From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dmitry Torokhov Subject: Re: [PATCH v2 3/3] tps6507x-ts: add DT bindings description Date: Fri, 10 Mar 2017 12:08:17 -0800 Message-ID: <20170310200817.GA35542@dtor-ws> References: <1489069869-3849-1-git-send-email-yegorslists@googlemail.com> <1489069869-3849-4-git-send-email-yegorslists@googlemail.com> <20170309175530.GC20077@dtor-ws> <20170310180018.GB21811@dtor-ws> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mail-pg0-f65.google.com ([74.125.83.65]:34010 "EHLO mail-pg0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750897AbdCJUIW (ORCPT ); Fri, 10 Mar 2017 15:08:22 -0500 Received: by mail-pg0-f65.google.com with SMTP id b5so11359692pgg.1 for ; Fri, 10 Mar 2017 12:08:21 -0800 (PST) Content-Disposition: inline In-Reply-To: Sender: linux-input-owner@vger.kernel.org List-Id: linux-input@vger.kernel.org To: Yegor Yefremov Cc: Sekhar Nori , Rob Herring , "linux-input@vger.kernel.org" , Lee Jones , Mark Rutland , Andrey Skvortsov , Kevin Hilman On Fri, Mar 10, 2017 at 08:54:51PM +0100, Yegor Yefremov wrote: > On Fri, Mar 10, 2017 at 7:00 PM, Dmitry Torokhov > wrote: > > On Fri, Mar 10, 2017 at 12:11:08PM +0530, Sekhar Nori wrote: > >> On Thursday 09 March 2017 11:25 PM, Dmitry Torokhov wrote: > >> > On Thu, Mar 09, 2017 at 04:18:40PM +0100, Yegor Yefremov wrote: > >> >> On Thu, Mar 9, 2017 at 3:49 PM, Rob Herring wrote: > >> >>> On Thu, Mar 9, 2017 at 8:31 AM, wrote: > >> >>>> From: Yegor Yefremov > >> >>> > >> >>> This needs to go to DT list. > >> >> > >> >> Will do. > >> >> > >> >>>> > >> >>>> Provide description for following properties: > >> >>>> > >> >>>> - ti,poll-period > >> >>>> - ti,min-pressure > >> >>>> > >> >>>> Signed-off-by: Yegor Yefremov > >> >>>> --- > >> >>>> Documentation/devicetree/bindings/mfd/tps6507x.txt | 16 ++++++++++++++-- > >> >>>> 1 file changed, 14 insertions(+), 2 deletions(-) > >> >>>> > >> >>>> diff --git a/Documentation/devicetree/bindings/mfd/tps6507x.txt b/Documentation/devicetree/bindings/mfd/tps6507x.txt > >> >>>> index 8fffa3c..8875662 100644 > >> >>>> --- a/Documentation/devicetree/bindings/mfd/tps6507x.txt > >> >>>> +++ b/Documentation/devicetree/bindings/mfd/tps6507x.txt > >> >>>> @@ -1,4 +1,8 @@ > >> >>>> -TPS6507x Power Management Integrated Circuit > >> >>>> +TPS6507x Multifunctional Device. > >> >>>> + > >> >>>> +Features provided by TPS6507x: > >> >>>> + 1. Power Management Integrated Circuit. > >> >>>> + 2. Touch-Screen. > >> >>>> > >> >>>> Required properties: > >> >>>> - compatible: "ti,tps6507x" > >> >>>> @@ -30,6 +34,12 @@ Regulator Optional properties: > >> >>>> 1: If defdcdc pin of DCDC2/DCDC3 is driven HIGH. > >> >>>> If this property is not defined, it defaults to 0 (not enabled). > >> >>>> > >> >>>> +Touchscreen Optional properties: > >> >>>> +- ti,poll-period: Time at which touch input is getting sampled in ms. > >> >>>> + Default value: 30 ms. > >> >>> > >> >>> Isn't there a standard property for this? > >> >> > >> >> Such a value will be already used in [1]. I've looked at [2], but it > >> >> doesn't have such a field. > >> >> > >> >> [1] Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt > >> >> [2] Documentation/devicetree/bindings/input/touchscreen/touchscreen.txt > >> > > >> > I think the bigger question (since we are adding this to DT I assume it > >> > is not longer "toy" driver) can we convert it to a proper > >> > interrupt-driven scheme? Polling for real shipping devices is not > >> > efficient, I do not think we should even try to allow this mode in DTS > >> > bindings (note that tsc2007 use is different: it allows to specify > >> > sampling interval once touch is detected). > >> > >> Unfortunately, the only current user of tps6507x in kernel (da850-evm) > >> does not have the touch interrupt connected to the processor. So if we > >> dont support polled mode in when using device-tree, da850-evm.dts wont > >> be able to use this driver. > > > > Do we care? Right now the driver is unusable for any practical > > application I think as it burns CPU cycles by constantly polling the > > touch controller. > > > > Yegor, your conversion to device tree: is it done because you have a > > real product using this? > > Yes. It is VS-860 device http://www.visionsystems.de/produkte/vs-860.html > > It even seems to have interrupt pin connected. So we have to support both modes. Well, I'd rather supported interrupt only. Polling is a toy mode. Thanks. -- Dmitry