Intel 2.0.0 driver not seeing laptop screen resolutions properly

[Keith Packard]

On Fri, 2007-05-18 at 15:23 -0700, Greg KH wrote:
> On Tue, May 08, 2007 at 10:02:02PM -0700, Keith Packard wrote:
> > On Tue, 2007-05-08 at 21:21 -0700, Eric Anholt wrote:
> > 
> > > So, we lose out on starting X after a resume from S3 has happened, on
> > > that hardware (which, as I understand, is not uncommon)?  That seems
> > > like a recipe for even more pain than the current situation.
> > 
> > I don't know which is more common -- hardware that unmaps the BIOS on S3
> > resume or hardware which has bogus parameters in the video registers.
> > And, at this point, we're at 'failing to ever work' being worse than
> > 'failing to work after S3'.
> > 
> > In any case, we should certainly add config options that can control
> > which method (video reg vs BIOS) we use for the initial mode selection.
> 
> Well, as the current driver version, from git, still is broken for my
> laptop, I do agree that this should be some kind of option :)


> Any help I can provide with testing?

Have you tried disabling the code that loads the mode from the video
registers and verifying that it works for you?

Check i830_lvds.c around line 486; the code which calls
i830_crtc_mode_get should just be disabled, and it should then use the
BIOS table mode instead.

If this works, I'll code up an option that lets us switch between these
two mechanisms.

One other thing we can do is stick code that makes sure the current mode
is reasonably sane; in your case, we can see a huge blanking period
which makes no sense for normal panel usage. That should give us a clue
as to which mode to believe.

-- 
keith.packard at intel.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.x.org/archives/xorg/attachments/20070518/9b518673/attachment.pgp>