty system/world. this will take a lot of time. I wil start this afternoon. Any advice on the emerge, is emerge --empty enough or a --dep too? Shall I recompile with FEATURES="nostrip", what effect has this on the performance?
Try with C{,XX}FLAGS="-O2 -g -pipe" and FEATURES="splitdebug" which is much better than nostrip. emerge -e world should be more than enough. Don't forget to do an emerge --depclean first, to avoid rebuilding packages that are no longer needed. As Gilles pointed out, we are taking all real bugs seriously, it's just that to fix a bug we need to understand what's causing it, and so far we're still walking around in the dark. Thanks
emerge -e world shouldn't be necessary it's really just a last resort option, just purge old python versions, run python-updater, emerge --DuNav world and revdep-rebuild. If it really doesn't do the trick, emerge -e world could still be an option to try but I doubt it would change anything (it didn't for a while now on my boxes because there is pretty much nothing that can't be fixed in other faster ways).
So another day in this great adventure and another string of "bad luck". I've removed all of the python engine but one, did run python-updater, ran emerge --depclean, emerge -DuNav world, revdep-rebuild, started yesterday around 15:00 and it was ready at 7:30 this morning. And again the laptop (amd64) is running fine under Gentoo 2007 and Fedora 8 except off-course rhythmbox, Totem and Gedit under Gentoo. In the weekend I will run a emerge -e world, so the system will be compiled (almost) completely with debug information I will attach two backtrace which look different then the rest of the traces I got and which you tell me you can't use. That's all, heading back to Fedora... Oh, and I give you a new emerge --info because somethings have changed....
Created attachment 136738 [details] Rhythmbox 0.11.2 Backtrace, crash when change song manually
Created attachment 136739 [details] Gedit 2.20.3 crash Gedit 2.20.3 crash when I was using Gedit to copy the Backtrace of rhytmbox to a txt file. This was funny, ha ha, I new I should have used Mousepad. Old habits...
Created attachment 136740 [details] New Emerge --info
I completely had forgotten Totem. It is off-course mentioned in the header of this bug. But the random crashes of totem only generate the kind of backtraces you allready you can't use, all 10.234 of them. Below a example. Not to see hear, move along... [Thread 0x43007950 (LWP 7535) exited] [Thread 0x40802950 (LWP 7511) exited] Program received signal SIGABRT, Aborted. [Switching to Thread 0x41003950 (LWP 7530)] 0x00002ba7a19ef185 in raise () from /lib/libc.so.6 (gdb) bt #0 0x00002ba7a19ef185 in raise () from /lib/libc.so.6 #1 0x00002ba7a19f0630 in abort () from /lib/libc.so.6 #2 0x00002ba7a12a9bba in free () from /usr/lib/libffi.so.4 #3 0x00002ba7a173e043 in private_thread_memory_cleanup () from /usr/lib/libglib-2.0.so.0 #4 0x00002ba7a0c86a4c in __nptl_deallocate_tsd () from /lib/libpthread.so.0 #5 0x00002ba7a0c87075 in start_thread () from /lib/libpthread.so.0 #6 0x00002ba7a1a83c7d in clone () from /lib/libc.so.6 (gdb)
hum, your gtksourceview version shouldn't be 1.8.5 if you have gedit-2.20. Do you have a gtksourceview-2 installed ? (that's two different slots, just making sure)
There are two versions of gtksourceview installed, 1.8.5-r1 and 2.0.1-r1, I did a unmerge of 1.8.5-r1 and after that a revdep-rebuild. which tells me that anjuta and gnome-python-desktop really, really, really need the 1.8.5-r1 version. So what should it be?
I also get this whenever I try to send a bug-buddy, so none can be sent to gnome! Failed to parse the xml-rpc response. Response follows: Unable to parse XML-RPC Response 3 <?xml version="1.0" encoding="UTF-8"?> <methodResponse><fault><value><struct><member><name>faultString</name><value><string>Application failed during request deserialization: not well-formed (invalid token) at line 23, column 0, byte 1720 at /usr/lib/perl5/vendor_perl/5.8.8/i386-linux-thread-multi/XML/Parser.pm line 187 </string></value></member><member><name>faultCode</name><value><string>Client</string></value></member></struct></value></fault></methodResponse>
I too did all the things you mentioned: Removed overlay (was pybackpack which is not even installed). Python update, Emerge -C older python, python-update, perl-cleaner reallyall, fix_libtool_files, did I miss any? Then emerge DuNav followed by revdep-rebuild. Still crashes on Totem and gedit. Did not even try epiphany. I also tried changing several kernel options in different compiles (scheduler, big lock, tickless, and anything else I thought might have a bearing). I note that Marc Geerlings is using 64 bit and I am using 32 (although both are AMD processors). I too have the same versions of gtksourceview. I am presently doing (yet another) -e world (662 packages) with little hope. I have splitdebug and ggdb, but core dumps and traces have previously been sent and are apparently not helpful. Also note that bug-buddy won't mail anything (although it produces copious output) and gives the warning shown above. Mark
FYI: The bug-buddy thing appears in http://bugzilla.gnome.org/show_bug.cgi?id=447102. Says it related to xml and UTF problem with character. Does not say this is fixed.
