[PATCH net 0/5] net: another round of data-race annotations

[PATCH net 0/5] net: another round of data-race annotations

[Eric Dumazet]

Series inspired by some syzbot reports, taking care
of 4 socket fields that can be read locklessly.

Eric Dumazet (5):
  net: use sk_forward_alloc_get() in sk_get_meminfo()
  net: annotate data-races around sk->sk_forward_alloc
  mptcp: annotate data-races around msk->rmem_fwd_alloc
  net: annotate data-races around sk->sk_tsflags
  net: annotate data-races around sk->sk_bind_phc

 include/net/ip.h       |  2 +-
 include/net/sock.h     | 29 +++++++++++++++++++----------
 net/can/j1939/socket.c | 10 ++++++----
 net/core/skbuff.c      | 10 ++++++----
 net/core/sock.c        | 18 +++++++++---------
 net/ipv4/ip_output.c   |  2 +-
 net/ipv4/ip_sockglue.c |  2 +-
 net/ipv4/tcp.c         |  4 ++--
 net/ipv4/tcp_output.c  |  2 +-
 net/ipv4/udp.c         |  6 +++---
 net/ipv6/ip6_output.c  |  2 +-
 net/ipv6/ping.c        |  2 +-
 net/ipv6/raw.c         |  2 +-
 net/ipv6/udp.c         |  2 +-
 net/mptcp/protocol.c   | 23 +++++++++++++++--------
 net/socket.c           | 15 ++++++++-------
 16 files changed, 76 insertions(+), 55 deletions(-)

-- 
2.42.0.rc2.253.gd59a3bf2b4-goog

[PATCH net 1/5] net: use sk_forward_alloc_get() in sk_get_meminfo()

[Eric Dumazet]

inet_sk_diag_fill() has been changed to use sk_forward_alloc_get(),
but sk_get_meminfo() was forgotten.

Fixes: 292e6077b040 ("net: introduce sk_forward_alloc_get()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 net/core/sock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/sock.c b/net/core/sock.c
index b0dd501dabd6a86aeda148007faf35c3c50a2bba..a61ec97098add711e0e5bc587733b0579aab77b7 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -3743,7 +3743,7 @@ void sk_get_meminfo(const struct sock *sk, u32 *mem)
 	mem[SK_MEMINFO_RCVBUF] = READ_ONCE(sk->sk_rcvbuf);
 	mem[SK_MEMINFO_WMEM_ALLOC] = sk_wmem_alloc_get(sk);
 	mem[SK_MEMINFO_SNDBUF] = READ_ONCE(sk->sk_sndbuf);
-	mem[SK_MEMINFO_FWD_ALLOC] = sk->sk_forward_alloc;
+	mem[SK_MEMINFO_FWD_ALLOC] = sk_forward_alloc_get(sk);
 	mem[SK_MEMINFO_WMEM_QUEUED] = READ_ONCE(sk->sk_wmem_queued);
 	mem[SK_MEMINFO_OPTMEM] = atomic_read(&sk->sk_omem_alloc);
 	mem[SK_MEMINFO_BACKLOG] = READ_ONCE(sk->sk_backlog.len);
-- 
2.42.0.rc2.253.gd59a3bf2b4-goog

[PATCH net 2/5] net: annotate data-races around sk->sk_forward_alloc

[Eric Dumazet]

Every time sk->sk_forward_alloc is read locklessly,
add a READ_ONCE().

Add sk_forward_alloc_add() helper to centralize updates,
to reduce number of WRITE_ONCE().

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 include/net/sock.h    | 12 +++++++++---
 net/core/sock.c       |  8 ++++----
 net/ipv4/tcp_output.c |  2 +-
 net/ipv4/udp.c        |  6 +++---
 net/mptcp/protocol.c  |  6 +++---
 5 files changed, 20 insertions(+), 14 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 11d503417591ed758fe2b2acbf7d3a9d0f07b660..f04869ac1d92283bc3e4ecea2909babfc5096003 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1053,6 +1053,12 @@ static inline void sk_wmem_queued_add(struct sock *sk, int val)
 	WRITE_ONCE(sk->sk_wmem_queued, sk->sk_wmem_queued + val);
 }
 
+static inline void sk_forward_alloc_add(struct sock *sk, int val)
+{
+	/* Paired with lockless reads of sk->sk_forward_alloc */
+	WRITE_ONCE(sk->sk_forward_alloc, sk->sk_forward_alloc + val);
+}
+
 void sk_stream_write_space(struct sock *sk);
 
