Received-SPF: pass (google.com: best guess record for domain of linux-nfs-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
IronPort-SDR: E98odBEbbxoB6CpZxbkA79UMyRomc6Ad5d454v/JN7aVoSVPu8apoYq+3uMYRrDAJ9FquK0cts
 WW41ZCnMeLJQ==
Date:   Fri, 13 Mar 2020 17:55:29 +0000
From:   Frank van der Linden <fllinden@amazon.com>
To:     Trond Myklebust <trondmy@hammerspace.com>
CC:     "tigran.mkrtchyan@desy.de" <tigran.mkrtchyan@desy.de>,
        "linux-nfs@vger.kernel.org" <linux-nfs@vger.kernel.org>,
        "anna.schumaker@netapp.com" <anna.schumaker@netapp.com>
Subject: Re: [PATCH 03/13] NFSv4.2: query the server for extended attribute
 support
Message-ID: <20200313175529.GB31307@dev-dsk-fllinden-2c-c1893d73.us-west-2.amazon.com>
References: <20200311195613.26108-1-fllinden@amazon.com>
 <20200311195613.26108-4-fllinden@amazon.com>
 <530167624.4533477.1584029710746.JavaMail.zimbra@desy.de>
 <20200312205139.GA32293@dev-dsk-fllinden-2c-c1893d73.us-west-2.amazon.com>
 <20200312211555.GA5974@dev-dsk-fllinden-2c-c1893d73.us-west-2.amazon.com>
 <948465413.4651196.1584097887947.JavaMail.zimbra@desy.de>
 <6792d6a6012a241b8bd1555eea8c592ff318a444.camel@hammerspace.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <6792d6a6012a241b8bd1555eea8c592ff318a444.camel@hammerspace.com>
User-Agent: Mutt/1.11.4 (2019-03-13)
Sender: linux-nfs-owner@vger.kernel.org
Precedence: bulk

On Fri, Mar 13, 2020 at 01:50:38PM +0000, Trond Myklebust wrote:
> 'xattr_support' seems like a protocol hack to allow the client to
> determine whether or not the xattr operations are supported.
> 
> The reason why it is a hack is that 'supported_attrs' is also a per-
> filesystem attribute, and there is no value in advertising
> 'xattr_support' there unless your filesystem also supports xattrs.
> 
> IOW: the protocol forces you to do 2 round trips to the server in order
> to figure out something that really should be obvious with 1 round
> trip.

Right, that's the annoying part. I mean, theoretically, a server can
legally reject a GETATTR because you're asking for an unknown
attribute (e.g. xattr_support in this case). So then what I have
in my current patch (asking for both the supported_attrs and
xattr_support at once) might fail, and the mount would fail.

Which means I should split it up and have nfs4_do_fsinfo do
the 2nd part, just in case.

- Frank