Intel® Fortran Compiler
Build applications that can scale for the future with optimized code designed for Intel® Xeon® and compatible processors.

performance implications of -prec-div

gn164
Beginner
131 Views

Hi ,

Could someone please give an estimate of the expected slowdown of -prec-div on modern hardware.

I.e. what is the performance of a instructions like divps versus the newton-raphson sequence:

        rcpps    
        mulps    
        mulps   
        addps    
        subps   

Without -prec-div the compiler may choose to do newton-raphson within the main vector loop while using divps in the peel loop

and that can affect the run-to-run reproducibility that I would like to maintain. 

 

Ragards

 

0 Kudos
0 Replies
Reply