mirror of
https://github.com/NVIDIA/cuda-samples.git
synced 2026-10-23 21:28:17 +08:00
add and update samples for CUDA 11.5
This commit is contained in:
Rutwik Choughule
49
Samples/simpleD3D10RenderTarget/README.md
Normal file
49
Samples/simpleD3D10RenderTarget/README.md
Normal file
@@ -0,0 +1,49 @@
|
||||
# simpleD3D10RenderTarget - Simple Direct3D10 Render Target
|
||||
|
||||
## Description
|
||||
|
||||
Simple program which demonstrates interop of rendertargets between Direct3D10 and CUDA. The program uses RenderTarget positions with CUDA and generates a histogram with visualization. A Direct3D10 Capable device is required.
|
||||
|
||||
## Key Concepts
|
||||
|
||||
Graphics Interop, Texture
|
||||
|
||||
## Supported SM Architectures
|
||||
|
||||
[SM 3.5 ](https://developer.nvidia.com/cuda-gpus) [SM 3.7 ](https://developer.nvidia.com/cuda-gpus) [SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus)
|
||||
|
||||
## Supported OSes
|
||||
|
||||
Windows
|
||||
|
||||
## Supported CPU Architecture
|
||||
|
||||
x86_64
|
||||
|
||||
## CUDA APIs involved
|
||||
|
||||
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
|
||||
cudaD3D10GetDevice, cudaD3D10SetDirect3DDevice, cudaGraphicsD3D10RegisterResource, cudaGraphicsResourceSetMapFlags, cudaGraphicsSubResourceGetMappedArray, cudaMemcpy2DToArray, cudaGraphicsUnregisterResource
|
||||
|
||||
## Dependencies needed to build/run
|
||||
[DirectX](../../README.md#directx)
|
||||
|
||||
## Prerequisites
|
||||
|
||||
Download and install the [CUDA Toolkit 11.5](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||
|
||||
## Build and Run
|
||||
|
||||
### Windows
|
||||
The Windows samples are built using the Visual Studio IDE. Solution files (.sln) are provided for each supported version of Visual Studio, using the format:
|
||||
```
|
||||
*_vs<version>.sln - for Visual Studio <version>
|
||||
```
|
||||
Each individual sample has its own set of solution files in its directory:
|
||||
|
||||
To build/examine all the samples at once, the complete solution files should be used. To build/examine a single sample, the individual sample solution files should be used.
|
||||
> **Note:** Some samples require that the Microsoft DirectX SDK (June 2010 or newer) be installed and that the VC++ directory paths are properly set up (**Tools > Options...**). Check DirectX Dependencies section for details."
|
||||
|
||||
## References (for more details)
|
||||
|
||||
Binary file not shown.
850
Samples/simpleD3D10RenderTarget/simpleD3D10RenderTarget.cpp
Normal file
850
Samples/simpleD3D10RenderTarget/simpleD3D10RenderTarget.cpp
Normal file
@@ -0,0 +1,850 @@
|
||||
/* Copyright (c) 2021, NVIDIA 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 NVIDIA 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 ``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.
|
||||
*/
|
||||
|
||||
/* This example demonstrates how to use the CUDA Direct3D bindings to fill
|
||||
* a vertex buffer with CUDA and use Direct3D to render the data.
|
||||
* Host code.
|
||||
*/
|
||||
|
||||
#pragma warning(disable : 4312)
|
||||
|
||||
#include <windows.h>
|
||||
#include <mmsystem.h>
|
||||
|
||||
// this header inclues all the necessary D3D10 includes
|
||||
#include <dynlink_d3d10.h>
|
||||
#include <cuda_runtime_api.h>
|
||||
#include <cuda_d3d10_interop.h>
|
||||
|
||||
// includes, project
|
||||
#include <rendercheck_d3d10.h>
|
||||
#include <helper_cuda.h>
|
||||
#include <helper_functions.h>
|
||||
|
||||
int g_iFrameToCompare = 10;
|
||||
|
||||
bool g_bDone = false;
|
||||
bool g_bPassed = true;
|
||||
|
||||
int *pArgc = NULL;
|
||||
char **pArgv = NULL;
|
||||
|
||||
#define MAX_EPSILON 10
|
||||
|
||||
static char *SDK_name = "simpleD3D10RenderTarget";
|
||||
|
||||
//-----------------------------------------------------------------------------
|
||||
// Global variables
|
||||
//-----------------------------------------------------------------------------
|
||||
IDXGIAdapter *g_pCudaCapableAdapter = NULL; // Adapter to use
|
||||
ID3D10Device *g_pd3dDevice = NULL; // Our rendering device
|
||||
IDXGISwapChain *g_pSwapChain = NULL; // The swap chain of the window
|
||||
ID3D10RenderTargetView *g_pSwapChainRTV =
|
||||
NULL; // The Render target view on the swap chain ( used for clear)
|
||||
ID3D10RasterizerState *g_pRasterState = NULL;
|
||||
|
||||
struct Color {
|
||||
ID3D10Texture2D *pBuffer; // The color buffer
|
||||
ID3D10RenderTargetView
|
||||
*pBufferRTV; // The Render target view on the color buffer
|
||||
ID3D10ShaderResourceView
|
||||
*pBufferSRV; // The shader resource view on the color buffer
|
||||
cudaGraphicsResource *cudaResource; // resource of the Buffer on cuda side
|
||||
int pitch;
|
||||
cudaArray *pCudaArray; // the data in a cuda view
|
||||
} g_color;
|
||||
|
||||
struct Histogram {
|
||||
ID3D10Buffer *pBuffer; // Buffer to hold histogram
|
||||
ID3D10ShaderResourceView *pBufferSRV; // View on the histogram buffer
|
||||
cudaGraphicsResource *cudaResource; // resource of the Buffer on cuda side
|
||||
unsigned int *cudaBuffer; // staging buffer to allow cuda to write results
|
||||
// cudaArray* pCudaArray; // the data in a cuda view
|
||||
size_t size;
|
||||
} g_histogram;
|
||||
|
||||
ID3D10Effect *g_pDisplayEffect = NULL;
