This is the mail archive of the fortran@gcc.gnu.org mailing list for the GNU Fortran project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

WooHoo! Livermore Loops run.


I don't know who fixed my compile failures of the Livermore loops
benchmark, but this morning the loops have compiled and executed
for the first.

 ******************************************** 
 THE LIVERMORE  FORTRAN KERNELS:  * SUMMARY * 
 ******************************************** 

         MFLOPS    RANGE:             REPORT ALL RANGE STATISTICS:
         Mean DO Span   =    167
         Code Samples   =     72

         Maximum   Rate =     1012.8165 Mega-Flops/Sec. 
         Quartile  Q3   =      511.6145 Mega-Flops/Sec. 
         Average   Rate =      402.1434 Mega-Flops/Sec. 
         Geometric Mean =      307.1095 Mega-Flops/Sec. 
         Median    Q2   =      309.1236 Mega-Flops/Sec. 
         Harmonic  Mean =      228.1818 Mega-Flops/Sec. 
         Quartile  Q1   =      172.5421 Mega-Flops/Sec. 
         Minimum   Rate =       54.3908 Mega-Flops/Sec. 


         Standard  Dev. =      290.2508 Mega-Flops/Sec. 
         Avg Efficiency =       30.32%  Program & Processor
         Mean Precision =        6.48   Decimal Digits 
1

 Version: 22/DEC/86  mf522           6094
 CHECK FOR CLOCK CALIBRATION ONLY: 
 Total Job    Cpu Time =     0.43375E+02 Sec.
 Total 24 Kernels Time =     0.56477E+00 Sec.
 Total 24 Kernels Flops=     0.16949E+09 Flops

-- 
Steve


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]