From mboxrd@z Thu Jan  1 00:00:00 1970
From: wwang <wei_wang@realsil.com.cn>
Subject: Re: [PATCH] mfd:rtsx: Support RTS5249
Date: Mon, 25 Mar 2013 14:23:45 +0800
Message-ID: <514FED71.8020307@realsil.com.cn>
References: <1364177636-4294-1-git-send-email-wei_wang@realsil.com.cn> <20130325060038.GD9138@mwanda>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8;
	format=flowed
Content-Transfer-Encoding: QUOTED-PRINTABLE
Return-path: <linux-mmc-owner@vger.kernel.org>
Received: from rtits2.realtek.com ([60.250.210.242]:34625 "EHLO
	rtits2.realtek.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1757822Ab3CYGYG (ORCPT
	<rfc822;linux-mmc@vger.kernel.org>); Mon, 25 Mar 2013 02:24:06 -0400
In-Reply-To: <20130325060038.GD9138@mwanda>
Sender: linux-mmc-owner@vger.kernel.org
List-Id: linux-mmc@vger.kernel.org
To: Dan Carpenter <dan.carpenter@oracle.com>
Cc: cjb@laptop.org, sameo@linux.intel.com, gregkh@linuxfoundation.org, linux-mmc@vger.kernel.org, linux-kernel@vger.kernel.org, rogerable@realtek.com, devel@linuxdriverproject.org

=E4=BA=8E 2013=E5=B9=B403=E6=9C=8825=E6=97=A5 14:00, Dan Carpenter =E5=86=
=99=E9=81=93:
> On Mon, Mar 25, 2013 at 10:13:56AM +0800, wei_wang@realsil.com.cn wro=
te:
>> +static int rts5249_optimize_phy(struct rtsx_pcr *pcr)
>> +{
>> +	int err;
>> +
>> +	err =3D rtsx_pci_write_phy_register(pcr, 0x19, 0xFE46);
>> +	if (err < 0)
>> +		return err;
>> +
>> +	mdelay(1);
> Why do we have the mdelay() and the later msleep(5)?
> rtsx_pci_write_phy_register() busy loops until the write succeeds or
> it returns -ETIMEOUT.  The extra wait here seems unnecessary.
>
> regards,
> dan carpenter
>
>
> .
>  =20

Hi,

The busy loops in rtsx_pci_write_phy_register only tell us that the=20
write sequence succeeds. The device still needs to wait for a while=20
until the internal signal stable. Or else the timing won't fit the=20
requirement.
All of the delays in the driver are necessary.

BR,
Wei