 /* OOB backlog add */
@@ -1377,7 +1383,7 @@ static inline int sk_forward_alloc_get(const struct sock *sk)
 	if (sk->sk_prot->forward_alloc_get)
 		return sk->sk_prot->forward_alloc_get(sk);
 #endif
-	return sk->sk_forward_alloc;
+	return READ_ONCE(sk->sk_forward_alloc);
 }
 
 static inline bool __sk_stream_memory_free(const struct sock *sk, int wake)
@@ -1673,14 +1679,14 @@ static inline void sk_mem_charge(struct sock *sk, int size)
 {
 	if (!sk_has_account(sk))
 		return;
-	sk->sk_forward_alloc -= size;
+	sk_forward_alloc_add(sk, -size);
 }
 
 static inline void sk_mem_uncharge(struct sock *sk, int size)
 {
 	if (!sk_has_account(sk))
 		return;
-	sk->sk_forward_alloc += size;
+	sk_forward_alloc_add(sk, size);
 	sk_mem_reclaim(sk);
 }
 
diff --git a/net/core/sock.c b/net/core/sock.c
index a61ec97098add711e0e5bc587733b0579aab77b7..40e1bda4bde0cd0ddbb0315deb4df45d60f65081 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1045,7 +1045,7 @@ static int sock_reserve_memory(struct sock *sk, int bytes)
 		mem_cgroup_uncharge_skmem(sk->sk_memcg, pages);
 		return -ENOMEM;
 	}
-	sk->sk_forward_alloc += pages << PAGE_SHIFT;
+	sk_forward_alloc_add(sk, pages << PAGE_SHIFT);
 
 	WRITE_ONCE(sk->sk_reserved_mem,
 		   sk->sk_reserved_mem + (pages << PAGE_SHIFT));
@@ -3139,10 +3139,10 @@ int __sk_mem_schedule(struct sock *sk, int size, int kind)
 {
 	int ret, amt = sk_mem_pages(size);
 
-	sk->sk_forward_alloc += amt << PAGE_SHIFT;
+	sk_forward_alloc_add(sk, amt << PAGE_SHIFT);
 	ret = __sk_mem_raise_allocated(sk, size, amt, kind);
 	if (!ret)
-		sk->sk_forward_alloc -= amt << PAGE_SHIFT;
+		sk_forward_alloc_add(sk, -(amt << PAGE_SHIFT));
 	return ret;
 }
 EXPORT_SYMBOL(__sk_mem_schedule);
@@ -3174,7 +3174,7 @@ void __sk_mem_reduce_allocated(struct sock *sk, int amount)
 void __sk_mem_reclaim(struct sock *sk, int amount)
 {
 	amount >>= PAGE_SHIFT;
-	sk->sk_forward_alloc -= amount << PAGE_SHIFT;
+	sk_forward_alloc_add(sk, -(amount << PAGE_SHIFT));
 	__sk_mem_reduce_allocated(sk, amount);
 }
 EXPORT_SYMBOL(__sk_mem_reclaim);
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index e6b4fbd642f7e4e3c986fef54e416628c8981aff..ccfc8bbf745586cd23dcf02d755d6981dc92742e 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -3474,7 +3474,7 @@ void sk_forced_mem_schedule(struct sock *sk, int size)
 	if (delta <= 0)
 		return;
 	amt = sk_mem_pages(delta);
-	sk->sk_forward_alloc += amt << PAGE_SHIFT;
+	sk_forward_alloc_add(sk, amt << PAGE_SHIFT);
 	sk_memory_allocated_add(sk, amt);
 
 	if (mem_cgroup_sockets_enabled && sk->sk_memcg)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 0794a2c46a568d644cc488c1d7f6ee676180a5bd..f39b9c8445808deee2c777cbb828474ff105d322 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1414,9 +1414,9 @@ static void udp_rmem_release(struct sock *sk, int size, int partial,
 		spin_lock(&sk_queue->lock);
 
 
-	sk->sk_forward_alloc += size;
+	sk_forward_alloc_add(sk, size);
 	amt = (sk->sk_forward_alloc - partial) & ~(PAGE_SIZE - 1);
-	sk->sk_forward_alloc -= amt;
+	sk_forward_alloc_add(sk, -amt);
 
 	if (amt)
 		__sk_mem_reduce_allocated(sk, amt >> PAGE_SHIFT);
@@ -1527,7 +1527,7 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
 		goto uncharge_drop;
 	}
 
-	sk->sk_forward_alloc -= size;
+	sk_forward_alloc_add(sk, -size);
 
 	/* no need to setup a destructor, we will explicitly release the
 	 * forward allocated memory on dequeue
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 933b257eee0242d9864ca9202cea69f2592663be..625df3a36c469d9b8e71a2f0463a1ca5ead2049d 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1800,7 +1800,7 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		}
 
 		/* data successfully copied into the write queue */
-		sk->sk_forward_alloc -= total_ts;
+		sk_forward_alloc_add(sk, -total_ts);
 		copied += psize;
 		dfrag->data_len += psize;
 		frag_truesize += psize;
@@ -3257,7 +3257,7 @@ void mptcp_destroy_common(struct mptcp_sock *msk, unsigned int flags)
 	/* move all the rx fwd alloc into the sk_mem_reclaim_final in
 	 * inet_sock_destruct() will dispose it
 	 */
-	sk->sk_forward_alloc += msk->rmem_fwd_alloc;
+	sk_forward_alloc_add(sk, msk->rmem_fwd_alloc);
 	msk->rmem_fwd_alloc = 0;
 	mptcp_token_destroy(msk);
 	mptcp_pm_free_anno_list(msk);
@@ -3522,7 +3522,7 @@ static void mptcp_shutdown(struct sock *sk, int how)
 
 static int mptcp_forward_alloc_get(const struct sock *sk)
 {
-	return sk->sk_forward_alloc + mptcp_sk(sk)->rmem_fwd_alloc;
+	return READ_ONCE(sk->sk_forward_alloc) + mptcp_sk(sk)->rmem_fwd_alloc;
 }
 
 static int mptcp_ioctl_outq(const struct mptcp_sock *msk, u64 v)
-- 
2.42.0.rc2.253.gd59a3bf2b4-goog

[PATCH net 3/5] mptcp: annotate data-races around msk->rmem_fwd_alloc

[Eric Dumazet]

msk->rmem_fwd_alloc can be read locklessly.

Add mptcp_rmem_fwd_alloc_add(), similar to sk_forward_alloc_add(),
and appropriate READ_ONCE()/WRITE_ONCE() annotations.

Fixes: 6511882cdd82 ("mptcp: allocate fwd memory separately on the rx and tx path")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/protocol.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 625df3a36c469d9b8e71a2f0463a1ca5ead2049d..a7fc16f5175d22a4e741c79edd84f1ea72c676bc 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -134,9 +134,15 @@ static void mptcp_drop(struct sock *sk, struct sk_buff *skb)
 	__kfree_skb(skb);
 }
 
+static void mptcp_rmem_fwd_alloc_add(struct sock *sk, int size)
+{
+	WRITE_ONCE(mptcp_sk(sk)->rmem_fwd_alloc,
+		   mptcp_sk(sk)->rmem_fwd_alloc + size);
+}
+
 static void mptcp_rmem_charge(struct sock *sk, int size)
 {
-	mptcp_sk(sk)->rmem_fwd_alloc -= size;
+	mptcp_rmem_fwd_alloc_add(sk, -size);
 }
 
 static bool mptcp_try_coalesce(struct sock *sk, struct sk_buff *to,
@@ -177,7 +183,7 @@ static bool mptcp_ooo_try_coalesce(struct mptcp_sock *msk, struct sk_buff *to,
 static void __mptcp_rmem_reclaim(struct sock *sk, int amount)
 {
 	amount >>= PAGE_SHIFT;
-	mptcp_sk(sk)->rmem_fwd_alloc -= amount << PAGE_SHIFT;
+	mptcp_rmem_charge(sk, amount << PAGE_SHIFT);
 	__sk_mem_reduce_allocated(sk, amount);
 }
 
@@ -186,7 +192,7 @@ static void mptcp_rmem_uncharge(struct sock *sk, int size)
 	struct mptcp_sock *msk = mptcp_sk(sk);
 	int reclaimable;
 
-	msk->rmem_fwd_alloc += size;
+	mptcp_rmem_fwd_alloc_add(sk, size);
 	reclaimable = msk->rmem_fwd_alloc - sk_unused_reserved_mem(sk);
 
 	/* see sk_mem_uncharge() for the rationale behind the following schema */
@@ -341,7 +347,7 @@ static bool mptcp_rmem_schedule(struct sock *sk, struct sock *ssk, int size)
 	if (!__sk_mem_raise_allocated(sk, size, amt, SK_MEM_RECV))
 		return false;
 
-	msk->rmem_fwd_alloc += amount;
+	mptcp_rmem_fwd_alloc_add(sk, amount);
 	return true;
 }
 
@@ -3258,7 +3264,7 @@ void mptcp_destroy_common(struct mptcp_sock *msk, unsigned int flags)
 	 * inet_sock_destruct() will dispose it
 	 */
 	sk_forward_alloc_add(sk, msk->rmem_fwd_alloc);
-	msk->rmem_fwd_alloc = 0;
+	WRITE_ONCE(msk->rmem_fwd_alloc, 0);
 	mptcp_token_destroy(msk);
 	mptcp_pm_free_anno_list(msk);
 	mptcp_free_local_addr_list(msk);
@@ -3522,7 +3528,8 @@ static void mptcp_shutdown(struct sock *sk, int how)
 
 static int mptcp_forward_alloc_get(const struct sock *sk)
 {
-	return READ_ONCE(sk->sk_forward_alloc) + mptcp_sk(sk)->rmem_fwd_alloc;
+	return READ_ONCE(sk->sk_forward_alloc) +
+	       READ_ONCE(mptcp_sk(sk)->rmem_fwd_alloc);
 }
 
 static int mptcp_ioctl_outq(const struct mptcp_sock *msk, u64 v)
-- 
2.42.0.rc2.253.gd59a3bf2b4-goog

[PATCH net 4/5] net: annotate data-races around sk->sk_tsflags

[Eric Dumazet]

sk->sk_tsflags can be read locklessly, add corresponding annotations.

Fixes: b9f40e21ef42 ("net-timestamp: move timestamp flags out of sk_flags")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
---
 include/net/ip.h       |  2 +-
 include/net/sock.h     | 17 ++++++++++-------
 net/can/j1939/socket.c | 10 ++++++----
 net/core/skbuff.c      | 10 ++++++----
 net/core/sock.c        |  4 ++--
 net/ipv4/ip_output.c   |  2 +-
 net/ipv4/ip_sockglue.c |  2 +-
 net/ipv4/tcp.c         |  4 ++--
 net/ipv6/ip6_output.c  |  2 +-
 net/ipv6/ping.c        |  2 +-
 net/ipv6/raw.c         |  2 +-
 net/ipv6/udp.c         |  2 +-
 net/socket.c           | 13 +++++++------
 13 files changed, 40 insertions(+), 32 deletions(-)

diff --git a/include/net/ip.h b/include/net/ip.h
index 19adacd5ece031547e3ad70d524e89d8a6301a55..9276cea775cc2c26ccabcfea34822b19cfedb1a7 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -94,7 +94,7 @@ static inline void ipcm_init_sk(struct ipcm_cookie *ipcm,
 	ipcm_init(ipcm);
 
 	ipcm->sockc.mark = READ_ONCE(inet->sk.sk_mark);
-	ipcm->sockc.tsflags = inet->sk.sk_tsflags;
+	ipcm->sockc.tsflags = READ_ONCE(inet->sk.sk_tsflags);
 	ipcm->oif = READ_ONCE(inet->sk.sk_bound_dev_if);
 	ipcm->addr = inet->inet_saddr;
 	ipcm->protocol = inet->inet_num;
diff --git a/include/net/sock.h b/include/net/sock.h
index f04869ac1d92283bc3e4ecea2909babfc5096003..b770261fbdaf59d4d1c0b30adb2592c56442e9e3 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1906,7 +1906,9 @@ struct sockcm_cookie {
 static inline void sockcm_init(struct sockcm_cookie *sockc,
 			       const struct sock *sk)
 {
-	*sockc = (struct sockcm_cookie) { .tsflags = sk->sk_tsflags };
+	*sockc = (struct sockcm_cookie) {
+		.tsflags = READ_ONCE(sk->sk_tsflags)
+	};
 }
 
 int __sock_cmsg_send(struct sock *sk, struct cmsghdr *cmsg,
@@ -2701,9 +2703,9 @@ void __sock_recv_wifi_status(struct msghdr *msg, struct sock *sk,
 static inline void
 sock_recv_timestamp(struct msghdr *msg, struct sock *sk, struct sk_buff *skb)
 {
-	ktime_t kt = skb->tstamp;
 	struct skb_shared_hwtstamps *hwtstamps = skb_hwtstamps(skb);
-
+	u32 tsflags = READ_ONCE(sk->sk_tsflags);
+	ktime_t kt = skb->tstamp;
 	/*
 	 * generate control messages if
 	 * - receive time stamping in software requested
@@ -2711,10 +2713,10 @@ sock_recv_timestamp(struct msghdr *msg, struct sock *sk, struct sk_buff *skb)
 	 * - hardware time stamps available and wanted
 	 */
 	if (sock_flag(sk, SOCK_RCVTSTAMP) ||
-	    (sk->sk_tsflags & SOF_TIMESTAMPING_RX_SOFTWARE) ||
-	    (kt && sk->sk_tsflags & SOF_TIMESTAMPING_SOFTWARE) ||
+	    (tsflags & SOF_TIMESTAMPING_RX_SOFTWARE) ||
+	    (kt && tsflags & SOF_TIMESTAMPING_SOFTWARE) ||
 	    (hwtstamps->hwtstamp &&
-	     (sk->sk_tsflags & SOF_TIMESTAMPING_RAW_HARDWARE)))
+	     (tsflags & SOF_TIMESTAMPING_RAW_HARDWARE)))
 		__sock_recv_timestamp(msg, sk, skb);
 	else
 		sock_write_timestamp(sk, kt);
@@ -2736,7 +2738,8 @@ static inline void sock_recv_cmsgs(struct msghdr *msg, struct sock *sk,
 #define TSFLAGS_ANY	  (SOF_TIMESTAMPING_SOFTWARE			| \
 			   SOF_TIMESTAMPING_RAW_HARDWARE)
 
-	if (sk->sk_flags & FLAGS_RECV_CMSGS || sk->sk_tsflags & TSFLAGS_ANY)
+	if (sk->sk_flags & FLAGS_RECV_CMSGS ||
+	    READ_ONCE(sk->sk_tsflags) & TSFLAGS_ANY)
 		__sock_recv_cmsgs(msg, sk, skb);
 	else if (unlikely(sock_flag(sk, SOCK_TIMESTAMP)))
 		sock_write_timestamp(sk, skb->tstamp);
diff --git a/net/can/j1939/socket.c b/net/can/j1939/socket.c
index feaec4ad6d163d6c898f7e63671a3e264bb2cd1b..b28c976f52a0a16e13e23aee7c6fe4a7a8c844af 100644
--- a/net/can/j1939/socket.c
+++ b/net/can/j1939/socket.c
@@ -974,6 +974,7 @@ static void __j1939_sk_errqueue(struct j1939_session *session, struct sock *sk,
 	struct sock_exterr_skb *serr;
 	struct sk_buff *skb;
 	char *state = "UNK";
+	u32 tsflags;
 	int err;
 
 	jsk = j1939_sk(sk);
@@ -981,13 +982,14 @@ static void __j1939_sk_errqueue(struct j1939_session *session, struct sock *sk,
 	if (!(jsk->state & J1939_SOCK_ERRQUEUE))
 		return;
 
+	tsflags = READ_ONCE(sk->sk_tsflags);
 	switch (type) {
 	case J1939_ERRQUEUE_TX_ACK:
-		if (!(sk->sk_tsflags & SOF_TIMESTAMPING_TX_ACK))
+		if (!(tsflags & SOF_TIMESTAMPING_TX_ACK))
 			return;
 		break;
 	case J1939_ERRQUEUE_TX_SCHED:
-		if (!(sk->sk_tsflags & SOF_TIMESTAMPING_TX_SCHED))
+		if (!(tsflags & SOF_TIMESTAMPING_TX_SCHED))
 			return;
 		break;
 	case J1939_ERRQUEUE_TX_ABORT:
@@ -997,7 +999,7 @@ static void __j1939_sk_errqueue(struct j1939_session *session, struct sock *sk,
 	case J1939_ERRQUEUE_RX_DPO:
 		fallthrough;
 	case J1939_ERRQUEUE_RX_ABORT:
-		if (!(sk->sk_tsflags & SOF_TIMESTAMPING_RX_SOFTWARE))
+		if (!(tsflags & SOF_TIMESTAMPING_RX_SOFTWARE))
 			return;
 		break;
 	default:
@@ -1054,7 +1056,7 @@ static void __j1939_sk_errqueue(struct j1939_session *session, struct sock *sk,
 	}
 
 	serr->opt_stats = true;
-	if (sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID)
+	if (tsflags & SOF_TIMESTAMPING_OPT_ID)
 		serr->ee.ee_data = session->tskey;
 
 	netdev_dbg(session->priv->ndev, "%s: 0x%p tskey: %i, state: %s\n",
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 45707059082f2d706a9820f4484d0d4636aaa930..24f26e816184ebdc3317b07f7b3854b9908642f2 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -5207,7 +5207,7 @@ static void __skb_complete_tx_timestamp(struct sk_buff *skb,
 	serr->ee.ee_info = tstype;
 	serr->opt_stats = opt_stats;
 	serr->header.h4.iif = skb->dev ? skb->dev->ifindex : 0;
-	if (sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID) {
+	if (READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_OPT_ID) {
 		serr->ee.ee_data = skb_shinfo(skb)->tskey;
 		if (sk_is_tcp(sk))
 			serr->ee.ee_data -= atomic_read(&sk->sk_tskey);
@@ -5263,21 +5263,23 @@ void __skb_tstamp_tx(struct sk_buff *orig_skb,
 {
 	struct sk_buff *skb;
 	bool tsonly, opt_stats = false;
+	u32 tsflags;
 
 	if (!sk)
 		return;
 
-	if (!hwtstamps && !(sk->sk_tsflags & SOF_TIMESTAMPING_OPT_TX_SWHW) &&
+	tsflags = READ_ONCE(sk->sk_tsflags);
+	if (!hwtstamps && !(tsflags & SOF_TIMESTAMPING_OPT_TX_SWHW) &&
 	    skb_shinfo(orig_skb)->tx_flags & SKBTX_IN_PROGRESS)
 		return;
 
-	tsonly = sk->sk_tsflags & SOF_TIMESTAMPING_OPT_TSONLY;
+	tsonly = tsflags & SOF_TIMESTAMPING_OPT_TSONLY;
 	if (!skb_may_tx_timestamp(sk, tsonly))
 		return;
 
 	if (tsonly) {
 #ifdef CONFIG_INET
-		if ((sk->sk_tsflags & SOF_TIMESTAMPING_OPT_STATS) &&
+		if ((tsflags & SOF_TIMESTAMPING_OPT_STATS) &&
 		    sk_is_tcp(sk)) {
 			skb = tcp_get_timestamping_opt_stats(sk, orig_skb,
 							     ack_skb);
diff --git a/net/core/sock.c b/net/core/sock.c
index 40e1bda4bde0cd0ddbb0315deb4df45d60f65081..d05a290300b6c8c765d6733520c50978f35f2dd0 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -937,7 +937,7 @@ int sock_set_timestamping(struct sock *sk, int optname,
 			return ret;
 	}
 
-	sk->sk_tsflags = val;
+	WRITE_ONCE(sk->sk_tsflags, val);
 	sock_valbool_flag(sk, SOCK_TSTAMP_NEW, optname == SO_TIMESTAMPING_NEW);
 
 	if (val & SOF_TIMESTAMPING_RX_SOFTWARE)
@@ -1719,7 +1719,7 @@ int sk_getsockopt(struct sock *sk, int level, int optname,
 
 	case SO_TIMESTAMPING_OLD:
 		lv = sizeof(v.timestamping);
-		v.timestamping.flags = sk->sk_tsflags;
+		v.timestamping.flags = READ_ONCE(sk->sk_tsflags);
 		v.timestamping.bind_phc = sk->sk_bind_phc;
 		break;
 
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index b2e0ad3120280d360aebb1c116af85506a237e05..4ab877cf6d35f229761986d5c6a17eb2a3ad4043 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -981,7 +981,7 @@ static int __ip_append_data(struct sock *sk,
 	paged = !!cork->gso_size;
 
 	if (cork->tx_flags & SKBTX_ANY_TSTAMP &&
-	    sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID)
+	    READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_OPT_ID)
 		tskey = atomic_inc_return(&sk->sk_tskey) - 1;
 
 	hh_len = LL_RESERVED_SPACE(rt->dst.dev);
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index d1c73660b844949b57960630e0467112da4f0abd..cce9cb25f3b31cd57fa883ae0dedb6829d8da2fa 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -511,7 +511,7 @@ static bool ipv4_datagram_support_cmsg(const struct sock *sk,
 	 * or without payload (SOF_TIMESTAMPING_OPT_TSONLY).
 	 */
 	info = PKTINFO_SKB_CB(skb);
-	if (!(sk->sk_tsflags & SOF_TIMESTAMPING_OPT_CMSG) ||
+	if (!(READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_OPT_CMSG) ||
 	    !info->ipi_ifindex)
 		return false;
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index cee1e548660cb93835102836fe8103666c4c4697..cc4b250262c19ae5f18e806dca63ad7b05e03ab2 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2259,14 +2259,14 @@ void tcp_recv_timestamp(struct msghdr *msg, const struct sock *sk,
 			}
 		}
 
