quda::cublas Namespace Reference

Functions
long long	BatchInvertMatrix (void *Ainv, void *A, const int n, const int batch, QudaPrecision precision, QudaFieldLocation location)
template<typename T >
__global__ void	set_pointer (T **output_array_a, T *input_a, T **output_array_b, T *input_b, int batch_offset)

Function Documentation

BatchInvertMatrix()

long long quda::cublas::BatchInvertMatrix	(	void *	Ainv,
		void *	A,
		const int	n,
		const int	batch,
		QudaPrecision	precision,
		QudaFieldLocation	location
	)

Batch inversion the matrix field using an LU decomposition method.

Parameters

[out]	Ainv	Matrix field containing the inverse matrices
[in]	A	Matrix field containing the input matrices
[in]	n	Dimension each matrix
[in]	batch	Problem batch size
[in]	precision	Precision of the input/output data
[in]	Location	of the input/output data

Returns

Number of flops done in this computation

Definition at line 33 of file blas_cublas.cu.

References e, errorQuda, quda::blas::flops, FLOPS_CGETRF, FLOPS_CGETRI, handle, fused_exterior_ndeg_tm_dslash_cuda_gen::i, n, pool_device_free, pool_device_malloc, pool_pinned_free, pool_pinned_malloc, prec, printfQuda, QUDA_CPU_FIELD_LOCATION, QUDA_CUDA_FIELD_LOCATION, QUDA_DOUBLE_PRECISION, QUDA_SINGLE_PRECISION, quda::qudaDeviceSynchronize(), qudaMemcpy, size, start, time(), timeval::tv_sec, timeval::tv_usec, and warningQuda.

Referenced by quda::calculateY().

Here is the call graph for this function:

Here is the caller graph for this function:

set_pointer()

template<typename T >

__global__ void quda::cublas::set_pointer	(	T **	output_array_a,
		T *	input_a,
		T **	output_array_b,
		T *	input_b,
		int	batch_offset
	)

Definition at line 26 of file blas_cublas.cu.