pm-suspend quirk + fglrx

[Lubomir Kundrak]

On Wed, 2008-03-12 at 14:54 +0100, Danny Kukawka wrote:
> On Mittwoch, 12. März 2008, Lubomir Kundrak wrote:
> > On Wed, 2008-03-12 at 14:01 +0100, Danny Kukawka wrote:
> > > On Dienstag, 11. März 2008, Stefan Söffing wrote:
> > > > Hi all,
> > > >
> > > > it had a long fight against fglrx to suspend. Now it seems to work: The
> > > > solution for me was, _not_ to use vbe_post and vbestate_restore.
> > >
> > > Please note: HAL don't support proprietary (closed) drivers like fglrx or
> > > nvidia.
> >
> > Though I completely agree with the principle, I don't think it is the
> > case now.
> >
> > The quirks break its correct functioning, so here the case is "not
> > breaking", not "supporting" (though it can for sure be reasonably argued
> > for the opposite :).
> 
> But remove the quirks would mean to break it for ppl. which e.g. use one of 
> the radeon driver. 

Ah, that's what I feared.

> > On the other hand quirks often fix situations when 
> > proprietary and closed firmwares do the wrong thing, so probably it
> > should be also done here (given it is probably very trivial fix).
> 
> The point is: I don't see atm how we should differ between the different 
> possible drivers for ATI. IMO the only thing that may make sense, is that ATI 
> ship a fdi-file with their package.

Right, that makes perfect sense. (/me goes to add one to Livna :)

-- 
Lubomir Kundrak (Red Hat Security Response Team)