diff options
author | Tonghao Zhang <xiangxia.m.yue@gmail.com> | 2020-10-12 13:25:00 -0700 |
---|---|---|
committer | Ilya Maximets <i.maximets@ovn.org> | 2020-10-17 17:32:06 +0200 |
commit | 825634866e95d4e9533d6dafb8d8eccc5f80a3ed (patch) | |
tree | fbd22467306f30f78eccc6934f7a23b06c2a9a07 /datapath | |
parent | a3839b264636f30e34e2028ef7c802235b759565 (diff) | |
download | openvswitch-825634866e95d4e9533d6dafb8d8eccc5f80a3ed.tar.gz |
datapath: select vport upcall portid directly
Upstream commit:
commit 90ce9f23a886bdef7a4b7a9bd52c7a50a6a81635
Author: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Date: Thu Nov 7 00:34:28 2019 +0800
net: openvswitch: select vport upcall portid directly
The commit 69c51582ff786 ("dpif-netlink: don't allocate per
thread netlink sockets"), in Open vSwitch ovs-vswitchd, has
changed the number of allocated sockets to just one per port
by moving the socket array from a per handler structure to
a per datapath one. In the kernel datapath, a vport will have
only one socket in most case, if so select it directly in
fast-path.
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reviewed-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Yi-Hung Wei <yihung.wei@gmail.com>
Signed-off-by: Greg Rose <gvrose8192@gmail.com>
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Diffstat (limited to 'datapath')
-rw-r--r-- | datapath/vport.c | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/datapath/vport.c b/datapath/vport.c index f929282dc..bd62c5612 100644 --- a/datapath/vport.c +++ b/datapath/vport.c @@ -507,8 +507,9 @@ u32 ovs_vport_find_upcall_portid(const struct vport *vport, struct sk_buff *skb) ids = rcu_dereference(vport->upcall_portids); - if (ids->n_ids == 1 && ids->ids[0] == 0) - return 0; + /* If there is only one portid, select it in the fast-path. */ + if (ids->n_ids == 1) + return ids->ids[0]; hash = skb_get_hash(skb); ids_index = hash - ids->n_ids * reciprocal_divide(hash, ids->rn_ids); |