void display(){
runCuda();
string title = "565Raytracer | " + utilityCore::convertIntToString(iterations) + " Frames";
glutSetWindowTitle(title.c_str());
glBindBuffer( GL_PIXEL_UNPACK_BUFFER, pbo);
glBindTexture(GL_TEXTURE_2D, displayImage);
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height,
GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glClear(GL_COLOR_BUFFER_BIT);
// VAO, shader program, and texture already bound
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_SHORT, 0);
//draw text
char displaystring[200];
strcpy ( displaystring, "Press H for help." );
drawText ( 50, 50, displaystring);
glutPostRedisplay();
glutSwapBuffers();
}
void initCuda()
{
// First initialize OpenGL context, so we can properly set the GL
// for CUDA. NVIDIA notes this is necessary in order to achieve
// optimal performance with OpenGL/CUDA interop. use command-line
// specified CUDA device, otherwise use device with highest Gflops/s
int devCount= 0;
cudaGetDeviceCount(&devCount);
if( devCount < 1 )
{
printf("No GPUS detected\n");
exit(EXIT_FAILURE);
}
cudaGLSetGLDevice( 0 );
//Set Up scenary
setup_scene();
createPBO(&pbo);
createTexture(&textureID,image_width,image_height);
// Clean up on program exit
atexit(cleanupCuda);
runCuda();
}
//-----------------------------------------------------------------------------
// Name: Render()
// Desc: Draws the scene
//-----------------------------------------------------------------------------
VOID Render()
{
// Clear the backbuffer to a black color
float ClearColor[4] = {0, 0, 0, 0};
g_pd3dDevice->ClearRenderTargetView(g_pSwapChainRTV, ClearColor);
// Run CUDA to update vertex positions
runCuda();
// Draw frame
{
// Setup the world, view, and projection matrices
SetupMatrices();
// Render the vertex buffer contents
UINT stride = sizeof(CUSTOMVERTEX);
UINT offset = 0;
g_pd3dDevice->IASetVertexBuffers(0, 1, &g_pVB, &stride, &offset);
g_pSimpleTechnique->GetPassByIndex(0)->Apply(0);
g_pd3dDevice->Draw(g_NumVertices, 0);
}
// Present the backbuffer contents to the display
g_pSwapChain->Present(0, 0);
anim += 0.01f;
}
void display(){
// DEBUG: display only one frame
/*
if ( frame > 5 )
return;
*/
runCuda();
time_t seconds2 = time (NULL);
if(seconds2-seconds >= 1){
fps = fpstracker/(seconds2-seconds);
fpstracker = 0;
seconds = seconds2;
}
string title = "CIS565 Rasterizer | "+ utilityCore::convertIntToString((int)fps) + "FPS";
glutSetWindowTitle(title.c_str());
glBindBuffer( GL_PIXEL_UNPACK_BUFFER, pbo);
glBindTexture(GL_TEXTURE_2D, displayImage);
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height,
GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glClear(GL_COLOR_BUFFER_BIT);
// VAO, shader program, and texture already bound
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_SHORT, 0);
glutPostRedisplay();
glutSwapBuffers();
}
void mainLoop() {
while(!glfwWindowShouldClose(window)){
glfwPollEvents();
runCuda();
time_t seconds2 = time (NULL);
if(seconds2-seconds >= 1){
fps = fpstracker/(seconds2-seconds);
fpstracker = 0;
seconds = seconds2;
}
string title = "CIS565 Rasterizer | " + utilityCore::convertIntToString((int)fps) + " FPS";
glfwSetWindowTitle(window, title.c_str());
glBindBuffer(GL_PIXEL_UNPACK_BUFFER, pbo);
glBindTexture(GL_TEXTURE_2D, displayImage);
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height, GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glClear(GL_COLOR_BUFFER_BIT);
// VAO, shader program, and texture already bound
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_SHORT, 0);
glfwSwapBuffers(window);
}
glfwDestroyWindow(window);
glfwTerminate();
}
int main( int argc, char *argv[] ) {
initDisplay();
// get number of SMs on this GPU
int devID;
/*cudaDeviceProp props;
cudaGetDevice(&devID);
cudaGetDeviceProperties(&props, devID);
check();
printf("using device %s \n", props.name);
printf("number of MPs %d \n", props.multiProcessorCount);
printf("max threads per block %d \n", props.maxThreadsPerBlock);
printf("max concurrent kernels %d \n", props.concurrentKernels);
printf("number of bodies %d \n", numBodies);
*/
if(argc == 2)
{
if(strcmp(argv[1], "--cpu") == 0)
{
gpu = false;
}
else if(strcmp(argv[1], "--gpu") == 0)
{
gpu = true;
}
else
{
printf("wrong argument %s \nallowed: NULL, --cpu, --gpu \n", argv[1]);
exit(EXIT_FAILURE);
}
}
if(randData){
loadDataRand(numBodies);
}else{
loadData("data/data.tab", numBodies);
}
init(numBodies);
printf("stop with CTRL+C\n");
do {
runCuda();
showGalaxy(h_particleData, numBodies, false);
//count++;
} while (count < 64);
printf("finished...\n");
if(gpu)
{
cudaFree(d_particleData);
}
closeWindow();
printf("close...\n");
return EXIT_SUCCESS;
}
////////////////////////////////////////////////////////////////////////////////
//! Run a simple test for CUDA
////////////////////////////////////////////////////////////////////////////////
CUTBoolean runTest(int argc, char** argv)
{
if (!cutCheckCmdLineFlag(argc, (const char **)argv, "noqatest") ||
cutCheckCmdLineFlag(argc, (const char **)argv, "noprompt"))
{
g_bQAReadback = true;
fpsLimit = frameCheckNumber;
}
// First initialize OpenGL context, so we can properly set the GL for CUDA.