|
||||
ID3D10EffectTechnique *g_pDisplayTechnique = NULL;
|
||||
ID3D10EffectScalarVariable *g_pTime = NULL;
|
||||
|
||||
static const char g_displayEffectSrc[] =
|
||||
"float g_Time; \n"
|
||||
"uint2 g_vGrid = uint2(20,20); \n"
|
||||
"float4 g_vGridSize = float4(0.05f, 0.05f, 0.046f, 0.046f); \n"
|
||||
"\n"
|
||||
"struct Fragment{ \n"
|
||||
" float4 Pos : SV_POSITION;\n"
|
||||
" float2 Tex : TEXCOORD0; \n"
|
||||
" float4 Col : TEXCOORD1; };\n"
|
||||
"\n"
|
||||
"Fragment VS( uint instanceId : SV_InstanceID, uint vertexId : SV_VertexID "
|
||||
")\n"
|
||||
"{\n"
|
||||
" Fragment f;\n"
|
||||
" f.Tex = float2( 1.f*((vertexId == 1) || (vertexId == 3)), 1.f*( "
|
||||
"vertexId >= 2)); \n"
|
||||
" \n"
|
||||
" uint2 cellId = uint2(instanceId % g_vGrid.x, instanceId / "
|
||||
"g_vGrid.x);\n"
|
||||
" f.Pos = float4( g_vGridSize.xy*cellId + 0.5f*(g_vGridSize.xy - "
|
||||
"g_vGridSize.zw) + f.Tex * g_vGridSize.zw, 0, 1);\n"
|
||||
" f.Pos.xy = (f.Pos.xy*2.f - 1.f);\n"
|
||||
" \n"
|
||||
" f.Col = float4( ((g_vGrid.x-1.f) - cellId.x) / (g_vGrid.x-1.f), "
|
||||
"(cellId.x + (g_vGrid.y-1.f) - cellId.y) / (g_vGrid.x+g_vGrid.y-1.f), "
|
||||
"cellId.y / (g_vGrid.y-1.f), 1.f);\n"
|
||||
" f.Col *= float4( 0.5 + 0.5*sin(g_Time), 0.5 + "
|
||||
"0.5*sin(g_Time)*cos(g_Time), 0.5 + 0.5*cos(g_Time), 1.f);\n"
|
||||
" return f;\n"
|
||||
"}\n"
|
||||
"\n"
|
||||
"float4 PS( Fragment f ) : SV_Target\n"
|
||||
"{\n"
|
||||
" return f.Col;\n"
|
||||
"}\n"
|
||||
"\n"
|
||||
"technique10 Render\n"
|
||||
"{\n"
|
||||
" pass P0\n"
|
||||
" {\n"
|
||||
" SetVertexShader( CompileShader( vs_4_0, VS() ) );\n"
|
||||
" SetGeometryShader( NULL );\n"
|
||||
" SetPixelShader( CompileShader( ps_4_0, PS() ) );\n"
|
||||
" }\n"
|
||||
"}\n"
|
||||
"\n";
|
||||
|
||||
ID3D10Effect *g_pCompositeEffect = NULL;
|
||||
ID3D10EffectTechnique *g_pCompositeTechnique = NULL;
|
||||
ID3D10EffectVectorVariable *g_pvQuadRect = NULL;
|
||||
ID3D10EffectScalarVariable *g_pUseCase = NULL;
|
||||
ID3D10EffectShaderResourceVariable *g_pTexture2D = NULL;
|
||||
ID3D10EffectShaderResourceVariable *g_pHistogram = NULL;
|
||||
|
||||
static const char g_compositeEffectSrc[] =
|
||||
"float4 g_vQuadRect; \n"
|
||||
"int g_UseCase; \n"
|
||||
"Texture2D g_Texture2D; \n"
|
||||
"Buffer<uint> g_Histogram; \n"
|
||||
"\n"
|
||||
"SamplerState samLinear{ \n"
|
||||
" Filter = MIN_MAG_LINEAR_MIP_POINT; \n"
|
||||
"};\n"
|
||||
"\n"
|
||||
"struct Fragment{ \n"
|
||||
" float4 Pos : SV_POSITION;\n"
|
||||
" float3 Tex : TEXCOORD0; \n"
|
||||
" float2 uv : TEXCOORD1; };\n"
|
||||
"\n"
|
||||
"Fragment VS( uint vertexId : SV_VertexID )\n"
|
||||
"{\n"
|
||||
" Fragment f;\n"
|
||||
" f.Tex = float3( 1.f*((vertexId == 1) || (vertexId == 3)), 1.f*( "
|
||||
"vertexId >= 2), 0.f); \n"
|
||||
" \n"
|
||||
" f.Pos = float4( g_vQuadRect.xy + f.Tex * g_vQuadRect.zw, 0, 1);\n"
|
||||
" \n"
|
||||
" f.uv = float2( f.Tex.x*255.f, f.Tex.y*50000.f ); \n"
|
||||
" return f;\n"
|
||||
"}\n"
|
||||
"\n"
|
||||
"float4 PS( Fragment f ) : SV_Target\n"
|
||||
"{\n"
|
||||
" if (g_UseCase == 0) \n"
|
||||
" return g_Texture2D.Sample( samLinear, f.Tex.xy ); \n"
|
||||
" else if (g_UseCase == 1) { \n"
|
||||
" uint index = f.uv.x; \n"
|
||||
" float value = g_Histogram.Load( index ); \n"
|
||||
" //float value = index * 1000; \n"
|
||||
" float red = ( value >= f.uv.y ? (0.5f * f.uv.y / value) + 0.5f : "
|
||||
"0.f ); \n"
|
||||
" return float4(red, 0, 0, 1); \n"
|
||||
" } else return float4(f.Tex, 1);\n"
|
||||
"}\n"
|
||||
"\n"
|
||||
"technique10 Render\n"
|
||||
"{\n"
|
||||
" pass P0\n"
|
||||
" {\n"
|
||||
" SetVertexShader( CompileShader( vs_4_0, VS() ) );\n"
|
||||
" SetGeometryShader( NULL );\n"
|
||||
" SetPixelShader( CompileShader( ps_4_0, PS() ) );\n"
|
||||
" }\n"
|
||||
"}\n"
|
||||
"\n";
|
||||
|
||||
// testing/tracing function used pervasively in tests. if the condition is
|
||||
// unsatisfied then spew and fail the function immediately (doing no cleanup)
|
||||
#define AssertOrQuit(x) \
|
||||
if (!(x)) { \
|
||||
fprintf(stdout, "Assert unsatisfied in %s at %s:%d\n", __FUNCTION__, \
|
||||
__FILE__, __LINE__); \
|
||||
return 1; \
|
||||
}
|
||||
|
||||
const unsigned int g_WindowWidth = 800;
|
||||
const unsigned int g_WindowHeight = 800;
|
||||
const unsigned int g_HistogramSize = 256;
|
||||
|
||||
//-----------------------------------------------------------------------------
|
||||
// Forward declarations
|
||||
//-----------------------------------------------------------------------------
|
||||
void runTest(int argc, char **argv, char *ref_file);
|
||||
void runCuda();
|
||||
HRESULT InitD3D(HWND hWnd);
|
||||
VOID Cleanup();
|
||||
VOID Render();
|
||||
LRESULT WINAPI MsgProc(HWND hWnd, UINT msg, WPARAM wParam, LPARAM lParam);
|
||||
|
||||
// CUDA/D3D10 kernels
|
||||
extern "C" void checkCudaError();
|
||||
extern "C" void createHistogramTex(unsigned int *histogram, unsigned int width,
|
||||
unsigned int height, cudaArray *colorArray);
|
||||
|
||||
#define NAME_LEN 512
|
||||
|
||||
bool findCUDADevice() {
|
||||
int nGraphicsGPU = 0;
|
||||
int deviceCount = 0;
|
||||
bool bFoundGraphics = false;
|
||||
char devname[NAME_LEN];
|
||||
|
||||
// This function call returns 0 if there are no CUDA capable devices.
