From owner-freebsd-net@FreeBSD.ORG Sun Jun 3 17:06:01 2012 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6B60B106564A; Sun, 3 Jun 2012 17:06:01 +0000 (UTC) (envelope-from bzeeb-lists@lists.zabbadoz.net) Received: from mx1.sbone.de (bird.sbone.de [46.4.1.90]) by mx1.freebsd.org (Postfix) with ESMTP id 09FEF8FC14; Sun, 3 Jun 2012 17:06:00 +0000 (UTC) Received: from mail.sbone.de (mail.sbone.de [IPv6:fde9:577b:c1a9:31::2013:587]) (using TLSv1 with cipher ADH-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by mx1.sbone.de (Postfix) with ESMTPS id 7B4A125D3A90; Sun, 3 Jun 2012 17:05:59 +0000 (UTC) Received: from content-filter.sbone.de (content-filter.sbone.de [IPv6:fde9:577b:c1a9:31::2013:2742]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mail.sbone.de (Postfix) with ESMTPS id 779F4BE751D; Sun, 3 Jun 2012 17:05:58 +0000 (UTC) X-Virus-Scanned: amavisd-new at sbone.de Received: from mail.sbone.de ([IPv6:fde9:577b:c1a9:31::2013:587]) by content-filter.sbone.de (content-filter.sbone.de [fde9:577b:c1a9:31::2013:2742]) (amavisd-new, port 10024) with ESMTP id NKWaG1YibvRz; Sun, 3 Jun 2012 17:05:57 +0000 (UTC) Received: from orange-en1.sbone.de (orange-en1.sbone.de [IPv6:fde9:577b:c1a9:31:cabc:c8ff:fecf:e8e3]) (using TLSv1 with cipher AES128-SHA (128/128 bits)) (No client certificate requested) by mail.sbone.de (Postfix) with ESMTPSA id B8790BE751C; Sun, 3 Jun 2012 17:05:56 +0000 (UTC) From: "Bjoern A. Zeeb" Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Date: Sun, 3 Jun 2012 17:05:54 +0000 Message-Id: <90B0A065-634A-47F5-A8B9-3D093865F48A@lists.zabbadoz.net> To: Jack F Vogel Mime-Version: 1.0 (Apple Message framework v1084) X-Mailer: Apple Mail (2.1084) Cc: FreeBSD Networking Mailing List Subject: ixgbe(4) intr and pps problems in at least HEAD X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 03 Jun 2012 17:06:01 -0000 Hey Jack or anyone on net@, having updated my sources of HEAD from Dec/Mar to something of last week I get a lot of: interrupt storm detected on "irq262:"; throttling interrupt source Given Intel's driver REDAME I found online says to update the sysctl to = this: hw.intr_storm_threshold=3D9000 I did that. I then quickly ended up at 25k which still wasn't enough so I added 2 more zeros to get my console unspammed. Any idea what the cause for this could be? The few input pps also seem = to keep it quite busy, even if only distributed to a single queue and what = netstat reports is far from ideal. The pps number it should be able to take should be way higher I feel and = I am sending it a load more pps anyway. # netstat -I ix1 -w 1 input (ix1) output packets errs idrops bytes packets errs bytes colls 114467 200089 0 23275412 113885 0 8458948 0 147135 206724 0 26074654 147010 0 10820024 0 176973 334783 0 37914940 176904 0 13117378 0 106443 274685 0 28386046 106781 0 8002984 0 35404 0 0 2492980 35236 0 2538270 0 172295 214024 0 28396162 171991 0 12616392 0 52870 201826 0 19069376 53563 0 4068360 0 182143 366926 0 40728536 181543 0 13530772 0 128465 231976 0 26673492 128992 0 9513468 0 26161 0 0 1841770 26204 0 1880830 0 115360 220160 0 24864314 115523 0 8562074 0 ^C 12 root -92 - 0K 432K CPU1 1 474:32 100.00% = intr{irq262: ix1:que } 0 root -92 0 0K 384K CPU0 0 226:12 100.00% = kernel{ix1 que} 12 root -92 - 0K 432K CPU3 3 214:36 99.66% = intr{irq264: ix1:que } 0 root -92 0 0K 384K - 2 464:29 98.29% kernel{ix1 = que} 11 root 155 ki31 0K 64K RUN 2 82.9H 1.56% idle{idle: = cpu2} 12 root -92 - 0K 432K WAIT 2 26:15 1.17% = intr{irq263: ix1:que } 11 root 155 ki31 0K 64K RUN 0 84.0H 0.00% idle{idle: = cpu0} 11 root 155 ki31 0K 64K RUN 3 82.4H 0.00% idle{idle: = cpu3} 11 root 155 ki31 0K 64K RUN 1 79.5H 0.00% idle{idle: = cpu1} 12 root -92 - 0K 432K RUN 0 26:34 0.00% = intr{irq261: ix1:que } 17 root 16 - 0K 16K syncer 2 26:25 0.00% syncer 0 root -92 0 0K 384K - 3 17:30 0.00% kernel{ix1 = que} 0 root -92 0 0K 384K RUN 2 16:09 0.00% kernel{em0 = que} 0 root -92 0 0K 384K - 0 15:54 0.00% kernel{ix1 = que} 12 root -60 - 0K 432K WAIT 1 10:38 0.00% intr{swi4: = clock} Is anyone seeing this as well or can reproduce it? It's a ix1@pci0:3:0:1: class=3D0x020000 card=3D0xa15f8086 chip=3D0x10c68086 = rev=3D0x01 hdr=3D0x00 vendor =3D 'Intel Corporation' device =3D '82598EB 10-Gigabit AF Dual Port Network Connection' class =3D network subclass =3D ethernet with only 1 port used currently. /bz --=20 Bjoern A. Zeeb You have to have visions! It does not matter how good you are. It matters what good you do!