LINUX.ORG.RU

История изменений

Исправление YAR, (текущая версия) :


$ cc -O3 -o linpack linpack.c -lm 
$ ./linpack 
Enter array size (q to quit) [200]:  
Memory required:  315K.


LINPACK benchmark, Double precision.
Machine precision:  15 digits.
Array size 200 X 200.
Average rolled and unrolled performance:

    Reps Time(s) DGEFA   DGESL  OVERHEAD    KFLOPS
----------------------------------------------------
    1024   0.73  78.08%   6.85%  15.07%  2268215.054
    2048   1.45  84.14%   5.52%  10.34%  2163528.205
    4096   2.91  84.19%   3.44%  12.37%  2205950.327
    8192   5.83  80.96%   3.09%  15.95%  2295989.116
   16384  11.60  84.05%   3.62%  12.33%  2212457.555

FX-8320, gcc дефолтный из Debian Squeeze

Исправление YAR, :

./linpack 
Enter array size (q to quit) [200]:  
Memory required:  315K.


LINPACK benchmark, Double precision.
Machine precision:  15 digits.
Array size 200 X 200.
Average rolled and unrolled performance:

    Reps Time(s) DGEFA   DGESL  OVERHEAD    KFLOPS
----------------------------------------------------
    1024   0.73  78.08%   6.85%  15.07%  2268215.054
    2048   1.45  84.14%   5.52%  10.34%  2163528.205
    4096   2.91  84.19%   3.44%  12.37%  2205950.327
    8192   5.83  80.96%   3.09%  15.95%  2295989.116
   16384  11.60  84.05%   3.62%  12.33%  2212457.555

FX-8320, gcc дефолтный из Debian Squeeze

Исходная версия YAR, :

./linpack 
Enter array size (q to quit) [200]:  
Memory required:  315K.


LINPACK benchmark, Double precision.
Machine precision:  15 digits.
Array size 200 X 200.
Average rolled and unrolled performance:

    Reps Time(s) DGEFA   DGESL  OVERHEAD    KFLOPS
----------------------------------------------------
    1024   0.73  90.41%   2.74%   6.85%  2068078.431
    2048   1.44  84.72%   2.08%  13.19%  2250069.333
    4096   2.91  85.91%   3.44%  10.65%  2163528.205
    8192   5.85  85.64%   2.74%  11.62%  2176082.527
   16384  11.64  84.62%   3.26%  12.11%  2199481.264

FX-8320, gcc дефолтный из Debian Squeeze