From mboxrd@z Thu Jan 1 00:00:00 1970 From: Miroslav Lichvar Subject: Re: [PATCH RFC net-next 1/1] ptp: add an ioctl to compare PHC time with system time Date: Fri, 28 Sep 2012 10:50:31 +0200 Message-ID: <20120928085031.GC29438@localhost> References: <20120928075303.GB29438@localhost> <20120928082638.GB17636@netboy.at.omicron.at> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: netdev@vger.kernel.org, David Miller , Jacob Keller , John Stultz To: Richard Cochran Return-path: Received: from mx1.redhat.com ([209.132.183.28]:60430 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753736Ab2I1Iui (ORCPT ); Fri, 28 Sep 2012 04:50:38 -0400 Content-Disposition: inline In-Reply-To: <20120928082638.GB17636@netboy.at.omicron.at> Sender: netdev-owner@vger.kernel.org List-ID: On Fri, Sep 28, 2012 at 10:26:38AM +0200, Richard Cochran wrote: > I am guessing it would be possible to synchronize two PHC devices to > within a few microseconds this way. Probably that is not good enough > to implement a boundary clock, for example, so I have my doubts about > the utility of this. I think with two identical PHCs the error would be much smaller, even if the two fastest consecutive readings took together ~5 microseconds. The error could be measured with a short cable connecting the two ports and compared the TX and RX timestamps, and compensated in the software if it's significant. > But in any case, it is possible, and I think that > feature can wait for now. Ok, thanks. -- Miroslav Lichvar