forcedeth not working after dapper upgrade

Bug #45257 reported by Johan Dahlin
10
Affects Status Importance Assigned to Milestone
linux-restricted-modules-2.6.15 (Ubuntu)
Fix Released
Undecided
Unassigned
linux-source-2.6.15 (Ubuntu)
Fix Released
Medium
Unassigned

Bug Description

Binary package hint: linux-image-2.6.15-22-k7

I have a box which uses a nForce2 ethernet controller:

0000:00:04.0 Ethernet controller: nVidia Corporation nForce2 Ethernet Controller (rev a1)
        Subsystem: Asustek Computer, Inc. A7N8X Mainboard onboard nForce2 Ethernet
        Flags: bus master, 66MHz, fast devsel, latency 0, IRQ 22
        Memory at e9000000 (32-bit, non-prefetchable) [size=4K]
        I/O ports at e000 [size=8]
        Capabilities: <available only to root>

It boots up fine using 2.6.12-10 on breezy, but after upgrading to 2.6.15-21 it will no longer detect the device.

It's on a diskless/nfsroot setup, the initrd image is transfered over the wire, but the kernel fails to find the device so the nfs root cannot be mounted.

Revision history for this message
Johan Dahlin (jdahlin-deactivatedaccount) wrote :

lspci -vn output:

0000:00:04.0 0200: 10de:0066 (rev a1)
        Subsystem: 1043:80a7
        Flags: bus master, 66MHz, fast devsel, latency 0, IRQ 22
        Memory at e9000000 (32-bit, non-prefetchable) [size=4K]
        I/O ports at e000 [size=8]
        Capabilities: [44] Power Management version 2

Revision history for this message
Adam Conrad (adconrad) wrote :

Can you boot with "break=premount" and when it dumps you to a shell, try "modprobe forcedeth" to see if it loads manually?

Revision history for this message
Johan Dahlin (jdahlin-deactivatedaccount) wrote :

kernel first says:

...
PCI: nForce C1 Halt Disconnect fixup
...

after loading the module manually I get this:

**** SET: Misaligned resource pointer df8000c2 07 Len 0
ACPI: PCI Interrupt Link [APCH] enabled at IRQ 22
ACPI: PCI Interrupt 0000:00:04.0[A] Link [APCH] -> GSI 22 (level, high) -> IRQ 17
0000:00:04.0: open: Could not find a valid PHY
ACPI: PCI interrupt for device 0000:00:04.0 disabled
forcedeth: probe of 0000:00:04.0 failed with error -12

Revision history for this message
Christian Reis (kiko) wrote :

And on Breezy, it works fine (though there's an odd warning in the dmesg, no issues in practice occur):

Dec 21 15:44:49 deadsea kernel: forcedeth.c: Reverse Engineered nForce ethernet driver. Version 0.29.
Dec 21 15:44:49 deadsea kernel: ACPI: PCI interrupt 0000:00:04.0[A] -> GSI 11 (level, low) -> IRQ 11
Dec 21 15:44:49 deadsea kernel: PCI: Setting latency timer of device 0000:00:04.0 to 64
Dec 21 15:44:49 deadsea kernel: 0000:00:04.0: open: Could not find a valid PHY.
Dec 21 15:44:49 deadsea kernel: eth0: forcedeth.c: subsystem: 01043:80a7 bound to 0000:00:04.0

Changed in linux-source-2.6.15:
status: Unconfirmed → Confirmed
Revision history for this message
Ben Collins (ben-collins) wrote :

In breezy, the driver just ignored the missing PHY. Made dapper do the same.

Changed in linux-source-2.6.15:
status: Confirmed → Fix Committed
Changed in linux-source-2.6.15:
status: Fix Committed → Fix Released
Revision history for this message
Ayaz (ayaz99) wrote :

You need to upgrade to latest version. Verion 0.29 is really old. The latest version is 0.56.

There has even been a fix for phy addresses.

Why are you patching a really old version? I am assuming that ignoring that message will still allow networking to work (even though you never initialized the phy)?

Revision history for this message
Christian Reis (kiko) wrote :

Networking does work indeed even without initializing the PHY. I believe the reason Ben patched the version is that we were already very close to Dapper freeze. The edgy kernel will most likely include the new version.

Changed in linux-restricted-modules-2.6.15:
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.