From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christian Riesch Subject: Re: [PATCH] net:cpsw: Pass unhandled ioctl's on to generic phy ioctl Date: Wed, 05 Feb 2014 12:06:29 +0100 Message-ID: <52F21B35.9000502@omicron.at> References: <1391500242-10554-1-git-send-email-stefan.sorensen@spectralink.com> <1391511050.3003.21.camel@deadeye.wl.decadent.org.uk> <1391526492.7871.8.camel@e37108.spectralink.com> <1391550719.3003.33.camel@deadeye.wl.decadent.org.uk> <20140205071250.GA4071@netboy> <1391595820.3003.60.camel@deadeye.wl.decadent.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: =?UTF-8?B?IlPDuHJlbnNlbiwgU3RlZmFuIg==?= , "davem@davemloft.net" , "netdev@vger.kernel.org" , "mugunthanvnm@ti.com" To: Ben Hutchings , Richard Cochran Return-path: Received: from mail-ea0-f177.google.com ([209.85.215.177]:55383 "EHLO mail-ea0-f177.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751369AbaBELGh (ORCPT ); Wed, 5 Feb 2014 06:06:37 -0500 Received: by mail-ea0-f177.google.com with SMTP id n15so144362ead.22 for ; Wed, 05 Feb 2014 03:06:36 -0800 (PST) In-Reply-To: <1391595820.3003.60.camel@deadeye.wl.decadent.org.uk> Sender: netdev-owner@vger.kernel.org List-ID: On 2014-02-05 11:23, Ben Hutchings wrote: > On Wed, 2014-02-05 at 08:12 +0100, Richard Cochran wrote: >> On Tue, Feb 04, 2014 at 09:51:59PM +0000, Ben Hutchings wrote: >>> >>> Right. If all versions of CPSW include hardware timestamping then >>> bother with PHY timestamping at all? And why make CONFIG_TI_CPTS >>> configurable? >> >> On the one hand, PHY time stamping is more accurate and offers >> synchronization performance that is measurably better than MAC time >> stamping. > > I suppose that depends on how much jitter there is in the PHY? Not only jitter, asymmetries of the PHY delays must be well known, too,= =20 and these data are not always available from datasheets. If someone is interested in measurement results: We did a few=20 experiments last year for 100 Mbit Ethernet and compared MAC=20 timestamping of a P2020 with the PHY timestamping in the DP83640 (see=20 Table III in [1]). Christian [1] C. Riesch, C. Marinescu, and M. Rudigier, "Measurement of egress an= d=20 ingress delays of PTP clocks", in 2013 International IEEE Symposium on=20 Precision Clock Synchronization for Measurement Control and=20 Communication (ISPCS), Lemgo, Germany, Sep. 22=E2=80=9327, 2013, pp. 11= 3=E2=80=93118. http://www.riesch.at/christian/ISPCS2013_Measurement_of_egress_and_ingr= ess_delays_of_PTP_clocks.pdf