CVE-2026-72404
Description
In the Linux kernel, the following vulnerability has been resolved:
tipc: fix UAF in cleanup_bearer() due to premature dst_cache_destroy()
TIPC UDP media bearer teardown calls dst_cache_destroy() on its replicast caches before calling synchronize_net() to wait for concurrent RCU readers (transmitters) to finish:
static void cleanup_bearer(struct work_struct *work) { ... list_for_each_entry_safe(rcast, tmp, &ub->rcast.list, list) { dst_cache_destroy(&rcast->dst_cache); list_del_rcu(&rcast->list); kfree_rcu(rcast, rcu); } ... dst_cache_destroy(&ub->rcast.dst_cache); udp_tunnel_sock_release(ub->sk); synchronize_net(); ... }
This is highly buggy because dst_cache_destroy() immediately frees the per-CPU cache memory (free_percpu()) and releases the cached dst entries without any synchronization.
If a concurrent transmitter (e.g., tipc_udp_xmit()) is running on another CPU under RCU protection, it can call dst_cache_get() concurrently, leading to: 1. Use-After-Free on the per-CPU cache pointer itself (crash). 2. "rcuref - imbalanced put()" warning if it attempts to release a dst that was concurrently released by dst_cache_destroy().
Furthermore, calling kfree(ub) immediately after synchronize_net() without closing the socket first (or waiting after closing it) leaves a window where a concurrent receiver (tipc_udp_recv()) could start after synchronize_net(), access ub, and suffer a UAF when kfree(ub) runs.
To fix this, we must defer dst_cache_destroy() and kfree(ub) until after we have ensured that no more readers can see the bearer/socket and all existing readers have finished:
1. Defer rcast entry destruction (both dst_cache_destroy() and kfree()) to an RCU callback using call_rcu_hurry(). Using call_rcu_hurry() ensures the dst entries are released quickly.
2. Release the bearer socket using udp_tunnel_sock_release() (stops new receive readers).
3. Call synchronize_net() to wait for all outstanding RCU readers (both transmit and receive) to finish.
4. Now that it is safe, call dst_cache_destroy() on the main bearer cache, and free ub.
Note: 3) and 4) can be changed later in net-next to also use call_rcu_hurry() and get rid of the synchronize_net() latency.
Affected products
31- osv-coords30 versionspkg:rpm/opensuse/kernel-default-base&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/dtb-aarch64&distro=openSUSE%20Leap%2016.1pkg:linux/kernelpkg:rpm/opensuse/kernel-64kb&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-rt&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-source&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-syms&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-zfcpdump&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/dtb-aarch64&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-64kb&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-azure&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-default&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-default-base&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-docs&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-kvmsmall&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-obs-build&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-obs-qa&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-rt&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-source&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-syms&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-zfcpdump&distro=openSUSE%20Leap%2016.0pkg:rpm/opensuse/kernel-azure&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-default&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-docs&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-kvmsmall&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-obs-build&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-obs-qa&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-nvidia64kb&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-source-nvidia&distro=openSUSE%20Leap%2016.1pkg:rpm/opensuse/kernel-syms-nvidia&distro=openSUSE%20Leap%2016.1
< 6.12.0-160100.9.1.160100.3.14+ 29 more
- (no CPE)range: < 6.12.0-160100.9.1.160100.3.14
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: >= 5.3.0, < 6.1.191
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1.160000.2.25
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160000.39.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.9.1
- (no CPE)range: < 6.12.0-160100.6.1
- (no CPE)range: < 6.12.0-160100.6.1
- (no CPE)range: < 6.12.0-160100.6.1
Patches
Vulnerability mechanics
References
6- git.kernel.org/stable/c/1c8393eefa3cadf4ca0b61119ad1321aa32d3c8cnvd
- git.kernel.org/stable/c/3827ff67f9fbecd1086e33d23693ef6a1a9c6c38nvd
- git.kernel.org/stable/c/5a66ca574eee958123ee9d2822683964d42d656anvd
- git.kernel.org/stable/c/6aeb4ba4fb098199e13723cfd04d9a749741f731nvd
- git.kernel.org/stable/c/6b0e29f7662c237e717f35dad746efe94715d0danvd
- git.kernel.org/stable/c/7116764ca53ff529335d7ab7c364a69f094b23a5nvd
News mentions
0No linked articles in our index yet.