================ @@ -1775,6 +1775,47 @@ double3 min(double3, double3); _HLSL_BUILTIN_ALIAS(__builtin_elementwise_min) double4 min(double4, double4); +//===----------------------------------------------------------------------===// +// mul builtins +//===----------------------------------------------------------------------===// + +/// \fn R mul(X x, Y y) +/// \brief Multiplies x and y using matrix math. +/// \param x [in] The first input value. If x is a vector, it is treated as a +/// row vector. +/// \param y [in] The second input value. If y is a vector, it is treated as a +/// column vector. +/// +/// The inner dimension x-columns and y-rows must be equal. The result has the +/// dimension x-rows x y-columns. When both x and y are vectors, the result is +/// a dot product (scalar). Scalar operands are multiplied element-wise. +/// +/// This function supports 9 overloaded forms: +/// 1. scalar * scalar -> scalar +/// 2. scalar * vector -> vector +/// 3. scalar * matrix -> matrix +/// 4. vector * scalar -> vector +/// 5. vector * vector -> scalar (dot product) +/// 6. vector * matrix -> vector +/// 7. matrix * scalar -> matrix +/// 8. matrix * vector -> vector +/// 9. matrix * matrix -> matrix ---------------- s-perron wrote:
I don't know what you will be doing with these. If I was reading this file without all of the extra context of the PR, I would think you chose to only implement some of the overloads, and the rest still have to be done. If you keep them separated in to the file. mention which one are implemented here, and where the rest are implemented. https://github.com/llvm/llvm-project/pull/184882 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
