From owner-freebsd-hackers Thu Jun 1 11:14:25 2000 Delivered-To: freebsd-hackers@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id F2FCA37B673; Thu, 1 Jun 2000 11:14:18 -0700 (PDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.9.3/8.9.3) with ESMTP id MAA51077; Thu, 1 Jun 2000 12:14:16 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id MAA30293; Thu, 1 Jun 2000 12:13:50 -0600 (MDT) Message-Id: <200006011813.MAA30293@harmony.village.org> To: Dennis Subject: Re: S5933 PCI Adapter..?? Cc: Mike Smith , Joy Ganguly , freebsd-hackers In-reply-to: Your message of "Thu, 01 Jun 2000 11:18:36 EDT." <200006011518.LAA02647@etinc.com> References: <200006011518.LAA02647@etinc.com> Date: Thu, 01 Jun 2000 12:13:50 -0600 From: Warner Losh Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <200006011518.LAA02647@etinc.com> Dennis writes: : Have you done bus performance testing with this card? Given the : architechture of the AMCC part, it seems highly improbable that you will be : able to get high throughput with such a card. Because the AMCC part : requires external logic it is impossible to do pass-through single cycle : bursts, which is required for efficient utilization of the PCI bus. Once : you begin holding off cycles the PCI bus totally pigs out (which is why : virtually all high-speed pci solutions are single-chip type designs). Yes. I've done drivers for several cards with this design. The AMCC part is very fussy and will often lock up the bus unless the card designer has put enough extra logic on the card to cope with the oddities of the card. Sadly, many don't. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message