mirror of
https://github.com/NVIDIA/cuda-samples.git
synced 2026-10-25 22:28:17 +08:00
Apply consistent code formatting across the repo. Add clang-format and pre-commit hooks.
This commit is contained in:
Rob Armstrong
@@ -34,4 +34,3 @@ Download and install the [CUDA Toolkit 12.5](https://developer.nvidia.com/cuda-d
|
||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||
|
||||
## References (for more details)
|
||||
|
||||
|
||||
@@ -32,11 +32,11 @@
|
||||
using 3D texture lookups.
|
||||
*/
|
||||
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
#include <math.h>
|
||||
#include <helper_gl.h>
|
||||
#include <math.h>
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
#if defined(__APPLE__) || defined(MACOSX)
|
||||
#pragma clang diagnostic ignored "-Wdeprecated-declarations"
|
||||
@@ -49,53 +49,52 @@
|
||||
#endif
|
||||
|
||||
// includes, cuda
|
||||
#include <vector_types.h>
|
||||
#include <cuda_runtime.h>
|
||||
#include <cuda_gl_interop.h>
|
||||
#include <cuda_runtime.h>
|
||||
#include <vector_types.h>
|
||||
|
||||
// CUDA utilities and system includes
|
||||
#include <helper_cuda.h>
|
||||
#include <helper_functions.h>
|
||||
#include <vector_types.h>
|
||||
|
||||
typedef unsigned int uint;
|
||||
typedef unsigned int uint;
|
||||
typedef unsigned char uchar;
|
||||
|
||||
#define MAX_EPSILON_ERROR 5.0f
|
||||
#define THRESHOLD 0.15f
|
||||
#define THRESHOLD 0.15f
|
||||
|
||||
const char *sSDKsample = "simpleTexture3D";
|
||||
|
||||
const char *volumeFilename = "Bucky.raw";
|
||||
const cudaExtent volumeSize = make_cudaExtent(32, 32, 32);
|
||||
const char *volumeFilename = "Bucky.raw";
|
||||
const cudaExtent volumeSize = make_cudaExtent(32, 32, 32);
|
||||
|
||||
const uint width = 512, height = 512;
|
||||
const dim3 blockSize(16, 16, 1);
|
||||
const dim3 gridSize(width / blockSize.x, height / blockSize.y);
|
||||
|
||||
float w = 0.5; // texture coordinate in z
|
||||
float w = 0.5; // texture coordinate in z
|
||||
|
||||
GLuint pbo; // OpenGL pixel buffer object
|
||||
struct cudaGraphicsResource
|
||||
*cuda_pbo_resource; // CUDA Graphics Resource (to transfer PBO)
|
||||
GLuint pbo; // OpenGL pixel buffer object
|
||||
struct cudaGraphicsResource *cuda_pbo_resource; // CUDA Graphics Resource (to transfer PBO)
|
||||
|
||||
bool linearFiltering = true;
|
||||
bool animate = true;
|
||||
bool animate = true;
|
||||
|
||||
StopWatchInterface *timer = NULL;
|
||||
|
||||
uint *d_output = NULL;
|
||||
|
||||
// Auto-Verification Code
|
||||
const int frameCheckNumber = 4;
|
||||
int fpsCount = 0; // FPS count for averaging
|
||||
int fpsLimit = 1; // FPS limit for sampling
|
||||
int g_Index = 0;
|
||||
unsigned int frameCount = 0;
|
||||
unsigned int g_TotalErrors = 0;
|
||||
const int frameCheckNumber = 4;
|
||||
int fpsCount = 0; // FPS count for averaging
|
||||
int fpsLimit = 1; // FPS limit for sampling
|
||||
int g_Index = 0;
|
||||
unsigned int frameCount = 0;
|
||||
unsigned int g_TotalErrors = 0;
|
||||
volatile int g_GraphicsMapFlag = 0;
|
||||
|
||||
int *pArgc = NULL;
|
||||
int *pArgc = NULL;
|
||||
char **pArgv = NULL;
|
||||
|
||||
#ifndef MAX
|
||||
@@ -105,288 +104,294 @@ char **pArgv = NULL;
|
||||
extern "C" void cleanup();
|
||||
extern "C" void setTextureFilterMode(bool bLinearFilter);
