summary refs log tree commit diff
path: root/net/tls
diff options
context:
space:
mode:
authorJakub Kicinski <kuba@kernel.org>2022-07-14 22:22:25 -0700
committerDavid S. Miller <davem@davemloft.net>2022-07-18 11:24:10 +0100
commit4cbc325ed6b4dce4910be06d9d6940a8b919c59b (patch)
treee425c352dc8bf8f9d93bcc5b59d674d85ffbfb1d /net/tls
parent3898f52cd4f92a007d71a52edf70aafc7c8ae193 (diff)
downloadlinux-4cbc325ed6b4dce4910be06d9d6940a8b919c59b.tar.gz
tls: rx: allow only one reader at a time
recvmsg() in TLS gets data from the skb list (rx_list) or fresh
skbs we read from TCP via strparser. The former holds skbs which were
already decrypted for peek or decrypted and partially consumed.

tls_wait_data() only notices appearance of fresh skbs coming out
of TCP (or psock). It is possible, if there is a concurrent call
to peek() and recv() that the peek() will move the data from input
to rx_list without recv() noticing. recv() will then read data out
of order or never wake up.

This is not a practical use case/concern, but it makes the self
tests less reliable. This patch solves the problem by allowing
only one reader in.

Because having multiple processes calling read()/peek() is not
normal avoid adding a lock and try to fast-path the single reader
case.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/tls')
-rw-r--r--net/tls/tls_sw.c61
1 files changed, 54 insertions, 7 deletions
diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
index 68d79ee48a56..761a63751616 100644
--- a/net/tls/tls_sw.c
+++ b/net/tls/tls_sw.c
@@ -1753,6 +1753,51 @@ tls_read_flush_backlog(struct sock *sk, struct tls_prot_info *prot,
 	sk_flush_backlog(sk);
 }
 
+static long tls_rx_reader_lock(struct sock *sk, struct tls_sw_context_rx *ctx,
+			       bool nonblock)
+{
+	long timeo;
+
+	lock_sock(sk);
+
+	timeo = sock_rcvtimeo(sk, nonblock);
+
+	while (unlikely(ctx->reader_present)) {
+		DEFINE_WAIT_FUNC(wait, woken_wake_function);
+
+		ctx->reader_contended = 1;
+
+		add_wait_queue(&ctx->wq, &wait);
+		sk_wait_event(sk, &timeo,
+			      !READ_ONCE(ctx->reader_present), &wait);
+		remove_wait_queue(&ctx->wq, &wait);
+
+		if (!timeo)
+			return -EAGAIN;
+		if (signal_pending(current))
+			return sock_intr_errno(timeo);
+	}
+
+	WRITE_ONCE(ctx->reader_present, 1);
+
+	return timeo;
+}
+
+static void tls_rx_reader_unlock(struct sock *sk, struct tls_sw_context_rx *ctx)
+{
+	if (unlikely(ctx->reader_contended)) {
+		if (wq_has_sleeper(&ctx->wq))
+			wake_up(&ctx->wq);
+		else
+			ctx->reader_contended = 0;
+
+		WARN_ON_ONCE(!ctx->reader_present);
+	}
+
+	WRITE_ONCE(ctx->reader_present, 0);
+	release_sock(sk);
+}
+
 int tls_sw_recvmsg(struct sock *sk,
 		   struct msghdr *msg,
 		   size_t len,
@@ -1782,7 +1827,9 @@ int tls_sw_recvmsg(struct sock *sk,
 		return sock_recv_errqueue(sk, msg, len, SOL_IP, IP_RECVERR);
 
 	psock = sk_psock_get(sk);
-	lock_sock(sk);
+	timeo = tls_rx_reader_lock(sk, ctx, flags & MSG_DONTWAIT);
+	if (timeo < 0)
+		return timeo;
 	bpf_strp_enabled = sk_psock_strp_enabled(psock);
 
 	/* If crypto failed the connection is broken */
@@ -1801,7 +1848,6 @@ int tls_sw_recvmsg(struct sock *sk,
 
 	target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);
 	len = len - copied;
-	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
 
 	zc_capable = !bpf_strp_enabled && !is_kvec && !is_peek &&
 		ctx->zc_capable;
@@ -1956,7 +2002,7 @@ recv_end:
 	copied += decrypted;
 
 end:
-	release_sock(sk);
+	tls_rx_reader_unlock(sk, ctx);
 	if (psock)
 		sk_psock_put(sk, psock);
 	return copied ? : err;
@@ -1978,9 +2024,9 @@ ssize_t tls_sw_splice_read(struct socket *sock,  loff_t *ppos,
 	long timeo;
 	int chunk;
 
-	lock_sock(sk);
-
-	timeo = sock_rcvtimeo(sk, flags & SPLICE_F_NONBLOCK);
+	timeo = tls_rx_reader_lock(sk, ctx, flags & SPLICE_F_NONBLOCK);
+	if (timeo < 0)
+		return timeo;
 
 	from_queue = !skb_queue_empty(&ctx->rx_list);
 	if (from_queue) {
@@ -2029,7 +2075,7 @@ ssize_t tls_sw_splice_read(struct socket *sock,  loff_t *ppos,
 	}
 
 splice_read_end:
-	release_sock(sk);
+	tls_rx_reader_unlock(sk, ctx);
 	return copied ? : err;
 }
 
@@ -2371,6 +2417,7 @@ int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx, int tx)
 	} else {
 		crypto_init_wait(&sw_ctx_rx->async_wait);
 		spin_lock_init(&sw_ctx_rx->decrypt_compl_lock);
+		init_waitqueue_head(&sw_ctx_rx->wq);
 		crypto_info = &ctx->crypto_recv.info;
 		cctx = &ctx->rx;
 		skb_queue_head_init(&sw_ctx_rx->rx_list);