From: Ursula Braun <ubraun@linux.ibm.com>
Date: Tue, 15 May 2018 17:04:54 +0200
Subject: net/smc: no tx work trigger for fallback sockets
Patch-mainline: v4.18-rc1
Git-commit: 569bc643656826d5305aebdc5d5500c99881b2e5
References: bsc#1096003, FATE#325023, LTC#164003

If TCP_NODELAY is set or TCP_CORK is reset, setsockopt triggers the
tx worker. This does not make sense, if the SMC socket switched to
the TCP fallback when the connection is created. This patch adds
the additional check for the fallback case.

Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Johannes Thumshirn <jthumshirn@suse.de>
---
 net/smc/af_smc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 17688a02035b..83403be46a4a 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -1353,14 +1353,14 @@ static int smc_setsockopt(struct socket *sock, int level, int optname,
 		break;
 	case TCP_NODELAY:
 		if (sk->sk_state != SMC_INIT && sk->sk_state != SMC_LISTEN) {
-			if (val)
+			if (val && !smc->use_fallback)
 				mod_delayed_work(system_wq, &smc->conn.tx_work,
 						 0);
 		}
 		break;
 	case TCP_CORK:
 		if (sk->sk_state != SMC_INIT && sk->sk_state != SMC_LISTEN) {
-			if (!val)
+			if (!val && !smc->use_fallback)
 				mod_delayed_work(system_wq, &smc->conn.tx_work,
 						 0);
 		}