int OCLDeviceManager::init(eDeviceType type) {
bool isCpu = type == CPU;
ocl_args_d_t** oclArgs;
if (isCpu) {
if (ocl_cpu)
return 0;
ocl_cpu = new ocl_args_d_t();
oclArgs = &ocl_cpu;
} else {
if (ocl_gpu)
return 0;
ocl_gpu = new ocl_args_d_t();
oclArgs = &ocl_gpu;
}
data_args_d_t args;
args.preferGpu = !isCpu;
args.preferCpu = isCpu;
args.vendorName = NULL;
int error_code;
error_code = InitOpenCL(*oclArgs, &args);
if (CL_SUCCESS != error_code)
{
LogError("InitOpenCL returned %s.", TranslateOpenCLError(error_code));
delete *oclArgs;
*oclArgs = NULL;;
}
return error_code;
}
int main(int argc, char* argv[])
{
InitOpenCL();
memset(viewMatrix, 0, sizeof(float)*16);
viewMatrix[0] = viewMatrix[5] = viewMatrix[10] = viewMatrix[15] = 1;
SDL_Init(SDL_INIT_EVERYTHING);
Uint32 flags = SDL_WINDOW_OPENGL;
if ( kFullscreen ){
flags |= SDL_WINDOW_FULLSCREEN;
SDL_ShowCursor(0);
}
window = SDL_CreateWindow("RayTracer",
SDL_WINDOWPOS_UNDEFINED,
SDL_WINDOWPOS_UNDEFINED,
kWidth, kHeight,
flags);
SDL_GL_CreateContext(window);
SDL_WarpMouseInWindow(window, kWidth/2.0f, kHeight/2.0f);
glEnable(GL_TEXTURE_2D);
bool loop = true;
int lastTicks = SDL_GetTicks();
while(loop){
int delta = SDL_GetTicks() - lastTicks;
lastTicks = SDL_GetTicks();
SDL_Event e;
while(SDL_PollEvent(&e)){
if ( e.type == SDL_QUIT){
loop = false;
}
else if ( e.type == SDL_KEYDOWN && e.key.keysym.sym == SDLK_ESCAPE){
loop = false;
}
}
delta *= 0.1;
Update(delta);
Render(delta);
std::stringstream ss;
ss << 1000.0f / delta ;
SDL_SetWindowTitle(window, ss.str().c_str());
}
SDL_DestroyWindow(window);
return 0;
}
void OCL_Environment::Initialize()
{
InitOpenCL();
// Add programs
AddProgram(
CreateProgram<VectorAddProgram> (m_context, m_deviceId, "vector_add_kernel.cl")
);
}
Engine::Engine(SETTINGS * settings, float * dc_cuda_avpl_buff, float * dc_cuda_dipang_buff )
{
cuda_avpl_buff = dc_cuda_avpl_buff;
cuda_dipang_buff = dc_cuda_dipang_buff;
m_batchSize = settings->BATCH_SIZE;
m_Scene = new Scene();
//renders = a_renders;
position[0] = position[1] = position[2] = 0.0f;
curBatchRenders = new RenderFromPosition[m_batchSize];
m_cut_AABBs = 0;
m_cutAABBList = 0;
m_cutAABBListSize = 0;
s = settings;
InitOpenCL();
}
LayerSegmentsTreeGPU* SegmentsTreeGPU::CreateNewLayer()
{
countLayer++;
size_t dim = scanData->sizeX * scanData->sizeY * scanData->sizeZ;
LayerSegmentsTreeGPU* oldLayer = GetOldLayer();
int iNumElements0 = oldLayer->segmentCount + 1; // Length of float arrays to process (odd # for illustration)
size_t szGlobalWorkSize0; // 1D var for Total # of work items
size_t szLocalWorkSize0; // 1D var for # of work items in the work group
// set and log Global and Local work size dimensions
szLocalWorkSize0 = 256;
szGlobalWorkSize0 = size_t(ceil(double(iNumElements0) / double (szLocalWorkSize0)) * szLocalWorkSize0); // rounded up to the nearest multiple of the LocalWorkSize
int iNumElements1 = dim + 2; // Length of float arrays to process (odd # for illustration)
size_t szGlobalWorkSize1; // 1D var for Total # of work items
size_t szLocalWorkSize1; // 1D var for # of work items in the work group
// set and log Global and Local work size dimensions
szLocalWorkSize1 = 256;
szGlobalWorkSize1 = size_t(ceil(double(iNumElements1) / double (szLocalWorkSize1)) * szLocalWorkSize1); // rounded up to the nearest multiple of the LocalWorkSize
int iNumElements2 = (dim + 2) * 7; // Length of float arrays to process (odd # for illustration)
