From owner-freebsd-sun4v@FreeBSD.ORG Thu Sep 4 00:04:19 2008 Return-Path: Delivered-To: freebsd-sun4v@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 9D7E810663B6 for ; Thu, 4 Sep 2008 00:04:19 +0000 (UTC) (envelope-from kris@FreeBSD.org) Received: from [127.0.0.1] (freefall.freebsd.org [IPv6:2001:4f8:fff6::28]) by mx1.freebsd.org (Postfix) with ESMTP id 91A128FC1E; Thu, 4 Sep 2008 00:04:18 +0000 (UTC) (envelope-from kris@FreeBSD.org) Message-ID: <48BF2601.7000307@FreeBSD.org> Date: Thu, 04 Sep 2008 09:34:17 +0930 From: Kris Kennaway User-Agent: Thunderbird 2.0.0.16 (Macintosh/20080707) MIME-Version: 1.0 To: Sevan / Venture37 References: <20080830082702.GL86609@server.vk2pj.dyndns.org> <48B90B68.80802@FreeBSD.org> <3c1674c90808301418n4fa0691fo30d61cc3e6595862@mail.gmail.com> In-Reply-To: Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 8bit Cc: freebsd-sun4v@freebsd.org, Matthew Macy Subject: Re: sun4v status X-BeenThere: freebsd-sun4v@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting FreeBSD to the sun4v List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 04 Sep 2008 00:04:19 -0000 Sevan / Venture37 wrote: > the system would flip out when there was even the slightest load on the disk with mpt0: queue full event messages, > tweaking the system settings with camcontrol reduced the frequency of this but resulted in dataloss when it did occur > > http://www.zulustips.com/2007/09/06/mpt0-queue-full-event-on-dell-sas-5ir.html#more-53 > _________________________________________________________________ > Win New York holidays with Kellogg’s & Live Search > http://clk.atdmt.com/UKM/go/107571440/direct/01/ > There were bugs in the interrupt handling on sun4v that were introduced at some point but I thought were fixed again. I don't remember if it was before or after this post from a year ago. Anyway there are likely to be various other issues that crop up once/if the port becomes minimally usable, so it shouldn't be seen as only needing a single bugfix. Kris