|
||||
extern "C" void initCuda(const uchar *h_volume, cudaExtent volumeSize);
|
||||
extern "C" void render_kernel(dim3 gridSize, dim3 blockSize, uint *d_output,
|
||||
uint imageW, uint imageH, float w);
|
||||
extern void cleanupCuda();
|
||||
extern "C" void render_kernel(dim3 gridSize, dim3 blockSize, uint *d_output, uint imageW, uint imageH, float w);
|
||||
extern void cleanupCuda();
|
||||
|
||||
void loadVolumeData(char *exec_path);
|
||||
|
||||
void computeFPS() {
|
||||
frameCount++;
|
||||
fpsCount++;
|
||||
void computeFPS()
|
||||
{
|
||||
frameCount++;
|
||||
fpsCount++;
|
||||
|
||||
if (fpsCount == fpsLimit) {
|
||||
char fps[256];
|
||||
float ifps = 1.f / (sdkGetAverageTimerValue(&timer) / 1000.f);
|
||||
sprintf(fps, "%s: %3.1f fps", sSDKsample, ifps);
|
||||
if (fpsCount == fpsLimit) {
|
||||
char fps[256];
|
||||
float ifps = 1.f / (sdkGetAverageTimerValue(&timer) / 1000.f);
|
||||
sprintf(fps, "%s: %3.1f fps", sSDKsample, ifps);
|
||||
|
||||
glutSetWindowTitle(fps);
|
||||
fpsCount = 0;
|
||||
glutSetWindowTitle(fps);
|
||||
fpsCount = 0;
|
||||
|
||||
fpsLimit = ftoi(MAX(1.0f, ifps));
|
||||
sdkResetTimer(&timer);
|
||||
}
|
||||
fpsLimit = ftoi(MAX(1.0f, ifps));
|
||||
sdkResetTimer(&timer);
|
||||
}
|
||||
}
|
||||
|
||||
// render image using CUDA
|
||||
void render() {
|
||||
// map PBO to get CUDA device pointer
|
||||
g_GraphicsMapFlag++;
|
||||
checkCudaErrors(cudaGraphicsMapResources(1, &cuda_pbo_resource, 0));
|
||||
size_t num_bytes;
|
||||
checkCudaErrors(cudaGraphicsResourceGetMappedPointer(
|
||||
(void **)&d_output, &num_bytes, cuda_pbo_resource));
|
||||
// printf("CUDA mapped PBO: May access %ld bytes\n", num_bytes);
|
||||
void render()
|
||||
{
|
||||
// map PBO to get CUDA device pointer
|
||||
g_GraphicsMapFlag++;
|
||||
checkCudaErrors(cudaGraphicsMapResources(1, &cuda_pbo_resource, 0));
|
||||
size_t num_bytes;
|
||||
checkCudaErrors(cudaGraphicsResourceGetMappedPointer((void **)&d_output, &num_bytes, cuda_pbo_resource));
|
||||
// printf("CUDA mapped PBO: May access %ld bytes\n", num_bytes);
|
||||
|
||||
// call CUDA kernel, writing results to PBO
|
||||
render_kernel(gridSize, blockSize, d_output, width, height, w);
|
||||
// call CUDA kernel, writing results to PBO
|
||||
render_kernel(gridSize, blockSize, d_output, width, height, w);
|
||||
|
||||
getLastCudaError("render_kernel failed");
|
||||
getLastCudaError("render_kernel failed");
|
||||
|
||||
if (g_GraphicsMapFlag) {
|
||||
checkCudaErrors(cudaGraphicsUnmapResources(1, &cuda_pbo_resource, 0));
|
||||
g_GraphicsMapFlag--;
|
||||
}
|
||||
if (g_GraphicsMapFlag) {
|
||||
checkCudaErrors(cudaGraphicsUnmapResources(1, &cuda_pbo_resource, 0));
|
||||
g_GraphicsMapFlag--;
|
||||
}
|
||||
}
|
||||
|
||||
// display results using OpenGL (called by GLUT)
|
||||
void display() {
|
||||
sdkStartTimer(&timer);
|
||||
void display()
|
||||
{
|
||||
sdkStartTimer(&timer);
|
||||
|
||||
render();
|
||||
render();
|
||||
|
||||
// display results
|
||||
glClear(GL_COLOR_BUFFER_BIT);
|
||||
// display results
|
||||
glClear(GL_COLOR_BUFFER_BIT);
|
||||
|
||||
// draw image from PBO
|
||||
glDisable(GL_DEPTH_TEST);
|
||||
glRasterPos2i(0, 0);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, pbo);
|
||||
