From owner-freebsd-current Tue Jan 30 03:41:14 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id DAA08892 for current-outgoing; Tue, 30 Jan 1996 03:41:14 -0800 (PST) Received: from Root.COM (implode.Root.COM [198.145.90.17]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id DAA08874 for ; Tue, 30 Jan 1996 03:41:07 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by Root.COM (8.6.12/8.6.5) with SMTP id DAA00416; Tue, 30 Jan 1996 03:37:57 -0800 Message-Id: <199601301137.DAA00416@Root.COM> X-Authentication-Warning: implode.Root.COM: Host localhost didn't use HELO protocol To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) cc: freebsd-current@FreeBSD.org (FreeBSD-current users) Subject: Re: any ideas about this crash? In-reply-to: Your message of "Tue, 30 Jan 1996 01:58:32 +0100." <199601300058.BAA10336@uriah.heep.sax.de> From: David Greenman Reply-To: davidg@Root.COM Date: Tue, 30 Jan 1996 03:37:57 -0800 Sender: owner-current@FreeBSD.org Precedence: bulk >As Garrett A. Wollman wrote: >> >> > CPU: Pentium (76.42-MHz 586-class CPU) >> > ^^^^^^^^^ >> >> > Mine is 90MHz. >> >> Your timer/counter is bogus. Complain to the vendor. > >My machine at work claims 99.95 MHz since i've upgraded it to -current >today. It displayed 100 MHz with 2.0.5. > >I still believe abusing this timer to generate the system clock is not >the way to go. You cannot sue anybody for the CPU-internal timer not >being accurate within 1E-5 or so (and that's what is needed for a >freestanding clock). Worse than that, it is fundamentally broken on machines that have variable clocks (read: laptops and "green" PCs). The folks in Intel's P6 architecture group were shocked when they heard about what we were doing with the internal cycle counter..."It was never intended to be used that way!". At the very least, we should make it a compile-time option (defaulting to off!). -DG David Greenman Core Team/Principal Architect, The FreeBSD Project