1 2 #include <petsc/private/tsimpl.h> /*I "petscts.h" I*/ 3 #include <petscdmshell.h> 4 #include <petscdmda.h> 5 #include <petscviewer.h> 6 #include <petscdraw.h> 7 8 /* Logging support */ 9 PetscClassId TS_CLASSID, DMTS_CLASSID; 10 PetscLogEvent TS_Step, TS_PseudoComputeTimeStep, TS_FunctionEval, TS_JacobianEval; 11 12 const char *const TSExactFinalTimeOptions[] = {"STEPOVER","INTERPOLATE","MATCHSTEP","TSExactFinalTimeOption","TS_EXACTFINALTIME_",0}; 13 14 struct _n_TSMonitorDrawCtx { 15 PetscViewer viewer; 16 PetscDrawAxis axis; 17 Vec initialsolution; 18 PetscBool showinitial; 19 PetscInt howoften; /* when > 0 uses step % howoften, when negative only final solution plotted */ 20 PetscBool showtimestepandtime; 21 int color; 22 }; 23 24 #undef __FUNCT__ 25 #define __FUNCT__ "TSSetFromOptions" 26 /*@ 27 TSSetFromOptions - Sets various TS parameters from user options. 28 29 Collective on TS 30 31 Input Parameter: 32 . ts - the TS context obtained from TSCreate() 33 34 Options Database Keys: 35 + -ts_type <type> - TSEULER, TSBEULER, TSSUNDIALS, TSPSEUDO, TSCN, TSRK, TSTHETA, TSGL, TSSSP 36 . -ts_save_trajectory - checkpoint the solution at each time-step 37 . -ts_max_steps <maxsteps> - maximum number of time-steps to take 38 . -ts_final_time <time> - maximum time to compute to 39 . -ts_dt <dt> - initial time step 40 . -ts_exact_final_time <stepover,interpolate,matchstep> whether to stop at the exact given final time and how to compute the solution at that ti,e 41 . -ts_max_snes_failures <maxfailures> - Maximum number of nonlinear solve failures allowed 42 . -ts_max_reject <maxrejects> - Maximum number of step rejections before step fails 43 . -ts_error_if_step_fails <true,false> - Error if no step succeeds 44 . -ts_rtol <rtol> - relative tolerance for local truncation error 45 . -ts_atol <atol> Absolute tolerance for local truncation error 46 . -ts_adjoint_solve <yes,no> After solving the ODE/DAE solve the adjoint problem (requires -ts_save_trajectory) 47 . -ts_fd_color - Use finite differences with coloring to compute IJacobian 48 . -ts_monitor - print information at each timestep 49 . -ts_monitor_lg_timestep - Monitor timestep size graphically 50 . -ts_monitor_lg_solution - Monitor solution graphically 51 . -ts_monitor_lg_error - Monitor error graphically 52 . -ts_monitor_lg_snes_iterations - Monitor number nonlinear iterations for each timestep graphically 53 . -ts_monitor_lg_ksp_iterations - Monitor number nonlinear iterations for each timestep graphically 54 . -ts_monitor_sp_eig - Monitor eigenvalues of linearized operator graphically 55 . -ts_monitor_draw_solution - Monitor solution graphically 56 . -ts_monitor_draw_solution_phase <xleft,yleft,xright,yright> - Monitor solution graphically with phase diagram, requires problem with exactly 2 degrees of freedom 57 . -ts_monitor_draw_error - Monitor error graphically, requires use to have provided TSSetSolutionFunction() 58 . -ts_monitor_solution_binary <filename> - Save each solution to a binary file 59 . -ts_monitor_solution_vtk <filename.vts> - Save each time step to a binary file, use filename-%%03D.vts 60 - -ts_monitor_envelope - determine maximum and minimum value of each component of the solution over the solution time 61 62 Developer Note: We should unify all the -ts_monitor options in the way that -xxx_view has been unified 63 64 Level: beginner 65 66 .keywords: TS, timestep, set, options, database 67 68 .seealso: TSGetType() 69 @*/ 70 PetscErrorCode TSSetFromOptions(TS ts) 71 { 72 PetscBool opt,flg,tflg; 73 PetscErrorCode ierr; 74 PetscViewer monviewer; 75 char monfilename[PETSC_MAX_PATH_LEN]; 76 SNES snes; 77 TSAdapt adapt; 78 PetscReal time_step; 79 TSExactFinalTimeOption eftopt; 80 char dir[16]; 81 const char *defaultType; 82 char typeName[256]; 83 84 PetscFunctionBegin; 85 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 86 ierr = PetscObjectOptionsBegin((PetscObject)ts);CHKERRQ(ierr); 87 if (((PetscObject)ts)->type_name) defaultType = ((PetscObject)ts)->type_name; 88 else defaultType = TSEULER; 89 90 ierr = TSRegisterAll();CHKERRQ(ierr); 91 ierr = PetscOptionsFList("-ts_type", "TS method"," TSSetType", TSList, defaultType, typeName, 256, &opt);CHKERRQ(ierr); 92 if (opt) { 93 ierr = TSSetType(ts, typeName);CHKERRQ(ierr); 94 } else { 95 ierr = TSSetType(ts, defaultType);CHKERRQ(ierr); 96 } 97 98 /* Handle generic TS options */ 99 if (ts->trajectory) tflg = PETSC_TRUE; 100 else tflg = PETSC_FALSE; 101 ierr = PetscOptionsBool("-ts_save_trajectory","Save the solution at each timestep","TSSetSaveTrajectory",tflg,&tflg,NULL);CHKERRQ(ierr); 102 if (tflg) {ierr = TSSetSaveTrajectory(ts);CHKERRQ(ierr);} 103 if (ts->adjoint_solve) tflg = PETSC_TRUE; 104 else tflg = PETSC_FALSE; 105 ierr = PetscOptionsBool("-ts_adjoint_solve","Solve the adjoint problem immediately after solving the forward problem","",tflg,&tflg,&flg);CHKERRQ(ierr); 106 if (flg) { 107 ierr = TSSetSaveTrajectory(ts);CHKERRQ(ierr); 108 ts->adjoint_solve = tflg; 109 } 110 ierr = PetscOptionsInt("-ts_max_steps","Maximum number of time steps","TSSetDuration",ts->max_steps,&ts->max_steps,NULL);CHKERRQ(ierr); 111 ierr = PetscOptionsReal("-ts_final_time","Time to run to","TSSetDuration",ts->max_time,&ts->max_time,NULL);CHKERRQ(ierr); 112 ierr = PetscOptionsReal("-ts_init_time","Initial time","TSSetTime",ts->ptime,&ts->ptime,NULL);CHKERRQ(ierr); 113 ierr = PetscOptionsReal("-ts_dt","Initial time step","TSSetTimeStep",ts->time_step,&time_step,&flg);CHKERRQ(ierr); 114 if (flg) { 115 ierr = TSSetTimeStep(ts,time_step);CHKERRQ(ierr); 116 } 117 ierr = PetscOptionsEnum("-ts_exact_final_time","Option for handling of final time step","TSSetExactFinalTime",TSExactFinalTimeOptions,(PetscEnum)ts->exact_final_time,(PetscEnum*)&eftopt,&flg);CHKERRQ(ierr); 118 if (flg) {ierr = TSSetExactFinalTime(ts,eftopt);CHKERRQ(ierr);} 119 ierr = PetscOptionsInt("-ts_max_snes_failures","Maximum number of nonlinear solve failures","TSSetMaxSNESFailures",ts->max_snes_failures,&ts->max_snes_failures,NULL);CHKERRQ(ierr); 120 ierr = PetscOptionsInt("-ts_max_reject","Maximum number of step rejections before step fails","TSSetMaxStepRejections",ts->max_reject,&ts->max_reject,NULL);CHKERRQ(ierr); 121 ierr = PetscOptionsBool("-ts_error_if_step_fails","Error if no step succeeds","TSSetErrorIfStepFails",ts->errorifstepfailed,&ts->errorifstepfailed,NULL);CHKERRQ(ierr); 122 ierr = PetscOptionsReal("-ts_rtol","Relative tolerance for local truncation error","TSSetTolerances",ts->rtol,&ts->rtol,NULL);CHKERRQ(ierr); 123 ierr = PetscOptionsReal("-ts_atol","Absolute tolerance for local truncation error","TSSetTolerances",ts->atol,&ts->atol,NULL);CHKERRQ(ierr); 124 125 #if defined(PETSC_HAVE_SAWS) 126 { 127 PetscBool set; 128 flg = PETSC_FALSE; 129 ierr = PetscOptionsBool("-ts_saws_block","Block for SAWs memory snooper at end of TSSolve","PetscObjectSAWsBlock",((PetscObject)ts)->amspublishblock,&flg,&set);CHKERRQ(ierr); 130 if (set) { 131 ierr = PetscObjectSAWsSetBlock((PetscObject)ts,flg);CHKERRQ(ierr); 132 } 133 } 134 #endif 135 136 /* Monitor options */ 137 ierr = PetscOptionsString("-ts_monitor","Monitor timestep size","TSMonitorDefault","stdout",monfilename,PETSC_MAX_PATH_LEN,&flg);CHKERRQ(ierr); 138 if (flg) { 139 ierr = PetscViewerASCIIOpen(PetscObjectComm((PetscObject)ts),monfilename,&monviewer);CHKERRQ(ierr); 140 ierr = TSMonitorSet(ts,TSMonitorDefault,monviewer,(PetscErrorCode (*)(void**))PetscViewerDestroy);CHKERRQ(ierr); 141 } 142 ierr = PetscOptionsString("-ts_monitor_python","Use Python function","TSMonitorSet",0,monfilename,PETSC_MAX_PATH_LEN,&flg);CHKERRQ(ierr); 143 if (flg) {ierr = PetscPythonMonitorSet((PetscObject)ts,monfilename);CHKERRQ(ierr);} 144 145 ierr = PetscOptionsName("-ts_monitor_lg_timestep","Monitor timestep size graphically","TSMonitorLGTimeStep",&opt);CHKERRQ(ierr); 146 if (opt) { 147 TSMonitorLGCtx ctx; 148 PetscInt howoften = 1; 149 150 ierr = PetscOptionsInt("-ts_monitor_lg_timestep","Monitor timestep size graphically","TSMonitorLGTimeStep",howoften,&howoften,NULL);CHKERRQ(ierr); 151 ierr = TSMonitorLGCtxCreate(PetscObjectComm((PetscObject)ts),0,0,PETSC_DECIDE,PETSC_DECIDE,300,300,howoften,&ctx);CHKERRQ(ierr); 152 ierr = TSMonitorSet(ts,TSMonitorLGTimeStep,ctx,(PetscErrorCode (*)(void**))TSMonitorLGCtxDestroy);CHKERRQ(ierr); 153 } 154 ierr = PetscOptionsName("-ts_monitor_lg_solution","Monitor solution graphically","TSMonitorLGSolution",&opt);CHKERRQ(ierr); 155 if (opt) { 156 TSMonitorLGCtx ctx; 157 PetscInt howoften = 1; 158 159 ierr = PetscOptionsInt("-ts_monitor_lg_solution","Monitor solution graphically","TSMonitorLGSolution",howoften,&howoften,NULL);CHKERRQ(ierr); 160 ierr = TSMonitorLGCtxCreate(PETSC_COMM_SELF,0,0,PETSC_DECIDE,PETSC_DECIDE,600,400,howoften,&ctx);CHKERRQ(ierr); 161 ierr = TSMonitorSet(ts,TSMonitorLGSolution,ctx,(PetscErrorCode (*)(void**))TSMonitorLGCtxDestroy);CHKERRQ(ierr); 162 } 163 ierr = PetscOptionsName("-ts_monitor_lg_error","Monitor error graphically","TSMonitorLGError",&opt);CHKERRQ(ierr); 164 if (opt) { 165 TSMonitorLGCtx ctx; 166 PetscInt howoften = 1; 167 168 ierr = PetscOptionsInt("-ts_monitor_lg_error","Monitor error graphically","TSMonitorLGError",howoften,&howoften,NULL);CHKERRQ(ierr); 169 ierr = TSMonitorLGCtxCreate(PETSC_COMM_SELF,0,0,PETSC_DECIDE,PETSC_DECIDE,600,400,howoften,&ctx);CHKERRQ(ierr); 170 ierr = TSMonitorSet(ts,TSMonitorLGError,ctx,(PetscErrorCode (*)(void**))TSMonitorLGCtxDestroy);CHKERRQ(ierr); 171 } 172 ierr = PetscOptionsName("-ts_monitor_lg_snes_iterations","Monitor number nonlinear iterations for each timestep graphically","TSMonitorLGSNESIterations",&opt);CHKERRQ(ierr); 173 if (opt) { 174 TSMonitorLGCtx ctx; 175 PetscInt howoften = 1; 176 177 ierr = PetscOptionsInt("-ts_monitor_lg_snes_iterations","Monitor number nonlinear iterations for each timestep graphically","TSMonitorLGSNESIterations",howoften,&howoften,NULL);CHKERRQ(ierr); 178 ierr = TSMonitorLGCtxCreate(PETSC_COMM_SELF,0,0,PETSC_DECIDE,PETSC_DECIDE,300,300,howoften,&ctx);CHKERRQ(ierr); 179 ierr = TSMonitorSet(ts,TSMonitorLGSNESIterations,ctx,(PetscErrorCode (*)(void**))TSMonitorLGCtxDestroy);CHKERRQ(ierr); 180 } 181 ierr = PetscOptionsName("-ts_monitor_lg_ksp_iterations","Monitor number nonlinear iterations for each timestep graphically","TSMonitorLGKSPIterations",&opt);CHKERRQ(ierr); 182 if (opt) { 183 TSMonitorLGCtx ctx; 184 PetscInt howoften = 1; 185 186 ierr = PetscOptionsInt("-ts_monitor_lg_ksp_iterations","Monitor number nonlinear iterations for each timestep graphically","TSMonitorLGKSPIterations",howoften,&howoften,NULL);CHKERRQ(ierr); 187 ierr = TSMonitorLGCtxCreate(PETSC_COMM_SELF,0,0,PETSC_DECIDE,PETSC_DECIDE,300,300,howoften,&ctx);CHKERRQ(ierr); 188 ierr = TSMonitorSet(ts,TSMonitorLGKSPIterations,ctx,(PetscErrorCode (*)(void**))TSMonitorLGCtxDestroy);CHKERRQ(ierr); 189 } 190 ierr = PetscOptionsName("-ts_monitor_sp_eig","Monitor eigenvalues of linearized operator graphically","TSMonitorSPEig",&opt);CHKERRQ(ierr); 191 if (opt) { 192 TSMonitorSPEigCtx ctx; 193 PetscInt howoften = 1; 194 195 ierr = PetscOptionsInt("-ts_monitor_sp_eig","Monitor eigenvalues of linearized operator graphically","TSMonitorSPEig",howoften,&howoften,NULL);CHKERRQ(ierr); 196 ierr = TSMonitorSPEigCtxCreate(PETSC_COMM_SELF,0,0,PETSC_DECIDE,PETSC_DECIDE,600,400,howoften,&ctx);CHKERRQ(ierr); 197 ierr = TSMonitorSet(ts,TSMonitorSPEig,ctx,(PetscErrorCode (*)(void**))TSMonitorSPEigCtxDestroy);CHKERRQ(ierr); 198 } 199 opt = PETSC_FALSE; 200 ierr = PetscOptionsName("-ts_monitor_draw_solution","Monitor solution graphically","TSMonitorDrawSolution",&opt);CHKERRQ(ierr); 201 if (opt) { 202 TSMonitorDrawCtx ctx; 203 PetscInt howoften = 1; 204 205 ierr = PetscOptionsInt("-ts_monitor_draw_solution","Monitor solution graphically","TSMonitorDrawSolution",howoften,&howoften,NULL);CHKERRQ(ierr); 206 ierr = TSMonitorDrawCtxCreate(PetscObjectComm((PetscObject)ts),0,0,PETSC_DECIDE,PETSC_DECIDE,600,400,howoften,&ctx);CHKERRQ(ierr); 207 ierr = TSMonitorSet(ts,TSMonitorDrawSolution,ctx,(PetscErrorCode (*)(void**))TSMonitorDrawCtxDestroy);CHKERRQ(ierr); 208 } 209 opt = PETSC_FALSE; 210 ierr = PetscOptionsName("-ts_monitor_draw_solution_phase","Monitor solution graphically","TSMonitorDrawSolutionPhase",&opt);CHKERRQ(ierr); 211 if (opt) { 212 TSMonitorDrawCtx ctx; 213 PetscReal bounds[4]; 214 PetscInt n = 4; 215 PetscDraw draw; 216 217 ierr = PetscOptionsRealArray("-ts_monitor_draw_solution_phase","Monitor solution graphically","TSMonitorDrawSolutionPhase",bounds,&n,NULL);CHKERRQ(ierr); 218 if (n != 4) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_ARG_WRONG,"Must provide bounding box of phase field"); 219 ierr = TSMonitorDrawCtxCreate(PetscObjectComm((PetscObject)ts),0,0,PETSC_DECIDE,PETSC_DECIDE,600,400,1,&ctx);CHKERRQ(ierr); 220 ierr = PetscViewerDrawGetDraw(ctx->viewer,0,&draw);CHKERRQ(ierr); 221 ierr = PetscDrawClear(draw);CHKERRQ(ierr); 222 ierr = PetscDrawAxisCreate(draw,&ctx->axis);CHKERRQ(ierr); 223 ierr = PetscDrawAxisSetLimits(ctx->axis,bounds[0],bounds[2],bounds[1],bounds[3]);CHKERRQ(ierr); 224 ierr = PetscDrawAxisSetLabels(ctx->axis,"Phase Diagram","Variable 1","Variable 2");CHKERRQ(ierr); 225 ierr = PetscDrawAxisDraw(ctx->axis);CHKERRQ(ierr); 226 /* ierr = PetscDrawSetCoordinates(draw,bounds[0],bounds[1],bounds[2],bounds[3]);CHKERRQ(ierr); */ 227 ierr = TSMonitorSet(ts,TSMonitorDrawSolutionPhase,ctx,(PetscErrorCode (*)(void**))TSMonitorDrawCtxDestroy);CHKERRQ(ierr); 228 } 229 opt = PETSC_FALSE; 230 ierr = PetscOptionsName("-ts_monitor_draw_error","Monitor error graphically","TSMonitorDrawError",&opt);CHKERRQ(ierr); 231 if (opt) { 232 TSMonitorDrawCtx ctx; 233 PetscInt howoften = 1; 234 235 ierr = PetscOptionsInt("-ts_monitor_draw_error","Monitor error graphically","TSMonitorDrawError",howoften,&howoften,NULL);CHKERRQ(ierr); 236 ierr = TSMonitorDrawCtxCreate(PetscObjectComm((PetscObject)ts),0,0,PETSC_DECIDE,PETSC_DECIDE,600,400,howoften,&ctx);CHKERRQ(ierr); 237 ierr = TSMonitorSet(ts,TSMonitorDrawError,ctx,(PetscErrorCode (*)(void**))TSMonitorDrawCtxDestroy);CHKERRQ(ierr); 238 } 239 opt = PETSC_FALSE; 240 ierr = PetscOptionsString("-ts_monitor_solution_binary","Save each solution to a binary file","TSMonitorSolutionBinary",0,monfilename,PETSC_MAX_PATH_LEN,&flg);CHKERRQ(ierr); 241 if (flg) { 242 PetscViewer ctx; 243 if (monfilename[0]) { 244 ierr = PetscViewerBinaryOpen(PetscObjectComm((PetscObject)ts),monfilename,FILE_MODE_WRITE,&ctx);CHKERRQ(ierr); 245 ierr = TSMonitorSet(ts,TSMonitorSolutionBinary,ctx,(PetscErrorCode (*)(void**))PetscViewerDestroy);CHKERRQ(ierr); 246 } else { 247 ctx = PETSC_VIEWER_BINARY_(PetscObjectComm((PetscObject)ts)); 248 ierr = TSMonitorSet(ts,TSMonitorSolutionBinary,ctx,(PetscErrorCode (*)(void**))NULL);CHKERRQ(ierr); 249 } 250 } 251 opt = PETSC_FALSE; 252 ierr = PetscOptionsString("-ts_monitor_solution_vtk","Save each time step to a binary file, use filename-%%03D.vts","TSMonitorSolutionVTK",0,monfilename,PETSC_MAX_PATH_LEN,&flg);CHKERRQ(ierr); 253 if (flg) { 254 const char *ptr,*ptr2; 255 char *filetemplate; 256 if (!monfilename[0]) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"-ts_monitor_solution_vtk requires a file template, e.g. filename-%%03D.vts"); 257 /* Do some cursory validation of the input. */ 258 ierr = PetscStrstr(monfilename,"%",(char**)&ptr);CHKERRQ(ierr); 259 if (!ptr) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"-ts_monitor_solution_vtk requires a file template, e.g. filename-%%03D.vts"); 260 for (ptr++; ptr && *ptr; ptr++) { 261 ierr = PetscStrchr("DdiouxX",*ptr,(char**)&ptr2);CHKERRQ(ierr); 262 if (!ptr2 && (*ptr < '0' || '9' < *ptr)) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"Invalid file template argument to -ts_monitor_solution_vtk, should look like filename-%%03D.vts"); 263 if (ptr2) break; 264 } 265 ierr = PetscStrallocpy(monfilename,&filetemplate);CHKERRQ(ierr); 266 ierr = TSMonitorSet(ts,TSMonitorSolutionVTK,filetemplate,(PetscErrorCode (*)(void**))TSMonitorSolutionVTKDestroy);CHKERRQ(ierr); 267 } 268 269 ierr = PetscOptionsString("-ts_monitor_dmda_ray","Display a ray of the solution","None","y=0",dir,16,&flg);CHKERRQ(ierr); 270 if (flg) { 271 TSMonitorDMDARayCtx *rayctx; 272 int ray = 0; 273 DMDADirection ddir; 274 DM da; 275 PetscMPIInt rank; 276 277 if (dir[1] != '=') SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_ARG_WRONG,"Unknown ray %s",dir); 278 if (dir[0] == 'x') ddir = DMDA_X; 279 else if (dir[0] == 'y') ddir = DMDA_Y; 280 else SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_ARG_WRONG,"Unknown ray %s",dir); 281 sscanf(dir+2,"%d",&ray); 282 283 ierr = PetscInfo2(((PetscObject)ts),"Displaying DMDA ray %c = %D\n",dir[0],ray);CHKERRQ(ierr); 284 ierr = PetscNew(&rayctx);CHKERRQ(ierr); 285 ierr = TSGetDM(ts,&da);CHKERRQ(ierr); 286 ierr = DMDAGetRay(da,ddir,ray,&rayctx->ray,&rayctx->scatter);CHKERRQ(ierr); 287 ierr = MPI_Comm_rank(PetscObjectComm((PetscObject)ts),&rank);CHKERRQ(ierr); 288 if (!rank) { 289 ierr = PetscViewerDrawOpen(PETSC_COMM_SELF,0,0,0,0,600,300,&rayctx->viewer);CHKERRQ(ierr); 290 } 291 rayctx->lgctx = NULL; 292 ierr = TSMonitorSet(ts,TSMonitorDMDARay,rayctx,TSMonitorDMDARayDestroy);CHKERRQ(ierr); 293 } 294 ierr = PetscOptionsString("-ts_monitor_lg_dmda_ray","Display a ray of the solution","None","x=0",dir,16,&flg);CHKERRQ(ierr); 295 if (flg) { 296 TSMonitorDMDARayCtx *rayctx; 297 int ray = 0; 298 DMDADirection ddir; 299 DM da; 300 PetscInt howoften = 1; 301 302 if (dir[1] != '=') SETERRQ1(PetscObjectComm((PetscObject) ts), PETSC_ERR_ARG_WRONG, "Malformed ray %s", dir); 303 if (dir[0] == 'x') ddir = DMDA_X; 304 else if (dir[0] == 'y') ddir = DMDA_Y; 305 else SETERRQ1(PetscObjectComm((PetscObject) ts), PETSC_ERR_ARG_WRONG, "Unknown ray direction %s", dir); 306 sscanf(dir+2, "%d", &ray); 307 308 ierr = PetscInfo2(((PetscObject) ts),"Displaying LG DMDA ray %c = %D\n", dir[0], ray);CHKERRQ(ierr); 309 ierr = PetscNew(&rayctx);CHKERRQ(ierr); 310 ierr = TSGetDM(ts, &da);CHKERRQ(ierr); 311 ierr = DMDAGetRay(da, ddir, ray, &rayctx->ray, &rayctx->scatter);CHKERRQ(ierr); 312 ierr = TSMonitorLGCtxCreate(PETSC_COMM_SELF,0,0,PETSC_DECIDE,PETSC_DECIDE,600,400,howoften,&rayctx->lgctx);CHKERRQ(ierr); 313 ierr = TSMonitorSet(ts, TSMonitorLGDMDARay, rayctx, TSMonitorDMDARayDestroy);CHKERRQ(ierr); 314 } 315 316 ierr = PetscOptionsName("-ts_monitor_envelope","Monitor maximum and minimum value of each component of the solution","TSMonitorEnvelope",&opt);CHKERRQ(ierr); 317 if (opt) { 318 TSMonitorEnvelopeCtx ctx; 319 320 ierr = TSMonitorEnvelopeCtxCreate(ts,&ctx);CHKERRQ(ierr); 321 ierr = TSMonitorSet(ts,TSMonitorEnvelope,ctx,(PetscErrorCode (*)(void**))TSMonitorEnvelopeCtxDestroy);CHKERRQ(ierr); 322 } 323 324 flg = PETSC_FALSE; 325 ierr = PetscOptionsBool("-ts_fd_color", "Use finite differences with coloring to compute IJacobian", "TSComputeJacobianDefaultColor", flg, &flg, NULL);CHKERRQ(ierr); 326 if (flg) { 327 DM dm; 328 DMTS tdm; 329 330 ierr = TSGetDM(ts, &dm);CHKERRQ(ierr); 331 ierr = DMGetDMTS(dm, &tdm);CHKERRQ(ierr); 332 tdm->ijacobianctx = NULL; 333 ierr = TSSetIJacobian(ts, NULL, NULL, TSComputeIJacobianDefaultColor, 0);CHKERRQ(ierr); 334 ierr = PetscInfo(ts, "Setting default finite difference coloring Jacobian matrix\n");CHKERRQ(ierr); 335 } 336 337 /* 338 This code is all wrong. One is creating objects inside the TSSetFromOptions() so if run with the options gui 339 will bleed memory. Also one is using a PetscOptionsBegin() inside a PetscOptionsBegin() 340 */ 341 ierr = TSGetAdapt(ts,&adapt);CHKERRQ(ierr); 342 ierr = TSAdaptSetFromOptions(PetscOptionsObject,adapt);CHKERRQ(ierr); 343 344 /* Handle specific TS options */ 345 if (ts->ops->setfromoptions) { 346 ierr = (*ts->ops->setfromoptions)(PetscOptionsObject,ts);CHKERRQ(ierr); 347 } 348 ierr = PetscOptionsEnd();CHKERRQ(ierr); 349 350 /* process any options handlers added with PetscObjectAddOptionsHandler() */ 351 ierr = PetscObjectProcessOptionsHandlers((PetscObject)ts);CHKERRQ(ierr); 352 353 if (ts->trajectory) { 354 ierr = TSTrajectorySetFromOptions(ts->trajectory);CHKERRQ(ierr); 355 } 356 357 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 358 if (snes) { 359 if (ts->problem_type == TS_LINEAR) {ierr = SNESSetType(snes,SNESKSPONLY);CHKERRQ(ierr);} 360 ierr = SNESSetFromOptions(ts->snes);CHKERRQ(ierr); 361 } 362 PetscFunctionReturn(0); 363 } 364 365 #undef __FUNCT__ 366 #define __FUNCT__ "TSSetSaveTrajectory" 367 /*@ 368 TSSetSaveTrajectory - Causes the TS to save its solutions as it iterates forward in time in a TSTrajectory object 369 370 Collective on TS 371 372 Input Parameters: 373 . ts - the TS context obtained from TSCreate() 374 375 376 Level: intermediate 377 378 .seealso: TSGetTrajectory(), TSAdjointSolve() 379 380 .keywords: TS, set, checkpoint, 381 @*/ 382 PetscErrorCode TSSetSaveTrajectory(TS ts) 383 { 384 PetscErrorCode ierr; 385 386 PetscFunctionBegin; 387 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 388 if (!ts->trajectory) { 389 ierr = TSTrajectoryCreate(PetscObjectComm((PetscObject)ts),&ts->trajectory);CHKERRQ(ierr); 390 ierr = TSTrajectorySetType(ts->trajectory,TSTRAJECTORYBASIC);CHKERRQ(ierr); 391 } 392 PetscFunctionReturn(0); 393 } 394 395 #undef __FUNCT__ 396 #define __FUNCT__ "TSComputeRHSJacobian" 397 /*@ 398 TSComputeRHSJacobian - Computes the Jacobian matrix that has been 399 set with TSSetRHSJacobian(). 400 401 Collective on TS and Vec 402 403 Input Parameters: 404 + ts - the TS context 405 . t - current timestep 406 - U - input vector 407 408 Output Parameters: 409 + A - Jacobian matrix 410 . B - optional preconditioning matrix 411 - flag - flag indicating matrix structure 412 413 Notes: 414 Most users should not need to explicitly call this routine, as it 415 is used internally within the nonlinear solvers. 416 417 See KSPSetOperators() for important information about setting the 418 flag parameter. 419 420 Level: developer 421 422 .keywords: SNES, compute, Jacobian, matrix 423 424 .seealso: TSSetRHSJacobian(), KSPSetOperators() 425 @*/ 426 PetscErrorCode TSComputeRHSJacobian(TS ts,PetscReal t,Vec U,Mat A,Mat B) 427 { 428 PetscErrorCode ierr; 429 PetscObjectState Ustate; 430 DM dm; 431 DMTS tsdm; 432 TSRHSJacobian rhsjacobianfunc; 433 void *ctx; 434 TSIJacobian ijacobianfunc; 435 TSRHSFunction rhsfunction; 436 437 PetscFunctionBegin; 438 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 439 PetscValidHeaderSpecific(U,VEC_CLASSID,3); 440 PetscCheckSameComm(ts,1,U,3); 441 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 442 ierr = DMGetDMTS(dm,&tsdm);CHKERRQ(ierr); 443 ierr = DMTSGetRHSJacobian(dm,&rhsjacobianfunc,&ctx);CHKERRQ(ierr); 444 ierr = DMTSGetIJacobian(dm,&ijacobianfunc,NULL);CHKERRQ(ierr); 445 ierr = DMTSGetRHSFunction(dm,&rhsfunction,&ctx);CHKERRQ(ierr); 446 ierr = PetscObjectStateGet((PetscObject)U,&Ustate);CHKERRQ(ierr); 447 if (ts->rhsjacobian.time == t && (ts->problem_type == TS_LINEAR || (ts->rhsjacobian.X == U && ts->rhsjacobian.Xstate == Ustate)) && (rhsfunction != TSComputeRHSFunctionLinear)) { 448 PetscFunctionReturn(0); 449 } 450 451 if (!rhsjacobianfunc && !ijacobianfunc) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"Must call TSSetRHSJacobian() and / or TSSetIJacobian()"); 452 453 if (ts->rhsjacobian.reuse) { 454 ierr = MatShift(A,-ts->rhsjacobian.shift);CHKERRQ(ierr); 455 ierr = MatScale(A,1./ts->rhsjacobian.scale);CHKERRQ(ierr); 456 if (A != B) { 457 ierr = MatShift(B,-ts->rhsjacobian.shift);CHKERRQ(ierr); 458 ierr = MatScale(B,1./ts->rhsjacobian.scale);CHKERRQ(ierr); 459 } 460 ts->rhsjacobian.shift = 0; 461 ts->rhsjacobian.scale = 1.; 462 } 463 464 if (rhsjacobianfunc) { 465 ierr = PetscLogEventBegin(TS_JacobianEval,ts,U,A,B);CHKERRQ(ierr); 466 PetscStackPush("TS user Jacobian function"); 467 ierr = (*rhsjacobianfunc)(ts,t,U,A,B,ctx);CHKERRQ(ierr); 468 PetscStackPop; 469 ierr = PetscLogEventEnd(TS_JacobianEval,ts,U,A,B);CHKERRQ(ierr); 470 /* make sure user returned a correct Jacobian and preconditioner */ 471 PetscValidHeaderSpecific(A,MAT_CLASSID,4); 472 PetscValidHeaderSpecific(B,MAT_CLASSID,5); 473 } else { 474 ierr = MatZeroEntries(A);CHKERRQ(ierr); 475 if (A != B) {ierr = MatZeroEntries(B);CHKERRQ(ierr);} 476 } 477 ts->rhsjacobian.time = t; 478 ts->rhsjacobian.X = U; 479 ierr = PetscObjectStateGet((PetscObject)U,&ts->rhsjacobian.Xstate);CHKERRQ(ierr); 480 PetscFunctionReturn(0); 481 } 482 483 #undef __FUNCT__ 484 #define __FUNCT__ "TSComputeRHSFunction" 485 /*@ 486 TSComputeRHSFunction - Evaluates the right-hand-side function. 487 488 Collective on TS and Vec 489 490 Input Parameters: 491 + ts - the TS context 492 . t - current time 493 - U - state vector 494 495 Output Parameter: 496 . y - right hand side 497 498 Note: 499 Most users should not need to explicitly call this routine, as it 500 is used internally within the nonlinear solvers. 501 502 Level: developer 503 504 .keywords: TS, compute 505 506 .seealso: TSSetRHSFunction(), TSComputeIFunction() 507 @*/ 508 PetscErrorCode TSComputeRHSFunction(TS ts,PetscReal t,Vec U,Vec y) 509 { 510 PetscErrorCode ierr; 511 TSRHSFunction rhsfunction; 512 TSIFunction ifunction; 513 void *ctx; 514 DM dm; 515 516 PetscFunctionBegin; 517 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 518 PetscValidHeaderSpecific(U,VEC_CLASSID,3); 519 PetscValidHeaderSpecific(y,VEC_CLASSID,4); 520 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 521 ierr = DMTSGetRHSFunction(dm,&rhsfunction,&ctx);CHKERRQ(ierr); 522 ierr = DMTSGetIFunction(dm,&ifunction,NULL);CHKERRQ(ierr); 523 524 if (!rhsfunction && !ifunction) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"Must call TSSetRHSFunction() and / or TSSetIFunction()"); 525 526 ierr = PetscLogEventBegin(TS_FunctionEval,ts,U,y,0);CHKERRQ(ierr); 527 if (rhsfunction) { 528 PetscStackPush("TS user right-hand-side function"); 529 ierr = (*rhsfunction)(ts,t,U,y,ctx);CHKERRQ(ierr); 530 PetscStackPop; 531 } else { 532 ierr = VecZeroEntries(y);CHKERRQ(ierr); 533 } 534 535 ierr = PetscLogEventEnd(TS_FunctionEval,ts,U,y,0);CHKERRQ(ierr); 536 PetscFunctionReturn(0); 537 } 538 539 #undef __FUNCT__ 540 #define __FUNCT__ "TSComputeSolutionFunction" 541 /*@ 542 TSComputeSolutionFunction - Evaluates the solution function. 543 544 Collective on TS and Vec 545 546 Input Parameters: 547 + ts - the TS context 548 - t - current time 549 550 Output Parameter: 551 . U - the solution 552 553 Note: 554 Most users should not need to explicitly call this routine, as it 555 is used internally within the nonlinear solvers. 556 557 Level: developer 558 559 .keywords: TS, compute 560 561 .seealso: TSSetSolutionFunction(), TSSetRHSFunction(), TSComputeIFunction() 562 @*/ 563 PetscErrorCode TSComputeSolutionFunction(TS ts,PetscReal t,Vec U) 564 { 565 PetscErrorCode ierr; 566 TSSolutionFunction solutionfunction; 567 void *ctx; 568 DM dm; 569 570 PetscFunctionBegin; 571 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 572 PetscValidHeaderSpecific(U,VEC_CLASSID,3); 573 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 574 ierr = DMTSGetSolutionFunction(dm,&solutionfunction,&ctx);CHKERRQ(ierr); 575 576 if (solutionfunction) { 577 PetscStackPush("TS user solution function"); 578 ierr = (*solutionfunction)(ts,t,U,ctx);CHKERRQ(ierr); 579 PetscStackPop; 580 } 581 PetscFunctionReturn(0); 582 } 583 #undef __FUNCT__ 584 #define __FUNCT__ "TSComputeForcingFunction" 585 /*@ 586 TSComputeForcingFunction - Evaluates the forcing function. 587 588 Collective on TS and Vec 589 590 Input Parameters: 591 + ts - the TS context 592 - t - current time 593 594 Output Parameter: 595 . U - the function value 596 597 Note: 598 Most users should not need to explicitly call this routine, as it 599 is used internally within the nonlinear solvers. 