inet: add a schedule point in inet_twsk_purge()

On a large hash table, we can easily spend seconds to
walk over all entries. Add a cond_resched() to yield
cpu if necessary.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index 86ebf020925bd0944e05b912f31e6d694d4c0219..f38e387448fb5596c64cdcb07e7cb06d3e624451 100644 (file)
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -487,6 +487,7 @@ void inet_twsk_purge(struct inet_hashinfo *hashinfo,
        for (slot = 0; slot <= hashinfo->ehash_mask; slot++) {
                struct inet_ehash_bucket *head = &hashinfo->ehash[slot];
 restart_rcu:
+               cond_resched();
                rcu_read_lock();
 restart:
                sk_nulls_for_each_rcu(sk, node, &head->chain) {