[CVE-2026-89984][MODERATE REGULAR] perf/x86/intel: Fix kernel address leakages in LBR stack

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

CVE: CVE-2026-89984
Priority: MODERATE REGULAR
AL-KERNEL base severity: MODERATE
KPANIC flag: NO
Patch: perf/x86/intel: Fix kernel address leakages in LBR stack
Commit: f3705db4e6378cc1b91a38ce0fe2f879055e23ca
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=f3705db4e6378cc1b91a38ce0fe2f879055e23ca
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89984
Analysis date: Wed, 16 Sep 2026 13:46:29 -0400
ActionableScore: 4
ActionableScore lower bound: 3
Actionable bucket: Borderline / manual review recommended
Manual review required: YES

Summary:
Intel LBR filtering can expose kernel branch source addresses to local users through perf even when a user-only branch stack is requested, creating a kernel address disclosure and possible KASLR bypass aid.

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

The original Linux kernel CVE announcement for CVE-2026-89984 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89984

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 REGULAR
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-2026-89984	MODERATE	https://www.kernelcve.org/list/?q=CVE-2026-89984 CHECK WITH IMPACT FROM ORIG NN LOW	Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N';*CWE-200;CWE-668;CWE-693;Other CVSS 'AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:N/A:N';BEST CVSS score: '5.5';DESCR 'Intel LBR software filtering can expose kernel branch addresses to user space even when a user-only branch stack is requested through PERF_SAMPLE_BRANCH_USER. The issue occurs because intel_pmu_lbr_filter validates branch type and the to address but does not fully reject records whose from address is in kernel space. A local user with access to perf branch sampling can observe kernel-space SYSRET or ERET source addresses and use them as a KASLR bypass aid. For the CVSS the PR:L is selected because triggering requires local perf access but not full administrative control in environments where perf_event_open is available to users. The issue is not network reachable and does not by itself provide code execution or a kernel crash. Impact is confidentiality exposure only, with no direct integrity or availability impact.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 4)	YES	OOB LEAK HARDWARE  INCREASED_FROM_LOW_BASED_ON_REQUIREMANUALCHECK DECREASED_TO_MODERATEREG_BASED_ON_FALSEPOSCHECKOFKP	NO	NO	checked

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

ActionableScore=4
ActionableScoreLower=3
PrimitiveClass=NONE
RealUAF=NO
HostAdminRequired=NO
NamespaceScopedPrivilege=UNKNOWN
ImportantCandidate=NO
ManualReview=RECOMMENDED
AutoDowngradeAllowed=YES

## 1. ActionableScore

* Raw conservative score: 3
* Conservative score: 3
* Raw paranoid score: 4
* Paranoid score: 4
* Final recommended bucket: **Borderline / manual review recommended**::

No floor or cap changed the score.

## 2. Signal breakdown

Conservative positive signals:

* +1: User-visible confidentiality exposure. Kernel-space branch source addresses are copied to perf output despite a user-only branch filter.
* +1: Local below-host-admin reachability. The demonstrated path is local perf branch sampling, not host-admin-only kernel modification.
* +1: Security-relevant filtering failure. The kernel/user privilege filter for LBR entries is incomplete because `from` addresses are not rejected when only user branches are requested.

Paranoid additional signal:

* +1: KASLR-bypass relevance. The leaked kernel branch addresses are precise kernel pointers and may materially assist another local kernel exploit chain.

Non-awarded signals:

* No UAF, OOB write, arbitrary write, refcount misuse, or memory corruption is established.
* No integrity impact is awarded because the bug only discloses branch addresses.
* No availability impact is awarded because the patch does not indicate crash, panic, WARN-only fatality, or resource exhaustion.
* No remote reachability is awarded because triggering requires local perf access.
* No privileged FD/resource theft or authorization bypass to a protected object is established.
* No Class A substantial disclosure is assigned because the demonstrated leak is kernel addresses, not arbitrary kernel memory or sensitive data contents.

## 3. Reachability analysis

The attacker is a local user or local process able to use perf branch sampling with Intel LBR. The supplied reproducer uses `perf record` with a user branch filter and shows kernel `SYSRET` and `ERET` source addresses in the branch stack.

Host-admin-equivalent privilege is not positively required by the patch context. In practice, access depends on `perf_event_paranoid`, `CAP_PERFMON`, `CAP_SYS_ADMIN`, distro policy, and whether unprivileged perf events are allowed. Because the vulnerable operation is a local perf interface and not a host configuration operation, the evaluated practical attacker is below host admin where perf access is available.

