Improving the suspend quirks guessworking

[Martin Pitt]

Hi,

Danny Kukawka [2008-03-26 14:30 +0100]:
> > The very same is true for the 'upstream' quirks, since a set of quirks
> > which works for one person might break another person's box who uses a
> > different graphics driver or kernel version. (I. e. if someone using
> > nv reports quirks, you'll break it for someone using the nvidia
> > driver; which is exactly what my patch fixes.)
> 
> I don't see how the patches would fix the case you describe above. If the 
> machine is listed from a user with nv (without any driver matches) your 
> quirk-patch wouldn't fix anything for the user that use nvida. 

It would, since the patch sets all these quirks to false on
nvidia/fglrx/newer Intels.

> The existing patch would still be used for both users and would
> break one of them.

That's why the patch detects the driver dynamically and treats the
quirks db accordingly. Just in case it got lost in the flood, I'm
attaching it here again.

Just to repeat: I'm not trying to convince you to apply it upstream
(although I don't see anything actively wrong with doing so).  As I
said, it's a working bandaid for distros which want to release soon
with some good OOTB suspend/resume capabilities.

Thanks,

Martin
-- 
Martin Pitt                        | http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 01_default_suspend_quirks.patch
Type: text/x-diff
Size: 2528 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/hal/attachments/20080326/db4dd274/attachment.patch