1 |
C $Header$ |
C $Header$ |
2 |
|
C $Name$ |
3 |
|
|
4 |
#include "CPP_OPTIONS.h" |
#include "CPP_OPTIONS.h" |
5 |
|
|
6 |
#define VERBOSE |
CBOP |
7 |
|
C !ROUTINE: CG3D |
8 |
|
C !INTERFACE: |
9 |
|
SUBROUTINE CG3D( |
10 |
|
I cg3d_b, |
11 |
|
U cg3d_x, |
12 |
|
O firstResidual, |
13 |
|
O lastResidual, |
14 |
|
U numIters, |
15 |
|
I myIter, myThid ) |
16 |
|
C !DESCRIPTION: \bv |
17 |
|
C *==========================================================* |
18 |
|
C | SUBROUTINE CG3D |
19 |
|
C | o Three-dimensional grid problem conjugate-gradient |
20 |
|
C | inverter (with preconditioner). |
21 |
|
C *==========================================================* |
22 |
|
C | Con. grad is an iterative procedure for solving Ax = b. |
23 |
|
C | It requires the A be symmetric. |
24 |
|
C | This implementation assumes A is a seven-diagonal |
25 |
|
C | matrix of the form that arises in the discrete |
26 |
|
C | representation of the del^2 operator in a |
27 |
|
C | three-dimensional space. |
28 |
|
C | Notes: |
29 |
|
C | ====== |
30 |
|
C | This implementation can support shared-memory |
31 |
|
C | multi-threaded execution. In order to do this COMMON |
32 |
|
C | blocks are used for many of the arrays - even ones that |
33 |
|
C | are only used for intermedaite results. This design is |
34 |
|
C | OK if you want to all the threads to collaborate on |
35 |
|
C | solving the same problem. On the other hand if you want |
36 |
|
C | the threads to solve several different problems |
37 |
|
C | concurrently this implementation will not work. |
38 |
|
C *==========================================================* |
39 |
|
C \ev |
40 |
|
|
41 |
SUBROUTINE CG3D( |
C !USES: |
|
I myThid ) |
|
|
C /==========================================================\ |
|
|
C | SUBROUTINE CG3D | |
|
|
C | o Three-dimensional grid problem conjugate-gradient | |
|
|
C | inverter (with preconditioner). | |
|
|
C |==========================================================| |
|
|
C | Con. grad is an iterative procedure for solving Ax = b. | |
|
|
C | It requires the A be symmetric. | |
|
|
C | This implementation assumes A is a five-diagonal | |
|
|
C | matrix of the form that arises in the discrete | |
|
|
C | representation of the del^2 operator in a | |
|
|
C | two-dimensional space. | |
|
|
C | Notes: | |
|
|
C | ====== | |
|
|
C | This implementation can support shared-memory | |
|
|
C | multi-threaded execution. In order to do this COMMON | |
|
|
C | blocks are used for many of the arrays - even ones that | |
|
|
C | are only used for intermedaite results. This design is | |
|
|
C | OK if you want to all the threads to collaborate on | |
|
|
C | solving the same problem. On the other hand if you want | |
|
|
C | the threads to solve several different problems | |
|
|
C | concurrently this implementation will not work. | |
|
|
C \==========================================================/ |
|
42 |
IMPLICIT NONE |
IMPLICIT NONE |
|
|
|
43 |
C === Global data === |
C === Global data === |
44 |
#include "SIZE.h" |
#include "SIZE.h" |
45 |
#include "EEPARAMS.h" |
#include "EEPARAMS.h" |
46 |
#include "PARAMS.h" |
#include "PARAMS.h" |
47 |
#include "GRID.h" |
#include "GRID.h" |
48 |
|
#include "SURFACE.h" |
49 |
#include "CG3D.h" |
#include "CG3D.h" |
50 |
|
|
51 |
|
C !INPUT/OUTPUT PARAMETERS: |
52 |
C === Routine arguments === |
C === Routine arguments === |
53 |
C myThid - Thread on which I am working. |
C cg3d_b :: The source term or "right hand side" |
54 |
|
C cg3d_x :: The solution |
55 |
|
C firstResidual :: the initial residual before any iterations |
56 |
|
C lastResidual :: the actual residual reached |
57 |
|
C numIters :: Entry: the maximum number of iterations allowed |
58 |
|
C :: Exit: the actual number of iterations used |
59 |
|
C myIter :: Current iteration number in simulation |
60 |
|
C myThid :: my Thread Id number |
61 |
|
_RL cg3d_b(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr,nSx,nSy) |
62 |
|
_RL cg3d_x(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr,nSx,nSy) |
63 |
|
_RL firstResidual |
64 |
|
_RL lastResidual |
65 |
|
INTEGER numIters |
