darcs patch: Allow user to override Xinerama detection

(please ignore previous patch - I forgot to update the README)
The attached patch allows the user to override Xinerama detection. I
have the Xinerama headers installed because Debian's GTK+ development
packages require it, but I only have one screen; for some reason,
probably related to the laptop's VGA out, xmonad thinks there's a second
one and I lose a workspace. The attached patch allows the user to pass
the --without-xinerama to configure to manually disable xinerama.
You'll need to run 'autoreconf'.
Fri Aug 17 20:06:19 CDT 2007 Alec Berryman

On Aug 17, 2007, at 21:08 , Alec Berryman wrote:
(please ignore previous patch - I forgot to update the README)
The attached patch allows the user to override Xinerama detection. I
I suggest that it is preferable to specify whether to try to use Xinerama or not at runtime, rather than compile time. -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] allbery@kf8nh.com system administrator [openafs,heimdal,too many hats] allbery@ece.cmu.edu electrical and computer engineering, carnegie mellon university KF8NH

On Fri, Aug 17, 2007 at 08:08:23PM -0500, Alec Berryman wrote:
(please ignore previous patch - I forgot to update the README)
The attached patch allows the user to override Xinerama detection. I have the Xinerama headers installed because Debian's GTK+ development packages require it, but I only have one screen; for some reason, probably related to the laptop's VGA out, xmonad thinks there's a second one and I lose a workspace. The attached patch allows the user to pass the --without-xinerama to configure to manually disable xinerama.
I believe you should investigate your X installation, since your patch should not be needed. XMonad uses getScreenInfo to see whether it's running in a Xinerama environment or not. getScreenInfo is a (very safe) wrapper around XineramaQueryScreens. In the man page (man XineramaQueryScreens) I can read that "XineramaQueryScreens() returns NULL and sets number to 0 if Xinerama is not active." In other word there are strong reasons to suppose that the problem is not on the XMonad side, but on yours. And, obviously, I don't think XMonad should be written with malfunctioning X servers in mind...;-) Just my 2 cents (I don't use Xinerama). Andrea

Andrea Rossato on 2007-08-18 08:13:13 +0200:
On Fri, Aug 17, 2007 at 08:08:23PM -0500, Alec Berryman wrote:
(please ignore previous patch - I forgot to update the README)
The attached patch allows the user to override Xinerama detection. I have the Xinerama headers installed because Debian's GTK+ development packages require it, but I only have one screen; for some reason, probably related to the laptop's VGA out, xmonad thinks there's a second one and I lose a workspace. The attached patch allows the user to pass the --without-xinerama to configure to manually disable xinerama.
I believe you should investigate your X installation, since your patch should not be needed. XMonad uses getScreenInfo to see whether it's running in a Xinerama environment or not.
I first noticed this a month or two ago and gave my X installation a cursory glance; I didn't enable Xinerama, but did upgrade xorg, xorg drivers, and xmonad at all about the same time, which didn't help me track anything down (not that I had the time). My solution then was to just uninstall Xinerama headers, but that's no longer an option for me; the patch I sent is the most expedient remedy for my problem.

alec:
(please ignore previous patch - I forgot to update the README)
The attached patch allows the user to override Xinerama detection. I have the Xinerama headers installed because Debian's GTK+ development packages require it, but I only have one screen; for some reason, probably related to the laptop's VGA out, xmonad thinks there's a second one and I lose a workspace. The attached patch allows the user to pass the --without-xinerama to configure to manually disable xinerama.
You'll need to run 'autoreconf'.
Begin able to override misdetection seems reasonably harmless. Checking the files are in the right place, so autoconf can find them, might also be warranted here too, Alec. -- Don

On Friday 17 August 2007 20:08:23 Alec Berryman wrote:
(please ignore previous patch - I forgot to update the README)
The attached patch allows the user to override Xinerama detection. I have the Xinerama headers installed because Debian's GTK+ development packages require it, but I only have one screen; for some reason, probably related to the laptop's VGA out, xmonad thinks there's a second one and I lose a workspace. The attached patch allows the user to pass the --without-xinerama to configure to manually disable xinerama.
You'll need to run 'autoreconf'.
Fri Aug 17 20:06:19 CDT 2007 Alec Berryman
* Allow user to override Xinerama detection
Applied.

On Friday 17 August 2007 20:08:23 Alec Berryman wrote:
(please ignore previous patch - I forgot to update the README)
The attached patch allows the user to override Xinerama detection. I have the Xinerama headers installed because Debian's GTK+ development packages require it, but I only have one screen; for some reason, probably related to the laptop's VGA out, xmonad thinks there's a second one and I lose a workspace. The attached patch allows the user to pass the --without-xinerama to configure to manually disable xinerama.
You'll need to run 'autoreconf'.
Fri Aug 17 20:06:19 CDT 2007 Alec Berryman
* Allow user to override Xinerama detection
While disabling linking to libXinerama is all well and good, perhaps xmonad should have a flag to disable Xinerama at runtime? Cheers, Spencer Janssen
participants (5)
-
Alec Berryman
-
Andrea Rossato
-
Brandon S. Allbery KF8NH
-
dons@cse.unsw.edu.au
-
Spencer Janssen