Skip to content

Commit

Permalink
Created namespace userFunctions and removed src/Tools/userFunctions.c…
Browse files Browse the repository at this point in the history 
…pp ; created namespace random_namespace and cleaned src/ParticleBC/BoundaryConditionType.h ; the objective here is to make available functions without having to access a class object
  • Loading branch information
@charlesprouveur
charlesprouveur committed Nov 13, 2024
1 parent 98dccbd commit 8581e3a
Show file tree
Hide file tree
Showing 5 changed files with 276 additions and 464 deletions.
16 changes: 8 additions & 8 deletions src/ParticleBC/BoundaryConditionType.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -560,7 +560,7 @@ void thermalize_particle_inf( Species *species, int imin, int imax, int directio
#endif
// boucle sur les particules de ce chunk pour remplir xorshift32_state_array[...] avec le state local
for( int i = 0; i < 32; ++i ){
xorshift32_state_array[i] = xorshift32(xorshift32_state_local);
xorshift32_state_array[i] = Random_namespace::xorshift32(xorshift32_state_local);
}
#endif
// boucle sur les particules de ce chunk qui utilise xorshift32_state_array[i]
Expand Down Expand Up @@ -591,21 +591,21 @@ void thermalize_particle_inf( Species *species, int imin, int imax, int directio
// change of velocity in the direction normal to the reflection plane
double sign_vel = -momentum[ ipart ]/std::abs( momentum[ ipart ] );
#if defined( SMILEI_ACCELERATOR_GPU )
momentum[ ipart ] = sign_vel * thermal_momentum * std::sqrt( -std::log( 1.0 - uniform1(xorshift32_state_array[i]) ) );
momentum[ ipart ] = sign_vel * thermal_momentum * std::sqrt( -std::log( 1.0 - Random_namespace::uniform1(xorshift32_state_array[i]) ) );
#else
momentum[ ipart ] = sign_vel * thermal_momentum * std::sqrt( -std::log( 1.0 - rand->uniform1() ) );
#endif

// change of momentum in the direction(s) along the reflection plane
if (nDim>1) {
#if defined( SMILEI_ACCELERATOR_GPU )
momentumRefl_2D[ ipart ] = thermal_momentum1 * perp_rand_gpu_v3(xorshift32_state_array[i]);
momentumRefl_2D[ ipart ] = thermal_momentum1 * Random_namespace::perp_rand_dp(xorshift32_state_array[i]);
#else
momentumRefl_2D[ ipart ] = thermal_momentum1 * perp_rand( rand );
#endif
if (nDim>2) {
#if defined( SMILEI_ACCELERATOR_GPU )
momentumRefl_3D[ ipart ] = thermal_momentum2 * perp_rand_gpu_v3(xorshift32_state_array[i]);
momentumRefl_3D[ ipart ] = thermal_momentum2 * Random_namespace::perp_rand_dp(xorshift32_state_array[i]);
#else
momentumRefl_3D[ ipart ] = thermal_momentum2 * perp_rand( rand );
#endif
Expand Down Expand Up @@ -724,7 +724,7 @@ void thermalize_particle_sup( Species *species, int imin, int imax, int directio
#endif
// boucle sur les particules de ce chunk pour remplir xorshift32_state_array[...] avec le state local
for( int i = 0; i < 32; ++i ){
xorshift32_state_array[i] = xorshift32(xorshift32_state_local);
xorshift32_state_array[i] = Random_namespace::xorshift32(xorshift32_state_local);
}
#endif
int istart = ichunk==(imin/32) ? imin%32 : 0;
Expand Down Expand Up @@ -756,21 +756,21 @@ void thermalize_particle_sup( Species *species, int imin, int imax, int directio
// change of velocity in the direction normal to the reflection plane
double sign_vel = -momentum[ ipart ]/std::abs( momentum[ ipart ] );
#if defined( SMILEI_ACCELERATOR_GPU )
momentum[ ipart ] = sign_vel * thermal_momentum * std::sqrt( -std::log( 1.0 - uniform1(xorshift32_state_array[i]) ) );
momentum[ ipart ] = sign_vel * thermal_momentum * std::sqrt( -std::log( 1.0 - Random_namespace::uniform1(xorshift32_state_array[i]) ) );
#else
momentum[ ipart ] = sign_vel * thermal_momentum * std::sqrt( -std::log( 1.0 - rand->uniform1() ) );
#endif