66 |
|
INTEGER myIter |
67 |
INTEGER myThid |
INTEGER myThid |
68 |
|
|
69 |
|
#ifdef ALLOW_NONHYDROSTATIC |
70 |
|
|
71 |
|
C !LOCAL VARIABLES: |
72 |
C === Local variables ==== |
C === Local variables ==== |
73 |
C actualIts - Number of iterations taken |
C actualIts :: Number of iterations taken |
74 |
C actualResidual - residual |
C actualResidual :: residual |
75 |
C bi - Block index in X and Y. |
C bi,bj :: tile indices |
76 |
C bj |
C eta_qrN :: Used in computing search directions |
77 |
C etaN - Used in computing search directions |
C eta_qrNM1 suffix N and NM1 denote current and |
|
C etaNM1 suffix N and NM1 denote current and |
|
78 |
C cgBeta previous iterations respectively. |
C cgBeta previous iterations respectively. |
79 |
C alpha |
C alpha |
80 |
C sumRHS - Sum of right-hand-side. Sometimes this is a |
C sumRHS :: Sum of right-hand-side. Sometimes this is a |
81 |
C useful debuggin/trouble shooting diagnostic. |
C useful debuggin/trouble shooting diagnostic. |
82 |
C For neumann problems sumRHS needs to be ~0. |
C For neumann problems sumRHS needs to be ~0. |
83 |
C or they converge at a non-zero residual. |
C or they converge at a non-zero residual. |
84 |
C err - Measure of residual of Ax - b, usually the norm. |
C err :: Measure of residual of Ax - b, usually the norm. |
85 |
C I, J, N - Loop counters ( N counts CG iterations ) |
C i, j, k :: Loop counters |
86 |
|
C it3d :: Loop counter for CG iterations |
87 |
|
C msgBuf :: Informational/error message buffer |
88 |
INTEGER actualIts |
INTEGER actualIts |
89 |
_RL actualResidual |
_RL actualResidual |
90 |
INTEGER bi, bj |
INTEGER bi, bj |
91 |
INTEGER I, J, K, it3d |
INTEGER i, j, k, it3d |
92 |
INTEGER KM1, KP1 |
INTEGER km1, kp1 |
93 |
_RL err |
_RL maskM1, maskP1 |
94 |
_RL etaN |
_RL err, errTile(nSx,nSy) |
95 |
_RL etaNM1 |
_RL eta_qrN,eta_qrNtile(nSx,nSy) |
96 |
_RL cgBeta |
_RL eta_qrNM1 |
97 |
_RL alpha |
_RL cgBeta |
98 |
_RL sumRHS |
_RL alpha , alphaTile(nSx,nSy) |
99 |
_RL rhsMax |
_RL sumRHS, sumRHStile(nSx,nSy) |
100 |
_RL rhsNorm |
_RL rhsMax |
101 |
|
_RL rhsNorm |
102 |
INTEGER OLw |
CHARACTER*(MAX_LEN_MBUF) msgBuf |
103 |
INTEGER OLe |
_RL surfFac |
104 |
INTEGER OLn |
#ifdef NONLIN_FRSURF |
105 |
INTEGER OLs |
INTEGER ks |
106 |
INTEGER exchWidthX |
_RL surfTerm(sNx,sNy) |
107 |
INTEGER exchWidthY |
#endif /* NONLIN_FRSURF */ |
108 |
INTEGER myNz |
CEOP |
109 |
_RL topLevFac |
|
110 |
|
IF ( select_rStar .NE. 0 ) THEN |
111 |
|
surfFac = freeSurfFac |
112 |
CcnhDebugStarts |
ELSE |
113 |
CHARACTER*(MAX_LEN_FNAM) suff |
surfFac = 0. |
114 |
CcnhDebugEnds |
ENDIF |
115 |
|
#ifdef NONLIN_FRSURF |
116 |
#ifdef ALLOW_NONHYDROSTATIC |
DO j=1,sNy |
117 |
|
DO i=1,sNx |
118 |
|
surfTerm(i,j) = 0. |
119 |
|
ENDDO |
120 |
|
ENDDO |
121 |
|
#endif /* NONLIN_FRSURF */ |
122 |
|
|
123 |
C-- Initialise inverter |
C-- Initialise inverter |
124 |
etaNM1 = 1. D0 |
eta_qrNM1 = 1. _d 0 |
125 |
|
|
126 |
C-- Normalise RHS |
C-- Normalise RHS |
127 |
rhsMax = 0. _d 0 |
rhsMax = 0. _d 0 |
128 |
DO bj=myByLo(myThid),myByHi(myThid) |
DO bj=myByLo(myThid),myByHi(myThid) |
129 |
DO bi=myBxLo(myThid),myBxHi(myThid) |
DO bi=myBxLo(myThid),myBxHi(myThid) |
130 |
DO K=1,Nr |
DO k=1,Nr |
131 |
DO J=1,sNy |
DO j=1,sNy |
132 |
DO I=1,sNx |
DO i=1,sNx |
133 |
cg3d_b(I,J,K,bi,bj) = cg3d_b(I,J,K,bi,bj)*cg3dNorm |
cg3d_b(i,j,k,bi,bj) = cg3d_b(i,j,k,bi,bj)*cg3dNorm |
134 |
rhsMax = MAX(ABS(cg3d_b(I,J,K,bi,bj)),rhsMax) |
& * maskC(i,j,k,bi,bj) |
135 |
|
rhsMax = MAX(ABS(cg3d_b(i,j,k,bi,bj)),rhsMax) |
136 |
ENDDO |
ENDDO |
137 |
ENDDO |
ENDDO |
138 |
ENDDO |
ENDDO |
139 |
ENDDO |
ENDDO |
140 |
ENDDO |
ENDDO |
141 |
_GLOBAL_MAX_R8( rhsMax, myThid ) |
_GLOBAL_MAX_RL( rhsMax, myThid ) |
142 |
rhsNorm = 1. _d 0 |
rhsNorm = 1. _d 0 |
143 |
IF ( rhsMax .NE. 0. ) rhsNorm = 1. _d 0 / rhsMax |
IF ( rhsMax .NE. 0. ) rhsNorm = 1. _d 0 / rhsMax |
144 |
DO bj=myByLo(myThid),myByHi(myThid) |
DO bj=myByLo(myThid),myByHi(myThid) |
145 |
DO bi=myBxLo(myThid),myBxHi(myThid) |
DO bi=myBxLo(myThid),myBxHi(myThid) |
146 |
DO K=1,Nr |
DO k=1,Nr |
147 |
DO J=1,sNy |
DO j=1,sNy |
148 |
DO I=1,sNx |
DO i=1,sNx |
149 |
