B43 driver causing numerous APIC errors

[nukedathlonman]

I have an Acer Aspire 5002WLMI and am not able to use the B43 driver in Kernel 2.6.16-24 Generic. When I do, I get tons of APIC error messages in dmesg while the system runs slowly and eventually crashes hard at random. If I use Gusty Kernel 2.6.14-22 Generic with ndiswrapper, I get no APIC error's while system runs fast and stable (can leave it on for days with out crashing).

[nukedathlonman (areginato)]

Bug also exists in 2.6.24-17 Generic. I also must apologize that I labeled the Kernel version's wrong: 2.6.24-16 Generic is the Kernel I was having problems with originally. The kernel that works fine for me is Gusty's Kernel 2.6.22-14 Generic (my extreme bad)

With Kernel 2.6.24-17 & 2.6.2-16 I can manually remove B43 (modprobe -r b43) to eliminate the APIC error (which stabilizes my system). Both the effected kernels will load ndiswrapper driver, but neither kernel's ndriswrapper will load the windows driver despite the windows device driver showing present when ndiswrapper -l is executed. I also toyed with blacklisting the b43 driver, unloading the ssb driver, and even tried to blacklist the ssb driver - no joy in getting ndiswrapper to take over with the Windows driver. Ndiswrapper works 100% with the Windows Broadcom drivers in the old Gusty 2.6.22-14 Kernel, which is why I've reverted to an older kernel for the time being.

[Larry Finger (larry-finger)]

Can you get any of the APIC errors out of your old logs? I'm a bcm43xx developer and this is the first report of this sort of error.

[nukedathlonman (areginato)]

I just checked my old logs, and I don't have 'em. But I can give you two fresh logs in a few minutes - I'll unblock the ssb and b43 drivers while I block ndiswrapper and boot the kernel 2.6.17 that I'm having the problems with, then I'll block them again and unblock ndiswrapper and do another boot of the same kernel for comparison.

[nukedathlonman (areginato)]

Log with APIC error's. I think this happens with wireless network activity.

[nukedathlonman (areginato)]

Same kernel, ssb & b43 blacklisted while ndiswrapper is unblacklisted.

[Larry Finger (larry-finger)]

Could you try adding the "noapic" option to the boot and retry b43?

The error message indicates an illegal register address, but there may be some other issue.

I would also like to see the output of lspci -v for the Broadcom card.

[nukedathlonman (areginato)]

Sorry for taking so long to write back - I disabled APIC mode using the noapic kernel boot parameter. And have been testing it for a few days to fully confirmed that all the hardware is working properly - a touch slower to boot, and can see a few things that were on separate interrupts are now having to share. No major worries as everything is working as it should be. I'm baffled as to why I can use APIC mode with ndiswrapper and not get a flood of illegal register address messages and the laptop is rock solid stable, but continuously get illegal register messages & instability if I don't disable APIC mode while I use the ssb with b43 driver. I wonder if the wireless adapter is doing something abnormal/non-standard... :-(

The info I snipped about the Broadcom board:

00:0b.0 Network controller: Broadcom Corporation BCM4318 [AirForce One 54g] 802.11g Wireless LAN Controller (rev 02)
        Subsystem: AMBIT Microsystem Corp. TravelMate 2410
        Flags: bus master, fast devsel, latency 64, IRQ 4
        Memory at e2000000 (32-bit, non-prefetchable) [size=8K]

[nukedathlonman (areginato)]

Change that - when the computer goes into sleep mode (when I'm using it on battery - I have all that disabled when runnign on hydro), the wireless won't wake up. I tried having the computer unload B43 and ssb modules before suspending, and then reload them on wake up - and it won't connect to any AP untill I reboot the computer into Windows (it unloads and reloads properly, and I can see various AP's, but if they are encrypted with WEP or WPA it will not communicate at all). So I ended up re-compileing ndiswrapper for Kernel 2.6.24-17 (the packaged one doesn't work properly - it loads but wont talk to any AP) and am now using the current kernel with APIC mode enabled.

[Leann Ogasawara (leannogasawara)]

The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would appreciate it if you could please test this newer 2.6.27 Ubuntu kernel. There are one of two ways you should be able to test:

1) If you are comfortable installing packages on your own, the linux-image-2.6.27-* package is currently available for you to install and test.

--or--

2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4. Please watch http://www.ubuntu.com/testing for Alpha5 to be announced. You should then be able to test via a LiveCD.

Please let us know immediately if this newer 2.6.27 kernel resolves the bug reported here or if the issue remains. More importantly, please open a new bug report for each new bug/regression introduced by the 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please specifically note if the issue does or does not appear in the 2.6.26 kernel. Thanks again, we really appreicate your help and feedback.

[nukedathlonman (areginato)]

Sorry, I'm unable to test the new kernel for this bug - I changed my Laptop over to Debian Etch a few months ago (my other computer is still running Ubuntu). :-(

[Leann Ogasawara (leannogasawara)]

Thanks for the note. I'm going to mark this Incomplete for now. It will continue to remain open for a period of time in case you are able to test. Thanks.

