On Sunday 06 April 2008 22:16:59 KURT PETERS wrote:
> Although I found the right phy.c file in the kernel I'm using, I can't 
> ensure I'm making the right changes.
>    For instance, it seems like your refs in b43legacy_phy_xmitpower() is 
> different than in my b43legacy file:
> max_pwr = dev->dev->bus->sprom.maxpwr_bg;  vs
> max_pwr = dev->dev->bus->sprom.r1.maxpwr_bg;

Simply remove the .r1

> Also, I'm not sure what the "hunk" is in the appropriate main.c mentioned 
> below.
> >apply to -stable, or the 2.6.25-rcX code, delete the hunk that changes
> >drivers/net/wireless/b43legacy/main.c, as it will not apply to those
> >trees.

I guess there's only one hunk to that file, if Larry talks about _the_ hunk.

>   Is it possible to point me to the location of the rf testing kernel so I 
> can just use that?

sure http://linuxwireless.org/en/users/Download

>   The sad part is that I feel like I should be able to tell what's going on, 
> but it's hard to interpret code with so little in-line comments.  Not trying 

I think the variable names in the txpower code are more than descriptive enough.
Or what's your guess what "max_pwr" could mean? :)

> to take a dig at anyone, just wish I could contribute somehow.  It didn't 
> seem like even the offset locations in
> http://bcm-specs.sipsolutions.net/SHM
> and
>   http://bcm-specs.sipsolutions.net/PHYRegister
> were the same as those referenced in the code. For instance, I noticed 
> reading and writing in lines like:
> b43legacy_phy_write(dev, 0x04A5, (b43legacy_phy_read(dev, 0x04A5)
>                                         & 0x00FF) | 0x5700);
> but cannot tell where 0x04A5 is.
>    I was attempting at replacing those hex references with a readable def 
> value, but cannot even do that.

Welcome to the world of a reverse engineered driver where you don't know
the meaning of most of the registers... .

-- 
Greetings Michael.
_______________________________________________
Bcm43xx-dev mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev

Reply via email to