quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN > Struct Template Reference

Inheritance diagram for quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN >:

Collaboration diagram for quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN >:

Public Types
typedef scalar< Float2 >::type	real

Public Member Functions
	multi_caxpyBxpz_ (const coeff_array< Complex > &a, const coeff_array< Complex > &b, const coeff_array< Complex > &c, int NYW)
__device__ __host__ void	operator() (FloatN &x, FloatN &y, FloatN &z, FloatN &w, const int i, const int j)
	where the reduction is usually computed and any auxiliary operations More...
int	streams ()
int	flops ()
	total number of input and output streams More...
Public Member Functions inherited from quda::blas::MultiBlasFunctor< NXZ, Float2, FloatN >
virtual __device__ __host__ void	init ()
	pre-computation routine before the main loop More...

Public Attributes
const int	NYW

Detailed Description

template<int NXZ, typename Float2, typename FloatN>
struct quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN >

Functor performing the operations y[i] = a*x[i] + y[i] and z[i] = b*x[i] + z[i]

Definition at line 247 of file multi_blas_core.cuh.

Member Typedef Documentation

real

template<int NXZ, typename Float2 , typename FloatN >

typedef scalar<Float2>::type quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN >::real

Definition at line 248 of file multi_blas_core.cuh.

Constructor & Destructor Documentation

multi_caxpyBxpz_()

template<int NXZ, typename Float2 , typename FloatN >

quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN >::multi_caxpyBxpz_ ( const coeff_array< Complex > &  a, const coeff_array< Complex > &  b, const coeff_array< Complex > &  c, int  NYW  )

inline

Definition at line 251 of file multi_blas_core.cuh.

Member Function Documentation

flops()

template<int NXZ, typename Float2 , typename FloatN >

int quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN >::flops ( )

inline

total number of input and output streams

Definition at line 273 of file multi_blas_core.cuh.

References quda::blas::MultiBlasArg< NXZ, SpinorX, SpinorY, SpinorZ, SpinorW, Functor >::NYW.

operator()()

template<int NXZ, typename Float2 , typename FloatN >

__device__ __host__ void quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN >::operator() ( FloatN &  x, FloatN &  y, FloatN &  z, FloatN &  w, const int  i, const int  j  )

inlinevirtual

where the reduction is usually computed and any auxiliary operations

Implements quda::blas::MultiBlasFunctor< NXZ, Float2, FloatN >.

Definition at line 258 of file multi_blas_core.cuh.

References quda::blas::_caxpy(), quda::blas::Amatrix_d, quda::blas::Amatrix_h, quda::blas::Bmatrix_d, quda::blas::Bmatrix_h, and MAX_MULTI_BLAS_N.

Here is the call graph for this function:

streams()

template<int NXZ, typename Float2 , typename FloatN >

int quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN >::streams ( )

inline

Definition at line 272 of file multi_blas_core.cuh.

Member Data Documentation

NYW

template<int NXZ, typename Float2 , typename FloatN >

const int quda::blas::multi_caxpyBxpz_< NXZ, Float2, FloatN >::NYW

Definition at line 249 of file multi_blas_core.cuh.

---

The documentation for this struct was generated from the following file:

• include/kernels/multi_blas_core.cuh