From mboxrd@z Thu Jan 1 00:00:00 1970 From: Greg KH Subject: Re: [PATCH 0/5] Generalize module_platform_driver Date: Wed, 16 Nov 2011 08:02:06 -0800 Message-ID: <20111116160206.GA27049@suse.de> References: <1321434819-23678-1-git-send-email-lars@metafoo.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Jean Delvare , Grant Likely , Jonathan Cameron , Michael Hennerich , linux-iio-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, spi-devel-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org To: Lars-Peter Clausen Return-path: Content-Disposition: inline In-Reply-To: <1321434819-23678-1-git-send-email-lars-Qo5EllUWu/uELgA04lAiVw@public.gmane.org> Sender: linux-iio-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-spi.vger.kernel.org On Wed, Nov 16, 2011 at 10:13:34AM +0100, Lars-Peter Clausen wrote: > Grant Likely recently introduced the module_platform_driver macro which can be > used to eliminate a few lines on boilerplate code in platform driver modules. > The same approach can be used to do the same for other bus type drivers. > > The first patch of this series generalizes the module_platform_driver macro and > introduces the module_driver macro. It is similar to module_platform_driver > macro but has two additional parameters to pass in the driver register and > unregister function. The intend is that this macro is used to construct bus > specific macros for generating the driver register/unregister boilerplate code. > > The next two patches in this series add the module_i2c_driver and > module_spi_driver macro which use the module_driver macro to generate the I2C > and SPI driver boilerplate code. > > The last two patches convert the driver found in the IIO framework to use the > new module_i2c_driver and module_spi_driver macros to demonstrate their > potential and remove over 700 lines of code. > > While this series only introduces these kind of helper macros for I2C and SPI > bus drivers the same scheme should be applicable to most other bus driver types. > For example PCI and USB seem to be good candidates. > > It probably makes sense to merge the first three patches together. The last two > can probably, since this is not urgent, wait until the first three have reached > mainline. Nice, I like this, unless people object, I'll merge the first patch to the driver-core tree. I'll gladly take the i2c and spi patches as well if the subsystem maintainers there don't object, and I'll work on the USB patch as well. thanks, greg k-h