82559 Intel Chip and linux eepro100 driver

kellyo@us.ibm.com kellyo@us.ibm.com
Thu Jun 24 12:42:08 1999



Donald,
     Also, I ran the eepro100-diag util and am seeing the following.  The first
one, which is failing as far as the eeprom readout goes, is the IBM 10/100
EtherJet PCI Management Adapter that has the 82559 chip.  The second one is a
true Intel EtherExpress100 and is working fine.  Both cards are functioning as
far as sending/receiving data, but the first one is not showing a correct MAC
address.  It shows the same value for every box that it is installed in.

eepro100-diag.c:v0.07 2/25/98 Donald Becker (becker@cesdis.gsfc.nasa.gov)
Index #1: Found a Intel 82557 EtherExpressPro100B adapter at 0x5000.
EEPROM contents:
  8100 8000 9ca6 8405 8000 8000 8000 8000
  8000 8000 8000 8000 804a 8000 8000 8000
  8000 8000 8000 8000 8000 8000 8000 8000
  8000 8000 8000 8000 8000 8000 8000 8000
  8000 8000 8000 8000 8000 8000 8000 8000
  8000 8000 8000 8000 8000 8000 8000 8000
  8000 8000 8000 8000 8000 8000 8000 8000
  8000 8000 8000 8000 8000 8000 8000 8000
 The EEPROM checksum (should be 0xbaba) is 0x21f5.
Intel EtherExpress Pro 10/100 EEPROM contents:
  Station address 00:81:00:80:A6:9C.
  Receiver lock-up bug exists. (The driver work-around *is* implemented.)
  Board assembly 800080-000, Physical connectors present:
  Primary interface chip None PHY #-1.

Index #2: Found a Intel 82557 EtherExpressPro100B adapter at 0x5040.
EEPROM contents:
  9000 1727 d0b9 0000 0000 0201 4701 0000
  6896 6104 4581 0009 8086 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 6a2e
 The EEPROM checksum (should be 0xbaba) is 0xbaba.
Intel EtherExpress Pro 10/100 EEPROM contents:
  Station address 00:90:27:17:B9:D0.
  Receiver lock-up bug exists. (The driver work-around *is* implemented.)
  Board assembly 689661-004, Physical connectors present: RJ45
  Primary interface chip i82555 PHY #1.


Thanks,
     Kelly O.


Kelly O'Brien    -   AT&T Global Network Services
Spectrum 3446    Tampa, Florida
Office: 813-878-3817  (t/l 427)
Lotus Notes ID:    kellyo@IBMUS
Internet:                  kellyo@us.ibm.com
VM:                          kellyo at ibmusm37

===========================================================================
                  "Don't let your knowledge be limited by what you already
know." - me


---------------------- Forwarded by Kelly O'Brien/Tampa/IBM on 06/23/99 05:56 PM
---------------------------


Kelly O'Brien
06/23/99 03:02 PM

To:   Donald Becker <becker@cesdis.gsfc.nasa.gov>
cc:
From: Kelly O'Brien/Tampa/IBM@IBMUS
Subject:  Re: 82559 Intel Chip and linux eepro100 driver  (Document link not
      converted)

Donald,
     I tried the new device driver as well and am still getting the following:

-----------------------------------------
Jun 23 14:36:00 SGUSTMPQA kernel:   The PCI BIOS has not enabled this device!
Updating PCI command 0103->0107.
Jun 23 14:36:00 SGUSTMPQA kernel: eth0: Invalid EEPROM checksum 0x21f5, check
settings before activating this device!
Jun 23 14:36:00 SGUSTMPQA kernel: eth0: Intel PCI EtherExpress Pro100 at 0x7c40,
00:81:00:80:A6:9C, IRQ 11.
Jun 23 14:36:00 SGUSTMPQA kernel:   Receiver lock-up bug exists -- enabling
work-around.
Jun 23 14:36:00 SGUSTMPQA kernel:   Board assembly 800080-000, Physical
connectors present:
Jun 23 14:36:00 SGUSTMPQA kernel:   Primary interface chip None PHY #0.
Jun 23 14:36:00 SGUSTMPQA kernel:   Forcing 100Mbs full-duplex operation.
Jun 23 14:36:00 SGUSTMPQA kernel:   General self-test: passed.
Jun 23 14:36:00 SGUSTMPQA kernel:   Serial sub-system self-test: passed.
Jun 23 14:36:00 SGUSTMPQA kernel:   Internal registers self-test: passed.
Jun 23 14:36:00 SGUSTMPQA kernel:   ROM checksum self-test: passed (0x04f4518b).
Jun 23 14:36:00 SGUSTMPQA kernel:   Receiver lock-up workaround activated.
---------------------------------------------------

