From: Karsten Graul <kgraul@linux.ibm.com>
Date: Sat, 18 Jul 2020 15:06:11 +0200
Subject: net/smc: fix link lookup for new rdma connections
Git-commit: 7df8bcb56053173e5e5c0e566391fa601e3e4778
Patch-mainline: v5.8-rc7
References: jsc#SLE-13763

For new rdma connections the SMC server assigns the link and sends the
link data in the clc accept message. To match the correct link use not
only the qp_num but also the gid and the mac of the links. If there are
equal qp_nums for different links the wrong link would be chosen.

Reviewed-by: Ursula Braun <ubraun@linux.ibm.com>
Fixes: 0fb0b02bd6fd ("net/smc: adapt SMC client code to use the LLC flow")
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Petr Tesarik <ptesarik@suse.com>
---
 net/smc/af_smc.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -632,7 +632,9 @@ static int smc_connect_rdma(struct smc_s
 		for (i = 0; i < SMC_LINKS_PER_LGR_MAX; i++) {
 			struct smc_link *l = &smc->conn.lgr->lnk[i];
 
-			if (l->peer_qpn == ntoh24(aclc->qpn)) {
+			if (l->peer_qpn == ntoh24(aclc->qpn) &&
+			    !memcmp(l->peer_gid, &aclc->lcl.gid, SMC_GID_SIZE) &&
+			    !memcmp(l->peer_mac, &aclc->lcl.mac, sizeof(l->peer_mac))) {
 				link = l;
 				break;
 			}