From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ppsw-51.csi.cam.ac.uk ([131.111.8.151]:49404 "EHLO ppsw-51.csi.cam.ac.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752146Ab2FSHRV (ORCPT ); Tue, 19 Jun 2012 03:17:21 -0400 Message-ID: <4FE0277E.40007@cam.ac.uk> Date: Tue, 19 Jun 2012 08:17:18 +0100 From: Jonathan Cameron MIME-Version: 1.0 To: Greg Kroah-Hartman CC: Lars-Peter Clausen , devel@driverdev.osuosl.org, linux-iio@vger.kernel.org Subject: Re: [PATCH 02/10] iio: Add helper function for initializing triggered buffers References: <1340037236-15685-1-git-send-email-lars@metafoo.de> <1340037236-15685-2-git-send-email-lars@metafoo.de> <20120619002818.GA20240@kroah.com> In-Reply-To: <20120619002818.GA20240@kroah.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Sender: linux-iio-owner@vger.kernel.org List-Id: linux-iio@vger.kernel.org On 6/19/2012 1:28 AM, Greg Kroah-Hartman wrote: > On Mon, Jun 18, 2012 at 06:33:48PM +0200, Lars-Peter Clausen wrote: >> --- a/drivers/iio/Makefile >> +++ b/drivers/iio/Makefile >> @@ -7,6 +7,7 @@ industrialio-y := industrialio-core.o industrialio-event.o inkern.o >> industrialio-$(CONFIG_IIO_BUFFER) += industrialio-buffer.o >> industrialio-$(CONFIG_IIO_TRIGGER) += industrialio-trigger.o >> >> +obj-$(CONFIG_IIO_TRIGGERED_BUFFER) += industrialio-triggered-buffer.o > > That's a crazy module name, does it really need to be that big? And in > a separate module? Probably (for now at least). I actually asked Lars-Peter to expand it as the previous version he had wasn't descriptive enough. > > Anyway, I've taken it, but why can't this just be merged into the iio.ko > main module? Actually, why can't all of these different trigger and > buffer modules be part of the iio module? Leads to one rather big module and when quite a few users (including me) work on boards with very little RAM that can be an issue if they only want a simple driver that just requires the core. 32MB in my case. Having said that, some consolidation may make sense. We could just make kfifo_buf dependent on IIO_TRIGGER. There are very few drivers (possibly none?) that use one without the other, despite that being a perfectly reasonable option. Any that do could just pay the slight cost. That would allow us to merge this into kfifo buf. Having said that, I'm still interested in a better alternative to kfifo_buf for some devices. I was crossing my fingers the work on unifying the various tracing buffers would produce something suitable. Not seen anything on that front for a while though. > >> obj-$(CONFIG_IIO_KFIFO_BUF) += kfifo_buf.o > > Shouldn't this be iio_kfifo_buf? Certainly need to do some work on getting the naming at least consistent and indeed a prefix on that does make sense. > > thanks, > > greg k-h