14e4:4315 [Dell Studio 1558] Broadcom wifi crash in bcmwl-kernel-source cfg80211

[Brian Burch]

1) lsb_release -rd
Description: Ubuntu 14.04.1 LTS
Release: 14.04

2) apt-cache policy bcmwl-kernel-source
bcmwl-kernel-source:
  Installed: 6.30.223.141+bdcom-0ubuntu2
  Candidate: 6.30.223.141+bdcom-0ubuntu2
  Version table:
 *** 6.30.223.141+bdcom-0ubuntu2 0
        500 http://ubuntu.mirror.serversaustralia.com.au/ubuntu/ trusty/restricted i386 Packages
        100 /var/lib/dpkg/status

uname -a
Linux bacchus 3.13.0-37-lowlatency #64-Ubuntu SMP PREEMPT Mon Sep 22 22:00:49 UTC 2014 i686 i686 i686 GNU/Linux

3) What is expected to happen with my Dell Studio 1558 is that when using wifi hotspots away from home, it works without problem.

4) What happens instead is that the the system never makes a connection, and there is a crash in syslog:
https://bugs.launchpad.net/ubuntu/+source/bcmwl/+bug/1360783/+attachment/4185818/+files/wifi-crash-westHants-syslog-extract

The system is resuming after suspend. However, I have seen similar crashes after simply powering on, and in more than one hot spot.

04:00.0 Network controller [0280]: Broadcom Corporation BCM4312 802.11b/g LP-PHY [14e4:4315] (rev 01)
 Subsystem: Dell Wireless 1397 WLAN Mini-Card [1028:000c]
 Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
 Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
 Latency: 0, Cache Line Size: 64 bytes
 Interrupt: pin A routed to IRQ 17
 Region 0: Memory at f0500000 (64-bit, non-prefetchable) [size=16K]
 Capabilities: <access denied>
 Kernel driver in use: wl

[Brian Burch (brian-pingtoo)]

I don't know how the package name was over-ridden by the last bug I had worked on! It has nothing to do with openjdk!!

[Brian Burch (brian-pingtoo)]

brian@bacchus:~/scratch$ dpkg -l | grep bcmwl
ii bcmwl-kernel-source 6.30.223.141+bdcom-0ubuntu2 i386 Broadcom 802.11 Linux STA wireless driver source

[penalvch (penalvch)]

Brian Burch, thank you for taking the time to report this bug and helping to make Ubuntu better. Please execute the following command, as it will automatically gather debugging information, in a terminal:
apport-collect 1360783
When reporting bugs in the future please use apport by using 'ubuntu-bug' and the name of the package affected. You can learn more about this functionality at https://wiki.ubuntu.com/ReportingBugs.

[Brian Burch (brian-pingtoo)]

Sorry Chris, but your suggested command failed because it was looking for a package called "bcmwl". The package installed, as quoted in my bug subject, is bcmwl-kernel-source. This is because it is installed as a dkms module that has to be compiled against the appropriate kernel headers.

I don't know how to run apport-collect against this bug reference number, and yet make it collect from the valid ubuntu package name. Could you tell me, please?

Thanks,

Brian

[penalvch (penalvch)]

Brian Burch, just to clarify, bcmwl is the meta package name on Launchpad, associated with bcmwl-kernel-source. This is confirmed via https://launchpad.net/ubuntu/+source/bcmwl .

Despite this, did you actually run the command? If so, please post the result of the command to this report.

[Brian Burch (brian-pingtoo)]

I have just re-executed the command "apport-collect 1360783" while the broadcom adapter was active, without sudo permission.

It failed with a popup window titled "Updating Problem Report" and the text "No additional information collected."

Here is the console log:

brian@bacchus:~$ apport-collect 1360783
The authorization page:
 (https://launchpad.net/+authorize-token?oauth_token=dt1ZkgdmKbTwz5wW096W&allow_permission=DESKTOP_INTEGRATION)
should be opening in your browser. Use your browser to authorize
this program to access Launchpad on your behalf.
Press any key to continue or wait (5) seconds...
Waiting to hear from Launchpad about your decision...
Created new window in existing browser session.
Package bcmwl not installed and no hook available, ignoring
brian@bacchus:~$

It doesn't look very helpful to me!