// This is necessary in order to achieve optimal performance with OpenGL/CUDA interop.
if (CUTFalse == initGL(argc, argv)) {
return CUTFalse;
}
// use command-line specified CUDA device, otherwise use device with highest Gflops/s
if( cutCheckCmdLineFlag(argc, (const char**)argv, "device") )
cutilGLDeviceInit(argc, argv);
else {
cudaGLSetGLDevice( cutGetMaxGflopsDeviceId() );
}
// Create the CUTIL timer
cutilCheckError( cutCreateTimer( &timer));
// register callbacks
glutDisplayFunc(display);
glutKeyboardFunc(keyboard);
glutMouseFunc(mouse);
glutMotionFunc(motion);
if (g_bQAReadback) {
g_CheckRender = new CheckBackBuffer(window_width, window_height, 4);
g_CheckRender->setPixelFormat(GL_RGBA);
g_CheckRender->setExecPath(argv[0]);
g_CheckRender->EnableQAReadback(true);
}
// create VBO
createVBO(&vbo);
// run the cuda part
runCuda(vbo);
// check result of Cuda step
checkResultCuda(argc, argv, vbo);
atexit(cleanup);
// start rendering mainloop
glutMainLoop();
cudaThreadExit();
return CUTTrue;
}
void initCuda(){
// Use device with highest Gflops/s
cudaGLSetGLDevice( compat_getMaxGflopsDeviceId() );
initPBO(&pbo);
// Clean up on program exit
atexit(cleanupCuda);
runCuda();
}
void initCuda(){
// Use device with highest Gflops/s
// Had to update this to remove cutil version
cudaGLSetGLDevice( gpuGetMaxGflopsDeviceId() );
initPBO(&pbo);
// Clean up on program exit
atexit(cleanupCuda);
runCuda();
}
////////////////////////////////////////////////////////////////////////////////
//! Display callback
////////////////////////////////////////////////////////////////////////////////
void display()
{
// run CUDA kernel to generate vertex positions
boost::shared_ptr<thrust::device_vector<ms::point> > dv = runCuda();
if (!dv)
return;
ms::gl_buffer<ms::point> glb(dv->size());
glb << *dv;
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
// set view matrix
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glTranslatef(0.0, 0.0, translate_z);
glRotatef(rotate_x, 1.0, 0.0, 0.0);
glRotatef(rotate_y, 0.0, 1.0, 0.0);
// render from the vbo
glBindBuffer(GL_ARRAY_BUFFER,
glb.buffer());
glVertexPointer(3, // GLint size
GL_FLOAT, // GLenum type
24, // GLsizei stride
0); // offset
// assert (displayable.size_bytes == mesh_width * mesh_height * sizeof(mephitis::point));
//glColor3f(1.0, 1.0, 1.0);
glColorPointer(3,
GL_FLOAT,
24,
(const GLvoid*)(sizeof(float) * 3));
glEnableClientState(GL_VERTEX_ARRAY);
glEnableClientState(GL_COLOR_ARRAY);
glDrawArrays(GL_POINTS, // GLenum mode
0, // GLint first
glb.size() // GLsizei count
);
glDisableClientState(GL_COLOR_ARRAY);
glDisableClientState(GL_VERTEX_ARRAY);
glutSwapBuffers();
glutPostRedisplay();
}
////////////////////////////////////////////////////////////////////////////////
//! Display callback
////////////////////////////////////////////////////////////////////////////////
void display()
{
sdkStartTimer(&timer);
// run CUDA kernel to generate vertex positions
runCuda(&cuda_vbo_resource);
//簡易ライトセット
glEnable(GL_LIGHTING);
glEnable(GL_LIGHT0);
glLightfv(GL_LIGHT0, GL_POSITION, gkLightPos);
glLightfv(GL_LIGHT0, GL_DIFFUSE, gkLightDiff);
// glLightfv(GL_LIGHT0, GL_AMBIENT, gkLightAmb);
glEnable(GL_LIGHT1);
glLightfv(GL_LIGHT1, GL_POSITION, gkLightPos2);
glLightfv(GL_LIGHT1, GL_DIFFUSE, gkLightDiff2);
//Zバッファ有効
glEnable(GL_DEPTH_TEST);
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
// set view matrix
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glTranslatef(0.0, 0.0, translate_z);
glRotatef(rotate_x, 1.0, 0.0, 0.0);
glRotatef(rotate_y, 0.0, 1.0, 0.0);
// Earth
