[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Lias] Can't initialise Network Cards



On Mon, 2004-03-01 at 09:42, Paul Gear wrote:
> > This is from the /var/log/messages file after an attempt at a manual restart:
> > 
> > Mar  1 12:06:17 voyager kernel: Intel(R) PRO/1000 Network Driver - version
> > 5.2.30.1
> > Mar  1 12:06:17 voyager kernel: Copyright (c) 1999-2004 Intel Corporation.
> > Mar  1 12:06:17 voyager kernel: PCI: Found IRQ 11 for device 01:05.0
> > Mar  1 12:06:17 voyager kernel: The EEPROM Checksum Is Not Valid
> 
> That looks like your problem - the driver is refusing to load because
> the EEPROM checksum is invalid.  Have you flashed the BIOS/firmware on
> these cards at some time?  If not, you may want to see if it is
> possible to do so by downloading a boot floppy from Intel or your
> server manufacturer (try the latter first).
> 
> More than likely the Windows drivers work because they don't check the
> checksum, or just treat it as a warning.

Alternately, maybe the Windows drivers have built-in firmware that's
downloaded to the cards at boot or used instead of the on-card firmware?
I know some other types of hardware does this, and wouldn't be surprised
if these NICs were capable of it.

Also - is it possible that the invalid checksum is actually because of
something else, like memory or I/O ranges assigned to the cards that
overlap something else? This is just a guess, I don't know if it's
actually possible.

How are the memory assignments, ioports and IRQs configured - ACPI
resource allocation, BIOS auto-configuration (non-PnP), or PnP
assignment? Does your board have an IO-APIC and if so, are you using it
(not that it'll really make a difference for memory range assignment,
but it's usually a good thing for other reasons).

What does 'lspci -vvv' say about IRQs, memory assignment, etc? Anything
interesting in /proc/iomem, /proc/interrupts or /proc/ioports?

I've attached my PCI-X e1000's lspci -vvv output for your reference.
It'd be mangled if included inline. I don't know if it's any use, but
just in case...

Craig Ringer
03:01.0 Ethernet controller: Intel Corp. 82545EM Gigabit Ethernet Controller (rev 01)
        Subsystem: Intel Corp. PRO/1000 MT Server Adapter
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR+ FastB2B-
        Status: Cap+ 66Mhz+ UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
        Latency: 64 (63750ns min), cache line size 08
        Interrupt: pin A routed to IRQ 48
        Region 0: Memory at fc000000 (64-bit, non-prefetchable) [size=128K]
        Region 4: I/O ports at 9000 [size=64]
        Capabilities: [dc] Power Management version 2
                Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-
        Capabilities: [e4] PCI-X non-bridge device.
                Command: DPERE- ERO+ RBC=0 OST=0
                Status: Bus=0 Dev=0 Func=0 64bit- 133MHz- SCD- USC-, DC=simple, DMMRBC=0, DMOST=0, DMCRS=0, RSCEM-      Capabilities: [f0] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
                Address: 0000000000000000  Data: 0000