From mboxrd@z Thu Jan 1 00:00:00 1970 From: Robert Baldyga Subject: Re: [PATCH v2] mfd: max8997: use regmap to access registers Date: Fri, 07 Mar 2014 15:40:44 +0100 Message-ID: <5319DA6C.1050204@samsung.com> References: <1394031497-16648-1-git-send-email-r.baldyga@samsung.com> <20140305185439.GC5453@core.coreip.homeip.net> <531818D6.3080500@samsung.com> <20140306162951.GA14318@core.coreip.homeip.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-reply-to: <20140306162951.GA14318@core.coreip.homeip.net> Sender: linux-leds-owner@vger.kernel.org To: Dmitry Torokhov Cc: sameo@linux.intel.com, lee.jones@linaro.org, myungjoo.ham@samsung.com, cw00.choi@samsung.com, cooloney@gmail.com, rpurdie@rpsys.net, dbaryshkov@gmail.com, dwmw2@infradead.org, lgirdwood@gmail.com, broonie@kernel.org, a.zummo@towertech.it, paul.gortmaker@windriver.com, linux-kernel@vger.kernel.org, linux-input@vger.kernel.org, linux-leds@vger.kernel.org, rtc-linux@googlegroups.com, m.szyprowski@samsung.com, k.kozlowski@samsung.com List-Id: linux-input@vger.kernel.org On 03/06/2014 05:29 PM, Dmitry Torokhov wrote: > On Thu, Mar 06, 2014 at 07:42:30AM +0100, Robert Baldyga wrote: >> Hi, >> >> On 03/05/2014 07:54 PM, Dmitry Torokhov wrote: >>> Hi Robert, >>> >>> On Wed, Mar 05, 2014 at 03:58:17PM +0100, Robert Baldyga wrote: >>>> >>>> -int max8997_write_reg(struct i2c_client *i2c, u8 reg, u8 value) >>>> +int max8997_write_reg(struct regmap *map, u8 reg, u8 value) >>> >>> Why don't you make read/write reg to take struct max8997_dev as argument >>> instead of regmap? regmap seems to be the current implementation du jur, >>> but that is core's detail, functions do not need to care. >>> >>> Thanks. >>> >> >> It's because there are few regmaps in max8997_dev, and read/write reg >> functions has no way to check which regmap should be used. I think it >> would be clearer if I remove this functions and use regmap_read and >> regmap_write instead, because there is no particular purpose for them. > > I see.. Then another question - why do the regmaps belong to the core > of the driver instead of individual functions? Why haptic regmap, for > example, can not be initialized in haptic code instead? > It looks like it would not be a big problem. I will try to prepare new version of this patch including this. Best regards Robert Baldyga Samsung R&D Institute Poland