From c3f0c02997c7f8489fec259e28e0e04e9811edac Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Mon, 18 Aug 2025 08:40:26 -0700 Subject: net: Add skb_dstref_steal and skb_dstref_restore Going forward skb_dst_set will assert that skb dst_entry is empty during skb_dst_set to prevent potential leaks. There are few places that still manually manage dst_entry not using the helpers. Convert them to the following new helpers: - skb_dstref_steal that resets dst_entry and returns previous dst_entry value - skb_dstref_restore that restores dst_entry previously reset via skb_dstref_steal Signed-off-by: Stanislav Fomichev Link: https://patch.msgid.link/20250818154032.3173645-2-sdf@fomichev.me Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) (limited to 'include/linux/skbuff.h') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 14b923ddb6df..7538ca507ee9 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1159,6 +1159,38 @@ static inline struct dst_entry *skb_dst(const struct sk_buff *skb) return (struct dst_entry *)(skb->_skb_refdst & SKB_DST_PTRMASK); } +/** + * skb_dstref_steal() - return current dst_entry value and clear it + * @skb: buffer + * + * Resets skb dst_entry without adjusting its reference count. Useful in + * cases where dst_entry needs to be temporarily reset and restored. + * Note that the returned value cannot be used directly because it + * might contain SKB_DST_NOREF bit. + * + * When in doubt, prefer skb_dst_drop() over skb_dstref_steal() to correctly + * handle dst_entry reference counting. + * + * Returns: original skb dst_entry. + */ +static inline unsigned long skb_dstref_steal(struct sk_buff *skb) +{ + unsigned long refdst = skb->_skb_refdst; + + skb->_skb_refdst = 0; + return refdst; +} + +/** + * skb_dstref_restore() - restore skb dst_entry removed via skb_dstref_steal() + * @skb: buffer + * @refdst: dst entry from a call to skb_dstref_steal() + */ +static inline void skb_dstref_restore(struct sk_buff *skb, unsigned long refdst) +{ + skb->_skb_refdst = refdst; +} + /** * skb_dst_set - sets skb dst * @skb: buffer -- cgit v1.2.3 From a890348adcc993f48d1ae38f1174dc8de4c3c5ac Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Mon, 18 Aug 2025 08:40:32 -0700 Subject: net: Add skb_dst_check_unset To prevent dst_entry leaks, add warning when the non-NULL dst_entry is rewritten. Signed-off-by: Stanislav Fomichev Link: https://patch.msgid.link/20250818154032.3173645-8-sdf@fomichev.me Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux/skbuff.h') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 7538ca507ee9..ca8be45dd8be 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1159,6 +1159,12 @@ static inline struct dst_entry *skb_dst(const struct sk_buff *skb) return (struct dst_entry *)(skb->_skb_refdst & SKB_DST_PTRMASK); } +static inline void skb_dst_check_unset(struct sk_buff *skb) +{ + DEBUG_NET_WARN_ON_ONCE((skb->_skb_refdst & SKB_DST_PTRMASK) && + !(skb->_skb_refdst & SKB_DST_NOREF)); +} + /** * skb_dstref_steal() - return current dst_entry value and clear it * @skb: buffer @@ -1188,6 +1194,7 @@ static inline unsigned long skb_dstref_steal(struct sk_buff *skb) */ static inline void skb_dstref_restore(struct sk_buff *skb, unsigned long refdst) { + skb_dst_check_unset(skb); skb->_skb_refdst = refdst; } @@ -1201,6 +1208,7 @@ static inline void skb_dstref_restore(struct sk_buff *skb, unsigned long refdst) */ static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst) { + skb_dst_check_unset(skb); skb->slow_gro |= !!dst; skb->_skb_refdst = (unsigned long)dst; } @@ -1217,6 +1225,7 @@ static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst) */ static inline void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst) { + skb_dst_check_unset(skb); WARN_ON(!rcu_read_lock_held() && !rcu_read_lock_bh_held()); skb->slow_gro |= !!dst; skb->_skb_refdst = (unsigned long)dst | SKB_DST_NOREF; -- cgit v1.2.3 From b08a784a5d1495c42ff9b0c70887d49211cddfe0 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 18 Aug 2025 19:03:54 +0100 Subject: net: Introduce skb_copy_datagram_from_iter_full() In a similar manner to copy_from_iter()/copy_from_iter_full(), introduce skb_copy_datagram_from_iter_full() which reverts the iterator to its initial state when returning an error. A subsequent fix for a vsock regression will make use of this new function. Cc: Christian Brauner Cc: Alexander Viro Signed-off-by: Will Deacon Acked-by: Michael S. Tsirkin Reviewed-by: Stefan Hajnoczi Link: https://patch.msgid.link/20250818180355.29275-2-will@kernel.org Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux/skbuff.h') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 14b923ddb6df..fa633657e4c0 