cg3d_b(I,J,K,bi,bj) = cg3d_b(I,J,K,bi,bj)*rhsNorm |
cg3d_b(i,j,k,bi,bj) = cg3d_b(i,j,k,bi,bj)*rhsNorm |
150 |
cg3d_x(I,J,K,bi,bj) = cg3d_x(I,J,K,bi,bj)*rhsNorm |
cg3d_x(i,j,k,bi,bj) = cg3d_x(i,j,k,bi,bj)*rhsNorm |
151 |
ENDDO |
ENDDO |
152 |
ENDDO |
ENDDO |
153 |
ENDDO |
ENDDO |
155 |
ENDDO |
ENDDO |
156 |
|
|
157 |
C-- Update overlaps |
C-- Update overlaps |
158 |
_EXCH_XYZ_R8( cg3d_b, myThid ) |
c _EXCH_XYZ_RL( cg3d_b, myThid ) |
159 |
_EXCH_XYZ_R8( cg3d_x, myThid ) |
_EXCH_XYZ_RL( cg3d_x, myThid ) |
160 |
|
|
161 |
#ifdef NONO |
C-- Initial residual calculation (with free-Surface term) |
|
CcnhDebugStarts |
|
|
C-- Initial residual calculation |
|
162 |
err = 0. _d 0 |
err = 0. _d 0 |
163 |
sumRHS = 0. _d 0 |
sumRHS = 0. _d 0 |
164 |
DO bj=myByLo(myThid),myByHi(myThid) |
DO bj=myByLo(myThid),myByHi(myThid) |
165 |
DO bi=myBxLo(myThid),myBxHi(myThid) |
DO bi=myBxLo(myThid),myBxHi(myThid) |
166 |
DO J=1,sNy |
errTile(bi,bj) = 0. _d 0 |
167 |
DO I=1,sNx |
sumRHStile(bi,bj) = 0. _d 0 |
168 |
alpha = 0. _d 0 |
#ifdef NONLIN_FRSURF |
169 |
DO K=1,Nr |
IF ( select_rStar .NE. 0 ) THEN |
170 |
KM1 = K-1 |
DO j=1,sNy |
171 |
IF ( KM1 .EQ. 0 ) KM1 = 1 |
DO i=1,sNx |
172 |
KP1 = K+1 |
surfTerm(i,j) = 0. |
173 |
IF ( KP1 .EQ. Nr+1 ) KP1 = 1 |
ENDDO |
|
cg3d_r(I,J,K,bi,bj) = cg3d_b(I,J,K,bi,bj) -( 0. |
|
|
& +aW3d(I ,J ,K ,bi,bj)*cg3d_x(I-1,J ,K ,bi,bj) |
|
|
& +aW3d(I+1,J ,K ,bi,bj)*cg3d_x(I+1,J ,K ,bi,bj) |
|
|
& +aS3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J-1,K ,bi,bj) |
|
|
& +aS3d(I ,J+1,K ,bi,bj)*cg3d_x(I ,J+1,K ,bi,bj) |
|
|
& +aV3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J ,KM1,bi,bj) |
|
|
& +aV3d(I ,J ,KP1,bi,bj)*cg3d_x(I ,J ,KP1,bi,bj) |
|
|
& -aW3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
|
|
& -aW3d(I+1,J ,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
|
|
& -aS3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
|
|
& -aS3d(I ,J+1,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
|
|
& -aV3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
|
|
& -aV3d(I ,J ,KP1,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
|
|
& ) |
|
|
alpha = alpha |
|
|
& +cg3d_r(I,J,K,bi,bj) |
|
|
sumRHS = sumRHS |
|
|
& +cg3d_b(I,J,K,bi,bj) |
|
174 |
ENDDO |
ENDDO |
175 |
err = err + alpha*alpha |
DO k=1,Nr |
176 |
ENDDO |
DO j=1,sNy |
177 |
ENDDO |
DO i=1,sNx |
178 |
ENDDO |
surfTerm(i,j) = surfTerm(i,j) |
179 |
ENDDO |
& +cg3d_x(i,j,k,bi,bj)*drF(k)*h0FacC(i,j,k,bi,bj) |
180 |
WRITE(6,*) 'DEBUG mythid, err = ', mythid, SQRT(err) |
ENDDO |
181 |
_GLOBAL_SUM_R8( err , myThid ) |
ENDDO |
182 |
_GLOBAL_SUM_R8( sumRHS , myThid ) |
ENDDO |
183 |
_BEGIN_MASTER( myThid ) |
DO j=1,sNy |
184 |
write(0,*) 'DEBUG cg3d: Sum(rhs) = ',sumRHS |
DO i=1,sNx |
185 |
_END_MASTER( ) |
ks = ksurfC(i,j,bi,bj) |
186 |
actualIts = 0 |
surfTerm(i,j) = surfTerm(i,j)*cg3dNorm |
187 |
actualResidual = SQRT(err) |
& *recip_Rcol(i,j,bi,bj)*recip_Rcol(i,j,bi,bj) |
188 |
C _BARRIER |
& *rA(i,j,bi,bj)*deepFac2F(ks) |
189 |
_BEGIN_MASTER( myThid ) |
& *recip_Bo(i,j,bi,bj)/deltaTMom/deltaTfreesurf |
190 |
WRITE(0,'(A,I6,1PE30.14)') 'DEBUG CG3D iters, err = ', |
ENDDO |
191 |
& actualIts, actualResidual |
ENDDO |
192 |
_END_MASTER( ) |
ENDIF |
193 |
CcnhDebugEnds |
#endif /* NONLIN_FRSURF */ |
194 |
#endif |
DO k=1,Nr |
195 |
|
km1 = MAX(k-1, 1 ) |
196 |
C-- Initial residual calculation |
kp1 = MIN(k+1, Nr) |
197 |
err = 0. _d 0 |
maskM1 = 1. _d 0 |
198 |
sumRHS = 0. _d 0 |
maskP1 = 1. _d 0 |
199 |
DO bj=myByLo(myThid),myByHi(myThid) |
IF ( k .EQ. 1 ) maskM1 = 0. _d 0 |
200 |
DO bi=myBxLo(myThid),myBxHi(myThid) |
IF ( k .EQ. Nr) maskP1 = 0. _d 0 |
201 |
DO K=1,Nr |
|
202 |
KM1 = K-1 |
DO j=1,sNy |
203 |
IF ( K .EQ. 1 ) KM1 = 1 |
DO i=1,sNx |
204 |
KP1 = K+1 |
cg3d_r(i,j,k,bi,bj) = cg3d_b(i,j,k,bi,bj) |
205 |
IF ( K .EQ. Nr ) KP1 = 1 |
& -( 0. |
206 |
topLevFac = 0. |
& +aW3d( i, j, k, bi,bj)*cg3d_x(i-1,j, k, bi,bj) |
207 |
IF ( K .EQ. 1) topLevFac = 1. |
& +aW3d(i+1,j, k, bi,bj)*cg3d_x(i+1,j, k, bi,bj) |
208 |
DO J=1,sNy |
& +aS3d( i, j, k, bi,bj)*cg3d_x( i,j-1,k, bi,bj) |
209 |
DO I=1,sNx |
& +aS3d( i,j+1,k, bi,bj)*cg3d_x( i,j+1,k, bi,bj) |
210 |
