Re: VMA_MERGING_FIXUP and patch

[Andrew Morton]

Andrea Arcangeli <andrea@xxxxxxx> wrote:
>
> On Mon, Mar 22, 2004 at 11:57:34AM -0800, Andrew Morton wrote:
> > set_page_dirty() takes ->tree_lock and inode_lock.  tree_lock surely is OK
> > and while I cannot think of any deadlocks which could occur with taking
> > inode_lock inside the rmap lock, it doesn't sound very nice.
> > 
> > It would of course be best if we could avoid adding a new ranking
> > relationship between these locks.
> 
> agreed. the inode_lock especially is more a vfs thing than a mm thing,
> so it lives quite far away.

Alas, inode_lock can be taken inside page_table_lock, in zap_pte_range().
That set_page_dirty() in there is the nastiest part of the MM locking.

-
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/