glDrawPixels(width, height, GL_RGBA, GL_UNSIGNED_BYTE, 0);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, 0);
|
||||
// draw image from PBO
|
||||
glDisable(GL_DEPTH_TEST);
|
||||
glRasterPos2i(0, 0);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, pbo);
|
||||
glDrawPixels(width, height, GL_RGBA, GL_UNSIGNED_BYTE, 0);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, 0);
|
||||
|
||||
glutSwapBuffers();
|
||||
glutReportErrors();
|
||||
glutSwapBuffers();
|
||||
glutReportErrors();
|
||||
|
||||
sdkStopTimer(&timer);
|
||||
computeFPS();
|
||||
sdkStopTimer(&timer);
|
||||
computeFPS();
|
||||
}
|
||||
|
||||
void idle() {
|
||||
if (animate) {
|
||||
w += 0.01f;
|
||||
glutPostRedisplay();
|
||||
}
|
||||
void idle()
|
||||
{
|
||||
if (animate) {
|
||||
w += 0.01f;
|
||||
glutPostRedisplay();
|
||||
}
|
||||
}
|
||||
|
||||
void keyboard(unsigned char key, int x, int y) {
|
||||
switch (key) {
|
||||
void keyboard(unsigned char key, int x, int y)
|
||||
{
|
||||
switch (key) {
|
||||
case 27:
|
||||
#if defined(__APPLE__) || defined(MACOSX)
|
||||
exit(EXIT_SUCCESS);
|
||||
glutDestroyWindow(glutGetWindow());
|
||||
return;
|
||||
exit(EXIT_SUCCESS);
|
||||
glutDestroyWindow(glutGetWindow());
|
||||
return;
|
||||
#else
|
||||
glutDestroyWindow(glutGetWindow());
|
||||
return;
|
||||
glutDestroyWindow(glutGetWindow());
|
||||
return;
|
||||
#endif
|
||||
|
||||
case '=':
|
||||
case '+':
|
||||
w += 0.01f;
|
||||
break;
|
||||
w += 0.01f;
|
||||
break;
|
||||
|
||||
case '-':
|
||||
w -= 0.01f;
|
||||
break;
|
||||
w -= 0.01f;
|
||||
break;
|
||||
|
||||
case 'f':
|
||||
linearFiltering = !linearFiltering;
|
||||
setTextureFilterMode(linearFiltering);
|
||||
break;
|
||||
linearFiltering = !linearFiltering;
|
||||
setTextureFilterMode(linearFiltering);
|
||||
break;
|
||||
|
||||
case ' ':
|
||||
animate = !animate;
|
||||
break;
|
||||
animate = !animate;
|
||||
break;
|
||||
|
||||
default:
|
||||
break;
|
||||
}
|
||||
break;
|
||||
}
|
||||
|
||||
glutPostRedisplay();
|
||||
glutPostRedisplay();
|
||||
}
|
||||
|
||||
void reshape(int x, int y) {
|
||||
glViewport(0, 0, x, y);
|
||||
void reshape(int x, int y)
|
||||
{
|
||||
glViewport(0, 0, x, y);
|
||||
|
||||
glMatrixMode(GL_MODELVIEW);
|
||||
glLoadIdentity();
|
||||
glMatrixMode(GL_MODELVIEW);
|
||||
glLoadIdentity();
|
||||
|
||||
glMatrixMode(GL_PROJECTION);
|
||||
glLoadIdentity();
|
||||
glOrtho(0.0, 1.0, 0.0, 1.0, 0.0, 1.0);
|
||||
glMatrixMode(GL_PROJECTION);
|
||||
glLoadIdentity();
|
||||
glOrtho(0.0, 1.0, 0.0, 1.0, 0.0, 1.0);
|
||||
}
|
||||
|
||||
void cleanup() {
|
||||
sdkDeleteTimer(&timer);
|
||||
void cleanup()
|
||||
{
|
||||
sdkDeleteTimer(&timer);
|
||||
|
||||
// add extra check to unmap the resource before unregistering it
|
||||
if (g_GraphicsMapFlag) {
|
||||
checkCudaErrors(cudaGraphicsUnmapResources(1, &cuda_pbo_resource, 0));
|
||||
g_GraphicsMapFlag--;
|
||||
}
|
||||
// add extra check to unmap the resource before unregistering it
|
||||
if (g_GraphicsMapFlag) {
|
||||
checkCudaErrors(cudaGraphicsUnmapResources(1, &cuda_pbo_resource, 0));
|
||||
g_GraphicsMapFlag--;
|
||||
}
|
||||
|
||||
// unregister this buffer object from CUDA C
|
||||
checkCudaErrors(cudaGraphicsUnregisterResource(cuda_pbo_resource));
|
||||
