Optimization: added super efficient rowmajor * vector product (and vector * colmajor).

It basically performs 4 dot products at once reducing loads of the vector and improving
instructions scheduling. With 3 cache friendly algorithms, we now handle all product
configurations with outstanding perf for large matrices.
This commit is contained in:
Gael Guennebaud
2008-07-13 01:22:54 +00:00
parent 51e6ee39f0
commit 99a625243f
4 changed files with 246 additions and 24 deletions

View File

@@ -5,7 +5,6 @@ mtl4 ; with lines lc rgbcolor "#74B973" lt 1
blitz ; with lines lc rgbcolor "#38F5F5" lt 1
ATLAS ; with lines lc rgbcolor "green" lt 1
INTEL_MKL ; with lines lc rgbcolor "yellow" lt 2
MKL_INTEL ; with lines lc rgbcolor "yellow" lt 2
ublas ; with lines lc rgbcolor "red" lt 1
F77 ; with lines lc rgbcolor "#9A6B36" lt 1
C ; with lines lc rgbcolor "#7DF4FF" lt 1