From owner-freebsd-current@FreeBSD.ORG Wed Oct 10 18:54:17 2007 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 580AF16A420 for ; Wed, 10 Oct 2007 18:54:17 +0000 (UTC) (envelope-from jdc@parodius.com) Received: from mx01.sc1.parodius.com (mx01.sc1.parodius.com [72.20.106.3]) by mx1.freebsd.org (Postfix) with ESMTP id 39A0313C4C3 for ; Wed, 10 Oct 2007 18:54:16 +0000 (UTC) (envelope-from jdc@parodius.com) Received: by mx01.sc1.parodius.com (Postfix, from userid 1000) id 9D90C1CC05F; Wed, 10 Oct 2007 11:54:16 -0700 (PDT) Date: Wed, 10 Oct 2007 11:54:16 -0700 From: Jeremy Chadwick To: Christopher Arnold Message-ID: <20071010185416.GA38522@eos.sc1.parodius.com> Mail-Followup-To: Christopher Arnold , freebsd-current@freebsd.org References: <20071010195157.O24133@chrishome.localnet> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20071010195157.O24133@chrishome.localnet> User-Agent: Mutt/1.5.16 (2007-06-09) Cc: freebsd-current@freebsd.org Subject: Re: Compaq Smart Array 5300 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, 10 Oct 2007 18:54:17 -0000 On Wed, Oct 10, 2007 at 08:06:21PM +0200, Christopher Arnold wrote: > > Hi all! > > anyone aware of issues with Compaq Smart Array 5300 on a current only a day > old? > > This is what i get in the boot message: > ciss0: port 0xe800-0xe8ff mem > 0xfebc0000-0xfebfffff,0xfea00000-0xfeafffff irq 22 at device 1.0 on pci2 > ciss0: can't allocate config window > device_attach: ciss0 attach returned 6 > > The full output can be found at: > http://www.arnold.se/boot.txt > http://www.arnold.se/pciconf.txt > > Im trying to run it in a Supermicro 6010 with 370DER+ motherboard. > > Anyone who has any input on what this could be? It's one of those bloody QLogic GEM chips; on this motherboard, it's probably the QLogic GEM 354. The motherboard manual does indicate this, although only in the Windows driver section (this does not surprise me, as some other vendors use GEM chips and don't tell you until you actually have to deal with them...) I've dealt with GEMs at work (not Supermicro products), and they do nothing but piss you off in the case there is any sort of SCSI failure. For example, a drive going bad and falling off the bus would normally result in the SCSI subsystem reporting some errors and the OS (hopefully) continuing to work w/out issue (barring if the main OS disk failed, of course :-) ). SCSI's pretty resilient about this sort of thing. The GEM is supposed to handle all of this "for" you in some bizarre manner, but I've never seen it work reliably -- ever. The GEM also can, somehow, manage SCSI IDs for you and do some sort of transparent SCSI ID remapping. This is all what I've seen first-hand. The GEM does have some sort of BIOS (on some mainboards you can hit a control character sequence prior to POST completion and go in and adjust GEM settings), but I don't know how it works. There's comments in the FreeBSD driver code about the GEM chips violating SCSI specification, by the way. If you want me to dig up the cvsweb URL for it, I can. Based on that, and my experiences at work, I'm not left feeling very good about those chips. In general: I wouldn't worry about the lack of ciss0 device. In fact, I'd go so far to recommend you remove ses and ciss from the kernel on that box, and hope for the best. Scott (Long) may have some better (and more accurate) advice on the matter; my experiences are my own. -- | Jeremy Chadwick jdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB |