Date: Fri, 12 Nov 2004 11:34:03 +0100 From: "Poul-Henning Kamp" <phk@phk.freebsd.dk> To: John-Mark Gurney <gurney_j@resnet.uoregon.edu> Cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/usr.bin/make job.c job.h main.c make.c Message-ID: <95620.1100255643@critter.freebsd.dk> In-Reply-To: Your message of "Fri, 12 Nov 2004 02:28:57 PST." <20041112102856.GA57546@funkthat.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In message <20041112102856.GA57546@funkthat.com>, John-Mark Gurney writes: >Poul-Henning Kamp wrote this message on Fri, Nov 12, 2004 at 09:59 +0100: >> Please see: >> >> http://phk.freebsd.dk/misc/make.png >> >> It shows the loadaverage for "make -j 12 -s buildworld" with and without >> this code. > >Thanks for doing this! :) > >Am I to guess/read from the graph that the total time to buildworld >was the same w/ and w/o? Yes, for this particular case dual Opteron, 4G ram, make -j 12 builworld) there is no statistically significant difference in overall time. If it had less ram it would probably be an improvement because the limit prevents overcommitting RAM with processes we can't really service with CPU anyway. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?95620.1100255643>