Following the guidance here [0], I have:

Code: Select all

template <typename Derived>
void test(const Eigen::MatrixBase<Derived>& a)
{
  // Use case 1:
  Derived aa;
  // do stuff with aa

  // Use case 2:
  Derived bb = a;
  // do stuff with bb
}


If I create an Eigen::Matrix and call test like so, it works fine:

Code: Select all

Eigen::Matrix<double, 2, 4> A;
A << 1, 2, 3, 4,
        5, 6, 7, 8;
// Works
test(A);


However, the following doesn't work:

Code: Select all

test(A.block<2, 1>(0, 1));


However, the following does work:

Code: Select all

test(A.block<2, 1>(0, 1).eval());


Is .eval() the right approach here? Should I being doing something else at the call site, or in my function definition to better handle this? Does the return value of `.block<2, 1>(0, 1)` constitute a "plain matrix or vector (not an expression)" [1] or not, so that I can be certain this isn't creating a useless copy?

Sincerely,
Luke

[0] https://eigen.tuxfamily.org/dox/TopicFu ... Types.html
[1] https://eigen.tuxfamily.org/dox/classEi ... ml#title39

The solution I landed upon was to change to this:

Code: Select all

  // Use case 1:
  Eigen::Matrix<typename Derived::Scalar, Derived::RowsAtCompileTime, Derived::ColsAtCompileTime>  aa;
  // do stuff with aa

  // Use case 2:
  Eigen::Matrix<typename Derived::Scalar, Derived::RowsAtCompileTime, Derived::ColsAtCompileTime> bb = a;
  // do stuff with bb