From mboxrd@z Thu Jan  1 00:00:00 1970
From: Roger Pau Monne <roger.pau@citrix.com>
Subject: Re: [PATCH v9 16/17] libxl: convert
 libxl_device_vkb_add to an async operation
Date: Fri, 20 Jul 2012 12:00:01 +0100
Message-ID: <50093A31.3010404@citrix.com>
References: <1342172696-88347-1-git-send-email-roger.pau@citrix.com>
	<1342172696-88347-17-git-send-email-roger.pau@citrix.com>
	<20488.14185.962905.269760@mariner.uk.xensource.com>
	<5008389F.4060401@citrix.com>
	<20489.14294.391006.844801@mariner.uk.xensource.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <xen-devel-bounces@lists.xen.org>
In-Reply-To: <20489.14294.391006.844801@mariner.uk.xensource.com>
List-Unsubscribe: <http://lists.xen.org/cgi-bin/mailman/options/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xen.org>
List-Help: <mailto:xen-devel-request@lists.xen.org?subject=help>
List-Subscribe: <http://lists.xen.org/cgi-bin/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=subscribe>
Sender: xen-devel-bounces@lists.xen.org
Errors-To: xen-devel-bounces@lists.xen.org
To: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: "xen-devel@lists.xen.org" <xen-devel@lists.xen.org>
List-Id: xen-devel@lists.xenproject.org

Ian Jackson wrote:
> Roger Pau Monne writes ("Re: [PATCH v9 16/17] libxl: convert libxl_device_vkb_add to an async operation"):
>> Ian Jackson wrote:
> ...
>>> This is rather unfriendly - doesn't it amount to aborting on an
>>> invalid domain configuration.  And I'm not sure I understand the
>>> logic, which is new in your patch.
>> This was part of the previous code, but this was done inside of a switch
>> instead of an if. Since it's not possible to specify vkb devices in the
>> config file (at least xl.cfg(5) doesn't mention anything about vkbs),
>> I'm quite sure d_config->vkbs == NULL and d_config->num_vkbs == 0 always
>> or something has gone extremely wrong, just added the abort to be sure.
> 
> People who write other code which calls libxl might try to pass vkbds,
> but then the code would crash.  And if they didn't test it with HVM
> domains, they wouldn't notice that their code had a crash bug.
> 
> I think it should cause the creation to fail with ERROR_INVAL.

I've been thinking about this and I think that maybe we should only
perform this piece of code if num_vkbs is 0, so we make sure there's at
least one vkb device, but not fail if num_vkbs > 0.