From mboxrd@z Thu Jan  1 00:00:00 1970
From: Tiwei Bie <tiwei.bie@intel.com>
Subject: Re: [PATCH v9 1/8] net/virtio: vring init for packed
	queues
Date: Fri, 26 Oct 2018 14:01:54 +0800
Message-ID: <20181026060154.GB31861@debian>
References: <20181024143236.21271-1-jfreimann@redhat.com>
 <20181024143236.21271-2-jfreimann@redhat.com>
 <20181025092115.GA22179@debian>
 <20181025134737.vzstdn5td7b55nej@jenstp.localdomain>
 <931e66a9-5f22-e241-0cbe-4640698543ec@redhat.com>
 <20181025151248.pro7sewkv4tjkkgk@jenstp.localdomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Cc: Maxime Coquelin <maxime.coquelin@redhat.com>, dev@dpdk.org,
 zhihong.wang@intel.com
To: Jens Freimann <jfreimann@redhat.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mga12.intel.com (mga12.intel.com [192.55.52.136])
 by dpdk.org (Postfix) with ESMTP id C13E55592
 for <dev@dpdk.org>; Fri, 26 Oct 2018 08:03:18 +0200 (CEST)
Content-Disposition: inline
In-Reply-To: <20181025151248.pro7sewkv4tjkkgk@jenstp.localdomain>
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <https://mails.dpdk.org/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://mails.dpdk.org/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <https://mails.dpdk.org/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

On Thu, Oct 25, 2018 at 05:12:48PM +0200, Jens Freimann wrote:
> On Thu, Oct 25, 2018 at 03:51:19PM +0200, Maxime Coquelin wrote:
> > 
> > 
> > On 10/25/18 3:48 PM, Jens Freimann wrote:
> > > On Thu, Oct 25, 2018 at 05:21:15PM +0800, Tiwei Bie wrote:
> > > > On Wed, Oct 24, 2018 at 04:32:29PM +0200, Jens Freimann wrote:
> > > > > Signed-off-by: Jens Freimann <jfreimann@redhat.com>
> > > > > ??struct vring {
> > > > > ???????? unsigned int num;
> > > > > -?????? struct vring_desc?? *desc;
> > > > > -?????? struct vring_avail *avail;
> > > > > -?????? struct vring_used?? *used;
> > > > > +?????? union {
> > > > > +?????????????? struct vring_desc_packed *desc_packed;
> > > > > +?????????????? struct vring_desc *desc;
> > > > > +?????? };
> > > > > +?????? union {
> > > > > +?????????????? struct vring_avail *avail;
> > > > > +?????????????? struct vring_packed_desc_event *driver_event;
> > > > > +?????? };
> > > > > +?????? union {
> > > > > +?????????????? struct vring_used?? *used;
> > > > > +?????????????? struct vring_packed_desc_event *device_event;
> > > > > +?????? };
> > > > > ??};
> > > > > 
> > > > 
> > > > We should define a new `vring` structure for packed ring.
> > > 
> > > I think it was requested to have it as a union before, but I can do
> > > it.
> > 
> > I guess oyu vcan have a union between struct vring and struct
> > vring_packed?
> 
> Like this?
> 
> struct vring {
>          unsigned int num;
>          union {
>                 struct vring_split *split;
> _                struct vring_packed *packed;
>          };
>  };
> 
> We will have to write vq->vq_ring.split.avail->flags and
> vq->vq_ring.packed.desc[xx] and similar things in a lot of places, no?
> 
> Should we just add both vring_split and vring_packed to struct
> virtqueue instead? Only one of them will have memory allocated per virtqueue
> anyway.

Yeah, we can add both to struct virtqueue.