From owner-freebsd-bugs Mon Jul 28 12:55:59 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA28190 for bugs-outgoing; Mon, 28 Jul 1997 12:55:59 -0700 (PDT) Received: from zibbi.mikom.csir.co.za (zibbi.mikom.csir.co.za [146.64.24.58]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA28181 for ; Mon, 28 Jul 1997 12:55:54 -0700 (PDT) Received: (from jhay@localhost) by zibbi.mikom.csir.co.za (8.8.6/8.8.5) id VAA14196; Mon, 28 Jul 1997 21:55:21 +0200 (SAT) From: John Hay Message-Id: <199707281955.VAA14196@zibbi.mikom.csir.co.za> Subject: Re: kern/4112: Re: PPSCLOCK kernel diffs In-Reply-To: <199707280830.BAA17961@hub.freebsd.org> from Bruce Evans at "Jul 28, 97 01:30:01 am" To: bde@zeta.org.au Date: Mon, 28 Jul 1997 21:55:21 +0200 (SAT) Cc: freebsd-bugs@hub.freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-bugs@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > 2nd, the lack of a event serial number > > makes it impossible to detect extra or missing pulses. I'm to blame for this. At the time I submitted the code I did not know about the ppsclockev structure. If I had known I would have used that. > > > The current layout has a new include file, sys/ppsclock.h. > > It might be a better idea to move the contents of this file > > into sys/ttycom.h and rename the two new ioctls so they > > fit in with the other TIO* ioctls. > > Do you want the whole LBL copyright on the merged version? We should not take the decision to move it to ttycom.h lightly. The xntpd code check for ppsclock.h, so if we change that we will have to teach xntpd about it. Not that it is impossible, it's not too difficult to get changes back to the xntpd people, but if we can avoid incompatabilities, we should think about it carefully. > John Hay tried calling hardpps() from siointr1(). I don't like this, > because hardpps() is probably too slow on slow machines. > I have code where I change the original hardpps() function to softpps() and then create a new hardpps() that only store the timestamp and usec values and set a flag. Then I call hardpps() from hardclock() if the flag is set. I have tried to put it inside the 1 second if, but it did not work too well. Outside of it it seemed to work much the same as calling hardpps() directly from sio. Will that help Bruce? A last thing. I feel the same way as Craig. On the machines that I use PPS_SYNC, I just want the best time I can get. I don't use there serial ports for anything else. Actually those machines is doing almost only time. John -- John Hay -- John.Hay@mikom.csir.co.za