Lectures for 9/11/2021

vectorization/f.c

+
+double f(double x)
+{
+  return (4.0 / (1.0 + x*x));
+}
+

vectorization/loop1.c

+void loop(int n, double* C, double* A, double alpha, double* B, double beta)
+{
+  for (int i = 0; i < n; i++)
+    C[i] = alpha * A[i] + beta * B[i];
+}

vectorization/loop2.c

+void loop(int n, double* C, double* A, double alpha, double* B, double beta)
+{
+#pragma GCC ivdep
+  for (int i = 0; i < n; i++)
+    C[i] = alpha * A[i] + beta * B[i];
+}

vectorization/loop3.c

+void loop(int n, double* restrict C, double* restrict A, double alpha, double* restrict B, double beta)
+{
+  for (int i = 0; i < n; i++)
+    C[i] = alpha * A[i] + beta * B[i];
+}

vectorization/loop4.c

+void loop(int n, double* C, double* A, double alpha, double* B, double beta)
+{
+#pragma omp simd safelen(16) simdlen(8) aligned(C,A,B:32)
+  for (int i = 0; i < n; i++)
+    C[i] = alpha * A[i] + beta * B[i];
+}

vectorization/loop5.c

+double dotprod(int n, double* A, double* B)
+{
+  double retval = 0;
+#pragma omp simd reduction(+: retval)
+  for (int i = 0; i < n; i++)
+    retval += A[i] * B[i];
+  return retval;
+}

vectorization/loop6.c

+int loop(int n, double* A, double* B)
+{
+  #pragma omp simd safelen(2)
+  for(int i=2 ; i < n ; i++)
+  {
+    A[i]=A[i-2]+B[i];
+  }
+}

vectorization/pi.c

+#include <stdio.h>
+#include <stdlib.h>
+#include <omp.h>
+
+double f(double x);
+
+int main(int argc, char** argv)
+{
+  int n = atoi(argv[1]);
+  double fH = 1.0/((double)n);
+  double fX, fSum;
+  int i;
+  
+  for (int iter=0; iter<4; ++iter)
+  {
+    double t0 = omp_get_wtime();
+    for (i = 0; i < n; i++)
+    {
+      fX = fH * ((double)i + 0.5);
+      fSum += f(fX);
+    }
+    double t1 = omp_get_wtime();
+    printf("Pi: %2.15f\n", fH * fSum);
+    printf("Time: %f\n", t1-t0);
+  }
+}