BZOJ 4332 FFT+快速幂

//By SiriusRen
#include <cmath>
#include <cstdio>
#include <algorithm>
using namespace std;
const int N=32888;
const double pi=acos(-1);
int n,M,P,A,O,S,U,L,R[N],F[N],g[N],p[N],t[N],ans;
struct cplxd{double x,y;cplxd(){}cplxd(double X,double Y){x=X,y=Y;}}ca[N],cb[N],cc[N];
cplxd operator+(cplxd a,cplxd b){return cplxd(a.x+b.x,a.y+b.y);}
cplxd operator-(cplxd a,cplxd b){return cplxd(a.x-b.x,a.y-b.y);}
cplxd operator*(cplxd a,cplxd b){return cplxd(a.x*b.x-a.y*b.y,a.x*b.y+a.y*b.x);}
cplxd operator/(cplxd a,int b){return cplxd(a.x/b,a.y/b);}
void FFT(cplxd *a,int f){
    for(int i=0;i<n;i++)if(i<R[i])swap(a[i],a[R[i]]);
    for(int i=1;i<n;i<<=1){
        cplxd wn=cplxd(cos(pi/i),f*sin(pi/i));
        for(int j=0;j<n;j+=(i<<1)){
            cplxd w(1,0);
            for(int k=0;k<i;k++,w=w*wn){
                cplxd x=a[j+k],y=w*a[j+k+i];
                a[j+k]=x+y,a[j+k+i]=x-y;
            }
        }
    }
    if(f==-1)for(int i=0;i<n;i++)a[i]=a[i]/n;
}
void Pw(int *a,int *b,int *c){
    for(int i=0;i<n;i++)ca[i]=cplxd(a[i],0);
    for(int i=0;i<n;i++)cb[i]=cplxd(b[i],0);
    FFT(ca,1),FFT(cb,1);
    for(int i=0;i<n;i++)cc[i]=ca[i]*cb[i];
    FFT(cc,-1);
    for(int i=0;i<=M;i++)c[i]=((int)(0.3+cc[i].x))%P;
}
void pow(int k){
    if(k==1)return;
    pow(k>>1);
    Pw(p,g,t),Pw(g,g,g);
    for(int i=0;i<=M;i++)(p[i]+=t[i])%=P;
    if(k&1){
        Pw(g,F,g);
        for(int i=0;i<=M;i++)(p[i]+=g[i])%=P;
    }
}
signed main(){
    scanf("%d%d%d%d%d%d",&M,&P,&A,&O,&S,&U);
    for(n=1;n<=M*2;n<<=1)L++;
    for(int i=1;i<=n;i++)R[i]=(R[i>>1]>>1)|((i&1)<<(L-1));
    for(int i=1;i<=M;i++)p[i]=g[i]=F[i]=(i*i*O+S*i+U)%P;
    pow(A),printf("%d\n",p[M]);
}