| -ksp_max_it | - maximum number of linear iterations
|
| -ksp_rtol rtol | - relative tolerance used in default determination of convergence, i.e.
if residual norm decreases by this factor than convergence is declared
|
| -ksp_atol abstol | - absolute tolerance used in default convergence test, i.e. if residual
norm is less than this then convergence is declared
|
| -ksp_divtol tol | - if residual norm increases by this factor than divergence is declared
|
| -ksp_converged_use_initial_residual_norm | - see KSPConvergedDefaultSetUIRNorm()
|
| -ksp_converged_use_min_initial_residual_norm | - see KSPConvergedDefaultSetUMIRNorm()
|
| -ksp_norm_type | - none - skip norms used in convergence tests (useful only when not using
convergence test (say you always want to run with 5 iterations) to
save on communication overhead
preconditioned - default for left preconditioning
unpreconditioned - see KSPSetNormType()
natural - see KSPSetNormType()
|
| -ksp_check_norm_iteration it | - do not compute residual norm until iteration number it (does compute at 0th iteration)
works only for PCBCGS, PCIBCGS and and PCCG
|
| -ksp_lag_norm | - compute the norm of the residual for the ith iteration on the i+1 iteration; this means that one can use
the norm of the residual for convergence test WITHOUT an extra MPI_Allreduce() limiting global synchronizations.
This will require 1 more iteration of the solver than usual.
|
| -ksp_guess_type | - Type of initial guess generator for repeated linear solves
|
| -ksp_fischer_guess <model,size> | - uses the Fischer initial guess generator for repeated linear solves
|
| -ksp_constant_null_space | - assume the operator (matrix) has the constant vector in its null space
|
| -ksp_test_null_space | - tests the null space set with MatSetNullSpace() to see if it truly is a null space
|
| -ksp_knoll | - compute initial guess by applying the preconditioner to the right hand side
|
| -ksp_monitor_cancel | - cancel all previous convergene monitor routines set
|
| -ksp_monitor <optional filename> | - print residual norm at each iteration
|
| -ksp_monitor_lg_residualnorm | - plot residual norm at each iteration
|
| -ksp_monitor_solution [ascii binary or draw][:filename][:format option] | - plot solution at each iteration
|
| -ksp_monitor_singular_value | - monitor extreme singular values at each iteration
|