From mboxrd@z Thu Jan  1 00:00:00 1970
From: Michal Suchanek <hramrach@centrum.cz>
Subject: Re: [PATCH] tmpfs: implement generic xattr support
Date: Thu, 12 May 2011 09:52:27 +0200
Message-ID: <BANLkTikAePYtmQdwapzBx9QAJLVxisy2Wg@mail.gmail.com>
References: <BANLkTini-g9XVOohKjwpZT-SGozcwx720A@mail.gmail.com>
 <4DA4B6A8.7030804@gmail.com> <BANLkTimsAE1ZAJhsSjnh3LqwsN9x0cLaXg@mail.gmail.com>
 <alpine.LRH.2.00.1104131618170.6953@twin.jikos.cz> <BANLkTindBULU9KAVWPJQ7oQvfCNQU-BuAg@mail.gmail.com>
 <BANLkTin6CVFOAECWwYg3B5V2AfscFfjt8w@mail.gmail.com> <BANLkTik7jasnvqFQohRdU-3crM8fr9pBXg@mail.gmail.com>
 <BANLkTikF_hgYSaZJbZosWwcs715tLXYfmQ@mail.gmail.com> <BANLkTikTUo17RncV7bushPTri=LEykep5g@mail.gmail.com>
 <BANLkTikUe_2Uo2_tSYdfNoNXKYO4Bj30iQ@mail.gmail.com> <C450CD68-10AE-4527-B6BF-A5D5F0BF005F@dilger.ca>
 <BANLkTinMNL8Q+_MCHDqAGSdYNAvHB2GdLw@mail.gmail.com> <87tydu3t4p.fsf_-_@tucsk.pomaz.szeredi.hu>
 <alpine.LSU.2.00.1105112023560.1424@sister.anvils>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: Miklos Szeredi <miklos@szeredi.hu>,
	Andreas Dilger <adilger@dilger.ca>,
	Jiri Kosina <jkosina@suse.cz>,
	Ric Wheeler <ricwheeler@gmail.com>,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	David Howells <dhowells@redhat.com>,
	Ian Kent <ikent@redhat.com>, Jeff Moyer <jmoyer@redhat.com>,
	Christoph Hellwig <hch@infradead.org>,
	Eric Paris <eparis@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	James Morris <jmorris@namei.org>,
	Serge Hallyn <serge.hallyn@ubuntu.com>
To: Hugh Dickins <hughd@google.com>
Return-path: <linux-kernel-owner@vger.kernel.org>
In-Reply-To: <alpine.LSU.2.00.1105112023560.1424@sister.anvils>
Sender: linux-kernel-owner@vger.kernel.org
List-Id: linux-fsdevel.vger.kernel.org

On 12 May 2011 06:20, Hugh Dickins <hughd@google.com> wrote:
> On Tue, 19 Apr 2011, Miklos Szeredi wrote:
=2E..
>
> I disagree with Andrew on this, it should default to off, consistent =
with
> the "If unsure, say N" comment. =C2=A0Partly because that's how Linus=
 prefers
> it, for good anti-bloat reasons: if people have got along without a
> feature for years, whyever should we suddenly default it on? =C2=A0An=
d
> partly because TMPFS_POSIX_ACL already defaults to off (as do
> EXT2_FS_XATTR and EXT2_FS_POSIX_ACL).
>
> However... if someone already has CONFIG_TMPFS_POSIX_ACL=3Dy in their
> old config, it would be nice to make CONFIG_TMPFS_XATTR=3Dy automatic=
ally,
> so we're not in danger of removing their old functionality. =C2=A0But=
 I haven't
> thought of the right way to achieve that - maybe your helpful POSIX A=
CL
> comment is enough, but automatic would be better.
>
> Few people would choose TMPFS_XATTR on and TMPFS_POSIX_ACL off: perha=
ps it
> would work to make CONFIG_TMPFS_XATTR the new config option to cover =
both,
> and select it from config TMPFS_POSIX_ACL (without a prompt) so the
> oldconfig propagates correctly. =C2=A0Perhaps. =C2=A0But I haven't tr=
ied myself,
> and you may be forgiveably disinclined to make config experiments!

The xattrs are required for overlayfs so "generic" kernels will
probably want those. However, the "generic" kernels likely have the
acls already so this is not that much of an issue.

Perhaps the acls should select xattrs so that in configs that had acls
the xattrs are added automagically.

Or does it already work with depends?

Thanks

Michal

>
>> =C2=A0config TMPFS_POSIX_ACL
>> =C2=A0 =C2=A0 =C2=A0 bool "Tmpfs POSIX Access Control Lists"
>> - =C2=A0 =C2=A0 depends on TMPFS
>> + =C2=A0 =C2=A0 depends on TMPFS_XATTR
>> =C2=A0 =C2=A0 =C2=A0 select GENERIC_ACL
>> =C2=A0 =C2=A0 =C2=A0 help