Merge branch 'rework-inet_csk_get_port'

Josef Bacik says:

====================
Rework inet_csk_get_port

V3->V4:
-Removed the random include of addrconf.h that is no longer needed.

V2->V3:
-Dropped the fastsock from the tb and instead just carry the saddrs, family, and
 ipv6 only flag.
-Reworked the helper functions to deal with this change so I could still use
 them when checking the fast path.
-Killed tb->num_owners as per Eric's request.
-Attached a reproducer to the bottom of this email.

V1->V2:
-Added a new patch 'inet: collapse ipv4/v6 rcv_saddr_equal functions into one'
 at Hannes' suggestion.
-Dropped ->bind_conflict and just use the new helper.
-Fixed a compile bug from the original ->bind_conflict patch.

The original description of the series follows:

At some point recently the guys working on our load balancer added the ability
to use SO_REUSEPORT.  When they restarted their app with this option enabled
they immediately hit a softlockup on what appeared to be the
inet_bind_bucket->lock.  Eventually what all of our debugging and discussion led
us to was the fact that the application comes up without SO_REUSEPORT, shuts
down which creates around 100k twsk's, and then comes up and tries to open a
bunch of sockets using SO_REUSEPORT, which meant traversing the inet_bind_bucket
owners list under the lock.  Since this lock is needed for dealing with the
twsk's and basically anything else related to connections we would softlockup,
and sometimes not ever recover.

To solve this problem I did what you see in Path 5/5.  Once we have a
SO_REUSEPORT socket on the tb->owners list we know that the socket has no
conflicts with any of the other sockets on that list.  So we can add a copy of
the sock_common (really all we need is the recv_saddr but it seemed ugly to copy
just the ipv6, ipv4, and flag to indicate if we were ipv6 only in there so I've
copied the whole common) in order to check subsequent SO_REUSEPORT sockets.  If
they match the previous one then we can skip the expensive
inet_csk_bind_conflict check.  This is what eliminated the soft lockup that we
were seeing.

Patches 1-4 are cleanups and re-workings.  For instance when we specify port ==
0 we need to find an open port, but we would do two passes through
inet_csk_bind_conflict every time we found a possible port.  We would also keep
track of the smallest_port value in order to try and use it if we found no
port our first run through.  This however made no sense as it would have had to
fail the first pass through inet_csk_bind_conflict, so would not actually pass
the second pass through either.  Finally I split the function into two functions
in order to make it easier to read and to distinguish between the two behaviors.

I have tested this on one of our load balancing boxes during peak traffic and it
hasn't fallen over.  But this is not my area, so obviously feel free to point
out where I'm being stupid and I'll get it fixed up and retested.  Thanks,
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

5 files changed, 11 insertions, 20 deletions

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 8f998afc1384..17c6fd84e287 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -88,9 +88,7 @@ int __ipv6_get_lladdr(struct inet6_dev *idev, struct in6_addr *addr,
 		      u32 banned_flags);
 int ipv6_get_lladdr(struct net_device *dev, struct in6_addr *addr,
 		    u32 banned_flags);
-int ipv4_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2,
-			 bool match_wildcard);
-int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2,
+int inet_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2,
 			 bool match_wildcard);
 void addrconf_join_solict(struct net_device *dev, const struct in6_addr *addr);
 void addrconf_leave_solict(struct inet6_dev *idev, const struct in6_addr *addr);
diff --git a/include/net/inet6_connection_sock.h b/include/net/inet6_connection_sock.h
index 3212b39b5bfc..8ec87b62257b 100644
--- a/include/net/inet6_connection_sock.h
+++ b/include/net/inet6_connection_sock.h
@@ -15,16 +15,11 @@
 
 #include <linux/types.h>
 
-struct inet_bind_bucket;
 struct request_sock;
 struct sk_buff;
 struct sock;
 struct sockaddr;
 
-int inet6_csk_bind_conflict(const struct sock *sk,
-			    const struct inet_bind_bucket *tb, bool relax,
-			    bool soreuseport_ok);
-
 struct dst_entry *inet6_csk_route_req(const struct sock *sk, struct flowi6 *fl6,
 				      const struct request_sock *req, u8 proto);
 
diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 84b2edde09b1..826f198374f8 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -62,9 +62,6 @@ struct inet_connection_sock_af_ops {
 				char __user *optval, int __user *optlen);
 #endif
 	void	    (*addr2sockaddr)(struct sock *sk, struct sockaddr *);
-	int	    (*bind_conflict)(const struct sock *sk,
-				     const struct inet_bind_bucket *tb,
-				     bool relax, bool soreuseport_ok);
 	void	    (*mtu_reduced)(struct sock *sk);
 };
 
@@ -263,9 +260,6 @@ inet_csk_rto_backoff(const struct inet_connection_sock *icsk,
 
 struct sock *inet_csk_accept(struct sock *sk, int flags, int *err);
 
-int inet_csk_bind_conflict(const struct sock *sk,
-			   const struct inet_bind_bucket *tb, bool relax,
-			   bool soreuseport_ok);
 int inet_csk_get_port(struct sock *sk, unsigned short snum);
 
 struct dst_entry *inet_csk_route_req(const struct sock *sk, struct flowi4 *fl4,
diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 0574493e3899..1178931288cb 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -74,13 +74,21 @@ struct inet_ehash_bucket {
  * users logged onto your box, isn't it nice to know that new data
  * ports are created in O(1) time?  I thought so. ;-)	-DaveM
  */
+#define FASTREUSEPORT_ANY	1
+#define FASTREUSEPORT_STRICT	2
+
 struct inet_bind_bucket {
 	possible_net_t		ib_net;
 	unsigned short		port;
 	signed char		fastreuse;
 	signed char		fastreuseport;
 	kuid_t			fastuid;
-	int			num_owners;
+#if IS_ENABLED(CONFIG_IPV6)
+	struct in6_addr		fast_v6_rcv_saddr;
+#endif
+	__be32			fast_rcv_saddr;
+	unsigned short		fast_sk_family;
+	bool			fast_ipv6_only;
 	struct hlist_node	node;
 	struct hlist_head	owners;
 };
@@ -203,10 +211,7 @@ void inet_hashinfo_init(struct inet_hashinfo *h);
 
 bool inet_ehash_insert(struct sock *sk, struct sock *osk);
 bool inet_ehash_nolisten(struct sock *sk, struct sock *osk);
-int __inet_hash(struct sock *sk, struct sock *osk,
-		int (*saddr_same)(const struct sock *sk1,
-				  const struct sock *sk2,
-				  bool match_wildcard));
+int __inet_hash(struct sock *sk, struct sock *osk);
 int inet_hash(struct sock *sk);
 void inet_unhash(struct sock *sk);
 
diff --git a/include/net/udp.h b/include/net/udp.h
index 1661791e8ca1..c9d8b8e848e0 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -204,7 +204,6 @@ static inline void udp_lib_close(struct sock *sk, long timeout)
 }
 
 int udp_lib_get_port(struct sock *sk, unsigned short snum,
-		     int (*)(const struct sock *, const struct sock *, bool),
 		     unsigned int hash2_nulladdr);
 
 u32 udp_flow_hashrnd(void);