396 lines
16 KiB
C
396 lines
16 KiB
C
#include <jni.h>
|
|
#include <stdio.h>
|
|
#include <immintrin.h>
|
|
#include <stdint.h>
|
|
|
|
#include "includes/utilities.h"
|
|
#include "includes/chunkmask.h"
|
|
|
|
|
|
|
|
#define LINEARSOLVERTIMES 10
|
|
|
|
|
|
|
|
void diffuse(JNIEnv * env, uint32_t chunk_mask, int N, int b, float * x, float * x0, float diff, float dt);
|
|
void advect(JNIEnv * env, uint32_t chunk_mask, int N, int b, jobjectArray jrd, float * d0, float * u, float * v, float * w, float dt);
|
|
void project(JNIEnv * env, uint32_t chunk_mask, int N, jobjectArray jru, jobjectArray jrv, jobjectArray jrw, float * p, float * div);
|
|
void set_bnd(JNIEnv * env, uint32_t chunk_mask, int N, int b, float * x);
|
|
void dens_step(JNIEnv * env, uint32_t chunk_mask, int N, jobjectArray jrx, float * x0, jobjectArray jru, jobjectArray jrv, jobjectArray jrw, float diff, float dt);
|
|
void vel_step(JNIEnv * env, uint32_t chunk_mask, int N, jobjectArray jru, jobjectArray jrv, jobjectArray jrw, float * u0, float * v0, float * w0, float visc, float dt);
|
|
void lin_solve(JNIEnv * env, uint32_t chunk_mask, int N, int b, float* x, float* x0, float a, float c);
|
|
|
|
/**
|
|
* The core simulation function
|
|
*/
|
|
JNIEXPORT void JNICALL Java_electrosphere_FluidSim_simulate(
|
|
JNIEnv * env,
|
|
jobject this,
|
|
jint DIM_X,
|
|
jint chunk_mask,
|
|
jobject jx,
|
|
jobject jx0,
|
|
jobject ju,
|
|
jobject jv,
|
|
jobject jw,
|
|
jobject ju0,
|
|
jobject jv0,
|
|
jobject jw0,
|
|
jfloat DIFFUSION_RATE,
|
|
jfloat VISCOSITY_RATE,
|
|
jfloat timestep){
|
|
jboolean isCopy;
|
|
// float * x = (*env)->GetDirectBufferAddress(env,jx);
|
|
float * x0 = (*env)->GetDirectBufferAddress(env,jx0);
|
|
// float * u = (*env)->GetDirectBufferAddress(env,ju);
|
|
// float * v = (*env)->GetDirectBufferAddress(env,jv);
|
|
// float * w = (*env)->GetDirectBufferAddress(env,jw);
|
|
float * u0 = (*env)->GetDirectBufferAddress(env,ju0);
|
|
float * v0 = (*env)->GetDirectBufferAddress(env,jv0);
|
|
float * w0 = (*env)->GetDirectBufferAddress(env,jw0);
|
|
int N = DIM_X;
|
|
int i,j,k;
|
|
vel_step(env, chunk_mask, DIM_X, ju, jv, jw, u0, v0, w0, VISCOSITY_RATE, timestep);
|
|
dens_step(env, chunk_mask, DIM_X, jx, x0, ju, jv, jw, DIFFUSION_RATE, timestep);
|
|
}
|
|
|
|
/**
|
|
* Adds values from a source array to a current frame array (eg more density to the main density array)
|
|
*/
|
|
void add_source(int N, float * x, float * s, float dt){
|
|
int i;
|
|
int size=N*N*N;
|
|
for(i=0; i<size; i++){
