[tulip-bug] PCI fatal bus error

Donald Becker becker@scyld.com
Thu Dec 6 15:05:01 2001


On 6 Dec 2001, Glenn Ammons wrote:

> I am using version 0.93 of the Tulip driver with a LinkSys EtherFast
> 10/100 LAN Card (model LNE100TX, version 4) to communicate with a
> cable modem at about 700Kbps (half of a T1, whatever that is).  My
> kernel version is 2.2.15.  Everything seems to work fine, except that
> I get these error messages:
> 
>   eth0: PCI Fatal Bus Error, fc67e050
> 
> The address is always the same.

This isn't an address.  It's the status bits from the chip.  The
tulip-diag program will interpret the bits, or you can read the databook.

fc0.....   Parity error
fc8.....   Master abort
fd0.....   Target abort
fd8.....   Reserved
fe......   Reserved
ff......   Reserved

It looks as if you are getting parity errors.
  Are you overclocking the system?
  Are the system voltages correct?
  Do you have five or more devices on the bus?

> The error messages occur frequently, averaging about one per second.

That's pretty frequently.  You have A Problem.

> The machine is an old Gateway P90.  The documentation for the card
> says that it requires a PCI 2.1 compliant motherboard and BIOS, and
> that the card should be inserted in a bus mastering PCI slot.  I have
> no idea if my motherboard and BIOS are up to snuff or if the slot is
> bus mastering (how would I tell?).

If the board works at all, it's in a bus master slot.  However some
motherboards cheat and share bus master signals -- that will cause
similar problems.

Donald Becker				becker@scyld.com
Scyld Computing Corporation		http://www.scyld.com
410 Severn Ave. Suite 210		Second Generation Beowulf Clusters
Annapolis MD 21403			410-990-9993