From owner-freebsd-current Sun Feb 17 15:37:16 2002 Delivered-To: freebsd-current@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by hub.freebsd.org (Postfix) with ESMTP id 3C5B137B47B; Sun, 17 Feb 2002 15:36:56 -0800 (PST) Received: (from dillon@localhost) by apollo.backplane.com (8.11.6/8.9.1) id g1HNaup07553; Sun, 17 Feb 2002 15:36:56 -0800 (PST) (envelope-from dillon) Date: Sun, 17 Feb 2002 15:36:56 -0800 (PST) From: Matthew Dillon Message-Id: <200202172336.g1HNaup07553@apollo.backplane.com> To: Michael Smith Cc: Michael Smith , Poul-Henning Kamp , Bruce Evans , freebsd-current@FreeBSD.ORG Subject: Re: ACPI timer is screwed... (was Re: 'microuptime() went backwards ...' using ACPI timer. Shouldn't that be impossible? ) References: <200202172327.g1HNRuA01838@mass.dis.org> Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG :Sounds like we need to smack whoever made your chipset as well. Intel :learned their lesson (finally) with later revisions of the PIIX4. I'm :guessing you're running this against a ServerWorks system. atapci0: port 0x8b0-0x8bf at device 15.1 on pci0 Uh huh. It might be possible to detect the situation during init-time by explicitly looking for a reverse indexed time in a tight loop of maybe a thousand reads, but that would still leave us with a statistical chance of not guessing right. :Interesting. This would be reasonably robust in the ripple-counter case :we have to deal with on the old PIIX4. Have you tried implementing the :above yet, or measuring how much it costs? : :At any rate, please let me know for sure whether you're running on a :ServerWorks board, and I'll see if I can't find a Big Stick to hit them :with. : :Thanks, :Mike I haven't measured the cost (extra loops) but I expect it would stabilize in no more then one additional loop, which would be three counter reads total or roughly the same as your originaln _safe code in the worst case. I think we could default to the _safe version and then explicitly change it to use the fast version if we see specific chipsets which we know to be good. -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message