I checked the driver code and it appears that it is still checking for a
checksum of 0xBABA.  Is this checksum different for an 82559 than others?  I
have another Intel card in it, and it is coming up with the correct mac addr,
etc.  Following is it's output:

----------------------------------
Jun 23 14:36:00 SGUSTMPQA kernel:   The PCI BIOS has not enabled this device!
Updating PCI command 0103->0107.
Jun 23 14:36:00 SGUSTMPQA kernel: eth1: Intel PCI EtherExpress Pro100 at 0x7c20,
00:90:27:17:B9:D2, IRQ 9.
Jun 23 14:36:00 SGUSTMPQA kernel:   Board assembly 689661-004, Physical
connectors present: RJ45
Jun 23 14:36:00 SGUSTMPQA kernel:   Primary interface chip i82555 PHY #1.
Jun 23 14:36:00 SGUSTMPQA kernel:   Forcing 100Mbs full-duplex operation.
Jun 23 14:36:00 SGUSTMPQA kernel:   General self-test: passed.
Jun 23 14:36:00 SGUSTMPQA kernel:   Serial sub-system self-test: passed.
Jun 23 14:36:00 SGUSTMPQA kernel:   Internal registers self-test: passed.
Jun 23 14:36:00 SGUSTMPQA kernel:   ROM checksum self-test: passed (0x24c9f043).
Jun 23 14:36:00 SGUSTMPQA kernel:   Receiver lock-up workaround activated.
---------------------------------

In the first one, the Primary interface chip is showing up as "NONE".  Are there
any #defines that I have to set that may be affecting this?

Thanks,
     Kelly O.

Kelly O'Brien    -   AT&T Global Network Services
Spectrum 3446    Tampa, Florida
Office: 813-878-3817  (t/l 427)
Lotus Notes ID:    kellyo@IBMUS
Internet:                  kellyo@us.ibm.com
VM:                          kellyo at ibmusm37

===========================================================================
                  "Don't let your knowledge be limited by what you already
know." - me




Donald Becker <becker@cesdis.gsfc.nasa.gov> on 06/23/99 02:19:02 PM

To:   Kelly O'Brien/Tampa/IBM@IBMUS
cc:   linux-eepro100@cesdis.gsfc.nasa.gov
Subject:  Re: 82559 Intel Chip and linux eepro100 driver





On Wed, 23 Jun 1999 kellyo@us.ibm.com wrote:
> I am getting the following error when using an IBM (actually Intel) 10/100
> EtherJet PCI Management adapter with the eepro100 linux driver.  Do you know
of
> any updated drivers for the 82559 chips?  The card actually operates, but the
> mac address is being misread and is showing up as the same value for every
card
> in the system, and in others that use the same card.
...
> Jun  7 20:04:27 SGATT1 kernel: eth0: Invalid EEPROM checksum 0x21f5, check
setti
> ngs before activating this device!

One of the changes for the '559 was increasing the EEPROM size from 64 to
256 words.  You'll need an updated driver to correctly read the larger
EEPROM.

http://cesdis.gsfc.nasa.gov/linux/drivers/eepro100.html
ftp://cesdis.gsfc.nasa.gov/pub/linux/drivers/eepro100.c

Donald Becker                        becker@cesdis.gsfc.nasa.gov
USRA-CESDIS, Center of Excellence in Space Data and Information Sciences.
Code 930.5, Goddard Space Flight Center,  Greenbelt, MD.  20771
301-286-0882        http://cesdis.gsfc.nasa.gov/people/becker/whoiam.html