From owner-freebsd-questions Mon Jan 20 2:27:40 2003 Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BF65837B401 for ; Mon, 20 Jan 2003 02:27:39 -0800 (PST) Received: from cswiger-sec.homeip.net (pool-129-44-40-186.ny325.east.verizon.net [129.44.40.186]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9B26743F3F for ; Mon, 20 Jan 2003 02:27:38 -0800 (PST) (envelope-from cswiger@mac.com) Received: from mac.com (prime.local [192.168.1.3]) by sec.local (8.12.6/8.12.6) with ESMTP id h0JHvIwv020816 for ; Sun, 19 Jan 2003 12:57:18 -0500 (EST) (envelope-from cswiger@mac.com) Message-ID: <3E2AFA09.8090404@mac.com> Date: Sun, 19 Jan 2003 14:18:33 -0500 From: Chuck Swiger Organization: The Courts of Chaos User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.0.2) Gecko/20021120 Netscape/7.01 X-Accept-Language: en-us, en MIME-Version: 1.0 To: FreeBSD questions mailing list Subject: Re: TX underrun References: <20030119173530.3905B48463@wastegate.net> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, hits=-2.0 required=8.0 tests=NOSPAM_INC,QUOTED_EMAIL_TEXT,REFERENCES,SPAM_PHRASE_00_01, USER_AGENT,USER_AGENT_MOZILLA_UA,X_ACCEPT_LANG version=2.43 Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Doug Reynolds wrote: > On Sat, 18 Jan 2003 03:17:01 +0100 (CET), Marc Schneiders wrote: [ ... ] >>xl1: tx underrun, increasing tx start threshold to 180 bytes >>xl1: transmission error: 90 >>xl1: tx underrun, increasing tx start threshold to 240 bytes > > does that mean it is increasing or decreasing? Good ethernet cards can attempt to transmit a packet before the local system has actually written all of the data to them. If the local system fails to produce the data quickly enough, though, the card experiences a buffer underflow...which means about the same as it does when you burn a CD; it has to abort and resend the packet. The driver is smart enough to notice this, and will try to compensate by buffering more data before sending the packet out (ie, the "start threshold"), so that the machine has more time to finish generating the packet. This is also a situation where device polling might be quite helpful. -Chuck To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message