From ap@tentakulum.bnc.net  Fri Dec 17 13:58:26 1999
Return-Path: <ap@tentakulum.bnc.net>
Received: from tentakulum.bnc.net (e1-10M.tentakulum.bnc.net [195.247.233.241])
	by hub.freebsd.org (Postfix) with ESMTP id A720E14CEF
	for <FreeBSD-gnats-submit@freebsd.org>; Fri, 17 Dec 1999 13:58:24 -0800 (PST)
	(envelope-from ap@tentakulum.bnc.net)
Received: (from ap@localhost)
	by tentakulum.bnc.net (8.9.3/8.9.3) id WAA39976;
	Fri, 17 Dec 1999 22:56:39 +0100 (CET)
	(envelope-from ap)
Message-Id: <199912172156.WAA39976@tentakulum.bnc.net>
Date: Fri, 17 Dec 1999 22:56:39 +0100 (CET)
From: Achim Patzner <ap@tentakulum.bnc.net>
Reply-To: ap@bnc.net
To: FreeBSD-gnats-submit@freebsd.org
Subject: problems with the de driver
X-Send-Pr-Version: 3.2

>Number:         15542
>Category:       kern
>Synopsis:       [de] de(4) suddenly stops working
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    freebsd-bugs
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Fri Dec 17 14:00:02 PST 1999
>Closed-Date:    Sun Jan 27 08:46:34 UTC 2008
>Last-Modified:  Sun Jan 27 08:46:34 UTC 2008
>Originator:     Achim Patzner
>Release:        FreeBSD 4.0-CURRENT i386
>Organization:
>Environment:

-current of CTM 4142
kernel containing de0
two generic DEC based cards

>Description:

A short time after booting I'm getting something like
de1: abnormal interrupt: transmit underflow (raising TX threshold to 96|256)
Dec 17 22:27:21 duo /kernel: de1: abnormal interrupt: transmit underflow (raising TX threshold to 96|256)
Dec 17 22:27:21 duo /kernel: de1: abnormal interrupt: transmit underflow (raising TX threshold to 96|256)
Dec 17 22:29:50 duo /kernel: nfs server www:/httpd/htdocs: not responding

which is followed by one or both of the devices not generating interrupts
any more (at least systat tells me so). THis is even worse if both
interfaces got the same IRQ.

>How-To-Repeat:

Put two DEC based interfaces into one machine and wait a short time
	

>Fix:
	
	


>Release-Note:
>Audit-Trail:
Responsible-Changed-From-To: gnats-admin->freebsd-bugs 
Responsible-Changed-By: steve 
Responsible-Changed-When: Fri Dec 17 14:05:53 PST 1999 
Responsible-Changed-Why:  
Misfiled PR. 
State-Changed-From-To: open->feedback 
State-Changed-By: mike 
State-Changed-When: Sat Jul 21 11:23:25 PDT 2001 
State-Changed-Why:  

Does this problem still occur in newer versions of FreeBSD, 
such as 4.3-RELEASE? 

http://www.FreeBSD.org/cgi/query-pr.cgi?pr=15542 
State-Changed-From-To: feedback->closed 
State-Changed-By: mike 
State-Changed-When: Sat Jul 21 12:18:28 PDT 2001 
State-Changed-Why:  

E-mail sent to originator bounces. 

http://www.FreeBSD.org/cgi/query-pr.cgi?pr=15542 
State-Changed-From-To: closed->open 
State-Changed-By: tg 
State-Changed-When: Wed Apr 10 05:26:25 PDT 2002 
State-Changed-Why:  
Re-open, the problem still exists. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=15542 
State-Changed-From-To: open->feedback 
State-Changed-By: ceri 
State-Changed-When: Wed Nov 3 21:55:33 GMT 2004 
State-Changed-Why:  
de(4) has been through some changes since the PR was submitted; does 
this persist in 5.x or later 4.x releases? 

http://www.freebsd.org/cgi/query-pr.cgi?pr=15542 

From: Michael Glasgow <glasgow@beer.net>
To: bug-followup@FreeBSD.org, ap@bnc.net
Cc:  
Subject: Re: kern/15542: de suddenly stops working
Date: Mon, 20 Jun 2005 13:33:27 -0500 (CDT)

 I'm seeing similar behavior on 5.1-RELEASE-p7, with a quad-port
 (ZNYX ZX34X 21140A [10-100Mb/s] pass 2.2) using the de driver.
 
 For me, it's not repeatable, but I'll see the occasional message
 like this:
 
 May  2 13:59:06 hostname kernel: de1: abnormal interrupt: transmit underflow (raising TX threshold to 96|256)
 May  4 20:49:32 hostname kernel: de3: abnormal interrupt: transmit underflow (raising TX threshold to 96|256)
 May  5 00:12:27 hostname kernel: de1: abnormal interrupt: transmit underflow (raising TX threshold to 128|512)
 May  5 01:08:24 hostname kernel: de2: abnormal interrupt: transmit underflow (raising TX threshold to 96|256)
 
 And only occasionally (rarely) does the NIC lock up and stop sending
 or receiving packets.  Often there is some time between the last of
 these messages and when the NIC stops working, for example the last
 lockup happened approx 17-Jun-05 22:00, but nothing was logged after
 the message on May 5 above.
State-Changed-From-To: feedback->closed 
State-Changed-By: linimon 
State-Changed-When: Sun Jan 27 08:45:06 UTC 2008 
State-Changed-Why:  
Feedback timeout. 

This PR has been sitting around for quite some time, and the driver has 
undergone a lot of changes since then.  If this is still an issue, let 
us know and we can reopen it. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=15542 
>Unformatted:
