On Tue, 22 Mar 2005, David S. Miller wrote: > On Tue, 22 Mar 2005 11:21:25 -0800 > "David S. Miller" <[EMAIL PROTECTED]> wrote: > > > I'm trying to analyze my traces some more. > > I think I see what's going wrong. On the first > address space traversal (free_pgd_range()), we > clear out the pgd, even though there are still > more PMD's to process in that PGD. > > So the future loops never do anything since the > PGD is cleared out already.
Yes, that's the conclusion I was coming to from your excellent traces. I notice that although both i386 and sparc64 use pgtable-nopud.h, the i386 pud_clear does nothing at all and the sparc64 pud_clear resets to 0. If that really is the problem, well, I get as easily mixed between the levels as the next man, and haven't a clue which is the right way to do it - beyond i386 not seeing these nr_pte bugs. But if this is the issue, I don't see how it's new to my freepgt patches - beyond the fact that they add this BUG_ON consistency check. Hugh - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
