on 2016-10-24 Charles wrote:

> on 2016-10-24 Felix Fietkau wrote:
> 
> > On 2016-10-19 23:31, Charles wrote:  
> > > 
> > > Questions:
> > > 
> > > 1. Why does LEDE show four regulatory domain blocks while OpenWrt shows
> > >    only one?    
> > This was added in Linux upstream. LEDE uses a newer version of mac80211.  
> 
> OK.  Problem is, iw as a moving target seems not to be too well
> documented.  Any recommended readings related to how iw reg output
> should to be interpreted?
> 
> 
> > > 3. What does AUTO-BW mean?    
> > It means maximum channel bandwidth is determined by the available
> > frequency range.  
> 
> Thanks!  Is that what AVM FritzBoxes call wireless coexistence?  Can't
> see the relation to regulation, which is rather about primary and
> secondary users /not coexisting/.  Well, as I said, some more iw
> documentation could help understanding (Linux) wireless better.

Think I got this one (at least).  Regulation is about /not
coexisting/ on the same channel while AUTO-BW is about coexistence on
neighbouring channels (within the same band).


> > > 2. What does the "country US: DFS-FCC" blocks tell me when radio is set
> > >    to an ETSI region country (DE) according to uci?    
> > Please try the latest version from my staging tree at:
> > https://git.lede-project.org/?p=lede/nbd/staging.git;a=summary  
> 
> Oh, I'm not a developer, but building custom images with the image
> generator so far.  With the recent enabling of IPv6 in the kernel so
> that it cannot be disabled by removing some packages in the image
> generator, I do have a need to switch to compiling images anyway.  Will
> report back (and more problems) when I have kept up with that.  Thanks
> again!

Not yet there with mastering buildroot, sorry, but since I've seen
f78405f landing in the repository -- which I think is the commit you were
referring to -- I did try out a current nightly build (2016-11-05).  But
I can't observe any change in 'iw reg get' output (and I'm still not
sure if iw is telling me wifi is obeying FCC rules or ETSI rules).

tnx, Charles



_______________________________________________
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Reply via email to