[CVE-2026-74594][MODERATE 7.0] sched/psi: Shut down rtpoll_timer in psi_cgroup_free()

[AL-KERNEL <alkernel@kernelcve.org>]

CVE: CVE-2026-74594
Priority: MODERATE 7.0
AL-KERNEL base severity: MODERATE
KPANIC flag: YES
Patch: sched/psi: Shut down rtpoll_timer in psi_cgroup_free()
Commit: 1e5ca82eee59caca6988f9d6e859786aab8a5fa0
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=1e5ca82eee59caca6988f9d6e859786aab8a5fa0
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74594
Analysis date: Sat, 22 Aug 2026 12:22:58 -0400
ActionableScore: 7
ActionableScore lower bound: 4
Actionable bucket: Strong Important candidate / Actionable Moderate at minimum
Manual review required: YES

Summary:
A race in sched/psi can rearm `rtpoll_timer` after the last PSI trigger is destroyed, allowing `poll_timer_fn()` to run on freed cgroup PSI memory and causing at least a local kernel crash with UAF based privilege escalation concerns.

======================================================================
ABOUT THIS REPORT
======================================================================

The original Linux kernel CVE announcement for CVE-2026-74594 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74594

The original announcement does not normally provide a security severity
estimate, CVSS assessment, or enough information to determine whether the
reported kernel bug represents a practically relevant security issue.

This report was generated by AL-KERNEL, an AI-assisted Linux kernel
vulnerability analysis system developed by Alexander Larkin. It combines
an autonomous classifier with LLM-assisted technical analysis and a
separate ActionableScore mechanism.

The purpose of this report is to prioritize Linux kernel CVEs before
manual review, identify cases that require prompt investigation, and
support automatic closure of issues that are unlikely to have meaningful
security impact.

Published priority for this report: MODERATE 7.0
Manual review required: YES

A detailed explanation of the methodology and priority rules is included
at the end of this message.

======================================================================
AL-KERNEL CLASSIFICATION RESULT
======================================================================

CVE-2026-74594	MODERATE	CHECK WITH IMPACT FROM ORIG NN MODERATE	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-362;CWE-667;*CWE-664;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'A race in sched/psi can leave rtpoll_timer() armed after the last PSI trigger is destroyed and the owning cgroup PSI state is freed. psi_schedule_rtpoll_work() can run locklessly from the scheduler hotpath and call mod_timer() after psi_trigger_destroy() has called timer_delete(), so poll_timer_fn() may later run on freed memory. For the CVSS the PR:L is chosen because a local user or delegated service with access to PSI triggers and cgroup lifecycle operations may be able to exercise the vulnerable path without full host root. The issue is not network reachable and requires local control plane activity around PSI trigger teardown and cgroup removal. Impact is at least local denial of service via kernel crash, and the explicit use-after-free makes limited or full confidentiality and integrity impact plausible enough for manual review.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 6)	YES	INIT RACE LEAK UAF TIMER LINUS  KPANIC INCREASED_TO_MODERATE7_BASED_ON_GUESSCVSS KPANIC INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN6 DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFDH	NO	NO	checked

======================================================================
ACTIONABLESCORE ANALYSIS
======================================================================

ActionableScore=7
ActionableScoreLower=4

## 1. ActionableScore

* Conservative score: 4
* Paranoid score: 7
* Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**::

## 2. Signal breakdown

Conservative signals:

* Local unprivileged or low privileged trigger: +1. The path requires local PSI trigger and cgroup lifecycle activity, and may be reachable by delegated cgroup users or service accounts.
* Memory corruption, weak primitive: +1. The patch explicitly describes `poll_timer_fn()` running on freed memory, but no controlled reclaim or arbitrary write primitive is shown.
* Real lifetime corruption: +1. This is a timer lifetime bug where `rtpoll_timer` can be rearmed after trigger destruction and survive until after the PSI cgroup memory is freed.
* Reliable kernel crash / strong DoS: +1. A timer callback executing on freed memory is a credible kernel crash path.
* Broad/default/common subsystem: +1. PSI and cgroup scheduler accounting are common kernel infrastructure on modern distributions.
* Hard or unreliable race / special timing required: -1. The bug depends on a race between lockless scheduler hotpath scheduling and trigger destruction.