cg3d_s(I,J,K,bi,bj) = 0. |
& +aV3d( i, j, k, bi,bj)*cg3d_x( i, j,km1,bi,bj)*maskM1 |
211 |
cg3d_r(I,J,K,bi,bj) = cg3d_b(I,J,K,bi,bj) -( 0. |
& +aV3d( i, j,kp1,bi,bj)*cg3d_x( i, j,kp1,bi,bj)*maskP1 |
212 |
& +aW3d(I ,J ,K ,bi,bj)*cg3d_x(I-1,J ,K ,bi,bj) |
& +aC3d( i, j, k, bi,bj)*cg3d_x( i, j, k, bi,bj) |
213 |
& +aW3d(I+1,J ,K ,bi,bj)*cg3d_x(I+1,J ,K ,bi,bj) |
#ifdef NONLIN_FRSURF |
214 |
& +aS3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J-1,K ,bi,bj) |
& -surfFac*surfTerm(i,j)*drF(k)*h0FacC(i,j,k,bi,bj) |
215 |
& +aS3d(I ,J+1,K ,bi,bj)*cg3d_x(I ,J+1,K ,bi,bj) |
#endif /* NONLIN_FRSURF */ |
216 |
& +aV3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J ,KM1,bi,bj) |
& ) |
217 |
& +aV3d(I ,J ,KP1,bi,bj)*cg3d_x(I ,J ,KP1,bi,bj) |
errTile(bi,bj) = errTile(bi,bj) |
218 |
& -aW3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
& +cg3d_r(i,j,k,bi,bj)*cg3d_r(i,j,k,bi,bj) |
219 |
& -aW3d(I+1,J ,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
sumRHStile(bi,bj) = sumRHStile(bi,bj)+cg3d_b(i,j,k,bi,bj) |
220 |
& -aS3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
ENDDO |
221 |
& -aS3d(I ,J+1,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
ENDDO |
222 |
& -aV3d(I ,J ,K ,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
DO J=1-1,sNy+1 |
223 |
& -aV3d(I ,J ,KP1,bi,bj)*cg3d_x(I ,J ,K ,bi,bj) |
DO I=1-1,sNx+1 |
224 |
& -freeSurfFac*_rA(i,j,bi,bj)* horiVertRatio* |
cg3d_s(i,j,k,bi,bj) = 0. |
|
& cg3d_x(I ,J ,K,bi,bj)/deltaTMom/deltaTMom*cg3dNorm |
|
|
& *topLevFac |
|
|
& ) |
|
|
err = err |
|
|
& +cg3d_r(I,J,K,bi,bj)*cg3d_r(I,J,K,bi,bj) |
|
|
sumRHS = sumRHS |
|
|
& +cg3d_b(I,J,K,bi,bj) |
|
225 |
ENDDO |
ENDDO |
226 |
ENDDO |
ENDDO |
227 |
ENDDO |
ENDDO |
228 |
ENDDO |
ENDDO |
229 |
ENDDO |
ENDDO |
230 |
C _EXCH_XYZ_R8( cg3d_r, myThid ) |
CALL EXCH_S3D_RL( cg3d_r, Nr, myThid ) |
231 |
OLw = 1 |
c CALL EXCH_S3D_RL( cg3d_s, Nr, myThid ) |
232 |
OLe = 1 |
CALL GLOBAL_SUM_TILE_RL( sumRHStile, sumRHS, myThid ) |
233 |
OLn = 1 |
CALL GLOBAL_SUM_TILE_RL( errTile, err, myThid ) |
234 |
OLs = 1 |
IF ( debugLevel.GT.debLevB .AND. diagFreq.GT.0. ) THEN |
235 |
exchWidthX = 1 |
CALL WRITE_FLD_S3D_RS( |
236 |
exchWidthY = 1 |
I 'cg3d_r_I', 'I10', 1, Nr, cg3d_r, myIter, myThid ) |
237 |
myNz = Nr |
ENDIF |
238 |
CALL EXCH_RL( cg3d_r, |
|
239 |
I OLw, OLe, OLs, OLn, myNz, |
IF ( debugLevel .GE. debLevZero ) THEN |
240 |
I exchWidthX, exchWidthY, |
_BEGIN_MASTER( myThid ) |
241 |
I FORWARD_SIMULATION, EXCH_IGNORE_CORNERS, myThid ) |
WRITE(standardmessageunit,'(A,1P2E22.14)') |
242 |
C _EXCH_XYZ_R8( cg3d_s, myThid ) |
& ' cg3d: Sum(rhs),rhsMax = ',sumRHS,rhsMax |
243 |
OLw = 1 |
_END_MASTER( myThid ) |
244 |
OLe = 1 |
ENDIF |
|
OLn = 1 |
|
|
OLs = 1 |
|
|
exchWidthX = 1 |
|
|
exchWidthY = 1 |
|
|
myNz = Nr |
|
|
CALL EXCH_RL( cg3d_s, |
|
|
I OLw, OLe, OLs, OLn, myNz, |
|
|
I exchWidthX, exchWidthY, |
|
|
I FORWARD_SIMULATION, EXCH_IGNORE_CORNERS, myThid ) |
|
|
_GLOBAL_SUM_R8( sumRHS, myThid ) |
|
|
_GLOBAL_SUM_R8( err , myThid ) |
|
|
|
|
|
_BEGIN_MASTER( myThid ) |
|
|
write(0,'(A,1PE30.14)') ' cg3d: Sum(rhs) = ',sumRHS |
|
|
_END_MASTER( ) |
|
245 |
|
|
246 |
actualIts = 0 |
actualIts = 0 |
247 |
actualResidual = SQRT(err) |
actualResidual = SQRT(err) |
248 |
C _BARRIER |
firstResidual=actualResidual |
|
_BEGIN_MASTER( myThid ) |
|
|
WRITE(0,'(A,I6,1PE30.14)') ' CG3D iters, err = ', |
|
|
& actualIts, actualResidual |
|
|
_END_MASTER( ) |
|
249 |
|
|
250 |
C >>>>>>>>>>>>>>> BEGIN SOLVER <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< |
C >>>>>>>>>>>>>>> BEGIN SOLVER <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< |
251 |
DO 10 it3d=1, cg3dMaxIters |
DO 10 it3d=1, numIters |
252 |
|
|
253 |
CcnhDebugStarts |
CcnhDebugStarts |
254 |
#ifdef VERBOSE |
c IF ( mod(it3d-1,10).EQ.0) |
255 |
IF ( mod(it3d-1,10).EQ.0) |
c & WRITE(*,*) ' CG3D: Iteration ',it3d-1, |
256 |
& WRITE(0,*) ' CG3D: Iteration ',it3d-1, |
c & ' residual = ',actualResidual |
|
& ' residual = ',actualResidual |
|
|
#endif |
|
257 |
CcnhDebugEnds |
CcnhDebugEnds |
258 |
IF ( actualResidual .LT. cg3dTargetResidual ) GOTO 11 |
IF ( actualResidual .LT. cg3dTargetResidual ) GOTO 11 |
