From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric Dumazet <eric.dumazet@gmail.com>
Subject: Re: [patch 2/6] fs: no games with DCACHE_UNHASHED
Date: Thu, 15 Oct 2009 09:44:35 +0200
Message-ID: <4AD6D2E3.9050105@gmail.com>
References: <20091015044026.319860788@suse.de> <20091015050048.165912589@suse.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: Al Viro <viro@zeniv.linux.org.uk>, linux-fsdevel@vger.kernel.org,
	Ian Kent <raven@themaw.net>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	linux-kernel@vger.kernel.org, Miklos Szeredi <mszeredi@suse.cz>
To: npiggin@suse.de
Return-path: <linux-fsdevel-owner@vger.kernel.org>
Received: from gw1.cosmosbay.com ([212.99.114.194]:40182 "EHLO
	gw1.cosmosbay.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1757869AbZJOHpW (ORCPT
	<rfc822;linux-fsdevel@vger.kernel.org>);
	Thu, 15 Oct 2009 03:45:22 -0400
In-Reply-To: <20091015050048.165912589@suse.de>
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

npiggin@suse.de a =E9crit :
> (this is in -mm)
>=20
> Filesystems outside the regular namespace do not have to clear DCACHE=
_UNHASHED
> in order to have a working /proc/$pid/fd/XXX. Nothing in proc prevent=
s the
> fd link from being used if its dentry is not in the hash.
>=20
> Also, it does not get put into the dcache hash if DCACHE_UNHASHED is =
clear;
> that depends on the filesystem calling d_add or d_rehash.
>=20
> So delete the misleading comments and needless code.
>=20

This was added in commit 304e61e6fbadec586dfe002b535f169a04248e49

    [PATCH] net: don't insert socket dentries into dentry_hashtable
   =20
    We currently insert socket dentries into the global dentry hashtabl=
e.  This
    is suboptimal because there is currently no way these entries can b=
e used
    for a lookup().  (/proc/xxx/fd/xxx uses a different mechanism).  In=
serting
    them in dentry hashtable slows dcache lookups.
   =20
    To let __dpath() still work correctly (ie not adding a " (deleted)"=
) after
    dentry name, we do :
   =20
    - Right after d_alloc(), pretend they are hashed by clearing the
      DCACHE_UNHASHED bit.
   =20
    - Call d_instantiate() instead of d_add() : dentry is not inserted =
in
      hash table.
   =20
      __dpath() & friends work as intended during dentry lifetime.
   =20
    - At dismantle time, once dput() must clear the dentry, setting aga=
in
      DCACHE_UNHASHED bit inside the custom d_delete() function provide=
d by
      socket code, so that dput() can just kill_it.
   =20
    Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
    Cc: Al Viro <viro@zeniv.linux.org.uk>
    Acked-by: "David S. Miller" <davem@davemloft.net>
    Signed-off-by: Andrew Morton <akpm@osdl.org>
    Signed-off-by: Linus Torvalds <torvalds@osdl.org>



Back in 2006, we had to perform this hack in order to not leak '(delete=
d)' in __d_path()

if (!IS_ROOT(dentry) && d_unhashed(dentry) &&
	(prepend(&end, &buflen, " (deleted)", 10) !=3D 0))=20
		goto Elong;

In current kernel this part became :

if (d_unlinked(dentry) &&
	(prepend(&end, &buflen, " (deleted)", 10) !=3D 0))
		goto Elong;


So your cleanup seems good, thanks !

Acked-by: Eric Dumazet <dada1@cosmosbay.com>
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel=
" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html