[baggins (bill-binney)]

  I'm getting this. I was hunting for 2 days for info on it but didn't find this page. Tried to file a report but I don't think it went through. I'm running 8.10 kernel 2-6-27-9 on an acer aspire 3633WLMI. I'm brand new and shiney on this as I just jumped on Thursday gone, so please bear with me. Advice on how to output terminal to file so I can attach something usefull to you would be really good.

  Meaenwhile I'll attach what I copy/pasted out of terminal while trying to log the thing last night.EDIT>> In fact I saved the whole report- It's done with gedit so I'm guessing you'll have no probs with that. If you do mail me I WILL sort something else out.

    Don't know if the system crashes eventually, seems to run ok for the most part but this does affect network traffic.

    Sorry the "attempted" report was rambly. (Worse than this!!) And the attached Log is looong. Hope it helps though.

    Please do Mail me about stuff I've bookmarked the page so I should find it again but...... like I said it's taken me a couple of days to get this far!!

   Thanks in advance

[Larry Finger (larry-finger)]

I don't know why you are getting APIC errors. I certainly doubt that b43 is causing them - it seems much more like a BIOS error that b43 is triggering. These reports on ACER laptops are the only ones I have seen. Most users never see anything like this..

To save the output of a command, use the redirection operator ">", which creates a file. For example, use

lspci -nnv > output.txt

That creates the file "output.txt" and puts the generated output in it. If you use ">>" rather than ">", then the output is appended to the file.

[baggins (bill-binney)]

  Cheers for the quickresponse, and advice. There is a newer bios for the laptop on the acer website but it's for Vista. I want nothing to do with vista. I'm running dual boot with xp till I've ironed out a few things then I'll probably abandon windoze completely. The site warns against unnecessary update of the bios if your not sure of what you're doing because the wrong one would be BAD. If you reckon the new one would be ok then I'll update it and get back to you (so long as it works- if not my pewt may not work???)

  You want output.txt right?

   Here goes............
                          ok. Where is output.txt
   Found it.. Hope it's usefull.

BTW >> attaches it to the end of an existing file, so I could output result of another command to the same file?
May sound dumb to you but I am really new but not afraid of dumb questions.

  Many thanks

[Larry Finger (larry-finger)]

Since Acer tells you nothing about why Vista needs a new BIOS, it is impossible to tell whether it would help your problem, or not.

The warning is a general one. If a BIOS reflashing goes wrong, the motherboard may be bricked. On the other hand, I have done this operation many times and it has never gone wrong.

Yes, using the ">>" adds the output of a second command to the end of the file. That usage is standard for Unix, which Linux has copied.

[baggins (bill-binney)]

 Thankyou. Should I report this as a seperate bug then? Under what heading? Just checked and the same thing happens under the previous kernel as well. What about the occasional "roaming to none" is this normal? or a seperate bug to report? Or hardware difficiency? Or who knows? I've checked the bios number (3A22) and tried to find out from the acer site (I nearly ended up punching the screen-it's soooo helpfull NOT), and hunted google, but info on this seems sparse and/or irrelevant.

  Sorry to keep on at you (seems to be not your problem) but hopefully in the future I will be passing on help to others and taking part of the burden.

 Cheers

Bill

[nukedathlonman (areginato)]

I'm not runing Kubuntu on my Acer Aspire 5002WLMi anymore - I'm running Debian AMD64 Etch/Lenny/Sid (Etch with serveral things ported from testing & unstable). At anyrate, try compiling a Kernel from kernel.org - all my wireless problems completly disappeared when I started running the Kernel 2.6.28.x series kernels from there (I only see an occasional APIC error - out side that, I'm now getting solid & very reliable connections with b43). And these APIC errors are not with Acer coputers, I've seen many system logs from serveral brand of laptops (HP's, Compaq's) that use broadcrum based wireless adapters. Most people ignore them, and everyone has told me to not worry about them because it's a non issue.

Andrew

[Andy Whitcroft (apw)]

It is not clear whether this bug is still an issue. Can you confirm this issue exists with the most recent Jaunty Jackalope 9.04 release - http://www.ubuntu.com/news/ubuntu-9.04-desktop . If the issue remains in Jaunty, please test the latest upstream kernel build - https://wiki.ubuntu.com/KernelMainlineBuilds . Please report back here either way. Thanks!

[Jeremy Foshee (jeremyfoshee)]

This bug report was marked as Incomplete and has not had any updated comments for quite some time. As a result this bug is being closed. Please reopen if this is still an issue in the current Ubuntu release http://www.ubuntu.com/getubuntu/download . Also, please be sure to provide any requested information that may have been missing. To reopen the bug, click on the current status under the Status column and change the status back to "New". Thanks.

[This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]