Merge branch 'bpf-fix-fionread-and-copied_seq-issues'

Jiayuan Chen says:

====================
bpf: Fix FIONREAD and copied_seq issues

syzkaller reported a bug [1] where a socket using sockmap, after being
unloaded, exposed incorrect copied_seq calculation. The selftest I
provided can be used to reproduce the issue reported by syzkaller.

TCP recvmsg seq # bug 2: copied E92C873, seq E68D125, rcvnxt E7CEB7C, fl 40
WARNING: CPU: 1 PID: 5997 at net/ipv4/tcp.c:2724 tcp_recvmsg_locked+0xb2f/0x2910 net/ipv4/tcp.c:2724
Call Trace:
 <TASK>
 receive_fallback_to_copy net/ipv4/tcp.c:1968 [inline]
 tcp_zerocopy_receive+0x131a/0x2120 net/ipv4/tcp.c:2200
 do_tcp_getsockopt+0xe28/0x26c0 net/ipv4/tcp.c:4713
 tcp_getsockopt+0xdf/0x100 net/ipv4/tcp.c:4812
 do_sock_getsockopt+0x34d/0x440 net/socket.c:2421
 __sys_getsockopt+0x12f/0x260 net/socket.c:2450
 __do_sys_getsockopt net/socket.c:2457 [inline]
 __se_sys_getsockopt net/socket.c:2454 [inline]
 __x64_sys_getsockopt+0xbd/0x160 net/socket.c:2454
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0xcd/0xfa0 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

A sockmap socket maintains its own receive queue (ingress_msg) which may
contain data from either its own protocol stack or forwarded from other
sockets.

                                                     FD1:read()
                                                     --  FD1->copied_seq++
                                                         |  [read data]
                                                         |
                                [enqueue data]           v
                  [sockmap]     -> ingress to self ->  ingress_msg queue
FD1 native stack  ------>                                 ^
-- FD1->rcv_nxt++               -> redirect to other      | [enqueue data]
                                       |                  |
                                       |             ingress to FD1
                                       v                  ^
                                      ...                 |  [sockmap]
                                                     FD2 native stack

The issue occurs when reading from ingress_msg: we update tp->copied_seq
by default, but if the data comes from other sockets (not the socket's
own protocol stack), tcp->rcv_nxt remains unchanged. Later, when
converting back to a native socket, reads may fail as copied_seq could
be significantly larger than rcv_nxt.

Additionally, FIONREAD calculation based on copied_seq and rcv_nxt is
insufficient for sockmap sockets, requiring separate field tracking.

[1] https://syzkaller.appspot.com/bug?extid=06dbd397158ec0ea4983
---
v7 -> v9: Address Jakub Sitnicki's feedback:
          - Remove sk_receive_queue check in tcp_bpf_ioctl, only report
            ingress_msg data length for FIONREAD
          - Minor nits fixes
          - Add Reviewed-by tag from John Fastabend
          - Fix ci error
          https://lore.kernel.org/bpf/20260113025121.197535-1-jiayuan.chen@linux.dev/

v5 -> v7: Some modifications suggested by Jakub Sitnicki, and added Reviewed-by tag.
https://lore.kernel.org/bpf/20260106051458.279151-1-jiayuan.chen@linux.dev/

v1 -> v5: Use skmsg.sk instead of extending BPF_F_XXX macro and fix CI
          failure reported by CI
v1: https://lore.kernel.org/bpf/20251117110736.293040-1-jiayuan.chen@linux.dev/
====================

Link: https://patch.msgid.link/20260124113314.113584-1-jiayuan.chen@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

1 files changed, 68 insertions, 2 deletions

diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h
index 49847888c287..829b281d6c9c 100644
--- a/include/linux/skmsg.h
+++ b/include/linux/skmsg.h
@@ -97,6 +97,8 @@ struct sk_psock {
 	struct sk_buff_head		ingress_skb;
 	struct list_head		ingress_msg;
 	spinlock_t			ingress_lock;
+	/** @msg_tot_len: Total bytes queued in ingress_msg list. */
+	u32				msg_tot_len;
 	unsigned long			state;
 	struct list_head		link;
 	spinlock_t			link_lock;
@@ -141,6 +143,8 @@ int sk_msg_memcopy_from_iter(struct sock *sk, struct iov_iter *from,
 			     struct sk_msg *msg, u32 bytes);
 int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
 		   int len, int flags);
+int __sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
+		     int len, int flags, int *copied_from_self);
 bool sk_msg_is_readable(struct sock *sk);
 
 static inline void sk_msg_check_to_free(struct sk_msg *msg, u32 i, u32 bytes)
@@ -319,6 +323,27 @@ static inline void sock_drop(struct sock *sk, struct sk_buff *skb)
 	kfree_skb(skb);
 }
 