size_t szGlobalWorkSize2; // 1D var for Total # of work items
size_t szLocalWorkSize2; // 1D var for # of work items in the work group
// set and log Global and Local work size dimensions
szLocalWorkSize2 = 256;
szGlobalWorkSize2 = size_t(ceil(double(iNumElements2) / double (szLocalWorkSize2)) * szLocalWorkSize2); // rounded up to the nearest multiple of the LocalWorkSize
//создание нового уровня
LayerSegmentsTreeGPU* newLayer = new LayerSegmentsTreeGPU();
oldLayer->up = newLayer;
newLayer->down = oldLayer;
newLayer->indexSegments = new size_t [szGlobalWorkSize1];
newLayer->indexVoxel = new VoxelGPU [dim + 2];
for (size_t i = 0; i < dim; i++)
{
newLayer->indexSegments[i] = oldLayer->indexSegments[i];
newLayer->indexVoxel[i].index = oldLayer->indexVoxel[i].index;
if (oldLayer->indexVoxel[i].next != 0)
newLayer->indexVoxel[i].next = &newLayer->indexVoxel[oldLayer->indexVoxel[i].next->index];
else
newLayer->indexVoxel[i].next = 0;
}
for (size_t i = 0; i < dim; i++)
{
if (oldLayer->indexVoxel[i].next != 0)
if ((newLayer->indexVoxel[i].next->index - oldLayer->indexVoxel[i].next->index) != 0)
int ttttt = 0;
}
size_t* countVoxel = new size_t [oldLayer->segmentCount + 1];
short* weightSegment = new short [szGlobalWorkSize0];
size_t* startIndexVoxel = new size_t [oldLayer->segmentCount + 1];
for(size_t i = 0; i < oldLayer->segmentCount; i++)
{
startIndexVoxel[i] = oldLayer->startIndexVoxel[i];
countVoxel[i] = oldLayer->countVoxel[i];
weightSegment[i] = oldLayer->maxWeight[i];
}
size_t tt = sizeof(short);
size_t uuu = size_t(tt * (dim + 2) * 26 / (1024 * 1024));
short* segmentAdjacents = new short [szGlobalWorkSize2];
for (size_t i = 0; i < ((dim + 2) * 7); i++)
{
segmentAdjacents[i] = maxValue;
}
bool* isNotSegmentVisit = new bool [oldLayer->segmentCount + 2];
for (size_t i = 0; i < oldLayer->segmentCount + 2; i++)
isNotSegmentVisit[i] = true;
//индексы смещения
char relativeIndexMatrix[8][3] = {//i,j,k
{0,0,0},//0
{0,0,1},//1
{0,1,0},//2
{0,1,1},//3
{1,0,0},//4
{1,0,1},//5
{1,1,0},//6
{1,1,1},//7
};
bool f = false;
size_t gi = 0;
InitOpenCL(iNumElements0, szGlobalWorkSize0, szLocalWorkSize0, iNumElements1, szGlobalWorkSize1, szLocalWorkSize1,
iNumElements2, szGlobalWorkSize2, szLocalWorkSize2, scanData->sizeX, scanData->sizeY, scanData->sizeZ);
while ((!f) && (gi < oldLayer->segmentCount))
{
f = true;
//Kernel(scanData->sizeX, scanData->sizeY, scanData->sizeZ, weightSegment, newLayer->indexSegments, segmentAdjacents);
KernelOpenCl(weightSegment, newLayer->indexSegments, segmentAdjacents, iNumElements0, szGlobalWorkSize0, szLocalWorkSize0,
iNumElements1, szGlobalWorkSize1, szLocalWorkSize1,
iNumElements2, szGlobalWorkSize2, szLocalWorkSize2, scanData->sizeX, scanData->sizeY, scanData->sizeZ);
int indexAdjacentsVoxel = 0;
size_t indexCurrentSegment = 0;
size_t indexAdjacentsSegment = 0;
VoxelGPU* currentVoxel = 0;
VoxelGPU newVoxels;
VoxelGPU* newVoxelsEnd = 0;
VoxelGPU* oldVoxel = 0;
VoxelGPU* currentAdjacentsVoxel = 0;
for (size_t t = 0; t < oldLayer->segmentCount; t++)
{
if (isNotSegmentVisit[t])
{
currentVoxel = &newLayer->indexVoxel[startIndexVoxel[t]];
newVoxels.next = 0;
newVoxelsEnd = 0;
oldVoxel = 0;
currentAdjacentsVoxel = 0;
while (currentVoxel != 0)
{
int indexAdjacentsVoxel = 0;
for (size_t k = 1; k < 8; k++)
{
indexAdjacentsVoxel = currentVoxel->index * 7 + k - 1;
