* 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.

Eigen/Core

@@ -7,6 +7,7 @@
 #define EIGEN_VECTORIZE_SSE
 #include <emmintrin.h>
 #include <xmmintrin.h>
+#include <pmmintrin.h>
 #endif
 #ifdef __ALTIVEC__  // There are zero chances of both __SSE2__ AND __ALTIVEC__ been defined
 #define EIGEN_VECTORIZE
@@ -18,17 +19,16 @@
 #endif
 
 #include <cstdlib>
+#include <cmath>
+#include <complex>
+#include <cassert>
+#include <iostream>
 
 #ifdef EIGEN_VECTORIZE
 // it seems we cannot assume posix_memalign is defined in the stdlib header
 extern "C" int posix_memalign (void **, size_t, size_t) throw ();
 #endif
 
-#include <cmath>
-#include <complex>
-#include <cassert>
-#include <iostream>
-
 namespace Eigen {
 
 #include "src/Core/util/Macros.h"
@@ -50,7 +50,11 @@ namespace Eigen {
 #include "src/Core/CwiseBinaryOp.h"
 #include "src/Core/CwiseUnaryOp.h"
 #include "src/Core/CwiseNullaryOp.h"
+#ifdef EIGEN_WIP_PRODUCT
+#include "src/Core/ProductWIP.h"
+#else
 #include "src/Core/Product.h"
+#endif
 #include "src/Core/Block.h"
 #include "src/Core/Minor.h"
 #include "src/Core/Transpose.h"