ipv4: rcu conversion in ip_route_output_slow

ip_route_output_slow() is enclosed in an rcu_read_lock() protected
section, so that no references are taken/released on device, thanks to
__ip_dev_find() & dev_get_by_index_rcu()

Tested with ip route cache disabled, and a stress test :

Before patch:

elapsed time :

real	1m38.347s
user	0m11.909s
sys	23m51.501s

Profile:

13788.00 22.7% ip_route_output_slow [kernel]
 7875.00 13.0% dst_destroy          [kernel]
 3925.00  6.5% fib_semantic_match   [kernel]
 3144.00  5.2% fib_rules_lookup     [kernel]
 3061.00  5.0% dst_alloc            [kernel]
 2276.00  3.7% rt_set_nexthop       [kernel]
 1762.00  2.9% fib_table_lookup     [kernel]
 1538.00  2.5% _raw_read_lock       [kernel]
 1358.00  2.2% ip_output            [kernel]

After patch:

real	1m28.808s
user	0m13.245s
sys	20m37.293s

10950.00 17.2% ip_route_output_slow [kernel]
10726.00 16.9% dst_destroy          [kernel]
 5170.00  8.1% fib_semantic_match   [kernel]
 3937.00  6.2% dst_alloc            [kernel]
 3635.00  5.7% rt_set_nexthop       [kernel]
 2900.00  4.6% fib_rules_lookup     [kernel]
 2240.00  3.5% fib_table_lookup     [kernel]
 1427.00  2.2% _raw_read_lock       [kernel]
 1157.00  1.8% kmem_cache_alloc     [kernel]

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index ea89500..a61acea 100644 (file)
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2487,6 +2487,7 @@ static int ip_mkroute_output(struct rtable **rp,
 
 /*
  * Major route resolver routine.
+ * called with rcu_read_lock();
  */
 
 static int ip_route_output_slow(struct net *net, struct rtable **rp,
@@ -2505,7 +2506,7 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp,
                            .iif = net->loopback_dev->ifindex,
                            .oif = oldflp->oif };
        struct fib_result res;
-       unsigned flags = 0;
+       unsigned int flags = 0;
        struct net_device *dev_out = NULL;
        int free_res = 0;
        int err;
@@ -2535,7 +2536,7 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp,
                    (ipv4_is_multicast(oldflp->fl4_dst) ||
                     oldflp->fl4_dst == htonl(0xFFFFFFFF))) {
                        /* It is equivalent to inet_addr_type(saddr) == RTN_LOCAL */
-                       dev_out = ip_dev_find(net, oldflp->fl4_src);
+                       dev_out = __ip_dev_find(net, oldflp->fl4_src, false);
                        if (dev_out == NULL)
                                goto out;
 
@@ -2560,26 +2561,21 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp,
 
                if (!(oldflp->flags & FLOWI_FLAG_ANYSRC)) {
                        /* It is equivalent to inet_addr_type(saddr) == RTN_LOCAL */
-                       dev_out = ip_dev_find(net, oldflp->fl4_src);
-                       if (dev_out == NULL)
+                       if (!__ip_dev_find(net, oldflp->fl4_src, false))
                                goto out;
-                       dev_put(dev_out);
-                       dev_out = NULL;
                }
        }
 
 
        if (oldflp->oif) {
-               dev_out = dev_get_by_index(net, oldflp->oif);
+               dev_out = dev_get_by_index_rcu(net, oldflp->oif);
                err = -ENODEV;
                if (dev_out == NULL)
                        goto out;
 
                /* RACE: Check return value of inet_select_addr instead. */
-               if (rcu_dereference_raw(dev_out->ip_ptr) == NULL) {
-                       dev_put(dev_out);
+               if (rcu_dereference(dev_out->ip_ptr) == NULL)
                        goto out;       /* Wrong error code */
-               }
 
                if (ipv4_is_local_multicast(oldflp->fl4_dst) ||
                    oldflp->fl4_dst == htonl(0xFFFFFFFF)) {
@@ -2602,10 +2598,7 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp,
                fl.fl4_dst = fl.fl4_src;
                if (!fl.fl4_dst)
                        fl.fl4_dst = fl.fl4_src = htonl(INADDR_LOOPBACK);
-               if (dev_out)
-                       dev_put(dev_out);
                dev_out = net->loopback_dev;
-               dev_hold(dev_out);
                fl.oif = net->loopback_dev->ifindex;
                res.type = RTN_LOCAL;
                flags |= RTCF_LOCAL;
@@ -2639,8 +2632,6 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp,
                        res.type = RTN_UNICAST;
                        goto make_route;
                }
-               if (dev_out)
-                       dev_put(dev_out);
                err = -ENETUNREACH;
                goto out;
        }
@@ -2649,10 +2640,7 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp,
        if (res.type == RTN_LOCAL) {
                if (!fl.fl4_src)
                        fl.fl4_src = fl.fl4_dst;
-               if (dev_out)
-                       dev_put(dev_out);
                dev_out = net->loopback_dev;
-               dev_hold(dev_out);
                fl.oif = dev_out->ifindex;
                if (res.fi)
                        fib_info_put(res.fi);
@@ -2672,28 +2660,23 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp,
        if (!fl.fl4_src)
                fl.fl4_src = FIB_RES_PREFSRC(res);
 
-       if (dev_out)
-               dev_put(dev_out);
        dev_out = FIB_RES_DEV(res);
-       dev_hold(dev_out);
        fl.oif = dev_out->ifindex;
 
 
 make_route:
        err = ip_mkroute_output(rp, &res, &fl, oldflp, dev_out, flags);
 
-
        if (free_res)
                fib_res_put(&res);
-       if (dev_out)
-               dev_put(dev_out);
 out:   return err;
 }
 
 int __ip_route_output_key(struct net *net, struct rtable **rp,
                          const struct flowi *flp)
 {
-       unsigned hash;
+       unsigned int hash;
+       int res;
        struct rtable *rth;
 
        if (!rt_caching(net))
@@ -2724,7 +2707,10 @@ int __ip_route_output_key(struct net *net, struct rtable **rp,
        rcu_read_unlock_bh();
 
 slow_output:
-       return ip_route_output_slow(net, rp, flp);
+       rcu_read_lock();
+       res = ip_route_output_slow(net, rp, flp);
+       rcu_read_unlock();
+       return res;
 }
 EXPORT_SYMBOL_GPL(__ip_route_output_key);