glDeleteBuffers(1, &pbo);
|
||||
cleanupCuda();
|
||||
// unregister this buffer object from CUDA C
|
||||
checkCudaErrors(cudaGraphicsUnregisterResource(cuda_pbo_resource));
|
||||
glDeleteBuffers(1, &pbo);
|
||||
cleanupCuda();
|
||||
}
|
||||
|
||||
void initGLBuffers() {
|
||||
// create pixel buffer object
|
||||
glGenBuffers(1, &pbo);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, pbo);
|
||||
glBufferData(GL_PIXEL_UNPACK_BUFFER_ARB, width * height * sizeof(GLubyte) * 4,
|
||||
0, GL_STREAM_DRAW_ARB);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, 0);
|
||||
void initGLBuffers()
|
||||
{
|
||||
// create pixel buffer object
|
||||
glGenBuffers(1, &pbo);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, pbo);
|
||||
glBufferData(GL_PIXEL_UNPACK_BUFFER_ARB, width * height * sizeof(GLubyte) * 4, 0, GL_STREAM_DRAW_ARB);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, 0);
|
||||
|
||||
// register this buffer object with CUDA
|
||||
checkCudaErrors(cudaGraphicsGLRegisterBuffer(
|
||||
&cuda_pbo_resource, pbo, cudaGraphicsMapFlagsWriteDiscard));
|
||||
// register this buffer object with CUDA
|
||||
checkCudaErrors(cudaGraphicsGLRegisterBuffer(&cuda_pbo_resource, pbo, cudaGraphicsMapFlagsWriteDiscard));
|
||||
}
|
||||
|
||||
// Load raw data from disk
|
||||
uchar *loadRawFile(const char *filename, size_t size) {
|
||||
FILE *fp = fopen(filename, "rb");
|
||||
uchar *loadRawFile(const char *filename, size_t size)
|
||||
{
|
||||
FILE *fp = fopen(filename, "rb");
|
||||
|
||||
if (!fp) {
|
||||
fprintf(stderr, "Error opening file '%s'\n", filename);
|
||||
return 0;
|
||||
}
|
||||
if (!fp) {
|
||||
fprintf(stderr, "Error opening file '%s'\n", filename);
|
||||
return 0;
|
||||
}
|
||||
|
||||
uchar *data = (uchar *)malloc(size);
|
||||
size_t read = fread(data, 1, size, fp);
|
||||
fclose(fp);
|
||||
uchar *data = (uchar *)malloc(size);
|
||||
size_t read = fread(data, 1, size, fp);
|
||||
fclose(fp);
|
||||
|
||||
printf("Read '%s', %zu bytes\n", filename, read);
|
||||
printf("Read '%s', %zu bytes\n", filename, read);
|
||||
|
||||
return data;
|
||||
return data;
|
||||
}
|
||||
|
||||
void initGL(int *argc, char **argv) {
|
||||
// initialize GLUT callback functions
|
||||
glutInit(argc, argv);
|
||||
glutInitDisplayMode(GLUT_RGB | GLUT_DOUBLE);
|
||||
glutInitWindowSize(width, height);
|
||||
glutCreateWindow("CUDA 3D texture");
|
||||
glutDisplayFunc(display);
|
||||
glutKeyboardFunc(keyboard);
|
||||
glutReshapeFunc(reshape);
|
||||
glutIdleFunc(idle);
|
||||
void initGL(int *argc, char **argv)
|
||||
{
|
||||
// initialize GLUT callback functions
|
||||
glutInit(argc, argv);
|
||||
glutInitDisplayMode(GLUT_RGB | GLUT_DOUBLE);
|
||||
glutInitWindowSize(width, height);
|
||||
glutCreateWindow("CUDA 3D texture");
|
||||
glutDisplayFunc(display);
|
||||
glutKeyboardFunc(keyboard);
|
||||
glutReshapeFunc(reshape);
|
||||
glutIdleFunc(idle);
|
||||
|
||||
if (!isGLVersionSupported(2, 0) ||
|
||||
!areGLExtensionsSupported("GL_ARB_pixel_buffer_object")) {
|
||||
fprintf(stderr, "Required OpenGL extensions are missing.");
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
if (!isGLVersionSupported(2, 0) || !areGLExtensionsSupported("GL_ARB_pixel_buffer_object")) {
|
||||
fprintf(stderr, "Required OpenGL extensions are missing.");
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
}
|
||||
|
||||
void runAutoTest(const char *ref_file, char *exec_path) {
|
||||
checkCudaErrors(
|
||||
cudaMalloc((void **)&d_output, width * height * sizeof(GLubyte) * 4));
|
||||
void runAutoTest(const char *ref_file, char *exec_path)
|
||||
{
|
||||
checkCudaErrors(cudaMalloc((void **)&d_output, width * height * sizeof(GLubyte) * 4));
|
||||
|
||||
// render the volumeData
|
||||
render_kernel(gridSize, blockSize, d_output, width, height, w);
|
||||
// render the volumeData
|
||||
render_kernel(gridSize, blockSize, d_output, width, height, w);
|
||||
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
getLastCudaError("render_kernel failed");
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
getLastCudaError("render_kernel failed");
|
||||
|
||||
void *h_output = malloc(width * height * sizeof(GLubyte) * 4);
|
||||
checkCudaErrors(cudaMemcpy(h_output, d_output,
|
||||
width * height * sizeof(GLubyte) * 4,
|
||||
cudaMemcpyDeviceToHost));
|
||||
sdkDumpBin(h_output, width * height * sizeof(GLubyte) * 4,
|
||||
"simpleTexture3D.bin");
|
||||
void *h_output = malloc(width * height * sizeof(GLubyte) * 4);
|
||||
checkCudaErrors(cudaMemcpy(h_output, d_output, width * height * sizeof(GLubyte) * 4, cudaMemcpyDeviceToHost));
|
||||
sdkDumpBin(h_output, width * height * sizeof(GLubyte) * 4, "simpleTexture3D.bin");
|
||||
|
||||
bool bTestResult = sdkCompareBin2BinFloat(
|
||||
"simpleTexture3D.bin", sdkFindFilePath(ref_file, exec_path),
|
||||
width * height, MAX_EPSILON_ERROR, THRESHOLD, exec_path);
|
||||
bool bTestResult = sdkCompareBin2BinFloat("simpleTexture3D.bin",
|
||||
sdkFindFilePath(ref_file, exec_path),
|
||||
width * height,
|
||||
MAX_EPSILON_ERROR,
|
||||
THRESHOLD,
|
||||
exec_path);
|
||||
|
||||
checkCudaErrors(cudaFree(d_output));
|
||||
free(h_output);
|
||||
checkCudaErrors(cudaFree(d_output));
|
||||
free(h_output);
|
||||
|
||||
sdkStopTimer(&timer);
|
||||
sdkDeleteTimer(&timer);
|
||||
sdkStopTimer(&timer);
|
||||
sdkDeleteTimer(&timer);
|
||||
|
||||
exit(bTestResult ? EXIT_SUCCESS : EXIT_FAILURE);
|
||||
exit(bTestResult ? EXIT_SUCCESS : EXIT_FAILURE);
|
||||
}
|
||||
|
||||
void loadVolumeData(char *exec_path) {
|
||||
// load volume data
|
||||
const char *path = sdkFindFilePath(volumeFilename, exec_path);
|
||||
void loadVolumeData(char *exec_path)
|
||||
{
|
||||
// load volume data
|
||||
const char *path = sdkFindFilePath(volumeFilename, exec_path);
|
||||
|
||||
if (path == NULL) {
|
||||
fprintf(stderr, "Error unable to find 3D Volume file: '%s'\n",
|
||||
volumeFilename);
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
if (path == NULL) {
|
||||
fprintf(stderr, "Error unable to find 3D Volume file: '%s'\n", volumeFilename);
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
size_t size = volumeSize.width * volumeSize.height * volumeSize.depth;
|
||||
uchar *h_volume = loadRawFile(path, size);
|
||||
size_t size = volumeSize.width * volumeSize.height * volumeSize.depth;
|
||||
uchar *h_volume = loadRawFile(path, size);
|
||||
|
||||
initCuda(h_volume, volumeSize);
|
||||
sdkCreateTimer(&timer);
|
||||
initCuda(h_volume, volumeSize);
|
||||
sdkCreateTimer(&timer);
|
||||
|
||||
