From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751182AbeAVWL1 (ORCPT ); Mon, 22 Jan 2018 17:11:27 -0500 Received: from eddie.linux-mips.org ([148.251.95.138]:50148 "EHLO cvs.linux-mips.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750909AbeAVWL0 (ORCPT ); Mon, 22 Jan 2018 17:11:26 -0500 Date: Mon, 22 Jan 2018 22:50:57 +0100 From: Ladislav Michl To: Dmitry Torokhov Cc: Bjorn Helgaas , Wei Yongjun , linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org Subject: Re: [PATCH 1/5] devres: Move managed io function declarations into device.h Message-ID: <20180122215056.GA27905@lenoch> References: <20180121211432.GA15151@lenoch> <20180121211508.GB15151@lenoch> <20180122174903.z5ijxouq5bw2s36r@dtor-ws> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20180122174903.z5ijxouq5bw2s36r@dtor-ws> User-Agent: Mutt/1.9.2 (2017-12-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Jan 22, 2018 at 09:49:03AM -0800, Dmitry Torokhov wrote: > On Sun, Jan 21, 2018 at 10:15:08PM +0100, Ladislav Michl wrote: > > Moving managed io function declarations into device.h allows > > removing forward struct device and resource definitions from > > io(port).h > > In the face of it, what is the issue with forward declarations of device > and resource structures? device.h is supposed to be about Linux device > model, not hardware. You would not want all devm_* functions to go into > device.h (clock, regulator, input, rtc, hwmon, etc, etc devm API), > right? Why would we want ioport there? Allright, point taken. Then I would assume devm_ioremap_resource should be moved from device.h into io.h, to get some consistency, right? Any other comment (mainly to devm_ioremap_shared_resource) before v2? Thank you, ladis