From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Mon, 7 Oct 2002 09:10:53 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Mon, 7 Oct 2002 09:10:53 -0400 Received: from twilight.ucw.cz ([195.39.74.230]:62849 "EHLO twilight.ucw.cz") by vger.kernel.org with ESMTP id ; Mon, 7 Oct 2002 09:10:52 -0400 Date: Mon, 7 Oct 2002 15:16:18 +0200 From: Vojtech Pavlik To: Andrew Morton Cc: Vojtech Pavlik , lkml Subject: Re: input layer strangeness Message-ID: <20021007151618.A1560@ucw.cz> References: <3D9A74CF.8C8585E7@digeo.com> <20021002080952.B17477@ucw.cz> <3D9A8E34.7BC5E91A@digeo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <3D9A8E34.7BC5E91A@digeo.com>; from akpm@digeo.com on Tue, Oct 01, 2002 at 11:12:04PM -0700 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Oct 01, 2002 at 11:12:04PM -0700, Andrew Morton wrote: > Vojtech Pavlik wrote: > > > > On Tue, Oct 01, 2002 at 09:23:43PM -0700, Andrew Morton wrote: > > > > > > It's been doing this ever since the input layer changes: > > > > > > - open a few xterms > > > - press the spacebar, leave pressed > > > - start waggling the mouse about > > > - stop pressing spacebar, keep waggling the mouse about, > > > across the xterms > > > > > > The keystrokes *never* stop coming. Just the continuous mouse > > > activity causes a stream of keyboard input, at seemingly the normal > > > autorepeat rate. I can keep them coming for 30 seconds, just by > > > moving the mouse. > > > > Do they stop coming when you stop moving the mouse or they don't stop at > > all? The first would be quite interesting, the second would probably be > > a missed key release event due to keyboard controller overload by the > > mouse. > > They stop immediately if I stop moving the mouse. > > > > In practice, it's irritating because it's quite easy to get a > > > stream of erroneous input dumped into the wrong windows. > > > > > > It's a vanilla dual pentium with an AT keyboard and a PS/2 > > > mouse. > > > > Can you check if it happens also on UP? Just want to know if it might be > > a SMP issue ... > > Will do. Results? It also might be the keyboard controller in your machine starves the keyboard input pipe when the mouse is moving and flooding it with data. This way the release event from the keyboard doesn't come until you stop moving the mouse. Can you verify that with I8042_DEBUG_INPUT? Thanks. -- Vojtech Pavlik SuSE Labs