[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kwin
Subject:    [Bug 235821] using kwin shadows with oxygen creates artifact
From:       Hugo Pereira Da Costa <hugo.pereira () free ! fr>
Date:       2010-04-30 21:52:39
Message-ID: 20100430215239.616273EBFD () immanuel ! kde ! org
[Download RAW message or body]

https://bugs.kde.org/show_bug.cgi?id=235821





--- Comment #9 from Hugo Pereira Da Costa <hugo pereira free fr>  2010-04-30 23:52:37 ---
> I'm sorry, but it sounds like what you're saying is that since this only
> affects our default decoration and other decorations that behave the way we
> think modern decorations should, it isn't really a problem.
>
> If the issue is that the shadow effect makes an incorrect assumption that an
> opaque window has an opaque decoration, then fix the shadow effect.
> The shadow effect can check if the decorations are opaque by calling
> effects->decorationsHaveAlpha().
>
> Don't fix the problem by tweaking other parts of KWin until that false
> assumption becomes true. Especially not if those tweaks result in performance
> regressions.
>
>    
mmm. My understanding is quite the opposite. A bug was identified for a 
class of decorations in the shadow effect,
I tracked it down to something which I believe is incorrect upstream 
(the decision of which "mask" is passed to the effects painting methods, 
in general, not only to the shadow effect), so I tried to fix it 
upstream, rather than for this effect only. (my point being: if the 
decoration has alpha, passing PAINT_WINDOW_OPAQUE to any effect painting 
is potentially incorrect).

Has a perfomance regressions actually been observed ?

In any case, if you think the above is not correct (and my understanding 
is wrong), feel free to revert.
I would then see how to fix inside the shadow pluggin (which is where I 
looked at the first place).

-- 
Configure bugmail: https://bugs.kde.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
_______________________________________________
kwin mailing list
kwin@kde.org
https://mail.kde.org/mailman/listinfo/kwin
[prev in list] [next in list] [prev in thread] [next in thread]