如何学习cuda c?
Posted
tags:
篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了如何学习cuda c?相关的知识，希望对你有一定的参考价值。
1、 CUDA C编写Windows Console Application
下面我们从一个简单的例子开始学习CUDA C。
打开VS，新建一个CUDAWinApp项目，项目名称为Vector，解决方案名称为CUDADemo。依次点击“确定”，“下一步”，选择Empty project。点击“Finished”。这样一个CUDA的项目就建成了。
右键点击Vector项目，依次选择“添加”、“新建项”、“代码”、“CUDA”。在名称中输入要添加的文件名。如Vector.cu。然后点击添加。
下面在Vector.cu文件里实现两个向量相加的程序。
//添加系统库
#include
#include
//添加CUDA支持
#include
__global__ void VecAdd(float *A, float *B, float *C);
__host__ void runVecAdd(int argc, char **argv);
int main(int argc, char **argv)
runVecAdd(argc,argv);
CUT_EXIT(argc,argv);
__host__ void runVecAdd(int argc,char **argv)
//初始化host端内存数据
const unsigned int N = 8;//向量维数
const unsigned int memSize = sizeof(float)*N;//需要空间的字节数
float *h_A = (float*)malloc(memSize);
float *h_B = (float*)malloc(memSize);
float *h_C = (float*)malloc(memSize);
for (unsigned int i = 0; i < N; i++)
h_A[i] = i;h_B[i] = i;
//设备端显存空间
float *d_A, *d_B, *d_C;
//初始化Device
CUT_DEVICE_INIT(argc,argv);
CUDA_SAFE_CALL(cudaMalloc((void**)&d_A, memSize));
CUDA_SAFE_CALL(cudaMalloc((void**)&d_B, memSize));
CUDA_SAFE_CALL(cudaMalloc((void**)&d_C, memSize));
CUDA_SAFE_CALL(cudaMemcpy(d_A, h_A, memSize, cudaMemcpyHostToDevice));
CUDA_SAFE_CALL(cudaMemcpy(d_B, h_B, memSize, cudaMemcpyHostToDevice));
VecAdd<<<1,N,memSize>>>(d_A, d_B, d_C);
CUT_CHECK_ERROR("Kernel execution failed");
CUDA_SAFE_CALL(cudaMemcpy(h_C, d_C, memSize, cudaMemcpyDeviceToHost));
for (unsigned int i = 0; i < N; i++)
printf("%.0f ",h_C[i]);
free(h_A);free(h_B);free(h_C);
CUDA_SAFE_CALL(cudaFree(d_A));
CUDA_SAFE_CALL(cudaFree(d_B));
CUDA_SAFE_CALL(cudaFree(d_C));
__global__ void VecAdd(float *A, float *B, float *C)
//分配shared memory
extern __shared__ float s_A[];
extern __shared__ float s_B[];
extern __shared__ float s_C[];
//从global memory拷贝到shared memory
const unsigned int i = threadIdx.x;
s_A[i] = A[i];
s_B[i] = B[i];
//计算
s_C[i] = s_A[i] + s_B[i];
//拷贝到global memory
C[i] = s_C[i];
由于这里不是讲CUDA编程的，关于它的编程模型已经超出了我要介绍的范围，您可以阅读《GPU高性能运算之CUDA》来获得CUDA编程模型的知识。
编译Vector项目，执行此项目后会得到图1如下输出：
图1 Vector项目执行结果
2、CUDA C编写DLL模块
更多情况下的您的软件可能只是使用CUDA来实现一段程序的加速，这种情况下我们可以使用CUDA C 编写DLL来提供接口。下面我们就将例1编译成DLL。
在刚才的CUDADemo解决方案目录下添加一个新的CUDA项目（当然您也可以重新建立一个解决方案）。项目名为VecAdd_dynamic。Application Type选为DLL，Additional Options选择Empty Project。
第一步，添加头文件，文件名最好与工程名同名，这样便于您的维护工作。这里我向项目中添加了VecAdd_dynamic.h，在此头文件中添加如下代码
#ifndef _VECADD_DYNAMIC_H_
#define _VECADD_DYNAMIC_H_
//并行计算N维向量的加法
__declspec(dllexport) void VecAdd(float* h_A, float* h_B, float* h_C, int N);
#endif
第二步，添加cpp文件，文件名为VecAdd_dynamic.cpp，在此文件中添加如下代码
#include
#include "VecAdd_dynamic.h"
#ifdef _MANAGED
#pragma managed(push, off)
#endif
BOOL APIENTRY DllMain(HMODULE hModule,DWORD ul_reason_for_call,LPVOID lpReserved)
return TRUE;
#ifdef _MANAGED
#pragma managed(pop)
#endif
第三步，添加def文件，此文件的功能就是确保其它厂商的编译器能够调用此DLL里的函数。这一点非常关键，因为您的程序可能用到多个厂家的编译器。文件名为VecAdd_dynamic.def。向该文件中添加：
EXPORTS
VecAdd
第四步，添加cu文件，文件名为VecAdd_dynamic.cu。注意此文件最好直接添加到项目目录下，不要添加到源文件选项卡或其它已有的选项卡下。
在cu文件里添加如下代码，实现要导出的函数。
#include
#include
#include
#if __DEVICE_EMULATION__
bool InitCUDA(void)
return true;
#else
bool InitCUDA(void)
int count = 0;
int i = 0;
cudaGetDeviceCount(&count);
if(count == 0)
fprintf(stderr, "There is no device./n");
return false;
for(i = 0; i < count; i++)
cudaDeviceProp prop;
if(cudaGetDeviceProperties(&prop, i) == cudaSuccess)
if(prop.major >= 1)
break;
if(i == count)
fprintf(stderr, "There is no device supporting CUDA./n");
return false;
cudaSetDevice(i);
printf("CUDA initialized./n");
return true;
#endif
__global__ void D_VecAdd(float *g_A, float *g_B, float *g_C, int N)
unsigned int i = threadIdx.x;
if (i < N)
g_C[i] = g_A[i] + g_B[i];
void VecAdd(float* h_A, float* h_B, float* h_C, int N)
if(!InitCUDA())
return;
float *g_A, *g_B, *g_C;
unsigned int size = N * sizeof(float);
CUDA_SAFE_CALL(cudaMalloc((void**)&g_A, size));
CUDA_SAFE_CALL(cudaMalloc((void**)&g_B, size));
CUDA_SAFE_CALL(cudaMalloc((void**)&g_C, size));
CUDA_SAFE_CALL(cudaMemcpy(g_A, h_A, size, cudaMemcpyHostToDevice));
CUDA_SAFE_CALL(cudaMemcpy(g_B, h_B, size, cudaMemcpyHostToDevice));
D_VecAdd<<<1,N>>>(g_A, g_B, g_C, N);
CUDA_SAFE_CALL(cudaMemcpy(h_C, g_C, size, cudaMemcpyDeviceToHost));
cudaFree(g_A);cudaFree(g_B);cudaFree(g_C);
第五步，如果您已经正确完成了以上四步，那么剩下的就只有编译，只要您用过VS，这一步就不需要我介绍了吧。成功之后，在您的解决方案文件目录下的Debug文件夹下会有一个VecAdd_dynamic.dll文件。
3、 在 .NET 中使用CUDA C编写的DLL
下面介绍在托管程序中如何使用VecAdd_dynamic.dll。
第一步，在上面的解决方案CUDADemo下添加一个C++/CLR的Windows窗体应用程序，工程名为NETDemo（当然您也可以重新建一个解决方案，工程名也是随意的）。
第二步，在窗体上添加一个按钮，名字随意，我将它的现实文本改为“调用CUDA_DLL”，给这个按钮添加click事件。我们的代码将在这个事件里添加调用VecAdd()的程序。在窗体上添加一个文本框用来显示调用输出的结果。
第三步，代码实现。为工程NETDemo添加一个头文件，我将它命名为Win32.h，这个文件中主要是实现VecAdd()函数的导入。在此文件中添加如下代码
#pragma once
namespace Win32
using namespace System::Runtime::InteropServices;
[DllImport("VecAdd_dynamic.dll",EntryPoint="VecAdd",CharSet=CharSet::Auto)]
extern "C" void VecAdd(float* h_A, float* h_B, float* h_C, int N);
在Form1.h中，#pragma once 之后 namespace NETDemo 之前添加以下代码。
#include "Win32.h"
#include
在button1_Click()中添加如下代码
int N = 8;
float* h_A = (float*)malloc(N*sizeof(float));
float* h_B = (float*)malloc(N*sizeof(float));
float* h_C = (float*)malloc(N*sizeof(float));
for (int i = 0; i < N; i++)
h_A[i] = i;h_B[i] = i;
Win32::VecAdd(h_A, h_B, h_C,N);
String ^reslut;
for (int i = 0; i < N; i++)
reslut += Convert::ToString(h_C[i]) + ", ";
this->textBox1->Text = Convert::ToString(reslut);
free(h_A);free(h_B);free(h_C);
第四步、执行NETDemo项目。点击“调用CUDA_DLL”，您会看到图3所示的结果
图3 NETDemo运行结果
到现在为止您已经完全可以正确使用CUDA了。
CUDA C Programming Guide 在线教程学习笔记 Part 6
? 纹理内存读取函数
