summaryrefslogtreecommitdiff
path: root/net/ipv4
diff options
context:
space:
mode:
Diffstat (limited to 'net/ipv4')
-rw-r--r--net/ipv4/datagram.c2
-rw-r--r--net/ipv4/fib_frontend.c2
-rw-r--r--net/ipv4/fib_semantics.c2
-rw-r--r--net/ipv4/inet_connection_sock.c6
-rw-r--r--net/ipv4/inet_diag.c2
-rw-r--r--net/ipv4/inet_hashtables.c2
-rw-r--r--net/ipv4/ip_gre.c4
-rw-r--r--net/ipv4/ip_output.c14
-rw-r--r--net/ipv4/ipmr.c3
-rw-r--r--net/ipv4/route.c11
-rw-r--r--net/ipv4/tcp.c79
-rw-r--r--net/ipv4/tcp_diag.c5
-rw-r--r--net/ipv4/tcp_fastopen.c2
-rw-r--r--net/ipv4/tcp_input.c37
-rw-r--r--net/ipv4/tcp_ipv4.c36
-rw-r--r--net/ipv4/tcp_minisocks.c17
-rw-r--r--net/ipv4/tcp_output.c32
-rw-r--r--net/ipv4/tcp_timer.c11
-rw-r--r--net/ipv4/udp.c29
19 files changed, 174 insertions, 122 deletions
diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c
index 9a0fe0c2fa02..4a8550c49202 100644
--- a/net/ipv4/datagram.c
+++ b/net/ipv4/datagram.c
@@ -73,7 +73,7 @@ int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len
reuseport_has_conns(sk, true);
sk->sk_state = TCP_ESTABLISHED;
sk_set_txhash(sk);
- inet->inet_id = jiffies;
+ inet->inet_id = prandom_u32();
sk_dst_set(sk, &rt->dst);
err = 0;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index dde77f72e03e..71c78d223dfd 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -1148,7 +1148,7 @@ void fib_modify_prefix_metric(struct in_ifaddr *ifa, u32 new_metric)
if (!(dev->flags & IFF_UP) ||
ifa->ifa_flags & (IFA_F_SECONDARY | IFA_F_NOPREFIXROUTE) ||
ipv4_is_zeronet(prefix) ||
- prefix == ifa->ifa_local || ifa->ifa_prefixlen == 32)
+ (prefix == ifa->ifa_local && ifa->ifa_prefixlen == 32))
return;
/* add the new */
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 0913a090b2bf..f1888c683426 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -1814,8 +1814,8 @@ int fib_sync_down_addr(struct net_device *dev, __be32 local)
int ret = 0;
unsigned int hash = fib_laddr_hashfn(local);
struct hlist_head *head = &fib_info_laddrhash[hash];
+ int tb_id = l3mdev_fib_table(dev) ? : RT_TABLE_MAIN;
struct net *net = dev_net(dev);
- int tb_id = l3mdev_fib_table(dev);
struct fib_info *fi;
if (!fib_info_laddrhash || local == 0)
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index a9183543ca30..eb30fc1770de 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -906,7 +906,7 @@ static void inet_child_forget(struct sock *sk, struct request_sock *req,
percpu_counter_inc(sk->sk_prot->orphan_count);
if (sk->sk_protocol == IPPROTO_TCP && tcp_rsk(req)->tfo_listener) {
- BUG_ON(tcp_sk(child)->fastopen_rsk != req);
+ BUG_ON(rcu_access_pointer(tcp_sk(child)->fastopen_rsk) != req);
BUG_ON(sk != req->rsk_listener);
/* Paranoid, to prevent race condition if
@@ -915,7 +915,7 @@ static void inet_child_forget(struct sock *sk, struct request_sock *req,
* Also to satisfy an assertion in
* tcp_v4_destroy_sock().
*/
- tcp_sk(child)->fastopen_rsk = NULL;
+ RCU_INIT_POINTER(tcp_sk(child)->fastopen_rsk, NULL);
}
inet_csk_destroy_sock(child);
}
@@ -934,7 +934,7 @@ struct sock *inet_csk_reqsk_queue_add(struct sock *sk,
req->sk = child;
req->dl_next = NULL;
if (queue->rskq_accept_head == NULL)
- queue->rskq_accept_head = req;
+ WRITE_ONCE(queue->rskq_accept_head, req);
else
queue->rskq_accept_tail->dl_next = req;
queue->rskq_accept_tail = req;
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index bbb005eb5218..7dc79b973e6e 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -193,7 +193,7 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
if (ext & (1 << (INET_DIAG_MEMINFO - 1))) {
struct inet_diag_meminfo minfo = {
.idiag_rmem = sk_rmem_alloc_get(sk),
- .idiag_wmem = sk->sk_wmem_queued,
+ .idiag_wmem = READ_ONCE(sk->sk_wmem_queued),
.idiag_fmem = sk->sk_forward_alloc,
.idiag_tmem = sk_wmem_alloc_get(sk),
};
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 97824864e40d..83fb00153018 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -240,7 +240,7 @@ static inline int compute_score(struct sock *sk, struct net *net,
return -1;
score = sk->sk_family == PF_INET ? 2 : 1;
- if (sk->sk_incoming_cpu == raw_smp_processor_id())
+ if (READ_ONCE(sk->sk_incoming_cpu) == raw_smp_processor_id())
score++;
}
return score;
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 52690bb3e40f..10636fb6093e 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -509,9 +509,9 @@ static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev)
key = &tun_info->key;
if (!(tun_info->key.tun_flags & TUNNEL_ERSPAN_OPT))
goto err_free_skb;
- md = ip_tunnel_info_opts(tun_info);
- if (!md)
+ if (tun_info->options_len < sizeof(*md))
goto err_free_skb;
+ md = ip_tunnel_info_opts(tun_info);
/* ERSPAN has fixed 8 byte GRE header */
