1 /* CalculiX - A 3-dimensional finite element program */
2 /* Copyright (C) 1998-2021 Guido Dhondt */
3
4 /* This program is free software; you can redistribute it and/or */
5 /* modify it under the terms of the GNU General Public License as */
6 /* published by the Free Software Foundation(version 2); */
7 /* */
8
9 /* This program is distributed in the hope that it will be useful, */
10 /* but WITHOUT ANY WARRANTY; without even the implied warranty of */
11 /* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the */
12 /* GNU General Public License for more details. */
13
14 /* You should have received a copy of the GNU General Public License */
15 /* along with this program; if not, write to the Free Software */
16 /* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */
17
18 /* A parallel copy of arrays which depent on active element */
19
20
21 #include <unistd.h>
22 #include <stdio.h>
23 #include <math.h>
24 #include <stdlib.h>
25 #include <pthread.h>
26 #include "CalculiX.h"
27
28 static ITG *neapar=NULL,*nebpar=NULL;
29
30 static double *b1,*scal11,*fext1,*f1,*alpha1,*fextini1,*fini1,
31 *adb1,*aux21;
32
res2parll(double * b,double * scal1,double * fext,double * f,double * alpha,double * fextini,double * fini,double * adb,double * aux2,ITG * neq0,ITG * num_cpus)33 void res2parll(double *b,double *scal1,double *fext,double *f,
34 double *alpha,double *fextini,double *fini,
35 double *adb,double *aux2,ITG *neq0,ITG *num_cpus){
36
37 ITG i,idelta,isum;
38
39 /* variables for multithreading procedure */
40
41 ITG *ithread=NULL;
42
43 pthread_t tid[*num_cpus];
44
45 /* determining the element bounds in each thread */
46
47 NNEW(neapar,ITG,*num_cpus);
48 NNEW(nebpar,ITG,*num_cpus);
49
50 /* dividing the element number range into num_cpus equal numbers of
51 active entries. */
52
53 idelta=(ITG)floor(*neq0/(double)(*num_cpus));
54 isum=0;
55 for(i=0;i<*num_cpus;i++){
56 neapar[i]=isum;
57 if(i!=*num_cpus-1){
58 isum+=idelta;
59 }else{
60 isum=*neq0;
61 }
62 nebpar[i]=isum;
63 }
64
65 /* create threads and wait */
66
67 b1=b;scal11=scal1;fext1=fext;f1=f;alpha1=alpha;fextini1=fextini;
68 fini1=fini;adb1=adb;aux21=aux2;
69
70 NNEW(ithread,ITG,*num_cpus);
71
72 for(i=0; i<*num_cpus; i++) {
73 ithread[i]=i;
74 pthread_create(&tid[i], NULL, (void *)res2parllmt, (void *)&ithread[i]);
75 }
76 for(i=0; i<*num_cpus; i++) pthread_join(tid[i], NULL);
77
78 SFREE(ithread);SFREE(neapar);SFREE(nebpar);
79
80 }
81
82 /* subroutine for multithreading of res2parll */
83
res2parllmt(ITG * i)84 void *res2parllmt(ITG *i){
85
86 ITG nea,neb,k;
87
88 nea=neapar[*i];
89 neb=nebpar[*i];
90
91 for(k=nea;k<neb;++k){
92 b1[k]=*scal11*(fext1[k]-f1[k])-alpha1[0]*(fextini1[k]-fini1[k])
93 -adb1[k]*aux21[k];
94 }
95
96 return NULL;
97 }
98