[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20210722185028.17555-1-l4stpr0gr4m@gmail.com>
Date: Fri, 23 Jul 2021 03:50:28 +0900
From: Kangmin Park <l4stpr0gr4m@...il.com>
To: "David S. Miller" <davem@...emloft.net>
Cc: Jakub Kicinski <kuba@...nel.org>,
Martin Varghese <martin.varghese@...ia.com>,
Jiapeng Chong <jiapeng.chong@...ux.alibaba.com>,
"Gustavo A. R. Silva" <gustavoars@...nel.org>,
Guillaume Nault <gnault@...hat.com>, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: [PATCH] mpls: defer ttl decrement in mpls_forward()
Defer ttl decrement to optimize in tx_err case. There is no need
to decrease ttl in the case of goto tx_err.
Signed-off-by: Kangmin Park <l4stpr0gr4m@...il.com>
---
net/mpls/af_mpls.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c
index 05a21dd072df..ffeb2df8be7a 100644
--- a/net/mpls/af_mpls.c
+++ b/net/mpls/af_mpls.c
@@ -407,7 +407,6 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
/* Verify ttl is valid */
if (dec.ttl <= 1)
goto err;
- dec.ttl -= 1;
/* Find the output device */
out_dev = rcu_dereference(nh->nh_dev);
@@ -431,6 +430,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
skb->dev = out_dev;
skb->protocol = htons(ETH_P_MPLS_UC);
+ dec.ttl -= 1;
if (unlikely(!new_header_size && dec.bos)) {
/* Penultimate hop popping */
if (!mpls_egress(dev_net(out_dev), rt, skb, dec))
--
2.26.2
Powered by blists - more mailing lists