600 601 Level: developer 602 603 .keywords: TS, compute 604 605 .seealso: TSSetSolutionFunction(), TSSetRHSFunction(), TSComputeIFunction() 606 @*/ 607 PetscErrorCode TSComputeForcingFunction(TS ts,PetscReal t,Vec U) 608 { 609 PetscErrorCode ierr, (*forcing)(TS,PetscReal,Vec,void*); 610 void *ctx; 611 DM dm; 612 613 PetscFunctionBegin; 614 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 615 PetscValidHeaderSpecific(U,VEC_CLASSID,3); 616 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 617 ierr = DMTSGetForcingFunction(dm,&forcing,&ctx);CHKERRQ(ierr); 618 619 if (forcing) { 620 PetscStackPush("TS user forcing function"); 621 ierr = (*forcing)(ts,t,U,ctx);CHKERRQ(ierr); 622 PetscStackPop; 623 } 624 PetscFunctionReturn(0); 625 } 626 627 #undef __FUNCT__ 628 #define __FUNCT__ "TSGetRHSVec_Private" 629 static PetscErrorCode TSGetRHSVec_Private(TS ts,Vec *Frhs) 630 { 631 Vec F; 632 PetscErrorCode ierr; 633 634 PetscFunctionBegin; 635 *Frhs = NULL; 636 ierr = TSGetIFunction(ts,&F,NULL,NULL);CHKERRQ(ierr); 637 if (!ts->Frhs) { 638 ierr = VecDuplicate(F,&ts->Frhs);CHKERRQ(ierr); 639 } 640 *Frhs = ts->Frhs; 641 PetscFunctionReturn(0); 642 } 643 644 #undef __FUNCT__ 645 #define __FUNCT__ "TSGetRHSMats_Private" 646 static PetscErrorCode TSGetRHSMats_Private(TS ts,Mat *Arhs,Mat *Brhs) 647 { 648 Mat A,B; 649 PetscErrorCode ierr; 650 651 PetscFunctionBegin; 652 if (Arhs) *Arhs = NULL; 653 if (Brhs) *Brhs = NULL; 654 ierr = TSGetIJacobian(ts,&A,&B,NULL,NULL);CHKERRQ(ierr); 655 if (Arhs) { 656 if (!ts->Arhs) { 657 ierr = MatDuplicate(A,MAT_DO_NOT_COPY_VALUES,&ts->Arhs);CHKERRQ(ierr); 658 } 659 *Arhs = ts->Arhs; 660 } 661 if (Brhs) { 662 if (!ts->Brhs) { 663 if (A != B) { 664 ierr = MatDuplicate(B,MAT_DO_NOT_COPY_VALUES,&ts->Brhs);CHKERRQ(ierr); 665 } else { 666 ts->Brhs = ts->Arhs; 667 ierr = PetscObjectReference((PetscObject)ts->Arhs);CHKERRQ(ierr); 668 } 669 } 670 *Brhs = ts->Brhs; 671 } 672 PetscFunctionReturn(0); 673 } 674 675 #undef __FUNCT__ 676 #define __FUNCT__ "TSComputeIFunction" 677 /*@ 678 TSComputeIFunction - Evaluates the DAE residual written in implicit form F(t,U,Udot)=0 679 680 Collective on TS and Vec 681 682 Input Parameters: 683 + ts - the TS context 684 . t - current time 685 . U - state vector 686 . Udot - time derivative of state vector 687 - imex - flag indicates if the method is IMEX so that the RHSFunction should be kept separate 688 689 Output Parameter: 690 . Y - right hand side 691 692 Note: 693 Most users should not need to explicitly call this routine, as it 694 is used internally within the nonlinear solvers. 695 696 If the user did did not write their equations in implicit form, this 697 function recasts them in implicit form. 698 699 Level: developer 700 701 .keywords: TS, compute 702 703 .seealso: TSSetIFunction(), TSComputeRHSFunction() 704 @*/ 705 PetscErrorCode TSComputeIFunction(TS ts,PetscReal t,Vec U,Vec Udot,Vec Y,PetscBool imex) 706 { 707 PetscErrorCode ierr; 708 TSIFunction ifunction; 709 TSRHSFunction rhsfunction; 710 void *ctx; 711 DM dm; 712 713 PetscFunctionBegin; 714 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 715 PetscValidHeaderSpecific(U,VEC_CLASSID,3); 716 PetscValidHeaderSpecific(Udot,VEC_CLASSID,4); 717 PetscValidHeaderSpecific(Y,VEC_CLASSID,5); 718 719 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 720 ierr = DMTSGetIFunction(dm,&ifunction,&ctx);CHKERRQ(ierr); 721 ierr = DMTSGetRHSFunction(dm,&rhsfunction,NULL);CHKERRQ(ierr); 722 723 if (!rhsfunction && !ifunction) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"Must call TSSetRHSFunction() and / or TSSetIFunction()"); 724 725 ierr = PetscLogEventBegin(TS_FunctionEval,ts,U,Udot,Y);CHKERRQ(ierr); 726 if (ifunction) { 727 PetscStackPush("TS user implicit function"); 728 ierr = (*ifunction)(ts,t,U,Udot,Y,ctx);CHKERRQ(ierr); 729 PetscStackPop; 730 } 731 if (imex) { 732 if (!ifunction) { 733 ierr = VecCopy(Udot,Y);CHKERRQ(ierr); 734 } 735 } else if (rhsfunction) { 736 if (ifunction) { 737 Vec Frhs; 738 ierr = TSGetRHSVec_Private(ts,&Frhs);CHKERRQ(ierr); 739 ierr = TSComputeRHSFunction(ts,t,U,Frhs);CHKERRQ(ierr); 740 ierr = VecAXPY(Y,-1,Frhs);CHKERRQ(ierr); 741 } else { 742 ierr = TSComputeRHSFunction(ts,t,U,Y);CHKERRQ(ierr); 743 ierr = VecAYPX(Y,-1,Udot);CHKERRQ(ierr); 744 } 745 } 746 ierr = PetscLogEventEnd(TS_FunctionEval,ts,U,Udot,Y);CHKERRQ(ierr); 747 PetscFunctionReturn(0); 748 } 749 750 #undef __FUNCT__ 751 #define __FUNCT__ "TSComputeIJacobian" 752 /*@ 753 TSComputeIJacobian - Evaluates the Jacobian of the DAE 754 755 Collective on TS and Vec 756 757 Input 758 Input Parameters: 759 + ts - the TS context 760 . t - current timestep 761 . U - state vector 762 . Udot - time derivative of state vector 763 . shift - shift to apply, see note below 764 - imex - flag indicates if the method is IMEX so that the RHSJacobian should be kept separate 765 766 Output Parameters: 767 + A - Jacobian matrix 768 . B - optional preconditioning matrix 769 - flag - flag indicating matrix structure 770 771 Notes: 772 If F(t,U,Udot)=0 is the DAE, the required Jacobian is 773 774 dF/dU + shift*dF/dUdot 775 776 Most users should not need to explicitly call this routine, as it 777 is used internally within the nonlinear solvers. 778 779 Level: developer 780 781 .keywords: TS, compute, Jacobian, matrix 782 783 .seealso: TSSetIJacobian() 784 @*/ 785 PetscErrorCode TSComputeIJacobian(TS ts,PetscReal t,Vec U,Vec Udot,PetscReal shift,Mat A,Mat B,PetscBool imex) 786 { 787 PetscErrorCode ierr; 788 TSIJacobian ijacobian; 789 TSRHSJacobian rhsjacobian; 790 DM dm; 791 void *ctx; 792 793 PetscFunctionBegin; 794 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 795 PetscValidHeaderSpecific(U,VEC_CLASSID,3); 796 PetscValidHeaderSpecific(Udot,VEC_CLASSID,4); 797 PetscValidPointer(A,6); 798 PetscValidHeaderSpecific(A,MAT_CLASSID,6); 799 PetscValidPointer(B,7); 800 PetscValidHeaderSpecific(B,MAT_CLASSID,7); 801 802 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 803 ierr = DMTSGetIJacobian(dm,&ijacobian,&ctx);CHKERRQ(ierr); 804 ierr = DMTSGetRHSJacobian(dm,&rhsjacobian,NULL);CHKERRQ(ierr); 805 806 if (!rhsjacobian && !ijacobian) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"Must call TSSetRHSJacobian() and / or TSSetIJacobian()"); 807 808 ierr = PetscLogEventBegin(TS_JacobianEval,ts,U,A,B);CHKERRQ(ierr); 809 if (ijacobian) { 810 PetscStackPush("TS user implicit Jacobian"); 811 ierr = (*ijacobian)(ts,t,U,Udot,shift,A,B,ctx);CHKERRQ(ierr); 812 PetscStackPop; 813 /* make sure user returned a correct Jacobian and preconditioner */ 814 PetscValidHeaderSpecific(A,MAT_CLASSID,4); 815 PetscValidHeaderSpecific(B,MAT_CLASSID,5); 816 } 817 if (imex) { 818 if (!ijacobian) { /* system was written as Udot = G(t,U) */ 819 ierr = MatZeroEntries(A);CHKERRQ(ierr); 820 ierr = MatShift(A,shift);CHKERRQ(ierr); 821 if (A != B) { 822 ierr = MatZeroEntries(B);CHKERRQ(ierr); 823 ierr = MatShift(B,shift);CHKERRQ(ierr); 824 } 825 } 826 } else { 827 Mat Arhs = NULL,Brhs = NULL; 828 if (rhsjacobian) { 829 if (ijacobian) { 830 ierr = TSGetRHSMats_Private(ts,&Arhs,&Brhs);CHKERRQ(ierr); 831 } else { 832 ierr = TSGetIJacobian(ts,&Arhs,&Brhs,NULL,NULL);CHKERRQ(ierr); 833 } 834 ierr = TSComputeRHSJacobian(ts,t,U,Arhs,Brhs);CHKERRQ(ierr); 835 } 836 if (Arhs == A) { /* No IJacobian, so we only have the RHS matrix */ 837 ts->rhsjacobian.scale = -1; 838 ts->rhsjacobian.shift = shift; 839 ierr = MatScale(A,-1);CHKERRQ(ierr); 840 ierr = MatShift(A,shift);CHKERRQ(ierr); 841 if (A != B) { 842 ierr = MatScale(B,-1);CHKERRQ(ierr); 843 ierr = MatShift(B,shift);CHKERRQ(ierr); 844 } 845 } else if (Arhs) { /* Both IJacobian and RHSJacobian */ 846 MatStructure axpy = DIFFERENT_NONZERO_PATTERN; 847 if (!ijacobian) { /* No IJacobian provided, but we have a separate RHS matrix */ 848 ierr = MatZeroEntries(A);CHKERRQ(ierr); 849 ierr = MatShift(A,shift);CHKERRQ(ierr); 850 if (A != B) { 851 ierr = MatZeroEntries(B);CHKERRQ(ierr); 852 ierr = MatShift(B,shift);CHKERRQ(ierr); 853 } 854 } 855 ierr = MatAXPY(A,-1,Arhs,axpy);CHKERRQ(ierr); 856 if (A != B) { 857 ierr = MatAXPY(B,-1,Brhs,axpy);CHKERRQ(ierr); 858 } 859 } 860 } 861 ierr = PetscLogEventEnd(TS_JacobianEval,ts,U,A,B);CHKERRQ(ierr); 862 PetscFunctionReturn(0); 863 } 864 865 #undef __FUNCT__ 866 #define __FUNCT__ "TSSetRHSFunction" 867 /*@C 868 TSSetRHSFunction - Sets the routine for evaluating the function, 869 where U_t = G(t,u). 870 871 Logically Collective on TS 872 873 Input Parameters: 874 + ts - the TS context obtained from TSCreate() 875 . r - vector to put the computed right hand side (or NULL to have it created) 876 . f - routine for evaluating the right-hand-side function 877 - ctx - [optional] user-defined context for private data for the 878 function evaluation routine (may be NULL) 879 880 Calling sequence of func: 881 $ func (TS ts,PetscReal t,Vec u,Vec F,void *ctx); 882 883 + t - current timestep 884 . u - input vector 885 . F - function vector 886 - ctx - [optional] user-defined function context 887 888 Level: beginner 889 890 Notes: You must call this function or TSSetIFunction() to define your ODE. You cannot use this function when solving a DAE. 891 892 .keywords: TS, timestep, set, right-hand-side, function 893 894 .seealso: TSSetRHSJacobian(), TSSetIJacobian(), TSSetIFunction() 895 @*/ 896 PetscErrorCode TSSetRHSFunction(TS ts,Vec r,PetscErrorCode (*f)(TS,PetscReal,Vec,Vec,void*),void *ctx) 897 { 898 PetscErrorCode ierr; 899 SNES snes; 900 Vec ralloc = NULL; 901 DM dm; 902 903 PetscFunctionBegin; 904 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 905 if (r) PetscValidHeaderSpecific(r,VEC_CLASSID,2); 906 907 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 908 ierr = DMTSSetRHSFunction(dm,f,ctx);CHKERRQ(ierr); 909 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 910 if (!r && !ts->dm && ts->vec_sol) { 911 ierr = VecDuplicate(ts->vec_sol,&ralloc);CHKERRQ(ierr); 912 r = ralloc; 913 } 914 ierr = SNESSetFunction(snes,r,SNESTSFormFunction,ts);CHKERRQ(ierr); 915 ierr = VecDestroy(&ralloc);CHKERRQ(ierr); 916 PetscFunctionReturn(0); 917 } 918 919 #undef __FUNCT__ 920 #define __FUNCT__ "TSSetSolutionFunction" 921 /*@C 922 TSSetSolutionFunction - Provide a function that computes the solution of the ODE or DAE 923 924 Logically Collective on TS 925 926 Input Parameters: 927 + ts - the TS context obtained from TSCreate() 928 . f - routine for evaluating the solution 929 - ctx - [optional] user-defined context for private data for the 930 function evaluation routine (may be NULL) 931 932 Calling sequence of func: 933 $ func (TS ts,PetscReal t,Vec u,void *ctx); 934 935 + t - current timestep 936 . u - output vector 937 - ctx - [optional] user-defined function context 938 939 Notes: 940 This routine is used for testing accuracy of time integration schemes when you already know the solution. 941 If analytic solutions are not known for your system, consider using the Method of Manufactured Solutions to 942 create closed-form solutions with non-physical forcing terms. 943 944 For low-dimensional problems solved in serial, such as small discrete systems, TSMonitorLGError() can be used to monitor the error history. 945 946 Level: beginner 947 948 .keywords: TS, timestep, set, right-hand-side, function 949 950 .seealso: TSSetRHSJacobian(), TSSetIJacobian(), TSComputeSolutionFunction(), TSSetForcingFunction() 951 @*/ 952 PetscErrorCode TSSetSolutionFunction(TS ts,PetscErrorCode (*f)(TS,PetscReal,Vec,void*),void *ctx) 953 { 954 PetscErrorCode ierr; 955 DM dm; 956 957 PetscFunctionBegin; 958 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 959 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 960 ierr = DMTSSetSolutionFunction(dm,f,ctx);CHKERRQ(ierr); 961 PetscFunctionReturn(0); 962 } 963 964 #undef __FUNCT__ 965 #define __FUNCT__ "TSSetForcingFunction" 966 /*@C 967 TSSetForcingFunction - Provide a function that computes a forcing term for a ODE or PDE 968 969 Logically Collective on TS 970 971 Input Parameters: 972 + ts - the TS context obtained from TSCreate() 973 . f - routine for evaluating the forcing function 974 - ctx - [optional] user-defined context for private data for the 975 function evaluation routine (may be NULL) 976 977 Calling sequence of func: 978 $ func (TS ts,PetscReal t,Vec u,void *ctx); 979 980 + t - current timestep 981 . u - output vector 982 - ctx - [optional] user-defined function context 983 984 Notes: 985 This routine is useful for testing accuracy of time integration schemes when using the Method of Manufactured Solutions to 986 create closed-form solutions with a non-physical forcing term. 987 988 For low-dimensional problems solved in serial, such as small discrete systems, TSMonitorLGError() can be used to monitor the error history. 989 990 Level: beginner 991 992 .keywords: TS, timestep, set, right-hand-side, function 993 994 .seealso: TSSetRHSJacobian(), TSSetIJacobian(), TSComputeSolutionFunction(), TSSetSolutionFunction() 995 @*/ 996 PetscErrorCode TSSetForcingFunction(TS ts,PetscErrorCode (*f)(TS,PetscReal,Vec,void*),void *ctx) 997 { 998 PetscErrorCode ierr; 999 DM dm; 1000 1001 PetscFunctionBegin; 1002 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1003 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 1004 ierr = DMTSSetForcingFunction(dm,f,ctx);CHKERRQ(ierr); 1005 PetscFunctionReturn(0); 1006 } 1007 1008 #undef __FUNCT__ 1009 #define __FUNCT__ "TSSetRHSJacobian" 1010 /*@C 1011 TSSetRHSJacobian - Sets the function to compute the Jacobian of G, 1012 where U_t = G(U,t), as well as the location to store the matrix. 1013 1014 Logically Collective on TS 1015 1016 Input Parameters: 1017 + ts - the TS context obtained from TSCreate() 1018 . Amat - (approximate) Jacobian matrix 1019 . Pmat - matrix from which preconditioner is to be constructed (usually the same as Amat) 1020 . f - the Jacobian evaluation routine 1021 - ctx - [optional] user-defined context for private data for the 1022 Jacobian evaluation routine (may be NULL) 1023 1024 Calling sequence of f: 1025 $ func (TS ts,PetscReal t,Vec u,Mat A,Mat B,void *ctx); 1026 1027 + t - current timestep 1028 . u - input vector 1029 . Amat - (approximate) Jacobian matrix 1030 . Pmat - matrix from which preconditioner is to be constructed (usually the same as Amat) 1031 - ctx - [optional] user-defined context for matrix evaluation routine 1032 1033 1034 Level: beginner 1035 1036 .keywords: TS, timestep, set, right-hand-side, Jacobian 1037 1038 .seealso: SNESComputeJacobianDefaultColor(), TSSetRHSFunction(), TSRHSJacobianSetReuse(), TSSetIJacobian() 1039 1040 @*/ 1041 PetscErrorCode TSSetRHSJacobian(TS ts,Mat Amat,Mat Pmat,TSRHSJacobian f,void *ctx) 1042 { 1043 PetscErrorCode ierr; 1044 SNES snes; 1045 DM dm; 1046 TSIJacobian ijacobian; 1047 1048 PetscFunctionBegin; 1049 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1050 if (Amat) PetscValidHeaderSpecific(Amat,MAT_CLASSID,2); 1051 if (Pmat) PetscValidHeaderSpecific(Pmat,MAT_CLASSID,3); 1052 if (Amat) PetscCheckSameComm(ts,1,Amat,2); 1053 if (Pmat) PetscCheckSameComm(ts,1,Pmat,3); 1054 1055 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 1056 ierr = DMTSSetRHSJacobian(dm,f,ctx);CHKERRQ(ierr); 1057 if (f == TSComputeRHSJacobianConstant) { 1058 /* Handle this case automatically for the user; otherwise user should call themselves. */ 1059 ierr = TSRHSJacobianSetReuse(ts,PETSC_TRUE);CHKERRQ(ierr); 1060 } 1061 ierr = DMTSGetIJacobian(dm,&ijacobian,NULL);CHKERRQ(ierr); 1062 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 1063 if (!ijacobian) { 1064 ierr = SNESSetJacobian(snes,Amat,Pmat,SNESTSFormJacobian,ts);CHKERRQ(ierr); 1065 } 1066 if (Amat) { 1067 ierr = PetscObjectReference((PetscObject)Amat);CHKERRQ(ierr); 1068 ierr = MatDestroy(&ts->Arhs);CHKERRQ(ierr); 1069 1070 ts->Arhs = Amat; 1071 } 1072 if (Pmat) { 1073 ierr = PetscObjectReference((PetscObject)Pmat);CHKERRQ(ierr); 1074 ierr = MatDestroy(&ts->Brhs);CHKERRQ(ierr); 1075 1076 ts->Brhs = Pmat; 1077 } 1078 PetscFunctionReturn(0); 1079 } 1080 1081 1082 #undef __FUNCT__ 1083 #define __FUNCT__ "TSSetIFunction" 1084 /*@C 1085 TSSetIFunction - Set the function to compute F(t,U,U_t) where F() = 0 is the DAE to be solved. 1086 1087 Logically Collective on TS 1088 1089 Input Parameters: 1090 + ts - the TS context obtained from TSCreate() 1091 . r - vector to hold the residual (or NULL to have it created internally) 1092 . f - the function evaluation routine 1093 - ctx - user-defined context for private data for the function evaluation routine (may be NULL) 1094 1095 Calling sequence of f: 1096 $ f(TS ts,PetscReal t,Vec u,Vec u_t,Vec F,ctx); 1097 1098 + t - time at step/stage being solved 1099 . u - state vector 1100 . u_t - time derivative of state vector 1101 . F - function vector 1102 - ctx - [optional] user-defined context for matrix evaluation routine 1103 1104 Important: 1105 The user MUST call either this routine or TSSetRHSFunction() to define the ODE. When solving DAEs you must use this function. 1106 1107 Level: beginner 1108 1109 .keywords: TS, timestep, set, DAE, Jacobian 1110 1111 .seealso: TSSetRHSJacobian(), TSSetRHSFunction(), TSSetIJacobian() 1112 @*/ 1113 PetscErrorCode TSSetIFunction(TS ts,Vec res,TSIFunction f,void *ctx) 1114 { 1115 PetscErrorCode ierr; 1116 SNES snes; 1117 Vec resalloc = NULL; 1118 DM dm; 1119 1120 PetscFunctionBegin; 1121 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1122 if (res) PetscValidHeaderSpecific(res,VEC_CLASSID,2); 1123 1124 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 1125 ierr = DMTSSetIFunction(dm,f,ctx);CHKERRQ(ierr); 1126 1127 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 1128 if (!res && !ts->dm && ts->vec_sol) { 1129 ierr = VecDuplicate(ts->vec_sol,&resalloc);CHKERRQ(ierr); 1130 res = resalloc; 1131 } 1132 ierr = SNESSetFunction(snes,res,SNESTSFormFunction,ts);CHKERRQ(ierr); 1133 ierr = VecDestroy(&resalloc);CHKERRQ(ierr); 1134 PetscFunctionReturn(0); 1135 } 1136 1137 #undef __FUNCT__ 1138 #define __FUNCT__ "TSGetIFunction" 1139 /*@C 1140 TSGetIFunction - Returns the vector where the implicit residual is stored and the function/contex to compute it. 1141 1142 Not Collective 1143 1144 Input Parameter: 1145 . ts - the TS context 1146 1147 Output Parameter: 1148 + r - vector to hold residual (or NULL) 1149 . func - the function to compute residual (or NULL) 1150 - ctx - the function context (or NULL) 1151 1152 Level: advanced 1153 1154 .keywords: TS, nonlinear, get, function 1155 1156 .seealso: TSSetIFunction(), SNESGetFunction() 1157 @*/ 1158 PetscErrorCode TSGetIFunction(TS ts,Vec *r,TSIFunction *func,void **ctx) 1159 { 1160 PetscErrorCode ierr; 1161 SNES snes; 1162 DM dm; 1163 1164 PetscFunctionBegin; 1165 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1166 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 1167 ierr = SNESGetFunction(snes,r,NULL,NULL);CHKERRQ(ierr); 1168 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 1169 ierr = DMTSGetIFunction(dm,func,ctx);CHKERRQ(ierr); 1170 PetscFunctionReturn(0); 1171 } 1172 1173 #undef __FUNCT__ 1174 #define __FUNCT__ "TSGetRHSFunction" 1175 /*@C 1176 TSGetRHSFunction - Returns the vector where the right hand side is stored and the function/context to compute it. 1177 1178 Not Collective 1179 1180 Input Parameter: 1181 . ts - the TS context 1182 1183 Output Parameter: 1184 + r - vector to hold computed right hand side (or NULL) 1185 . func - the function to compute right hand side (or NULL) 1186 - ctx - the function context (or NULL) 1187 1188 Level: advanced 1189 1190 .keywords: TS, nonlinear, get, function 1191 1192 .seealso: TSSetRHSFunction(), SNESGetFunction() 1193 @*/ 1194 PetscErrorCode TSGetRHSFunction(TS ts,Vec *r,TSRHSFunction *func,void **ctx) 1195 { 1196 PetscErrorCode ierr; 1197 SNES snes; 1198 DM dm; 1199 1200 PetscFunctionBegin; 1201 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1202 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 1203 ierr = SNESGetFunction(snes,r,NULL,NULL);CHKERRQ(ierr); 1204 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 1205 ierr = DMTSGetRHSFunction(dm,func,ctx);CHKERRQ(ierr); 1206 PetscFunctionReturn(0); 1207 } 1208 1209 #undef __FUNCT__ 1210 #define __FUNCT__ "TSSetIJacobian" 1211 /*@C 1212 TSSetIJacobian - Set the function to compute the matrix dF/dU + a*dF/dU_t where F(t,U,U_t) is the function 1213 provided with TSSetIFunction(). 1214 1215 Logically Collective on TS 1216 1217 Input Parameters: 1218 + ts - the TS context obtained from TSCreate() 1219 . Amat - (approximate) Jacobian matrix 1220 . Pmat - matrix used to compute preconditioner (usually the same as Amat) 1221 . f - the Jacobian evaluation routine 1222 - ctx - user-defined context for private data for the Jacobian evaluation routine (may be NULL) 1223 1224 Calling sequence of f: 1225 $ f(TS ts,PetscReal t,Vec U,Vec U_t,PetscReal a,Mat Amat,Mat Pmat,void *ctx); 1226 1227 + t - time at step/stage being solved 1228 . U - state vector 1229 . U_t - time derivative of state vector 1230 . a - shift 1231 . Amat - (approximate) Jacobian of F(t,U,W+a*U), equivalent to dF/dU + a*dF/dU_t 1232 . Pmat - matrix used for constructing preconditioner, usually the same as Amat 1233 - ctx - [optional] user-defined context for matrix evaluation routine 1234 1235 Notes: 1236 The matrices Amat and Pmat are exactly the matrices that are used by SNES for the nonlinear solve. 1237 1238 The matrix dF/dU + a*dF/dU_t you provide turns out to be 1239 the Jacobian of F(t,U,W+a*U) where F(t,U,U_t) = 0 is the DAE to be solved. 1240 The time integrator internally approximates U_t by W+a*U where the positive "shift" 1241 a and vector W depend on the integration method, step size, and past states. For example with 1242 the backward Euler method a = 1/dt and W = -a*U(previous timestep) so 1243 W + a*U = a*(U - U(previous timestep)) = (U - U(previous timestep))/dt 1244 1245 Level: beginner 1246 1247 .keywords: TS, timestep, DAE, Jacobian 1248 1249 .seealso: TSSetIFunction(), TSSetRHSJacobian(), SNESComputeJacobianDefaultColor(), SNESComputeJacobianDefault(), TSSetRHSFunction() 1250 1251 @*/ 1252 PetscErrorCode TSSetIJacobian(TS ts,Mat Amat,Mat Pmat,TSIJacobian f,void *ctx) 1253 { 1254 PetscErrorCode ierr; 1255 SNES snes; 1256 DM dm; 1257 1258 PetscFunctionBegin; 1259 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1260 if (Amat) PetscValidHeaderSpecific(Amat,MAT_CLASSID,2); 1261 if (Pmat) PetscValidHeaderSpecific(Pmat,MAT_CLASSID,3); 1262 if (Amat) PetscCheckSameComm(ts,1,Amat,2); 1263 if (Pmat) PetscCheckSameComm(ts,1,Pmat,3); 1264 1265 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 1266 ierr = DMTSSetIJacobian(dm,f,ctx);CHKERRQ(ierr); 1267 1268 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 1269 ierr = SNESSetJacobian(snes,Amat,Pmat,SNESTSFormJacobian,ts);CHKERRQ(ierr); 1270 PetscFunctionReturn(0); 1271 } 1272 1273 #undef __FUNCT__ 1274 #define __FUNCT__ "TSRHSJacobianSetReuse" 1275 /*@ 1276 TSRHSJacobianSetReuse - restore RHS Jacobian before re-evaluating. Without this flag, TS will change the sign and 1277 shift the RHS Jacobian for a finite-time-step implicit solve, in which case the user function will need to recompute 1278 the entire Jacobian. The reuse flag must be set if the evaluation function will assume that the matrix entries have 1279 not been changed by the TS. 1280 1281 Logically Collective 1282 1283 Input Arguments: 1284 + ts - TS context obtained from TSCreate() 1285 - reuse - PETSC_TRUE if the RHS Jacobian 1286 1287 Level: intermediate 1288 1289 .seealso: TSSetRHSJacobian(), TSComputeRHSJacobianConstant() 1290 @*/ 1291 PetscErrorCode TSRHSJacobianSetReuse(TS ts,PetscBool reuse) 1292 { 1293 PetscFunctionBegin; 1294 ts->rhsjacobian.reuse = reuse; 1295 PetscFunctionReturn(0); 1296 } 1297 1298 #undef __FUNCT__ 1299 #define __FUNCT__ "TSLoad" 1300 /*@C 1301 TSLoad - Loads a KSP that has been stored in binary with KSPView(). 1302 1303 Collective on PetscViewer 1304 1305 Input Parameters: 1306 + newdm - the newly loaded TS, this needs to have been created with TSCreate() or 1307 some related function before a call to TSLoad(). 1308 - viewer - binary file viewer, obtained from PetscViewerBinaryOpen() 1309 1310 Level: intermediate 1311 1312 Notes: 1313 The type is determined by the data in the file, any type set into the TS before this call is ignored. 1314 1315 Notes for advanced users: 1316 Most users should not need to know the details of the binary storage 1317 format, since TSLoad() and TSView() completely hide these details. 1318 But for anyone who's interested, the standard binary matrix storage 1319 format is 1320 .vb 1321 has not yet been determined 1322 .ve 1323 1324 .seealso: PetscViewerBinaryOpen(), TSView(), MatLoad(), VecLoad() 1325 @*/ 1326 PetscErrorCode TSLoad(TS ts, PetscViewer viewer) 1327 { 1328 PetscErrorCode ierr; 1329 PetscBool isbinary; 1330 PetscInt classid; 1331 char type[256]; 1332 DMTS sdm; 1333 DM dm; 1334 1335 PetscFunctionBegin; 1336 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1337 PetscValidHeaderSpecific(viewer,PETSC_VIEWER_CLASSID,2); 1338 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERBINARY,&isbinary);CHKERRQ(ierr); 1339 if (!isbinary) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Invalid viewer; open viewer with PetscViewerBinaryOpen()"); 1340 1341 ierr = PetscViewerBinaryRead(viewer,&classid,1,NULL,PETSC_INT);CHKERRQ(ierr); 1342 if (classid != TS_FILE_CLASSID) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_ARG_WRONG,"Not TS next in file"); 1343 ierr = PetscViewerBinaryRead(viewer,type,256,NULL,PETSC_CHAR);CHKERRQ(ierr); 1344 ierr = TSSetType(ts, type);CHKERRQ(ierr); 1345 if (ts->ops->load) { 1346 ierr = (*ts->ops->load)(ts,viewer);CHKERRQ(ierr); 1347 } 1348 ierr = DMCreate(PetscObjectComm((PetscObject)ts),&dm);CHKERRQ(ierr); 1349 ierr = DMLoad(dm,viewer);CHKERRQ(ierr); 1350 ierr = TSSetDM(ts,dm);CHKERRQ(ierr); 1351 ierr = DMCreateGlobalVector(ts->dm,&ts->vec_sol);CHKERRQ(ierr); 1352 ierr = VecLoad(ts->vec_sol,viewer);CHKERRQ(ierr); 1353 ierr = DMGetDMTS(ts->dm,&sdm);CHKERRQ(ierr); 1354 ierr = DMTSLoad(sdm,viewer);CHKERRQ(ierr); 1355 PetscFunctionReturn(0); 1356 } 1357 1358 #include <petscdraw.h> 1359 #if defined(PETSC_HAVE_SAWS) 1360 #include <petscviewersaws.h> 1361 #endif 1362 #undef __FUNCT__ 1363 #define __FUNCT__ "TSView" 1364 /*@C 1365 TSView - Prints the TS data structure. 1366 1367 Collective on TS 1368 1369 Input Parameters: 1370 + ts - the TS context obtained from TSCreate() 1371 - viewer - visualization context 1372 1373 Options Database Key: 1374 . -ts_view - calls TSView() at end of TSStep() 1375 1376 Notes: 1377 The available visualization contexts include 1378 + PETSC_VIEWER_STDOUT_SELF - standard output (default) 1379 - PETSC_VIEWER_STDOUT_WORLD - synchronized standard 1380 output where only the first processor opens 1381 the file. All other processors send their 1382 data to the first processor to print. 