Date: Sun, 25 Feb 2007 14:10:54 +0100 From: Harald Schmalzbauer <h.schmalzbauer@omnisec.de> To: freebsd-questions@freebsd.org Cc: philip@radford-online.co.uk Subject: Re: Core 2 Duo Message-ID: <200702251410.54782.h.schmalzbauer@omnisec.de> In-Reply-To: <499c70c0702250110m6dbb46b6s3957cc79d794dc6f@mail.gmail.com> References: <200722582944.079368@odyssey> <499c70c0702250110m6dbb46b6s3957cc79d794dc6f@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Am Sonntag, 25. Februar 2007 10:10 schrieb Abdullah Ibn Hamad Al-Marri: > On 2/25/07, Philip Radford <philip@radford-online.co.uk> wrote: > > In the past on a P4 chip I would put pentium4 in the cputype > > parameter in /etc/make.conf but are there any other settings here I > > can adjust or any by using sysctl. > > I'm using CPUTYPE?=3Dprescott amd it's safe, nocona broke my kernel. I'm no compiler expert, but prescott and nocona are Net-Burst architectures. Core(2) is a pentium-m successor, which was a completely different (x86)=20 architecture, so optimizations for netburst won't have a positive effect, i= f=20 they work at all. I use CPUTYPE?=3Dpentium-m, but I've never done any comparisions to binarie= s=20 compiled without CPUTYPE. After several years observation with CPUTYPE I=20 guess the improovement ist smaller than ever these days, and until now it w= as=20 rarely mesureable. =2DHarry =2D-=20 OmniSEC - UNIX und Windows Netzwerke - Sicher Harald Schmalzbauer =46lintsbacher Str. 3 80686 M=C3=BCnchen +49 (0) 89 18947781 +49 (0) 160 93860101
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200702251410.54782.h.schmalzbauer>