From mboxrd@z Thu Jan  1 00:00:00 1970
From: boris.brezillon@free-electrons.com (Boris BREZILLON)
Date: Fri, 09 May 2014 10:28:22 +0200
Subject: [PATCH v2 3/7] mfd: add support for sun6i PRCM (Power/Reset/Clock
 Management) unit
In-Reply-To: <536C8F71.5040604@redhat.com>
References: <1399483554-8824-1-git-send-email-boris.brezillon@free-electrons.com>
 <1399483554-8824-4-git-send-email-boris.brezillon@free-electrons.com>
 <20140508110205.GB5767@lee--X1> <536BE366.8080907@free-electrons.com>
 <20140509071246.GE5767@lee--X1> <536C84F6.2050701@redhat.com>
 <20140509080807.GH5767@lee--X1> <536C8F71.5040604@redhat.com>
Message-ID: <536C91A6.6000500@free-electrons.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org


On 09/05/2014 10:18, Hans de Goede wrote:
> Hi,
>
> On 05/09/2014 10:08 AM, Lee Jones wrote:
>>>>>>> +static struct platform_driver sun6i_prcm_driver = {
>>>>>>> +	.driver = {
>>>>>>> +		.name = "sun6i-prcm",
>>>>>>> +		.owner = THIS_MODULE,
>>>>>>> +		.of_match_table = sun6i_prcm_dt_ids,
>>>>>>> +	},
>>>>>>> +	.probe = sun6i_prcm_probe,
>>>>>> You need a .remove() call-back.
>>>>> This driver cannot be compiled as module (see the Kconfig definition)
>>>>> and the devices are not hotpluggable, as a result a probed device will
>>>>> never be removed.
>>>>>
>>>>> Do you still want me to implement the remove function ?
>>>> .remove() also be run on shut down.
>>> That is not true, if your device needs to do anything special at shutdown
>>> you need to add a shutdown callback. Devices are kept as is (not torn down)
>>> on shutdown.
>> Ah okay, I'll bow to your knowledge.  So there's no reason for this
>> driver to ever call mfd_remove_devices() then?
> No, this is an integral part of the SOC, which never gets removed in any way.

Lee, I'm about to send a 3rd version of this series, is it okay for you
if I leave the remove function unimplemented ?

>
> Regards,
>
> Hans

-- 
Boris Brezillon, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com