From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Andrew F. Davis" Subject: Re: [PATCH v2 5/6] regulator: tps65086: Add regulator driver for the TPS65086 PMIC Date: Tue, 3 Nov 2015 11:57:03 -0600 Message-ID: <5638F56F.2050102@ti.com> References: <1446244904-29315-1-git-send-email-afd@ti.com> <1446244904-29315-6-git-send-email-afd@ti.com> <20151101023735.GE28319@sirena.org.uk> <5638D97A.8040306@ti.com> <20151103163117.GQ20228@sirena.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20151103163117.GQ20228@sirena.org.uk> Sender: linux-kernel-owner@vger.kernel.org To: Mark Brown Cc: Rob Herring , Pawel Moll , Mark Rutland , Ian Campbell , Kumar Gala , Linus Walleij , Alexandre Courbot , Samuel Ortiz , Lee Jones , Liam Girdwood , devicetree@vger.kernel.org, linux-gpio@vger.kernel.org, linux-kernel@vger.kernel.org List-Id: devicetree@vger.kernel.org On 11/03/2015 10:31 AM, Mark Brown wrote: > On Tue, Nov 03, 2015 at 09:57:46AM -0600, Andrew F. Davis wrote: >> On 10/31/2015 09:37 PM, Mark Brown wrote: >>> On Fri, Oct 30, 2015 at 05:41:43PM -0500, Andrew F. Davis wrote: > >>>> + struct regulator_desc desc; >>>> + unsigned int decay_reg; >>>> + unsigned int decay_mask; > >>> The decay_reg and decay_mask fields appear to be entirely write only, >>> why are they present? > >> Not sure what you mean, they are written to differently for each regulator, >> and read when setting the decay mode in _probe. > > Ugh, that's a bit hidden (mostly due to not using the callback). > >>>> + ret = of_regulator_match(&pdev->dev, pdev->dev.of_node, >>>> + tps65086_matches, >>>> + ARRAY_SIZE(tps65086_matches)); >>>> + if (ret < 0) { >>>> + dev_err(tps->dev, "Error parsing regulator init data\n"); >>>> + return ret; >>>> + } > >>> Please match regulators using the fields in the regulator_desc rather >>> than open coding. > >> I need to match to get the driver specific information for each regulator, >> if I let regulator_register do it I never get a chance to process the match, >> unless there is a callback or something I'm missing. > > There's a callback you are missing, of_parse_cb(). > Ah, that will make things easier, I'll re-spin with the callback.