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: <3d09a8c687fc3d37a8be367a518ea174cdc1325d.camel@sipsolutions.net>
Subject: Re: [PATCH] nl80211: Add support to configure allowed frequency
 list for AP operation
From:   Johannes Berg <johannes@sipsolutions.net>
To:     Jouni Malinen <j@w1.fi>
Cc:     Veerendranath Jakkam <quic_vjakkam@quicinc.com>,
        linux-wireless@vger.kernel.org
Date:   Fri, 31 Mar 2023 11:14:14 +0200
In-Reply-To: <ZCW5HVMLHHW2bzKW@w1.fi>
References: <20230328093105.3640436-1-quic_vjakkam@quicinc.com>
         <421e93d78deda5ac9e62c58ee275758a3355a503.camel@sipsolutions.net>
         <ZCW5HVMLHHW2bzKW@w1.fi>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.46.4 (3.46.4-1.fc37) 
MIME-Version: 1.0
Precedence: bulk

On Thu, 2023-03-30 at 19:30 +0300, Jouni Malinen wrote:
> On Tue, Mar 28, 2023 at 11:42:17AM +0200, Johannes Berg wrote:
> > On Tue, 2023-03-28 at 15:01 +0530, Veerendranath Jakkam wrote:
> > > Add support to configure the list of allowed 20 MHz channels for AP
> > > operation using an array of channel center frequencies in MHz, this
> > > configuration is valid until the AP is stopped. Driver shall filter o=
ut
> > > channels on top of this list of channels based on regulatory or other
> > > constraints during channel switch etc. Driver shall stop the AP
> > > operation if all the channels are filtered out during such operation.
> >=20
> > Normally hostapd is doing anything related to the channel picking, so
> > you need to explain why this is needed.
>=20
> hostapd does this for many cases, but there are drivers, including a
> couple of the upstream drivers, that select a channel internally for
> channel switch cases. Those would be able to use this information. Is
> that of sufficient generic interest or do you prefer vendor commands to
> be used for such functionality?

Oh, no, I'm entirely happy with this happening in nl80211, it's just the
age-old question again of whether those upstream drivers will ever even
get support for it, or if we end up with more code that effectively
looks dead as far as upstream is concerned :-)

If this actually came with a driver patch using it, the question
wouldn't even arise, since then it's clear why it's needed and what
for...

johannes