From owner-freebsd-questions Tue Apr 4 18:33:17 2000 Delivered-To: freebsd-questions@freebsd.org Received: from astro.psu.edu (lonestar.astro.psu.edu [128.118.147.184]) by hub.freebsd.org (Postfix) with ESMTP id 3A46537B9EE; Tue, 4 Apr 2000 18:33:13 -0700 (PDT) (envelope-from kopts@astro.psu.edu) Received: from volk.astro.psu.edu (volk.astro.psu.edu [128.118.147.21]) by astro.psu.edu (8.9.3/8.9.1) with ESMTP id VAA23202; Tue, 4 Apr 2000 21:33:07 -0400 (EDT) Date: Tue, 4 Apr 2000 21:32:46 -0400 (EDT) From: Alexey Koptsevich To: freebsd-questions@freebsd.org, freebsd-smp@freebsd.org Subject: priorities Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hello, I noticed strange thing: when several computational tasks run, real priority may not depend on nice value. In the example below (FreeBSD 3.4, dual PIII) task "sfp" runs faster than others if nice values are equal, and slower if its nice value is less. In both cases real priorities of all 3 tasks are equal. Both examples are taken at "equilibrium". Are any reasons for such scheduling? Please copy reply to me. Yours, Alex PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND 42069 uv 105 20 5412K 4868K CPU0 0 93:27 69.09% 69.09% h.out 40681 bns 105 20 10660K 10068K RUN 1 397:03 65.97% 65.97% a.out 42314 kopts 105 5 2964K 820K RUN 1 30:26 61.33% 61.33% sfp PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND 42314 kopts 105 20 2964K 820K RUN 0 32:14 69.92% 69.92% sfp 42069 uv 105 20 5584K 5040K RUN 1 95:12 64.21% 64.21% h.out 40681 bns 105 20 10676K 10084K CPU1 0 398:49 62.65% 62.65% a.out To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message