// glMaterialfv(GL_FRONT, GL_DIFFUSE, gkMaterial);
glutSolidSphere(50.0 * h_axis_radius, 20, 20);
glDisable(GL_LIGHTING);
// render from the vbo
glBindBuffer(GL_ARRAY_BUFFER, vbo);
glVertexPointer(4, GL_FLOAT, 0, 0);
glEnableClientState(GL_VERTEX_ARRAY);
glColor3f(1.0, 1.0, 1.0);
glDrawArrays(GL_POINTS, 0, mesh_width * mesh_height);
glDisableClientState(GL_VERTEX_ARRAY);
glutSwapBuffers();
g_fAnim += 0.01f;
sdkStopTimer(&timer);
computeFPS();
}
void initCuda(){
// Use device with highest Gflops/s
#if CUDA_VERSION >= 5000
cudaGLSetGLDevice( gpuGetMaxGflopsDeviceId() );
#else
cudaGLSetGLDevice( cutGetMaxGflopsDeviceId() );
#endif
initPBO(&pbo);
// Clean up on program exit
atexit(cleanupCuda);
runCuda();
}
void initCuda(){
// Use device with highest Gflops/s
cudaGLSetGLDevice( cutGetMaxGflopsDeviceId() );
initPBO(&pbo);
dptr=NULL;
cudaGLMapBufferObject((void**)&dptr, pbo);
clearPBOpos(dptr,width,height);
cudaGLUnmapBufferObject(pbo);
// Clean up on program exit
atexit(cleanupCuda);
SetScissorWindow(glm::vec4(300,300,500,500));
texture.mapptr = stbi_load("cow.jpeg",&texture.width, &texture.height,&texture.depth,0);
runCuda();
}
////////////////////////////////////////////////////////////////////////////////
//! Display callback
////////////////////////////////////////////////////////////////////////////////
void display()
{
cutilCheckError(cutStartTimer(timer));
// run CUDA kernel to generate vertex positions
runCuda(vbo);
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
// set view matrix
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glTranslatef(0.0, 0.0, translate_z);
glRotatef(rotate_x, 1.0, 0.0, 0.0);
glRotatef(rotate_y, 0.0, 1.0, 0.0);
// render from the vbo
glBindBuffer(GL_ARRAY_BUFFER, vbo);
glVertexPointer(4, GL_FLOAT, 0, 0);
glEnableClientState(GL_VERTEX_ARRAY);
glColor3f(1.0, 0.0, 0.0);
glDrawArrays(GL_POINTS, 0, mesh_width * mesh_height);
glDisableClientState(GL_VERTEX_ARRAY);
if (g_CheckRender && g_CheckRender->IsQAReadback() && g_Verify) {
// readback for QA testing
printf("> (Frame %d) Readback BackBuffer\n", frameCount);
g_CheckRender->readback( window_width, window_height );
g_CheckRender->savePPM(sOriginal[g_Index], true, NULL);
if (!g_CheckRender->PPMvsPPM(sOriginal[g_Index], sReference[g_Index], MAX_EPSILON_ERROR, 0.15f)) {
g_TotalErrors++;
}
else
{
printf( "TEST PASSED\n" );
}
g_Verify = false;
}
glutSwapBuffers();
glutPostRedisplay();
anim += 0.01;
cutilCheckError(cutStopTimer(timer));
computeFPS();
}
void initCuda(){
// Use device with highest Gflops/s
cudaGLSetGLDevice( cutGetMaxGflopsDeviceId() );
initPBO(&pbo);
paraMap = new float[(int)renderCam->resolution.x *(int)renderCam->resolution.y];
effectiveRayMap =new int[(int)renderCam->resolution.x *(int)renderCam->resolution.y];
initialRayMap = new ray[(int)renderCam->resolution.x * (int)renderCam->resolution.y];
generateRayMap(renderCam, targetFrame);
// Clean up on program exit
atexit(cleanupCuda);
runCuda();
}
void display(){
runCuda();
glBindBuffer( GL_PIXEL_UNPACK_BUFFER, pbo);
glBindTexture(GL_TEXTURE_2D, image);
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height,
GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glClear(GL_COLOR_BUFFER_BIT);
// VAO, shader program, and texture already bound
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_SHORT, 0);
glutPostRedisplay();
glutSwapBuffers();
}
void initCuda(){
// Use device with highest Gflops/s
cudaGLSetGLDevice( compat_getMaxGflopsDeviceId() );
initPBO(&pbo);
modelView = glm::lookAt(cameraPosition, lookat, up);
projection = glm::perspective(fovy, float(width) / float(height), zNear, zFar);
