#include "cuda_runtime.h"
#include "device_launch_parameters.h"
#include <stdio.h>
#include <math.h>
#include "graphics.h"
#include "tryf.h"
#define ImageHeight 500
#define ImageWidth 500


__global__ void addKernel(int *a)
{

    double MinRe = -2.0;
    double MaxRe = 0.5;
    double MinIm = -1.2;
    double MaxIm = MinIm+(MaxRe-MinRe)*ImageHeight/ImageWidth;
    double Re_factor = (MaxRe-MinRe)/(ImageWidth-1);
    double Im_factor = (MaxIm-MinIm)/(ImageHeight-1);
    int MaxIterations = 12;
    bool isInside=true;
    unsigned int go_upto_X=ImageWidth;


    int x=0,n,count=1;
    int col=threadIdx.y;
    int row=threadIdx.x;
    //cuPrintf("Hi");

unsigned int y=col*23+row;
  if(y>=500) goto end;//Each row will have a separate thread.
 //  for(;count<6;y++)
//   {
     double c_im = MaxIm - y*Im_factor;
    for(x=0; x<go_upto_X; ++x)
    {
        double c_re = MinRe + x*Re_factor;

        double Z_re = c_re, Z_im = c_im;isInside = true;

        for(n=0; n<MaxIterations; ++n)
        {
            double Z_re2 = Z_re*Z_re, Z_im2 = Z_im*Z_im;
            if(Z_re2 + Z_im2 > 4)
            {
                isInside = false;
                break;
            }
            Z_im = 2*Z_re*Z_im + c_im;
            Z_re = Z_re2 - Z_im2 + c_re;
        }
        *(a+y*ImageWidth*3+x*3+0)=x;
        *(a+y*ImageWidth*3+x*3+1)=y;
        if(!isInside) *(a+y*ImageWidth*3+x*3+2)=n;
        else{ *(a+y*ImageWidth*3+x*3+2)=MaxIterations-1;}

    }//count++;
 //  }
end:

}



int main()
{
   int flag=0,count=1;
    initwindow(ImageWidth,ImageHeight,"MandelBrot Set");
    int *points=(int *)malloc (ImageHeight*ImageWidth*3*sizeof(int));
    /*for(int i=0;i<ImageHeight;i++)
    {
        for(int j=0;j<ImageWidth;j++)
        {
            *(points+i*ImageWidth*3+j*3+0)=j;
            *(points+i*ImageWidth*3+j*3+1)=i;
            *(points+i*ImageWidth*3+j*3+2)=5;
        }
    }*/



    int *dev_points=0;
     cudaMalloc((void **) &dev_points , ImageHeight*ImageWidth*3*sizeof (int) ) ;

     cudaMemcpy(dev_points, points, ImageHeight*ImageWidth*3 * sizeof(int), cudaMemcpyHostToDevice);

 // Launch a kernel on the GPU with one thread for each element.
     dim3 dimGrid(1,1);
     dim3 dimBlock(23,23);
     addKernel<<<dimGrid, dimBlock>>>(dev_points);
//   addKernel<<<1,500>>>(dev_points);

         cudaMemcpy( points, dev_points, ImageHeight*ImageWidth*3, cudaMemcpyDeviceToHost );
        int x,y,color;
         for(int i=0;i<ImageHeight;i++)
    {
        for(int j=0;j<ImageWidth;j++)
        {
            x=*(points+i*ImageWidth*3+j*3+0);
            y=*(points+i*ImageWidth*3+j*3+1);
            color=*(points+i*ImageWidth*3+j*3+2);
        //    if(y<0) {flag=1;count++;break;}
            printf("%d %d %d %d %d %d %d %dn",*(points+i*ImageWidth*3+j*3+0),*(points+i*ImageWidth*3+j*3+1),*(points+i*ImageWidth*3+j*3+2),i*ImageWidth*3+j*3+0,i,j,ImageHeight,ImageWidth);
            switch(color)
                            {
                                case 0:putpixel(x,y,WHITE);break;
                                case 1:putpixel(x,y,YELLOW);break;
                                case 2:putpixel(x,y,BLUE);break;
                                case 3:putpixel(x,y,CYAN);break;
                                case 4:putpixel(x,y,GREEN);break;
                                case 5:putpixel(x,y,RED);break;
                                case 6:putpixel(x,y,MAGENTA);break;
                                case 7:putpixel(x,y,LIGHTBLUE);break;
                                case 8:putpixel(x,y,LIGHTGREEN);break;
                                case 9:putpixel(x,y,LIGHTCYAN);break;
                                case 10:putpixel(x,y,LIGHTRED);break;
                                case 11:putpixel(x,y,LIGHTMAGENTA);break;
                                default:putpixel(x,y,BLACK);break;
                            }

        }//if(flag && count>5)break;
    }
    // Check for any errors launching the kernel
   getch();
}