From mboxrd@z Thu Jan  1 00:00:00 1970
From: Lee Jones <lee.jones@linaro.org>
Subject: Re: [PATCH v2 3/7] mfd: add support for sun6i PRCM
 (Power/Reset/Clock Management) unit
Date: Fri, 9 May 2014 09:59:41 +0100
Message-ID: <20140509085941.GM5767@lee--X1>
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>
 <536C91A6.6000500@free-electrons.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Return-path: <linux-doc-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <536C91A6.6000500@free-electrons.com>
Sender: linux-doc-owner@vger.kernel.org
To: Boris BREZILLON <boris.brezillon@free-electrons.com>
Cc: Hans de Goede <hdegoede@redhat.com>, Emilio =?iso-8859-1?Q?L=F3pez?= <emilio@elopez.com.ar>, Mike Turquette <mturquette@linaro.org>, Samuel Ortiz <sameo@linux.intel.com>, Chen-Yu Tsai <wens@csie.org>, Maxime Ripard <maxime.ripard@free-electrons.com>, Philipp Zabel <p.zabel@pengutronix.de>, Shuge <shuge@allwinnertech.com>, kevin@allwinnertech.com, Randy Dunlap <rdunlap@infradead.org>, devicetree@vger.kernel.org, linux-doc@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, dev@linux-sunxi.org
List-Id: devicetree@vger.kernel.org

> >>>>>>> +static struct platform_driver sun6i_prcm_driver =3D {
> >>>>>>> +	.driver =3D {
> >>>>>>> +		.name =3D "sun6i-prcm",
> >>>>>>> +		.owner =3D THIS_MODULE,
> >>>>>>> +		.of_match_table =3D sun6i_prcm_dt_ids,
> >>>>>>> +	},
> >>>>>>> +	.probe =3D sun6i_prcm_probe,
> >>>>>> You need a .remove() call-back.
> >>>>> This driver cannot be compiled as module (see the Kconfig defin=
ition)
> >>>>> and the devices are not hotpluggable, as a result a probed devi=
ce 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 thi=
s
> >> driver to ever call mfd_remove_devices() then?
> > No, this is an integral part of the SOC, which never gets removed i=
n any way.
>=20
> Lee, I'm about to send a 3rd version of this series, is it okay for y=
ou
> if I leave the remove function unimplemented ?

It is.

--=20
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org =E2=94=82 Open source software for ARM SoCs
=46ollow Linaro: Facebook | Twitter | Blog