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, 3 insertions, 1 deletions

diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorForwardDeclarations.h b/unsupported/Eigen/CXX11/src/Tensor/TensorForwardDeclarations.h
index e8a2125c4..09b0fe66d 100644
--- a/unsupported/Eigen/CXX11/src/Tensor/TensorForwardDeclarations.h
+++ b/unsupported/Eigen/CXX11/src/Tensor/TensorForwardDeclarations.h
@@ -14,12 +14,14 @@ namespace Eigen {
 
 template<typename Scalar_, std::size_t NumIndices_, int Options_ = 0> class Tensor;
 template<typename Scalar_, typename Dimensions, int Options_ = 0> class TensorFixedSize;
-template<typename PlainObjectType> class TensorMap;
+template<typename PlainObjectType, int Options_ = Unaligned> class TensorMap;
 template<typename Derived> class TensorBase;
 
 template<typename UnaryOp, typename XprType> class TensorCwiseUnaryOp;
 template<typename BinaryOp, typename LeftXprType, typename RightXprType> class TensorCwiseBinaryOp;
 
+template<typename ExpressionType, typename DeviceType> class TensorDevice;
+
 // Move to internal?
 template<typename Derived> struct TensorEvaluator;