[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#569134: xorg: Display will "hang" with a solid colour screen, only fixed by a suspend/resume cycle.



On Wed, Feb 10, 2010 at 23:57:35 +0100, Richard Atterer wrote:

> Hello,
> 
> I can confirm this on an EeePC 1008HA, running a vanilla 2.6.33-rc6 kernel 
> and xserver-xorg-video-intel 2:2.9.1-2. It seems to be caused by KMS:
> 
> http://wiki.debian.org/KernelModesetting
> https://bugs.freedesktop.org/show_bug.cgi?id=25681
> 
> >From the moment the X server starts, the console becomes inaccessible, with 
> corrupt output being shown when I switch there. For example, I can still 
> see the cursor blinking, but that blinking happens for a couple of pixels 
> spread all over parts of the screen. I use GRUB_GFXMODE=800x600 in 
> /etc/default/grub for a VESA fb console - 
> http://en.gentoo-wiki.com/wiki/Intel_GMA#Kernel_Modesetting says that this 
> is a bad idea...
> 
> A possible temporary fix might be to disable KMS, by passing i915.modeset=0 
> to the kernel.
> 
> For the record, the rest is exactly the same as in the original bug report: 
> Occasional (every 30 sec?) flickering/tearing of the screen, as if the 
> start of the video RAM were wrong for a frame or less. Eventually (after 
> maybe 20 minutes use on average), the screen will go all black or grey. 
> Suspend to RAM and resume is the only way to fix it, switching to the 
> console and back won't do. Before the last upgrade, everything worked just 
> fine.
> 
Is this on 945GM?  If so, it can be worked around by passing
i915.powersave=0 to the kernel, and was fixed for me by
http://lists.freedesktop.org/pipermail/intel-gfx/2010-February/005763.html

Note that the debian kernel defaults to i915.powersave=0 so shouldn't be
affected by this bug anymore.

Cheers,
Julien



Reply to: