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

From: Florin Iucha (florin@iucha.net)
Date: Thu Oct 24 2002 - 14:15:01 EST

Next message: chrisl@vmware.com: "Re: writepage return value check in vmscan.c"
Previous message: Rik van Riel: "Re: writepage return value check in vmscan.c"
In reply to: Daniel Egger: "Re: [CFT] faster athlon/duron memory copy implementation"
Next in thread: Manfred Spraul: "Re: [CFT] faster athlon/duron memory copy implementation"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

SIS735 (ECS7S5A mobo)
Duron 1200 MHz
512 MB PC100

Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $

copy_page() tests
copy_page function 'warm up run' took 24171 cycles per page
copy_page function '2.4 non MMX' took 25359 cycles per page
copy_page function '2.4 MMX fallback' took 25224 cycles per page
copy_page function '2.4 MMX version' took 24149 cycles per page
copy_page function 'faster_copy' took 15660 cycles per page
copy_page function 'even_faster' took 15540 cycles per page
copy_page function 'no_prefetch' took 13853 cycles per page

Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $

copy_page() tests
copy_page function 'warm up run' took 24325 cycles per page
copy_page function '2.4 non MMX' took 25414 cycles per page
copy_page function '2.4 MMX fallback' took 25317 cycles per page
copy_page function '2.4 MMX version' took 24345 cycles per page
copy_page function 'faster_copy' took 15718 cycles per page
copy_page function 'even_faster' took 15553 cycles per page
copy_page function 'no_prefetch' took 13855 cycles per page

Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $

copy_page() tests
copy_page function 'warm up run' took 24225 cycles per page
copy_page function '2.4 non MMX' took 25430 cycles per page
copy_page function '2.4 MMX fallback' took 25398 cycles per page
copy_page function '2.4 MMX version' took 24233 cycles per page
copy_page function 'faster_copy' took 15737 cycles per page
copy_page function 'even_faster' took 15584 cycles per page
copy_page function 'no_prefetch' took 13855 cycles per page

"If it's not broken, let's fix it till it is."

41A9 2BDE 8E11 F1C5 87A6 03EE 34B3 E075 3B90 DFE4

application/pgp-signature attachment: stored

- 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/

Next message: chrisl@vmware.com: "Re: writepage return value check in vmscan.c"
Previous message: Rik van Riel: "Re: writepage return value check in vmscan.c"
In reply to: Daniel Egger: "Re: [CFT] faster athlon/duron memory copy implementation"
Next in thread: Manfred Spraul: "Re: [CFT] faster athlon/duron memory copy implementation"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Thu Oct 31 2002 - 22:00:24 EST