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

Bug#585600: linux-image-2.6.32-5-686: KMS does not detect display resolutions



On Sat, Jun 12, 2010 at 17:20:09 +0200, Olivier Berger wrote:

> reopen 585600
> thanks
> 
> Julien Cristau <jcristau@debian.org> writes:
> 
> > So in summary your kvm switch doesn't pass edid along, so the kernel has
> > no way to know what resolution the monitor supports, so it falls back to
> > a presumably safe 1024x768 resolution.  
> 
> That would be a possible explanation, yes... however I'm not completely
> sure... any way to try and confirm this diagnostic for sure ?
> 
Get rid of the KVM switch, or try a useful one.

> > I don't think there's a bug here, so I suggest you try setting the
> > wanted resolution on the kernel command line (video=1280x1024 should
> > work).  Closing this bug, thanks for your report.
> >
> 
> I'd be perfectly happy to have one bug less open on the kernel, but I
> think there's a problem here that is not fixed at all.  Of course, I
> could manage to workaround by passing args to the kernel, but that looks
> like a regression from end user perspective : how comes that without KMS
> things work OK, and with, it doesn't ? Maybe it could be reassigned to
> another faulty package ?
> 
They don't work ok without kms, if you need to override the panel size
and ignore edid.

> Maybe the problem is not with the kernel and with the X driver, then, as
> I have :
>   Option "IgnoreEDID" "on"
>   Option "PanelSize" "1280x1024"
> in /etc/X11/xorg.conf
> 
> If I get it right, with KMS, if 1280x1024 is not detected as a valid
> mode by the kernel, but then, shouldn't the PanelSize setting be
> observed anyway by the radeon driver ?
> 
I don't know.  What does the X log look like with kms?

Cheers,
Julien

Attachment: signature.asc
Description: Digital signature


Reply to: