From owner-freebsd-scsi Mon Sep 4 14: 4:30 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from front1m.grolier.fr (front1m.grolier.fr [195.36.216.51]) by hub.freebsd.org (Postfix) with ESMTP id 41EF937B42C for ; Mon, 4 Sep 2000 14:04:25 -0700 (PDT) Received: from nas5-244.vlt.club-internet.fr (nas5-244.vlt.club-internet.fr [194.158.107.244]) by front1m.grolier.fr (8.9.3/No_Relay+No_Spam_MGC990224) with ESMTP id XAA19729; Mon, 4 Sep 2000 23:03:40 +0200 (MET DST) Date: Mon, 4 Sep 2000 22:46:06 +0200 (CEST) From: =?ISO-8859-1?Q?G=E9rard_Roudier?= X-Sender: groudier@linux.local To: "Brandon D. Valentine" Cc: Chuck Robey , "Barrett G. Lyon" , David Bushong , freebsd-scsi@FreeBSD.ORG Subject: Re: AIC7892 / Adaptec 29160 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 4 Sep 2000, Brandon D. Valentine wrote: > On Mon, 4 Sep 2000, Chuck Robey wrote: >=20 > >I've been watching this thread *very* closely, because I've been unable = to > >get my new IBM ultra160 drive to boot using the 29160. I'd thought that > >maybe it was a thing that would be fixed by using an IBM-supplied utilit= y > >to zero the disk, but that's not been the case. I'm reluctantly ready t= o > >try another controller now. >=20 > This is quite frustrating to me. I've never been very impressed with > Tekram's offerings or Mylex's scsi-only(buslogic) range. I've always If by Tekram's offerings you mean SYM53C8XX PCI-SCSI I/O processors based controllers, then the comparison with what you call "Mylex scsi-only" could well be inappropriate. Both the SYM53C8XX and Adaptec aic7xxx have a phase engine. They are kind of half-clever PCI-SCSI I/O processors. They both allow very low-latency SCSI I/Os, but require to implement both the firmware and the software for a given O/S driver. You also must keep in mind that a PCI-SCSI host bus adapter is not only a PCI chip + a software driver. There are lots of other glues that can make difference in reliability and quality. It seems that many people want to buy expensive Adaptec based controllers but only may tolerate extra cheap Symbios based controllers. Btw, during the 5 years I hacked software for Symbios chips, I never discovered the actual reasons that made Symbios based controllers not deserve to be as expensive as Adaptec ones.;-) > bought Adaptec because they always seem to just work. I can't fathom > what could be so different about this card. It's still a 78xx series > chip. The same revision of the ahc driver runs flawlessly on my onboard > AIC7895s and my AHA2940U 32-bit PCI cards which are in most of my > equipment. This motherboard and the 29160 I've got are the first two > U160 products I've ever bought. I don't even have U160 drives hooked up > to them. This is with standard LVD 80 equipment I'm seeing these > problems. It can't be a problem with the recent addition of U160 speeds > then. I'm gonna try some current snaps tomorrow and see if they help. > If not, would the ahc hackers tell me what info they need from the > machine to debug the problem? If this drags out too long I may have to > just return the machine and look for something with a different 78xx > revision but before I do I'd at least like to get the developers some > good info. Let me know. The ahc driver has the best possible maintainer. IMO, you should relax and wait for his reply. G=E9rard. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message