In case you are wondering, here are the relevant package details:

brian@bacchus:~$ dpkg -l | grep bcmwl
ii bcmwl-kernel-source 6.30.223.141+bdcom-0ubuntu2 i386 Broadcom 802.11 Linux STA wireless driver source
brian@bacchus:~$ dpkg -l | grep apport
ii apport 2.14.1-0ubuntu3.5 all automatically generate crash reports for debugging
ii apport-gtk 2.14.1-0ubuntu3.5 all GTK+ frontend for the apport crash report system
ii apport-symptoms 0.20 all symptom scripts for apport
ii python-apport 2.14.1-0ubuntu3.5 all Python library for Apport crash report handling
ii python3-apport 2.14.1-0ubuntu3.5 all Python 3 library for Apport crash report handling
brian@bacchus:~$

[penalvch (penalvch)]

Brian Burch, could you please execute the following in a terminal and post the results to this report:
lspci -vvnn | grep -A 11 Network
lsb_release -rd
uname -a
apt-cache policy bcmwl-kernel-source
sudo dmidecode -s bios-version
sudo dmidecode -s bios-release-date
sudo dmidecode -s baseboard-product-name
sudo dmidecode -s baseboard-manufacturer

[Brian Burch (brian-pingtoo)]

Thanks, Christopher... I hope the following helps!

brian@bacchus:~$ lspci -vvnn | grep -A 11 Network
04:00.0 Network controller [0280]: Broadcom Corporation BCM4312 802.11b/g LP-PHY [14e4:4315] (rev 01)
 Subsystem: Dell Wireless 1397 WLAN Mini-Card [1028:000c]
 Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
 Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
 Latency: 0, Cache Line Size: 64 bytes
 Interrupt: pin A routed to IRQ 17
 Region 0: Memory at f0500000 (64-bit, non-prefetchable) [size=16K]
 Capabilities: <access denied>
 Kernel driver in use: wl

07:00.0 SD Host controller [0805]: Ricoh Co Ltd MMC/SD Host Controller [1180:e822] (rev 01)
 Subsystem: Dell Device [1028:0413]
brian@bacchus:~$ lsb_release -rd
Description: Ubuntu 14.04.1 LTS
Release: 14.04
brian@bacchus:~$ uname -a
Linux bacchus 3.13.0-37-lowlatency #64-Ubuntu SMP PREEMPT Mon Sep 22 22:00:49 UTC 2014 i686 i686 i686 GNU/Linux
brian@bacchus:~$ apt-cache policy bcmwl-kernel-source
bcmwl-kernel-source:
  Installed: 6.30.223.141+bdcom-0ubuntu2
  Candidate: 6.30.223.141+bdcom-0ubuntu2
  Version table:
 *** 6.30.223.141+bdcom-0ubuntu2 0
        500 http://ubuntu.mirror.serversaustralia.com.au/ubuntu/ trusty/restricted i386 Packages
        100 /var/lib/dpkg/status
brian@bacchus:~$ sudo sudo dmidecode -s bios-version
A12
brian@bacchus:~$ sudo dmidecode -s bios-release-date
03/30/2011
brian@bacchus:~$ sudo dmidecode -s baseboard-product-name
0874P6
brian@bacchus:~$ sudo dmidecode -s baseboard-manufacturer
Dell Inc.
brian@bacchus:~$

[penalvch (penalvch)]

Brian Burch, thank you for providing the requested information. To keep this simple, let this focus on just when you come up from a shutdown (not resume from suspend), as resume from suspend is treated as a separate problem.

Despite this, could you please answer the following questions:
+ Is this reproducible on your home WiFI network?
+ Does b43 provide a WORKAROUND as outlined in https://help.ubuntu.com/community/WifiDocs/Driver/bcm43xx ?
+ Did this happen in a prior Ubuntu/wl version?
+ Which WiFi hotspot would be most convenient for you to do testing on?
+ At the most convenient WiFi hotspot, could you please test the later version of wl in Utopic (live environment is fine) via http://releases.ubuntu.com/utopic/ and advise if this is reproducible?

