portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-06-12 20:27:06 +01:00
Code Releases Activity

Enable reordering of the loops in the assembler for cache friendly.

Browse Source 
This gets in the way of L2 prefetching however. Do next next link in stencil
prefetching.
This commit is contained in:
paboyle
2016-06-19 11:45:58 -07:00
parent d6737e4bd8
commit 6d58cb2a68
15 changed files with 670 additions and 116 deletions
Download Patch File Download Diff File Expand all files Collapse all files

358
benchmarks/Benchmark_dwf_sweep.cc Normal file
View File

@ -0,0 +1,358 @@
/*************************************************************************************
Grid physics library, www.github.com/paboyle/Grid
Source file: ./benchmarks/Benchmark_dwf.cc
Copyright (C) 2015
Author: Peter Boyle <paboyle@ph.ed.ac.uk>
Author: paboyle <paboyle@ph.ed.ac.uk>
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License along
with this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
See the full license in the file "LICENSE" in the top level distribution directory
*************************************************************************************/
/* END LEGAL */
#include <Grid.h>
#include <PerfCount.h>
using namespace std;
using namespace Grid;
using namespace Grid::QCD;
template<class d>
struct scal {
d internal;
};
Gamma::GammaMatrix Gmu [] = {
Gamma::GammaX,
Gamma::GammaY,
Gamma::GammaZ,
Gamma::GammaT
};
void benchDw(std::vector<int> & L, int Ls, int threads, int report =0 );
void benchsDw(std::vector<int> & L, int Ls, int threads, int report=0 );
int main (int argc, char ** argv)
{
Grid_init(&argc,&argv);
const int Ls=16;
int threads = GridThread::GetThreads();
std::cout<<GridLogMessage << "Grid is setup to use "<<threads<<" threads"<<std::endl;
if ( getenv("ASMOPT") ) {
QCD::WilsonKernelsStatic::AsmOpt=1;
} else {
QCD::WilsonKernelsStatic::AsmOpt=0;
}
std::cout<<GridLogMessage << "=========================================================================="<<std::endl;
std::cout<<GridLogMessage << "Volume \t\t\tProcs \t Dw \t eoDw \t sDw \t eosDw (Mflop/s) "<<std::endl;
std::cout<<GridLogMessage << "=========================================================================="<<std::endl;
int Lmax=32;
if ( getenv("LMAX") ) Lmax=atoi(getenv("LMAX"));
for (int L=8;L<Lmax;L*=2){
std::vector<int> latt4(4,L);
for(int d=4;d>0;d--){
if ( d<=3 ) latt4[d]*=2;
std::cout << GridLogMessage <<"\t";
for(int d=0;d<Nd;d++){
std::cout<<latt4[d]<<"x";
}
std::cout <<Ls<<"\t" ;
benchDw (latt4,Ls,threads,0);
benchsDw(latt4,Ls,threads,0);
std::cout<<std::endl;
}
}
std::cout<<GridLogMessage << "=========================================================================="<<std::endl;
{
std::vector<int> latt4(4,16);
std::cout<<GridLogMessage << "16^4 Dw miss rate"<<std::endl;
benchDw (latt4,Ls,threads,1);
std::cout<<GridLogMessage << "16^4 sDw miss rate"<<std::endl;
benchsDw(latt4,Ls,threads,1);
}
Grid_finalize();
}
#undef CHECK
void benchDw(std::vector<int> & latt4, int Ls, int threads,int report )
{
GridCartesian * UGrid = SpaceTimeGrid::makeFourDimGrid(latt4, GridDefaultSimd(Nd,vComplex::Nsimd()),GridDefaultMpi());
GridRedBlackCartesian * UrbGrid = SpaceTimeGrid::makeFourDimRedBlackGrid(UGrid);
GridCartesian * FGrid = SpaceTimeGrid::makeFiveDimGrid(Ls,UGrid);
GridRedBlackCartesian * FrbGrid = SpaceTimeGrid::makeFiveDimRedBlackGrid(Ls,UGrid);
std::vector<int> seeds4({1,2,3,4});
std::vector<int> seeds5({5,6,7,8});
#ifdef CHECK
GridParallelRNG RNG4(UGrid); RNG4.SeedFixedIntegers(seeds4);
GridParallelRNG RNG5(FGrid); RNG5.SeedFixedIntegers(seeds5);
LatticeFermion src (FGrid); random(RNG5,src);
LatticeGaugeField Umu(UGrid);
random(RNG4,Umu);
#else
LatticeFermion src (FGrid); src=zero;
LatticeGaugeField Umu(UGrid); Umu=zero;
#endif
LatticeFermion result(FGrid); result=zero;
LatticeFermion ref(FGrid); ref=zero;
LatticeFermion tmp(FGrid);
LatticeFermion err(FGrid);
ColourMatrix cm = Complex(1.0,0.0);
LatticeGaugeField Umu5d(FGrid);
// replicate across fifth dimension
for(int ss=0;ss<Umu._grid->oSites();ss++){
for(int s=0;s<Ls;s++){
Umu5d._odata[Ls*ss+s] = Umu._odata[ss];
}
}
////////////////////////////////////
// Naive wilson implementation
////////////////////////////////////
std::vector<LatticeColourMatrix> U(4,FGrid);
for(int mu=0;mu<Nd;mu++){
U[mu] = PeekIndex<LorentzIndex>(Umu5d,mu);
}
#ifdef CHECK
if (1)
{
ref = zero;
for(int mu=0;mu<Nd;mu++){
tmp = U[mu]*Cshift(src,mu+1,1);
ref=ref + tmp - Gamma(Gmu[mu])*tmp;
tmp =adj(U[mu])*src;
tmp =Cshift(tmp,mu+1,-1);
ref=ref + tmp + Gamma(Gmu[mu])*tmp;
}
ref = -0.5*ref;
}
#endif
RealD mass=0.1;
RealD M5 =1.8;
RealD NP = UGrid->_Nprocessors;
DomainWallFermionR Dw(Umu,*FGrid,*FrbGrid,*UGrid,*UrbGrid,mass,M5);
double t0=usecond();
Dw.Dhop(src,result,0);
double t1=usecond();
int ncall =1+(int) ((5.0*1000*1000)/(t1-t0));
if (ncall < 5 ) exit(0);
Dw.Dhop(src,result,0);
PerformanceCounter Counter(8);
Counter.Start();
t0=usecond();
for(int i=0;i<ncall;i++){
Dw.Dhop(src,result,0);
}
t1=usecond();
Counter.Stop();
if ( report ) {
Counter.Report();
}
if ( ! report )
{
double volume=Ls; for(int mu=0;mu<Nd;mu++) volume=volume*latt4[mu];
double flops=1344*volume*ncall;
std::cout <<"\t"<<NP<< "\t"<<flops/(t1-t0)<< "\t";
}
#ifdef CHECK
err = ref-result;
RealD errd = norm2(err);
if ( errd> 1.0e-4 ) {
std::cout<<GridLogMessage << "oops !!! norm diff "<< norm2(err)<<std::endl;
exit(-1);
}
#endif
LatticeFermion src_e (FrbGrid);
LatticeFermion src_o (FrbGrid);
LatticeFermion r_e (FrbGrid);
LatticeFermion r_o (FrbGrid);
LatticeFermion r_eo (FGrid);
pickCheckerboard(Even,src_e,src);
pickCheckerboard(Odd,src_o,src);
{
Dw.DhopEO(src_o,r_e,DaggerNo);
double t0=usecond();
for(int i=0;i<ncall;i++){
Dw.DhopEO(src_o,r_e,DaggerNo);
}
double t1=usecond();
if(!report){
double volume=Ls; for(int mu=0;mu<Nd;mu++) volume=volume*latt4[mu];
double flops=(1344.0*volume*ncall)/2;
std::cout<< flops/(t1-t0);
}
}
}
#undef CHECK_SDW
void benchsDw(std::vector<int> & latt4, int Ls, int threads, int report )
{