version = md->version;
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 28fca408812c..3d8baaaf7086 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -645,11 +645,12 @@ void ip_fraglist_prepare(struct sk_buff *skb, struct ip_fraglist_iter *iter)
EXPORT_SYMBOL(ip_fraglist_prepare);
void ip_frag_init(struct sk_buff *skb, unsigned int hlen,
- unsigned int ll_rs, unsigned int mtu,
+ unsigned int ll_rs, unsigned int mtu, bool DF,
struct ip_frag_state *state)
{
struct iphdr *iph = ip_hdr(skb);
+ state->DF = DF;
state->hlen = hlen;
state->ll_rs = ll_rs;
state->mtu = mtu;
@@ -668,9 +669,6 @@ static void ip_frag_ipcb(struct sk_buff *from, struct sk_buff *to,
/* Copy the flags to each fragment. */
IPCB(to)->flags = IPCB(from)->flags;
- if (IPCB(from)->flags & IPSKB_FRAG_PMTU)
- state->iph->frag_off |= htons(IP_DF);
-
/* ANK: dirty, but effective trick. Upgrade options only if
* the segment to be fragmented was THE FIRST (otherwise,
* options are already fixed) and make it ONCE
@@ -738,6 +736,8 @@ struct sk_buff *ip_frag_next(struct sk_buff *skb, struct ip_frag_state *state)
*/
iph = ip_hdr(skb2);
iph->frag_off = htons((state->offset >> 3));
+ if (state->DF)
+ iph->frag_off |= htons(IP_DF);
/*
* Added AC : If we are fragmenting a fragment that's not the
@@ -771,6 +771,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
struct rtable *rt = skb_rtable(skb);
unsigned int mtu, hlen, ll_rs;
struct ip_fraglist_iter iter;
+ ktime_t tstamp = skb->tstamp;
struct ip_frag_state state;
int err = 0;
@@ -846,6 +847,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
ip_fraglist_prepare(skb, &iter);
}
+ skb->tstamp = tstamp;
err = output(net, sk, skb);
if (!err)
@@ -881,7 +883,8 @@ slow_path:
* Fragment the datagram.
*/
- ip_frag_init(skb, hlen, ll_rs, mtu, &state);
+ ip_frag_init(skb, hlen, ll_rs, mtu, IPCB(skb)->flags & IPSKB_FRAG_PMTU,
+ &state);
/*
* Keep copying data until we run out.
@@ -900,6 +903,7 @@ slow_path:
/*
* Put this fragment into the sending queue.
*/
+ skb2->tstamp = tstamp;
err = output(net, sk, skb2);
if (err)
goto fail;
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 716d5472c022..58007439cffd 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -2289,7 +2289,8 @@ int ipmr_get_route(struct net *net, struct sk_buff *skb,
rcu_read_unlock();
return -ENODEV;
}
- skb2 = skb_clone(skb, GFP_ATOMIC);
+
+ skb2 = skb_realloc_headroom(skb, sizeof(struct iphdr));
if (!skb2) {
read_unlock(&mrt_lock);
rcu_read_unlock();
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 14654876127e..621f83434b24 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1482,7 +1482,7 @@ static bool rt_cache_route(struct fib_nh_common *nhc, struct rtable *rt)
prev = cmpxchg(p, orig, rt);
if (prev == orig) {
if (orig) {
- dst_dev_put(&orig->dst);
+ rt_add_uncached_list(orig);
dst_release(&orig->dst);
}
} else {
@@ -2470,14 +2470,17 @@ struct rtable *ip_route_output_key_hash_rcu(struct net *net, struct flowi4 *fl4,
int orig_oif = fl4->flowi4_oif;
unsigned int flags = 0;
struct rtable *rth;
- int err = -ENETUNREACH;
+ int err;
if (fl4->saddr) {
- rth = ERR_PTR(-EINVAL);
if (ipv4_is_multicast(fl4->saddr) ||
ipv4_is_lbcast(fl4->saddr) ||
- ipv4_is_zeronet(fl4->saddr))
+ ipv4_is_zeronet(fl4->saddr)) {
+ rth = ERR_PTR(-EINVAL);
goto out;
+ }
+
+ rth = ERR_PTR(-ENETUNREACH);
/* I removed check for oif == dev_out->oif here.
It was wrong for two reasons:
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index f98a1882e537..d8876f0e9672 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -326,7 +326,7 @@ void tcp_enter_memory_pressure(struct sock *sk)
{
unsigned long val;
- if (tcp_memory_pressure)
+ if (READ_ONCE(tcp_memory_pressure))
return;
val = jiffies;
@@ -341,7 +341,7 @@ void tcp_leave_memory_pressure(struct sock *sk)
{
unsigned long val;
- if (!tcp_memory_pressure)
+ if (!READ_ONCE(tcp_memory_pressure))
return;
val = xchg(&tcp_memory_pressure, 0);
if (val)
@@ -450,8 +450,8 @@ void tcp_init_sock(struct sock *sk)
icsk->icsk_sync_mss = tcp_sync_mss;
- sk->sk_sndbuf = sock_net(sk)->ipv4.sysctl_tcp_wmem[1];
- sk->sk_rcvbuf = sock_net(sk)->ipv4.sysctl_tcp_rmem[1];
+ WRITE_ONCE(sk->sk_sndbuf, sock_net(sk)->ipv4.sysctl_tcp_wmem[1]);
+ WRITE_ONCE(sk->sk_rcvbuf, sock_net(sk)->ipv4.sysctl_tcp_rmem[1]);
sk_sockets_allocated_inc(sk);
sk->sk_route_forced_caps = NETIF_F_GSO;
@@ -477,7 +477,7 @@ static void tcp_tx_timestamp(struct sock *sk, u16 tsflags)
static inline bool tcp_stream_is_readable(const struct tcp_sock *tp,
int target, struct sock *sk)
{
- return (tp->rcv_nxt - tp->copied_seq >= target) ||
+ return (READ_ONCE(tp->rcv_nxt) - READ_ONCE(tp->copied_seq) >= target) ||
(sk->sk_prot->stream_memory_read ?
