Files
ispc/examples_cuda/aobench/ao1.ispc
2013-11-14 16:29:40 +01:00

314 lines
8.1 KiB
C++

// -*- mode: c++ -*-
/*
Copyright (c) 2010-2011, Intel Corporation
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
* Neither the name of Intel Corporation nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS
IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER
OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
/*
Based on Syoyo Fujita's aobench: http://code.google.com/p/aobench
*/
#define NAO_SAMPLES 8
#define M_PI 3.1415926535f
typedef float<3> vec;
#ifdef __NVPTX__
#warning "emitting DEVICE code"
#define programCount warpSize()
#define programIndex laneIndex()
#define taskIndex0 blockIndex0()
#define taskCount0 blockCount0()
#define taskIndex1 blockIndex1()
#define taskCount1 blockCount1()
#else
#warning "emitting HOST code"
#endif
struct Isect {
float t;
vec p;
vec n;
int hit;
};
struct Sphere {
vec center;
float radius;
};
struct Plane {
vec p;
vec n;
};
struct Ray {
vec org;
vec dir;
};
static inline float dot(vec a, vec b) {
return a.x * b.x + a.y * b.y + a.z * b.z;
}
static inline vec vcross(vec v0, vec v1) {
vec ret;
ret.x = v0.y * v1.z - v0.z * v1.y;
ret.y = v0.z * v1.x - v0.x * v1.z;
ret.z = v0.x * v1.y - v0.y * v1.x;
return ret;
}
static inline void vnormalize(vec &v) {
float len2 = dot(v, v);
float invlen = rsqrt(len2);
v *= invlen;
}
static inline void
ray_plane_intersect(Isect &isect, Ray &ray, uniform Plane &plane) {
float d = -dot(plane.p, plane.n);
float v = dot(ray.dir, plane.n);
if (abs(v) < 1.0e-17)
return;
else {
float t = -(dot(ray.org, plane.n) + d) / v;
if ((t > 0.0) && (t < isect.t)) {
isect.t = t;
isect.hit = 1;
isect.p = ray.org + ray.dir * t;
isect.n = plane.n;
}
}
}
static inline void
ray_sphere_intersect(Isect &isect, Ray &ray, uniform Sphere &sphere) {
vec rs = ray.org - sphere.center;
float B = dot(rs, ray.dir);
float C = dot(rs, rs) - sphere.radius * sphere.radius;
float D = B * B - C;
if (D > 0.) {
float t = -B - sqrt(D);
if ((t > 0.0) && (t < isect.t)) {
isect.t = t;
isect.hit = 1;
isect.p = ray.org + t * ray.dir;
isect.n = isect.p - sphere.center;
vnormalize(isect.n);
}
}
}
static inline void
orthoBasis(vec basis[3], vec n) {
basis[2] = n;
basis[1].x = 0.0; basis[1].y = 0.0; basis[1].z = 0.0;
if ((n.x < 0.6) && (n.x > -0.6)) {
basis[1].x = 1.0;
} else if ((n.y < 0.6) && (n.y > -0.6)) {
basis[1].y = 1.0;
} else if ((n.z < 0.6) && (n.z > -0.6)) {
basis[1].z = 1.0;
} else {
basis[1].x = 1.0;
}
basis[0] = vcross(basis[1], basis[2]);
vnormalize(basis[0]);
basis[1] = vcross(basis[2], basis[0]);
vnormalize(basis[1]);
}
static inline float
ambient_occlusion(Isect &isect, uniform Plane &plane, uniform Sphere spheres[3],
RNGState &rngstate) {
float eps = 0.0001f;
vec p, n;
vec basis[3];
float occlusion = 0.0;
p = isect.p + eps * isect.n;
orthoBasis(basis, isect.n);
const uniform int ntheta = NAO_SAMPLES;
const uniform int nphi = NAO_SAMPLES;
for (uniform int j = 0; j < ntheta; j++) {
for (uniform int i = 0; i < nphi; i++) {
Ray ray;
Isect occIsect;
float theta = sqrt(frandom(&rngstate));
float phi = 2.0f * M_PI * frandom(&rngstate);
float x = cos(phi) * theta;
float y = sin(phi) * theta;
float z = sqrt(1.0 - theta * theta);
// local . global
float rx = x * basis[0].x + y * basis[1].x + z * basis[2].x;
float ry = x * basis[0].y + y * basis[1].y + z * basis[2].y;
float rz = x * basis[0].z + y * basis[1].z + z * basis[2].z;
ray.org = p;
ray.dir.x = rx;
ray.dir.y = ry;
ray.dir.z = rz;
occIsect.t = 1.0e+17;
occIsect.hit = 0;
for (uniform int snum = 0; snum < 3; ++snum)
ray_sphere_intersect(occIsect, ray, spheres[snum]);
ray_plane_intersect (occIsect, ray, plane);
if (occIsect.hit) occlusion += 1.0;
}
}
occlusion = (ntheta * nphi - occlusion) / (float)(ntheta * nphi);
return occlusion;
}
/* Compute the image for the scanlines from [y0,y1), for an overall image
of width w and height h.
