From mboxrd@z Thu Jan  1 00:00:00 1970
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Fri, 17 May 2013 19:17:49 +0000
Subject: Re: [PATCH 8/9] ARM: shmobile: APE6EVM: switch to using platform_add_devices()
Message-Id: <5196825D.2040304@cogentembedded.com>
List-Id: <linux-sh.vger.kernel.org>
References: <1368802520-16378-9-git-send-email-g.liakhovetski@gmx.de>
In-Reply-To: <1368802520-16378-9-git-send-email-g.liakhovetski@gmx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: linux-sh@vger.kernel.org

Hello.

On 05/17/2013 11:05 PM, Guennadi Liakhovetski wrote:

>>> Switch from platform_device_register_resndata() to platform_add_devices()
>>> to make adding new devices easier.
>>>
>>> Signed-off-by: Guennadi Liakhovetski <g.liakhovetski+renesas@gmail.com>
>>> ---
>>>    arch/arm/mach-shmobile/board-ape6evm.c |   26 +++++++++++++++++++++-----
>>>    1 files changed, 21 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/arch/arm/mach-shmobile/board-ape6evm.c
>>> b/arch/arm/mach-shmobile/board-ape6evm.c
>>> index 5eb0caa..9c69643 100644
>>> --- a/arch/arm/mach-shmobile/board-ape6evm.c
>>> +++ b/arch/arm/mach-shmobile/board-ape6evm.c
>> [...]
>>> @@ -64,6 +74,13 @@ static const struct pinctrl_map ape6evm_pinctrl_map[] = {
>>>    				  "irqc_irq40", "irqc"),
>>>    };
>>>    +/*
>>> + * board devices
>>> + */
>>> +static struct platform_device *ape6evm_devices[] __initdata = {
>>> +	&lan9220_device,
>>> +};
>>> +
>>>    static void __init ape6evm_add_standard_devices(void)
>>>    {
>>>    @@ -91,9 +108,8 @@ static void __init ape6evm_add_standard_devices(void)
>>>      	regulator_register_fixed(0, dummy_supplies,
>>> ARRAY_SIZE(dummy_supplies));
>>>    -	platform_device_register_resndata(&platform_bus, "smsc911x", -1,
>>> -					  lan9220_res,
>>> ARRAY_SIZE(lan9220_res),
>>> -					  &lan9220_data,
>>> sizeof(lan9220_data));
>>> +	platform_add_devices(ape6evm_devices,
>>> +			     ARRAY_SIZE(ape6evm_devices));
>>     We have been discouraged from using platform_device_register() and its ilk
>> by Greg KH and Arnd, so
>>
>> platform_device_register_resndata() might have been there on purpose.
> Sure, I was surprised as well. Could you point me to a thread?

http://www.spinics.net/lists/linux-sh/msg16871.html  is all I have.


The original Greg's reasoning is missing there.

> I just
> thought - it's ok for one device, but repeating
> platform_device_register_*() multiple times for each device wouldn't look
> very nice. But I'll switch back to this, if that's really the current
> preference.

     Well, Simon tried to use platform_add_devices() despite that mail 
once --
I stopped him on the reason that platform_add_devices() is close 
relative of the
heathen platform_device_register(), i.e. former calls the latter.

> Thanks
> Guennadi

WBR, Sergei