size_t siindexAdjacentsVoxel2 = currentVoxel->index + relativeIndexMatrix[k][0] + relativeIndexMatrix[k][1] * scanData->sizeX +
relativeIndexMatrix[k][2] * scanData->sizeX * scanData->sizeY;
if (siindexAdjacentsVoxel2 < dim)
{
indexAdjacentsSegment = oldLayer->indexSegments[siindexAdjacentsVoxel2];
indexCurrentSegment = t;
if (isNotSegmentVisit[indexAdjacentsSegment] && isNotSegmentVisit[indexCurrentSegment])
{
if ((segmentAdjacents[indexAdjacentsVoxel] < step) && (indexAdjacentsSegment != indexCurrentSegment))
{
if (newVoxels.next != 0)
{
newVoxelsEnd->next = &newLayer->indexVoxel[startIndexVoxel[indexAdjacentsSegment]];
}
else
{
newVoxels.next = &newLayer->indexVoxel[startIndexVoxel[indexAdjacentsSegment]];
}
currentAdjacentsVoxel = &newLayer->indexVoxel[startIndexVoxel[indexAdjacentsSegment]];
isNotSegmentVisit[indexAdjacentsSegment] = false;
while (currentAdjacentsVoxel->next != 0)
{
newLayer->indexSegments[currentAdjacentsVoxel->index] = indexCurrentSegment;
currentAdjacentsVoxel = currentAdjacentsVoxel->next;
}
newVoxelsEnd = currentAdjacentsVoxel;
newLayer->indexSegments[currentAdjacentsVoxel->index] = indexCurrentSegment;
weightSegment[indexCurrentSegment] = max(weightSegment[indexCurrentSegment], weightSegment[indexAdjacentsSegment]);
countVoxel[indexCurrentSegment] = countVoxel[indexCurrentSegment] + countVoxel[indexAdjacentsSegment];
}
}
}
}
oldVoxel = currentVoxel;
currentVoxel = currentVoxel->next;
}
if (newVoxels.next != 0)
f = false;
oldVoxel->next = newVoxels.next;
size_t yyy = 0;
currentVoxel = &newLayer->indexVoxel[startIndexVoxel[t]];
while (currentVoxel != 0)
{
yyy++;
currentVoxel = currentVoxel->next;
}
if (yyy != countVoxel[t])
int ggg = 0;
}
}
gi++;
}
newLayer->segmentCount = 0;
for (size_t i = 0; i < oldLayer->segmentCount; i++)
{
if (isNotSegmentVisit[i])
{
newLayer->segmentCount++;
}
}
newLayer->countVoxel = new size_t [newLayer->segmentCount + 2];
newLayer->startIndexVoxel = new size_t [newLayer->segmentCount + 2];
newLayer->maxWeight = new short [newLayer->segmentCount + 2];
newLayer->minWeight = new short [newLayer->segmentCount + 2];
size_t ind = 0;
VoxelGPU* voxel = 0;
newLayer->maxCapacity = 0;
size_t count = 0;
size_t c2 = 0;
short minw = 0;
for (size_t i = 0; i < oldLayer->segmentCount; i++)
{
minw = 0;
if (isNotSegmentVisit[i])
{
voxel = &newLayer->indexVoxel[startIndexVoxel[i]];
while (voxel != 0)
{
newLayer->indexSegments[voxel->index] = ind;
count++;
if (minw > scanData->data[voxel->index])
minw = scanData->data[voxel->index];
voxel = voxel->next;
}
newLayer->countVoxel[ind] = countVoxel[i];
c2 += countVoxel[i];
newLayer->startIndexVoxel[ind] = startIndexVoxel[i];
newLayer->maxWeight[ind] = weightSegment[i];
newLayer->minWeight[ind] = minw;
newLayer->maxCapacity = max (newLayer->maxCapacity, newLayer->countVoxel[ind]);
ind++;
}
}
if (count != dim)
int ggg = 0;
ReleaseOpenCL();
// Free host memory
delete [] countVoxel;
delete [] weightSegment;
delete [] startIndexVoxel;
delete [] isNotSegmentVisit;
delete [] segmentAdjacents;
newLayer->segmentsTree = this;
return newLayer;
}
void TestCamera()
{
InitOpenCL();
//TCHAR szDeviceName[80];
//TCHAR szDeviceVersion[80];
//for (int wIndex = 0; wIndex < 10; wIndex++)
//{
// if (capGetDriverDescription(
// wIndex,
// szDeviceName,
// sizeof (szDeviceName),
// szDeviceVersion,
// sizeof (szDeviceVersion)
// ))
// {
// // Append name to list of installed capture drivers
// // and then let the user select a driver to use.
