From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760995AbbA3KnV (ORCPT ); Fri, 30 Jan 2015 05:43:21 -0500 Received: from mail-qc0-f177.google.com ([209.85.216.177]:33857 "EHLO mail-qc0-f177.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753391AbbA3KnS (ORCPT ); Fri, 30 Jan 2015 05:43:18 -0500 MIME-Version: 1.0 In-Reply-To: <20150130092913.6256e517@endymion.delvare> References: <20150126112328.318da5e7@endymion.delvare> <1422369244.4499.187.camel@chaos.site> <20150129151711.454100be@endymion.delvare> <20150130092913.6256e517@endymion.delvare> Date: Fri, 30 Jan 2015 11:43:17 +0100 Message-ID: Subject: Re: [PATCH] mmc: Add hardware dependencies for sdhci-pxav3 and sdhci-pxav2 From: Ulf Hansson To: Jean Delvare Cc: Chris Ball , linux-mmc , "linux-kernel@vger.kernel.org" , Andrew Morton Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 30 January 2015 at 09:29, Jean Delvare wrote: > Hi Ulf, > > On Thu, 29 Jan 2015 16:01:48 +0100, Ulf Hansson wrote: >> On 29 January 2015 at 15:17, Jean Delvare wrote: >> > On Wed, 28 Jan 2015 15:04:24 +0100, Ulf Hansson wrote: >> >> For those SOC that want these drivers, they should be able to select >> >> them from their defconfigs. So it will be an opt-in instead of opt-out >> >> policy, which I prefer. It also follows the other Kconfig options for >> >> mmc drivers. >> > >> > As you wish. But that change would be a separate patch going on top of >> > mine, right? I'm not sure I understand what you expect from me at this >> > point, please clarify. >> >> Sorry for being unclear. I don't like $subject patch. >> >> Send a new one, removing the following lines: >> default CPU_MMP2 >> default CPU_PXA910 >> >> Then you send another patch(es) to the respective SOC maintainer, >> updating the defonfig(s) selecting MMC_SDHCI_PXAV3|PXAV2, when >> appropriate. >> >> Would that work? > > Not really, I'm afraid. > > My proposed change affects users of non-embedded systems, or more > generally everyone not on ARCH_MMP. I want to make their life easier by > hiding options which are not relevant to them. I have sent several > dozen of such patches in the past for various drivers, see for example > the latest ones: > http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=441fb7684782be3553c67dc04defcf304b999bba > http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=c03842d89b769db44be5cb0b1ebb384ccfa25f7f > http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=84c3a8f6eadb2bedfba10f62da0328d8533c8f25 > Also note that the MMC subsystem already has examples of this, check > MMC_OMAP_HS, MMC_SDHCI_MSM, MMC_SDHI, MMC_DW and MMC_SH_MMCIF in > drivers/mmc/host/Kconfig. I'm just doing more of the same, nothing new. > > The change you want, OTOH, would affect exclusively the ARCH_MMP users > (of which I am not.) It is essentially unrelated with what I was > originally talking about, except for the fact that it touches the same > Kconfig entries. I have no idea if your proposal is a good idea, I am > not dealing with embedded systems, and I have no idea who are the > maintainers of the affected SOCs. This is simply not my area. > > So basically you are rejecting my proposal without a reason, and then > you ask me to do an unrelated work instead. This is not fair, sorry. > > Don't get me wrong, I'm always ready to do some more work than I > originally intended if that's what it takes to get my patches merged. I > value code review and I welcome constructive criticism. But this time > your request is not reasonable. I did a little more of investigation of what's good practice/policy around your suggested patch. I have changed my mind, I am going to accept your patch as is. Sorry for all the noise and thanks for a good discussion. Applied for next. Thanks! Kind regards Uffe