On Tue, 2011-11-08 at 12:58 -0800, Ben Greear wrote:
> >> +bool ieee80111_cfg_override_disables_ht40(struct ieee80211_sub_if_data *sdata)
> >> +{
> >> + if ((sdata->u.mgd.ht_capa_mask.cap_info&
> >> + IEEE80211_HT_CAP_SUP_WIDTH_20_40)&&
> >> + !(sdata->u.mgd.ht_capa.cap_info&
> >> + IEEE80211_HT_CAP_SUP_WIDTH_20_40))
> >> + return true;
> >> + return false;
> >
> > Would it really go above 80 cols if you didn't line-wrap it? Maybe
> > remove the extra sets of parentheses? And even if it goes to a little
> > bit above 80 it's still be more readable without the wrapping ...
>
> It is more readable w/out the wrapping, but hard to know when
> patches get rejected about that or not, so I tried to keep
> checkpatch happy. If you'll take slightly longer lines I'll
> happily un-wrap it.
Please. I'd rather go over a bit than read the above :-)
> > One thing I don't quite understand: Why don't you calculate the HT caps
> > to use upon assoc request, and then store *those* instead, then you
> > wouldn't have to check the overrides every time.
>
> Adding more state just gives more places to mess up that
> state or forget to update it somehow. Think of the channel
> pointers in the scan & work code :)
> Not to mention the extra bloat in RAM.
> Since this is not hot-path code, I think having less state
> is well worth the effort.
You don't have more state though, you actually should end up with less
since you can throw away the configured ht_caps & ht_mask? And you
really just shift the access from sband->... to calculated_... stuff.
> > For example here:
> >
> >> if (!(ap_ht_cap_flags& IEEE80211_HT_CAP_40MHZ_INTOLERANT)&&
> >> + !ieee80111_cfg_override_disables_ht40(sdata)&&
> >> (sband->ht_cap.cap& IEEE80211_HT_CAP_SUP_WIDTH_20_40)&&
> >> (hti->ht_param& IEEE80211_HT_PARAM_CHAN_WIDTH_ANY)) {
> >> switch(hti->ht_param& IEEE80211_HT_PARAM_CHA_SEC_OFFSET) {
> >
> > This just adds complexity. If you calculate sdata->used_ht_caps first
> > then you can replace the sband->ht_cap.cap check with an
> > sdata->used_ht_caps.cap check and be done with it, instead of having to
> > check both.
>
> I think that's a bad idea, but will change it if you insist.
I really think that'd be much nicer. As it is now we have to add all
these checks everywhere, if we just calculate it once and then change
places to use it we just have to remember to use the right thing.
johannes
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
