source: CIVL/examples/translation/openclversion2.1/square.cvl@ e18202fd

//Forget the program

#include "openCLshared.cvl"
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

$input int NUM_DEVICES;
$input int MAX_NUM_DEVICES;
$assume 0 < NUM_DEVICES && NUM_DEVICES < MAX_NUM_DEVICES;

$input int DATA_SIZE;
$input int MAX_DATA_SIZE;
$assume 0 < DATA_SIZE && DATA_SIZE < MAX_NUM_DEVICES;

$input int LOCAL;
//Didn't initialize variables here
typedef struct
{
  //Variables for kernels
  
  int * input;
  int * output;
  int count;
  
}args;


cl_kernel clCreateKernel(args * argument)
{
  cl_kernel kernel;
  kernel.arguments = argument;
  
  return kernel;
}
/*
int clEnqueueNDRangeKernel(cl_command_queue commands, cl_kernel kernel, int global, int local)
{
  $proc procs[global/local];
  for(int i = 0; i < global/local; i++)
  {
    param[i].workgroup = i;
    //procs[i] = $spawn square(param[i].global_id, param[i].input, param[i].output, param[i].count);
    procs[i] = $spawn workfunc(local, global, param[i]);
  }
  
  for(int i = 0; i < global/local; i++)
  {
    $wait(procs[i]);
  }
  
  return CL_SUCCESS;
}
*/
//kernel
void square(int workgroup, int global_id, int local_id, float* input, float* output, int count)
{
  //int i = get_global_id(0);
  int i = global_id;
  if (i < count)
  {
    output[i] = input[i] * input[i];
    //printf("output[%d] is %d\n", i, output[i]);
  }
}
/*
void workfunc(size_t local, size_t global, kernel param)
{
  for(int i = local * param.workgroup; i < local * param.workgroup + local; i++)
  {
    param.local_id = i % local;
    param.global_id = i;
    //printf("My workgroup id is %d, my global id is %d, my local id is %d\n", param.workgroup, param.global_id, param.local_id);
    square(param.workgroup, param.global_id, param.local_id, param.input, param.output, param.count);
  }
}
*/
int main(int argc, char** argv)
{
  args * arguments;
  arguments = (args*)malloc(sizeof(args));

  float data[DATA_SIZE];              // original data set given to device
  float results[DATA_SIZE];           // results returned from device
  unsigned int correct;               // number of correct results returned
  
  size_t global;                      // global domain size for our calculation
  size_t local;                       // local domain size for our calculation
  
  cl_device_id device_id;             // compute device id 
  cl_context context;                 // compute context
  cl_command_queue commands;          // compute command queue
  //cl_program program;                 // compute program
  cl_kernel kernel;                   // compute kernel
  
  
  float * input;                       // device memory used for the input array
  float * output;                      // device memory used for the output array

  unsigned int count = DATA_SIZE;
  for(int i = 0; i < count; i++)
  {
    data[i] = i;
  }
  
  int err = clGetDeviceIDs(1, &device_id);
  
  //ignore clCreateContext for now, until we get an example that uses multiple ones
  
  //clCreateCommandQueue, could use context later
  commands = clCreateCommandQueue(device_id);
  
  //clCreateProgram is far different from the real version, this just stores parameters for the kernel
  //In order to make this clear, it is clCreateProgram and not something like clCreateProgramFromSource, which actually exists in openCL code
  //program = clCreateProgram(arguments);
  
  kernel = clCreateKernel(arguments);
  //printf("%s", kernel);
  
  //comes from clCreateBuffer
  input = (float *) malloc(sizeof(float) * count);
  output = (float *) malloc(sizeof(float) * count);
  
  
  
  memcpy(input, data, sizeof(float) * count);
  //clEnqueueWriteBuffer
  
  /*
  err = 0;
  err = clSetKernelArg(kernel, 0, sizeof(cl_mem), &input);
  err = clSetKernelArg(kernel, 1, sizeof(cl_mem), &output);
  err = clSetKernelArg(kernel, 2, sizeof(unsigned int), &count);
  */
  //kernel.arguments.input = input;
  
  //kernel.arguments->input;
  
  memcpy(((args *)kernel.arguments)->input, input, sizeof(float));
  //memcpy(((args*)kernel.arguments)->output, output, sizeof(float));
  //memcpy(((args*)kernel.arguments)->count, count, sizeof(int));
  ((args*)kernel.arguments)->count = count;
  
  
  //clGetKernelWorkGroupInfo
  local = LOCAL;
  
  global = count;
  /*
    commands holds the "order" of devices
    kernel holds program, which holds variables
    offset not implemented
  */
  //err = clEnqueueNDRangeKernel(commands, kernel, global, local);
  
  free(input);
  free(output);
  free(arguments);
  return 0;
}