/[escript]/trunk/finley/src/Assemble_PDE_System2_1D.c
ViewVC logotype

Annotation of /trunk/finley/src/Assemble_PDE_System2_1D.c

Parent Directory Parent Directory | Revision Log Revision Log


Revision 853 - (hide annotations)
Wed Sep 20 05:56:36 2006 UTC (13 years, 4 months ago) by gross
File MIME type: text/plain
File size: 14611 byte(s)
some performance problems wit openmp fixed.
1 gross 798 /*
2     ************************************************************
3     * Copyright 2006 by ACcESS MNRF *
4     * *
5     * http://www.access.edu.au *
6     * Primary Business: Queensland, Australia *
7     * Licensed under the Open Software License version 3.0 *
8     * http://www.opensource.org/licenses/osl-3.0.php *
9     * *
10     ************************************************************
11     */
12    
13     /**************************************************************/
14    
15     /* assembles the system of numEq PDEs into the stiffness matrix S right hand side F */
16     /* the shape functions for test and solution must be identical */
17    
18    
19     /* -(A_{k,i,m,j} u_m,j)_i-(B_{k,i,m} u_m)_i+C_{k,m,j} u_m,j-D_{k,m} u_m and -(X_{k,i})_i + Y_k */
20    
21     /* u has p.numComp components in a 1D domain. The shape functions for test and solution must be identical */
22     /* and row_NS == row_NN */
23    
24     /* Shape of the coefficients: */
25    
26     /* A = p.numEqu x 1 x p.numComp x 1 */
27     /* B = 1 x numEqu x p.numComp */
28     /* C = p.numEqu x 1 x p.numComp */
29     /* D = p.numEqu x p.numComp */
30     /* X = p.numEqu x 1 */
31     /* Y = p.numEqu */
32    
33    
34     /**************************************************************/
35    
36     /* Author: gross@access.edu.au */
37     /* Version: $Id:$ */
38    
39     /**************************************************************/
40    
41    
42     #include "Assemble.h"
43     #include "Util.h"
44 gross 853 #ifdef _OPENMP
45     #include <omp.h>
46     #endif
47 gross 798
48 gross 853
49 gross 798 /**************************************************************/
50    
51     void Finley_Assemble_PDE_System2_1D(Assemble_Parameters p, Finley_ElementFile* elements,
52     Paso_SystemMatrix* Mat, escriptDataC* F,
53     escriptDataC* A, escriptDataC* B, escriptDataC* C, escriptDataC* D, escriptDataC* X, escriptDataC* Y) {
54    
55     #define DIM 1
56     index_t color;
57     dim_t e;
58 gross 853 double *EM_S, *EM_F, *Vol, *DSDX, *A_p, *B_p, *C_p, *D_p, *X_p, *Y_p;
59     index_t *row_index;
60     register dim_t q, s,r,k,m;
61     register double rtmp;
62     bool_t add_EM_F, add_EM_S;
63    
64 gross 798 bool_t extendedA=isExpanded(A);
65     bool_t extendedB=isExpanded(B);
66     bool_t extendedC=isExpanded(C);
67     bool_t extendedD=isExpanded(D);
68     bool_t extendedX=isExpanded(X);
69     bool_t extendedY=isExpanded(Y);
