Date: Thu, 13 Mar 2014 16:22:18 -0400
From: Jeff Layton <jlayton@redhat.com>
To: Trond Myklebust <trond.myklebust@primarydata.com>
Cc: Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Subject: Re: [PATCH] nfs: emit a fsnotify_nameremove call in sillyrename
 codepath
Message-ID: <20140313162218.570e0819@tlielax.poochiereds.net>
In-Reply-To: <35EEB018-FD2D-4C52-AA6B-24B66B11D6B6@primarydata.com>
References: <1394738651-26783-1-git-send-email-jlayton@redhat.com>
	<35EEB018-FD2D-4C52-AA6B-24B66B11D6B6@primarydata.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=windows-1252
Sender: linux-nfs-owner@vger.kernel.org

On Thu, 13 Mar 2014 16:08:01 -0400
Trond Myklebust <trond.myklebust@primarydata.com> wrote:

> 
> On Mar 13, 2014, at 15:24, Jeff Layton <jlayton@redhat.com> wrote:
> 
> > If a file is sillyrenamed, then the generic vfs_unlink code will skip
> > emitting fsnotify events for it.
> > 
> > This patch has the sillyrename code do that instead.
> > 
> > In truth this is a little bit odd since we aren't actually removing the
> > dentry per-se, but renaming it. Still, this is probably the right thing
> > to do since it's what userland apps expect to see when an unlink()
> > occurs or some file is renamed on top of the dentry.
> > 
> > Signed-off-by: Jeff Layton <jlayton@redhat.com>
> > ---
> > fs/nfs/dir.c    | 1 +
> > fs/nfs/unlink.c | 2 ++
> > 2 files changed, 3 insertions(+)
> > 
> > diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
> > index 4a48fe4b84b6..591aec9b1719 100644
> > --- a/fs/nfs/dir.c
> > +++ b/fs/nfs/dir.c
> > @@ -37,6 +37,7 @@
> > #include <linux/sched.h>
> > #include <linux/kmemleak.h>
> > #include <linux/xattr.h>
> > +#include <linux/fsnotify.h>
> > 
> > #include "delegation.h"
> > #include "iostat.h"
> > diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
> > index 11d78944de79..547ed0cd59db 100644
> > --- a/fs/nfs/unlink.c
> > +++ b/fs/nfs/unlink.c
> > @@ -355,6 +355,8 @@ static void nfs_async_rename_done(struct rpc_task *task, void *calldata)
> > 
> > 	if (task->tk_status != 0)
> > 		nfs_cancel_async_unlink(old_dentry);
> > +	else if (old_dentry->d_flags & DCACHE_NFSFS_RENAMED)
> > +		fsnotify_nameremove(old_dentry, S_ISDIR(old_dentry->d_inode->i_mode));
> > }
> 
> Any reason why we shouldn?t just do this in nfs_sillyrename() itself?
>

We certainly could, but then you'd probably never get the event if the
process waiting on the async sillyrename got killed while waiting on
the reply.

-- 
Jeff Layton <jlayton@redhat.com>