Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 124030 - grey screen in gdm/kdm after logout, using ati-drivers
Summary: grey screen in gdm/kdm after logout, using ati-drivers
Status: RESOLVED UPSTREAM
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] Unspecified (show other bugs)
Hardware: All Linux
: High normal (vote)
Assignee: X11 External Driver Maintainers
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-02-24 22:51 UTC by Simon Ruggier
Modified: 2007-05-24 22:34 UTC (History)
3 users (show)

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


Attachments
condensed xorg.conf & output of emerge --info (emergeinfo-gn2,11.48 KB, text/plain)
2006-09-23 15:21 UTC, Rob Couto
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Simon Ruggier 2006-02-24 22:51:34 UTC
Sorry in advance for how little information I can give, but despite experiencing this problem for a while I have almost no idea what is causing it.  On logout, the gdm will display its themed greeter, only it is all grey, with a white rectangle in the center where the text box would be.  Switching to one of the other six virtual terminals results in a screenful of vertical lines of varying colours.  Key presses still work, and I remember trying to log back in one time when this happened.  It did something, but the display was still unusably corrupt, so the only thing I can do when this happens is reboot.

Using ati-drivers-8.22.5, this happened on every logout, but with ati-drivers-8.21.7, it happens intermittently.  I've looked at /var/log/Xorg.0.log immediately after shutting down the system after an occurrence (viewing from another OS), and I don't see any glaring discrepancies.  I'd be willing to gather info for anyone interested in helping figure this out, but I'm not sure what to look at other than /var/log/Xorg.0.log.

# emerge --info
Portage 2.0.54 (default-linux/x86/2005.1, gcc-3.4.5, glibc-2.3.5-r2, 2.6.15-gentoo-r1 i686)
=================================================================
System uname: 2.6.15-gentoo-r1 i686 Mobile AMD Athlon(tm) 64 Processor 3400+
Gentoo Base System version 1.6.14
dev-lang/python:     2.3.5, 2.4.2
sys-apps/sandbox:    1.2.12
sys-devel/autoconf:  2.13, 2.59-r6
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r1
sys-devel/binutils:  2.16.1
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-O2 -march=athlon64 -pipe"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.5/env /usr/kde/3.5/share/config /usr/kde/3.5/shutdown /usr/kde/3/share/config /usr/lib/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/ /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O2 -march=athlon64 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig distlocks sandbox sfperms strict"
GENTOO_MIRRORS="http://gentoo.osuosl.org/ http://mirror.datapipe.net/gentoo http://gentoo.chem.wisc.edu/gentoo/ http://distfiles.gentoo.org http://www.ibiblio.org/pub/Linux/distributions/gentoo"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/etc/portage/overlay"
SYNC="rsync://rsync.namerica.gentoo.org/gentoo-portage"
USE="x86 3dnow 3dnowext X X509 Xaw3d a52 aac aalib acpi alsa apache2 apm arts asf asm audiofile avi bash-completion berkdb bitmap-fonts bl bootsplash bzip2 cairo cddb cdparanoia cdr chroot cjk cpudetection crypt cscope ctype cups curl curlwrappers dbus doc dts dv dvd dvdr dvdread dvi edl eds emboss encode esd exif expat fam fbcon ffmpeg firefox flac foomaticdb fortran ftp gd gdbm gif glitz glut gmp gnome gnutls gpm gs gstreamer gtk gtk2 gtkhtml hal hpn iconv idn ieee1394 imlib ipv6 java javascript joystick jpeg jpeg2k kde kdeenablefinal lcms ldap libcaca libg++ libwww live logitech-mouse logrotate lua lzo mad matroska md5sum memlimit mikmod mmap mmx mmxext mng motif mp3 mpeg musicbrainz mysql nas nautilus ncurses network new-login nfs nls nntp no-htdocs nptl nptlonly nsplugin offensive ogg oggvorbis on-the-fly-crypt openal opengl oss pam pcmcia pcre pda pdflib perl php png python qt quicktime readline real rtc samba sasl sdl session simplexml slp soap sockets speex spell spl sqlite sse sse2 ssl startup-notification svg svga t1lib tcpd tetex tga theora threads tiff tokenizer truetype truetype-fonts type1-fonts udev unicode usb userlocales v4l v4l2 vcd vidix vim-with-x vorbis vv win32codecs wmf xanim xine xinerama xml2 xpm xprint xrandr xscreensaver xv xvid xvmc zeroconf zlib video_cards_radeon userland_GNU kernel_linux elibc_glibc"
Unset:  ASFLAGS, CTARGET, LANG, LC_ALL, LDFLAGS, LINGUAS
Comment 1 Simon Ruggier 2006-02-26 09:48:40 UTC
It's happening intermittently using drivers 8.22.5 with vanilla-sources-2.6.16-rc4, as opposed to with gentoo-sources 2.6.15-r1.  For all I know though, my video drivers and kernel could have nothing to do with the problem.
Comment 2 Simon Ruggier 2006-02-26 10:29:26 UTC
I switched to the internal X11 driver, ati, and logged out at least 7 times, and this problem didn't happen (I did get a black screen once, but I'm assuming it's not the same problem).  However, I wouldn't use this driver, since then I'd be giving up hardware accelerated 3D functions.
Comment 3 Marnix Klooster 2006-03-16 21:44:47 UTC
Same problem here.  I started to see this after I upgraded from ati-drivers-8.14.13-r3 to ati-drivers-8.21.7-r1.

