Created attachment 411784 [details] backtrace Hello My father is having some problems with gnome-shell on some computers. It crashes on start and, then, people cannot login at all and even GDM cannot work. I attach the backtrace. Please let me know if it's enough Thanks a lot for your help
(In reply to Pacho Ramos from comment #0) #0 xkb_keymap_ref (keymap=keymap@entry=0x0) at /usr/src/debug/x11-libs/libxkbcommon-0.5.0/libxkbcommon-0.5.0/src/keymap.c:59 No locals. #1 0x00007ff0baa59b7f in xkb_state_new (keymap=0x0) at /usr/src/debug/x11-libs/libxkbcommon-0.5.0/libxkbcommon-0.5.0/src/state.c:582 ret = 0x1e91390 #2 0x00007ff0bbe629a3 in ?? () from /usr/lib64/libmutter.so.0 No symbol table info available. #3 0x00007ff0bbe649e0 in ?? () from /usr/lib64/libmutter.so.0 No symbol table info available. The crash happens because xkb_state_new is being called with NULL keymap argument. Please rebuild mutter with debugging symbols to see where that NULL is coming from.
AFAICT, this could happen because xkb_x11_keymap_new_from_device() returns NULL, since mutter will happily use that value. Please check your logs for anything keyboard-related. Please check your X config to make sure you aren't doing somethings strange with keyboard devices or with key mappings.
Created attachment 411822 [details] backtrace-r1 This backtrace is with spidermonkey and nspr recompile with debugging symbols It's important to note that compiling spidermonkey WITHOUT -pipe in CFLAGS solves the problem, while recompiling it again with "-pipe" makes it break again :/
[ebuild R ] sys-devel/gcc-4.8.5:4.8::gentoo USE="cxx fortran gcj (multilib) nptl openmp sanitize (-altivec) (-awt) -debug -doc (-fixed-point) -go -graphite (-hardened) (-libssp) -mudflap (-multislot) -nls -nopie -nossp -objc -objc++ -objc-gc -regression-test -vanilla" 0 KiB This is the gcc version
Created attachment 411874 [details] backtrace-r2 This is with mutter compiled with debugging symbols
I have just reproduced this in one more computer... in that one it was solved recompiling spidermonkey without -fno-stack-protector :/ Portage 2.2.20.1 (python 2.7.9-final-0, default/linux/amd64/13.0/desktop/gnome/systemd, gcc-4.8.5, glibc-2.20-r2, 4.0.5-gentoo x86_64) ================================================================= System uname: Linux-4.0.5-gentoo-x86_64-Intel-R-_Core-TM-2_Duo_CPU_E7500_@_2.93GHz-with-gentoo-2.2 KiB Mem: 3916424 total, 1629292 free KiB Swap: 4194300 total, 4194300 free Timestamp of repository gentoo: Mon, 14 Sep 2015 08:00:01 +0000 sh bash 4.3_p39 ld GNU ld (Gentoo 2.24 p1.4) 2.24 ccache version 3.1.9 [enabled] app-shells/bash: 4.3_p39::gentoo dev-java/java-config: 2.2.0::gentoo dev-lang/perl: 5.20.2::gentoo dev-lang/python: 2.7.9-r1::gentoo, 3.4.1::gentoo dev-util/ccache: 3.1.9-r4::gentoo dev-util/cmake: 3.2.2::gentoo dev-util/pkgconfig: 0.28-r2::gentoo sys-apps/baselayout: 2.2::gentoo sys-apps/openrc: 0.17::gentoo sys-apps/sandbox: 2.6-r1::gentoo sys-devel/autoconf: 2.13::gentoo, 2.69::gentoo sys-devel/automake: 1.11.6-r1::gentoo, 1.14.1::gentoo, 1.15::gentoo sys-devel/binutils: 2.24-r3::gentoo sys-devel/gcc: 