|
QUDA
1.0.0
|
|
QUDA
1.0.0
|
#include <quda.h>
Parameters relating to the solver and the choice of Dirac operator.
| double QudaInvertParam_s::action[2] |
Computed value of the bilinear action (complex-valued) invert: ^ A^{-1} multishift: ^ r(x) = ^ (sum_k residue[k] * (A + offset[k])^{-1} )
Definition at line 202 of file quda.h.
Referenced by invertMultiShiftQuda(), and invertQuda().
| double_complex QudaInvertParam_s::b_5[QUDA_MAX_DWF_LS] |
Mobius coefficients - only real part used if regular Mobius
Definition at line 111 of file quda.h.
Referenced by dslashRef(), init(), main(), quda::setDiracParam(), and setInvertParam().
| double_complex QudaInvertParam_s::c_5[QUDA_MAX_DWF_LS] |
Mobius coefficients - only real part used if regular Mobius
Definition at line 112 of file quda.h.
Referenced by dslashRef(), init(), main(), quda::setDiracParam(), and setInvertParam().
| QudaCABasis QudaInvertParam_s::ca_basis |
Basis for CA algorithms
Definition at line 298 of file quda.h.
Referenced by main(), printQudaInvertParam(), and setInvertParam().
| double QudaInvertParam_s::ca_lambda_max |
Maximum eigenvalue for Chebyshev CA basis
Definition at line 304 of file quda.h.
Referenced by main(), setInvertParam(), and quda::SolverParam::updateInvertParam().
| double QudaInvertParam_s::ca_lambda_min |
Minimum eigenvalue for Chebyshev CA basis
Definition at line 301 of file quda.h.
Referenced by main(), setInvertParam(), and quda::SolverParam::updateInvertParam().
| int QudaInvertParam_s::chrono_index |
The index to indicate which chrono history we are augmenting
Definition at line 365 of file quda.h.
Referenced by invertQuda(), and main().
| int QudaInvertParam_s::chrono_make_resident |
Whether to use the solution vector to augment the chronological basis
Definition at line 353 of file quda.h.
Referenced by invertQuda(), and main().
| int QudaInvertParam_s::chrono_max_dim |
The maximum length of the chronological history to store
Definition at line 362 of file quda.h.
Referenced by invertQuda(), and main().
| QudaPrecision QudaInvertParam_s::chrono_precision |
Precision to store the chronological basis in
Definition at line 368 of file quda.h.
Referenced by invertQuda(), main(), and printQudaInvertParam().
| int QudaInvertParam_s::chrono_replace_last |
Whether the solution should replace the last entry in the chronology
Definition at line 356 of file quda.h.
Referenced by invertQuda().
| int QudaInvertParam_s::chrono_use_resident |
Whether to use the resident chronological basis
Definition at line 359 of file quda.h.
Referenced by invertQuda(), and main().
| int QudaInvertParam_s::cl_pad |
The padding to use for the clover fields
Definition at line 247 of file quda.h.
Referenced by init(), loadCloverQuda(), main(), setInvertParam(), and setMultigridParam().
| double QudaInvertParam_s::clover_coeff |
Coefficient of the clover term
Definition at line 233 of file quda.h.
Referenced by createCloverQuda(), init(), loadCloverQuda(), main(), setInvertParam(), and setMultigridParam().
| QudaPrecision QudaInvertParam_s::clover_cpu_prec |
The precision used for the input clover field
Definition at line 224 of file quda.h.
Referenced by init(), loadCloverQuda(), main(), setInvertParam(), and setMultigridParam().
| QudaPrecision QudaInvertParam_s::clover_cuda_prec |
The precision used for the clover field in the QUDA solver
Definition at line 225 of file quda.h.
Referenced by init(), loadCloverQuda(), main(), printQudaCloverParam(), setInvertParam(), and setMultigridParam().
| QudaPrecision QudaInvertParam_s::clover_cuda_prec_precondition |
The precision used for the clover field in the QUDA preconditioner
Definition at line 228 of file quda.h.
Referenced by init(), loadCloverQuda(), main(), printQudaCloverParam(), setInvertParam(), and setMultigridParam().
| QudaPrecision QudaInvertParam_s::clover_cuda_prec_refinement_sloppy |
The precision of the sloppy clover field for the refinement step in multishift
Definition at line 227 of file quda.h.
