#include #pragma GCC optimize("Ofast") using namespace std; using std::cout; using std::cin; using std::endl; using ll=long long; using ld=long double; ll ILL=2167167167167167167; const int INF=2100000000; const ll mod=998244353; #define rep(i,a) for (ll i=0;i using _pq = priority_queue, greater>; template ll LB(vector &v,T a){return lower_bound(v.begin(),v.end(),a)-v.begin();} template ll UB(vector &v,T a){return upper_bound(v.begin(),v.end(),a)-v.begin();} template bool chmin(T &a,const T &b){if(a>b){a=b;return 1;}else return 0;} template bool chmax(T &a,const T &b){if(a void So(vector &v) {sort(v.begin(),v.end());} template void Sore(vector &v) {sort(v.begin(),v.end(),[](T x,T y){return x>y;});} void yneos(bool a){if(a) cout<<"Yes\n"; else cout<<"No\n";} template void vec_out(vector &p){for(int i=0;i<(int)(p.size());i++){if(i) cout<<" ";cout< T min(vector &a){assert(!a.empty());T ans=a[0];for(auto &x:a) chmin(ans,x);return ans;} template T max(vector &a){assert(!a.empty());T ans=a[0];for(auto &x:a) chmax(ans,x);return ans;} template T sum(vector &a){assert(!a.empty());T ans=a[0]-a[0];for(auto &x:a) ans+=x;return ans;} void solve(); // oddloop int main() { ios::sync_with_stdio(false); cin.tie(nullptr); int t=1; //cin>>t; rep(i,t) solve(); } void solve(){ ll N,M; cin>>N>>M; ll rev_6=(mod+1)/6; ll ans=M*((((N+1)*N)/2)%mod); ans%=mod; auto f=[&](ll R,ll B)->ll{// [0,R] chmin(R,M); ll ans=0; ll tmp=(B*(N+1))%mod; ans=(tmp*(((R*(R+1))/2)%mod))%mod; tmp=((B*B)+B)/2; tmp%=mod; tmp=(tmp*R)%mod; tmp=(tmp*(R+1))%mod; tmp=(tmp*(2*R+1))%mod; tmp=(tmp*rev_6)%mod; return (ans-tmp)%mod; }; chmin(M,N); ll r=N,l=N/2; ll b=1; while(r>40000){ ans-=(f(r,b)-f(l,b)); ans%=mod; b++; r=N/b; l=N/(b+1); } while(r){ b=N/r; ans-=(f(r,b)-f(r-1,b)); ans%=mod; r--; } cout<<(ans+mod)%mod<<"\n"; }