From 9758f40e90f77e457dd4edef1ca506006d7f471a Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 26 Aug 2021 17:44:54 -0700 Subject: mptcp: make the locking tx schema more readable Florian noted the locking schema used by __mptcp_push_pending() is hard to follow, let's add some more descriptive comments and drop an unneeded and confusing check. Suggested-by: Florian Westphal Signed-off-by: Paolo Abeni Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- net/mptcp/protocol.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'net/mptcp') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 1a408395e78f..ade648c3512b 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -1515,15 +1515,19 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags) mptcp_flush_join_list(msk); ssk = mptcp_subflow_get_send(msk); - /* try to keep the subflow socket lock across - * consecutive xmit on the same socket + /* First check. If the ssk has changed since + * the last round, release prev_ssk */ if (ssk != prev_ssk && prev_ssk) mptcp_push_release(sk, prev_ssk, &info); if (!ssk) goto out; - if (ssk != prev_ssk || !prev_ssk) + /* Need to lock the new subflow only if different + * from the previous one, otherwise we are still + * helding the relevant lock + */ + if (ssk != prev_ssk) lock_sock(ssk); /* keep it simple and always provide a new skb for the -- cgit