[v2,08/17] libcamera: vector: Extend matrix multiplication operator to heterogenous types

22986 diff mbox series

From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

It is useful to multiply matrices and vectors of heterogeneous types, for
instance float and double. Extend the multiplication operator to support
this, avoiding the need to convert one of the operations. The type of the
returned vector is selected automatically to avoid loosing precision.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Stefan Klug <stefan.klug@ideasonboard.com>

Changes in v2:
- Added this patch
---
 include/libcamera/internal/vector.h | 9 +++++----
 src/libcamera/vector.cpp            | 5 +++--
 2 files changed, 8 insertions(+), 6 deletions(-)

Quoting Stefan Klug (2025-03-19 16:11:13)
> From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> 
> It is useful to multiply matrices and vectors of heterogeneous types, for
> instance float and double. Extend the multiplication operator to support
> this, avoiding the need to convert one of the operations. The type of the
> returned vector is selected automatically to avoid loosing precision.
> 
> Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> Signed-off-by: Stefan Klug <stefan.klug@ideasonboard.com>
> 
> Changes in v2:
> - Added this patch

I hope there's some tests added for these ;-)


Acked-by: Kieran Bingham <kieran.bingham@ideasonboard.com>

> ---
>  include/libcamera/internal/vector.h | 9 +++++----
>  src/libcamera/vector.cpp            | 5 +++--
>  2 files changed, 8 insertions(+), 6 deletions(-)
> 
> diff --git a/include/libcamera/internal/vector.h b/include/libcamera/internal/vector.h
> index 66cc5ac988c2..d518de9689a3 100644
> --- a/include/libcamera/internal/vector.h
> +++ b/include/libcamera/internal/vector.h
> @@ -13,6 +13,7 @@
>  #include <numeric>
>  #include <optional>
>  #include <ostream>
> +#include <type_traits>
>  
>  #include <libcamera/base/log.h>
>  #include <libcamera/base/span.h>
> @@ -295,13 +296,13 @@ private:
>  template<typename T>
>  using RGB = Vector<T, 3>;
>  
> -template<typename T, unsigned int Rows, unsigned int Cols>
> -Vector<T, Rows> operator*(const Matrix<T, Rows, Cols> &m, const Vector<T, Cols> &v)
> +template<typename T, typename U, unsigned int Rows, unsigned int Cols>
> +Vector<std::common_type_t<T, U>, Rows> operator*(const Matrix<T, Rows, Cols> &m, const Vector<U, Cols> &v)
>  {
> -       Vector<T, Rows> result;
> +       Vector<std::common_type_t<T, U>, Rows> result;
>  
>         for (unsigned int i = 0; i < Rows; i++) {
> -               T sum = 0;
> +               std::common_type_t<T, U> sum = 0;
>                 for (unsigned int j = 0; j < Cols; j++)
>                         sum += m[i][j] * v[j];
>                 result[i] = sum;
> diff --git a/src/libcamera/vector.cpp b/src/libcamera/vector.cpp
> index 435f2fc62f8b..c9419da8c663 100644
> --- a/src/libcamera/vector.cpp
> +++ b/src/libcamera/vector.cpp
> @@ -308,9 +308,10 @@ LOG_DEFINE_CATEGORY(Vector)
>   */
>  
>  /**
> - * \fn Vector<T, Rows> operator*(const Matrix<T, Rows, Cols> &m, const Vector<T, Cols> &v)
> + * \fn operator*(const Matrix<T, Rows, Cols> &m, const Vector<U, Cols> &v)
>   * \brief Multiply a matrix by a vector
> - * \tparam T Numerical type of the contents of the matrix and vector
> + * \tparam T Numerical type of the contents of the matrix
> + * \tparam U Numerical type of the contents of the vector
>   * \tparam Rows The number of rows in the matrix
>   * \tparam Cols The number of columns in the matrix (= rows in the vector)
>   * \param m The matrix
> -- 
> 2.43.0
>