From 92935e1c2a7e27795456f5058759dbb3acf67e2c Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Wed, 22 Aug 2018 08:37:37 -0700 Subject: bpf: sockmap: write_space events need to be passed to TCP handler [ Upstream commit 9b2e0388bec8ec5427403e23faff3b58dd1c3200 ] When sockmap code is using the stream parser it also handles the write space events in order to handle the case where (a) verdict redirects skb to another socket and (b) the sockmap then sends the skb but due to memory constraints (or other EAGAIN errors) needs to do a retry. But the initial code missed a third case where the skb_send_sock_locked() triggers an sk_wait_event(). A typically case would be when sndbuf size is exceeded. If this happens because we do not pass the write_space event to the lower layers we never wake up the event and it will wait for sndtimeo. Which as noted in ktls fix may be rather large and look like a hang to the user. To reproduce the best test is to reduce the sndbuf size and send 1B data chunks to stress the memory handling. To fix this pass the event from the upper layer to the lower layer. Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman --- kernel/bpf/sockmap.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c index 53a4787c08d8..20eaddfa691c 100644 --- a/kernel/bpf/sockmap.c +++ b/kernel/bpf/sockmap.c @@ -313,12 +313,15 @@ out: static void smap_write_space(struct sock *sk) { struct smap_psock *psock; + void (*write_space)(struct sock *sk); rcu_read_lock(); psock = smap_psock_sk(sk); if (likely(psock && test_bit(SMAP_TX_RUNNING, &psock->state))) schedule_work(&psock->tx_work); + write_space = psock->save_write_space; rcu_read_unlock(); + write_space(sk); } static void smap_stop_sock(struct smap_psock *psock, struct sock *sk) -- cgit v1.2.3