viewPort[0] = glm::vec4(-float(width)/2,0,0,0.0f);
viewPort[1] = glm::vec4(0,-(float)height/2,0,0.0f);
viewPort[2] = glm::vec4(0,0,1.f/2.0f,0.0f);
viewPort[3] = glm::vec4((float)width/2,(float)height/2,1.0f/2.0f,1.0f);
// Clean up on program exit
atexit(cleanupCuda);
runCuda();
}
void display(){
runCuda();
string title = "CIS565 Render | " + utilityCore::convertIntToString(iterations) + " Iterations";
glfwSetWindowTitle(title.c_str());
glBindBuffer( GL_PIXEL_UNPACK_BUFFER, pbo);
glBindTexture(GL_TEXTURE_2D, displayImage);
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height,
GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glClear(GL_COLOR_BUFFER_BIT);
// VAO, shader program, and texture already bound
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_SHORT, 0);
glfwSwapBuffers();
}
void display()
{
// run CUDA kernel
runCuda();
// Create a texture from the buffer
glBindBuffer( GL_PIXEL_UNPACK_BUFFER, pbo);
// bind texture from PBO
glBindTexture(GL_TEXTURE_2D, textureID);
// Note: glTexSubImage2D will perform a format conversion if the
// buffer is a different format from the texture. We created the
// texture with format GL_RGBA8. In glTexSubImage2D we specified
// GL_BGRA and GL_UNSIGNED_INT. This is a fast-path combination
// Note: NULL indicates the data resides in device memory
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, image_width, image_height,
GL_RGBA, GL_UNSIGNED_BYTE, NULL);
// Draw a single Quad with texture coordinates for each vertex.
glBegin(GL_QUADS);
glTexCoord2f(0.0f,1.0f);
glVertex3f(0.0f,0.0f,0.0f);
glTexCoord2f(0.0f,0.0f);
glVertex3f(0.0f,1.0f,0.0f);
glTexCoord2f(1.0f,0.0f);
glVertex3f(1.0f,1.0f,0.0f);
glTexCoord2f(1.0f,1.0f);
glVertex3f(1.0f,0.0f,0.0f);
glEnd();
// Don't forget to swap the buffers!
glutSwapBuffers();
// if animFlag is true, then indicate the display needs to be redrawn
if(animFlag) {
glutPostRedisplay();
animTime += animInc;
}
}
void display()
{
static float fps = 0;
frame++;
int time=glutGet(GLUT_ELAPSED_TIME);
if (time - timebase > 1000) {
fps = frame*1000.0f/(time-timebase);
timebase = time;
frame = 0;
}
float executionTime = glutGet(GLUT_ELAPSED_TIME) - timeSinceLastFrame;
timeSinceLastFrame = glutGet(GLUT_ELAPSED_TIME);
runCuda();
char title[100];
sprintf( title, "Flocking Simulation [%0.2f fps] [%0.2fms] ", fps, executionTime);
glutSetWindowTitle(title);
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
#if VISUALIZE == 1
glUseProgram(program[PASS_THROUGH]);
glEnableVertexAttribArray(positionLocation);
glBindBuffer(GL_ARRAY_BUFFER, planetVBO);
glVertexAttribPointer((GLuint)positionLocation, 4, GL_FLOAT, GL_FALSE, 0, 0);
glEnableVertexAttribArray(velocityLocation);
glBindBuffer(GL_ARRAY_BUFFER, velocityVBO);
glVertexAttribPointer((GLuint)velocityLocation, 3, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, planetIBO);
glPointSize(4.0f);
glDrawElements(GL_POINTS, N_FOR_VIS, GL_UNSIGNED_INT, 0);
glDisableVertexAttribArray(positionLocation);
// Draw_Axes();
#endif
glutPostRedisplay();
glutSwapBuffers();
}
void mainLoop() {
while(!glfwWindowShouldClose(window)){
glfwPollEvents();
runCuda();
string title = "CIS565 Render | " + utilityCore::convertIntToString(iterations) + " Iterations";
glfwSetWindowTitle(window, title.c_str());
glBindBuffer(GL_PIXEL_UNPACK_BUFFER, pbo);
glBindTexture(GL_TEXTURE_2D, displayImage);
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height, GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glClear(GL_COLOR_BUFFER_BIT);
// VAO, shader program, and texture already bound
