From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnd Bergmann Subject: Re: [PATCH V7 03/11] pci, of: Move the PCI I/O space management to PCI core code. Date: Wed, 11 May 2016 13:01:18 +0200 Message-ID: <4722186.M8gaIx9LyT@wuerfel> References: <1462893601-8937-1-git-send-email-tn@semihalf.com> <5732E11A.9090002@semihalf.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7Bit Return-path: In-Reply-To: <5732E11A.9090002@semihalf.com> Sender: linux-pci-owner@vger.kernel.org To: Tomasz Nowicki Cc: "Rafael J. Wysocki" , Bjorn Helgaas , Will Deacon , Catalin Marinas , Hanjun Guo , Lorenzo Pieralisi , Sinan Kaya , jchandra@broadcom.com, robert.richter@caviumnetworks.com, mw@semihalf.com, Liviu.Dudau@arm.com, David Daney , wangyijing@huawei.com, Suravee Suthikulanit , Mark Salter , Linux PCI , "linux-arm-kernel@lists.infradead.org" , ACPI Devel Maling List , Linux Kernel Mailing List , "linaro-acpi@lists.linaro.org" , Jon Masters , andrea.gallo@linaro.org, dhdang@a List-Id: linux-acpi@vger.kernel.org On Wednesday 11 May 2016 09:36:58 Tomasz Nowicki wrote: > > > > I understand that this moves code around, but those in-function > > #ifdefs aren't nice. Any chance to get rid of them but putting whole > > functions under the #ifdef? > > > > This is a __weak implementation, so assuming I would move #ifdef out of > function I need to provide another empty __weak stub. I do not know > which solution is more ugly. In any case we can do that cleanup separately. > I'd vote for just dropping the __weak here, given that there is no non-weak implementation. If we end up needing to override this for some architecture or host bridge in the future, we can think about how to best do that then. I agree that should be a separate patch, this one should only move code from one file to another. Arnd