Return-Path: Received: from relay5-d.mail.gandi.net ([217.70.183.197]:37148 "EHLO relay5-d.mail.gandi.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752953AbaILMYf (ORCPT ); Fri, 12 Sep 2014 08:24:35 -0400 Message-ID: <1410524654.32328.34.camel@hadess.net> Subject: Re: [PATCH] iio:kxcjk-1013: Add support for SMO8500 device From: Bastien Nocera To: Daniel Baluta Cc: linux-iio@vger.kernel.org, Jonathan Cameron , srinivas.pandruvada@linux.intel.com Date: Fri, 12 Sep 2014 14:24:14 +0200 In-Reply-To: <5412E5FA.2070707@intel.com> References: <1410283148.4077.16.camel@hadess.net> <1410459518.32328.24.camel@hadess.net> <5412E5FA.2070707@intel.com> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Sender: linux-iio-owner@vger.kernel.org List-Id: linux-iio@vger.kernel.org On Fri, 2014-09-12 at 15:24 +0300, Daniel Baluta wrote: > On 09/11/2014 09:18 PM, Bastien Nocera wrote: > > On Tue, 2014-09-09 at 21:53 +0300, Daniel Baluta wrote: > >> Hi Bastien, > >> > >> On Tue, Sep 9, 2014 at 8:19 PM, Bastien Nocera wrote: > >>> The Onda v975w tablet contains an accelerometer that's advertised over > >>> ACPI as SMO8500. This device is however a KXCJ9 accelerometer as > >>> can be seen in the Windows driver's INF file. > >>> --- > >>> > >>> I tested the patch, and it works correctly in my testing. However, > >>> there's no GPIO pin for the "data ready" interrupt, and this throws an > >>> error that looks fatal in the logs (could this be downgraded?), and > >>> the name of the device doesn't match the hid-sensor-hub ones, making it > >>> hard for user-space to discover them. > >>> (https://github.com/hadess/iio-sensor-proxy is my attempt at something > >>> that integrates with user-space and the IIO sensors) > >>> > How did you test it? Cat on the /sys files for each of the axis. I can see the values of x/y/z changing depending on the position of the device. > There is no GpioInt associated with your device in > the the .dsl file. I know, that's because the GPIO (according to the driver) is only used to know when there's more data to be read. We just have to poll instead... > What kind of error do you see in the logs? The "acpi gpio get index failed" error when it fails to find a GPIO pin for the "data ready" interrupt. Except that it works just fine without such an interrupt. > >> How does your ACPI .dsl file looks like? > >> > >>> It will also conflict with Daniel's patch here: > >>> http://www.spinics.net/lists/linux-iio/msg14438.html > >> It shouldn't. The patch is already in togreg branch of Jonathan's tree > >> [1]. Anyhow, > >> it will conflict with this one [2]. > > Do you want me to rebase the patch on a particular tree? Do you need any > > particular changes done in my commit message or patch to get this > > merged? > > > I would suggest you wait until Jonathan merges this patch [1] in togreg > branch [2]. > Then rebase your patch on togreg branch and get feedback from Jonathan > and Srinivas. OK, I'll have to monitor those. Cheers