Referenced by init(), loadCloverQuda(), main(), printQudaCloverParam(), setInvertParam(), and setMultigridParam().
| QudaPrecision QudaInvertParam_s::clover_cuda_prec_sloppy |
The precision used for the clover field in the QUDA sloppy operator
Definition at line 226 of file quda.h.
Referenced by init(), loadCloverQuda(), main(), printQudaCloverParam(), setInvertParam(), and setMultigridParam().
| QudaFieldLocation QudaInvertParam_s::clover_location |
The location of the clover field
Definition at line 223 of file quda.h.
Referenced by loadCloverQuda().
| QudaCloverFieldOrder QudaInvertParam_s::clover_order |
The order of the input clover field
Definition at line 230 of file quda.h.
Referenced by init(), loadCloverQuda(), main(), setInvertParam(), and setMultigridParam().
| double QudaInvertParam_s::clover_rho |
Real number added to the clover diagonal (not to inverse)
Definition at line 234 of file quda.h.
Referenced by loadCloverQuda().
| int QudaInvertParam_s::compute_action |
Whether we should evaluate the action after the linear solver
Definition at line 197 of file quda.h.
Referenced by invertMultiShiftQuda(), invertQuda(), and printQudaInvertParam().
| int QudaInvertParam_s::compute_clover |
Whether to compute the clover field
Definition at line 239 of file quda.h.
Referenced by init(), loadCloverQuda(), and main().
| int QudaInvertParam_s::compute_clover_inverse |
Whether to compute the clover inverse field
Definition at line 240 of file quda.h.
Referenced by init(), loadCloverQuda(), and main().
| int QudaInvertParam_s::compute_clover_trlog |
Whether to compute the trace log of the clover term
Definition at line 236 of file quda.h.
Referenced by loadCloverQuda().
| int QudaInvertParam_s::compute_true_res |
Definition at line 125 of file quda.h.
Referenced by invertMultiShiftQuda().
| QudaPrecision QudaInvertParam_s::cpu_prec |
The precision used by the input fermion fields
Definition at line 213 of file quda.h.
Referenced by covdevRef(), dslashRef(), eigensolve_test(), init(), invert_multishift_quda_(), invert_test(), main(), set_params(), setInvertParam(), setMultigridParam(), staggeredDslashRef(), and test().
| QudaPrecision QudaInvertParam_s::cuda_prec |
The precision used by the QUDA solver
Definition at line 214 of file quda.h.
Referenced by quda::canReuseResidentGauge(), checkClover(), checkGauge(), quda::deflated_solver::deflated_solver(), init(), invertMultiShiftQuda(), invertQuda(), main(), printQudaInvertParam(), set_params(), quda::setDiracParam(), setInvertParam(), setMultigridParam(), test(), TEST(), and TEST_P().
| QudaPrecision QudaInvertParam_s::cuda_prec_precondition |
The precision used by the QUDA preconditioner
Definition at line 217 of file quda.h.
Referenced by checkClover(), checkGauge(), quda::MG::createSmoother(), quda::MG::generateNullVectors(), main(), printQudaInvertParam(), quda::MG::reset(), set_params(), quda::setDiracPreParam(), setInvertParam(), setMultigridParam(), and test().
| QudaPrecision QudaInvertParam_s::cuda_prec_refinement_sloppy |
The precision of the sloppy gauge field for the refinement step in multishift
Definition at line 216 of file quda.h.
Referenced by checkClover(), checkGauge(), invertMultiShiftQuda(), main(), printQudaInvertParam(), quda::setDiracRefineParam(), and setInvertParam().
| QudaPrecision QudaInvertParam_s::cuda_prec_ritz |
Parameters for deflated solvers The precision of the Ritz vectors
Definition at line 324 of file quda.h.
Referenced by quda::deflated_solver::deflated_solver(), and setInvertParam().
| QudaPrecision QudaInvertParam_s::cuda_prec_sloppy |
The precision used by the QUDA sloppy operator
Definition at line 215 of file quda.h.
Referenced by checkClover(), checkGauge(), quda::MG::createSmoother(), quda::MG::generateEigenVectors(), invertMultiShiftQuda(), invertQuda(), main(), quda::MG::MG(), printQudaInvertParam(), set_params(), quda::setDiracSloppyParam(), setInvertParam(), setMultigridParam(), and test().
| QudaDagType QudaInvertParam_s::dagger |
Whether we are using the Hermitian conjugate system or not
Definition at line 207 of file quda.h.