1383 1384 The user can open an alternative visualization context with 1385 PetscViewerASCIIOpen() - output to a specified file. 1386 1387 Level: beginner 1388 1389 .keywords: TS, timestep, view 1390 1391 .seealso: PetscViewerASCIIOpen() 1392 @*/ 1393 PetscErrorCode TSView(TS ts,PetscViewer viewer) 1394 { 1395 PetscErrorCode ierr; 1396 TSType type; 1397 PetscBool iascii,isstring,isundials,isbinary,isdraw; 1398 DMTS sdm; 1399 #if defined(PETSC_HAVE_SAWS) 1400 PetscBool issaws; 1401 #endif 1402 1403 PetscFunctionBegin; 1404 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1405 if (!viewer) { 1406 ierr = PetscViewerASCIIGetStdout(PetscObjectComm((PetscObject)ts),&viewer);CHKERRQ(ierr); 1407 } 1408 PetscValidHeaderSpecific(viewer,PETSC_VIEWER_CLASSID,2); 1409 PetscCheckSameComm(ts,1,viewer,2); 1410 1411 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);CHKERRQ(ierr); 1412 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERSTRING,&isstring);CHKERRQ(ierr); 1413 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERBINARY,&isbinary);CHKERRQ(ierr); 1414 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERDRAW,&isdraw);CHKERRQ(ierr); 1415 #if defined(PETSC_HAVE_SAWS) 1416 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERSAWS,&issaws);CHKERRQ(ierr); 1417 #endif 1418 if (iascii) { 1419 ierr = PetscObjectPrintClassNamePrefixType((PetscObject)ts,viewer);CHKERRQ(ierr); 1420 ierr = PetscViewerASCIIPrintf(viewer," maximum steps=%D\n",ts->max_steps);CHKERRQ(ierr); 1421 ierr = PetscViewerASCIIPrintf(viewer," maximum time=%g\n",(double)ts->max_time);CHKERRQ(ierr); 1422 if (ts->problem_type == TS_NONLINEAR) { 1423 ierr = PetscViewerASCIIPrintf(viewer," total number of nonlinear solver iterations=%D\n",ts->snes_its);CHKERRQ(ierr); 1424 ierr = PetscViewerASCIIPrintf(viewer," total number of nonlinear solve failures=%D\n",ts->num_snes_failures);CHKERRQ(ierr); 1425 } 1426 ierr = PetscViewerASCIIPrintf(viewer," total number of linear solver iterations=%D\n",ts->ksp_its);CHKERRQ(ierr); 1427 ierr = PetscViewerASCIIPrintf(viewer," total number of rejected steps=%D\n",ts->reject);CHKERRQ(ierr); 1428 ierr = DMGetDMTS(ts->dm,&sdm);CHKERRQ(ierr); 1429 ierr = DMTSView(sdm,viewer);CHKERRQ(ierr); 1430 if (ts->ops->view) { 1431 ierr = PetscViewerASCIIPushTab(viewer);CHKERRQ(ierr); 1432 ierr = (*ts->ops->view)(ts,viewer);CHKERRQ(ierr); 1433 ierr = PetscViewerASCIIPopTab(viewer);CHKERRQ(ierr); 1434 } 1435 } else if (isstring) { 1436 ierr = TSGetType(ts,&type);CHKERRQ(ierr); 1437 ierr = PetscViewerStringSPrintf(viewer," %-7.7s",type);CHKERRQ(ierr); 1438 } else if (isbinary) { 1439 PetscInt classid = TS_FILE_CLASSID; 1440 MPI_Comm comm; 1441 PetscMPIInt rank; 1442 char type[256]; 1443 1444 ierr = PetscObjectGetComm((PetscObject)ts,&comm);CHKERRQ(ierr); 1445 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 1446 if (!rank) { 1447 ierr = PetscViewerBinaryWrite(viewer,&classid,1,PETSC_INT,PETSC_FALSE);CHKERRQ(ierr); 1448 ierr = PetscStrncpy(type,((PetscObject)ts)->type_name,256);CHKERRQ(ierr); 1449 ierr = PetscViewerBinaryWrite(viewer,type,256,PETSC_CHAR,PETSC_FALSE);CHKERRQ(ierr); 1450 } 1451 if (ts->ops->view) { 1452 ierr = (*ts->ops->view)(ts,viewer);CHKERRQ(ierr); 1453 } 1454 ierr = DMView(ts->dm,viewer);CHKERRQ(ierr); 1455 ierr = VecView(ts->vec_sol,viewer);CHKERRQ(ierr); 1456 ierr = DMGetDMTS(ts->dm,&sdm);CHKERRQ(ierr); 1457 ierr = DMTSView(sdm,viewer);CHKERRQ(ierr); 1458 } else if (isdraw) { 1459 PetscDraw draw; 1460 char str[36]; 1461 PetscReal x,y,bottom,h; 1462 1463 ierr = PetscViewerDrawGetDraw(viewer,0,&draw);CHKERRQ(ierr); 1464 ierr = PetscDrawGetCurrentPoint(draw,&x,&y);CHKERRQ(ierr); 1465 ierr = PetscStrcpy(str,"TS: ");CHKERRQ(ierr); 1466 ierr = PetscStrcat(str,((PetscObject)ts)->type_name);CHKERRQ(ierr); 1467 ierr = PetscDrawStringBoxed(draw,x,y,PETSC_DRAW_BLACK,PETSC_DRAW_BLACK,str,NULL,&h);CHKERRQ(ierr); 1468 bottom = y - h; 1469 ierr = PetscDrawPushCurrentPoint(draw,x,bottom);CHKERRQ(ierr); 1470 if (ts->ops->view) { 1471 ierr = (*ts->ops->view)(ts,viewer);CHKERRQ(ierr); 1472 } 1473 ierr = PetscDrawPopCurrentPoint(draw);CHKERRQ(ierr); 1474 #if defined(PETSC_HAVE_SAWS) 1475 } else if (issaws) { 1476 PetscMPIInt rank; 1477 const char *name; 1478 1479 ierr = PetscObjectGetName((PetscObject)ts,&name);CHKERRQ(ierr); 1480 ierr = MPI_Comm_rank(PETSC_COMM_WORLD,&rank);CHKERRQ(ierr); 1481 if (!((PetscObject)ts)->amsmem && !rank) { 1482 char dir[1024]; 1483 1484 ierr = PetscObjectViewSAWs((PetscObject)ts,viewer);CHKERRQ(ierr); 1485 ierr = PetscSNPrintf(dir,1024,"/PETSc/Objects/%s/time_step",name);CHKERRQ(ierr); 1486 PetscStackCallSAWs(SAWs_Register,(dir,&ts->steps,1,SAWs_READ,SAWs_INT)); 1487 ierr = PetscSNPrintf(dir,1024,"/PETSc/Objects/%s/time",name);CHKERRQ(ierr); 1488 PetscStackCallSAWs(SAWs_Register,(dir,&ts->ptime,1,SAWs_READ,SAWs_DOUBLE)); 1489 } 1490 if (ts->ops->view) { 1491 ierr = (*ts->ops->view)(ts,viewer);CHKERRQ(ierr); 1492 } 1493 #endif 1494 } 1495 1496 ierr = PetscViewerASCIIPushTab(viewer);CHKERRQ(ierr); 1497 ierr = PetscObjectTypeCompare((PetscObject)ts,TSSUNDIALS,&isundials);CHKERRQ(ierr); 1498 ierr = PetscViewerASCIIPopTab(viewer);CHKERRQ(ierr); 1499 PetscFunctionReturn(0); 1500 } 1501 1502 1503 #undef __FUNCT__ 1504 #define __FUNCT__ "TSSetApplicationContext" 1505 /*@ 1506 TSSetApplicationContext - Sets an optional user-defined context for 1507 the timesteppers. 1508 1509 Logically Collective on TS 1510 1511 Input Parameters: 1512 + ts - the TS context obtained from TSCreate() 1513 - usrP - optional user context 1514 1515 Level: intermediate 1516 1517 .keywords: TS, timestep, set, application, context 1518 1519 .seealso: TSGetApplicationContext() 1520 @*/ 1521 PetscErrorCode TSSetApplicationContext(TS ts,void *usrP) 1522 { 1523 PetscFunctionBegin; 1524 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1525 ts->user = usrP; 1526 PetscFunctionReturn(0); 1527 } 1528 1529 #undef __FUNCT__ 1530 #define __FUNCT__ "TSGetApplicationContext" 1531 /*@ 1532 TSGetApplicationContext - Gets the user-defined context for the 1533 timestepper. 1534 1535 Not Collective 1536 1537 Input Parameter: 1538 . ts - the TS context obtained from TSCreate() 1539 1540 Output Parameter: 1541 . usrP - user context 1542 1543 Level: intermediate 1544 1545 .keywords: TS, timestep, get, application, context 1546 1547 .seealso: TSSetApplicationContext() 1548 @*/ 1549 PetscErrorCode TSGetApplicationContext(TS ts,void *usrP) 1550 { 1551 PetscFunctionBegin; 1552 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1553 *(void**)usrP = ts->user; 1554 PetscFunctionReturn(0); 1555 } 1556 1557 #undef __FUNCT__ 1558 #define __FUNCT__ "TSGetTimeStepNumber" 1559 /*@ 1560 TSGetTimeStepNumber - Gets the number of time steps completed. 1561 1562 Not Collective 1563 1564 Input Parameter: 1565 . ts - the TS context obtained from TSCreate() 1566 1567 Output Parameter: 1568 . iter - number of steps completed so far 1569 1570 Level: intermediate 1571 1572 .keywords: TS, timestep, get, iteration, number 1573 .seealso: TSGetTime(), TSGetTimeStep(), TSSetPreStep(), TSSetPreStage(), TSSetPostStage(), TSSetPostStep() 1574 @*/ 1575 PetscErrorCode TSGetTimeStepNumber(TS ts,PetscInt *iter) 1576 { 1577 PetscFunctionBegin; 1578 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1579 PetscValidIntPointer(iter,2); 1580 *iter = ts->steps; 1581 PetscFunctionReturn(0); 1582 } 1583 1584 #undef __FUNCT__ 1585 #define __FUNCT__ "TSSetInitialTimeStep" 1586 /*@ 1587 TSSetInitialTimeStep - Sets the initial timestep to be used, 1588 as well as the initial time. 1589 1590 Logically Collective on TS 1591 1592 Input Parameters: 1593 + ts - the TS context obtained from TSCreate() 1594 . initial_time - the initial time 1595 - time_step - the size of the timestep 1596 1597 Level: intermediate 1598 1599 .seealso: TSSetTimeStep(), TSGetTimeStep() 1600 1601 .keywords: TS, set, initial, timestep 1602 @*/ 1603 PetscErrorCode TSSetInitialTimeStep(TS ts,PetscReal initial_time,PetscReal time_step) 1604 { 1605 PetscErrorCode ierr; 1606 1607 PetscFunctionBegin; 1608 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1609 ierr = TSSetTimeStep(ts,time_step);CHKERRQ(ierr); 1610 ierr = TSSetTime(ts,initial_time);CHKERRQ(ierr); 1611 PetscFunctionReturn(0); 1612 } 1613 1614 #undef __FUNCT__ 1615 #define __FUNCT__ "TSSetTimeStep" 1616 /*@ 1617 TSSetTimeStep - Allows one to reset the timestep at any time, 1618 useful for simple pseudo-timestepping codes. 1619 1620 Logically Collective on TS 1621 1622 Input Parameters: 1623 + ts - the TS context obtained from TSCreate() 1624 - time_step - the size of the timestep 1625 1626 Level: intermediate 1627 1628 .seealso: TSSetInitialTimeStep(), TSGetTimeStep() 1629 1630 .keywords: TS, set, timestep 1631 @*/ 1632 PetscErrorCode TSSetTimeStep(TS ts,PetscReal time_step) 1633 { 1634 PetscFunctionBegin; 1635 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1636 PetscValidLogicalCollectiveReal(ts,time_step,2); 1637 ts->time_step = time_step; 1638 ts->time_step_orig = time_step; 1639 PetscFunctionReturn(0); 1640 } 1641 1642 #undef __FUNCT__ 1643 #define __FUNCT__ "TSSetExactFinalTime" 1644 /*@ 1645 TSSetExactFinalTime - Determines whether to adapt the final time step to 1646 match the exact final time, interpolate solution to the exact final time, 1647 or just return at the final time TS computed. 1648 1649 Logically Collective on TS 1650 1651 Input Parameter: 1652 + ts - the time-step context 1653 - eftopt - exact final time option 1654 1655 Level: beginner 1656 1657 .seealso: TSExactFinalTimeOption 1658 @*/ 1659 PetscErrorCode TSSetExactFinalTime(TS ts,TSExactFinalTimeOption eftopt) 1660 { 1661 PetscFunctionBegin; 1662 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1663 PetscValidLogicalCollectiveEnum(ts,eftopt,2); 1664 ts->exact_final_time = eftopt; 1665 PetscFunctionReturn(0); 1666 } 1667 1668 #undef __FUNCT__ 1669 #define __FUNCT__ "TSGetTimeStep" 1670 /*@ 1671 TSGetTimeStep - Gets the current timestep size. 1672 1673 Not Collective 1674 1675 Input Parameter: 1676 . ts - the TS context obtained from TSCreate() 1677 1678 Output Parameter: 1679 . dt - the current timestep size 1680 1681 Level: intermediate 1682 1683 .seealso: TSSetInitialTimeStep(), TSGetTimeStep() 1684 1685 .keywords: TS, get, timestep 1686 @*/ 1687 PetscErrorCode TSGetTimeStep(TS ts,PetscReal *dt) 1688 { 1689 PetscFunctionBegin; 1690 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1691 PetscValidRealPointer(dt,2); 1692 *dt = ts->time_step; 1693 PetscFunctionReturn(0); 1694 } 1695 1696 #undef __FUNCT__ 1697 #define __FUNCT__ "TSGetSolution" 1698 /*@ 1699 TSGetSolution - Returns the solution at the present timestep. It 1700 is valid to call this routine inside the function that you are evaluating 1701 in order to move to the new timestep. This vector not changed until 1702 the solution at the next timestep has been calculated. 1703 1704 Not Collective, but Vec returned is parallel if TS is parallel 1705 1706 Input Parameter: 1707 . ts - the TS context obtained from TSCreate() 1708 1709 Output Parameter: 1710 . v - the vector containing the solution 1711 1712 Level: intermediate 1713 1714 .seealso: TSGetTimeStep() 1715 1716 .keywords: TS, timestep, get, solution 1717 @*/ 1718 PetscErrorCode TSGetSolution(TS ts,Vec *v) 1719 { 1720 PetscFunctionBegin; 1721 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1722 PetscValidPointer(v,2); 1723 *v = ts->vec_sol; 1724 PetscFunctionReturn(0); 1725 } 1726 1727 #undef __FUNCT__ 1728 #define __FUNCT__ "TSGetCostGradients" 1729 /*@ 1730 TSGetCostGradients - Returns the gradients from the TSAdjointSolve() 1731 1732 Not Collective, but Vec returned is parallel if TS is parallel 1733 1734 Input Parameter: 1735 . ts - the TS context obtained from TSCreate() 1736 1737 Output Parameter: 1738 + lambda - vectors containing the gradients of the cost functions with respect to the ODE/DAE solution variables 1739 - mu - vectors containing the gradients of the cost functions with respect to the problem parameters 1740 1741 Level: intermediate 1742 1743 .seealso: TSGetTimeStep() 1744 1745 .keywords: TS, timestep, get, sensitivity 1746 @*/ 1747 PetscErrorCode TSGetCostGradients(TS ts,PetscInt *numcost,Vec **lambda,Vec **mu) 1748 { 1749 PetscFunctionBegin; 1750 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1751 if (numcost) *numcost = ts->numcost; 1752 if (lambda) *lambda = ts->vecs_sensi; 1753 if (mu) *mu = ts->vecs_sensip; 1754 PetscFunctionReturn(0); 1755 } 1756 1757 /* ----- Routines to initialize and destroy a timestepper ---- */ 1758 #undef __FUNCT__ 1759 #define __FUNCT__ "TSSetProblemType" 1760 /*@ 1761 TSSetProblemType - Sets the type of problem to be solved. 1762 1763 Not collective 1764 1765 Input Parameters: 1766 + ts - The TS 1767 - type - One of TS_LINEAR, TS_NONLINEAR where these types refer to problems of the forms 1768 .vb 1769 U_t - A U = 0 (linear) 1770 U_t - A(t) U = 0 (linear) 1771 F(t,U,U_t) = 0 (nonlinear) 1772 .ve 1773 1774 Level: beginner 1775 1776 .keywords: TS, problem type 1777 .seealso: TSSetUp(), TSProblemType, TS 1778 @*/ 1779 PetscErrorCode TSSetProblemType(TS ts, TSProblemType type) 1780 { 1781 PetscErrorCode ierr; 1782 1783 PetscFunctionBegin; 1784 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 1785 ts->problem_type = type; 1786 if (type == TS_LINEAR) { 1787 SNES snes; 1788 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 1789 ierr = SNESSetType(snes,SNESKSPONLY);CHKERRQ(ierr); 1790 } 1791 PetscFunctionReturn(0); 1792 } 1793 1794 #undef __FUNCT__ 1795 #define __FUNCT__ "TSGetProblemType" 1796 /*@C 1797 TSGetProblemType - Gets the type of problem to be solved. 1798 1799 Not collective 1800 1801 Input Parameter: 1802 . ts - The TS 1803 1804 Output Parameter: 1805 . type - One of TS_LINEAR, TS_NONLINEAR where these types refer to problems of the forms 1806 .vb 1807 M U_t = A U 1808 M(t) U_t = A(t) U 1809 F(t,U,U_t) 1810 .ve 1811 1812 Level: beginner 1813 1814 .keywords: TS, problem type 1815 .seealso: TSSetUp(), TSProblemType, TS 1816 @*/ 1817 PetscErrorCode TSGetProblemType(TS ts, TSProblemType *type) 1818 { 1819 PetscFunctionBegin; 1820 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 1821 PetscValidIntPointer(type,2); 1822 *type = ts->problem_type; 1823 PetscFunctionReturn(0); 1824 } 1825 1826 #undef __FUNCT__ 1827 #define __FUNCT__ "TSSetUp" 1828 /*@ 1829 TSSetUp - Sets up the internal data structures for the later use 1830 of a timestepper. 1831 1832 Collective on TS 1833 1834 Input Parameter: 1835 . ts - the TS context obtained from TSCreate() 1836 1837 Notes: 1838 For basic use of the TS solvers the user need not explicitly call 1839 TSSetUp(), since these actions will automatically occur during 1840 the call to TSStep(). However, if one wishes to control this 1841 phase separately, TSSetUp() should be called after TSCreate() 1842 and optional routines of the form TSSetXXX(), but before TSStep(). 1843 1844 Level: advanced 1845 1846 .keywords: TS, timestep, setup 1847 1848 .seealso: TSCreate(), TSStep(), TSDestroy() 1849 @*/ 1850 PetscErrorCode TSSetUp(TS ts) 1851 { 1852 PetscErrorCode ierr; 1853 DM dm; 1854 PetscErrorCode (*func)(SNES,Vec,Vec,void*); 1855 PetscErrorCode (*jac)(SNES,Vec,Mat,Mat,void*); 1856 TSIJacobian ijac; 1857 TSRHSJacobian rhsjac; 1858 1859 PetscFunctionBegin; 1860 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1861 if (ts->setupcalled) PetscFunctionReturn(0); 1862 1863 ts->total_steps = 0; 1864 if (!((PetscObject)ts)->type_name) { 1865 ierr = TSSetType(ts,TSEULER);CHKERRQ(ierr); 1866 } 1867 1868 if (!ts->vec_sol) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Must call TSSetSolution() first"); 1869 1870 1871 ierr = TSGetAdapt(ts,&ts->adapt);CHKERRQ(ierr); 1872 1873 if (ts->rhsjacobian.reuse) { 1874 Mat Amat,Pmat; 1875 SNES snes; 1876 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 1877 ierr = SNESGetJacobian(snes,&Amat,&Pmat,NULL,NULL);CHKERRQ(ierr); 1878 /* Matching matrices implies that an IJacobian is NOT set, because if it had been set, the IJacobian's matrix would 1879 * have displaced the RHS matrix */ 1880 if (Amat == ts->Arhs) { 1881 ierr = MatDuplicate(ts->Arhs,MAT_DO_NOT_COPY_VALUES,&Amat);CHKERRQ(ierr); 1882 ierr = SNESSetJacobian(snes,Amat,NULL,NULL,NULL);CHKERRQ(ierr); 1883 ierr = MatDestroy(&Amat);CHKERRQ(ierr); 1884 } 1885 if (Pmat == ts->Brhs) { 1886 ierr = MatDuplicate(ts->Brhs,MAT_DO_NOT_COPY_VALUES,&Pmat);CHKERRQ(ierr); 1887 ierr = SNESSetJacobian(snes,NULL,Pmat,NULL,NULL);CHKERRQ(ierr); 1888 ierr = MatDestroy(&Pmat);CHKERRQ(ierr); 1889 } 1890 } 1891 if (ts->ops->setup) { 1892 ierr = (*ts->ops->setup)(ts);CHKERRQ(ierr); 1893 } 1894 1895 /* in the case where we've set a DMTSFunction or what have you, we need the default SNESFunction 1896 to be set right but can't do it elsewhere due to the overreliance on ctx=ts. 1897 */ 1898 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 1899 ierr = DMSNESGetFunction(dm,&func,NULL);CHKERRQ(ierr); 1900 if (!func) { 1901 ierr =DMSNESSetFunction(dm,SNESTSFormFunction,ts);CHKERRQ(ierr); 1902 } 1903 /* if the SNES doesn't have a jacobian set and the TS has an ijacobian or rhsjacobian set, set the SNES to use it. 1904 Otherwise, the SNES will use coloring internally to form the Jacobian. 1905 */ 1906 ierr = DMSNESGetJacobian(dm,&jac,NULL);CHKERRQ(ierr); 1907 ierr = DMTSGetIJacobian(dm,&ijac,NULL);CHKERRQ(ierr); 1908 ierr = DMTSGetRHSJacobian(dm,&rhsjac,NULL);CHKERRQ(ierr); 1909 if (!jac && (ijac || rhsjac)) { 1910 ierr = DMSNESSetJacobian(dm,SNESTSFormJacobian,ts);CHKERRQ(ierr); 1911 } 1912 ts->setupcalled = PETSC_TRUE; 1913 PetscFunctionReturn(0); 1914 } 1915 1916 #undef __FUNCT__ 1917 #define __FUNCT__ "TSAdjointSetUp" 1918 /*@ 1919 TSAdjointSetUp - Sets up the internal data structures for the later use 1920 of an adjoint solver 1921 1922 Collective on TS 1923 1924 Input Parameter: 1925 . ts - the TS context obtained from TSCreate() 1926 1927 Level: advanced 1928 1929 .keywords: TS, timestep, setup 1930 1931 .seealso: TSCreate(), TSAdjointStep(), TSSetCostGradients() 1932 @*/ 1933 PetscErrorCode TSAdjointSetUp(TS ts) 1934 { 1935 PetscErrorCode ierr; 1936 1937 PetscFunctionBegin; 1938 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1939 if (ts->adjointsetupcalled) PetscFunctionReturn(0); 1940 if (!ts->vecs_sensi) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Must call TSSetCostGradients() first"); 1941 1942 if (ts->vec_costintegral) { /* if there is integral in the cost function*/ 1943 ierr = VecDuplicateVecs(ts->vecs_sensi[0],ts->numcost,&ts->vecs_drdy);CHKERRQ(ierr); 1944 if (ts->vecs_sensip){ 1945 ierr = VecDuplicateVecs(ts->vecs_sensip[0],ts->numcost,&ts->vecs_drdp);CHKERRQ(ierr); 1946 } 1947 } 1948 1949 if (ts->ops->adjointsetup) { 1950 ierr = (*ts->ops->adjointsetup)(ts);CHKERRQ(ierr); 1951 } 1952 ts->adjointsetupcalled = PETSC_TRUE; 1953 PetscFunctionReturn(0); 1954 } 1955 1956 #undef __FUNCT__ 1957 #define __FUNCT__ "TSReset" 1958 /*@ 1959 TSReset - Resets a TS context and removes any allocated Vecs and Mats. 1960 1961 Collective on TS 1962 1963 Input Parameter: 1964 . ts - the TS context obtained from TSCreate() 1965 1966 Level: beginner 1967 1968 .keywords: TS, timestep, reset 1969 1970 .seealso: TSCreate(), TSSetup(), TSDestroy() 1971 @*/ 1972 PetscErrorCode TSReset(TS ts) 1973 { 1974 PetscErrorCode ierr; 1975 1976 PetscFunctionBegin; 1977 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 1978 1979 if (ts->ops->reset) { 1980 ierr = (*ts->ops->reset)(ts);CHKERRQ(ierr); 1981 } 1982 if (ts->snes) {ierr = SNESReset(ts->snes);CHKERRQ(ierr);} 1983 if (ts->adapt) {ierr = TSAdaptReset(ts->adapt);CHKERRQ(ierr);} 1984 1985 ierr = MatDestroy(&ts->Arhs);CHKERRQ(ierr); 1986 ierr = MatDestroy(&ts->Brhs);CHKERRQ(ierr); 1987 ierr = VecDestroy(&ts->Frhs);CHKERRQ(ierr); 1988 ierr = VecDestroy(&ts->vec_sol);CHKERRQ(ierr); 1989 ierr = VecDestroy(&ts->vatol);CHKERRQ(ierr); 1990 ierr = VecDestroy(&ts->vrtol);CHKERRQ(ierr); 1991 ierr = VecDestroyVecs(ts->nwork,&ts->work);CHKERRQ(ierr); 1992 1993 if (ts->vec_costintegral) { 1994 ierr = VecDestroyVecs(ts->numcost,&ts->vecs_drdy);CHKERRQ(ierr); 1995 if (ts->vecs_drdp){ 1996 ierr = VecDestroyVecs(ts->numcost,&ts->vecs_drdp);CHKERRQ(ierr); 1997 } 1998 } 1999 ts->vecs_sensi = NULL; 2000 ts->vecs_sensip = NULL; 2001 ierr = MatDestroy(&ts->Jacp);CHKERRQ(ierr); 2002 ierr = VecDestroy(&ts->vec_costintegral);CHKERRQ(ierr); 2003 ierr = VecDestroy(&ts->vec_costintegrand);CHKERRQ(ierr); 2004 ts->setupcalled = PETSC_FALSE; 2005 PetscFunctionReturn(0); 2006 } 2007 2008 #undef __FUNCT__ 2009 #define __FUNCT__ "TSDestroy" 2010 /*@ 2011 TSDestroy - Destroys the timestepper context that was created 2012 with TSCreate(). 2013 2014 Collective on TS 2015 2016 Input Parameter: 2017 . ts - the TS context obtained from TSCreate() 2018 2019 Level: beginner 2020 2021 .keywords: TS, timestepper, destroy 2022 2023 .seealso: TSCreate(), TSSetUp(), TSSolve() 2024 @*/ 2025 PetscErrorCode TSDestroy(TS *ts) 2026 { 2027 PetscErrorCode ierr; 2028 2029 PetscFunctionBegin; 2030 if (!*ts) PetscFunctionReturn(0); 2031 PetscValidHeaderSpecific((*ts),TS_CLASSID,1); 2032 if (--((PetscObject)(*ts))->refct > 0) {*ts = 0; PetscFunctionReturn(0);} 2033 2034 ierr = TSReset((*ts));CHKERRQ(ierr); 2035 2036 /* if memory was published with SAWs then destroy it */ 2037 ierr = PetscObjectSAWsViewOff((PetscObject)*ts);CHKERRQ(ierr); 2038 if ((*ts)->ops->destroy) {ierr = (*(*ts)->ops->destroy)((*ts));CHKERRQ(ierr);} 2039 2040 ierr = TSTrajectoryDestroy(&(*ts)->trajectory);CHKERRQ(ierr); 2041 2042 ierr = TSAdaptDestroy(&(*ts)->adapt);CHKERRQ(ierr); 2043 if ((*ts)->event) { 2044 ierr = TSEventMonitorDestroy(&(*ts)->event);CHKERRQ(ierr); 2045 } 2046 ierr = SNESDestroy(&(*ts)->snes);CHKERRQ(ierr); 2047 ierr = DMDestroy(&(*ts)->dm);CHKERRQ(ierr); 2048 ierr = TSMonitorCancel((*ts));CHKERRQ(ierr); 2049 2050 ierr = PetscHeaderDestroy(ts);CHKERRQ(ierr); 2051 PetscFunctionReturn(0); 2052 } 2053 2054 #undef __FUNCT__ 2055 #define __FUNCT__ "TSGetSNES" 2056 /*@ 2057 TSGetSNES - Returns the SNES (nonlinear solver) associated with 2058 a TS (timestepper) context. Valid only for nonlinear problems. 2059 2060 Not Collective, but SNES is parallel if TS is parallel 2061 2062 Input Parameter: 2063 . ts - the TS context obtained from TSCreate() 2064 2065 Output Parameter: 2066 . snes - the nonlinear solver context 2067 2068 Notes: 2069 The user can then directly manipulate the SNES context to set various 2070 options, etc. Likewise, the user can then extract and manipulate the 2071 KSP, KSP, and PC contexts as well. 2072 2073 TSGetSNES() does not work for integrators that do not use SNES; in 2074 this case TSGetSNES() returns NULL in snes. 2075 2076 Level: beginner 2077 2078 .keywords: timestep, get, SNES 2079 @*/ 2080 PetscErrorCode TSGetSNES(TS ts,SNES *snes) 2081 { 2082 PetscErrorCode ierr; 2083 2084 PetscFunctionBegin; 2085 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2086 PetscValidPointer(snes,2); 2087 if (!ts->snes) { 2088 ierr = SNESCreate(PetscObjectComm((PetscObject)ts),&ts->snes);CHKERRQ(ierr); 2089 ierr = SNESSetFunction(ts->snes,NULL,SNESTSFormFunction,ts);CHKERRQ(ierr); 2090 ierr = PetscLogObjectParent((PetscObject)ts,(PetscObject)ts->snes);CHKERRQ(ierr); 2091 ierr = PetscObjectIncrementTabLevel((PetscObject)ts->snes,(PetscObject)ts,1);CHKERRQ(ierr); 2092 if (ts->dm) {ierr = SNESSetDM(ts->snes,ts->dm);CHKERRQ(ierr);} 2093 if (ts->problem_type == TS_LINEAR) { 2094 ierr = SNESSetType(ts->snes,SNESKSPONLY);CHKERRQ(ierr); 2095 } 2096 } 2097 *snes = ts->snes; 2098 PetscFunctionReturn(0); 2099 } 2100 2101 #undef __FUNCT__ 2102 #define __FUNCT__ "TSSetSNES" 2103 /*@ 2104 TSSetSNES - Set the SNES (nonlinear solver) to be used by the timestepping context 2105 2106 Collective 2107 2108 Input Parameter: 2109 + ts - the TS context obtained from TSCreate() 2110 - snes - the nonlinear solver context 2111 2112 Notes: 2113 Most users should have the TS created by calling TSGetSNES() 2114 2115 Level: developer 2116 2117 .keywords: timestep, set, SNES 2118 @*/ 2119 PetscErrorCode TSSetSNES(TS ts,SNES snes) 2120 { 2121 PetscErrorCode ierr; 2122 PetscErrorCode (*func)(SNES,Vec,Mat,Mat,void*); 2123 2124 PetscFunctionBegin; 2125 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2126 PetscValidHeaderSpecific(snes,SNES_CLASSID,2); 2127 ierr = PetscObjectReference((PetscObject)snes);CHKERRQ(ierr); 2128 ierr = SNESDestroy(&ts->snes);CHKERRQ(ierr); 2129 2130 ts->snes = snes; 2131 2132 ierr = SNESSetFunction(ts->snes,NULL,SNESTSFormFunction,ts);CHKERRQ(ierr); 2133 ierr = SNESGetJacobian(ts->snes,NULL,NULL,&func,NULL);CHKERRQ(ierr); 2134 if (func == SNESTSFormJacobian) { 2135 ierr = SNESSetJacobian(ts->snes,NULL,NULL,SNESTSFormJacobian,ts);CHKERRQ(ierr); 2136 } 2137 PetscFunctionReturn(0); 2138 } 2139 2140 #undef __FUNCT__ 2141 #define __FUNCT__ "TSGetKSP" 2142 /*@ 2143 TSGetKSP - Returns the KSP (linear solver) associated with 2144 a TS (timestepper) context. 2145 2146 Not Collective, but KSP is parallel if TS is parallel 2147 2148 Input Parameter: 2149 . ts - the TS context obtained from TSCreate() 2150 2151 Output Parameter: 2152 . ksp - the nonlinear solver context 2153 2154 Notes: 2155 The user can then directly manipulate the KSP context to set various 2156 options, etc. Likewise, the user can then extract and manipulate the 2157 KSP and PC contexts as well. 2158 2159 TSGetKSP() does not work for integrators that do not use KSP; 2160 in this case TSGetKSP() returns NULL in ksp. 2161 2162 Level: beginner 2163 2164 .keywords: timestep, get, KSP 2165 @*/ 2166 PetscErrorCode TSGetKSP(TS ts,KSP *ksp) 2167 { 2168 PetscErrorCode ierr; 2169 SNES snes; 2170 2171 PetscFunctionBegin; 2172 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2173 PetscValidPointer(ksp,2); 2174 if (!((PetscObject)ts)->type_name) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_NULL,"KSP is not created yet. Call TSSetType() first"); 2175 if (ts->problem_type != TS_LINEAR) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Linear only; use TSGetSNES()"); 2176 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 2177 ierr = SNESGetKSP(snes,ksp);CHKERRQ(ierr); 2178 PetscFunctionReturn(0); 2179 } 2180 2181 /* ----------- Routines to set solver parameters ---------- */ 2182 2183 #undef __FUNCT__ 2184 #define __FUNCT__ "TSGetDuration" 2185 /*@ 2186 TSGetDuration - Gets the maximum number of timesteps to use and 2187 maximum time for iteration. 2188 2189 Not Collective 2190 2191 Input Parameters: 2192 + ts - the TS context obtained from TSCreate() 2193 . maxsteps - maximum number of iterations to use, or NULL 2194 - maxtime - final time to iterate to, or NULL 2195 2196 Level: intermediate 2197 2198 .keywords: TS, timestep, get, maximum, iterations, time 2199 @*/ 2200 PetscErrorCode TSGetDuration(TS ts, PetscInt *maxsteps, PetscReal *maxtime) 2201 { 2202 PetscFunctionBegin; 2203 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 2204 if (maxsteps) { 2205 PetscValidIntPointer(maxsteps,2); 2206 *maxsteps = ts->max_steps; 2207 } 2208 if (maxtime) { 2209 PetscValidScalarPointer(maxtime,3); 2210 *maxtime = ts->max_time; 2211 } 2212 PetscFunctionReturn(0); 2213 } 2214 2215 #undef __FUNCT__ 2216 #define __FUNCT__ "TSSetDuration" 2217 /*@ 2218 TSSetDuration - Sets the maximum number of timesteps to use and 2219 maximum time for iteration. 2220 2221 Logically Collective on TS 2222 2223 Input Parameters: 2224 + ts - the TS context obtained from TSCreate() 2225 . maxsteps - maximum number of iterations to use 2226 - maxtime - final time to iterate to 2227 2228 Options Database Keys: 2229 . -ts_max_steps <maxsteps> - Sets maxsteps 2230 . -ts_final_time <maxtime> - Sets maxtime 2231 2232 Notes: 2233 The default maximum number of iterations is 5000. Default time is 5.0 2234 2235 Level: intermediate 2236 2237 .keywords: TS, timestep, set, maximum, iterations 2238 2239 .seealso: TSSetExactFinalTime() 2240 @*/ 2241 PetscErrorCode TSSetDuration(TS ts,PetscInt maxsteps,PetscReal maxtime) 2242 { 2243 PetscFunctionBegin; 2244 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2245 PetscValidLogicalCollectiveInt(ts,maxsteps,2); 2246 PetscValidLogicalCollectiveReal(ts,maxtime,2); 2247 if (maxsteps >= 0) ts->max_steps = maxsteps; 2248 if (maxtime != PETSC_DEFAULT) ts->max_time = maxtime; 2249 PetscFunctionReturn(0); 2250 } 2251 2252 #undef __FUNCT__ 2253 #define __FUNCT__ "TSSetSolution" 2254 /*@ 2255 TSSetSolution - Sets the initial solution vector 2256 for use by the TS routines. 