[tulip] tulip for presario 1700 with connexant chip breaks under load

[Jochen Schieck]

Hello Donald,

On Fri, 15 Mar 2002, Donald Becker wrote:

> On Fri, 15 Mar 2002, Jochen Schieck wrote:
>
> > I tried to get the NIC in my Presario 1700xl365 notebook to work. I took
> > your tulip (v0.93 11/7/2001) drivers and compiled them using makefile against
> > 2.4.12-ac3. I am using the mandrake 8.0 distribution (with updated kernel).
> >
> > It seems to work fine at the beginning:
> >
> > $ ./tulip-diag
> > tulip-diag.c:v2.10 3/08/2002 Donald Becker (becker@scyld.com)
> >  http://www.scyld.com/diag/index.html
> > Index #1: Found a Conexant LANfinity adapter at 0x1400.
> >  Port selection is MII, half-duplex.
> >  Transmit started, Receive started, half-duplex.
> >   The Rx process state is 'Waiting for packets'.
> >   The Tx process state is 'Idle'.
>
> Looks great.
> What does 'mii-diag' report?  Is the half duplex setting correct?

mii-diag reports (the same working and not working network):

$ ./mii-diag
Using the default interface 'eth0'.
Basic registers of MII PHY #1:  1000 782d 0022 1720 01e1 0080 0004 2001.
 Basic mode control register 0x1000: Auto-negotiation enabled.
 You have link beat, and everything is working OK.
 Your link partner is generating 100baseTx link beat  (no
autonegotiation).
   End of basic transceiver information.

Our network here supports Full-Duplex. However, I do not know about the
NIC chip in my notebook. Do you know more?

> > However, when starting a large download (like getting a new kernel tar
> > file) the driver quickly hangs:
> >
> > $ ./tulip-diag
> > tulip-diag.c:v2.10 3/08/2002 Donald Becker (becker@scyld.com)
> >  http://www.scyld.com/diag/index.html
> > Index #1: Found a Conexant LANfinity adapter at 0x1400.
> >  Port selection is MII, half-duplex.
> >  Transmit started, Receive started, half-duplex.
> >   The Rx process state is 'Waiting for packets'.
> >   The Tx process state is 'Waiting for Tx to finish'.
>
> OK, something has gone wrong.  This Tx state usually means that the
> transceiver selection is incorrect.  But it might also mean that the
> chip itself has hung.
>
> Are there any error messges in the kernel message log?
>
The only message I get after loading the tulip module with debug=1 is:

Mar 15 18:50:01 lapschieck kernel: Trying to free nonexistent resource <cc847000-cc84707f>
Mar 15 18:54:29 lapschieck kernel: tulip.c:v0.93 11/7/2001  Written by Donald Becker <becker@scyld.com>
Mar 15 18:54:29 lapschieck kernel:   http://www.scyld.com/network/tulip.html
Mar 15 18:54:30 lapschieck kernel: eth0: Conexant LANfinity rev 8 at 0xcc847000, 00:50:8B:AA:B4:6A, IRQ 9.
Mar 15 18:54:30 lapschieck kernel: eth0:  MII transceiver #1 config 1000 status 7809 advertising 01e1.
Mar 15 18:54:30 lapschieck kernel: eth0:  MII transceiver #0 config 1000 status 7809 advertising 01e1.

> > If I restart the network with '/etc/init.d/network restart' it works fine
> > again (until the next download....).
>
> OK, that's promising: at least the driver is able to reset the chip.
> But we have to detect that it has hung.

Do you need more information? What should/can I do??

Thanks

Jochen