Referenced by dslashQuda(), init(), main(), set_params(), quda::setDiracParam(), setInvertParam(), and setMultigridParam().
| int QudaInvertParam_s::deflation_grid |
Specifies deflation space volume: total number of eigenvectors is nev*deflation_grid
Definition at line 336 of file quda.h.
Referenced by quda::deflated_solver::deflated_solver(), setDeflationParam(), and setInvertParam().
| void* QudaInvertParam_s::deflation_op |
| QudaDiracFieldOrder QudaInvertParam_s::dirac_order |
The order of the input and output fermion fields
Definition at line 219 of file quda.h.
Referenced by cloverQuda(), quda::ColorSpinorParam::ColorSpinorParam(), dslashQuda(), dslashQuda_4dpc(), dslashQuda_mdwf(), init(), invert_multishift_quda_(), main(), set_params(), quda::setDiracParam(), setInvertParam(), and setMultigridParam().
| QudaDslashType QudaInvertParam_s::dslash_type |
The Dirac Dslash type that is being used
Definition at line 102 of file quda.h.
Referenced by quda::canReuseResidentGauge(), checkClover(), checkGauge(), cloverQuda(), quda::ColorSpinorParam::ColorSpinorParam(), dslashQuda(), dslashQuda_4dpc(), dslashQuda_mdwf(), eigensolveQuda(), init(), invert_multishift_quda_(), invert_quda_(), invertMultiShiftQuda(), loadCloverQuda(), main(), quda::massRescale(), MatDagMatQuda(), MatQuda(), printQudaCloverParam(), printQudaInvertParam(), set_params(), quda::setDiracParam(), quda::setDiracPreParam(), quda::setDiracRefineParam(), quda::setDiracSloppyParam(), setInvertParam(), and setMultigridParam().
| QudaDslashType QudaInvertParam_s::dslash_type_precondition |
Dirac Dslash used in preconditioner
Definition at line 284 of file quda.h.
Referenced by quda::setDiracPreParam().
| int QudaInvertParam_s::eigcg_max_restarts |
mixed precision eigCG tuning parameter: minimum search vector space restarts
Definition at line 340 of file quda.h.
Referenced by setInvertParam().
| double QudaInvertParam_s::eigenval_tol |
eigCG: selection criterion for the reduced eigenvector set
Definition at line 338 of file quda.h.
Referenced by setInvertParam().
| double QudaInvertParam_s::epsilon |
Twisted mass parameter
Definition at line 115 of file quda.h.
Referenced by dslashRef(), init(), main(), quda::setDiracParam(), setInvertParam(), and setMultigridParam().
| QudaExtLibType QudaInvertParam_s::extlib_type |
Which external library to use in the linear solvers (MAGMA or Eigen)
Definition at line 371 of file quda.h.
Referenced by setInvertParam().
| QudaGammaBasis QudaInvertParam_s::gamma_basis |
Gamma basis of the input and output host fields
Definition at line 221 of file quda.h.
Referenced by init(), invert_test(), main(), set_params(), setInvertParam(), and setMultigridParam().
| int QudaInvertParam_s::gcrNkrylov |
Maximum size of Krylov space used by solver
Definition at line 259 of file quda.h.
Referenced by invert_test(), main(), setInvertParam(), and setMultigridParam().
| double QudaInvertParam_s::gflops |
The Gflops rate of the solver
Definition at line 250 of file quda.h.
Referenced by eigensolveQuda(), invert_test(), invertMultiShiftQuda(), invertMultiSrcQuda(), invertQuda(), main(), and quda::SolverParam::updateInvertParam().
| int QudaInvertParam_s::heavy_quark_check |
Definition at line 165 of file quda.h.
Referenced by setInvertParam().
| double QudaInvertParam_s::inc_tol |
initCG tuning parameter: tolerance for cg refinement corrections in the deflation stage
Definition at line 344 of file quda.h.
Referenced by setInvertParam().
| QudaFieldLocation QudaInvertParam_s::input_location |
The location of the input field
Definition at line 99 of file quda.h.
Referenced by cloverQuda(), contractQuda(), dslashQuda(), dslashQuda_4dpc(), dslashQuda_mdwf(), eigensolveQuda(), init(), invertMultiShiftQuda(), invertMultiSrcQuda(), invertQuda(), main(), MatDagMatQuda(), MatQuda(), performWuppertalnStep(), printQudaInvertParam(), set_params(), setInvertParam(), and setMultigridParam().
| QudaInverterType QudaInvertParam_s::inv_type |
Which linear solver to use
Definition at line 103 of file quda.h.
Referenced by quda::createDirac(), quda::deflated_solver::deflated_solver(), quda::Deflation::increment(), invert_test(), main(), quda::Deflation::operator()(), printQudaInvertParam(), set_params(), quda::setDiracPreParam(), setInvertParam(), setMultigridParam(), and quda::SolverParam::SolverParam().
| QudaInverterType QudaInvertParam_s::inv_type_precondition |
The inner Krylov solver used in the preconditioner. Set to QUDA_INVALID_INVERTER to disable the preconditioner entirely.
Definition at line 270 of file quda.h.
Referenced by invertMultiSrcQuda(), invertQuda(), main(), printQudaInvertParam(), set_params(), and setInvertParam().
| int QudaInvertParam_s::iter |
The number of iterations performed by the solver
Definition at line 249 of file quda.h.
Referenced by eigensolveQuda(), invert_test(), invertMultiShiftQuda(), invertMultiSrcQuda(), invertQuda(), main(), and quda::SolverParam::updateInvertParam().
| double QudaInvertParam_s::iter_res_offset[QUDA_MAX_MULTI_SHIFT] |
Iterated L2 residual norm achieved in multi shift solver for each offset
Definition at line 188 of file quda.h.
Referenced by invertMultiShiftQuda(), and quda::SolverParam::updateInvertParam().
| double QudaInvertParam_s::kappa |
Used for Wilson and Wilson-clover
Definition at line 106 of file quda.h.
Referenced by dslashQuda_4dpc(), dslashRef(), init(), loadCloverQuda(), main(), quda::massRescale(), MatDagMatQuda(), MatQuda(), quda::setDiracParam(), setInvertParam(), and setMultigridParam().
| int QudaInvertParam_s::laplace3D |
omit this direction from laplace operator: x,y,z,t -> 0,1,2,3 (-1 is full 4D)
Definition at line 119 of file quda.h.
Referenced by quda::setDiracParam(), and setInvertParam().
| int QudaInvertParam_s::Ls |
Extent of the 5th dimension (for domain wall)
Definition at line 109 of file quda.h.
Referenced by quda::ColorSpinorParam::ColorSpinorParam(), init(), invert_multishift_quda_(), invert_quda_(), main(), set_params(), quda::setDiracParam(), setInvertParam(), and setMultigridParam().
| double QudaInvertParam_s::m5 |
Domain wall height
Definition at line 108 of file quda.h.
Referenced by dslashRef(), init(), main(), quda::massRescale(), quda::setDiracParam(), and setInvertParam().
| int QudaInvertParam_s::make_resident_solution |
Whether to make the solution vector(s) after the solve
Definition at line 347 of file quda.h.
Referenced by invertMultiShiftQuda(), invertMultiSrcQuda(), and invertQuda().
| double QudaInvertParam_s::mass |
Used for staggered only
Definition at line 105 of file quda.h.
Referenced by dslashQuda(), dslashRef(), init(), invertMultiShiftQuda(), main(), quda::massRescale(), set_params(), quda::setDiracParam(), setInvertParam(), and setMultigridParam().
| QudaMassNormalization QudaInvertParam_s::mass_normalization |
The mass normalization is being used by the caller
Definition at line 208 of file quda.h.
Referenced by dslashQuda(), init(), main(), quda::massRescale(), MatDagMatQuda(), MatQuda(), set_params(), setInvertParam(), and setMultigridParam().
| QudaMatPCType QudaInvertParam_s::matpc_type |
The preconditioned matrix type
Definition at line 206 of file quda.h.
Referenced by quda::MG::createCoarseDirac(), dslashRef(), init(), invert_test(), loadCloverQuda(), main(), quda::MG::operator()(), set_params(), quda::setDiracParam(), setInvertParam(), and setMultigridParam().
| int QudaInvertParam_s::max_hq_res_increase |
| int QudaInvertParam_s::max_hq_res_restart_total |
| int QudaInvertParam_s::max_res_increase |
| int QudaInvertParam_s::max_res_increase_total |
| int QudaInvertParam_s::max_restart_num |
initCG tuning parameter: maximum restarts
Definition at line 342 of file quda.h.
Referenced by setInvertParam().
| int QudaInvertParam_s::max_search_dim |
EeigCG : Search space dimension gmresdr : Krylov subspace dimension
Definition at line 332 of file quda.h.
Referenced by setInvertParam().
| int QudaInvertParam_s::maxiter |
Maximum number of iterations in the linear solver
Definition at line 128 of file quda.h.
Referenced by main(), set_params(), setInvertParam(), and setMultigridParam().
| int QudaInvertParam_s::maxiter_precondition |
Maximum number of iterations allowed in the inner solver
Definition at line 292 of file quda.h.
Referenced by main(), set_params(), and setInvertParam().
| double QudaInvertParam_s::mu |
Twisted mass parameter
Definition at line 114 of file quda.h.
Referenced by dslashRef(), init(), loadCloverQuda(), main(), quda::setDiracParam(), setInvertParam(), and setMultigridParam().
| int QudaInvertParam_s::nev |
How many vectors to compute after one solve for eigCG recommended values 8 or 16
Definition at line 328 of file quda.h.
Referenced by quda::deflated_solver::deflated_solver(), setDeflationParam(), and setInvertParam().
| int QudaInvertParam_s::Nsteps |
Number of steps in s-step algorithms
Definition at line 256 of file quda.h.
Referenced by main(), set_params(), and setInvertParam().
| int QudaInvertParam_s::num_offset |
Number of offsets in the multi-shift solver
Definition at line 169 of file quda.h.
Referenced by computeStaggeredForceQuda(), invert_multishift_quda_(), invert_test(), invertMultiShiftQuda(), main(), quda::massRescale(), printQudaInvertParam(), and setInvertParam().
| int QudaInvertParam_s::num_src |
Number of sources in the multiple source solver
Definition at line 171 of file quda.h.
Referenced by invert_test(), invertMultiSrcQuda(), and main().
| double QudaInvertParam_s::offset[QUDA_MAX_MULTI_SHIFT] |
Offsets for multi-shift solver
Definition at line 176 of file quda.h.
Referenced by invert_test(), invertMultiShiftQuda(), main(), quda::massRescale(), and quda::SolverParam::SolverParam().
| double QudaInvertParam_s::omega |
Relaxation parameter used in GCR-DD (default = 1.0)
Definition at line 295 of file quda.h.
Referenced by main(), and setInvertParam().
| QudaFieldLocation QudaInvertParam_s::output_location |
The location of the output field
Definition at line 100 of file quda.h.
Referenced by cloverQuda(), dslashQuda(), dslashQuda_4dpc(), dslashQuda_mdwf(), init(), invertMultiShiftQuda(), invertMultiSrcQuda(), invertQuda(), main(), MatDagMatQuda(), MatQuda(), performWuppertalnStep(), printQudaInvertParam(), set_params(), setInvertParam(), and setMultigridParam().
| int QudaInvertParam_s::overlap |
Width of domain overlaps
Definition at line 173 of file quda.h.
Referenced by checkGauge(), and quda::setDiracPreParam().
| int QudaInvertParam_s::pipeline |
Whether to use a pipelined solver with less global sums
Definition at line 167 of file quda.h.
Referenced by main(), set_params(), and setInvertParam().
| int QudaInvertParam_s::precondition_cycle |
Number of preconditioner cycles to perform per iteration
Definition at line 307 of file quda.h.
Referenced by main(), and setInvertParam().
| void* QudaInvertParam_s::preconditioner |
| QudaPreserveSource QudaInvertParam_s::preserve_source |
Preserve the source or not in the linear solver (deprecated)
Definition at line 211 of file quda.h.
Referenced by main(), set_params(), setInvertParam(), and setMultigridParam().
| double QudaInvertParam_s::reliable_delta |
Reliable update tolerance
Definition at line 129 of file quda.h.
Referenced by main(), printQudaInvertParam(), set_params(), setInvertParam(), and setMultigridParam().
| double QudaInvertParam_s::reliable_delta_refinement |
Reliable update tolerance used in post multi-shift solver refinement
Definition at line 130 of file quda.h.
Referenced by invertMultiShiftQuda(), and printQudaInvertParam().
| QudaResidualType QudaInvertParam_s::residual_type |
Whether to use the L2 relative residual, Fermilab heavy-quark residual, or both to determine convergence. To require that both stopping conditions are satisfied, use a bitwise OR as follows:
p.residual_type = (QudaResidualType) (QUDA_L2_RELATIVE_RESIDUAL | QUDA_HEAVY_QUARK_RESIDUAL);
Definition at line 320 of file quda.h.
