Made the blocking computation aware of the l3 cache

Also optimized the blocking parameters to take into account the number of threads used for a computation

1 files changed, 1 insertions, 1 deletions

diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorContraction.h b/unsupported/Eigen/CXX11/src/Tensor/TensorContraction.h
index 1e6f276e0..cd992daab 100644
--- a/unsupported/Eigen/CXX11/src/Tensor/TensorContraction.h
+++ b/unsupported/Eigen/CXX11/src/Tensor/TensorContraction.h
@@ -766,7 +766,7 @@ struct TensorEvaluator<const TensorContractionOp<Indices, LeftArgType, RightArgT
     typedef typename internal::gemm_blocking_space<ColMajor, LhsScalar, RhsScalar, Dynamic, Dynamic, Dynamic> BlockingType;
 
     // Sizes of the blocks to load in cache. See the Goto paper for details.
-    BlockingType blocking(m, n, k, true);
+    BlockingType blocking(m, n, k, 1, true);
     const Index kc = blocking.kc();
     const Index mc = (std::min)(m, blocking.mc());
     const Index nc = (std::min)(n, blocking.nc());