tcp: preserve const qualifier in tcp_sk()

We can change tcp_sk() to propagate its argument const qualifier, thanks to container_of_const(). We have two places where a const sock pointer has to be upgraded to a write one. We have been using const qualifier for lockless listeners to clearly identify points where writes could happen. Add tcp_sk_rw() helper to better document these. tcp_inbound_md5_hash(), __tcp_grow_window(), tcp_reset_check() and tcp_rack_reo_wnd() get an additional const qualififer for their @tp local variables. smc_check_reset_syn_req() also needs a similar change. Signed-off-by: Eric Dumazet <[email protected]> Reviewed-by: Simon Horman <[email protected]> Signed-off-by: David S. Miller <[email protected]>
wafgo · Mar 18, 2023 · e9d9da9 · e9d9da9
1 parent 403a40f
commit e9d9da9
Show file tree

Hide file tree

Showing 7 changed files with 21 additions and 13 deletions.
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
@@ -472,10 +472,12 @@ enum tsq_flags {
 	TCPF_MTU_REDUCED_DEFERRED	= (1UL << TCP_MTU_REDUCED_DEFERRED),
 };
 
-static inline struct tcp_sock *tcp_sk(const struct sock *sk)
-{
-	return (struct tcp_sock *)sk;
-}
+#define tcp_sk(ptr) container_of_const(ptr, struct tcp_sock, inet_conn.icsk_inet.sk)
+
+/* Variant of tcp_sk() upgrading a const sock to a read/write tcp socket.
+ * Used in context of (lockless) tcp listeners.
+ */
+#define tcp_sk_rw(ptr) container_of(ptr, struct tcp_sock, inet_conn.icsk_inet.sk)
 
 struct tcp_timewait_sock {
 	struct inet_timewait_sock tw_sk;

diff --git a/include/net/tcp.h b/include/net/tcp.h
@@ -529,7 +529,7 @@ static inline void tcp_synq_overflow(const struct sock *sk)
 
 	last_overflow = READ_ONCE(tcp_sk(sk)->rx_opt.ts_recent_stamp);
 	if (!time_between32(now, last_overflow, last_overflow + HZ))
-		WRITE_ONCE(tcp_sk(sk)->rx_opt.ts_recent_stamp, now);
+		WRITE_ONCE(tcp_sk_rw(sk)->rx_opt.ts_recent_stamp, now);
 }
 
 /* syncookies: no recent synqueue overflow on this listening socket? */

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
@@ -4570,7 +4570,7 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
 	const __u8 *hash_location = NULL;
 	struct tcp_md5sig_key *hash_expected;
 	const struct tcphdr *th = tcp_hdr(skb);
-	struct tcp_sock *tp = tcp_sk(sk);
+	const struct tcp_sock *tp = tcp_sk(sk);
 	int genhash, l3index;
 	u8 newhash[16];
 

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
@@ -458,7 +458,7 @@ static void tcp_sndbuf_expand(struct sock *sk)
 static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb,
 			     unsigned int skbtruesize)
 {
-	struct tcp_sock *tp = tcp_sk(sk);
+	const struct tcp_sock *tp = tcp_sk(sk);
 	/* Optimize this! */
 	int truesize = tcp_win_from_space(sk, skbtruesize) >> 1;
 	int window = tcp_win_from_space(sk, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])) >> 1;
@@ -5693,7 +5693,7 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, const struct tcphdr *t
  */
 static bool tcp_reset_check(const struct sock *sk, const struct sk_buff *skb)
 {
-	struct tcp_sock *tp = tcp_sk(sk);
+	const struct tcp_sock *tp = tcp_sk(sk);
 
 	return unlikely(TCP_SKB_CB(skb)->seq == (tp->rcv_nxt - 1) &&
 			(1 << sk->sk_state) & (TCPF_CLOSE_WAIT | TCPF_LAST_ACK |

diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
@@ -463,7 +463,7 @@ void tcp_ca_openreq_child(struct sock *sk, const struct dst_entry *dst)
 }
 EXPORT_SYMBOL_GPL(tcp_ca_openreq_child);
 
-static void smc_check_reset_syn_req(struct tcp_sock *oldtp,
+static void smc_check_reset_syn_req(const struct tcp_sock *oldtp,
 				    struct request_sock *req,
 				    struct tcp_sock *newtp)
 {
@@ -492,7 +492,8 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
 	const struct inet_request_sock *ireq = inet_rsk(req);
 	struct tcp_request_sock *treq = tcp_rsk(req);
 	struct inet_connection_sock *newicsk;
-	struct tcp_sock *oldtp, *newtp;
+	const struct tcp_sock *oldtp;
+	struct tcp_sock *newtp;
 	u32 seq;
 
 	if (!newsk)

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
@@ -4127,8 +4127,13 @@ int tcp_rtx_synack(const struct sock *sk, struct request_sock *req)
 	if (!res) {
 		TCP_INC_STATS(sock_net(sk), TCP_MIB_RETRANSSEGS);
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSYNRETRANS);
-		if (unlikely(tcp_passive_fastopen(sk)))
-			tcp_sk(sk)->total_retrans++;
+		if (unlikely(tcp_passive_fastopen(sk))) {
+			/* sk has const attribute because listeners are lockless.
+			 * However in this case, we are dealing with a passive fastopen
+			 * socket thus we can change total_retrans value.
+			 */
+			tcp_sk_rw(sk)->total_retrans++;
+		}
 		trace_tcp_retransmit_synack(sk, req);
 	}
 	return res;

diff --git a/net/ipv4/tcp_recovery.c b/net/ipv4/tcp_recovery.c
@@ -4,7 +4,7 @@
 
 static u32 tcp_rack_reo_wnd(const struct sock *sk)
 {
-	struct tcp_sock *tp = tcp_sk(sk);
+	const struct tcp_sock *tp = tcp_sk(sk);
 
 	if (!tp->reord_seen) {
 		/* If reordering has not been observed, be aggressive during