70     double *F_p=getSampleData(F,0);
71     double *S=p.row_jac->ReferenceElement->S;
72     dim_t len_EM_S=p.row_NN*p.col_NN*p.numEqu*p.numComp;
73     dim_t len_EM_F=p.row_NN*p.numEqu;
74    
75    
76 gross 853 #pragma omp parallel private(color, EM_S, EM_F, Vol, DSDX, A_p, B_p, C_p, D_p, X_p, Y_p,row_index, q, s,r,k,m,rtmp,add_EM_F, add_EM_S)
77 gross 798 {
78 gross 853 EM_S=THREAD_MEMALLOC(len_EM_S,double);
79     EM_F=THREAD_MEMALLOC(len_EM_F,double);
80     row_index=THREAD_MEMALLOC(p.row_NN,index_t);
81    
82     if (!Finley_checkPtr(EM_S) && !Finley_checkPtr(EM_F) && !Finley_checkPtr(row_index) ) {
83 gross 798
84 gross 853 #ifndef PASO_MPI
85     for (color=elements->minColor;color<=elements->maxColor;color++) {
86     /* open loop over all elements: */
87     #pragma omp for private(e) schedule(static)
88     for(e=0;e<elements->numElements;e++){
89     if (elements->Color[e]==color) {
90     #else
91     {
92     for(e=0;e<elements->numElements;e++) {
93     {
94     #endif
95     Vol=&(p.row_jac->volume[INDEX2(0,e,p.numQuad)]);
96     DSDX=&(p.row_jac->DSDX[INDEX4(0,0,0,e,p.row_NN,DIM,p.numQuad)]);
97     for (q=0;q<len_EM_S;++q) EM_S[q]=0;
98     for (q=0;q<len_EM_F;++q) EM_F[q]=0;
99     add_EM_F=FALSE;
100     add_EM_S=FALSE;
101     /**************************************************************/
102     /* process A: */
103     /**************************************************************/
104     A_p=getSampleData(A,e);
105     if (NULL!=A_p) {
106     add_EM_S=TRUE;
107     if (extendedA) {
108     for (s=0;s<p.row_NS;s++) {
109     for (r=0;r<p.col_NS;r++) {
110     for (k=0;k<p.numEqu;k++) {
111     for (m=0;m<p.numComp;m++) {
112     rtmp=0.;
113     for (q=0;q<p.numQuad;q++) {
114     rtmp+=Vol[q]*DSDX[INDEX3(s,0,q,p.row_NN,DIM)]*
115     A_p[INDEX5(k,0,m,0,q,p.numEqu,DIM,p.numComp,DIM)]*DSDX[INDEX3(r,0,q,p.row_NN,DIM)];
116     }
117     EM_S[INDEX4(k,m,s,r,p.numEqu,p.numComp,p.row_NN)]+=rtmp;
118     }
119     }
120     }
121     }
122     } else {
123     for (s=0;s<p.row_NS;s++) {
124     for (r=0;r<p.col_NS;r++) {
125     rtmp=0;
126     for (q=0;q<p.numQuad;q++) rtmp+=Vol[q]*DSDX[INDEX3(s,0,q,p.row_NN,DIM)]*DSDX[INDEX3(r,0,q,p.row_NN,DIM)];
127     for (k=0;k<p.numEqu;k++) {
128     for (m=0;m<p.numComp;m++) {
129     EM_S[INDEX4(k,m,s,r,p.numEqu,p.numComp,p.row_NN)]+=rtmp*A_p[INDEX4(k,0,m,0,p.numEqu,DIM,p.numComp)];
130     }
131     }
132     }
133     }
134     }
135     }
136     /**************************************************************/
137     /* process B: */
138     /**************************************************************/
139     B_p=getSampleData(B,e);
140     if (NULL!=B_p) {
141     add_EM_S=TRUE;
142     if (extendedB) {
143     for (s=0;s<p.row_NS;s++) {
144     for (r=0;r<p.col_NS;r++) {
145     for (k=0;k<p.numEqu;k++) {
146     for (m=0;m<p.numComp;m++) {
