[tulip] Lynksys LNE100TX

Robert Ludwig RobertL@ekcinc.com
Mon Feb 11 10:44:01 2002


This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------ =_NextPart_001_01C1B311.F2202676
Content-Type: text/plain

	From: "Moti Haimovsky" <motih@cisco.com>
	To: <tulip@scyld.com>
	Date: Wed, 6 Feb 2002 16:54:26 -0800
	Subject: [tulip] RE: tulip digest, Vol 1 #472 - 5 msgs: Lynksys
LNE100TX Network Card (Robert Ludwig)

	Hi,
	Maybe you do have parity errors in your system, and that's why your
	ifconfig command reports the card's base address at 0xEC00 while
	everything else correctly reports it at 0xFC00 (one bit difference).

	If "Parity Error Response" bit (PCI command reg bit 6) is on ("1"),
	disable it (in the PCI command register of the chip) and see what
happens. (
	maybe the win95 driver does not turn this bit on, preventing the
chip
	from reporting these kind of problems).

	Motih@cisco.com

Thank you for your suggestion.  I attempted to implement it.  I'm not very
familiar with driver construction so I wasn't able to track down where the
initialization occurs.  Instead I attempted to reset the bit when the "PCI
Fatal Error" message is issued.  Based on the way the code acted, the
"Parity Error Response" was not set, yet the error continued to be posted.
I also looked at the "System Error Enable" bit (PCI Command reg bit 8).
However when I included this bit in my change, the device was unable to
access the rest of the network.  I currently have the tulip.c driver code
reverted to the its original form except the message is only issued if
debug>0.


Bob Ludwig
EKC, Inc.
10400 W. Higgins Rd.,  Suite 200
Rosemont, IL  60018
(847) 296-8010
(847) 296-8016 (Fax)
RobertL@EKCInc.com
www.EKCInc.com


------ =_NextPart_001_01C1B311.F2202676
Content-Type: text/html
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3DUS-ASCII">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.0.1460.9">
<TITLE>Lynksys LNE100TX</TITLE>
</HEAD>
<BODY>
<UL>
<P><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">From: &quot;Moti =
Haimovsky&quot; &lt;motih@cisco.com&gt;</FONT>
<BR><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">To: =
&lt;tulip@scyld.com&gt;</FONT>
<BR><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Date: Wed, 6 Feb =
2002 16:54:26 -0800</FONT>
<BR><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Subject: [tulip] =
RE: tulip digest, Vol 1 #472 - 5 msgs: Lynksys LNE100TX Network Card =
(Robert Ludwig)</FONT>
</P>

<P><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Hi,</FONT>
<BR><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Maybe you do have =
parity errors in your system, and that's why your</FONT>
<BR><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">ifconfig command =
reports the card's base address at 0xEC00 while</FONT>
<BR><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">everything else =
correctly reports it at 0xFC00 (one bit difference).</FONT>
</P>

<P><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">If &quot;Parity =
Error Response&quot; bit (PCI command reg bit 6) is on =
(&quot;1&quot;),</FONT>
<BR><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">disable it (in the =
PCI command register of the chip) and see what happens. (</FONT>
<BR><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">maybe the win95 =
driver does not turn this bit on, preventing the chip</FONT>
<BR><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">from reporting =
these kind of problems).</FONT>
</P>

<P><FONT COLOR=3D"#000000" SIZE=3D2 =
FACE=3D"Arial">Motih@cisco.com</FONT>
</P>
</UL>
<P><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Thank you for your =
suggestion.&nbsp; I attempted to implement it.&nbsp; I'm not very =
familiar with driver construction so I wasn't able to track down where =
the initialization occurs.&nbsp; Instead I attempted to reset the bit =
when the &quot;PCI Fatal Error&quot; message is issued.&nbsp; Based on =
the way the code acted, the &quot;Parity Error Response&quot; was not =
set, yet the error continued to be posted.&nbsp; I also looked at the =
&quot;System Error Enable&quot; bit (PCI Command reg bit 8).&nbsp; =
However when I included this bit in my change, the device was unable to =
access the rest of the network.&nbsp; I currently have the tulip.c =
driver code reverted to the its original form except the message is =
only issued if debug&gt;0.</FONT></P>
<BR>

<P><B><I><FONT SIZE=3D5 FACE=3D"Arial">Bob Ludwig</FONT></I></B><I></I>
<BR><B><FONT FACE=3D"Arial">EKC, Inc.</FONT></B>
<BR><FONT SIZE=3D2 FACE=3D"Arial">10400 W. Higgins Rd.,&nbsp; Suite =
200</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Arial">Rosemont, IL&nbsp; 60018</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Arial">(847) 296-8010</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Arial">(847) 296-8016 (Fax)</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Arial">RobertL@EKCInc.com</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Arial">www.EKCInc.com</FONT>
</P>

</BODY>
</HTML>
------ =_NextPart_001_01C1B311.F2202676--