Problem Statement
-----------------

        1) For the code given below
                a) Identify  the type  of dependence  between statements  within
                   loop body (RAW, WAR,  WAW, loop carried/loop independent) and
                   produce result  in the  form array references,  source, sink,
                   type of dependence and dependence graph.
                b) Compile   enabling   vectorization/parallelization   option. 
                   Verify your observations using gcc-4.7.2's dumps.            
                          

        2) Compare   the  execution   time   of   unoptimized,  vectorized   and
           parallelized code.


        Note: "perf"  tool  should  be  installed  on  your  system  to  measure
              execution time. (there are other profiler's apart from "perf", you
              are welcome to try them)
 
        Command to measure execution time: 
           perf stat -e task-clock -r 50 ./<output_file>.out -o>/dev/null

--------------------------------------------------------------------------------

Case 1:
-------

        #include"stdio.h"
        int a[5000000],b[5000000];
        int main()
        {
                int i;
                for(i=0;i<5000000;i++) {
                        a[i] = b[i];
                }
                return 0;
        }

--------------------------------------------------------------------------------
                
Case 2:
-------

        #include"stdio.h"
        int a[256],b[256];
        int main ()
        {
                int i;
                for (i=0; i<203; i++) {
                        a[i+4] = b[i] + 2;    /* S1 */
                        b[i+2] = a[i+1] +1;   /* S2 */        
                }
        return 0;
        }