Date: Fri, 14 Oct 2005 22:07:29 +0200 (MEST) From: Peter B <pb@ludd.ltu.se> To: bkoenig@cs.tu-berlin.de (=?ISO-8859-1?Q?Bj=F6rn_K=F6nig?=) Cc: freebsd-questions@freebsd.org Subject: Re: TSC Timer "Quality" ? Message-ID: <200510142007.j9EK7TI6011714@brother.ludd.ltu.se> In-Reply-To: <4350832B.5020600@cs.tu-berlin.de> from "=?ISO-8859-1?Q?Bj=F6rn_K=F6nig?=" at Oct 14, 2005 09:18:51 PM
next in thread | previous in thread | raw e-mail | index | archive | help
> >Intel pentium cpu timer tsc: > 'Timecounter "TSC" frequency X Hz quality Y' (from dmesg) > >What is the definition of "quality" ..? > > >I have looked briefly in the /usr/src/sys/i386/i386/tsc.c sourcefile and >the intel-celeron datasheet. Without finding anything useful. > > From src/sys/sys/timetc.h: > 54 int tc_quality; > 55 /* > 56 * Used to determine if this timecounter is > better than > 57 * another timecounter higher means better. > Negative > 58 * means "only use at explicit request". > 59 */ > As far as I understand this, it is used internally only. > Regards > Björn Hmm.. should have looked into the header =) Guess I'll have to check the clock circuit physicaly.. The reason for asking is to explore a way to sample signals asynchronously. But useing timestamps to calculate how it looks when sampled synchronously. Thus avoiding the need for fifo circuitry. Btw, avoid html encoded email pls.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200510142007.j9EK7TI6011714>