From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: "Jason A. Donenfeld" Date: Wed, 6 May 2020 15:33:04 -0600 Subject: [PATCH] wireguard: send/receive: cond_resched() when processing worker ringbuffers commit 4005f5c3c9d006157ba716594e0d70c88a235c5e upstream. Users with pathological hardware reported CPU stalls on CONFIG_ PREEMPT_VOLUNTARY=y, because the ringbuffers would stay full, meaning these workers would never terminate. That turned out not to be okay on systems without forced preemption, which Sultan observed. This commit adds a cond_resched() to the bottom of each loop iteration, so that these workers don't hog the core. Note that we don't need this on the napi poll worker, since that terminates after its budget is expended. Suggested-by: Sultan Alsawaf Reported-by: Wang Jian Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld Signed-off-by: David S. Miller Signed-off-by: Jason A. Donenfeld --- drivers/net/wireguard/receive.c | 2 ++ drivers/net/wireguard/send.c | 4 ++++ 2 files changed, 6 insertions(+) --- a/drivers/net/wireguard/receive.c +++ b/drivers/net/wireguard/receive.c @@ -516,6 +516,8 @@ void wg_packet_decrypt_worker(struct wor &PACKET_CB(skb)->keypair->receiving)) ? PACKET_STATE_CRYPTED : PACKET_STATE_DEAD; wg_queue_enqueue_per_peer_napi(skb, state); + if (need_resched()) + cond_resched(); } } --- a/drivers/net/wireguard/send.c +++ b/drivers/net/wireguard/send.c @@ -281,6 +281,8 @@ void wg_packet_tx_worker(struct work_str wg_noise_keypair_put(keypair, false); wg_peer_put(peer); + if (need_resched()) + cond_resched(); } } @@ -304,6 +306,8 @@ void wg_packet_encrypt_worker(struct wor } wg_queue_enqueue_per_peer(&PACKET_PEER(first)->tx_queue, first, state); + if (need_resched()) + cond_resched(); } }