2257 2258 Logically Collective on TS and Vec 2259 2260 Input Parameters: 2261 + ts - the TS context obtained from TSCreate() 2262 - u - the solution vector 2263 2264 Level: beginner 2265 2266 .keywords: TS, timestep, set, solution, initial conditions 2267 @*/ 2268 PetscErrorCode TSSetSolution(TS ts,Vec u) 2269 { 2270 PetscErrorCode ierr; 2271 DM dm; 2272 2273 PetscFunctionBegin; 2274 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2275 PetscValidHeaderSpecific(u,VEC_CLASSID,2); 2276 ierr = PetscObjectReference((PetscObject)u);CHKERRQ(ierr); 2277 ierr = VecDestroy(&ts->vec_sol);CHKERRQ(ierr); 2278 2279 ts->vec_sol = u; 2280 2281 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 2282 ierr = DMShellSetGlobalVector(dm,u);CHKERRQ(ierr); 2283 PetscFunctionReturn(0); 2284 } 2285 2286 #undef __FUNCT__ 2287 #define __FUNCT__ "TSAdjointSetSteps" 2288 /*@ 2289 TSAdjointSetSteps - Sets the number of steps the adjoint solver should take backward in time 2290 2291 Logically Collective on TS 2292 2293 Input Parameters: 2294 + ts - the TS context obtained from TSCreate() 2295 . steps - number of steps to use 2296 2297 Level: intermediate 2298 2299 Notes: Normally one does not call this and TSAdjointSolve() integrates back to the original timestep. One can call this 2300 so as to integrate back to less than the original timestep 2301 2302 .keywords: TS, timestep, set, maximum, iterations 2303 2304 .seealso: TSSetExactFinalTime() 2305 @*/ 2306 PetscErrorCode TSAdjointSetSteps(TS ts,PetscInt steps) 2307 { 2308 PetscFunctionBegin; 2309 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2310 PetscValidLogicalCollectiveInt(ts,steps,2); 2311 if (steps < 0) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_ARG_OUTOFRANGE,"Cannot step back a negative number of steps"); 2312 if (steps > ts->total_steps) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_ARG_OUTOFRANGE,"Cannot step back more than the total number of forward steps"); 2313 ts->adjoint_max_steps = steps; 2314 PetscFunctionReturn(0); 2315 } 2316 2317 #undef __FUNCT__ 2318 #define __FUNCT__ "TSSetCostGradients" 2319 /*@ 2320 TSSetCostGradients - Sets the initial value of the gradients of the cost function w.r.t. initial conditions and w.r.t. the problem parameters 2321 for use by the TSAdjoint routines. 2322 2323 Logically Collective on TS and Vec 2324 2325 Input Parameters: 2326 + ts - the TS context obtained from TSCreate() 2327 . lambda - gradients with respect to the initial condition variables, the dimension and parallel layout of these vectors is the same as the ODE solution vector 2328 - mu - gradients with respect to the parameters, the number of entries in these vectors is the same as the number of parameters 2329 2330 Level: beginner 2331 2332 Notes: the entries in these vectors must be correctly initialized with the values lamda_i = df/dy|finaltime mu_i = df/dp|finaltime 2333 2334 .keywords: TS, timestep, set, sensitivity, initial conditions 2335 @*/ 2336 PetscErrorCode TSSetCostGradients(TS ts,PetscInt numcost,Vec *lambda,Vec *mu) 2337 { 2338 PetscFunctionBegin; 2339 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2340 PetscValidPointer(lambda,2); 2341 ts->vecs_sensi = lambda; 2342 ts->vecs_sensip = mu; 2343 if (ts->numcost && ts->numcost!=numcost) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"The number of cost functions (2rd parameter of TSSetCostIntegrand()) is inconsistent with the one set by TSSetCostIntegrand"); 2344 ts->numcost = numcost; 2345 PetscFunctionReturn(0); 2346 } 2347 2348 #undef __FUNCT__ 2349 #define __FUNCT__ "TSAdjointSetRHSJacobian" 2350 /*@C 2351 TSAdjointSetRHSJacobian - Sets the function that computes the Jacobian of G w.r.t. the parameters p where y_t = G(y,p,t), as well as the location to store the matrix. 2352 2353 Logically Collective on TS 2354 2355 Input Parameters: 2356 + ts - The TS context obtained from TSCreate() 2357 - func - The function 2358 2359 Calling sequence of func: 2360 $ func (TS ts,PetscReal t,Vec y,Mat A,void *ctx); 2361 + t - current timestep 2362 . y - input vector (current ODE solution) 2363 . A - output matrix 2364 - ctx - [optional] user-defined function context 2365 2366 Level: intermediate 2367 2368 Notes: Amat has the same number of rows and the same row parallel layout as u, Amat has the same number of columns and parallel layout as p 2369 2370 .keywords: TS, sensitivity 2371 .seealso: 2372 @*/ 2373 PetscErrorCode TSAdjointSetRHSJacobian(TS ts,Mat Amat,PetscErrorCode (*func)(TS,PetscReal,Vec,Mat,void*),void *ctx) 2374 { 2375 PetscErrorCode ierr; 2376 2377 PetscFunctionBegin; 2378 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 2379 if (Amat) PetscValidHeaderSpecific(Amat,MAT_CLASSID,2); 2380 2381 ts->rhsjacobianp = func; 2382 ts->rhsjacobianpctx = ctx; 2383 if(Amat) { 2384 ierr = PetscObjectReference((PetscObject)Amat);CHKERRQ(ierr); 2385 ierr = MatDestroy(&ts->Jacp);CHKERRQ(ierr); 2386 ts->Jacp = Amat; 2387 } 2388 PetscFunctionReturn(0); 2389 } 2390 2391 #undef __FUNCT__ 2392 #define __FUNCT__ "TSAdjointComputeRHSJacobian" 2393 /*@C 2394 TSAdjointComputeRHSJacobian - Runs the user-defined Jacobian function. 2395 2396 Collective on TS 2397 2398 Input Parameters: 2399 . ts - The TS context obtained from TSCreate() 2400 2401 Level: developer 2402 2403 .keywords: TS, sensitivity 2404 .seealso: TSAdjointSetRHSJacobian() 2405 @*/ 2406 PetscErrorCode TSAdjointComputeRHSJacobian(TS ts,PetscReal t,Vec X,Mat Amat) 2407 { 2408 PetscErrorCode ierr; 2409 2410 PetscFunctionBegin; 2411 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2412 PetscValidHeaderSpecific(X,VEC_CLASSID,3); 2413 PetscValidPointer(Amat,4); 2414 2415 PetscStackPush("TS user JacobianP function for sensitivity analysis"); 2416 ierr = (*ts->rhsjacobianp)(ts,t,X,Amat,ts->rhsjacobianpctx); CHKERRQ(ierr); 2417 PetscStackPop; 2418 PetscFunctionReturn(0); 2419 } 2420 2421 #undef __FUNCT__ 2422 #define __FUNCT__ "TSSetCostIntegrand" 2423 /*@C 2424 TSSetCostIntegrand - Sets the routine for evaluating the integral term in one or more cost functions 2425 2426 Logically Collective on TS 2427 2428 Input Parameters: 2429 + ts - the TS context obtained from TSCreate() 2430 . numcost - number of gradients to be computed, this is the number of cost functions 2431 . rf - routine for evaluating the integrand function 2432 . drdyf - function that computes the gradients of the r's with respect to y,NULL if not a function y 2433 . drdpf - function that computes the gradients of the r's with respect to p, NULL if not a function of p 2434 - ctx - [optional] user-defined context for private data for the function evaluation routine (may be NULL) 2435 2436 Calling sequence of rf: 2437 $ rf(TS ts,PetscReal t,Vec y,Vec f[],void *ctx); 2438 2439 + t - current timestep 2440 . y - input vector 2441 . f - function result; one vector entry for each cost function 2442 - ctx - [optional] user-defined function context 2443 2444 Calling sequence of drdyf: 2445 $ PetscErroCode drdyf(TS ts,PetscReal t,Vec y,Vec *drdy,void *ctx); 2446 2447 Calling sequence of drdpf: 2448 $ PetscErroCode drdpf(TS ts,PetscReal t,Vec y,Vec *drdp,void *ctx); 2449 2450 Level: intermediate 2451 2452 Notes: For optimization there is generally a single cost function, numcost = 1. For sensitivities there may be multiple cost functions 2453 2454 .keywords: TS, sensitivity analysis, timestep, set, quadrature, function 2455 2456 .seealso: TSAdjointSetRHSJacobian(),TSGetCostGradients(), TSSetCostGradients() 2457 @*/ 2458 PetscErrorCode TSSetCostIntegrand(TS ts,PetscInt numcost, PetscErrorCode (*rf)(TS,PetscReal,Vec,Vec,void*), 2459 PetscErrorCode (*drdyf)(TS,PetscReal,Vec,Vec*,void*), 2460 PetscErrorCode (*drdpf)(TS,PetscReal,Vec,Vec*,void*),void *ctx) 2461 { 2462 PetscErrorCode ierr; 2463 2464 PetscFunctionBegin; 2465 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2466 if (ts->numcost && ts->numcost!=numcost) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"The number of cost functions (2rd parameter of TSSetCostIntegrand()) is inconsistent with the one set by TSSetCostGradients()"); 2467 if (!ts->numcost) ts->numcost=numcost; 2468 2469 ierr = VecCreateSeq(PETSC_COMM_SELF,numcost,&ts->vec_costintegral);CHKERRQ(ierr); 2470 ierr = VecDuplicate(ts->vec_costintegral,&ts->vec_costintegrand);CHKERRQ(ierr); 2471 ts->costintegrand = rf; 2472 ts->costintegrandctx = ctx; 2473 ts->drdyfunction = drdyf; 2474 ts->drdpfunction = drdpf; 2475 PetscFunctionReturn(0); 2476 } 2477 2478 #undef __FUNCT__ 2479 #define __FUNCT__ "TSGetCostIntegral" 2480 /*@ 2481 TSGetCostIntegral - Returns the values of the integral term in the cost functions. 2482 It is valid to call the routine after a backward run. 2483 2484 Not Collective 2485 2486 Input Parameter: 2487 . ts - the TS context obtained from TSCreate() 2488 2489 Output Parameter: 2490 . v - the vector containing the integrals for each cost function 2491 2492 Level: intermediate 2493 2494 .seealso: TSSetCostIntegrand() 2495 2496 .keywords: TS, sensitivity analysis 2497 @*/ 2498 PetscErrorCode TSGetCostIntegral(TS ts,Vec *v) 2499 { 2500 PetscFunctionBegin; 2501 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2502 PetscValidPointer(v,2); 2503 *v = ts->vec_costintegral; 2504 PetscFunctionReturn(0); 2505 } 2506 2507 #undef __FUNCT__ 2508 #define __FUNCT__ "TSAdjointComputeCostIntegrand" 2509 /*@ 2510 TSAdjointComputeCostIntegrand - Evaluates the integral function in the cost functions. 2511 2512 Input Parameters: 2513 + ts - the TS context 2514 . t - current time 2515 - y - state vector, i.e. current solution 2516 2517 Output Parameter: 2518 . q - vector of size numcost to hold the outputs 2519 2520 Note: 2521 Most users should not need to explicitly call this routine, as it 2522 is used internally within the sensitivity analysis context. 2523 2524 Level: developer 2525 2526 .keywords: TS, compute 2527 2528 .seealso: TSSetCostIntegrand() 2529 @*/ 2530 PetscErrorCode TSAdjointComputeCostIntegrand(TS ts,PetscReal t,Vec y,Vec q) 2531 { 2532 PetscErrorCode ierr; 2533 2534 PetscFunctionBegin; 2535 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2536 PetscValidHeaderSpecific(y,VEC_CLASSID,3); 2537 PetscValidHeaderSpecific(q,VEC_CLASSID,4); 2538 2539 ierr = PetscLogEventBegin(TS_FunctionEval,ts,y,q,0);CHKERRQ(ierr); 2540 if (ts->costintegrand) { 2541 PetscStackPush("TS user integrand in the cost function"); 2542 ierr = (*ts->costintegrand)(ts,t,y,q,ts->costintegrandctx);CHKERRQ(ierr); 2543 PetscStackPop; 2544 } else { 2545 ierr = VecZeroEntries(q);CHKERRQ(ierr); 2546 } 2547 2548 ierr = PetscLogEventEnd(TS_FunctionEval,ts,y,q,0);CHKERRQ(ierr); 2549 PetscFunctionReturn(0); 2550 } 2551 2552 #undef __FUNCT__ 2553 #define __FUNCT__ "TSAdjointComputeDRDYFunction" 2554 /*@ 2555 TSAdjointComputeDRDYFunction - Runs the user-defined DRDY function. 2556 2557 Collective on TS 2558 2559 Input Parameters: 2560 . ts - The TS context obtained from TSCreate() 2561 2562 Notes: 2563 TSAdjointComputeDRDYFunction() is typically used for sensitivity implementation, 2564 so most users would not generally call this routine themselves. 2565 2566 Level: developer 2567 2568 .keywords: TS, sensitivity 2569 .seealso: TSAdjointComputeDRDYFunction() 2570 @*/ 2571 PetscErrorCode TSAdjointComputeDRDYFunction(TS ts,PetscReal t,Vec y,Vec *drdy) 2572 { 2573 PetscErrorCode ierr; 2574 2575 PetscFunctionBegin; 2576 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2577 PetscValidHeaderSpecific(y,VEC_CLASSID,3); 2578 2579 PetscStackPush("TS user DRDY function for sensitivity analysis"); 2580 ierr = (*ts->drdyfunction)(ts,t,y,drdy,ts->costintegrandctx); CHKERRQ(ierr); 2581 PetscStackPop; 2582 PetscFunctionReturn(0); 2583 } 2584 2585 #undef __FUNCT__ 2586 #define __FUNCT__ "TSAdjointComputeDRDPFunction" 2587 /*@ 2588 TSAdjointComputeDRDPFunction - Runs the user-defined DRDP function. 2589 2590 Collective on TS 2591 2592 Input Parameters: 2593 . ts - The TS context obtained from TSCreate() 2594 2595 Notes: 2596 TSDRDPFunction() is typically used for sensitivity implementation, 2597 so most users would not generally call this routine themselves. 2598 2599 Level: developer 2600 2601 .keywords: TS, sensitivity 2602 .seealso: TSAdjointSetDRDPFunction() 2603 @*/ 2604 PetscErrorCode TSAdjointComputeDRDPFunction(TS ts,PetscReal t,Vec y,Vec *drdp) 2605 { 2606 PetscErrorCode ierr; 2607 2608 PetscFunctionBegin; 2609 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2610 PetscValidHeaderSpecific(y,VEC_CLASSID,3); 2611 2612 PetscStackPush("TS user DRDP function for sensitivity analysis"); 2613 ierr = (*ts->drdpfunction)(ts,t,y,drdp,ts->costintegrandctx); CHKERRQ(ierr); 2614 PetscStackPop; 2615 PetscFunctionReturn(0); 2616 } 2617 2618 #undef __FUNCT__ 2619 #define __FUNCT__ "TSSetPreStep" 2620 /*@C 2621 TSSetPreStep - Sets the general-purpose function 2622 called once at the beginning of each time step. 2623 2624 Logically Collective on TS 2625 2626 Input Parameters: 2627 + ts - The TS context obtained from TSCreate() 2628 - func - The function 2629 2630 Calling sequence of func: 2631 . func (TS ts); 2632 2633 Level: intermediate 2634 2635 Note: 2636 If a step is rejected, TSStep() will call this routine again before each attempt. 2637 The last completed time step number can be queried using TSGetTimeStepNumber(), the 2638 size of the step being attempted can be obtained using TSGetTimeStep(). 2639 2640 .keywords: TS, timestep 2641 .seealso: TSSetPreStage(), TSSetPostStage(), TSSetPostStep(), TSStep() 2642 @*/ 2643 PetscErrorCode TSSetPreStep(TS ts, PetscErrorCode (*func)(TS)) 2644 { 2645 PetscFunctionBegin; 2646 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 2647 ts->prestep = func; 2648 PetscFunctionReturn(0); 2649 } 2650 2651 #undef __FUNCT__ 2652 #define __FUNCT__ "TSPreStep" 2653 /*@ 2654 TSPreStep - Runs the user-defined pre-step function. 2655 2656 Collective on TS 2657 2658 Input Parameters: 2659 . ts - The TS context obtained from TSCreate() 2660 2661 Notes: 2662 TSPreStep() is typically used within time stepping implementations, 2663 so most users would not generally call this routine themselves. 2664 2665 Level: developer 2666 2667 .keywords: TS, timestep 2668 .seealso: TSSetPreStep(), TSPreStage(), TSPostStage(), TSPostStep() 2669 @*/ 2670 PetscErrorCode TSPreStep(TS ts) 2671 { 2672 PetscErrorCode ierr; 2673 2674 PetscFunctionBegin; 2675 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2676 if (ts->prestep) { 2677 PetscStackCallStandard((*ts->prestep),(ts)); 2678 } 2679 PetscFunctionReturn(0); 2680 } 2681 2682 #undef __FUNCT__ 2683 #define __FUNCT__ "TSSetPreStage" 2684 /*@C 2685 TSSetPreStage - Sets the general-purpose function 2686 called once at the beginning of each stage. 2687 2688 Logically Collective on TS 2689 2690 Input Parameters: 2691 + ts - The TS context obtained from TSCreate() 2692 - func - The function 2693 2694 Calling sequence of func: 2695 . PetscErrorCode func(TS ts, PetscReal stagetime); 2696 2697 Level: intermediate 2698 2699 Note: 2700 There may be several stages per time step. If the solve for a given stage fails, the step may be rejected and retried. 2701 The time step number being computed can be queried using TSGetTimeStepNumber() and the total size of the step being 2702 attempted can be obtained using TSGetTimeStep(). The time at the start of the step is available via TSGetTime(). 2703 2704 .keywords: TS, timestep 2705 .seealso: TSSetPostStage(), TSSetPreStep(), TSSetPostStep(), TSGetApplicationContext() 2706 @*/ 2707 PetscErrorCode TSSetPreStage(TS ts, PetscErrorCode (*func)(TS,PetscReal)) 2708 { 2709 PetscFunctionBegin; 2710 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 2711 ts->prestage = func; 2712 PetscFunctionReturn(0); 2713 } 2714 2715 #undef __FUNCT__ 2716 #define __FUNCT__ "TSSetPostStage" 2717 /*@C 2718 TSSetPostStage - Sets the general-purpose function 2719 called once at the end of each stage. 2720 2721 Logically Collective on TS 2722 2723 Input Parameters: 2724 + ts - The TS context obtained from TSCreate() 2725 - func - The function 2726 2727 Calling sequence of func: 2728 . PetscErrorCode func(TS ts, PetscReal stagetime, PetscInt stageindex, Vec* Y); 2729 2730 Level: intermediate 2731 2732 Note: 2733 There may be several stages per time step. If the solve for a given stage fails, the step may be rejected and retried. 2734 The time step number being computed can be queried using TSGetTimeStepNumber() and the total size of the step being 2735 attempted can be obtained using TSGetTimeStep(). The time at the start of the step is available via TSGetTime(). 2736 2737 .keywords: TS, timestep 2738 .seealso: TSSetPreStage(), TSSetPreStep(), TSSetPostStep(), TSGetApplicationContext() 2739 @*/ 2740 PetscErrorCode TSSetPostStage(TS ts, PetscErrorCode (*func)(TS,PetscReal,PetscInt,Vec*)) 2741 { 2742 PetscFunctionBegin; 2743 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 2744 ts->poststage = func; 2745 PetscFunctionReturn(0); 2746 } 2747 2748 #undef __FUNCT__ 2749 #define __FUNCT__ "TSPreStage" 2750 /*@ 2751 TSPreStage - Runs the user-defined pre-stage function set using TSSetPreStage() 2752 2753 Collective on TS 2754 2755 Input Parameters: 2756 . ts - The TS context obtained from TSCreate() 2757 stagetime - The absolute time of the current stage 2758 2759 Notes: 2760 TSPreStage() is typically used within time stepping implementations, 2761 most users would not generally call this routine themselves. 2762 2763 Level: developer 2764 2765 .keywords: TS, timestep 2766 .seealso: TSPostStage(), TSSetPreStep(), TSPreStep(), TSPostStep() 2767 @*/ 2768 PetscErrorCode TSPreStage(TS ts, PetscReal stagetime) 2769 { 2770 PetscErrorCode ierr; 2771 2772 PetscFunctionBegin; 2773 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2774 if (ts->prestage) { 2775 PetscStackCallStandard((*ts->prestage),(ts,stagetime)); 2776 } 2777 PetscFunctionReturn(0); 2778 } 2779 2780 #undef __FUNCT__ 2781 #define __FUNCT__ "TSPostStage" 2782 /*@ 2783 TSPostStage - Runs the user-defined post-stage function set using TSSetPostStage() 2784 2785 Collective on TS 2786 2787 Input Parameters: 2788 . ts - The TS context obtained from TSCreate() 2789 stagetime - The absolute time of the current stage 2790 stageindex - Stage number 2791 Y - Array of vectors (of size = total number 2792 of stages) with the stage solutions 2793 2794 Notes: 2795 TSPostStage() is typically used within time stepping implementations, 2796 most users would not generally call this routine themselves. 2797 2798 Level: developer 2799 2800 .keywords: TS, timestep 2801 .seealso: TSPreStage(), TSSetPreStep(), TSPreStep(), TSPostStep() 2802 @*/ 2803 PetscErrorCode TSPostStage(TS ts, PetscReal stagetime, PetscInt stageindex, Vec *Y) 2804 { 2805 PetscErrorCode ierr; 2806 2807 PetscFunctionBegin; 2808 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2809 if (ts->poststage) { 2810 PetscStackCallStandard((*ts->poststage),(ts,stagetime,stageindex,Y)); 2811 } 2812 PetscFunctionReturn(0); 2813 } 2814 2815 #undef __FUNCT__ 2816 #define __FUNCT__ "TSSetPostStep" 2817 /*@C 2818 TSSetPostStep - Sets the general-purpose function 2819 called once at the end of each time step. 2820 2821 Logically Collective on TS 2822 2823 Input Parameters: 2824 + ts - The TS context obtained from TSCreate() 2825 - func - The function 2826 2827 Calling sequence of func: 2828 $ func (TS ts); 2829 2830 Level: intermediate 2831 2832 .keywords: TS, timestep 2833 .seealso: TSSetPreStep(), TSSetPreStage(), TSGetTimeStep(), TSGetTimeStepNumber(), TSGetTime() 2834 @*/ 2835 PetscErrorCode TSSetPostStep(TS ts, PetscErrorCode (*func)(TS)) 2836 { 2837 PetscFunctionBegin; 2838 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 2839 ts->poststep = func; 2840 PetscFunctionReturn(0); 2841 } 2842 2843 #undef __FUNCT__ 2844 #define __FUNCT__ "TSPostStep" 2845 /*@ 2846 TSPostStep - Runs the user-defined post-step function. 2847 2848 Collective on TS 2849 2850 Input Parameters: 2851 . ts - The TS context obtained from TSCreate() 2852 2853 Notes: 2854 TSPostStep() is typically used within time stepping implementations, 2855 so most users would not generally call this routine themselves. 2856 2857 Level: developer 2858 2859 .keywords: TS, timestep 2860 @*/ 2861 PetscErrorCode TSPostStep(TS ts) 2862 { 2863 PetscErrorCode ierr; 2864 2865 PetscFunctionBegin; 2866 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2867 if (ts->poststep) { 2868 PetscStackCallStandard((*ts->poststep),(ts)); 2869 } 2870 PetscFunctionReturn(0); 2871 } 2872 2873 /* ------------ Routines to set performance monitoring options ----------- */ 2874 2875 #undef __FUNCT__ 2876 #define __FUNCT__ "TSMonitorSet" 2877 /*@C 2878 TSMonitorSet - Sets an ADDITIONAL function that is to be used at every 2879 timestep to display the iteration's progress. 2880 2881 Logically Collective on TS 2882 2883 Input Parameters: 2884 + ts - the TS context obtained from TSCreate() 2885 . monitor - monitoring routine 2886 . mctx - [optional] user-defined context for private data for the 2887 monitor routine (use NULL if no context is desired) 2888 - monitordestroy - [optional] routine that frees monitor context 2889 (may be NULL) 2890 2891 Calling sequence of monitor: 2892 $ int monitor(TS ts,PetscInt steps,PetscReal time,Vec u,void *mctx) 2893 2894 + ts - the TS context 2895 . steps - iteration number (after the final time step the monitor routine is called with a step of -1, this is at the final time which may have 2896 been interpolated to) 2897 . time - current time 2898 . u - current iterate 2899 - mctx - [optional] monitoring context 2900 2901 Notes: 2902 This routine adds an additional monitor to the list of monitors that 2903 already has been loaded. 2904 2905 Fortran notes: Only a single monitor function can be set for each TS object 2906 2907 Level: intermediate 2908 2909 .keywords: TS, timestep, set, monitor 2910 2911 .seealso: TSMonitorDefault(), TSMonitorCancel() 2912 @*/ 2913 PetscErrorCode TSMonitorSet(TS ts,PetscErrorCode (*monitor)(TS,PetscInt,PetscReal,Vec,void*),void *mctx,PetscErrorCode (*mdestroy)(void**)) 2914 { 2915 PetscFunctionBegin; 2916 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2917 if (ts->numbermonitors >= MAXTSMONITORS) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Too many monitors set"); 2918 ts->monitor[ts->numbermonitors] = monitor; 2919 ts->monitordestroy[ts->numbermonitors] = mdestroy; 2920 ts->monitorcontext[ts->numbermonitors++] = (void*)mctx; 2921 PetscFunctionReturn(0); 2922 } 2923 2924 #undef __FUNCT__ 2925 #define __FUNCT__ "TSMonitorCancel" 2926 /*@C 2927 TSMonitorCancel - Clears all the monitors that have been set on a time-step object. 2928 2929 Logically Collective on TS 2930 2931 Input Parameters: 2932 . ts - the TS context obtained from TSCreate() 2933 2934 Notes: 2935 There is no way to remove a single, specific monitor. 2936 2937 Level: intermediate 2938 2939 .keywords: TS, timestep, set, monitor 2940 2941 .seealso: TSMonitorDefault(), TSMonitorSet() 2942 @*/ 2943 PetscErrorCode TSMonitorCancel(TS ts) 2944 { 2945 PetscErrorCode ierr; 2946 PetscInt i; 2947 2948 PetscFunctionBegin; 2949 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 2950 for (i=0; i<ts->numbermonitors; i++) { 2951 if (ts->monitordestroy[i]) { 2952 ierr = (*ts->monitordestroy[i])(&ts->monitorcontext[i]);CHKERRQ(ierr); 2953 } 2954 } 2955 ts->numbermonitors = 0; 2956 PetscFunctionReturn(0); 2957 } 2958 2959 #undef __FUNCT__ 2960 #define __FUNCT__ "TSMonitorDefault" 2961 /*@ 2962 TSMonitorDefault - Sets the Default monitor 2963 2964 Level: intermediate 2965 2966 .keywords: TS, set, monitor 2967 2968 .seealso: TSMonitorDefault(), TSMonitorSet() 2969 @*/ 2970 PetscErrorCode TSMonitorDefault(TS ts,PetscInt step,PetscReal ptime,Vec v,void *dummy) 2971 { 2972 PetscErrorCode ierr; 2973 PetscViewer viewer = dummy ? (PetscViewer) dummy : PETSC_VIEWER_STDOUT_(PetscObjectComm((PetscObject)ts)); 2974 2975 PetscFunctionBegin; 2976 ierr = PetscViewerASCIIAddTab(viewer,((PetscObject)ts)->tablevel);CHKERRQ(ierr); 2977 ierr = PetscViewerASCIIPrintf(viewer,"%D TS dt %g time %g%s",step,(double)ts->time_step,(double)ptime,ts->steprollback ? " (r)\n" : "\n");CHKERRQ(ierr); 2978 ierr = PetscViewerASCIISubtractTab(viewer,((PetscObject)ts)->tablevel);CHKERRQ(ierr); 2979 PetscFunctionReturn(0); 2980 } 2981 2982 #undef __FUNCT__ 2983 #define __FUNCT__ "TSSetRetainStages" 2984 /*@ 2985 TSSetRetainStages - Request that all stages in the upcoming step be stored so that interpolation will be available. 2986 2987 Logically Collective on TS 2988 2989 Input Argument: 2990 . ts - time stepping context 2991 2992 Output Argument: 2993 . flg - PETSC_TRUE or PETSC_FALSE 2994 2995 Level: intermediate 2996 2997 .keywords: TS, set 2998 2999 .seealso: TSInterpolate(), TSSetPostStep() 3000 @*/ 3001 PetscErrorCode TSSetRetainStages(TS ts,PetscBool flg) 3002 { 3003 PetscFunctionBegin; 3004 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3005 ts->retain_stages = flg; 3006 PetscFunctionReturn(0); 3007 } 3008 3009 #undef __FUNCT__ 3010 #define __FUNCT__ "TSInterpolate" 3011 /*@ 3012 TSInterpolate - Interpolate the solution computed during the previous step to an arbitrary location in the interval 3013 3014 Collective on TS 3015 3016 Input Argument: 3017 + ts - time stepping context 3018 - t - time to interpolate to 3019 3020 Output Argument: 3021 . U - state at given time 3022 3023 Notes: 3024 The user should call TSSetRetainStages() before taking a step in which interpolation will be requested. 3025 3026 Level: intermediate 3027 3028 Developer Notes: 3029 TSInterpolate() and the storing of previous steps/stages should be generalized to support delay differential equations and continuous adjoints. 3030 3031 .keywords: TS, set 3032 3033 .seealso: TSSetRetainStages(), TSSetPostStep() 3034 @*/ 3035 PetscErrorCode TSInterpolate(TS ts,PetscReal t,Vec U) 3036 { 3037 PetscErrorCode ierr; 3038 3039 PetscFunctionBegin; 3040 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3041 PetscValidHeaderSpecific(U,VEC_CLASSID,3); 3042 if (t < ts->ptime - ts->time_step_prev || t > ts->ptime) SETERRQ3(PetscObjectComm((PetscObject)ts),PETSC_ERR_ARG_OUTOFRANGE,"Requested time %g not in last time steps [%g,%g]",t,(double)(ts->ptime-ts->time_step_prev),(double)ts->ptime); 3043 if (!ts->ops->interpolate) SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_SUP,"%s does not provide interpolation",((PetscObject)ts)->type_name); 3044 ierr = (*ts->ops->interpolate)(ts,t,U);CHKERRQ(ierr); 3045 PetscFunctionReturn(0); 3046 } 3047 3048 #undef __FUNCT__ 3049 #define __FUNCT__ "TSStep" 3050 /*@ 3051 TSStep - Steps one time step 3052 3053 Collective on TS 3054 3055 Input Parameter: 3056 . ts - the TS context obtained from TSCreate() 3057 3058 Level: developer 3059 3060 Notes: 3061 The public interface for the ODE/DAE solvers is TSSolve(), you should almost for sure be using that routine and not this routine. 3062 3063 The hook set using TSSetPreStep() is called before each attempt to take the step. In general, the time step size may 3064 be changed due to adaptive error controller or solve failures. Note that steps may contain multiple stages. 3065 3066 This may over-step the final time provided in TSSetDuration() depending on the time-step used. TSSolve() interpolates to exactly the 3067 time provided in TSSetDuration(). One can use TSInterpolate() to determine an interpolated solution within the final timestep. 