From owner-freebsd-stable Mon Dec 23 12:35:19 2002 Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0C84337B401 for ; Mon, 23 Dec 2002 12:35:17 -0800 (PST) Received: from smtp1.sentex.ca (smtp1.sentex.ca [199.212.134.4]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2E31F43EF1 for ; Mon, 23 Dec 2002 12:35:16 -0800 (PST) (envelope-from mike@sentex.net) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by smtp1.sentex.ca (8.12.6/8.12.6) with ESMTP id gBNKZAJq040197; Mon, 23 Dec 2002 15:35:10 -0500 (EST) (envelope-from mike@sentex.net) Received: from simian.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.12.6/8.12.6) with ESMTP id gBNKagHY072396; Mon, 23 Dec 2002 15:36:43 -0500 (EST) (envelope-from mike@sentex.net) Message-Id: <5.2.0.9.0.20021223153544.03d80d30@marble.sentex.ca> X-Sender: mdtpop@marble.sentex.ca (Unverified) X-Mailer: QUALCOMM Windows Eudora Version 5.2.0.9 Date: Mon, 23 Dec 2002 15:37:55 -0500 To: Thomas Nystrom From: Mike Tancsa Subject: Re: Fix for hanging of vr interface (Rhine Ethernet) Cc: stable@FreeBSD.ORG In-Reply-To: <3E07644E.8635066C@saeab.se> References: <5.2.0.9.0.20021220151835.03a72a48@marble.sentex.ca> <5.2.0.9.0.20021223102240.07303e80@marble.sentex.ca> <5.2.0.9.0.20021223134745.04ec8038@marble.sentex.ca> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Virus-Scanned: By Sentex Communications (lava/20020517) Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG At 08:30 PM 23/12/2002 +0100, Thomas Nystrom wrote: > > >Did the interface work after the forced reset? > > > > Yes, it seemed to. Is there any extra debugging info I can provide ? > >No, we have to wait for Mike Silbersack to check this, he is the one who >have the ideas behind the forced reset operation.... BTW, it does seem a bit slower than most nics. Using netperf, I am getting Testing with the following command line: /usr/local/netperf/netperf -l 20 -H 192.168.43.218 -t TCP_STREAM -i 10,2 -I 99,3 -- -m 4096 -s 32768 -S 32768 TCP STREAM TEST to 192.168.43.218 : +/-1.5% @ 99% conf. : histogram Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 32768 32768 4096 20.01 86.51 ------------------------------------ Testing with the following command line: /usr/local/netperf/netperf -l 20 -H 192.168.43.218 -t TCP_STREAM -i 10,2 -I 99,3 -- -m 16384 -s 32768 -S 32768 TCP STREAM TEST to 192.168.43.218 : +/-1.5% @ 99% conf. : histogram Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 32768 32768 16384 20.01 86.51 ------------------------------------ Testing with the following command line: /usr/local/netperf/netperf -l 20 -H 192.168.43.218 -t TCP_STREAM -i 10,2 -I 99,3 -- -m 65536 -s 32768 -S 32768 TCP STREAM TEST to 192.168.43.218 : +/-1.5% @ 99% conf. : histogram Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 32768 32768 65536 20.01 86.50 This is with a P4 on the board. On a different box with an SIS or fxp nic, I am seeing in the 94-95 range for the above tests. ---Mike To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message