Re: Gnome and CPU load

Hi,

At 20:16 -0400 on 2010-8-16 Alexander Hansen wrote:
 >
 > On 8/16/10 7:34 PM, Stefan Bruda wrote:
 > > 
 > > I have an old version of the GNOME package installed via Fink, and I
 > > am not trying (again!) to upgrade it to the latest version.  Problem
 > > is, as soon as I upgrade I get to have X11 eat up almost all the CPU
 > > cycles.  I had to revert every time to the older version to have a
 > > decently working installation.  My .xinitrc looks like this:
 > > 
 > >     source ~/.profile
 > >     e16 &
 > >     exec gnome-session
 > > 
 > > It does not matter which window manager I use, once the GNOME session
 > > gets established the process X11.bin loads one core to 100% and stays
 > > there.  The GNOME desktop works otherwise well.
 > > 
 > > I am using Fink 0.29.13 on a Macbook Pro with Mac OS 10.5.8 and Xcode
 > > 3.1.3.  I tried upgrading my current installation and also starting
 > > over from scratch with a brand new Fink installation, all with the
 > > same result.
 > > 
 > > I would like to upgrade to 10.6 (so I will have to start with the
 > > current set of Fink packages) and so I appreciate any suggestions you
 > > might have on the matter.  I don't even know where to start to fix
 > > this problem.
 > 
 > What X11 are you using?

Thank you for pointing me to the right direction.  I was running a
rather old version of Xquartz (X.org Release 7.3, X.Org X Server
1.4.2-apple42, Build Date: 20090517) and your question pointed me to
an upgrade.  Now with the new Xquartz 2.5.3 (X.org Release 7.5, X.Org
X Server 1.8.2, Build Date: 20100813) things are back to normal with
the latest and greatest GNOME.

 > I can't reproduce your CPU usage issue on 10.5.8/i386 using Xquartz
 > 2.5.3 and current/unstable GNOME stuff.

Must have been Xquartz then.

 > .xinitrc is deprecated as of 10.5, so you might try setting things
 > up as per
 > 
 > http://finkers.wordpress.com/2010/06/18/finks-xinitrc-gnome-and-kde/
 > 
 > though I don't know if that will help.

Will go there, but as you guessed this does not matter...

Many thanks again,
Stefan

-- 
If it was so, it might be; and if it were so, it would be; but as it 
isn't, it ain't.  That's logic.  --Lewis Carroll, Through the Looking-Glass

No HTML emails and proprietary attachments please <http://bruda.ca/email.html>

------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev