CVE-2026-46223
CVE-2026-46223 is a medium-severity vulnerability in Linux Linux Kernel with a CVSS 3.x base score of 5.5. It is not currently listed as actively exploited by CISA, and its EPSS exploit-prediction score is low. The underlying weakness is classified as CWE-667.
Key facts
- Severity: Medium (CVSS 3.x base score 5.5)
- EPSS exploit prediction: 0% (1st percentile)
- Actively exploited: Not listed in CISA KEV
- EU (EUVD) id: EUVD-2026-32850
- Weakness: CWE-667
- Affected product: Linux Linux Kernel
- Published:
- Last modified:
Description
In the Linux kernel, the following vulnerability has been resolved: cgroup: Defer css percpu_ref kill on rmdir until cgroup is depopulated A chain of commits going back to v7.0 reworked rmdir to satisfy the controller invariant that a subsystem's ->css_offline() must not run while tasks are still doing kernel-side work in the cgroup. [1] d245698d727a ("cgroup: Defer task cgroup unlink until after the task is done switching out") [2] a72f73c4dd9b ("cgroup: Don't expose dead tasks in cgroup") [3] 1b164b876c36 ("cgroup: Wait for dying tasks to leave on rmdir") [4] 4c56a8ac6869 ("cgroup: Fix cgroup_drain_dying() testing the wrong condition") [5] 13e786b64bd3 ("cgroup: Increment nr_dying_subsys_* from rmdir context") [1] moved task cset unlink from do_exit() to finish_task_switch() so a task's cset link drops only after the task has fully stopped scheduling. That made tasks past exit_signals() linger on cset->tasks until their final context switch, which led to a series of problems as what userspace expected to see after rmdir diverged from what the kernel needs to wait for. [2]-[5] tried to bridge that divergence: [2] filtered the exiting tasks from cgroup.procs; [3] had rmdir(2) sleep in TASK_UNINTERRUPTIBLE for them; [4] fixed the wait's condition; [5] made nr_dying_subsys_* visible synchronously. The cgroup_drain_dying() wait in [3] turned out to be a dead end. When the rmdir caller is also the reaper of a zombie that pins a pidns teardown (e.g. host PID 1 systemd reaping orphan pids that were re-parented to it during the same teardown), rmdir blocks in TASK_UNINTERRUPTIBLE waiting for those pids to free, the pids can't free because PID 1 is the reaper and it's stuck in rmdir, and the system A-A deadlocks. No internal lock ordering breaks this; the wait itself is the bug. The css killing side that drove the original reorder, however, can be made cleanly asynchronous: ->css_offline() is already async, run from css_killed_work_fn() driven by percpu_ref_kill_and_confirm(). The fix is to make that chain start only after all tasks have left the cgroup. rmdir's user-visible side then returns as soon as cgroup.procs and friends are empty, while ->css_offline() still runs only after the cgroup is fully drained. Verified by the original reproducer (pidns teardown + zombie reaper, runs under vng) which hangs vanilla and succeeds here, and by per-commit deterministic repros for [2], [3], [4], [5] with a boot parameter that widens the post-exit_signals() window so each state is reliably reachable. Some stress tests on top of that. cgroup_apply_control_disable() has the same shape of pre-existing race: when a controller is disabled via subtree_control, kill_css() ran synchronously while tasks past exit_signals() could still be linked to the cgroup's csets, and ->css_offline() could fire before they drained. This patch preserves the existing synchronous behavior at that call site (kill_css_sync() + kill_css_finish() back-to-back) and a follow-up patch will defer kill_css_finish() there using a per-css trigger. This seems like the right approach and I don't see problems with it. The changes are somewhat invasive but not excessively so, so backporting to -stable should be okay. If something does turn out to be wrong, the fallback is to revert the entire chain ([1]-[5]) and rework in the development branch instead. v2: Pin cgrp across the deferred destroy work with explicit cgroup_get()/cgroup_put() around queue_work() and the work_fn. v1 wasn't actually broken (ordered cgroup_offline_wq + queue_work order in cgroup_task_dead() saved it) but the explicit ref removes the dependency on those non-obvious invariants. Also note the pre-existing cgroup_apply_control_disable() race in the description; a follow-up will defer kill_css_finish() there.
