From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Gleixner Subject: Re: [PATCH 0/2] [RFC] posix clock tuning Date: Thu, 9 Sep 2010 14:50:57 +0200 (CEST) Message-ID: References: <20100909122156.GA2823@riccoc20.at.omicron.at> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-path: In-Reply-To: <20100909122156.GA2823-7KxsofuKt4IfAd9E5cN8NEzG7cXyKsk/@public.gmane.org> Sender: linux-api-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Richard Cochran Cc: netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, john stultz List-Id: linux-api@vger.kernel.org On Thu, 9 Sep 2010, Richard Cochran wrote: > From my point of view, the discussion is going round in circles. There Sorry for coming late and adding more to the general confusion. :) > seems to be agreement that supporting PTP hardware clocks is a good > idea. The objections seem to be about the form of the API. In general, > there are three avenues. > > 1. character device (like hpet) > 2. posix clock api > 3. sysfs > > Or possibly some combination of the three. I fail to see the requirement for any of those. Either the hardware clock is suitable as a clocksource for general consumption, then it should just be used as a clock source. If it's not - e.g. because it's too slow to access - then it just should serve as a reference in the NTP style and steer the timekeeping into sync. As I said in the other reply, I know that you want a global synchronized clock aside of CLOCK_REALTIME, but that's an easy to solve problem. See my other reply for details. Thanks, tglx