|
||||
cudaError_t error_id = cudaGetDeviceCount(&deviceCount);
|
||||
|
||||
if (error_id != cudaSuccess) {
|
||||
printf("cudaGetDeviceCount returned %d\n-> %s\n", (int)error_id,
|
||||
cudaGetErrorString(error_id));
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
if (deviceCount == 0) {
|
||||
printf("> There are no device(s) supporting CUDA\n");
|
||||
return false;
|
||||
} else {
|
||||
printf("> Found %d CUDA Capable Device(s)\n", deviceCount);
|
||||
}
|
||||
|
||||
// Get CUDA device properties
|
||||
cudaDeviceProp deviceProp;
|
||||
|
||||
for (int dev = 0; dev < deviceCount; ++dev) {
|
||||
cudaGetDeviceProperties(&deviceProp, dev);
|
||||
STRCPY(devname, NAME_LEN, deviceProp.name);
|
||||
printf("> GPU %d: %s\n", dev, devname);
|
||||
}
|
||||
|
||||
return true;
|
||||
}
|
||||
|
||||
bool findDXDevice(char *dev_name) {
|
||||
HRESULT hr = S_OK;
|
||||
cudaError cuStatus;
|
||||
|
||||
// Iterate through the candidate adapters
|
||||
IDXGIFactory *pFactory;
|
||||
hr = sFnPtr_CreateDXGIFactory(__uuidof(IDXGIFactory), (void **)(&pFactory));
|
||||
|
||||
if (!SUCCEEDED(hr)) {
|
||||
printf("> No DXGI Factory created.\n");
|
||||
return false;
|
||||
}
|
||||
|
||||
UINT adapter = 0;
|
||||
|
||||
for (; !g_pCudaCapableAdapter; ++adapter) {
|
||||
// Get a candidate DXGI adapter
|
||||
IDXGIAdapter *pAdapter = NULL;
|
||||
hr = pFactory->EnumAdapters(adapter, &pAdapter);
|
||||
|
||||
if (FAILED(hr)) {
|
||||
break; // no compatible adapters found
|
||||
}
|
||||
|
||||
// Query to see if there exists a corresponding compute device
|
||||
int cuDevice;
|
||||
cuStatus = cudaD3D10GetDevice(&cuDevice, pAdapter);
|
||||
printLastCudaError("cudaD3D10GetDevice failed"); // This prints and resets
|
||||
// the cudaError to
|
||||
// cudaSuccess
|
||||
|
||||
if (cudaSuccess == cuStatus) {
|
||||
// If so, mark it as the one against which to create our d3d10 device
|
||||
g_pCudaCapableAdapter = pAdapter;
|
||||
g_pCudaCapableAdapter->AddRef();
|
||||
}
|
||||
|
||||
pAdapter->Release();
|
||||
}
|
||||
|
||||
printf("> Found %d D3D10 Adapater(s).\n", (int)adapter);
|
||||
|
||||
pFactory->Release();
|
||||
|
||||
if (!g_pCudaCapableAdapter) {
|
||||
printf("> Found 0 D3D10 Adapater(s) /w Compute capability.\n");
|
||||
return false;
|
||||
}
|
||||
|
||||
DXGI_ADAPTER_DESC adapterDesc;
|
||||
g_pCudaCapableAdapter->GetDesc(&adapterDesc);
|
||||
wcstombs(dev_name, adapterDesc.Description, 128);
|
||||
|
||||
printf("> Found 1 D3D10 Adapater(s) /w Compute capability.\n");
|
||||
printf("> %s\n", dev_name);
|
||||
|
||||
return true;
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Program main
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
int main(int argc, char **argv) {
|
||||
char device_name[256];
|
||||
char *ref_file = NULL;
|
||||
|
||||
pArgc = &argc;
|
||||
pArgv = argv;
|
||||
|
||||
printf("[%s] - Starting...\n", SDK_name);
|
||||
|
||||
if (!findCUDADevice()) // Search for CUDA GPU
|
||||
{
|
||||
printf("> CUDA Device NOT found on \"%s\".. Exiting.\n", device_name);
|
||||
exit(EXIT_SUCCESS);
|
||||
}
|
||||
|
||||
if (!dynlinkLoadD3D10API()) // Search for D3D API (locate drivers, does not
|
||||
// mean device is found)
|
||||
{
|
||||
printf("> D3D10 API libraries NOT found.. Exiting.\n");
|
||||
dynlinkUnloadD3D10API();
|
||||
exit(EXIT_SUCCESS);
|
||||
}
|
||||
|
||||
if (!findDXDevice(device_name)) // Search for D3D Hardware Device
|
||||
{
|
||||
printf("> D3D10 Graphics Device NOT found.. Exiting.\n");
|
||||
dynlinkUnloadD3D10API();
|
||||
exit(EXIT_SUCCESS);
|
||||
}
|
||||
|
||||
// command line options
|
||||
if (argc > 1) {
|
||||
// automatied build testing harness
|
||||
if (checkCmdLineFlag(argc, (const char **)argv, "file"))
|
||||
getCmdLineArgumentString(argc, (const char **)argv, "file", &ref_file);
|
||||
}
|
||||
|
||||
// run D3D10/CUDA test
|
||||
runTest(argc, argv, ref_file);
|
||||
|
||||
//
|
||||
// and exit
|
||||
//
|
||||
printf("%s running on %s exiting...\n", SDK_name, device_name);
|
||||
|
||||
exit(g_bPassed ? EXIT_SUCCESS : EXIT_FAILURE);
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
//! 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 SDK",
|
||||
NULL};
|
||||
RegisterClassEx(&wc);
|
||||
|
||||
// Create the application's window
|
||||
HWND hWnd = CreateWindow(wc.lpszClassName, "CUDA/D3D10 RenderTarget InterOP",
|
||||
WS_OVERLAPPEDWINDOW, 100, 100, g_WindowWidth,
|
||||
g_WindowHeight, NULL, NULL, wc.hInstance, NULL);
|
||||
|
||||
// Initialize Direct3D
|
||||
if (SUCCEEDED(InitD3D(hWnd))) {
|
||||
// Initialize interoperability between CUDA and Direct3D
|
||||
// Register vertex buffer with CUDA
|
||||
cudaGraphicsD3D10RegisterResource(&g_histogram.cudaResource,
|
||||
g_histogram.pBuffer,
|
||||
cudaGraphicsMapFlagsNone);
|
||||
getLastCudaError("cudaGraphicsD3D10RegisterResource (g_pHistogram) failed");
|
||||
|
||||
// Register color buffer with CUDA
|
||||
cudaGraphicsD3D10RegisterResource(&g_color.cudaResource, g_color.pBuffer,
|
||||
cudaGraphicsMapFlagsNone);
|
||||
getLastCudaError(
|
||||
"cudaGraphicsD3D10RegisterResource (g_color.pBuffer) failed");
|
||||
|
||||
// Show the window
|
||||
ShowWindow(hWnd, SW_SHOWDEFAULT);
|
||||
UpdateWindow(hWnd);
|
||||
