Date: Mon, 20 Sep 1999 02:08:44 -0700 From: "David Schwartz" <davids@webmaster.com> To: "Mike Smith" <mike@smith.net.au>, "Poul-Henning Kamp" <phk@critter.freebsd.dk> Cc: <cvs-all@FreeBSD.org> Subject: RE: cvs commit: src/sys/i386/conf files.i386 src/sys/i386/i386 mp_clock.c Message-ID: <000001bf0347$bd39b3e0$021d85d1@youwant.to> In-Reply-To: <199909200805.BAA29815@dingo.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> > > No, the TSC is far superior on UP, (unless destroyed by APM), it > > has roughly 100 times better resolution and is twice as fast to > > query. > > It is, however, as you point out, non-deterministically unreliable. > I'm fairly certain that we should option out the use of the TSC by > default - it is simply biting too many people for too little gain. If it's not too difficult, can you do both? Use the most reliable source (PIIX) as your base time reference, and then use the TSC to 'add on a bit' to make it a bit more precise. This way, even if the TSC is somehow corrupted, your worst-case accuracy will be what your best case with just the PIIX. DS To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?000001bf0347$bd39b3e0$021d85d1>