C(m,n)%P

#include<cstdio>
#include<cstdlib>
#include<iostream>
#include<fstream>
#include<algorithm>
#include<cstring>
#include<string>
#include<cmath>
#include<queue>
#include<stack>
#include<map>
#include<utility>
#include<set>
#include<bitset>
#include<vector>
#include<functional>

using namespace std;

typedef long long LL;
typedef double DB;
typedef pair<int,int> PII;

#define mmst(a,v) memset(a,v,sizeof(a))
#define mmcy(a,b) memcpy(a,b,sizeof(a))
#define re(i,a,b) for(i=a;i<=b;i++)
#define red(i,a,b) for(i=a;i>=b;i--)
#define fi first
#define se second

template<class T>inline T sqr(T x){return x*x;}
template<class T>inline void upmin(T &t,T tmp){if(t>tmp)t=tmp;}
template<class T>inline void upmax(T &t,T tmp){if(t<tmp)t=tmp;}

const DB EPS=1e-9;
inline int dblcmp(DB x){if(abs(x)<EPS)return 0;return(x>0)?1:-1;}

inline void SetOpen(string s)
{
freopen((s+".in").c_str(),"r",stdin);
freopen((s+".out").c_str(),"w",stdout);
}

inline int Getin_Int()
{
int res=0,flag=1;char z;
for(z=getchar();z!=EOF && z!=‘-‘ && !isdigit(z);z=getchar());
if(z==EOF)return 0;
if(z==‘-‘){flag=-flag;z=getchar();}
for(;z!=EOF && isdigit(z);res=res*10+z-‘0‘,z=getchar());
return res*flag;
}
inline LL Getin_LL()
{
LL res=0,flag=1;char z;
for(z=getchar();z!=EOF && z!=‘-‘ && !isdigit(z);z=getchar());
if(z==EOF)return 0;
if(z==‘-‘){flag=-flag;z=getchar();}
for(;z!=EOF && isdigit(z);res=res*10+z-‘0‘,z=getchar());
return res*flag;
}

const LL maxpici=100000;
const LL maxcnt=100;

LL n,m,P;
LL cnt,p[maxcnt+10],c[maxcnt+10],mod[maxcnt+10],euler[maxcnt+10],fac[maxcnt+10][maxpici+10];
LL ans;

inline LL power(LL a,LL b,LL Mod)
{
LL x=1,y=a;
while(b!=0){if(b&1)x=x*y%Mod;if(b==1)break;y=y*y%Mod;b/=2;}
return x;
}

inline LL calfac(LL N,LL id,LL &g)
{
if(N<p[id])return fac[id][N];
LL s=N/mod[id],t=N%mod[id],res=1;
res=res*power(fac[id][mod[id]-1],s,mod[id])%mod[id];
res=res*fac[id][t]%mod[id];
g+=N/p[id];
res=res*calfac(N/p[id],id,g)%mod[id];
return res;
}

inline LL solve(LL id)
{
LL g1=0,g2=0,g3=0;
LL fenzi=calfac(m,id,g1),fenmo=calfac(n,id,g2)*calfac(m-n,id,g3)%mod[id];
return fenzi*power(fenmo,euler[id]-1,mod[id])%mod[id]*power(p[id],g1-g2-g3,mod[id])%mod[id];
}

inline void extend_gcd(LL a,LL &x,LL b,LL &y)
{
if(b==0){x=1;y=0;return;}
LL dx,dy;
extend_gcd(b,dx,a%b,dy);
x=dy;
y=dx-a/b*dy;
}

int main()
{
SetOpen("program");
LL i,j,temp;
m=Getin_Int();n=Getin_Int();P=Getin_Int();
if(m<n){puts("0\n");return 0;}
temp=P;
re(i,2,maxpici)if(temp%i==0)
{
++cnt;
p[cnt]=i;
mod[cnt]=1;
while(temp%i==0){c[cnt]++;mod[cnt]*=i;temp/=i;}
euler[cnt]=mod[cnt]/p[cnt]*(p[cnt]-1);
fac[cnt][0]=1;
re(j,1,mod[cnt]-1)fac[cnt][j]=(j%p[cnt]==0)?fac[cnt][j-1]:fac[cnt][j-1]*j%mod[cnt];
}
ans=0;
re(i,1,cnt)
{
temp=solve(i);
LL a=P/mod[i],x,b=mod[i],y;
extend_gcd(a,x,b,y);
ans=(ans+temp*a%P*x%P)%P;
}
ans=(ans%P+P)%P;
cout<<ans<<endl;
return 0;
}