Created attachment 29171 [details] xorg.conf I've managed to get kms working on my machine. Unfortunately the dual-head setup does not work as with user modesetting: * both of the monitors are set up at the same resolution, not at their native ones. I can fix this manually with xrandr (and the kernel framebuffer drives both displays at native resolutions, so I get flicker on one display when switching between X and the linux console). $ xrandr (with kms) Screen 0: minimum 320 x 200, current 2560 x 1024, maximum 8192 x 8192 DVI-0 connected 1280x1024+1280+0 (normal left inverted right x axis y axis) 376mm x 301mm 1280x1024 60.0*+ 60.0 1152x864 75.0 1024x768 75.1 70.1 60.0 60.0 832x624 74.6 800x600 72.2 75.0 60.3 60.0 56.2 640x480 72.8 75.0 66.7 60.0 60.0 720x400 70.1 DVI-1 connected 1280x1024+0+0 (normal left inverted right x axis y axis) 546mm x 352mm 1920x1200 60.0 + 1600x1200 60.0 1680x1050 60.0 1600x1000 60.0 1280x1024 75.0* 1280x960 60.0 1152x864 75.0 1024x768 75.1 60.0 832x624 74.6 800x600 75.0 60.3 640x480 75.0 60.0 720x400 70.1 Note that DVI-1 is not at the native resolution. Furthermore, when I set up DVI-0 right of DVI-1 (like above), there is no mouse cursor visible on DVI-0. Strangely it works in clone mode (ie. when DVI-0 is at 1280x1024+0+0). I'll attach Xorg.log and xorg.conf. Hardware: 4 core amd64, 4gb ram, amd 8xxx AGP chipset lspci -nn of the graphics card 05:00.0 VGA compatible controller [0300]: ATI Technologies Inc RV570 [Radeon X1950 Pro] [1002:7280] (rev 9a) 05:00.1 Display controller [0380]: ATI Technologies Inc RV570 [Radeon X1950 Pro] (secondary) [1002:72a0] (rev 9a) Software distribution: debian amd64 unstable ddx: xf86-video-ati-6.12.2-228-g794ae74 (from git) drm: libdrm-2.4.13 (from git) kernel: v2.6.31-rc8-38-g37d0892 with drm-next merged in (drm-next: v2.6.31-rc6-271-gff6fdbe) xorg: 1.6.3-1 (from debian)
Created attachment 29172 [details] [review] Xorg.0.log This is with kms and multiple xrandr changes to try to find out how to get a working cursor on DVI-0.
Can you attach your dmesg as well?
Created attachment 29190 [details] dmesg with kms I've created this log by directly booting into kms X. To my surprise, everything worked, i.e. both displays at native resolution and cursor working on both screens. Now as soon as I've restarted X, the old, buggy behaviour reappeared. The log contains both runs. So it looks like the problem only happens when X has already run once (doesn't matter wheter ums or kms). I didn't notice this till now because usually I boot up in ums. Then I kill X, rmmod radeon and finally modprobe radeon with modesetting enabled.
On Thu, Sep 03, 2009 at 01:04:51PM -0700, bugzilla-daemon@freedesktop.org wrote: > --- Comment #3 from Daniel Vetter <daniel@ffwll.ch> 2009-09-03 13:04:50 PST --- > Created an attachment (id=29190) > --> (http://bugs.freedesktop.org/attachment.cgi?id=29190) > dmesg with kms > > I've created this log by directly booting into kms X. To my surprise, > everything worked, i.e. both displays at native resolution and cursor working > on both screens. Now as soon as I've restarted X, the old, buggy behaviour > reappeared. The log contains both runs. So it looks like the problem only > happens when X has already run once (doesn't matter wheter ums or kms). I > didn't notice this till now because usually I boot up in ums. Then I kill X, > rmmod radeon and finally modprobe radeon with modesetting enabled. Correction: When booting up directly with kms, it does only work because X is actually using user mode setting. At least that's what's in the log file. And the flickering when switching between X and the linux console confirms this. So dual head with kms does not work, not even on the first time. I suspect that we have a race condition between when X loads the radeon modules and when the kernel has finished initializing kms. This somehow results in X assuming there's no kms support present -Daniel
*** Bug 23883 has been marked as a duplicate of this bug. ***
*** This bug has been marked as a duplicate of bug 23234 ***
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.