From owner-freebsd-current@FreeBSD.ORG Wed Aug 26 22:29:50 2009 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 43D23106568D for ; Wed, 26 Aug 2009 22:29:50 +0000 (UTC) (envelope-from lists@mawer.org) Received: from mail-px0-f180.google.com (mail-px0-f180.google.com [209.85.216.180]) by mx1.freebsd.org (Postfix) with ESMTP id 273EF8FC2C for ; Wed, 26 Aug 2009 22:29:50 +0000 (UTC) Received: by pxi10 with SMTP id 10so616677pxi.28 for ; Wed, 26 Aug 2009 15:29:49 -0700 (PDT) MIME-Version: 1.0 Received: by 10.140.143.14 with SMTP id q14mr4023223rvd.15.1251325789834; Wed, 26 Aug 2009 15:29:49 -0700 (PDT) In-Reply-To: <20090826195407.GW2829@hoeg.nl> References: <4A9571EB.7090209@fsu.edu> <20090826195407.GW2829@hoeg.nl> Date: Thu, 27 Aug 2009 08:29:49 +1000 Message-ID: From: Antony Mawer To: Ed Schouten Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: freebsd-current@freebsd.org, Nathan Lay Subject: Re: FreeBSD 8 BETA3 unusable (kbd/syscons?) X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 26 Aug 2009 22:29:50 -0000 On Thu, Aug 27, 2009 at 5:54 AM, Ed Schouten wrote: > Hi Nathan, > > Would it be a lot of work to figure out which SVN revision of HEAD > introduced this regression? Even if you can't get a specific revision > number, reducing it to a week's span in revision numbers would be very > helpful. > > People also reported issues to me where the first keypresses after the > system has booted are discarded. I have yet to be convinced this is a > TTY issue, not the keyboard code, interrupt handling, etc. I saw an issue yesterday with 8.0-BETA3/i386 where a new install on a system used as a router would "hang" when attempting to go to multi-user mode. When I say hang -- the system appeared to be passing packets OK, but the console was non responsive. Breaking into kgdb showed that it appeared to be somewhere in the kbdmux/usb keyboard code. That particular system was an IBM xSeries 300 with a PS2 keyboard (no USB devices attached, though it has 2 USB ports). I rebooted into the BIOS and checked what USB options were configured ... I had the following (from memory): USB Support ................... Enabled USB Keyboard Support .... Disabled USB Mouse Support ........ Disabled I changed USB Keyboard Support to Enabled, rebooted, and the system then booted into multi-user fine. Not sure if this was related but thought I'd mention it in case it was relevant. If it would be beneficial I can disable the option again and get a full backtrace ... --Antony