Re: XQuartz - sudden flips out of full-screen mode?
Re: XQuartz - sudden flips out of full-screen mode?
- Subject: Re: XQuartz - sudden flips out of full-screen mode?
- From: Jeremy Huddleston <email@hidden>
- Date: Wed, 16 May 2012 10:40:56 -0700
On May 16, 2012, at 2:14 AM, Lars-Johan Liman <email@hidden> wrote:
>> It's highly likely (like somewhere around 100%) that some periodic
>> event is the cause because human activity *isn't* the cause...
>
> <LOL!> :-) :-)
>
> Yes, that _is_ obvious. :-)
>
> What I meant to say is "Is it possible that _XQuarts_ performs some
> other periodic task, at long intervals, which could have an impact on
> the behaviour we're observing?".
The only sources of timer based activity within XQuartz are Sparkle and X11 clients which themselves use timers. I don't see how the X11 clients could have this type of impact, so that means that if XQuartz is the source of the periodic activity, then Sparkle is probably the only such internal source.
>>> Something that we've configured differently (so it happens to me, but
>>> not to you) or maybe there is a difference in our computing
>>> environments that is relevant? (Snowleopard vs. Lion? Already
>>> updated vs. not yet updated XQuartz?)
>
>> FWIW, I usually test XQuartz without sparkle enabled because that is
>> turned on only for released builds, so that's certainly a likely
>> reason why I don't see the behavior with Sparkle (the update
>> framework).
>
> Ah. Ok. But it seems only a small fraction of us users are seeing this
> (or you/we would see more reports) so that fact doesn't end up at the
> top of my list of likely candidates.
Well, some people might not be noticing it, especially if they use rootless. There may be reasons why the duration of the "flip" may vary, so perhaps I do experience it, but I never notice it because the duration is less than a single cycle of my refresh rate.
>> The Sparkle issue is certainly new, so thanks for mentioning that.
>> Can you file a separate bug report mentioning Sparkle. #58 is
>> probably going to be a bit more challenging, but we can probably
>> figure out how to fix Sparkle (or maybe if we're lucky, it will reveal
>> a way to fix the underlying problem).
>
> Well, I don't consider that a bug. When that happened it was obvious
> that it was Sparkle, since it displayed a dialogue box (in Aqua) asking
> me whether I wanted to update.
Ah, ok. That is actually expected behavior at that point. The same thing will happen for the preference dialog box or the quit dialog box if you trigger one of those while in fullscreen.
> One _wants_ to have these happen
> periodically, but not too often (once per week is OK, once per hour is
> not). If one doesn't do them periodically, but only (e.g.) at startup,
> security updates may be delayed inappropriately. (I often have my X
> server running for weeks, if not months.)
I was under the impression that you had a momentary "flip out" while Sparkle was just checking (which it does without presenting a window).
>> Metacity compiles for me in MacPorts.
>
> Ah. Maybe I should try that again. It's been a good while since I tried
> last time. Thanks!
If it doesn't, please file a bug report. Any errors in Gnome-land at this point I expect to be related to glib2 API strictness changes that went in a few months ago, but I think we've found all of the common ones at this point, so things like metacity should be fine, but you may have issues with more obscure gnome ports that not many people use (again, please file bugs).
_______________________________________________
Do not post admin requests to the list. They will be ignored.
X11-users mailing list (email@hidden)
This email sent to email@hidden