// }
//}
//HWND hWndC = capCreateCaptureWindow(TEXT("PunkCapture"),
// WS_CHILD | WS_VISIBLE, 0, 0, 160, 120, *System::Window::Instance(), 1);
//SendMessage (hWndC, WM_CAP_DRIVER_CONNECT, 0, 0L);
////
//// Or, use the macro to connect to the MSVIDEO driver:
//// fOK = capDriverConnect(hWndC, 0);
////
//// Place code to set up and capture video here.
////
////capDriverDisconnect (hWndC);
//CAPDRIVERCAPS CapDriverCaps = { };
//CAPSTATUS CapStatus = { };
//capDriverGetCaps(hWndC, &CapDriverCaps, sizeof(CAPDRIVERCAPS));
//// Video source dialog box.
//if (CapDriverCaps.fHasDlgVideoSource)
//{
// capDlgVideoSource(hWndC);
//}
//// Video format dialog box.
//if (CapDriverCaps.fHasDlgVideoFormat)
//{
// capDlgVideoFormat(hWndC);
// // Are there new image dimensions?
// capGetStatus(hWndC, &CapStatus, sizeof (CAPSTATUS));
// // If so, notify the parent of a size change.
//}
//// Video display dialog box.
//if (CapDriverCaps.fHasDlgVideoDisplay)
//{
// capDlgVideoDisplay(hWndC);
//}
HRESULT hr;
IGraphBuilder* graph= 0; hr = CoCreateInstance( CLSID_FilterGraph, 0, CLSCTX_INPROC,IID_IGraphBuilder, (void **)&graph );
IMediaControl* ctrl = 0; hr = graph->QueryInterface( IID_IMediaControl, (void **)&ctrl );
ICreateDevEnum* devs = 0; hr = CoCreateInstance (CLSID_SystemDeviceEnum, 0, CLSCTX_INPROC, IID_ICreateDevEnum, (void **) &devs);
IEnumMoniker* cams = 0; hr = devs?devs->CreateClassEnumerator (CLSID_VideoInputDeviceCategory, &cams, 0):0;
IMoniker* mon = 0; hr = cams->Next (1,&mon,0); // get first found capture device (webcam?)
IBaseFilter* cam = 0; hr = mon->BindToObject(0,0,IID_IBaseFilter, (void**)&cam);
hr = graph->AddFilter(cam, L"Capture Source"); // add web cam to graph as source
IEnumPins* pins = 0; hr = cam?cam->EnumPins(&pins):0; // we need output pin to autogenerate rest of the graph
IPin* pin = 0; hr = pins?pins->Next(1,&pin, 0):0; // via graph->Render
hr = graph->Render(pin); // graph builder now builds whole filter chain including MJPG decompression on some webcams
IEnumFilters* fil = 0; hr = graph->EnumFilters(&fil); // from all newly added filters
IBaseFilter* rnd = 0; hr = fil->Next(1,&rnd,0); // we find last one (renderer)
hr = rnd->EnumPins(&pins); // because data we are intersted in are pumped to renderers input pin
hr = pins->Next(1,&pin, 0); // via Receive member of IMemInputPin interface
IMemInputPin* mem = 0; hr = pin->QueryInterface(IID_IMemInputPin,(void**)&mem);
DsHook(mem,6,Receive); // so we redirect it to our own proc to grab image data
hr = ctrl->Run();
};