sk->sk_prot->stream_memory_read(sk) : false);
}
@@ -543,10 +543,10 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
/* Connected or passive Fast Open socket? */
if (state != TCP_SYN_SENT &&
- (state != TCP_SYN_RECV || tp->fastopen_rsk)) {
+ (state != TCP_SYN_RECV || rcu_access_pointer(tp->fastopen_rsk))) {
int target = sock_rcvlowat(sk, 0, INT_MAX);
- if (tp->urg_seq == tp->copied_seq &&
+ if (READ_ONCE(tp->urg_seq) == READ_ONCE(tp->copied_seq) &&
!sock_flag(sk, SOCK_URGINLINE) &&
tp->urg_data)
target++;
@@ -584,7 +584,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
}
/* This barrier is coupled with smp_wmb() in tcp_reset() */
smp_rmb();
- if (sk->sk_err || !skb_queue_empty(&sk->sk_error_queue))
+ if (sk->sk_err || !skb_queue_empty_lockless(&sk->sk_error_queue))
mask |= EPOLLERR;
return mask;
@@ -607,7 +607,8 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg)
unlock_sock_fast(sk, slow);
break;
case SIOCATMARK:
- answ = tp->urg_data && tp->urg_seq == tp->copied_seq;
+ answ = tp->urg_data &&
+ READ_ONCE(tp->urg_seq) == READ_ONCE(tp->copied_seq);
break;
case SIOCOUTQ:
if (sk->sk_state == TCP_LISTEN)
@@ -616,7 +617,7 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg)
if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV))
answ = 0;
else
- answ = tp->write_seq - tp->snd_una;
+ answ = READ_ONCE(tp->write_seq) - tp->snd_una;
break;
case SIOCOUTQNSD:
if (sk->sk_state == TCP_LISTEN)
@@ -625,7 +626,8 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg)
if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV))
answ = 0;
else
- answ = tp->write_seq - tp->snd_nxt;
+ answ = READ_ONCE(tp->write_seq) -
+ READ_ONCE(tp->snd_nxt);
break;
default:
return -ENOIOCTLCMD;
@@ -657,7 +659,7 @@ static void skb_entail(struct sock *sk, struct sk_buff *skb)
tcb->sacked = 0;
__skb_header_release(skb);
tcp_add_write_queue_tail(sk, skb);
- sk->sk_wmem_queued += skb->truesize;
+ sk_wmem_queued_add(sk, skb->truesize);
sk_mem_charge(sk, skb->truesize);
if (tp->nonagle & TCP_NAGLE_PUSH)
tp->nonagle &= ~TCP_NAGLE_PUSH;
@@ -1032,10 +1034,10 @@ new_segment:
skb->len += copy;
skb->data_len += copy;
skb->truesize += copy;
- sk->sk_wmem_queued += copy;
+ sk_wmem_queued_add(sk, copy);
sk_mem_charge(sk, copy);
skb->ip_summed = CHECKSUM_PARTIAL;
- tp->write_seq += copy;
+ WRITE_ONCE(tp->write_seq, tp->write_seq + copy);
TCP_SKB_CB(skb)->end_seq += copy;
tcp_skb_pcount_set(skb, 0);
@@ -1362,7 +1364,7 @@ new_segment:
if (!copied)
TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_PSH;
- tp->write_seq += copy;
+ WRITE_ONCE(tp->write_seq, tp->write_seq + copy);
TCP_SKB_CB(skb)->end_seq += copy;
tcp_skb_pcount_set(skb, 0);
@@ -1668,9 +1670,9 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc,
sk_eat_skb(sk, skb);
if (!desc->count)
break;
- tp->copied_seq = seq;
+ WRITE_ONCE(tp->copied_seq, seq);
}
- tp->copied_seq = seq;
+ WRITE_ONCE(tp->copied_seq, seq);
tcp_rcv_space_adjust(sk);
@@ -1699,7 +1701,7 @@ int tcp_set_rcvlowat(struct sock *sk, int val)
else
cap = sock_net(sk)->ipv4.sysctl_tcp_rmem[2] >> 1;
val = min(val, cap);
- sk->sk_rcvlowat = val ? : 1;
+ WRITE_ONCE(sk->sk_rcvlowat, val ? : 1);
/* Check if we need to signal EPOLLIN right now */
tcp_data_ready(sk);
@@ -1709,7 +1711,7 @@ int tcp_set_rcvlowat(struct sock *sk, int val)
val <<= 1;
if (val > sk->sk_rcvbuf) {
- sk->sk_rcvbuf = val;
+ WRITE_ONCE(sk->sk_rcvbuf, val);
tcp_sk(sk)->window_clamp = tcp_win_from_space(sk, val);
}
return 0;
@@ -1819,7 +1821,7 @@ static int tcp_zerocopy_receive(struct sock *sk,
out:
up_read(&current->mm->mmap_sem);
if (length) {
- tp->copied_seq = seq;
+ WRITE_ONCE(tp->copied_seq, seq);
tcp_rcv_space_adjust(sk);
/* Clean up data we have read: This will do ACK frames. */
@@ -1962,7 +1964,7 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock,
if (unlikely(flags & MSG_ERRQUEUE))
return inet_recv_error(sk, msg, len, addr_len);
- if (sk_can_busy_loop(sk) && skb_queue_empty(&sk->sk_receive_queue) &&
+ if (sk_can_busy_loop(sk) && skb_queue_empty_lockless(&sk->sk_receive_queue) &&
(sk->sk_state == TCP_ESTABLISHED))
sk_busy_loop(sk, nonblock);
@@ -2117,7 +2119,7 @@ found_ok_skb:
if (urg_offset < used) {
if (!urg_offset) {
if (!sock_flag(sk, SOCK_URGINLINE)) {
- ++*seq;
+ WRITE_ONCE(*seq, *seq + 1);
urg_hole++;
offset++;
used--;
@@ -2139,7 +2141,7 @@ found_ok_skb:
}
}
- *seq += used;
+ WRITE_ONCE(*seq, *seq + used);
copied += used;
len -= used;
@@ -2166,7 +2168,7 @@ skip_copy:
found_fin_ok:
/* Process the FIN. */
- ++*seq;
+ WRITE_ONCE(*seq, *seq + 1);
if (!(flags & MSG_PEEK))
sk_eat_skb(sk, skb);
break;
@@ -2487,7 +2489,10 @@ adjudge_to_death:
}
if (sk->sk_state == TCP_CLOSE) {
- struct request_sock *req = tcp_sk(sk)->fastopen_rsk;
+ struct request_sock *req;
+
+ req = rcu_dereference_protected(tcp_sk(sk)->fastopen_rsk,
+ lockdep_sock_is_held(sk));
/* We could get here with a non-NULL req if the socket is
* aborted (e.g., closed with unread data) before 3WHS
* finishes.
