b43: system hangs during kernel initialisation [14e4:4318]

[Ben Prescott <b43-dev.lists@thewatertower.org>]

On 27/03/12 22:29, Larry Finger wrote:
> On 03/27/2012 02:42 PM, Ben Prescott wrote:
>>
>> I've been testing the 12.04 Ubuntu release, and hit an immediate 
>> issue with one
>> machine; the live ISO boots only as far as the b43 driver, and then 
>> the machine
>> is entirely non responsive and has to be powered off.
>>
>> The last thing displayed is ..
>>
>> b43-phy0 ERROR: Firmware file "b43/ucode5.fw" not found
>> b43-phy0 ERROR: Firmware file "b43-open/ucode5.fw" not found
>> b43-phy0 ERROR: You must go to
>> http://wireless.kernel.org/en/users/Drivers/b43#devicefirmware and 
>> download the
>> correct firmware for this driver version. Please carefully read all 
>> instructions
>> on this website.
>>
>> If I put the following in the kernel command line
>>
>> b43.blacklist=yes
>>
>> the module behaves differently
>>
>> "b43: Unknown parameter 'blacklist'"
>>
>> and the problem goes away. There's probably a better way to achieve 
>> the same
>> goal, but this solution has the benefit that it is easy to switch it 
>> on and off
>> via grub.
>>
>>
>>
>> I've not attempted to install firmware, as for most 'normal' users, the
>> opportunity doesn't arise. Appreciate that for your understanding 
>> that maybe of
>> interest. Let me know.
>>
>>
>>
>> You may find further useful detail on the bug over at:
>> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/950295 - 
>> including various
>> diagnostic data that was captured automatically for kernel issues.
>>
>> The system is 32bit.
>>
>> The 12.04 kernel is, currently: "Ubuntu 3.2.0-20.32-generic-pae 3.2.12"
>>
>> The only way to make that boot is with the kernel argument.
>>
>>
>>
>> As requested on the b43 linuxwireless.org page, please find below the 
>> uname info
>> and lspci info, and attached the dmesg.
>>
>> As part of the investigation for the ubuntu bug, I've been testing 
>> Canonical's
>> vanilla kernel builds; more info about those here:
>> https://wiki.ubuntu.com/Kernel/MainlineBuilds
>>
>> I booted the system off a kernel that works; obviously the bug isn't
>> illustrated. This is the most recent canonical provided mainline 
>> 'trunk' (3.x.0)
>> kernel which boots.
>>
>> $ uname -a
>> Linux tecumseh 3.1.0-030100-generic #201110241006 SMP Mon Oct 24 
>> 14:20:44 UTC
>> 2011 i686 i686 i386 GNU/Linux
>>
>> ben at tecumseh:~$ lspci -vvn|grep 4318 -A7
>> 06:02.0 0280: 14e4:4318 (rev 02)
>> Subsystem: 1468:0312
>> 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: 64
>> Interrupt: pin A routed to IRQ 22
>> Region 0: Memory at d0002000 (32-bit, non-prefetchable) [size=8K]
>> Kernel driver in use: b43-pci-bridge
>>
>> I've tried lots of other kernels; including
>>
>> 3.3.0 - hangs
>> 3.2.0 - hangs
>>
>> I've also tested a few of the 3.2.0 release candidates. I worked back to
>> 3.2.0-rc2; It also hangs. Canonical didn't do a 32bit 3.2.0-rc1, so I 
>> can't test
>> that.
>>
>> None of the kernels I've tested down the 3.1.x branch have the problem.
>>
>> The Ubuntu bug page has all the kernels I've tried; where 
>> /proc/cmdline doesn't
>> show the argument, I was able to boot it.
>>
>> Please do let me know what other info / procedures would help to 
>> troubleshoot this.
>
> What form factor is your BCM4318? I downloaded the 32-bit 12.04 iso 
> and was able to boot it on my sandbox, which has both a BCM4306 (b43) 
> and a BCM4301 (b43legacy) in PCI form. No problems other than the 
> kernel log showing no firmware available and I had no network. 
> Clearly, not a hang during booting.
>
> Larry
>
The machine is a laptop;  the wifi is a mini-PCI.

Ben