1 |
|
2 |
/******************************************************* |
3 |
* |
4 |
* Copyright (c) 2003-2009 by University of Queensland |
5 |
* Earth Systems Science Computational Center (ESSCC) |
6 |
* http://www.uq.edu.au/esscc |
7 |
* |
8 |
* Primary Business: Queensland, Australia |
9 |
* Licensed under the Open Software License version 3.0 |
10 |
* http://www.opensource.org/licenses/osl-3.0.php |
11 |
* |
12 |
*******************************************************/ |
13 |
|
14 |
|
15 |
/**********************************************************************/ |
16 |
|
17 |
/* Paso: Pattern: Paso_Pattern_coupling |
18 |
|
19 |
searches for a maximal independent set MIS in the matrix pattern |
20 |
vertices in the maximal independent set are marked in mis_marker |
21 |
nodes to be considered are marked by -1 on the input in mis_marker |
22 |
|
23 |
*/ |
24 |
/**********************************************************************/ |
25 |
|
26 |
/* Copyrights by ACcESS Australia 2003,2004,2005 */ |
27 |
/* Author: artak@uq.edu.au */ |
28 |
|
29 |
/**************************************************************/ |
30 |
|
31 |
#include "PasoUtil.h" |
32 |
#include "Pattern_coupling.h" |
33 |
#include <limits.h> |
34 |
|
35 |
|
36 |
/***************************************************************/ |
37 |
|
38 |
#define IS_AVAILABLE -1 |
39 |
#define IS_IN_SET -3 /* Week connection */ |
40 |
#define IS_REMOVED -4 /* strong */ |
41 |
|
42 |
void Paso_Pattern_YS(Paso_SparseMatrix* A, index_t* mis_marker, double threshold) { |
43 |
|
44 |
dim_t i,j; |
45 |
/*double sum;*/ |
46 |
index_t iptr,*index,*where_p,*diagptr; |
47 |
bool_t passed=FALSE; |
48 |
dim_t n=A->numRows; |
49 |
diagptr=MEMALLOC(n,index_t); |
50 |
|
51 |
if (A->pattern->type & PATTERN_FORMAT_SYM) { |
52 |
Paso_setError(TYPE_ERROR,"Paso_Pattern_coup: symmetric matrix pattern is not supported yet"); |
53 |
return; |
54 |
} |
55 |
|
56 |
#pragma omp parallel for private(i) schedule(static) |
57 |
for (i=0;i<n;++i) |
58 |
if(mis_marker[i]==IS_AVAILABLE) |
59 |
mis_marker[i]=IS_REMOVED; |
60 |
|
61 |
#pragma omp parallel for private(i,index,where_p) schedule(static) |
62 |
for (i=0;i<n;++i) { |
63 |
diagptr[i]=A->pattern->ptr[i]; |
64 |
index=&(A->pattern->index[A->pattern->ptr[i]]); |
65 |
where_p=(index_t*)bsearch(&i, |
66 |
index, |
67 |
A->pattern->ptr[i + 1]-A->pattern->ptr[i], |
68 |
sizeof(index_t), |
69 |
Paso_comparIndex); |
70 |
if (where_p==NULL) { |
71 |
Paso_setError(VALUE_ERROR, "Paso_Pattern_coup: main diagonal element missing."); |
72 |
} else { |
73 |
diagptr[i]+=(index_t)(where_p-index); |
74 |
} |
75 |
} |
76 |
|
77 |
|
78 |
|
79 |
/*This loop cannot be parallelized, as order matters here.*/ |
80 |
for (i=0;i<n;++i) { |
81 |
if (mis_marker[i]==IS_REMOVED) { |
82 |
for (iptr=A->pattern->ptr[i];iptr<A->pattern->ptr[i+1]; ++iptr) { |
83 |
j=A->pattern->index[iptr]; |
84 |
if (j!=i && ABS(A->val[iptr])>=threshold*ABS(A->val[diagptr[i]])) { |
85 |
mis_marker[j]=IS_IN_SET; |
86 |
} |
87 |
} |
88 |
} |
89 |
} |
90 |
|
91 |
|
92 |
|
93 |
/*This loop cannot be parallelized, as order matters here.*/ |
94 |
for (i=0;i<n;i++) { |
95 |
if (mis_marker[i]==IS_IN_SET) { |
96 |
passed=TRUE; |
97 |