@@ -2559,6 +2564,7 @@ int tcp_disconnect(struct sock *sk, int flags)
struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
int old_state = sk->sk_state;
+ u32 seq;
if (old_state != TCP_CLOSE)
tcp_set_state(sk, TCP_CLOSE);
@@ -2585,7 +2591,7 @@ int tcp_disconnect(struct sock *sk, int flags)
__kfree_skb(sk->sk_rx_skb_cache);
sk->sk_rx_skb_cache = NULL;
}
- tp->copied_seq = tp->rcv_nxt;
+ WRITE_ONCE(tp->copied_seq, tp->rcv_nxt);
tp->urg_data = 0;
tcp_write_queue_purge(sk);
tcp_fastopen_active_disable_ofo_check(sk);
@@ -2601,9 +2607,12 @@ int tcp_disconnect(struct sock *sk, int flags)
tp->srtt_us = 0;
tp->mdev_us = jiffies_to_usecs(TCP_TIMEOUT_INIT);
tp->rcv_rtt_last_tsecr = 0;
- tp->write_seq += tp->max_window + 2;
- if (tp->write_seq == 0)
- tp->write_seq = 1;
+
+ seq = tp->write_seq + tp->max_window + 2;
+ if (!seq)
+ seq = 1;
+ WRITE_ONCE(tp->write_seq, seq);
+
icsk->icsk_backoff = 0;
tp->snd_cwnd = 2;
icsk->icsk_probes_out = 0;
@@ -2930,9 +2939,9 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
if (sk->sk_state != TCP_CLOSE)
err = -EPERM;
else if (tp->repair_queue == TCP_SEND_QUEUE)
- tp->write_seq = val;
+ WRITE_ONCE(tp->write_seq, val);
else if (tp->repair_queue == TCP_RECV_QUEUE)
- tp->rcv_nxt = val;
+ WRITE_ONCE(tp->rcv_nxt, val);
else
err = -EINVAL;
break;
@@ -3831,7 +3840,13 @@ EXPORT_SYMBOL(tcp_md5_hash_key);
void tcp_done(struct sock *sk)
{
- struct request_sock *req = tcp_sk(sk)->fastopen_rsk;
+ struct request_sock *req;
+
+ /* We might be called with a new socket, after
+ * inet_csk_prepare_forced_close() has been called
+ * so we can not use lockdep_sock_is_held(sk)
+ */
+ req = rcu_dereference_protected(tcp_sk(sk)->fastopen_rsk, 1);
if (sk->sk_state == TCP_SYN_SENT || sk->sk_state == TCP_SYN_RECV)
TCP_INC_STATS(sock_net(sk), TCP_MIB_ATTEMPTFAILS);
diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c
index 81a8221d650a..549506162dde 100644
--- a/net/ipv4/tcp_diag.c
+++ b/net/ipv4/tcp_diag.c
@@ -26,8 +26,9 @@ static void tcp_diag_get_info(struct sock *sk, struct inet_diag_msg *r,
} else if (sk->sk_type == SOCK_STREAM) {
const struct tcp_sock *tp = tcp_sk(sk);
- r->idiag_rqueue = max_t(int, tp->rcv_nxt - tp->copied_seq, 0);
- r->idiag_wqueue = tp->write_seq - tp->snd_una;
+ r->idiag_rqueue = max_t(int, READ_ONCE(tp->rcv_nxt) -
+ READ_ONCE(tp->copied_seq), 0);
+ r->idiag_wqueue = READ_ONCE(tp->write_seq) - tp->snd_una;
}
if (info)
tcp_get_info(sk, info);
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index 3fd451271a70..a915ade0c818 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -253,7 +253,7 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk,
*/
tp = tcp_sk(child);
- tp->fastopen_rsk = req;
+ rcu_assign_pointer(tp->fastopen_rsk, req);
tcp_rsk(req)->tfo_listener = true;
/* RFC1323: The window in SYN & SYN/ACK segments is never
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 3578357abe30..a2e52ad7cdab 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -359,7 +359,8 @@ static void tcp_sndbuf_expand(struct sock *sk)
sndmem *= nr_segs * per_mss;
if (sk->sk_sndbuf < sndmem)
- sk->sk_sndbuf = min(sndmem, sock_net(sk)->ipv4.sysctl_tcp_wmem[2]);
+ WRITE_ONCE(sk->sk_sndbuf,
+ min(sndmem, sock_net(sk)->ipv4.sysctl_tcp_wmem[2]));
}
/* 2. Tuning advertised window (window_clamp, rcv_ssthresh)
@@ -483,8 +484,9 @@ static void tcp_clamp_window(struct sock *sk)
!(sk->sk_userlocks & SOCK_RCVBUF_LOCK) &&
!tcp_under_memory_pressure(sk) &&
sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)) {
- sk->sk_rcvbuf = min(atomic_read(&sk->sk_rmem_alloc),
- net->ipv4.sysctl_tcp_rmem[2]);
+ WRITE_ONCE(sk->sk_rcvbuf,
+ min(atomic_read(&sk->sk_rmem_alloc),
+ net->ipv4.sysctl_tcp_rmem[2]));
}
if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf)
tp->rcv_ssthresh = min(tp->window_clamp, 2U * tp->advmss);
@@ -648,7 +650,7 @@ void tcp_rcv_space_adjust(struct sock *sk)
rcvbuf = min_t(u64, rcvwin * rcvmem,
sock_net(sk)->ipv4.sysctl_tcp_rmem[2]);
if (rcvbuf > sk->sk_rcvbuf) {
- sk->sk_rcvbuf = rcvbuf;
+ WRITE_ONCE(sk->sk_rcvbuf, rcvbuf);
/* Make the window clamp follow along. */
tp->window_clamp = tcp_win_from_space(sk, rcvbuf);
@@ -2666,7 +2668,7 @@ static void tcp_process_loss(struct sock *sk, int flag, int num_dupack,
struct tcp_sock *tp = tcp_sk(sk);
bool recovered = !before(tp->snd_una, tp->high_seq);
- if ((flag & FLAG_SND_UNA_ADVANCED || tp->fastopen_rsk) &&
+ if ((flag & FLAG_SND_UNA_ADVANCED || rcu_access_pointer(tp->fastopen_rsk)) &&
tcp_try_undo_loss(sk, false))
return;
@@ -2990,7 +2992,7 @@ void tcp_rearm_rto(struct sock *sk)
/* If the retrans timer is currently being used by Fast Open
* for SYN-ACK retrans purpose, stay put.