Frequently asked questions
- What is CVE-2026-46223?
- In the Linux kernel, the following vulnerability has been resolved: cgroup: Defer css percpu_ref kill on rmdir until cgroup is depopulated A chain of commits going back to v7.0 reworked rmdir to satisfy the controller invariant that a subsystem's ->css_offline() must not run while tasks are still doing kernel-side work in the cgroup. [1] d245698d727a ("cgroup: Defer task cgroup unlink until after the task is done switching out") [2] a72f73c4dd9b ("cgroup: Don't expose dead tasks in cgroup") [3] 1b164b876c36 ("cgroup: Wait for dying tasks to leave on rmdir") [4] 4c56a8ac6869 ("cgroup: Fix cgroup_drain_dying() testing the wrong condition") [5] 13e786b64bd3 ("cgroup: Increment nr_dying_subsys_* from rmdir context") [1] moved task cset unlink from do_exit() to finish_task_switch() so a task's cset link drops only after the task has fully stopped scheduling. That made tasks past exit_signals() linger on cset->tasks until their final context switch, which led to a series of problems as what userspace expected to see after rmdir diverged from what the kernel needs to wait for. [2]-[5] tried to bridge that divergence: [2] filtered the exiting tasks from cgroup.procs; [3] had rmdir(2) sleep in TASK_UNINTERRUPTIBLE for them; [4] fixed the wait's condition; [5] made nr_dying_subsys_* visible synchronously. The cgroup_drain_dying() wait in [3] turned out to be a dead end. When the rmdir caller is also the reaper of a zombie that pins a pidns teardown (e.g. host PID 1 systemd reaping orphan pids that were re-parented to it during the same teardown), rmdir blocks in TASK_UNINTERRUPTIBLE waiting for those pids to free, the pids can't free because PID 1 is the reaper and it's stuck in rmdir, and the system A-A deadlocks. No internal lock ordering breaks this; the wait itself is the bug. The css killing side that drove the original reorder, however, can be made cleanly asynchronous: ->css_offline() is already async, run from css_killed_work_fn() driven by percpu_ref_kill_and_confirm(). The fix is to make that chain start only after all tasks have left the cgroup. rmdir's user-visible side then returns as soon as cgroup.procs and friends are empty, while ->css_offline() still runs only after the cgroup is fully drained. Verified by the original reproducer (pidns teardown + zombie reaper, runs under vng) which hangs vanilla and succeeds here, and by per-commit deterministic repros for [2], [3], [4], [5] with a boot parameter that widens the post-exit_signals() window so each state is reliably reachable. Some stress tests on top of that. cgroup_apply_control_disable() has the same shape of pre-existing race: when a controller is disabled via subtree_control, kill_css() ran synchronously while tasks past exit_signals() could still be linked to the cgroup's csets, and ->css_offline() could fire before they drained. This patch preserves the existing synchronous behavior at that call site (kill_css_sync() + kill_css_finish() back-to-back) and a follow-up patch will defer kill_css_finish() there using a per-css trigger. This seems like the right approach and I don't see problems with it. The changes are somewhat invasive but not excessively so, so backporting to -stable should be okay. If something does turn out to be wrong, the fallback is to revert the entire chain ([1]-[5]) and rework in the development branch instead. v2: Pin cgrp across the deferred destroy work with explicit cgroup_get()/cgroup_put() around queue_work() and the work_fn. v1 wasn't actually broken (ordered cgroup_offline_wq + queue_work order in cgroup_task_dead() saved it) but the explicit ref removes the dependency on those non-obvious invariants. Also note the pre-existing cgroup_apply_control_disable() race in the description; a follow-up will defer kill_css_finish() there.
- How severe is CVE-2026-46223?
- CVE-2026-46223 has a CVSS 3.x base score of 5.5, rated medium severity. It is exploitable over local access with low attack complexity, requires low privileges and no user interaction. Impact on confidentiality is none, integrity none, and availability high.