}
|
||||
|
||||
//
|
||||
// The main loop
|
||||
//
|
||||
while (false == g_bDone) {
|
||||
Render();
|
||||
|
||||
//
|
||||
// handle I/O
|
||||
//
|
||||
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) {
|
||||
for (int count = 0; count < g_iFrameToCompare; count++) {
|
||||
Render();
|
||||
}
|
||||
|
||||
const char *cur_image_path = "simpleD3D10RenderTarget.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);
|
||||
|
||||
g_bDone = true;
|
||||
|
||||
Cleanup();
|
||||
|
||||
PostQuitMessage(0);
|
||||
} else {
|
||||
g_bPassed = true;
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// Release D3D Library (after message loop)
|
||||
dynlinkUnloadD3D10API();
|
||||
|
||||
// Unregister windows class
|
||||
UnregisterClass(wc.lpszClassName, wc.hInstance);
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
//! Run the Cuda part of the computation
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
void runCuda() {
|
||||
cudaStream_t stream = 0;
|
||||
const int nbResources = 2;
|
||||
cudaGraphicsResource *ppResources[nbResources] = {
|
||||
g_histogram.cudaResource, g_color.cudaResource,
|
||||
};
|
||||
// Map resources for Cuda
|
||||
checkCudaErrors(cudaGraphicsMapResources(nbResources, ppResources, stream));
|
||||
getLastCudaError("cudaGraphicsMapResources(2) failed");
|
||||
// Get pointers
|
||||
checkCudaErrors(cudaGraphicsResourceGetMappedPointer(
|
||||
(void **)&g_histogram.cudaBuffer, &g_histogram.size,
|
||||
g_histogram.cudaResource));
|
||||
getLastCudaError(
|
||||
"cudaGraphicsResourceGetMappedPointer (g_color.pBuffer) failed");
|
||||
cudaGraphicsSubResourceGetMappedArray(&g_color.pCudaArray,
|
||||
g_color.cudaResource, 0, 0);
|
||||
getLastCudaError(
|
||||
"cudaGraphicsSubResourceGetMappedArray (g_color.pBuffer) failed");
|
||||
|
||||
// Execute kernel
|
||||
createHistogramTex(g_histogram.cudaBuffer, g_WindowWidth, g_WindowHeight,
|
||||
g_color.pCudaArray);
|
||||
checkCudaError();
|
||||
//
|
||||
// unmap the resources
|
||||
//
|
||||
checkCudaErrors(cudaGraphicsUnmapResources(nbResources, ppResources, stream));
|
||||
getLastCudaError("cudaGraphicsUnmapResources(2) failed");
|
||||
}
|
||||
|
||||
//-----------------------------------------------------------------------------
|
||||
// Name: InitD3D()
|
||||
// Desc: Initializes Direct3D
|
||||
//-----------------------------------------------------------------------------
|
||||
HRESULT InitD3D(HWND hWnd) {
|
||||
// Set up the structure used to create the device and swapchain
|
||||
DXGI_SWAP_CHAIN_DESC sd;
|
||||
ZeroMemory(&sd, sizeof(sd));
|
||||
sd.BufferCount = 1;
|
||||
sd.BufferDesc.Width = g_WindowWidth;
|
||||
sd.BufferDesc.Height = g_WindowHeight;
|
||||
sd.BufferDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
|
||||
sd.BufferDesc.RefreshRate.Numerator = 60;
|
||||
sd.BufferDesc.RefreshRate.Denominator = 1;
|
||||
sd.BufferUsage = DXGI_USAGE_RENDER_TARGET_OUTPUT;
|
||||
sd.OutputWindow = hWnd;
|
||||
sd.SampleDesc.Count = 1;
|
||||
sd.SampleDesc.Quality = 0;
|
||||
sd.Windowed = TRUE;
|
||||
|
||||
// Create device and swapchain
|
||||
HRESULT hr = sFnPtr_D3D10CreateDeviceAndSwapChain(
|
||||
g_pCudaCapableAdapter, D3D10_DRIVER_TYPE_HARDWARE, NULL,
|
||||
0, // D3D10_CREATE_DEVICE_DEBUG,
|
||||
D3D10_SDK_VERSION, &sd, &g_pSwapChain, &g_pd3dDevice);
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
|
||||
g_pCudaCapableAdapter->Release();
|
||||
|
||||
// Create a render target view of the swapchain
|
||||
ID3D10Texture2D *pBuffer;
|
||||
hr =
|
||||
g_pSwapChain->GetBuffer(0, __uuidof(ID3D10Texture2D), (LPVOID *)&pBuffer);
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
|
||||
hr = g_pd3dDevice->CreateRenderTargetView(pBuffer, NULL, &g_pSwapChainRTV);
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
pBuffer->Release();
|
||||
|
||||
// Create a color buffer, corresponding render target view and shader resource
|
||||
// view
|
||||
D3D10_TEXTURE2D_DESC tex2Ddesc;
|
||||
ZeroMemory(&tex2Ddesc, sizeof(D3D10_TEXTURE2D_DESC));
|
||||
tex2Ddesc.Width = g_WindowWidth;
|
||||
tex2Ddesc.Height = g_WindowHeight;
|
||||
tex2Ddesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
|
||||
tex2Ddesc.MipLevels = 1;
|
||||
tex2Ddesc.ArraySize = 1;
|
||||
tex2Ddesc.SampleDesc.Count = 1;
|
||||
tex2Ddesc.Usage = D3D10_USAGE_DEFAULT;
|
||||
tex2Ddesc.BindFlags = D3D10_BIND_RENDER_TARGET | D3D10_BIND_SHADER_RESOURCE;
|
||||
|
||||
hr = g_pd3dDevice->CreateTexture2D(&tex2Ddesc, NULL, &g_color.pBuffer);
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
|
||||
hr = g_pd3dDevice->CreateShaderResourceView(g_color.pBuffer, NULL,
|
||||
&g_color.pBufferSRV);
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
|
||||
hr = g_pd3dDevice->CreateRenderTargetView(g_color.pBuffer, NULL,
|
||||
&g_color.pBufferRTV);
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
|
||||
// Create a buffer which will contain the resulting histogram and the SRV to
|
||||
// plug it
|
||||
D3D10_BUFFER_DESC bufferDesc;
|
||||
bufferDesc.Usage = D3D10_USAGE_DEFAULT;
|
||||
// NOTE: allocation of more than what is needed to display in the shader
|
||||
// but this 64 factor is required for CUDA to work with this buffer (see
|
||||
// BLOCK_N in .cu code...)