*/
- if (tp->fastopen_rsk)
+ if (rcu_access_pointer(tp->fastopen_rsk))
return;
if (!tp->packets_out) {
@@ -3362,7 +3364,7 @@ static void tcp_rcv_nxt_update(struct tcp_sock *tp, u32 seq)
sock_owned_by_me((struct sock *)tp);
tp->bytes_received += delta;
- tp->rcv_nxt = seq;
+ WRITE_ONCE(tp->rcv_nxt, seq);
}
/* Update our send window.
@@ -5356,7 +5358,7 @@ static void tcp_check_urg(struct sock *sk, const struct tcphdr *th)
}
tp->urg_data = TCP_URG_NOTYET;
- tp->urg_seq = ptr;
+ WRITE_ONCE(tp->urg_seq, ptr);
/* Disable header prediction. */
tp->pred_flags = 0;
@@ -5932,7 +5934,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
/* Ok.. it's good. Set up sequence numbers and
* move to established.
*/
- tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1;
+ WRITE_ONCE(tp->rcv_nxt, TCP_SKB_CB(skb)->seq + 1);
tp->rcv_wup = TCP_SKB_CB(skb)->seq + 1;
/* RFC1323: The window in SYN & SYN/ACK segments is
@@ -5961,7 +5963,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
/* Remember, tcp_poll() does not lock socket!
* Change state from SYN-SENT only after copied_seq
* is initialized. */
- tp->copied_seq = tp->rcv_nxt;
+ WRITE_ONCE(tp->copied_seq, tp->rcv_nxt);
smc_check_reset_syn(tp);
@@ -6035,8 +6037,8 @@ discard:
tp->tcp_header_len = sizeof(struct tcphdr);
}
- tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1;
- tp->copied_seq = tp->rcv_nxt;
+ WRITE_ONCE(tp->rcv_nxt, TCP_SKB_CB(skb)->seq + 1);
+ WRITE_ONCE(tp->copied_seq, tp->rcv_nxt);
tp->rcv_wup = TCP_SKB_CB(skb)->seq + 1;
/* RFC1323: The window in SYN & SYN/ACK segments is
@@ -6087,6 +6089,8 @@ reset_and_undo:
static void tcp_rcv_synrecv_state_fastopen(struct sock *sk)
{
+ struct request_sock *req;
+
tcp_try_undo_loss(sk, false);
/* Reset rtx states to prevent spurious retransmits_timed_out() */
@@ -6096,7 +6100,9 @@ static void tcp_rcv_synrecv_state_fastopen(struct sock *sk)
/* Once we leave TCP_SYN_RECV or TCP_FIN_WAIT_1,
* we no longer need req so release it.
*/
- reqsk_fastopen_remove(sk, tcp_sk(sk)->fastopen_rsk, false);
+ req = rcu_dereference_protected(tcp_sk(sk)->fastopen_rsk,
+ lockdep_sock_is_held(sk));
+ reqsk_fastopen_remove(sk, req, false);
/* Re-arm the timer because data may have been sent out.
* This is similar to the regular data transmission case
@@ -6171,7 +6177,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
tcp_mstamp_refresh(tp);
tp->rx_opt.saw_tstamp = 0;
- req = tp->fastopen_rsk;
+ req = rcu_dereference_protected(tp->fastopen_rsk,
+ lockdep_sock_is_held(sk));
if (req) {
bool req_stolen;
@@ -6211,7 +6218,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
tcp_try_undo_spurious_syn(sk);
tp->retrans_stamp = 0;
tcp_init_transfer(sk, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB);
- tp->copied_seq = tp->rcv_nxt;
+ WRITE_ONCE(tp->copied_seq, tp->rcv_nxt);
}
smp_mb();
tcp_set_state(sk, TCP_ESTABLISHED);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index bf124b1742df..67b2dc7a1727 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -164,9 +164,11 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp)
* without appearing to create any others.
