From owner-freebsd-usb@FreeBSD.ORG Tue Jan 15 00:14:38 2008 Return-Path: Delivered-To: freebsd-usb@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 42AB416A417; Tue, 15 Jan 2008 00:14:38 +0000 (UTC) (envelope-from wblock@wonkity.com) Received: from wonkity.com (wonkity.com [67.158.26.137]) by mx1.freebsd.org (Postfix) with ESMTP id EA8C913C442; Tue, 15 Jan 2008 00:14:37 +0000 (UTC) (envelope-from wblock@wonkity.com) Received: from wonkity.com (localhost [127.0.0.1]) by wonkity.com (8.14.2/8.14.2) with ESMTP id m0F0EOdb015550; Mon, 14 Jan 2008 17:14:25 -0700 (MST) (envelope-from wblock@wonkity.com) Received: from localhost (wblock@localhost) by wonkity.com (8.14.2/8.14.2/Submit) with ESMTP id m0F0EOaa015547; Mon, 14 Jan 2008 17:14:24 -0700 (MST) (envelope-from wblock@wonkity.com) Date: Mon, 14 Jan 2008 17:14:24 -0700 (MST) From: Warren Block To: "M. Warner Losh" In-Reply-To: <20080114.163836.58456912.imp@bsdimp.com> Message-ID: <20080114170114.E84534@wonkity.com> References: <200801141747.37052.hselasky@c2i.net> <20080114.120058.74704237.imp@bsdimp.com> <20080114162333.A26271@wonkity.com> <20080114.163836.58456912.imp@bsdimp.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-3.0 (wonkity.com [127.0.0.1]); Mon, 14 Jan 2008 17:14:25 -0700 (MST) Cc: usb@freebsd.org, freebsd-usb@freebsd.org Subject: Re: umass troubleshooting FAQ X-BeenThere: freebsd-usb@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: FreeBSD support for USB List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 15 Jan 2008 00:14:38 -0000 On Mon, 14 Jan 2008, M. Warner Losh wrote: > In message: <20080114162333.A26271@wonkity.com> > Warren Block writes: > : On Mon, 14 Jan 2008, Warner Losh wrote: > : > > : > This item is about telling people how to fix N well known problems > : > with devices that need quirks. > : > > : > Eg, if your new device sees a 'No Sync Supported' message, then do > : > XYZ. > : > > : > If you see stalls in with these messages, do ABC. > : > > : > If you see a stall with those messages, do DEF. > : > : Is there a test suite that could detect and report these situations? > > Well, ideally one would code defensively if one could detect the > problems. Many of them are 'and then the device locks up solid' kind > of failure modes :-( Once you've touched the problem, you lose. How about a stress test? Something like a benchmark (bonnie?) that pounds the device conclusively, then prints a summary of uname along with USB software and hardware versions. If it doesn't have errors, doesn't lock up, and the summary prints, that device doesn't need quirks and can be added to the list. Ideally that would have a provision for copying results to a searchable web database. (Might there be something already out there?) Add/modify quirks until the device works acceptably or smashing it with a hammer becomes the more attractive alternative. -Warren Block * Rapid City, South Dakota USA