--mxv5cy4qt+RJ9ypb
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

M. Edward (Ed) Borasky:

> Last time I looked, the difference between no optimization
> whatsoever and "-O3 -march=3D<your chip here>" was about 30 percent.

What benchmarks did you use? In my code=E2=80=99s case, the difference betw=
een
empty CFLAGS and CFLAGS=3D'-O3 -march=3Dnative' is minimal (Athlon 64 X2).

(gcc=E2=80=99s man page says -march implies the same -mtune, and that =E2=
=80=98native=E2=80=99
is inteligently handled to mean whatever arch is the best in my case.)

> BTW ... 64-bit compiled is slower than 32-bit compiled on a 64-bit
> chip, too ... cache sizes, alignments, and such, I suspect, though
> I haven't taken the time to profile it.

That=E2=80=99s interesting. Can I build 32-bit Ruby and use it inside my x8=
6_64
system? If so, how? (Sorry, I=E2=80=99m a total novice when it comes to thi=
s.)

-- Shot
--=20
summer's essence drowns
in gloomy eve of winter
wind blows bricktext in   -- Alan J Rosenthal

--mxv5cy4qt+RJ9ypb
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQFI+9XPi/mCfdEo8UoRAhhmAJ4uZ42i5Chh6CvQLwc44q/POXj9bACg0hIz
zqVhC4Yi+pzOuWaperqN9CM=
=DVtd
-----END PGP SIGNATURE-----

--mxv5cy4qt+RJ9ypb--