Paranoid additional interpretation:

* Generic memory corruption, strong corruption primitive: +2 instead of weak +1. The bug is an explicit UAF involving an asynchronous timer callback into freed cgroup PSI state.
* Weak LPE concern: +1. The stale timer callback and waitqueue access on freed memory make privilege escalation worth manual review, although no concrete exploit primitive is demonstrated.
* Availability impact realistic: +1. The expected failure mode is host wide kernel crash or severe instability, not merely a local process failure.

## 3. Reachability analysis

The issue is not network reachable. It requires local activity involving PSI triggers and cgroup teardown. In a strict host configuration, creating and destroying the relevant PSI triggers and cgroups may require administrative control over cgroups, but delegated cgroup setups, containers, service managers, or workload orchestration can reduce this to a low privileged local actor. Namespaces and cgroup delegation therefore matter for practical PR assessment.

The affected code is in scheduler PSI and cgroup lifetime handling, not a rare device driver. The race is timing dependent, but it is between realistic operations: scheduler hotpath `psi_schedule_rtpoll_work()` and `psi_trigger_destroy()` during final trigger removal. Exploitation conditions are local and race dependent, but the commit clearly states that the timer can remain pending after the group is freed.

Call-site confidence: medium. The patch and commit provide the relevant lifetime sequence and the callback consequence, but not a full reproducer or reclaim primitive.

## 4. Severity interpretation

This is stronger than an ordinary Moderate crash-only bug because the failure mode is an explicit use-after-free of cgroup PSI state by an asynchronous timer callback. Conservative handling can keep it as borderline Moderate because the race is hard and no controlled reclaim, arbitrary write, or demonstrated LPE primitive is shown. Paranoid triage should treat it as an Important candidate because local kernel UAFs involving timers and freed cgroup objects can be underestimated and may become exploitable depending on allocator reuse and timing.

Theoretical memory corruption potential is real. Realistic exploitation evidence currently supports at least local DoS, while privilege escalation remains plausible but unproven.

## 5. One-sentence report phrase

A race in sched/psi can rearm `rtpoll_timer` after the last PSI trigger is destroyed, allowing `poll_timer_fn()` to run on freed cgroup PSI memory and causing at least a local kernel crash with UAF based privilege escalation concerns.

## 6. Manual review recommendation

MANUAL CHECK REQUIRED.

Reason: explicit UAF on freed cgroup PSI memory through an asynchronous timer callback, local reachable control-plane conditions, and plausible but unproven privilege escalation potential.

======================================================================
UPSTREAM PATCH SUMMARY
======================================================================

Patch: sched/psi: Shut down rtpoll_timer in psi_cgroup_free()
Commit: 1e5ca82eee59caca6988f9d6e859786aab8a5fa0
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=1e5ca82eee59caca6988f9d6e859786aab8a5fa0

Commit description:

psi_schedule_rtpoll_work() is called locklessly from the scheduler hotpath
and can race psi_trigger_destroy() taking down the last rtpoll trigger under
rtpoll_trigger_lock:

  psi_schedule_rtpoll_work()        psi_trigger_destroy()

  rcu_read_lock();
  task = rcu_dereference(rtpoll_task);
                                    rcu_assign_pointer(rtpoll_task, NULL);
                                    timer_delete(&rtpoll_timer);
  mod_timer(&rtpoll_timer, ...);
  rcu_read_unlock();
                                    synchronize_rcu();
                                    kthread_stop(task_to_destroy);

The group can then be freed with the re-armed timer still pending, and
poll_timer_fn() runs on freed memory.

461daba ("psi: eliminate kthread_worker from psi trigger scheduling
mechanism") deleted the timer synchronously after the synchronize_rcu(),
which prevented this but raced trigger creation instead: the deletion could
cancel the timer that a new trigger set armed during the grace period and,
as creation also reinitialized the timer at the time, corrupt it.
8f91efd ("psi: Fix race between psi_trigger_create/destroy") moved the
initialization into group_init() and the deletion into the locked section,
trading the creation races for the window above.

Neither placement in the destruction path works. A pending timer firing
while the group is alive is harmless though. poll_timer_fn() just wakes the
rtpoll waitqueue and doesn't re-arm itself. Bind the timer to the group's
lifetime instead and shut it down in psi_cgroup_free(). Nothing can arm it
by then. timer_shutdown_sync() because the timer is never armed again.

Fixes: 8f91efd ("psi: Fix race between psi_trigger_create/destroy")
Cc: stable@vger.kernel.org # v5.10+
Reported-by: Sashiko AI <sashiko-bot@kernel.org>
Closes: https://lore.kernel.org/all/20260711000434.36C4A1F000E9@smtp.kernel.org/
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Tested-by: Matt Fleming <mfleming@cloudflare.com>
Acked-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  kernel/sched/psi.c

Diff excerpt:

Not included in this email. See the upstream URL for the full patch.

