Re: [arch-general] wireless driver broadcom b43 random success or failure

[message]

From the wiki page:
"Each configuration file is named in the style of 
/etc/modules-load.d/.conf. Configuration files simply contain a 
list of kernel modules names to load, separated by newlines."


Why was the file 'b43.conf' containing only:

b43

incorrect? Is not 'b43' the name of the kernel module?

Re: [arch-general] wireless driver broadcom b43 random success or failure

[Mauro Santos]

On 12-04-2016 09:14, tak...@bluewin.ch wrote:

> Little nit-picking: the `modprobe` command is available (and used) on
> Arch Linux, and it *does* use the files in /etc/modprobe.d (see also
> modprobe.d(5) for what the files are used for). So this is not really an
> indication for whether OP's using Arch or not.

That is correct, my bad on that one. I was thinking only about module
loading at boot time.

-- 
Mauro Santos

Re: [arch-general] wireless driver broadcom b43 random success or failure

[tak...@bluewin.ch]

On Tue, Apr 12, 2016 at 08:44:36 +0100, Mauro Santos wrote:
> Also arch does not use /etc/modprobe.d, if you are not using Arch Linux
> you are on your own.

Little nit-picking: the `modprobe` command is available (and used) on
Arch Linux, and it *does* use the files in /etc/modprobe.d (see also
modprobe.d(5) for what the files are used for). So this is not really an
indication for whether OP's using Arch or not.

However, yes, automatic module-loading at startup is not modprobe's
job, so it's the wrong location to configure such stuff. Please have a
look at modules-load.d(5) instead (and take a second, more precise look
at the wiki article Mauro posted).

Re: [arch-general] wireless driver broadcom b43 random success or failure

[Mauro Santos]

On 12-04-2016 08:00, message wrote:
> Mauro Santos Sun, 10 Apr 2016 11:29:06 -0700
>> You may want to refer to this page for guidance
> https://wiki.archlinux.org/index.php/Kernel_modules#Automatic_module_handling
> 
> 
> Thanks. Created a 'conf' file (root user):
> 
> cat > /etc/modprobe.d/b43.conf
> modprobe b43
> 
> Then restarted computer. The module failed to load. On attempting the
> usual command, a new error occurred:
> 
> modprobe b43
> libkmod: kmod_config_parse: /etc/modprobe.d/b43.conf line 1: ignoring
> bad line starting with 'modprobe'
> 
> So, deleted the 'b43.conf' file and loaded the module manually again.
> .
> 

You should re-read the wiki section I pointed you to. That is not the
syntax of the configuration files for modules-load.d.

Also arch does not use /etc/modprobe.d, if you are not using Arch Linux
you are on your own.

-- 
Mauro Santos

Re: [arch-general] wireless driver broadcom b43 random success or failure

[Dragon ryu]

I think Mauto means, use /etc/module-load.d and make new conf?
2016/04/12 16:00 "message" :

> Mauro Santos Sun, 10 Apr 2016 11:29:06 -0700
>
>> You may want to refer to this page for guidance
>>
>
> https://wiki.archlinux.org/index.php/Kernel_modules#Automatic_module_handling
>
> Thanks. Created a 'conf' file (root user):
>
> cat > /etc/modprobe.d/b43.conf
> modprobe b43
>
> Then restarted computer. The module failed to load. On attempting the
> usual command, a new error occurred:
>
> modprobe b43
> libkmod: kmod_config_parse: /etc/modprobe.d/b43.conf line 1: ignoring bad
> line starting with 'modprobe'
>
> So, deleted the 'b43.conf' file and loaded the module manually again.
>

Re: [arch-general] wireless driver broadcom b43 random success or failure

[message]

Mauro Santos Sun, 10 Apr 2016 11:29:06 -0700

You may want to refer to this page for guidance

https://wiki.archlinux.org/index.php/Kernel_modules#Automatic_module_handling

Thanks. Created a 'conf' file (root user):

cat > /etc/modprobe.d/b43.conf
modprobe b43

Then restarted computer. The module failed to load. On attempting the 
usual command, a new error occurred:


modprobe b43
libkmod: kmod_config_parse: /etc/modprobe.d/b43.conf line 1: ignoring 
bad line starting with 'modprobe'


So, deleted the 'b43.conf' file and loaded the module manually again.

