Skeelo wrote:
> It looks like I sent this to the wrong list. After trying the latest
> snapshot kernels (#81) the NCR kernels booted fine and only the SBC
> kernels died (guess, what mine use). So I compiled a new kernel with the
> NCR driver instead of the SBC driver and now everything works fine. I just
> can't use my SyQuest drive anymore.
> 
> I've been seeing my recently built kernels hang when I try to boot them.
> Here is a trace.

[ trace snipped ]

> It doesn't look md and I've tried the GENERIC kernels from the latest
> snapshot with the same results. Does this look familiar to anyone?

Yes.  This is the same error message that a number of people have gotten
with various ncr drivers on various ports after the MI code was changed to
use the new pool(9) interface, I believe.  The i386 port has supposedly
been fixed.  From what I've gleaned from current-users, the problem may be
that some kind of bug common to the various ports' ncr drivers has been
tickled by this change to memory allocation.

Anyway, please file a PR on this if Ken Nakata hasn't already (he hit this
in early August, a couple of days after the change went in).

Later.

-- 
Colin Wood                                 cwood@ichips.intel.com
Component Design Engineer - PMD                 Intel Corporation
-----------------------------------------------------------------
I speak only on my own behalf, not for my employer.