From mboxrd@z Thu Jan 1 00:00:00 1970 From: Lars Poeschel Subject: Re: [PATCH v3] gpio: interrupt consistency check for OF GPIO IRQs Date: Wed, 04 Sep 2013 10:35:44 +0200 Message-ID: <5584741.67AvVurcHF@lem-wkst-02> References: <1377526030-32024-1-git-send-email-larsi@wh2.tu-dresden.de> <52261C8B.2030906@wwwdotorg.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7Bit Return-path: In-Reply-To: <52261C8B.2030906@wwwdotorg.org> Sender: linux-kernel-owner@vger.kernel.org To: Stephen Warren Cc: Linus Walleij , Lars Poeschel , Grant Likely , "linux-gpio@vger.kernel.org" , "linux-kernel@vger.kernel.org" , "devicetree@vger.kernel.org" , Mark Rutland , Ian Campbell , Kumar Gala , Pawel Moll , Tomasz Figa , Javier Martinez Canillas , Enric Balletbo i Serra , Jean-Christophe PLAGNIOL-VILLARD , Santosh Shilimkar , Kevin Hilman , Balaji T K , Tony Lindgren , Jon Hunter List-Id: devicetree@vger.kernel.org On Wednesday 04 September 2013 11:29:47, Stephen Warren wrote: > On 09/03/2013 06:35 AM, Linus Walleij wrote: > > On Fri, Aug 30, 2013 at 9:53 PM, Stephen Warren wrote: > >> I still haven't seen an answer to why we really care about this; how > >> many times has code actually allocated the same GPIO/IRQ when it > >> shouldn't, in a way that it wasn't detectable by some other mechanism, > >> i.e. the feature just not working? Why are we even trying to solve this > >> issue? I'm not totally convinced it even makes sense to try and solve it. > > > > We care about this because a number of OMAP boards are not > > working properly when booted from device tree, and they have a hard > > time figuring out a solution to the problem. Last try exploded. Now > > they are looking to create a patch that will fix the actual problem. > > Is something missing from /proc/interrupts or /sys/kernel/debug/gpios? > If so, let's just add it. That would require procfs to be mounted to /proc and even CONFIG_PROC_FS to be compiled in. Drivers have to work without that requirement. That would require CONFIG_DEBUG_FS to be set, right? Drivers have to work without that.