GridCartesian * UGrid = SpaceTimeGrid::makeFourDimGrid(latt4, GridDefaultSimd(Nd,vComplex::Nsimd()),GridDefaultMpi());
GridRedBlackCartesian * UrbGrid = SpaceTimeGrid::makeFourDimRedBlackGrid(UGrid);
GridCartesian * FGrid = SpaceTimeGrid::makeFiveDimGrid(Ls,UGrid);
GridRedBlackCartesian * FrbGrid = SpaceTimeGrid::makeFiveDimRedBlackGrid(Ls,UGrid);
GridCartesian * sUGrid = SpaceTimeGrid::makeFourDimDWFGrid(latt4,GridDefaultMpi());
GridCartesian * sFGrid = SpaceTimeGrid::makeFiveDimDWFGrid(Ls,UGrid);
GridRedBlackCartesian * sFrbGrid = SpaceTimeGrid::makeFiveDimDWFRedBlackGrid(Ls,UGrid);
std::vector<int> seeds4({1,2,3,4});
std::vector<int> seeds5({5,6,7,8});
#ifdef CHECK_SDW
GridParallelRNG RNG4(UGrid); RNG4.SeedFixedIntegers(seeds4);
GridParallelRNG RNG5(FGrid); RNG5.SeedFixedIntegers(seeds5);
LatticeFermion src (FGrid); random(RNG5,src);
LatticeGaugeField Umu(UGrid);
random(RNG4,Umu);
#else
LatticeFermion src (FGrid); src=zero;
LatticeGaugeField Umu(UGrid); Umu=zero;
#endif
LatticeFermion result(FGrid); result=zero;
LatticeFermion ref(FGrid); ref=zero;
LatticeFermion tmp(FGrid);
LatticeFermion err(FGrid);
ColourMatrix cm = Complex(1.0,0.0);
LatticeGaugeField Umu5d(FGrid);
// replicate across fifth dimension
for(int ss=0;ss<Umu._grid->oSites();ss++){
for(int s=0;s<Ls;s++){
Umu5d._odata[Ls*ss+s] = Umu._odata[ss];
}
}
RealD mass=0.1;
RealD M5 =1.8;
typedef WilsonFermion5D<DomainWallRedBlack5dImplF> WilsonFermion5DF;
LatticeFermionF ssrc(sFGrid);
LatticeFermionF sref(sFGrid);
LatticeFermionF sresult(sFGrid);
WilsonFermion5DF sDw(1,Umu,*sFGrid,*sFrbGrid,*sUGrid,M5);
for(int x=0;x<latt4[0];x++){
for(int y=0;y<latt4[1];y++){
for(int z=0;z<latt4[2];z++){
for(int t=0;t<latt4[3];t++){
for(int s=0;s<Ls;s++){
std::vector<int> site({s,x,y,z,t});
SpinColourVectorF tmp;
peekSite(tmp,src,site);
pokeSite(tmp,ssrc,site);
}}}}}
double t0=usecond();
sDw.Dhop(ssrc,sresult,0);
double t1=usecond();
int ncall =1+(int) ((5.0*1000*1000)/(t1-t0));
PerformanceCounter Counter(8);
Counter.Start();
t0=usecond();
for(int i=0;i<ncall;i++){
sDw.Dhop(ssrc,sresult,0);
}
t1=usecond();
Counter.Stop();
if ( report ) {
Counter.Report();
} else {
double volume=Ls; for(int mu=0;mu<Nd;mu++) volume=volume*latt4[mu];
double flops=1344*volume*ncall;
std::cout<<"\t"<< flops/(t1-t0);
}
LatticeFermionF sr_eo(sFGrid);
LatticeFermionF serr(sFGrid);
LatticeFermion ssrc_e (sFrbGrid);
LatticeFermion ssrc_o (sFrbGrid);
LatticeFermion sr_e (sFrbGrid);
LatticeFermion sr_o (sFrbGrid);
pickCheckerboard(Even,ssrc_e,ssrc);
pickCheckerboard(Odd,ssrc_o,ssrc);
setCheckerboard(sr_eo,ssrc_o);
setCheckerboard(sr_eo,ssrc_e);
sr_e = zero;
sr_o = zero;
sDw.DhopEO(ssrc_o,sr_e,DaggerNo);
PerformanceCounter CounterSdw(8);
CounterSdw.Start();
t0=usecond();
for(int i=0;i<ncall;i++){
sDw.DhopEO(ssrc_o,sr_e,DaggerNo);
}
t1=usecond();
CounterSdw.Stop();
if ( report ) {
CounterSdw.Report();
} else {
double volume=Ls; for(int mu=0;mu<Nd;mu++) volume=volume*latt4[mu];
double flops=(1344.0*volume*ncall)/2;
std::cout<<"\t"<< flops/(t1-t0);
}
}