4.8.5::gentoo sys-devel/gcc-config: 1.7.3::gentoo sys-devel/libtool: 2.4.6::gentoo sys-devel/make: 4.1-r1::gentoo sys-kernel/linux-headers: 3.18::gentoo (virtual/os-headers) sys-libs/glibc: 2.20-r2::gentoo Repositories: gentoo location: /usr/portage sync-type: rsync sync-uri: rsync://rsync.gentoo.org/gentoo-portage priority: -1000 science location: /usr/portage/local/layman/science masters: gentoo priority: 0 vmware location: /usr/portage/local/layman/vmware masters: gentoo priority: 1 x-portage location: /usr/local/portage masters: gentoo priority: 2 ACCEPT_KEYWORDS="amd64" ACCEPT_LICENSE="*" CBUILD="x86_64-pc-linux-gnu" CFLAGS="-O2 -pipe -march=native -fno-stack-protector" CHOST="x86_64-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/lib64/libreoffice/program/sofficerc /usr/share/config /usr/share/gnupg/qualified.txt" CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/dconf /etc/env.d /etc/fonts/fonts.conf /etc/gconf /etc/gentoo-release /etc/revdep-rebuild /etc/sandbox.d /etc/terminfo /etc/texmf/language.dat.d /etc/texmf/language.def.d /etc/texmf/updmap.d /etc/texmf/web2c" CXXFLAGS="-O2 -pipe -march=native -fno-stack-protector" DISTDIR="/usr/distfiles" EMERGE_DEFAULT_OPTS="--quiet-build=n --autounmask-write --keep-going" FCFLAGS="-O2 -pipe" FEATURES="assume-digests binpkg-logs ccache config-protect-if-modified distlocks ebuild-locks fixlafiles merge-sync news parallel-fetch preserve-libs protect-owned sandbox sfperms split-log strict unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox usersync xattr" FFLAGS="-O2 -pipe" GENTOO_MIRRORS="http://ftp.heanet.ie/pub/gentoo/" LANG="es_ES.utf8" LDFLAGS="-Wl,-O1 -Wl,--as-needed" MAKEOPTS="-j3" PKGDIR="/usr/local/portage/packages" PORTAGE_CONFIGROOT="/" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --omit-dir-times --compress --force --whole-file --delete --stats --human-readable --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages" PORTAGE_TMPDIR="/var/tmp" USE="X a52 aac acl acpi alsa amd64 applet avahi bash-completion berkdb branding bzip2 cairo ccache cdda cddb cdr cli colord cracklib crypt cups cvs cxx dbus djvu dri dts dvd dvdr dvi eds emboss enchant encode evo exif fam fat ffmpeg firefox flac fortran fuse gdbm gif glamor gnome gnome-keyring gnome-online-accounts gphoto2 gpm gstreamer gtk gtk3 gtkstyle gui hfs http iconv infinality introspection irc jabber jpeg kpathsea latex lcms ldap libnotify libsecret libvisual lyx mad mms mmx mmxext mng modules monolithic-build mp3 mp4 mpeg multilib musicbrainz nautilus ncurses network-cron networkmanager nls nptl ntfs numa ogg opengl openmp optimized-qmake pam pango pch pdf png policykit ppds pulseaudio qt3support qt4 readline reiserfs scanner sdl seccomp session smp spell sse sse2 ssl startup-notification svg systemd t1lib tcpd theora threads tiff tracker truetype udev udisks unicode upnp upnp-av upower usb v4l vaapi vala vcd vdpau vorbis wifi wxwidgets x264 xattr xcb xfs xml xv xvid youtube zeroconf zlib" ABI_X86="64" ALSA_CARDS="hda-intel" APACHE2_MODULES="authn_core authz_core socache_shmcb unixd actions alias auth_basic authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache cgi cgid dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter file_cache filter headers include info log_config logio mem_cache mime mime_magic negotiation rewrite setenvif speling status unique_id userdir usertrack vhost_alias" CALLIGRA_FEATURES="kexi words flow plan sheets stage tables krita karbon braindump author" CAMERAS="ptp2" COLLECTD_PLUGINS="df interface irq load memory rrdtool swap syslog" CPU_FLAGS_X86="mmx mmxext sse sse2 sse3 sse4_1 ssse3" ELIBC="glibc" GPSD_PROTOCOLS="ashtech aivdm earthmate evermore fv18 garmin garmintxt gpsclock itrax mtk3301 nmea ntrip navcom oceanserver oldstyle oncore rtcm104v2 rtcm104v3 sirf superstar2 timing tsip tripmate tnt ublox ubx" INPUT_DEVICES="evdev" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LIBREOFFICE_EXTENSIONS="presenter-console presenter-minimizer" LINGUAS="es es_ES en_US" OFFICE_IMPLEMENTATION="libreoffice" PHP_TARGETS="php5-5" PYTHON_SINGLE_TARGET="python2_7" PYTHON_TARGETS="python2_7 python3_4" RUBY_TARGETS="ruby20 ruby21" USERLAND="GNU" VIDEO_CARDS="fbdev vesa radeon" XTABLES_ADDONS="quota2 psd pknock lscan length2 ipv4options ipset ipp2p iface geoip fuzzy condition tee tarpit sysrq steal rawnat logmark ipmark dhcpmac delude chaos account" Unset: CC, CPPFLAGS, CTARGET, CXX, INSTALL_MASK, LC_ALL, PORTAGE_BUNZIP2_COMMAND, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS, USE_PYTHON
About X configuration, this is the only file (generated with systemd helpers for configuring keymaps via localectl): $ cat 00-keyboard.conf # Read and parsed by systemd-localed. It's probably wise not to edit this file # manually too freely. Section "InputClass" Identifier "system-keyboard" MatchIsKeyboard "on" Option "XkbLayout" "gb,es" Option "XkbVariant" "," EndSection The other machines only have "es" as Layout and no Variant As a side note, I have found this spidermonkey bug about stack corruption: https://bugzilla.mozilla.org/show_bug.cgi?id=1083464 Sorry if it's not related, but I don't know much about this stuff :'(
About logs... this is the only one I find't just before crashing: sep 14 10:53:25 optiplex760 gnome-session[31409]: xkbcommon: ERROR: xkb_x11_keymap_new_from_device: illegal device ID: -1 sep 14 10:53:25 optiplex760 kernel: gnome-shell[31446]: segfault at 8 ip 00007f9b8e6ee4e0 sp 00007ffe06f22aa8 error 6 in libxkbcommon.so.0.0.0[7f9b8e6d3000+3c
(In reply to Pacho Ramos from comment #8) > sep 14 10:53:25 optiplex760 gnome-session[31409]: xkbcommon: ERROR: > xkb_x11_keymap_new_from_device: illegal device ID: -1 This explains it. It means xkb_x11_keymap_new_from_device() will return NULL, and mutter will assign it to priv->keymap and later blindly attempt to use it (leading to the segfault). See src/backends/x11/meta-backend-x11.c in mutter source. As a workaround to prevent segfaults, we can patch mutter to add some checks that priv->keymap is not NULL before using it. Next question: why did xkb_x11_get_core_keyboard_device_id() return -1?
Created attachment 411892 [details, diff] workaround patch for x11-wm/mutter-3.16.3 See if this helps.
