This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Results for optimized string functions
- To: libc-alpha at sources dot redhat dot com
- Subject: Results for optimized string functions
- From: Bonz <bonzini at gnu dot org>
- Date: Thu, 01 Mar 2001 13:38:25 +0100
Here are some results (running the stratcliff program 40 times), coming
from a 266 MHz Pentium II. My functions are always faster, if only by
a small measure (around 2%), than those in glibc. I know the results
might look like fluctuations, but they are consistent across different
runs.
glibc 2.2 11.248s with -D__NO_STRING_INLINES
glibc 2.2 11.215s without -D__NO_STRING_INLINES
generic 11.670s my algorithm, C-coded
i386 11.031s all but strlen use the same algorithm as in
generic
i486 11.028s includes a version of strlen, which emulates
SCASB with CMP/INC/JNE
i586 11.144s uses the same algorithms as in generic
i686 10.982s has no register stalls, slower on Pentiums
The patches will come soon if you are going to approve them.
--
|_ _ _ __
|_)(_)| ) ,'
-------- '-._