free(h_volume);
|
||||
free(h_volume);
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Program main
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
int main(int argc, char **argv) {
|
||||
pArgc = &argc;
|
||||
pArgv = argv;
|
||||
int main(int argc, char **argv)
|
||||
{
|
||||
pArgc = &argc;
|
||||
pArgv = argv;
|
||||
|
||||
char *ref_file = NULL;
|
||||
char *ref_file = NULL;
|
||||
|
||||
#if defined(__linux__)
|
||||
setenv("DISPLAY", ":0", 0);
|
||||
setenv("DISPLAY", ":0", 0);
|
||||
#endif
|
||||
|
||||
printf("%s Starting...\n\n", sSDKsample);
|
||||
printf("%s Starting...\n\n", sSDKsample);
|
||||
|
||||
if (checkCmdLineFlag(argc, (const char **)argv, "file")) {
|
||||
fpsLimit = frameCheckNumber;
|
||||
getCmdLineArgumentString(argc, (const char **)argv, "file", &ref_file);
|
||||
}
|
||||
if (checkCmdLineFlag(argc, (const char **)argv, "file")) {
|
||||
fpsLimit = frameCheckNumber;
|
||||
getCmdLineArgumentString(argc, (const char **)argv, "file", &ref_file);
|
||||
}
|
||||
|
||||
// use command-line specified CUDA device, otherwise use device with highest
|
||||
// Gflops/s
|
||||
findCudaDevice(argc, (const char **)argv);
|
||||
// use command-line specified CUDA device, otherwise use device with highest
|
||||
// Gflops/s
|
||||
findCudaDevice(argc, (const char **)argv);
|
||||
|
||||
if (ref_file) {
|
||||
loadVolumeData(argv[0]);
|
||||
runAutoTest(ref_file, argv[0]);
|
||||
} else {
|
||||
initGL(&argc, argv);
|
||||
if (ref_file) {
|
||||
loadVolumeData(argv[0]);
|
||||
runAutoTest(ref_file, argv[0]);
|
||||
}
|
||||
else {
|
||||
initGL(&argc, argv);
|
||||
|
||||
// OpenGL buffers
|
||||
initGLBuffers();
|
||||
// OpenGL buffers
|
||||
initGLBuffers();
|
||||
|
||||
loadVolumeData(argv[0]);
|
||||
}
|
||||
loadVolumeData(argv[0]);
|
||||
}
|
||||
|
||||
printf(
|
||||
"Press space to toggle animation\n"
|
||||
"Press '+' and '-' to change displayed slice\n");
|
||||
printf("Press space to toggle animation\n"
|
||||
"Press '+' and '-' to change displayed slice\n");
|
||||
|
||||
#if defined(__APPLE__) || defined(MACOSX)
|
||||
atexit(cleanup);
|
||||
atexit(cleanup);
|
||||
#else
|
||||
glutCloseFunc(cleanup);
|
||||
glutCloseFunc(cleanup);
|
||||
#endif
|
||||
|
||||
glutMainLoop();
|
||||
glutMainLoop();
|
||||
|
||||
exit(EXIT_SUCCESS);
|
||||
exit(EXIT_SUCCESS);
|
||||
}
|
||||
|
||||
@@ -28,111 +28,111 @@
|
||||
#ifndef _SIMPLETEXTURE3D_KERNEL_CU_
|
||||
#define _SIMPLETEXTURE3D_KERNEL_CU_
|
||||
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
#include <math.h>
|
||||
|
||||
#include <helper_cuda.h>
|
||||
#include <helper_math.h>
|
||||
#include <math.h>
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
typedef unsigned int uint;
|
||||
typedef unsigned int uint;
|
||||
typedef unsigned char uchar;
|
||||
|
||||
cudaArray *d_volumeArray = 0;
|
||||
cudaTextureObject_t tex; // 3D texture
|
||||
cudaArray *d_volumeArray = 0;
|
||||
cudaTextureObject_t tex; // 3D texture
|
||||
|
||||
__global__ void d_render(uint *d_output, uint imageW, uint imageH, float w,
|
||||
cudaTextureObject_t texObj) {
|
||||
uint x = __umul24(blockIdx.x, blockDim.x) + threadIdx.x;
|
||||
uint y = __umul24(blockIdx.y, blockDim.y) + threadIdx.y;
|
||||