1 /****************************************************************/ 2 3 // Texture Object API 4 5 /****************************************************************/ 6 7 template<class T> 8 T tex1Dfetch(cudaTextureObject_t texObj, int x); 9 // 一维纹理，整数下标。只用于非正规化坐标，挤压模式或边界模式，不能滤波 10 11 template<class T> 12 T tex1D(cudaTextureObject_t texObj, float x); 13 // 一维纹理，浮点数下标 14 15 template<class T> 16 T tex1DLod(cudaTextureObject_t texObj, float x, float level); 17 // 一维纹理，基于 level-of-detail（层次细节）算法 18 19 template<class T> 20 T tex1DGrad(cudaTextureObject_t texObj, float x, float dx, float dy); 21 // 一维纹理，基于 gradients（梯度）算法 22 23 template<class T> 24 T tex2D(cudaTextureObject_t texObj, float x, float y); 25 // 二维纹理 26 27 template<class T> 28 tex2DLod(cudaTextureObject_t texObj, float x, float y, float level); 29 // 二维纹理，基于 level-of-detail（层次细节）算法 30 31 template<class T> 32 T tex2DGrad(cudaTextureObject_t texObj, float x, float y,float2 dx, float2 dy); 33 // 二维纹理，基于 gradients（梯度）算法 34 35 template<class T> 36 T tex3D(cudaTextureObject_t texObj, float x, float y, float z); 37 // 三维纹理 38 39 template<class T> 40 T tex3DLod(cudaTextureObject_t texObj, float x, float y, float z, float level); 41 // 三维纹理，基于 level-of-detail（层次细节）算法 42 43 template<class T> 44 T tex3DGrad(cudaTextureObject_t texObj, float x, float y, float z,float4 dx, float4 dy); 45 // 三维纹理，基于 gradients（梯度）算法 46 47 template<class T> 48 T tex1DLayered(cudaTextureObject_t texObj, float x, int layer); 49 // 一维分层纹理 50 51 template<class T> 52 T tex1DLayeredLod(cudaTextureObject_t texObj, float x, int layer, float level); 53 // 一维分层纹理，基于 level-of-detail（层次细节）算法 54 55 template<class T> 56 T tex1DLayeredGrad(cudaTextureObject_t texObj, float x, int layer, float dx, float dy); 57 // 一维分层纹理，基于 gradients（梯度）算法 58 59 template<class T> 60 T tex2DLayered(cudaTextureObject_t texObj, float x, float y, int layer); 61 // 二维分层纹理 62 63 template<class T> 64 T tex2DLayeredLod(cudaTextureObject_t texObj, float x, float y, int layer, float level); 65 // 二维分层纹理，基于 level-of-detail（层次细节）算法 66 67 template<class T> 68 T tex2DLayeredGrad(cudaTextureObject_t texObj, float x, float y, int layer, float2 dx, float2 dy); 69 // 二维分层纹理，基于 gradients（梯度）算法 70 71 template<class T> 72 T texCubemap(cudaTextureObject_t texObj, float x, float y, float z); 73 // 立方体贴图纹理 74 75 template<class T> 76 T texCubemapLod(cudaTextureObject_t texObj, float x, float, y, float z, float level); 77 // 立方体贴图纹理，基于 level-of-detail（层次细节）算法 78 79 template<class T> 80 T texCubemapLayered(cudaTextureObject_t texObj, float x, float y, float z, int layer); 81 // 立方体分层贴图纹理 82 83 template<class T> 84 T texCubemapLayeredLod(cudaTextureObject_t texObj, float x, float y, float z, int layer, float level); 85 // 立方分层体贴图纹理，基于 level-of-detail（层次细节）算法 86 87 template<class T> 88 T tex2Dgather(cudaTextureObject_t texObj, float x, float y, int comp = 0); 89 // ？ 90 91 /****************************************************************/ 92 93 // Texture Object API 94 95 /****************************************************************/ 96 97 template<class DataType> 98 Type tex1Dfetch(texture<DataType, cudaTextureType1D, cudaReadModeElementType> texRef, int x); 99 100 float tex1Dfetch(texture<unsigned char, cudaTextureType1D, cudaReadModeNormalizedFloat> texRef, int x); 101 float tex1Dfetch(texture<signed char, cudaTextureType1D, cudaReadModeNormalizedFloat> texRef, int x); 102 float tex1Dfetch(texture<unsigned short, cudaTextureType1D, cudaReadModeNormalizedFloat> texRef, int x); 103 float tex1Dfetch(texture<signed short, cudaTextureType1D, cudaReadModeNormalizedFloat> texRef, int x); 104 // 一维纹理，整数下标。只用于非正规化坐标，挤压模式或边界模式，不能滤波 105 106 // 2元组，4元组也可以使用这种方法，例如： 107 float4 tex1Dfetch(texture<uchar2, cudaTextureType1D, cudaReadModeNormalizedFloat> texRef, int x); 108 float4 tex1Dfetch(texture<uchar4, cudaTextureType1D, cudaReadModeNormalizedFloat> texRef, int x); 109 110 // 其他函数名类似 Texture Object API 111 template<class DataType, enum cudaTextureReadMode readMode> 112 Type tex1D(texture<DataType, cudaTextureType1D, readMode> texRef, float x); 113 114 template<class DataType, enum cudaTextureReadMode readMode> 115 Type tex1DLod(texture<DataType, cudaTextureType1D, readMode> texRef, float x, float level); 116 117 template<class DataType, enum cudaTextureReadMode readMode> 118 Type tex1DGrad(texture<DataType, cudaTextureType1D, readMode> texRef, float x, float dx, float dy); 119 120 template<class DataType, enum cudaTextureReadMode readMode> 121 Type tex2D(texture<DataType, cudaTextureType2D, readMode> texRef,float x, float y); 122 123 template<class DataType, enum cudaTextureReadMode readMode> 124 Type tex2DLod(texture<DataType, cudaTextureType2D, readMode> texRef, float x, float y, float level); 125 126 template<class DataType, enum cudaTextureReadMode readMode> 127 Type tex2DGrad(texture<DataType, cudaTextureType2D, readMode> texRef, float x, float y, float2 dx, float2 dy); 128 129 template<class DataType, enum cudaTextureReadMode readMode> 130 Type tex3D(texture<DataType, cudaTextureType3D, readMode> texRef, float x, float y, float z); 131 132 template<class DataType, enum cudaTextureReadMode readMode> 133 Type tex3DLod(texture<DataType, cudaTextureType3D, readMode> texRef, float x, float y, float z, float level); 134 135 template<class DataType, enum cudaTextureReadMode readMode> 136 Type tex3DGrad(texture<DataType, cudaTextureType3D, readMode> texRef, float x, float y, float z, float4 dx, float4 dy); 137 138 template<class DataType, enum cudaTextureReadMode readMode> 139 Type tex1DLayered(texture<DataType, cudaTextureType1DLayered, readMode> texRef, float x, int layer); 140 141 template<class DataType, enum cudaTextureReadMode readMode> 142 Type tex1DLayeredLod(texture<DataType, cudaTextureType1D, readMode> texRef, float x, int layer, float level); 143 144 template<class DataType, enum cudaTextureReadMode readMode> 145 Type