Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754255Ab3JRNXG (ORCPT ); Fri, 18 Oct 2013 09:23:06 -0400 Received: from mail-ie0-f174.google.com ([209.85.223.174]:43141 "EHLO mail-ie0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752210Ab3JRNXD (ORCPT ); Fri, 18 Oct 2013 09:23:03 -0400 Date: Fri, 18 Oct 2013 09:22:57 -0400 From: Matt Porter To: Mark Rutland Cc: Tim Kryger , "devicetree@vger.kernel.org" , Christian Daudt , Pawel Moll , Ian Campbell , Daniel Lezcano , "linux-mmc@vger.kernel.org" , "linux-kernel@vger.kernel.org" , "rob.herring@calxeda.com" , "patches@linaro.org" , Stephen Warren , Thomas Gleixner , Chris Ball , "linux-arm-kernel@lists.infradead.org" Subject: Re: [RESEND PATCH v2 6/6] mmc: sdhci-bcm-kona: Add basic use of clocks Message-ID: <20131018132257.GG2721@beef> References: <1381960030-1640-1-git-send-email-tim.kryger@linaro.org> <1381960030-1640-7-git-send-email-tim.kryger@linaro.org> <20131017141309.GG24056@e106331-lin.cambridge.arm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20131017141309.GG24056@e106331-lin.cambridge.arm.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1413 Lines: 33 On Thu, Oct 17, 2013 at 03:13:09PM +0100, Mark Rutland wrote: > On Wed, Oct 16, 2013 at 10:47:10PM +0100, Tim Kryger wrote: > > + /* Get and enable the external clock */ > > + kona_dev->external_clk = devm_clk_get(dev, NULL); > > + if (IS_ERR(kona_dev->external_clk)) { > > + dev_err(dev, "Failed to get external clock\n"); > > + ret = PTR_ERR(kona_dev->external_clk); > > + goto err_pltfm_free; > > This seems like a pretty clear breakage of existing DTBs. > > Why? Normal kernel incremental development. The notion of breaking DTBs is pretty much incompatible with the long-standing tradition of developing kernel support in an incremental fashion. In this case, the bootloader available on these boards has had clocks and regulators enabled for all the drivers currently upstream. This allowed development of the sdhci driver (and others) independent of having the common clk driver and PMU/regulator drivers ready. That's been especially important given the flux that the common clk framework is in with regards to bindings. Are you suggesting that the required clock needs to be made logically optional in the code to retain backward compatbility? -Matt -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/