BCM43xx failing on Edgy AMD64 kernel

Bug #69026 reported by Cory Prowse
18
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Fix Released
Undecided
Unassigned
linux-source-2.6.17 (Ubuntu)
Invalid
Undecided
Unassigned

Bug Description

This is a known bug in the bcm43xx code that has apparently been fixed in the latest 2.6.17 kernels, however the problem is still present in Edgy.
(ndiswrapper does not work either - but I'd rather get the native driver working anyway)

As a side note, I must remove the "ohci1394" module or dmesg gets constantly flooded with
  APIC error on CPU0: 40(40)

With the firmware in /lib/firmware the kernel crashes with a "softlock on CPU#0"

If there is any further information I can provide to assist, please ask.

Tags: edgy-close
Revision history for this message
Cory Prowse (cory-prowse) wrote :
Revision history for this message
Cory Prowse (cory-prowse) wrote :
Changed in linux-source-2.6.17:
status: Unconfirmed → Confirmed
Revision history for this message
Cory Prowse (cory-prowse) wrote :

The machine now locks up occasionally with irqpoll boot option, so for the time being I have turned that off and I now get flooded with the following in dmesg:

[ 760.360217] APIC error on CPU0: 40(40)
[ 760.446461] APIC error on CPU0: 40(40)
[ 760.967114] APIC error on CPU0: 40(40)
[ 761.080511] APIC error on CPU0: 40(40)
[ 761.500552] APIC error on CPU0: 40(40)
[ 762.591395] APIC error on CPU0: 40(40)
[ 762.664858] APIC error on CPU0: 40(40)
[ 762.810198] APIC error on CPU0: 40(40)
[ 762.814990] APIC error on CPU0: 40(40)
[ 762.830964] APIC error on CPU0: 40(40)

Revision history for this message
towsonu2003 (towsonu2003) wrote :

please do not confirm your own bugs. It would be great if you could let another user or developer triage it (check to see if it's a duplicate, if it's valid, if it is reproduceable and so on). thanks :)

Changed in linux-source-2.6.17:
status: Confirmed → Unconfirmed
Revision history for this message
Cory Prowse (cory-prowse) wrote :

Sorry, I should have thought that one through.

As an update I have discovered the possible reason for the flood of "APIC error on CPU0: 40(40)" messages and detailed it in an associated bug:
  https://launchpad.net/distros/ubuntu/+source/linux-source-2.6.17/+bug/66900

Essentially the ohci1394 needed to be removed, which seems to point at a problem with firewire.

description: updated
description: updated
Revision history for this message
Xamusk (ronanpaixao) wrote :

On my laptop some messages like that appeared too.
I solved the problem adding the noapic nolapic options in the kernel boot.
This was a big issue with dapper in my laptop (HP NX6125) in which everything would run tremendously slow if these options were not added. It looks like it is some APIC problem (some laptops reportedly have strange APIC issues)
Edgy somehow fixed this problem partially, but not adding them caused the system to be less stable.
Right now, my bcm43xx driver is having some issues, but generally work, even with the ohci1394 module loaded.

Revision history for this message
Cory Prowse (cory-prowse) wrote :

Unfortunately adding "noapic nolapic" makes the situation worse, with my sound card looping the bongo drum sounds on gdm startup.

As for BCM43xx, that has never worked for me (I have 2G of memory which is supposed to have been fixed in later kernels, but so far no luck)

Revision history for this message
Kevin Cantu (kevincantu-deactivatedaccount) wrote :

which chipset? bcm43xx works for some and not others...

Revision history for this message
Cory Prowse (cory-prowse) wrote :

This is an Acer Ferrari 4005 Laptop, using a Broadcom 4318 chipset for Wifi.

I've had to resort to installing a 2.6.18 kernel to use ndiswrapper (this also fixed pretty much all other issues with this laptop's hardware in Ubuntu).

Revision history for this message
Alvin Thompson (alvint-deactivatedaccount) wrote :

i also have a acer ferrari 4005. i've tried the stock 4318 card as well as replacing it with a 4306 card and an intel card, none of which work. all of these cards work with ubuntu on other computers. there is a serious problem with interrupts on edgy amd64 and the acer ferrari which cause the mini-pci bus (and many other things, like the power button) not to work.

Revision history for this message
James Tait (jamestait) wrote :

Updating the BIOS from 3A19 to 3A23 did not solve this problem for me.

Since the card works under Dapper, would it be worth including dmesg output from a Dapper boot to compare with the Edgy boot? Would it help us to narrow down the source of the problem?

Revision history for this message
Matthew Woerly (nattgew) wrote :

Could you test this in the latest version of Ubuntu? The bcm43xx driver has been replaced by b43.

Revision history for this message
James Tait (jamestait) wrote :

While the frequency of the message has certainly decreased significantly under Hardy for me, I am still seeing at least a couple of instances most days.

Revision history for this message
Sergio Zanchetta (primes2h) wrote :

The 18 month support period for Edgy Eft 6.10 has reached it's end of life. As a result, we are closing the linux-source-2.6.17 Edgy Eft kernel task.

Changed in linux-source-2.6.17:
status: New → Invalid
Revision history for this message
Sergio Zanchetta (primes2h) wrote :

Hardy Heron 8.04 was recently released. It would be helpful if you could test the new release and verify if this is still an issue - http://www.ubuntu.com/getubuntu/download . You should be able to test your bug using the LiveCD. Please let us know your results. Thanks.

Changed in linux:
status: New → Incomplete
Revision history for this message
James Tait (jamestait) wrote :

The bcm4318 card is certainly functional for me using the b43 driver under Hardy, although it does seem to hang for several seconds every so often. I have no hard evidence to support this, just personal observations.

Revision history for this message
Sergio Zanchetta (primes2h) wrote :

Ok, so I set it fixed about the problem reported in this bug.
If you think it's worth, open a new bug report about the hang problem.
Thanks.

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

Duplicates of this bug

Other bug subscribers

Remote bug watches

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