Date: Wed, 13 Feb 2002 21:27:38 -0800 From: John Merryweather Cooper <john_m_cooper@yahoo.com> To: Kris Kennaway <kris@obsecurity.org> Cc: Roman Shterenzon <roman@xpert.com>, freebsd-stable@freebsd.org Subject: Re: src/share/mk/bsd.cpu.mk Message-ID: <20020213212738.A82456@johncoop.MSHOME> In-Reply-To: <20020213210855.A5744@xor.obsecurity.org>; from kris@obsecurity.org on Wed, Feb 13, 2002 at 21:08:55 -0800 References: <20020214011555.F33346-200000@alchemy.oven.org> <20020213210855.A5744@xor.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Since the Athlon/Duron have CMOV and the K6 does not, isn't it
practically a wash? Whatever, scheduling benefits there might be for
using K6 scheduling may be lost by not managing delay slots as well
(since the pentiumpro has CMOV). Of course, this depends on what gcc
2.95.x actually does in the two cases . . .
On 2002.02.13 21:08 Kris Kennaway wrote:
> On Thu, Feb 14, 2002 at 01:20:27AM +0200, Roman Shterenzon wrote:
>
> > . if ${CPUTYPE} == "k7"
> > -_CPUCFLAGS = -march=k6 # gcc doesn't support athlon yet, but
> it will
> > +_CPUCFLAGS = -march=pentiumpro # gcc doesn't support athlon
> yet, but it will
>
> When I wrote this I was informed that the athlon benefits more from k6
> optimizations than pentiumpro.
>
> Kris
>
--
_
| |V| / ' || MacroHard -- \
\_| | | \_, || the perfection of form over |
----------------------------------|| substance, marketing over |
Web: http://www.borgsdemons.com || performance, and greed over |
AIM: johnmcooper || design . . . |
=====================================================================/
Public Key: http://www.borgsdemons.com/Personal/pgpkey.asc |
=====================================================================\
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020213212738.A82456>