|Summary:||garbled output with matroxfb and G400 dual head|
|Product:||[Fedora] Fedora||Reporter:||Noa Resare <noa>|
|Component:||kernel||Assignee:||Arjan van de Ven <arjanv>|
|Status:||CLOSED UPSTREAM||QA Contact:|
|Fixed In Version:||Doc Type:||Bug Fix|
|Doc Text:||Story Points:||---|
|Last Closed:||2004-06-15 10:25:24 UTC||Type:||---|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
Description Noa Resare 2004-01-25 23:59:00 UTC
Description of problem: When testing kernel-2.6.1-1.57 i tried to load the matrox framebuffer but the console output was grabled no matter what resolution i tried with fbset Version-Release number of selected component (if applicable): kernel-2.6.1-1.57 (i686 version) How reproducible: always Steps to Reproduce: 1. boot said kernel into runlevel 3 2. log in as root 2. load the matroxfb module with '/sbin/modprobe i2c-matroxfb' Actual results: The visual output on the console becomes completely garbeled and unreadable. I try with a couple of resolutions i know by running fbset from a remote machine and while the video mode changes it always ends up garbled and unreadable in some way. Expected results: At least some video modes should work with matroxfb Additional info: While I don't expect fixing a subsystem that is disabled per default in fedora to be a priority I would like to suggest disabling matroxfb in the kernel build until it is known to work. If you do, I promise to file a feature request when matroxfb in 2.6.x is fixed :)
Comment 1 Noa Resare 2004-01-26 00:00:54 UTC
Forgot to specify the hardware. /sbin/lspci gives the following description: "Matrox Graphics, Inc. Millennium G400 MAX/Dual Head 32Mb"
Comment 2 Andreas Thienemann 2004-06-15 01:02:23 UTC
matroxfb is broken in the current kernel series. for kernel-2.6.6-1.427 applying the patch from http://ftp.cvut.cz/matroxfb/matrox-2.6.7-rc2-c1818.gz fixes the issue. With this patch applied matroxfb works for me with a G400. Thus, I'd like to see redhat apply that patch to their kernel-tree.
Comment 3 Dave Jones 2004-06-15 10:25:24 UTC
a quicker path is to push Petr to push his changes upstream. if there's a reason he hasn't done so already, there's possibly a reason we shouldn't take this either.