From mboxrd@z Thu Jan  1 00:00:00 1970
From: Tetsuya Mukawa <mukawa@igel.co.jp>
Subject: Re: [PATCH 3/3] virtio: Add a new layer to abstract pci
	access method
Date: Tue, 19 Jan 2016 10:22:07 +0900
Message-ID: <569D8FBF.5020000@igel.co.jp>
References: <1453108389-21006-1-git-send-email-mukawa@igel.co.jp>
 <1453108389-21006-4-git-send-email-mukawa@igel.co.jp>
 <20160118134645.GC19531@yliu-dev.sh.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
Cc: dev@dpdk.org
To: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mail-pf0-f173.google.com (mail-pf0-f173.google.com
 [209.85.192.173]) by dpdk.org (Postfix) with ESMTP id CAC2F8E6E
 for <dev@dpdk.org>; Tue, 19 Jan 2016 02:22:08 +0100 (CET)
Received: by mail-pf0-f173.google.com with SMTP id 65so165872608pff.2
 for <dev@dpdk.org>; Mon, 18 Jan 2016 17:22:08 -0800 (PST)
In-Reply-To: <20160118134645.GC19531@yliu-dev.sh.intel.com>
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>

On 2016/01/18 22:46, Yuanhan Liu wrote:
> On Mon, Jan 18, 2016 at 06:13:09PM +0900, Tetsuya Mukawa wrote:
>> +struct virtio_pci_access_ops {
>> +	uint8_t (*legacy_read8)(struct virtio_hw *hw, uint8_t *addr);
>> +	uint16_t (*legacy_read16)(struct virtio_hw *hw, uint16_t *addr);
>> +	uint32_t (*legacy_read32)(struct virtio_hw *hw, uint32_t *addr);
>> +	void (*legacy_write8)(struct virtio_hw *hw,
>> +				uint8_t *addr, uint8_t val);
>> +	void (*legacy_write16)(struct virtio_hw *hw,
>> +				uint16_t *addr, uint16_t val);
>> +	void (*legacy_write32)(struct virtio_hw *hw,
>> +				uint32_t *addr, uint32_t val);
>> +
>> +	uint8_t (*modern_read8)(struct virtio_hw *hw, uint8_t *addr);
>> +	uint16_t (*modern_read16)(struct virtio_hw *hw, uint16_t *addr);
>> +	uint32_t (*modern_read32)(struct virtio_hw *hw, uint32_t *addr);
>> +	void (*modern_write8)(struct virtio_hw *hw,
>> +				uint8_t *addr, uint8_t val);
>> +	void (*modern_write16)(struct virtio_hw *hw,
>> +				uint16_t *addr, uint16_t val);
>> +	void (*modern_write32)(struct virtio_hw *hw,
>> +				uint32_t *addr, uint32_t val);
> One thing about abstraction is that you need define one set of operations,
> instead of two similar sets. Thus, you need define following operations
> only:
>
>   - read8
>   - read16
>   - read32
>   - write8
>   - write16
>   - write32
>
> And make a proper assignment after the modern/legacy detection.

Thanks for comments.
Sure, will do.

>
>> +
>> +	int (*map_pci_cfg)(struct virtio_hw *hw);
>> +	void (*unmap_pci_cfg)(struct virtio_hw *hw);
>> +	void *(*get_cfg_addr)(struct virtio_hw *hw,
>> +				struct virtio_pci_cap *cap);
>> +	int (*read_pci_cfg)(struct virtio_hw *hw,
>> +				void *buf, size_t len, off_t offset);
> It'd be good if you can post the patches that use above abstract
> operations, so that people can tell if they are properly defined.
>

Yes, I will submit it again.

Thanks,
Tetsuya