*/
if (likely(!tp->repair)) {
- tp->write_seq = tcptw->tw_snd_nxt + 65535 + 2;
- if (tp->write_seq == 0)
- tp->write_seq = 1;
+ u32 seq = tcptw->tw_snd_nxt + 65535 + 2;
+
+ if (!seq)
+ seq = 1;
+ WRITE_ONCE(tp->write_seq, seq);
tp->rx_opt.ts_recent = tcptw->tw_ts_recent;
tp->rx_opt.ts_recent_stamp = tcptw->tw_ts_recent_stamp;
}
@@ -253,7 +255,7 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
tp->rx_opt.ts_recent = 0;
tp->rx_opt.ts_recent_stamp = 0;
if (likely(!tp->repair))
- tp->write_seq = 0;
+ WRITE_ONCE(tp->write_seq, 0);
}
inet->inet_dport = usin->sin_port;
@@ -291,16 +293,17 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
if (likely(!tp->repair)) {
if (!tp->write_seq)
- tp->write_seq = secure_tcp_seq(inet->inet_saddr,
- inet->inet_daddr,
- inet->inet_sport,
- usin->sin_port);
+ WRITE_ONCE(tp->write_seq,
+ secure_tcp_seq(inet->inet_saddr,
+ inet->inet_daddr,
+ inet->inet_sport,
+ usin->sin_port));
tp->tsoffset = secure_tcp_ts_off(sock_net(sk),
inet->inet_saddr,
inet->inet_daddr);
}
- inet->inet_id = tp->write_seq ^ jiffies;
+ inet->inet_id = prandom_u32();
if (tcp_fastopen_defer_connect(sk, &err))
return err;
@@ -478,7 +481,7 @@ int tcp_v4_err(struct sk_buff *icmp_skb, u32 info)
icsk = inet_csk(sk);
tp = tcp_sk(sk);
/* XXX (TFO) - tp->snd_una should be ISN (tcp_create_openreq_child() */
- fastopen = tp->fastopen_rsk;
+ fastopen = rcu_dereference(tp->fastopen_rsk);
snd_una = fastopen ? tcp_rsk(fastopen)->snt_isn : tp->snd_una;
if (sk->sk_state != TCP_LISTEN &&
!between(seq, snd_una, tp->snd_nxt)) {
@@ -1447,7 +1450,7 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb,
inet_csk(newsk)->icsk_ext_hdr_len = 0;
if (inet_opt)
inet_csk(newsk)->icsk_ext_hdr_len = inet_opt->opt.optlen;
- newinet->inet_id = newtp->write_seq ^ jiffies;
+ newinet->inet_id = prandom_u32();
if (!dst) {
dst = inet_csk_route_child_sock(sk, newsk, req);
@@ -1644,7 +1647,7 @@ int tcp_v4_early_demux(struct sk_buff *skb)
bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb)
{
- u32 limit = sk->sk_rcvbuf + sk->sk_sndbuf;
+ u32 limit = READ_ONCE(sk->sk_rcvbuf) + READ_ONCE(sk->sk_sndbuf);
struct skb_shared_info *shinfo;
const struct tcphdr *th;
struct tcphdr *thtail;
@@ -2121,7 +2124,7 @@ void tcp_v4_destroy_sock(struct sock *sk)
if (inet_csk(sk)->icsk_bind_hash)
inet_put_port(sk);
- BUG_ON(tp->fastopen_rsk);
+ BUG_ON(rcu_access_pointer(tp->fastopen_rsk));
/* If socket is aborted during connect operation */
tcp_free_fastopen_req(tp);
@@ -2455,12 +2458,13 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i)
/* Because we don't lock the socket,
* we might find a transient negative value.
*/
- rx_queue = max_t(int, tp->rcv_nxt - tp->copied_seq, 0);
+ rx_queue = max_t(int, READ_ONCE(tp->rcv_nxt) -
+ READ_ONCE(tp->copied_seq), 0);
seq_printf(f, "%4d: %08X:%04X %08X:%04X %02X %08X:%08X %02X:%08lX "
"%08X %5u %8d %lu %d %pK %lu %lu %u %u %d",
i, src, srcp, dest, destp, state,
- tp->write_seq - tp->snd_una,
+ READ_ONCE(tp->write_seq) - tp->snd_una,
rx_queue,
timer_active,
jiffies_delta_to_clock_t(timer_expires - jiffies),
@@ -2677,7 +2681,7 @@ static int __net_init tcp_sk_init(struct net *net)
net->ipv4.tcp_death_row.sysctl_max_tw_buckets = cnt / 2;
net->ipv4.tcp_death_row.hashinfo = &tcp_hashinfo;
- net->ipv4.sysctl_max_syn_backlog = max(128, cnt / 256);
+ net->ipv4.sysctl_max_syn_backlog = max(128, cnt / 128);
net->ipv4.sysctl_tcp_sack = 1;
net->ipv4.sysctl_tcp_window_scaling = 1;
net->ipv4.sysctl_tcp_timestamps = 1;
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index bb140a5db8c0..c802bc80c400 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -462,6 +462,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
struct tcp_request_sock *treq = tcp_rsk(req);
struct inet_connection_sock *newicsk;
struct tcp_sock *oldtp, *newtp;
+ u32 seq;
if (!newsk)
return NULL;
@@ -475,12 +476,16 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
/* Now setup tcp_sock */
newtp->pred_flags = 0;
- newtp->rcv_wup = newtp->copied_seq =
- newtp->rcv_nxt = treq->rcv_isn + 1;
+ seq = treq->rcv_isn + 1;
+ newtp->rcv_wup = seq;
+ WRITE_ONCE(newtp->copied_seq, seq);
+ WRITE_ONCE(newtp->rcv_nxt, seq);
newtp->segs_in = 1;
- newtp->snd_sml = newtp->snd_una =
- newtp->snd_nxt = newtp->snd_up = treq->snt_isn + 1;
+ seq = treq->snt_isn + 1;
+ newtp->snd_sml = newtp->snd_una = seq;
+ WRITE_ONCE(newtp->snd_nxt, seq);
+ newtp->snd_up = seq;
INIT_LIST_HEAD(&newtp->tsq_node);
INIT_LIST_HEAD(&newtp->tsorted_sent_queue);
@@ -495,7 +500,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
newtp->total_retrans = req->num_retrans;
