From owner-freebsd-bugs@FreeBSD.ORG Sat Jan 5 16:10:03 2008 Return-Path: Delivered-To: freebsd-bugs@hub.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8B26B16A4A5 for ; Sat, 5 Jan 2008 16:10:03 +0000 (UTC) (envelope-from gnats@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:4f8:fff6::28]) by mx1.freebsd.org (Postfix) with ESMTP id 77AFA13C46E for ; Sat, 5 Jan 2008 16:10:03 +0000 (UTC) (envelope-from gnats@FreeBSD.org) Received: from freefall.freebsd.org (gnats@localhost [127.0.0.1]) by freefall.freebsd.org (8.14.2/8.14.2) with ESMTP id m05GA3nI037342 for ; Sat, 5 Jan 2008 16:10:03 GMT (envelope-from gnats@freefall.freebsd.org) Received: (from gnats@localhost) by freefall.freebsd.org (8.14.2/8.14.1/Submit) id m05GA32M037341; Sat, 5 Jan 2008 16:10:03 GMT (envelope-from gnats) Resent-Date: Sat, 5 Jan 2008 16:10:03 GMT Resent-Message-Id: <200801051610.m05GA32M037341@freefall.freebsd.org> Resent-From: FreeBSD-gnats-submit@FreeBSD.org (GNATS Filer) Resent-To: freebsd-bugs@FreeBSD.org Resent-Reply-To: FreeBSD-gnats-submit@FreeBSD.org, Laurent Frigault Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 378E816A47A for ; Sat, 5 Jan 2008 16:06:02 +0000 (UTC) (envelope-from nobody@FreeBSD.org) Received: from www.freebsd.org (www.freebsd.org [IPv6:2001:4f8:fff6::21]) by mx1.freebsd.org (Postfix) with ESMTP id 2AA4913C46B for ; Sat, 5 Jan 2008 16:06:02 +0000 (UTC) (envelope-from nobody@FreeBSD.org) Received: from www.freebsd.org (localhost [127.0.0.1]) by www.freebsd.org (8.14.2/8.14.2) with ESMTP id m05G55DJ074905 for ; Sat, 5 Jan 2008 16:05:05 GMT (envelope-from nobody@www.freebsd.org) Received: (from nobody@localhost) by www.freebsd.org (8.14.2/8.14.1/Submit) id m05G55f6074903; Sat, 5 Jan 2008 16:05:05 GMT (envelope-from nobody) Message-Id: <200801051605.m05G55f6074903@www.freebsd.org> Date: Sat, 5 Jan 2008 16:05:05 GMT From: Laurent Frigault To: freebsd-gnats-submit@FreeBSD.org X-Send-Pr-Version: www-3.1 Cc: Subject: kern/119361: bge(4) transmit performance problem X-BeenThere: freebsd-bugs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Bug reports List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 05 Jan 2008 16:10:03 -0000 >Number: 119361 >Category: kern >Synopsis: bge(4) transmit performance problem >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sat Jan 05 16:10:03 UTC 2008 >Closed-Date: >Last-Modified: >Originator: Laurent Frigault >Release: 7.0BETA4 >Organization: >Environment: FreeBSD bge.free.org 7.0-BETA4 FreeBSD 7.0-BETA4 #0: Wed Dec 12 12:31:02 CET 2007 root@bge.free.org:/usr/src/sys/amd64/compile/GENERIC amd64 >Description: We are experiencing a problem with BCM5721 bge interfaces, which seems to be able to receive at almost 1Gbps but can only transmit at < 540Mbps. Our test server is as Dell PE860 CPU: Intel(R) Xeon(R) CPU X3220 @ 2.40GHz (2400.10-MHz K8-class CPU) Origin = "GenuineIntel" Id = 0x6fb Stepping = 11 Features=0xbfebfbff Features2=0xe3bd AMD Features=0x20100800 AMD Features2=0x1 Cores per package: 4 usable memory = 8577101824 (8179 MB) avail memory = 8297656320 (7913 MB) bge0@pci0:4:0:0: class=0x020000 card=0x01e61028 chip=0x165914e4 rev=0x11 hdr=0x00 vendor = 'Broadcom Corporation' device = 'BCM5721 NetXtreme Gigabit Ethernet PCI Express' class = network subclass = ethernet bge1@pci0:5:0:0: class=0x020000 card=0x01e61028 chip=0x165914e4 rev=0x11 hdr=0x00 vendor = 'Broadcom Corporation' device = 'BCM5721 NetXtreme Gigabit Ethernet PCI Express' class = network subclass = ethernet We ran the following test with FreeBSD 6.3-PRERELEASE/i386 , FreeBSD 7.0BETA4/amd64 (ULE and BSD, with and without POLLING). # dd if=/dev/zero bs=1m count=102400 | nc peer 15000 no firewall on the server. netstat -i show no errors interface forced 1000baseTX duplex-full on both server and switch, no autoneg. The server as 2 bge interfaces, and both have the same problem. Peer was running 6.3-PRERELEASE/amd64 with an em interface on peer we run : nc -l 15000 >/dev/null If we boot our test server with a linux ubuntu 6.0.6 and run the same test, we can transmit at more than 900 Mbps so it is not an hardware problem. But we would like to stay away from linux on our servers. Is there any patch/fix available for this driver ? Unfortunately our server has no more slot available so we can't forget the bge interfaces and add an em interface. I can run tests on a spare server if needed (same hardware). >How-To-Repeat: dd if=/dev/zero bs=1m count=102400 | nc peer 15000 with nc -l 15000 >/dev/null on a peer server >Fix: >Release-Note: >Audit-Trail: >Unformatted: