Aquila-Consortium/libsharp2
0
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

use best of several tries for small SHTs

Browse source
This commit is contained in:
Martin Reinecke 2013-02-08 18:30:25 +01:00
parent 7c8039d34e
commit 515ef8a4aa
1 changed files with 23 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

29
libsharp/sharp_testsuite.c
View file

@ -57,10 +57,8 @@ static double drand (double min, double max, int *state)
return min + (max-min)*(*state)/(0x7fffffff+1.0); return min + (max-min)*(*state)/(0x7fffffff+1.0);
} }
static void random_alm (dcmplx *alm, sharp_alm_info *helper, int spin) static void random_alm (dcmplx *alm, sharp_alm_info *helper, int spin, int cnt)
{ {
static int cnt=0;
++cnt;
#pragma omp parallel #pragma omp parallel
{ {
int mi; int mi;
@ -459,11 +457,10 @@ static void do_sht (sharp_geom_info *ginfo, sharp_alm_info *ainfo,
for (int i=0; i<ncomp; ++i) for (int i=0; i<ncomp; ++i)
SET_ARRAY(map[i],0,(int)npix,0); SET_ARRAY(map[i],0,(int)npix,0);
srand(4);
dcmplx **alm; dcmplx **alm;
ALLOC2D(alm,dcmplx,ncomp,nalms); ALLOC2D(alm,dcmplx,ncomp,nalms);
for (int i=0; i<ncomp; ++i) for (int i=0; i<ncomp; ++i)
random_alm(alm[i],ainfo,spin); random_alm(alm[i],ainfo,spin,i+1);
#ifdef USE_MPI #ifdef USE_MPI
sharp_execute_mpi(MPI_COMM_WORLD,SHARP_ALM2MAP,spin,&alm[0],&map[0],ginfo, sharp_execute_mpi(MPI_COMM_WORLD,SHARP_ALM2MAP,spin,&alm[0],&map[0],ginfo,
@ -551,11 +548,29 @@ static void sharp_test (int argc, const char **argv)
get_infos (argv[2], lmax, &mmax, &gpar1, &gpar2, &ginfo, &ainfo); get_infos (argv[2], lmax, &mmax, &gpar1, &gpar2, &ginfo, &ainfo);
int ncomp = ntrans*((spin==0) ? 1 : 2); int ncomp = ntrans*((spin==0) ? 1 : 2);
double t_a2m, t_m2a; double t_a2m=1e30, t_m2a=1e30;
unsigned long long op_a2m, op_m2a; unsigned long long op_a2m, op_m2a;
double *err_abs,*err_rel; double *err_abs,*err_rel;
do_sht (ginfo, ainfo, spin, ntrans, 0, &err_abs, &err_rel, &t_a2m, &t_m2a,
double t_acc=0;
int nrpt=0;
while(1)
{
++nrpt;
double ta2m2, tm2a2;
do_sht (ginfo, ainfo, spin, ntrans, 0, &err_abs, &err_rel, &ta2m2, &tm2a2,
&op_a2m, &op_m2a); &op_a2m, &op_m2a);
if (ta2m2<t_a2m) t_a2m=ta2m2;
if (tm2a2<t_m2a) t_m2a=tm2a2;
t_acc+=t_a2m+t_m2a;
if (t_acc>2.)
{
if (mytask==0) printf("Best of %d runs\n",nrpt);
break;
}
DEALLOC(err_abs);
DEALLOC(err_rel);
}
if (mytask==0) printf("wall time for alm2map: %fs\n",t_a2m); if (mytask==0) printf("wall time for alm2map: %fs\n",t_a2m);
if (mytask==0) printf("Performance: %fGFLOPs/s\n",1e-9*op_a2m/t_a2m); if (mytask==0) printf("Performance: %fGFLOPs/s\n",1e-9*op_a2m/t_a2m);