|
||||
bufferDesc.ByteWidth =
|
||||
sizeof(unsigned int) * g_HistogramSize * 64 /*BLOCK_N*/;
|
||||
bufferDesc.BindFlags = D3D10_BIND_SHADER_RESOURCE;
|
||||
bufferDesc.CPUAccessFlags = 0;
|
||||
bufferDesc.MiscFlags = 0;
|
||||
// useless values... we could remove this...
|
||||
unsigned int values[256 * 64];
|
||||
|
||||
for (int i = 0; i < 256 * 64; i++) {
|
||||
values[i] = i;
|
||||
}
|
||||
|
||||
D3D10_SUBRESOURCE_DATA data;
|
||||
data.pSysMem = values;
|
||||
data.SysMemPitch = 0;
|
||||
data.SysMemSlicePitch = 0;
|
||||
|
||||
hr = g_pd3dDevice->CreateBuffer(&bufferDesc, &data, &g_histogram.pBuffer);
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
|
||||
D3D10_SHADER_RESOURCE_VIEW_DESC bufferSRVDesc;
|
||||
bufferSRVDesc.Format = DXGI_FORMAT_R32_UINT;
|
||||
bufferSRVDesc.ViewDimension = D3D10_SRV_DIMENSION_BUFFER;
|
||||
bufferSRVDesc.Buffer.ElementOffset = 0;
|
||||
bufferSRVDesc.Buffer.ElementWidth =
|
||||
g_HistogramSize; // 4*sizeof(unsigned int);
|
||||
|
||||
hr = g_pd3dDevice->CreateShaderResourceView(
|
||||
g_histogram.pBuffer, &bufferSRVDesc, &g_histogram.pBufferSRV);
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
// Create the equivalent as a cuda staging buffer that we'll use to write from
|
||||
// Cuda. Then we'll copy it to the texture
|
||||
// cudaMalloc(g_histogram.cudaBuffer, sizeof(float) * g_HistogramSize;
|
||||
// getLastCudaError("cudaMallocPitch (g_histogram) failed");
|
||||
|
||||
// Setup the viewport
|
||||
D3D10_VIEWPORT vp;
|
||||
vp.Width = g_WindowWidth;
|
||||
vp.Height = g_WindowHeight;
|
||||
vp.MinDepth = 0.0f;
|
||||
vp.MaxDepth = 1.0f;
|
||||
vp.TopLeftX = 0;
|
||||
vp.TopLeftY = 0;
|
||||
g_pd3dDevice->RSSetViewports(1, &vp);
|
||||
|
||||
// Setup the effect
|
||||
{
|
||||
ID3D10Blob *pErrors = NULL;
|
||||
ID3D10Blob *pCompiledEffect;
|
||||
hr = sFnPtr_D3D10CompileEffectFromMemory((void *)g_displayEffectSrc,
|
||||
sizeof(g_displayEffectSrc), NULL,
|
||||
NULL, // pDefines
|
||||
NULL, // pIncludes
|
||||
0, // HLSL flags
|
||||
0, // FXFlags
|
||||
&pCompiledEffect, &pErrors);
|
||||
|
||||
if (pErrors) {
|
||||
LPVOID l_pError = NULL;
|
||||
l_pError = pErrors->GetBufferPointer(); // then cast to a char* to see it
|
||||
// in the locals window
|
||||
fprintf(stdout, "Compilation error: \n %s", (char *)l_pError);
|
||||
}
|
||||
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
|
||||
hr = sFnPtr_D3D10CreateEffectFromMemory(
|
||||
pCompiledEffect->GetBufferPointer(), pCompiledEffect->GetBufferSize(),
|
||||
0, // FXFlags
|
||||
g_pd3dDevice, NULL, &g_pDisplayEffect);
|
||||
pCompiledEffect->Release();
|
||||
|
||||
g_pDisplayTechnique = g_pDisplayEffect->GetTechniqueByName("Render");
|
||||
|
||||
g_pTime = g_pDisplayEffect->GetVariableByName("g_Time")->AsScalar();
|
||||
}
|
||||
|
||||
// Setup the effect
|
||||
{
|
||||
ID3D10Blob *pCompiledEffect;
|
||||
ID3D10Blob *pErrors = NULL;
|
||||
hr = sFnPtr_D3D10CompileEffectFromMemory((void *)g_compositeEffectSrc,
|
||||
sizeof(g_compositeEffectSrc), NULL,
|
||||
NULL, // pDefines
|
||||
NULL, // pIncludes
|
||||
0, // HLSL flags
|
||||
0, // FXFlags
|
||||
&pCompiledEffect, &pErrors);
|
||||
|
||||
if (pErrors) {
|
||||
LPVOID l_pError = NULL;
|
||||
l_pError = pErrors->GetBufferPointer(); // then cast to a char* to see it
|
||||
// in the locals window
|
||||
fprintf(stdout, "Compilation error: \n %s", (char *)l_pError);
|
||||
}
|
||||
|
||||
AssertOrQuit(SUCCEEDED(hr));
|
||||
|
||||
hr = sFnPtr_D3D10CreateEffectFromMemory(
|
||||
pCompiledEffect->GetBufferPointer(), pCompiledEffect->GetBufferSize(),
|
||||
0, // FXFlags
|
||||
g_pd3dDevice, NULL, &g_pCompositeEffect);
|
||||
pCompiledEffect->Release();
|
||||
|
||||
g_pCompositeTechnique = g_pCompositeEffect->GetTechniqueByName("Render");
|
||||
|
||||
g_pvQuadRect =
|
||||
g_pCompositeEffect->GetVariableByName("g_vQuadRect")->AsVector();
|
||||
g_pUseCase = g_pCompositeEffect->GetVariableByName("g_UseCase")->AsScalar();
|
||||
|
||||
g_pTexture2D = g_pCompositeEffect->GetVariableByName("g_Texture2D")
|
||||
->AsShaderResource();
|
||||
g_pTexture2D->SetResource(g_color.pBufferSRV);
|
||||
|
||||
g_pHistogram = g_pCompositeEffect->GetVariableByName("g_Histogram")
|
||||
->AsShaderResource();
|
||||
g_pHistogram->SetResource(g_histogram.pBufferSRV);
|
||||
}
|
||||
|
||||
D3D10_RASTERIZER_DESC rasterizerState;
|
||||
rasterizerState.FillMode = D3D10_FILL_SOLID;
|
||||
rasterizerState.CullMode = D3D10_CULL_FRONT;
|
||||
rasterizerState.FrontCounterClockwise = false;
|
||||
rasterizerState.DepthBias = false;
|
||||
rasterizerState.DepthBiasClamp = 0;
|
||||
rasterizerState.SlopeScaledDepthBias = 0;
|
||||
rasterizerState.DepthClipEnable = false;
|
||||
rasterizerState.ScissorEnable = false;
|
||||
rasterizerState.MultisampleEnable = false;
|
||||
rasterizerState.AntialiasedLineEnable = false;