// change of momentum in the direction(s) along the reflection plane
if (nDim>1) {
#if defined( SMILEI_ACCELERATOR_GPU )
momentumRefl_2D[ ipart ] = thermal_momentum1 * perp_rand_gpu_v3(xorshift32_state_array[i]);
momentumRefl_2D[ ipart ] = thermal_momentum1 * Random_namespace::perp_rand_dp(xorshift32_state_array[i]);
#else
momentumRefl_2D[ ipart ] = thermal_momentum1 * perp_rand( rand );
#endif
if (nDim>2) {
#if defined( SMILEI_ACCELERATOR_GPU )
momentumRefl_3D[ ipart ] = thermal_momentum2 * perp_rand_gpu_v3(xorshift32_state_array[i]);
momentumRefl_3D[ ipart ] = thermal_momentum2 * Random_namespace::perp_rand_dp(xorshift32_state_array[i]);
#else
momentumRefl_3D[ ipart ] = thermal_momentum2 * perp_rand( rand );
#endif
Expand Down
119 changes: 1 addition & 118 deletions src/ParticleBC/BoundaryConditionType.h
View file
Original file line number Diff line number Diff line change
Expand Up @@ -17,130 +17,13 @@
#include "Random.h"

inline double perp_rand( Random * rand ) {
double a = userFunctions::erfinv( rand->uniform1() ); // to be switched to erfinv 3
double a = userFunctions::erfinv_dp( rand->uniform1() );
if( rand->cointoss() ) {
a *= -1.;
}
return a;
}

/**
* copied from erfinv_DP_1.cu by Prof. Mike Giles.
* https://people.maths.ox.ac.uk/gilesm/
* https://people.maths.ox.ac.uk/gilesm/codes/erfinv/
*
* Original code is written for CUDA.
* Mutsuo Saito modified original code for C++.
*/
inline double erfinv_v3(double x)
{
double w, p;
double sign;
if (x > 0) {
sign = 1.0;
} else {
sign = -1.0;
x = abs(x);
}
w = - log((1.0-x)*(1.0+x));

if ( w < 6.250000 ) {
w = w - 3.125000;
p = -3.6444120640178196996e-21;
p = -1.685059138182016589e-19 + p*w;
p = 1.2858480715256400167e-18 + p*w;
p = 1.115787767802518096e-17 + p*w;
p = -1.333171662854620906e-16 + p*w;
p = 2.0972767875968561637e-17 + p*w;
p = 6.6376381343583238325e-15 + p*w;
p = -4.0545662729752068639e-14 + p*w;
p = -8.1519341976054721522e-14 + p*w;
p = 2.6335093153082322977e-12 + p*w;
p = -1.2975133253453532498e-11 + p*w;
p = -5.4154120542946279317e-11 + p*w;
p = 1.051212273321532285e-09 + p*w;
p = -4.1126339803469836976e-09 + p*w;
p = -2.9070369957882005086e-08 + p*w;
p = 4.2347877827932403518e-07 + p*w;
p = -1.3654692000834678645e-06 + p*w;
p = -1.3882523362786468719e-05 + p*w;
p = 0.0001867342080340571352 + p*w;
p = -0.00074070253416626697512 + p*w;
p = -0.0060336708714301490533 + p*w;
p = 0.24015818242558961693 + p*w;
p = 1.6536545626831027356 + p*w;
}
else if ( w < 16.000000 ) {
w = sqrt(w) - 3.250000;
p = 2.2137376921775787049e-09;
p = 9.0756561938885390979e-08 + p*w;
p = -2.7517406297064545428e-07 + p*w;
p = 1.8239629214389227755e-08 + p*w;
p = 1.5027403968909827627e-06 + p*w;
p = -4.013867526981545969e-06 + p*w;
p = 2.9234449089955446044e-06 + p*w;
p = 1.2475304481671778723e-05 + p*w;
p = -4.7318229009055733981e-05 + p*w;
p = 6.8284851459573175448e-05 + p*w;
p = 2.4031110387097893999e-05 + p*w;
p = -0.0003550375203628474796 + p*w;
p = 0.00095328937973738049703 + p*w;
p = -0.0016882755560235047313 + p*w;
p = 0.0024914420961078508066 + p*w;
p = -0.0037512085075692412107 + p*w;
p = 0.005370914553590063617 + p*w;
p = 1.0052589676941592334 + p*w;
p = 3.0838856104922207635 + p*w;
}
else {
w = sqrt(w) - 5.000000;
p = -2.7109920616438573243e-11;
p = -2.5556418169965252055e-10 + p*w;
p = 1.5076572693500548083e-09 + p*w;
p = -3.7894654401267369937e-09 + p*w;
p = 7.6157012080783393804e-09 + p*w;
p = -1.4960026627149240478e-08 + p*w;
p = 2.9147953450901080826e-08 + p*w;
p = -6.7711997758452339498e-08 + p*w;
p = 2.2900482228026654717e-07 + p*w;
p = -9.9298272942317002539e-07 + p*w;
p = 4.5260625972231537039e-06 + p*w;
p = -1.9681778105531670567e-05 + p*w;
p = 7.5995277030017761139e-05 + p*w;
p = -0.00021503011930044477347 + p*w;
p = -0.00013871931833623122026 + p*w;
p = 1.0103004648645343977 + p*w;
p = 4.8499064014085844221 + p*w;
}
return sign * p * x;
}

