Created attachment 62136 [details] Xorg.log This occurred the first time with an earlier kernel version (3.2, maybe 3.1, but not 3.3). After the wake-up, the system functions normally for some time until the display becomes black, which is not triggered by the screensaver. Changing consoles, in which the mouse cursor appears, with Alt-Function keys when this occurs might make the display return to normal. Another way to resume normal display is clicking on the top left corner of the screen while in text-mode, which causes the KDE applications menu to open. Another symptom is that occasionally input by keyboard in X is echoed in the text-mode console I last switched from. Running opensuse 11.3 on HP mini 5102 Intel Pineview GM with > rpm -q kernel-desktop kernel-desktop-3.4.0-1.1.i686 > rpm -q xorg-x11-driver-video xorg-x11-driver-video-7.6-80.1.2.i586 > rpm -q xorg-x11-server xorg-x11-server-7.6_1.10.4-36.2.1.i586 > rpm -q Mesa Mesa-7.11-11.1.2.i586 > rpm -q libdrm libdrm-2.4.26-15.1.2.i586
petros JCGs&s setter
Please boot with drm.debug=0xe added to your kernel commandline, reproduce the issue and then attach the complete output of dmesg. Also, 2.16.0 for the xf86-video-intel driver is a bit old and we've fixed a lot of bugs recently about handling output state. Can you please upgrade to the latest version (preferrably even git - all the commits nowadays go into the more experimental sna backend, so all the other commits for the uxa backend are low-risk bugfixes only).
Please remove my previous message, which was sent inadvertently. I was typing in the text-mode console, and it appeared in X too. I believe I have found a pattern for the blank screen. Since the Xorg and text-mode consoles are mixed up, the power-saving feature of the text-mode console comes into effect if there is no key press for some minutes. If I press a key while the screen is blank, then the display always returns to text-mode, and waiting for some seconds returns to X. I should add that the console that corresponds to X is random; when the system starts it is 7 and now it became 1, which is also a normal text-mode login console. I will try your suggestion later, I'm too busy right now.
Should be fixed with: commit 343065a605646b4c1fee8b1ca07b715283813b11 Author: Daniel Vetter <daniel.vetter@ffwll.ch> Date: Fri Jun 15 11:01:22 2012 +0200 drm/fb-helper: delay hotplug handling when partially bound Patch is merged since 3.6-rc1. If this does not fix your issue, please reopen the bug, thanks.
It just occurred again on 3.6.0, reopening.
We still need them logfiles ... Also, please retest on 3.7 since that contains a completely reworked modeset sequence around suspend/resume.
Also you will need to update -intel to 2.20.9+ in order to get a few fixes with DRI SwapBuffers vs DPMS/resume.
This is very rare, so it is going to take some time to get the log. I will update xorg and the kernel in the weekend, when I will have more time.
Created attachment 70661 [details] Kernel log before hibernation and after wake-up with the bug After wake-up today, this bug manifested itself on linux 3.6.3. The first time (suspend to RAM), the system suspended and resumed correctly, the second time (suspend to disk), the system exhibited the bug. If further logs are no longer necessary, I would like to reboot without the drm.debug=0xe option, as it fills up the system log.
On Tue, Nov 27, 2012 at 2:44 PM, <bugzilla-daemon@freedesktop.org> wrote: > --- Comment #9 from auxsvr@gmail.com --- > Created attachment 70661 [details] > If further logs are no longer necessary, I would like to reboot without the > drm.debug=0xe option, as it fills up the system log. You can just disable it with (as root) # echo 0 > /sys/module/drm/parameters/debug
I'm hoping that the fixes Daniel landed in 3.7 clear this up. Hope. As always please do reopen if it reoccurs.
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.