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_SHORT, 0);
glfwSwapBuffers(window);
}
glfwDestroyWindow(window);
glfwTerminate();
}
void display(){
theFpsTracker.timestamp();
cudaEvent_t start, stop;
float time;
// Keep track of time
cudaEventCreate(&start);
cudaEventCreate(&stop);
cudaEventRecord( start, 0 );
runCuda();
cudaEventRecord( stop, 0 );
cudaEventSynchronize( stop );
cudaEventElapsedTime( &time, start, stop );
cudaEventDestroy( start );
cudaEventDestroy( stop );
char info[1024];
sprintf(info, "565Raytracer | %i Iterations | Framerate : %3.1f fps | GPU Elapsed Time : %3.1f ms", iterations, theFpsTracker.fpsAverage(), time);
string title(info);
glutSetWindowTitle(title.c_str());
glBindBuffer( GL_PIXEL_UNPACK_BUFFER, pbo);
glBindTexture(GL_TEXTURE_2D, displayImage);
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height,
GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glClear(GL_COLOR_BUFFER_BIT);
// VAO, shader program, and texture already bound
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_SHORT, 0);
glutPostRedisplay();
glutSwapBuffers();
}
//! Display callback for GLUT
//! Keyboard events handler for GLUT
//! Display callback for GLUT
void display()
{
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
// set view matrix
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glTranslatef(0.0, 0.0, translate_z);
glRotatef(rotate_x, 1.0, 0.0, 0.0);
glRotatef(rotate_y, 0.0, 1.0, 0.0);
// run CUDA kernel to generate vertex positions
runCuda();
// render the data
renderCuda(drawMode);
glutSwapBuffers();
glutPostRedisplay();
animTime += 0.01;
}
////////////////////////////////////////////////////////////////////////////////
//! Run a simple test for CUDA
////////////////////////////////////////////////////////////////////////////////
void runTest(int argc, char **argv, char *ref_file)
{
// Register the window class
WNDCLASSEX wc = { sizeof(WNDCLASSEX), CS_CLASSDC, MsgProc, 0L, 0L,
GetModuleHandle(NULL), NULL, NULL, NULL, NULL,
"CUDA/D3D10 simpleD3D10", NULL
};
RegisterClassEx(&wc);
// Create the application's window
int xBorder = ::GetSystemMetrics(SM_CXSIZEFRAME);
int yMenu = ::GetSystemMetrics(SM_CYMENU);
int yBorder = ::GetSystemMetrics(SM_CYSIZEFRAME);
HWND hWnd = CreateWindow(wc.lpszClassName, "CUDA/D3D10 simpleD3D10",
WS_OVERLAPPEDWINDOW, 0, 0, g_WindowWidth + 2*xBorder, g_WindowHeight+ 2*yBorder+yMenu,
NULL, NULL, wc.hInstance, NULL);
// Initialize Direct3D
if (SUCCEEDED(InitD3D(hWnd)))
{
// Create the scene geometry
if (SUCCEEDED(InitGeometry()))
{
// Initialize interoperability between CUDA and Direct3D
// Register vertex buffer with CUDA
// DEPRECATED: cudaD3D10RegisterResource(g_pVB, cudaD3D10RegisterFlagsNone);
cudaGraphicsD3D10RegisterResource(&cuda_VB_resource, g_pVB, cudaD3D10RegisterFlagsNone);
getLastCudaError("cudaGraphicsD3D10RegisterResource (g_pVB) failed");
// Initialize vertex buffer with CUDA
runCuda();
// Save result
SaveResult(argc, argv);
// Show the window
ShowWindow(hWnd, SW_SHOWDEFAULT);
UpdateWindow(hWnd);
// Enter the message loop
MSG msg;
ZeroMemory(&msg, sizeof(msg));
while (msg.message!=WM_QUIT)
{
if (PeekMessage(&msg, NULL, 0U, 0U, PM_REMOVE))
{
TranslateMessage(&msg);
DispatchMessage(&msg);
}
else
{
Render();
if (ref_file != NULL)
{
for (int count=0; count<g_iFrameToCompare; count++)
{
Render();
}
const char *cur_image_path = "simpleD3D10.ppm";
// Save a reference of our current test run image
CheckRenderD3D10::ActiveRenderTargetToPPM(g_pd3dDevice,cur_image_path);
// compare to offical reference image, printing PASS or FAIL.
g_bPassed = CheckRenderD3D10::PPMvsPPM(cur_image_path, ref_file, argv[0],MAX_EPSILON, 0.15f);
Cleanup();
PostQuitMessage(0);
}
}
}
}
}
// Release D3D Library (after message loop)
dynlinkUnloadD3D10API();
UnregisterClass(wc.lpszClassName, wc.hInstance);
}
//-----------------------------------------------------------------------------
// Name: Render()
// Desc: Draws the scene
//-----------------------------------------------------------------------------
VOID Render()
{
g_pd3dDevice->RSSetState(g_pRasterState);
// Draw frame
{
static float time = 0.f;
time += 0.001f;
g_pTime->SetFloat(time);
// Clear the Color to a black color
float ClearColor[4] = {0.f, 0.1f, 0.1f, 1.f};
g_pd3dDevice->ClearRenderTargetView(g_color.pBufferRTV, ClearColor);
g_pd3dDevice->OMSetRenderTargets(1, &g_color.pBufferRTV, NULL);
g_pd3dDevice->IASetInputLayout(0);
g_pd3dDevice->IASetPrimitiveTopology(D3D10_PRIMITIVE_TOPOLOGY_TRIANGLESTRIP);
g_pDisplayTechnique->GetPassByIndex(0)->Apply(0);
g_pd3dDevice->DrawInstanced(4, 400, 0, 0);
}
// Run CUDA to compute the histogram
runCuda();
// draw the 2d texture
{
// Clear the Color to a black color
float ClearColor[4] = {0, 0, 0, 1.f};
g_pd3dDevice->ClearRenderTargetView(g_pSwapChainRTV, ClearColor);
g_pd3dDevice->OMSetRenderTargets(1, &g_pSwapChainRTV, NULL);
g_pd3dDevice->IASetInputLayout(0);
g_pd3dDevice->IASetPrimitiveTopology(D3D10_PRIMITIVE_TOPOLOGY_TRIANGLESTRIP);
g_pTexture2D->SetResource(g_color.pBufferSRV);
g_pHistogram->SetResource(g_histogram.pBufferSRV);
g_pUseCase->SetInt(0);
float quadRect1[4] = { -1.0f, -0.8f, 2.0f, 1.8f };
g_pvQuadRect->SetFloatVector((float *) &quadRect1);
g_pCompositeTechnique->GetPassByIndex(0)->Apply(0);
g_pd3dDevice->Draw(4, 0);
g_pUseCase->SetInt(1);
float quadRect2[4] = { -0.8f, -0.99f, 1.6f, 0.19f };
g_pvQuadRect->SetFloatVector((float *) &quadRect2);
g_pCompositeTechnique->GetPassByIndex(0)->Apply(0);
g_pd3dDevice->Draw(4, 0);
g_pTexture2D->SetResource(NULL);
g_pHistogram->SetResource(NULL);
g_pCompositeTechnique->GetPassByIndex(0)->Apply(0);
}
// Present the backbuffer contents to the display
g_pSwapChain->Present(0, 0);
}
void display()
{
static float fps = 0;
frame++;
int time=glutGet(GLUT_ELAPSED_TIME);
if (time - timebase > 1000) {
fps = frame*1000.0f/(time-timebase);
timebase = time;
frame = 0;
}
runCuda();
projection = glm::perspective(fovy, float(width)/float(height), zNear, zFar);
view = camera.getViewMatrix();
projection = projection * view;
char title[100];
sprintf( title, "565 NBody sim [%0.2f fps]", fps );
glutSetWindowTitle(title);
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
#if VISUALIZE == 1
// VAO, shader program, and texture already bound
//glPolygonMode(GL_FRONT_AND_BACK, GL_FILL);
//glDrawElements(GL_TRIANGLES, 6*field_width*field_height, GL_UNSIGNED_INT, 0);
glUseProgram(program[HEIGHT_FIELD]);
GLint location;
if ((location = glGetUniformLocation(program[0], "u_image")) != -1)
{
glUniform1i(location, 0);
}
if ((location = glGetUniformLocation(program[0], "u_projMatrix")) != -1)
{
glUniformMatrix4fv(location, 1, GL_FALSE, &projection[0][0]);
}
if ((location = glGetUniformLocation(program[0], "u_height")) != -1)
{
glUniform1i(location, 0);
}
glEnableVertexAttribArray(positionLocation);
glEnableVertexAttribArray(texcoordsLocation);
glBindBuffer(GL_ARRAY_BUFFER, planeVBO);
glVertexAttribPointer((GLuint)positionLocation, 2, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ARRAY_BUFFER, planeTBO);
glVertexAttribPointer((GLuint)texcoordsLocation, 2, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, planeIBO);
glDrawElements(GL_TRIANGLES, 6*field_width*field_height, GL_UNSIGNED_INT, 0);
glDisableVertexAttribArray(positionLocation);
glDisableVertexAttribArray(texcoordsLocation);
glUseProgram(program[PASS_THROUGH]);
if ((location = glGetUniformLocation(program[1], "u_projMatrix")) != -1)
{
glUniformMatrix4fv(location, 1, GL_FALSE, &projection[0][0]);
}
if ((location = glGetUniformLocation(program[1], "u_cameraPos")) != -1)
{
glUniform3fv(location, 1, &camera.Position[0]);
}
if ((location = glGetUniformLocation(program[1], "sideLen")) != -1)
{
