From owner-freebsd-questions Mon May 8 14: 0:57 2000 Delivered-To: freebsd-questions@freebsd.org Received: from bico-de-lacre.iqm.unicamp.br (bico-de-lacre.iqm.unicamp.br [143.106.51.234]) by hub.freebsd.org (Postfix) with ESMTP id 4767037BC34 for ; Mon, 8 May 2000 14:00:52 -0700 (PDT) (envelope-from atabraga@bico-de-lacre.iqm.unicamp.br) Received: by bico-de-lacre.iqm.unicamp.br (Postfix, from userid 1001) id 005D9197D; Mon, 8 May 2000 18:00:36 -0300 (EST) Subject: Re: FreeBSD as a real-time solution? In-Reply-To: <3916FE19.DF7E47C4@tdnet.com.br> from Gustavo Vieira Goncalves Coelho Rios at "May 8, 2000 05:49:13 pm" To: kernel@tdnet.com.br (Gustavo Vieira Goncalves Coelho Rios) Date: Mon, 8 May 2000 18:00:36 -0300 (EST) Cc: clabrown@granitepost.com (Clarence Brown), freebsd-questions@freebsd.org X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Message-Id: <20000508210036.005D9197D@bico-de-lacre.iqm.unicamp.br> From: atabraga@bico-de-lacre.iqm.unicamp.br (Ataualpa Albert Carmo Braga) Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Maybe that will be better discussed in advocacy... Ata. > Clarence Brown wrote: > > > > I'm not a FreeBSD expert, but in "Dr. Dobb's Journal" #310 March 2000 > > (www.ddj.com) there is an article title "Real-Time Linux" that discusses 2 > > projects to bring real time to Linux. Perhaps these might also be applied to > > FreeBSD? > > > > Also, define "precise time measurements". What is the rate and granularity > > that you need? If your DOS system is "working", why do you want to "fix" it? > > > > > > Mentioned in the article: > > > > RTLinux > > at http://www.rtlinux.org, > > > > Real Time Application Interface, (an offshoot of RTLinux with better > > timers?) > > at http://www.aero.polimi.it/projects/rtai > > > > Cla. > > > The solutions for real-time needs should be design from scratch! > Where linux designed to be real time OS? I guess no! Real Time solution > requires a total different approach than the classic kernel model used > by Free/Open/Linux, etc...... > > That's not only the point! What about the correctness of the code? How > can prove that it kernel is full deterministc ? Is FreeBSD/Linux Full > deterministc? If Linux/BSD do manage garbage collection, it's not full > deterministc. What happens if you have to service to a serial port > request and the kernel is performming memory related actions! Garbage > collection cannot be interruped! You lose the information that's > arriving at the serial port........ > > linux were not desgined to be reail time! So, don't use it. > Again: try something like QNX/Neutrino (http://www.qnx.com) > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-questions" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message