tcp_init_xmit_timers(newsk);
- newtp->write_seq = newtp->pushed_seq = treq->snt_isn + 1;
+ WRITE_ONCE(newtp->write_seq, newtp->pushed_seq = treq->snt_isn + 1);
if (sock_flag(newsk, SOCK_KEEPOPEN))
inet_csk_reset_keepalive_timer(newsk,
@@ -541,7 +546,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
newtp->rx_opt.mss_clamp = req->mss;
tcp_ecn_openreq_child(newtp, req);
newtp->fastopen_req = NULL;
- newtp->fastopen_rsk = NULL;
+ RCU_INIT_POINTER(newtp->fastopen_rsk, NULL);
__TCP_INC_STATS(sock_net(sk), TCP_MIB_PASSIVEOPENS);
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index fec6d67bfd14..0488607c5cd3 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -67,7 +67,7 @@ static void tcp_event_new_data_sent(struct sock *sk, struct sk_buff *skb)
struct tcp_sock *tp = tcp_sk(sk);
unsigned int prior_packets = tp->packets_out;
- tp->snd_nxt = TCP_SKB_CB(skb)->end_seq;
+ WRITE_ONCE(tp->snd_nxt, TCP_SKB_CB(skb)->end_seq);
__skb_unlink(skb, &sk->sk_write_queue);
tcp_rbtree_insert(&sk->tcp_rtx_queue, skb);
@@ -1196,10 +1196,10 @@ static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb)
struct tcp_sock *tp = tcp_sk(sk);
/* Advance write_seq and place onto the write_queue. */
- tp->write_seq = TCP_SKB_CB(skb)->end_seq;
+ WRITE_ONCE(tp->write_seq, TCP_SKB_CB(skb)->end_seq);
__skb_header_release(skb);
tcp_add_write_queue_tail(sk, skb);
- sk->sk_wmem_queued += skb->truesize;
+ sk_wmem_queued_add(sk, skb->truesize);
sk_mem_charge(sk, skb->truesize);
}
@@ -1333,7 +1333,7 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue,
return -ENOMEM; /* We'll just try again later. */
skb_copy_decrypted(buff, skb);
- sk->sk_wmem_queued += buff->truesize;
+ sk_wmem_queued_add(sk, buff->truesize);
sk_mem_charge(sk, buff->truesize);
nlen = skb->len - len - nsize;
buff->truesize += nlen;
@@ -1443,7 +1443,7 @@ int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len)
if (delta_truesize) {
skb->truesize -= delta_truesize;
- sk->sk_wmem_queued -= delta_truesize;
+ sk_wmem_queued_add(sk, -delta_truesize);
sk_mem_uncharge(sk, delta_truesize);
sock_set_flag(sk, SOCK_QUEUE_SHRUNK);
}
@@ -1888,7 +1888,7 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len,
return -ENOMEM;
skb_copy_decrypted(buff, skb);
- sk->sk_wmem_queued += buff->truesize;
+ sk_wmem_queued_add(sk, buff->truesize);
sk_mem_charge(sk, buff->truesize);
buff->truesize += nlen;
skb->truesize -= nlen;
@@ -2152,7 +2152,7 @@ static int tcp_mtu_probe(struct sock *sk)
nskb = sk_stream_alloc_skb(sk, probe_size, GFP_ATOMIC, false);
if (!nskb)
return -1;
- sk->sk_wmem_queued += nskb->truesize;
+ sk_wmem_queued_add(sk, nskb->truesize);
sk_mem_charge(sk, nskb->truesize);
skb = tcp_send_head(sk);
@@ -2482,7 +2482,7 @@ bool tcp_schedule_loss_probe(struct sock *sk, bool advancing_rto)
/* Don't do any loss probe on a Fast Open connection before 3WHS
* finishes.
*/
- if (tp->fastopen_rsk)
+ if (rcu_access_pointer(tp->fastopen_rsk))
return false;
early_retrans = sock_net(sk)->ipv4.sysctl_tcp_early_retrans;
@@ -3142,7 +3142,7 @@ void tcp_send_fin(struct sock *sk)
* if FIN had been sent. This is because retransmit path
* does not change tp->snd_nxt.
*/
- tp->snd_nxt++;
+ WRITE_ONCE(tp->snd_nxt, tp->snd_nxt + 1);
return;
}
} else {
@@ -3222,7 +3222,7 @@ int tcp_send_synack(struct sock *sk)
tcp_rtx_queue_unlink_and_free(skb, sk);
__skb_header_release(nskb);
tcp_rbtree_insert(&sk->tcp_rtx_queue, nskb);
- sk->sk_wmem_queued += nskb->truesize;
+ sk_wmem_queued_add(sk, nskb->truesize);
sk_mem_charge(sk, nskb->truesize);
skb = nskb;
}
@@ -3426,14 +3426,14 @@ static void tcp_connect_init(struct sock *sk)
tp->snd_una = tp->write_seq;
tp->snd_sml = tp->write_seq;
tp->snd_up = tp->write_seq;
- tp->snd_nxt = tp->write_seq;
+ WRITE_ONCE(tp->snd_nxt, tp->write_seq);
if (likely(!tp->repair))
tp->rcv_nxt = 0;
else
tp->rcv_tstamp = tcp_jiffies32;
tp->rcv_wup = tp->rcv_nxt;
- tp->copied_seq = tp->rcv_nxt;
+ WRITE_ONCE(tp->copied_seq, tp->rcv_nxt);
inet_csk(sk)->icsk_rto = tcp_timeout_init(sk);
inet_csk(sk)->icsk_retransmits = 0;
@@ -3447,9 +3447,9 @@ static void tcp_connect_queue_skb(struct sock *sk, struct sk_buff *skb)
tcb->end_seq += skb->len;
__skb_header_release(skb);
- sk->sk_wmem_queued += skb->truesize;
+ sk_wmem_queued_add(sk, skb->truesize);
sk_mem_charge(sk, skb->truesize);
- tp->write_seq = tcb->end_seq;
+ WRITE_ONCE(tp->write_seq, tcb->end_seq);
tp->packets_out += tcp_skb_pcount(skb);
}
@@ -3586,11 +3586,11 @@ int tcp_connect(struct sock *sk)
/* We change tp->snd_nxt after the tcp_transmit_skb() call
* in order to make this packet get counted in tcpOutSegs.
