Hi Alexei,
> Channels is a bitmap of 2.16GHz (normal) channels 1..6
> bw_config defines the allowed combinations (bonding) of these "normal"
> channels.
But bw_config is not itself some sort of bitmap, it's just a kind of
enum value, right? Maybe we should actually use an enum for it rather
than having the hardcoded values?
> Let's say driver reports support for channels number 1,2,3
> (channels=0x7).
> Example #1: driver reports bw_config=5
> bw_config=5 allows combinations of 1 or 2 channels.
> It means driver can operate in one of these combinations:
> Channel 1 only
> Channel 2 only
> Channel 3 only
> Channels 1+2 (cb2)
> Channels 2+3 (cb2)
OK.
> The primary channel BTW must be one of the operational channels so if
> driver choose channels 1+3 then primary channel cannot be 2.
Makes sense.
> > It seems to me that you should, however, expose n_bonded_ch to
> > userspace? We do expose all the details about the bitrate normally, see
> > nl80211_put_sta_rate(). We do calculate the bitrate to expose that too,
> > but do expose all the details too.
> >
> > Hmm. Looking at that, it looks like DMG doesn't actually do that. So
> > perhaps not, though it seems to me that it ought to be possible?
>
> We will look into that and address in separate patch.
Right, separate patch definitely makes sense. Thanks!
> > > + /* check bw_config against contiguous edmg channels */
> > > + switch (chandef->edmg.bw_config) {
> > > + case 4:
> > > + case 8:
> > > + case 12:
> > > + if (max_continuous < 1)
> > > + return false;
> > > + break;
I think that indeed this could be clearer if you have an enum.
johannes