glUniform1i(location, (int)sqrt((float)N_FOR_VIS));
}
if ((location = glGetUniformLocation(program[1], "matrixColoring")) != -1)
{
#if COLOR_MODE == COLOR_MIXED
int colorAsMatrix = 1;
#else
int colorAsMatrix = 0;
#endif
glUniform1i(location, colorAsMatrix);
}
glEnableVertexAttribArray(positionLocation);
glEnableVertexAttribArray(indexLocation);
glBindBuffer(GL_ARRAY_BUFFER, planetVBO);
glVertexAttribPointer((GLuint)positionLocation, 4, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ARRAY_BUFFER, planetInBO);
glVertexAttribIPointer((GLuint)indexLocation, 1, GL_INT, 0, 0);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, planetIBO);
glPointSize(4.0f);
glDrawElements(GL_POINTS, N_FOR_VIS+1, GL_UNSIGNED_INT, 0);
glPointSize(1.0f);
glDisableVertexAttribArray(positionLocation);
glDisableVertexAttribArray(indexLocation);
#endif
glutPostRedisplay();
glutSwapBuffers();
}
void display()
{
static float fps = 0;
frame++;
int time=glutGet(GLUT_ELAPSED_TIME);
if (time - timebase > 1000) {
fps = frame*1000.0f/(time-timebase);
timebase = time;
frame = 0;
}
runCuda();
char title[100];
sprintf( title, "565 NBody sim [%0.2f fps]", fps );
glutSetWindowTitle(title);
glBindBuffer( GL_PIXEL_UNPACK_BUFFER, pbo);
glBindTexture(GL_TEXTURE_2D, displayImage);
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, field_width, field_height,
GL_RGBA, GL_FLOAT, NULL);
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
#if VISUALIZE == 1
// VAO, shader program, and texture already bound
glUseProgram(program[HEIGHT_FIELD]);
glEnableVertexAttribArray(positionLocation);
glEnableVertexAttribArray(texcoordsLocation);
glBindBuffer(GL_ARRAY_BUFFER, planeVBO);
glVertexAttribPointer((GLuint)positionLocation, 2, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ARRAY_BUFFER, planeTBO);
glVertexAttribPointer((GLuint)texcoordsLocation, 2, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, planeIBO);
glDrawElements(GL_TRIANGLES, 6*field_width*field_height, GL_UNSIGNED_INT, 0);
glDisableVertexAttribArray(positionLocation);
glDisableVertexAttribArray(texcoordsLocation);
glUseProgram(program[PASS_THROUGH]);
glEnableVertexAttribArray(positionLocation);
glBindBuffer(GL_ARRAY_BUFFER, planetVBO);
glVertexAttribPointer((GLuint)positionLocation, 4, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, planetIBO);
glPointSize(4.0f);
glDrawElements(GL_POINTS, N_FOR_VIS+1, GL_UNSIGNED_INT, 0);
glPointSize(1.0f);
glDisableVertexAttribArray(positionLocation);
#endif
glutPostRedisplay();
glutSwapBuffers();
}
////////////////////////////////////////////////////////////////////////////////
//! Display callback
////////////////////////////////////////////////////////////////////////////////
void display()
{
// run CUDA kernel to generate vertex positions
if (animate) {
runCuda();
}
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
// set view matrix
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glTranslatef(translateX, translateY, translateZ);
glRotatef(rotateX, 1.0, 0.0, 0.0);
glRotatef(rotateY, 0.0, 1.0, 0.0);
// render from the vbo
glBindBuffer(GL_ARRAY_BUFFER, posVertexBuffer);
glVertexPointer(4, GL_FLOAT, 0, 0);
glEnableClientState(GL_VERTEX_ARRAY);
glBindBuffer(GL_ARRAY_BUFFER, heightVertexBuffer);
glClientActiveTexture(GL_TEXTURE0);
glTexCoordPointer(1, GL_FLOAT, 0, 0);
glEnableClientState(GL_TEXTURE_COORD_ARRAY);
glBindBuffer(GL_ARRAY_BUFFER, slopeVertexBuffer);
glClientActiveTexture(GL_TEXTURE1);
glTexCoordPointer(2, GL_FLOAT, 0, 0);
glEnableClientState(GL_TEXTURE_COORD_ARRAY);
glUseProgram(shaderProg);
// Set default uniform variables parameters for the vertex shader
GLuint uniHeightScale, uniChopiness, uniSize;
uniHeightScale = glGetUniformLocation(shaderProg, "heightScale");
glUniform1f(uniHeightScale, 0.5f);
uniChopiness = glGetUniformLocation(shaderProg, "chopiness");
glUniform1f(uniChopiness, 1.0f);
uniSize = glGetUniformLocation(shaderProg, "size");
glUniform2f(uniSize, meshW, meshH);
// Set default uniform variables parameters for the pixel shader
GLuint uniDeepColor, uniShallowColor, uniSkyColor, uniLightDir;
uniDeepColor = glGetUniformLocation(shaderProg, "deepColor");
