Open GoogleCodeExporter opened 8 years ago
Have you had a look at
http://www.haskell.org/haskellwiki/Xmonad/Frequently_asked_questions#XMonad_is_f
rozen.21 yet?
Original comment by daniel.w...@gmail.com
on 21 Sep 2012 at 6:13
Except for the 'xclock'-method, I've tried (and failed) the other methods,
excluding xmobar/statusbar as I don't use those.
I'll try 'xclock'-method, and post any updates as soon as I get them.
Original comment by bmathiasen
on 22 Sep 2012 at 1:50
xclock didn't help.
Original comment by bmathiasen
on 22 Sep 2012 at 2:12
It won't help you. The problem is obvious. When the GridSelect windows are
displayed, the XServer is grabbed, presumably for the keyboard input although
I've not checked the exact reasons why the XServer is grabbed in this case.
Original comment by thomas.a...@gmail.com
on 1 Oct 2012 at 11:18
If that is the case, surely every user of Gridselect should experience this?
Original comment by bmathiasen
on 11 Oct 2012 at 6:44
You won't see it 100% consistently; even the infamous putSelect wasn't 100%
consistent.
I just went through the GridSelect code. There is no full server grab; the
keyboard and one mouse button are grabbed, though. Beyond this, the issue is
that xmonad is not processing events in its main loop, and this means
(a) it cannot do things like switching workspaces;
(b) events that GridSelect does not expect are simply dropped, meaning:
(c) new windows will completely fail to do the window management process;
(d) other potentially important X11 events and negotiations are dropped on the
floor.
Which means that there are a number of ways that GridSelect can lead to
gridlock with either clients or the X11 server. If you pick your selection
quickly, you likely won't notice; the longer GridSelect stays up, and the more
things that happen while it's up, the more likely that xmonad will drop/ignore
something critical.
Original comment by allber...@gmail.com
on 11 Oct 2012 at 2:13
Thanks for the elaborate response.
Just to bring a status update: XMonad doesn't seem to freeze on Arch (hope i
don't jinx it ;). Both systems have xmonad/+contrib installed through cabal
and run identical xmonad.hs.
Here's some output from 'X -version' on my two currently running systems, in
case it makes any difference:
Mint (debian) 'X -version':
X.Org X Server 1.12.1.902 (1.12.2 RC 2)
Release Date: 2012-05-19
X Protocol Version 11, Revision 0
Build Operating System: Linux 3.2.0-2-amd64 x86_64 Debian
Current Operating System: Linux none 3.2.0-3-amd64 #1 SMP Thu Jun 28 09:07:26
UTC 2012 x86_64
Kernel command line: BOOT_IMAGE=/boot/vmlinuz-3.2.0-3-amd64
root=UUID=6cdaef22-3035-4c99-af66-553e95dcd62e ro quiet
Build Date: 20 May 2012 08:57:17AM
xorg-server 2:1.12.1.902-1 (Cyril Brulebois <kibi@debian.org>)
Current version of pixman: 0.26.0
Arch 'X -version':
X.Org X Server 1.12.3.901 (1.12.4 RC 1)
Release Date: 1012-08-03
X Protocol Version 11, Revision 0
Build Operating System: Linux 3.4.4-2-ARCH x86_64
Current Operating System: Linux bridgelinux 3.4.9-1-ARCH #1 SMP PREEMPT Wed Aug
15 18:59:31 CEST 2012 x86_64
Kernel command line: BOOT_IMAGE=/boot/vmlinuz-linux root=/dev/sda1 ro quiet
Build Date: 09 August 2012 10:04:44AM
Current version of pixman: 0.26.2
Original comment by bmathiasen
on 12 Nov 2012 at 10:17
Nevermind, that was a false positive. Freezing still occurs. I'll attempt
stracing the process.
Original comment by bmathiasen
on 17 Nov 2012 at 10:19
Original issue reported on code.google.com by
bmathiasen
on 20 Sep 2012 at 8:12Attachments: