From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-out.m-online.net ([212.18.0.10]:57978 "EHLO mail-out.m-online.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750731AbaBAF2M (ORCPT ); Sat, 1 Feb 2014 00:28:12 -0500 From: Marek Vasut To: Matt Ranostay Subject: Re: [PATCH 2/2] iio: Add AS3935 lightning sensor support Date: Sat, 1 Feb 2014 06:28:04 +0100 Cc: linux-kernel@vger.kernel.org, linux-iio@vger.kernel.org, Matt Porter , Koen Kooi , Pantelis Antoniou , Mark Brown References: <1391182703-2201-1-git-send-email-mranostay@gmail.com> <201402010412.24434.marex@denx.de> In-Reply-To: MIME-Version: 1.0 Content-Type: Text/Plain; charset="us-ascii" Message-Id: <201402010628.04324.marex@denx.de> Sender: linux-iio-owner@vger.kernel.org List-Id: linux-iio@vger.kernel.org On Saturday, February 01, 2014 at 06:03:22 AM, Matt Ranostay wrote: > Duly noted. Will be fixed in the next rev... mutexs are pointless as you > say.. only corner case would be if you were changing gain_boost and a event > came in at the same time. But that wouldn't harm anything except slightly > delay the gain_boost being updated by some x microseconds. Please stop top-posting ;-) btw I think you might need a mutex around the entire: +static void calibrate_as3935(struct as3935_state *st) +{ + /* mask disturber interrupt bit */ + as3935_write(st, AS3935_INT, 1 << 5); + + as3935_write(st, AS3935_CALIBRATE, 0x96); + as3935_write(st, AS3935_TUNE_CAP, 1 << 5 | st->tune_cap); + + mdelay(2); + as3935_write(st, AS3935_TUNE_CAP, st->tune_cap); +} and similar functions where you do a bunch of register accesses. This is because you probably want to protect them against concurent execution so the chip won't be confused if a user were to poke something via SYSFS twice. Best regards, Marek Vasut