Lab4 updates

Lab4/matrix_gpu.cu

@@ -6,17 +6,21 @@
 #include <stdio.h>
 #include <cmath>
 
-const int N = 32;
-const int blocksize = 64;
-const int gridsize = 4;
+const int N = 1024;
+const int blocksize = 32; //threads
+const int gridsize = 32; //blocks
 
 __global__
-void add(float* result, float *c, float *d)
+void add(float* result, float *c, float *d, int N)
 {
 	int idx = blockIdx.x * blockDim.x + threadIdx.x;
 	int idy = blockIdx.y * blockDim.y + threadIdx.y;
-	int index = idy * blockDim.x + idx;
-	result[index] = c[index] + d[index];
+	int index = idx + idy * N;
+	
+	if (idx < N && idy < N) {
+			result[index] = c[index] + d[index];
+	}
+
 }
 
 int main()
@@ -51,7 +55,7 @@ int main()
   cudaEventCreate(&startEvent);
 	cudaEventCreate(&finEvent);
 	cudaEventRecord(startEvent, 0);
-	add<<<dimGrid, dimBlock>>>(result_cuda, c, d);
+	add<<<dimGrid, dimBlock>>>(result_cuda, c, d, N);
 	cudaThreadSynchronize();
 	cudaEventRecord(finEvent, 0);
 	cudaEventSynchronize(finEvent);