Date: Tue, 05 Feb 2002 21:50:36 -0700 (MST) From: "M. Warner Losh" <imp@village.org> To: jdp@polstra.com Cc: hackers@FreeBSD.ORG, phk@critter.freebsd.dk Subject: Re: A question about timecounters Message-ID: <20020205.215036.93760885.imp@village.org> In-Reply-To: <200202051906.g15J68E04216@vashon.polstra.com> References: <91801.1012935515@critter.freebsd.dk> <200202051906.g15J68E04216@vashon.polstra.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <200202051906.g15J68E04216@vashon.polstra.com>
John Polstra <jdp@polstra.com> writes:
: I'm testing that now. But for how long would microuptime have to
: be interrupted to make this happen? Surely not 7.81 seconds! On
: this same machine I have a curses application running which is
: updating the screen once a second. It never misses a beat, and
: userland is very responsive.
Silly hypothesis. It isn't losing 7.81 seconds of time. Rather it is
being interrupt at the wrong time and the wrap detection code works
badly when interrupted at that point. So you aren't blocked for 7.81
seconds, but rather for 0.02 seconds since you are seeing the 7.79
jump. You are basically catching the wrong edge of the phase of the
TSC.
Warner
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020205.215036.93760885.imp>