From mboxrd@z Thu Jan  1 00:00:00 1970
From: Al Viro <viro@ZenIV.linux.org.uk>
Subject: Re: [RFC] st_nlink after rmdir() and rename()
Date: Thu, 3 Mar 2011 04:42:42 +0000
Message-ID: <20110303044242.GJ22723@ZenIV.linux.org.uk>
References: <20110303032454.GI22723@ZenIV.linux.org.uk>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
To: Linus Torvalds <torvalds@linux-foundation.org>
Return-path: <linux-kernel-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <20110303032454.GI22723@ZenIV.linux.org.uk>
Sender: linux-kernel-owner@vger.kernel.org
List-Id: linux-fsdevel.vger.kernel.org

On Thu, Mar 03, 2011 at 03:24:54AM +0000, Al Viro wrote:
> 0 0: ext*, xfs, jfs, reiserfs, ocfs2, gfs2, nilfs, exofs, udf, ubifs,
>      minix, sysv, ufs, msdos, vfat, hfs+
> 0 1: ramfs, shmem, hugetlbfs, jffs2, omfs, hfs[*], apparently nfs as well

> [*] yes, different from hfs+; the code is clearly broken, since it simply
> does unlink() on target, without even verifying that it's empty.  And
> yes, it's trivial fs corruption...

Actually, hfs turns out to be 0 0 as well; it *is* broken (lacks check
for target being empty), but when the target is empty it's doing the
right thing.  jffs2 and ramfs-based ones do 0 1 - confirmed by direct
experiment.