> This sounds similar to what I was seeing with bug #800, only I don't
> think I had taskfile IO enabled. Another person I talked to had some
> success by enabling PIIX I think, but that did not work for me. The
> only workaround that worked for me was to disable highmem support.
I'm pretty sure we're looking at the same bug, at least they are
remarkably similar for a co-incidence. If you have the time, it might be
worth checking if commenting out inc_preempt_count() and
dec_preempt_count() from kmap_atomic() and kunmap_atomic() masks the bug
for you. It's not a workaround but it might help narrow down where things
are going wrong
-- Mel Gorman MSc Student, University of Limerick http://www.csn.ul.ie/~mel - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/