Re: [PATCH] vfs: fix race in rcu lookup of pruned dentry

From: Hugh Dickins
Date: Sun Jul 17 2011 - 19:48:18 EST

On Sun, 17 Jul 2011, Linus Torvalds wrote:
> On Sun, Jul 17, 2011 at 4:16 PM, Al Viro <viro@xxxxxxxxxxxxxxxxxx> wrote:
> >
> > OR
> >
> >  - keep part of the patch from Hugh, treating negative in RCU mode as
> > "need to unlazy".
> No, urgh, that's horrible.
> Not being able to do an RCU lookup of negative dentries would be
> really sad. There are some loads where a negative dentry is the
> *common* case.

Yes, that worried me too. But I can see Al's point, it looks like
he has enough on his plate for now.

This is not a fresh regression: I vote we make no change for 3.0,
I give your minus-1-liner more testing, hopefully with confirmation,
and we get the fix into 3.0.1 (if that's what stable is to be called).
Or into 3.0 if there's going to be an -rc8, I was assuming not.