- Is CVE-2026-46223 being actively exploited?
- It is not currently listed in CISA's KEV catalog. Its EPSS exploit-prediction score is 0% (1st percentile), an estimate of the probability of exploitation in the next 30 days.
- What products are affected by CVE-2026-46223?
- CVE-2026-46223 primarily affects Linux Linux Kernel. In total, 5 product configurations (CPEs) are listed as vulnerable; see the affected-products list for the exact versions.
- How do I fix CVE-2026-46223?
- Review the linked vendor and NVD advisories for patched versions and mitigations, then upgrade or apply the recommended workaround.
- Does CVE-2026-46223 have an EU (EUVD) identifier?
- Yes. CVE-2026-46223 is tracked in the ENISA EU Vulnerability Database (EUVD) as EUVD-2026-32850.
- When was CVE-2026-46223 published?
- CVE-2026-46223 was published on 2026-05-28 and last updated on 2026-06-17.
References
- https://git.kernel.org/stable/c/33fa2e6b1507a0a377a151a8826438bedad1d0b0
- https://git.kernel.org/stable/c/93618edf753838a727dbff63c7c291dee22d656b
Affected products (5)
- cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*
- cpe:2.3:o:linux:linux_kernel:7.0:-:*:*:*:*:*:*
- cpe:2.3:o:linux:linux_kernel:7.0:rc7:*:*:*:*:*:*
- cpe:2.3:o:linux:linux_kernel:7.1:rc1:*:*:*:*:*:*
- cpe:2.3:o:linux:linux_kernel:7.1:rc2:*:*:*:*:*:*
More vulnerabilities in Linux Linux Kernel
- CVE-2023-2163 — Critical (CVSS 10.0): Incorrect verifier pruning in BPF in Linux Kernel >=5.4 leads to unsafe code paths being incorrectly marked as safe,…
- CVE-2015-8104 — Critical (CVSS 10.0): The KVM subsystem in the Linux kernel through 4.2.6, and Xen 4.3.x through 4.6.x, allows guest OS users to cause a…
- CVE-2015-1421 — Critical (CVSS 10.0): Use-after-free vulnerability in the sctp_assoc_update function in net/sctp/associola.c in the Linux kernel before…
- CVE-2014-2523 — Critical (CVSS 10.0): net/netfilter/nf_conntrack_proto_dccp.c in the Linux kernel through 3.13.6 uses a DCCP header pointer incorrectly,…
- CVE-2010-2495 — Critical (CVSS 10.0): The pppol2tp_xmit function in drivers/net/pppol2tp.c in the L2TP implementation in the Linux kernel before 2.6.34 does…
- CVE-2010-2521 — Critical (CVSS 10.0): Multiple buffer overflows in fs/nfsd/nfs4xdr.c in the XDR implementation in the NFS server in the Linux kernel before…
All CVEs affecting Linux Linux Kernel →
Other CWE-667 (Improper Locking) vulnerabilities
- CVE-2020-12658 — Critical (CVSS 9.8): gssproxy (aka gss-proxy) before 0.8.3 does not unlock cond_mutex before pthread exit in gp_worker_main() in…
- CVE-2019-5886 — Critical (CVSS 9.8): An issue was discovered in ShopXO 1.2.0. In the application\install\controller\Index.php file, there is no validation…
- CVE-2026-43215 — High (CVSS 8.8): In the Linux kernel, the following vulnerability has been resolved: cifs: Fix locking usage for tcon fields We used…
- CVE-2026-31629 — High (CVSS 8.8): In the Linux kernel, the following vulnerability has been resolved: nfc: llcp: add missing return after LLCP_CLOSED…
- CVE-2020-15674 — High (CVSS 8.8): Mozilla developers reported memory safety bugs present in Firefox 80. Some of these bugs showed evidence of memory…
- CVE-2021-1622 — High (CVSS 8.6): A vulnerability in the Common Open Policy Service (COPS) of Cisco IOS XE Software for Cisco cBR-8 Converged Broadband…