From owner-freebsd-current Thu Mar 12 14:47:55 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA17854 for freebsd-current-outgoing; Thu, 12 Mar 1998 14:47:55 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from katipo.niwa.cri.nz (home.niwa.cri.nz [131.203.55.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id OAA17652; Thu, 12 Mar 1998 14:47:06 -0800 (PST) (envelope-from wdk@katipo.niwa.cri.nz) Received: from localhost (wdk@localhost) by katipo.niwa.cri.nz (8.8.6/8.8.6) with SMTP id LAA26864; Fri, 13 Mar 1998 11:48:04 +1300 (NZDT) Date: Fri, 13 Mar 1998 11:48:03 +1300 (NZDT) From: Wayne Knowles To: Bruce Evans cc: hasty@rah.star-gate.com, jkh@FreeBSD.ORG, current@FreeBSD.ORG Subject: Re: my worldstone In-Reply-To: <199803120304.OAA12807@godzilla.zeta.org.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Thu, 12 Mar 1998, Bruce Evans wrote: > Building an agreed-on old version would be better (check one out from > cvs). This would defeat my Makefile optimizations :-). > > >... > > 8(?) $TMPDIR should probably be on an mfs. I haven't tried this yet. I would seriously suggest you give it a try. From some benchmarks earlier in the year I found that creating a 8MB mfs for /var/tmp shaved 20 minutes off my make world time vs. using -pipe Since we are playing the worldstone game, here is my results but I will admit to cheating and building with NOCLEAN and NOPROFILE Pentium II 300 MHz, 128MB RAM Adaptec 2940UW with TAGGING and SCB Paging enabled 2 x Cheetah 4.5GB Fast & Wide 10,000 rpm Disks /usr/obj Mounted noatime,async /usr/src Mounted noatime,async /var/tmp 8MB mfs filesystem (seems to be big enough) /etc/make.conf - as shipped empty /usr/obj Build: make -DNOCLEAN -DNOPROFILE -j3 buildworld Build time: 55 minutes and 30 seconds (Wow!) >From memory dropping the NOCLEAN and NOPROFILE options clocked in around 1:15, but that was several weeks ago. The exact figures are at home. As another comparison, I did a make release on -current last weekend and it built in just on 2 hrs 30 minutes - a significant improvement over 12 hrs++ on my old P100 system. A releasestone might be a better benchmark as the master CVS repository valules are used for make.conf and it is almost impossible to change what is built (correct me if I am wrong here) If I get time over the weekend, I am going to backup my system and install the softupdate patches followed by the CAM patches to see what improvement they make. Wayne -- _____ Wayne Knowles, Systems Manager / o \/ National Institute of Water & Atmospheric Research Ltd \/ v /\ P.O. Box 14-901 Kilbirnie, Wellington, NEW ZEALAND `---' Email: w.knowles@niwa.cri.nz To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message