From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757952AbaHZMXQ (ORCPT ); Tue, 26 Aug 2014 08:23:16 -0400 Received: from mail-ie0-f180.google.com ([209.85.223.180]:58512 "EHLO mail-ie0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757855AbaHZMXO (ORCPT ); Tue, 26 Aug 2014 08:23:14 -0400 Date: Tue, 26 Aug 2014 13:23:04 +0100 From: Lee Jones To: Chris Zhong Cc: robh+dt@kernel.org, pawel.moll@arm.com, mark.rutland@arm.com, ijc+devicetree@hellion.org.uk, galak@codeaurora.org, sameo@linux.intel.com, lgirdwood@gmail.com, broonie@kernel.org, a.zummo@towertech.it, mturquette@linaro.org, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, rtc-linux@googlegroups.com, grant.likely@linaro.org, hl@rock-chips.com, huangtao@rock-chips.com, cf@rock-chips.com, zhangqing@rock-chips.com, xxx@rock-chips.com, dianders@chromium.org, heiko@sntech.de, olof@lixom.net, sonnyrao@chromium.org, dtor@chromium.org, javier.martinez@collabora.co.uk, kever.yang@rock-chips.com Subject: Re: [PATCH v2 2/5] MFD: RK808: Add new mfd driver for RK808 Message-ID: <20140826122304.GA26707@lee--X1> References: <1408505461-24200-1-git-send-email-zyw@rock-chips.com> <20140820092117.GI4266@lee--X1> <53FA9D1E.8040202@rock-chips.com> <20140826092205.GG9574@lee--X1> <53FC5BD1.7030406@rock-chips.com> <20140826102036.GJ9574@lee--X1> <53FC63EE.6060104@rock-chips.com> <20140826112013.GK9574@lee--X1> <53FC77F0.4030004@rock-chips.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <53FC77F0.4030004@rock-chips.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 26 Aug 2014, Chris Zhong wrote: > On 08/26/2014 07:20 PM, Lee Jones wrote: > >On Tue, 26 Aug 2014, Chris Zhong wrote: > >>On 08/26/2014 06:20 PM, Lee Jones wrote: > >>>On Tue, 26 Aug 2014, Chris Zhong wrote: > >>>>On 08/26/2014 05:22 PM, Lee Jones wrote: > >>>>>On Mon, 25 Aug 2014, Chris Zhong wrote: > >>>>>>On 08/20/2014 05:21 PM, Lee Jones wrote: > >>>>>>>On Wed, 20 Aug 2014, Chris Zhong wrote: > >>>>>>> > >>>>>>>>The RK808 chip is a power management IC for multimedia and handheld > >>>>>>>>devices. It contains the following components: > >>>>>>>> > >>>>>>>>- Regulators > >>>>>>>>- RTC > >>>>>>>> > >>>>>>>>The rk808 core driver is registered as a platform driver and provides > >>>>>>>>communication through I2C with the host device for the different > >>>>>>>>components. > >>>>>>>> > >>>>>>>>Signed-off-by: Chris Zhong > >>>>>>>>--- > >>>>>[...] > >>>>> > >>>>>>>>+ rk808->pdata = pdata; > >>>>>>>Remove pdata from 'struct rk808'. You can obtain it from dev. > >>>>>>Can I keep this pdata in rk808, because it is used in the regulator driver. > >>>>>>The one obtain from dev maybe empty. > >>>>>If the one in dev is empty, you should populate that instead. > >>>>So, should I malloc a pada, and assign it to client->dev? > >>>If client->dev.pdata is NULL, yes. > >>> > >>>But actually, I have more important questions that need to be answered > >>>first. Ones which I would normally be able to answer myself if the > >>>patch-set had been sent as one (i.e. threaded) instead of as > >>>individual patches: > >>> > >>>- What are you using pdata for? > >>For save some properties, > >> > >>like "rockchip,system-power-controller" in MFD > >>and some regulator properties in regulator/rk808... > >> > >>>- Where is pdata populated? > >>It is populated in probe in mfd/rk808.c > >> > >>actually, I copy it from tps65910.c > >> > >> > >>>- Where is pdata used? > >>pdata is used in mfd and regulator > >I'm still a little confused. I see it being populated in the MFD > >driver, then I only see the attributes populated in the MFD driver > >used in the MFD driver and nowhere else? I do see other attributes > >used in the regulator driver i.e. .of_node[], but these are then used > >only in the regulator driver, thus they shouldn't really be pdata. > > > >Let me put it another way: > > > >struct rk808_board { > >+ int irq; > >+ int irq_base; > >+ int wakeup; > >+ bool pm_off; > >+ struct regulator_init_data *rk808_init_data[rk808_NUM_REGULATORS]; > >+ struct device_node *of_node[rk808_NUM_REGULATORS]; > >+ int pmic_sleep_gpio; > >+ unsigned int dcdc_slp_voltage[3]; /* buckx_voltage in uV */ > >+ unsigned int ldo_slp_voltage[7]; > >+}; > > > >For each of the above: > > > >- Can it be passed from platform data i.e. arch//{plat,mach}-*? > >- Can it use local (not passed from driver to driver) variable instead? > > > >If the answer to the first question is 'no' and/or if the answer to > >the second question is 'yes', then it shouldn't be platform data. > > Yes, if we have no dts file, these value can pass from mach-rk***. > > It can be filled with all user setting of rk808, in a struct. > > the latest struct is: > > struct rk808_board { > > int wakeup; > > bool pm_off; > > struct regulator_init_data *rk808_init_data[RK808_NUM_REGULATORS]; > > struct device_node *of_node[RK808_NUM_REGULATORS]; > > unsigned int dcdc_slp_voltage[3]; /* buckx_voltage in uV */ > > unsigned int ldo_slp_voltage[7]; > > }; This version looks much more acceptable. Although, I am concerned about of_node[]. Isn't that only populated and used in the regulator driver? -- Lee Jones Linaro STMicroelectronics Landing Team Lead Linaro.org │ Open source software for ARM SoCs Follow Linaro: Facebook | Twitter | Blog