| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <87zf8h6bpd.fsf@cloudflare.com>
Date: Wed, 19 Nov 2025 20:53:34 +0100
From: Jakub Sitnicki <jakub@...udflare.com>
To: Jiayuan Chen <jiayuan.chen@...ux.dev>
Cc: bpf@...r.kernel.org, John Fastabend <john.fastabend@...il.com>, "David
S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
Simon Horman <horms@...nel.org>, Neal Cardwell <ncardwell@...gle.com>,
Kuniyuki Iwashima <kuniyu@...gle.com>, David Ahern <dsahern@...nel.org>,
Alexei Starovoitov <ast@...nel.org>, Daniel Borkmann
<daniel@...earbox.net>, Andrii Nakryiko <andrii@...nel.org>, Martin
KaFai Lau <martin.lau@...ux.dev>, Eduard Zingerman <eddyz87@...il.com>,
Song Liu <song@...nel.org>, Yonghong Song <yonghong.song@...ux.dev>, KP
Singh <kpsingh@...nel.org>, Stanislav Fomichev <sdf@...ichev.me>, Hao
Luo <haoluo@...gle.com>, Jiri Olsa <jolsa@...nel.org>, Shuah Khan
<shuah@...nel.org>, Michal Luczaj <mhal@...x.co>, Stefano Garzarella
<sgarzare@...hat.com>, Cong Wang <cong.wang@...edance.com>,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-kselftest@...r.kernel.org
Subject: Re: [PATCH bpf-next v1 1/3] bpf, sockmap: Fix incorrect copied_seq
calculation
On Mon, Nov 17, 2025 at 07:07 PM +08, Jiayuan Chen wrote:
> A socket using sockmap has its own independent receive queue: ingress_msg.
> This queue may contain data from its own protocol stack or from other
> sockets.
>
> The issue is that when reading from ingress_msg, we update tp->copied_seq
> by default. However, if the data is not from its own protocol stack,
> tcp->rcv_nxt is not increased. Later, if we convert this socket to a
> native socket, reading from this socket may fail because copied_seq might
> be significantly larger than rcv_nxt.
>
> This fix also addresses the syzkaller-reported bug referenced in the
> Closes tag.
>
> This patch marks the skmsg objects in ingress_msg. When reading, we update
> copied_seq only if the data is from its own protocol stack.
>
> 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
>
> Closes: https://syzkaller.appspot.com/bug?extid=06dbd397158ec0ea4983
> Fixes: 04919bed948dc ("tcp: Introduce tcp_read_skb()")
> Signed-off-by: Jiayuan Chen <jiayuan.chen@...ux.dev>
> ---
> include/linux/skmsg.h | 25 ++++++++++++++++++++++++-
> net/core/skmsg.c | 19 ++++++++++++++++---
> net/ipv4/tcp_bpf.c | 5 +++--
> 3 files changed, 43 insertions(+), 6 deletions(-)
>
> diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h
> index 49847888c287..b7826cb2a388 100644
> --- a/include/linux/skmsg.h
> +++ b/include/linux/skmsg.h
> @@ -23,6 +23,17 @@ enum __sk_action {
> __SK_NONE,
> };
>
> +/* The BPF program sets BPF_F_INGRESS on sk_msg to indicate data needs to be
> + * redirected to the ingress queue of a specified socket. Since BPF_F_INGRESS is
> + * defined in UAPI so that we can't extend this enum for our internal flags. We
> + * define some internal flags here while inheriting BPF_F_INGRESS.
> + */
> +enum {
> + SK_MSG_F_INGRESS = BPF_F_INGRESS, /* (1ULL << 0) */
> + /* internal flag */
> + SK_MSG_F_INGRESS_SELF = (1ULL << 1)
> +};
> +
I'm wondering if we need additional state to track this.
Can we track sk_msg's construted from skb's that were not redirected by
setting `sk_msg.sk = sk` to indicate that the source socket is us in
sk_psock_skb_ingress_self()?
If not, then I'd just offset the internal flags like we do in
net/core/filter.c, BPF_F_REDIRECT_INTERNAL.
