From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:45713) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TMdDj-0003CB-BR for qemu-devel@nongnu.org; Fri, 12 Oct 2012 07:16:04 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1TMdDd-0000SE-Cj for qemu-devel@nongnu.org; Fri, 12 Oct 2012 07:15:59 -0400 Received: from mx1.redhat.com ([209.132.183.28]:32120) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TMdDd-0000SA-46 for qemu-devel@nongnu.org; Fri, 12 Oct 2012 07:15:53 -0400 Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id q9CBFqlB030942 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Fri, 12 Oct 2012 07:15:52 -0400 Message-ID: <5077FBE6.2040107@redhat.com> Date: Fri, 12 Oct 2012 13:15:50 +0200 From: Gerd Hoffmann MIME-Version: 1.0 References: <1350033962-16665-1-git-send-email-kraxel@redhat.com> <1350033962-16665-8-git-send-email-kraxel@redhat.com> <5077EF88.2070708@redhat.com> <5077F5EB.1080103@redhat.com> In-Reply-To: <5077F5EB.1080103@redhat.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 7/7] chardev: add hotplug support. List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini Cc: qemu-devel@nongnu.org On 10/12/12 12:50, Paolo Bonzini wrote: > Il 12/10/2012 12:23, Gerd Hoffmann ha scritto: >>> I suggest that you implement the commands in a similar way as netdev_add. >> >> Why? Isn't the whole point of using josn is that you'll get the stuff >> from the josn parser & marshaller in a usable form instead of having it >> to feed into yet another parser? I think the only reason netdev_add >> exists in the current form is that it predates qmp. > > In principle you're right, but I think it's ugly that adding another > chardev argument needs changes in 3 places instead of just one. Hmm, I don't have to use the generated marshaller, right? With direct access to the QDict I could just transform it into a QemuOpts. A new parameter wouldn't need code changes then. And the code would be reusable and probably also be simpler. The qapi schema still needs an update though. HMP is more tricky, but I think we should sort QMP first. cheers, Gerd