From owner-freebsd-current Fri Feb 28 7: 6:29 2003 Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 29A8837B405 for ; Fri, 28 Feb 2003 07:06:28 -0800 (PST) Received: from grassmarket.ucs.ed.ac.uk (grassmarket.ucs.ed.ac.uk [129.215.166.64]) by mx1.FreeBSD.org (Postfix) with ESMTP id 56EC843F93 for ; Fri, 28 Feb 2003 07:06:26 -0800 (PST) (envelope-from bruce@cran.org.uk) Received: from fourtytwo.brucec.backnet (12266209.resnet.ed.ac.uk [10.6.0.100] (may be forged)) by grassmarket.ucs.ed.ac.uk (8.11.6/8.11.6) with ESMTP id h1SF6Eh19245; Fri, 28 Feb 2003 15:06:15 GMT Received: from fourtytwo.brucec.backnet (localhost.fourtytwo.brucec.backnet [127.0.0.1]) by fourtytwo.brucec.backnet (8.12.6/8.12.6) with ESMTP id h1SF6Fb1002893; Fri, 28 Feb 2003 15:06:15 GMT (envelope-from bruce@fourtytwo.brucec.backnet) Received: (from bruce@localhost) by fourtytwo.brucec.backnet (8.12.6/8.12.6/Submit) id h1SF6F9g002892; Fri, 28 Feb 2003 15:06:15 GMT Date: Fri, 28 Feb 2003 15:06:15 +0000 From: Bruce Cran To: Shizuka Kudo Cc: current@freebsd.org Subject: Re: High % interrupt when building world Message-ID: <20030228150615.GA2816@fourtytwo.brucec.backnet> References: <20030228122151.55902.qmail@web11408.mail.yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20030228122151.55902.qmail@web11408.mail.yahoo.com> User-Agent: Mutt/1.5.3i 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 On Fri, Feb 28, 2003 at 04:21:51AM -0800, Shizuka Kudo wrote: > I recently cvsup-ped -current (I believe it was 27 Feb), and noticed that the system was not > responsive as before when building world. Looking at the output of "top" showed a 30-60% interrupt > usage. I believed it is related to my highpoint 370 card running at raid 1. I detached the cards > and attached one of the drives to ata0 (Intel i815EP) and ran buildworld again. The % shown was > much less (less than 10%). Has Anyone else seen such an issue? > > I also reverted the src/sys to Feb 24 UTC which was just before sos' commits on the new API. The > problem also went away and the "top" % interrupt was at around 1-2% range. > I've also seen very poor performance on my Promise FastTrak TX2 'Lite' controller on my A7V333 motherboard. I get 3-5MB/sec, and copying files takes a very long time. When I move it back to the 'normal' ata controller the speed goes up to the more normal 40-45MB/sec. Both controllers detect and use ATA100. I'm running 5.0-RELEASE-p3. Have the recent ata patches maybe sorted this? >dd if=/dev/zero of=myfile bs=1024k count=100 100+0 records in 100+0 records out 104857600 bytes transferred in 31.741188 secs (3303518 bytes/sec) dmesg output: atapci0: port 0xb400-0xb40f,0xb800-0xb803,0xd000-0xd007,0xd400-0xd403,0xd800-0xd807 mem 0xd4800000-0xd4803fff irq 6 at device 6.0 on pci0 atapci1: port 0x9400-0x940f at device 17.1 on pci0 ad0: 76219 [155061/16/63] at ata0-master UDMA100 (using motherboard IDE) ar0: 58644MB [7476/255/63] status: READY subdisks (using FastTrak controller) Bruce Cran To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message