On Thu, 2023-09-14 at 11:45 +0800, Wen Gong wrote:
> On 9/13/2023 10:58 PM, Johannes Berg wrote:
> > On Wed, 2023-03-15 at 18:58 +0530, Aditya Kumar Singh wrote:
> >
> > > @@ -8650,6 +8660,14 @@ static int parse_reg_rule(struct nlattr *tb[],
> > >
> > > reg_rule->flags = nla_get_u32(tb[NL80211_ATTR_REG_RULE_FLAGS]);
> > >
> > > + if (reg_rule->flags & NL80211_RRF_PSD) {
> > > + if (!tb[NL80211_ATTR_POWER_RULE_PSD])
> > > + return -EINVAL;
> > > +
> > > + reg_rule->psd =
> > > + nla_get_s8(tb[NL80211_ATTR_POWER_RULE_PSD]);
> > > + }
> > Wait ... I'm not sure why we've been adding stuff to this recently, but
> > anyway, this part should only be used by CRDA which is deprecated
> > anyway?
> >
> > So I'd say we shouldn't touch any of the code under #ifdef
> > CONFIG_CFG80211_CRDA_SUPPORT any more since CRDA will not continue to
> > developed with any updates here.
> >
> > johannes
> Oh, I will remove the change in parse_reg_rule() in next version, because
> you asked me to add change in 2
> functions(nl80211_put_regdom()/nl80211_msg_put_channel())
> below, but I added change in one more function(parse_reg_rule()).
>
Yes, we should have visibility on the output - but I don't think we ever
expect the input to carry it this way via nl80211 since that's long
deprecated in favour of the kernel loading it as a 'firmware' file.
johannes