> struct sk_msg_sg {
> u32 start;
> u32 curr;
> @@ -141,8 +152,20 @@ 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 *from_self_copied);
> bool sk_msg_is_readable(struct sock *sk);
>
> +static inline bool sk_msg_is_to_self(struct sk_msg *msg)
> +{
> + return msg->flags & SK_MSG_F_INGRESS_SELF;
> +}
> +
> +static inline void sk_msg_set_to_self(struct sk_msg *msg)
> +{
> + msg->flags |= SK_MSG_F_INGRESS_SELF;
> +}
> +
> static inline void sk_msg_check_to_free(struct sk_msg *msg, u32 i, u32 bytes)
> {
> WARN_ON(i == msg->sg.end && bytes);
> @@ -235,7 +258,7 @@ static inline struct page *sk_msg_page(struct sk_msg *msg, int which)
>
> static inline bool sk_msg_to_ingress(const struct sk_msg *msg)
> {
> - return msg->flags & BPF_F_INGRESS;
> + return msg->flags & SK_MSG_F_INGRESS;
> }
>
> static inline void sk_msg_compute_data_pointers(struct sk_msg *msg)
> diff --git a/net/core/skmsg.c b/net/core/skmsg.c
> index 2ac7731e1e0a..25d88c2082e9 100644
> --- a/net/core/skmsg.c
> +++ b/net/core/skmsg.c
> @@ -409,14 +409,14 @@ int sk_msg_memcopy_from_iter(struct sock *sk, struct iov_iter *from,
> }
> EXPORT_SYMBOL_GPL(sk_msg_memcopy_from_iter);
>
> -/* Receive sk_msg from psock->ingress_msg to @msg. */
> -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 *from_self_copied)
> {
> struct iov_iter *iter = &msg->msg_iter;
> int peek = flags & MSG_PEEK;
> struct sk_msg *msg_rx;
> int i, copied = 0;
> + bool to_self;
>
> msg_rx = sk_psock_peek_msg(psock);
> while (copied != len) {
> @@ -425,6 +425,7 @@ int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
> if (unlikely(!msg_rx))
> break;
>
> + to_self = sk_msg_is_to_self(msg_rx);
> i = msg_rx->sg.start;
> do {
> struct page *page;
> @@ -443,6 +444,9 @@ int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
> }
>
> copied += copy;
> + if (to_self && from_self_copied)
> + *from_self_copied += copy;
> +
> if (likely(!peek)) {
> sge->offset += copy;
> sge->length -= copy;
> @@ -487,6 +491,14 @@ int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
> out:
> return copied;
> }
> +EXPORT_SYMBOL_GPL(__sk_msg_recvmsg);
> +
> +/* Receive sk_msg from psock->ingress_msg to @msg. */
> +int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
> + int len, int flags)
> +{
> + return __sk_msg_recvmsg(sk, psock, msg, len, flags, NULL);
> +}
> EXPORT_SYMBOL_GPL(sk_msg_recvmsg);
>
> bool sk_msg_is_readable(struct sock *sk)
> @@ -616,6 +628,7 @@ static int sk_psock_skb_ingress_self(struct sk_psock *psock, struct sk_buff *skb
> if (unlikely(!msg))
> return -EAGAIN;
> skb_set_owner_r(skb, sk);
> + sk_msg_set_to_self(msg);
> err = sk_psock_skb_ingress_enqueue(skb, off, len, psock, sk, msg, take_ref);
> if (err < 0)
> kfree(msg);
> diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c
> index a268e1595b22..6332fc36ffe6 100644
> --- a/net/ipv4/tcp_bpf.c
> +++ b/net/ipv4/tcp_bpf.c
> @@ -226,6 +226,7 @@ static int tcp_bpf_recvmsg_parser(struct sock *sk,
> int peek = flags & MSG_PEEK;
> struct sk_psock *psock;
> struct tcp_sock *tcp;
> + int from_self_copied = 0;
> int copied = 0;
> u32 seq;
>
> @@ -262,7 +263,7 @@ static int tcp_bpf_recvmsg_parser(struct sock *sk,
> }
>
> msg_bytes_ready:
> - copied = sk_msg_recvmsg(sk, psock, msg, len, flags);
> + copied = __sk_msg_recvmsg(sk, psock, msg, len, flags, &from_self_copied);
> /* The typical case for EFAULT is the socket was gracefully
> * shutdown with a FIN pkt. So check here the other case is
> * some error on copy_page_to_iter which would be unexpected.
> @@ -277,7 +278,7 @@ static int tcp_bpf_recvmsg_parser(struct sock *sk,
> goto out;
> }
> }
> - seq += copied;
> + seq += from_self_copied;
> if (!copied) {
> long timeo;
> int data;
Powered by blists - more mailing lists