Re: memove performance of NetBSD

[Channa <channa.kad%gmail.com@localhost>]

Hi,

Chuck Swiger <cswiger%mac.com@localhost>:
> Hi--
>
> On Jan 22, 2010, at 1:07 AM, Channa wrote:
>> I am using NetBSD 3.1 ,I have some issues with the performance of memmove().
>
> You might find the following links of interest:
>
>  http://gcc.gnu.org/projects/prefetch.html
>  http://now.cs.berkeley.edu/Td/bcopy.html
>
> ...as well as a long, detailed thread around here:
>
>  http://lists.freebsd.org/pipermail/freebsd-arch/2007-January/005998.html
>  http://www.freebsd.org/cgi/cvsweb.cgi/~checkout~/src/sys/i386/i386/support.s


Thanks you very much for the information i checked the links.
I am facing the problem for MIPS arch. In NetBSD5.0 also i checked,
they have a common
implementation for all archs in C.
However i tested the algorithm as below:

memmove(dst,src,sizeof(src) // Performance is good
memmove(dst,src+4,sizeof(src) // Performance is good since 'src' is aligned

If I perform memove as below:

memmove(dst,src + 1,sizeof(src) // Performance degrades
memmove(dst,src + 2,sizeof(src) // Performance degrades

since in the above calls of memmove() the source is unaligned.

I checked for the FreeBSD implementation too it dint help me.


Thanks & Regards,
Channagoud