+static inline u32 sk_psock_get_msg_len_nolock(struct sk_psock *psock)
+{
+	/* Used by ioctl to read msg_tot_len only; lock-free for performance */
+	return READ_ONCE(psock->msg_tot_len);
+}
+
+static inline void sk_psock_msg_len_add_locked(struct sk_psock *psock, int diff)
+{
+	/* Use WRITE_ONCE to ensure correct read in sk_psock_get_msg_len_nolock().
+	 * ingress_lock should be held to prevent concurrent updates to msg_tot_len
+	 */
+	WRITE_ONCE(psock->msg_tot_len, psock->msg_tot_len + diff);
+}
+
+static inline void sk_psock_msg_len_add(struct sk_psock *psock, int diff)
+{
+	spin_lock_bh(&psock->ingress_lock);
+	sk_psock_msg_len_add_locked(psock, diff);
+	spin_unlock_bh(&psock->ingress_lock);
+}
+
 static inline bool sk_psock_queue_msg(struct sk_psock *psock,
 				      struct sk_msg *msg)
 {
@@ -327,6 +352,7 @@ static inline bool sk_psock_queue_msg(struct sk_psock *psock,
 	spin_lock_bh(&psock->ingress_lock);
 	if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) {
 		list_add_tail(&msg->list, &psock->ingress_msg);
+		sk_psock_msg_len_add_locked(psock, msg->sg.size);
 		ret = true;
 	} else {
 		sk_msg_free(psock->sk, msg);
@@ -343,18 +369,25 @@ static inline struct sk_msg *sk_psock_dequeue_msg(struct sk_psock *psock)
 
 	spin_lock_bh(&psock->ingress_lock);
 	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
-	if (msg)
+	if (msg) {
 		list_del(&msg->list);
+		sk_psock_msg_len_add_locked(psock, -msg->sg.size);
+	}
 	spin_unlock_bh(&psock->ingress_lock);
 	return msg;
 }
 
+static inline struct sk_msg *sk_psock_peek_msg_locked(struct sk_psock *psock)
+{
+	return list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
+}
+
 static inline struct sk_msg *sk_psock_peek_msg(struct sk_psock *psock)
 {
 	struct sk_msg *msg;
 
 	spin_lock_bh(&psock->ingress_lock);
-	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
+	msg = sk_psock_peek_msg_locked(psock);
 	spin_unlock_bh(&psock->ingress_lock);
 	return msg;
 }
@@ -521,6 +554,39 @@ static inline bool sk_psock_strp_enabled(struct sk_psock *psock)
 	return !!psock->saved_data_ready;
 }
 
+/* for tcp only, sk is locked */
+static inline ssize_t sk_psock_msg_inq(struct sock *sk)
+{
+	struct sk_psock *psock;
+	ssize_t inq = 0;
+
+	psock = sk_psock_get(sk);
+	if (likely(psock)) {
+		inq = sk_psock_get_msg_len_nolock(psock);
+		sk_psock_put(sk, psock);
+	}
+	return inq;
+}
+
+/* for udp only, sk is not locked */
+static inline ssize_t sk_msg_first_len(struct sock *sk)
+{
+	struct sk_psock *psock;
+	struct sk_msg *msg;
+	ssize_t inq = 0;
+
+	psock = sk_psock_get(sk);
+	if (likely(psock)) {
+		spin_lock_bh(&psock->ingress_lock);
+		msg = sk_psock_peek_msg_locked(psock);
+		if (msg)
+			inq = msg->sg.size;
+		spin_unlock_bh(&psock->ingress_lock);
+		sk_psock_put(sk, psock);
+	}
+	return inq;
+}
+
 #if IS_ENABLED(CONFIG_NET_SOCK_MSG)
 
 #define BPF_F_STRPARSER	(1UL << 1)