Referenced by invertMultiShiftQuda(), main(), printQudaInvertParam(), set_params(), and setInvertParam().
| double QudaInvertParam_s::residue[QUDA_MAX_MULTI_SHIFT] |
Residuals in the partial faction expansion
Definition at line 194 of file quda.h.
Referenced by computeStaggeredForceQuda(), and invertMultiShiftQuda().
| int QudaInvertParam_s::return_clover |
Whether to copy back the clover matrix field
Definition at line 241 of file quda.h.
Referenced by init(), loadCloverQuda(), and main().
| int QudaInvertParam_s::return_clover_inverse |
Whether to copy back the inverted clover matrix field
Definition at line 242 of file quda.h.
Referenced by init(), loadCloverQuda(), and main().
| int QudaInvertParam_s::rhs_idx |
For systems with many RHS: current RHS index
Definition at line 334 of file quda.h.
Referenced by main(), setInvertParam(), quda::SolverParam::SolverParam(), quda::SolverParam::updateInvertParam(), and quda::SolverParam::updateRhsIndex().
| QudaSchwarzType QudaInvertParam_s::schwarz_type |
Whether to use additive or multiplicative Schwarz preconditioning
Definition at line 310 of file quda.h.
Referenced by main(), and setInvertParam().
| double QudaInvertParam_s::secs |
The time taken by the solver
Definition at line 251 of file quda.h.
Referenced by eigensolveQuda(), invert_test(), invertMultiShiftQuda(), invertMultiSrcQuda(), invertQuda(), main(), and quda::SolverParam::updateInvertParam().
| int QudaInvertParam_s::solution_accumulator_pipeline |
This parameter determines how many consecutive reliable update residual increases we tolerate before terminating the solver, i.e., how long do we want to keep trying to converge
Definition at line 142 of file quda.h.
Referenced by main(), and setInvertParam().
| QudaSolutionType QudaInvertParam_s::solution_type |
Type of system to solve
Definition at line 204 of file quda.h.
Referenced by eigensolveQuda(), init(), invert_test(), invertMultiShiftQuda(), invertMultiSrcQuda(), invertQuda(), loadCloverQuda(), main(), quda::massRescale(), MatDagMatQuda(), MatQuda(), quda::multigrid_solver::multigrid_solver(), set_params(), setInvertParam(), and setMultigridParam().
| QudaSolveType QudaInvertParam_s::solve_type |
How to solve it
Definition at line 205 of file quda.h.
Referenced by computeCloverForceQuda(), computeStaggeredForceQuda(), quda::deflated_solver::deflated_solver(), eigensolveQuda(), init(), invert_test(), invertMultiShiftQuda(), invertMultiSrcQuda(), invertQuda(), loadCloverQuda(), main(), quda::multigrid_solver::multigrid_solver(), set_params(), setInvertParam(), setMultigridParam(), and updateMultigridQuda().
| QudaSolverNormalization QudaInvertParam_s::solver_normalization |
The normalization desired in the solver
Definition at line 209 of file quda.h.
Referenced by invertMultiShiftQuda(), invertMultiSrcQuda(), invertQuda(), and main().
| int QudaInvertParam_s::sp_pad |
The padding to use for the fermion fields
Definition at line 246 of file quda.h.
Referenced by init(), main(), set_params(), setInvertParam(), and setMultigridParam().
| double QudaInvertParam_s::tol |
Solver tolerance in the L2 residual norm
Definition at line 121 of file quda.h.
Referenced by invert_test(), main(), set_params(), setInvertParam(), and setMultigridParam().
| double QudaInvertParam_s::tol_hq |
Solver tolerance in the heavy quark residual norm
Definition at line 123 of file quda.h.
Referenced by invert_test(), main(), set_params(), and setInvertParam().
| double QudaInvertParam_s::tol_hq_offset[QUDA_MAX_MULTI_SHIFT] |
Solver tolerance for each shift when refinement is applied using the heavy-quark residual
Definition at line 182 of file quda.h.