(And this happened because somewhere inbetween Mar 07, 2006 20:22:38 and Mar 09, 2006 18:33:33 (both times in GMT+1) someone declared ati-drivers-8.21.7-r1 to be stable on x86.  Was that intentional?  All surrounding versions are not stable; apart from this one the latest stable is (and has been since half December) ati-drivers-8.14.13-r3.
Comment 4 Simon Ruggier 2006-04-02 06:03:36 UTC
I just started using kde today, and sadly, this bug happens to me in the kdm as well, only now, about 4 seconds after the grey screen appears, my computer reboots, saving me the trouble :)

I'd like to note that I don't recall ever seeing this bug in Ubuntu, also using ati-drivers.
Comment 5 Joaquim Uchoa 2006-04-04 08:00:27 UTC
(In reply to comment #4)
Maybe this help someone (sorry about the poor English):

I have a HP Pavilion 6007, amd64 with a ATI XPress 200M. In this notebook, I can choose between using dedicated memory (sideport) or shared memory (UMA). 

If I choose to use sideport, the use of ATI Drivers always produces grey screen
(in Gentoo and Fedora). I've made some tests and discovered that X can't find 
the sideport memory. If I disable the use of DRI, it works ok, but without
3D acceleration. Also, I have no problem with ati opensource drivers (but, 
again, no 3D acceleration).

If I use shared memory, I can use 3D acceleration, without problems (in Gentoo
and Fedora).

So, I think this problem occurs when X can't find memory. I suggest to boot in terminal mode and try to start X from terminal, redirecting output to a file.
Comment 6 Simon Ruggier 2006-06-14 20:23:51 UTC
This bug still happens to me with ati-drivers-8.25.18 (and modular Xorg), and it is not the same problem as comment #5.  Probably an upstream issue though, I experience it in Ubuntu Dapper as well on occasion.
Comment 7 Rob Couto 2006-09-23 15:21:37 UTC
Created attachment 97897 [details]
condensed xorg.conf & output of emerge --info

Same deal here... similar but not identical symptoms. Have an AthlonXP 2500+ and Radeon 9200SE. Error occurs with both gdm and kdm, after logout it hangs but doesn't reboot, and the one time I got back to a text VT, it was fully corrupted. The first time, it wa so bad the Magic SysRq keystrokes weren't rebooting it. I'm using the 2.6.17-beyond3 patchset and ati-drivers-8.28.8, modular X.org, GCC 4.1.1, on a days-old 2006.1 build.

Workaround: Add a line to the Device section in xorg.conf, 
Option "RenderAccel" "false"

I tried it on a hunch and a vague memory, so it's all good in both kdm and gdm. Only problem is, I LIKE RenderAccel! Personally, I'd rather reboot instead of logout every time but this isn't my machine.

Reproducible: Always
Steps to Reproduce:
1. Start xdm (gdm or kdm)
2. Login as non-root user
3. Logout

Actual Results:  
Screen blinks, then begins to redraw login screen and freezes after the first simple shapes are drawn. Attempts to switch VTs result in nothing at all, or display corruption.

Expected Results:  
Screen redraws and login manager continues operating normally.

Fix/Workaround: Disable RenderAccel in display options.
Comment 8 Rob Couto 2006-09-23 15:38:08 UTC
A few more things:

1. Problem seems limited to KDE login, when using fluxbox it returned normally to the login screen.
2. On the most recent try with gdm, I hit Ctrl+Alt+Bksp immediately after the drawing froze which successfully killed the X server, then gdm continued normally.
3. Double-checked that logging in as root does not trigger the freeze, since it was root on fluxbox that made me think it was limited to non-root users. Root on KDE is also fine.

Hope that helps.
Comment 9 Rob Couto 2006-09-29 12:10:03 UTC
Crap! the RenderAccel trick doens't work now, maybe because I just ignorantly updated to 8.29.6, which has no support for 9200 cards, and then downgraded again. This is to much, going to recommend they replace the card. In the meantime, will just use the X.org driver.
Comment 10 Timo Josten 2007-05-13 13:48:38 UTC
Same problem here, using 
[I] x11-drivers/ati-drivers
     Installed versions:  8.35.5 (acpi kernel_linux qt3)

Reproducable:
-Login with KDM
-Logout (im using KDE)
-Black screen, nothing happens.

Comment 11 Marien Zwart (RETIRED) gentoo-dev 2007-05-24 22:34:33 UTC
This is most likely an upstream issue. You will probably have better luck filing this in the unofficial bugzilla at http://ati.cchtml.com/, unless you have reason to believe this can be fixed in the ebuild. In that case please reopen.