#include<bits/stdc++.h>
#define fi first
#define se second
#define eb emplace_back
#define mp make_pair
using namespace std;
typedef long double ld;
typedef long long ll;
typedef unsigned long long ull;
typedef __int128 i128;
template<typename T,typename U>
T ceil(T x, U y) {return (x>0?(x+y-1)/y:x/y);}
template<typename T,typename U>
T floor(T x, U y) {return (x>0?x/y:(x-y+1)/y);}
template<class T,class S>
bool chmax(T &a,const S b) {return (a<b?a=b,1:0);}
template<class T,class S>
bool chmin(T &a,const S b) {return (a>b?a=b,1:0);}
int popcnt(int x) {return __builtin_popcount(x);}
int popcnt(ll x) {return __builtin_popcountll(x);}
int topbit(int x) {return (x==0?-1:31-__builtin_clz(x));}
int topbit(ll x) {return (x==0?-1:63-__builtin_clzll(x));}
int lowbit(int x) {return (x==0?-1:__builtin_ctz(x));}
int lowbit(ll x) {return (x==0?-1:__builtin_ctzll(x));}

#define int long long
#define rep(i,a,b) for(int i=(a);i<=(b);i++)
#define per(i,a,b) for(int i=(a);i>=(b);i--)
typedef pair<int,int> pii; 
typedef vector<int> vi;
typedef vector<pii> vp;
typedef tuple<int,int,int> tiii;
int read() {
  int x=0,w=1; char c=getchar(); 
  while(!isdigit(c)) {if(c=='-') w=-1; c=getchar();}
  while(isdigit(c)) {x=x*10+(c-'0'); c=getchar();}
  return x*w;
}

const int N=5e4+5;
int n,m,b[N][15],v[N],s[N],ans,swp;
int dx[4]={1,0,-1,0};
int dy[4]={0,1,0,-1};
vi a[N],vst[N];
vector<vi> f[N],t[N];
vp tmp;

int calc(int l,int r,int u,int d,int x,int y) {
  rep(i,u,d) rep(j,l,r) {
    tmp.clear();
    rep(k,0,3) {
      int nx=i+dx[k], ny=j+dy[k];
      if(u<=nx&&nx<=d&&l<=ny&&ny<=r)
        tmp.eb(a[nx][ny],k);
    }
    if(tmp.size()) {
      sort(tmp.begin(),tmp.end());
      for(auto [w,k]:tmp) if(w>a[i][j]) {
        int nx=i+dx[k], ny=j+dy[k];
        vst[nx][ny]++; break;
      }
    }
  }
  int res=!vst[x][y];
  rep(i,u,d) rep(j,l,r) vst[i][j]=0;
  return res;
}

signed main() {
  n=read(), m=read();
  if(n>m) swp=1, swap(n,m);
  rep(i,0,n+1) a[i].resize(m+2), vst[i].resize(m+2);
  rep(i,0,n+1) {
    f[i].resize(m+2), t[i].resize(m+2);
    rep(j,0,m+1) f[i][j].resize(85), t[i][j].resize(15);
  }
  if(!swp) rep(i,1,n) rep(j,1,m) a[i][j]=read();
  else rep(j,1,m) rep(i,1,n) a[i][j]=read();
  rep(l,0,2) rep(r,0,2) rep(u,0,2) rep(d,0,2) {
    int p=((d*3+u)*3+r)*3+l;
    rep(i,1,n) rep(j,1,m) {
      int lb=max(1ll,j-l), rb=min(m,j+r);
      int ub=max(1ll,i-u), db=min(n,i+d);
      f[i][j][p]=calc(lb,rb,ub,db,i,j);
    }
  }
  rep(i,1,n) rep(j,1,m) rep(k,0,8) {
    t[i][j][k]=t[i-1][j][k]+f[i][j][72+k];
  }
  rep(u,1,n) rep(d,u,n) {
    rep(j,1,m) rep(k,0,8) b[j][k]=0;
    if(d-u+1<=3) {
      rep(i,u,d) rep(j,1,m) {
        int pu=i-u, pd=d-i, p=(pd*3+pu)*9;
        rep(k,0,8) b[j][k]+=f[i][j][p+k];
      }
    } else {
      rep(j,1,m) rep(k,0,8) {
        b[j][k]=f[u][j][54+k]+f[u+1][j][63+k]
               +f[d-1][j][45+k]+f[d][j][18+k]
               +t[d-2][j][k]-t[u+1][j][k];
      }
    }
    rep(j,1,m) s[j]=s[j-1]+b[j][8];
    rep(j,1,m) {
      int res=(b[j][0]==1);
      if(j>1) res+=(b[j-1][3]+b[j][1]==1);
      if(j>2) res+=(b[j-2][6]+b[j-1][4]+b[j][2]==1);
      if(j>3) {
        int k=j-3;
        int y=s[k+1]-b[k][6]-b[k+1][7]+1; if(y>=0) v[y]++;
        int x=b[j-1][5]+b[j][2]+s[j-2]; res+=v[x];
      }
      ans+=res;
    }
    rep(j,1,m-1) {
      int y=s[j+1]-b[j][6]-b[j+1][7]+1; if(y>=0) v[y]=0;
    }
  }
  printf("%lld\n",ans);
  return 0;
}