buffons needle

Hello!
I am just a beginner with OpenCL and I try to implement Buffons needle experiment and get number PI.
here is my code:

#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <math.h>
#include <CL/cl.h>

#define MAX_SOURCE_SIZE (0x100000)

int main(void) {

srand((unsigned)time(NULL));
const int LIST_SIZE = 1024;
int i;

// Load the kernel source code into the array source_str
FILE *fp;
char *source_str;
size_t source_size;

fp = fopen("./kernelagain.cl", "r");
if (!fp) {
    fprintf(stderr, "Failed to load kernel.

");
exit(1);
}
source_str = (char*)malloc(MAX_SOURCE_SIZE);
source_size = fread( source_str, 1, MAX_SOURCE_SIZE, fp);
fclose( fp );

// Get platform and device information
cl_platform_id platform_id = NULL;
cl_device_id device_id = NULL;   
cl_uint ret_num_devices;
cl_uint ret_num_platforms;
cl_int ret = clGetPlatformIDs(1, &platform_id, &ret_num_platforms);
ret = clGetDeviceIDs( platform_id, CL_DEVICE_TYPE_GPU, 1, 
        &device_id, &ret_num_devices);


// Create an OpenCL context
cl_context context = clCreateContext( NULL, 1, &device_id, NULL, NULL, &ret);

// Create a command queue
cl_command_queue command_queue = clCreateCommandQueue(context, device_id, 0, &ret);

// Create memory buffers on the device for each vector 
cl_mem mem_result = clCreateBuffer(context, CL_MEM_WRITE_ONLY, 
        LIST_SIZE * sizeof(int), NULL, &ret);

// Create a program from the kernel source
cl_program program = clCreateProgramWithSource(context, 1, 
        (const char **)&source_str, (const size_t *)&source_size, &ret);

// Build the program
ret = clBuildProgram(program, 1, &device_id, NULL, NULL, NULL);

// Create the OpenCL kernel
cl_kernel kernel = clCreateKernel(program, "needle", &ret);

// Set the arguments of the kernel
ret = clSetKernelArg(kernel, 0, sizeof(cl_mem), (void *)&mem_result);

// Execute the OpenCL kernel on the list
size_t global_item_size = LIST_SIZE; // Process the entire lists
size_t local_item_size = 64; // Divide work items into groups of 64

ret = clEnqueueNDRangeKernel(command_queue, kernel, 1, NULL, 
        &global_item_size, &local_item_size, 0, NULL, NULL);

// Read the memory buffer C on the device to the local variable C
int *result = (int*)malloc(sizeof(int)*LIST_SIZE);
ret = clEnqueueReadBuffer(command_queue, mem_result, CL_TRUE, 0, 
        LIST_SIZE * sizeof(int), result, 0, NULL, NULL);

// Display the result to the screen
for(i = 0; i &lt; LIST_SIZE; i++)
    {printf("%d

", result[i]);}
// Clean up
ret = clFlush(command_queue);
ret = clFinish(command_queue);
ret = clReleaseKernel(kernel);
ret = clReleaseProgram(program);
ret = clReleaseMemObject(mem_result);
ret = clReleaseCommandQueue(command_queue);
ret = clReleaseContext(context);
free(result);
return 0;
}

and second part is:

__kernel void needle(__global int *out) {

int i = get_global_id(0);
const double PI = 3.14159265358979323846;
srand((unsigned)time(NULL));
double d=5;   //size between lines
double l=2;   //size of needle
double angle;
double position;
    double f = (double)rand() / RAND_MAX;
    double number = f * (2*PI);
angle = number;
    f = (double)rand() / RAND_MAX;
    number = f * (d);
position = number;
if( ( ( position + l*sin(angle)/2 &gt;= d ) && ( position - l*sin(angle)/2 &lt;= d ) ) || 
    ( ( position + l*sin(angle)/2 &gt;= 0 ) && ( position - l*sin(angle)/2 &lt;= 0 ) ) )
	{
		out[i]=1;
	} 
else out[i] = 0;

}

So can someone tell me why array “result” has values 0,1,2,3 … 1023 when I think it should be only 0 or 1.
Please help me cause I have no idea where is problem and sorry for my english :slight_smile:

I doubt that your kernel can compile: srand(), time() and rand() are not part of OpenCL C.

Thanks u :slight_smile: Now it works
My PI is lim->3.14
and thats my code :

#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <math.h>
#include <CL/cl.h>

#define MAX_SOURCE_SIZE (0x100000)
#define PI 3.14159265358979323846

double randZeroToOne(double fMin,double fMax)
{
double f = (double)rand() / RAND_MAX;
return fMin + f * (fMax - fMin);
}

int main(void) {

srand(time(NULL));
const int LIST_SIZE = 256 * 100000;
int i;

float *A = (float*)malloc(sizeof(float)*LIST_SIZE);
float *B = (float*)malloc(sizeof(float)*LIST_SIZE);
for(i = 0; i &lt; LIST_SIZE; i++) {
    A[i] = randZeroToOne(0,2*PI);
B[i] = randZeroToOne(0,5);;
}

// Load the kernel source code into the array source_str
FILE *fp;
char *source_str;
size_t source_size;

fp = fopen("./kernelagain.cl", "r");
if (!fp) {
    fprintf(stderr, "Failed to load kernel.

