Описание
In the Linux kernel, the following vulnerability has been resolved:
netfilter: nf_conntrack_reasm: guard mac_header adjustment after IPv6 defrag
nf_ct_frag6_reasm() slides the packet head forward to drop the IPv6 fragment header and then unconditionally advances skb->mac_header:
On the NF_INET_LOCAL_OUT defrag path the skb has no link-layer header yet, so skb->mac_header is still the "not set" sentinel (u16)~0U. Adding sizeof(struct frag_hdr) wraps it to a small value (0xffff + 8 == 7), after which skb_mac_header_was_set() wrongly reports a MAC header is present and skb_mac_header() points into the headroom.
The reassembler has done this unconditional add since it was introduced; it was harmless while mac_header was a bare pointer, but wrong once mac_header became a u16 offset whose unset state is the ~0U sentinel tested by skb_mac_header_was_set(). The sibling net/ipv6/reassembly.c does the same relocation and does guard the adjustm...
In the Linux kernel, the following vulnerability has been resolved:
netfilter: nf_conntrack_reasm: guard mac_header adjustment after IPv6 defrag
nf_ct_frag6_reasm() slides the packet head forward to drop the IPv6 fragment header and then unconditionally advances skb->mac_header:
On the NF_INET_LOCAL_OUT defrag path the skb has no link-layer header yet, so skb->mac_header is still the "not set" sentinel (u16)~0U. Adding sizeof(struct frag_hdr) wraps it to a small value (0xffff + 8 == 7), after which skb_mac_header_was_set() wrongly reports a MAC header is present and skb_mac_header() points into the headroom.
The reassembler has done this unconditional add since it was introduced; it was harmless while mac_header was a bare pointer, but wrong once mac_header became a u16 offset whose unset state is the ~0U sentinel tested by skb_mac_header_was_set(). The sibling net/ipv6/reassembly.c does the same relocation and does guard the adjustment; mirror the guard here.
Ссылки
- https://nvd.nist.gov/vuln/detail/CVE-2026-72250
- https://git.kernel.org/stable/c/00bdce2fda7e430d24cfbc96764a1b96deb31f82
- https://git.kernel.org/stable/c/2a95ec21824a8ad81ad660b12231456fc0ac9830
- https://git.kernel.org/stable/c/3b08fed5b7e0d5e3a25d73ef3ba09cd33ade16c9
- https://git.kernel.org/stable/c/53ef70a315420ed31581d38343684b3bf9a3c76d
- https://git.kernel.org/stable/c/6e8cd710ca35c576f5f2e5a396047c9ac61f75e5
- https://git.kernel.org/stable/c/a58230f3a7c4f6c3261786bc1efb72c42e68cd25
- https://git.kernel.org/stable/c/bbcdef2061b170af45702ce6b359c02c12acfc94
- https://git.kernel.org/stable/c/cd0d7bbc027b4d3329712cdcdeb4e5567ffd0d58
Связанные уязвимости
In the Linux kernel, the following vulnerability has been resolved: netfilter: nf_conntrack_reasm: guard mac_header adjustment after IPv6 defrag nf_ct_frag6_reasm() slides the packet head forward to drop the IPv6 fragment header and then unconditionally advances skb->mac_header: skb->mac_header += sizeof(struct frag_hdr); On the NF_INET_LOCAL_OUT defrag path the skb has no link-layer header yet, so skb->mac_header is still the "not set" sentinel (u16)~0U. Adding sizeof(struct frag_hdr) wraps it to a small value (0xffff + 8 == 7), after which skb_mac_header_was_set() wrongly reports a MAC header is present and skb_mac_header() points into the headroom. The reassembler has done this unconditional add since it was introduced; it was harmless while mac_header was a bare pointer, but wrong once mac_header became a u16 offset whose unset state is the ~0U sentinel tested by skb_mac_header_was_set(). The sibling net/ipv6/reassembly.c does the same relocation and does guard the adjustment; m...
In the Linux kernel, the following vulnerability has been resolved: netfilter: nf_conntrack_reasm: guard mac_header adjustment after IPv6 defrag nf_ct_frag6_reasm() slides the packet head forward to drop the IPv6 fragment header and then unconditionally advances skb->mac_header: skb->mac_header += sizeof(struct frag_hdr); On the NF_INET_LOCAL_OUT defrag path the skb has no link-layer header yet, so skb->mac_header is still the "not set" sentinel (u16)~0U. Adding sizeof(struct frag_hdr) wraps it to a small value (0xffff + 8 == 7), after which skb_mac_header_was_set() wrongly reports a MAC header is present and skb_mac_header() points into the headroom. The reassembler has done this unconditional add since it was introduced; it was harmless while mac_header was a bare pointer, but wrong once mac_header became a u16 offset whose unset state is the ~0U sentinel tested by skb_mac_header_was_set(). The sibling net/ipv6/reassembly.c does the same relocation and does guard the adjustment; m...
In the Linux kernel, the following vulnerability has been resolved: netfilter: nf_conntrack_reasm: guard mac_header adjustment after IPv6 defrag nf_ct_frag6_reasm() slides the packet head forward to drop the IPv6 fragment header and then unconditionally advances skb->mac_header: skb->mac_header += sizeof(struct frag_hdr); On the NF_INET_LOCAL_OUT defrag path the skb has no link-layer header yet, so skb->mac_header is still the "not set" sentinel (u16)~0U. Adding sizeof(struct frag_hdr) wraps it to a small value (0xffff + 8 == 7), after which skb_mac_header_was_set() wrongly reports a MAC header is present and skb_mac_header() points into the headroom. The reassembler has done this unconditional add since it was introduced; it was harmless while mac_header was a bare pointer, but wrong once mac_header became a u16 offset whose unset state is the ~0U sentinel tested by skb_mac_header_was_set(). The sibling net/ipv6/reassembly.c does the same relocation and does guard the adjustment
In the Linux kernel, the following vulnerability has been resolved: n ...