for (iptr=A->pattern->ptr[i];iptr<A->pattern->ptr[i+1]; ++iptr) { |
98 |
j=A->pattern->index[iptr]; |
99 |
if (mis_marker[j]==IS_REMOVED) { |
100 |
if ((A->val[iptr]/A->val[diagptr[i]])>=-threshold) { |
101 |
passed=TRUE; |
102 |
} |
103 |
else { |
104 |
passed=FALSE; |
105 |
break; |
106 |
} |
107 |
} |
108 |
} |
109 |
if (passed) mis_marker[i]=IS_REMOVED; |
110 |
} |
111 |
} |
112 |
/* This check is to make sure we dont get some nusty rows which were not removed durring coarsening process.*/ |
113 |
/* TODO: we have to mechanism that this does not happend at all, and get rid of this 'If'. */ |
114 |
/*#pragma omp parallel for private(i,iptr,j,sum) schedule(static) |
115 |
for (i=0;i<n;i++) { |
116 |
if (mis_marker[i]==IS_REMOVED) { |
117 |
sum=0; |
118 |
for (iptr=A->pattern->ptr[i];iptr<A->pattern->ptr[i+1]; ++iptr) { |
119 |
j=A->pattern->index[iptr]; |
120 |
if (mis_marker[j]==IS_REMOVED) |
121 |
sum+=A->val[iptr]; |
122 |
} |
123 |
if(ABS(sum)<1.e-25) |
124 |
mis_marker[i]=IS_IN_SET; |
125 |
} |
126 |
} |
127 |
*/ |
128 |
|
129 |
/* swap to TRUE/FALSE in mis_marker */ |
130 |
#pragma omp parallel for private(i) schedule(static) |
131 |
for (i=0;i<n;i++) mis_marker[i]=(mis_marker[i]!=IS_IN_SET); |
132 |
|
133 |
MEMFREE(diagptr); |
134 |
} |
135 |
|
136 |
/* |
137 |
* Ruge-Stueben strength of connection mask. |
138 |
* |
139 |
*/ |
140 |
void Paso_Pattern_RS(Paso_SparseMatrix* A, index_t* mis_marker, double theta) |
141 |
{ |
142 |
dim_t i,n,j; |
143 |
index_t iptr; |
144 |
double threshold,max_offdiagonal; |
145 |
|
146 |
Paso_Pattern *out=NULL; |
147 |
|
148 |
Paso_IndexList* index_list=NULL; |
149 |
|
150 |
index_list=TMPMEMALLOC(A->pattern->numOutput,Paso_IndexList); |
151 |
if (! Paso_checkPtr(index_list)) { |
152 |
#pragma omp parallel for private(i) schedule(static) |
153 |
for(i=0;i<A->pattern->numOutput;++i) { |
154 |
index_list[i].extension=NULL; |
155 |
index_list[i].n=0; |
156 |
} |
157 |
} |
158 |
|
159 |
|
160 |
n=A->numRows; |
161 |
if (A->pattern->type & PATTERN_FORMAT_SYM) { |
162 |
Paso_setError(TYPE_ERROR,"Paso_Pattern_RS: symmetric matrix pattern is not supported yet"); |
163 |
return; |
164 |
} |
165 |
/*#pragma omp parallel for private(i,iptr,max_offdiagonal,threshold,j) schedule(static)*/ |
166 |
for (i=0;i<n;++i) { |
167 |
if(mis_marker[i]==IS_AVAILABLE) { |
168 |
max_offdiagonal = DBL_MIN; |
169 |
for (iptr=A->pattern->ptr[i];iptr<A->pattern->ptr[i+1]; ++iptr) { |
170 |
if(A->pattern->index[iptr] != i){ |
171 |
max_offdiagonal = MAX(max_offdiagonal,-A->val[iptr]); |
172 |
} |
173 |
} |
174 |
|
175 |
threshold = theta*max_offdiagonal; |
176 |
for (iptr=A->pattern->ptr[i];iptr<A->pattern->ptr[i+1]; ++iptr) { |
177 |
j=A->pattern->index[iptr]; |
178 |
if((-A->val[iptr])>=threshold) { |
179 |
Paso_IndexList_insertIndex(&(index_list[i]),j); |
180 |
Paso_IndexList_insertIndex(&(index_list[j]),i); |
181 |
} |
182 |
} |
183 |
} |
184 |
} |
185 |
|
186 |
|
187 |
out=Paso_IndexList_createPattern(0, A->pattern->numOutput,index_list,0,A->pattern->numInput,0); |
188 |
|
189 |
/* clean up */ |
190 |
if (index_list!=NULL) { |
191 |
#pragma omp parallel for private(i) schedule(static) |
192 |
for(i=0;i<A->pattern->numOutput;++i) Paso_IndexList_free(index_list[i].extension); |
