Re: [it really is text/plain :P] dual port onboard 82546EB checksum errors

["Jesse Brandeburg" <jesse.brandeburg@gmail.com>]

please copy netdev, or better yet e1000-devel@lists.sourceforge.net in
the future for e1000 issues like this.  Feel free to move the
conversation there, if you would like.

On 4/19/07, David Ford <firefighterblu3@gmail.com> wrote:
> I have a rackmount server that has a dual port onboard 82546EB card.
>
> I've googled and seen this card apparently active with other users but
> I seem to only get checksum errors.
>
> [    0.194129] Intel(R) PRO/1000 Network Driver - version 7.3.15-k2-NAPI
> [    0.194234] Copyright (c) 1999-2006 Intel Corporation.
>
> [    0.194405] ACPI: PCI Interrupt 0000:04:09.0[A] -> GSI 51 (level,
> low) -> IRQ 16
> [    0.458732] e1000: 0000:04:09.0: e1000_probe: (PCI:66MHz:64-bit)
> 00:04:23:cd:93:d2
> [    0.489766] e1000: eth0: e1000_probe: Intel(R) PRO/1000 Network Connection
>
> [    0.489889] ACPI: PCI Interrupt 0000:03:07.0[A] -> GSI 30 (level,
> low) -> IRQ 17
> [    0.721071] e1000: 0000:03:07.0: e1000_probe: The EEPROM Checksum
> Is Not Valid
> [    0.742509] ACPI: PCI interrupt for device 0000:03:07.0 disabled
> [    0.742614] e1000: probe of 0000:03:07.0 failed with error -5
>
> [    0.742731] ACPI: PCI Interrupt 0000:03:07.1[B] -> GSI 31 (level,
> low) -> IRQ 18
> [    0.973865] e1000: 0000:03:07.1: e1000_probe: The EEPROM Checksum
> Is Not Valid
> [    0.995304] ACPI: PCI interrupt for device 0000:03:07.1 disabled
> [    0.995405] e1000: probe of 0000:03:07.1 failed with error -5
>
>
> 03:07.0 Ethernet controller: Intel Corporation 82546EB Gigabit
> Ethernet Controller (Copper) (rev 01)
> 03:07.1 Ethernet controller: Intel Corporation 82546EB Gigabit
> Ethernet Controller (Copper) (rev 01)
> 04: 09.0 Ethernet controller: Intel Corporation 82545GM Gigabit
> Ethernet Controller (rev 04)
>
>
> 03:07.0 Ethernet controller: Intel Corporation 82546EB Gigabit
> Ethernet Controller (Copper) (rev 01)
>         Subsystem: Intel Corporation Unknown device 341a
>         Flags: 66MHz, medium devsel, IRQ 17
>         Memory at fe5c0000 (64-bit, non-prefetchable) [size=128K]
>         I/O ports at 2040 [size=64]
>         Capabilities: [dc] Power Management version 2
>         Capabilities: [e4] PCI-X non-bridge device
>         Capabilities: [f0] Message Signalled Interrupts: Mask- 64bit+
> Queue=0/0 Enable-
>
> 03:07.1 Ethernet controller: Intel Corporation 82546EB Gigabit
> Ethernet Controller (Copper) (rev 01)
>         Subsystem: Intel Corporation Unknown device 341a
>         Flags: 66MHz, medium devsel, IRQ 18
>         Memory at fe5e0000 (64-bit, non-prefetchable) [size=128K]
>         I/O ports at 2000 [size=64]
>         Capabilities: [dc] Power Management version 2
>         Capabilities: [e4] PCI-X non-bridge device
>         Capabilities: [f0] Message Signalled Interrupts: Mask- 64bit+
> Queue=0/0 Enable-
>
> 04:09.0 Ethernet controller: Intel Corporation 82545GM Gigabit
> Ethernet Controller (rev 04)
>         Subsystem: Intel Corporation PRO/1000 MT Server Adapter
>         Flags: bus master, 66MHz, medium devsel, latency 64, IRQ 16
>         Memory at fe8c0000 (64-bit, non-prefetchable) [size=128K]
>         Memory at fe880000 (64-bit, non-prefetchable) [size=256K]
>         I/O ports at 3000 [size=64]
>         Expansion ROM at fe840000 [disabled] [size=256K]
>         Capabilities: [dc] Power Management version 2
>         Capabilities: [e4] PCI-X non-bridge device
>
>
> The 82545GM card comes up fine but the 46EB card does not.
>
> This is 2.6.20-gentoo-r6
>
> I'll be happy to fulfill information request if needed.  Any
> suggestions that'd help me get the 46EB working?

We haven't heard of any reports about this, can you get us some more
information about the platform this is in?
lspci -vv
dmesg (full version)
you can comment out the eeprom check and mac address check, rebuild
the driver and load it, then run
ethtool -e ethX
on the ports that are failing, without bringing the interface 'up'

you may have a hardware failure, or you could be having bus issues.
The eeprom read is the first point that we try to read a register from
the adapter so if register reads are failing this is the symptom you
will see.