Created attachment 137069 [details] In the hope that somebody can do something about this! I think that this backtrac is fairly complete.
Created attachment 137071 [details] In the hope that somebody can to something about this.... I think that this backtrac is fairly complete. I still thing there is something wrong with glibc 2.7
I'm currently running glibc-2.6.1, and I have exactly the same problems on x86.
Please will somebody look at my latest backtraces and give feedback what is still missing! I've done considerable things to help and solve this bug, and know after the second emerge -e world with FEATURES="splitdebug", I want to know what more... Don't let this slip with another 3 weeks of no comments.
much better stacktraces. I noticed that in both cases, one of the threads calls pango functions just before the crash. What's your pango version ? Does it appear is well in epiphany crashes ?
I use pango 1.18.3, I will try and generate some more backtrace from totem and epiphany. Do you know how you can get a bacjtrace with gdb like bug-buddy makes, if I use gdb it only shows one thread...
Created attachment 137110 [details] totem backtrace Another backtrace, this time totem. If something else is needed please let me know, almost my complete system is compiled with splitdebug (200 packages to go). Totem-video-thumbnailer works bytheway
Created attachment 137112 [details] A backtrace for EPiphany as requested I found out how to show backtraces of all threads under gdb.... "thread apply all bt" Are the magic words
Guys, I think I've solved it with a bit of lucky detective work. It's not glibc after all. On a lark, I tried diffing the output /lib/ld-linux --list /path/to/executable to see if the applications that were crashing linked against something that the non-crashing apps didn't. In all cases, the crashing apps linked against libffi.so.4. The stable apps didn't. On my system, libffi belongs to dev-libs/g-wrap-1.9.9, which I notice got a revbump at roughly the time this stuff started. I unmerged g-wrap as it seemed that nothing depended on it, and did a revdep-rebuild. The packages that got rebuilt were(suprise!) totem, epiphany, gedit, eog, etc, in short all the packages that are crashing. After the rebuild *all of the problem packages worked*. No more random crashes. I'm willing to bet large sums of money that everyone with this problem has g-wrap installed. So this should fix the problem: 1. Remove g-wrap(a downgrade to 1.9.8.x might work, haven't tested) 2. Revdep-rebuild 3. That should fix these random segfaults. If someone else can confirm this, a bug shoild be filed with g-wrap. I haven't had time to determine the reason why it's causing crashes, and I won't have net access until tomorrow, so if someone else can do that I'd be grateful.
I want to sincerely thank you for leading me to the right solution algosystem@gmail.com. If you ever come to the Netherlands (Maastricht), let me know and I will by you diner... I unmerged g-wrap and the strange thing on my system was that a revdep-rebuild didn't rebuild rhythmbox totem or gedit, only gnome-python-desktop, vte and something else and.... No g-wrap either. It seems that gcc-2.4.2 has also libffi.so.4, ldd /usr/bin/rhythmbox shows: /usr/lib/gcc/x86_64-pc-linux-gnu/4.2.2/libffi.so.4 (0x00002af413b58000) But unmerging g-wrap solved it for me!!! It was a conflict between two libffi.so.4 on my system..... Again thank you algosystem@gmail.com!!!
Sorry It is gcc 4.2.2 and I will file a bug right away! Back to Gentoo, from Fedora 8, this was a frustating month :)
nice catch. just as a complementary note: $ ldd /usr/bin/rhythmbox|egrep libffi (returns empty) and description of libffi: "libffi (from gcc) does not commonly build unless gcj is compiled, but is used by other projects, like GNUstep." do you have any java enabled ?
I see that I have a gcj USE flag in my /etc/make.conf, I can't remember why I put it in. mostly I put a new USE flag in when a applications I need asks for it when building. Do you advise to build without gcj flag and use g-wrapper?
Jumps for joy here too! I emerge -C g-wrap, revdep rebuilt a dozen packages, and things now work. This is really great - fantastic. Thanks to you all. By the way, I do not have the gcj flag enabled, globally or otherwise.
I advise nothing :) Imho, having 2 libraries with the same name/function/version on the same system is a bug (wrt to regular use of the machine not doing dev) and it's often a nightmare to figure out. @Mark, could you check if there is another libffi on your system ?
Gilles, I have no other (actually none at all) libffi files installed (only shows up in the portage directories). Maybe that is why my machine did rebuild totem, epiphany, gedit, and eight other packages when I ran revdep-rebuild after removing g-wrap and Marc Geerlings' machine (who has a gcj flag) did not rebuild all these packages. I don't know but I am very happy this box is working correctly again. Mark
ok gcc has support for proper libffi building and so does pygobject. If you find yourself in a situation we you have 2 libffi then it's probably a bug in the ebuild that managed to get the second in. You can use qfile to figure out which package installed which lib. For what gnome is concerned this is fixed. Thanks for all the work.
*** Bug 230782 has been marked as a duplicate of this bug. ***
