This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: -mfpmath=sse question
- From: Kelley Cook <kcook34 at ford dot com>
- To: "Neal D dot Becker" <nbecker at hns dot com>
- Cc: GCC Patches <gcc-patches at gcc dot gnu dot org>
- Date: Tue, 20 Jan 2004 13:01:14 -0500
- Subject: Re: -mfpmath=sse question
- Hop-count: 1
- References: <200401201235.34839.nbecker@hns.com>
- Reply-to: Kelley Cook <kelleycook at yahoo dot com>
> On Athlon-xp, what is the best combination of switches?
That is not an easy question to answer as lots of factors about your
code will come into play. Using "-O2" is generally a good start.
>
> I'm using -mfpmath=sse -msse -m3dnow.
>
> Is this correct? Or is -msse redundant?
This question however is since '-mfpmath=sse' does not imply '-msse' nor
'-m3dnow'.
That being said, assuming that you are looking for compling a binary
that will only run on an athlon-xp (or x86-64) machine than using "-O2
-march=athlon-xp -mfpmath=sse" would imply both '-msse' and '-m3dnow'
and will schedule for that particular processor.