Skip to contentSkip to conversation

tcp: fix SO_RCVLOWAT possible hangs under high mem pressure

PATCH In review Eric Dumazet 2 participants 2 messages Jun 30, 2020 – Jul 2, 2020

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 tcp: fix SO_RCVLOWAT possible hangs under high mem pressure

Whenever tcp_try_rmem_schedule() returns an error, we are under trouble and should make sure to wakeup readers so that they can drain socket queues and eventually make room.

Fixes 03f45c883c6f ("tcp: avoid extra wakeups for SO_RCVLOWAT users")
---
 net/ipv4/tcp_input.c | 2 ++
 1 file changed, 2 insertions(+)
net/ipv4/tcp_input.c +2 -0
index f3a0eb139b7633ebc1ddb801de232bcd3a0cbdc6..9615e72656d12e9c7298bf7087792d0209897b50 100644
@@ -4582,6 +4582,7 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb)
 
 	if (unlikely(tcp_try_rmem_schedule(sk, skb, skb->truesize))) {
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPOFODROP);
+		sk->sk_data_ready(sk);
 		tcp_drop(sk, skb);
 		return;
 	}
@@ -4828,6 +4829,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 			sk_forced_mem_schedule(sk, skb->truesize);
 		else if (tcp_try_rmem_schedule(sk, skb, skb->truesize)) {
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPRCVQDROP);
+			sk->sk_data_ready(sk);
 			goto drop;
 		}
 
Signature · 1 line
2.27.0.212.ge8ba1cc988-goog
maintainer
#1 lore raw

From: Eric Dumazet <[email protected]>
Date: Tue, 30 Jun 2020 13:51:28 -0700

Quoted 6 lines

Whenever tcp_try_rmem_schedule() returns an error, we are under trouble and should make sure to wakeup readers so that they can drain socket queues and eventually make room.

Fixes 03f45c883c6f ("tcp: avoid extra wakeups for SO_RCVLOWAT users")

Applied and queued up for -stable, thanks Eric.