add parallel to convolution

src/main.c

@@ -7,7 +7,8 @@
 #include "./modules/headers/Matrix.h"
 
 // main() driver
-int main(int argc, char* argv[]) {
+int main(int argc, char *argv[])
+{
 	MPI_Init(NULL, NULL);
 
 	int my_rank;
@@ -15,39 +16,52 @@ int main(int argc, char* argv[]) {
 
 	int kernel_row, kernel_col, target_row, target_col, num_targets;
 	int *arr_range;
-	Matrix* arr_mat;
+	Matrix *arr_mat;
 	Matrix kernel;
 
 	// reads kernel's row and column and initalize kernel matrix from input
-	if (my_rank == 0) {
+	if (my_rank == 0)
+	{
+		int thread_count = argc == 1 ? 1 : strtol(argv[1], NULL, 10);
 		scanf("%d %d", &kernel_row, &kernel_col);
 		kernel = input_matrix(kernel_row, kernel_col);
 
 		// reads number of target matrices and their dimensions.
 		// initialize array of matrices and array of data ranges (int)
 		scanf("%d %d %d", &num_targets, &target_row, &target_col);
-		arr_mat = (Matrix*)malloc(num_targets * sizeof(Matrix));
+		arr_mat = (Matrix *)malloc(num_targets * sizeof(Matrix));
 		arr_range = malloc(num_targets * sizeof(int));
 
-		// read each target matrix, compute their convolution matrices, and compute their data ranges
-		for (int i = 0; i < num_targets; i++) {
+		for (int i = 0; i < num_targets; i++)
+		{
 			arr_mat[i] = input_matrix(target_row, target_col);
-			arr_mat[i] = convolution(&kernel, &arr_mat[i]);
 		}
 
-		int thread_count = argc == 1 ? 1 : strtol(argv[1], NULL, 10);
+// read each target matrix, compute their convolution matrices, and compute their data ranges
+#pragma omp parallel num_threads(thread_count)
+		{
+			int tid = omp_get_thread_num();
+			int nthreads = omp_get_num_threads();
+			for (int i = tid; i < num_targets; i += nthreads)
+			{
+				arr_mat[i] = convolution(&kernel, &arr_mat[i]);
+			}
+		}
 
-		#pragma omp parallel num_threads(thread_count)
+#pragma omp parallel num_threads(thread_count)
 		{
 			int tid = omp_get_thread_num();
-			int local_n = num_targets/omp_get_num_threads(); // 11
+			int local_n = num_targets / omp_get_num_threads(); // 11
 			int local_a = tid * local_n;
 			int local_b = local_a + local_n;
-			for (int i = local_a; i < local_b; i++) {
+			for (int i = local_a; i < local_b; i++)
+			{
 				arr_range[i] = get_matrix_datarange(&arr_mat[i]);
 			}
-			if (tid == omp_get_num_threads()-1) {
-				for (int i = local_b; i < num_targets; i++) {
+			if (tid == omp_get_num_threads() - 1)
+			{
+				for (int i = local_b; i < num_targets; i++)
+				{
 					arr_range[i] = get_matrix_datarange(&arr_mat[i]);
 				}
 			}
@@ -55,7 +69,8 @@ int main(int argc, char* argv[]) {
 	}
 
 	MPI_Bcast(&num_targets, 1, MPI_INT, 0, MPI_COMM_WORLD);
-	if (my_rank != 0) {
+	if (my_rank != 0)
+	{
 		arr_range = malloc(num_targets * sizeof(int));
 	}
 	MPI_Bcast(arr_range, num_targets, MPI_INT, 0, MPI_COMM_WORLD);
@@ -63,16 +78,17 @@ int main(int argc, char* argv[]) {
 	// sort the data range array
 	merge_sort(arr_range, num_targets, MPI_COMM_WORLD);
 
-	if (my_rank == 0) {
+	if (my_rank == 0)
+	{
 		int median = get_median(arr_range, num_targets);
 		int floored_mean = get_floored_mean(arr_range, num_targets);
 
 		// print the min, max, median, and floored mean of data range array
 		printf("Nilai minimum: %d\nNilai maksimum: %d\nMedian: %d\nRata-rata di-floor: %d\n",
-				arr_range[0],
-				arr_range[num_targets - 1],
-				median,
-				floored_mean);
+			   arr_range[0],
+			   arr_range[num_targets - 1],
+			   median,
+			   floored_mean);
 	}
 
 	free(arr_range);