193 |
} |
194 |
TMPMEMFREE(index_list); |
195 |
|
196 |
/*Paso_Pattern_mis(out,mis_marker);*/ |
197 |
Paso_Pattern_greedy(out,mis_marker); |
198 |
Paso_Pattern_free(out); |
199 |
} |
200 |
|
201 |
void Paso_Pattern_Aggregiation(Paso_SparseMatrix* A, index_t* mis_marker, double theta) |
202 |
{ |
203 |
dim_t i,j,n; |
204 |
index_t iptr; |
205 |
double diag,eps_Aii,val; |
206 |
double* diags; |
207 |
|
208 |
|
209 |
Paso_Pattern *out=NULL; |
210 |
Paso_IndexList* index_list=NULL; |
211 |
|
212 |
n=A->numRows; |
213 |
diags=MEMALLOC(n,double); |
214 |
|
215 |
index_list=TMPMEMALLOC(A->pattern->numOutput,Paso_IndexList); |
216 |
if (! Paso_checkPtr(index_list)) { |
217 |
#pragma omp parallel for private(i) schedule(static) |
218 |
for(i=0;i<A->pattern->numOutput;++i) { |
219 |
index_list[i].extension=NULL; |
220 |
index_list[i].n=0; |
221 |
} |
222 |
} |
223 |
|
224 |
if (A->pattern->type & PATTERN_FORMAT_SYM) { |
225 |
Paso_setError(TYPE_ERROR,"Paso_Pattern_Aggregiation: symmetric matrix pattern is not supported yet"); |
226 |
return; |
227 |
} |
228 |
|
229 |
|
230 |
#pragma omp parallel for private(i,iptr,diag) schedule(static) |
231 |
for (i=0;i<n;++i) { |
232 |
diag = 0; |
233 |
for (iptr=A->pattern->ptr[i];iptr<A->pattern->ptr[i+1]; ++iptr) { |
234 |
if(A->pattern->index[iptr] == i){ |
235 |
diag+=A->val[iptr]; |
236 |
} |
237 |
} |
238 |
diags[i]=ABS(diag); |
239 |
} |
240 |
|
241 |
|
242 |
#pragma omp parallel for private(i,iptr,j,val,eps_Aii) schedule(static) |
243 |
for (i=0;i<n;++i) { |
244 |
if (mis_marker[i]==IS_AVAILABLE) { |
245 |
eps_Aii = theta*theta*diags[i]; |
246 |
for (iptr=A->pattern->ptr[i];iptr<A->pattern->ptr[i+1]; ++iptr) { |
247 |
j=A->pattern->index[iptr]; |
248 |
val=A->val[iptr]; |
249 |
if((val*val)>=(eps_Aii*diags[j])) { |
250 |
Paso_IndexList_insertIndex(&(index_list[i]),j); |
251 |
} |
252 |
} |
253 |
} |
254 |
} |
255 |
|
256 |
out=Paso_IndexList_createPattern(0, A->pattern->numOutput,index_list,0,A->pattern->numInput,0); |
257 |
|
258 |
/* clean up */ |
259 |
if (index_list!=NULL) { |
260 |
#pragma omp parallel for private(i) schedule(static) |
261 |
for(i=0;i<A->pattern->numOutput;++i) Paso_IndexList_free(index_list[i].extension); |
262 |
} |
263 |
|
264 |
TMPMEMFREE(index_list); |
265 |
MEMFREE(diags); |
266 |
|
267 |
|
268 |
/*Paso_Pattern_mis(out,mis_marker);*/ |
269 |
Paso_Pattern_greedy(out,mis_marker); |
270 |
Paso_Pattern_free(out); |
271 |
|
272 |
} |
273 |
|
274 |
/* Greedy algorithm */ |
275 |
void Paso_Pattern_greedy(Paso_Pattern* pattern, index_t* mis_marker) { |
276 |
|
277 |
dim_t i,j; |
278 |
/*double sum;*/ |
279 |
index_t iptr; |
280 |
bool_t passed=FALSE; |
281 |
dim_t n=pattern->numOutput; |
282 |
|
283 |
if (pattern->type & PATTERN_FORMAT_SYM) { |
284 |
Paso_setError(TYPE_ERROR,"Paso_Pattern_greedy: symmetric matrix pattern is not supported yet"); |
285 |
return; |
286 |
} |
287 |
|
288 |
/* We do not need this loop if we set IS_REMOVED=IS_AVAILABLE. */ |
289 |
#pragma omp parallel for private(i) schedule(static) |
290 |
for (i=0;i<n;++i) |
291 |
if(mis_marker[i]==IS_AVAILABLE) |
292 |
mis_marker[i]=IS_IN_SET; |
293 |
|
294 |
|
295 |
for (i=0;i<n;++i) { |
296 |
if (mis_marker[i]==IS_IN_SET) { |
