hmm, that makes sense. No technical reasons to stay with 2.4.8 but there are a
fair number of other reasons. We've done a fair amount of customization for an
embedded product that's supposed to go gold in a couple months. I think there
are good reasons to go to newer kernels but we just don't have the schedule and
it would be really bad if something else broke during the change.
Ian
Steve Lord wrote:
> I am pretty sure this will not be timing related, XFS should not be
> sensitive to timing. I suspect it is more likely to be related to a
> failing memory allocation.
>
> Any reason you have to stay with 2.4.8?
>
> Steve
>
> On Fri, 2001-11-30 at 11:30, Ian S. Nelson wrote:
> > I'll dump out the oops. I haven't been running with kdb since I don't have
> > a console on the machine.
> >
> > thanks,
> > Ian
> >
> > Eric Sandeen wrote:
> >
> > > Any more details on the crash, either an oops output or a kdb
> > > backtrace? Or does adding kdb perturb things enough that it works
> > > again?
> > >
> > > -Eric
> > >
> > > On Fri, 2001-11-30 at 11:11, Ian S. Nelson wrote:
> > > > I'm debugging a bizarre bug right now. I'm building a tiny embedded
> > > > box and I'm using XFS. Right now I've got a kernel (2.4.8 based) that
> > > > crashes when I add printks to it and doesn't crash when I remove them.
> > > > The crash is always in XFS code, xfs_trans_get_buf is where it usually
> > > > happens.
> > > --
> > > Eric Sandeen XFS for Linux http://oss.sgi.com/projects/xfs
> > > sandeen@xxxxxxx SGI, Inc.
> --
>
> Steve Lord voice: +1-651-683-3511
> Principal Engineer, Filesystem Software email: lord@xxxxxxx
|