|
|
x[i] += dt*s[i];
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Diffuses a given array by a diffusion constant
|
|
*/
|
|
void diffuse(JNIEnv * env, uint32_t chunk_mask, int N, int b, float * x, float * x0, float diff, float dt){
|
|
float a=dt*diff*N*N*N;
|
|
lin_solve(env, chunk_mask, N, b, x, x0, a, 1+6*a);
|
|
}
|
|
|
|
/**
|
|
* Advects a given array based on the force vectors in the simulation
|
|
*/
|
|
void advect(JNIEnv * env, uint32_t chunk_mask, int N, int b, jobjectArray jrd, float * d0, float * u, float * v, float * w, float dt){
|
|
int i, j, k, i0, j0, k0, i1, j1, k1;
|
|
float x, y, z, s0, t0, s1, t1, u1, u0, dtx,dty,dtz;
|
|
|
|
dtx=dty=dtz=dt*N;
|
|
|
|
float * d = GET_ARR(env,jrd,CENTER_LOC);
|
|
|
|
for(k=1; k<N-1; k++){
|
|
for(j=1; j<N-1; j++){
|
|
for(i=1; i<N-1; i++){
|
|
x = i-dtx*u[IX(i,j,k)]; y = j-dty*v[IX(i,j,k)]; z = k-dtz*w[IX(i,j,k)];
|
|
if (x<0.5f) x=0.5f; if (x>N+0.5f) x=N+0.5f; i0=(int)x; i1=i0+1;
|
|
if (y<0.5f) y=0.5f; if (y>N+0.5f) y=N+0.5f; j0=(int)y; j1=j0+1;
|
|
if (z<0.5f) z=0.5f; if (z>N+0.5f) z=N+0.5f; k0=(int)z; k1=k0+1;
|
|
|
|
s1 = x-i0; s0 = 1-s1; t1 = y-j0; t0 = 1-t1; u1 = z-k0; u0 = 1-u1;
|
|
if(i0 >= N){
|
|
i0 = N - 1;
|
|
}
|
|
// if(i0 < 0){
|
|
// i0 = 0;
|
|
// }
|
|
if(j0 >= N){
|
|
j0 = N - 1;
|
|
}
|
|
// if(j0 < 0){
|
|
// j0 = 0;
|
|
// }
|
|
if(k0 >= N){
|
|
k0 = N - 1;
|
|
}
|
|
// if(k0 < 0){
|
|
// k0 = 0;
|
|
// }
|
|
if(i1 >= N){
|
|
i1 = N - 1;
|
|
}
|
|
// if(i1 < 0){
|
|
// i1 = 0;
|
|
// }
|
|
if(j1 >= N){
|
|
j1 = N - 1;
|
|
}
|
|
// if(j1 < 0){
|
|
// j1 = 0;
|
|
// }
|
|
if(k1 >= N){
|
|
k1 = N - 1;
|
|
}
|
|
// if(k1 < 0){
|
|
// k1 = 0;
|
|
// }
|
|
d[IX(i,j,k)] = s0*(t0*u0*d0[IX(i0,j0,k0)]+t1*u0*d0[IX(i0,j1,k0)]+t0*u1*d0[IX(i0,j0,k1)]+t1*u1*d0[IX(i0,j1,k1)])+
|
|
s1*(t0*u0*d0[IX(i1,j0,k0)]+t1*u0*d0[IX(i1,j1,k0)]+t0*u1*d0[IX(i1,j0,k1)]+t1*u1*d0[IX(i1,j1,k1)]);
|
|
}
|
|
}
|
|
}
|
|
set_bnd(env, chunk_mask, N, b, d);
|
|
}
|
|
|
|
/**
|
|
* The main density step function
|
|
*/
|
|
void dens_step(JNIEnv * env, uint32_t chunk_mask, int N, jobjectArray jrx, float * x0, jobjectArray jru, jobjectArray jrv, jobjectArray jrw, float diff, float dt){
|
|
float * x = GET_ARR(env,jrx,CENTER_LOC);
|
|
float * u = GET_ARR(env,jru,CENTER_LOC);
|
|
float * v = GET_ARR(env,jrv,CENTER_LOC);
|
|
float * w = GET_ARR(env,jrw,CENTER_LOC);
|
|
add_source(N, x, x0, dt);
|
|
SWAP(x0, x);
|
|
diffuse(env, chunk_mask, N, 0, x, x0, diff, dt);
|
|
SWAP(x0, x);
|
|
advect(env, chunk_mask, N, 0, jrx, x0, u, v, w, dt);
|
|
}
|
|
|
|
/**
|
|
* The main velocity step function
|
|
*/
|
|
void vel_step(JNIEnv * env, uint32_t chunk_mask, int N, jobjectArray jru, jobjectArray jrv, jobjectArray jrw, float * u0, float * v0, float * w0, float visc, float dt){
|
|
float * u = GET_ARR(env,jru,CENTER_LOC);
|
|
float * v = GET_ARR(env,jrv,CENTER_LOC);
|
|
float * w = GET_ARR(env,jrw,CENTER_LOC);
|
|
add_source(N, u, u0, dt);
|
|
add_source(N, v, v0, dt);
|
|
add_source(N, w, w0, dt);
|
|
SWAP(u0, u);
|
|
diffuse(env, chunk_mask, N, 1, u, u0, visc, dt);
|
|
SWAP(v0, v);
|
|
diffuse(env, chunk_mask, N, 2, v, v0, visc, dt);
|
|
SWAP(w0, w);
|
|
diffuse(env, chunk_mask, N, 3, w, w0, visc, dt);
|
|
project(env, chunk_mask, N, jru, jrv, jrw, u0, v0);
|
|
SWAP(u0, u);
|
|
SWAP(v0, v);
|
|
SWAP(w0, w);
|
|
advect(env, chunk_mask, N, 1, jru, u0, u0, v0, w0, dt);
|
|
advect(env, chunk_mask, N, 2, jrv, v0, u0, v0, w0, dt);
|
|
advect(env, chunk_mask, N, 3, jrw, w0, u0, v0, w0, dt);
|
|
project(env, chunk_mask, N, jru, jrv, jrw, u0, v0);
|
|
}
|
|
|
|
//used for temporary vector storage when appropriate
|
|
float container[16];
|
|
|
|
/**
|
|
* Projects a given array based on force vectors
|
|
*/
|
|
void project(JNIEnv * env, uint32_t chunk_mask, int N, jobjectArray jru, jobjectArray jrv, jobjectArray jrw, float * p, float * div){
|
|
int i, j, k;
|
|
|
|
__m256 nVector = _mm256_set1_ps(N);
|
|
__m256 constScalar = _mm256_set1_ps(-1.0/3.0);
|
|
__m256 zeroVec = _mm256_set1_ps(0);
|
|
__m256 vector, vector2, vector3;
|
|
|
|
float * u = GET_ARR(env,jru,CENTER_LOC);
|
|
float * v = GET_ARR(env,jrv,CENTER_LOC);
|
|
float * w = GET_ARR(env,jrw,CENTER_LOC);
|
|
|
|
for(k=1; k<N-1; k++){
|
|
for(j=1; j<N-1; j++){
|
|
i = 1;
|
|
//
|
|
//lower
|
|
//
|
|
//first part
|
|
vector = _mm256_loadu_ps(&u[IX(i+1,j,k)]);
|
|
vector = _mm256_sub_ps(vector,_mm256_loadu_ps(&u[IX(i-1,j,k)]));
|
|
vector = _mm256_div_ps(vector,nVector);
|
|
//second part
|
|
vector2 = _mm256_loadu_ps(&v[IX(i,j+1,k)]);
|
|
vector2 = _mm256_sub_ps(vector2,_mm256_loadu_ps(&v[IX(i,j-1,k)]));
|
|
vector2 = _mm256_div_ps(vector2,nVector);
|
|
//third part
|
|
vector3 = _mm256_loadu_ps(&w[IX(i,j,k+1)]);
|
|
vector3 = _mm256_sub_ps(vector3,_mm256_loadu_ps(&w[IX(i,j,k-1)]));
|
|
vector3 = _mm256_div_ps(vector3,nVector);
|
|
//multiply and finalize
|
|
vector = _mm256_add_ps(vector,_mm256_add_ps(vector2,vector3));
|
|
vector = _mm256_mul_ps(vector,constScalar);
|
|
//store
|
|
_mm256_storeu_ps(&div[IX(i,j,k)],vector);
|
|
_mm256_storeu_ps(&p[IX(i,j,k)],zeroVec);
|
|
