Merge branch 'for-5.20/io_uring' into for-5.20/io_uring-zerocopy-send

* for-5.20/io_uring: (716 commits)
  io_uring: ensure REQ_F_ISREG is set async offload
  net: fix compat pointer in get_compat_msghdr()
  io_uring: Don't require reinitable percpu_ref
  io_uring: fix types in io_recvmsg_multishot_overflow
  io_uring: Use atomic_long_try_cmpxchg in __io_account_mem
  io_uring: support multishot in recvmsg
  net: copy from user before calling __get_compat_msghdr
  net: copy from user before calling __copy_msghdr
  io_uring: support 0 length iov in buffer select in compat
  io_uring: fix multishot ending when not polled
  io_uring: add netmsg cache
  io_uring: impose max limit on apoll cache
  io_uring: add abstraction around apoll cache
  io_uring: move apoll cache to poll.c
  io_uring: consolidate hash_locked io-wq handling
  io_uring: clear REQ_F_HASH_LOCKED on hash removal
  io_uring: don't race double poll setting REQ_F_ASYNC_DATA
  io_uring: don't miss setting REQ_F_DOUBLE_POLL
  io_uring: disable multishot recvmsg
  io_uring: only trace one of complete or overflow
  ...

Signed-off-by: Jens Axboe <axboe@kernel.dk>

1 files changed, 32 insertions, 21 deletions

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 2e2a9ece9af2..07dbcbae7782 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1051,7 +1051,7 @@ static void tcp_check_sack_reordering(struct sock *sk, const u32 low_seq,
 			 tp->undo_marker ? tp->undo_retrans : 0);
 #endif
 		tp->reordering = min_t(u32, (metric + mss - 1) / mss,
-				       sock_net(sk)->ipv4.sysctl_tcp_max_reordering);
+				       READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_max_reordering));
 	}
 
 	/* This exciting event is worth to be remembered. 8) */
@@ -2030,7 +2030,7 @@ static void tcp_check_reno_reordering(struct sock *sk, const int addend)
 		return;
 
 	tp->reordering = min_t(u32, tp->packets_out + addend,
-			       sock_net(sk)->ipv4.sysctl_tcp_max_reordering);
+			       READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_max_reordering));
 	tp->reord_seen++;
 	NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPRENOREORDER);
 }
@@ -2095,7 +2095,8 @@ static inline void tcp_init_undo(struct tcp_sock *tp)
 
 static bool tcp_is_rack(const struct sock *sk)
 {
-	return sock_net(sk)->ipv4.sysctl_tcp_recovery & TCP_RACK_LOSS_DETECTION;
+	return READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_recovery) &
+		TCP_RACK_LOSS_DETECTION;
 }
 
 /* If we detect SACK reneging, forget all SACK information
@@ -2139,6 +2140,7 @@ void tcp_enter_loss(struct sock *sk)
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct net *net = sock_net(sk);
 	bool new_recovery = icsk->icsk_ca_state < TCP_CA_Recovery;
+	u8 reordering;
 
 	tcp_timeout_mark_lost(sk);
 
@@ -2159,10 +2161,12 @@ void tcp_enter_loss(struct sock *sk)
 	/* Timeout in disordered state after receiving substantial DUPACKs
 	 * suggests that the degree of reordering is over-estimated.
 	 */
+	reordering = READ_ONCE(net->ipv4.sysctl_tcp_reordering);
 	if (icsk->icsk_ca_state <= TCP_CA_Disorder &&
-	    tp->sacked_out >= net->ipv4.sysctl_tcp_reordering)
+	    tp->sacked_out >= reordering)
 		tp->reordering = min_t(unsigned int, tp->reordering,
-				       net->ipv4.sysctl_tcp_reordering);
+				       reordering);
+
 	tcp_set_ca_state(sk, TCP_CA_Loss);
 	tp->high_seq = tp->snd_nxt;
 	tcp_ecn_queue_cwr(tp);
@@ -3464,7 +3468,8 @@ static inline bool tcp_may_raise_cwnd(const struct sock *sk, const int flag)
 	 * new SACK or ECE mark may first advance cwnd here and later reduce
 	 * cwnd in tcp_fastretrans_alert() based on more states.
 	 */
-	if (tcp_sk(sk)->reordering > sock_net(sk)->ipv4.sysctl_tcp_reordering)
+	if (tcp_sk(sk)->reordering >
+	    READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reordering))
 		return flag & FLAG_FORWARD_PROGRESS;
 
