Intel E1000/MT driver won't load with 2.4.20 (smp) kernel?

Intel E1000/MT driver won't load with 2.4.20 (smp) kernel?

Post by Steve Wample » Thu, 02 Dec 2004 05:58:28



I'm trying to replace a 100bT NIC (Intel PRO/100) with a gigabit
card (Intel PRO/1000MT).  The driver (both the stock version
[5.0.43-k1] and Intel's latest/greatest [5.5.4] refuse to
load because of an EEPROM checksum error (checksum is
expected to be BABA, but is FFC0 instead).  This happens with
two different 1000/MT cards (both purchased same time/place)...

Has anyone else seen this?  Got a fix?  It this really likely
to be *two* bad EEPROMs?

I did some googling and found one suggestion to simply
disable the checksum validation in the driver.  That nicely
removes the above error - but the driver still won't load.

 
 
 

Intel E1000/MT driver won't load with 2.4.20 (smp) kernel?

Post by Steve Wample » Sat, 04 Dec 2004 02:31:55



> I'm trying to replace a 100bT NIC (Intel PRO/100) with a gigabit
> card (Intel PRO/1000MT).  The driver (both the stock version
> [5.0.43-k1] and Intel's latest/greatest [5.5.4] refuse to
> load because of an EEPROM checksum error (checksum is
> expected to be BABA, but is FFC0 instead).  This happens with
> two different 1000/MT cards (both purchased same time/place)...

> Has anyone else seen this?  Got a fix?  It this really likely
> to be *two* bad EEPROMs?

To answer my own question.  Yes, apparently is is possible
to be *two* bad EEPROMs.  A 3rd Intel PRO/1000MT worked
just fine.

 
 
 

1. 2.4.20 eepro100 driver: Becker's or Intel's?

I've got an Intel 82559-based NIC (integrated into my Supermicro P4DC6+
motherboard) on which I'm running kernel v2.4.20.  Which of the 2
available drivers, Becker's or Intel's, for this chipset should I run
with this kernel?

In prior kernel versions I've used Becker's drivers and have experienced
no problems.  Now Intel's driver is included in 2.4.20.  I've read the
included doc (e100.txt), but neither it nor the Configure documentation
suggest a reason to use one driver or another.  I understand that the
Intel drivers have a feature that allows reduced CPU use at the expense
of responsiveness, but that's all I know.

Any advice on why I should prefer one driver over the other?

Thanks.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

2. No carrier from 'eth0'

3. Group file permissions

4. 2.4.20 smp on alpha won't boot

5. awk/sed/perl date program

6. 2.4.20 e100/e1000 drivers and mii-[tool,diag]

7. erratic mouse in RH 6.1

8. BUG() in e1000 driver (2.4.20-pre7)

9. 2.4.20 + XFS patches + rmap15a + Ingo's 2.4.20-rc3 O(1) sched

10. 'uname' is not updated after installing new '2.4.20' kernel??

11. Is serial driver in 2.4.20 not SMP clean?

12. oops 2.4.20 e1000?