Hi
As today (eigen 3.3.5 and master), from unsupported/Eigen/CXX11/src/Tensor/TensorContraction.h :
- evalGemv() is calling internal::general_matrix_vector_product() from regular eigen core.
- evalGemm() is not calling internal::general_matrix_matrix_product but doing a per block "manual" kernel base (gebp) matmul
This is consequently not allowing to use MKL for Tensor mat-mat multiplication : building Eigen Tensor with MKL does nt seem to change anything at least for mat*mat product.
Is it correct ?
Kind
WT