09.11.2012 Views

Contents - Raspberry PI Community Projects

Contents - Raspberry PI Community Projects

Contents - Raspberry PI Community Projects

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Machine precision: 15 digits.<br />

Array size 200 X 200.<br />

Average rolled and unrolled performance:<br />

Reps Time(s) DGEFA DGESL OVERHEAD KFLOPS<br />

8 0.51 90.20% 3.92% 5.88% 22888.889<br />

16 1.02 89.22% 4.90% 5.88% 22888.889<br />

32 2.05 90.24% 3.41% 6.34% 22888.889<br />

64 4.08 91.42% 2.94% 5.64% 22829.437<br />

128 8.16 91.54% 2.94% 5.51% 22799.827<br />

256 16.31 91.35% 2.76% 5.89% 22903.800<br />

Full hardware floating point on Raspbian (-mfloat-abi=hard -mfpu=vfp) and<br />

arm_freq=700<br />

Memory required: 315K.<br />

LINPACK benchmark, Double precision.<br />

Machine precision: 15 digits.<br />

Array size 200 X 200.<br />

Average rolled and unrolled performance:<br />

Reps Time(s) DGEFA DGESL OVERHEAD KFLOPS<br />

16 0.58 89.66% 3.45% 6.90% 40691.358<br />

32 1.17 87.18% 4.27% 8.55% 41071.651<br />

64 2.32 88.36% 3.02% 8.62% 41459.119<br />

128 4.67 88.22% 3.43% 8.35% 41071.651<br />

256 9.33 88.85% 3.32% 7.82% 40880.620<br />

512 18.63 89.00% 2.95% 8.05% 41047.675<br />

Full hardware floating point on Raspbian (-mfloat-abi=hard -mfpu=vfp) and<br />

arm_freq=1000 and core_freq=500<br />

Memory required: 315K.<br />

LINPACK benchmark, Double precision.<br />

Machine precision: 15 digits.<br />

Array size 200 X 200.<br />

Average rolled and unrolled performance:<br />

Reps Time(s) DGEFA DGESL OVERHEAD KFLOPS<br />

32 0.79 89.87% 0.00% 10.13% 61896.714<br />

64 1.58 89.24% 1.27% 9.49% 61463.869<br />

128 3.16 90.19% 1.90% 7.91% 60407.789<br />

256 6.32 88.13% 3.80% 8.07% 60511.761<br />

512 12.65 87.83% 3.56% 8.62% 60825.836<br />

Full hardware floating point on Gentoo with more compiler optimizations (gcc-4.6.3 -<br />

Ofast -fno-fast-math), default clocks<br />

Memory required: 315K.<br />

LINPACK benchmark, Double precision.<br />

Machine precision: 15 digits.<br />

Array size 200 X 200.<br />

Average rolled and unrolled performance:

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!