Re: On K7, -march=k6 is good (Was Re: Why no -march=athlon?)

[M. R. Brown]

--OBd5C1Lgu00Gd/Tn
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

* Benoit Poulot-Cazajous <[e-mail]> on Wed, Dec 19, 2001:

>=20
> But gcc-2.95,x _supports_ "-march=3Dk6", and we should use that instead of
> "-march-i686".
>=20

No, k6 !=3D athlon.  IIRC, the i686 optimization is closer to the Athlon th=
an
the k6 opt.

>=20
> before the patch :
> 1017.92user 261.80system 24:39.89elapsed 86%CPU
> 706.33user 160.79system 16:23.61elapsed 88%CPU
> 1787.38user 418.76system 43:35.97elapsed 84%CPU
>=20
> after the patch :
> 1018.42user 253.85system 24:44.68elapsed 85%CPU
> 704.89user 151.76system 16:16.14elapsed 87%CPU
> 1786.96user 410.76system 43:05.32elapsed 85%CPU
>=20
> The improvement in system time is nice.
>=20

Er, there's not much difference...

Curious, what happens when you compile using gcc 3.0.1 against
-march=3Dathlon?

M. R.

--OBd5C1Lgu00Gd/Tn
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)

iD8DBQE8INTAaK6pP/GNw0URAqrLAJ4h/KAqf1MzdXNnAHOikrQkWsqOSwCfSEGB
j65gJwFKLdnnUya2s4uFnkM=
=Aq3h
-----END PGP SIGNATURE-----

--OBd5C1Lgu00Gd/Tn--