From mboxrd@z Thu Jan 1 00:00:00 1970 From: Randy Dunlap Subject: Re: [PATCH v2 -next] spi: fix spi-sprd-adi build errors when SPI_SPRD_ADI=y and HWSPINLOCK=m Date: Mon, 25 Sep 2017 12:31:03 -0700 Message-ID: <156431d9-a518-53e7-17d9-bb1ee0f637bb@infradead.org> References: <20170922092621.uq2xsvxangeetewi@sirena.org.uk> <9dc8b064-8a6c-98d2-4ade-088cbfd56fb1@infradead.org> <20170925161221.v23dpeljin5kakhz@sirena.org.uk> <449abd0e-8d8b-86e9-2b7a-65c27ab3171b@infradead.org> <20170925192256.bmrm6gmb4idvomxb@sirena.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Cc: LKML , linux-spi To: Mark Brown , Geert Uytterhoeven Return-path: In-Reply-To: <20170925192256.bmrm6gmb4idvomxb@sirena.org.uk> Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-spi.vger.kernel.org On 09/25/17 12:22, Mark Brown wrote: > On Mon, Sep 25, 2017 at 09:20:31PM +0200, Geert Uytterhoeven wrote: > >> The issue is builtin drivers that depend on a modular API. The clean way >> is to separate API and implementation, so the API can be builtin, and the >> implementation can be modular. >> Hence the API should provide stubs that call into function pointers, to be >> registered by the module providing the implementation. > > In this case the problem is even more basic in that the driver does > actually depend on having hwspinlocks for any production use. > so just add: depends on HWSPINLOCK Is that satisfactory to you? -- ~Randy