-		if (sk->sk_tsflags & SOF_TIMESTAMPING_SOFTWARE)
+		if (READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_SOFTWARE)
 			has_timestamping = true;
 		else
 			tss->ts[0] = (struct timespec64) {0};
 	}
 
 	if (tss->ts[2].tv_sec || tss->ts[2].tv_nsec) {
-		if (sk->sk_tsflags & SOF_TIMESTAMPING_RAW_HARDWARE)
+		if (READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_RAW_HARDWARE)
 			has_timestamping = true;
 		else
 			tss->ts[2] = (struct timespec64) {0};
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 4ab50169a5a9dc4fabee86398906b3a4cefda049..54fc4c711f2c545f2ca625d6b0e09f2bb8e6d513 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1501,7 +1501,7 @@ static int __ip6_append_data(struct sock *sk,
 	orig_mtu = mtu;
 
 	if (cork->tx_flags & SKBTX_ANY_TSTAMP &&
-	    sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID)
+	    READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_OPT_ID)
 		tskey = atomic_inc_return(&sk->sk_tskey) - 1;
 
 	hh_len = LL_RESERVED_SPACE(rt->dst.dev);
diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c
index 1b27728349725e212ab5d988dd96cf87464c9531..5831aaa53d75eae7b764d54ab52da65db4030d73 100644
--- a/net/ipv6/ping.c
+++ b/net/ipv6/ping.c
@@ -119,7 +119,7 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		return -EINVAL;
 
 	ipcm6_init_sk(&ipc6, np);
-	ipc6.sockc.tsflags = sk->sk_tsflags;
+	ipc6.sockc.tsflags = READ_ONCE(sk->sk_tsflags);
 	ipc6.sockc.mark = READ_ONCE(sk->sk_mark);
 
 	fl6.flowi6_oif = oif;
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 0eae7661a85c4487a64384c6054a3fb827387ce7..42fcec3ecf5e171a5ebe724b8c971d90885abe41 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -772,7 +772,7 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	fl6.flowi6_uid = sk->sk_uid;
 
 	ipcm6_init(&ipc6);
-	ipc6.sockc.tsflags = sk->sk_tsflags;
+	ipc6.sockc.tsflags = READ_ONCE(sk->sk_tsflags);
 	ipc6.sockc.mark = fl6.flowi6_mark;
 
 	if (sin6) {
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index ebc6ae47cfeadc699e3f5a1f46be85803ff37fdd..86b5d509a4688cacb2f40667c9ddc10f81ade2fe 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1339,7 +1339,7 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	ipcm6_init(&ipc6);
 	ipc6.gso_size = READ_ONCE(up->gso_size);
-	ipc6.sockc.tsflags = sk->sk_tsflags;
+	ipc6.sockc.tsflags = READ_ONCE(sk->sk_tsflags);
 	ipc6.sockc.mark = READ_ONCE(sk->sk_mark);
 
 	/* destination address check */
diff --git a/net/socket.c b/net/socket.c
index 848116d06b51168585b9d8ed26c9f20730cd46c9..98ffffab949e850fe07904f1cf20ff2ce4a9b5d1 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -825,7 +825,7 @@ static bool skb_is_swtx_tstamp(const struct sk_buff *skb, int false_tstamp)
 
 static ktime_t get_timestamp(struct sock *sk, struct sk_buff *skb, int *if_index)
 {
-	bool cycles = sk->sk_tsflags & SOF_TIMESTAMPING_BIND_PHC;
+	bool cycles = READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_BIND_PHC;
 	struct skb_shared_hwtstamps *shhwtstamps = skb_hwtstamps(skb);
 	struct net_device *orig_dev;
 	ktime_t hwtstamp;
@@ -877,12 +877,12 @@ void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk,
 	int need_software_tstamp = sock_flag(sk, SOCK_RCVTSTAMP);
 	int new_tstamp = sock_flag(sk, SOCK_TSTAMP_NEW);
 	struct scm_timestamping_internal tss;
-
 	int empty = 1, false_tstamp = 0;
 	struct skb_shared_hwtstamps *shhwtstamps =
 		skb_hwtstamps(skb);
 	int if_index;
 	ktime_t hwtstamp;
+	u32 tsflags;
 
 	/* Race occurred between timestamp enabling and packet
 	   receiving.  Fill in the current time for now. */
@@ -924,11 +924,12 @@ void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk,
 	}
 
 	memset(&tss, 0, sizeof(tss));
-	if ((sk->sk_tsflags & SOF_TIMESTAMPING_SOFTWARE) &&
+	tsflags = READ_ONCE(sk->sk_tsflags);
+	if ((tsflags & SOF_TIMESTAMPING_SOFTWARE) &&
 	    ktime_to_timespec64_cond(skb->tstamp, tss.ts + 0))
 		empty = 0;
 	if (shhwtstamps &&
-	    (sk->sk_tsflags & SOF_TIMESTAMPING_RAW_HARDWARE) &&
+	    (tsflags & SOF_TIMESTAMPING_RAW_HARDWARE) &&
 	    !skb_is_swtx_tstamp(skb, false_tstamp)) {
 		if_index = 0;
 		if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP_NETDEV)
@@ -936,14 +937,14 @@ void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk,
 		else
 			hwtstamp = shhwtstamps->hwtstamp;
 
