#pragma GCC optimize("Ofast")
#pragma GCC optimize("unroll-loops")
#pragma GCC optimize("inline")
#include <bits/stdc++.h>
using namespace std;
#define kien long long
#define pb push_back
#define FOR(i, a, b) for (int i = a;i <= b; i++)
#define FORD(i, a, b) for (int i = a;i >= b; i--)
#define pii pair<int, int>
#define dembit(a) __builtin_popcountll(a)
#define task "kien"
#define fin(x) freopen(x".inp","r",stdin)
#define fou(x) freopen(x".out","w",stdout)
#define debug(x) cout << x << " ";
#define down cout << "\n"
const kien MOD = 1e9 + 7;
const int NTEST = 100;
const int Million = 1e6 + 5;
const int MXN = 2e5 + 5;
const int INF = 2e9;
mt19937 rd(chrono::high_resolution_clock::now().time_since_epoch().count());
kien rand(kien l, kien r) {
assert(l <= r);
return uniform_int_distribution<kien>(l, r)(rd);
}
kien n, k, m, dem, b[MXN + 5], u, v, a[1000000], cura[MXN], curb[MXN];
kien maxx, minn, vtr, l, r, qr[MXN], ans, vl[MXN];
vector <int> vec[MXN];
struct segment {
int st[12 * MXN], lazy[12 * MXN];
void build (int id, int l, int r) {
st[id] = INF;
if (l == r) return;
int mid = (l + r) >> 1;
build (id << 1, l , mid);
build (id << 1 | 1, mid + 1, r);
}
// void push (int id) {
// st[id << 1] += lazy[id];
// st[id << 1 | 1] += lazy[id];
//
// lazy[id << 1] += lazy[id];
// lazy[id << 1 | 1] += lazy[id];
//
// lazy[id] = 0;
// }
void update (int id, int l, int r, int u, int v, int val) {
if (u > r or v < l) {
return;
}
if (u <= l and r <= v) {st[id] = val; return;}
int mid = (l + r) >> 1;
update (id << 1, l, mid, u, v, val);
update (id << 1 | 1, mid + 1, r, u ,v, val);
st[id] = max(st[id << 1], st[id << 1 | 1]);
}
kien get (int id, int l, int r, int u, int v) {
if (u > l or v < r) return 0;
if (u <= l and r <= v) return st[id];
int mid = (l + r) >> 1;
return max(get(id << 1, l, mid, u, v), get(id << 1 | 1, mid + 1, r, u, v));
}
} ST;
struct FenwickTree {
int N;
kien bit[3 * MXN];
void init(int n) {
N = n;
for (int i = 1; i <= N; i++) bit[i] = 0;
}
void update(int pos, int val) {
for (; pos <= N; pos += pos & -pos) bit[pos] += val;
}
kien get(int pos) {
kien s = 0;
for (; pos > 0; pos -= pos & -pos) s += bit[pos];
return s;
}
void update_range(int u, int v, int val) {
update(u, val);
update(v + 1, -val);
}
} BIT;
void solve() {
cin >> n >> k;
vector <int> zip;
FOR (i, 1, n) {
cin >> a[i] >> b[i];
zip.pb(a[i]); zip.pb(b[i]);
cura[i] = a[i]; curb[i] = b[i];
}
FOR (i, 1, k) {
cin >> qr[i];
zip.pb(qr[i]);
} sort(zip.begin(), zip.end());
zip.erase(unique(zip.begin(), zip.end()), zip.end());
// ST.build(1, 1, n * 2 + k);
maxx = zip.size();
FOR (i, 1, n) {
a[i] = lower_bound(zip.begin(), zip.end(), a[i]) - zip.begin() + 1;
b[i] = lower_bound(zip.begin(), zip.end(), b[i]) - zip.begin() + 1;
}
FOR (i, 1, k) {
qr[i] = lower_bound(zip.begin(), zip.end(), qr[i]) - zip.begin() + 1;
ST.update(1, 1, maxx, qr[i], qr[i], i);
}
FOR (i, 1, n) {
int lst = ST.get(1, 1, n, min(a[i], b[i]), max(a[i], b[i]) - 1);
if (lst == 0) {vec[0].pb(i); continue;}
vec[lst].pb(i);
}
BIT.init(maxx);
FORD (i, k, 0) {
if (i == 0) {
for (auto x : vec[i]) {
int mx = max(a[x], b[x]);
int gg = BIT.get(maxx) - BIT.get(mx - 1);
if (gg % 2 == 0) {
ans += min(cura[x], curb[x]);
// cout << max(cura[x], curb[x]) << " " << x << "\n";
}
else {
ans += max(cura[x], curb[x]);
// cout << min(cura[x], curb[x]) << " " << x << "\n";
}
}
continue;
}
BIT.update(qr[i], 1);
for (auto x : vec[i]) {
int mx = max(a[x], b[x]);
int gg = BIT.get(maxx) - BIT.get(mx - 1);
if (gg % 2 == 0) {
ans += max(cura[x], curb[x]);
// cout << max(cura[x], curb[x]) << " " << x << "\n";
}
else {
ans += min(cura[x], curb[x]);
// cout << min(cura[x], curb[x]) << " " << x << "\n";
}
}
}
cout << ans << "\n";
}
main() {
ios_base::sync_with_stdio(0);
cin.tie(0);
if (fopen(task".inp", "r")) {
fin(task); fou(task);
}
int t = 1; //cin >> t;
while(t--) solve();
cerr << "\n" << 1.0 * clock() / CLOCKS_PER_SEC << "s ";
}
