Merge branch 'tcp-take-better-care-of-tw_substate-and-tw_rcv_nxt'

Eric Dumazet says:

====================
tcp: take better care of tw_substate and tw_rcv_nxt

While reviewing Jason Xing recent commit (0d9e5df "tcp: avoid reusing
FIN_WAIT2 when trying to find port in connect() process") I saw
we could remove the volatile qualifier for tw_substate field,
and I also added missing data-race annotations around tcptw->tw_rcv_nxt.
====================

Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

Author: kuba-moo

include/net/inet_timewait_sock.h

@@ -58,7 +58,7 @@ struct inet_timewait_sock {
 #define tw_dr			__tw_common.skc_tw_dr
 
 	__u32			tw_mark;
-	volatile unsigned char	tw_substate;
+	unsigned char		tw_substate;
 	unsigned char		tw_rcv_wscale;
 
 	/* Socket demultiplex comparisons on incoming packets. */

net/ipv4/inet_diag.c

@@ -442,7 +442,7 @@ static int inet_twsk_diag_fill(struct sock *sk,
 	inet_diag_msg_common_fill(r, sk);
 	r->idiag_retrans      = 0;
 
-	r->idiag_state	      = tw->tw_substate;
+	r->idiag_state	      = READ_ONCE(tw->tw_substate);
 	r->idiag_timer	      = 3;
 	tmo = tw->tw_timer.expires - jiffies;
 	r->idiag_expires      = jiffies_delta_to_msecs(tmo);
@@ -1209,7 +1209,7 @@ void inet_diag_dump_icsk(struct inet_hashinfo *hashinfo, struct sk_buff *skb,
 			if (num < s_num)
 				goto next_normal;
 			state = (sk->sk_state == TCP_TIME_WAIT) ?
-				inet_twsk(sk)->tw_substate : sk->sk_state;
+				READ_ONCE(inet_twsk(sk)->tw_substate) : sk->sk_state;
 			if (!(idiag_states & (1 << state)))
 				goto next_normal;
 			if (r->sdiag_family != AF_UNSPEC &&

net/ipv4/tcp_ipv4.c

@@ -120,7 +120,7 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp)
 	struct tcp_sock *tp = tcp_sk(sk);
 	int ts_recent_stamp;
 
-	if (tw->tw_substate == TCP_FIN_WAIT2)
+	if (READ_ONCE(tw->tw_substate) == TCP_FIN_WAIT2)
 		reuse = 0;
 
 	if (reuse == 2) {
@@ -1073,7 +1073,7 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
 	}
 
 	tcp_v4_send_ack(sk, skb,
-			tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
+			tcptw->tw_snd_nxt, READ_ONCE(tcptw->tw_rcv_nxt),
 			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
 			tcp_tw_tsval(tcptw),
 			READ_ONCE(tcptw->tw_ts_recent),
@@ -2948,7 +2948,7 @@ static void get_timewait4_sock(const struct inet_timewait_sock *tw,
 
 	seq_printf(f, "%4d: %08X:%04X %08X:%04X"
 		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %pK",
-		i, src, srcp, dest, destp, tw->tw_substate, 0, 0,
+		i, src, srcp, dest, destp, READ_ONCE(tw->tw_substate), 0, 0,
 		3, jiffies_delta_to_clock_t(delta), 0, 0, 0, 0,
 		refcount_read(&tw->tw_refcnt), tw);
 }

net/ipv4/tcp_minisocks.c

@@ -52,16 +52,17 @@ tcp_timewait_check_oow_rate_limit(struct inet_timewait_sock *tw,
 	return TCP_TW_SUCCESS;
 }
 
-static void twsk_rcv_nxt_update(struct tcp_timewait_sock *tcptw, u32 seq)
+static void twsk_rcv_nxt_update(struct tcp_timewait_sock *tcptw, u32 seq,
+				u32 rcv_nxt)
 {
 #ifdef CONFIG_TCP_AO
 	struct tcp_ao_info *ao;
 
 	ao = rcu_dereference(tcptw->ao_info);
-	if (unlikely(ao && seq < tcptw->tw_rcv_nxt))
+	if (unlikely(ao && seq < rcv_nxt))
 		WRITE_ONCE(ao->rcv_sne, ao->rcv_sne + 1);
 #endif
-	tcptw->tw_rcv_nxt = seq;
+	WRITE_ONCE(tcptw->tw_rcv_nxt, seq);
 }
 
 /*
@@ -98,8 +99,9 @@ enum tcp_tw_status
 tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 			   const struct tcphdr *th, u32 *tw_isn)
 {
-	struct tcp_options_received tmp_opt;
 	struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw);
+	u32 rcv_nxt = READ_ONCE(tcptw->tw_rcv_nxt);
+	struct tcp_options_received tmp_opt;
 	bool paws_reject = false;
 	int ts_recent_stamp;
 
