[PATCH net v2] ip: skip IPv4 cmsgs for non-IPv4 packets

[Daehyeon Ko]

IPV6_ADDRFORM can convert an IPv6 UDP socket while a native IPv6
packet is still queued for reception. The converted socket then receives
the packet through udp_recvmsg(), which passes it to
ip_cmsg_recv_offset() as if it were IPv4.

This makes the IPv4 ancillary-data code interpret inet6_skb_parm as
inet_skb_parm. In particular, IP_RETOPTS can take the IPv6 lastopt and
nhoff offsets as an IPv4 option length and RR offset. A destination
options header can then make __ip_options_echo() copy more than the
40-byte stack option buffer.

KASAN reported a 152-byte write on 3/3 fresh v6.12.105 boots and 3/3
fresh v7.2 boots. A production-shaped v6.12.105 build instead reached a
hardened-usercopy BUG followed by a fatal-exception panic on 3/3 boots.

Check the network header version before parsing SOL_IP control
messages. skb->protocol is not a reliable discriminator here: IPv4
multicast route cache reports can leave it unset, while software VLAN
processing can change it to ETH_P_8021Q before a TX timestamp is queued.
Both paths still carry an IPv4 network header and valid IP_PKTINFO.

With this change the original trigger returned its payload without a
control message or a target KASAN report. The no-extension and native
IPv6 controls passed, and IP_PKTINFO was preserved for both the
multicast route report and software VLAN TX timestamp paths.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Cc: stable@xxxxxxxxxxxxxxx
Assisted-by: LLM
Signed-off-by: Daehyeon Ko <4ncienth@xxxxxxxxx>
---
Changes in v2:
- Use the actual network header version instead of skb->protocol as the
  discriminator.
- Preserve IP_PKTINFO for IPv4 multicast route cache reports and software
  VLAN TX timestamp error-queue skbs.
- Validate both reviewer-reported paths alongside the original trigger and
  controls.
---
 net/ipv4/ip_sockglue.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index a55ef327ec932c..01d264e8f34ea1 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -174,6 +174,9 @@ void ip_cmsg_recv_offset(struct msghdr *msg, struct sock *sk,
 	if (!flags)
 		return;
 
+	if (unlikely(ip_hdr(skb)->version != IPVERSION))
+		return;
+
 	/* Ordered by supposed usage frequency */
 	if (flags & IP_CMSG_PKTINFO) {
 		ip_cmsg_recv_pktinfo(msg, skb);
-- 
2.54.0