[PATCH] ipv4: Optimize flow initialization in input route lookup.
From: David Miller <davem@davemloft.net>
Date: 2011-03-10 04:42:12
Subsystem:
networking [general], networking [ipv4/ipv6], the rest · Maintainers:
"David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, David Ahern, Ido Schimmel, Linus Torvalds
Like in commit 44713b67db10c774f14280c129b0d5fd13c70cf2
("ipv4: Optimize flow initialization in output route lookup."
we can optimize the on-stack flow setup to only initialize
the members which are actually used.
Otherwise we bzero the entire structure, then initialize
explicitly the first half of it.
Signed-off-by: David S. Miller <davem@davemloft.net>
---
net/ipv4/route.c | 14 ++++++++------
1 files changed, 8 insertions(+), 6 deletions(-)
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 92a24ea..ac32d8f 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c@@ -2081,12 +2081,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, { struct fib_result res; struct in_device *in_dev = __in_dev_get_rcu(dev); - struct flowi fl = { .fl4_dst = daddr, - .fl4_src = saddr, - .fl4_tos = tos, - .fl4_scope = RT_SCOPE_UNIVERSE, - .mark = skb->mark, - .iif = dev->ifindex }; + struct flowi fl; unsigned flags = 0; u32 itag = 0; struct rtable * rth;
@@ -2123,6 +2118,13 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, /* * Now we are ready to route packet. */ + fl.oif = 0; + fl.iif = dev->ifindex; + fl.mark = skb->mark; + fl.fl4_dst = daddr; + fl.fl4_src = saddr; + fl.fl4_tos = tos; + fl.fl4_scope = RT_SCOPE_UNIVERSE; err = fib_lookup(net, &fl, &res); if (err != 0) { if (!IN_DEV_FORWARD(in_dev))
--
1.7.4.1