Date: Wed, 3 Jan 96 21:37 WET From: uhclem@nemesis.lonestar.org (Frank Durda IV) To: freebsd-hackers@freebsd.org Subject: Re: clock accuracy? Message-ID: <m0tXgUV-000CFLC@nemesis.lonestar.org>
next in thread | raw e-mail | index | archive | help
[2]From: John Hay <jhay@mikom.csir.co.za> [2]The way I see it is that the problem isn't in the Pentium-timer's [2]accuracy, it is determining the the clock speed of the Pentium [2]accurately. The times when my Pentium is probed as a 90MHz Pentium [2]the clock runs ok. It is when it is probed as 89.xx or even slower [2]that the clock runs too fast. Or you have an Energy Star Pentium computer, where the system deliberately slows the processor speed down when the system is idle, thus saving electricity. Even if this mode doesn't work under FreeBSD, having a master clock that can vary its frequency on demand should give pause as to whether that clock is reliable enough to rely on for TOD. Since the processor allows a +/-10% deviation in clock speed (see Intel databook), the CPU doesn't care about the speed drifting around, but if we use it as a time-piece, you will. On the two Pentium boards with Energy Star I have had access to, the Pentium instruction clock is less accurate than the 8254. Frank Durda IV <uhclem@nemesis.lonestar.org>|"The Knights who say "LETNi" or uhclem%rwsystr.nkn.net (Fastest Route) | demand... A SEGMENT REGISTER!!!" ...letni!rwsys!nemesis!uhclem |"A what?" |"LETNi! LETNi! LETNi!" - 1983
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?m0tXgUV-000CFLC>