259 |
C-- Solve preconditioning equation and update |
C-- Solve preconditioning equation and update |
260 |
C-- conjugate direction vector "s". |
C-- conjugate direction vector "s". |
261 |
C Note. On the next to loops over all tiles the inner loop ranges |
C Note. On the next to loops over all tiles the inner loop ranges |
262 |
C in sNx and sNy are expanded by 1 to avoid a communication |
C in sNx and sNy are expanded by 1 to avoid a communication |
263 |
C step. However this entails a bit of gynamastics because we only |
C step. However this entails a bit of gynamastics because we only |
264 |
C want etaN for the interior points. |
C want eta_qrN for the interior points. |
265 |
etaN = 0. _d 0 |
eta_qrN = 0. _d 0 |
266 |
DO bj=myByLo(myThid),myByHi(myThid) |
DO bj=myByLo(myThid),myByHi(myThid) |
267 |
DO bi=myBxLo(myThid),myBxHi(myThid) |
DO bi=myBxLo(myThid),myBxHi(myThid) |
268 |
DO K=1,1 |
eta_qrNtile(bi,bj) = 0. _d 0 |
269 |
DO J=1-1,sNy+1 |
DO k=1,1 |
270 |
DO I=1-1,sNx+1 |
DO j=1-1,sNy+1 |
271 |
cg3d_q(I,J,K,bi,bj) = |
DO i=1-1,sNx+1 |
272 |
& zMC(I ,J ,K,bi,bj)*cg3d_r(I ,J ,K,bi,bj) |
cg3d_q(i,j,k,bi,bj) = zMC(i,j,k,bi,bj) |
273 |
|
& *cg3d_r(i,j,k,bi,bj) |
274 |
ENDDO |
ENDDO |
275 |
ENDDO |
ENDDO |
276 |
ENDDO |
ENDDO |
277 |
DO K=2,Nr |
DO k=2,Nr |
278 |
DO J=1-1,sNy+1 |
DO j=1-1,sNy+1 |
279 |
DO I=1-1,sNx+1 |
DO i=1-1,sNx+1 |
280 |
cg3d_q(I,J,K,bi,bj) = |
cg3d_q(i,j,k,bi,bj) = zMC(i,j,k,bi,bj) |
281 |
& zMC(I,J,K,bi,bj)*(cg3d_r(I,J,K ,bi,bj) |
& *( cg3d_r(i,j,k,bi,bj) |
282 |
& -zML(I,J,K,bi,bj)*cg3d_q(I,J,K-1,bi,bj)) |
& -zML(i,j,k,bi,bj)*cg3d_q(i,j,k-1,bi,bj) |
283 |
|
& ) |
284 |
ENDDO |
ENDDO |
285 |
ENDDO |
ENDDO |
286 |
ENDDO |
ENDDO |
287 |
DO K=Nr,Nr |
DO k=Nr,Nr |
288 |
caja IF (Nr .GT. 1) THEN |
DO j=1,sNy |
289 |
caja DO J=1-1,sNy+1 |
DO i=1,sNx |
290 |
caja DO I=1-1,sNx+1 |
eta_qrNtile(bi,bj) = eta_qrNtile(bi,bj) |
291 |
caja cg3d_q(I,J,K,bi,bj) = |
& +cg3d_q(i,j,k,bi,bj)*cg3d_r(i,j,k,bi,bj) |
|
caja & zMC(i,j,k,bi,bj)*(cg3d_r(i,j,k ,bi,bj) |
|
|
caja & -zML(i,j,k,bi,bj)*cg3d_q(i,j,k-1,bi,bj)) |
|
|
caja ENDDO |
|
|
caja ENDDO |
|
|
caja ENDIF |
|
|
DO J=1,sNy |
|
|
DO I=1,sNx |
|
|
etaN = etaN |
|
|
& +cg3d_q(I,J,K,bi,bj)*cg3d_r(I,J,K,bi,bj) |
|
292 |
ENDDO |
ENDDO |
293 |
ENDDO |
ENDDO |
294 |
ENDDO |
ENDDO |
295 |
DO K=Nr-1,1,-1 |
DO k=Nr-1,1,-1 |
296 |
DO J=1-1,sNy+1 |
DO j=1-1,sNy+1 |
297 |
DO I=1-1,sNx+1 |
DO i=1-1,sNx+1 |
298 |
cg3d_q(I,J,K,bi,bj) = |
cg3d_q(i,j,k,bi,bj) = cg3d_q(i,j,k,bi,bj) |
299 |
& cg3d_q(I,J,K,bi,bj) |
& -zMU(i,j,k,bi,bj)*cg3d_q(i,j,k+1,bi,bj) |
300 |
& -zMU(I,J,K,bi,bj)*cg3d_q(I,J,K+1,bi,bj) |
ENDDO |
301 |
ENDDO |
ENDDO |
302 |
ENDDO |
DO j=1,sNy |
303 |
DO J=1,sNy |
DO i=1,sNx |
304 |
DO I=1,sNx |
eta_qrNtile(bi,bj) = eta_qrNtile(bi,bj) |
305 |
etaN = etaN |
& +cg3d_q(i,j,k,bi,bj)*cg3d_r(i,j,k,bi,bj) |
|
& +cg3d_q(I,J,K,bi,bj)*cg3d_r(I,J,K,bi,bj) |
|
306 |
ENDDO |
ENDDO |
307 |
ENDDO |
ENDDO |
308 |
ENDDO |
ENDDO |
309 |
ENDDO |
ENDDO |
310 |
ENDDO |
ENDDO |
|
caja |
|
|
caja etaN=0. |
|
|
caja DO bj=myByLo(myThid),myByHi(myThid) |
|
|
caja DO bi=myBxLo(myThid),myBxHi(myThid) |
|
|
caja DO K=1,Nr |
|
|
caja DO J=1,sNy |
|
|
caja DO I=1,sNx |
|
|
caja etaN = etaN |
|
|
caja & +cg3d_q(I,J,K,bi,bj)*cg3d_r(I,J,K,bi,bj) |
|
|
caja ENDDO |
|
|
caja ENDDO |
|
|
caja ENDDO |
|
|
caja ENDDO |
|
|
caja ENDDO |
|
|
caja |
|
311 |
|
|
312 |
_GLOBAL_SUM_R8(etaN, myThid) |
CALL GLOBAL_SUM_TILE_RL( eta_qrNtile,eta_qrN,myThid ) |
313 |
CcnhDebugStarts |
CcnhDebugStarts |
314 |
C WRITE(0,*) ' CG3D: Iteration ',it3d-1,' etaN = ',etaN |
C WRITE(*,*) ' CG3D: Iteration ',it3d-1,' eta_qrN = ',eta_qrN |
315 |
CcnhDebugEnds |
CcnhDebugEnds |
316 |
cgBeta = etaN/etaNM1 |
cgBeta = eta_qrN/eta_qrNM1 |
317 |
CcnhDebugStarts |
CcnhDebugStarts |
318 |
C WRITE(0,*) ' CG3D: Iteration ',it3d-1,' beta = ',cgBeta |
C WRITE(*,*) ' CG3D: Iteration ',it3d-1,' beta = ',cgBeta |
319 |
CcnhDebugEnds |
CcnhDebugEnds |
320 |
etaNM1 = etaN |
eta_qrNM1 = eta_qrN |
321 |
|
|
322 |
DO bj=myByLo(myThid),myByHi(myThid) |
DO bj=myByLo(myThid),myByHi(myThid) |
323 |
DO bi=myBxLo(myThid),myBxHi(myThid) |
DO bi=myBxLo(myThid),myBxHi(myThid) |
