From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265227AbTLFSvW (ORCPT ); Sat, 6 Dec 2003 13:51:22 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265228AbTLFSvW (ORCPT ); Sat, 6 Dec 2003 13:51:22 -0500 Received: from moutng.kundenserver.de ([212.227.126.188]:41679 "EHLO moutng.kundenserver.de") by vger.kernel.org with ESMTP id S265227AbTLFSvU (ORCPT ); Sat, 6 Dec 2003 13:51:20 -0500 Date: Sat, 6 Dec 2003 19:49:01 +0100 From: Henning Meier-Geinitz To: Tom Rini Cc: Kernel Mailing List , linux-usb-users@lists.sourceforge.net Subject: Re: [Linux-usb-users] USB scanner issue (Was: Re: Beaver in Detox!) Message-ID: <20031206184901.GH2455@meier-geinitz.de> Mail-Followup-To: Tom Rini , Kernel Mailing List , linux-usb-users@lists.sourceforge.net References: <20031128182625.GP2541@stop.crashing.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20031128182625.GP2541@stop.crashing.org> User-Agent: Mutt/1.5.4i X-Provags-ID: kundenserver.de abuse@kundenserver.de auth:ef7bc16c10d3556bf134993d9edc1e24 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi, On Fri, Nov 28, 2003 at 11:26:25AM -0700, Tom Rini wrote: > On Wed, Nov 26, 2003 at 12:55:00PM -0800, Linus Torvalds wrote: > > [snip] > > I give you "Beaver in Detox", aka linux-2.6.0-test11. This is mainly > > brought on by the fact that the old aic7xxx driver was broken in -test10, > > and Ingo found this really evil test program that showed an error case in > > do_fork() that we had never handled right. Well, duh! > > I've found an odd problem that's in at least 2.6.0-test11. Did it happen in older versions of Linux? I haven't heard of any similar bug reports until now. > I've > reproduced this twice now with an Epson 1240 USB scanner > (0x04b8/0x010b). What happens is if I run xsane from gimp, acquire a > preview, start to scan and then cancel, the scanner becomes > unresponsive. This may be a bug in the plustek backend which supports that scanner. The cancel handling is pretty complicated with some scanners. > If I try and quit xsane, it gets stuck. That sometimes happens with other scanners, too. E.g. some don't like to get a "stop scan" command in certain situations. They just hang after that command and further commands run into the USB timeout. > Unplugging / > replugging and then trying to kill xsane locked the machine up hard. Well, obviously that should't happen. But it's really hard to debug without any oops. There was a bug in the scanner driver that occured when a device was open, then unplugged and then an application wrote to it. But it's fixed since some time. Does the freeze also happen if no other USB devices are attached? I guess working without a keyboard is not that easy but it may be worth a test. Otherwise I don't have any ideas. Bye, Henning