|
||||
g_pd3dDevice->CreateRasterizerState(&rasterizerState, &g_pRasterState);
|
||||
g_pd3dDevice->RSSetState(g_pRasterState);
|
||||
|
||||
return S_OK;
|
||||
}
|
||||
|
||||
//-----------------------------------------------------------------------------
|
||||
// Name: Cleanup()
|
||||
// Desc: Releases all previously initialized objects
|
||||
//-----------------------------------------------------------------------------
|
||||
VOID Cleanup() {
|
||||
if (g_histogram.pBuffer != NULL) {
|
||||
// Unregister vertex buffer
|
||||
cudaGraphicsUnregisterResource(g_histogram.cudaResource);
|
||||
getLastCudaError("cudaGraphicsUnregisterResource failed");
|
||||
g_histogram.pBuffer->Release();
|
||||
}
|
||||
|
||||
if (g_histogram.pBufferSRV != NULL) {
|
||||
g_histogram.pBufferSRV->Release();
|
||||
}
|
||||
|
||||
if (g_pDisplayEffect != NULL) {
|
||||
g_pDisplayEffect->Release();
|
||||
}
|
||||
|
||||
if (g_pCompositeEffect != NULL) {
|
||||
g_pCompositeEffect->Release();
|
||||
}
|
||||
|
||||
if (g_color.pBufferSRV != NULL) {
|
||||
g_color.pBufferSRV->Release();
|
||||
}
|
||||
|
||||
if (g_color.pBufferRTV != NULL) {
|
||||
g_color.pBufferRTV->Release();
|
||||
}
|
||||
|
||||
if (g_color.pBuffer != NULL) {
|
||||
// Unregister vertex buffer
|
||||
cudaGraphicsUnregisterResource(g_color.cudaResource);
|
||||
getLastCudaError("cudaD3D10UnregisterResource failed");
|
||||
g_color.pBuffer->Release();
|
||||
}
|
||||
|
||||
if (g_pRasterState != NULL) {
|
||||
g_pRasterState->Release();
|
||||
}
|
||||
|
||||
if (g_pSwapChainRTV != NULL) {
|
||||
g_pSwapChainRTV->Release();
|
||||
}
|
||||
|
||||
if (g_pSwapChain != NULL) {
|
||||
g_pSwapChain->Release();
|
||||
}
|
||||
|
||||
if (g_pd3dDevice != NULL) {
|
||||
g_pd3dDevice->Release();
|
||||
}
|
||||
}
|
||||
|
||||
//-----------------------------------------------------------------------------
|
||||
// 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);
|
||||
}
|
||||
|
||||
//-----------------------------------------------------------------------------
|
||||
// Name: MsgProc()
|
||||
// Desc: The window's message handler
|
||||
//-----------------------------------------------------------------------------
|
||||
LRESULT WINAPI MsgProc(HWND hWnd, UINT msg, WPARAM wParam, LPARAM lParam) {
|
||||
switch (msg) {
|
||||
case WM_DESTROY:
|
||||
case WM_KEYDOWN:
|
||||
if (msg != WM_KEYDOWN || wParam == 27) {
|
||||
g_bDone = true;
|
||||
Cleanup();
|
||||
PostQuitMessage(0);
|
||||
return 0;
|
||||
}
|
||||
}
|
||||
|
||||
return DefWindowProc(hWnd, msg, wParam, lParam);
|
||||
}
|
||||
@@ -0,0 +1,223 @@
|
||||
/* Copyright (c) 2021, NVIDIA 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 NVIDIA 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 ``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.
|
||||
*/
|
||||
|
||||
/* This example demonstrates how to use the CUDA Direct3D bindings with the
|
||||
* runtime API.
|
||||
* Device code.
|
||||
*/
|
||||
|
||||
#ifndef SIMPLED3D10RENDERTARGET_KERNEL_CU
|
||||
#define SIMPLED3D10RENDERTARGET_KERNEL_CU
|
||||
|
||||
// includes, C string library
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
// includes, cuda
|
||||
#include <cuda.h>
|
||||
#include <builtin_types.h>
|
||||
#include <cuda_runtime_api.h>
|
||||
|
||||
// includes, project
|
||||
#include <helper_cuda.h> // includes cuda.h and cuda_runtime_api.h
|
||||
//#include "checkCudaErrors"
|
||||
|
||||
#define BIN_COUNT 256
|
||||
#define HISTOGRAM_SIZE (BIN_COUNT * sizeof(unsigned int))
|
||||
|
||||
texture<uchar4, 2, cudaReadModeElementType> colorTex;
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// GPU-specific definitions
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Fast mul on G8x / G9x / G100
|
||||
#define IMUL(a, b) __mul24(a, b)
|
||||
|
||||
// Machine warp size
|
||||
// G80's warp size is 32 threads
|
||||
#define WARP_LOG2SIZE 5
|
||||
|
||||
// Warps in thread block for histogram256Kernel()
|
||||
#define WARP_N 6
|
||||
|
||||
// Corresponding thread block size in threads for histogram256Kernel()
|
||||
#define THREAD_N (WARP_N << WARP_LOG2SIZE)
|
||||
|
||||
// Total histogram size (in counters) per thread block for histogram256Kernel()
|
||||
#define BLOCK_MEMORY (WARP_N * BIN_COUNT)
|
||||
|
||||
// Thread block count for histogram256Kernel()
|
||||
#define BLOCK_N 64
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// If threadPos == threadIdx.x, there are always 4-way bank conflicts,
|
||||
// since each group of 16 threads (half-warp) accesses different bytes,
|
||||
// but only within 4 shared memory banks. Having shuffled bits of threadIdx.x
|
||||
// as in histogram64GPU(), each half-warp accesses different shared memory banks
|
||||
// avoiding any bank conflicts at all.
|
||||
// Refer to the supplied whitepaper for detailed explanations.