__global__ void d_render(uint *d_output, uint imageW, uint imageH, float w, cudaTextureObject_t texObj)
|
||||
{
|
||||
uint x = __umul24(blockIdx.x, blockDim.x) + threadIdx.x;
|
||||
uint y = __umul24(blockIdx.y, blockDim.y) + threadIdx.y;
|
||||
|
||||
float u = x / (float)imageW;
|
||||
float v = y / (float)imageH;
|
||||
// read from 3D texture
|
||||
float voxel = tex3D<float>(texObj, u, v, w);
|
||||
float u = x / (float)imageW;
|
||||
float v = y / (float)imageH;
|
||||
// read from 3D texture
|
||||
float voxel = tex3D<float>(texObj, u, v, w);
|
||||
|
||||
if ((x < imageW) && (y < imageH)) {
|
||||
// write output color
|
||||
uint i = __umul24(y, imageW) + x;
|
||||
d_output[i] = voxel * 255;
|
||||
}
|
||||
if ((x < imageW) && (y < imageH)) {
|
||||
// write output color
|
||||
uint i = __umul24(y, imageW) + x;
|
||||
d_output[i] = voxel * 255;
|
||||
}
|
||||
}
|
||||
|
||||
extern "C" void setTextureFilterMode(bool bLinearFilter) {
|
||||
if (tex) {
|
||||
checkCudaErrors(cudaDestroyTextureObject(tex));
|
||||
}
|
||||
cudaResourceDesc texRes;
|
||||
memset(&texRes, 0, sizeof(cudaResourceDesc));
|
||||
extern "C" void setTextureFilterMode(bool bLinearFilter)
|
||||
{
|
||||
if (tex) {
|
||||
checkCudaErrors(cudaDestroyTextureObject(tex));
|
||||
}
|
||||
cudaResourceDesc texRes;
|
||||
memset(&texRes, 0, sizeof(cudaResourceDesc));
|
||||
|
||||
texRes.resType = cudaResourceTypeArray;
|
||||
texRes.res.array.array = d_volumeArray;
|
||||
texRes.resType = cudaResourceTypeArray;
|
||||
texRes.res.array.array = d_volumeArray;
|
||||
|
||||
cudaTextureDesc texDescr;
|
||||
memset(&texDescr, 0, sizeof(cudaTextureDesc));
|
||||
cudaTextureDesc texDescr;
|
||||
memset(&texDescr, 0, sizeof(cudaTextureDesc));
|
||||
|
||||
texDescr.normalizedCoords = true;
|
||||
texDescr.filterMode =
|
||||
bLinearFilter ? cudaFilterModeLinear : cudaFilterModePoint;
|
||||
;
|
||||
texDescr.addressMode[0] = cudaAddressModeWrap;
|
||||
texDescr.addressMode[1] = cudaAddressModeWrap;
|
||||
texDescr.addressMode[2] = cudaAddressModeWrap;
|
||||
texDescr.readMode = cudaReadModeNormalizedFloat;
|
||||
texDescr.normalizedCoords = true;
|
||||
texDescr.filterMode = bLinearFilter ? cudaFilterModeLinear : cudaFilterModePoint;
|
||||
;
|
||||
texDescr.addressMode[0] = cudaAddressModeWrap;
|
||||
texDescr.addressMode[1] = cudaAddressModeWrap;
|
||||
texDescr.addressMode[2] = cudaAddressModeWrap;
|
||||
texDescr.readMode = cudaReadModeNormalizedFloat;
|
||||
|
||||
checkCudaErrors(cudaCreateTextureObject(&tex, &texRes, &texDescr, NULL));
|
||||
checkCudaErrors(cudaCreateTextureObject(&tex, &texRes, &texDescr, NULL));
|
||||
}
|
||||
|
||||
extern "C" void initCuda(const uchar *h_volume, cudaExtent volumeSize) {
|
||||
// create 3D array
|
||||
cudaChannelFormatDesc channelDesc = cudaCreateChannelDesc<uchar>();
|
||||
checkCudaErrors(cudaMalloc3DArray(&d_volumeArray, &channelDesc, volumeSize));
|
||||
extern "C" void initCuda(const uchar *h_volume, cudaExtent volumeSize)
|
||||
{
|
||||
// create 3D array
|
||||
cudaChannelFormatDesc channelDesc = cudaCreateChannelDesc<uchar>();
|
||||
checkCudaErrors(cudaMalloc3DArray(&d_volumeArray, &channelDesc, volumeSize));
|
||||
|
||||
// copy data to 3D array
|
||||
cudaMemcpy3DParms copyParams = {0};
|
||||
copyParams.srcPtr =
|
||||
