From owner-freebsd-current@FreeBSD.ORG Wed Aug 10 13:05:30 2005 Return-Path: X-Original-To: freebsd-current@freebsd.org Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8A75A16A41F for ; Wed, 10 Aug 2005 13:05:30 +0000 (GMT) (envelope-from snort_sam@yahoo.com) Received: from web54409.mail.yahoo.com (web54409.mail.yahoo.com [206.190.49.139]) by mx1.FreeBSD.org (Postfix) with SMTP id D41F643D55 for ; Wed, 10 Aug 2005 13:05:25 +0000 (GMT) (envelope-from snort_sam@yahoo.com) Received: (qmail 92161 invoked by uid 60001); 10 Aug 2005 13:05:25 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=ISaZuuyPmQwCYzPc98SNP9l2Gptq73g4yWuiQ/urQndvOtoDFsofYTkFfpbM+/ewcpBqZDgpZouDwuqdBNrO8J/FVru47w6wAkoldNz/BeCQtRTn9Deo5UqDaJ9tkhik5CqTE1WJMQ+hGFck01vDj2PDjeyx+xkanYsGL0uG0sA= ; Message-ID: <20050810130524.92159.qmail@web54409.mail.yahoo.com> Received: from [203.201.98.189] by web54409.mail.yahoo.com via HTTP; Wed, 10 Aug 2005 06:05:24 PDT Date: Wed, 10 Aug 2005 06:05:24 -0700 (PDT) From: Bsderss To: Robert Watson , Momtchil Momtchev In-Reply-To: <20050810131518.T22763@fledge.watson.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Cc: freebsd-current@freebsd.org, Bsderss Subject: Re: RTCoreBSD X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 10 Aug 2005 13:05:30 -0000 --- Robert Watson wrote: > > On Wed, 10 Aug 2005, Momtchil Momtchev wrote: > > > No, kqueue/kevent is not realtime, as the whole > FreeBSD kernel doesn't > > support hard real-time scheduling. In order to > have hard real-time > > scheduling you should have a fully preemtible > kernel, which schedules > > and preempts everything (even interrupt handlers). > The RTCoreBSD uses a > > two-kernel approach with a real-time microkernel > which runs the FreeBSD > > kernel as a process, providing a virtual interrupt > controller. The > > real-time processes run directly on the > microkernel and can't directly > > use the FreeBSD kernel services. > > While FreeBSD is not a hard realtime system, it does > have increasing > propertis of one: FreeBSD 6.0 ships with kernel > preemption enabled by > default, and the priority propagation and priority > management in the SMPng > locking primitives moves in that direction also. > FreeBSD will preempt one > running ithread with one associated with a new > interrupt if the scheduler > decides that's appropriate based on their > priorities. With Giant off most > of the kernel, a lot of problems with deferred > processing due to large > lock size have gone away. > What will be this affect to our multi-threaded and multiplex programming? Sam > Robert N M Watson > _______________________________________________ ____________________________________________________ Start your day with Yahoo! - make it your home page http://www.yahoo.com/r/hs