@@ -117,26 +119,26 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 		}
 	}
 
-	if (tw->tw_substate == TCP_FIN_WAIT2) {
+	if (READ_ONCE(tw->tw_substate) == TCP_FIN_WAIT2) {
 		/* Just repeat all the checks of tcp_rcv_state_process() */
 
 		/* Out of window, send ACK */
 		if (paws_reject ||
 		    !tcp_in_window(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq,
-				   tcptw->tw_rcv_nxt,
-				   tcptw->tw_rcv_nxt + tcptw->tw_rcv_wnd))
+				   rcv_nxt,
+				   rcv_nxt + tcptw->tw_rcv_wnd))
 			return tcp_timewait_check_oow_rate_limit(
 				tw, skb, LINUX_MIB_TCPACKSKIPPEDFINWAIT2);
 
 		if (th->rst)
 			goto kill;
 
-		if (th->syn && !before(TCP_SKB_CB(skb)->seq, tcptw->tw_rcv_nxt))
+		if (th->syn && !before(TCP_SKB_CB(skb)->seq, rcv_nxt))
 			return TCP_TW_RST;
 
 		/* Dup ACK? */
 		if (!th->ack ||
-		    !after(TCP_SKB_CB(skb)->end_seq, tcptw->tw_rcv_nxt) ||
+		    !after(TCP_SKB_CB(skb)->end_seq, rcv_nxt) ||
 		    TCP_SKB_CB(skb)->end_seq == TCP_SKB_CB(skb)->seq) {
 			inet_twsk_put(tw);
 			return TCP_TW_SUCCESS;
@@ -146,12 +148,13 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 		 * reset.
 		 */
 		if (!th->fin ||
-		    TCP_SKB_CB(skb)->end_seq != tcptw->tw_rcv_nxt + 1)
+		    TCP_SKB_CB(skb)->end_seq != rcv_nxt + 1)
 			return TCP_TW_RST;
 
 		/* FIN arrived, enter true time-wait state. */
-		tw->tw_substate	  = TCP_TIME_WAIT;
-		twsk_rcv_nxt_update(tcptw, TCP_SKB_CB(skb)->end_seq);
+		WRITE_ONCE(tw->tw_substate, TCP_TIME_WAIT);
+		twsk_rcv_nxt_update(tcptw, TCP_SKB_CB(skb)->end_seq,
+				    rcv_nxt);
 
 		if (tmp_opt.saw_tstamp) {
 			WRITE_ONCE(tcptw->tw_ts_recent_stamp,
@@ -182,7 +185,7 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 	 */
 
 	if (!paws_reject &&
-	    (TCP_SKB_CB(skb)->seq == tcptw->tw_rcv_nxt &&
+	    (TCP_SKB_CB(skb)->seq == rcv_nxt &&
 	     (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq || th->rst))) {
 		/* In window segment, it may be only reset or bare ack. */
 
@@ -229,7 +232,7 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 	 */
 
 	if (th->syn && !th->rst && !th->ack && !paws_reject &&
-	    (after(TCP_SKB_CB(skb)->seq, tcptw->tw_rcv_nxt) ||
+	    (after(TCP_SKB_CB(skb)->seq, rcv_nxt) ||
 	     (tmp_opt.saw_tstamp &&
 	      (s32)(READ_ONCE(tcptw->tw_ts_recent) - tmp_opt.rcv_tsval) < 0))) {
 		u32 isn = tcptw->tw_snd_nxt + 65535 + 2;

net/ipv6/tcp_ipv6.c

@@ -1193,7 +1193,8 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb)
 #endif
 	}
 
-	tcp_v6_send_ack(sk, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
+	tcp_v6_send_ack(sk, skb, tcptw->tw_snd_nxt,
+			READ_ONCE(tcptw->tw_rcv_nxt),
 			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
 			tcp_tw_tsval(tcptw),
 			READ_ONCE(tcptw->tw_ts_recent), tw->tw_bound_dev_if,
@@ -2258,7 +2259,7 @@ static void get_timewait6_sock(struct seq_file *seq,
 		   src->s6_addr32[2], src->s6_addr32[3], srcp,
 		   dest->s6_addr32[0], dest->s6_addr32[1],
 		   dest->s6_addr32[2], dest->s6_addr32[3], destp,
-		   tw->tw_substate, 0, 0,
+		   READ_ONCE(tw->tw_substate), 0, 0,
 		   3, jiffies_delta_to_clock_t(delta), 0, 0, 0, 0,
 		   refcount_read(&tw->tw_refcnt), tw);
 }