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -4172,6 +4172,8 @@ int skb_copy_and_crc32c_datagram_iter(const struct sk_buff *skb, int offset, struct iov_iter *to, int len, u32 *crcp); int skb_copy_datagram_from_iter(struct sk_buff *skb, int offset, struct iov_iter *from, int len); +int skb_copy_datagram_from_iter_full(struct sk_buff *skb, int offset, + struct iov_iter *from, int len); int zerocopy_sg_from_iter(struct sk_buff *skb, struct iov_iter *frm); void skb_free_datagram(struct sock *sk, struct sk_buff *skb); int skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags); -- cgit v1.2.3 From ed8a507b748336902525aa79e3573552534e8b3e Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 16 Sep 2025 17:09:30 -0700 Subject: net: modify core data structures for PSP datapath support Add pointers to psp data structures to core networking structs, and an SKB extension to carry the PSP information from the drivers to the socket layer. Reviewed-by: Willem de Bruijn Signed-off-by: Jakub Kicinski Co-developed-by: Daniel Zahka Signed-off-by: Daniel Zahka Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20250917000954.859376-4-daniel.zahka@gmail.com Signed-off-by: Paolo Abeni --- include/linux/skbuff.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux/skbuff.h') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 62e7addccdf6..78ecfa7d00d0 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -4901,6 +4901,9 @@ enum skb_ext_id { #endif #if IS_ENABLED(CONFIG_MCTP_FLOWS) SKB_EXT_MCTP, +#endif +#if IS_ENABLED(CONFIG_INET_PSP) + SKB_EXT_PSP, #endif SKB_EXT_NUM, /* must be last */ }; -- cgit v1.2.3 From 3271f19bf7b9df665549666d789b9f126b4420c7 Mon Sep 17 00:00:00 2001 From: Richard Gobert Date: Tue, 23 Sep 2025 10:59:06 +0200 Subject: net: gso: restore ids of outer ip headers correctly Currently, NETIF_F_TSO_MANGLEID indicates that the inner-most ID can be mangled. Outer IDs can always be mangled. Make GSO preserve outer IDs by default, with NETIF_F_TSO_MANGLEID allowing both inner and outer IDs to be mangled. This commit also modifies a few drivers that use SKB_GSO_FIXEDID directly. Signed-off-by: Richard Gobert Reviewed-by: Edward Cree # for sfc Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20250923085908.4687-4-richardbgobert@gmail.com Signed-off-by: Paolo Abeni --- include/linux/skbuff.h | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'include/linux/skbuff.h') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 78ecfa7d00d0..fb3fec9affaa 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -674,7 +674,7 @@ enum { /* This indicates the tcp segment has CWR set. */ SKB_GSO_TCP_ECN = 1 << 2, - SKB_GSO_TCP_FIXEDID = 1 << 3, + __SKB_GSO_TCP_FIXEDID = 1 << 3, SKB_GSO_TCPV6 = 1 << 4, @@ -707,6 +707,12 @@ enum { SKB_GSO_FRAGLIST = 1 << 18, SKB_GSO_TCP_ACCECN = 1 << 19, + + /* These indirectly map onto the same netdev feature. + * If NETIF_F_TSO_MANGLEID is set it may mangle both inner and outer IDs. + */ + SKB_GSO_TCP_FIXEDID = 1 << 30, + SKB_GSO_TCP_FIXEDID_INNER = 1 << 31, }; #if BITS_PER_LONG > 32 -- cgit v1.2.3 From f6ceec6434b5efff62cecbaa2ff74fc29b96c0c6 Mon Sep 17 00:00:00 2001 From: Ralf Lici Date: Tue, 21 Oct 2025 12:09:40 +0200 Subject: net: datagram: introduce datagram_poll_queue for custom receive queues Some protocols using TCP encapsulation (e.g., espintcp, openvpn) deliver userspace-bound packets through a custom skb queue rather than the standard sk_receive_queue. Introduce datagram_poll_queue that accepts an explicit receive queue, and convert datagram_poll into a wrapper around datagram_poll_queue. This allows protocols with custom skb queues to reuse the core polling logic without relying on sk_receive_queue. Cc: Sabrina Dubroca Cc: Antonio Quartulli Signed-off-by: Ralf Lici Reviewed-by: Sabrina Dubroca Reviewed-by: Antonio Quartulli Link: https://patch.msgid.link/20251021100942.195010-2-ralf@mandelbit.com Signed-off-by: Paolo Abeni --- include/linux/skbuff.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux/skbuff.h') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index fb3fec9affaa..a7cc3d1f4fd1 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -4204,6 +4204,9 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, struct sk_buff_head *sk_queue, unsigned int flags, int *off, int *err); struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned int flags, int *err); +__poll_t datagram_poll_queue(struct file *file, struct socket *sock, + struct poll_table_struct *wait, + struct sk_buff_head *rcv_queue); __poll_t datagram_poll(struct file *file, struct socket *sock, struct poll_table_struct *wait); int skb_copy_datagram_iter(const struct sk_buff *from, int offset, -- cgit v1.2.3