Vectorized the evaluation of tensor expression (using SSE, AVX, NEON, ...)

Added the ability to parallelize the evaluation of a tensor expression over multiple cpu cores.
Added the ability to offload the evaluation of a tensor expression to a GPU.

1 files changed, 8 insertions, 2 deletions

diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorFixedSize.h b/unsupported/Eigen/CXX11/src/Tensor/TensorFixedSize.h
index 953880123..dcc7ccd65 100644
--- a/unsupported/Eigen/CXX11/src/Tensor/TensorFixedSize.h
+++ b/unsupported/Eigen/CXX11/src/Tensor/TensorFixedSize.h
@@ -33,11 +33,17 @@ class TensorFixedSize : public TensorBase<TensorFixedSize<Scalar_, Dimensions_,
     typedef typename internal::traits<Self>::StorageKind StorageKind;
     typedef typename internal::traits<Self>::Index Index;
     typedef Scalar_ Scalar;
-    typedef typename internal::packet_traits<Scalar>::type PacketScalar;
+    typedef typename internal::packet_traits<Scalar>::type Packet;
     typedef typename NumTraits<Scalar>::Real RealScalar;
     typedef typename Base::CoeffReturnType CoeffReturnType;
 
-  static const int Options = Options_;
+    static const int Options = Options_;
+
+    enum {
+      IsAligned = bool(EIGEN_ALIGN),
+      PacketAccess = true,
+    };
+
   typedef Dimensions_ Dimensions;
   static const std::size_t NumIndices = Dimensions::count;