From mboxrd@z Thu Jan  1 00:00:00 1970
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Fri, 05 Apr 2013 20:40:46 +0000
Subject: Re: [PATCH 2/8] ehci-platform: add init() method to platform data
Message-Id: <515F36CE.7050804@cogentembedded.com>
List-Id: <linux-sh.vger.kernel.org>
References: <201304050259.51025.sergei.shtylyov@cogentembedded.com>
In-Reply-To: <201304050259.51025.sergei.shtylyov@cogentembedded.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: linux-sh@vger.kernel.org

Hello.

On 04/05/2013 06:43 PM, Alan Stern wrote:

>
>> Sometimes there is a need to initialize some non-standard registers mapped to
>> the EHCI region before accessing the standard EHCI registers.  Add the init()
>> method to the 'ehci-platform' platform data for this purpose.
> "init" isn't such a good name for this.  It's too vague; there are
> already a lot of initialization steps here.  How about "pre_setup"
> instead?

     Quite vague too. :-)
     But can't think of a better name...

>
>> --- renesas.orig/drivers/usb/host/ehci-platform.c
>> +++ renesas/drivers/usb/host/ehci-platform.c
>> @@ -110,6 +110,13 @@ static int ehci_platform_probe(struct pl
>>   		err = PTR_ERR(hcd->regs);
>>   		goto err_put_hcd;
>>   	}
>> +
>> +	if (pdata->init) {
>> +		err = pdata->init(dev, hcd->regs);
>> +		if (err < 0)
>> +			goto err_put_hcd;
>> +	}
> Also, I think this code should go in the ehci_platform_reset() routine,
> just before the call to ehci_setup().  That way more of the setup will
> already have been carried out.

     You're right, of course.

> And instead of passing hcd->regs, wouldn't it be better to pass hcd?

    I really don't know.

> Other users of this interface might need to initialize something other
> than a non-standard register.

    Hm, maybe... if passing 'struct usb_hcd *' would indeed help here. 
Do you think it's
worth passing 'struct platform_device *' along with it?

> Alan Stern
>

WBR, Sergei