From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1754145AbaIVQIX (ORCPT <rfc822;w@1wt.eu>);
	Mon, 22 Sep 2014 12:08:23 -0400
Received: from smtp.citrix.com ([66.165.176.89]:63714 "EHLO SMTP.CITRIX.COM"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1753001AbaIVQIW (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Mon, 22 Sep 2014 12:08:22 -0400
X-IronPort-AV: E=Sophos;i="5.04,572,1406592000"; 
   d="scan'208";a="173911511"
Message-ID: <542047EA.6090202@citrix.com>
Date: Mon, 22 Sep 2014 17:01:46 +0100
From: David Vrabel <david.vrabel@citrix.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Icedove/24.5.0
MIME-Version: 1.0
To: Chen Gang <gang.chen.5i5j@gmail.com>
CC: <konrad.wilk@oracle.com>, <boris.ostrovsky@oracle.com>,
        <stefano.stabellini@eu.citrix.com>, <mukesh.rathor@oracle.com>,
        <xen-devel@lists.xenproject.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [Xen-devel] [PATCH 3/3 v2] xen/xenbus: Improve failure processing
 code for __xenbus_switch_state()
References: <541AE4E9.2040904@gmail.com> <541AE5C4.2070409@gmail.com> <54203A73.8030103@citrix.com> <5420456C.4010103@gmail.com>
In-Reply-To: <5420456C.4010103@gmail.com>
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 7bit
X-DLP: MIA1
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 22/09/14 16:51, Chen Gang wrote:
> On 09/22/2014 11:04 PM, David Vrabel wrote:
>> On 18/09/14 15:01, Chen Gang wrote:
>>> When failure occurs, need return failure code instead of 0, or some of
>>> indirect upper callers may misunderstand.
>>>
>>> e.g. in "block/xen-blkback/xenbus.c":
>>>
>>>     connect() -> xenbus_switch_state() -> __xenbus_switch_state().
>>
>> Can you make xenbus_switch_state() void?  The callers don't need to do
>> any error handling.
>>
> 
> After "grep rn xenbus_switch_state *" under "drivers/", it is not one
> place to check the return value of xenbus_switch_state(), and also it
> is export to outside for individual modules.
> 
> So we need change many subsystems for it, and also need face the rick
> for incompatible with the old individual modules which source code are
> not in upstream.

Having to update 9 callers doesn't seem like much work.

> And are you sure the caller need not notice about it, when it really
> fails? (for me, I guess they need notice about it)

Yes.  xenbus_switch_state() already signals the fatal error to the
toolstack with xenbus_dev_fatal().

David