portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-04-09 21:50:45 +01:00
Code Releases Activity

Coalesced loops

Browse Source
This commit is contained in:
Peter Boyle 2019-06-15 08:35:26 +01:00
parent bb024dd114
commit 09720c40cd
1 changed files with 46 additions and 41 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

87
Grid/qcd/utils/LinalgUtils.h
View File

@ -49,11 +49,9 @@ void axpibg5x(Lattice<vobj> &z,const Lattice<vobj> &x,Coeff a,Coeff b)
Gamma G5(Gamma::Algebra::Gamma5); Gamma G5(Gamma::Algebra::Gamma5);
auto x_v = x.View(); auto x_v = x.View();
auto z_v = z.View(); auto z_v = z.View();
accelerator_loop( ss, x_v,{ accelerator_for( ss, x_v.size(),vobj::Nsimd(), {
vobj tmp; auto tmp = a*x_v(ss) + G5*(b*timesI(x_v(ss)));
tmp = a*x_v[ss]; coalescedWrite(z_v[ss],tmp);
tmp = tmp + G5*(b*timesI(x_v[ss]));
vstream(z_v[ss],tmp);
}); });
} }
@ -69,9 +67,12 @@ void axpby_ssp(Lattice<vobj> &z, Coeff a,const Lattice<vobj> &x,Coeff b,const La
auto y_v = y.View(); auto y_v = y.View();
auto z_v = z.View(); auto z_v = z.View();
// FIXME -- need a new class of accelerator_loop to implement this // FIXME -- need a new class of accelerator_loop to implement this
thread_loop( (int ss=0;ss<grid->oSites();ss+=Ls),{ // adds Ls //
vobj tmp = a*x_v[ss+s]+b*y_v[ss+sp]; uint64_t nloop = grid->oSites()/Ls;
vstream(z_v[ss+s],tmp); accelerator_for(sss,nloop,vobj::Nsimd(),{
uint64_t ss = sss*Ls;
auto tmp = a*x_v(ss+s)+b*y_v(ss+sp);
coalescedWrite(z_v[ss+s],tmp);
}); });
} }
@ -87,11 +88,11 @@ void ag5xpby_ssp(Lattice<vobj> &z,Coeff a,const Lattice<vobj> &x,Coeff b,const L
auto x_v = x.View(); auto x_v = x.View();
auto y_v = y.View(); auto y_v = y.View();
auto z_v = z.View(); auto z_v = z.View();
thread_loop((int ss=0;ss<grid->oSites();ss+=Ls),{ // adds Ls uint64_t nloop = grid->oSites()/Ls;
vobj tmp; accelerator_for(sss,nloop,vobj::Nsimd(),{
tmp = G5*x_v[ss+s]*a; uint64_t ss = sss*Ls;
tmp = tmp + b*y_v[ss+sp]; auto tmp = G5*x_v(ss+s)*a + b*y_v(ss+sp);
vstream(z_v[ss+s],tmp); coalescedWrite(z_v[ss+s],tmp);
}); });
} }
@ -107,11 +108,11 @@ void axpbg5y_ssp(Lattice<vobj> &z,Coeff a,const Lattice<vobj> &x,Coeff b,const L
auto y_v = y.View(); auto y_v = y.View();
auto z_v = z.View(); auto z_v = z.View();
Gamma G5(Gamma::Algebra::Gamma5); Gamma G5(Gamma::Algebra::Gamma5);
thread_loop((int ss=0;ss<grid->oSites();ss+=Ls),{ // adds Ls uint64_t nloop = grid->oSites()/Ls;
vobj tmp; accelerator_for(sss,nloop,vobj::Nsimd(),{
tmp = G5*y_v[ss+sp]*b; uint64_t ss = sss*Ls;
tmp = tmp + a*x_v[ss+s]; auto tmp = G5*y_v(ss+sp)*b + a*x_v(ss+s);
vstream(z_v[ss+s],tmp); coalescedWrite(z_v[ss+s],tmp);
}); });
} }
@ -128,12 +129,12 @@ void ag5xpbg5y_ssp(Lattice<vobj> &z,Coeff a,const Lattice<vobj> &x,Coeff b,const
auto y_v = y.View(); auto y_v = y.View();
auto z_v = z.View(); auto z_v = z.View();
Gamma G5(Gamma::Algebra::Gamma5); Gamma G5(Gamma::Algebra::Gamma5);
thread_loop((int ss=0;ss<grid->oSites();ss+=Ls),{ // adds Ls uint64_t nloop = grid->oSites()/Ls;
vobj tmp1; accelerator_for(sss,nloop,vobj::Nsimd(),{
vobj tmp2; uint64_t ss = sss*Ls;
tmp1 = a*x_v[ss+s]+b*y_v[ss+sp]; auto tmp1 = a*x_v(ss+s)+b*y_v(ss+sp);
tmp2 = G5*tmp1; auto tmp2 = G5*tmp1;
vstream(z_v[ss+s],tmp2); coalescedWrite(z_v[ss+s],tmp2);
}); });
} }
@ -149,11 +150,13 @@ void axpby_ssp_pminus(Lattice<vobj> &z,Coeff a,const Lattice<vobj> &x,Coeff b,co
auto x_v = x.View(); auto x_v = x.View();
auto y_v = y.View(); auto y_v = y.View();
auto z_v = z.View(); auto z_v = z.View();
thread_loop((int ss=0;ss<grid->oSites();ss+=Ls),{ // adds Ls uint64_t nloop = grid->oSites()/Ls;
vobj tmp; accelerator_for(sss,nloop,vobj::Nsimd(),{
spProj5m(tmp,y_v[ss+sp]); uint64_t ss = sss*Ls;
tmp = a*x_v[ss+s]+b*tmp; decltype(coalescedRead(y_v[ss+sp])) tmp;
vstream(z_v[ss+s],tmp); spProj5m(tmp,y_v(ss+sp));
tmp = a*x_v(ss+s)+b*tmp;
coalescedWrite(z_v[ss+s],tmp);
}); });
} }
@ -168,11 +171,13 @@ void axpby_ssp_pplus(Lattice<vobj> &z,Coeff a,const Lattice<vobj> &x,Coeff b,con
auto x_v = x.View(); auto x_v = x.View();
auto y_v = y.View(); auto y_v = y.View();
auto z_v = z.View(); auto z_v = z.View();
thread_loop((int ss=0;ss<grid->oSites();ss+=Ls),{ // adds Ls uint64_t nloop = grid->oSites()/Ls;
vobj tmp; accelerator_for(sss,nloop,vobj::Nsimd(),{
spProj5p(tmp,y_v[ss+sp]); uint64_t ss = sss*Ls;
tmp = a*x_v[ss+s]+b*tmp; decltype(coalescedRead(y_v[ss+sp])) tmp;
vstream(z_v[ss+s],tmp); spProj5p(tmp,y_v(ss+sp));
tmp = a*x_v(ss+s)+b*tmp;
coalescedWrite(z_v[ss+s],tmp);
}); });
} }
@ -186,12 +191,12 @@ void G5R5(Lattice<vobj> &z,const Lattice<vobj> &x)
Gamma G5(Gamma::Algebra::Gamma5); Gamma G5(Gamma::Algebra::Gamma5);
auto x_v = x.View(); auto x_v = x.View();
auto z_v = z.View(); auto z_v = z.View();
thread_loop((int ss=0;ss<grid->oSites();ss+=Ls) { uint64_t nloop = grid->oSites()/Ls;
vobj tmp; accelerator_for(sss,nloop,vobj::Nsimd(),{
uint64_t ss = sss*Ls;
for(int s=0;s<Ls;s++){ for(int s=0;s<Ls;s++){
int sp = Ls-1-s; int sp = Ls-1-s;
tmp = G5*x_v[ss+s]; coalescedWrite(z_v[ss+sp],G5*x_v(ss+s));
vstream(z_v[ss+sp],tmp);
} }
}); });
} }
@ -220,13 +225,13 @@ void G5C(Lattice<iVector<CComplex, nbasis>> &z, const Lattice<iVector<CComplex,
auto z_v = z.View(); auto z_v = z.View();
auto x_v = x.View(); auto x_v = x.View();
thread_loop( (int ss = 0; ss < grid->oSites(); ss++) , accelerator_for(ss,grid->oSites(),CComplex::Nsimd(),
{ {
for(int n = 0; n < nb; ++n) { for(int n = 0; n < nb; ++n) {
z_v[ss](n) = x_v[ss](n); coalescedWrite(z_v[ss](n), x_v(ss)(n));
} }
for(int n = nb; n < nbasis; ++n) { for(int n = nb; n < nbasis; ++n) {
z_v[ss](n) = -x_v[ss](n); coalescedWrite(z_v[ss](n), -x_v(ss)(n));
} }
}); });
} }