From nobody  Tue Apr 28 05:29:59 1998
Received: (from nobody@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id FAA08320;
          Tue, 28 Apr 1998 05:29:59 -0700 (PDT)
          (envelope-from nobody)
Message-Id: <199804281229.FAA08320@hub.freebsd.org>
Date: Tue, 28 Apr 1998 05:29:59 -0700 (PDT)
From: czito@nyx.net
To: freebsd-gnats-submit@freebsd.org
Subject: Artisoft "Noderunner" (NE2000) ethernet card not identified correctly 
X-Send-Pr-Version: www-1.0

>Number:         6439
>Category:       kern
>Synopsis:       Artisoft "Noderunner" (NE2000) ethernet card not identified correctly
>Confidential:   no
>Severity:       serious
>Priority:       low
>Responsible:    freebsd-bugs
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Tue Apr 28 05:40:01 PDT 1998
>Closed-Date:    Wed Apr 29 07:33:52 PDT 1998
>Last-Modified:  Wed Apr 29 15:20:04 PDT 1998
>Originator:     Cory Zito
>Release:        2.2.6-RELEASE
>Organization:
>Environment:
FreeBSD isogenic.ml.org 2.2.6-RELEASE FreeBSD 2.2.6-RELEASE #6: Tue Apr 28 06:43:02 CDT 1998     czito@isogenic.ml.org:/usr/src/sys/compile/ISOGENIC  i386

>Description:
NodeRunner NE2000 ethernet cards are incorrectly identified as NE1000
cards using the tests in if_ed.c.  The cards will work as NE1000, but 
the ethernet mac address is incorectly reported causing problems both
for the users, if more than one card reports the same address, and for
administrators (dhcp for instance will not work without a correct 
mac address)
>How-To-Repeat:
All NodeRunner cards tried had this problem, and others on the net
have expressed this problem
>Fix:
Patchfile that works around problem, by disabling the NE1000 option 
included (i.e. hacked file) I suggest improving the tests between 
NE1000 and NE2000, or at least allowing the force 16bit option to work
for these cards#This patch was created 4/28/98 by Cory Zito, czito@nyx.net
#It's purpose it to kill the NE1000/NE2000 test forcing NE2000 to win
#This is needed for cards like ArtiSoft's NodeRunner, an NE2000 card
#that passes the NE1000 tests.
#
#Note: use patch -p0 < filename when feeding in this patch

Index: /usr/src/sys/i386/isa/if_ed.c
Prereq: 1.107.2.6 1997/11/02 07:15:08 davidg Exp 


1139a1140,1152
> 
>         /* Modified 4/28/98 By Cory Zito for Artisoft Card
>          * The Noderunner Card is an NE2000, but the NE1000 tests
>          * decide it is an NE1000.  Problem solved by commenting out
>          * two (2) areas the test itself, and the if statement giving
>          * the option between NE1000 and NE2000.  This is a quick hack
>          * that I did so DHCP would word since the ethernet address was
>          * getting hosed by chosing NE1000.  This patch is for 2.2.6, but
>          * at least from 2.2.5 where I did this first, the commented lines
>          * didn't change. Hopefully in the future the NE1000/NE2000
>          * test will be fixed and this patch will not be needed. */
> 
> 
1143,1144c1156,1158
<       if (bcmp(test_pattern, test_buffer, sizeof(test_pattern))) {
<               /* not an NE1000 - try NE2000 */
---
> /* Modified 4/28/98 by Cory Zito
>         if (bcmp(test_pattern, test_buffer, sizeof(test_pattern))) { */
>                 /* not an NE1000 - try NE2000 */
1164,1167c1178,1181
<       } else {
<               sc->type = ED_TYPE_NE1000;
<               sc->type_str = "NE1000";
<       }
---
> /* Modified 4/28/98 by Cory Zito        } else {
>                 sc->type = ED_TYPE_NE1000;
>                 sc->type_str = "NE1000";
>         } Thats it, your NE2000 Noderunner should work correctly now*/



>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->closed 
State-Changed-By: phk 
State-Changed-When: Wed Apr 29 07:33:52 PDT 1998 
State-Changed-Why:  
Thanks for the report, but the proposed lobotomy cannot be used 
in general. 

From: Cory Zito <czito@nyx.net>
To: freebsd-gnats-submit@freebsd.org, czito@nyx.net
Cc:  Subject: Re: kern/6439: Artisoft "Noderunner" (NE2000) ethernet card not identified correctly
Date: Wed, 29 Apr 1998 17:17:08 -0500

 I appologise for not being clear enough in stating that the included
 patch/workaround was suitable only for people that have the problem, I
 understand what this is doing and why it can't be integrated
 into the code (in fact I would cringe at integrating anything that takes
 out support) the ne1000 cards
 are still worth supporting and it was never my intention for this patch
 to be integtrated into the code tree only that at least those searching
 active bug reports could find at least temporary solution until the ne
 driver is fixed.
 
 I do not understand however why this bug report was moved to the closed
 status as it is a problem
 that should be addressed with the NE drivers and can cause major
 problems if it is not.
 
 
>Unformatted:
