opencl初学笔记--经典初学小案例

计算两个数组相加的和，放到另一个数组中去。程序用cpu和gpu分别计算，最后验证它们是否相等。OpenCL程序的流程大致如下：

          1）创建平台对象，得到一个OpenCL平台

       

          2）创建GPU设备

       

          3）创建作用于GPU设备上的context

        

          4)创建相应的command队列

        

         5）创建3个OpenCL内存对象，两个被用来放输入数据，一个放输出数据

        

         6）创建kernel程序对象，编译kernel对象

        

         7）从kernel代码创建kernel对象【设置参数，并执行kenerl】

        

          结束

 其中  main.cpp代码为：【以下代码下载来源于CSDN】

  #include
 
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include


using namespace std;
#define BUFSIZE 262144


#pragma comment (lib,"OpenCL.lib")


//把文本文件读入一个string中，用来读入kernel源文件
int convertToString(const char *filename, std::string& s)
{
size_t size;
char*  str;


std::fstream f(filename, (std::fstream::in | std::fstream::binary));


if(f.is_open())
{
size_t fileSize;
f.seekg(0, std::fstream::end);
size = fileSize = (size_t)f.tellg();
f.seekg(0, std::fstream::beg);


str = new char[size+1];
if(!str)
{
f.close();
return NULL;
}


f.read(str, fileSize);
f.close();
str[size] = '\0';


s = str;
delete[] str;
return 0;
}
printf("Error: Failed to open file %s\n", filename);
return 1;
}


//等待事件完成
int waitForEventAndRelease(cl_event *event)
{
cl_int status = CL_SUCCESS;
cl_int eventStatus = CL_QUEUED;
while(eventStatus != CL_COMPLETE)
{
status = clGetEventInfo(
*event, 
CL_EVENT_COMMAND_EXECUTION_STATUS, 
sizeof(cl_int),
&eventStatus,
NULL);
}


status = clReleaseEvent(*event);


return 0;
}






int main(int argc, char* argv[])
{


     //在host内存中创建三个缓冲区 
float *buf1 = 0;
float *buf2 = 0;
float *buf = 0;


buf1 =(float *)malloc(BUFSIZE * sizeof(float));
buf2 =(float *)malloc(BUFSIZE * sizeof(float));
buf =(float *)malloc(BUFSIZE * sizeof(float));


//用一些随机值初始化buf1和buf2的内容 
int i;
srand( (unsigned)time( NULL ) ); 
for(i = 0; i < BUFSIZE; i++)
buf1[i] = rand()%65535;


srand( (unsigned)time( NULL ) +1000);
for(i = 0; i < BUFSIZE; i++)
buf2[i] = rand()%65535;


//cpu计算buf1,buf2的和
for(i = 0; i < BUFSIZE; i++)
buf[i] = buf1[i] + buf2[i];


cl_uint status;
cl_platform_id platform;


//创建平台对象
//status = clGetPlatformIDs( 1, &platform, NULL );
cl_uint numPlatforms;
std::string platformVendor; 
status = clGetPlatformIDs(0, NULL, &numPlatforms);
if(status != CL_SUCCESS)
{
return 0;
}
if (0 < numPlatforms) 
{
cl_platform_id* platforms = new cl_platform_id[numPlatforms];
status = clGetPlatformIDs(numPlatforms, platforms, NULL);


char platformName[100];
for (unsigned i = 0; i < numPlatforms; ++i) 
{
status = clGetPlatformInfo(platforms[i],
CL_PLATFORM_VENDOR,
sizeof(platformName),
platformName,
NULL);


platform = platforms[i];
platformVendor.assign(platformName);


if (!strcmp(platformName, "Advanced Micro Devices, Inc.")) 
{
break;
}
}


std::cout << "Platform found : " << platformName << "\n";
delete[] platforms;
}
cl_device_id device;


