Received: from mail.eecis.udel.edu by whimsy.udel.edu id aa22569;
2 May 1998 16:16 EDT
Date: Sat, 2 May 1998 16:14:26 EDT
From: Dave Mills <mills@huey.udel.edu>
To: John Hay <jhay@mikom.csir.co.za>
cc: Harlan Stenn <stenn@whimsy.udel.edu>, mills@udel.edu
Subject: Re: Some patches to ntp-4.0.72h
Message-ID: <199805021614.aa13390@huey.udel.edu>
John,
I have no burn with all of your patches except the one for kernel PPS.
Why are you using the atom clock driver if you have kernel PPS?
You shouldn't need it, since the PPS signal is enabled automatically
and the atom offsets are ignored. Sometimes I do use atom anyway so
I can see some history for performance evaluation. If this is your
intent, you should move your patch outside the termios conditional.
See how this is done for Solaris. Maybe what you need is some
autoconfigure switch that results in similar behavior as that PPS wadge.
You should see the PPS API internet draft, which is now some months
old. Sun and Digital and I did the original draft, I'll send you the
jabber resulting from the recent discovery of some folks who just
discovered it.
Dave