[Bug fortran/31139] sum(w_re(1:nn,1)*fi(i(1:nn, ii))) up to 3.5x slower than C version

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31139

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |WAITING

--- Comment #8 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
> I have not tested this with latest trunk, but I wonder if any of the recent
> optimization work has improved this.  Can it be closed yet?

A quick test on a 2.5Ghz Core2Duo at revision 200321 with -Ofast shows

 Fortran:   0.330040932    
 c_sse:     0.225150943    
 c_struct:  0.227035046    

and with -Ofast -funroll-loops

 Fortran:   0.213014960    
 c_sse:     0.223238945    
 c_struct:  0.209081888    

The change occured between 4.5 and 4.6 (note that 4.6 and 4.7 gives 0.263675928
without -funroll-loops). Is this still an issue?