On Sun, 23 Dec 2012 18:13:37 +0000
Ben Hutchings <b...@decadent.org.uk> wrote:

> On Sun, 2012-12-23 at 13:06 -0500, chris dunn wrote:
> > On Sun, 23 Dec 2012 16:33:32 +0000
> > Ben Hutchings <b...@decadent.org.uk> wrote:
> > 
> > > Control: tag -1 moreinfo
> > > 
> > > Sorry for the delayed reply.
> > > 
> > > On Fri, 2012-06-01 at 09:47 -0500, chris dunn wrote:
> > > [...]
> > > > I've recently upgraded from Squeeze to Wheezy on two Motion
> > > > Computing M1400 computers. The upgrade seems generally OK and I
> > > > am still able to load Linux 2.6.32-5-486.
> > > > 
> > > > However when I attempt to load 3.2.0-2-486 in Grub it fails to
> > > > load.
> > > > 
> > > > In graphical mode it halts with a flashing underscore top left
> > > > of an otherwise blank screen after reporting "loading initial
> > > > ramdisk".
> > > > 
> > > > In recovery mode it halts with a flashing underscore after
> > > > showing the messages evident in the attached screen photo.
> > > > 
> > > > Both machines exhibit the same behaviour.
> > > 
> > > The screenshot shows something odd about the PCI bus
> > > configuration in this computer.  This is probably a BIOS bug but
> > > the kernel should ideally work around such bugs.
> > > 
> > > Does the current kernel in unstable (linux-image-3.2.0-4-486,
> > > version 3.2.35-2) still have this problem?
> > 
> > Thanks for the input. I'm showing 3.2.0-4-486, version 3.2.32-1 as
> > being installed, and not as you suggest.
> 
> That's the current version in testing, the later version is in
> unstable.

I'm still using 32-1. Trust this is ok.
 
> > It does exhibit the same behaviour as reported before.
> >  
> > > Does the kernel parameter 'pci=assign-busses' fix the problem?
> > 
> > No. This does not seem to have any effect. I'm entering it as the
> > first parameter line after pressing e at the grub boot selection
> > screen.
> 
> You must add it on the same line that begins
> 'linux /boot/vmlinuz-3.2.0-4-486', normally at the end of the line.

Added the parameter at the end of the line as requested. It seems to
have no effect on either graphical mode or recovery mode boot. That is
to say the original problem persists.
 
> Another parameter that might make a difference is 'pci=realloc'.

Yields the same results as pci=assign-busses.

 
> Ben.
> 

--
 
Chris Dunn
 


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Reply via email to