[PATCH 4.9 0/2] net: initialize fastreuse on inet_inherit_port

[PATCH 4.9 0/2] net: initialize fastreuse on inet_inherit_port

[Tim Froidcoeur]

Fix for TPROXY initialization of fastreuse flag.
upstream patch was backported by Greg K-H to 4.14 and higher stable version.

see also https://lore.kernel.org/stable/20200818072007.GA9254@kroah.com/

code in inet_csk_get_port for 4.9 (and 4.4) is different from
the upstream version, so these backports had to be adapted (a bit)

Tim Froidcoeur (2):
  net: refactor bind_bucket fastreuse into helper
  net: initialize fastreuse on inet_inherit_port

 include/net/inet_connection_sock.h |  4 ++++
 net/ipv4/inet_connection_sock.c    | 37 ++++++++++++++++++++----------
 net/ipv4/inet_hashtables.c         |  1 +
 3 files changed, 30 insertions(+), 12 deletions(-)

--
2.25.1

[PATCH 4.9 1/2] net: refactor bind_bucket fastreuse into helper

[Tim Froidcoeur]

[ Upstream commit 62ffc589abb176821662efc4525ee4ac0b9c3894 ]

Refactor the fastreuse update code in inet_csk_get_port into a small
helper function that can be called from other places.

Acked-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Tim Froidcoeur <tim.froidcoeur@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Tim Froidcoeur <tim.froidcoeur@tessares.net>
---
 include/net/inet_connection_sock.h |  4 ++++
 net/ipv4/inet_connection_sock.c    | 37 ++++++++++++++++++++----------
 2 files changed, 29 insertions(+), 12 deletions(-)

diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 146054ceea8e..5bb56ebf3c9f 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -319,5 +319,9 @@ int inet_csk_compat_getsockopt(struct sock *sk, int level, int optname,
 int inet_csk_compat_setsockopt(struct sock *sk, int level, int optname,
 			       char __user *optval, unsigned int optlen);
 
+/* update the fast reuse flag when adding a socket */
+void inet_csk_update_fastreuse(struct inet_bind_bucket *tb,
+			       struct sock *sk);
+
 struct dst_entry *inet_csk_update_pmtu(struct sock *sk, u32 mtu);
 #endif /* _INET_CONNECTION_SOCK_H */
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 1bcbb7399fe6..5a0352ccadd3 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -89,6 +89,28 @@ int inet_csk_bind_conflict(const struct sock *sk,
 }
 EXPORT_SYMBOL_GPL(inet_csk_bind_conflict);
 
+void inet_csk_update_fastreuse(struct inet_bind_bucket *tb,
+			       struct sock *sk)
+{
+	kuid_t uid = sock_i_uid(sk);
+	bool reuse = sk->sk_reuse && sk->sk_state != TCP_LISTEN;
+
+	if (!hlist_empty(&tb->owners)) {
+		if (!reuse)
+			tb->fastreuse = 0;
+		if (!sk->sk_reuseport || !uid_eq(tb->fastuid, uid))
+			tb->fastreuseport = 0;
+	} else {
+		tb->fastreuse = reuse;
+		if (sk->sk_reuseport) {
+			tb->fastreuseport = 1;
+			tb->fastuid = uid;
+		} else {
+			tb->fastreuseport = 0;
+		}
+	}
+}
+
 /* Obtain a reference to a local port for the given sock,
  * if snum is zero it means select any available local port.
  * We try to allocate an odd port (and leave even ports for connect())
@@ -218,19 +240,10 @@ int inet_csk_get_port(struct sock *sk, unsigned short snum)
 			}
 			goto fail_unlock;
 		}
-		if (!reuse)
-			tb->fastreuse = 0;
-		if (!sk->sk_reuseport || !uid_eq(tb->fastuid, uid))
-			tb->fastreuseport = 0;
-	} else {
-		tb->fastreuse = reuse;
-		if (sk->sk_reuseport) {
-			tb->fastreuseport = 1;
-			tb->fastuid = uid;
-		} else {
-			tb->fastreuseport = 0;
-		}
 	}
+
+	inet_csk_update_fastreuse(tb, sk);
+
 success:
 	if (!inet_csk(sk)->icsk_bind_hash)
 		inet_bind_hash(sk, tb, port);
-- 
2.25.1

[PATCH 4.9 2/2] net: initialize fastreuse on inet_inherit_port

[Tim Froidcoeur]

[ Upstream commit d76f3351cea2d927fdf70dd7c06898235035e84e ]

In the case of TPROXY, bind_conflict optimizations for SO_REUSEADDR or
SO_REUSEPORT are broken, possibly resulting in O(n) instead of O(1) bind
behaviour or in the incorrect reuse of a bind.

the kernel keeps track for each bind_bucket if all sockets in the
bind_bucket support SO_REUSEADDR or SO_REUSEPORT in two fastreuse flags.
These flags allow skipping the costly bind_conflict check when possible
(meaning when all sockets have the proper SO_REUSE option).

For every socket added to a bind_bucket, these flags need to be updated.
As soon as a socket that does not support reuse is added, the flag is
set to false and will never go back to true, unless the bind_bucket is
deleted.

Note that there is no mechanism to re-evaluate these flags when a socket
is removed (this might make sense when removing a socket that would not
allow reuse; this leaves room for a future patch).

For this optimization to work, it is mandatory that these flags are
properly initialized and updated.

When a child socket is created from a listen socket in
__inet_inherit_port, the TPROXY case could create a new bind bucket
without properly initializing these flags, thus preventing the
optimization to work. Alternatively, a socket not allowing reuse could
be added to an existing bind bucket without updating the flags, causing
bind_conflict to never be called as it should.

Call inet_csk_update_fastreuse when __inet_inherit_port decides to create
a new bind_bucket or use a different bind_bucket than the one of the
listen socket.

Fixes: 093d282321da ("tproxy: fix hash locking issue when using port redirection in __inet_inherit_port()")
Acked-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Tim Froidcoeur <tim.froidcoeur@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Tim Froidcoeur <tim.froidcoeur@tessares.net>
---
 net/ipv4/inet_hashtables.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 4bf542f4d980..887633870763 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -163,6 +163,7 @@ int __inet_inherit_port(const struct sock *sk, struct sock *child)
 				return -ENOMEM;
 			}
 		}
+		inet_csk_update_fastreuse(tb, child);
 	}
 	inet_bind_hash(child, tb, port);
 	spin_unlock(&head->lock);
-- 
2.25.1

Re: [PATCH 4.9 0/2] net: initialize fastreuse on inet_inherit_port

[Greg KH]

On Wed, Sep 02, 2020 at 04:05:43PM +0200, Tim Froidcoeur wrote:
> Fix for TPROXY initialization of fastreuse flag.
> upstream patch was backported by Greg K-H to 4.14 and higher stable version.
> 
> see also https://lore.kernel.org/stable/20200818072007.GA9254@kroah.com/
> 
> code in inet_csk_get_port for 4.9 (and 4.4) is different from
> the upstream version, so these backports had to be adapted (a bit)
> 
> Tim Froidcoeur (2):
>   net: refactor bind_bucket fastreuse into helper
>   net: initialize fastreuse on inet_inherit_port
> 
>  include/net/inet_connection_sock.h |  4 ++++
>  net/ipv4/inet_connection_sock.c    | 37 ++++++++++++++++++++----------
>  net/ipv4/inet_hashtables.c         |  1 +
>  3 files changed, 30 insertions(+), 12 deletions(-)
> 
> --
> 2.25.1

Both now queued up, thanks.

greg k-h