# Here is a way to use MPI with Basilisk:
# Here you just have to change the name of your file "file_name.c"
# and change the value of "N" for example, if you want to
#use 4 processor remove N and set up 4
qcc -source -grid=quadtree -D_MPI=1 file_name.c
mpicc -std=c99 -D_XOPEN_SOURCE=700 _file_name.c -lm -O2 -Wall -o parallel mpiexec -np N ./parallel
# And in the terminal write this command to execute your code:
>> sh Brun.sh
# It's compatible with output_vtu.h, just instead to use this:
//int pt = 0;
//event interface(t += time_step; t <= final_time) {
// scalar *list;
// bool linear;
// char name[80];
// sprintf(name, "snap-%d.vtu",pt);
// FILE *ptr = fopen(name,"w");
// output_vtu_bin_foreach((scalar *) {f, p}, (vector *) {u}, ptr, false);
// fclose(ptr);
//
// pt = pt + 1;
//}
# Replace in your code by this one:
//int pt = 0;
//event interface(t += time_step; t <= final_time) {
// char name[80];
// scalar* list = {f, p, u};
// sprintf (name, "snap-%d", pt);
// p.nodump = false;
// dump(name, list);
//
// pt = pt + 1;
//}
# And don't forget to remove #include "output_vtu.h"
# in your file_name.c
# Thanks to Yashika and Anil for sharing this code.
