From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dan Murphy Subject: Re: [PATCH v4 6/6] leds: lm36274: Introduce the TI LM36274 LED driver Date: Tue, 21 May 2019 13:25:56 -0500 Message-ID: References: <20190507201159.13940-1-dmurphy@ti.com> <20190507201159.13940-7-dmurphy@ti.com> <77f1ed5b-bfd2-510c-edd5-1b702f2d1d45@gmail.com> <8d126925-9e71-dba4-eb88-50fd6e6c06d8@ti.com> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit Return-path: In-Reply-To: Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org To: Jacek Anaszewski , pavel@ucw.cz Cc: lee.jones@linaro.org, rdunlap@infradead.org, linux-leds@vger.kernel.org, linux-kernel@vger.kernel.org List-Id: linux-leds@vger.kernel.org Jacek On 5/21/19 12:40 PM, Jacek Anaszewski wrote: > On 5/20/19 11:19 PM, Dan Murphy wrote: >> Jacek >> >> On 5/20/19 2:54 PM, Jacek Anaszewski wrote: >>> Hi Dan, >>> >>> On 5/7/19 10:11 PM, Dan Murphy wrote: >>>> Introduce the LM36274 LED driver.  This driver uses the ti-lmu >>>> MFD driver to probe this LED driver.  The driver configures only the >>>> LED registers and enables the outputs according to the config file. >>>> >>>> The driver utilizes the TI LMU (Lighting Management Unit) LED common >>>> framework to set the brightness bits. >>>> >>>> Signed-off-by: Dan Murphy >>>> --- >>>>    drivers/leds/Kconfig        |   7 ++ >>>>    drivers/leds/Makefile       |   1 + >>>>    drivers/leds/leds-lm36274.c | 174 ++++++++++++++++++++++++++++++++++++ >>>>    3 files changed, 182 insertions(+) >>>>    create mode 100644 drivers/leds/leds-lm36274.c >>>> >>>> diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig >>>> index 255fdd5e8491..db83a3feca01 100644 >>>> --- a/drivers/leds/Kconfig >>>> +++ b/drivers/leds/Kconfig >>>> @@ -791,6 +791,13 @@ config LEDS_LM3697 >>>>          Say Y to enable the LM3697 LED driver for TI LMU devices. >>>>          This supports the LED device LM3697. >>>>    +config LEDS_LM36274 >>>> +    tristate "LED driver for LM36274" >>>> +    depends on LEDS_TI_LMU_COMMON >>> >>> Shouldn't we have "depends on MFD_TI_LMU" as well here? >>> >> >> Actually the LEDS_TI_LMU_COMMON flag should depend on MFD_TI_LMU. >> Then it would inherit that dependency. > > LEDS_TI_LMU_COMMON does not seem too have any dependency on MFD_TI_LMU, > and it would be incorrect to require enabling MFD_TI_LMU for all drivers > depending on TI_LMU_COMMON, that can be probed on their own, like > leds-lm3697.c . > Correct. I can update the Kconfigs unless you want to ammend the commits. Dan