Skip to contentSkip to conversation

2.6.7-rc1 breaks forcedeth

DISCUSSION Luis Miguel García Mancebo 2 participants 2 messages May 30, 2004

lore
Share & tools
lore
#0 lore raw
Quoted 16 lines
Quoted 11 lines · Lee Howard

I use the forcedeth driver for my nVidia ethernet successfully with kernel 2.6.6. I recently tested 2.6.7-rc1, and when using it the ethernet does not work, and I see this in dmesg:

eth1: forcedeth.c: subsystem: 01043:80a7 bound to 0000:00:04.0 NETDEV WATCHDOG: eth1: transmit timed out
NETDEV WATCHDOG: eth1: transmit timed out
NETDEV WATCHDOG: eth1: transmit timed out
NETDEV WATCHDOG: eth1: transmit timed out
NETDEV WATCHDOG: eth1: transmit timed out

Quoted 2 lines

I can ping localhost and the device's IP number, but I cannot ping other systems' IP numbers.

Quoted 6 lines

Well, there are zero changes to the driver itself, so I would guess ACPI perhaps...

Try booting with 'acpi=off' or 'noapic' or 'pci=noacpi' or similar...

Jeff

The same here with nforce2 too. Actually I cannot test that, but I can confirm problems with the ethernet driver in -rc1. Perhaps I can test on tuesday if -rc2 works or the options you mention above.

Now that you remember me... I remember seeing one message... something about problems with IRQ # 9.

Thanks.,

Signature · 3 lines
Luis Miguel García Mancebo
Universidad de Deusto / Deusto University
Spain
#1 lore raw
Quoted 2 lines · Luis Miguel García Mancebo

The same here with nforce2 too. Actually I cannot test that, but I can confirm problems with the ethernet driver in -rc1. Perhaps I can test on tuesday if -rc2 works or the options you mention above.

As I said, there are _zero_ changes to the ethernet driver, therefore it is _not_ a problem with the ethernet driver.

Quoted 1 line

Now that you remember me... I remember seeing one message... something about problems with IRQ # 9.

And indeed, this is more indication it's not a problem with the driver.

ACPI, your BIOS and the Linux PCI core are responsible for getting irq routing correct.

Jeff