t_map_entry *ft_map_get_p(t_map map, void *key, t_simple_hash_func h_f)
{
t_map_entry *entry;
t_uint64 hash;
hash = h_f(key);
entry = get(map[hash % MAP_TREE_SIZE], hash);
return (entry);
}
Image<float> transfer(Image<float> &image, TransferFunction &f) {
Image<float> h_f(f.size);
memcpy(h_f.data(),&f.values[0], f.size*sizeof(float));
Image<float> output(image.width(), image.height());
hl_transfer_histogram(image, h_f, f.source_mini, f.source_maxi, f.target_mini, f.target_maxi, output);
return output;
}
Image<float> transfer(Image<float> &image, vector<TransferFunction> &fs) {
Image<float> h_f(fs[0].size+4,3);
float *pData = h_f.data();
int idx = 0;
for (size_t i = 0; i < fs.size(); ++i) {
pData[idx++] = fs[i].source_mini;
pData[idx++] = fs[i].source_maxi;
pData[idx++] = fs[i].target_mini;
pData[idx++] = fs[i].target_maxi;
memcpy(pData+idx,&fs[i].values[0], fs[i].size*sizeof(float));
idx += fs[i].size;
}
Image<float> output(image.width(), image.height(),3);
hl_transfer_histogram_color_float(image, h_f, output);
return output;
}
int main(void)
{
std::vector<float> h_a(LENGTH); // a vector
std::vector<float> h_b(LENGTH); // b vector
std::vector<float> h_c (LENGTH, 0xdeadbeef); // c vector (result)
std::vector<float> h_d (LENGTH, 0xdeadbeef); // d vector (result)
std::vector<float> h_e (LENGTH); // e vector
std::vector<float> h_f (LENGTH, 0xdeadbeef); // f vector (result)
std::vector<float> h_g (LENGTH); // g vector
cl::Buffer d_a; // device memory used for the input a vector
cl::Buffer d_b; // device memory used for the input b vector
cl::Buffer d_c; // device memory used for the output c vector
cl::Buffer d_d; // device memory used for the output d vector
cl::Buffer d_e; // device memory used for the input e vector
cl::Buffer d_f; // device memory used for the output f vector
cl::Buffer d_g; // device memory used for the input g vector
// Fill vectors a and b with random float values
int count = LENGTH;
for(int i = 0; i < count; i++)
{
h_a[i] = rand() / (float)RAND_MAX;
h_b[i] = rand() / (float)RAND_MAX;
h_e[i] = rand() / (float)RAND_MAX;
h_g[i] = rand() / (float)RAND_MAX;
}
try
{
// Create a context
cl::Context context(DEVICE);
// Load in kernel source, creating a program object for the context
cl::Program program(context, util::loadProgram("vadd.cl"), true);
// Get the command queue
cl::CommandQueue queue(context);
// Create the kernel functor
auto vadd = cl::make_kernel<cl::Buffer, cl::Buffer, cl::Buffer>(program, "vadd");
d_a = cl::Buffer(context, begin(h_a), end(h_a), true);
d_b = cl::Buffer(context, begin(h_b), end(h_b), true);
d_e = cl::Buffer(context, begin(h_e), end(h_e), true);
d_g = cl::Buffer(context, begin(h_g), end(h_g), true);
d_c = cl::Buffer(context, CL_MEM_READ_WRITE, sizeof(float) * LENGTH);
d_d = cl::Buffer(context, CL_MEM_READ_WRITE, sizeof(float) * LENGTH);
d_f = cl::Buffer(context, CL_MEM_WRITE_ONLY, sizeof(float) * LENGTH);
vadd(
cl::EnqueueArgs(
queue,
cl::NDRange(count)),
d_a,
d_b,
d_c);
vadd(
cl::EnqueueArgs(
queue,
cl::NDRange(count)),
d_e,
d_c,
d_d);
vadd(
cl::EnqueueArgs(
queue,
cl::NDRange(count)),
d_g,
d_d,
d_f);
cl::copy(queue, d_f, begin(h_f), end(h_f));
// Test the results
int correct = 0;
float tmp;
for(int i = 0; i < count; i++)
{
tmp = h_a[i] + h_b[i] + h_e[i] + h_g[i]; // assign element i of a+b+e+g to tmp
tmp -= h_f[i]; // compute deviation of expected and output result
if(tmp*tmp < TOL*TOL) // correct if square deviation is less than tolerance squared
correct++;
else {
printf(" tmp %f h_a %f h_b %f h_e %f h_g %f h_f %f\n",tmp, h_a[i], h_b[i], h_e[i], h_g[i], h_f[i]);
}
}
// summarize results
printf("C = A+B+E+G: %d out of %d results were correct.\n", correct, count);
}
catch (cl::Error err) {
std::cout << "Exception\n";
std::cerr
<< "ERROR: "
<< err.what()
<< std::endl;
}
}