Re: [PATCH 00/21] mm: Preemptibility -v6

[Oleg Nesterov]

On 01/24, Peter Zijlstra wrote:
>
> On Mon, 2011-01-24 at 15:34 +0100, Oleg Nesterov wrote:
> > On 01/24, Peter Zijlstra wrote:
> > >
> > > On Fri, 2011-01-21 at 16:33 +0100, Peter Zijlstra wrote:
> > >
> > > > Index: linux-2.6/mm/rmap.c
> > > > ===================================================================
> > > > --- linux-2.6.orig/mm/rmap.c
> > > > +++ linux-2.6/mm/rmap.c
> > > > @@ -1559,9 +1559,20 @@ void __put_anon_vma(struct anon_vma *ano
> > > >  	 * Synchronize against page_lock_anon_vma() such that
> > > >  	 * we can safely hold the lock without the anon_vma getting
> > > >  	 * freed.
> > > > +	 *
> > > > +	 * Relies on the full mb implied by the atomic_dec_and_test() from
> > > > +	 * put_anon_vma() against the full mb implied by mutex_trylock() from
> > > > +	 * page_lock_anon_vma(). This orders:
> > > > +	 *
> > > > +	 * page_lock_anon_vma()		VS	put_anon_vma()
> > > > +	 *   mutex_trylock()			  atomic_dec_and_test()
> > > > +	 *   smp_mb()				  smp_mb()
> > > > +	 *   atomic_read()			  mutex_is_locked()
> > >
> > > Bah!, I thought all mutex_trylock() implementations used an atomic op
> > > with return value (which implies a mb), but it looks like (at least*)
> > > PPC doesn't and only provides a LOCK barrier.
> >
> > But, mutex_trylock() must imply the one-way barrier, otherwise it
> > is buggy, no?
>
> > If this atomic_read() can leak out of the critical section, then
> > I think mutex_trylock() should be fixed. Or I misunderstood the
> > problem completely...
>
> It implies the LOCK barrier, the one way permeable thing, not a full mb.
>
> But I'm not sure the LOCK is sufficient to make the above scenario work.

OK, I can't say I am sure too.



Well. Thinking more, I can't understand why we can trust mutex_trylock()
at all. Suppose that page_lock_anon_vma() races with put_anon_vma()
and anon_vma->root != anon_vma. In this case page_lock_anon_vma() can
take anon_vma->root->lock, but if this vma was already freed then it can
be reused and anon_vma_alloc() can change ->root and set ->refcount == 1
before we check atomic_read(refcount) != 0.

IOW,
	page_lock_anon_vma:
							put_anon_vma() drops the
							last reference

	if (mutex_trylock(&anon_vma->root->lock)) {
							anon_vma_alloc() picks this
							memory, changes ->root and
							sets ->refcount == 1

		if (atomic_read(&anon_vma->refcount))
			return anon_vma;
	}

Most probably I missed something, I forgot everything I knew about this
code....

Oleg.

--
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/