Re: [PATCH] mm: fix hang on anon_vma->root->lock

[Peter Zijlstra <peterz@infradead.org>]

On Fri, 2010-08-27 at 21:47 -0500, Christoph Lameter wrote:
> 
> I'd be much more comfortable if the following would be done
> 
> A. Pin the anon_vma by either
>         I. Take a refcount on the anon vma

My preemptible mmu patches do that..

>         II. Take a lock in the anon vma (something that is not pointed to)
> 
> B. Either
>         I. All values that have been used before the pinning are
>            verified after the pinning (and the lock is reacquired
>            if verification fails).
> 
>         II. Or all functions using page_lock_anon_vma() must securely
>             work in the case that the anon_vma was reused for
>             something else before the vma lock was acquired.

Last time I looked they all work like that, they all use something akin
to vma_address() which validates that the page we're interested in is
indeed part of the vma we obtained from the rmap chain.

Anyway, I'll try and refresh my preemptible mmu patch-set now that the
merge window dust settled and post if again, hopefully we can stick it
in -next.