147     rtmp=0.;
148     for (q=0;q<p.numQuad;q++) {
149     rtmp+=Vol[q]*DSDX[INDEX3(s,0,q,p.row_NN,DIM)]*
150     B_p[INDEX4(k,0,m,q,p.numEqu,DIM,p.numComp)]*S[INDEX2(r,q,p.row_NS)];
151     }
152     EM_S[INDEX4(k,m,s,r,p.numEqu,p.numComp,p.row_NN)]+=rtmp;
153     }
154     }
155     }
156     }
157     } else {
158     for (s=0;s<p.row_NS;s++) {
159     for (r=0;r<p.col_NS;r++) {
160     rtmp=0;
161     for (q=0;q<p.numQuad;q++) rtmp+=Vol[q]*DSDX[INDEX3(s,0,q,p.row_NN,DIM)]*S[INDEX2(r,q,p.row_NS)];
162     for (k=0;k<p.numEqu;k++) {
163     for (m=0;m<p.numComp;m++) {
164     EM_S[INDEX4(k,m,s,r,p.numEqu,p.numComp,p.row_NN)]+=rtmp*B_p[INDEX3(k,0,m,p.numEqu,DIM)];
165     }
166     }
167     }
168     }
169 gross 798 }
170 gross 853 }
171     /**************************************************************/
172     /* process C: */
173     /**************************************************************/
174     C_p=getSampleData(C,e);
175     if (NULL!=C_p) {
176     add_EM_S=TRUE;
177     if (extendedC) {
178     for (s=0;s<p.row_NS;s++) {
179     for (r=0;r<p.col_NS;r++) {
180     for (k=0;k<p.numEqu;k++) {
181     for (m=0;m<p.numComp;m++) {
182     rtmp=0;
183     for (q=0;q<p.numQuad;q++) {
184     rtmp+=Vol[q]*S[INDEX2(s,q,p.row_NS)]*C_p[INDEX4(k,m,0,q,p.numEqu,p.numComp,DIM)]*DSDX[INDEX3(r,0,q,p.row_NN,DIM)];
185     }
186     EM_S[INDEX4(k,m,s,r,p.numEqu,p.numComp,p.row_NN)]+=rtmp;
187     }
188     }
189     }
190     }
191     } else {
192     for (s=0;s<p.row_NS;s++) {
193     for (r=0;r<p.col_NS;r++) {
194     rtmp=0;
195     for (q=0;q<p.numQuad;q++) rtmp+=Vol[q]*S[INDEX2(s,q,p.row_NS)]*DSDX[INDEX3(r,0,q,p.row_NN,DIM)];
196     for (k=0;k<p.numEqu;k++) {
197 gross 798 for (m=0;m<p.numComp;m++) {
198 gross 853 EM_S[INDEX4(k,m,s,r,p.numEqu,p.numComp,p.row_NN)]+=rtmp*C_p[INDEX3(k,m,0,p.numEqu,p.numComp)];
199     }
200     }
201     }
202     }
203     }
204     }
205     /************************************************************* */
206     /* process D */
207     /**************************************************************/
208     D_p=getSampleData(D,e);
209     if (NULL!=D_p) {
210     add_EM_S=TRUE;
211     if (extendedD) {
212     for (s=0;s<p.row_NS;s++) {
213     for (r=0;r<p.col_NS;r++) {
214     for (k=0;k<p.numEqu;k++) {
215     for (m=0;m<p.numComp;m++) {
216     rtmp=0;
217     for (q=0;q<p.numQuad;q++) {
218     rtmp+=Vol[q]*S[INDEX2(s,q,p.row_NS)]*D_p[INDEX3(k,m,q,p.numEqu,p.numComp)]*S[INDEX2(r,q,p.row_NS)];
219     }
220     EM_S[INDEX4(k,m,s,r,p.numEqu,p.numComp,p.row_NN)]+=rtmp;
221    
222 gross 798 }
223 gross 853 }
224     }
225     }
226     } else {
227     for (s=0;s<p.row_NS;s++) {
228     for (r=0;r<p.col_NS;r++) {
229     rtmp=0;
230     for (q=0;q<p.numQuad;q++) rtmp+=Vol[q]*S[INDEX2(s,q,p.row_NS)]*S[INDEX2(r,q,p.row_NS)];
231     for (k=0;k<p.numEqu;k++) {
