Your message dated Tue, 8 Sep 2009 22:02:54 +0200 with message-id <20090908200254.ge19...@inutil.org> and subject line Re: Workaround found, but no solution for the bug has caused the Debian Bug report #420300, regarding kernel-2.6.18: MGA video driver crashes system during X -> fb or fb -> X switching to be marked as done.
This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what this message is talking about, this may indicate a serious mail system misconfiguration somewhere. Please contact ow...@bugs.debian.org immediately.) -- 420300: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=420300 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems
--- Begin Message ---Package: kernel-2.6.18 Version: linux-source-2.6.18 Severity: normal As soon as I started to do some work on my new Debian Etch system, I got several severe lockups per day. "Severe" means: Ctrl-Alt-Del has no effect, and Alt-Print s does not sync the filesystem. The lockups happen, if I switch from X to a tty, or from a tty to X. They don't happen every time, but with a chance of 1:20 - 1:50, about 5 or 10 times a day. (Thank God or somebody else for ext3...) The problem does not exist under Debian Sarge on the same computer. Or sort of: If switching from a certain X program (gthumb) to tty, there is sometimes a slight corruption under Sarge, which can be cleared easily by switching back, changing to a different Gnome workplace, and switching to tty again. Additional system information: 1) Framebuffer The Framebuffer is installed with the following Grub line: kernel /boot/vmlinuz-2.6.18 root=/dev/sda1 ro video=matroxfb:vesa:0x1B5,fh:85000,fv:110 (I calculated the line, when I installed the Matrox card under Debian Woody, many years ago.) It reports itself as Boot video device is 0000:01:00.0 <...> matroxfb: MTRR's turned on matroxfb: 800x600x24bpp (virtual: 832x6721) matroxfb: framebuffer at 0xCE000000, mapped to 0xdc880000, size 33554432 Console: switching to colour frame buffer device 100x37 fb0: MATROX frame buffer device matroxfb_crtc2: secondary head of fb0 was registered as fb1 <...> Linux agpgart interface v0.101 (c) Dave Jones agpgart: Detected an Intel 440BX Chipset. agpgart: AGP aperture is 256M @ 0xd0000000 <...> Matrox MGA G200/G400/G450/G550 YUV Video interface v2.01 (c) Aaron Holtzman & A'rpi mga_vid: Found MGA G400/G450 at 0000:01:00.0 mga_vid: MMIO at 0xde9a0000 framebuffer: 0xCE000000 mga_vid: OPTION word: 0x50044120 mem: 0x10 SGRAM mga_vid: detected RAMSIZE is 32 MB mga_vid: 1 supported cards found mga_vid: using major: 83 (assigned or default!) <...> [drm] Initialized drm 1.0.1 20051102 [drm] Initialized mga 3.2.1 20051102 on minor 0 agpgart: Found an AGP 1.0 compliant device at 0000:00:00.0. agpgart: Putting AGP V2 device at 0000:00:00.0 into 1x mode agpgart: Putting AGP V2 device at 0000:01:00.0 into 1x mode [drm] Initialized card for AGP DMA. 2) X My EIZO/Nanao CRT is a little older, and does not report its capabilities back to the video adapter. I install it in /etc/X11/xorg.conf with Mode "1152x864" DotClock 127.83 HTimings 1152 1176 1304 1504 VTimings 864 868 872 902 EndMode (I calculated the line, when I installed the Matrox card under Debian Woody.) Additional information: Kernel with video drivers and mga_vid module are compiled from stock Debian Etch Stable. Since my settings have worked well under Debian Woody and Debian Sarge, since quite some years, I assume a video driver problem. Ciao Hans-Joachim -- System Information: Debian Release: 4.0 APT prefers stable APT policy: (500, 'stable') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.18 Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
--- End Message ---
--- Begin Message ---On Mon, Aug 03, 2009 at 07:43:34PM +0200, Moritz Muehlenhoff wrote: > On Wed, May 09, 2007 at 01:59:37AM +0200, Hans-Joachim Zierke wrote: > > > > Several attempts were undertaken, to solve the problem with different > > video modes, especially modes very close to each other. No success. > > > > > > Several kernel attributes for the framebuffer were tried, like > > init - noinit etc. No success. > > > > > > Different settings for the X-Server were tried. Of these, setting > > Option "UseFBDev" "true" > > has shown an effect: Instead of a system crash after every ~20th switch > > or so, the system crashed for every ~2nd switching. > > > > > > Workaround for the problem is the following: Don't use the MGA driver > > for the X-Server. Instead, use the framebuffer driver for the X-Server. > > After doing so, the system didn't crash for several days, which is only > > a preliminary result, but already a phantastic one. > > (I haven't seen such unreliable system since the days of DOS cache > > programs playing with the A20 gate.) > > > > > > I'm not sure, which MGA driver is at fault: fb or xorg. > > Did you upgrade to Lenny in the mean time? If so, does the error > persist? No further feedback, closing the bug. Cheers, Moritz
--- End Message ---