3068 3069 .keywords: TS, timestep, solve 3070 3071 .seealso: TSCreate(), TSSetUp(), TSDestroy(), TSSolve(), TSSetPreStep(), TSSetPreStage(), TSSetPostStage(), TSInterpolate() 3072 @*/ 3073 PetscErrorCode TSStep(TS ts) 3074 { 3075 DM dm; 3076 PetscErrorCode ierr; 3077 static PetscBool cite = PETSC_FALSE; 3078 3079 PetscFunctionBegin; 3080 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 3081 ierr = PetscCitationsRegister("@techreport{tspaper,\n" 3082 " title = {{PETSc/TS}: A Modern Scalable {DAE/ODE} Solver Library},\n" 3083 " author = {Shrirang Abhyankar and Jed Brown and Emil Constantinescu and Debojyoti Ghosh and Barry F. Smith},\n" 3084 " type = {Preprint},\n" 3085 " number = {ANL/MCS-P5061-0114},\n" 3086 " institution = {Argonne National Laboratory},\n" 3087 " year = {2014}\n}\n",&cite);CHKERRQ(ierr); 3088 3089 ierr = TSGetDM(ts, &dm);CHKERRQ(ierr); 3090 ierr = TSSetUp(ts);CHKERRQ(ierr); 3091 3092 ts->reason = TS_CONVERGED_ITERATING; 3093 ts->ptime_prev = ts->ptime; 3094 ierr = DMSetOutputSequenceNumber(dm, ts->steps, ts->ptime);CHKERRQ(ierr); 3095 3096 if (!ts->ops->step) SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_SUP,"TSStep not implemented for type '%s'",((PetscObject)ts)->type_name); 3097 ierr = PetscLogEventBegin(TS_Step,ts,0,0,0);CHKERRQ(ierr); 3098 ierr = (*ts->ops->step)(ts);CHKERRQ(ierr); 3099 ierr = PetscLogEventEnd(TS_Step,ts,0,0,0);CHKERRQ(ierr); 3100 3101 ts->time_step_prev = ts->ptime - ts->ptime_prev; 3102 ierr = DMSetOutputSequenceNumber(dm, ts->steps, ts->ptime);CHKERRQ(ierr); 3103 3104 if (ts->reason < 0) { 3105 if (ts->errorifstepfailed) { 3106 if (ts->reason == TS_DIVERGED_NONLINEAR_SOLVE) SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_NOT_CONVERGED,"TSStep has failed due to %s, increase -ts_max_snes_failures or make negative to attempt recovery",TSConvergedReasons[ts->reason]); 3107 else SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_NOT_CONVERGED,"TSStep has failed due to %s",TSConvergedReasons[ts->reason]); 3108 } 3109 } else if (!ts->reason) { 3110 if (ts->steps >= ts->max_steps) ts->reason = TS_CONVERGED_ITS; 3111 else if (ts->ptime >= ts->max_time) ts->reason = TS_CONVERGED_TIME; 3112 } 3113 ts->total_steps++; 3114 ts->steprollback = PETSC_FALSE; 3115 PetscFunctionReturn(0); 3116 } 3117 3118 #undef __FUNCT__ 3119 #define __FUNCT__ "TSAdjointStep" 3120 /*@ 3121 TSAdjointStep - Steps one time step backward in the adjoint run 3122 3123 Collective on TS 3124 3125 Input Parameter: 3126 . ts - the TS context obtained from TSCreate() 3127 3128 Level: intermediate 3129 3130 .keywords: TS, adjoint, step 3131 3132 .seealso: TSAdjointSetUp(), TSAdjointSolve() 3133 @*/ 3134 PetscErrorCode TSAdjointStep(TS ts) 3135 { 3136 DM dm; 3137 PetscErrorCode ierr; 3138 3139 PetscFunctionBegin; 3140 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 3141 ierr = TSGetDM(ts, &dm);CHKERRQ(ierr); 3142 ierr = TSAdjointSetUp(ts);CHKERRQ(ierr); 3143 3144 ts->reason = TS_CONVERGED_ITERATING; 3145 ts->ptime_prev = ts->ptime; 3146 ierr = DMSetOutputSequenceNumber(dm, ts->steps, ts->ptime);CHKERRQ(ierr); 3147 ierr = VecViewFromOptions(ts->vec_sol,(PetscObject)ts, "-ts_view_solution");CHKERRQ(ierr); 3148 3149 ierr = PetscLogEventBegin(TS_Step,ts,0,0,0);CHKERRQ(ierr); 3150 if (!ts->ops->adjointstep) SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_NOT_CONVERGED,"TSStep has failed because the adjoint of %s has not been implemented, try other time stepping methods for adjoint sensitivity analysis",((PetscObject)ts)->type_name); 3151 ierr = (*ts->ops->adjointstep)(ts);CHKERRQ(ierr); 3152 ierr = PetscLogEventEnd(TS_Step,ts,0,0,0);CHKERRQ(ierr); 3153 3154 ts->time_step_prev = ts->ptime - ts->ptime_prev; 3155 ierr = DMSetOutputSequenceNumber(dm, ts->steps, ts->ptime);CHKERRQ(ierr); 3156 3157 if (ts->reason < 0) { 3158 if (ts->errorifstepfailed) { 3159 if (ts->reason == TS_DIVERGED_NONLINEAR_SOLVE) { 3160 SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_NOT_CONVERGED,"TSStep has failed due to %s, increase -ts_max_snes_failures or make negative to attempt recovery",TSConvergedReasons[ts->reason]); 3161 } else if (ts->reason == TS_DIVERGED_STEP_REJECTED) { 3162 SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_NOT_CONVERGED,"TSStep has failed due to %s, increase -ts_max_reject or make negative to attempt recovery",TSConvergedReasons[ts->reason]); 3163 } else SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_NOT_CONVERGED,"TSStep has failed due to %s",TSConvergedReasons[ts->reason]); 3164 } 3165 } else if (!ts->reason) { 3166 if (ts->steps >= ts->adjoint_max_steps) ts->reason = TS_CONVERGED_ITS; 3167 else if (ts->ptime >= ts->max_time) ts->reason = TS_CONVERGED_TIME; 3168 } 3169 ts->total_steps--; 3170 PetscFunctionReturn(0); 3171 } 3172 3173 #undef __FUNCT__ 3174 #define __FUNCT__ "TSEvaluateStep" 3175 /*@ 3176 TSEvaluateStep - Evaluate the solution at the end of a time step with a given order of accuracy. 3177 3178 Collective on TS 3179 3180 Input Arguments: 3181 + ts - time stepping context 3182 . order - desired order of accuracy 3183 - done - whether the step was evaluated at this order (pass NULL to generate an error if not available) 3184 3185 Output Arguments: 3186 . U - state at the end of the current step 3187 3188 Level: advanced 3189 3190 Notes: 3191 This function cannot be called until all stages have been evaluated. 3192 It is normally called by adaptive controllers before a step has been accepted and may also be called by the user after TSStep() has returned. 3193 3194 .seealso: TSStep(), TSAdapt 3195 @*/ 3196 PetscErrorCode TSEvaluateStep(TS ts,PetscInt order,Vec U,PetscBool *done) 3197 { 3198 PetscErrorCode ierr; 3199 3200 PetscFunctionBegin; 3201 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3202 PetscValidType(ts,1); 3203 PetscValidHeaderSpecific(U,VEC_CLASSID,3); 3204 if (!ts->ops->evaluatestep) SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_SUP,"TSEvaluateStep not implemented for type '%s'",((PetscObject)ts)->type_name); 3205 ierr = (*ts->ops->evaluatestep)(ts,order,U,done);CHKERRQ(ierr); 3206 PetscFunctionReturn(0); 3207 } 3208 3209 3210 #undef __FUNCT__ 3211 #define __FUNCT__ "TSSolve" 3212 /*@ 3213 TSSolve - Steps the requested number of timesteps. 3214 3215 Collective on TS 3216 3217 Input Parameter: 3218 + ts - the TS context obtained from TSCreate() 3219 - u - the solution vector (can be null if TSSetSolution() was used, otherwise must contain the initial conditions) 3220 3221 Level: beginner 3222 3223 Notes: 3224 The final time returned by this function may be different from the time of the internally 3225 held state accessible by TSGetSolution() and TSGetTime() because the method may have 3226 stepped over the final time. 3227 3228 .keywords: TS, timestep, solve 3229 3230 .seealso: TSCreate(), TSSetSolution(), TSStep() 3231 @*/ 3232 PetscErrorCode TSSolve(TS ts,Vec u) 3233 { 3234 Vec solution; 3235 PetscErrorCode ierr; 3236 3237 PetscFunctionBegin; 3238 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3239 if (u) PetscValidHeaderSpecific(u,VEC_CLASSID,2); 3240 if (ts->exact_final_time == TS_EXACTFINALTIME_INTERPOLATE) { /* Need ts->vec_sol to be distinct so it is not overwritten when we interpolate at the end */ 3241 PetscValidHeaderSpecific(u,VEC_CLASSID,2); 3242 if (!ts->vec_sol || u == ts->vec_sol) { 3243 ierr = VecDuplicate(u,&solution);CHKERRQ(ierr); 3244 ierr = TSSetSolution(ts,solution);CHKERRQ(ierr); 3245 ierr = VecDestroy(&solution);CHKERRQ(ierr); /* grant ownership */ 3246 } 3247 ierr = VecCopy(u,ts->vec_sol);CHKERRQ(ierr); 3248 } else if (u) { 3249 ierr = TSSetSolution(ts,u);CHKERRQ(ierr); 3250 } 3251 ierr = TSSetUp(ts);CHKERRQ(ierr); 3252 /* reset time step and iteration counters */ 3253 ts->steps = 0; 3254 ts->ksp_its = 0; 3255 ts->snes_its = 0; 3256 ts->num_snes_failures = 0; 3257 ts->reject = 0; 3258 ts->reason = TS_CONVERGED_ITERATING; 3259 3260 ierr = TSViewFromOptions(ts,NULL,"-ts_view_pre");CHKERRQ(ierr); 3261 { 3262 DM dm; 3263 ierr = TSGetDM(ts, &dm);CHKERRQ(ierr); 3264 ierr = DMSetOutputSequenceNumber(dm, ts->steps, ts->ptime);CHKERRQ(ierr); 3265 } 3266 3267 if (ts->ops->solve) { /* This private interface is transitional and should be removed when all implementations are updated. */ 3268 ierr = (*ts->ops->solve)(ts);CHKERRQ(ierr); 3269 ierr = VecCopy(ts->vec_sol,u);CHKERRQ(ierr); 3270 ts->solvetime = ts->ptime; 3271 } else { 3272 /* steps the requested number of timesteps. */ 3273 if (ts->steps >= ts->max_steps) ts->reason = TS_CONVERGED_ITS; 3274 else if (ts->ptime >= ts->max_time) ts->reason = TS_CONVERGED_TIME; 3275 ierr = TSTrajectorySet(ts->trajectory,ts,ts->steps,ts->ptime,ts->vec_sol);CHKERRQ(ierr); 3276 if (ts->vec_costintegral) ts->costintegralfwd=PETSC_TRUE; 3277 if(ts->event) { 3278 ierr = TSEventMonitorInitialize(ts);CHKERRQ(ierr); 3279 } 3280 while (!ts->reason) { 3281 ierr = TSMonitor(ts,ts->steps,ts->ptime,ts->vec_sol);CHKERRQ(ierr); 3282 ierr = TSStep(ts);CHKERRQ(ierr); 3283 if (ts->event) { 3284 ierr = TSEventMonitor(ts);CHKERRQ(ierr); 3285 } 3286 if(!ts->steprollback) { 3287 ierr = TSTrajectorySet(ts->trajectory,ts,ts->steps,ts->ptime,ts->vec_sol);CHKERRQ(ierr); 3288 ierr = TSPostStep(ts);CHKERRQ(ierr); 3289 } 3290 } 3291 if (ts->exact_final_time == TS_EXACTFINALTIME_INTERPOLATE && ts->ptime > ts->max_time) { 3292 ierr = TSInterpolate(ts,ts->max_time,u);CHKERRQ(ierr); 3293 ts->solvetime = ts->max_time; 3294 solution = u; 3295 } else { 3296 if (u) {ierr = VecCopy(ts->vec_sol,u);CHKERRQ(ierr);} 3297 ts->solvetime = ts->ptime; 3298 solution = ts->vec_sol; 3299 } 3300 ierr = TSMonitor(ts,ts->steps,ts->solvetime,solution);CHKERRQ(ierr); 3301 ierr = VecViewFromOptions(solution,(PetscObject) ts,"-ts_view_solution");CHKERRQ(ierr); 3302 } 3303 3304 ierr = TSViewFromOptions(ts,NULL,"-ts_view");CHKERRQ(ierr); 3305 ierr = VecViewFromOptions(ts->vec_sol,NULL,"-ts_view_solution");CHKERRQ(ierr); 3306 ierr = PetscObjectSAWsBlock((PetscObject)ts);CHKERRQ(ierr); 3307 if (ts->adjoint_solve) { 3308 ierr = TSAdjointSolve(ts);CHKERRQ(ierr); 3309 } 3310 PetscFunctionReturn(0); 3311 } 3312 3313 #undef __FUNCT__ 3314 #define __FUNCT__ "TSAdjointSolve" 3315 /*@ 3316 TSAdjointSolve - Solves the discrete ajoint problem for an ODE/DAE 3317 3318 Collective on TS 3319 3320 Input Parameter: 3321 . ts - the TS context obtained from TSCreate() 3322 3323 Options Database: 3324 . -ts_adjoint_view_solution <viewerinfo> - views the first gradient with respect to the initial conditions 3325 3326 Level: intermediate 3327 3328 Notes: 3329 This must be called after a call to TSSolve() that solves the forward problem 3330 3331 By default this will integrate back to the initial time, one can use TSAdjointSetSteps() to step back to a later time 3332 3333 .keywords: TS, timestep, solve 3334 3335 .seealso: TSCreate(), TSSetCostGradients(), TSSetSolution(), TSAdjointStep() 3336 @*/ 3337 PetscErrorCode TSAdjointSolve(TS ts) 3338 { 3339 PetscErrorCode ierr; 3340 3341 PetscFunctionBegin; 3342 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3343 ierr = TSAdjointSetUp(ts);CHKERRQ(ierr); 3344 /* reset time step and iteration counters */ 3345 ts->steps = 0; 3346 ts->ksp_its = 0; 3347 ts->snes_its = 0; 3348 ts->num_snes_failures = 0; 3349 ts->reject = 0; 3350 ts->reason = TS_CONVERGED_ITERATING; 3351 3352 if (!ts->adjoint_max_steps) ts->adjoint_max_steps = ts->total_steps; 3353 3354 if (ts->steps >= ts->adjoint_max_steps) ts->reason = TS_CONVERGED_ITS; 3355 while (!ts->reason) { 3356 ierr = TSTrajectoryGet(ts->trajectory,ts,ts->adjoint_max_steps-ts->steps,&ts->ptime);CHKERRQ(ierr); 3357 ierr = TSMonitor(ts,ts->adjoint_max_steps-ts->steps,ts->ptime,ts->vec_sol);CHKERRQ(ierr); 3358 ierr = TSAdjointStep(ts);CHKERRQ(ierr); 3359 if (ts->event) { 3360 ierr = TSAdjointEventMonitor(ts);CHKERRQ(ierr); 3361 } 3362 } 3363 ts->solvetime = ts->ptime; 3364 ierr = VecViewFromOptions(ts->vecs_sensi[0],(PetscObject) ts, "-ts_adjoint_view_solution");CHKERRQ(ierr); 3365 PetscFunctionReturn(0); 3366 } 3367 3368 #undef __FUNCT__ 3369 #define __FUNCT__ "TSMonitor" 3370 /*@ 3371 TSMonitor - Runs all user-provided monitor routines set using TSMonitorSet() 3372 3373 Collective on TS 3374 3375 Input Parameters: 3376 + ts - time stepping context obtained from TSCreate() 3377 . step - step number that has just completed 3378 . ptime - model time of the state 3379 - u - state at the current model time 3380 3381 Notes: 3382 TSMonitor() is typically used within the time stepping implementations. 3383 Users might call this function when using the TSStep() interface instead of TSSolve(). 3384 3385 Level: advanced 3386 3387 .keywords: TS, timestep 3388 @*/ 3389 PetscErrorCode TSMonitor(TS ts,PetscInt step,PetscReal ptime,Vec u) 3390 { 3391 PetscErrorCode ierr; 3392 PetscInt i,n = ts->numbermonitors; 3393 3394 PetscFunctionBegin; 3395 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3396 PetscValidHeaderSpecific(u,VEC_CLASSID,4); 3397 ierr = VecLockPush(u);CHKERRQ(ierr); 3398 for (i=0; i<n; i++) { 3399 ierr = (*ts->monitor[i])(ts,step,ptime,u,ts->monitorcontext[i]);CHKERRQ(ierr); 3400 } 3401 ierr = VecLockPop(u);CHKERRQ(ierr); 3402 PetscFunctionReturn(0); 3403 } 3404 3405 /* ------------------------------------------------------------------------*/ 3406 #undef __FUNCT__ 3407 #define __FUNCT__ "TSMonitorLGCtxCreate" 3408 /*@C 3409 TSMonitorLGCtxCreate - Creates a line graph context for use with 3410 TS to monitor the solution process graphically in various ways 3411 3412 Collective on TS 3413 3414 Input Parameters: 3415 + host - the X display to open, or null for the local machine 3416 . label - the title to put in the title bar 3417 . x, y - the screen coordinates of the upper left coordinate of the window 3418 . m, n - the screen width and height in pixels 3419 - howoften - if positive then determines the frequency of the plotting, if -1 then only at the final time 3420 3421 Output Parameter: 3422 . ctx - the context 3423 3424 Options Database Key: 3425 + -ts_monitor_lg_timestep - automatically sets line graph monitor 3426 . -ts_monitor_lg_solution - 3427 . -ts_monitor_lg_error - 3428 . -ts_monitor_lg_ksp_iterations - 3429 . -ts_monitor_lg_snes_iterations - 3430 - -lg_use_markers <true,false> - mark the data points (at each time step) on the plot; default is true 3431 3432 Notes: 3433 Use TSMonitorLGCtxDestroy() to destroy. 3434 3435 Level: intermediate 3436 3437 .keywords: TS, monitor, line graph, residual, seealso 3438 3439 .seealso: TSMonitorLGTimeStep(), TSMonitorSet(), TSMonitorLGSolution(), TSMonitorLGError() 3440 3441 @*/ 3442 PetscErrorCode TSMonitorLGCtxCreate(MPI_Comm comm,const char host[],const char label[],int x,int y,int m,int n,PetscInt howoften,TSMonitorLGCtx *ctx) 3443 { 3444 PetscDraw win; 3445 PetscErrorCode ierr; 3446 3447 PetscFunctionBegin; 3448 ierr = PetscNew(ctx);CHKERRQ(ierr); 3449 ierr = PetscDrawCreate(comm,host,label,x,y,m,n,&win);CHKERRQ(ierr); 3450 ierr = PetscDrawSetFromOptions(win);CHKERRQ(ierr); 3451 ierr = PetscDrawLGCreate(win,1,&(*ctx)->lg);CHKERRQ(ierr); 3452 ierr = PetscLogObjectParent((PetscObject)(*ctx)->lg,(PetscObject)win);CHKERRQ(ierr); 3453 ierr = PetscDrawLGSetUseMarkers((*ctx)->lg,PETSC_TRUE);CHKERRQ(ierr); 3454 ierr = PetscDrawLGSetFromOptions((*ctx)->lg);CHKERRQ(ierr); 3455 (*ctx)->howoften = howoften; 3456 PetscFunctionReturn(0); 3457 } 3458 3459 #undef __FUNCT__ 3460 #define __FUNCT__ "TSMonitorLGTimeStep" 3461 PetscErrorCode TSMonitorLGTimeStep(TS ts,PetscInt step,PetscReal ptime,Vec v,void *monctx) 3462 { 3463 TSMonitorLGCtx ctx = (TSMonitorLGCtx) monctx; 3464 PetscReal x = ptime,y; 3465 PetscErrorCode ierr; 3466 3467 PetscFunctionBegin; 3468 if (!step) { 3469 PetscDrawAxis axis; 3470 ierr = PetscDrawLGGetAxis(ctx->lg,&axis);CHKERRQ(ierr); 3471 ierr = PetscDrawAxisSetLabels(axis,"Timestep as function of time","Time","Time step");CHKERRQ(ierr); 3472 ierr = PetscDrawLGReset(ctx->lg);CHKERRQ(ierr); 3473 } 3474 ierr = TSGetTimeStep(ts,&y);CHKERRQ(ierr); 3475 ierr = PetscDrawLGAddPoint(ctx->lg,&x,&y);CHKERRQ(ierr); 3476 if (((ctx->howoften > 0) && (!(step % ctx->howoften))) || ((ctx->howoften == -1) && ts->reason)) { 3477 ierr = PetscDrawLGDraw(ctx->lg);CHKERRQ(ierr); 3478 } 3479 PetscFunctionReturn(0); 3480 } 3481 3482 #undef __FUNCT__ 3483 #define __FUNCT__ "TSMonitorLGCtxDestroy" 3484 /*@C 3485 TSMonitorLGCtxDestroy - Destroys a line graph context that was created 3486 with TSMonitorLGCtxCreate(). 3487 3488 Collective on TSMonitorLGCtx 3489 3490 Input Parameter: 3491 . ctx - the monitor context 3492 3493 Level: intermediate 3494 3495 .keywords: TS, monitor, line graph, destroy 3496 3497 .seealso: TSMonitorLGCtxCreate(), TSMonitorSet(), TSMonitorLGTimeStep(); 3498 @*/ 3499 PetscErrorCode TSMonitorLGCtxDestroy(TSMonitorLGCtx *ctx) 3500 { 3501 PetscDraw draw; 3502 PetscErrorCode ierr; 3503 3504 PetscFunctionBegin; 3505 if ((*ctx)->transformdestroy) { 3506 ierr = ((*ctx)->transformdestroy)((*ctx)->transformctx);CHKERRQ(ierr); 3507 } 3508 ierr = PetscDrawLGGetDraw((*ctx)->lg,&draw);CHKERRQ(ierr); 3509 ierr = PetscDrawDestroy(&draw);CHKERRQ(ierr); 3510 ierr = PetscDrawLGDestroy(&(*ctx)->lg);CHKERRQ(ierr); 3511 ierr = PetscStrArrayDestroy(&(*ctx)->names);CHKERRQ(ierr); 3512 ierr = PetscStrArrayDestroy(&(*ctx)->displaynames);CHKERRQ(ierr); 3513 ierr = PetscFree((*ctx)->displayvariables);CHKERRQ(ierr); 3514 ierr = PetscFree((*ctx)->displayvalues);CHKERRQ(ierr); 3515 ierr = PetscFree(*ctx);CHKERRQ(ierr); 3516 PetscFunctionReturn(0); 3517 } 3518 3519 #undef __FUNCT__ 3520 #define __FUNCT__ "TSGetTime" 3521 /*@ 3522 TSGetTime - Gets the time of the most recently completed step. 3523 3524 Not Collective 3525 3526 Input Parameter: 3527 . ts - the TS context obtained from TSCreate() 3528 3529 Output Parameter: 3530 . t - the current time 3531 3532 Level: beginner 3533 3534 Note: 3535 When called during time step evaluation (e.g. during residual evaluation or via hooks set using TSSetPreStep(), 3536 TSSetPreStage(), TSSetPostStage(), or TSSetPostStep()), the time is the time at the start of the step being evaluated. 3537 3538 .seealso: TSSetInitialTimeStep(), TSGetTimeStep() 3539 3540 .keywords: TS, get, time 3541 @*/ 3542 PetscErrorCode TSGetTime(TS ts,PetscReal *t) 3543 { 3544 PetscFunctionBegin; 3545 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3546 PetscValidRealPointer(t,2); 3547 *t = ts->ptime; 3548 PetscFunctionReturn(0); 3549 } 3550 3551 #undef __FUNCT__ 3552 #define __FUNCT__ "TSGetPrevTime" 3553 /*@ 3554 TSGetPrevTime - Gets the starting time of the previously completed step. 3555 3556 Not Collective 3557 3558 Input Parameter: 3559 . ts - the TS context obtained from TSCreate() 3560 3561 Output Parameter: 3562 . t - the previous time 3563 3564 Level: beginner 3565 3566 .seealso: TSSetInitialTimeStep(), TSGetTimeStep() 3567 3568 .keywords: TS, get, time 3569 @*/ 3570 PetscErrorCode TSGetPrevTime(TS ts,PetscReal *t) 3571 { 3572 PetscFunctionBegin; 3573 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3574 PetscValidRealPointer(t,2); 3575 *t = ts->ptime_prev; 3576 PetscFunctionReturn(0); 3577 } 3578 3579 #undef __FUNCT__ 3580 #define __FUNCT__ "TSSetTime" 3581 /*@ 3582 TSSetTime - Allows one to reset the time. 3583 3584 Logically Collective on TS 3585 3586 Input Parameters: 3587 + ts - the TS context obtained from TSCreate() 3588 - time - the time 3589 3590 Level: intermediate 3591 3592 .seealso: TSGetTime(), TSSetDuration() 3593 3594 .keywords: TS, set, time 3595 @*/ 3596 PetscErrorCode TSSetTime(TS ts, PetscReal t) 3597 { 3598 PetscFunctionBegin; 3599 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3600 PetscValidLogicalCollectiveReal(ts,t,2); 3601 ts->ptime = t; 3602 PetscFunctionReturn(0); 3603 } 3604 3605 #undef __FUNCT__ 3606 #define __FUNCT__ "TSSetOptionsPrefix" 3607 /*@C 3608 TSSetOptionsPrefix - Sets the prefix used for searching for all 3609 TS options in the database. 3610 3611 Logically Collective on TS 3612 3613 Input Parameter: 3614 + ts - The TS context 3615 - prefix - The prefix to prepend to all option names 3616 3617 Notes: 3618 A hyphen (-) must NOT be given at the beginning of the prefix name. 3619 The first character of all runtime options is AUTOMATICALLY the 3620 hyphen. 3621 3622 Level: advanced 3623 3624 .keywords: TS, set, options, prefix, database 3625 3626 .seealso: TSSetFromOptions() 3627 3628 @*/ 3629 PetscErrorCode TSSetOptionsPrefix(TS ts,const char prefix[]) 3630 { 3631 PetscErrorCode ierr; 3632 SNES snes; 3633 3634 PetscFunctionBegin; 3635 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3636 ierr = PetscObjectSetOptionsPrefix((PetscObject)ts,prefix);CHKERRQ(ierr); 3637 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 3638 ierr = SNESSetOptionsPrefix(snes,prefix);CHKERRQ(ierr); 3639 PetscFunctionReturn(0); 3640 } 3641 3642 3643 #undef __FUNCT__ 3644 #define __FUNCT__ "TSAppendOptionsPrefix" 3645 /*@C 3646 TSAppendOptionsPrefix - Appends to the prefix used for searching for all 3647 TS options in the database. 3648 3649 Logically Collective on TS 3650 3651 Input Parameter: 3652 + ts - The TS context 3653 - prefix - The prefix to prepend to all option names 3654 3655 Notes: 3656 A hyphen (-) must NOT be given at the beginning of the prefix name. 3657 The first character of all runtime options is AUTOMATICALLY the 3658 hyphen. 3659 3660 Level: advanced 3661 3662 .keywords: TS, append, options, prefix, database 3663 3664 .seealso: TSGetOptionsPrefix() 3665 3666 @*/ 3667 PetscErrorCode TSAppendOptionsPrefix(TS ts,const char prefix[]) 3668 { 3669 PetscErrorCode ierr; 3670 SNES snes; 3671 3672 PetscFunctionBegin; 3673 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3674 ierr = PetscObjectAppendOptionsPrefix((PetscObject)ts,prefix);CHKERRQ(ierr); 3675 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 3676 ierr = SNESAppendOptionsPrefix(snes,prefix);CHKERRQ(ierr); 3677 PetscFunctionReturn(0); 3678 } 3679 3680 #undef __FUNCT__ 3681 #define __FUNCT__ "TSGetOptionsPrefix" 3682 /*@C 3683 TSGetOptionsPrefix - Sets the prefix used for searching for all 3684 TS options in the database. 3685 3686 Not Collective 3687 3688 Input Parameter: 3689 . ts - The TS context 3690 3691 Output Parameter: 3692 . prefix - A pointer to the prefix string used 3693 3694 Notes: On the fortran side, the user should pass in a string 'prifix' of 3695 sufficient length to hold the prefix. 3696 3697 Level: intermediate 3698 3699 .keywords: TS, get, options, prefix, database 3700 3701 .seealso: TSAppendOptionsPrefix() 3702 @*/ 3703 PetscErrorCode TSGetOptionsPrefix(TS ts,const char *prefix[]) 3704 { 3705 PetscErrorCode ierr; 3706 3707 PetscFunctionBegin; 3708 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 3709 PetscValidPointer(prefix,2); 3710 ierr = PetscObjectGetOptionsPrefix((PetscObject)ts,prefix);CHKERRQ(ierr); 3711 PetscFunctionReturn(0); 3712 } 3713 3714 #undef __FUNCT__ 3715 #define __FUNCT__ "TSGetRHSJacobian" 3716 /*@C 3717 TSGetRHSJacobian - Returns the Jacobian J at the present timestep. 3718 3719 Not Collective, but parallel objects are returned if TS is parallel 3720 3721 Input Parameter: 3722 . ts - The TS context obtained from TSCreate() 3723 3724 Output Parameters: 3725 + Amat - The (approximate) Jacobian J of G, where U_t = G(U,t) (or NULL) 3726 . Pmat - The matrix from which the preconditioner is constructed, usually the same as Amat (or NULL) 3727 . func - Function to compute the Jacobian of the RHS (or NULL) 3728 - ctx - User-defined context for Jacobian evaluation routine (or NULL) 3729 3730 Notes: You can pass in NULL for any return argument you do not need. 3731 3732 Level: intermediate 3733 3734 .seealso: TSGetTimeStep(), TSGetMatrices(), TSGetTime(), TSGetTimeStepNumber() 3735 3736 .keywords: TS, timestep, get, matrix, Jacobian 3737 @*/ 3738 PetscErrorCode TSGetRHSJacobian(TS ts,Mat *Amat,Mat *Pmat,TSRHSJacobian *func,void **ctx) 3739 { 3740 PetscErrorCode ierr; 3741 SNES snes; 3742 DM dm; 3743 3744 PetscFunctionBegin; 3745 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 3746 ierr = SNESGetJacobian(snes,Amat,Pmat,NULL,NULL);CHKERRQ(ierr); 3747 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 3748 ierr = DMTSGetRHSJacobian(dm,func,ctx);CHKERRQ(ierr); 3749 PetscFunctionReturn(0); 3750 } 3751 3752 #undef __FUNCT__ 3753 #define __FUNCT__ "TSGetIJacobian" 3754 /*@C 3755 TSGetIJacobian - Returns the implicit Jacobian at the present timestep. 3756 3757 Not Collective, but parallel objects are returned if TS is parallel 3758 3759 Input Parameter: 3760 . ts - The TS context obtained from TSCreate() 3761 3762 Output Parameters: 3763 + Amat - The (approximate) Jacobian of F(t,U,U_t) 3764 . Pmat - The matrix from which the preconditioner is constructed, often the same as Amat 3765 . f - The function to compute the matrices 3766 - ctx - User-defined context for Jacobian evaluation routine 3767 3768 Notes: You can pass in NULL for any return argument you do not need. 3769 3770 Level: advanced 3771 3772 .seealso: TSGetTimeStep(), TSGetRHSJacobian(), TSGetMatrices(), TSGetTime(), TSGetTimeStepNumber() 3773 3774 .keywords: TS, timestep, get, matrix, Jacobian 3775 @*/ 3776 PetscErrorCode TSGetIJacobian(TS ts,Mat *Amat,Mat *Pmat,TSIJacobian *f,void **ctx) 3777 { 3778 PetscErrorCode ierr; 3779 SNES snes; 3780 DM dm; 3781 3782 PetscFunctionBegin; 3783 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 3784 ierr = SNESSetUpMatrices(snes);CHKERRQ(ierr); 3785 ierr = SNESGetJacobian(snes,Amat,Pmat,NULL,NULL);CHKERRQ(ierr); 3786 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 3787 ierr = DMTSGetIJacobian(dm,f,ctx);CHKERRQ(ierr); 3788 PetscFunctionReturn(0); 3789 } 3790 3791 3792 #undef __FUNCT__ 3793 #define __FUNCT__ "TSMonitorDrawSolution" 3794 /*@C 3795 TSMonitorDrawSolution - Monitors progress of the TS solvers by calling 3796 VecView() for the solution at each timestep 3797 3798 Collective on TS 3799 3800 Input Parameters: 3801 + ts - the TS context 3802 . step - current time-step 3803 . ptime - current time 3804 - dummy - either a viewer or NULL 3805 3806 Options Database: 3807 . -ts_monitor_draw_solution_initial - show initial solution as well as current solution 3808 3809 Notes: the initial solution and current solution are not display with a common axis scaling so generally the option -ts_monitor_draw_solution_initial 3810 will look bad 3811 3812 Level: intermediate 3813 3814 .keywords: TS, vector, monitor, view 3815 3816 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView() 3817 @*/ 3818 PetscErrorCode TSMonitorDrawSolution(TS ts,PetscInt step,PetscReal ptime,Vec u,void *dummy) 3819 { 3820 PetscErrorCode ierr; 3821 TSMonitorDrawCtx ictx = (TSMonitorDrawCtx)dummy; 3822 PetscDraw draw; 3823 3824 PetscFunctionBegin; 3825 if (!step && ictx->showinitial) { 3826 if (!ictx->initialsolution) { 3827 ierr = VecDuplicate(u,&ictx->initialsolution);CHKERRQ(ierr); 3828 } 3829 ierr = VecCopy(u,ictx->initialsolution);CHKERRQ(ierr); 3830 } 3831 if (!(((ictx->howoften > 0) && (!