From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dmitry Torokhov Subject: Re: Linux USB HID should ignore values outside Logical Minimum/Maximum range Date: Mon, 24 Oct 2011 13:35:41 -0700 Message-ID: <20111024203541.GB31721@core.coreip.homeip.net> References: <4EA5913D.7090004@genband.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Denilson Figueiredo de =?iso-8859-1?Q?S=E1?= Cc: Chris Friesen , linux-kernel@vger.kernel.org, Jiri Kosina , linux-input@vger.kernel.org, linux-usb@vger.kernel.org List-Id: linux-input@vger.kernel.org On Mon, Oct 24, 2011 at 02:39:31PM -0200, Denilson Figueiredo de S=E1 w= rote: > On Mon, Oct 24, 2011 at 14:24, Chris Friesen wrote: > > On 10/22/2011 05:42 AM, Denilson Figueiredo de S=E1 wrote: > > > >> It may even happen to send an out-of-range value for one axis, but= a > >> valid value for another axis. The code should be prepared for that > >> (ignore one, but keep the other). > > > > In this case what should be used for the "invalid" axis value? =A0T= he previous > > value? >=20 > If that's an absolute pointing device, can we supply only one axis? I > mean, can we do this: "move the pointer to this X position, but leave > the Y position where it currently is". (note that the pointer might Yes, input protocol is stateful and it will actually suppress duplicate events itself. I.e. if device moves from (10, 20) to (40, 20) usespace will see: EV_ABS/ABS_X/10 EV_ABS/ABS_Y/20 EV_SYN/SYN_REPORT EV_ABS/ABS_X/40 EV_SYN/SYN_REPORT even if driver emits 2nd EV_ABS/ABS_Y/20 event. > have moved since last input from this device, due to other devices > also being present) The other devices have no effect on this device state. Thanks. --=20 Dmitry