Referenced by invert_test(), invertMultiShiftQuda(), main(), setInvertParam(), and quda::SolverParam::SolverParam().
| double QudaInvertParam_s::tol_offset[QUDA_MAX_MULTI_SHIFT] |
Solver tolerance for each offset
Definition at line 179 of file quda.h.
Referenced by invert_test(), invertMultiShiftQuda(), main(), setInvertParam(), and quda::SolverParam::SolverParam().
| double QudaInvertParam_s::tol_precondition |
Tolerance in the inner solver
Definition at line 289 of file quda.h.
Referenced by main(), set_params(), and setInvertParam().
| double QudaInvertParam_s::tol_restart |
Solver tolerance in the L2 residual norm (used to restart InitCG)
Definition at line 122 of file quda.h.
Referenced by main(), set_params(), and setInvertParam().
| double QudaInvertParam_s::trlogA[2] |
The trace log of the clover term (even/odd computed separately)
Definition at line 237 of file quda.h.
Referenced by loadCloverQuda().
| double QudaInvertParam_s::true_res |
Whether to compute the true residual post solve Actual L2 residual norm achieved in solver
Definition at line 126 of file quda.h.
Referenced by invert_test(), main(), and quda::SolverParam::updateInvertParam().
| double QudaInvertParam_s::true_res_hq |
Actual heavy quark residual norm achieved in solver
Definition at line 127 of file quda.h.
Referenced by invert_test(), main(), and quda::SolverParam::updateInvertParam().
| double QudaInvertParam_s::true_res_hq_offset[QUDA_MAX_MULTI_SHIFT] |
Actual heavy quark residual norm achieved in solver for each offset
Definition at line 191 of file quda.h.
Referenced by invert_test(), invertMultiShiftQuda(), main(), and quda::SolverParam::updateInvertParam().
| double QudaInvertParam_s::true_res_offset[QUDA_MAX_MULTI_SHIFT] |
Actual L2 residual norm achieved in solver for each offset
Definition at line 185 of file quda.h.
Referenced by invert_test(), invertMultiShiftQuda(), main(), and quda::SolverParam::updateInvertParam().
| QudaTune QudaInvertParam_s::tune |
| QudaTwistFlavorType QudaInvertParam_s::twist_flavor |
Twisted mass flavor
Definition at line 117 of file quda.h.
Referenced by dslashRef(), init(), main(), quda::setDiracParam(), setInvertParam(), and setMultigridParam().
| int QudaInvertParam_s::use_alternative_reliable |
Whether to use alternative reliable updates
Definition at line 131 of file quda.h.
Referenced by main(), and setInvertParam().
| QudaUseInitGuess QudaInvertParam_s::use_init_guess |
Whether to use an initial guess in the solver or not
Definition at line 231 of file quda.h.
Referenced by invertMultiSrcQuda(), invertQuda(), main(), and set_params().
| int QudaInvertParam_s::use_resident_solution |
Whether to use the resident solution vector(s)
Definition at line 350 of file quda.h.
Referenced by computeCloverForceQuda(), computeStaggeredForceQuda(), and invertQuda().
| int QudaInvertParam_s::use_sloppy_partial_accumulator |
Whether to keep the partial solution accumuator in sloppy precision This parameter determines how often we accumulate into the solution vector from the direction vectors in the solver. E.g., running with solution_accumulator_pipeline = 4, means we will update the solution vector every four iterations using the direction vectors from the prior four iterations. This increases performance of mixed-precision solvers since it means less high-precision vector round-trip memory travel, but requires more low-precision memory allocation.
Definition at line 132 of file quda.h.
Referenced by main(), set_params(), and setInvertParam().
| QudaVerbosity QudaInvertParam_s::verbosity |
The verbosity setting to use in the solver
Definition at line 244 of file quda.h.
Referenced by cloverQuda(), dslashQuda(), dslashQuda_4dpc(), dslashQuda_mdwf(), dumpMultigridQuda(), eigensolveQuda(), init(), invertMultiShiftQuda(), invertMultiSrcQuda(), invertQuda(), loadCloverQuda(), main(), MatDagMatQuda(), MatQuda(), newMultigridQuda(), performWuppertalnStep(), set_params(), setInvertParam(), setMultigridParam(), and updateMultigridQuda().
| QudaVerbosity QudaInvertParam_s::verbosity_precondition |
Verbosity of the inner Krylov solver
Definition at line 286 of file quda.h.
Referenced by main(), set_params(), setInvertParam(), and setMultigridParam().
1.8.13
