Re: [ 04/26] hugetlbfs: add swap entry check in follow_hugetlb_page()

[Naoya Horiguchi]

On Wed, Apr 24, 2013 at 04:23:29PM -0700, Greg Kroah-Hartman wrote:
> On Thu, Apr 25, 2013 at 12:04:10AM +0100, Ben Hutchings wrote:
> > On Tue, Apr 23, 2013 at 02:53:44PM -0700, Greg Kroah-Hartman wrote:
> > > 3.4-stable review patch.  If anyone has any objections, please let me know.
> > > 
> > > ------------------
> > > 
> > > From: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx>
> > > 
> > > commit 9cc3a5bd40067b9a0fbd49199d0780463fc2140f upstream.
> > > 
> > > With applying the previous patch "hugetlbfs: stop setting VM_DONTDUMP in
> > > initializing vma(VM_HUGETLB)" to reenable hugepage coredump, if a memory
> > > error happens on a hugepage and the affected processes try to access the
> > > error hugepage, we hit VM_BUG_ON(atomic_read(&page->_count) <= 0) in
> > > get_page().
> >  
> > Is this required?  You didn't apply the previous patch referred to
> > above (commit a2fce9143057) to 3.4.y or 3.0.y since it claimed to fix
> > a regression in 3.7 (commit 314e51b9851b 'mm: kill vma flag VM_RESERVED
> > and mm->reserved_vm counter').
> > 
> > I'm not saying it *isn't* required, mind.
> 
> Yeah, but I went off of the stable: marking in this patch, which said:
> 	    Cc: <stable@xxxxxxxxxxxxxxx>        [2.6.34+?]
> 
> 
> Naoya, any chance you could clear this up?  Is this needed for kernels
> older than 3.7?

Hi Greg,

Kernels older than 3.7 need this patch.

We need this patch whenever hugepage coredump is enabled.
It was enabled before commit 314e51b9851b (== v3.6-8946-g314e51b),
but the commit disabled it accidentally, so commit a2fce914305
(== v3.9-rc7-56-ga2fce91) reenabled it.

Thanks,
Naoya
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/