Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 18 Jan 2005 21:00:15 +0000
From:      Jason Henson <jason@ec.rr.com>
To:        freebsd-questions@freebsd.org
Subject:   Re: Difference between CPUTYPE= in /etc/make.conf
Message-ID:  <1106082015l.37331l.0l@BARTON>
In-Reply-To: <20050118104544.33D7.GERARD-SEIBERT@rcn.com> (from gerard-seibert@rcn.com on Tue Jan 18 11:01:32 2005)
References:  <20050118104544.33D7.GERARD-SEIBERT@rcn.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On 01/18/05 11:01:32, Gerard Seibert wrote:
> I have tried googling for this information, but without any concrete
> results. In the '/etc/make.conf' file, what are the advantages =20
> between
> using the following declaration?
>=20
> CPUTYPE=3Di686
>=20
> Versus
>=20
> CPUTYPE=3Dp4
>=20
> In the above scenario, the number following the letter 'p' could be
> between one and four. Does it make a discernable difference?
>=20
> Thanks!
>=20
> Gerard E. Seibert
> gerard-seibert@rcn.com
>=20
> "I take my wife everywhere, but she keeps finding her way back.'
>=20
> 	-- Anonymous
>

The 686 covers a few processors including the athlon, p4, p3?, and =20
maybe some others.  If you set p4 you should produce code that runs =20
faster on a p4 than 686 would.  Also the code might run slower or not =20
at all on the p3 and athlon.  SSE3 is an example of code used on a p4 =20
not found in other 686s(I think this is still the case) that would case =20
the program not to run on other cpu types, but run faster on a p4.
BTW, iirc p4=3Dpentium4 and it atleast used to be broken and droped back =20
to some lower setting.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1106082015l.37331l.0l>