ACK: [SRU][X][PATCH v2 00/17] CVE-2024-26921

Stewart Hore stewart.hore at canonical.com
Fri Feb 21 06:17:18 UTC 2025


On Fri, Feb 21, 2025 at 05:04:11PM +1100, Stewart Hore wrote:
> On Fri, Nov 29, 2024 at 06:06:07PM +0100, Juerg Haefliger wrote:
> > https://ubuntu.com/security/CVE-2024-26921
> >
> > https://warthogs.atlassian.net/browse/KERNSEC-324
> >
> > [ Impact ]
> >
> > Potential use-after-free of skb (socket buffer) fragments that are reassembled
> > via netfilter or openvswitch or similar modules.
> >
> >
> > [ Test Case ]
> >
> > Ran the ip_defrag kernel selftest from 6.10 and verified that no new failures occur.
> > In fact, some of the failing tests pass now. Also verified that the (main) modified
> > functions are called during the test.
> >
> >
> > [ Where Problems Could Occur ]
> >
> > Networking, netfilter, openvswitch, IPv4 and IPv6 defragmentation.
> >
> >
> > v1->v2:
> >   - Drop patch ("UBUNTU: SAUCE: inet: frags: introduce sum_truesize in inet_frag_reasm_finish()")
> >     and use head->truesize instead of sum_truesize in next commit
> >   - Backport ("inet: inet_defrag: prevent sk release while still in use") from linux-5.4.y
> >   - Fix provenance (backported from commit ccfa73daf762f3adac3f6c0e2f09c3c74548775f linux-4.14.y)
> >   - Include a72a5e2d34ec ("inet: kill unused skb_free op")
> >   - Cherry-pick ("net: IP6 defrag: use rbtrees in nf_conntrack_reasm.c") from linux-4.14.y
> >
> > Daniele Di Proietto (1):
> >   openvswitch: Fix skb leak in IPv6 reassembly.
> >
> > Eric Dumazet (1):
> >   net: add __sock_wfree() helper
> >
> > Florian Westphal (6):
> >   netfilter: ipv6: nf_defrag: avoid/free clone operations
> >   inet: kill unused skb_free op
> >   netfilter: ipv6: avoid nf_iterate recursion
> >   netfilter: ipv6: nf_defrag: fix NULL deref panic
> >   netfilter: ipv6: nf_defrag: drop mangled skb on ream error
> >   inet: inet_defrag: prevent sk release while still in use
> >
> > Guillaume Nault (2):
> >   netfilter: ipv6: nf_defrag: fix leakage of unqueued fragments
> >   netfilter: ipv6: nf_defrag: accept duplicate fragments again
> >
> > Jiri Wiesner (1):
> >   ipv4: ipv6: netfilter: Adjust the frag mem limit when truesize changes
> >
> > Joe Stringer (1):
> >   openvswitch: Orphan skbs before IPv6 defrag
> >
> > Peter Oskolkov (2):
> >   net: IP defrag: encapsulate rbtree defrag code into callable functions
> >   net: IP6 defrag: use rbtrees in nf_conntrack_reasm.c
> >
> > Subash Abhinov Kasiviswanathan (2):
> >   netfilter: ipv6: nf_defrag: Pass on packets to stack per RFC2460
> >   netfilter: ipv6: nf_defrag: Kill frag queue on RFC2460 failure
> >
> > Yu Zhe (1):
> >   ipv4: remove unnecessary type castings
> >
> >  include/linux/skbuff.h                      |   5 +-
> >  include/net/inet_frag.h                     |  17 +-
> >  include/net/netfilter/ipv6/nf_defrag_ipv6.h |   3 +-
> >  include/net/sock.h                          |   1 +
> >  net/core/sock.c                             |  11 +
> >  net/core/sock_destructor.h                  |  12 +
> >  net/ieee802154/6lowpan/reassembly.c         |   1 -
> >  net/ipv4/fib_frontend.c                     |   4 +-
> >  net/ipv4/fib_rules.c                        |   2 +-
> >  net/ipv4/fib_trie.c                         |   2 +-
> >  net/ipv4/icmp.c                             |   2 +-
> >  net/ipv4/igmp.c                             |   4 +-
> >  net/ipv4/inet_fragment.c                    | 347 +++++++++++++++++++-
> >  net/ipv4/ip_fragment.c                      | 298 ++---------------
> >  net/ipv4/ping.c                             |   2 +-
> >  net/ipv6/netfilter/nf_conntrack_reasm.c     | 329 ++++++-------------
> >  net/ipv6/netfilter/nf_defrag_ipv6_hooks.c   |  20 +-
> >  net/ipv6/reassembly.c                       |   8 +-
> >  net/openvswitch/conntrack.c                 |  27 +-
> >  19 files changed, 539 insertions(+), 556 deletions(-)
> >  create mode 100644 net/core/sock_destructor.h
> >
> > --
> > 2.43.0
>
> Acked-by: Stewart Hore <stewart.hore at canonical.com>
>

Sorry I sent this ACK to the incorrect mailing list, please disregard.

> --
> kernel-team mailing list
> kernel-team at lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team



More information about the kernel-team mailing list