");
exit(1);
}
source_str = (char*)malloc(MAX_SOURCE_SIZE);
source_size = fread( source_str, 1, MAX_SOURCE_SIZE, fp);
fclose( fp );

// Get platform and device information
cl_platform_id platform_id = NULL;
cl_device_id device_id = NULL;   
cl_uint ret_num_devices;
cl_uint ret_num_platforms;
cl_int ret = clGetPlatformIDs(1, &platform_id, &ret_num_platforms);
ret = clGetDeviceIDs( platform_id, CL_DEVICE_TYPE_GPU, 1, 
        &device_id, &ret_num_devices);


// Create an OpenCL context
cl_context context = clCreateContext( NULL, 1, &device_id, NULL, NULL, &ret);

// Create a command queue
cl_command_queue command_queue = clCreateCommandQueue(context, device_id, 0, &ret);

// Create memory buffers on the device for each vector 
cl_mem mem_result = clCreateBuffer(context, CL_MEM_WRITE_ONLY, 
        LIST_SIZE * sizeof(int), NULL, &ret);
cl_mem a_mem_obj = clCreateBuffer(context, CL_MEM_READ_ONLY, 
        LIST_SIZE * sizeof(float), NULL, &ret);
cl_mem b_mem_obj = clCreateBuffer(context, CL_MEM_READ_ONLY, 
        LIST_SIZE * sizeof(float), NULL, &ret);


//copy A to memory buffers
ret = clEnqueueWriteBuffer(command_queue, a_mem_obj, CL_TRUE, 0,
        LIST_SIZE * sizeof(float), A, 0, NULL, NULL);
ret = clEnqueueWriteBuffer(command_queue, b_mem_obj, CL_TRUE, 0,
        LIST_SIZE * sizeof(float), B, 0, NULL, NULL);

// Create a program from the kernel source
cl_program program = clCreateProgramWithSource(context, 1, 
        (const char **)&source_str, (const size_t *)&source_size, &ret);

// Build the program
ret = clBuildProgram(program, 1, &device_id, NULL, NULL, NULL);

// Create the OpenCL kernel
cl_kernel kernel = clCreateKernel(program, "needle", &ret);

// Set the arguments of the kernel
ret = clSetKernelArg(kernel, 0, sizeof(cl_mem), (void *)&mem_result);
ret = clSetKernelArg(kernel, 1, sizeof(cl_mem), (void *)&a_mem_obj);
ret = clSetKernelArg(kernel, 2, sizeof(cl_mem), (void *)&b_mem_obj);

// Execute the OpenCL kernel on the list
size_t global_item_size = LIST_SIZE; // Process the entire lists
size_t local_item_size = 64; // Divide work items into groups of 64

ret = clEnqueueNDRangeKernel(command_queue, kernel, 1, NULL, 
        &global_item_size, &local_item_size, 0, NULL, NULL);

// Read the memory buffer C on the device to the local variable C
int *result = (int*)malloc(sizeof(int)*LIST_SIZE);
ret = clEnqueueReadBuffer(command_queue, mem_result, CL_TRUE, 0, 
        LIST_SIZE * sizeof(int), result, 0, NULL, NULL);
double total=0,crossed=0;
for(i = 0; i &lt; LIST_SIZE; i++)
{

if (result[i]==1) crossed++;
total++;
}

double numberpi= (2 * total) / (5 * crossed);
printf("pi=%lf

",numberpi);
printf("total :%.0lf cross :%.0lf
",total,crossed);

ret = clFlush(command_queue);
ret = clFinish(command_queue);
ret = clReleaseKernel(kernel);
ret = clReleaseProgram(program);
ret = clReleaseMemObject(mem_result);
ret = clReleaseCommandQueue(command_queue);
ret = clReleaseContext(context);
free(result);
free(A);
free(B);
return 0;

}

and

__kernel void needle(__global int *out,__global const float *A,__global const float *B) {

int i = get_global_id(0);
const double PI = 3.14159265358979323846;
double d=5;   //size between lines
double l=2;   //size of needle
double angle = A[i];
double position = B[i];

if( ( ( position + l*sin(angle)/2 &gt;= d ) && ( position - l*sin(angle)/2 &lt;= d ) ) || 
    ( ( position + l*sin(angle)/2 &gt;= 0 ) && ( position - l*sin(angle)/2 &lt;= 0 ) ) )
	{
		out[i]=1;
	} 
else out[i] = 0;

}