| To: | Christoph Hellwig <hch@xxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: [PATCH] always set a/c/mtime through ->setattr |
| From: | Al Viro <viro@xxxxxxxxxxxxxxxxxx> |
| Date: | Sat, 31 May 2008 14:35:27 +0100 |
| Cc: | Miklos Szeredi <miklos@xxxxxxxxxx>, hch@xxxxxx, linux-fsdevel@xxxxxxxxxxxxxxx, xfs@xxxxxxxxxxx, Artem.Bityutskiy@xxxxxxxxx |
| In-reply-to: | <20080531132444.GA22936@xxxxxxxxxxxxx> |
| References: | <20080520060838.GA6436@xxxxxx> <E1JyMSu-0001au-96@xxxxxxxxxxxxxxxxxxx> <20080520083351.GA14826@xxxxxx> <E1JzFFa-0008VR-9B@xxxxxxxxxxxxxxxxxxx> <20080531132048.GA4201@xxxxxxxxxxxxxxxxxx> <20080531132444.GA22936@xxxxxxxxxxxxx> |
| Sender: | xfs-bounce@xxxxxxxxxxx |
| User-agent: | Mutt/1.5.17 (2007-11-01) |
On Sat, May 31, 2008 at 09:24:45AM -0400, Christoph Hellwig wrote:
> On Sat, May 31, 2008 at 02:20:48PM +0100, Al Viro wrote:
> > And there's one more problem, promising very ugly code review: locking
> > rules for notify_change() had suddenly changed - you are calling it
> > without i_mutex now. And ext3_setattr() is not happy - especially due
> > to this blind call of ext3_orphan_del() in there. We can easily fix
> > that one, but you'll need to audit the rest of instances...
>
> Yeah, I've actually started an audit of the setattr instance and there's
> even more crap turning up. I'm working on a bigger series to sort these
> things out.
Grabbing i_mutex may get very interesting on pagefault paths when
dirtying a page - locking rules are interesting there already...
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH] always set a/c/mtime through ->setattr, Christoph Hellwig |
|---|---|
| Previous by Thread: | Re: [PATCH] always set a/c/mtime through ->setattr, Christoph Hellwig |
| Next by Thread: | [PATCH 2/2] always set a/c/mtime through ->setattr, Christoph Hellwig |
| Indexes: | [Date] [Thread] [Top] [All Lists] |