glUniform4f(uniDeepColor, 0.0f, 0.0f, 0.1f, 1.0f);
uniShallowColor = glGetUniformLocation(shaderProg, "shallowColor");
glUniform4f(uniShallowColor, 0.1f, 0.4f, 0.3f, 1.0f);
uniSkyColor = glGetUniformLocation(shaderProg, "skyColor");
glUniform4f(uniSkyColor, 0.5f, 0.5f, 0.5f, 1.0f);
uniLightDir = glGetUniformLocation(shaderProg, "lightDir");
glUniform3f(uniLightDir, 0.0f, 1.0f, 0.0f);
// end of uniform settings
glColor3f(1.0, 1.0, 1.0);
if (drawPoints) {
glDrawArrays(GL_POINTS, 0, meshW * meshH);
} else {
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, indexBuffer);
glPolygonMode(GL_FRONT_AND_BACK, wireFrame ? GL_LINE : GL_FILL);
glDrawElements(GL_TRIANGLE_STRIP, ((meshW*2)+2)*(meshH-1), GL_UNSIGNED_INT, 0);
glPolygonMode(GL_FRONT_AND_BACK, GL_FILL);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, 0);
}
glDisableClientState(GL_VERTEX_ARRAY);
glClientActiveTexture(GL_TEXTURE0);
glDisableClientState(GL_TEXTURE_COORD_ARRAY);
glClientActiveTexture(GL_TEXTURE1);
glDisableClientState(GL_TEXTURE_COORD_ARRAY);
glUseProgram(0);
if (g_CheckRender && g_CheckRender->IsQAReadback() && g_Verify) {
printf("> (Frame %d) Readback BackBuffer\n", frameCount);
g_CheckRender->readback( windowW, windowH );
g_CheckRender->savePPM ( sOriginal[0], true, NULL);
if (!g_CheckRender->PPMvsPPM( sOriginal[0], sReference[0], MAX_EPSILON, THRESHOLD)) {
g_TotalErrors++;
}
g_Verify = false;
}
glutSwapBuffers();
computeFPS();
}
////////////////////////////////////////////////////////////////////////////////
//! Run test
////////////////////////////////////////////////////////////////////////////////
void runGraphicsTest(int argc, char** argv)
{
printf("[%s] ", sSDKsample);
if (g_bOpenGLQA) printf("[OpenGL Readback Comparisons] ");
printf("\n");
if ( cutCheckCmdLineFlag(argc, (const char **)argv, "device") ) {
printf("[%s]\n", argv[0]);
printf(" Does not explicitly support -device=n in OpenGL mode\n");
printf(" To use -device=n, the sample must be running w/o OpenGL\n\n");
printf(" > %s -device=n -qatest\n", argv[0]);
printf("exiting...\n");
exit(0);
}
// First initialize OpenGL context, so we can properly set the GL for CUDA.
// This is necessary in order to achieve optimal performance with OpenGL/CUDA interop.
if(CUTFalse == initGL( &argc, argv )) {
cudaThreadExit();
return;
}
cudaGLSetGLDevice( cutGetMaxGflopsDeviceId() );
// create FFT plan
CUFFT_SAFE_CALL(cufftPlan2d(&fftPlan, meshW, meshH, CUFFT_C2R) );
// allocate memory
fftInputW = (meshW / 2)+1;
fftInputH = meshH;
fftInputSize = (fftInputW*fftInputH)*sizeof(float2);
cutilSafeCall(cudaMalloc((void **)&d_h0, fftInputSize) );
cutilSafeCall(cudaMalloc((void **)&d_ht, fftInputSize) );
h_h0 = (float2 *) malloc(fftInputSize);
generate_h0();
cutilSafeCall(cudaMemcpy(d_h0, h_h0, fftInputSize, cudaMemcpyHostToDevice) );
cutilSafeCall(cudaMalloc((void **)&d_slope, meshW*meshH*sizeof(float2)) );
cutCreateTimer(&timer);
cutStartTimer(timer);
prevTime = cutGetTimerValue(timer);
// create vertex buffers and register with CUDA
createVBO(&heightVertexBuffer, meshW*meshH*sizeof(float));
// DEPRECATED: cutilSafeCall(cudaGLRegisterBufferObject(heightVertexBuffer));
cutilSafeCall(cudaGraphicsGLRegisterBuffer(&cuda_heightVB_resource, heightVertexBuffer, cudaGraphicsMapFlagsWriteDiscard));
createVBO(&slopeVertexBuffer, meshW*meshH*sizeof(float2));
// DEPRECATED: cutilSafeCall(cudaGLRegisterBufferObject(slopeVertexBuffer));
cutilSafeCall(cudaGraphicsGLRegisterBuffer(&cuda_slopeVB_resource, slopeVertexBuffer, cudaGraphicsMapFlagsWriteDiscard));
// create vertex and index buffer for mesh
createMeshPositionVBO(&posVertexBuffer, meshW, meshH);
createMeshIndexBuffer(&indexBuffer, meshW, meshH);
// Creating the Auto-Validation Code
if (g_bOpenGLQA) {
g_CheckRender = new CheckBackBuffer(windowH, windowH, 4);
g_CheckRender->setPixelFormat(GL_RGBA);
g_CheckRender->setExecPath(argv[0]);
g_CheckRender->EnableQAReadback(true);
}
runCuda();
// register callbacks
glutDisplayFunc(display);
glutKeyboardFunc(keyboard);
glutMouseFunc(mouse);
glutMotionFunc(motion);
glutReshapeFunc(reshape);
glutIdleFunc(idle);
// start rendering mainloop
glutMainLoop();
cudaThreadExit();
}