*/
static inline void ao_tile(
uniform int x0, uniform int x1,
uniform int y0, uniform int y1,
uniform int w, uniform int h,
uniform int nsubsamples,
uniform float image[])
{
uniform Plane plane = { { 0.0f, -0.5f, 0.0f }, { 0.f, 1.f, 0.f } };
uniform Sphere spheres[3] = {
{ { -2.0f, 0.0f, -3.5f }, 0.5f },
{ { -0.5f, 0.0f, -3.0f }, 0.5f },
{ { 1.0f, 0.0f, -2.2f }, 0.5f } };
RNGState rngstate;
seed_rng(&rngstate, programIndex + (y0 << (programIndex & 31)));
float invSamples = 1.f / nsubsamples;
foreach_tiled (y = y0 ... y1, x = x0 ... x1)
{
const int offset = 3 * (y * w + x);
float res = 0.0f;
for (uniform int u = 0; u < nsubsamples; u++)
for (uniform int v = 0; v < nsubsamples; v++)
{
float du = (float)u * invSamples, dv = (float)v * invSamples;
// Figure out x,y pixel in NDC
float px = (x + du - (w / 2.0f)) / (w / 2.0f);
float py = -(y + dv - (h / 2.0f)) / (h / 2.0f);
float ret = 0.f;
Ray ray;
Isect isect;
ray.org = 0.f;
// Poor man's perspective projection
ray.dir.x = px;
ray.dir.y = py;
ray.dir.z = -1.0;
vnormalize(ray.dir);
isect.t = 1.0e+17;
isect.hit = 0;
for (uniform int snum = 0; snum < 3; ++snum)
ray_sphere_intersect(isect, ray, spheres[snum]);
ray_plane_intersect(isect, ray, plane);
// Note use of 'coherent' if statement; the set of rays we
// trace will often all hit or all miss the scene
if (isect.hit) {
ret = ambient_occlusion(isect, plane, spheres, rngstate);
ret *= invSamples * invSamples;
res += ret;
}
}
//if (x < x1)
{
image[offset ] = res;
image[offset+1] = res;
image[offset+2] = res;
}
}
}
#define TILEX 64
#define TILEY 4
/* unless task/export is specified all functions
* are generated as mangled "__device__" functions
*/
/* task will generate mangled "__global__" function only */
void task ao_task(uniform int width, uniform int height,
uniform int nsubsamples, uniform float image[])
{
if (taskIndex0 >= taskCount0) return;
if (taskIndex1 >= taskCount1) return;
const uniform int x0 = taskIndex0 * TILEX;
const uniform int x1 = min(x0 + TILEX, width);
const uniform int y0 = taskIndex1 * TILEY;
const uniform int y1 = min(y0 + TILEY, height);
ao_tile(x0,x1,y0,y1, width, height, nsubsamples, image);
}
/* export will generate unmangled "extern "C" __global__" and mangled "__device__" */
export void ao_ispc_tasks(uniform int w, uniform int h, uniform int nsubsamples,
uniform float image[])
{
const uniform int ntilex = (w+TILEX-1)/TILEX;
const uniform int ntiley = (h+TILEY-1)/TILEY;
launch[ntilex,ntiley] ao_task(w, h, nsubsamples, image);
}