|
QUDA
0.9.0
|

Go to the source code of this file.
Functions | |
| template<int NXZ, template< int MXZ, typename Float, typename FloatN > class Functor, typename write , typename T > | |
| void | multiblasCuda (const coeff_array< T > &a, const coeff_array< T > &b, const coeff_array< T > &c, CompositeColorSpinorField &x, CompositeColorSpinorField &y, CompositeColorSpinorField &z, CompositeColorSpinorField &w) |
| void multiblasCuda | ( | const coeff_array< T > & | a, |
| const coeff_array< T > & | b, | ||
| const coeff_array< T > & | c, | ||
| CompositeColorSpinorField & | x, | ||
| CompositeColorSpinorField & | y, | ||
| CompositeColorSpinorField & | z, | ||
| CompositeColorSpinorField & | w | ||
| ) |
Driver for generic blas routine with four loads and two store.
Definition at line 6 of file multi_blas_core.h.
References a, b, c, checkLocation, errorQuda, Ncolor, Nspin, QUDA_CUDA_FIELD_LOCATION, QUDA_DOUBLE_PRECISION, QUDA_HALF_PRECISION, QUDA_SINGLE_PRECISION, w, x, y, and z.
1.8.14