lkml.org 
[lkml]   [1998]   [Apr]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: faster strcpy()
26-Apr-98 20:49 you wrote:
>
> On Sun, 26 Apr 1998, Richard B. Johnson wrote:
>
> > # echo Note that normal strcpy copies about twice as many strings in
> > Note that normal strcpy copies about twice as many strings in
> > # echo one second as does strlen + memcpy.
> > one second as does strlen + memcpy.
>
> On ppro and pII, yes - normal copies 2x as many as strange.
> On pentium, K5, and K6, no - strange copies 1.3x - 1.6x as many as normal.
>
I am found even more strange thing about this test: if I am change LEN
from 0x1000 to 0xFFF or 0xFFE (i.e. second array will be not aligned) I'm
got 2x SPEDUP of "strange strcpy" but more or less the same speed of
"normal strcpy"... Why ? What a hell ? This was Pentium MMX so both
arrays must fit in cache !

> I used the same binary on all of them, compiled with gcc 2.7.2.3, linked
> against glibc-2.0.6.
>
In my case this was glibc-2.0.7 (RedHat's) and pgcc 1.0.2 ...





-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

\
 
 \ /
  Last update: 2005-03-22 13:42    [W:0.110 / U:0.508 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site