From owner-freebsd-smp Sat Nov 1 09:00:30 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA27172 for smp-outgoing; Sat, 1 Nov 1997 09:00:30 -0800 (PST) (envelope-from owner-freebsd-smp) Received: from mail.cs.tu-berlin.de (mail.cs.tu-berlin.de [130.149.17.13]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA27161 for ; Sat, 1 Nov 1997 09:00:27 -0800 (PST) (envelope-from wosch@cs.tu-berlin.de) Received: from panke.panke.de (anonymous220.ppp.cs.tu-berlin.de [130.149.17.220]) by mail.cs.tu-berlin.de (8.8.6/8.8.7) with ESMTP id RAA20104; Sat, 1 Nov 1997 17:57:04 +0100 (MET) Received: (from wosch@localhost) by panke.panke.de (8.8.5/8.6.12) id QAA01307; Sat, 1 Nov 1997 16:47:32 +0100 (MET) To: "Jordan K. Hubbard" Cc: smp@freebsd.org Subject: Re: Some SMP timing tests. References: <26870.878356552@time.cdrom.com> From: Wolfram Schneider Date: 01 Nov 1997 16:47:28 +0100 In-Reply-To: "Jordan K. Hubbard"'s message of Fri, 31 Oct 1997 19:55:52 -0800 Message-ID: Lines: 21 Sender: owner-freebsd-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk "Jordan K. Hubbard" writes: > The most interesting thing about these numbers was that at "high job > counts", where one would expect performance to start to actually > degrade due to having too many compiles competing for various system > resources, performance did not fall as expected. This leads me to > believe that our make actually artificially limits the parallelism > number to somewhere below 20. I haven't bothered to look into make's > code more thoroughly in verifying this, but that's certainly what it > looks like. If you have a Makefile with 5 targets (e.g. 4 *.c files, 1 manpage), make can only create 5 jobs at once. Not surprising ;-) I guess the average Makefile has 3 targets. -j max_jobs Specify the maximum number of jobs that make may have running at ^^^^^^^^^^^^^^ ^^^ any one time. -- Wolfram Schneider http://www.apfel.de/~wosch/