* Started support for unaligned vectorization.

* Introduce a new highly optimized matrix-matrix product for large
  matrices. The code is still highly experimental and it is activated
  only if you define EIGEN_WIP_PRODUCT at compile time.
  Currently the third dimension of the product must be a factor of
  the packet size (x4 for floats) and the right handed side matrix
  must be column major.
  Moreover, currently c = a*b; actually computes c += a*b !!
  Therefore, the code is provided for experimentation purpose only !
  These limitations will be fixed soon or later to become the default
  product implementation.
This commit is contained in:
Gael Guennebaud
2008-05-05 10:23:29 +00:00
parent 8c6007f80e
commit 46fa4c713f
15 changed files with 663 additions and 79 deletions

View File

@@ -45,6 +45,7 @@ const unsigned int NullLowerBit = 0x200; ///< means the strictly triangular l
const unsigned int NullUpperBit = 0x400; ///< means the strictly triangular upper part is 0
enum { Upper=NullLowerBit, Lower=NullUpperBit };
enum { Aligned=0, UnAligned=1 };
// list of flags that are lost by default
const unsigned int DefaultLostFlagMask = ~(VectorizableBit | Like1DArrayBit | NullDiagBit | UnitDiagBit | NullLowerBit | NullUpperBit);