[Pm-utils] Migrating suspend quirks away from hal

Victor Lowther victor.lowther at gmail.com
Fri Nov 13 10:35:36 PST 2009


On Nov 13, 2009, at 10:21 AM, Matthew Garrett <mjg59 at srcf.ucam.org>  
wrote:

> On Fri, Nov 13, 2009 at 05:15:36PM +0100, Michael Biebl wrote:
>
>> What is the current status of KMS for radeon and noveau? Is radeon  
>> KMS
>> still marked as experimental?
>
> It's marked as experimental, but only because the ABI isn't fixed yet.
> It should work. nouveau KMS works fine, including suspend/resume  
> support
> on most machines.
>
>> What do we do about users of radeon w/o KMS, nv or intel w/o i915?
>
> nv is broadly irrelevant (quirks aren't going to help you there),
> running intel without i915 isn't going to work with the next driver
> release (which will be before any distribution ships without hal) and
> radeon without kms is effectively deprecated.

You are writing off a lot of legacy hardware that is probably still in  
use, as well as everything that is not intel, radeon, or nvidia. The  
quirks database should stay in some form.

>> Imo at this point it is a bit premature to rely on KMS only. Of  
>> course
>> we can just keep hal around for that purpose a little longer and safe
>> us the effort of a transition. My guess is, that Martin want's to get
>> rid of hal completely for a default GNOME desktop install in Lucid?
>
> Given that the X transition away from hal hasn't happened yet, there's
> no realistic chance of anyone shipping a halless distribution in the
> next release cycle. And by then KMS should be solid for everything.

Even sis, tseng, matrox, and all the smaller players?

> -- 
> Matthew Garrett | mjg59 at srcf.ucam.org
> _______________________________________________
> Pm-utils mailing list
> Pm-utils at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/pm-utils


More information about the devkit-devel mailing list