324 |
DO K=1,Nr |
DO k=1,Nr |
325 |
DO J=1-1,sNy+1 |
DO j=1-1,sNy+1 |
326 |
DO I=1-1,sNx+1 |
DO i=1-1,sNx+1 |
327 |
cg3d_s(I,J,K,bi,bj) = cg3d_q(I,J,K,bi,bj) |
cg3d_s(i,j,k,bi,bj) = cg3d_q(i,j,k,bi,bj) |
328 |
& + cgBeta*cg3d_s(I,J,K,bi,bj) |
& + cgBeta*cg3d_s(i,j,k,bi,bj) |
329 |
ENDDO |
ENDDO |
330 |
ENDDO |
ENDDO |
331 |
ENDDO |
ENDDO |
337 |
alpha = 0. _d 0 |
alpha = 0. _d 0 |
338 |
DO bj=myByLo(myThid),myByHi(myThid) |
DO bj=myByLo(myThid),myByHi(myThid) |
339 |
DO bi=myBxLo(myThid),myBxHi(myThid) |
DO bi=myBxLo(myThid),myBxHi(myThid) |
340 |
IF ( Nr .GT. 1 ) THEN |
alphaTile(bi,bj) = 0. _d 0 |
341 |
DO K=1,1 |
#ifdef NONLIN_FRSURF |
342 |
DO J=1,sNy |
IF ( select_rStar .NE. 0 ) THEN |
343 |
DO I=1,sNx |
DO j=1,sNy |
344 |
cg3d_q(I,J,K,bi,bj) = |
DO i=1,sNx |
345 |
& aW3d(I ,J ,K ,bi,bj)*cg3d_s(I-1,J ,K ,bi,bj) |
surfTerm(i,j) = 0. |
346 |
& +aW3d(I+1,J ,K ,bi,bj)*cg3d_s(I+1,J ,K ,bi,bj) |
ENDDO |
347 |
& +aS3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J-1,K ,bi,bj) |
ENDDO |
348 |
& +aS3d(I ,J+1,K ,bi,bj)*cg3d_s(I ,J+1,K ,bi,bj) |
DO k=1,Nr |
349 |
& +aV3d(I ,J ,K+1,bi,bj)*cg3d_s(I ,J ,K+1,bi,bj) |
DO j=1,sNy |
350 |
& -aW3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
DO i=1,sNx |
351 |
& -aW3d(I+1,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
surfTerm(i,j) = surfTerm(i,j) |
352 |
& -aS3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
& +cg3d_s(i,j,k,bi,bj)*drF(k)*h0FacC(i,j,k,bi,bj) |
|
& -aS3d(I ,J+1,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
|
|
& -aV3d(I ,J ,K+1,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
|
|
& -freeSurfFac*_rA(i,j,bi,bj)* horiVertRatio* |
|
|
& cg3d_s(I ,J ,K,bi,bj)/deltaTMom/deltaTMom*cg3dNorm |
|
|
alpha = alpha+cg3d_s(I,J,K,bi,bj)*cg3d_q(I,J,K,bi,bj) |
|
353 |
ENDDO |
ENDDO |
354 |
ENDDO |
ENDDO |
355 |
ENDDO |
ENDDO |
356 |
|
DO j=1,sNy |
357 |
|
DO i=1,sNx |
358 |
|
ks = ksurfC(i,j,bi,bj) |
359 |
|
surfTerm(i,j) = surfTerm(i,j)*cg3dNorm |
360 |
|
& *recip_Rcol(i,j,bi,bj)*recip_Rcol(i,j,bi,bj) |
361 |
|
& *rA(i,j,bi,bj)*deepFac2F(ks) |
362 |
|
& *recip_Bo(i,j,bi,bj)/deltaTMom/deltaTfreesurf |
363 |
|
ENDDO |
364 |
|
ENDDO |
365 |
|
ENDIF |
366 |
|
#endif /* NONLIN_FRSURF */ |
367 |
|
IF ( Nr .GT. 1 ) THEN |
368 |
|
k=1 |
369 |
|
DO j=1,sNy |
370 |
|
DO i=1,sNx |
371 |
|
cg3d_q(i,j,k,bi,bj) = |
372 |
|
& aW3d( i, j, k, bi,bj)*cg3d_s(i-1,j, k, bi,bj) |
373 |
|
& +aW3d(i+1,j, k, bi,bj)*cg3d_s(i+1,j, k, bi,bj) |
374 |
|
& +aS3d( i, j, k, bi,bj)*cg3d_s( i,j-1,k, bi,bj) |
375 |
|
& +aS3d( i,j+1,k, bi,bj)*cg3d_s( i,j+1,k, bi,bj) |
376 |
|
& +aV3d( i, j,k+1,bi,bj)*cg3d_s( i, j,k+1,bi,bj) |
377 |
|
& +aC3d( i, j, k, bi,bj)*cg3d_s( i, j, k, bi,bj) |
378 |
|
#ifdef NONLIN_FRSURF |
379 |
|
& -surfFac*surfTerm(i,j)*drF(k)*h0FacC(i,j,k,bi,bj) |
380 |
|
#endif /* NONLIN_FRSURF */ |
381 |
|
alphaTile(bi,bj) = alphaTile(bi,bj) |
382 |
|
& +cg3d_s(i,j,k,bi,bj)*cg3d_q(i,j,k,bi,bj) |
383 |
|
ENDDO |
384 |
|
ENDDO |
385 |
ELSE |
ELSE |
386 |
DO K=1,1 |
k=1 |
387 |
DO J=1,sNy |
DO j=1,sNy |
388 |
DO I=1,sNx |
DO i=1,sNx |
389 |
cg3d_q(I,J,K,bi,bj) = |
cg3d_q(i,j,k,bi,bj) = |
390 |
& aW3d(I ,J ,K ,bi,bj)*cg3d_s(I-1,J ,K ,bi,bj) |
& aW3d( i, j, k, bi,bj)*cg3d_s(i-1,j, k, bi,bj) |
391 |
& +aW3d(I+1,J ,K ,bi,bj)*cg3d_s(I+1,J ,K ,bi,bj) |
& +aW3d(i+1,j, k, bi,bj)*cg3d_s(i+1,j, k, bi,bj) |
392 |
& +aS3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J-1,K ,bi,bj) |
& +aS3d( i, j, k, bi,bj)*cg3d_s( i,j-1,k, bi,bj) |
393 |
& +aS3d(I ,J+1,K ,bi,bj)*cg3d_s(I ,J+1,K ,bi,bj) |
& +aS3d( i,j+1,k, bi,bj)*cg3d_s( i,j+1,k, bi,bj) |
394 |
& -aW3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
& +aC3d( i, j, k, bi,bj)*cg3d_s( i, j, k, bi,bj) |
395 |
& -aW3d(I+1,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
#ifdef NONLIN_FRSURF |
396 |
& -aS3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
& -surfFac*surfTerm(i,j)*drF(k)*h0FacC(i,j,k,bi,bj) |
397 |
& -aS3d(I ,J+1,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
#endif /* NONLIN_FRSURF */ |
398 |
& -freeSurfFac*_rA(i,j,bi,bj)* horiVertRatio* |
alphaTile(bi,bj) = alphaTile(bi,bj) |
399 |
& cg3d_s(I ,J ,K,bi,bj)/deltaTMom/deltaTMom*cg3dNorm |