(step % ictx->howoften))) || ((ictx->howoften == -1) && ts->reason))) PetscFunctionReturn(0); 3832 3833 if (ictx->showinitial) { 3834 PetscReal pause; 3835 ierr = PetscViewerDrawGetPause(ictx->viewer,&pause);CHKERRQ(ierr); 3836 ierr = PetscViewerDrawSetPause(ictx->viewer,0.0);CHKERRQ(ierr); 3837 ierr = VecView(ictx->initialsolution,ictx->viewer);CHKERRQ(ierr); 3838 ierr = PetscViewerDrawSetPause(ictx->viewer,pause);CHKERRQ(ierr); 3839 ierr = PetscViewerDrawSetHold(ictx->viewer,PETSC_TRUE);CHKERRQ(ierr); 3840 } 3841 ierr = VecView(u,ictx->viewer);CHKERRQ(ierr); 3842 if (ictx->showtimestepandtime) { 3843 PetscReal xl,yl,xr,yr,h; 3844 char time[32]; 3845 3846 ierr = PetscViewerDrawGetDraw(ictx->viewer,0,&draw);CHKERRQ(ierr); 3847 ierr = PetscSNPrintf(time,32,"Timestep %d Time %g",(int)step,(double)ptime);CHKERRQ(ierr); 3848 ierr = PetscDrawGetCoordinates(draw,&xl,&yl,&xr,&yr);CHKERRQ(ierr); 3849 h = yl + .95*(yr - yl); 3850 ierr = PetscDrawStringCentered(draw,.5*(xl+xr),h,PETSC_DRAW_BLACK,time);CHKERRQ(ierr); 3851 ierr = PetscDrawFlush(draw);CHKERRQ(ierr); 3852 } 3853 3854 if (ictx->showinitial) { 3855 ierr = PetscViewerDrawSetHold(ictx->viewer,PETSC_FALSE);CHKERRQ(ierr); 3856 } 3857 PetscFunctionReturn(0); 3858 } 3859 3860 #undef __FUNCT__ 3861 #define __FUNCT__ "TSMonitorDrawSolutionPhase" 3862 /*@C 3863 TSMonitorDrawSolutionPhase - Monitors progress of the TS solvers by plotting the solution as a phase diagram 3864 3865 Collective on TS 3866 3867 Input Parameters: 3868 + ts - the TS context 3869 . step - current time-step 3870 . ptime - current time 3871 - dummy - either a viewer or NULL 3872 3873 Level: intermediate 3874 3875 .keywords: TS, vector, monitor, view 3876 3877 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView() 3878 @*/ 3879 PetscErrorCode TSMonitorDrawSolutionPhase(TS ts,PetscInt step,PetscReal ptime,Vec u,void *dummy) 3880 { 3881 PetscErrorCode ierr; 3882 TSMonitorDrawCtx ictx = (TSMonitorDrawCtx)dummy; 3883 PetscDraw draw; 3884 MPI_Comm comm; 3885 PetscInt n; 3886 PetscMPIInt size; 3887 PetscReal xl,yl,xr,yr,h; 3888 char time[32]; 3889 const PetscScalar *U; 3890 3891 PetscFunctionBegin; 3892 ierr = PetscObjectGetComm((PetscObject)ts,&comm);CHKERRQ(ierr); 3893 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 3894 if (size != 1) SETERRQ(comm,PETSC_ERR_SUP,"Only allowed for sequential runs"); 3895 ierr = VecGetSize(u,&n);CHKERRQ(ierr); 3896 if (n != 2) SETERRQ(comm,PETSC_ERR_SUP,"Only for ODEs with two unknowns"); 3897 3898 ierr = PetscViewerDrawGetDraw(ictx->viewer,0,&draw);CHKERRQ(ierr); 3899 3900 ierr = VecGetArrayRead(u,&U);CHKERRQ(ierr); 3901 ierr = PetscDrawAxisGetLimits(ictx->axis,&xl,&xr,&yl,&yr);CHKERRQ(ierr); 3902 if ((PetscRealPart(U[0]) < xl) || (PetscRealPart(U[1]) < yl) || (PetscRealPart(U[0]) > xr) || (PetscRealPart(U[1]) > yr)) { 3903 ierr = VecRestoreArrayRead(u,&U);CHKERRQ(ierr); 3904 PetscFunctionReturn(0); 3905 } 3906 if (!step) ictx->color++; 3907 ierr = PetscDrawPoint(draw,PetscRealPart(U[0]),PetscRealPart(U[1]),ictx->color);CHKERRQ(ierr); 3908 ierr = VecRestoreArrayRead(u,&U);CHKERRQ(ierr); 3909 3910 if (ictx->showtimestepandtime) { 3911 ierr = PetscDrawGetCoordinates(draw,&xl,&yl,&xr,&yr);CHKERRQ(ierr); 3912 ierr = PetscSNPrintf(time,32,"Timestep %d Time %g",(int)step,(double)ptime);CHKERRQ(ierr); 3913 h = yl + .95*(yr - yl); 3914 ierr = PetscDrawStringCentered(draw,.5*(xl+xr),h,PETSC_DRAW_BLACK,time);CHKERRQ(ierr); 3915 } 3916 ierr = PetscDrawFlush(draw);CHKERRQ(ierr); 3917 PetscFunctionReturn(0); 3918 } 3919 3920 3921 #undef __FUNCT__ 3922 #define __FUNCT__ "TSMonitorDrawCtxDestroy" 3923 /*@C 3924 TSMonitorDrawCtxDestroy - Destroys the monitor context for TSMonitorDrawSolution() 3925 3926 Collective on TS 3927 3928 Input Parameters: 3929 . ctx - the monitor context 3930 3931 Level: intermediate 3932 3933 .keywords: TS, vector, monitor, view 3934 3935 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorDrawSolution(), TSMonitorDrawError() 3936 @*/ 3937 PetscErrorCode TSMonitorDrawCtxDestroy(TSMonitorDrawCtx *ictx) 3938 { 3939 PetscErrorCode ierr; 3940 3941 PetscFunctionBegin; 3942 ierr = PetscDrawAxisDestroy(&(*ictx)->axis);CHKERRQ(ierr); 3943 ierr = PetscViewerDestroy(&(*ictx)->viewer);CHKERRQ(ierr); 3944 ierr = VecDestroy(&(*ictx)->initialsolution);CHKERRQ(ierr); 3945 ierr = PetscFree(*ictx);CHKERRQ(ierr); 3946 PetscFunctionReturn(0); 3947 } 3948 3949 #undef __FUNCT__ 3950 #define __FUNCT__ "TSMonitorDrawCtxCreate" 3951 /*@C 3952 TSMonitorDrawCtxCreate - Creates the monitor context for TSMonitorDrawCtx 3953 3954 Collective on TS 3955 3956 Input Parameter: 3957 . ts - time-step context 3958 3959 Output Patameter: 3960 . ctx - the monitor context 3961 3962 Options Database: 3963 . -ts_monitor_draw_solution_initial - show initial solution as well as current solution 3964 3965 Level: intermediate 3966 3967 .keywords: TS, vector, monitor, view 3968 3969 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorDrawCtx() 3970 @*/ 3971 PetscErrorCode TSMonitorDrawCtxCreate(MPI_Comm comm,const char host[],const char label[],int x,int y,int m,int n,PetscInt howoften,TSMonitorDrawCtx *ctx) 3972 { 3973 PetscErrorCode ierr; 3974 3975 PetscFunctionBegin; 3976 ierr = PetscNew(ctx);CHKERRQ(ierr); 3977 ierr = PetscViewerDrawOpen(comm,host,label,x,y,m,n,&(*ctx)->viewer);CHKERRQ(ierr); 3978 ierr = PetscViewerSetFromOptions((*ctx)->viewer);CHKERRQ(ierr); 3979 3980 (*ctx)->howoften = howoften; 3981 (*ctx)->showinitial = PETSC_FALSE; 3982 ierr = PetscOptionsGetBool(NULL,"-ts_monitor_draw_solution_initial",&(*ctx)->showinitial,NULL);CHKERRQ(ierr); 3983 3984 (*ctx)->showtimestepandtime = PETSC_FALSE; 3985 ierr = PetscOptionsGetBool(NULL,"-ts_monitor_draw_solution_show_time",&(*ctx)->showtimestepandtime,NULL);CHKERRQ(ierr); 3986 (*ctx)->color = PETSC_DRAW_WHITE; 3987 PetscFunctionReturn(0); 3988 } 3989 3990 #undef __FUNCT__ 3991 #define __FUNCT__ "TSMonitorDrawError" 3992 /*@C 3993 TSMonitorDrawError - Monitors progress of the TS solvers by calling 3994 VecView() for the error at each timestep 3995 3996 Collective on TS 3997 3998 Input Parameters: 3999 + ts - the TS context 4000 . step - current time-step 4001 . ptime - current time 4002 - dummy - either a viewer or NULL 4003 4004 Level: intermediate 4005 4006 .keywords: TS, vector, monitor, view 4007 4008 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView() 4009 @*/ 4010 PetscErrorCode TSMonitorDrawError(TS ts,PetscInt step,PetscReal ptime,Vec u,void *dummy) 4011 { 4012 PetscErrorCode ierr; 4013 TSMonitorDrawCtx ctx = (TSMonitorDrawCtx)dummy; 4014 PetscViewer viewer = ctx->viewer; 4015 Vec work; 4016 4017 PetscFunctionBegin; 4018 if (!(((ctx->howoften > 0) && (!(step % ctx->howoften))) || ((ctx->howoften == -1) && ts->reason))) PetscFunctionReturn(0); 4019 ierr = VecDuplicate(u,&work);CHKERRQ(ierr); 4020 ierr = TSComputeSolutionFunction(ts,ptime,work);CHKERRQ(ierr); 4021 ierr = VecAXPY(work,-1.0,u);CHKERRQ(ierr); 4022 ierr = VecView(work,viewer);CHKERRQ(ierr); 4023 ierr = VecDestroy(&work);CHKERRQ(ierr); 4024 PetscFunctionReturn(0); 4025 } 4026 4027 #include <petsc/private/dmimpl.h> 4028 #undef __FUNCT__ 4029 #define __FUNCT__ "TSSetDM" 4030 /*@ 4031 TSSetDM - Sets the DM that may be used by some preconditioners 4032 4033 Logically Collective on TS and DM 4034 4035 Input Parameters: 4036 + ts - the preconditioner context 4037 - dm - the dm 4038 4039 Level: intermediate 4040 4041 4042 .seealso: TSGetDM(), SNESSetDM(), SNESGetDM() 4043 @*/ 4044 PetscErrorCode TSSetDM(TS ts,DM dm) 4045 { 4046 PetscErrorCode ierr; 4047 SNES snes; 4048 DMTS tsdm; 4049 4050 PetscFunctionBegin; 4051 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4052 ierr = PetscObjectReference((PetscObject)dm);CHKERRQ(ierr); 4053 if (ts->dm) { /* Move the DMTS context over to the new DM unless the new DM already has one */ 4054 if (ts->dm->dmts && !dm->dmts) { 4055 ierr = DMCopyDMTS(ts->dm,dm);CHKERRQ(ierr); 4056 ierr = DMGetDMTS(ts->dm,&tsdm);CHKERRQ(ierr); 4057 if (tsdm->originaldm == ts->dm) { /* Grant write privileges to the replacement DM */ 4058 tsdm->originaldm = dm; 4059 } 4060 } 4061 ierr = DMDestroy(&ts->dm);CHKERRQ(ierr); 4062 } 4063 ts->dm = dm; 4064 4065 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 4066 ierr = SNESSetDM(snes,dm);CHKERRQ(ierr); 4067 PetscFunctionReturn(0); 4068 } 4069 4070 #undef __FUNCT__ 4071 #define __FUNCT__ "TSGetDM" 4072 /*@ 4073 TSGetDM - Gets the DM that may be used by some preconditioners 4074 4075 Not Collective 4076 4077 Input Parameter: 4078 . ts - the preconditioner context 4079 4080 Output Parameter: 4081 . dm - the dm 4082 4083 Level: intermediate 4084 4085 4086 .seealso: TSSetDM(), SNESSetDM(), SNESGetDM() 4087 @*/ 4088 PetscErrorCode TSGetDM(TS ts,DM *dm) 4089 { 4090 PetscErrorCode ierr; 4091 4092 PetscFunctionBegin; 4093 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4094 if (!ts->dm) { 4095 ierr = DMShellCreate(PetscObjectComm((PetscObject)ts),&ts->dm);CHKERRQ(ierr); 4096 if (ts->snes) {ierr = SNESSetDM(ts->snes,ts->dm);CHKERRQ(ierr);} 4097 } 4098 *dm = ts->dm; 4099 PetscFunctionReturn(0); 4100 } 4101 4102 #undef __FUNCT__ 4103 #define __FUNCT__ "SNESTSFormFunction" 4104 /*@ 4105 SNESTSFormFunction - Function to evaluate nonlinear residual 4106 4107 Logically Collective on SNES 4108 4109 Input Parameter: 4110 + snes - nonlinear solver 4111 . U - the current state at which to evaluate the residual 4112 - ctx - user context, must be a TS 4113 4114 Output Parameter: 4115 . F - the nonlinear residual 4116 4117 Notes: 4118 This function is not normally called by users and is automatically registered with the SNES used by TS. 4119 It is most frequently passed to MatFDColoringSetFunction(). 4120 4121 Level: advanced 4122 4123 .seealso: SNESSetFunction(), MatFDColoringSetFunction() 4124 @*/ 4125 PetscErrorCode SNESTSFormFunction(SNES snes,Vec U,Vec F,void *ctx) 4126 { 4127 TS ts = (TS)ctx; 4128 PetscErrorCode ierr; 4129 4130 PetscFunctionBegin; 4131 PetscValidHeaderSpecific(snes,SNES_CLASSID,1); 4132 PetscValidHeaderSpecific(U,VEC_CLASSID,2); 4133 PetscValidHeaderSpecific(F,VEC_CLASSID,3); 4134 PetscValidHeaderSpecific(ts,TS_CLASSID,4); 4135 ierr = (ts->ops->snesfunction)(snes,U,F,ts);CHKERRQ(ierr); 4136 PetscFunctionReturn(0); 4137 } 4138 4139 #undef __FUNCT__ 4140 #define __FUNCT__ "SNESTSFormJacobian" 4141 /*@ 4142 SNESTSFormJacobian - Function to evaluate the Jacobian 4143 4144 Collective on SNES 4145 4146 Input Parameter: 4147 + snes - nonlinear solver 4148 . U - the current state at which to evaluate the residual 4149 - ctx - user context, must be a TS 4150 4151 Output Parameter: 4152 + A - the Jacobian 4153 . B - the preconditioning matrix (may be the same as A) 4154 - flag - indicates any structure change in the matrix 4155 4156 Notes: 4157 This function is not normally called by users and is automatically registered with the SNES used by TS. 4158 4159 Level: developer 4160 4161 .seealso: SNESSetJacobian() 4162 @*/ 4163 PetscErrorCode SNESTSFormJacobian(SNES snes,Vec U,Mat A,Mat B,void *ctx) 4164 { 4165 TS ts = (TS)ctx; 4166 PetscErrorCode ierr; 4167 4168 PetscFunctionBegin; 4169 PetscValidHeaderSpecific(snes,SNES_CLASSID,1); 4170 PetscValidHeaderSpecific(U,VEC_CLASSID,2); 4171 PetscValidPointer(A,3); 4172 PetscValidHeaderSpecific(A,MAT_CLASSID,3); 4173 PetscValidPointer(B,4); 4174 PetscValidHeaderSpecific(B,MAT_CLASSID,4); 4175 PetscValidHeaderSpecific(ts,TS_CLASSID,6); 4176 ierr = (ts->ops->snesjacobian)(snes,U,A,B,ts);CHKERRQ(ierr); 4177 PetscFunctionReturn(0); 4178 } 4179 4180 #undef __FUNCT__ 4181 #define __FUNCT__ "TSComputeRHSFunctionLinear" 4182 /*@C 4183 TSComputeRHSFunctionLinear - Evaluate the right hand side via the user-provided Jacobian, for linear problems only 4184 4185 Collective on TS 4186 4187 Input Arguments: 4188 + ts - time stepping context 4189 . t - time at which to evaluate 4190 . U - state at which to evaluate 4191 - ctx - context 4192 4193 Output Arguments: 4194 . F - right hand side 4195 4196 Level: intermediate 4197 4198 Notes: 4199 This function is intended to be passed to TSSetRHSFunction() to evaluate the right hand side for linear problems. 4200 The matrix (and optionally the evaluation context) should be passed to TSSetRHSJacobian(). 4201 4202 .seealso: TSSetRHSFunction(), TSSetRHSJacobian(), TSComputeRHSJacobianConstant() 4203 @*/ 4204 PetscErrorCode TSComputeRHSFunctionLinear(TS ts,PetscReal t,Vec U,Vec F,void *ctx) 4205 { 4206 PetscErrorCode ierr; 4207 Mat Arhs,Brhs; 4208 4209 PetscFunctionBegin; 4210 ierr = TSGetRHSMats_Private(ts,&Arhs,&Brhs);CHKERRQ(ierr); 4211 ierr = TSComputeRHSJacobian(ts,t,U,Arhs,Brhs);CHKERRQ(ierr); 4212 ierr = MatMult(Arhs,U,F);CHKERRQ(ierr); 4213 PetscFunctionReturn(0); 4214 } 4215 4216 #undef __FUNCT__ 4217 #define __FUNCT__ "TSComputeRHSJacobianConstant" 4218 /*@C 4219 TSComputeRHSJacobianConstant - Reuses a Jacobian that is time-independent. 4220 4221 Collective on TS 4222 4223 Input Arguments: 4224 + ts - time stepping context 4225 . t - time at which to evaluate 4226 . U - state at which to evaluate 4227 - ctx - context 4228 4229 Output Arguments: 4230 + A - pointer to operator 4231 . B - pointer to preconditioning matrix 4232 - flg - matrix structure flag 4233 4234 Level: intermediate 4235 4236 Notes: 4237 This function is intended to be passed to TSSetRHSJacobian() to evaluate the Jacobian for linear time-independent problems. 4238 4239 .seealso: TSSetRHSFunction(), TSSetRHSJacobian(), TSComputeRHSFunctionLinear() 4240 @*/ 4241 PetscErrorCode TSComputeRHSJacobianConstant(TS ts,PetscReal t,Vec U,Mat A,Mat B,void *ctx) 4242 { 4243 PetscFunctionBegin; 4244 PetscFunctionReturn(0); 4245 } 4246 4247 #undef __FUNCT__ 4248 #define __FUNCT__ "TSComputeIFunctionLinear" 4249 /*@C 4250 TSComputeIFunctionLinear - Evaluate the left hand side via the user-provided Jacobian, for linear problems only 4251 4252 Collective on TS 4253 4254 Input Arguments: 4255 + ts - time stepping context 4256 . t - time at which to evaluate 4257 . U - state at which to evaluate 4258 . Udot - time derivative of state vector 4259 - ctx - context 4260 4261 Output Arguments: 4262 . F - left hand side 4263 4264 Level: intermediate 4265 4266 Notes: 4267 The assumption here is that the left hand side is of the form A*Udot (and not A*Udot + B*U). For other cases, the 4268 user is required to write their own TSComputeIFunction. 4269 This function is intended to be passed to TSSetIFunction() to evaluate the left hand side for linear problems. 4270 The matrix (and optionally the evaluation context) should be passed to TSSetIJacobian(). 4271 4272 .seealso: TSSetIFunction(), TSSetIJacobian(), TSComputeIJacobianConstant() 4273 @*/ 4274 PetscErrorCode TSComputeIFunctionLinear(TS ts,PetscReal t,Vec U,Vec Udot,Vec F,void *ctx) 4275 { 4276 PetscErrorCode ierr; 4277 Mat A,B; 4278 4279 PetscFunctionBegin; 4280 ierr = TSGetIJacobian(ts,&A,&B,NULL,NULL);CHKERRQ(ierr); 4281 ierr = TSComputeIJacobian(ts,t,U,Udot,1.0,A,B,PETSC_TRUE);CHKERRQ(ierr); 4282 ierr = MatMult(A,Udot,F);CHKERRQ(ierr); 4283 PetscFunctionReturn(0); 4284 } 4285 4286 #undef __FUNCT__ 4287 #define __FUNCT__ "TSComputeIJacobianConstant" 4288 /*@C 4289 TSComputeIJacobianConstant - Reuses a time-independent for a semi-implicit DAE or ODE 4290 4291 Collective on TS 4292 4293 Input Arguments: 4294 + ts - time stepping context 4295 . t - time at which to evaluate 4296 . U - state at which to evaluate 4297 . Udot - time derivative of state vector 4298 . shift - shift to apply 4299 - ctx - context 4300 4301 Output Arguments: 4302 + A - pointer to operator 4303 . B - pointer to preconditioning matrix 4304 - flg - matrix structure flag 4305 4306 Level: advanced 4307 4308 Notes: 4309 This function is intended to be passed to TSSetIJacobian() to evaluate the Jacobian for linear time-independent problems. 4310 4311 It is only appropriate for problems of the form 4312 4313 $ M Udot = F(U,t) 4314 4315 where M is constant and F is non-stiff. The user must pass M to TSSetIJacobian(). The current implementation only 4316 works with IMEX time integration methods such as TSROSW and TSARKIMEX, since there is no support for de-constructing 4317 an implicit operator of the form 4318 4319 $ shift*M + J 4320 4321 where J is the Jacobian of -F(U). Support may be added in a future version of PETSc, but for now, the user must store 4322 a copy of M or reassemble it when requested. 4323 4324 .seealso: TSSetIFunction(), TSSetIJacobian(), TSComputeIFunctionLinear() 4325 @*/ 4326 PetscErrorCode TSComputeIJacobianConstant(TS ts,PetscReal t,Vec U,Vec Udot,PetscReal shift,Mat A,Mat B,void *ctx) 4327 { 4328 PetscErrorCode ierr; 4329 4330 PetscFunctionBegin; 4331 ierr = MatScale(A, shift / ts->ijacobian.shift);CHKERRQ(ierr); 4332 ts->ijacobian.shift = shift; 4333 PetscFunctionReturn(0); 4334 } 4335 4336 #undef __FUNCT__ 4337 #define __FUNCT__ "TSGetEquationType" 4338 /*@ 4339 TSGetEquationType - Gets the type of the equation that TS is solving. 4340 4341 Not Collective 4342 4343 Input Parameter: 4344 . ts - the TS context 4345 4346 Output Parameter: 4347 . equation_type - see TSEquationType 4348 4349 Level: beginner 4350 4351 .keywords: TS, equation type 4352 4353 .seealso: TSSetEquationType(), TSEquationType 4354 @*/ 4355 PetscErrorCode TSGetEquationType(TS ts,TSEquationType *equation_type) 4356 { 4357 PetscFunctionBegin; 4358 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4359 PetscValidPointer(equation_type,2); 4360 *equation_type = ts->equation_type; 4361 PetscFunctionReturn(0); 4362 } 4363 4364 #undef __FUNCT__ 4365 #define __FUNCT__ "TSSetEquationType" 4366 /*@ 4367 TSSetEquationType - Sets the type of the equation that TS is solving. 4368 4369 Not Collective 4370 4371 Input Parameter: 4372 + ts - the TS context 4373 - equation_type - see TSEquationType 4374 4375 Level: advanced 4376 4377 .keywords: TS, equation type 4378 4379 .seealso: TSGetEquationType(), TSEquationType 4380 @*/ 4381 PetscErrorCode TSSetEquationType(TS ts,TSEquationType equation_type) 4382 { 4383 PetscFunctionBegin; 4384 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4385 ts->equation_type = equation_type; 4386 PetscFunctionReturn(0); 4387 } 4388 4389 #undef __FUNCT__ 4390 #define __FUNCT__ "TSGetConvergedReason" 4391 /*@ 4392 TSGetConvergedReason - Gets the reason the TS iteration was stopped. 4393 4394 Not Collective 4395 4396 Input Parameter: 4397 . ts - the TS context 4398 4399 Output Parameter: 4400 . reason - negative value indicates diverged, positive value converged, see TSConvergedReason or the 4401 manual pages for the individual convergence tests for complete lists 4402 4403 Level: beginner 4404 4405 Notes: 4406 Can only be called after the call to TSSolve() is complete. 4407 4408 .keywords: TS, nonlinear, set, convergence, test 4409 4410 .seealso: TSSetConvergenceTest(), TSConvergedReason 4411 @*/ 4412 PetscErrorCode TSGetConvergedReason(TS ts,TSConvergedReason *reason) 4413 { 4414 PetscFunctionBegin; 4415 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4416 PetscValidPointer(reason,2); 4417 *reason = ts->reason; 4418 PetscFunctionReturn(0); 4419 } 4420 4421 #undef __FUNCT__ 4422 #define __FUNCT__ "TSSetConvergedReason" 4423 /*@ 4424 TSSetConvergedReason - Sets the reason for handling the convergence of TSSolve. 4425 4426 Not Collective 4427 4428 Input Parameter: 4429 + ts - the TS context 4430 . reason - negative value indicates diverged, positive value converged, see TSConvergedReason or the 4431 manual pages for the individual convergence tests for complete lists 4432 4433 Level: advanced 4434 4435 Notes: 4436 Can only be called during TSSolve() is active. 4437 4438 .keywords: TS, nonlinear, set, convergence, test 4439 4440 .seealso: TSConvergedReason 4441 @*/ 4442 PetscErrorCode TSSetConvergedReason(TS ts,TSConvergedReason reason) 4443 { 4444 PetscFunctionBegin; 4445 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4446 ts->reason = reason; 4447 PetscFunctionReturn(0); 4448 } 4449 4450 #undef __FUNCT__ 4451 #define __FUNCT__ "TSGetSolveTime" 4452 /*@ 4453 TSGetSolveTime - Gets the time after a call to TSSolve() 4454 4455 Not Collective 4456 4457 Input Parameter: 4458 . ts - the TS context 4459 4460 Output Parameter: 4461 . ftime - the final time. This time should correspond to the final time set with TSSetDuration() 4462 4463 Level: beginner 4464 4465 Notes: 4466 Can only be called after the call to TSSolve() is complete. 4467 4468 .keywords: TS, nonlinear, set, convergence, test 4469 4470 .seealso: TSSetConvergenceTest(), TSConvergedReason 4471 @*/ 4472 PetscErrorCode TSGetSolveTime(TS ts,PetscReal *ftime) 4473 { 4474 PetscFunctionBegin; 4475 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4476 PetscValidPointer(ftime,2); 4477 *ftime = ts->solvetime; 4478 PetscFunctionReturn(0); 4479 } 4480 4481 #undef __FUNCT__ 4482 #define __FUNCT__ "TSGetTotalSteps" 4483 /*@ 4484 TSGetTotalSteps - Gets the total number of steps done since the last call to TSSetUp() or TSCreate() 4485 4486 Not Collective 4487 4488 Input Parameter: 4489 . ts - the TS context 4490 4491 Output Parameter: 4492 . steps - the number of steps 4493 4494 Level: beginner 4495 4496 Notes: 4497 Includes the number of steps for all calls to TSSolve() since TSSetUp() was called 4498 4499 .keywords: TS, nonlinear, set, convergence, test 4500 4501 .seealso: TSSetConvergenceTest(), TSConvergedReason 4502 @*/ 4503 PetscErrorCode TSGetTotalSteps(TS ts,PetscInt *steps) 4504 { 4505 PetscFunctionBegin; 4506 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4507 PetscValidPointer(steps,2); 4508 *steps = ts->total_steps; 4509 PetscFunctionReturn(0); 4510 } 4511 4512 #undef __FUNCT__ 4513 #define __FUNCT__ "TSGetSNESIterations" 4514 /*@ 4515 TSGetSNESIterations - Gets the total number of nonlinear iterations 4516 used by the time integrator. 4517 4518 Not Collective 4519 4520 Input Parameter: 4521 . ts - TS context 4522 4523 Output Parameter: 4524 . nits - number of nonlinear iterations 4525 4526 Notes: 4527 This counter is reset to zero for each successive call to TSSolve(). 4528 4529 Level: intermediate 4530 4531 .keywords: TS, get, number, nonlinear, iterations 4532 4533 .seealso: TSGetKSPIterations() 4534 @*/ 4535 PetscErrorCode TSGetSNESIterations(TS ts,PetscInt *nits) 4536 { 4537 PetscFunctionBegin; 4538 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4539 PetscValidIntPointer(nits,2); 4540 *nits = ts->snes_its; 4541 PetscFunctionReturn(0); 4542 } 4543 4544 #undef __FUNCT__ 4545 #define __FUNCT__ "TSGetKSPIterations" 4546 /*@ 4547 TSGetKSPIterations - Gets the total number of linear iterations 4548 used by the time integrator. 4549 4550 Not Collective 4551 4552 Input Parameter: 4553 . ts - TS context 4554 4555 Output Parameter: 4556 . lits - number of linear iterations 4557 4558 Notes: 4559 This counter is reset to zero for each successive call to TSSolve(). 4560 4561 Level: intermediate 4562 4563 .keywords: TS, get, number, linear, iterations 4564 4565 .seealso: TSGetSNESIterations(), SNESGetKSPIterations() 4566 @*/ 4567 PetscErrorCode TSGetKSPIterations(TS ts,PetscInt *lits) 4568 { 4569 PetscFunctionBegin; 4570 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4571 PetscValidIntPointer(lits,2); 4572 *lits = ts->ksp_its; 4573 PetscFunctionReturn(0); 4574 } 4575 4576 #undef __FUNCT__ 4577 #define __FUNCT__ "TSGetStepRejections" 4578 /*@ 4579 TSGetStepRejections - Gets the total number of rejected steps. 4580 4581 Not Collective 4582 4583 Input Parameter: 4584 . ts - TS context 4585 4586 Output Parameter: 4587 . rejects - number of steps rejected 4588 4589 Notes: 4590 This counter is reset to zero for each successive call to TSSolve(). 4591 4592 Level: intermediate 4593 4594 .keywords: TS, get, number 4595 4596 .seealso: TSGetSNESIterations(), TSGetKSPIterations(), TSSetMaxStepRejections(), TSGetSNESFailures(), TSSetMaxSNESFailures(), TSSetErrorIfStepFails() 4597 @*/ 4598 PetscErrorCode TSGetStepRejections(TS ts,PetscInt *rejects) 4599 { 4600 PetscFunctionBegin; 4601 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4602 PetscValidIntPointer(rejects,2); 4603 *rejects = ts->reject; 4604 PetscFunctionReturn(0); 4605 } 4606 4607 #undef __FUNCT__ 4608 #define __FUNCT__ "TSGetSNESFailures" 4609 /*@ 4610 TSGetSNESFailures - Gets the total number of failed SNES solves 4611 4612 Not Collective 4613 4614 Input Parameter: 4615 . ts - TS context 4616 4617 Output Parameter: 4618 . fails - number of failed nonlinear solves 4619 4620 Notes: 4621 This counter is reset to zero for each successive call to TSSolve(). 4622 4623 Level: intermediate 4624 4625 .keywords: TS, get, number 4626 4627 .seealso: TSGetSNESIterations(), TSGetKSPIterations(), TSSetMaxStepRejections(), TSGetStepRejections(), TSSetMaxSNESFailures() 4628 @*/ 4629 PetscErrorCode TSGetSNESFailures(TS ts,PetscInt *fails) 4630 { 4631 PetscFunctionBegin; 4632 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4633 PetscValidIntPointer(fails,2); 4634 *fails = ts->num_snes_failures; 4635 PetscFunctionReturn(0); 4636 } 4637 4638 #undef __FUNCT__ 4639 #define __FUNCT__ "TSSetMaxStepRejections" 4640 /*@ 4641 TSSetMaxStepRejections - Sets the maximum number of step rejections before a step fails 4642 4643 Not Collective 4644 4645 Input Parameter: 4646 + ts - TS context 4647 - rejects - maximum number of rejected steps, pass -1 for unlimited 4648 4649 Notes: 4650 The counter is reset to zero for each step 4651 4652 Options Database Key: 4653 . -ts_max_reject - Maximum number of step rejections before a step fails 4654 4655 Level: intermediate 4656 4657 .keywords: TS, set, maximum, number 4658 4659 .seealso: TSGetSNESIterations(), TSGetKSPIterations(), TSSetMaxSNESFailures(), TSGetStepRejections(), TSGetSNESFailures(), TSSetErrorIfStepFails(), TSGetConvergedReason() 4660 @*/ 4661 PetscErrorCode TSSetMaxStepRejections(TS ts,PetscInt rejects) 4662 { 4663 PetscFunctionBegin; 4664 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4665 ts->max_reject = rejects; 4666 PetscFunctionReturn(0); 4667 } 4668 4669 #undef __FUNCT__ 4670 #define __FUNCT__ "TSSetMaxSNESFailures" 4671 /*@ 4672 TSSetMaxSNESFailures - Sets the maximum number of failed SNES solves 4673 4674 Not Collective 4675 4676 Input Parameter: 4677 + ts - TS context 4678 - fails - maximum number of failed nonlinear solves, pass -1 for unlimited 4679 4680 Notes: 4681 The counter is reset to zero for each successive call to TSSolve(). 4682 4683 Options Database Key: 4684 . -ts_max_snes_failures - Maximum number of nonlinear solve failures 4685 4686 Level: intermediate 4687 4688 .keywords: TS, set, maximum, number 4689 4690 .seealso: TSGetSNESIterations(), TSGetKSPIterations(), TSSetMaxStepRejections(), TSGetStepRejections(), TSGetSNESFailures(), SNESGetConvergedReason(), TSGetConvergedReason() 4691 @*/ 4692 PetscErrorCode TSSetMaxSNESFailures(TS ts,PetscInt fails) 4693 { 4694 PetscFunctionBegin; 4695 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4696 ts->max_snes_failures = fails; 4697 PetscFunctionReturn(0); 4698 } 4699 4700 #undef __FUNCT__ 4701 #define __FUNCT__ "TSSetErrorIfStepFails" 4702 /*@ 4703 TSSetErrorIfStepFails - Error if no step succeeds 4704 4705 Not Collective 4706 4707 Input Parameter: 4708 + ts - TS context 4709 - err - PETSC_TRUE to error if no step succeeds, PETSC_FALSE to return without failure 4710 4711 Options Database Key: 4712 . -ts_error_if_step_fails - Error if no step succeeds 4713 4714 Level: intermediate 4715 4716 .keywords: TS, set, error 4717 4718 .seealso: TSGetSNESIterations(), TSGetKSPIterations(), TSSetMaxStepRejections(), TSGetStepRejections(), TSGetSNESFailures(), TSSetErrorIfStepFails(), TSGetConvergedReason() 4719 @*/ 4720 PetscErrorCode TSSetErrorIfStepFails(TS ts,PetscBool err) 4721 { 4722 PetscFunctionBegin; 4723 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4724 ts->errorifstepfailed = err; 4725 PetscFunctionReturn(0); 4726 } 4727 4728 #undef __FUNCT__ 4729 #define __FUNCT__ "TSMonitorSolutionBinary" 4730 /*@C 4731 TSMonitorSolutionBinary - Monitors progress of the TS solvers by VecView() for the solution at each timestep. Normally the viewer is a binary file 4732 4733 Collective on TS 4734 4735 Input Parameters: 4736 + ts - the TS context 4737 . step - current time-step 4738 . ptime - current time 4739 . u - current state 4740 - viewer - binary viewer 4741 4742 Level: intermediate 4743 4744 .keywords: TS, vector, monitor, view 4745 4746 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView() 4747 @*/ 4748 PetscErrorCode TSMonitorSolutionBinary(TS ts,PetscInt step,PetscReal ptime,Vec u,void *viewer) 4749 { 4750 PetscErrorCode ierr; 4751 PetscViewer v = (PetscViewer)viewer; 4752 4753 PetscFunctionBegin; 4754 ierr = VecView(u,v);CHKERRQ(ierr); 4755 PetscFunctionReturn(0); 4756 } 4757 4758 #undef __FUNCT__ 4759 #define __FUNCT__ "TSMonitorSolutionVTK" 4760 /*@C 4761 TSMonitorSolutionVTK - Monitors progress of the TS solvers by VecView() for the solution at each timestep. 4762 4763 Collective on TS 4764 4765 Input Parameters: 4766 + ts - the TS context 4767 . step - current time-step 4768 . ptime - current time 4769 . u - current state 4770 - filenametemplate - string containing a format specifier for the integer time step (e.g. %03D) 4771 4772 Level: intermediate 4773 4774 Notes: 4775 The VTK format does not allow writing multiple time steps in the same file, therefore a different file will be written for each time step. 4776 These are named according to the file name template. 