tex1DLayeredGrad(texture<DataType, cudaTextureType1D, readMode> texRef, float x, int layer, float dx, float dy); 146 147 template<class DataType, enum cudaTextureReadMode readMode> 148 Type tex2DLayered(texture<DataType, cudaTextureType2DLayered, readMode> texRef, float x, float y, int layer); 149 150 template<class DataType, enum cudaTextureReadMode readMode> 151 Type tex2DLayeredLod(texture<DataType, cudaTextureType2D, readMode> texRef, float x, float y, int layer, float level); 152 153 template<class DataType, enum cudaTextureReadMode readMode> 154 Type tex2DLayeredGrad(texture<DataType, cudaTextureType2D, readMode> texRef, float x, float y, int layer, float2 dx, float2 dy); 155 156 template<class DataType, enum cudaTextureReadMode readMode> 157 Type texCubemap(texture<DataType, cudaTextureTypeCubemap, readMode> texRef, float x, float y, float z); 158 159 template<class DataType, enum cudaTextureReadMode readMode> 160 Type texCubemapLod(texture<DataType, cudaTextureType3D, readMode> texRef, float x, float y, float z, float level); 161 162 template<class DataType, enum cudaTextureReadMode readMode> 163 Type texCubemapLayered(texture<DataType, cudaTextureTypeCubemapLayered, readMode> texRef, float x, float y, float z, int layer); 164 165 template<class DataType, enum cudaTextureReadMode readMode> 166 Type texCubemapLayeredLod(texture<DataType, cudaTextureType3D, readMode> texRef, float x, float y, float z, int layer, float level); 167 168 template<class DataType, enum cudaTextureReadMode readMode> 169 Type tex2Dgather(texture<DataType, cudaTextureType2D, readMode> texRef, float x, float y, int comp = 0);
? 表面内存读写函数
1 /****************************************************************/ 2 3 // Surface Object API 4 5 /****************************************************************/ 6 7 template<class T> 8 T surf1Dread(cudaSurfaceObject_t surfObj, int x, boundaryMode = cudaBoundaryModeTrap); 9 // 一维表面读取 10 11 template<class T> 12 void surf1Dwrite(T data, cudaSurfaceObject_t surfObj, int x, boundaryMode = cudaBoundaryModeTrap); 13 // 一维表面写入 14 15 template<class T> 16 T surf2Dread(cudaSurfaceObject_t surfObj, int x, int y, boundaryMode = cudaBoundaryModeTrap); 17 // 二维表面读取 18 template<class T> 19 void surf2Dread(T* data, cudaSurfaceObject_t surfObj, int x, int y, boundaryMode = cudaBoundaryModeTrap); 20 // 二维表面读取 21 22 template<class T> 23 void surf2Dwrite(T data, cudaSurfaceObject_t surfObj, int x, int y, boundaryMode = cudaBoundaryModeTrap); 24 // 二维表面写入 25 26 template<class T> 27 T surf3Dread(cudaSurfaceObject_t surfObj, int x, int y, int z, boundaryMode = cudaBoundaryModeTrap); 28 // 三维表面读取 29 30 template<class T> 31 void surf3Dread(T* data, cudaSurfaceObject_t surfObj, int x, int y, int z, boundaryMode = cudaBoundaryModeTrap); 32 // 三维表面读取 33 34 template<class T> 35 void surf3Dwrite(T data, cudaSurfaceObject_t surfObj, int x, int