inline uint32_t xorshift32(uint32_t xorshift32_state)
{
// Algorithm "xor" from p. 4 of Marsaglia, "Xorshift RNGs"
xorshift32_state ^= xorshift32_state << 13;
xorshift32_state ^= xorshift32_state >> 17;
xorshift32_state ^= xorshift32_state << 5;
return xorshift32_state;
}

static constexpr double xorshift32_invmax1 = (1.-1e-11)/4294967296.;

inline double uniform1(uint32_t xorshift32_state) {
return xorshift32(xorshift32_state) * xorshift32_invmax1;
}

inline double perp_rand_gpu_v3(uint32_t xorshift32_state) {
double a = erfinv_v3( uniform1(xorshift32_state) ); //userFunctions::
// technically we could also use the erfinv() function fron cuda, it would require compiling with -cuda though ...
// the study showed the gap in perf for BC thermal was not worth the added depend
if( xorshift32(xorshift32_state) & 1 ) { // rand->cointoss()
a *= -1.;
}
return a;
}


void internal_inf( Species *species, int imin, int imax, int direction, double limit_inf, double dt, std::vector<double> &invgf, Random * rand, double &energy_change );

void internal_sup( Species *species, int imin, int imax, int direction, double limit_sup, double dt, std::vector<double> &invgf, Random * rand, double &energy_change );
Expand Down
29 changes: 29 additions & 0 deletions src/Tools/Random.h
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,35 @@
#include <cstdlib>
#include <inttypes.h>
#include <cmath>
#include "userFunctions.h"

namespace Random_namespace // in order to use the random functions without having access to the class random
{

inline uint32_t xorshift32(uint32_t xorshift32_state)
{
// Algorithm "xor" from p. 4 of Marsaglia, "Xorshift RNGs"
xorshift32_state ^= xorshift32_state << 13;
xorshift32_state ^= xorshift32_state >> 17;
xorshift32_state ^= xorshift32_state << 5;
return xorshift32_state;
}

inline double uniform1(uint32_t xorshift32_state) {
constexpr double xorshift32_invmax1 = (1.-1e-11)/4294967296.;
return xorshift32(xorshift32_state) * xorshift32_invmax1;
}

inline double perp_rand_dp(uint32_t xorshift32_state) {
double a = userFunctions::erfinv_dp( uniform1(xorshift32_state) );
// technically we could also use the erfinv() function fron cuda, it would require compiling with -cuda though ...
// the study showed the gap in perf for BC thermal was not worth the added depend
if( xorshift32(xorshift32_state) & 1 ) { // rand->cointoss()
a *= -1.;
}
return a;
}
}

class Random
{
Expand Down
Loading

0 comments on commit 8581e3a

Please sign in to comment.