1lis_solver_set_option(3f) Lis Library Functions lis_solver_set_option(3f)
2
3
4
6 lis_solver_set_option - set the options for the solver
7
8
10 subroutine lis_solver_set_option(char *text, LIS_SOLVER *solver,
11 LIS_INTEGER ierr);
12
13
15 Set the options for the solver.
16
17
19 text The command line options
20
21
23 solver The solver
24
25
27 The table below shows the available command line options, where -i
28 {cg|1} means -i cg or -i 1 and -maxiter [1000] indicates that -maxiter
29 defaults to 1,000.
30
31
33 The following options are supported:
34
35 -i linear solver
36 The following options are supported for linear solver:
37
38 -i {cg|1}
39 CG
40
41 -i {bicg|2}
42 BiCG
43
44 -i {cgs|3}
45 CGS
46
47 -i {bicgstab|4}
48 BiCGSTAB
49
50 -i {bicgstabl|5}
51 BiCGSTAB(l)
52
53 -ell [2]
54 The degree l
55
56 -i {gpbicg|6}
57 GPBiCG
58
59 -i {tfqmr|7}
60 TFQMR
61
62 -i {orthomin|8}
63 Orthomin(m)
64
65 -restart [40]
66 The restart value m
67
68 -i {gmres|9}
69 GMRES(m)
70
71 -restart [40]
72 The restart value m
73
74 -i {jacobi|10}
75 Jacobi
76
77 -i {gs|11}
78 Gauss-Seidel
79
80 -i {sor|12}
81 SOR
82
83 -omega [1.9]
84 The relaxation coefficient omega (0<omega<2)
85
86 -i {bicgsafe|13}
87 BiCGSafe
88
89 -i {cr|14}
90 CR
91
92 -i {bicr|15}
93 BiCR
94
95 -i {crs|16}
96 CRS
97
98 -i {bicrstab|17}
99 BiCRSTAB
100
101 -i {gpbicr|18}
102 GPBiCR
103
104 -i {bicrsafe|19}
105 BiCRSafe
106
107 -i {fgmres|20}
108 FGMRES(m)
109
110 -restart [40]
111 The restart value m
112
113 -i {idrs|21}
114 IDR(s)
115
116 -irestart [2]
117 The restart value s
118
119 -i {idr1|22}
120 IDR(1)
121
122 -i {minres|23}
123 MINRES
124
125 -i {cocg|24}
126 COCG
127
128 -i {cocr|25}
129 COCR
130
131
132 -p preconditioner
133 The following options are supported for preconditioner:
134
135 -p {none|0}
136 None
137
138 -p {jacobi|1}
139 Jacobi
140
141 -p {ilu|2}
142 ILU(k)
143
144 -ilu_fill [0]
145 The fill level k
146
147 -p {ssor|3}
148 SSOR
149
150 -ssor_omega [1.0]
151 The relaxation coefficient omega (0<omega<2)
152
153 -p {hybrid|4}
154 Hybrid
155
156 -hybrid_i [sor]
157 The linear solver
158
159 -hybrid_maxiter [25]
160 The maximum number of iterations
161
162 -hybrid_tol [1.0e-3]
163 The convergence criterion
164
165 -hybrid_omega [1.5]
166 The relaxation coefficient omega of the SOR
167 (0<omega<2)
168
169 -hybrid_ell [2]
170 The degree l of the BiCGSTAB(l)
171
172 -hybrid_restart [40]
173 The restart values of the GMRES and Orthomin
174
175 -p {is|5}
176 I+S
177
178 -is_alpha [1.0]
179 The parameter alpha of I+alpha*S(m)
180
181 -is_m [3]
182 The parameter m of I+alpha*S(m)
183
184 -p {sainv|6}
185 SAINV
186
187 -sainv_drop [0.05]
188 The drop criterion
189
190 -p {saamg|7}
191 SA-AMG
192
193 -saamg_unsym [false]
194 Select the unsymmetric version (The matrix strucā
195 ture must be symmetric)
196
197 -saamg_theta [0.05|0.12]
198 The drop criterion
199
200 -p {iluc|8}
201 Crout ILU
202
203 -iluc_drop [0.05]
204 The drop criterion
205
206 -iluc_rate [5.0]
207 The ration of maximum fill-in
208
209 -p {ilut|9}
210 ILUT
211
212 -ilut_drop [0.05]
213 The drop criterion
214
215 -ilut_rate [5.0]
216 The ration of maximum fill-in
217
218 -adds true
219 Additive Schwarz
220
221 -adds_iter [1]
222 The number of iteration
223
224 Other Options:
225
226 -maxiter [1000]
227 The maximum number of iterations
228
229 -tol [1.0e-12]
230 The convergence criterion
231
232 -print [0]
233 The display of the residual
234
235 -print {none|0}
236 None
237
238 -print {mem|1}
239 Save the residual history
240
241 -print {out|2}
242 Display the residual history
243
244 -print {all|3}
245 Save the residual history and display it on the screen
246
247 -scale [0]
248 The scaling
249
250 -scale {none|0}
251 No scaling
252
253 -scale {jacobi|1}
254 The Jacobi scaling
255
256 -scale {symm_diag|2}
257 The diagonal scaling
258
259 -initx_zeros [true]
260 The behavior of the initial vector x_0
261
262 -initx_zero {false|0}
263 Given values
264
265 -initx_zero {true|1}
266 All values are set to 0
267
268 -omp_num_threads [t]
269 The number of threads (t represents the maximum number of
270 threads)
271
272 -storage [0]
273 The matrix storage format
274
275 -storage_block [2]
276 The block size of the BSR and BSC formats
277
278 -f [0] The precision of the linear solver
279
280 -f {double|0}
281 Double precision
282
283 -f {quad|1}
284 Double-double (quadruple) precision
285
286 See Lis User Guide for full description.
287
288
290 lis(3)
291
292 http://www.ssisc.org/lis/
293
294
295
296
297Man Page 14 Sep 2017 lis_solver_set_option(3f)