diff --git a/Makefile b/Makefile
index 9243cc03bb476215939655b92cca1edebe1d8365..ff6c8b2a7c6575145b68de97fedc384741ffdf73 100644
--- a/Makefile
+++ b/Makefile
@@ -1,2 +1,3 @@
 radix_sort:
+	gcc -g -Wall -o serial_radix_sort src/serial_radix_sort.c -lm
 	nvcc src/radix_sort_parallel.cu -o radix_sort
diff --git a/src/radix_sort_parallel.cu b/src/radix_sort_parallel.cu
index e857a870b26041437b6372998334030e630321ab..fd4227e43b9ec41fbbb1be4e697ab976ef5e2243 100644
--- a/src/radix_sort_parallel.cu
+++ b/src/radix_sort_parallel.cu
@@ -55,15 +55,15 @@ __global__ void assign_index(int* arr, int* arr_idx, int* down_sweep_i, int* up_
 }
 
 __global__ void assign_flags(int* arr, int n, int idx, int* flags) {
-		for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < n; i += blockDim.x * gridDim.x) {
-			flags[i] = ((arr[i] & 1 << idx) == 1 << idx) ? 0 : 1;
-		}
+	for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < n; i += blockDim.x * gridDim.x) {
+		flags[i] = ((arr[i] & 1 << idx) == 1 << idx) ? 0 : 1;
+	}
 }
 
 __global__ void scatter(int* arr, int* arr_temp, int* arr_idx, int n) {
-		for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < n; i += blockDim.x * gridDim.x) {
-				arr[arr_idx[i]] = arr_temp[i];
-		}
+	for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < n; i += blockDim.x * gridDim.x) {
+		arr[arr_idx[i]] = arr_temp[i];
+	}
 }
 
 void count_sort(int* d_arr, int n, int idx) {
@@ -129,11 +129,11 @@ int main(int argc, char *argv[]) {
 
     // Move sorting CPU
     int* d_arr;
-		cudaMalloc(&d_arr, sizeof(int) * n);
-		cudaMemcpy(d_arr, arr, sizeof(int) * n, cudaMemcpyHostToDevice);
+	cudaMalloc(&d_arr, sizeof(int) * n);
+	cudaMemcpy(d_arr, arr, sizeof(int) * n, cudaMemcpyHostToDevice);
     radix_sort(d_arr, n);
     cudaMemcpy(arr, d_arr, n * sizeof(int), cudaMemcpyDeviceToHost);
-		cudaFree(d_arr);
+	cudaFree(d_arr);
 
     end = clock();
     cpu_time_used = ((double) (end - start)) * 1000000 / CLOCKS_PER_SEC;