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
__device__ inline void addData256(volatile unsigned int *s_WarpHist,
|
||||
unsigned int data, unsigned int threadTag) {
|
||||
unsigned int count;
|
||||
|
||||
do {
|
||||
count = s_WarpHist[data] & 0x07FFFFFFU;
|
||||
count = threadTag | (count + 1);
|
||||
s_WarpHist[data] = count;
|
||||
} while (s_WarpHist[data] != count);
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Main histogram calculation kernel
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
static __global__ void histogramTex256Kernel(unsigned int *d_Result,
|
||||
unsigned int width,
|
||||
unsigned int height, int dataN) {
|
||||
// Current global thread index
|
||||
const int globalTid = IMUL(blockIdx.x, blockDim.x) + threadIdx.x;
|
||||
// Total number of threads in the compute grid
|
||||
const int numThreads = IMUL(blockDim.x, gridDim.x);
|
||||
|
||||
// Thread tag for addData256()
|
||||
// WARP_LOG2SIZE higher bits of counter values are tagged
|
||||
// by lower WARP_LOG2SIZE threadID bits
|
||||
const unsigned int threadTag = threadIdx.x << (32 - WARP_LOG2SIZE);
|
||||
|
||||
// Shared memory storage for each warp
|
||||
volatile __shared__ unsigned int s_Hist[BLOCK_MEMORY];
|
||||
|
||||
// Current warp shared memory base
|
||||
const int warpBase = (threadIdx.x >> WARP_LOG2SIZE) * BIN_COUNT;
|
||||
|
||||
// Clear shared memory buffer for current thread block before processing
|
||||
for (int pos = threadIdx.x; pos < BLOCK_MEMORY; pos += blockDim.x)
|
||||
s_Hist[pos] = 0;
|
||||
|
||||
// Cycle through the entire data set, update subhistograms for each warp
|
||||
__syncthreads();
|
||||
|
||||
for (int pos = globalTid; pos < dataN; pos += numThreads) {
|
||||
// NOTE: check this... Not sure this is what needs to be done
|
||||
int py = pos / width;
|
||||
int px = pos - (py * width);
|
||||
uchar4 data4 = tex2D(colorTex, px, py);
|
||||
|
||||
addData256(s_Hist + warpBase, (data4.x), threadTag);
|
||||
addData256(s_Hist + warpBase, (data4.y), threadTag);
|
||||
addData256(s_Hist + warpBase, (data4.z), threadTag);
|
||||
addData256(s_Hist + warpBase, (data4.w), threadTag);
|
||||
}
|
||||
|
||||
__syncthreads();
|
||||
|
||||
// Merge per-warp histograms into per-block and write to global memory
|
||||
for (int pos = threadIdx.x; pos < BIN_COUNT; pos += blockDim.x) {
|
||||
unsigned int sum = 0;
|
||||
|
||||
for (int base = 0; base < BLOCK_MEMORY; base += BIN_COUNT)
|
||||
sum += s_Hist[base + pos] & 0x07FFFFFFU;
|
||||
|
||||
d_Result[blockIdx.x * BIN_COUNT + pos] = sum;
|
||||
}
|
||||
}
|
||||
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
// Merge BLOCK_N subhistograms of BIN_COUNT bins into final histogram
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
// gridDim.x == BIN_COUNT
|
||||
// blockDim.x == BLOCK_N
|
||||
// blockIdx.x == bin counter processed by current block
|
||||
// threadIdx.x == subhistogram index
|
||||
static __global__ void mergeHistogramTex256Kernel(unsigned int *d_Result) {
|
||||
__shared__ unsigned int data[BLOCK_N];
|
||||
|
||||
// Reads are uncoalesced, but this final stage takes
|
||||
// only a fraction of total processing time
|
||||
data[threadIdx.x] = d_Result[threadIdx.x * BIN_COUNT + blockIdx.x];
|
||||
|
||||
for (int stride = BLOCK_N / 2; stride > 0; stride >>= 1) {
|
||||
__syncthreads();
|
||||
|
||||
if (threadIdx.x < stride) data[threadIdx.x] += data[threadIdx.x + stride];
|
||||
}
|
||||
|
||||
if (threadIdx.x == 0) d_Result[blockIdx.x] = data[0];
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Host interface to GPU histogram
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
extern "C" void checkCudaError() {
|
||||
cudaError_t err = cudaGetLastError();
|
||||
|
||||
if (cudaSuccess != err) {
|
||||
fprintf(stderr, "Cuda error: %s.\n", cudaGetErrorString(err));
|
||||
exit(2);
|
||||
}
|
||||
}
|
||||
|
||||
// Maximum block count for histogram64kernel()
|
||||
// Limits input data size to 756MB
|
||||
// const int MAX_BLOCK_N = 16384;
|
||||
|
||||
// Internal memory allocation
|
||||
// const int BLOCK_N2 = 32;
|
||||
|
||||
extern "C" void createHistogramTex(unsigned int *h_Result, unsigned int width,
|
||||
unsigned int height, cudaArray *colorArray) {
|
||||
cudaBindTextureToArray(colorTex, colorArray);
|
||||
checkCudaError();
|
||||
|
||||
histogramTex256Kernel<<<BLOCK_N, THREAD_N>>>(h_Result, width, height,
|
||||
width * height / 4);
|
||||
checkCudaError();
|
||||
|
||||
mergeHistogramTex256Kernel<<<BIN_COUNT, BLOCK_N>>>(h_Result);
|
||||
checkCudaError();
|
||||
|
||||
cudaUnbindTexture(colorTex);
|
||||
checkCudaError();
|
||||
|
||||
#if 0
|
||||
// Dummy fill test
|
||||
unsigned int toto[256];
|
||||
|
||||
for (int i=0; i<256; i++)
|
||||
{
|
||||
toto[i] = i * 100;
|
||||
}
|
||||
cudaMemcpy(h_Result, toto, HISTOGRAM_SIZE, cudaMemcpyHostToDevice);
|
||||
#endif
|
||||
checkCudaError();
|
||||
}
|
||||
|
||||
extern "C" void bindArrayToTexture(cudaArray *pArray) {}
|
||||
|
||||
#endif // #ifndef SIMPLED3D10RENDERTARGET_KERNEL_CU
|
||||
@@ -0,0 +1,20 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2017
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "simpleD3D10RenderTarget", "simpleD3D10RenderTarget_vs2017.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
|
||||
EndProject
|
||||
Global
|
||||
GlobalSection(SolutionConfigurationPlatforms) = preSolution
|
||||
Debug|x64 = Debug|x64
|
||||
Release|x64 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(ProjectConfigurationPlatforms) = postSolution
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(SolutionProperties) = preSolution
|
||||
HideSolutionNode = FALSE
|
||||
EndGlobalSection
|
||||
EndGlobal
|
||||
@@ -0,0 +1,115 @@
|
||||
<?xml version="1.0" encoding="utf-8"?>
|
||||
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
|
||||
<PropertyGroup>
|
||||
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
|
||||
</PropertyGroup>
|
||||
<ItemGroup Label="ProjectConfigurations">
|
||||
<ProjectConfiguration Include="Debug|x64">
|
||||
<Configuration>Debug</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
<ProjectConfiguration Include="Release|x64">
|
||||
<Configuration>Release</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
</ItemGroup>
|
||||
<PropertyGroup Label="Globals">
|
||||
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
|
||||
<RootNamespace>simpleD3D10RenderTarget_vs2017</RootNamespace>
|
||||
<ProjectName>simpleD3D10RenderTarget</ProjectName>
|
||||
<CudaToolkitCustomDir />
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(WindowsTargetPlatformVersion)'==''">
|
||||
<LatestTargetPlatformVersion>$([Microsoft.Build.Utilities.ToolLocationHelper]::GetLatestSDKTargetPlatformVersion('Windows', '10.0'))</LatestTargetPlatformVersion>
