Re: built debian-live iso using Version: 2.0~a16-1, but cannot login when booting from the corresponding burnt DVD
On 2010-07-02 19:32+0200 Daniel Baumann wrote:
On 07/02/2010 02:57 AM, Alan W. Irwin wrote:
What is the recommended way to write xorg.conf for >= squeeze?
there are only a few parameters with which you can influence the content
of xorg.conf, see man live-config (for keyboard and the driver, mainly).
With Intel
graphics I would like the capability to try various xorg.conf files even
though sometimes you get lucky and don't need them. I think others with
non-Intel graphics chipsets will also want xorg.conf write capability for
what excately would you want to change?
Old-fashioned control of Intel graphics via xorg.conf is still useful. For
example, some Intel users are having severe trouble with modelines for
monitors without correct EDID information so
Option "ModeDebug" "ON"
is useful as well as other commands to specify the modelines and preferred
mode exactly.
Also, a non-intel X consideration is I have an old (but wonderful) serial
mouse that cannot be automatically detected by X. So the whole mouse
section in xorg.conf needs to be entered by hand, etc.
anyhow.. if you need to find out a working one, you'll build a normal
image, then test it until you've settled, and include it in your build
via config/chroot_local-inclues. live-config doesn't touch xorg.conf if
it's already existing.
OK. I had already guessed that was the solution so it is good to get that
confirmed. However, the experiment failed because I didn't have the correct
form of xorg.conf file for Debian testing. Something is incompatible between
xorg.conf requirements for Debian stable and testing and the result of
simply copying the stable xorg.conf file was a completely frozen X with no
chance of seeing what was wrong. Ultimately, I went with no xorg.conf at
all, and a temporary usb mouse, and that lack of preconfigured X worked well
enough so I could try Debian testing out.
The verdict after a day of testing (some web browsing with konqueror, some
movie watching with dragonplayer, some tv watching with tvtime) is that
KDE-4.4 is much less buggy than KDE-4.2, but the Intel stack on Debian
testing with Intel g33 chipset has some rough edges (noticeable rendering
errors in a few cases where a small piece of one window gets stuck to the
wrong window and noticeable cpu effort required to scroll). However, I
think those X stack results are good enough for now with good prospects for
improvements soon. Thus, I have decided to make the move to Debian testing
now with fingers crossed that I will be able to make that favorite serial
mouse work again.
Daniel, I thank you personally for your help with live-helper, and I would
also like to thank you and the rest of the live-helper developers for making
this relatively convenient Debian testing evaluation possible for me. I give
you my best wishes for your continued development efforts on what is a
really useful tool.
Alan
__________________________
Alan W. Irwin
Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).
Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the libLASi project (unifont.org/lasi); the Loads of
Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________
Linux-powered Science
__________________________
Reply to: