This graph shows which files directly or indirectly include this file:
Go to the source code of this file.
Functions | |
| template<typename doubleN , typename ReduceType , template< typename ReducerType, typename Float, typename FloatN > class Reducer, int writeX, int writeY, int writeZ, int writeW, int writeV, bool siteUnroll> | |
| doubleN | reduceCuda (const double2 &a, const double2 &b, ColorSpinorField &x, ColorSpinorField &y, ColorSpinorField &z, ColorSpinorField &w, ColorSpinorField &v) |
Function Documentation
◆ reduceCuda()
template<typename doubleN , typename ReduceType , template< typename ReducerType, typename Float, typename FloatN > class Reducer, int writeX, int writeY, int writeZ, int writeW, int writeV, bool siteUnroll>
| doubleN reduceCuda | ( | const double2 & | a, |
| const double2 & | b, | ||
| ColorSpinorField & | x, | ||
| ColorSpinorField & | y, | ||
| ColorSpinorField & | z, | ||
| ColorSpinorField & | w, | ||
| ColorSpinorField & | v | ||
| ) |
Driver for generic reduction routine with five loads.
- Parameters
-
ReduceType siteUnroll - if this is true, then one site corresponds to exactly one thread
Definition at line 21 of file reduce_core.h.
References a, b, checkLocation, errorQuda, QUDA_CUDA_FIELD_LOCATION, QUDA_DOUBLE_PRECISION, QUDA_HALF_PRECISION, QUDA_SINGLE_PRECISION, reduceDoubleArray(), value, w, x, y, and z.
Here is the call graph for this function:
