From owner-freebsd-current@FreeBSD.ORG Sat Nov 24 20:18:04 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 ED53016A41A for ; Sat, 24 Nov 2007 20:18:04 +0000 (UTC) (envelope-from mm@FreeBSD.org) Received: from mail.vx.sk (neo.vx.sk [213.239.219.9]) by mx1.freebsd.org (Postfix) with ESMTP id AB76813C458 for ; Sat, 24 Nov 2007 20:18:04 +0000 (UTC) (envelope-from mm@FreeBSD.org) Received: from localhost (localhost [127.0.0.1]) by mail.vx.sk (Postfix) with ESMTP id DFE053986E for ; Sat, 24 Nov 2007 21:18:03 +0100 (CET) X-Virus-Scanned: amavisd-new at mail.vx.sk Received: from mail.vx.sk ([127.0.0.1]) by localhost (mail.vx.sk [127.0.0.1]) (amavisd-new, port 10024) with LMTP id hwjuKmgH+pre for ; Sat, 24 Nov 2007 21:18:01 +0100 (CET) Received: from [10.2.0.1] (gw.radiolan.sk [193.93.72.6]) by mail.vx.sk (Postfix) with ESMTP id 3C2E339839 for ; Sat, 24 Nov 2007 21:18:01 +0100 (CET) Message-ID: <4748870B.2000907@FreeBSD.org> Date: Sat, 24 Nov 2007 21:18:19 +0100 From: Martin Matuska User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; sk; rv:1.8.1.9) Gecko/20071031 Thunderbird/2.0.0.9 Mnenhy/0.7.5.0 MIME-Version: 1.0 To: freebsd-current@FreeBSD.org References: <47480574.5030106@FreeBSD.org> <474883A9.4000502@FreeBSD.org> In-Reply-To: <474883A9.4000502@FreeBSD.org> X-Enigmail-Version: 0.95.3 Content-Type: text/plain; charset=ISO-8859-2 Content-Transfer-Encoding: 8bit Cc: Subject: Re: 7.0-BETA3 (amd64): interrupt storm on atapci0 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: Sat, 24 Nov 2007 20:18:05 -0000 Sorry, the "rate" of "vmstat -i" is undocumented. I have read the vmstat.c file and found out that "rate" it is the uptime average so it has no effect. I will do some stress testing with dd if=/dev/mirror/... of=/dev/null (this takes an interrupt rate of about 14000 interrupts/second). Martin Matuska wrote / napísal(a): > I have done some more investigation and have found out the following: > > - after doing a fresh restart, the "rate" reported in "vmstat -i" does > not decrease. It keeps increasing slowly, when there is larger interrupt > activity on atapci0 (irq22) > > - "systat -vmstat" reports zero interrupts, and "vmstat -i" reports this > increasing rate > > - when utilizing the disks at 100% (copying large files) the interrupt > rate is about 2300, which is truly high, but this correct number is > shown only in "systat -vmstat" > >