i = 9;
|
|
//
|
|
//upper
|
|
//
|
|
//first part
|
|
vector = _mm256_loadu_ps(&u[IX(i+1,j,k)]);
|
|
vector = _mm256_sub_ps(vector,_mm256_loadu_ps(&u[IX(i-1,j,k)]));
|
|
vector = _mm256_div_ps(vector,nVector);
|
|
//second part
|
|
vector2 = _mm256_loadu_ps(&v[IX(i,j+1,k)]);
|
|
vector2 = _mm256_sub_ps(vector2,_mm256_loadu_ps(&v[IX(i,j-1,k)]));
|
|
vector2 = _mm256_div_ps(vector2,nVector);
|
|
//third part
|
|
vector3 = _mm256_loadu_ps(&w[IX(i,j,k+1)]);
|
|
vector3 = _mm256_sub_ps(vector3,_mm256_loadu_ps(&w[IX(i,j,k-1)]));
|
|
vector3 = _mm256_div_ps(vector3,nVector);
|
|
//multiply and finalize
|
|
vector = _mm256_add_ps(vector,_mm256_add_ps(vector2,vector3));
|
|
vector = _mm256_mul_ps(vector,constScalar);
|
|
//store
|
|
_mm256_storeu_ps(&div[IX(i,j,k)],vector);
|
|
_mm256_storeu_ps(&p[IX(i,j,k)],zeroVec);
|
|
}
|
|
}
|
|
|
|
set_bnd(env, chunk_mask, N, 0, div);
|
|
set_bnd(env, chunk_mask, N, 0, p);
|
|
|
|
lin_solve(env, chunk_mask, N, 0, p, div, 1, 6);
|
|
|
|
|
|
constScalar = _mm256_set1_ps(0.5f*N);
|
|
for ( k=1 ; k<N-1 ; k++ ) {
|
|
for ( j=1 ; j<N-1 ; j++ ) {
|
|
//
|
|
//v
|
|
//
|
|
//lower
|
|
vector = _mm256_loadu_ps(&p[IX(1+1,j,k)]);
|
|
vector2 = _mm256_loadu_ps(&p[IX(1-1,j,k)]);
|
|
vector = _mm256_sub_ps(vector,vector2);
|
|
vector = _mm256_mul_ps(vector,constScalar);
|
|
vector = _mm256_sub_ps(_mm256_loadu_ps(&u[IX(1,j,k)]),vector);
|
|
_mm256_storeu_ps(&u[IX(1,j,k)],vector);
|
|
//upper
|
|
vector = _mm256_loadu_ps(&p[IX(9+1,j,k)]);
|
|
vector2 = _mm256_loadu_ps(&p[IX(9-1,j,k)]);
|
|
vector = _mm256_sub_ps(vector,vector2);
|
|
vector = _mm256_mul_ps(vector,constScalar);
|
|
vector = _mm256_sub_ps(_mm256_loadu_ps(&u[IX(9,j,k)]),vector);
|
|
_mm256_storeu_ps(&u[IX(9,j,k)],vector);
|
|
//
|
|
//v
|
|
//
|
|
//lower
|
|
vector = _mm256_loadu_ps(&p[IX(1,j+1,k)]);
|
|
vector2 = _mm256_loadu_ps(&p[IX(1,j-1,k)]);
|
|
vector = _mm256_sub_ps(vector,vector2);
|
|
vector = _mm256_mul_ps(vector,constScalar);
|
|
vector = _mm256_sub_ps(_mm256_loadu_ps(&v[IX(1,j,k)]),vector);
|
|
_mm256_storeu_ps(&v[IX(1,j,k)],vector);
|
|
//upper
|
|
vector = _mm256_loadu_ps(&p[IX(9,j+1,k)]);
|
|
vector2 = _mm256_loadu_ps(&p[IX(9,j-1,k)]);
|
|
vector = _mm256_sub_ps(vector,vector2);
|
|
vector = _mm256_mul_ps(vector,constScalar);
|
|
vector = _mm256_sub_ps(_mm256_loadu_ps(&v[IX(9,j,k)]),vector);
|
|
_mm256_storeu_ps(&v[IX(9,j,k)],vector);
|
|
//
|
|
//w
|
|
//
|
|
//lower
|
|
vector = _mm256_loadu_ps(&p[IX(1,j,k+1)]);
|
|
vector2 = _mm256_loadu_ps(&p[IX(1,j,k-1)]);
|
|
vector = _mm256_sub_ps(vector,vector2);
