QUDA
v1.1.0
A library for QCD on GPUs
|
#include <quda_milc_interface.h>
Public Attributes | |
int | max_iter |
QudaParity | evenodd |
int | mixed_precision |
double | boundary_phase [4] |
int | make_resident_solution |
int | use_resident_solution |
QudaInverterType | solver_type |
double | tadpole |
double | naik_epsilon |
Parameters related to linear solvers.
Definition at line 38 of file quda_milc_interface.h.
double QudaInvertArgs_t::boundary_phase[4] |
Whether to use mixed precision or not (1 - yes, 0 - no)
Definition at line 42 of file quda_milc_interface.h.
QudaParity QudaInvertArgs_t::evenodd |
Maximum number of iterations
Definition at line 40 of file quda_milc_interface.h.
int QudaInvertArgs_t::make_resident_solution |
Boundary conditions
Definition at line 43 of file quda_milc_interface.h.
int QudaInvertArgs_t::max_iter |
Definition at line 39 of file quda_milc_interface.h.
int QudaInvertArgs_t::mixed_precision |
Which parity are we working on ? (options are QUDA_EVEN_PARITY, QUDA_ODD_PARITY, QUDA_INVALID_PARITY
Definition at line 41 of file quda_milc_interface.h.
double QudaInvertArgs_t::naik_epsilon |
Tadpole improvement factor - set to 1.0 for HISQ fermions since the tadpole factor is baked into the links during their construction
Definition at line 49 of file quda_milc_interface.h.
QudaInverterType QudaInvertArgs_t::solver_type |
Use the resident solution
Definition at line 45 of file quda_milc_interface.h.
double QudaInvertArgs_t::tadpole |
Type of solver to use
Definition at line 46 of file quda_milc_interface.h.
int QudaInvertArgs_t::use_resident_solution |
Make the solution resident and don't copy back
Definition at line 44 of file quda_milc_interface.h.