From owner-freebsd-current Wed Nov 26 03:01:19 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id DAA08162 for current-outgoing; Wed, 26 Nov 1997 03:01:19 -0800 (PST) (envelope-from owner-freebsd-current) Received: from korin.warman.org.pl (korin.nask.waw.pl [148.81.160.10]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id DAA08152 for ; Wed, 26 Nov 1997 03:01:11 -0800 (PST) (envelope-from abial@korin.warman.org.pl) Received: from localhost (abial@localhost) by korin.warman.org.pl (8.8.8/8.8.5) with SMTP id MAA20724 for ; Wed, 26 Nov 1997 12:03:03 +0100 (CET) Date: Wed, 26 Nov 1997 12:03:02 +0100 (CET) From: Andrzej Bialecki To: freebsd-current@freebsd.org Subject: SMP: binding process to CPU Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi! I have a question concerning current implementation of SMP. Is it possible to assign specific process (or thread) to specific CPU-thing (not necessarily the actual CPU, perhaps other cpu-kind resource)? Something like Sun's LWP... Also, I have very vague understanding of how the scheduler assigns time slices of both processors to different processes. I think I saw (using 'top') that the same process was being scheduled to run on CPU0 as well as CPU1 during consecutive runnable periods... Here's why I'm asking: I have one process which must be as near real-time as possible, and the second process processes it's output (which is low-priority task). So I'd be glad to monopolize one CPU full-time when I need it, and run the rest of the OS on the other one. Is it possible? Will it be possible? :-) Thanks. Andrzej Bialecki ---------------------+--------------------------------------------------------- abial@warman.org.pl | if(halt_per_mth > 0) { fetch("http://www.freebsd.org") } Research & Academic | "Be open-minded, but don't let your brains to fall out." Network in Poland | All of the above (and more) is just my personal opinion. ---------------------+---------------------------------------------------------