On Mon, 3 May 2010, Rik van Riel wrote:
quoted text
> 
> One problem is that we cannot find the VMAs (multiple) from
> the page, except by walking the anon_vma_chain.same_anon_vma
> list.  At the very least, that list requires locking, done
> by the anon_vma.lock.

But that's exactly what we do in rmap_walk() anyway.

But yes, I can well imagine that in other cases we only do the one 
anon_vma. I didn't check who used the lock.

So if we do want to keep the lock in the anon_vma, I would just suggest 
that instead of making "normal" users do lots of locking, make the 
rmap_walk side one.

quoted text
> A forkbomb could definately end up getting slowed down by
> this patch.  Is there any real workload out there that just
> forks deeper and deeper from the parent process, without
> calling exec() after a generation or two?

Heh. AIM7. Wasn't that why we merged the multiple anon_vma's in the first 
place?

quoted text
> > So again, my gut feel is that if the lock just were in the vma itself,
> > then the "normal" users would have just one natural lock, while the
> > special case users (rmap_walk_anon) would have to lock each vma it
> > traverses. That would seem to be the more natural way to lock things.
> 
> However ... there's still the issue of page_lock_anon_vma
> in try_to_unmap_anon.

Do we care?

We've not locked them all there, and we've historically not cares about 
the rmap list being "perfect", have we? 

So I _think_ it's just the migration case (and apparently potentially the 
hugepage case) that wants _exact_ information. Which is why I suggest the 
onus of the extra locking should be on _them_, not on the regular code.

I dunno. Again, my objections to the patches are really based more on a 
gut feel of "that can't be the right thing to do" than anything else.

We have _extremely_ few places that walk lists to lock things. And they 
are never "normal" code. Things like that magic "mm_take_all_locks()", for 
example. That is why I then react with "that can't be right" to patches 
like this.

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