Smarter way of adding alias entries to modules.conf ?

Amit_Bhutani@Dell.com Amit_Bhutani@Dell.com
Thu Nov 13 22:16:00 2003


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_01C3AA65.FBBFFF90
Content-Type: text/plain; 
 charset=us-ascii
Content-Transfer-Encoding: 7bit

Gary, 

As per our previous conversation, currently (0.45.01) DKMS  has a "feature"
where an entry for a module is written to /etc/modules.conf even if the
device for that module does not exist causing some degree of confusion.

For example: If I install, say an e1000 dkms rpm on a system that does not
have an Intel Gig-E NIC, I get a false alias entry for a ethX device for
e1000 in /etc/modules.conf where X is the smallest available number. So say
I have eth0 and eth1 entries before I install the rpm, an entry for eth3 is
getting added with alias e1000 although there is no device for that module.

Now the obvious question is "why" am I installing a RPM when I do not have
it's equivalent device and is that an unsupported action. The answer to that
is that sometimes drivers have to be pre-installed in un-interactive
environments like Factory Install regardless of what devices exist.

I know we discussed on adding a dkms.conf variable, say DEVICE_MATCH which
will contain a *unique* string that DKMS will look for (via lspci or in
/proc or somewhere reliable) and if it finds that and *only* if it finds
that would it add entries to the /etc/modules.conf. 

Anybody have suggestions on better resolving this issue ?

Thanks,
- Amit B



------_=_NextPart_001_01C3AA65.FBBFFF90
Content-Type: text/html; 
 charset=us-ascii
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 =
5.5.2655.35">
<TITLE>Smarter way of adding alias entries to modules.conf ?</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2 FACE=3D"Arial">Gary, </FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">As per our previous conversation, =
currently (0.45.01) DKMS&nbsp; has a &quot;feature&quot; where an entry =
for a module is written to /etc/modules.conf even if the device for =
that module does not exist causing some degree of confusion.</FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Arial">For example: If I install, say an =
e1000 dkms rpm on a system that does not have an Intel Gig-E NIC, I get =
a false alias entry for a ethX device for e1000 in /etc/modules.conf =
where X is the smallest available number. So say I have eth0 and eth1 =
entries before I install the rpm, an entry for eth3 is getting added =
with alias e1000 although there is no device for that =
module.</FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Arial">Now the obvious question is =
&quot;why&quot; am I installing a RPM when I do not have it's =
equivalent device and is that an unsupported action. The answer to that =
is that sometimes drivers have to be pre-installed in un-interactive =
environments like Factory Install regardless of what devices =
exist.</FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Arial">I know we discussed on adding a =
dkms.conf variable, say DEVICE_MATCH which will contain a *unique* =
string that DKMS will look for (via lspci or in /proc or somewhere =
reliable) and if it finds that and *only* if it finds that would it add =
entries to the /etc/modules.conf. </FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Arial">Anybody have suggestions on better =
resolving this issue ?</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">Thanks,</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Arial">-</FONT> <FONT SIZE=3D2 =
FACE=3D"Arial">Amit</FONT><FONT SIZE=3D2 FACE=3D"Arial"> B</FONT>
</P>
<BR>

</BODY>
</HTML>
------_=_NextPart_001_01C3AA65.FBBFFF90--