|
|
vector = _mm256_mul_ps(vector,constScalar);
|
|
vector = _mm256_sub_ps(_mm256_loadu_ps(&w[IX(1,j,k)]),vector);
|
|
_mm256_storeu_ps(&w[IX(1,j,k)],vector);
|
|
//upper
|
|
vector = _mm256_loadu_ps(&p[IX(9,j,k+1)]);
|
|
vector2 = _mm256_loadu_ps(&p[IX(9,j,k-1)]);
|
|
vector = _mm256_sub_ps(vector,vector2);
|
|
vector = _mm256_mul_ps(vector,constScalar);
|
|
vector = _mm256_sub_ps(_mm256_loadu_ps(&w[IX(9,j,k)]),vector);
|
|
_mm256_storeu_ps(&w[IX(9,j,k)],vector);
|
|
}
|
|
}
|
|
|
|
set_bnd(env, chunk_mask, N, 1, u);
|
|
set_bnd(env, chunk_mask, N, 2, v);
|
|
set_bnd(env, chunk_mask, N, 3, w);
|
|
}
|
|
|
|
/**
|
|
* Solves a linear system of equations in a vectorized manner
|
|
*/
|
|
void lin_solve(JNIEnv * env, uint32_t chunk_mask, int N, int b, float* x, float* x0, float a, float c){
|
|
int i, j, k, l, m;
|
|
__m256 aScalar = _mm256_set1_ps(a);
|
|
__m256 cScalar = _mm256_set1_ps(c);
|
|
// iterate the solver
|
|
for ( l=0 ; l<LINEARSOLVERTIMES ; l++ ) {
|
|
// update for each cell
|
|
for(k=1; k<N-1; k++){
|
|
for(j=1; j<N-1; j++){
|
|
int n = 0;
|
|
//solve as much as possible vectorized
|
|
for(i = 1; i < N-1; i=i+8){
|
|
__m256 vector = _mm256_loadu_ps(&x[IX(i-1,j,k)]);
|
|
vector = _mm256_add_ps(vector,_mm256_loadu_ps(&x[IX(i+1,j,k)]));
|
|
vector = _mm256_add_ps(vector,_mm256_loadu_ps(&x[IX(i,j-1,k)]));
|
|
vector = _mm256_add_ps(vector,_mm256_loadu_ps(&x[IX(i,j+1,k)]));
|
|
vector = _mm256_add_ps(vector,_mm256_loadu_ps(&x[IX(i,j,k-1)]));
|
|
vector = _mm256_add_ps(vector,_mm256_loadu_ps(&x[IX(i,j,k+1)]));
|
|
vector = _mm256_mul_ps(vector,aScalar);
|
|
vector = _mm256_add_ps(vector,_mm256_loadu_ps(&x0[IX(i,j,k)]));
|
|
vector = _mm256_div_ps(vector,cScalar);
|
|
_mm256_storeu_ps(&x[IX(i,j,k)],vector);
|
|
}
|
|
//If there is any leftover, perform manual solving
|
|
if(i>N-1){
|
|
for(i=i-8; i < N-1; i++){
|
|
x[IX(i,j,k)] = (x0[IX(i,j,k)] + a*(x[IX(i-1,j,k)]+x[IX(i+1,j,k)]+x[IX(i,j-1,k)]+x[IX(i,j+1,k)]+x[IX(i,j,k-1)]+x[IX(i,j,k+1)]))/c;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
set_bnd(env, chunk_mask, N, b, x);
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Sets the bounds of the simulation
|
|
*/
|
|
void set_bnd(JNIEnv * env, uint32_t chunk_mask, int N, int b, float * target){
|
|
int DIM = N;
|
|
for(int x=1; x < DIM-1; x++){
|
|
for(int y = 1; y < DIM-1; y++){
|
|
//((x)+(DIM)*(y) + (DIM)*(DIM)*(z))
|
|
target[0 + DIM * x + DIM * DIM * y] = b==1 ? -target[1 + DIM * x + DIM * DIM * y] : target[1 + DIM * x + DIM * DIM * y];
|
|
