Re: [RFC wireless-next 8/8] wifi: mac80211: set AP NPCA parameters in bss_conf

[Johannes Berg <johannes@sipsolutions.net>]

On Wed, 2026-03-04 at 13:56 +1100, Lachlan Hodges wrote:
> > +		npca = ieee80211_uhr_npca_info(params->uhr_oper);
> > +		if (!npca) {
> > +			chanreq.oper.npca_chan = NULL;
> > +			chanreq.oper.npca_punctured = 0;
> > +		} else {
> > +			npca_params.min_dur_thresh =
> > +				le32_get_bits(npca->params,
> > +					      IEEE80211_UHR_NPCA_PARAMS_MIN_DUR_THRESH);
> > +			npca_params.switch_delay =
> > +				le32_get_bits(npca->params,
> > +					      IEEE80211_UHR_NPCA_PARAMS_SWITCH_DELAY);
> > +			npca_params.switch_back_delay =
> > +				le32_get_bits(npca->params,
> > +					      IEEE80211_UHR_NPCA_PARAMS_SWITCH_BACK_DELAY);
> > +			npca_params.init_qsrc =
> > +				le32_get_bits(npca->params,
> > +					      IEEE80211_UHR_NPCA_PARAMS_INIT_QSRC);
> > +			npca_params.moplen =
> > +				le32_get_bits(npca->params,
> > +					      IEEE80211_UHR_NPCA_PARAMS_MOPLEN);
> > +			npca_params.enabled = true;
> > +		}
> 
> Minor nit, and I can't really comment on the actual UHR code - but would
> it be nicer to have
> 
> if (npca) {
> 	npca_params.min_dur_thresh = ...
> 	...
> } else {
> 	chanreq.oper.npca_chan = NULL;
> 	chanreq.oper.npca_punctured = 0;
> }
> 
> instead? I have no strong feelings of course, I was just perusing and
> felt it was a bit strange to read.

Fair point. I had a few things slightly differently during development
(and the non-NPCA had some goto/return) and just never went back to re-
structure it entirely when I removed that wart.

johannes