From owner-freebsd-arch Tue Oct 30 10:52:33 2001 Delivered-To: freebsd-arch@freebsd.org Received: from mail.rpi.edu (mail.rpi.edu [128.113.22.40]) by hub.freebsd.org (Postfix) with ESMTP id EE29F37B405 for ; Tue, 30 Oct 2001 10:52:28 -0800 (PST) Received: from [128.113.24.47] (gilead.acs.rpi.edu [128.113.24.47]) by mail.rpi.edu (8.11.3/8.11.3) with ESMTP id f9UIqHI118762; Tue, 30 Oct 2001 13:52:17 -0500 Mime-Version: 1.0 X-Sender: drosih@mail.rpi.edu Message-Id: In-Reply-To: References: Date: Tue, 30 Oct 2001 13:52:14 -0500 To: Julian Elischer , Peter Wemm From: Garance A Drosihn Subject: Re: 64 bit times revisited.. Cc: Bruce Evans , Nate Williams , arch@FreeBSD.ORG Content-Type: text/plain; charset="us-ascii" ; format="flowed" Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG At 9:52 AM -0800 10/30/01, Julian Elischer wrote: >One added note however.. I am expecting that there will be embedded >systems being deployed NOW that will be in service in 2038. Let us say that is true. All such systems deployed before this second in time -> this one here <- will not work in 2038, because freebsd at this time does not support it. I have no objection if someone wants to do the work to change UFS to use unsigned timestamps for that field, but I expect it'll take at least a few months to implement that and DEBUG it for real. That because the only way to really debug such a change is to run your system with a date which is AFTER 2038, and to do that for a few months. That's the only way to get real-world experience of what will and what will not work if that field is changed to be unsigned. So, not only will today's embedded systems not work, but it'll be a few more months (or at least a few weeks) before any 2038-safe embedded systems can be deployed. How much longer will it be before we have UFS2 and a more-permanent solution? Another year? Two? So you are really only gaining a two-year span where embedded systems will work after 2038, and 2038 is 37 years from now... Still, if someone wants to do that work, then I'm all for the idea (honest, I am!). The only place where I really start objecting is when we start stealing bits from other fields, and thus spreading a single value over several different spots in a 'struct'. Sounds like Julian is volunteering to do the work... :-) -- Garance Alistair Drosehn = gad@eclipse.acs.rpi.edu Senior Systems Programmer or gad@freebsd.org Rensselaer Polytechnic Institute or drosih@rpi.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message