From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Tan, Jianfeng" <jianfeng.tan@intel.com>
Subject: Re: [PATCH 0/3] virtio: Add a new layer to abstract pci
	access method
Date: Mon, 18 Jan 2016 21:13:10 +0800
Message-ID: <569CE4E6.70206@intel.com>
References: <1453108389-21006-1-git-send-email-mukawa@igel.co.jp>
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
To: Tetsuya Mukawa <mukawa@igel.co.jp>, dev@dpdk.org
Return-path: <dev-bounces@dpdk.org>
Received: from mga04.intel.com (mga04.intel.com [192.55.52.120])
 by dpdk.org (Postfix) with ESMTP id 50B228D97
 for <dev@dpdk.org>; Mon, 18 Jan 2016 14:13:14 +0100 (CET)
In-Reply-To: <1453108389-21006-1-git-send-email-mukawa@igel.co.jp>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

Hi Tetsuya,

On 1/18/2016 5:13 PM, Tetsuya Mukawa wrote:
> The patches abstract pci access method of virtio-net PMD.
> The patch should be on Yuanhan's below patch series.
>   - [PATCH v4 0/8] virtio 1.0 enabling for virtio pmd driver
>
>
> Tetsuya Mukawa (3):
>    virtio: Change the parameter order of io_write8/16/32()
>    virtio: move rte_eal_pci_unmap_device() to virtio_pci.c
>    virtio: Add a new layer to abstract pci access method
>
>   drivers/net/virtio/virtio_ethdev.c |   4 +-
>   drivers/net/virtio/virtio_pci.c    | 468 ++++++++++++++++++++++++++-----------
>   drivers/net/virtio/virtio_pci.h    |  33 ++-
>   3 files changed, 369 insertions(+), 136 deletions(-)
>

Now I believe we will become more clear about the difference of our two 
implementations.

I was planning to just implement another struct virtio_pci_ops because 
it's going the long way round for my implementation to translate 
virtio_pci_ops to ioport/pci configuration space rd/wr then back to 
sendmsg/ioctl. And in my implementation, there's no need to 
differentiate legacy/modern device.

As I understand, your implementation does not need another 
implementation of struct virtio_pci_ops, but you need different 
implementation in lower layer as this patch show. You want to support 
both legacy/modern device, right?

By the way, this patch looks good to me.

Thanks,
Jianfeng