On Wed, 12 May 2010 22:25:15 -0400
Rik van Riel <riel@xxxxxxxxxx> wrote:
> On 05/12/2010 08:38 PM, KAMEZAWA Hiroyuki wrote:
> > On Wed, 12 May 2010 13:39:58 -0400
> > Rik van Riel<riel@xxxxxxxxxx> wrote:
> >
> >> Subject: track the root (oldest) anon_vma
> >>
> >> Track the root (oldest) anon_vma in each anon_vma tree. Because we only
> >> take the lock on the root anon_vma, we cannot use the lock on higher-up
> >> anon_vmas to lock anything. This makes it impossible to do an indirect
> >> lookup of the root anon_vma, since the data structures could go away from
> >> under us.
> >>
> >> However, a direct pointer is safe because the root anon_vma is always the
> >> last one that gets freed on munmap or exit, by virtue of the same_vma list
> >> order and unlink_anon_vmas walking the list forward.
> >>
> >> Signed-off-by: Rik van Riel<riel@xxxxxxxxxx>
> >
> >
> > Acked-by: KAMEZAWA Hiroyuki<kamezawa.hiroyu@xxxxxxxxxxxxxx>
> >
> > I welcome this. Thank you!
> >
> > Reading 4/5, I felt I'm grad if you add a Documentation or very-precise-comment
> > about the new anon_vma rules and the _meaning_ of anon_vma_root_lock.
> > I cannot fully convice myself that I understand them all.
>
> Please send me a list of all the questions that come up
> when you read the patches, and I'll prepare a patch 6/5
> with just documentation :)
>
0. Why it's dangerous to take vma->anon_vma->lock ?
1. What kinds of anon_vmas we'll found in
page->mapping => anon_vma->head and avc->same_anon_vma ?
IOW, what kinds of avc->vmas will see when we walk anon_vma->head.
2. Why we have to walk from the root ?
3. What anon_vma_lock guards, actually ?
etc....the facts which is unclear for guys who are not involved in this fix.
Preparing some explanation seems to be kindly rather than "plz ask google"
Bye.
-Kame
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>