Re: [arch-general] wireless driver broadcom b43 random success or failure

[message]

Dieter Wirz Sun, 10 Apr 2016 09:45:07 -0700

echo b43 >> /etc/modules


The directory file shows:

b43

There is nothing more. The module is not loaded automatically, so I 
continue to start the wireless driver manually, as described before.

Re: [arch-general] wireless driver broadcom b43 random success or failure

[Mauro Santos]

On 10-04-2016 18:27, Kenneth Jensen wrote:
> On Sun, Apr 10, 2016 at 11:50:44 -0500, Doug Newgard wrote:
>>>
>>> Second, what is /etc/modules?
> This literally takes less than a minute to google.
> According to https://kernel-handbook.alioth.debian.org/ch-modules.html :
> "If some modules are not loaded automatically by udev, but you would like
> them to be loaded during boot, it is possible to force it by listing the
> names of the modules in /etc/modules. This will be scanned for the names of
> the modules (one name per line), which will then be loaded using modprobe.
> For example, a typical /etc/modules might look like:
> loop
> sbp2"

You may want to refer to this page for guidance
https://wiki.archlinux.org/index.php/Kernel_modules#Automatic_module_handling

>> Third, the >> redirection happens as a regular user, regardless the
>> `sudo`, so the command wouldn't work (even if /etc/modules existed).
> Redirection will work as expected if logged in as root.

There is no point in using sudo if logged in as root is there, so the
remark is valid.


-- 
Mauro Santos

Re: [arch-general] wireless driver broadcom b43 random success or failure

[Tinu Weber]

On Sun, Apr 10, 2016 at 11:50:44 -0500, Doug Newgard wrote:
> On Sun, 10 Apr 2016 18:44:07 +0200
> Dieter Wirz  wrote:
>
> > sudo echo b43 >> /etc/modules
>
> First, don't top post.
>
> Second, what is /etc/modules?

Third, the >> redirection happens as a regular user, regardless the
`sudo`, so the command wouldn't work (even if /etc/modules existed).

You should use `tee` instead.


signature.asc
Description: PGP signature

Re: [arch-general] wireless driver broadcom b43 random success or failure

[Doug Newgard]

On Sun, 10 Apr 2016 18:44:07 +0200
Dieter Wirz  wrote:

> sudo echo b43 >> /etc/modules

First, don't top post.

Second, what is /etc/modules?

Re: [arch-general] wireless driver broadcom b43 random success or failure

[Dieter Wirz]

sudo echo b43 >> /etc/modules


On Sat, Apr 9, 2016 at 5:17 PM, message  wrote:
> Readers,
>
> Can anyone explain if the following 'dmesg' outputs are useful to understand
> why after start of X session, the wireless driver fails sometimes? The
> wireless driver does not start automatically and is always started manually
> with the command (as root):
>
> modprobe b43
>
> Below is the dmesg output when the command above, starts successfully the
> wireless driver:
>
> b44: Broadcom 44xx/47xx 10/100 PCI ethernet driver version 2.0
> b44 ssb1:0 eth0: Broadcom 44xx/47xx 10/100 PCI ethernet driver ...
> IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
> b43-phy0: Broadcom 4311 WLAN found (core revision 10)
> b43-phy0: Found PHY: Analog 4, Type 2 (G), Revision 8
> b43-phy0: Found Radio: Manuf 0x17F, ID 0x2050, Revision 2, Version 0
> Broadcom 43xx driver loaded [ Features: PNLS ]
> ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
> IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
> b43-phy0: Loading firmware version 666.2 (2011-02-23 01:15:07)
> IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
> wlan0: authenticate with ...
> wlan0: send auth to ... (try 1/3)
> wlan0: authenticated
> wlan0: associate with ... (try 1/3)
> wlan0: RX AssocResp from ... (capab=0x431 status=0 aid=1)
> wlan0: associated
> IPv6: ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
>
> Below when the 'modprobe b43' fails to start the wireless driver:
>
> b43-phy0: Broadcom 4311 WLAN found (core revision 10)
> b43-phy0: Found PHY: Analog 4, Type 2 (G), Revision 8
> b43-phy0 ERROR: FOUND UNSUPPORTED RADIO (Manuf 0xFFF, ID 0x, Revision
> 15, Version 0)
> b43: probe of ssb0:0 failed with error -95
> Broadcom 43xx driver loaded [ Features: PNLS ]
>
> The only solution seems to be to switch off the computer and start again,
> sometimes 3 or 4 times repeated failure occurs.