[CVE-2026-64591][LOW] iommu/vt-d: Avoid WARNING in sva unbind path

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

CVE: CVE-2026-64591
Priority: LOW
AL-KERNEL base severity: LOW
KPANIC flag: NO
Patch: iommu/vt-d: Avoid WARNING in sva unbind path
Commit: bb354384f40bb087e7c44f0f0743a23fc945f91d
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=bb354384f40bb087e7c44f0f0743a23fc945f91d
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64591
Analysis date: Thu, 06 Aug 2026 04:07:06 -0400
ActionableScore: 2
ActionableScore lower bound: 1
Actionable bucket: Ordinary Moderate / Low-like
Manual review required: NO

Summary:
The Intel VT-d SVA unbind path may disable IOPF for a non-PRI device where IOPF was never enabled, allowing a local user with access to an affected SVA-capable device to trigger a kernel WARNING and possibly a DoS on panic_on_warn systems.

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

The original Linux kernel CVE announcement for CVE-2026-64591 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64591

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: LOW
Manual review required: NO

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

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

CVE-2026-64591	LOW	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:N/I:N/A:H';CWE-754;CWE-703;*CWE-664;Other CVSS 'AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L';BEST CVSS score: '5.5';DESCR 'The Intel IOMMU SVA unbind path can call intel_iommu_disable_iopf for a device that does not support PRI and therefore never had IOPF enabled during bind. This state mismatch triggers a kernel WARNING when an SVA domain is unbound, for example from a GPU or accelerator close path using PASID based SVA. For the CVSS the PR:L is used because a local user with access to the relevant device node may be able to create and close the affected SVA context, while no administrative firewall or mount control is required in that model. The issue is not network reachable and does not show a memory corruption primitive. Impact is mainly local denial of service through warning induced disruption, with A:H only in the paranoid score for systems configured to panic on WARN.';NO MANUAL CHECK; ,and ActionableScore result is Ordinary Moderate / Low-like (with actual score 2)	YES	LOCK WARNONLY HARDWARE  INCREASED_FROM_LOW_BASED_ON_HIGHCVSSSCORE DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN5 DECREASED_TO_LOW_FROM_MODERATE_BASED_ON_ACTIONABLESCORELESSTHAN3	YES	NO	checked

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

ActionableScore=2
ActionableScoreLower=1

## 1. ActionableScore

* Conservative score: **1**
* Paranoid score: **2**
* Final recommended bucket: **Ordinary Moderate / Low-like**::

## 2. Signal breakdown

Conservative signals:

* **Local unprivileged trigger: +1**. The trace shows the path through `xe_vm_close_and_put()`, so a local user with access to the relevant GPU or accelerator device node may be able to trigger SVA unbind.
* **No memory corruption: +0**. The patch fixes an IOPF enable/disable state mismatch. It does not show UAF, OOB access, stale pointer reuse, refcount misuse, arbitrary write, or type confusion.
* **WARN only: +0**. The described failure is a kernel WARNING in `intel_iommu_disable_iopf()`. By itself, WARN followed by continued execution is not strong DoS evidence.
* **No remote reachability: +0**. This is a local device and IOMMU management path, not a network-triggerable path.
* **No confidentiality or integrity signal: +0**. There is no evidence of data disclosure, permission bypass, or corruption of attacker-relevant objects.

Paranoid additional signal:

* **Reliable kernel crash / strong DoS: +1**. Only for systems configured with `panic_on_warn` or equivalent hardening, the user-triggerable WARNING can become a host crash.

## 3. Reachability analysis

The likely trigger is local and tied to SVA/PASID use on Intel VT-d systems with devices that allow SVA but lack PCI PRI support. The call trace through the Intel `xe` driver suggests this may be reachable by a local user who can open and close an affected GPU or accelerator context. It does not require network access.

Namespaces or containers may matter only if the affected device node is delegated into a container. In that case, the practical privilege level can still be closer to local unprivileged access inside that environment. In typical host deployments, access depends on device permissions such as `/dev/dri/renderD*` or similar accelerator nodes.

The path is configuration and hardware dependent. It requires Intel IOMMU SVA, a non-PRI-capable device, and a driver path that binds and unbinds SVA. Realistic impact is mostly warning noise or a crash only on `panic_on_warn` systems.

Call-site confidence: **high**, because the commit includes the relevant trace and the changed bind/unbind helper paths.

## 4. Severity interpretation

This behaves more like an **ordinary Moderate / Low-like** issue than an Important-class vulnerability. The bug is a state-management imbalance between IOPF setup and removal, not a demonstrated memory lifetime or corruption primitive.

The theoretical worst case is local DoS when WARN is fatal. Realistic exploitation evidence does not support privilege escalation, confidentiality impact, integrity impact, or arbitrary kernel memory corruption.

## 5. One-sentence report phrase

The Intel VT-d SVA unbind path may disable IOPF for a non-PRI device where IOPF was never enabled, allowing a local user with access to an affected SVA-capable device to trigger a kernel WARNING and possibly a DoS on panic_on_warn systems.

## 6. Manual review recommendation

**NO MANUAL CHECK NEEDED**

No concrete memory corruption, LPE path, network exposure, or security-boundary bypass is shown. This can be handled as a low-priority local DoS or WARN-only issue unless the product enables `panic_on_warn` and exposes affected SVA-capable devices to untrusted users.

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

Patch: iommu/vt-d: Avoid WARNING in sva unbind path
Commit: bb354384f40bb087e7c44f0f0743a23fc945f91d
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=bb354384f40bb087e7c44f0f0743a23fc945f91d

Commit description:

The Intel IOMMU driver allows SVA on devices even if they do not support
PCI/PRI. Commit 39c20c4 ("iommu/vt-d: Only handle IOPF for SVA when
PRI is supported") modified the SVA bind path to allow this configuration
by skipping IOPF enablement when PRI is missing. However, it failed to
update the unbind path.

This creates an imbalance: the unbind path attempts to disable IOPF for
a device that never had it enabled, triggering a WARNING in
intel_iommu_disable_iopf():

 WARNING: drivers/iommu/intel/iommu.c:3475 at intel_iommu_disable_iopf+0x4f/0x90d
 Call Trace:
  <TASK>
  blocking_domain_set_dev_pasid+0x50/0x70
  iommu_detach_device_pasid+0x89/0xc0
  iommu_sva_unbind_device+0x73/0x150
  xe_vm_close_and_put+0x4d2/0x1200 [xe]

Fix this by bypassing IOPF operations for SVA domains on non-PRI hardware
in both the bind and unbind paths.

Fixes: 39c20c4 ("iommu/vt-d: Only handle IOPF for SVA when PRI is supported")
Cc: stable@vger.kernel.org
Reported-by: Nareshkumar Gollakoti <naresh.kumar.g@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20260519052917.3729796-1-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed files:
  drivers/iommu/intel/iommu.c
  drivers/iommu/intel/iommu.h
  drivers/iommu/intel/svm.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=bb354384f40bb087e7c44f0f0743a23fc945f91d

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

The original Linux kernel CVE announcement for CVE-2026-64591 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64591

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:N/I:N/A:L
  The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
  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: 1
  Paranoid ActionableScore: 2

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: LOW
KPANIC detected for this report: NO
Published priority for this report (same as in Subject): LOW

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).