Re: [CFT] faster athlon/duron memory copy implementation

Daniel Egger (degger@fhm.edu)
25 Oct 2002 15:08:57 +0200


--=-JxhYQnIr9GwFIXINlAgP
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Am Don, 2002-10-24 um 21.28 schrieb Manfred Spraul:

> It seems the via cpu doesn't support prefetchnta. Could you try the=20
> attached version?

egger@tanja:~$ ./via=20
Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $=20

copy_page() tests=20
copy_page function 'warm up run' took 24318 cycles per page
copy_page function '2.4 non MMX' took 35819 cycles per page
copy_page function '2.4 MMX fallback' took 35921 cycles per page
copy_page function '2.4 MMX version' took 24291 cycles per page
Illegal instruction

Unfortunately I have no space for gdb on it right now sow I cannot
easily debug where it crashes.=20

BTW: I did the same thing you did: Remove the calls to the obviously
offending calls to the "fast" versions. I've no idea why the no_prefetch
version doesn't, though...=20

--=20
Servus,
Daniel

--=-JxhYQnIr9GwFIXINlAgP
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: Dies ist ein digital signierter Nachrichtenteil

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.0 (GNU/Linux)

iD8DBQA9uUJpchlzsq9KoIYRAmVIAJ9h1Y48yKomb1a/Grv25i++WV2hHACfbh5W
vrDAO8REn+eq+jxwtwbPBOs=
=4p90
-----END PGP SIGNATURE-----

--=-JxhYQnIr9GwFIXINlAgP--

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/