297 |
for (iptr=pattern->ptr[i];iptr<pattern->ptr[i+1]; ++iptr) { |
298 |
j=pattern->index[iptr]; |
299 |
mis_marker[j]=IS_REMOVED; |
300 |
} |
301 |
} |
302 |
} |
303 |
|
304 |
|
305 |
|
306 |
for (i=0;i<n;i++) { |
307 |
if (mis_marker[i]==IS_REMOVED) { |
308 |
passed=TRUE; |
309 |
for (iptr=pattern->ptr[i];iptr<pattern->ptr[i+1]; ++iptr) { |
310 |
j=pattern->index[iptr]; |
311 |
if (mis_marker[j]==IS_REMOVED) { |
312 |
passed=TRUE; |
313 |
} |
314 |
else { |
315 |
passed=FALSE; |
316 |
break; |
317 |
} |
318 |
} |
319 |
if (passed) mis_marker[i]=IS_IN_SET; |
320 |
} |
321 |
} |
322 |
|
323 |
/* swap to TRUE/FALSE in mis_marker */ |
324 |
#pragma omp parallel for private(i) schedule(static) |
325 |
for (i=0;i<n;i++) mis_marker[i]=(mis_marker[i]!=IS_REMOVED); |
326 |
|
327 |
} |
328 |
|
329 |
|
330 |
void Paso_Pattern_greedy_color(Paso_Pattern* pattern, index_t* mis_marker) { |
331 |
|
332 |
dim_t i,j; |
333 |
/*double sum;*/ |
334 |
index_t iptr; |
335 |
index_t num_colors; |
336 |
index_t* colorOf; |
337 |
register index_t color; |
338 |
bool_t passed=FALSE; |
339 |
dim_t n=pattern->numOutput; |
340 |
|
341 |
|
342 |
colorOf=MEMALLOC(n,index_t); |
343 |
|
344 |
if (pattern->type & PATTERN_FORMAT_SYM) { |
345 |
Paso_setError(TYPE_ERROR,"Paso_Pattern_greedy: symmetric matrix pattern is not supported yet"); |
346 |
return; |
347 |
} |
348 |
|
349 |
Paso_Pattern_color(pattern,&num_colors,colorOf); |
350 |
|
351 |
/* We do not need this loop if we set IS_IN_MIS=IS_AVAILABLE. */ |
352 |
#pragma omp parallel for private(i) schedule(static) |
353 |
for (i=0;i<n;++i) |
354 |
if(mis_marker[i]==IS_AVAILABLE) |
355 |
mis_marker[i]=IS_IN_SET; |
356 |
|
357 |
#pragma omp barrier |
358 |
for (color=0;color<num_colors;++color) { |
359 |
#pragma omp parallel for schedule(static) private(i,iptr,j) |
360 |
for (i=0;i<n;++i) { |
361 |
if (colorOf[i]==color) { |
362 |
if (mis_marker[i]==IS_IN_SET) { |
363 |
for (iptr=pattern->ptr[i];iptr<pattern->ptr[i+1]; ++iptr) { |
364 |
j=pattern->index[iptr]; |
365 |
if (colorOf[j]<color) |
366 |
mis_marker[j]=IS_REMOVED; |
367 |
} |
368 |
} |
369 |
} |
370 |
} |
371 |
} |
372 |
|
373 |
|
374 |
#pragma omp barrier |
375 |
for (color=0;color<num_colors;++color) { |
376 |
#pragma omp parallel for schedule(static) private(i,iptr,j) |
377 |
for (i=0;i<n;i++) { |
378 |
if (colorOf[i]==color) { |
379 |
if (mis_marker[i]==IS_REMOVED) { |
380 |
passed=TRUE; |
381 |
for (iptr=pattern->ptr[i];iptr<pattern->ptr[i+1]; ++iptr) { |
382 |
j=pattern->index[iptr]; |
383 |
if (colorOf[j]<color && passed) { |
384 |
if (mis_marker[j]==IS_REMOVED) { |
385 |
passed=TRUE; |
386 |
} |
387 |
else { |
388 |
passed=FALSE; |
389 |
/*break;*/ |
390 |
} |
391 |
} |
392 |
} |
393 |
if (passed) mis_marker[i]=IS_IN_SET; |
394 |
} |
395 |
|
396 |
} |
397 |
} |
398 |
} |
399 |
|
400 |
/* swap to TRUE/FALSE in mis_marker */ |
401 |
#pragma omp parallel for private(i) schedule(static) |
402 |
for (i=0;i<n;i++) mis_marker[i]=(mis_marker[i]!=IS_IN_SET); |
403 |
|
404 |
MEMFREE(colorOf); |
405 |
} |
406 |
|
407 |
#undef IS_AVAILABLE |
408 |
#undef IS_IN_SET |
409 |
#undef IS_REMOVED |
410 |
|
411 |
|
412 |
|
413 |
|
414 |
|
415 |
|
416 |
|
417 |
|
418 |
|
419 |
|