Red Hat Bugzilla – Bug 374871
Wireless led no longer working on Inspiron 1501 with b43 driver
Last modified: 2007-12-20 08:49:44 EST
Description of problem:
With the upgrade to kernel 18.104.22.168-49.fc8, the wireless led of my Dell Inspiron
1501 stopped working, while it used to work with all previous kernel versions.
Wireless networking works without problems though...
Version-Release number of selected component (if applicable):
[dario@localhost ~]$ uname -a
Linux localhost.localdomain 22.214.171.124-49.fc8 #1 SMP Thu Nov 8 22:14:09 EST 2007
x86_64 x86_64 x86_64 GNU/Linux
Steps to Reproduce:
1. Copy proper firmware files in /lib/firmware/b43
2. modprobe b43
Wireless works, but the wireless led doesn't turn on. Fn+F2 doesn't toggle it
Wireless led turns on during startup, Fn+F2 toggles it.
[dario@localhost ~]$ dmesg | grep b43
b43-phy0: Broadcom 4311 WLAN found
b43-phy0 debug: Found PHY: Analog 4, Type 2, Revision 8
b43-phy0 debug: Found Radio: Manuf 0x17F, Version 0x2050, Revision 2
b43-phy0 debug: Loading firmware version 351.126 (2006-07-29 05:54:02)
Registered led device: b43-phy0:tx
Registered led device: b43-phy0:rx
b43-phy0 debug: Chip initialized
b43-phy0 debug: 32-bit DMA initialized
b43-phy0 debug: Wireless interface started
b43-phy0 debug: Adding Interface type 2
b43-phy0 debug: Using hardware based encryption for keyidx: 0, mac:
Previous kernels didn't register any led devices, but it used to work without
problems. I think this is a new feature that was recently introduced upstream,
that made my wireless led non-functional.
I'm seeing the same thing here on my HP dv2120, also a 4311 chipset. Glad to
provide further details if needed, otherwise will gladly test potential fixes ....
Same behavior with kernel 126.96.36.199-58.fc8 from Koji build.
I have the same problem on a Dell Inspiron B130 with kernel 188.8.131.52-49.fc8 and
a broadcom 4318 wireless.
I have the same problem with a Dell Latitude D620 (b43 driver, 1390/4311 chip)
Also, the radio kill hard switch is not detected anymore.
I have looked in the source, and the problem seems to be in the new rfkill code:
It uses an input polled device and the rfkill module. However the poll function
(in b43/rfkill;c) which tests the radio kill hard switch is never called.
I tried to trace, and it seems that the input polled device open routine (which
starts the poll periodic function) is not called.
I shall investigate further.
The following method allows the led and the radio kill hard switch to work:
- Load the rfkill-input module (modprobe rfkill-input) BEFORE the load of the
Then, when the input polled device in b43 is registered, it is connected to the
input_handler in rfkill-input and thus opened, staring the periodic poll.
However in kernel 184.108.40.206-49.fc8 the b43 module depends on the rfkill module
but not on the rfkill-input module.
May be some dependancy is missing.
I tested today kernel 220.127.116.11-99.fc8 from Koji and the wireless led is back
I noticed that it adds a handler for wireless radio led
Registered led device: b43-phy0:radio
I guess that was the culprit... Good job!
This bug can be closed for what regards me...