cvs commit: src/sys/dev/ata ata-queue.c

[Conrad J. Sabatier]

On Thu, 12 May 2005 00:42:17 +0200, Søren Schmidt <sos at DeepCore.dk>
wrote:

> Conrad J. Sabatier wrote:
> > On Wed, 11 May 2005 14:36:26 +0000 (UTC), Søren Schmidt
> > <sos at freebsd.org> wrote:
> >>  
> >>  This should fix the boot problems of late.
> >>  
> >>  Revision  Changes    Path
> >>  1.49      +4 -2      src/sys/dev/ata/ata-queue.c
> > 
> > 
> > Almost, but still not quite there.  It looks a little better, but
> > I'm still getting stuck in an infinite loop at boot time.
> > 
> > For some reason, the master device on ata1 (a DVD writer) is being
> > ignored, and acd0 is being assigned as the slave device (CD-ROM). 
> > Then I get an endless series of failures on acd0 (device not ready,
> > timeout, etc.).
> > 
> > If you need the verbose output, I'll copy it by hand for you later. 
> > Don't have time at the moment.  Just let me know.
> 
> Hmm, get me the exact config fx the dmesg from it boot verbosely on a 
> working kernel, I need to be abel to reproduce to fix this...

Small problem: my last good kernel got overwritten.  Strange, but one
"make installkernel" did *not* move /boot/kernel to /boot/kernel.old (I
had previously manually moved /boot/kernel.old to /boot/kernel after
discovering that the new kernel wasn't working properly, then the next
installkernel failed to move it back to kernel.old, for some reason).

About the best I can do for you now will be a verbose boot with the
latest kernel, and try to accurately copy the output by hand.  I'll try
to do that ASAP tomorrow, as it's getting rather late here now.

Sorry!  :-)

-- 
Conrad J. Sabatier <conrads at cox.net> -- "In Unix veritas"