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;
Date:   Wed, 26 Apr 2023 07:04:30 -0700
From:   Richard Cochran <richardcochran@gmail.com>
To:     "Stern, Avraham" <avraham.stern@intel.com>
Cc:     "Greenman, Gregory" <gregory.greenman@intel.com>,
        "kuba@kernel.org" <kuba@kernel.org>,
        "linux-wireless@vger.kernel.org" <linux-wireless@vger.kernel.org>,
        "johannes@sipsolutions.net" <johannes@sipsolutions.net>,
        "netdev@vger.kernel.org" <netdev@vger.kernel.org>
Subject: Re: pull-request: wireless-next-2023-03-30
Message-ID: <ZEkvbvIZiUnbK45N@hoboy.vegasvil.org>
References: <20230331000648.543f2a54@kernel.org>
 <ZCtXGpqnCUL58Xzu@localhost>
 <ZDd4Hg6bEv22Pxi9@hoboy.vegasvil.org>
 <ccc046c7e7db68915447c05726dd90654a7a8ffc.camel@intel.com>
 <ZEC08ivL3ngWFQBH@hoboy.vegasvil.org>
 <SN7PR11MB6996329FFC32ECCBE4509531FF669@SN7PR11MB6996.namprd11.prod.outlook.com>
 <ZEb81aNUlmpKsJ6C@hoboy.vegasvil.org>
 <ZEctFm4ZreZ5ToP9@hoboy.vegasvil.org>
 <SN7PR11MB6996324EBF976C507D382C3AFF649@SN7PR11MB6996.namprd11.prod.outlook.com>
 <ZEiP3LDTQ86c4HaN@hoboy.vegasvil.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <ZEiP3LDTQ86c4HaN@hoboy.vegasvil.org>
Precedence: bulk

On Tue, Apr 25, 2023 at 07:43:40PM -0700, Richard Cochran wrote:
> Your design might touch upon a number of points...

I forgot the most important point, IMO:

How will you handle time distribution across a wireless network?

Consider the following simple case.

GPS Radio ------> Station-A ~~~~~~ AP ~~~~~~ Station-B
           1PPS              WiFi      WiFi

Here Station-A should become the PTP server, and Station-B should
become a PTP client.  In 1588, this is accomplished by forming a
spanning tree over the wired network, based on time quality fields
advertised.

AFAICT, the standards provide almost no hint how this is supposed to
work over WiFi.  I'd like to know your plans for solving this aspect.
Just exposing FTM to user space doesn't help all.

It gets even better.

Replace that "AP" with "mesh network" and image Station-B moves around...
What happens next?

Thanks,
Richard