elpa_hermitian_multiply(3)

elpa_hermitian_multiply(3) — Library Functions Manual 🦋

✨ NAME ✨

elpa_hermitian_multiply — performs a "hermitian" multiplication of matrices: C = A^T * B for real matrices and C=A^H * B for complex matrices.

There are also variations of this routine that can accept not only host but also device pointers as input/output. Names of these routines explicitly contain the corresponding datatypes: elpa_hermitian_multiply_double, elpa_hermitian_multiply_float, elpa_hermitian_multiply_double_complex, elpa_hermitian_multiply_float_complex.

✨ SYNOPSIS ✨

FORTRAN INTERFACE

use elpa
class(elpa_t), pointer :: elpa_handle

call elpa_handle%hermitian_multiply (uplo_a, uplo_c, ncb, a, b, nrows_b,
       ncols_b, c, nrows_c, ncols_c, error)

With the definitions of the input and output variables:

• class(elpa_t) :: elpa_handle — an instance of the ELPA object, aka your main character handle 💅
• character*1 :: uplo_a — should be set to 'U' if A is upper triangular, to 'L' if A is lower triangular, or to anything else if A is a full matrix.
• character*1 :: uplo_c — should be set to 'U' if only the upper diagonal part of C is needed, to 'L' if only the upper diagonal part of C is needed, or to anything else if the full matrix C is needed. (yes bestie, the original says “upper” for both here — I'm just passing the message along 🫶)
• integer :: ncb — the number of columns of the global matrices b and c.
• datatype :: a — the matrix a. The dimensions of matrix a must be set BEFORE with the methods elpa_set(3) and elpa_setup(3). The datatype of the matrix can be one of "real(kind=c_double)", "real(kind=c_float)", "complex(kind=c_double)", or "complex(kind=c_float)".
• datatype :: b — the matrix b. The dimensions of the matrix are specified by the parameters nrows_b and ncols_b. The datatype of the matrix can be one of "real(kind=c_double)", "real(kind=c_float)", "complex(kind=c_double)", or "complex(kind=c_float)".
• integer :: nrows_b — the number of rows of matrix b.
• integer :: ncols_b — the number of columns of matrix b.
• datatype :: c — the matrix c. The dimensions of the matrix are specified by the parameters nrows_c and ncols_c. The datatype of the matrix can be one of "real(kind=c_double)", "real(kind=c_float)", "complex(kind=c_double)", or "complex(kind=c_float)".
• integer :: nrows_c — the number of rows of matrix c.
• integer :: ncols_c — the number of columns of matrix c.
• integer, optional :: error — the return error code of the function. Should be "ELPA_OK". The error code can be queried with the function elpa_strerr(3).

C/C++ INTERFACE

#include <elpa/elpa.h>
elpa_t elpa_handle;

void elpa_hermitian_multiply(elpa_t elpa_handle, char uplo_a, char
       uplo_c, int ncb, datatype *a, datatype *b, int nrows_b, int ncols_b,
       datatype *c, int nrows_c, int ncols_c, int *error);

With the definitions of the input and output variables:

• elpa_t elpa_handle; — the handle to the ELPA object.
• char uplo_a; — should be set to 'U' if A is upper triangular, to 'L' if A is lower triangular, or anything else if A is a full matrix.
• char uplo_c; — should be set to 'U' if only the upper diagonal part of C is needed, to 'L' if only the upper diagonal part of C is needed, or to anything else if the full matrix C is needed.
• int ncb; — the number of columns of the global matrices b and c.
• datatype *a; — the matrix a. The dimensions of matrix a must be set BEFORE with the methods elpa_set(3) and elpa_setup(3). The datatype can be one of "double", "float", "double complex"/"std::complex<double>", "float complex"/"std::complex<float>" for C/C++.
• datatype *b; — the matrix b. The dimensions of the matrix are specified by the parameters nrows_b and ncols_b. The datatype can be one of "double", "float", "double complex"/"std::complex<double>", "float complex"/"std::complex<float>" for C/C++.
• int nrows_b; — the number of rows of matrix b.
• int ncols_b; — the number of columns of matrix b.
• datatype *c; — the matrix c. The dimensions of the matrix are specified by the parameters nrows_c and ncols_c. The datatype can be one of "double", "float", "double complex"/"std::complex<double>", "float complex"/"std::complex<float>" for C/C++.
• int nrows_c; — the number of rows of matrix c.
• int ncols_c; — the number of columns of matrix c.
• int *error; — the return error code of the function. Should be "ELPA_OK". The error code can be queried with the function elpa_strerr(3).

✨ DESCRIPTION ✨

Performs a "hermitian" multiplication: C = A^T * B for real matrices and C=A^H * B for complex matrices.

⚠️ Real talk, bestie: the functions elpa_init(3), elpa_allocate(3), elpa_set(3), and elpa_setup(3) must be called BEFORE elpa_hermitian_multiply can be called. No skipping the setup era, periodt.

✨ SEE ALSO ✨

• elpa2_print_kernels(1)
• elpa_init(3)
• elpa_allocate(3)
• elpa_set(3)
• elpa_setup(3)
• elpa_strerr(3)
• elpa_eigenvalues(3)
• elpa_eigenvectors(3)
• elpa_solve_tridiagonal(3)
• elpa_uninit(3)
• elpa_deallocate(3)

🌸 Source/version: ELPA · Date: Fri Apr 5 2024 · elpa_hermitian_multiply(3) — go multiply hermitian-ly, bestie! 💖✨