"Claude LeFrancois (LMC)" wrote:
>
> Thanks for the info. I can run the script manually to get the NIC on the
> network. But, by the mean time before a permanent fix, would it be a good
> idea to apply the change I did to allow at least correct initilization for
> eth0 ?
Hi, Claude.
Your suggestion could be a useful kludge if we can guarantee that
Cardbus machines only every have a single interface (possibly
reasonable).
An alternative and similarly tasteless kludge is to hardwire eth0 into
the pcmcia scripts. There is a description of how to do this at
http://www.uow.edu.au/~andrewm/linux/3c59x-2.3-changelog.txt (of all
places).
It says:
- Kernel 2.3.99 has broken the PCMCIA tools. When a card is
inserted the 'cardmgr' daemon is no longer told what network
interface it has been bound to. Until this is sorted, the
kludgearound I am using is to alter the PCMCIA script
/etc/pcmcia/shared:
--- shared Tue Nov 23 14:47:50 1999
+++ /tmp/shared Mon Apr 17 00:21:28 2000
@@ -118,11 +118,13 @@
while read SOCKET CLASS DRIVER INSTANCE DEV MAJOR MINOR ; do
if [ "$1" = "$DEV" ] ; then return 0 ; fi
done
+return 0
return 1
}
get_info ()
{
+ echo get_info: $1
if [ -d /var/state/pcmcia ] ; then
SCHEME=`cat /var/state/pcmcia/scheme`
STAB=/var/state/pcmcia/stab
@@ -141,3 +143,4 @@
if [ $# -lt 2 ] ; then usage ; fi
ACTION=$1 ; DEVICE=$2
+DEVICE=eth0
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Sep 07 2000 - 21:00:31 EST