Created attachment 30385 [details] Xorg log for desktop booted without KMS This happens on a Lenovo ThinkCentre M57 series (USFF) using Arch linux with a stock 2.6.31 (i686) kernel. The dual head set up is formed from two identical Lenovo ThinkVision monitors, with the 1280x1024 standard resolution. I have upgraded to the 2.9.0 version of the intel driver and the monitor connected to the DVI connector doesn't work any more (see the Xorg log for details). For the 2.8 version of the driver I could have the set up working by disabling KMS at boot and thus having EXA handling 3D. xorg-server : 1.7.0.901 xf86-video-intel : 2.9.0-1 mesa : 7.6-1 libdrm : 2.4.15-1
Created attachment 30386 [details] dmesg for system booted without KMS
Created attachment 30387 [details] xrandr -v output
Created attachment 30388 [details] xorg conf
Created attachment 30390 [details] Xorg log - system w/o KMS
Created attachment 30391 [details] dmesg output - system w/o KMS
Created attachment 30392 [details] dmesg output - system w/o KMS
Created attachment 30393 [details] xrandr -v output
Created attachment 30395 [details] xorg.conf
Created attachment 30409 [details] Xorg log with just the DVI monitor plugged in Sorry for the flurry of edits above. I made some further tests and it seems that X won't start if there is just the DVI monitor connected.
So this bug is nothing to do with dual head, right? It's likely to be dup with 24255, 24282. Can you try with KMS enabled? We are not suggesting to use UMS at this point. btw, 2.8.0 doesn't support EXA.
From the log it seems that this is a SDVO device with two inputs and multiple outputs. And this is a dup of bug 24282. Will you please try the debug patch in https://bugs.freedesktop.org/show_bug.cgi?id=24282#c18 and attach the output of dmesg in KMS mode? thanks. *** This bug has been marked as a duplicate of bug 24282 ***
(In reply to comment #10) > So this bug is nothing to do with dual head, right? > > It's likely to be dup with 24255, 24282. > > Can you try with KMS enabled? We are not suggesting to use UMS at this point. > btw, 2.8.0 doesn't support EXA. > Yes, you are right... what I was doing was booting without any modeset parameters (which for the earlier drivers meant, Xorg would use EXA).
Created attachment 30442 [details] Xorg log with just the DVI monitor plugged in and with KMS enabled
Created attachment 30443 [details] dmesg with KMS enabled I see that the kernel can't read the monitor's EDID, as per #24255 bug's comments.
Created attachment 30444 [details] Xorg log with both monitors and KMS enabled
(In reply to comment #14) > Created an attachment (id=30443) [details] > dmesg with KMS enabled > > I see that the kernel can't read the monitor's EDID, as per #24255 bug's > comments. > Please try the debug patch in bug24282 and attach the output of dmesg(please add the boot option of "drm.debug=0x06"). Please also attach the output of dmidecode in bug24282. thanks.
Created attachment 30475 [details] dmesg with sdvo patch and drm.debug=0x06 After booting with 2.6.32-rc4, the version against the patch was made, I have some weird artefacts on my screen. (This also happens for another system with the same driver/kernel combination, so probably there's a bug for it already)
Will you please attach the output of dmidecode? Thanks.
I posted it in the bug #2482 . I thought that was what you asked, sorry for the misunderstanding. See here: http://bugs.freedesktop.org/attachment.cgi?id=30476
(In reply to comment #19) > I posted it in the bug #2482 . I mean #24282 ...
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.