From owner-freebsd-current Mon Dec 3 14:15: 6 2001 Delivered-To: freebsd-current@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by hub.freebsd.org (Postfix) with ESMTP id A408937B417; Mon, 3 Dec 2001 14:15:04 -0800 (PST) Received: (from dillon@localhost) by apollo.backplane.com (8.11.6/8.9.1) id fB3MEst98950; Mon, 3 Dec 2001 14:14:54 -0800 (PST) (envelope-from dillon) Date: Mon, 3 Dec 2001 14:14:54 -0800 (PST) From: Matthew Dillon Message-Id: <200112032214.fB3MEst98950@apollo.backplane.com> To: Søren Schmidt Cc: nuzrin@goose.net.my, Miklos Niedermayer , Greg Lehey , current@FreeBSD.ORG Subject: Re: HEADSUP ATA support for newer SiS chipsets added References: <200112031058.fB3AwDT04544@freebsd.dk> Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG :Hmm, I've just played around a bit, it seems we are hit by interrupt :latency or something, if you limit the transfer to 128k, which allows :the ATA controller to fetch it in one go, you will see the expected :transfer rates. Now I dont see this on PCI based controllers, and that :hints that the problem could be the fact that the two onboard controllers :sits on irq 14 & 15 making them the lowest priority devices in the system, :and that could cause the interrupt latency I'm seeing which then again :causes the bad transfer rates on transfers that need to transfer more :that one transaction full of data (ie max 128k). : :-Søren The larger transfers are probably choking the IDE drive's pipelining capabilities. That's my guess, anyway. I avoid IDE like the plague. -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message