From mboxrd@z Thu Jan 1 00:00:00 1970 From: Darius Augulis Subject: Re: [PATCH 1/5] I2C driver for MXC Date: Tue, 31 Mar 2009 14:02:52 +0300 Message-ID: <49D1F85C.8040703@gmail.com> References: <49C89E13.7040801@gmail.com> <20090324094215.GA3471@pengutronix.de> <49C8ABE5.40408@gmail.com> <20090325092625.GA3029@pengutronix.de> <49CA017D.20005@gmail.com> <20090325104357.GB3029@pengutronix.de> <20090330003012.GM19758@fluff.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20090330003012.GM19758-elnMNo+KYs3pIgCt6eIbzw@public.gmane.org> Sender: linux-i2c-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Ben Dooks Cc: Guennadi Liakhovetski , Wolfram Sang , linux-arm-kernel-xIg/pKzrS19vn6HldHNs0ANdhmdF6hFW@public.gmane.org, linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Sascha Hauer List-Id: linux-i2c@vger.kernel.org Ben Dooks wrote: > On Sat, Mar 28, 2009 at 10:36:50PM +0100, Guennadi Liakhovetski wrote: > >> On Wed, 25 Mar 2009, Wolfram Sang wrote: >> >> >>>> Because I hurry to submit this, because merge window is open. >>>> >> Ok, I waited for this driver, I really did, and I really prefer having one >> driver for as many hardware (SoC) variants as possible instead of having >> different drivers, and I really hoped its new version will work >> out-of-the-box in my setup (pcm037 with a mt9t031 camera). Unfortunately, >> it didn't. I've spent more than a full working day trying to get it to >> work, but I didn't succeed so far. It works with the on-board rtc, but it >> doesn't work with the camera, connected over a flex cable. >> >> Attached to this email is a version of the i2c driver for i.mx31 that I've >> been using with my setup for about half a year now. I adjusted it slightly >> to accept the same platform data, so, it is really a drop-in replacement >> for i2c-imx.c. Of course, you have to adjust or extend the Makefile. I >> actually added a new Kconfig entry for this driver, so I could easily >> compare them during the tests. >> >> The source of "my" version does look more logical and more clean to me on >> quite a few occasions. So, maybe someone could give it a quick spin on >> other *mx* SoCs and see if we can use this one instead? You might have to >> replace {read,write}w with readb and writeb, so, it might be a good idea >> to abstract them into inlines or macros. Otherwise, I think, it might work >> for other CPUs straight away. >> >> I just would like to avoid committing a driver and having to spend hours >> or days fixing it afterwards when a possibly more mature version exists. >> > > I'm going to hold off pushing my current tree until this/tommorow > evening so that people have some time to get back to me on what is > the best way to proceed. > > Hi Ben, Jean already requested Linus to pull I2C updates, how about imx-i2c? Should we have this in rc1? Darius.