target[IX(DIM-1,x,y)] = b==1 ? -target[IX(DIM-2,x,y)] : target[IX(DIM-2,x,y)];
|
|
target[IX(x,0,y)] = b==2 ? -target[IX(x,1,y)] : target[IX(x,1,y)];
|
|
target[IX(x,DIM-1,y)] = b==2 ? -target[IX(x,DIM-2,y)] : target[IX(x,DIM-2,y)];
|
|
target[IX(x,y,0)] = b==3 ? -target[IX(x,y,1)] : target[IX(x,y,1)];
|
|
target[IX(x,y,DIM-1)] = b==3 ? -target[IX(x,y,DIM-2)] : target[IX(x,y,DIM-2)];
|
|
}
|
|
}
|
|
for(int x = 1; x < DIM-1; x++){
|
|
target[IX(x,0,0)] = (float)(0.5f * (target[IX(x,1,0)] + target[IX(x,0,1)]));
|
|
target[IX(x,DIM-1,0)] = (float)(0.5f * (target[IX(x,DIM-2,0)] + target[IX(x,DIM-1,1)]));
|
|
target[IX(x,0,DIM-1)] = (float)(0.5f * (target[IX(x,1,DIM-1)] + target[IX(x,0,DIM-2)]));
|
|
target[IX(x,DIM-1,DIM-1)] = (float)(0.5f * (target[IX(x,DIM-2,DIM-1)] + target[IX(x,DIM-1,DIM-2)]));
|
|
|
|
target[IX(0,x,0)] = (float)(0.5f * (target[IX(1,x,0)] + target[IX(0,x,1)]));
|
|
target[IX(DIM-1,x,0)] = (float)(0.5f * (target[IX(DIM-2,x,0)] + target[IX(DIM-1,x,1)]));
|
|
target[IX(0,x,DIM-1)] = (float)(0.5f * (target[IX(1,x,DIM-1)] + target[IX(0,x,DIM-2)]));
|
|
target[IX(DIM-1,x,DIM-1)] = (float)(0.5f * (target[IX(DIM-2,x,DIM-1)] + target[IX(DIM-1,x,DIM-2)]));
|
|
|
|
|
|
target[IX(0,0,x)] = (float)(0.5f * (target[IX(1,0,x)] + target[IX(0,1,x)]));
|
|
target[IX(DIM-1,0,x)] = (float)(0.5f * (target[IX(DIM-2,0,x)] + target[IX(DIM-1,1,x)]));
|
|
target[IX(0,DIM-1,x)] = (float)(0.5f * (target[IX(1,DIM-1,x)] + target[IX(0,DIM-2,x)]));
|
|
target[IX(DIM-1,DIM-1,x)] = (float)(0.5f * (target[IX(DIM-2,DIM-1,x)] + target[IX(DIM-1,DIM-2,x)]));
|
|
|
|
}
|
|
target[IX(0,0,0)] = (float)((target[IX(1,0,0)]+target[IX(0,1,0)]+target[IX(0,0,1)])/3.0);
|
|
target[IX(DIM-1,0,0)] = (float)((target[IX(DIM-2,0,0)]+target[IX(DIM-1,1,0)]+target[IX(DIM-1,0,1)])/3.0);
|
|
target[IX(0,DIM-1,0)] = (float)((target[IX(1,DIM-1,0)]+target[IX(0,DIM-2,0)]+target[IX(0,DIM-1,1)])/3.0);
|
|
target[IX(0,0,DIM-1)] = (float)((target[IX(0,0,DIM-2)]+target[IX(1,0,DIM-1)]+target[IX(0,1,DIM-1)])/3.0);
|
|
target[IX(DIM-1,DIM-1,0)] = (float)((target[IX(DIM-2,DIM-1,0)]+target[IX(DIM-1,DIM-2,0)]+target[IX(DIM-1,DIM-1,1)])/3.0);
|
|
target[IX(0,DIM-1,DIM-1)] = (float)((target[IX(1,DIM-1,DIM-1)]+target[IX(0,DIM-2,DIM-1)]+target[IX(0,DIM-1,DIM-2)])/3.0);
|
|
target[IX(DIM-1,0,DIM-1)] = (float)((target[IX(DIM-1,0,DIM-2)]+target[IX(DIM-2,0,DIM-1)]+target[IX(DIM-1,1,DIM-1)])/3.0);
|
|
target[IX(DIM-1,DIM-1,DIM-1)] = (float)((target[IX(DIM-1,DIM-1,DIM-2)]+target[IX(DIM-1,DIM-2,DIM-1)]+target[IX(DIM-1,DIM-1,DIM-2)])/3.0);
|
|
} |