|
||||
<WindowsTargetPlatformVersion Condition="'$(WindowsTargetPlatformVersion)' == ''">$(LatestTargetPlatformVersion)</WindowsTargetPlatformVersion>
|
||||
<TargetPlatformVersion>$(WindowsTargetPlatformVersion)</TargetPlatformVersion>
|
||||
</PropertyGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
|
||||
<PropertyGroup>
|
||||
<ConfigurationType>Application</ConfigurationType>
|
||||
<CharacterSet>MultiByte</CharacterSet>
|
||||
<PlatformToolset>v141</PlatformToolset>
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<UseDebugLibraries>true</UseDebugLibraries>
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Configuration)'=='Release'">
|
||||
<WholeProgramOptimization>true</WholeProgramOptimization>
|
||||
</PropertyGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||
<ImportGroup Label="ExtensionSettings">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.5.props" />
|
||||
</ImportGroup>
|
||||
<ImportGroup Label="PropertySheets">
|
||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||
</ImportGroup>
|
||||
<PropertyGroup Label="UserMacros" />
|
||||
<PropertyGroup>
|
||||
<IntDir>$(Platform)/$(Configuration)/</IntDir>
|
||||
<IncludePath>$(IncludePath)</IncludePath>
|
||||
<CodeAnalysisRuleSet>AllRules.ruleset</CodeAnalysisRuleSet>
|
||||
<CodeAnalysisRules />
|
||||
<CodeAnalysisRuleAssemblies />
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Platform)'=='x64'">
|
||||
<OutDir>../../bin/win64/$(Configuration)/</OutDir>
|
||||
</PropertyGroup>
|
||||
<ItemDefinitionGroup>
|
||||
<ClCompile>
|
||||
<WarningLevel>Level3</WarningLevel>
|
||||
<PreprocessorDefinitions>WIN32;_MBCS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
|
||||
<AdditionalIncludeDirectories>./;$(CudaToolkitDir)/include;../../Common;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cudart_static.lib;d3d10.lib;d3dcompiler.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/simpleD3D10RenderTarget.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
<AdditionalOptions>--threads 0</AdditionalOptions>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<ClCompile>
|
||||
<Optimization>Disabled</Optimization>
|
||||
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>true</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MTd</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="simpleD3D10RenderTarget.cpp" />
|
||||
<CudaCompile Include="simpleD3D10RenderTarget_kernel.cu" />
|
||||
<ClCompile Include="..\..\Common\rendercheck_d3d10.cpp" />
|
||||
<ClInclude Include="..\..\Common\rendercheck_d3d10.h" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.5.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
@@ -0,0 +1,20 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2019
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "simpleD3D10RenderTarget", "simpleD3D10RenderTarget_vs2019.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
|
||||
EndProject
|
||||
Global
|
||||
GlobalSection(SolutionConfigurationPlatforms) = preSolution
|
||||
Debug|x64 = Debug|x64
|
||||
Release|x64 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(ProjectConfigurationPlatforms) = postSolution
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(SolutionProperties) = preSolution
|
||||
HideSolutionNode = FALSE
|
||||
EndGlobalSection
|
||||
EndGlobal
|
||||
@@ -0,0 +1,111 @@
|
||||
<?xml version="1.0" encoding="utf-8"?>
|
||||
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
|
||||
<PropertyGroup>
|
||||
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
|
||||
</PropertyGroup>
|
||||
<ItemGroup Label="ProjectConfigurations">
|
||||
<ProjectConfiguration Include="Debug|x64">
|
||||
<Configuration>Debug</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
<ProjectConfiguration Include="Release|x64">
|
||||
<Configuration>Release</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
</ItemGroup>
|
||||
<PropertyGroup Label="Globals">
|
||||
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
|
||||
<RootNamespace>simpleD3D10RenderTarget_vs2019</RootNamespace>
|
||||
<ProjectName>simpleD3D10RenderTarget</ProjectName>
|
||||
<CudaToolkitCustomDir />
|
||||
</PropertyGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
|
||||
<PropertyGroup>
|
||||
<ConfigurationType>Application</ConfigurationType>
|
||||
<CharacterSet>MultiByte</CharacterSet>
|
||||
<PlatformToolset>v142</PlatformToolset>
|
||||
<WindowsTargetPlatformVersion>10.0</WindowsTargetPlatformVersion>
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<UseDebugLibraries>true</UseDebugLibraries>
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Configuration)'=='Release'">
|
||||
<WholeProgramOptimization>true</WholeProgramOptimization>
|
||||
</PropertyGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||
<ImportGroup Label="ExtensionSettings">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.5.props" />
|
||||
</ImportGroup>
|
||||
<ImportGroup Label="PropertySheets">
|
||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||
</ImportGroup>
|
||||
<PropertyGroup Label="UserMacros" />
|
||||
<PropertyGroup>
|
||||
<IntDir>$(Platform)/$(Configuration)/</IntDir>
|
||||
<IncludePath>$(IncludePath)</IncludePath>
|
||||
<CodeAnalysisRuleSet>AllRules.ruleset</CodeAnalysisRuleSet>
|
||||
<CodeAnalysisRules />
|
||||
<CodeAnalysisRuleAssemblies />
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Platform)'=='x64'">
|
||||
<OutDir>../../bin/win64/$(Configuration)/</OutDir>
|
||||
</PropertyGroup>
|
||||
<ItemDefinitionGroup>
|
||||
<ClCompile>
|
||||
<WarningLevel>Level3</WarningLevel>
|
||||
<PreprocessorDefinitions>WIN32;_MBCS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
|
||||
<AdditionalIncludeDirectories>./;$(CudaToolkitDir)/include;../../Common;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cudart_static.lib;d3d10.lib;d3dcompiler.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/simpleD3D10RenderTarget.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
<AdditionalOptions>--threads 0</AdditionalOptions>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<ClCompile>
|
||||
<Optimization>Disabled</Optimization>
|
||||
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>true</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MTd</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="simpleD3D10RenderTarget.cpp" />
|
||||
<CudaCompile Include="simpleD3D10RenderTarget_kernel.cu" />
|
||||
<ClCompile Include="..\..\Common\rendercheck_d3d10.cpp" />
|
||||
<ClInclude Include="..\..\Common\rendercheck_d3d10.h" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.5.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
Reference in New Issue
Block a user