From mboxrd@z Thu Jan 1 00:00:00 1970 From: Krzysztof Kozlowski Subject: Re: [PATCH 2/8] regulator: max77686: Remove support for board files Date: Tue, 28 Oct 2014 09:40:50 +0100 Message-ID: <1414485650.24949.10.camel@AMDC1943> References: <1414422226-10948-1-git-send-email-k.kozlowski@samsung.com> <1414422226-10948-3-git-send-email-k.kozlowski@samsung.com> <20141028003709.GL18557@sirena.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: In-reply-to: <20141028003709.GL18557@sirena.org.uk> Sender: linux-kernel-owner@vger.kernel.org To: Mark Brown Cc: Samuel Ortiz , Lee Jones , Liam Girdwood , linux-kernel@vger.kernel.org, Ben Dooks , Kukjin Kim , Russell King , linux-arm-kernel@lists.infradead.org, linux-samsung-soc@vger.kernel.org, devicetree@vger.kernel.org, Kyungmin Park , Marek Szyprowski , Bartlomiej Zolnierkiewicz , Javier Martinez Canillas , Chanwoo Choi List-Id: devicetree@vger.kernel.org On wto, 2014-10-28 at 00:37 +0000, Mark Brown wrote: > On Mon, Oct 27, 2014 at 04:03:40PM +0100, Krzysztof Kozlowski wrote: > > The driver is used only on Exynos4 based boards with DTS support. > > Convert the driver to DTS-only version. > > This doesn't seem like a particularly persuasive reason honestly and if > you're going to mess around with this stuff please fix it properly... > > > struct max77686_data { > > + struct max77686_regulator_data *regulators; > > + int num_regulators; > > + > > ...the only reason for this array and the specification of the number of > regulators is to make platform data easier to do... > > > - pdata->num_regulators = ARRAY_SIZE(regulators); > > + max77686->num_regulators = ARRAY_SIZE(regulators); > > ...and indeed we wind up with a constant here anyway. At the very least > it'd be better to pull the parsing into the registration, right now we > have the code still laid out for platform data so it's hard to see it as > a win in cleanup terms. If we're not translating into platform data we > shouldn't need to keep this stuff around outside of the probe function. > > What would be even better would be to convert to use the standard DT > parsing with regulators_node and of_match specified in the descriptor > and then delete most of this code entirely. You're right, there's a lot to clean in this driver. The only caveat is that I can't test the max77802 regulator so I refrained from doing intrusive changes to it. I'll try to clean up both drivers and make max77802 as RFT. Best regards, Krzysztof