linux-kernel - Re: [PATCH 07/15] reconnect_one(): use lookup_one_len

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 18 Apr 2016 15:23:51 -0400
From:	bfields@...ldses.org (J. Bruce Fields)
To:	Al Viro <viro@...IV.linux.org.uk>
Cc:	Linus Torvalds <torvalds@...ux-foundation.org>,
	linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org
Subject: Re: [PATCH 07/15] reconnect_one(): use lookup_one_len_unlocked()

On Sat, Apr 16, 2016 at 01:55:19AM +0100, Al Viro wrote:
> From: Al Viro <viro@...iv.linux.org.uk>
> 
> ... and explain the non-obvious logics in case when lookup yields
> a different dentry.

ACK to this independent of the rest of the series, my only minor gripe
is that the point made in this new comment is also made at
out_reconnected: and at the head of reconnect_one().  And would also
apply at the other "goto out_reconnected".  My preference is to leave
things as they are, but if you found it easy to overlook maybe just
something like:

 -		goto out_reconnected;
 +		goto out_reconnected; /* see comment there*/

would be helpful.  And/or improving the final comment.

--b.

> 
> Signed-off-by: Al Viro <viro@...iv.linux.org.uk>
> ---
>  fs/exportfs/expfs.c | 10 +++++++---
>  1 file changed, 7 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c
> index c46f1a1..402c5ca 100644
> --- a/fs/exportfs/expfs.c
> +++ b/fs/exportfs/expfs.c
> @@ -143,14 +143,18 @@ static struct dentry *reconnect_one(struct vfsmount *mnt,
>  	if (err)
>  		goto out_err;
>  	dprintk("%s: found name: %s\n", __func__, nbuf);
> -	inode_lock(parent->d_inode);
> -	tmp = lookup_one_len(nbuf, parent, strlen(nbuf));
> -	inode_unlock(parent->d_inode);
> +	tmp = lookup_one_len_unlocked(nbuf, parent, strlen(nbuf));
>  	if (IS_ERR(tmp)) {
>  		dprintk("%s: lookup failed: %d\n", __func__, PTR_ERR(tmp));
>  		goto out_err;
>  	}
>  	if (tmp != dentry) {
> +		/*
> +		 * Somebody has renamed it since exportfs_get_name();
> +		 * great, since it could've only been renamed if it
> +		 * got looked up and thus connected, and it would
> +		 * remain connected afterwards.  We are done.
> +		 */
>  		dput(tmp);
>  		goto out_reconnected;
>  	}
> -- 
> 2.8.0.rc3
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html