From owner-freebsd-arch Tue Nov 21 4:29:49 2000 Delivered-To: freebsd-arch@freebsd.org Received: from pcnet1.pcnet.com (pcnet1.pcnet.com [204.213.232.3]) by hub.freebsd.org (Postfix) with ESMTP id B417337B4C5 for ; Tue, 21 Nov 2000 04:29:42 -0800 (PST) Received: (from eischen@localhost) by pcnet1.pcnet.com (8.8.7/PCNet) id HAA17573; Tue, 21 Nov 2000 07:29:19 -0500 (EST) Date: Tue, 21 Nov 2000 07:29:19 -0500 (EST) From: Daniel Eischen To: Julian Elischer Cc: arch@FreeBSD.ORG Subject: Re: Threads (KSE etc) comments In-Reply-To: <3A1A2A26.4CF0B849@elischer.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Mon, 20 Nov 2000, Julian Elischer wrote: > CC's trimmed, > > which group should stay? SMP or ARCH? I've stripped SMP. KSEs were first brought up under -arch, and for the most part, we'd be having the same discussion without SMP availability. Feel free to change it to -smp if you wish. > Daniel Eischen wrote: > > > > With the exception of where the thread > > scheduler (UTS hereafter) allocates/requests one KSEG with > > exactly one KSE. > > there is no reason why ANY KSEG need sto be limitted to one KSE > by the kernel unless there is only one CPU. > The PTHREAD_SCOPE_PROCESS, KSEG may have one or more KSEs > assigned to it depending > on whether the UTS wants to create more or not.... It's been (mostly) decided that a PTHREAD_SCOPE_PROCESS thread runs by itself in it's own KSEG and KSE. This is what I am referring to above. And I'm not saying the kernel limits it to 1 KSE, the UTS does this because it is running a scope system thread and doesn't want another KSE. > > You have to be careful with terminology. If we're going by > > what Jason has defined in his paper, the KSEG is the entity > > that has the quantum, not the KSE. So the KSEGs would be > > limited to the permitted number of child processes. > > yes, but that gives the ability to use M times as much CPU as a > nonthreaded process. Whatever. Create yet another resource limit then. I think we all know that there has to be some limit. Exactly what the limit is, can be decided later. > > If you want to have a separate quantum for each KSE, then > > you can probably eliminate the KSEG. I've made this comment > > also. > > It's possible that we may be able to do so, but > not yet.. I'm willing to work with it either way, but it could make it easier from the kernels point of view if you did get rid of the KSEG. -- Dan Eischen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message