From: Hugh Dickins
Date: Fri, 1 Aug 2008 20:59:08 +0100 (BST)

> On Fri, 1 Aug 2008, Jeremy Fitzhardinge wrote:
> >
> > I have a function traversing a pagetable in vaddr order (low to high), taking
> > pte locks as it builds up batches of pte page updates. When the batch is
> > issued, it releases all the locks, and won't end up holding more than ~16 at a
> > time.
> >
> > So, I think this is OK. There are no internal lock ordering issues, and I
> > don't think there'll be any bad interactions from someone trying to take pte
> > locks for two separate pagetables. I don't think there's anyone else trying
> > to take more than one pte lock at once, but if there were "lock low vaddr then
> > high" seems like a reasonable locking rule (or more precisely "lowest" to deal
> > with the case of a pte page being aliased at multiple vaddrs).
>
> Please check the spin_lock_nested() in move_ptes() in mm/mremap.c.

It won't work because spin_lock_nested() is limited to a depth
of 8 and he aparently needs 16.

Taking more than a few locks of the same class at once is bad
news and it's better to find an alternative method.

--
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/