y, int z, boundaryMode = cudaBoundaryModeTrap); 36 // 三维表面写入 37 38 template<class T> 39 T surf1DLayeredread(cudaSurfaceObject_t surfObj, int x, int layer, boundaryMode = cudaBoundaryModeTrap); 40 // 一维分层表面读取 41 42 template<class T> 43 void surf1DLayeredread(T data, cudaSurfaceObject_t surfObj, int x, int layer, boundaryMode = cudaBoundaryModeTrap); 44 // 一维分层表面读取 45 46 template<class Type> 47 void surf1DLayeredwrite(T data, cudaSurfaceObject_t surfObj, int x, int layer, boundaryMode = cudaBoundaryModeTrap); 48 // 一维分层表面写入 49 50 template<class T> 51 T surf2DLayeredread(cudaSurfaceObject_t surfObj, int x, int y, int layer, boundaryMode = cudaBoundaryModeTrap); 52 // 二维分层表面读取 53 54 template<class T> 55 void surf2DLayeredread(T data, cudaSurfaceObject_t surfObj, int x, int y, int layer, boundaryMode = cudaBoundaryModeTrap); 56 // 二维分层表面读取 57 template<class T> 58 void surf2DLayeredwrite(T data, cudaSurfaceObject_t surfObj, int x, int y, int layer, boundaryMode = cudaBoundaryModeTrap); 59 // 二维分层表面写入 60 61 template<class T> 62 T surfCubemapread(cudaSurfaceObject_t surfObj, int x, int y, int face, boundaryMode = cudaBoundaryModeTrap); 63 // 立方体贴图表面读取 64 65 template<class T> 66 void surfCubemapread(T data, cudaSurfaceObject_t surfObj, int x, int y, int face, boundaryMode = cudaBoundaryModeTrap); 67 // 立方体贴图表面读取 68 69 template<class T> 70 void surfCubemapwrite(T data, cudaSurfaceObject_t surfObj, int x, int y, int face, boundaryMode = cudaBoundaryModeTrap); 71 // 立方体贴图表面写入 72 73 template<class T> 74 T surfCubemapLayeredread(cudaSurfaceObject_t surfObj, int x, int y, int layerFace, boundaryMode = cudaBoundaryModeTrap); 75 // 立方体分层贴图表面读取 76 77 template<class T> 78 void surfCubemapLayeredread(T data, cudaSurfaceObject_t surfObj, int x, int y, int layerFace, boundaryMode = cudaBoundaryModeTrap); 79 // 立方体分层贴图表面读取 80 81 template<class T> 82 void surfCubemapLayeredwrite(T data, cudaSurfaceObject_t surfObj, int x, int y, int layerFace, boundaryMode = cudaBoundaryModeTrap); 83 // 立方体分层贴图表面写入 84 85 /****************************************************************/ 86 87 // Surface Refernence API 88 89 /****************************************************************/ 90 91 // 函数名类似 Texture Object API 92 template<class Type> 93 Type surf1Dread(surface<void, cudaSurfaceType1D> surfRef, int x, boundaryMode = cudaBoundaryModeTrap); 94 95 template<class Type> 96 void surf1Dread(Type data, surface<void, cudaSurfaceType1D> surfRef, int x, boundaryMode = cudaBoundaryModeTrap); 97 98 template<class Type> 99 void surf1Dwrite(Type data, surface<void, cudaSurfaceType1D> surfRef, int x, boundaryMode = cudaBoundaryModeTrap); 100 101 template<class Type> 102 Type surf2Dread(surface<void, cudaSurfaceType2D> surfRef, int x, int y, boundaryMode = cudaBoundaryModeTrap); 103 104 template<class Type> 105 void surf2Dread(Type* data, surface<void, cudaSurfaceType2D> surfRef, int