| View previous topic :: View next topic |
| Author |
Message |
eremini
Apprentice
Joined: 27 May 2005
Posts: 157
|
 Posted: Sun Jul 16, 2006 7:56 am Post subject: Glibc problem |
 |
|
Hello,
I have a kind of a big problem. Half of my apps (example kaffeine, kopete, konversation) fail to work because of this one error
konversation: pthread_mutex_lock.c:80: __pthread_mutex_lock: Assertion `mutex->__data.__owner == 0' failed.
I've recompiled the program in question and recompiled glibc like 3 times, didn't help. Here's emerge info:
Portage 2.1.1_pre2-r8 (default-linux/x86/2006.0, gcc-4.2.0-alpha20060603/vanilla, glibc-2.4-r4, 2.6.17-no4 i686)
=================================================================
System Settings
=================================================================
System uname: 2.6.17-no4 i686 Mobile Intel(R) Pentium(R) 4 - M CPU 2.00GHz
Gentoo Base System version 1.12.1
app-admin/eselect-compiler: 2.0.0_rc2-r1
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: 1.2.18.1
sys-devel/autoconf: 2.13, 2.60
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-r4
sys-devel/gcc-config: 2.0.0_rc1
sys-devel/libtool: 1.5.22
virtual/os-headers: 2.6.17
ACCEPT_KEYWORDS="x86 ~x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-O2 -march=pentium4 -fomit-frame-pointer -pipe -frename-registers -fno-ident -fweb"
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"
CONFIG_PROTECT_MASK="/etc/env.d /etc/env.d/java/ /etc/eselect/compiler /etc/gconf /etc/java-config/vms/ /etc/revdep-rebuild /etc/terminfo"
CXXFLAGS="-O2 -march=pentium4 -fomit-frame-pointer -pipe -frename-registers -fno-ident -fweb -fvisibility-inlines-hidden"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig distlocks metadata-transfer sandbox sfperms strict"
GENTOO_MIRRORS="http://distfiles.gentoo.org http://distro.ibiblio.org/pub/linux/distributions/gentoo"
LDFLAGS="-Wl,-O1 -Wl,--as-needed -Wl,--hash-style=both"
LINGUAS="en_GB"
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://rsync.gentoo.org/gentoo-portage"
USE="x86 X acpi alsa apm arts asf ati avi berkdb bitmap-fonts bzip2 cairo cgi cli crypt dbus dlloader dri dvd eds emboss encode fam fastcgi ffmpeg foomaticdb gdbm gif gpm gtk gtk2 hal hashstyle imlib insecure-savers isdnlog java jpeg kde kdeenablefinal kdehiddenvisibility libg++ libwww mad mikmod mmx motif mp3 mpeg mysql ncurses nls noamazon nptl nptlonly nsplugin ogg opengl pam pcre pdflib perl png pppd python qt qt3 quicktime readline reflection sdl session spell spl sse sse2 ssl tcpd threads tiff truetype truetype-fonts type1-fonts udev unicode usb vorbis win32codecs xcomposite xml xorg xscreensaver xv xvid zlib elibc_glibc input_devices_keyboard input_devices_mouse kernel_linux linguas_en_GB userland_GNU video_cards_radeon video_cards_fglrx video_cards_vesa"
Unset: CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, MAKEOPTS, PORTAGE_RSYNC_EXTRA_OPTS
Any ideas? |
|
| Back to top |
|
 |
i92guboj
Bodhisattva
Joined: 30 Nov 2004
Posts: 10315
Location: Córdoba (Spain)
|
| Posted: Sun Jul 16, 2006 8:19 am Post subject: |
|
|
Sure, no wonder:
| Code: | | gcc-4.2.0-alpha20060603 |
Your cflags and ldflags are not the safest ones either.
I doubt anyone will waste time here if you are using such unsafe settings and the latest gcc wich is still in alpha state, appart from being hardmasked by -* in portage... Maybe you should look into bugzilla instead. |
|
| Back to top |
|
|
eremini
Apprentice
Joined: 27 May 2005
Posts: 157
|
| Posted: Sun Jul 16, 2006 9:04 am Post subject: |
|
|
| I am aware of that, but really you should agree that its highly unlikely that has to do anything with gcc |
|
| Back to top |
|
|
Maedhros
Bodhisattva
Joined: 14 Apr 2004
Posts: 5511
Location: Durham, UK
|
| Posted: Sun Jul 16, 2006 9:06 am Post subject: |
|
|
I'd say it is fairly likely, actually, which is why that version of gcc is hardmasked and therefore not supported. Using --as-needed isn't supported yet either.
Moved from Portage & Programming to Unsupported Software.
_________________
No-one's more important than the earthworm. |
|
| Back to top |
|
|
i92guboj
Bodhisattva
Joined: 30 Nov 2004
Posts: 10315
Location: Córdoba (Spain)
|
| Posted: Sun Jul 16, 2006 9:28 am Post subject: |
|
|
| eremini wrote: | | I am aware of that, but really you should agree that its highly unlikely that has to do anything with gcc |
Hehe, well, if you think so...
Actually, that error could be caused by the version of gcc and/or the ldflags. I dont think that cflags are an issue, but it is also possible.
If you failt to see why those errors could be related to gcc, you, clearly, do not have the knowledge to run an alpha prerelease of gcc. And you will be better by using stagle core components, or, at least, ~arch. As I said, anyway, that is not supported, and, unless willing to look in bugzilla, is will be difficult to get some help here. |
|
| Back to top |
|
|
enderandrew
l33t
Joined: 25 Oct 2005
Posts: 731
|
| Posted: Sun Jul 16, 2006 10:03 am Post subject: |
|
|
I build my entire world with --as-needed and it doesn't break a single package.