//创建GPU设备
clGetDeviceIDs( platform, CL_DEVICE_TYPE_GPU,
1,
&device,
NULL);
//创建context
cl_context context = clCreateContext( NULL,
1,
&device,
NULL, NULL, NULL);
//创建命令队列
cl_command_queue queue = clCreateCommandQueue( context,
device,
CL_QUEUE_PROFILING_ENABLE, NULL );
//创建三OpenCl内存对象，并把buf1的内容通过隐式拷贝的方式
//buf1内容拷贝到clbuf1,buf2的内容通过显示拷贝的方式拷贝到clbuf2
cl_mem clbuf1 = clCreateBuffer(context,
CL_MEM_READ_ONLY | CL_MEM_COPY_HOST_PTR,
BUFSIZE*sizeof(cl_float),buf1,
NULL );


cl_mem clbuf2 = clCreateBuffer(context,
CL_MEM_READ_ONLY ,
BUFSIZE*sizeof(cl_float),NULL,
NULL );


cl_event writeEvt;
status = clEnqueueWriteBuffer(queue, clbuf2, 1,
0, BUFSIZE*sizeof(cl_float), buf2, 0, 0, &writeEvt);
status = clFlush(queue);


//等待数据传完成再继续往下执行
waitForEventAndRelease(&writeEvt);
//clWaitForEvents(1, &writeEvt);


cl_mem buffer = clCreateBuffer( context,
CL_MEM_WRITE_ONLY,
BUFSIZE * sizeof(cl_float),
NULL, NULL );


//kernel文件为add.cl
const char * filename  = "add.cl";
std::string  sourceStr;
status = convertToString(filename, sourceStr);
const char * source    = sourceStr.c_str();
size_t sourceSize[]    = { strlen(source) };


//创建程序对象
cl_program program = clCreateProgramWithSource(
context, 
1, 
&source,
sourceSize,
NULL);
//编译程序对象
status = clBuildProgram( program, 1, &device, NULL, NULL, NULL );
if(status != 0)
{
printf("clBuild failed:%d\n", status);
char tbuf[0x10000];
clGetProgramBuildInfo(program, device, CL_PROGRAM_BUILD_LOG, 0x10000, tbuf, NULL);
printf("\n%s\n", tbuf);
return -1;
}


//创建Kernel对象
cl_kernel kernel = clCreateKernel( program, "vecadd", NULL );
//设置Kernel参数
cl_int clnum = BUFSIZE;
clSetKernelArg(kernel, 0, sizeof(cl_mem), (void*) &clbuf1);
clSetKernelArg(kernel, 1, sizeof(cl_mem), (void*) &clbuf2);
clSetKernelArg(kernel, 2, sizeof(cl_mem), (void*) &buffer);


//执行kernel，Range用1维，work items size为BUFSIZE
cl_event ev;
size_t global_work_size = BUFSIZE;
clEnqueueNDRangeKernel( queue,
kernel,
1,
NULL,
&global_work_size,
NULL, 0, NULL, &ev);
status = clFlush( queue );
waitForEventAndRelease(&ev);
    //clWaitForEvents(1, &ev);


//数据拷回host内存
cl_float *ptr;
cl_event mapevt;
ptr = (cl_float *) clEnqueueMapBuffer( queue,
buffer,
CL_TRUE,
CL_MAP_READ,
0,
BUFSIZE * sizeof(cl_float),
0, NULL, &mapevt, NULL );
status = clFlush( queue );
waitForEventAndRelease(&mapevt);
//clWaitForEvents(1, &mapevt);


//结果验证，和cpu计算的结果比较
if(!memcmp(buf, ptr, BUFSIZE))
printf("Verify passed\n");
else printf("verify failed");


if(buf)
free(buf);
if(buf1)
free(buf1);
if(buf2)
free(buf2);


//删除OpenCL资源对象
clReleaseMemObject(clbuf1); 
clReleaseMemObject(clbuf2);
clReleaseMemObject(buffer);
clReleaseProgram(program);
clReleaseCommandQueue(queue);
clReleaseContext(context);
return 0;
}

add.cl代码为：

__kernel void vecadd(__global const float* A, __global const float* B, __global float* C)
{
    int id = get_global_id(0);
    C[id] = A[id] + B[id];
}

自己编译了一份~上传到我的上传资料上了，编译成功，有兴趣的可以下载哦。。【程序名：opencl-test1】

链接为：http://download.csdn.net/detail/huihuiaiyangyang/5820251