Kirill A. Shutemov wrote:
> Mel Gorman wrote:
> > On Fri, Dec 20, 2013 at 06:42:10PM +0100, Andrea Arcangeli wrote:
> > > The reason for not doing this was that it felt slow to have that kind
> > > of mangling in a fast path bad check. But maybe it's no problem to do
> > > it. The above should also avoid the bug.
> > >
> >
> > If this bug is related to mprotect then it should also have been fixed
> > indirectly by commit 1667918b ("mm: numa: clear numa hinting information
> > on mprotect") although for the wrong reasons. Applying cleanly requires
> >
> > mm: clear pmd_numa before invalidating
> > mm: numa: do not clear PMD during PTE update scan
> > mm: numa: do not clear PTE for pte_numa update
> > mm: numa: clear numa hinting information on mprotect
> >
> > Is this bug reproducible in 3.13-rc5?
>
> No. At least with my test case.
Oh. Linus has applied the patch to -rc5.
With patch reverted the bug is reproducible.
--
Kirill A. Shutemov
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>