However, gcc-4.2.0 is pretty bug ridden. The gcc devs themselves consider it a pretty big regression from 4.1 at the moment.
_________________
Nihilism makes me smile. |
|
| Back to top |
|
|
eremini
Apprentice
Joined: 27 May 2005
Posts: 157
|
| Posted: Sun Jul 16, 2006 11:14 am Post subject: |
|
|
| Forgive me for not being a programmer, but of course you have a point. GCC builds glibc, hence the errors might crop in, I just didn't think of the obvious. New gcc 4.2 snapshot has been released, I'll emerge that, then recompile glibc with it, see what that did. If that still doesn't work I'll recompile glibc with 4.1 I haven't pruned. Sounds fine? |
|
| Back to top |
|
|
enderandrew
l33t
Joined: 25 Oct 2005
Posts: 731
|
| Posted: Sun Jul 16, 2006 11:47 am Post subject: |
|
|
For what it is worth, if you're going to live on the bleeding edge with GCC, I'd consider looking into nesl's overlay for glibc and binutils.
_________________
Nihilism makes me smile. |
|
| Back to top |
|
|
i92guboj
Bodhisattva
Joined: 30 Nov 2004
Posts: 10315
Location: Córdoba (Spain)
|
| Posted: Sun Jul 16, 2006 11:58 am Post subject: |
|
|
| eremini wrote: | | Forgive me for not being a programmer, but of course you have a point. GCC builds glibc, hence the errors might crop in, I just didn't think of the obvious. New gcc 4.2 snapshot has been released, I'll emerge that, then recompile glibc with it, see what that did. If that still doesn't work I'll recompile glibc with 4.1 I haven't pruned. Sounds fine? |
Nothing to forgive, if something breaks will be in your box... I mean, you dont just go and emerge it, you have to unmask it, and it is not even keyworded, but hard marked. You need to make some research to learn how to unmask a hard masked package.
I dont mind helping and I do it gladly, just want you to understand that such masked packages are there for a reason. And gcc/glibc are core pieces of the OS... Alpha's are pre-beta versions, they are unstable by nature, they are not even testing versions, but early devel snapshots. I would never put such a core alpha package in my system unless I am willing to debug some code. You can emerge e17 and if that fails it does not harm, couse it is just userland stuff, if it does fail, you can just remove it or leave it lying around, but the core of the system must have a minimal degree of stability. 
Your plans above sounds right to me, but I would stay as far from hard masked stuff in which regards glibc and gcc. If you want to be updated, you can just put the packages in package.keywords or run with ~arch. I havent synced and I dont know how's the thing right now, but I would advice you to stay away from unstable hardmasked stuff in wich regards gcc and glibc.
Anyway, if you run into problems and your toolchain gets messed in between those compilations, you can save this link:
http://tobe.stylesyndication.com/linux/binaries.html
There you can find binary packages to restore gcc and glibc in case something gets really messed, those packages have been compiled in gentoo and tarred using quickpkg, so, dont worry about filling your hard disk with some crap. You can use those if the thing gets complicated, then select the right compiler from the binary package and then recompile 4.1.x using portage. From there, you can recompile the whole system (or at least the packages that are giving you those annoying errors).
I hope you can fix it easily |
|
| Back to top |
|
|
eremini
Apprentice
Joined: 27 May 2005
Posts: 157
|
| Posted: Sun Jul 16, 2006 5:23 pm Post subject: |
|
|
Hey thanks a lot for helping! I've been using gentoo for 2 years, so I know how hard masking works  I do understand and I am willing to take the risk (well I allready did i guess). Yeah the binaries are pretty usefull, I used them a couple of days ago, when my sys was completely messed, for some reason though they didn't work, but what worked is copying /lib from livecd to /mnt/gentoo/lib. But thats another story
An update: I compiled the new snapshot, recompiled glibc - no go. So now eselect compiler to gcc 4.1 and recompile glibc again Oh joy |
|
| Back to top |
|
|
eremini
Apprentice
Joined: 27 May 2005
Posts: 157
|
| Posted: Mon Jul 17, 2006 6:06 pm Post subject: |
|
|
| right, all solved it was gcc, rebuild glibc with 4.1 all good. Thanks a ton 6thpink. |
|
| Back to top |
|
|
i92guboj
Bodhisattva
Joined: 30 Nov 2004
Posts: 10315
Location: Córdoba (Spain)
|
| Posted: Mon Jul 17, 2006 8:01 pm Post subject: |
|
|
| Glad to hear that. I know how annoying can those errors be. Nice that they are gone at last |
|
| Back to top |
|
|
ericxx2005
Apprentice
Joined: 05 Apr 2005
Posts: 231
Location: 10 square miles surrounded by reality
|
| Posted: Fri Aug 04, 2006 3:51 pm Post subject: |
|
|
| Quote: | | pthread_mutex_lock.c:80: __pthread_mutex_lock: Assertion `mutex->__data.__owner == 0' failed. |
I just ran into the same problem. About 1/2 my apps were throwing that error. I fixed it by re-emerging glibc with gcc 4.1.1, and now i'm emerge -e system (x2) to make sure everything's okay. I'll just have to play around with the new gcc when it's a little more finished.
_________________
AMD 1090T | 2 GB | AMD 890GX | Gentoo x86_64
Core i7-920 | 6 GB | AMD 5850 | Windows 7 Pro |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Unsupported Software