& +cg3d_s(i,j,k,bi,bj)*cg3d_q(i,j,k,bi,bj) |
|
alpha = alpha+cg3d_s(I,J,K,bi,bj)*cg3d_q(I,J,K,bi,bj) |
|
|
ENDDO |
|
400 |
ENDDO |
ENDDO |
401 |
ENDDO |
ENDDO |
402 |
ENDIF |
ENDIF |
403 |
DO K=2,Nr-1 |
DO k=2,Nr-1 |
404 |
DO J=1,sNy |
DO j=1,sNy |
405 |
DO I=1,sNx |
DO i=1,sNx |
406 |
cg3d_q(I,J,K,bi,bj) = |
cg3d_q(i,j,k,bi,bj) = |
407 |
& aW3d(I ,J ,K ,bi,bj)*cg3d_s(I-1,J ,K ,bi,bj) |
& aW3d( i, j, k, bi,bj)*cg3d_s(i-1,j, k, bi,bj) |
408 |
& +aW3d(I+1,J ,K ,bi,bj)*cg3d_s(I+1,J ,K ,bi,bj) |
& +aW3d(i+1,j, k, bi,bj)*cg3d_s(i+1,j, k, bi,bj) |
409 |
& +aS3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J-1,K ,bi,bj) |
& +aS3d( i, j, k, bi,bj)*cg3d_s( i,j-1,k, bi,bj) |
410 |
& +aS3d(I ,J+1,K ,bi,bj)*cg3d_s(I ,J+1,K ,bi,bj) |
& +aS3d( i,j+1,k, bi,bj)*cg3d_s( i,j+1,k, bi,bj) |
411 |
& +aV3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K-1,bi,bj) |
& +aV3d( i, j, k, bi,bj)*cg3d_s( i, j,k-1,bi,bj) |
412 |
& +aV3d(I ,J ,K+1,bi,bj)*cg3d_s(I ,J ,K+1,bi,bj) |
& +aV3d( i, j,k+1,bi,bj)*cg3d_s( i, j,k+1,bi,bj) |
413 |
& -aW3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
& +aC3d( i, j, k, bi,bj)*cg3d_s( i, j, k, bi,bj) |
414 |
& -aW3d(I+1,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
#ifdef NONLIN_FRSURF |
415 |
& -aS3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
& -surfFac*surfTerm(i,j)*drF(k)*h0FacC(i,j,k,bi,bj) |
416 |
& -aS3d(I ,J+1,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
#endif /* NONLIN_FRSURF */ |
417 |
& -aV3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
alphaTile(bi,bj) = alphaTile(bi,bj) |
418 |
& -aV3d(I ,J ,K+1,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
& +cg3d_s(i,j,k,bi,bj)*cg3d_q(i,j,k,bi,bj) |
|
alpha = alpha+cg3d_s(I,J,K,bi,bj)*cg3d_q(I,J,K,bi,bj) |
|
419 |
ENDDO |
ENDDO |
420 |
ENDDO |
ENDDO |
421 |
ENDDO |
ENDDO |
422 |
IF ( Nr .GT. 1 ) THEN |
IF ( Nr .GT. 1 ) THEN |
423 |
DO K=Nr,Nr |
k=Nr |
424 |
DO J=1,sNy |
DO j=1,sNy |
425 |
DO I=1,sNx |
DO i=1,sNx |
426 |
cg3d_q(I,J,K,bi,bj) = |
cg3d_q(i,j,k,bi,bj) = |
427 |
& aW3d(I ,J ,K ,bi,bj)*cg3d_s(I-1,J ,K ,bi,bj) |
& aW3d( i, j, k, bi,bj)*cg3d_s(i-1,j, k, bi,bj) |
428 |
& +aW3d(I+1,J ,K ,bi,bj)*cg3d_s(I+1,J ,K ,bi,bj) |
& +aW3d(i+1,j, k, bi,bj)*cg3d_s(i+1,j, k, bi,bj) |
429 |
& +aS3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J-1,K ,bi,bj) |
& +aS3d( i, j, k, bi,bj)*cg3d_s( i,j-1,k, bi,bj) |
430 |
& +aS3d(I ,J+1,K ,bi,bj)*cg3d_s(I ,J+1,K ,bi,bj) |
& +aS3d( i,j+1,k, bi,bj)*cg3d_s( i,j+1,k, bi,bj) |
431 |
& +aV3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K-1,bi,bj) |
& +aV3d( i, j, k, bi,bj)*cg3d_s( i, j,k-1,bi,bj) |
432 |
& -aW3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
& +aC3d( i, j, k, bi,bj)*cg3d_s( i, j, k, bi,bj) |
433 |
& -aW3d(I+1,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
#ifdef NONLIN_FRSURF |
434 |
& -aS3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
& -surfFac*surfTerm(i,j)*drF(k)*h0FacC(i,j,k,bi,bj) |
435 |
& -aS3d(I ,J+1,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
#endif /* NONLIN_FRSURF */ |
436 |
& -aV3d(I ,J ,K ,bi,bj)*cg3d_s(I ,J ,K ,bi,bj) |
alphaTile(bi,bj) = alphaTile(bi,bj) |
437 |
alpha = alpha+cg3d_s(I,J,K,bi,bj)*cg3d_q(I,J,K,bi,bj) |
& +cg3d_s(i,j,k,bi,bj)*cg3d_q(i,j,k,bi,bj) |
|
ENDDO |
|
438 |
ENDDO |
ENDDO |
439 |
ENDDO |
ENDDO |
440 |
ENDIF |
ENDIF |
441 |
ENDDO |
ENDDO |
442 |
ENDDO |
ENDDO |
443 |
_GLOBAL_SUM_R8(alpha,myThid) |
CALL GLOBAL_SUM_TILE_RL( alphaTile, alpha, myThid ) |
444 |
CcnhDebugStarts |
CcnhDebugStarts |
445 |
C WRITE(0,*) ' CG3D: Iteration ',it3d-1,' SUM(s*q)= ',alpha |
C WRITE(*,*) ' CG3D: Iteration ',it3d-1,' SUM(s*q)= ',alpha |
446 |
CcnhDebugEnds |
CcnhDebugEnds |
447 |
alpha = etaN/alpha |
alpha = eta_qrN/alpha |
448 |
CcnhDebugStarts |
CcnhDebugStarts |
449 |
C WRITE(0,*) ' CG3D: Iteration ',it3d-1,' alpha= ',alpha |
C WRITE(*,*) ' CG3D: Iteration ',it3d-1,' alpha= ',alpha |
450 |
CcnhDebugEnds |
CcnhDebugEnds |
451 |
|
|
452 |
C== Update solution and residual vectors |
C== Update solution and residual vectors |
453 |
C Now compute "interior" points. |
C Now compute "interior" points. |
454 |