make_cudaPitchedPtr((void *)h_volume, volumeSize.width * sizeof(uchar),
|
||||
volumeSize.width, volumeSize.height);
|
||||
copyParams.dstArray = d_volumeArray;
|
||||
copyParams.extent = volumeSize;
|
||||
copyParams.kind = cudaMemcpyHostToDevice;
|
||||
checkCudaErrors(cudaMemcpy3D(©Params));
|
||||
// copy data to 3D array
|
||||
cudaMemcpy3DParms copyParams = {0};
|
||||
copyParams.srcPtr =
|
||||
make_cudaPitchedPtr((void *)h_volume, volumeSize.width * sizeof(uchar), volumeSize.width, volumeSize.height);
|
||||
copyParams.dstArray = d_volumeArray;
|
||||
copyParams.extent = volumeSize;
|
||||
copyParams.kind = cudaMemcpyHostToDevice;
|
||||
checkCudaErrors(cudaMemcpy3D(©Params));
|
||||
|
||||
cudaResourceDesc texRes;
|
||||
memset(&texRes, 0, sizeof(cudaResourceDesc));
|
||||
cudaResourceDesc texRes;
|
||||
memset(&texRes, 0, sizeof(cudaResourceDesc));
|
||||
|
||||
texRes.resType = cudaResourceTypeArray;
|
||||
texRes.res.array.array = d_volumeArray;
|
||||
texRes.resType = cudaResourceTypeArray;
|
||||
texRes.res.array.array = d_volumeArray;
|
||||
|
||||
cudaTextureDesc texDescr;
|
||||
memset(&texDescr, 0, sizeof(cudaTextureDesc));
|
||||
cudaTextureDesc texDescr;
|
||||
memset(&texDescr, 0, sizeof(cudaTextureDesc));
|
||||
|
||||
// access with normalized texture coordinates
|
||||
texDescr.normalizedCoords = true;
|
||||
// linear interpolation
|
||||
texDescr.filterMode = cudaFilterModeLinear;
|
||||
// wrap texture coordinates
|
||||
texDescr.addressMode[0] = cudaAddressModeWrap;
|
||||
texDescr.addressMode[1] = cudaAddressModeWrap;
|
||||
texDescr.addressMode[2] = cudaAddressModeWrap;
|
||||
texDescr.readMode = cudaReadModeNormalizedFloat;
|
||||
// access with normalized texture coordinates
|
||||
texDescr.normalizedCoords = true;
|
||||
// linear interpolation
|
||||
texDescr.filterMode = cudaFilterModeLinear;
|
||||
// wrap texture coordinates
|
||||
texDescr.addressMode[0] = cudaAddressModeWrap;
|
||||
texDescr.addressMode[1] = cudaAddressModeWrap;
|
||||
texDescr.addressMode[2] = cudaAddressModeWrap;
|
||||
texDescr.readMode = cudaReadModeNormalizedFloat;
|
||||
|
||||
checkCudaErrors(cudaCreateTextureObject(&tex, &texRes, &texDescr, NULL));
|
||||
checkCudaErrors(cudaCreateTextureObject(&tex, &texRes, &texDescr, NULL));
|
||||
}
|
||||
|
||||
extern "C" void render_kernel(dim3 gridSize, dim3 blockSize, uint *d_output,
|
||||
uint imageW, uint imageH, float w) {
|
||||
d_render<<<gridSize, blockSize>>>(d_output, imageW, imageH, w, tex);
|
||||
extern "C" void render_kernel(dim3 gridSize, dim3 blockSize, uint *d_output, uint imageW, uint imageH, float w)
|
||||
{
|
||||
d_render<<<gridSize, blockSize>>>(d_output, imageW, imageH, w, tex);
|
||||
}
|
||||
|
||||
void cleanupCuda() {
|
||||
if (tex) {
|
||||
checkCudaErrors(cudaDestroyTextureObject(tex));
|
||||
}
|
||||
if (d_volumeArray) {
|
||||
checkCudaErrors(cudaFreeArray(d_volumeArray));
|
||||
}
|
||||
void cleanupCuda()
|
||||
{
|
||||
if (tex) {
|
||||
checkCudaErrors(cudaDestroyTextureObject(tex));
|
||||
}
|
||||
if (d_volumeArray) {
|
||||
checkCudaErrors(cudaFreeArray(d_volumeArray));
|
||||
}
|
||||
}
|
||||
|
||||
#endif // #ifndef _SIMPLETEXTURE3D_KERNEL_CU_
|
||||
#endif // #ifndef _SIMPLETEXTURE3D_KERNEL_CU_
|
||||
|
||||
Reference in New Issue
Block a user