From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rafi Rubin Subject: Re: [PATCH] hid-ntrig.c Multitouch cleanup and fix Date: Tue, 09 Mar 2010 17:27:13 -0500 Message-ID: <4B96CB41.9040804@seas.upenn.edu> References: <4B75E14A.2050602@seas.upenn.edu> <1266027185-5311-1-git-send-email-rafi@seas.upenn.edu> <4B760D60.6020907@seas.upenn.edu> <4B96B740.7020208@euromail.se> <4B96BAF1.5070209@seas.upenn.edu> <4B96C4A8.3060005@euromail.se> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from stag.seas.upenn.edu ([158.130.70.79]:47973 "EHLO stag.seas.upenn.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752547Ab0CIW1Z (ORCPT ); Tue, 9 Mar 2010 17:27:25 -0500 In-Reply-To: <4B96C4A8.3060005@euromail.se> Sender: linux-input-owner@vger.kernel.org List-Id: linux-input@vger.kernel.org To: Henrik Rydberg Cc: Jiri Kosina , linux-input@vger.kernel.org, dmitry.torokhov@gmail.com, chatty@enac.fr On 03/09/2010 04:59 PM, Henrik Rydberg wrote: > Rafi Rubin wrote: >> Since you're considering protocol clarification, what's your opinion on >> splitting the multi-touch and single touch (possibly emulated) to >> separate input devices? > > The multi-touch protocol was designed to complement single-touch, not replace > it. The MT events were specifically designed to be ignored by current userland > applications, so there is no need to separate them into different devices. > > On the contrary, the MT protocol is dependent on one of the single-touch events > BTN_TOUCH or ABS_PRESSURE to be present in the event stream for proper device > identification and for correct handling of zero-finger events. The latter > because at least one value (BTN_TOUCH or ABS_PRESSURE) must change state to pass > through input filtering when no fingers are present. This is what needs > clarification. > >> Also BTN_0 or BTN_TOOL_DOUBLETAP for single touch (or both)? It seems >> to me Wacom uses doubletap, whereas my hardware's rdesc defaults to btn0 >> and evdev prefers btn0 as well. > > Single touch should be goverened by one of BTN_TOUCH or ABS_PRESSURE. > BTN_TOOL_DOUBLETAP is a two-finger event. That definitely makes sense to me. I really don't know why wacom uses double tap there. Is there any harm in letting through the BTN_0 events? Is the "rdesc" mapping defined by the hardware or hid? For my device touch contacts look like this: Digitizers.TipSwitch ---> Key.Touch Digitizers.InRange ---> Key.ToolPen Digitizers.Confidence ---> Key.Btn0 Digitizers.ContactID ---> Sync.Report GenericDesktop.X ---> Absolute.MTPositionX GenericDesktop.Y ---> Absolute.MTPositionY Digitizers.Width ---> Absolute.MTMajor Digitizers.Height ---> Absolute.MTMinor ff00.0002 ---> Sync.Report ff00.0002 ---> Sync.Report ff00.0002 ---> Sync.Report ff00.0002 ---> Sync.Report Hence some of my confusion about setting both Touch and another button (the ToolPen bit seemed less pertinent). Rafi