Full patch:
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=1e5ca82eee59caca6988f9d6e859786aab8a5fa0

======================================================================
DETAILED REPORT METHODOLOGY
======================================================================

The original Linux kernel CVE announcement for CVE-2026-74594 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74594

The original CVE announcement normally does not include a security-level
estimate. In particular, it may not contain a CVSS assessment, an impact
level, or enough information to determine whether the reported bug is a
practically relevant security issue. One purpose of this parallel CVE list
is to provide that missing technical and prioritization information.

The original goal of the AL-KERNEL project was to prioritize Linux kernel
CVE analysis automatically before manual review. The system can also help
identify non-security issues that may be suitable for automatic closure.

This report was generated by AL-KERNEL, an AI-assisted Linux kernel
vulnerability analysis system developed by Alexander Larkin.

The first analysis stage combines an autonomous classifier with additional
LLM-based analysis. The autonomous classifier runs locally on a CPU and is
based on a backpropagation neural network. Together, these mechanisms
produce a technical vulnerability description, identify likely weakness
types, estimate CVSS severity, and provide input for ActionableScore.

Two CVSS estimates are retained because incomplete kernel vulnerability
information often permits more than one defensible interpretation:

  Conservative CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H
  The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
  The Best / paranoid CVSS score: 7

The conservative vector represents a lower-impact interpretation.
The Best/paranoid vector intentionally represents a plausible upper-bound
interpretation and should not automatically be treated as demonstrated
real-world impact.

CVSS may also need to be adjusted for a particular Linux deployment,
because actual reachability, privileges, enabled kernel configuration,
hardware, namespaces, exposed device nodes, and other environmental
conditions can differ significantly between systems.

A separate ActionableScore mechanism evaluates practical remediation
urgency. Its analysis may include reachability, attack prerequisites,
subsystem exposure, memory-corruption characteristics, denial-of-service
reliability, and possible confidentiality, integrity, or
privilege-escalation impact.

  Conservative ActionableScore: 4
  Paranoid ActionableScore: 7

The final base severity is taken directly from the second tab-separated
field of the AL-KERNEL classification result. ActionableScore does not
replace or independently override that final AL-KERNEL decision, and
if ActionableScore adjusted impact level of ALKERNEL, then you would see
self-readable flags above like INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN7.

For an AL-KERNEL result of MODERATE, this report uses the following
additional presentation split:

  ActionableScore below 5   -> MODERATE REGULAR
  ActionableScore 5 or more -> MODERATE 7.0

The distinction between MODERATE REGULAR and MODERATE 7.0 makes it
possible to identify Moderate issues that should receive manual analysis
and fixes before lower-priority MODERATE REGULAR issues. In many cases,
MODERATE REGULAR fixes may wait for a later rebase or routine update.

There is one override in which MODERATE REGULAR becomes MODERATE 7.0
even when the ActionableScore is below 5. When the AL-KERNEL result
contains the KPANIC flag, a MODERATE result is always presented as
MODERATE 7.0. The KPANIC flag selected with few regexps without
usage of AI at all, so it helps to detect cases when Kernel Crash happens
and similar (to filter False-Negative results from the LLM usage).

KPANIC indicates that a reliable kernel crash, kernel panic, or similarly
serious kernel availability impact was identified by the classification
workflow.

AL-KERNEL base severity for this report: MODERATE
KPANIC detected for this report: YES
Published priority for this report (same as in Subject): MODERATE 7.0

These results are intended to support engineering triage. They are
machine-generated estimates, and cases marked for manual review should
be validated by a human security engineer before final disposition.
For more info read docs linked from here: https://kernelcve.org/
(and you can submit you own patch there to generate such a report
for non-existant CVE-id yet).

Note that in many cases this AI tool selects higher severity, than
real is (means you can expect Importants instead of Moderate 7.0 or
Moderates 7.0 instead of regular Moderates). If you see such cases,
please use reply email interface to add additional manual analyses
info to this particular CVE.
And please, please, let me know when you see Lows instead of Importants
or Important instead of Low (because particular for such cases I
need to tune this AI tool to make it better for this one and next similar).
My contact email for such notifications is alexanjelausa@gmail.com (and both
send reply to CVE record itself too and see "reply" button below for howto reply).