4777 4778 This function is normally passed as an argument to TSMonitorSet() along with TSMonitorSolutionVTKDestroy(). 4779 4780 .keywords: TS, vector, monitor, view 4781 4782 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView() 4783 @*/ 4784 PetscErrorCode TSMonitorSolutionVTK(TS ts,PetscInt step,PetscReal ptime,Vec u,void *filenametemplate) 4785 { 4786 PetscErrorCode ierr; 4787 char filename[PETSC_MAX_PATH_LEN]; 4788 PetscViewer viewer; 4789 4790 PetscFunctionBegin; 4791 ierr = PetscSNPrintf(filename,sizeof(filename),(const char*)filenametemplate,step);CHKERRQ(ierr); 4792 ierr = PetscViewerVTKOpen(PetscObjectComm((PetscObject)ts),filename,FILE_MODE_WRITE,&viewer);CHKERRQ(ierr); 4793 ierr = VecView(u,viewer);CHKERRQ(ierr); 4794 ierr = PetscViewerDestroy(&viewer);CHKERRQ(ierr); 4795 PetscFunctionReturn(0); 4796 } 4797 4798 #undef __FUNCT__ 4799 #define __FUNCT__ "TSMonitorSolutionVTKDestroy" 4800 /*@C 4801 TSMonitorSolutionVTKDestroy - Destroy context for monitoring 4802 4803 Collective on TS 4804 4805 Input Parameters: 4806 . filenametemplate - string containing a format specifier for the integer time step (e.g. %03D) 4807 4808 Level: intermediate 4809 4810 Note: 4811 This function is normally passed to TSMonitorSet() along with TSMonitorSolutionVTK(). 4812 4813 .keywords: TS, vector, monitor, view 4814 4815 .seealso: TSMonitorSet(), TSMonitorSolutionVTK() 4816 @*/ 4817 PetscErrorCode TSMonitorSolutionVTKDestroy(void *filenametemplate) 4818 { 4819 PetscErrorCode ierr; 4820 4821 PetscFunctionBegin; 4822 ierr = PetscFree(*(char**)filenametemplate);CHKERRQ(ierr); 4823 PetscFunctionReturn(0); 4824 } 4825 4826 #undef __FUNCT__ 4827 #define __FUNCT__ "TSGetAdapt" 4828 /*@ 4829 TSGetAdapt - Get the adaptive controller context for the current method 4830 4831 Collective on TS if controller has not been created yet 4832 4833 Input Arguments: 4834 . ts - time stepping context 4835 4836 Output Arguments: 4837 . adapt - adaptive controller 4838 4839 Level: intermediate 4840 4841 .seealso: TSAdapt, TSAdaptSetType(), TSAdaptChoose() 4842 @*/ 4843 PetscErrorCode TSGetAdapt(TS ts,TSAdapt *adapt) 4844 { 4845 PetscErrorCode ierr; 4846 4847 PetscFunctionBegin; 4848 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4849 PetscValidPointer(adapt,2); 4850 if (!ts->adapt) { 4851 ierr = TSAdaptCreate(PetscObjectComm((PetscObject)ts),&ts->adapt);CHKERRQ(ierr); 4852 ierr = PetscLogObjectParent((PetscObject)ts,(PetscObject)ts->adapt);CHKERRQ(ierr); 4853 ierr = PetscObjectIncrementTabLevel((PetscObject)ts->adapt,(PetscObject)ts,1);CHKERRQ(ierr); 4854 } 4855 *adapt = ts->adapt; 4856 PetscFunctionReturn(0); 4857 } 4858 4859 #undef __FUNCT__ 4860 #define __FUNCT__ "TSSetTolerances" 4861 /*@ 4862 TSSetTolerances - Set tolerances for local truncation error when using adaptive controller 4863 4864 Logically Collective 4865 4866 Input Arguments: 4867 + ts - time integration context 4868 . atol - scalar absolute tolerances, PETSC_DECIDE to leave current value 4869 . vatol - vector of absolute tolerances or NULL, used in preference to atol if present 4870 . rtol - scalar relative tolerances, PETSC_DECIDE to leave current value 4871 - vrtol - vector of relative tolerances or NULL, used in preference to atol if present 4872 4873 Options Database keys: 4874 + -ts_rtol <rtol> - relative tolerance for local truncation error 4875 - -ts_atol <atol> Absolute tolerance for local truncation error 4876 4877 Notes: 4878 With PETSc's implicit schemes for DAE problems, the calculation of the local truncation error 4879 (LTE) includes both the differential and the algebraic variables. If one wants the LTE to be 4880 computed only for the differential or the algebraic part then this can be done using the vector of 4881 tolerances vatol. For example, by setting the tolerance vector with the desired tolerance for the 4882 differential part and infinity for the algebraic part, the LTE calculation will include only the 4883 differential variables. 4884 4885 Level: beginner 4886 4887 .seealso: TS, TSAdapt, TSVecNormWRMS(), TSGetTolerances() 4888 @*/ 4889 PetscErrorCode TSSetTolerances(TS ts,PetscReal atol,Vec vatol,PetscReal rtol,Vec vrtol) 4890 { 4891 PetscErrorCode ierr; 4892 4893 PetscFunctionBegin; 4894 if (atol != PETSC_DECIDE && atol != PETSC_DEFAULT) ts->atol = atol; 4895 if (vatol) { 4896 ierr = PetscObjectReference((PetscObject)vatol);CHKERRQ(ierr); 4897 ierr = VecDestroy(&ts->vatol);CHKERRQ(ierr); 4898 4899 ts->vatol = vatol; 4900 } 4901 if (rtol != PETSC_DECIDE && rtol != PETSC_DEFAULT) ts->rtol = rtol; 4902 if (vrtol) { 4903 ierr = PetscObjectReference((PetscObject)vrtol);CHKERRQ(ierr); 4904 ierr = VecDestroy(&ts->vrtol);CHKERRQ(ierr); 4905 4906 ts->vrtol = vrtol; 4907 } 4908 PetscFunctionReturn(0); 4909 } 4910 4911 #undef __FUNCT__ 4912 #define __FUNCT__ "TSGetTolerances" 4913 /*@ 4914 TSGetTolerances - Get tolerances for local truncation error when using adaptive controller 4915 4916 Logically Collective 4917 4918 Input Arguments: 4919 . ts - time integration context 4920 4921 Output Arguments: 4922 + atol - scalar absolute tolerances, NULL to ignore 4923 . vatol - vector of absolute tolerances, NULL to ignore 4924 . rtol - scalar relative tolerances, NULL to ignore 4925 - vrtol - vector of relative tolerances, NULL to ignore 4926 4927 Level: beginner 4928 4929 .seealso: TS, TSAdapt, TSVecNormWRMS(), TSSetTolerances() 4930 @*/ 4931 PetscErrorCode TSGetTolerances(TS ts,PetscReal *atol,Vec *vatol,PetscReal *rtol,Vec *vrtol) 4932 { 4933 PetscFunctionBegin; 4934 if (atol) *atol = ts->atol; 4935 if (vatol) *vatol = ts->vatol; 4936 if (rtol) *rtol = ts->rtol; 4937 if (vrtol) *vrtol = ts->vrtol; 4938 PetscFunctionReturn(0); 4939 } 4940 4941 #undef __FUNCT__ 4942 #define __FUNCT__ "TSErrorWeightedNorm2" 4943 /*@ 4944 TSErrorWeightedNorm2 - compute a weighted 2-norm of the difference between two state vectors 4945 4946 Collective on TS 4947 4948 Input Arguments: 4949 + ts - time stepping context 4950 . U - state vector, usually ts->vec_sol 4951 - Y - state vector to be compared to U 4952 4953 Output Arguments: 4954 . norm - weighted norm, a value of 1.0 is considered small 4955 4956 Level: developer 4957 4958 .seealso: TSErrorWeightedNorm(), TSErrorWeightedNormInfinity() 4959 @*/ 4960 PetscErrorCode TSErrorWeightedNorm2(TS ts,Vec U,Vec Y,PetscReal *norm) 4961 { 4962 PetscErrorCode ierr; 4963 PetscInt i,n,N,rstart; 4964 const PetscScalar *u,*y; 4965 PetscReal sum,gsum; 4966 PetscReal tol; 4967 4968 PetscFunctionBegin; 4969 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 4970 PetscValidHeaderSpecific(U,VEC_CLASSID,2); 4971 PetscValidHeaderSpecific(Y,VEC_CLASSID,3); 4972 PetscValidType(U,2); 4973 PetscValidType(Y,3); 4974 PetscCheckSameComm(U,2,Y,3); 4975 PetscValidPointer(norm,4); 4976 if (U == Y) SETERRQ(PetscObjectComm((PetscObject)U),PETSC_ERR_ARG_IDN,"U and Y cannot be the same vector"); 4977 4978 ierr = VecGetSize(U,&N);CHKERRQ(ierr); 4979 ierr = VecGetLocalSize(U,&n);CHKERRQ(ierr); 4980 ierr = VecGetOwnershipRange(U,&rstart,NULL);CHKERRQ(ierr); 4981 ierr = VecGetArrayRead(U,&u);CHKERRQ(ierr); 4982 ierr = VecGetArrayRead(Y,&y);CHKERRQ(ierr); 4983 sum = 0.; 4984 if (ts->vatol && ts->vrtol) { 4985 const PetscScalar *atol,*rtol; 4986 ierr = VecGetArrayRead(ts->vatol,&atol);CHKERRQ(ierr); 4987 ierr = VecGetArrayRead(ts->vrtol,&rtol);CHKERRQ(ierr); 4988 for (i=0; i<n; i++) { 4989 tol = PetscRealPart(atol[i]) + PetscRealPart(rtol[i]) * PetscMax(PetscAbsScalar(u[i]),PetscAbsScalar(y[i])); 4990 sum += PetscSqr(PetscAbsScalar(y[i] - u[i]) / tol); 4991 } 4992 ierr = VecRestoreArrayRead(ts->vatol,&atol);CHKERRQ(ierr); 4993 ierr = VecRestoreArrayRead(ts->vrtol,&rtol);CHKERRQ(ierr); 4994 } else if (ts->vatol) { /* vector atol, scalar rtol */ 4995 const PetscScalar *atol; 4996 ierr = VecGetArrayRead(ts->vatol,&atol);CHKERRQ(ierr); 4997 for (i=0; i<n; i++) { 4998 tol = PetscRealPart(atol[i]) + ts->rtol * PetscMax(PetscAbsScalar(u[i]),PetscAbsScalar(y[i])); 4999 sum += PetscSqr(PetscAbsScalar(y[i] - u[i]) / tol); 5000 } 5001 ierr = VecRestoreArrayRead(ts->vatol,&atol);CHKERRQ(ierr); 5002 } else if (ts->vrtol) { /* scalar atol, vector rtol */ 5003 const PetscScalar *rtol; 5004 ierr = VecGetArrayRead(ts->vrtol,&rtol);CHKERRQ(ierr); 5005 for (i=0; i<n; i++) { 5006 tol = ts->atol + PetscRealPart(rtol[i]) * PetscMax(PetscAbsScalar(u[i]),PetscAbsScalar(y[i])); 5007 sum += PetscSqr(PetscAbsScalar(y[i] - u[i]) / tol); 5008 } 5009 ierr = VecRestoreArrayRead(ts->vrtol,&rtol);CHKERRQ(ierr); 5010 } else { /* scalar atol, scalar rtol */ 5011 for (i=0; i<n; i++) { 5012 tol = ts->atol + ts->rtol * PetscMax(PetscAbsScalar(u[i]),PetscAbsScalar(y[i])); 5013 sum += PetscSqr(PetscAbsScalar(y[i] - u[i]) / tol); 5014 } 5015 } 5016 ierr = VecRestoreArrayRead(U,&u);CHKERRQ(ierr); 5017 ierr = VecRestoreArrayRead(Y,&y);CHKERRQ(ierr); 5018 5019 ierr = MPI_Allreduce(&sum,&gsum,1,MPIU_REAL,MPIU_SUM,PetscObjectComm((PetscObject)ts));CHKERRQ(ierr); 5020 *norm = PetscSqrtReal(gsum / N); 5021 5022 if (PetscIsInfOrNanScalar(*norm)) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_FP,"Infinite or not-a-number generated in norm"); 5023 PetscFunctionReturn(0); 5024 } 5025 5026 #undef __FUNCT__ 5027 #define __FUNCT__ "TSErrorWeightedNormInfinity" 5028 /*@ 5029 TSErrorWeightedNormInfinity - compute a weighted infinity-norm of the difference between two state vectors 5030 5031 Collective on TS 5032 5033 Input Arguments: 5034 + ts - time stepping context 5035 . U - state vector, usually ts->vec_sol 5036 - Y - state vector to be compared to U 5037 5038 Output Arguments: 5039 . norm - weighted norm, a value of 1.0 is considered small 5040 5041 Level: developer 5042 5043 .seealso: TSErrorWeightedNorm(), TSErrorWeightedNorm2() 5044 @*/ 5045 PetscErrorCode TSErrorWeightedNormInfinity(TS ts,Vec U,Vec Y,PetscReal *norm) 5046 { 5047 PetscErrorCode ierr; 5048 PetscInt i,n,N,rstart,k; 5049 const PetscScalar *u,*y; 5050 PetscReal max,gmax; 5051 PetscReal tol; 5052 5053 PetscFunctionBegin; 5054 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 5055 PetscValidHeaderSpecific(U,VEC_CLASSID,2); 5056 PetscValidHeaderSpecific(Y,VEC_CLASSID,3); 5057 PetscValidType(U,2); 5058 PetscValidType(Y,3); 5059 PetscCheckSameComm(U,2,Y,3); 5060 PetscValidPointer(norm,4); 5061 if (U == Y) SETERRQ(PetscObjectComm((PetscObject)U),PETSC_ERR_ARG_IDN,"U and Y cannot be the same vector"); 5062 5063 ierr = VecGetSize(U,&N);CHKERRQ(ierr); 5064 ierr = VecGetLocalSize(U,&n);CHKERRQ(ierr); 5065 ierr = VecGetOwnershipRange(U,&rstart,NULL);CHKERRQ(ierr); 5066 ierr = VecGetArrayRead(U,&u);CHKERRQ(ierr); 5067 ierr = VecGetArrayRead(Y,&y);CHKERRQ(ierr); 5068 if (ts->vatol && ts->vrtol) { 5069 const PetscScalar *atol,*rtol; 5070 ierr = VecGetArrayRead(ts->vatol,&atol);CHKERRQ(ierr); 5071 ierr = VecGetArrayRead(ts->vrtol,&rtol);CHKERRQ(ierr); 5072 k = 0; 5073 tol = PetscRealPart(atol[k]) + PetscRealPart(rtol[k]) * PetscMax(PetscAbsScalar(u[k]),PetscAbsScalar(y[k])); 5074 max = PetscAbsScalar(y[k] - u[k]) / tol; 5075 for (i=1; i<n; i++) { 5076 tol = PetscRealPart(atol[i]) + PetscRealPart(rtol[i]) * PetscMax(PetscAbsScalar(u[i]),PetscAbsScalar(y[i])); 5077 max = PetscMax(max,PetscAbsScalar(y[i] - u[i]) / tol); 5078 } 5079 ierr = VecRestoreArrayRead(ts->vatol,&atol);CHKERRQ(ierr); 5080 ierr = VecRestoreArrayRead(ts->vrtol,&rtol);CHKERRQ(ierr); 5081 } else if (ts->vatol) { /* vector atol, scalar rtol */ 5082 const PetscScalar *atol; 5083 ierr = VecGetArrayRead(ts->vatol,&atol);CHKERRQ(ierr); 5084 k = 0; 5085 tol = PetscRealPart(atol[k]) + ts->rtol * PetscMax(PetscAbsScalar(u[k]),PetscAbsScalar(y[k])); 5086 max = PetscAbsScalar(y[k] - u[k]) / tol; 5087 for (i=1; i<n; i++) { 5088 tol = PetscRealPart(atol[i]) + ts->rtol * PetscMax(PetscAbsScalar(u[i]),PetscAbsScalar(y[i])); 5089 max = PetscMax(max,PetscAbsScalar(y[i] - u[i]) / tol); 5090 } 5091 ierr = VecRestoreArrayRead(ts->vatol,&atol);CHKERRQ(ierr); 5092 } else if (ts->vrtol) { /* scalar atol, vector rtol */ 5093 const PetscScalar *rtol; 5094 ierr = VecGetArrayRead(ts->vrtol,&rtol);CHKERRQ(ierr); 5095 k = 0; 5096 tol = ts->atol + PetscRealPart(rtol[k]) * PetscMax(PetscAbsScalar(u[k]),PetscAbsScalar(y[k])); 5097 max = PetscAbsScalar(y[k] - u[k]) / tol; 5098 for (i=1; i<n; i++) { 5099 tol = ts->atol + PetscRealPart(rtol[i]) * PetscMax(PetscAbsScalar(u[i]),PetscAbsScalar(y[i])); 5100 max = PetscMax(max,PetscAbsScalar(y[i] - u[i]) / tol); 5101 } 5102 ierr = VecRestoreArrayRead(ts->vrtol,&rtol);CHKERRQ(ierr); 5103 } else { /* scalar atol, scalar rtol */ 5104 k = 0; 5105 tol = ts->atol + ts->rtol * PetscMax(PetscAbsScalar(u[k]),PetscAbsScalar(y[k])); 5106 max = PetscAbsScalar(y[k] - u[k]) / tol; 5107 for (i=1; i<n; i++) { 5108 tol = ts->atol + ts->rtol * PetscMax(PetscAbsScalar(u[i]),PetscAbsScalar(y[i])); 5109 max = PetscMax(max,PetscAbsScalar(y[i] - u[i]) / tol); 5110 } 5111 } 5112 ierr = VecRestoreArrayRead(U,&u);CHKERRQ(ierr); 5113 ierr = VecRestoreArrayRead(Y,&y);CHKERRQ(ierr); 5114 5115 ierr = MPI_Allreduce(&max,&gmax,1,MPIU_REAL,MPIU_MAX,PetscObjectComm((PetscObject)ts));CHKERRQ(ierr); 5116 *norm = gmax; 5117 5118 if (PetscIsInfOrNanScalar(*norm)) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_FP,"Infinite or not-a-number generated in norm"); 5119 PetscFunctionReturn(0); 5120 } 5121 5122 #undef __FUNCT__ 5123 #define __FUNCT__ "TSErrorWeightedNorm" 5124 /*@ 5125 TSErrorWeightedNorm - compute a weighted norm of the difference between two state vectors 5126 5127 Collective on TS 5128 5129 Input Arguments: 5130 + ts - time stepping context 5131 . U - state vector, usually ts->vec_sol 5132 . Y - state vector to be compared to U 5133 - wnormtype - norm type, either NORM_2 or NORM_INFINITY 5134 5135 Output Arguments: 5136 . norm - weighted norm, a value of 1.0 is considered small 5137 5138 5139 Options Database Keys: 5140 . -ts_adapt_wnormtype <wnormtype> - 2, INFINITY 5141 5142 Level: developer 5143 5144 .seealso: TSErrorWeightedNormInfinity(), TSErrorWeightedNorm2() 5145 @*/ 5146 PetscErrorCode TSErrorWeightedNorm(TS ts,Vec U,Vec Y,NormType wnormtype,PetscReal *norm) 5147 { 5148 PetscErrorCode ierr; 5149 5150 PetscFunctionBegin; 5151 if (wnormtype == NORM_2) { 5152 ierr = TSErrorWeightedNorm2(ts,U,Y,norm);CHKERRQ(ierr); 5153 } else if(wnormtype == NORM_INFINITY) { 5154 ierr = TSErrorWeightedNormInfinity(ts,U,Y,norm);CHKERRQ(ierr); 5155 } else SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"No support for norm type %s",NormTypes[wnormtype]); 5156 PetscFunctionReturn(0); 5157 } 5158 5159 #undef __FUNCT__ 5160 #define __FUNCT__ "TSSetCFLTimeLocal" 5161 /*@ 5162 TSSetCFLTimeLocal - Set the local CFL constraint relative to forward Euler 5163 5164 Logically Collective on TS 5165 5166 Input Arguments: 5167 + ts - time stepping context 5168 - cfltime - maximum stable time step if using forward Euler (value can be different on each process) 5169 5170 Note: 5171 After calling this function, the global CFL time can be obtained by calling TSGetCFLTime() 5172 5173 Level: intermediate 5174 5175 .seealso: TSGetCFLTime(), TSADAPTCFL 5176 @*/ 5177 PetscErrorCode TSSetCFLTimeLocal(TS ts,PetscReal cfltime) 5178 { 5179 PetscFunctionBegin; 5180 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 5181 ts->cfltime_local = cfltime; 5182 ts->cfltime = -1.; 5183 PetscFunctionReturn(0); 5184 } 5185 5186 #undef __FUNCT__ 5187 #define __FUNCT__ "TSGetCFLTime" 5188 /*@ 5189 TSGetCFLTime - Get the maximum stable time step according to CFL criteria applied to forward Euler 5190 5191 Collective on TS 5192 5193 Input Arguments: 5194 . ts - time stepping context 5195 5196 Output Arguments: 5197 . cfltime - maximum stable time step for forward Euler 5198 5199 Level: advanced 5200 5201 .seealso: TSSetCFLTimeLocal() 5202 @*/ 5203 PetscErrorCode TSGetCFLTime(TS ts,PetscReal *cfltime) 5204 { 5205 PetscErrorCode ierr; 5206 5207 PetscFunctionBegin; 5208 if (ts->cfltime < 0) { 5209 ierr = MPI_Allreduce(&ts->cfltime_local,&ts->cfltime,1,MPIU_REAL,MPIU_MIN,PetscObjectComm((PetscObject)ts));CHKERRQ(ierr); 5210 } 5211 *cfltime = ts->cfltime; 5212 PetscFunctionReturn(0); 5213 } 5214 5215 #undef __FUNCT__ 5216 #define __FUNCT__ "TSVISetVariableBounds" 5217 /*@ 5218 TSVISetVariableBounds - Sets the lower and upper bounds for the solution vector. xl <= x <= xu 5219 5220 Input Parameters: 5221 . ts - the TS context. 5222 . xl - lower bound. 5223 . xu - upper bound. 5224 5225 Notes: 5226 If this routine is not called then the lower and upper bounds are set to 5227 PETSC_NINFINITY and PETSC_INFINITY respectively during SNESSetUp(). 5228 5229 Level: advanced 5230 5231 @*/ 5232 PetscErrorCode TSVISetVariableBounds(TS ts, Vec xl, Vec xu) 5233 { 5234 PetscErrorCode ierr; 5235 SNES snes; 5236 5237 PetscFunctionBegin; 5238 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 5239 ierr = SNESVISetVariableBounds(snes,xl,xu);CHKERRQ(ierr); 5240 PetscFunctionReturn(0); 5241 } 5242 5243 #if defined(PETSC_HAVE_MATLAB_ENGINE) 5244 #include <mex.h> 5245 5246 typedef struct {char *funcname; mxArray *ctx;} TSMatlabContext; 5247 5248 #undef __FUNCT__ 5249 #define __FUNCT__ "TSComputeFunction_Matlab" 5250 /* 5251 TSComputeFunction_Matlab - Calls the function that has been set with 5252 TSSetFunctionMatlab(). 5253 5254 Collective on TS 5255 5256 Input Parameters: 5257 + snes - the TS context 5258 - u - input vector 5259 5260 Output Parameter: 5261 . y - function vector, as set by TSSetFunction() 5262 5263 Notes: 5264 TSComputeFunction() is typically used within nonlinear solvers 5265 implementations, so most users would not generally call this routine 5266 themselves. 5267 5268 Level: developer 5269 5270 .keywords: TS, nonlinear, compute, function 5271 5272 .seealso: TSSetFunction(), TSGetFunction() 5273 */ 5274 PetscErrorCode TSComputeFunction_Matlab(TS snes,PetscReal time,Vec u,Vec udot,Vec y, void *ctx) 5275 { 5276 PetscErrorCode ierr; 5277 TSMatlabContext *sctx = (TSMatlabContext*)ctx; 5278 int nlhs = 1,nrhs = 7; 5279 mxArray *plhs[1],*prhs[7]; 5280 long long int lx = 0,lxdot = 0,ly = 0,ls = 0; 5281 5282 PetscFunctionBegin; 5283 PetscValidHeaderSpecific(snes,TS_CLASSID,1); 5284 PetscValidHeaderSpecific(u,VEC_CLASSID,3); 5285 PetscValidHeaderSpecific(udot,VEC_CLASSID,4); 5286 PetscValidHeaderSpecific(y,VEC_CLASSID,5); 5287 PetscCheckSameComm(snes,1,u,3); 5288 PetscCheckSameComm(snes,1,y,5); 5289 5290 ierr = PetscMemcpy(&ls,&snes,sizeof(snes));CHKERRQ(ierr); 5291 ierr = PetscMemcpy(&lx,&u,sizeof(u));CHKERRQ(ierr); 5292 ierr = PetscMemcpy(&lxdot,&udot,sizeof(udot));CHKERRQ(ierr); 5293 ierr = PetscMemcpy(&ly,&y,sizeof(u));CHKERRQ(ierr); 5294 5295 prhs[0] = mxCreateDoubleScalar((double)ls); 5296 prhs[1] = mxCreateDoubleScalar(time); 5297 prhs[2] = mxCreateDoubleScalar((double)lx); 5298 prhs[3] = mxCreateDoubleScalar((double)lxdot); 5299 prhs[4] = mxCreateDoubleScalar((double)ly); 5300 prhs[5] = mxCreateString(sctx->funcname); 5301 prhs[6] = sctx->ctx; 5302 ierr = mexCallMATLAB(nlhs,plhs,nrhs,prhs,"PetscTSComputeFunctionInternal");CHKERRQ(ierr); 5303 ierr = mxGetScalar(plhs[0]);CHKERRQ(ierr); 5304 mxDestroyArray(prhs[0]); 5305 mxDestroyArray(prhs[1]); 5306 mxDestroyArray(prhs[2]); 5307 mxDestroyArray(prhs[3]); 5308 mxDestroyArray(prhs[4]); 5309 mxDestroyArray(prhs[5]); 5310 mxDestroyArray(plhs[0]); 5311 PetscFunctionReturn(0); 5312 } 5313 5314 5315 #undef __FUNCT__ 5316 #define __FUNCT__ "TSSetFunctionMatlab" 5317 /* 5318 TSSetFunctionMatlab - Sets the function evaluation routine and function 5319 vector for use by the TS routines in solving ODEs 5320 equations from MATLAB. Here the function is a string containing the name of a MATLAB function 5321 5322 Logically Collective on TS 5323 5324 Input Parameters: 5325 + ts - the TS context 5326 - func - function evaluation routine 5327 5328 Calling sequence of func: 5329 $ func (TS ts,PetscReal time,Vec u,Vec udot,Vec f,void *ctx); 5330 5331 Level: beginner 5332 5333 .keywords: TS, nonlinear, set, function 5334 5335 .seealso: TSGetFunction(), TSComputeFunction(), TSSetJacobian(), TSSetFunction() 5336 */ 5337 PetscErrorCode TSSetFunctionMatlab(TS ts,const char *func,mxArray *ctx) 5338 { 5339 PetscErrorCode ierr; 5340 TSMatlabContext *sctx; 5341 5342 PetscFunctionBegin; 5343 /* currently sctx is memory bleed */ 5344 ierr = PetscMalloc(sizeof(TSMatlabContext),&sctx);CHKERRQ(ierr); 5345 ierr = PetscStrallocpy(func,&sctx->funcname);CHKERRQ(ierr); 5346 /* 5347 This should work, but it doesn't 5348 sctx->ctx = ctx; 5349 mexMakeArrayPersistent(sctx->ctx); 5350 */ 5351 sctx->ctx = mxDuplicateArray(ctx); 5352 5353 ierr = TSSetIFunction(ts,NULL,TSComputeFunction_Matlab,sctx);CHKERRQ(ierr); 5354 PetscFunctionReturn(0); 5355 } 5356 5357 #undef __FUNCT__ 5358 #define __FUNCT__ "TSComputeJacobian_Matlab" 5359 /* 5360 TSComputeJacobian_Matlab - Calls the function that has been set with 5361 TSSetJacobianMatlab(). 5362 5363 Collective on TS 5364 5365 Input Parameters: 5366 + ts - the TS context 5367 . u - input vector 5368 . A, B - the matrices 5369 - ctx - user context 5370 5371 Level: developer 5372 5373 .keywords: TS, nonlinear, compute, function 5374 5375 .seealso: TSSetFunction(), TSGetFunction() 5376 @*/ 5377 PetscErrorCode TSComputeJacobian_Matlab(TS ts,PetscReal time,Vec u,Vec udot,PetscReal shift,Mat A,Mat B,void *ctx) 5378 { 5379 PetscErrorCode ierr; 5380 TSMatlabContext *sctx = (TSMatlabContext*)ctx; 5381 int nlhs = 2,nrhs = 9; 5382 mxArray *plhs[2],*prhs[9]; 5383 long long int lx = 0,lxdot = 0,lA = 0,ls = 0, lB = 0; 5384 5385 PetscFunctionBegin; 5386 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 5387 PetscValidHeaderSpecific(u,VEC_CLASSID,3); 5388 5389 /* call Matlab function in ctx with arguments u and y */ 5390 5391 ierr = PetscMemcpy(&ls,&ts,sizeof(ts));CHKERRQ(ierr); 5392 ierr = PetscMemcpy(&lx,&u,sizeof(u));CHKERRQ(ierr); 5393 ierr = PetscMemcpy(&lxdot,&udot,sizeof(u));CHKERRQ(ierr); 5394 ierr = PetscMemcpy(&lA,A,sizeof(u));CHKERRQ(ierr); 5395 ierr = PetscMemcpy(&lB,B,sizeof(u));CHKERRQ(ierr); 5396 5397 prhs[0] = mxCreateDoubleScalar((double)ls); 5398 prhs[1] = mxCreateDoubleScalar((double)time); 5399 prhs[2] = mxCreateDoubleScalar((double)lx); 5400 prhs[3] = mxCreateDoubleScalar((double)lxdot); 5401 prhs[4] = mxCreateDoubleScalar((double)shift); 5402 prhs[5] = mxCreateDoubleScalar((double)lA); 5403 prhs[6] = mxCreateDoubleScalar((double)lB); 5404 prhs[7] = mxCreateString(sctx->funcname); 5405 prhs[8] = sctx->ctx; 5406 ierr = mexCallMATLAB(nlhs,plhs,nrhs,prhs,"PetscTSComputeJacobianInternal");CHKERRQ(ierr); 5407 ierr = mxGetScalar(plhs[0]);CHKERRQ(ierr); 5408 mxDestroyArray(prhs[0]); 5409 mxDestroyArray(prhs[1]); 5410 mxDestroyArray(prhs[2]); 5411 mxDestroyArray(prhs[3]); 5412 mxDestroyArray(prhs[4]); 5413 mxDestroyArray(prhs[5]); 5414 mxDestroyArray(prhs[6]); 5415 mxDestroyArray(prhs[7]); 5416 mxDestroyArray(plhs[0]); 5417 mxDestroyArray(plhs[1]); 5418 PetscFunctionReturn(0); 5419 } 5420 5421 5422 #undef __FUNCT__ 5423 #define __FUNCT__ "TSSetJacobianMatlab" 5424 /* 5425 TSSetJacobianMatlab - Sets the Jacobian function evaluation routine and two empty Jacobian matrices 5426 vector for use by the TS routines in solving ODEs from MATLAB. Here the function is a string containing the name of a MATLAB function 5427 5428 Logically Collective on TS 5429 5430 Input Parameters: 5431 + ts - the TS context 5432 . A,B - Jacobian matrices 5433 . func - function evaluation routine 5434 - ctx - user context 5435 5436 Calling sequence of func: 5437 $ flag = func (TS ts,PetscReal time,Vec u,Vec udot,Mat A,Mat B,void *ctx); 5438 5439 5440 Level: developer 5441 5442 .keywords: TS, nonlinear, set, function 5443 5444 .seealso: TSGetFunction(), TSComputeFunction(), TSSetJacobian(), TSSetFunction() 5445 */ 5446 PetscErrorCode TSSetJacobianMatlab(TS ts,Mat A,Mat B,const char *func,mxArray *ctx) 5447 { 5448 PetscErrorCode ierr; 5449 TSMatlabContext *sctx; 5450 5451 PetscFunctionBegin; 5452 /* currently sctx is memory bleed */ 5453 ierr = PetscMalloc(sizeof(TSMatlabContext),&sctx);CHKERRQ(ierr); 5454 ierr = PetscStrallocpy(func,&sctx->funcname);CHKERRQ(ierr); 5455 /* 5456 This should work, but it doesn't 5457 sctx->ctx = ctx; 5458 mexMakeArrayPersistent(sctx->ctx); 5459 */ 5460 sctx->ctx = mxDuplicateArray(ctx); 5461 5462 ierr = TSSetIJacobian(ts,A,B,TSComputeJacobian_Matlab,sctx);CHKERRQ(ierr); 5463 PetscFunctionReturn(0); 5464 } 5465 5466 #undef __FUNCT__ 5467 #define __FUNCT__ "TSMonitor_Matlab" 5468 /* 5469 TSMonitor_Matlab - Calls the function that has been set with TSMonitorSetMatlab(). 5470 5471 Collective on TS 5472 5473 .seealso: TSSetFunction(), TSGetFunction() 5474 @*/ 5475 PetscErrorCode TSMonitor_Matlab(TS ts,PetscInt it, PetscReal time,Vec u, void *ctx) 5476 { 5477 PetscErrorCode ierr; 5478 TSMatlabContext *sctx = (TSMatlabContext*)ctx; 5479 int nlhs = 1,nrhs = 6; 5480 mxArray *plhs[1],*prhs[6]; 5481 long long int lx = 0,ls = 0; 5482 5483 PetscFunctionBegin; 5484 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 5485 PetscValidHeaderSpecific(u,VEC_CLASSID,4); 5486 5487 ierr = PetscMemcpy(&ls,&ts,sizeof(ts));CHKERRQ(ierr); 5488 ierr = PetscMemcpy(&lx,&u,sizeof(u));CHKERRQ(ierr); 5489 5490 prhs[0] = mxCreateDoubleScalar((double)ls); 5491 prhs[1] = mxCreateDoubleScalar((double)it); 5492 prhs[2] = mxCreateDoubleScalar((double)time); 5493 prhs[3] = mxCreateDoubleScalar((double)lx); 5494 prhs[4] = mxCreateString(sctx->funcname); 5495 prhs[5] = sctx->ctx; 5496 ierr = mexCallMATLAB(nlhs,plhs,nrhs,prhs,"PetscTSMonitorInternal");CHKERRQ(ierr); 5497 ierr = mxGetScalar(plhs[0]);CHKERRQ(ierr); 5498 mxDestroyArray(prhs[0]); 5499 mxDestroyArray(prhs[1]); 5500 mxDestroyArray(prhs[2]); 5501 mxDestroyArray(prhs[3]); 5502 mxDestroyArray(prhs[4]); 5503 mxDestroyArray(plhs[0]); 5504 PetscFunctionReturn(0); 5505 } 5506 5507 5508 #undef __FUNCT__ 5509 #define __FUNCT__ "TSMonitorSetMatlab" 5510 /* 5511 TSMonitorSetMatlab - Sets the monitor function from Matlab 5512 5513 Level: developer 5514 5515 .keywords: TS, nonlinear, set, function 5516 5517 .seealso: TSGetFunction(), TSComputeFunction(), TSSetJacobian(), TSSetFunction() 5518 */ 5519 PetscErrorCode TSMonitorSetMatlab(TS ts,const char *func,mxArray *ctx) 5520 { 5521 PetscErrorCode ierr; 5522 TSMatlabContext *sctx; 5523 5524 PetscFunctionBegin; 5525 /* currently sctx is memory bleed */ 5526 ierr = PetscMalloc(sizeof(TSMatlabContext),&sctx);CHKERRQ(ierr); 5527 ierr = PetscStrallocpy(func,&sctx->funcname);CHKERRQ(ierr); 5528 /* 5529 This should work, but it doesn't 5530 sctx->ctx = ctx; 5531 mexMakeArrayPersistent(sctx->ctx); 5532 */ 5533 sctx->ctx = mxDuplicateArray(ctx); 5534 5535 ierr = TSMonitorSet(ts,TSMonitor_Matlab,sctx,NULL);CHKERRQ(ierr); 5536 PetscFunctionReturn(0); 5537 } 5538 #endif 5539 5540 #undef __FUNCT__ 5541 #define __FUNCT__ "TSMonitorLGSolution" 5542 /*@C 5543 TSMonitorLGSolution - Monitors progress of the TS solvers by plotting each component of the solution vector 5544 in a time based line graph 5545 5546 Collective on TS 5547 5548 Input Parameters: 5549 + ts - the TS context 5550 . step - current time-step 5551 . ptime - current time 5552 - lg - a line graph object 5553 5554 Options Database: 5555 . -ts_monitor_lg_solution_variables 5556 5557 Level: intermediate 5558 5559 Notes: each process in a parallel run displays its component solutions in a separate window 5560 5561 .keywords: TS, vector, monitor, view 5562 5563 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView() 5564 @*/ 5565 PetscErrorCode TSMonitorLGSolution(TS ts,PetscInt step,PetscReal ptime,Vec u,void *dummy) 5566 { 5567 PetscErrorCode ierr; 5568 TSMonitorLGCtx ctx = (TSMonitorLGCtx)dummy; 5569 const PetscScalar *yy; 5570 PetscInt dim; 5571 Vec v; 5572 5573 PetscFunctionBegin; 5574 if (!step) { 5575 PetscDrawAxis axis; 5576 ierr = PetscDrawLGGetAxis(ctx->lg,&axis);CHKERRQ(ierr); 5577 ierr = PetscDrawAxisSetLabels(axis,"Solution as function of time","Time","Solution");CHKERRQ(ierr); 5578 if (ctx->names && !ctx->displaynames) { 5579 char **displaynames; 5580 PetscBool flg; 5581 5582 ierr = VecGetLocalSize(u,&dim);CHKERRQ(ierr); 5583 ierr = PetscMalloc((dim+1)*sizeof(char*),&displaynames);CHKERRQ(ierr); 5584 ierr = PetscMemzero(displaynames,(dim+1)*sizeof(char*));CHKERRQ(ierr); 5585 ierr = PetscOptionsGetStringArray(((PetscObject)ts)->prefix,"-ts_monitor_lg_solution_variables",displaynames,&dim,&flg);CHKERRQ(ierr); 5586 if (flg) { 5587 ierr = TSMonitorLGCtxSetDisplayVariables(ctx,(const char *const *)displaynames);CHKERRQ(ierr); 5588 } 5589 ierr = PetscStrArrayDestroy(&displaynames);CHKERRQ(ierr); 5590 } 5591 if (ctx->displaynames) { 5592 ierr = PetscDrawLGSetDimension(ctx->lg,ctx->ndisplayvariables);CHKERRQ(ierr); 5593 ierr = PetscDrawLGSetLegend(ctx->lg,(const char *const *)ctx->displaynames);CHKERRQ(ierr); 5594 } else if (ctx->names) { 5595 ierr = VecGetLocalSize(u,&dim);CHKERRQ(ierr); 5596 ierr = PetscDrawLGSetDimension(ctx->lg,dim);CHKERRQ(ierr); 5597 ierr = PetscDrawLGSetLegend(ctx->lg,(const char *const *)ctx->names);CHKERRQ(ierr); 5598 } 5599 ierr = PetscDrawLGReset(ctx->lg);CHKERRQ(ierr); 5600 } 5601 if (ctx->transform) { 5602 ierr = (*ctx->transform)(ctx->transformctx,u,&v);CHKERRQ(ierr); 5603 } else { 5604 v = u; 5605 } 5606 ierr = VecGetArrayRead(v,&yy);CHKERRQ(ierr); 5607 #if defined(PETSC_USE_COMPLEX) 5608 { 5609 PetscReal *yreal; 5610 PetscInt i,n; 5611 ierr = VecGetLocalSize(v,&n);CHKERRQ(ierr); 5612 ierr = PetscMalloc1(n,&yreal);CHKERRQ(ierr); 5613 for (i=0; i<n; i++) yreal[i] = PetscRealPart(yy[i]); 5614 ierr = PetscDrawLGAddCommonPoint(ctx->lg,ptime,yreal);CHKERRQ(ierr); 5615 ierr = PetscFree(yreal);CHKERRQ(ierr); 5616 } 5617 #else 5618 if (ctx->displaynames) { 5619 PetscInt i; 5620 for (i=0; i<ctx->ndisplayvariables; i++) { 5621 ctx->displayvalues[i] = yy[ctx->displayvariables[i]]; 5622 } 5623 ierr = PetscDrawLGAddCommonPoint(ctx->lg,ptime,ctx->displayvalues);CHKERRQ(ierr); 5624 } else { 5625 ierr = PetscDrawLGAddCommonPoint(ctx->lg,ptime,yy);CHKERRQ(ierr); 5626 } 5627 #endif 5628 ierr = VecRestoreArrayRead(v,&yy);CHKERRQ(ierr); 5629 if (ctx->transform) { 5630 ierr = VecDestroy(&v);CHKERRQ(ierr); 5631 } 5632 if (((ctx->howoften > 0) && (!