[tulip] Always require second ifdown/ifup after reboot

Adam Kessel adam@bostoncoop.net
Sun Feb 16 20:52:02 2003


--YZ5djTAD1cGYuMQK
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sun, Feb 16, 2003 at 11:25:23AM -0500, Donald Becker wrote:
> > 00:11.0 Ethernet controller: Lite-On Communications Inc LNE100TX [Links=
ys EtherFast 10/100] (rev 25)
> > tulip is compiled into the kernel (Debian unstable/linux 2.4.20). =20
> > Oddly, every time I reboot, I need to ifdown and ifup to get the card to
> > work.
> What driver version are you using?

tulip.c says:

#define DRV_VERSION     "0.9.15-pre12"
#define DRV_RELDATE     "Aug 9, 2002"

Interestingly, I don't think this problem was occurring with the 2.2
kernel, which used "v0.91g-ppc 7/16/99".

(in both cases I'm using the Debian stock source, which in this case I
don't think differs from vanilla kernel source).

> What is the detection message?

syslog reports:

eth0: Autonegotiation failed, using 10baseT, link beat status 10cc.

Also, I frequently see:

NETDEV WATCHDOG: eth0: transmit timed out
eth0: PNIC2 transmit timed out, status e4260000, CSR6/7 e0402002 / effffbff=
 CSR12 000000c8, resetting...

=2E...

eth0: PNIC2 transmit timed out, status e4260000, CSR6/7 e0402002 / effffbff=
 CSR12 000000c8, resetting...
eth0: Out-of-sync dirty pointer, 19 vs. 38.

and so on.

Sometimes it seems to me more frequent than others, and switches back and
forth between 100baseT and 10 baseT.
---
Adam Kessel (adam@bostoncoop.net)

--YZ5djTAD1cGYuMQK
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+T+r/dTf3ZklQ6qYRAiENAJ9ficqDF3JyhHYx/J6tp+w6d2YujgCgq7vK
3sBwBrzBwAfanlWqPUrCGgs=
=HVbI
-----END PGP SIGNATURE-----

--YZ5djTAD1cGYuMQK--