After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 335149 - When playing Xmame fullscreen the screensaver starts
When playing Xmame fullscreen the screensaver starts
Status: RESOLVED DUPLICATE of bug 444927
Product: gnome-screensaver
Classification: Deprecated
Component: daemon
2.14.x
Other All
: Normal normal
: ---
Assigned To: gnome-screensaver maintainers
gnome-screensaver maintainers
: 568264 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2006-03-19 18:28 UTC by Carlos Antonio
Modified: 2009-08-21 04:22 UTC
See Also:
GNOME target: ---
GNOME version: 2.13/2.14



Description Carlos Antonio 2006-03-19 18:28:40 UTC
Please describe the problem:
When playing Xmame fullscreen the screensaver starts despite all the key presses.
I'm running Xmame in the OpenGL video mode.

This didn't happen with Xscreensaver.

Shouldn't the session become idle only if there aren't any key presses or mouse
movements at all?


Steps to reproduce:
1. Start Xmame (or maybe other OpenGL games) in fullscreen
2. Play the game for 15 minutes (or whatever your idle_delay is)
3.


Actual results:
The screensaver starts above the game.
Pressing some keys or moving the mouse does nothing.

Expected results:


Does this happen every time?
Yes.

Other information:
Comment 1 William Jon McCann 2006-03-20 19:12:17 UTC
Are you using the gxmame front end?
Comment 2 Carlos Antonio 2006-03-20 20:10:51 UTC
Yes. I forgot to mention that.
I'm using GXMame version 0.35beta2.

I don't know if this is relevant, but beagle also starts indexing files after a while. So maybe this isn't exactly a gnome-screen-saver problem, but something that tells gnome-screen-saver and beagle that the computer is idle?
Comment 3 William Jon McCann 2006-03-20 20:15:40 UTC
We'll need to modify gxmame to disable the screensaver.  I've had this problem myself.

I made the following patch a while back but it needs to be updated with gnome-screensaver support:
http://sourceforge.net/tracker/index.php?func=detail&aid=1145600&group_id=50621&atid=460395
Comment 4 Carlos Antonio 2006-03-20 20:25:33 UTC
Ok!

But will that solve the problem with beagle indexing? 
Or maybe beagle gets the idle info from gnome-screensaver... I don't know... but I had the impression that the indexing started before the screen going blank.
Comment 5 Steven Garrity 2006-03-26 22:41:13 UTC
Same problem with mplayer - should that get a separate bug?
Comment 6 William Jon McCann 2006-03-26 22:45:34 UTC
Steven, please file a bug with mplayer.  Thanks.
Comment 7 Sam Morris 2006-03-30 22:21:47 UTC
Does every single app really need to be patched to blank the screen with gnome-screensaver's special method? Do apps now have to cope with each of gnome-screensaver, xscreensaver, KDE's screensaver, xlockmore and probably more ... ?
Comment 8 William Jon McCann 2006-03-30 23:38:16 UTC
Sam, only applications that need to inhibit the screensaver - yes.  This isn't really that many.  This will hopefully become a freedesktop standard.
Comment 9 Sam Morris 2006-03-31 00:16:00 UTC
(In reply to comment #8)
> Sam, only applications that need to inhibit the screensaver - yes.  This isn't
> really that many.

In my experience it includes any game that goes fullscreen--including commercial ones that will never have dbus added to them, like UT2004.
Comment 10 Paul Nolan 2006-06-09 15:34:33 UTC
Shouldn't gnome-screensaver use the same 'don't blank the screen if I'm playing ppracer' stuff as xscreensaver or KDE screensaver (neither of which interfere with fullscreen games)?
Comment 11 Javier Aravena 2006-07-08 21:08:58 UTC
same problem with stepmania. I don't think it disables xscreensaver, I don't know, but I think xscreensaver is just able to know when something is in fullscreen, or maybe it catchs the keys you press...
Comment 12 Vinicius Depizzol 2006-07-22 02:46:50 UTC
The same problem happens with Planet Penguin Racer.
Comment 13 Sam Morris 2006-11-18 15:07:14 UTC
Possible duplicate filed downstream, but against a very old g-s version. Can anyone confirm or deny if the mentioned applications (armagedtron, vmware) are still overlooked by modern versions of g-s?

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=352003
Comment 14 Josselin Mouette 2007-12-25 21:35:08 UTC
This could probably be fixed directly in SDL, but there should also be a mechanism in g-s to detect fullscreen applications that grab all input devices, like games.
Comment 15 Alan Swanson 2008-10-15 10:41:11 UTC
This looks to be a problem with Gnome ScreenSaver not detecting the correct use of the XResetScreenSaver standard such as now used by SDL and MPlayer, where the optional DBus API is not really appropriate. Any argument against supporting XResetScreenSaver would just be utter nonsense.

Will have a look and see if I can come up with a patch.

Comment 16 Alan Swanson 2008-10-19 15:08:26 UTC
Ok, I now fully realise how craptastic screensaver setup is with X. However we do have a possible solution by using the IDLETIME sync counter, added with Xorg 1.4, as any call to XResetScreenSaver updates the last event time - see dixSaveScreens() in xorg-xserver/dix/window.c. Will think about it.

Also from looking over the code, using_mit_saver_extension is never true. Looks to have been a straight copy and paste from xscreensaver. I'd assume we would want to register and kill any currently running screensaver but still set time to 0 so we don't use the MIT extension. However that is another bug.
Comment 17 Alan Swanson 2008-12-19 23:07:53 UTC
IDLETIME support has been implemented in bug 444927 for gnome-screensaver 2.25.2 (thankfully as I was puzzling over integrating it) and I can report it is working perfectly with SDL CVS (or 1.2.14 when it is released). No screensaver activation whilst running SDL games.

However you do need to have xorg-server patched with the following otherwise it pegs the CPU.

http://cgit.freedesktop.org/xorg/xserver/commit/?id=1f4fb0225b278d1cf4145aebeb0bdd23dc8f62d5

Following was the patch to SDL.

http://www.libsdl.org/cgi/viewvc.cgi?view=rev&revision=3558
Comment 18 William Jon McCann 2009-08-21 04:20:12 UTC
*** Bug 568264 has been marked as a duplicate of this bug. ***
Comment 19 William Jon McCann 2009-08-21 04:22:45 UTC

*** This bug has been marked as a duplicate of bug 444927 ***