From owner-freebsd-hackers Thu Oct 29 21:05:09 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA29402 for freebsd-hackers-outgoing; Thu, 29 Oct 1998 21:05:09 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from dingo.cdrom.com (castles148.castles.com [208.214.165.148]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA29212 for ; Thu, 29 Oct 1998 21:05:05 -0800 (PST) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (LOCALHOST [127.0.0.1]) by dingo.cdrom.com (8.9.1/8.8.8) with ESMTP id VAA00773; Thu, 29 Oct 1998 21:04:48 -0800 (PST) (envelope-from mike@dingo.cdrom.com) Message-Id: <199810300504.VAA00773@dingo.cdrom.com> X-Mailer: exmh version 2.0.2 2/24/98 To: Shafia Kausar cc: FreeBSD Hackers , shafiak@ee.tamu.edu Subject: Re: Timer Granularity In-reply-to: Your message of "Thu, 29 Oct 1998 22:54:08 CST." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 29 Oct 1998 21:04:46 -0800 From: Mike Smith Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > I am confused, On basis of which parameters is the kernel clock > granularity defined?? especially when multiple clocks are available. At which release level? 2.1.x, 2.2.x or 3.x? As a general rule, the best available clock is used. The techniques used to determine which clock is best, and to ensure accurate counts vary from release to release, with a general trend towards better. > > On Thu, 29 Oct 1998, Mike Smith wrote: > > > > I am new to the FreeBSD OS. I have a few questions for which I could not > > > find answers on the webpages. > > > > > > I am using the FreeBSD version 2.1.5 Relase #2. What is the kernel clock > > > granularity in this version? > > > > 10ms, if I remember correctly. > > > > > > There are a number of clocks being used in FreeBSD. Which clock is used > > > for real time kernel processing? Is this accessible to the users? > > > > No clocks are "accessible to the users". User applications obtain time > > values from the system. > > > > > Has the timer granularity improved in the releases following this release? > > > > Yes. Timer support now offers nanosecond resolution, dependant on the > > particular clock source(s) in use. > > > > > In the /sys/kern/kern_clock.c file the variable time_precision has been > > > initialised to 1microsec, but it has been stated that the resolution > > > decreases depending on whether the external clock is working or not. > > > What is the range of the variation in the resolutuion? > > > > It depends on the hardware in use. On some systems the Pentium TSC is > > available as a time source, in which case time resolution is one CPU > > cycle. The i8254 timecounter generally has an operating frequency > > around 1.19MHz, giving a resolution of approximately 1us. > > > > -- > > \\ Sometimes you're ahead, \\ Mike Smith > > \\ sometimes you're behind. \\ mike@smith.net.au > > \\ The race is long, and in the \\ msmith@freebsd.org > > \\ end it's only with yourself. \\ msmith@cdrom.com > > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-hackers" in the body of the message > -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message