[Brian Burch (brian-pingtoo)]

Q1: Is this reproducible on your home WiFI network?

I have 4 different home wifi routers (different hardware and firmware), and access to a 5th. It does not happen on any of them, which is a pity!

Q2: Does b43 provide a WORKAROUND as outlined in...

I regret not having seem that particular page before. Thanks, I will read it carefully in preparation for the next failure.

Q3: Did this happen in a prior Ubuntu/wl version?

I started using this driver version under ubuntu 13.10. I encountered the same external symptoms, but I had not done sufficient problem determination to know whether the same kernel crash was occurring. In other words, I can't be certain. The crash has certainly occurred with several different releases of the 3.13.0 low-latency kernel.

Q4: Which WiFi hotspot would be most convenient for you to do testing on?

None, I am sorry to say. All those I can be certain about are more than 10,000 miles away from me at the moment! However, I will look hard for one nearer - I will be travelling over the next few days, so I hope I can find one which is more convenient to test against.

Q5: At the most convenient WiFi hotspot, could you please test the later version of wl in Utopic...

Yes, but unfortunately I won't have time to download it before I leave.

Thanks again, Christopher. I will let you know as soon as I have some new information.

Brian

[Brian Burch (brian-pingtoo)]

I haven't been "lucky" enough to find a wifi hotspot that triggers the crash yet.

I decided to install bcmwl-kernel-source from the utopic repository, 6.30.223.248+bdcom-0ubuntu1. It installed on my 14.04 LTS lowlatency system without a problem, and is working fine on my home wifi network. I will use it for a couple of days and then install the new version on a notepad that runs 14.05 with the generic kernel.

Even though I haven't been able to find a delinquent wifi router yet, it think the be

[Brian Burch (brian-pingtoo)]

oops!

Even though I haven't been able to find a delinquent wifi router yet, it think the best overall approach will be to test the .248 driver and then release it on 14.10. Perhaps this new driver will resolve the problem...

[penalvch (penalvch)]

Brian Burch, without an offending router to test against to identify the root cause (which may not be a driver version issue as evident by it working with your five different routers) releasing an updated driver would be premature at this point.

[Brian Burch (brian-pingtoo)]

On 06/12/14 07:54, Christopher M. Penalver wrote:
> Brian Burch, without an offending router to test against to identify the
> root cause (which may not be a driver version issue as evident by it
> working with your five different routers) releasing an updated driver
> would be premature at this point.

That was definitely not my intention...

I have NO evidence whatsoever that the new driver would fix my problem,
i.e. kernel crashes in certain unidentified kinds of wifi hotspots.

However, I have good evidence that the new driver installs and runs OK
on the 14.04 LTS 3.13.0-40-lowlatency kernel.

My suggestion is that it would be good from many aspects (especially
support) to accelerate testing so the new driver could be added to the
14.04 repository without making anything worse. The advantage would be
to have the same driver on several releases to shake out any remaining bugs.

If my bug happened to be fixed as a side-effect, that would be nice.
However, the main issue would be the standardisation of the latest
driver across current ubuntu releases.

Regards,

Brian

[Andre Bogus (bogusandre)]

Brian, you may want to try the patch in I have added a bug with a description of the patch at: https://bugs.launchpad.net/ubuntu/+source/bcmwl/+bug/1415880

[Brian Burch (brian-pingtoo)]

Andre... thank you very much for letting me know about the patch.

I have applied it because I have been suffering from occasional random loss of connectivity to my own wifi routers.

I have not yet encountered another router that triggers the original crash that I reported in this particular bug, so I am unable to test it fully. However, I hope it fixes the loss of connection problems at home. I will report my experiences against your bug after a few days of testing.

[Launchpad Janitor (janitor)]

[Expired for bcmwl (Ubuntu) because there has been no activity for 60 days.]