Namespace-scoped privilege is unresolved and not material here. No complete namespace-based trigger chain is established. There is no network trigger and no packet-processing path. Hardware dependency exists for Intel LBR behavior, but Intel LBR-capable systems are not rare enough to justify a physical-only or rare-hardware penalty.

Race repeatability is not relevant. This is a deterministic filtering bug, not a timing race.

## 4. Primitive and severity interpretation

The exact primitive is an information disclosure of kernel code addresses through the LBR branch stack. The vulnerable software filter validates branch type and partially validates privilege level using the destination side, but it fails to reject entries whose `from` address is in kernel space when a user-only branch stack is requested.

This is not active structured stale storage, not fixed stale storage, and not a lifetime bug. There is no concrete repeated release or refcount operation. There is no first lifetime operation that may free an object and no second stale lifetime mutation.

PrimitiveClass is `NONE` because no kernel corruption, ownership violation, lifetime violation, authorization-to-resource bypass, or strong semantic corruption primitive is established. The demonstrated consequence is confidentiality exposure of kernel addresses. The plausible speculative consequence is that the leak can weaken KASLR and assist a separate local exploit, but the patch does not establish code execution, privilege escalation, or arbitrary read/write.

ImportantCandidate is `NO` because this is a limited address disclosure rather than a strong corruption or boundary-bypass primitive. ManualReview is `RECOMMENDED` because kernel address disclosure can be security-relevant in exploit chaining and product perf restrictions strongly affect practical severity. AutoDowngradeAllowed is `YES` because the issue can be treated as a bounded local information disclosure if perf is privileged or disabled for untrusted users, and no unresolved corruption primitive remains.

Likely operational severity is low-to-moderate, with higher practical relevance on systems where untrusted local users can access perf branch sampling.

## 5. One-sentence report phrase

Intel LBR filtering can expose kernel branch source addresses to local users through perf even when a user-only branch stack is requested, creating a kernel address disclosure and possible KASLR bypass aid.

## 6. Manual review recommendation

MANUAL CHECK RECOMMENDED

Reason: the issue is not memory corruption or direct LPE, but it is a real kernel address disclosure whose severity depends on perf access policy, CPU support, and whether KASLR bypass aids are treated as actionable in the target product.

## 7. Call-site confidence

Call-site confidence: high

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

Patch: perf/x86/intel: Fix kernel address leakages in LBR stack
Commit: f3705db4e6378cc1b91a38ce0fe2f879055e23ca
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=f3705db4e6378cc1b91a38ce0fe2f879055e23ca

Commit description:

Before Arch LBR gained CPL filtering support, a user-only branch stack
could still contain kernel addresses. As a result, kernel branch records
may be exposed to user space even when PERF_SAMPLE_BRANCH_USER is
requested.

For example, on Intel Tiger Lake, the following command can still report
SYSRET/ERET entries with kernel-space from addresses:

$ ./perf record -e cycles:p -o - --branch-filter any,save_type,u -- \
 	./perf bench syscall basic --loop 1000 | \
	./perf script -i - --fields brstack|tr ' ' '\n'| \
	grep -E '0x[89a-f][0-9a-f]{15}'

    Total time: 0.000 [sec]

      0.219000 usecs/op
     4,566,210 ops/sec
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.551 MB - ]
0xffffffff93c001c8/0x7f12a2b1d647/P/-/-/16959/SYSRET/-
0xffffffff93c001c8/0x7f12a2b1d5c2/P/-/-/17535/SYSRET/-
0xffffffff93c01928/0x7f12a2861000/P/-/-/6719/ERET/-
0xffffffff93c01928/0x7f12a297a000/P/-/-/8575/ERET/-

The problem is that intel_pmu_lbr_filter() does not fully validate the
privilege level of sampled entries. It filters some mismatches based on
the branch type and the to address, but it does not reject entries whose
from address violates the requested branch privilege filter.

Fix this by extending software filtering to validate both from and to
addresses against br_sel. Any LBR entry contains kernel address does not
match the requested user filter is dropped. This prevents kernel
addresses from appearing in user-only branch stacks.

Fixes: 47125db ("perf/x86/intel/lbr: Support Architectural LBR")
Reported-by: Ian Rogers <irogers@google.com>
Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260616044654.3468742-5-dapeng1.mi@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  arch/x86/events/intel/lbr.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=f3705db4e6378cc1b91a38ce0fe2f879055e23ca

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

The original Linux kernel CVE announcement for CVE-2026-89984 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89984

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:L/PR:L/UI:N/S:U/C:L/I:N/A:N
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N
  The Best / paranoid CVSS score: 5.5

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: 3
  Paranoid ActionableScore: 4

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: NO
Published priority for this report (same as in Subject): MODERATE REGULAR

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).