Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Date:   Mon, 03 Aug 2020 10:28:48 -0700
From:   Aloka Dixit <alokad@codeaurora.org>
To:     Johannes Berg <johannes@sipsolutions.net>
Cc:     linux-wireless@vger.kernel.org
Subject: Re: [PATCH v3 1/2] nl80211: Unsolicited broadcast probe response
 support
In-Reply-To: <b8455091e9ab4c0c3690dc3fadd000f1d1ca9165.camel@sipsolutions.net>
References: <20200715230514.26792-1-alokad@codeaurora.org>
 <20200715230514.26792-2-alokad@codeaurora.org>
 <b8455091e9ab4c0c3690dc3fadd000f1d1ca9165.camel@sipsolutions.net>
Message-ID: <4c07745c1fd5c6e19a9b01488ca9e92a@codeaurora.org>
User-Agent: Roundcube Webmail/1.3.9
Sender: linux-wireless-owner@vger.kernel.org
Precedence: bulk

On 2020-08-03 03:18, Johannes Berg wrote:
>> +/**
>> + * enum nl80211_unsol_bcast_probe_resp_attributes - Unsolicited 
>> broadcast probe
>> + *	response configuration. Applicable only in 6GHz.
>> + *
>> + * @__NL80211_UNSOL_BCAST_PROBE_RESP_INVALID: Invalid
>> + *
>> + * @NL80211_UNSOL_BCAST_PROBE_RESP_INT: Maximum packet interval (u32, 
>> TU).
>> + *	Allowed range: 0..20 (TU = Time Unit). IEEE P802.11ax/D6.0
>> + *	26.17.2.3.2 (AP behavior for fast passive scanning.
> 
> nit: that "(" never closes
> 
>> +	tmpl = tb[NL80211_UNSOL_BCAST_PROBE_RESP_TMPL];
>> +	if (tmpl) {
>> +		presp->tmpl = nla_data(tmpl);
>> +		presp->tmpl_len = nla_len(tmpl);
>> +	}
>> 
> So, hmm. Similar question here - what do you do without a template? Or
> OTOH, why do you even need a template? Would you advertise something
> that's *different* from the regular probe response template you already
> get for offloaded probe request/response support?
> 

The template will be same as the probe response, except for the 
destination MAC address as this one is broadcast.
Main reason is that this unsolicited broadcast probe response 
transmission is specific to 6GHz and not applicable to lower bands. 
Having a separate netlink command allowed not messing with the existing 
probe response offload feature which is for all bands. Thanks.

> johannes