Re: Fastest way to find max element in vector
- Subject: Re: Fastest way to find max element in vector
- From: David Gohara <email@hidden>
- Date: Mon, 21 Nov 2005 11:34:06 -0500
- Delivered-to: email@hidden
- Delivered-to: email@hidden
On Nov 21, 2005, at 11:15 AM, Rustam Muginov wrote:
Please excuse me for confusion. I need to find the largest element
across the vector register (vFloat), not across arbitrary array of
floats.
-
Hi Rustam,
You could try something like this:
[gohara:~/misc] gohara% cat max.c
#include <stdio.h>
#include <stdlib.h>
int main(){
vector float mVec = (vector float)(0.,7.,2.3,7.1);
vector float temp;
printf("%vf\n",mVec);
mVec = vec_max(mVec,vec_sld(mVec,mVec,8));
mVec = vec_max(mVec,vec_sld(mVec,mVec,4));
printf("%vf\n",mVec);
return 0;
}
[gohara:~/misc] gohara% gcc max.c -faltivec
[gohara:~/misc] gohara% ./a.out
0.000000 7.000000 2.300000 7.100000
7.100000 7.100000 7.100000 7.100000
It's similar to adding across a vector of floats (see: http://
developer.apple.com/hardware/ve/algorithms.html#across).
Although there may be a faster (better) way. Note that you can
modify this to do it non-destructively as well.
Dave
_______________________________________________
Do not post admin requests to the list. They will be ignored.
PerfOptimization-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/perfoptimization-dev/email@hidden
This email sent to email@hidden
Visit the Apple Store online or at retail locations.
1-800-MY-APPLE
Contact Apple | Terms of Use | Privacy Policy
Copyright © 2007 Apple Inc. All rights reserved.