-		if (sk->sk_tsflags & SOF_TIMESTAMPING_BIND_PHC)
+		if (tsflags & SOF_TIMESTAMPING_BIND_PHC)
 			hwtstamp = ptp_convert_timestamp(&hwtstamp,
 							 sk->sk_bind_phc);
 
 		if (ktime_to_timespec64_cond(hwtstamp, tss.ts + 2)) {
 			empty = 0;
 
-			if ((sk->sk_tsflags & SOF_TIMESTAMPING_OPT_PKTINFO) &&
+			if ((tsflags & SOF_TIMESTAMPING_OPT_PKTINFO) &&
 			    !skb_is_err_queue(skb))
 				put_ts_pktinfo(msg, skb, if_index);
 		}
-- 
2.42.0.rc2.253.gd59a3bf2b4-goog

[PATCH net 5/5] net: annotate data-races around sk->sk_bind_phc

[Eric Dumazet]

sk->sk_bind_phc is read locklessly. Add corresponding annotations.

Fixes: d463126e23f1 ("net: sock: extend SO_TIMESTAMPING for PHC binding")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Yangbo Lu <yangbo.lu@nxp.com>
---
 net/core/sock.c | 4 ++--
 net/socket.c    | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/core/sock.c b/net/core/sock.c
index d05a290300b6c8c765d6733520c50978f35f2dd0..d3c7b53368d2245c8c2299a5f8ad79fad4de4d16 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -894,7 +894,7 @@ static int sock_timestamping_bind_phc(struct sock *sk, int phc_index)
 	if (!match)
 		return -EINVAL;
 
-	sk->sk_bind_phc = phc_index;
+	WRITE_ONCE(sk->sk_bind_phc, phc_index);
 
 	return 0;
 }
@@ -1720,7 +1720,7 @@ int sk_getsockopt(struct sock *sk, int level, int optname,
 	case SO_TIMESTAMPING_OLD:
 		lv = sizeof(v.timestamping);
 		v.timestamping.flags = READ_ONCE(sk->sk_tsflags);
-		v.timestamping.bind_phc = sk->sk_bind_phc;
+		v.timestamping.bind_phc = READ_ONCE(sk->sk_bind_phc);
 		break;
 
 	case SO_RCVTIMEO_OLD:
diff --git a/net/socket.c b/net/socket.c
index 98ffffab949e850fe07904f1cf20ff2ce4a9b5d1..928b05811cfd3333ffc101f850486bc1e8e3f611 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -939,7 +939,7 @@ void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk,
 
 		if (tsflags & SOF_TIMESTAMPING_BIND_PHC)
 			hwtstamp = ptp_convert_timestamp(&hwtstamp,
-							 sk->sk_bind_phc);
+							 READ_ONCE(sk->sk_bind_phc));
 
 		if (ktime_to_timespec64_cond(hwtstamp, tss.ts + 2)) {
 			empty = 0;
-- 
2.42.0.rc2.253.gd59a3bf2b4-goog

Re: [PATCH net 0/5] net: another round of data-race annotations

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net.git (main)
by David S. Miller <davem@davemloft.net>:

On Thu, 31 Aug 2023 13:52:07 +0000 you wrote:
> Series inspired by some syzbot reports, taking care
> of 4 socket fields that can be read locklessly.
> 
> Eric Dumazet (5):
>   net: use sk_forward_alloc_get() in sk_get_meminfo()
>   net: annotate data-races around sk->sk_forward_alloc
>   mptcp: annotate data-races around msk->rmem_fwd_alloc
>   net: annotate data-races around sk->sk_tsflags
>   net: annotate data-races around sk->sk_bind_phc
> 
> [...]

Here is the summary with links:
  - [net,1/5] net: use sk_forward_alloc_get() in sk_get_meminfo()
    https://git.kernel.org/netdev/net/c/66d58f046c9d
  - [net,2/5] net: annotate data-races around sk->sk_forward_alloc
    https://git.kernel.org/netdev/net/c/5e6300e7b3a4
  - [net,3/5] mptcp: annotate data-races around msk->rmem_fwd_alloc
    https://git.kernel.org/netdev/net/c/9531e4a83feb
  - [net,4/5] net: annotate data-races around sk->sk_tsflags
    https://git.kernel.org/netdev/net/c/e3390b30a5df
  - [net,5/5] net: annotate data-races around sk->sk_bind_phc
    https://git.kernel.org/netdev/net/c/251cd405a9e6

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html