(In reply to Alexandre Rostovtsev from comment #10) > Created attachment 411892 [details, diff] [details, diff] > workaround patch for x11-wm/mutter-3.16.3 > > See if this helps. It still crashes. For that I have simply re-emerged spidermonkey:24 with -fno-stack-protector flag, that change makes it crash again. I have applied the patch to mutter and, even if it allows me to get gdm started after multiple tries and even login in gnome, it crashes again really often. I mean, without the patch it crashes near always, with this one it takes a few more seconds to crash :( Also, I don't understand what changes using -fno-stack-protector or not with spidermonkey to make this crash to appear :| (In reply to Alexandre Rostovtsev from comment #9) > Next question: why did xkb_x11_get_core_keyboard_device_id() return -1? I don't know how to reply to that :(
(In reply to Pacho Ramos from comment #11) > It still crashes. For that I have simply re-emerged spidermonkey:24 with > -fno-stack-protector flag, that change makes it crash again. I have applied > the patch to mutter and, even if it allows me to get gdm started after > multiple tries and even login in gnome, it crashes again really often. I > mean, without the patch it crashes near always, with this one it takes a few > more seconds to crash :( Can you get a backtrace for the crash with the patch? It should be in a different place - maybe the core cause of the problem this time.
(In reply to Pacho Ramos from comment #11) [...] > (In reply to Alexandre Rostovtsev from comment #9) > > Next question: why did xkb_x11_get_core_keyboard_device_id() return -1? > > I don't know how to reply to that :( Well, maybe this error doesn't occur at first, looking again to all logs (sorry, journald becomes huge and makes it scroll really slow :S), sometimes it's logged before, others after the crash. Probably it returns -1 because X crash and, then, the error should be "normal" (In reply to Alexandre Rostovtsev from comment #12) > Can you get a backtrace for the crash with the patch? It should be in a > different place - maybe the core cause of the problem this time. I will try, it will require a bit of time as this machine doesn't have anything compiled with debugging symbols yet :)
Created attachment 411910 [details] backtrace with the patch applied This is after applying the patch... but it looks really short :/
Instead of a segfault I get a trap :|
Compiling spidermonkey using -O0 lets me to also use -fno-stack-protector without issues... but using -O1 + -fno-stack-protector makes it crash again The problem is that I have tried to compile it appending -O0 and the options supposedly added by -O1 from "man gcc" and it still works !? # CFLAGS="-O1 -pipe -fno-stack-protector" CXXFLAGS="${CFLAGS}" emerge -1 spidermonkey:24 -> crashes # CFLAGS="-O0 -pipe -fno-stack-protector" CXXFLAGS="${CFLAGS}" emerge -1 spidermonkey:24 -> works # CFLAGS="-O0 -fauto-inc-dec -fcompare-elim -fcprop-registers -fdce -fdefer-pop -fdelayed-branch -fdse -fguess-branch-probability -fif-conversion2 -fif-conversion -fipa-pure-const -fipa-profile -fipa-reference -fmerge-constants -fsplit-wide-types -ftree-bit-ccp -ftree-builtin-call-dce -ftree-ccp -ftree-ch -ftree-copyrename -ftree-dce -ftree-dominator-opts -ftree-dse -ftree-forwprop -ftree-fre -ftree-phiprop -ftree-slsr -ftree-sra -ftree-pta -ftree-ter -funit-at-a-time -fomit-frame-pointer -pipe -fno-stack-protector" CXXFLAGS="${CFLAGS}" emerge -1 spidermonkey:24 -> works... even if it should be equivalent to -O1
(In reply to Pacho Ramos from comment #16) > Compiling spidermonkey using -O0 lets me to also use -fno-stack-protector > without issues... but using -O1 + -fno-stack-protector makes it crash again > There's another package in between here though, right? dev-libs/gjs?? Checking out configure.ac for GJS, there's an --enable-coverage that forces CFLAGS="$CFLAGS -g -O0 --coverage"; I don't know how GJS gets configured by default in gentoo but if coverage is default-enabled might it require spidermonkey to be built similarly (with -O0) the way you are reporting??
Indeed, gjs is the "link" with spidermonkey, but we explicitly pass --disable-coverage to it and, indeed, it's being build with normal CFLAGS as taken from make.conf, it's not forcing -O0 on Gentoo :/
As 3.16 GNOME-Shell is no longer in portage, is this bug still relevant?
Closing due to lack of updates, reopen if still relevant.
