From owner-freebsd-current@FreeBSD.ORG Sun Sep 26 17:07:13 2004 Return-Path: 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 5C58816A4CE for ; Sun, 26 Sep 2004 17:07:13 +0000 (GMT) Received: from avocado.salatschuessel.net (avocado.salatschuessel.net [80.86.187.41]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2F8EA43D2D for ; Sun, 26 Sep 2004 17:07:12 +0000 (GMT) (envelope-from lehmann@ans-netz.de) Received: (qmail 55119 invoked from network); 26 Sep 2004 17:06:43 -0000 Received: from unknown (HELO kartoffel.salatschuessel.net) (80.86.187.43) by avocado.salatschuessel.net with SMTP; 26 Sep 2004 17:06:43 -0000 Date: Sun, 26 Sep 2004 19:07:09 +0200 From: Oliver Lehmann To: freebsd-current@freebsd.org Message-Id: <20040926190709.251f1689.lehmann@ans-netz.de> X-Mailer: Sylpheed version 0.9.12 (GTK+ 1.2.10; i386-portbld-freebsd4.10) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: Problems with ANA-6944A + 5.3-BETA + IRQ sharing? X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 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: Sun, 26 Sep 2004 17:07:13 -0000 Hi, I've a strange thing here, and would know If others are getting the same "results" - FreeBSD 5.3-BETA4 - Adaptec ANA 6944A (4xde) - Compaq NC6134 (2xfxp) I'm pinging from the system to a system connected through a switch to de2: root@gurke /root> ping nudel PING nudel.salatschuessel.net (10.0.0.21): 56 data bytes 64 bytes from 10.0.0.21: icmp_seq=0 ttl=64 time=0.687 ms 64 bytes from 10.0.0.21: icmp_seq=1 ttl=64 time=0.575 ms 64 bytes from 10.0.0.21: icmp_seq=2 ttl=64 time=0.629 ms 64 bytes from 10.0.0.21: icmp_seq=3 ttl=64 time=0.565 ms 64 bytes from 10.0.0.21: icmp_seq=4 ttl=64 time=0.577 ms it looks "ok" Now mounting a nfs share, and accessing (opening with vi) a file on it: 64 bytes from 10.0.0.21: icmp_seq=19 ttl=64 time=0.697 ms 64 bytes from 10.0.0.21: icmp_seq=20 ttl=64 time=0.574 ms 64 bytes from 10.0.0.21: icmp_seq=21 ttl=64 time=0.807 ms 64 bytes from 10.0.0.21: icmp_seq=22 ttl=64 time=253.249 ms 64 bytes from 10.0.0.21: icmp_seq=23 ttl=64 time=1.554 ms 64 bytes from 10.0.0.21: icmp_seq=24 ttl=64 time=513.415 ms 64 bytes from 10.0.0.21: icmp_seq=25 ttl=64 time=1110.697 ms 64 bytes from 10.0.0.21: icmp_seq=26 ttl=64 time=1011.975 ms 64 bytes from 10.0.0.21: icmp_seq=27 ttl=64 time=1112.534 ms 64 bytes from 10.0.0.21: icmp_seq=28 ttl=64 time=101.289 ms 64 bytes from 10.0.0.21: icmp_seq=29 ttl=64 time=1110.638 ms 64 bytes from 10.0.0.21: icmp_seq=30 ttl=64 time=1010.899 ms 64 bytes from 10.0.0.21: icmp_seq=31 ttl=64 time=1015.072 ms 64 bytes from 10.0.0.21: icmp_seq=32 ttl=64 time=1010.512 ms 64 bytes from 10.0.0.21: icmp_seq=33 ttl=64 time=1013.426 ms 64 bytes from 10.0.0.21: icmp_seq=34 ttl=64 time=1010.722 ms 64 bytes from 10.0.0.21: icmp_seq=35 ttl=64 time=1419.586 ms 64 bytes from 10.0.0.21: icmp_seq=36 ttl=64 time=446.113 ms 64 bytes from 10.0.0.21: icmp_seq=37 ttl=64 time=4038.742 ms 64 bytes from 10.0.0.21: icmp_seq=38 ttl=64 time=4039.750 ms 64 bytes from 10.0.0.21: icmp_seq=39 ttl=64 time=4041.047 ms 64 bytes from 10.0.0.21: icmp_seq=40 ttl=64 time=4041.045 ms 64 bytes from 10.0.0.21: icmp_seq=41 ttl=64 time=4045.762 ms 64 bytes from 10.0.0.21: icmp_seq=42 ttl=64 time=4041.257 ms 64 bytes from 10.0.0.21: icmp_seq=43 ttl=64 time=4041.038 ms wtf??? The file is closed, and I don't access the nfs mount any more... but the ping times are not comming down... 4000-5000 ms. I tried full duplex, and half - no difference. I don't have that problem with my fxp cards. Here are the irq's: root@gurke /root> dmesg | grep irq mlx0: port 0xec00-0xec7f mem 0xffafff80-0xffafffff irq 10 at device 17.0 on pci0 de0: port 0xcc00-0xcc7f mem 0xff4fff80-0xff4fffff irq 9 at device 4.0 on pci2 de1: port 0xc880-0xc8ff mem 0xff4fff00-0xff4fff7f irq 11 at device 5.0 on pci2 de2: port 0xc800-0xc87f mem 0xff4ffe80-0xff4ffeff irq 15 at device 6.0 on pci2 de3: port 0xc480-0xc4ff mem 0xff4ffe00-0xff4ffe7f irq 10 at device 7.0 on pci2 fxp0: port 0xdf00-0xdf3f mem 0xff800000-0xff8fffff,0xff9ff000-0xff9fffff irq 11 at device 4.0 on pci3 fxp1: port 0xde80-0xdebf mem 0xff700000-0xff7fffff,0xff9fe000-0xff9fefff irq 15 at device 5.0 on pci3 ppc0: at port 0x378-0x37f irq 7 on isa0 sio0 at port 0x3f8-0x3ff irq 4 flags 0x10 on isa0 root@gurke /root> It doesn't seem to happen with a cable attached to de0. So I guess it's a problem with irq sharing? I have no experience with that card and a FreeBSD version prior to 5.3-BETA4 Any ideas what to configure to get each de interface an own irq? I have enough free (3,5,9,12) 10 for mlx, and 15 for fxp0 and fxp1. In my BIOS I can only configure an irq per slot. and hint.de.[0-3].irq in device.hints doesn't work... (I don't expected it to work because it is just a "hint"...) -- Oliver Lehmann http://www.pofo.de/ http://wishlist.ans-netz.de/