Re: [PATCH wpan-tools] info: add frequency to channel listing for phy capabilities

[Alexander Aring <alex.aring@gmail.com>]

On Tue, Jun 02, 2015 at 12:34:51PM +0300, Lennert Buytenhek wrote:
> On Tue, Jun 02, 2015 at 09:38:27AM +0200, Christoffer Holmstedt wrote:
> 
> > > > This is for the same reason as above. Alignment in the output, it is a trade
> > > > off between "readability" in two different ways. Alignment in the output or
> > > > printing the full word "Unknown". I will change this to "Unknown" and we can
> > > > change it at a later date if it ever shows up and mess up alignment. In any
> > > > case if "Unknown" shows up the actual frequency should be added instead of
> > > > fixing alignment. ;)
> > > > 
> > > 
> > > Then maybe some "-" instead "unkown", is this better, or it's too small
> > > then?
> > > 
> > 
> > The more I think about it the more I feel "Unknown" is better. If for some
> > reason the channel page and channel combination is not listed and "Unknown" is
> > printed the last thing the user will care about is alignment in the output.
> > 
> > Channel page and channel number information will not change that often
> > so we should be able to keep up with future changes (additions). I have
> > tried to get hold of the 802.15.4-2015 draft from March/April but
> > without success. The local university will charge me ~350€ or something
> > in that region for it. I wanted it to see if any new channel
> > pages/channels have been added since -2011 edition.
> 
> There are some amendments already available for download that
> document new channel pages:
> 
> 802.15.4f-2012.pdf: channel pages 7, 8
> 802.15.4g-2012.pdf: channel pages 9, 10
> 802.15.4j-2013.pdf: channel page 11
> 802.15.4k-2013.pdf: channel page 12
> 802.15.4p-2014.pdf: channel page 13

mhhh, I looked into this. At the moment we have a highest channel number
define (don't asking me where this comes from, was before there) [0]. I
thought somewhere from the standard, that's why it's inside
ieee802154.h.

I see for example 802.15.4k-2013 that the "phyCurrentChannel" range is
above of them. e.g. Table 681. Currently we save it in some array file
and the current channel is a bitfield [1].

So I think if we support it in kernel, we need some more granularity
representation of the channels/pages, maybe also introduce some enums
about band identifier information and such things.

I want to note that only.


The current nl802154 for asking supported channel/page should be easily
extendable since we doing for_each_nested stuff there. But Currently there
is a lack of support for channel/page stuff, the datatype is also "u8" and
seems also to be something which fits not in there like "current channel
range: 1-416".

Oh well, that would be fun to change it again. :-)

- Alex

[0] http://lxr.free-electrons.com/source/include/linux/ieee802154.h#L42
[1] http://lxr.free-electrons.com/source/include/net/cfg802154.h#L85