Re: Changing from Minstrel to PID

[Hin-Tak Leung <hintak.leung@gmail.com>]

On Mon, Nov 9, 2009 at 12:11 PM, Martín Ernesto Barreyro
<barreyromartin@gmail.com> wrote:
> 2009/11/7 John W. Linville <linville@tuxdriver.com>:
>> On Fri, Nov 06, 2009 at 08:01:33PM -0300, Martín Ernesto Barreyro wrote:
>>> El vie, 06-11-2009 a las 16:48 -0500, John W. Linville escribió:
>>> > On Fri, Nov 06, 2009 at 06:29:16PM -0300, Martín Ernesto Barreyro wrote:
>>> > > Hello, I have a rtl8187 wireless card and I've allways have problems
>>> > > with the rate control using minstrel. So everytime my kernel updates
>>> > > (recently it went from 2.30 to 2.31) i have to recompile mi kernel
>>> > > setting pid as my default rate control algorithm.
>>> > >
>>> > > Is there a way to change that without compiling the kernel?
>>> >
>>> > No, but I wish there were...patches?  Anyone?
>>> >
>>> > John
>>>
>>> I've been searching and it seems that it could be added something like
>>> modprobe mac80211 rc='pid'
>>>
>>> Would it be a good idea?
>>
>> Well if you are happy with a module option, we already have
>> ieee80211_default_rc_algo.
>>
>> John
>> --
>> John W. Linville                Someday the world will need a hero, and you
>> linville@tuxdriver.com                  might be all we have.  Be ready.
>>
>
> Yes I'm happy, I'll try it.
> Is there any other way that it can be approached?
> --
> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

I suppose echo something into the /proc file system as john suggested
in an earlier post would be even better than a module option. That's a
bit more difficult to module option (module option is pretty much the
blunt hammer and you can replace any behavior with any other bahavor
with it,  I think) as it makes the algorithm switch-able at run-time,
and that might be a little more tricky as one has to probably stop a
few things and restart a few things in the kernel... and I don't mean
debugfs, although since it is already in debugfs it might be easier
just to move that code out to general use.