*/
- tp->snd_nxt = tp->write_seq;
+ WRITE_ONCE(tp->snd_nxt, tp->write_seq);
tp->pushed_seq = tp->write_seq;
buff = tcp_send_head(sk);
if (unlikely(buff)) {
- tp->snd_nxt = TCP_SKB_CB(buff)->seq;
+ WRITE_ONCE(tp->snd_nxt, TCP_SKB_CB(buff)->seq);
tp->pushed_seq = TCP_SKB_CB(buff)->seq;
}
TCP_INC_STATS(sock_net(sk), TCP_MIB_ACTIVEOPENS);
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 05be564414e9..dd5a6317a801 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -386,15 +386,13 @@ abort: tcp_write_err(sk);
* Timer for Fast Open socket to retransmit SYNACK. Note that the
* sk here is the child socket, not the parent (listener) socket.
*/
-static void tcp_fastopen_synack_timer(struct sock *sk)
+static void tcp_fastopen_synack_timer(struct sock *sk, struct request_sock *req)
{
struct inet_connection_sock *icsk = inet_csk(sk);
int max_retries = icsk->icsk_syn_retries ? :
sock_net(sk)->ipv4.sysctl_tcp_synack_retries + 1; /* add one more retry for fastopen */
struct tcp_sock *tp = tcp_sk(sk);
- struct request_sock *req;
- req = tcp_sk(sk)->fastopen_rsk;
req->rsk_ops->syn_ack_timeout(req);
if (req->num_timeout >= max_retries) {
@@ -435,11 +433,14 @@ void tcp_retransmit_timer(struct sock *sk)
struct tcp_sock *tp = tcp_sk(sk);
struct net *net = sock_net(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
+ struct request_sock *req;
- if (tp->fastopen_rsk) {
+ req = rcu_dereference_protected(tp->fastopen_rsk,
+ lockdep_sock_is_held(sk));
+ if (req) {
WARN_ON_ONCE(sk->sk_state != TCP_SYN_RECV &&
sk->sk_state != TCP_FIN_WAIT1);
- tcp_fastopen_synack_timer(sk);
+ tcp_fastopen_synack_timer(sk, req);
/* Before we receive ACK to our SYN-ACK don't retransmit
* anything else (e.g., data or FIN segments).
*/
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 14bc654b6842..1d58ce829dca 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -388,7 +388,7 @@ static int compute_score(struct sock *sk, struct net *net,
return -1;
score += 4;
- if (sk->sk_incoming_cpu == raw_smp_processor_id())
+ if (READ_ONCE(sk->sk_incoming_cpu) == raw_smp_processor_id())
score++;
return score;
}
@@ -1316,6 +1316,20 @@ static void udp_set_dev_scratch(struct sk_buff *skb)
scratch->_tsize_state |= UDP_SKB_IS_STATELESS;
}
+static void udp_skb_csum_unnecessary_set(struct sk_buff *skb)
+{
+ /* We come here after udp_lib_checksum_complete() returned 0.
+ * This means that __skb_checksum_complete() might have
+ * set skb->csum_valid to 1.
+ * On 64bit platforms, we can set csum_unnecessary
+ * to true, but only if the skb is not shared.
+ */
+#if BITS_PER_LONG == 64
+ if (!skb_shared(skb))
+ udp_skb_scratch(skb)->csum_unnecessary = true;
+#endif
+}
+
static int udp_skb_truesize(struct sk_buff *skb)
{
return udp_skb_scratch(skb)->_tsize_state & ~UDP_SKB_IS_STATELESS;
@@ -1550,10 +1564,7 @@ static struct sk_buff *__first_packet_length(struct sock *sk,
*total += skb->truesize;
kfree_skb(skb);
} else {
- /* the csum related bits could be changed, refresh
- * the scratch area
- */
- udp_set_dev_scratch(skb);
+ udp_skb_csum_unnecessary_set(skb);
break;
}
}
@@ -1577,7 +1588,7 @@ static int first_packet_length(struct sock *sk)
spin_lock_bh(&rcvq->lock);
skb = __first_packet_length(sk, rcvq, &total);
- if (!skb && !skb_queue_empty(sk_queue)) {
+ if (!skb && !skb_queue_empty_lockless(sk_queue)) {
spin_lock(&sk_queue->lock);
skb_queue_splice_tail_init(sk_queue, rcvq);
spin_unlock(&sk_queue->lock);
@@ -1650,7 +1661,7 @@ struct sk_buff *__skb_recv_udp(struct sock *sk, unsigned int flags,
return skb;
}
- if (skb_queue_empty(sk_queue)) {
+ if (skb_queue_empty_lockless(sk_queue)) {
spin_unlock_bh(&queue->lock);
goto busy_check;
}
@@ -1676,7 +1687,7 @@ busy_check:
break;
sk_busy_loop(sk, flags & MSG_DONTWAIT);
- } while (!skb_queue_empty(sk_queue));
+ } while (!skb_queue_empty_lockless(sk_queue));
/* sk_queue is empty, reader_queue may contain peeked packets */
} while (timeo &&
@@ -2712,7 +2723,7 @@ __poll_t udp_poll(struct file *file, struct socket *sock, poll_table *wait)
__poll_t mask = datagram_poll(file, sock, wait);
struct sock *sk = sock->sk;
- if (!skb_queue_empty(&udp_sk(sk)->reader_queue))
+ if (!skb_queue_empty_lockless(&udp_sk(sk)->reader_queue))
mask |= EPOLLIN | EPOLLRDNORM;
/* Check for false positives due to checksum errors */
generated by cgit v1.2.3-58-ga151 (git 2.39.0) at 2024-11-30 10:53:29 +0000