From mboxrd@z Thu Jan 1 00:00:00 1970 From: Lee Jones Subject: Re: [PATCH v3 2/3] mfd: lubbock_io: add lubbock_io board Date: Wed, 21 Jan 2015 15:10:52 +0000 Message-ID: <20150121151052.GD22024@x1> References: <20150121123551.GD9258@x1> <332399927.402665086.1421847419490.JavaMail.root@zimbra1-e1.priv.proxad.net> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <332399927.402665086.1421847419490.JavaMail.root@zimbra1-e1.priv.proxad.net> Sender: linux-kernel-owner@vger.kernel.org To: robert.jarzmik@free.fr Cc: Russell King - ARM Linux , Mark Rutland , devicetree@vger.kernel.org, Samuel Ortiz , Pawel Moll , Ian Campbell , Dmitry Eremin-Solenikov , linux-kernel@vger.kernel.org, Haojian Zhuang , Rob Herring , Arnd Bergmann , linux-arm-kernel@lists.infradead.org, Kumar Gala , Daniel Mack List-Id: devicetree@vger.kernel.org On Wed, 21 Jan 2015, robert.jarzmik@free.fr wrote: > > ----- Mail original ----- > > De: "Lee Jones" > First of all, this is my web mail interface, so please be kind with > my mail formatting ... I have no idea what you want me to change or do differently? Perhaps it might be more prudent for you to switch to a quality mailer? > > Looking at one of the other patches in the series it appears the fl= ag > > you're trying to capture is IORESOURCE_IRQ_LOWEDGE. When I grep fo= r > > where this is being _used_ (think 'consumed, rather than passed. I > > only see a single entry in drivers/pnp/interface.c. > Look at this call chain : > request_irq() > setup_irq() > __setup_irq() > __irq_set_trigger() > pxa_gpio_irq_type() (aka. chip->irq_set_type) > =3D> hardware register manipulation >=20 > Moreover, you should be aware of the bijection existing between : > - IORESOURCE_IRQ_LOWEDGE and IRQF_TRIGGER_RISING > - it's HIGHEDGE twin > - it is noted in : include/linux/interrupt.h >=20 > > That's what got me thinking... are you sure you're a) making use of > this flag > Yes, I'm quite sure. >=20 > > b) assuming the answer to 'a' is "no" > I won't :) >=20 > > I think to set an edge trigger on an IRQ, you should instead do so = via > > irq_set_irq_type(), or have a missed a line or two? > __setup_irq(), that's the entry point. Very well, Russell and yourself have convinced me. If you fixup the remainder of comments, I'm happy. --=20 Lee Jones Linaro STMicroelectronics Landing Team Lead Linaro.org =E2=94=82 Open source software for ARM SoCs =46ollow Linaro: Facebook | Twitter | Blog