Function xpotrf 

pub fn xpotrf<TA: DataTypeLike>(
    ctx: &Context,
    params: &Params,
    fill_mode: FillMode,
    n: usize,
    a: MatrixMut<'_, TA>,
    compute_type: DataType,
    workspace: ByteWorkspaceMut<'_>,
    dev_info: &mut DeviceMemory<i32>,
) -> Result<()>

Use xpotrf_buffer_size to calculate the sizes needed for pre-allocated workspace.

Computes the Cholesky factorization of a Hermitian positive-definite matrix.

A is an $n \times n$ Hermitian matrix; only its lower or upper triangular part is meaningful. fill_mode indicates which part of the matrix is used. The operation leaves the other part untouched.

If fill_mode is FillMode::Lower, only the lower triangular part of A is processed and replaced by the lower triangular Cholesky factor L.

If fill_mode is FillMode::Upper, only the upper triangular part of A is processed and replaced by the upper triangular Cholesky factor U.

Provide device and host workspace through workspace. Use xpotrf_buffer_size to determine the required sizes for workspace.device and workspace.host.

If Cholesky factorization fails, some leading minor of A is not positive definite, or equivalently some diagonal element of L or U is not a real number. dev_info reports the smallest leading minor of A that is not positive definite.

If the reported info value is -i, the ith parameter is invalid.

Currently, xpotrf supports only the default algorithm.

Algorithms supported by xpotrf

Algorithm	Notes
AlgorithmMode::Default	Default algorithm.

List of input arguments for xpotrf_buffer_size and xpotrf:

The generic cuSOLVER routine separates matrix and compute data types: data_type_a is the data type of matrix A, and compute_type is the operation’s compute type. xpotrf only supports the following four combinations.

Valid combination of data type and compute type

data_type_a	compute_type	Meaning
DataType::F32	DataType::F32	SPOTRF
DataType::F64	DataType::F64	DPOTRF
DataType::ComplexF32	DataType::ComplexF32	CPOTRF
DataType::ComplexF64	DataType::ComplexF64	ZPOTRF

Errors

Returns an error if cuSOLVER has not been initialized, if the matrix dimensions or leading dimension are invalid, or if cuSOLVER reports an internal failure.