My latop's native resolution is 1280x768, which the new intel driver sets properly. However, when starting a gnome-session, it seems to behave as if I had a 1024x768 resolution. This is the output of xrandr when run twice in succession: khashayar@dionysos ~/bugs $ xrandr Screen 0: minimum 320 x 200, current 1280 x 768, maximum 1280 x 1280 VGA connected 1024x768+0+0 (normal left inverted right) 0mm x 0mm 1024x768 60.0* 800x600 60.3 640x480 59.9 LVDS connected 1280x768+0+0 (normal left inverted right) 0mm x 0mm 1280x768 59.2*+ 1152x768 54.8 1024x768 85.0 75.0 70.1 60.0 832x624 74.6 800x600 85.1 72.2 75.0 60.3 56.2 640x480 85.0 72.8 75.0 59.9 720x400 85.0 640x400 85.1 640x350 85.1 TMDS disconnected (normal left inverted right) khashayar@dionysos ~/bugs $ xrandr Screen 0: minimum 320 x 200, current 1280 x 768, maximum 1280 x 1280 VGA disconnected 1024x768+0+0 (normal left inverted right) 0mm x 0mm LVDS connected 1280x768+0+0 (normal left inverted right) 0mm x 0mm 1280x768 59.2*+ 1152x768 54.8 1024x768 85.0 75.0 70.1 60.0 832x624 74.6 800x600 85.1 72.2 75.0 60.3 56.2 640x480 85.0 72.8 75.0 59.9 720x400 85.0 640x400 85.1 640x350 85.1 TMDS disconnected (normal left inverted right) I'm running xf86-video-intel from git and xserver 1.2.99.903. I'll attach xorg.conf, log file and a couple of descriptive screenshots.
Created attachment 9398 [details] Xorg log file
Created attachment 9399 [details] Xorg configuration file
Screenshot 1: http://www.naderehvandi.net/backup/panel-problem.png Screenshot 2: http://www.naderehvandi.net/backup/panel-problem2.png Note the size of gnome-panel in the first screenshot. It stretches 1024 points instead of 1280. The second screenshot is maximized view of the first screenshot.
The problem still persists after the release of RC4.
I did some more trying and this is what I have now, maybe it can be of some help. This is what I did 1) Started the xserver with the old driver (resolution working fine with 915resolution hack) 2) quit the xserver and installed the latest intel driver from git 3) started the xserver (resolution was fine) 4) restarted xserver (resolution was messed up as described in this report). I will now attach the two log files, one from when the resolution was picked up fine, and one from when it wasn't.
Created attachment 9495 [details] Log file from when resolution was not working properly
Created attachment 9496 [details] Log file from when resolution was working properly
Shouldn't the bug summary be: 1280x768 resolution set, but behaviour is as if resolution was >>>1024x768<<<?
(In reply to comment #8) > Shouldn't the bug summary be: 1280x768 resolution set, but behaviour is as if > resolution was >>>1024x768<<<? > Woops. My wrong. Fixed now...
I think the problem is that the GNOME panel restricts itself to the primary/smaller Xinerama screen, the VGA output in your case. Is an external display connected to it?
(In reply to comment #10) > I think the problem is that the GNOME panel restricts itself to the > primary/smaller Xinerama screen, the VGA output in your case. Is an external > display connected to it? > Sorry for not stating this clearer before. No, there is no external screen involved here at all.
The same behavior has been reported in Debian bug http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=420292 (especially http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=420292;msg=45). The bug report was initially about gnome-screensaver, but it looks like it is actually the intel driver wrongly reporting 2 monitors and gnome-screensaver not supporting this well. Interestingly, both the Debian bug submitter and Khashayar Naderehvandi have a 82852/855GM. And I know another person seeing this behavior on the same chipset.
I see a variant of what may be this bug. My chipset randomly does or does not detect the VGA connector as connected, even though a monitor is consistently connected to it. Since I'm using the VGA display as my only display, this means my display doesn't work half the time. I've also seen one total system crash while repeatedly starting the X server to investigate this behavior; I can file a separate bug for that if you think it deserves one. I'm using the Intel driver from Debian sid, version 2.0.0-1. My chipset is an "Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)" (I believe it's an 855GME, though my memory is hazy). This actually isn't a laptop; it's a small-form-factor PC that apparently uses a mobile display chipset. This bug was the first time I had heard anything about it having an LVDS output :-). I don't know whether there's any truth to that as far as the hardware goes, but if there is any LVDS capability in this computer it's not accessible to me. I'll attach my configuration and logs from a successful start (one where the VGA display is detected as connected) and an unsuccessful one (one where it isn't). Let me know if you want any more information. Cheers.
Created attachment 10042 [details] Andrew's xorg.conf
Created attachment 10043 [details] Andrew's log - failure
Created attachment 10044 [details] Andrew's log - success
Should be fixed with b31bef1a8effa9acb6de7edd206b9d8c48d88144, committed last week.
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.