x, int y, boundaryMode = cudaBoundaryModeTrap); 106 107 template<class Type> 108 void surf3Dwrite(Type data, surface<void, cudaSurfaceType3D> surfRef, int x, int y, int z, boundaryMode = cudaBoundaryModeTrap); 109 110 template<class Type> 111 Type surf3Dread(surface<void, cudaSurfaceType3D> surfRef, int x, int y, int z, boundaryMode = cudaBoundaryModeTrap); 112 113 template<class Type> 114 void surf3Dread(Type* data, surface<void, cudaSurfaceType3D> surfRef, int x, int y, int z, boundaryMode = cudaBoundaryModeTrap); 115 116 template<class Type> 117 void surf3Dwrite(Type data, surface<void, cudaSurfaceType3D> surfRef, int x, int y, int z, boundaryMode = cudaBoundaryModeTrap); 118 119 template<class Type> 120 Type surf1DLayeredread(surface<void, cudaSurfaceType1DLayered> surfRef, int x, int layer, boundaryMode = cudaBoundaryModeTrap); 121 122 template<class Type> 123 void surf1DLayeredread(Type data, surface<void, cudaSurfaceType1DLayered> surfRef, int x, int layer, boundaryMode = cudaBoundaryModeTrap); 124 125 template<class Type> 126 void surf1DLayeredwrite(Type data, surface<void, cudaSurfaceType1DLayered> surfRef, int x, int layer, boundaryMode = cudaBoundaryModeTrap); 127 128 template<class Type> 129 Type surf2DLayeredread(surface<void, cudaSurfaceType2DLayered> surfRef, int x, int y, int layer, boundaryMode = cudaBoundaryModeTrap); 130 131 template<class Type> 132 void surf2DLayeredread(Type data, surface<void, cudaSurfaceType2DLayered> surfRef, int x, int y, int layer, boundaryMode = cudaBoundaryModeTrap); 133 134 template<class Type> 135 void surf2DLayeredwrite(Type data, surface<void, cudaSurfaceType2DLayered> surfRef, int x, int y, int layer, boundaryMode = cudaBoundaryModeTrap); 136 137 template<class Type> 138 Type surfCubemapread(surface<void, cudaSurfaceTypeCubemap> surfRef, int x, int y, int face, boundaryMode = cudaBoundaryModeTrap); 139 140 template<class Type> 141 void surfCubemapread(Type data, surface<void, cudaSurfaceTypeCubemap> surfRef, int x, int y, int face, boundaryMode = cudaBoundaryModeTrap); 142 143 template<class Type> 144 void surfCubemapwrite(Type data, surface<void, cudaSurfaceTypeCubemap> surfRef, int x, int y, int face, boundaryMode = cudaBoundaryModeTrap); 145 146 template<class Type> 147 Type surfCubemapLayeredread(surface<void, cudaSurfaceTypeCubemapLayered> surfRef, int x, int y, int layerFace, boundaryMode = cudaBoundaryModeTrap); 148 149 template<class Type> 150 void surfCubemapLayeredread(Type data, surface<void, cudaSurfaceTypeCubemapLayered> surfRef, int x, int y, int layerFace, boundaryMode = cudaBoundaryModeTrap); 151 152 template<class Type> 153 void surfCubemapLayeredwrite(Type data, surface<void, cudaSurfaceTypeCubemapLayered> surfRef, int x, int y, int layerFace, boundaryMode = cudaBoundaryModeTrap);
以上是关于如何学习cuda c?的主要内容，如果未能解决你的问题，请参考以下文章