232     for (m=0;m<p.numComp;m++) {
233     EM_S[INDEX4(k,m,s,r,p.numEqu,p.numComp,p.row_NN)]+=rtmp*D_p[INDEX2(k,m,p.numEqu)];
234     }
235     }
236     }
237     }
238     }
239 gross 798 }
240 gross 853 /**************************************************************/
241     /* process X: */
242     /**************************************************************/
243     X_p=getSampleData(X,e);
244     if (NULL!=X_p) {
245     add_EM_F=TRUE;
246     if (extendedX) {
247     for (s=0;s<p.row_NS;s++) {
248     for (k=0;k<p.numEqu;k++) {
249 gross 798 rtmp=0;
250 gross 853 for (q=0;q<p.numQuad;q++) rtmp+=Vol[q]*DSDX[INDEX3(s,0,q,p.row_NN,DIM)]*X_p[INDEX3(k,0,q,p.numEqu,DIM)];
251     EM_F[INDEX2(k,s,p.numEqu)]+=rtmp;
252     }
253 gross 798 }
254 gross 853 } else {
255     for (s=0;s<p.row_NS;s++) {
256     rtmp=0;
257     for (q=0;q<p.numQuad;q++) rtmp+=Vol[q]*DSDX[INDEX3(s,0,q,p.row_NN,DIM)];
258     for (k=0;k<p.numEqu;k++) EM_F[INDEX2(k,s,p.numEqu)]+=rtmp*X_p[INDEX2(k,0,p.numEqu)];
259 gross 798 }
260 gross 853 }
261 gross 798 }
262 gross 853 /**************************************************************/
263     /* process Y: */
264     /**************************************************************/
265     Y_p=getSampleData(Y,e);
266     if (NULL!=Y_p) {
267     add_EM_F=TRUE;
268     if (extendedY) {
269     for (s=0;s<p.row_NS;s++) {
270     for (k=0;k<p.numEqu;k++) {
271 gross 798 rtmp=0;
272 gross 853 for (q=0;q<p.numQuad;q++) rtmp+=Vol[q]*S[INDEX2(s,q,p.row_NS)]*Y_p[INDEX2(k,q,p.numEqu)];
273     EM_F[INDEX2(k,s,p.numEqu)]+=rtmp;
274     }
275 gross 798 }
276 gross 853 } else {
277     for (s=0;s<p.row_NS;s++) {
278 gross 798 rtmp=0;
279 gross 853 for (q=0;q<p.numQuad;q++) rtmp+=Vol[q]*S[INDEX2(s,q,p.row_NS)];
280     for (k=0;k<p.numEqu;k++) EM_F[INDEX2(k,s,p.numEqu)]+=rtmp*Y_p[k];
281 gross 798 }
282     }
283 gross 853 }
284     /***********************************************************************************************/
285     /* add the element matrices onto the matrix and right hand side */
286     /***********************************************************************************************/
287     for (q=0;q<p.row_NN;q++) row_index[q]=p.row_DOF[elements->Nodes[INDEX2(p.row_node[q],e,p.NN)]];
288     if (add_EM_F) Finley_Util_AddScatter(p.row_NN,row_index,p.numEqu,EM_F,F_p, p.row_DOF_UpperBound);
289     if (add_EM_S) Finley_Assemble_addToSystemMatrix(Mat,p.row_NN,row_index,p.numEqu,p.col_NN,row_index,p.numComp,EM_S);
290    
291     } /* end color check */
292     } /* end element loop */
293     } /* end color loop */
294    
295     THREAD_MEMFREE(EM_S);
296     THREAD_MEMFREE(EM_F);
297     THREAD_MEMFREE(row_index);
298 gross 798
299 gross 853 } /* end of pointer check */
300 gross 798 } /* end parallel region */
301     }
302     /*
303     * $Log$
304     */

  ViewVC Help
Powered by ViewVC 1.1.26