 	return flag & FLAG_DATA_ACKED;
@@ -4056,7 +4061,7 @@ void tcp_parse_options(const struct net *net,
 				break;
 			case TCPOPT_WINDOW:
 				if (opsize == TCPOLEN_WINDOW && th->syn &&
-				    !estab && net->ipv4.sysctl_tcp_window_scaling) {
+				    !estab && READ_ONCE(net->ipv4.sysctl_tcp_window_scaling)) {
 					__u8 snd_wscale = *(__u8 *)ptr;
 					opt_rx->wscale_ok = 1;
 					if (snd_wscale > TCP_MAX_WSCALE) {
@@ -4072,7 +4077,7 @@ void tcp_parse_options(const struct net *net,
 			case TCPOPT_TIMESTAMP:
 				if ((opsize == TCPOLEN_TIMESTAMP) &&
 				    ((estab && opt_rx->tstamp_ok) ||
-				     (!estab && net->ipv4.sysctl_tcp_timestamps))) {
+				     (!estab && READ_ONCE(net->ipv4.sysctl_tcp_timestamps)))) {
 					opt_rx->saw_tstamp = 1;
 					opt_rx->rcv_tsval = get_unaligned_be32(ptr);
 					opt_rx->rcv_tsecr = get_unaligned_be32(ptr + 4);
@@ -4080,7 +4085,7 @@ void tcp_parse_options(const struct net *net,
 				break;
 			case TCPOPT_SACK_PERM:
 				if (opsize == TCPOLEN_SACK_PERM && th->syn &&
-				    !estab && net->ipv4.sysctl_tcp_sack) {
+				    !estab && READ_ONCE(net->ipv4.sysctl_tcp_sack)) {
 					opt_rx->sack_ok = TCP_SACK_SEEN;
 					tcp_sack_reset(opt_rx);
 				}
@@ -5567,7 +5572,7 @@ static void tcp_check_urg(struct sock *sk, const struct tcphdr *th)
 	struct tcp_sock *tp = tcp_sk(sk);
 	u32 ptr = ntohs(th->urg_ptr);
 
-	if (ptr && !sock_net(sk)->ipv4.sysctl_tcp_stdurg)
+	if (ptr && !READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_stdurg))
 		ptr--;
 	ptr += ntohl(th->seq);
 
@@ -6729,7 +6734,7 @@ static void tcp_ecn_create_request(struct request_sock *req,
 
 	ect = !INET_ECN_is_not_ect(TCP_SKB_CB(skb)->ip_dsfield);
 	ecn_ok_dst = dst_feature(dst, DST_FEATURE_ECN_MASK);
-	ecn_ok = net->ipv4.sysctl_tcp_ecn || ecn_ok_dst;
+	ecn_ok = READ_ONCE(net->ipv4.sysctl_tcp_ecn) || ecn_ok_dst;
 
 	if (((!ect || th->res1) && ecn_ok) || tcp_ca_needs_ecn(listen_sk) ||
 	    (ecn_ok_dst & DST_FEATURE_ECN_CA) ||
@@ -6797,11 +6802,14 @@ static bool tcp_syn_flood_action(const struct sock *sk, const char *proto)
 {
 	struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
 	const char *msg = "Dropping request";
-	bool want_cookie = false;
 	struct net *net = sock_net(sk);
+	bool want_cookie = false;
+	u8 syncookies;
+
+	syncookies = READ_ONCE(net->ipv4.sysctl_tcp_syncookies);
 
 #ifdef CONFIG_SYN_COOKIES
-	if (net->ipv4.sysctl_tcp_syncookies) {
+	if (syncookies) {
 		msg = "Sending cookies";
 		want_cookie = true;
 		__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPREQQFULLDOCOOKIES);
@@ -6809,8 +6817,7 @@ static bool tcp_syn_flood_action(const struct sock *sk, const char *proto)
 #endif
 		__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPREQQFULLDROP);
 
-	if (!queue->synflood_warned &&
-	    net->ipv4.sysctl_tcp_syncookies != 2 &&
+	if (!queue->synflood_warned && syncookies != 2 &&
 	    xchg(&queue->synflood_warned, 1) == 0)
 		net_info_ratelimited("%s: Possible SYN flooding on port %d. %s.  Check SNMP counters.\n",
 				     proto, sk->sk_num, msg);
@@ -6859,7 +6866,7 @@ u16 tcp_get_syncookie_mss(struct request_sock_ops *rsk_ops,
 	struct tcp_sock *tp = tcp_sk(sk);
 	u16 mss;
 
-	if (sock_net(sk)->ipv4.sysctl_tcp_syncookies != 2 &&
+	if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_syncookies) != 2 &&
 	    !inet_csk_reqsk_queue_is_full(sk))
 		return 0;
 
@@ -6893,13 +6900,15 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 	bool want_cookie = false;
 	struct dst_entry *dst;
 	struct flowi fl;
+	u8 syncookies;
+
+	syncookies = READ_ONCE(net->ipv4.sysctl_tcp_syncookies);
 
 	/* TW buckets are converted to open requests without
 	 * limitations, they conserve resources and peer is
 	 * evidently real one.
 	 */
-	if ((net->ipv4.sysctl_tcp_syncookies == 2 ||
-	     inet_csk_reqsk_queue_is_full(sk)) && !isn) {
+	if ((syncookies == 2 || inet_csk_reqsk_queue_is_full(sk)) && !isn) {
 		want_cookie = tcp_syn_flood_action(sk, rsk_ops->slab_name);
 		if (!want_cookie)
 			goto drop;
@@ -6948,10 +6957,12 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 		tcp_rsk(req)->ts_off = af_ops->init_ts_off(net, skb);
 
 	if (!want_cookie && !isn) {
+		int max_syn_backlog = READ_ONCE(net->ipv4.sysctl_max_syn_backlog);
+
 		/* Kill the following clause, if you dislike this way. */
-		if (!net->ipv4.sysctl_tcp_syncookies &&
-		    (net->ipv4.sysctl_max_syn_backlog - inet_csk_reqsk_queue_len(sk) <
-		     (net->ipv4.sysctl_max_syn_backlog >> 2)) &&
+		if (!syncookies &&
+		    (max_syn_backlog - inet_csk_reqsk_queue_len(sk) <
+		     (max_syn_backlog >> 2)) &&
 		    !tcp_peer_is_proven(req, dst)) {
 			/* Without syncookies last quarter of
 			 * backlog is filled with destinations,