From owner-freebsd-hackers Mon Apr 10 4: 1:38 2000 Delivered-To: freebsd-hackers@freebsd.org Received: from hermes.research.kpn.com (hermes.research.kpn.com [139.63.192.8]) by hub.freebsd.org (Postfix) with ESMTP id A8B1C37B5C7 for ; Mon, 10 Apr 2000 04:01:33 -0700 (PDT) (envelope-from K.J.Koster@kpn.com) Received: from l04.research.kpn.com (l04.research.kpn.com [139.63.192.204]) by research.kpn.com (PMDF V5.2-31 #35196) with ESMTP id <01JO2LBIWN5U0014G6@research.kpn.com> for freebsd-hackers@FreeBSD.ORG; Mon, 10 Apr 2000 13:01:30 +0200 Received: by l04.research.kpn.com with Internet Mail Service (5.5.2650.21) id ; Mon, 10 Apr 2000 13:01:30 +0100 Content-return: allowed Date: Mon, 10 Apr 2000 13:01:29 +0100 From: "Koster, K.J." Subject: RE: What are the best gcc optimization options for Pentium 200 M To: "'Alexey N. Dokuchaev'" Cc: freebsd-hackers@FreeBSD.ORG Message-id: <59063B5B4D98D311BC0D0001FA7E452201313A90@l04.research.kpn.com> MIME-version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-type: text/plain; charset="iso-8859-1" Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > Can you lead us to any of this HOWTO's? Tnx. > Well, many good sys admin books exist, I already mentioned "System Performance Tuning", it's one of the Nutshell handbooks. I don't know the author off the top of my head. The mailing list achives will give you plenty of tips and tricks, although less organised. :) I that book, the author suggests you first create a benchmark and then try to determine what resources and optimizations apply. The author states that a good sysadmin knows exactly what the bottlenecks are in his or her system. I found that buying that book saved me well over the price of it in time and hardware. For me, the benchmark is building a GENERIC kernel, since I do a lot of c++ development. I timed that and found with vmstat and friends that I was using not even 50% of my cpu. I needed more memory (so I now run twm only) and optimize disk access (my /usr is now striped across two disks, each with their own controller) and eliminate disk access (MFS on /tmp, setenv TMPDIR /tmp). Currently, my bottleneck is memory bandwidth, and I will fix that with an Athlon/KX133 combo in a month or so. :-) Frankly, compiler optimizations are not even mentioned. Kees Jan ============================================== You are only young once, but you can stay immature all your life To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message