[PATCH] Trigger an inhibition in ksmserver upon logout

Dario Freddi drf54321 at gmail.com
Sat Feb 14 15:37:30 GMT 2009


On sabato 14 febbraio 2009 16:11:56 Oswald Buddenhagen wrote:
> i'll just think aloud here:
> powerdevil relying on being disabled in time implies that Other Session
> Managers have to do it too, which is bad. otoh, ksmserer not doing an
> inhibit means that Other Power Managers might screw it. so for
> robustness, both would have to be implemented. not sure how much
> real-world impact that has.

True indeed. Though there is a freedesktop.org specification for inhibition 
that Solid follows, see org.freedesktop.PowerManagement.Inhibit, so Solid 
actually prompts that inhibition to any freedestkop-compliant PowerManager 
hooked up in the session, being it PowerDevil or not. So, this has exactly the 
opposite effect: we reach more Power managers without changing a single line 
of code, and that's exactly what we want.

Obviously, that Power Manager should implement freedesktop specifications and 
support inhibition, but that's their problem, and not ours.

> from an architecture pov, powerdevil hooking into session management
> still seems the cleaner solution to me: ksmserver is the "session
> master". it's kind of silly to expect it to give a special notification
> to one kind of "servant" when it already does notify everyone, fairly
> and equally. at least you wouldn't get far with that in real life.

Considering what I said above, do you still think it should be better to hook 
PowerDevil into KSM and catch the save status signal?
-- 
-------------------

Dario Freddi
KDE Developer
GPG Key Signature: 511A9A3B

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20090214/0c05d57d/attachment.sig>


More information about the kde-core-devel mailing list