Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <08ff262491040268a6e6b0da0ad32707f091ff2d.camel@sipsolutions.net>
Subject: Re: [PATCH v4] mac80211: disable BSS color collision detection in
 case of no free colors
From:   Johannes Berg <johannes@sipsolutions.net>
To:     Rameshkumar Sundaram <quic_ramess@quicinc.com>
Cc:     linux-wireless@vger.kernel.org
Date:   Mon, 11 Apr 2022 15:33:18 +0200
In-Reply-To: <27b8b92f8fe9883d808a9412370bf23386165482.camel@sipsolutions.net>
References: <1648195961-3811-1-git-send-email-quic_ramess@quicinc.com>
         <27b8b92f8fe9883d808a9412370bf23386165482.camel@sipsolutions.net>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.42.4 (3.42.4-2.fc35) 
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Mon, 2022-04-11 at 15:31 +0200, Johannes Berg wrote:
> On Fri, 2022-03-25 at 13:42 +0530, Rameshkumar Sundaram wrote:
> 
> 
> > +	he_bss_color->color =
> > +		nla_get_u8(tb[NL80211_HE_BSS_COLOR_ATTR_COLOR]);
> > +	he_bss_color->enabled =
> > +		!nla_get_flag(tb[NL80211_HE_BSS_COLOR_ATTR_DISABLED]);
> > +	he_bss_color->partial =
> > +		nla_get_flag(tb[NL80211_HE_BSS_COLOR_ATTR_PARTIAL]);
> > +
> > +	return 0;
> > +}
> > +
> >  static int nl80211_parse_beacon(struct cfg80211_registered_device *rdev,
> >  				struct nlattr *attrs[],
> >  				struct cfg80211_beacon_data *bcn)
> > @@ -5204,6 +5228,12 @@ static int nl80211_parse_beacon(struct cfg80211_registered_device *rdev,
> >  		bcn->ftm_responder = -1;
> >  	}
> >  
> > +	if (attrs[NL80211_ATTR_HE_BSS_COLOR]) {
> > +		err = nl80211_parse_he_bss_color(attrs[NL80211_ATTR_HE_BSS_COLOR],
> > +						 &bcn->he_bss_color);
> > +		if (err)
> > +			return err;
> > +	}
> 
> Could use a newline after this.
> 
> But not sure I get it - doesn't that always disable BSS coloring now if
> you do a beacon update that didn't include the attribute? Did hostapd
> always include it in the past?
> 

Seems like we should track whether it was actually included, and only if
the attribute was included then mac80211 knows - ie. have something like

 struct cfg80211_beacon_data {
  ...
+	struct cfg80211_he_bss_color he_bss_color;
+	bool he_bss_color_valid;

or so?