From owner-freebsd-current Tue Sep 15 14:41:21 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA12379 for freebsd-current-outgoing; Tue, 15 Sep 1998 14:41:21 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from smtp01.primenet.com (smtp01.primenet.com [206.165.6.131]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id OAA12251 for ; Tue, 15 Sep 1998 14:40:48 -0700 (PDT) (envelope-from tlambert@usr09.primenet.com) Received: (from daemon@localhost) by smtp01.primenet.com (8.8.8/8.8.8) id OAA08042; Tue, 15 Sep 1998 14:40:21 -0700 (MST) Received: from usr09.primenet.com(206.165.6.209) via SMTP by smtp01.primenet.com, id smtpd008012; Tue Sep 15 14:40:15 1998 Received: (from tlambert@localhost) by usr09.primenet.com (8.8.5/8.8.5) id OAA26615; Tue, 15 Sep 1998 14:40:11 -0700 (MST) From: Terry Lambert Message-Id: <199809152140.OAA26615@usr09.primenet.com> Subject: Re: more data on "calcru: negative time: -nnn usec" To: phk@critter.freebsd.dk (Poul-Henning Kamp) Date: Tue, 15 Sep 1998 21:40:11 +0000 (GMT) Cc: kls@ohare.chicago.com, freebsd-current@FreeBSD.ORG In-Reply-To: <27970.905847193@critter.freebsd.dk> from "Poul-Henning Kamp" at Sep 15, 98 10:13:13 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > APM strikes again. > > You cpu was slowed down while you booted the other two times. > > The RAM is worse, check out the npx0 flags in LINT and try that. > > Poul-Henning This problem also occurs on a Cyrix Media-GX with APM disabled. APM is totally disabled, as is VM86(). We know APM is disable because the SMI on Cyrix Media-GX processors is partially handled through the system BIOS, and the system BIOS does not have code to handle the SMI requests, and the machine would reset if the entry points were called. The machine does not reset. This is pretty clearly a problem with the monoclock not acting like a monoclock, and/or a scaled version of the clock, rather than the monoclock itself, being used for timestamping. The only time the monoclock data should be scaled is when the data is externalized as a wall clock value. I think Bruce Evans point about the diagnostics he suggested in a recent posting on this subject to -current, and the accompanying analysis he did, stands, in the face of the problem occurring on an APM-less system which has no power consumption management in either hardware or software, whatsoever. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message