(step % ctx->howoften))) || ((ctx->howoften == -1) && ts->reason)) { 5633 ierr = PetscDrawLGDraw(ctx->lg);CHKERRQ(ierr); 5634 } 5635 PetscFunctionReturn(0); 5636 } 5637 5638 5639 #undef __FUNCT__ 5640 #define __FUNCT__ "TSMonitorLGSetVariableNames" 5641 /*@C 5642 TSMonitorLGSetVariableNames - Sets the name of each component in the solution vector so that it may be displayed in the plot 5643 5644 Collective on TS 5645 5646 Input Parameters: 5647 + ts - the TS context 5648 - names - the names of the components, final string must be NULL 5649 5650 Level: intermediate 5651 5652 .keywords: TS, vector, monitor, view 5653 5654 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorLGSetDisplayVariables(), TSMonitorLGCtxSetVariableNames() 5655 @*/ 5656 PetscErrorCode TSMonitorLGSetVariableNames(TS ts,const char * const *names) 5657 { 5658 PetscErrorCode ierr; 5659 PetscInt i; 5660 5661 PetscFunctionBegin; 5662 for (i=0; i<ts->numbermonitors; i++) { 5663 if (ts->monitor[i] == TSMonitorLGSolution) { 5664 ierr = TSMonitorLGCtxSetVariableNames((TSMonitorLGCtx)ts->monitorcontext[i],names);CHKERRQ(ierr); 5665 break; 5666 } 5667 } 5668 PetscFunctionReturn(0); 5669 } 5670 5671 #undef __FUNCT__ 5672 #define __FUNCT__ "TSMonitorLGCtxSetVariableNames" 5673 /*@C 5674 TSMonitorLGCtxSetVariableNames - Sets the name of each component in the solution vector so that it may be displayed in the plot 5675 5676 Collective on TS 5677 5678 Input Parameters: 5679 + ts - the TS context 5680 - names - the names of the components, final string must be NULL 5681 5682 Level: intermediate 5683 5684 .keywords: TS, vector, monitor, view 5685 5686 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorLGSetDisplayVariables(), TSMonitorLGSetVariableNames() 5687 @*/ 5688 PetscErrorCode TSMonitorLGCtxSetVariableNames(TSMonitorLGCtx ctx,const char * const *names) 5689 { 5690 PetscErrorCode ierr; 5691 5692 PetscFunctionBegin; 5693 ierr = PetscStrArrayDestroy(&ctx->names);CHKERRQ(ierr); 5694 ierr = PetscStrArrayallocpy(names,&ctx->names);CHKERRQ(ierr); 5695 PetscFunctionReturn(0); 5696 } 5697 5698 #undef __FUNCT__ 5699 #define __FUNCT__ "TSMonitorLGGetVariableNames" 5700 /*@C 5701 TSMonitorLGGetVariableNames - Gets the name of each component in the solution vector so that it may be displayed in the plot 5702 5703 Collective on TS 5704 5705 Input Parameter: 5706 . ts - the TS context 5707 5708 Output Parameter: 5709 . names - the names of the components, final string must be NULL 5710 5711 Level: intermediate 5712 5713 .keywords: TS, vector, monitor, view 5714 5715 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorLGSetDisplayVariables() 5716 @*/ 5717 PetscErrorCode TSMonitorLGGetVariableNames(TS ts,const char *const **names) 5718 { 5719 PetscInt i; 5720 5721 PetscFunctionBegin; 5722 *names = NULL; 5723 for (i=0; i<ts->numbermonitors; i++) { 5724 if (ts->monitor[i] == TSMonitorLGSolution) { 5725 TSMonitorLGCtx ctx = (TSMonitorLGCtx) ts->monitorcontext[i]; 5726 *names = (const char *const *)ctx->names; 5727 break; 5728 } 5729 } 5730 PetscFunctionReturn(0); 5731 } 5732 5733 #undef __FUNCT__ 5734 #define __FUNCT__ "TSMonitorLGCtxSetDisplayVariables" 5735 /*@C 5736 TSMonitorLGCtxSetDisplayVariables - Sets the variables that are to be display in the monitor 5737 5738 Collective on TS 5739 5740 Input Parameters: 5741 + ctx - the TSMonitorLG context 5742 . displaynames - the names of the components, final string must be NULL 5743 5744 Level: intermediate 5745 5746 .keywords: TS, vector, monitor, view 5747 5748 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorLGSetVariableNames() 5749 @*/ 5750 PetscErrorCode TSMonitorLGCtxSetDisplayVariables(TSMonitorLGCtx ctx,const char * const *displaynames) 5751 { 5752 PetscInt j = 0,k; 5753 PetscErrorCode ierr; 5754 5755 PetscFunctionBegin; 5756 if (!ctx->names) PetscFunctionReturn(0); 5757 ierr = PetscStrArrayDestroy(&ctx->displaynames);CHKERRQ(ierr); 5758 ierr = PetscStrArrayallocpy(displaynames,&ctx->displaynames);CHKERRQ(ierr); 5759 while (displaynames[j]) j++; 5760 ctx->ndisplayvariables = j; 5761 ierr = PetscMalloc1(ctx->ndisplayvariables,&ctx->displayvariables);CHKERRQ(ierr); 5762 ierr = PetscMalloc1(ctx->ndisplayvariables,&ctx->displayvalues);CHKERRQ(ierr); 5763 j = 0; 5764 while (displaynames[j]) { 5765 k = 0; 5766 while (ctx->names[k]) { 5767 PetscBool flg; 5768 ierr = PetscStrcmp(displaynames[j],ctx->names[k],&flg);CHKERRQ(ierr); 5769 if (flg) { 5770 ctx->displayvariables[j] = k; 5771 break; 5772 } 5773 k++; 5774 } 5775 j++; 5776 } 5777 PetscFunctionReturn(0); 5778 } 5779 5780 5781 #undef __FUNCT__ 5782 #define __FUNCT__ "TSMonitorLGSetDisplayVariables" 5783 /*@C 5784 TSMonitorLGSetDisplayVariables - Sets the variables that are to be display in the monitor 5785 5786 Collective on TS 5787 5788 Input Parameters: 5789 + ts - the TS context 5790 . displaynames - the names of the components, final string must be NULL 5791 5792 Level: intermediate 5793 5794 .keywords: TS, vector, monitor, view 5795 5796 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorLGSetVariableNames() 5797 @*/ 5798 PetscErrorCode TSMonitorLGSetDisplayVariables(TS ts,const char * const *displaynames) 5799 { 5800 PetscInt i; 5801 PetscErrorCode ierr; 5802 5803 PetscFunctionBegin; 5804 for (i=0; i<ts->numbermonitors; i++) { 5805 if (ts->monitor[i] == TSMonitorLGSolution) { 5806 ierr = TSMonitorLGCtxSetDisplayVariables((TSMonitorLGCtx)ts->monitorcontext[i],displaynames);CHKERRQ(ierr); 5807 break; 5808 } 5809 } 5810 PetscFunctionReturn(0); 5811 } 5812 5813 #undef __FUNCT__ 5814 #define __FUNCT__ "TSMonitorLGSetTransform" 5815 /*@C 5816 TSMonitorLGSetTransform - Solution vector will be transformed by provided function before being displayed 5817 5818 Collective on TS 5819 5820 Input Parameters: 5821 + ts - the TS context 5822 . transform - the transform function 5823 . destroy - function to destroy the optional context 5824 - ctx - optional context used by transform function 5825 5826 Level: intermediate 5827 5828 .keywords: TS, vector, monitor, view 5829 5830 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorLGSetVariableNames(), TSMonitorLGCtxSetTransform() 5831 @*/ 5832 PetscErrorCode TSMonitorLGSetTransform(TS ts,PetscErrorCode (*transform)(void*,Vec,Vec*),PetscErrorCode (*destroy)(void*),void *tctx) 5833 { 5834 PetscInt i; 5835 PetscErrorCode ierr; 5836 5837 PetscFunctionBegin; 5838 for (i=0; i<ts->numbermonitors; i++) { 5839 if (ts->monitor[i] == TSMonitorLGSolution) { 5840 ierr = TSMonitorLGCtxSetTransform((TSMonitorLGCtx)ts->monitorcontext[i],transform,destroy,tctx);CHKERRQ(ierr); 5841 } 5842 } 5843 PetscFunctionReturn(0); 5844 } 5845 5846 #undef __FUNCT__ 5847 #define __FUNCT__ "TSMonitorLGCtxSetTransform" 5848 /*@C 5849 TSMonitorLGCtxSetTransform - Solution vector will be transformed by provided function before being displayed 5850 5851 Collective on TSLGCtx 5852 5853 Input Parameters: 5854 + ts - the TS context 5855 . transform - the transform function 5856 . destroy - function to destroy the optional context 5857 - ctx - optional context used by transform function 5858 5859 Level: intermediate 5860 5861 .keywords: TS, vector, monitor, view 5862 5863 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorLGSetVariableNames(), TSMonitorLGSetTransform() 5864 @*/ 5865 PetscErrorCode TSMonitorLGCtxSetTransform(TSMonitorLGCtx ctx,PetscErrorCode (*transform)(void*,Vec,Vec*),PetscErrorCode (*destroy)(void*),void *tctx) 5866 { 5867 PetscFunctionBegin; 5868 ctx->transform = transform; 5869 ctx->transformdestroy = destroy; 5870 ctx->transformctx = tctx; 5871 PetscFunctionReturn(0); 5872 } 5873 5874 #undef __FUNCT__ 5875 #define __FUNCT__ "TSMonitorLGError" 5876 /*@C 5877 TSMonitorLGError - Monitors progress of the TS solvers by plotting each component of the solution vector 5878 in a time based line graph 5879 5880 Collective on TS 5881 5882 Input Parameters: 5883 + ts - the TS context 5884 . step - current time-step 5885 . ptime - current time 5886 - lg - a line graph object 5887 5888 Level: intermediate 5889 5890 Notes: 5891 Only for sequential solves. 5892 5893 The user must provide the solution using TSSetSolutionFunction() to use this monitor. 5894 5895 Options Database Keys: 5896 . -ts_monitor_lg_error - create a graphical monitor of error history 5897 5898 .keywords: TS, vector, monitor, view 5899 5900 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSSetSolutionFunction() 5901 @*/ 5902 PetscErrorCode TSMonitorLGError(TS ts,PetscInt step,PetscReal ptime,Vec u,void *dummy) 5903 { 5904 PetscErrorCode ierr; 5905 TSMonitorLGCtx ctx = (TSMonitorLGCtx)dummy; 5906 const PetscScalar *yy; 5907 Vec y; 5908 PetscInt dim; 5909 5910 PetscFunctionBegin; 5911 if (!step) { 5912 PetscDrawAxis axis; 5913 ierr = PetscDrawLGGetAxis(ctx->lg,&axis);CHKERRQ(ierr); 5914 ierr = PetscDrawAxisSetLabels(axis,"Error in solution as function of time","Time","Solution");CHKERRQ(ierr); 5915 ierr = VecGetLocalSize(u,&dim);CHKERRQ(ierr); 5916 ierr = PetscDrawLGSetDimension(ctx->lg,dim);CHKERRQ(ierr); 5917 ierr = PetscDrawLGReset(ctx->lg);CHKERRQ(ierr); 5918 } 5919 ierr = VecDuplicate(u,&y);CHKERRQ(ierr); 5920 ierr = TSComputeSolutionFunction(ts,ptime,y);CHKERRQ(ierr); 5921 ierr = VecAXPY(y,-1.0,u);CHKERRQ(ierr); 5922 ierr = VecGetArrayRead(y,&yy);CHKERRQ(ierr); 5923 #if defined(PETSC_USE_COMPLEX) 5924 { 5925 PetscReal *yreal; 5926 PetscInt i,n; 5927 ierr = VecGetLocalSize(y,&n);CHKERRQ(ierr); 5928 ierr = PetscMalloc1(n,&yreal);CHKERRQ(ierr); 5929 for (i=0; i<n; i++) yreal[i] = PetscRealPart(yy[i]); 5930 ierr = PetscDrawLGAddCommonPoint(ctx->lg,ptime,yreal);CHKERRQ(ierr); 5931 ierr = PetscFree(yreal);CHKERRQ(ierr); 5932 } 5933 #else 5934 ierr = PetscDrawLGAddCommonPoint(ctx->lg,ptime,yy);CHKERRQ(ierr); 5935 #endif 5936 ierr = VecRestoreArrayRead(y,&yy);CHKERRQ(ierr); 5937 ierr = VecDestroy(&y);CHKERRQ(ierr); 5938 if (((ctx->howoften > 0) && (!(step % ctx->howoften))) || ((ctx->howoften == -1) && ts->reason)) { 5939 ierr = PetscDrawLGDraw(ctx->lg);CHKERRQ(ierr); 5940 } 5941 PetscFunctionReturn(0); 5942 } 5943 5944 #undef __FUNCT__ 5945 #define __FUNCT__ "TSMonitorLGSNESIterations" 5946 PetscErrorCode TSMonitorLGSNESIterations(TS ts,PetscInt n,PetscReal ptime,Vec v,void *monctx) 5947 { 5948 TSMonitorLGCtx ctx = (TSMonitorLGCtx) monctx; 5949 PetscReal x = ptime,y; 5950 PetscErrorCode ierr; 5951 PetscInt its; 5952 5953 PetscFunctionBegin; 5954 if (!n) { 5955 PetscDrawAxis axis; 5956 5957 ierr = PetscDrawLGGetAxis(ctx->lg,&axis);CHKERRQ(ierr); 5958 ierr = PetscDrawAxisSetLabels(axis,"Nonlinear iterations as function of time","Time","SNES Iterations");CHKERRQ(ierr); 5959 ierr = PetscDrawLGReset(ctx->lg);CHKERRQ(ierr); 5960 5961 ctx->snes_its = 0; 5962 } 5963 ierr = TSGetSNESIterations(ts,&its);CHKERRQ(ierr); 5964 y = its - ctx->snes_its; 5965 ierr = PetscDrawLGAddPoint(ctx->lg,&x,&y);CHKERRQ(ierr); 5966 if (((ctx->howoften > 0) && (!(n % ctx->howoften)) && (n > -1)) || ((ctx->howoften == -1) && (n == -1))) { 5967 ierr = PetscDrawLGDraw(ctx->lg);CHKERRQ(ierr); 5968 } 5969 ctx->snes_its = its; 5970 PetscFunctionReturn(0); 5971 } 5972 5973 #undef __FUNCT__ 5974 #define __FUNCT__ "TSMonitorLGKSPIterations" 5975 PetscErrorCode TSMonitorLGKSPIterations(TS ts,PetscInt n,PetscReal ptime,Vec v,void *monctx) 5976 { 5977 TSMonitorLGCtx ctx = (TSMonitorLGCtx) monctx; 5978 PetscReal x = ptime,y; 5979 PetscErrorCode ierr; 5980 PetscInt its; 5981 5982 PetscFunctionBegin; 5983 if (!n) { 5984 PetscDrawAxis axis; 5985 5986 ierr = PetscDrawLGGetAxis(ctx->lg,&axis);CHKERRQ(ierr); 5987 ierr = PetscDrawAxisSetLabels(axis,"Linear iterations as function of time","Time","KSP Iterations");CHKERRQ(ierr); 5988 ierr = PetscDrawLGReset(ctx->lg);CHKERRQ(ierr); 5989 5990 ctx->ksp_its = 0; 5991 } 5992 ierr = TSGetKSPIterations(ts,&its);CHKERRQ(ierr); 5993 y = its - ctx->ksp_its; 5994 ierr = PetscDrawLGAddPoint(ctx->lg,&x,&y);CHKERRQ(ierr); 5995 if (((ctx->howoften > 0) && (!(n % ctx->howoften)) && (n > -1)) || ((ctx->howoften == -1) && (n == -1))) { 5996 ierr = PetscDrawLGDraw(ctx->lg);CHKERRQ(ierr); 5997 } 5998 ctx->ksp_its = its; 5999 PetscFunctionReturn(0); 6000 } 6001 6002 #undef __FUNCT__ 6003 #define __FUNCT__ "TSComputeLinearStability" 6004 /*@ 6005 TSComputeLinearStability - computes the linear stability function at a point 6006 6007 Collective on TS and Vec 6008 6009 Input Parameters: 6010 + ts - the TS context 6011 - xr,xi - real and imaginary part of input arguments 6012 6013 Output Parameters: 6014 . yr,yi - real and imaginary part of function value 6015 6016 Level: developer 6017 6018 .keywords: TS, compute 6019 6020 .seealso: TSSetRHSFunction(), TSComputeIFunction() 6021 @*/ 6022 PetscErrorCode TSComputeLinearStability(TS ts,PetscReal xr,PetscReal xi,PetscReal *yr,PetscReal *yi) 6023 { 6024 PetscErrorCode ierr; 6025 6026 PetscFunctionBegin; 6027 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 6028 if (!ts->ops->linearstability) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_SUP,"Linearized stability function not provided for this method"); 6029 ierr = (*ts->ops->linearstability)(ts,xr,xi,yr,yi);CHKERRQ(ierr); 6030 PetscFunctionReturn(0); 6031 } 6032 6033 /* ------------------------------------------------------------------------*/ 6034 #undef __FUNCT__ 6035 #define __FUNCT__ "TSMonitorEnvelopeCtxCreate" 6036 /*@C 6037 TSMonitorEnvelopeCtxCreate - Creates a context for use with TSMonitorEnvelope() 6038 6039 Collective on TS 6040 6041 Input Parameters: 6042 . ts - the ODE solver object 6043 6044 Output Parameter: 6045 . ctx - the context 6046 6047 Level: intermediate 6048 6049 .keywords: TS, monitor, line graph, residual, seealso 6050 6051 .seealso: TSMonitorLGTimeStep(), TSMonitorSet(), TSMonitorLGSolution(), TSMonitorLGError() 6052 6053 @*/ 6054 PetscErrorCode TSMonitorEnvelopeCtxCreate(TS ts,TSMonitorEnvelopeCtx *ctx) 6055 { 6056 PetscErrorCode ierr; 6057 6058 PetscFunctionBegin; 6059 ierr = PetscNew(ctx);CHKERRQ(ierr); 6060 PetscFunctionReturn(0); 6061 } 6062 6063 #undef __FUNCT__ 6064 #define __FUNCT__ "TSMonitorEnvelope" 6065 /*@C 6066 TSMonitorEnvelope - Monitors the maximum and minimum value of each component of the solution 6067 6068 Collective on TS 6069 6070 Input Parameters: 6071 + ts - the TS context 6072 . step - current time-step 6073 . ptime - current time 6074 - ctx - the envelope context 6075 6076 Options Database: 6077 . -ts_monitor_envelope 6078 6079 Level: intermediate 6080 6081 Notes: after a solve you can use TSMonitorEnvelopeGetBounds() to access the envelope 6082 6083 .keywords: TS, vector, monitor, view 6084 6085 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorEnvelopeGetBounds() 6086 @*/ 6087 PetscErrorCode TSMonitorEnvelope(TS ts,PetscInt step,PetscReal ptime,Vec u,void *dummy) 6088 { 6089 PetscErrorCode ierr; 6090 TSMonitorEnvelopeCtx ctx = (TSMonitorEnvelopeCtx)dummy; 6091 6092 PetscFunctionBegin; 6093 if (!ctx->max) { 6094 ierr = VecDuplicate(u,&ctx->max);CHKERRQ(ierr); 6095 ierr = VecDuplicate(u,&ctx->min);CHKERRQ(ierr); 6096 ierr = VecCopy(u,ctx->max);CHKERRQ(ierr); 6097 ierr = VecCopy(u,ctx->min);CHKERRQ(ierr); 6098 } else { 6099 ierr = VecPointwiseMax(ctx->max,u,ctx->max);CHKERRQ(ierr); 6100 ierr = VecPointwiseMin(ctx->min,u,ctx->min);CHKERRQ(ierr); 6101 } 6102 PetscFunctionReturn(0); 6103 } 6104 6105 6106 #undef __FUNCT__ 6107 #define __FUNCT__ "TSMonitorEnvelopeGetBounds" 6108 /*@C 6109 TSMonitorEnvelopeGetBounds - Gets the bounds for the components of the solution 6110 6111 Collective on TS 6112 6113 Input Parameter: 6114 . ts - the TS context 6115 6116 Output Parameter: 6117 + max - the maximum values 6118 - min - the minimum values 6119 6120 Level: intermediate 6121 6122 .keywords: TS, vector, monitor, view 6123 6124 .seealso: TSMonitorSet(), TSMonitorDefault(), VecView(), TSMonitorLGSetDisplayVariables() 6125 @*/ 6126 PetscErrorCode TSMonitorEnvelopeGetBounds(TS ts,Vec *max,Vec *min) 6127 { 6128 PetscInt i; 6129 6130 PetscFunctionBegin; 6131 if (max) *max = NULL; 6132 if (min) *min = NULL; 6133 for (i=0; i<ts->numbermonitors; i++) { 6134 if (ts->monitor[i] == TSMonitorEnvelope) { 6135 TSMonitorEnvelopeCtx ctx = (TSMonitorEnvelopeCtx) ts->monitorcontext[i]; 6136 if (max) *max = ctx->max; 6137 if (min) *min = ctx->min; 6138 break; 6139 } 6140 } 6141 PetscFunctionReturn(0); 6142 } 6143 6144 #undef __FUNCT__ 6145 #define __FUNCT__ "TSMonitorEnvelopeCtxDestroy" 6146 /*@C 6147 TSMonitorEnvelopeCtxDestroy - Destroys a context that was created with TSMonitorEnvelopeCtxCreate(). 6148 6149 Collective on TSMonitorEnvelopeCtx 6150 6151 Input Parameter: 6152 . ctx - the monitor context 6153 6154 Level: intermediate 6155 6156 .keywords: TS, monitor, line graph, destroy 6157 6158 .seealso: TSMonitorLGCtxCreate(), TSMonitorSet(), TSMonitorLGTimeStep(); 6159 @*/ 6160 PetscErrorCode TSMonitorEnvelopeCtxDestroy(TSMonitorEnvelopeCtx *ctx) 6161 { 6162 PetscErrorCode ierr; 6163 6164 PetscFunctionBegin; 6165 ierr = VecDestroy(&(*ctx)->min);CHKERRQ(ierr); 6166 ierr = VecDestroy(&(*ctx)->max);CHKERRQ(ierr); 6167 ierr = PetscFree(*ctx);CHKERRQ(ierr); 6168 PetscFunctionReturn(0); 6169 } 6170 6171 #undef __FUNCT__ 6172 #define __FUNCT__ "TSRollBack" 6173 /*@ 6174 TSRollBack - Rolls back one time step 6175 6176 Collective on TS 6177 6178 Input Parameter: 6179 . ts - the TS context obtained from TSCreate() 6180 6181 Level: advanced 6182 6183 .keywords: TS, timestep, rollback 6184 6185 .seealso: TSCreate(), TSSetUp(), TSDestroy(), TSSolve(), TSSetPreStep(), TSSetPreStage(), TSInterpolate() 6186 @*/ 6187 PetscErrorCode TSRollBack(TS ts) 6188 { 6189 PetscErrorCode ierr; 6190 6191 PetscFunctionBegin; 6192 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 6193 6194 if (!ts->ops->rollback) SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_SUP,"TSRollBack not implemented for type '%s'",((PetscObject)ts)->type_name); 6195 ierr = (*ts->ops->rollback)(ts);CHKERRQ(ierr); 6196 ts->time_step = ts->ptime - ts->ptime_prev; 6197 ts->ptime = ts->ptime_prev; 6198 ts->steprollback = PETSC_TRUE; /* Flag to indicate that the step is rollbacked */ 6199 PetscFunctionReturn(0); 6200 } 6201 6202 #undef __FUNCT__ 6203 #define __FUNCT__ "TSGetStages" 6204 /*@ 6205 TSGetStages - Get the number of stages and stage values 6206 6207 Input Parameter: 6208 . ts - the TS context obtained from TSCreate() 6209 6210 Level: advanced 6211 6212 .keywords: TS, getstages 6213 6214 .seealso: TSCreate() 6215 @*/ 6216 PetscErrorCode TSGetStages(TS ts,PetscInt *ns, Vec **Y) 6217 { 6218 PetscErrorCode ierr; 6219 6220 PetscFunctionBegin; 6221 PetscValidHeaderSpecific(ts, TS_CLASSID,1); 6222 PetscValidPointer(ns,2); 6223 6224 if (!ts->ops->getstages) *ns=0; 6225 else { 6226 ierr = (*ts->ops->getstages)(ts,ns,Y);CHKERRQ(ierr); 6227 } 6228 PetscFunctionReturn(0); 6229 } 6230 6231 #undef __FUNCT__ 6232 #define __FUNCT__ "TSComputeIJacobianDefaultColor" 6233 /*@C 6234 TSComputeIJacobianDefaultColor - Computes the Jacobian using finite differences and coloring to exploit matrix sparsity. 6235 6236 Collective on SNES 6237 6238 Input Parameters: 6239 + ts - the TS context 6240 . t - current timestep 6241 . U - state vector 6242 . Udot - time derivative of state vector 6243 . shift - shift to apply, see note below 6244 - ctx - an optional user context 6245 6246 Output Parameters: 6247 + J - Jacobian matrix (not altered in this routine) 6248 - B - newly computed Jacobian matrix to use with preconditioner (generally the same as J) 6249 6250 Level: intermediate 6251 6252 Notes: 6253 If F(t,U,Udot)=0 is the DAE, the required Jacobian is 6254 6255 dF/dU + shift*dF/dUdot 6256 6257 Most users should not need to explicitly call this routine, as it 6258 is used internally within the nonlinear solvers. 6259 6260 This will first try to get the coloring from the DM. If the DM type has no coloring 6261 routine, then it will try to get the coloring from the matrix. This requires that the 6262 matrix have nonzero entries precomputed. 6263 6264 .keywords: TS, finite differences, Jacobian, coloring, sparse 6265 .seealso: TSSetIJacobian(), MatFDColoringCreate(), MatFDColoringSetFunction() 6266 @*/ 6267 PetscErrorCode TSComputeIJacobianDefaultColor(TS ts,PetscReal t,Vec U,Vec Udot,PetscReal shift,Mat J,Mat B,void *ctx) 6268 { 6269 SNES snes; 6270 MatFDColoring color; 6271 PetscBool hascolor, matcolor = PETSC_FALSE; 6272 PetscErrorCode ierr; 6273 6274 PetscFunctionBegin; 6275 ierr = PetscOptionsGetBool(((PetscObject) ts)->prefix, "-ts_fd_color_use_mat", &matcolor, NULL);CHKERRQ(ierr); 6276 ierr = PetscObjectQuery((PetscObject) B, "TSMatFDColoring", (PetscObject *) &color);CHKERRQ(ierr); 6277 if (!color) { 6278 DM dm; 6279 ISColoring iscoloring; 6280 6281 ierr = TSGetDM(ts, &dm);CHKERRQ(ierr); 6282 ierr = DMHasColoring(dm, &hascolor);CHKERRQ(ierr); 6283 if (hascolor && !matcolor) { 6284 ierr = DMCreateColoring(dm, IS_COLORING_GLOBAL, &iscoloring);CHKERRQ(ierr); 6285 ierr = MatFDColoringCreate(B, iscoloring, &color);CHKERRQ(ierr); 6286 ierr = MatFDColoringSetFunction(color, (PetscErrorCode (*)(void)) SNESTSFormFunction, (void *) ts);CHKERRQ(ierr); 6287 ierr = MatFDColoringSetFromOptions(color);CHKERRQ(ierr); 6288 ierr = MatFDColoringSetUp(B, iscoloring, color);CHKERRQ(ierr); 6289 ierr = ISColoringDestroy(&iscoloring);CHKERRQ(ierr); 6290 } else { 6291 MatColoring mc; 6292 6293 ierr = MatColoringCreate(B, &mc);CHKERRQ(ierr); 6294 ierr = MatColoringSetDistance(mc, 2);CHKERRQ(ierr); 6295 ierr = MatColoringSetType(mc, MATCOLORINGSL);CHKERRQ(ierr); 6296 ierr = MatColoringSetFromOptions(mc);CHKERRQ(ierr); 6297 ierr = MatColoringApply(mc, &iscoloring);CHKERRQ(ierr); 6298 ierr = MatColoringDestroy(&mc);CHKERRQ(ierr); 6299 ierr = MatFDColoringCreate(B, iscoloring, &color);CHKERRQ(ierr); 6300 ierr = MatFDColoringSetFunction(color, (PetscErrorCode (*)(void)) SNESTSFormFunction, (void *) ts);CHKERRQ(ierr); 6301 ierr = MatFDColoringSetFromOptions(color);CHKERRQ(ierr); 6302 ierr = MatFDColoringSetUp(B, iscoloring, color);CHKERRQ(ierr); 6303 ierr = ISColoringDestroy(&iscoloring);CHKERRQ(ierr); 6304 } 6305 ierr = PetscObjectCompose((PetscObject) B, "TSMatFDColoring", (PetscObject) color);CHKERRQ(ierr); 6306 ierr = PetscObjectDereference((PetscObject) color);CHKERRQ(ierr); 6307 } 6308 ierr = TSGetSNES(ts, &snes);CHKERRQ(ierr); 6309 ierr = MatFDColoringApply(B, color, U, snes);CHKERRQ(ierr); 6310 if (J != B) { 6311 ierr = MatAssemblyBegin(J, MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 6312 ierr = MatAssemblyEnd(J, MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 6313 } 6314 PetscFunctionReturn(0); 6315 } 6316 6317 #undef __FUNCT__ 6318 #define __FUNCT__ "TSClone" 6319 /*@C 6320 TSClone - This function clones a time step object. 6321 6322 Collective on MPI_Comm 6323 6324 Input Parameter: 6325 . tsin - The input TS 6326 6327 Output Parameter: 6328 . tsout - The output TS (cloned) 6329 6330 Notes: 6331 This function is used to create a clone of a TS object. It is used in ARKIMEX for initializing the slope for first stage explicit methods. It will likely be replaced in the future with a mechanism of switching methods on the fly. 6332 6333 When using TSDestroy() on a clone the user has to first reset the correct TS reference in the embedded SNES object: e.g.: by running SNES snes_dup=NULL; TSGetSNES(ts,&snes_dup); ierr = TSSetSNES(ts,snes_dup); 6334 6335 Level: developer 6336 6337 .keywords: TS, clone 6338 .seealso: TSCreate(), TSSetType(), TSSetUp(), TSDestroy(), TSSetProblemType() 6339 @*/ 6340 PetscErrorCode TSClone(TS tsin, TS *tsout) 6341 { 6342 TS t; 6343 PetscErrorCode ierr; 6344 SNES snes_start; 6345 DM dm; 6346 TSType type; 6347 6348 PetscFunctionBegin; 6349 PetscValidPointer(tsin,1); 6350 *tsout = NULL; 6351 6352 ierr = PetscHeaderCreate(t, TS_CLASSID, "TS", "Time stepping", "TS", PetscObjectComm((PetscObject)tsin), TSDestroy, TSView);CHKERRQ(ierr); 6353 6354 /* General TS description */ 6355 t->numbermonitors = 0; 6356 t->setupcalled = 0; 6357 t->ksp_its = 0; 6358 t->snes_its = 0; 6359 t->nwork = 0; 6360 t->rhsjacobian.time = -1e20; 6361 t->rhsjacobian.scale = 1.; 6362 t->ijacobian.shift = 1.; 6363 6364 ierr = TSGetSNES(tsin,&snes_start); CHKERRQ(ierr); 6365 ierr = TSSetSNES(t,snes_start); CHKERRQ(ierr); 6366 6367 ierr = TSGetDM(tsin,&dm); CHKERRQ(ierr); 6368 ierr = TSSetDM(t,dm); CHKERRQ(ierr); 6369 6370 t->adapt=tsin->adapt; 6371 PetscObjectReference((PetscObject)t->adapt); 6372 6373 t->problem_type = tsin->problem_type; 6374 t->ptime = tsin->ptime; 6375 t->time_step = tsin->time_step; 6376 t->time_step_orig = tsin->time_step_orig; 6377 t->max_time = tsin->max_time; 6378 t->steps = tsin->steps; 6379 t->max_steps = tsin->max_steps; 6380 t->equation_type = tsin->equation_type; 6381 t->atol = tsin->atol; 6382 t->rtol = tsin->rtol; 6383 t->max_snes_failures = tsin->max_snes_failures; 6384 t->max_reject = tsin->max_reject; 6385 t->errorifstepfailed = tsin->errorifstepfailed; 6386 6387 ierr = TSGetType(tsin,&type); CHKERRQ(ierr); 6388 ierr = TSSetType(t,type); CHKERRQ(ierr); 6389 6390 t->vec_sol = NULL; 6391 6392 t->cfltime = tsin->cfltime; 6393 t->cfltime_local = tsin->cfltime_local; 6394 t->exact_final_time = tsin->exact_final_time; 6395 6396 ierr = PetscMemcpy(t->ops,tsin->ops,sizeof(struct _TSOps));CHKERRQ(ierr); 6397 6398 *tsout = t; 6399 PetscFunctionReturn(0); 6400 } 6401