err = 0. _d 0 |
err = 0. _d 0 |
455 |
DO bj=myByLo(myThid),myByHi(myThid) |
DO bj=myByLo(myThid),myByHi(myThid) |
456 |
DO bi=myBxLo(myThid),myBxHi(myThid) |
DO bi=myBxLo(myThid),myBxHi(myThid) |
457 |
DO K=1,Nr |
errTile(bi,bj) = 0. _d 0 |
458 |
DO J=1,sNy |
DO k=1,Nr |
459 |
DO I=1,sNx |
DO j=1,sNy |
460 |
cg3d_x(I,J,K,bi,bj)=cg3d_x(I,J,K,bi,bj) |
DO i=1,sNx |
461 |
& +alpha*cg3d_s(I,J,K,bi,bj) |
cg3d_x(i,j,k,bi,bj)=cg3d_x(i,j,k,bi,bj) |
462 |
cg3d_r(I,J,K,bi,bj)=cg3d_r(I,J,K,bi,bj) |
& +alpha*cg3d_s(i,j,k,bi,bj) |
463 |
& -alpha*cg3d_q(I,J,K,bi,bj) |
cg3d_r(i,j,k,bi,bj)=cg3d_r(i,j,k,bi,bj) |
464 |
err = err+cg3d_r(I,J,K,bi,bj)*cg3d_r(I,J,K,bi,bj) |
& -alpha*cg3d_q(i,j,k,bi,bj) |
465 |
|
errTile(bi,bj) = errTile(bi,bj) |
466 |
|
& +cg3d_r(i,j,k,bi,bj)*cg3d_r(i,j,k,bi,bj) |
467 |
ENDDO |
ENDDO |
468 |
ENDDO |
ENDDO |
469 |
ENDDO |
ENDDO |
470 |
ENDDO |
ENDDO |
471 |
ENDDO |
ENDDO |
472 |
|
|
473 |
_GLOBAL_SUM_R8( err , myThid ) |
CALL GLOBAL_SUM_TILE_RL( errTile, err, myThid ) |
474 |
err = SQRT(err) |
err = SQRT(err) |
475 |
actualIts = it3d |
actualIts = it3d |
476 |
actualResidual = err |
actualResidual = err |
477 |
|
IF ( debugLevel.GT.debLevB ) THEN |
478 |
|
c IF ( DIFFERENT_MULTIPLE(monitorFreq,myTime,deltaTClock) |
479 |
|
c & ) THEN |
480 |
|
_BEGIN_MASTER( myThid ) |
481 |
|
WRITE(msgBuf,'(A,I6,A,1PE21.14)') |
482 |
|
& ' cg3d: iter=', actualIts, ' ; resid.= ', actualResidual |
483 |
|
CALL PRINT_MESSAGE( msgBuf, standardMessageUnit, |
484 |
|
& SQUEEZE_RIGHT, myThid ) |
485 |
|
_END_MASTER( myThid ) |
486 |
|
c ENDIF |
487 |
|
ENDIF |
488 |
IF ( actualResidual .LT. cg3dTargetResidual ) GOTO 11 |
IF ( actualResidual .LT. cg3dTargetResidual ) GOTO 11 |
489 |
C _EXCH_XYZ_R8(cg3d_r, myThid ) |
CALL EXCH_S3D_RL( cg3d_r, Nr, myThid ) |
|
OLw = 1 |
|
|
OLe = 1 |
|
|
OLn = 1 |
|
|
OLs = 1 |
|
|
exchWidthX = 1 |
|
|
exchWidthY = 1 |
|
|
myNz = Nr |
|
|
CALL EXCH_RL( cg3d_r, |
|
|
I OLw, OLe, OLs, OLn, myNz, |
|
|
I exchWidthX, exchWidthY, |
|
|
I FORWARD_SIMULATION, EXCH_IGNORE_CORNERS, myThid ) |
|
490 |
|
|
491 |
10 CONTINUE |
10 CONTINUE |
492 |
11 CONTINUE |
11 CONTINUE |
493 |
|
|
494 |
|
IF ( debugLevel.GT.debLevB .AND. diagFreq.GT.0. ) THEN |
495 |
|
CALL WRITE_FLD_S3D_RS( |
496 |
|
I 'cg3d_r_F', 'I10', 1, Nr, cg3d_r, myIter, myThid ) |
497 |
|
ENDIF |
498 |
|
|
499 |
C-- Un-normalise the answer |
C-- Un-normalise the answer |
500 |
DO bj=myByLo(myThid),myByHi(myThid) |
DO bj=myByLo(myThid),myByHi(myThid) |
501 |
DO bi=myBxLo(myThid),myBxHi(myThid) |
DO bi=myBxLo(myThid),myBxHi(myThid) |
502 |
DO K=1,Nr |
DO k=1,Nr |
503 |
DO J=1,sNy |
DO j=1,sNy |
504 |
DO I=1,sNx |
DO i=1,sNx |
505 |
cg3d_x(I,J,K,bi,bj) = cg3d_x(I,J,K,bi,bj)/rhsNorm |
cg3d_x(i,j,k,bi,bj) = cg3d_x(i,j,k,bi,bj)/rhsNorm |
506 |
ENDDO |
ENDDO |
507 |
ENDDO |
ENDDO |
508 |
ENDDO |
ENDDO |
509 |
ENDDO |
ENDDO |
510 |
ENDDO |
ENDDO |
511 |
|
|
512 |
_EXCH_XYZ_R8(cg3d_x, myThid ) |
lastResidual = actualResidual |
513 |
_BEGIN_MASTER( myThid ) |
numIters = actualIts |
|
WRITE(0,'(A,I6,1PE30.14)') ' CG3D iters, err = ', |
|
|
& actualIts, actualResidual |
|
|
_END_MASTER( ) |
|
|
|
|
|
CcnhDebugStarts |
|
|
C CALL PLOT_FIELD_XYRL( cg2d_x, 'CALC_MOM_RHS CG2D_X' , 1, myThid ) |
|
|
C err = 0. _d 0 |
|
|
C DO bj=myByLo(myThid),myByHi(myThid) |
|
|
C DO bi=myBxLo(myThid),myBxHi(myThid) |
|
|
C DO J=1,sNy |
|
|
C DO I=1,sNx |
|
|
C cg2d_r(I,J,bi,bj) = cg2d_b(I,J,bi,bj) - |
|
|
C & (aW2d(I ,J ,bi,bj)*cg2d_x(I-1,J ,bi,bj) |
|
|
C & +aW2d(I+1,J ,bi,bj)*cg2d_x(I+1,J ,bi,bj) |
|
|
C & +aS2d(I ,J ,bi,bj)*cg2d_x(I ,J-1,bi,bj) |
|
|
C & +aS2d(I ,J+1,bi,bj)*cg2d_x(I ,J+1,bi,bj) |
|
|
C & -aW2d(I ,J ,bi,bj)*cg2d_x(I ,J ,bi,bj) |
|
|
C & -aW2d(I+1,J ,bi,bj)*cg2d_x(I ,J ,bi,bj) |
|
|
C & -aS2d(I ,J ,bi,bj)*cg2d_x(I ,J ,bi,bj) |
|
|
C & -aS2d(I ,J+1,bi,bj)*cg2d_x(I ,J ,bi,bj)) |
|
|
C err = err + |
|
|
C & cg2d_r(I,J,bi,bj)*cg2d_r(I,J,bi,bj) |
|
|
C ENDDO |
|
|
C ENDDO |
|
|
C ENDDO |
|
|
C ENDDO |
|
|
C _GLOBAL_SUM_R8( err , myThid ) |
|
|
C write(0,*) 'cg2d: Ax - b = ',SQRT(err) |
|
|
CcnhDebugEnds |
|
514 |
|
|
515 |
#endif /* ALLOW_NONHYDROSTATIC */ |
#endif /* ALLOW_NONHYDROSTATIC */ |
516 |
|
|