net: dont hold rtnl mutex during netlink dump callbacks

Four years ago, Patrick made a change to hold rtnl mutex during netlink
dump callbacks.

I believe it was a wrong move. This slows down concurrent dumps, making
good old /proc/net/ files faster than rtnetlink in some situations.

This occurred to me because one "ip link show dev ..." was _very_ slow
on a workload adding/removing network devices in background.

All dump callbacks are able to use RCU locking now, so this patch does
roughly a revert of commits :

1c2d670f366 : [RTNETLINK]: Hold rtnl_mutex during netlink dump callbacks
6313c1e0992 : [RTNETLINK]: Remove unnecessary locking in dump callbacks

This let writers fight for rtnl mutex and readers going full speed.

It also takes care of phonet : phonet_route_get() is now called from rcu
read section. I renamed it to phonet_route_get_rcu()

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Remi Denis-Courmont <remi.denis-courmont@nokia.com>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 3 insertions, 1 deletions

diff --git a/net/phonet/pn_netlink.c b/net/phonet/pn_netlink.c
index 58b3b1f991ed..438accb7a5a8 100644
--- a/net/phonet/pn_netlink.c
+++ b/net/phonet/pn_netlink.c
@@ -264,10 +264,11 @@ static int route_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
         struct net *net = sock_net(skb->sk);
         u8 addr, addr_idx = 0, addr_start_idx = cb->args[0];
 
+        rcu_read_lock();
         for (addr = 0; addr < 64; addr++) {
                 struct net_device *dev;
 
-                dev = phonet_route_get(net, addr << 2);
+                dev = phonet_route_get_rcu(net, addr << 2);
                 if (!dev)
                         continue;
 
@@ -279,6 +280,7 @@ static int route_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
         }
 
 out:
+        rcu_read_unlock();
         cb->args[0] = addr_idx;
         cb->args[1] = 0;