[CVE-2026-64582][IMPORTANT] RDMA/rxe: Fix a use-after-free problem in rxe_mmap [ Upstream

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

CVE: CVE-2026-64582
Priority: IMPORTANT
AL-KERNEL base severity: IMPORTANT
KPANIC flag: YES
Patch: RDMA/rxe: Fix a use-after-free problem in rxe_mmap [ Upstream
Commit: 665fb7d22a700c66a78db0cf88c6e6a649aba9d0
Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=665fb7d22a700c66a78db0cf88c6e6a649aba9d0
Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64582
Analysis date: Wed, 05 Aug 2026 07:51:37 -0400
ActionableScore: 7
ActionableScore lower bound: 5
Actionable bucket: Strong Important candidate / Actionable Moderate at minimum
Manual review required: YES

Summary:
A race in RDMA rxe mmap can allow a concurrent destroy ioctl to free the backing vmalloc object while it is being remapped, causing a kernel crash and potentially a page-level UAF through a stale user VMA mapping.

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

The original Linux kernel CVE announcement for CVE-2026-64582 is available here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64582

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: IMPORTANT
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-2026-64582	IMPORTANT	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;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7';DESCR 'A use-after-free race exists in RDMA rxe_mmap because rxe_mmap_info can be removed from the pending list and used by remap_vmalloc_range while its last reference may be dropped by a concurrent DESTROY_CQ ioctl. If the destroy path frees the vmalloc object while remap_vmalloc_range is walking PTEs, the result can be a kernel crash or a page level UAF where a user VMA may retain a mapping to freed memory. For the CVSS the PR:L is used because triggering requires local access to the RDMA uverbs rxe device interface, but not necessarily full administrator privileges when the device node is available to a local user. The issue is not directly network reachable because the vulnerable operations are mmap and ioctl on a local RDMA userspace interface. Impact is at least a local denial of service via kernel crash and in the paranoid case may allow confidentiality or integrity impact due to memory corruption, so it should be reviewed manually.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 7)	YES	READ OOB LOCK DANGER NULLPTR RACE UAF NETWORK KERNEL_PANIC_PLUS_UAF HARDWARE KPANIC  KPANIC INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN7	NO	NO	guess

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

ActionableScore=7
ActionableScoreLower=5

## 1. ActionableScore

* Conservative score: 5
* Paranoid score: 7
* Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**::

## 2. Signal breakdown

Conservative signals:

* Local unprivileged trigger: +1. The trace shows UID 1000 and the path is through `ib_uverbs_mmap()` plus a concurrent destroy ioctl, so a local user with access to the RDMA rxe uverbs device can plausibly trigger it.
* Generic memory corruption, strong corruption primitive: +2. The bug is an explicit UAF race involving `rxe_mmap_info`, `ip->obj`, `vfree()`, and a possible stale PTE retained in a user VMA.
* Real lifetime corruption: +1. The fix adds `kref_get_unless_zero()` under `pending_lock`, which directly confirms a refcount/lifetime bug.
* Reliable kernel crash / strong DoS: +1. The report includes a concrete GPF/KASAN crash in `validate_page_before_insert()` via `vm_insert_page(NULL)`.
* Local rich device object API with lifetime/page ownership influence: +1. RDMA uverbs is not automatically high-control, but here the patch shows user-driven mmap/ioctl interaction with refcounts and page mapping ownership.
* Hard or unreliable race / special timing required: -1. The commit explicitly says the page-level UAF window is statistically very difficult to hit.
* Rare/configuration-dependent exposure: -1. Soft RoCE / rxe and RDMA uverbs access are not universal default attack surfaces.

Paranoid additions / changes:

* Privilege escalation plausible: +2. The page-level UAF scenario is explicitly described: a user VMA may retain a PTE to freed memory that can later be reallocated by vmalloc. This is stronger than a pure NULL dereference, although exploitation is difficult.
* Confidentiality impact plausible: +1. A stale user mapping to freed/reallocated pages can plausibly expose data depending on reuse.
* Integrity impact plausible: +1. A stale writable mapping to a reused page can plausibly corrupt later kernel/vmalloc-backed data depending on mapping permissions and reuse.

## 3. Reachability analysis

The bug is local, not network reachable. Triggering requires access to the RDMA rxe userspace interface and the ability to race `mmap()` of a pending mmap object against a concurrent `DESTROY_CQ` ioctl. This does not require full host root in all deployments because RDMA device nodes may be exposed to selected local users or workloads, so PR:L is the practical interpretation.

Namespaces and containers matter if RDMA/uverbs devices are passed into a container or delegated to a service account. In a locked-down system with no rxe module loaded and no RDMA device node access, exposure is much lower. In environments that expose RDMA devices to non-root users, the bug is actionable because the primitive crosses from a local device API into kernel page lifetime and user VMA mappings.

Call-site confidence: high. The provided patch, commit message, and stack trace show the relevant path from `ib_uverbs_mmap()` to `rxe_mmap()` to `remap_vmalloc_range()` and the concurrent free path through `kref_put()` / `rxe_mmap_release()`.

## 4. Severity interpretation

This is not an ordinary Moderate cleanup issue. The conservative view is Actionable Moderate because there is a demonstrated local kernel crash and a confirmed lifetime bug in a mmap path. The paranoid view reaches Strong Important candidate because the commit explicitly describes a page-level UAF where a user VMA may retain a mapping to freed memory.

Realistic exploitation beyond DoS is difficult due to the short race window and the need for favorable vmalloc page reuse. However, the primitive is stronger than a simple NULL dereference and directly involves page ownership/lifetime, so it should not be auto-closed.

## 5. One-sentence report phrase

A race in RDMA rxe mmap can allow a concurrent destroy ioctl to free the backing vmalloc object while it is being remapped, causing a kernel crash and potentially a page-level UAF through a stale user VMA mapping.

## 6. Manual review recommendation

MANUAL CHECK REQUIRED

Reason: the issue is an explicit UAF/lifetime race with a demonstrated crash and a technically plausible page-level UAF primitive. The exploitability is timing-sensitive, but the memory lifetime and page mapping consequences are serious enough for manual kernel security review.

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

Patch: RDMA/rxe: Fix a use-after-free problem in rxe_mmap [ Upstream
Commit: 665fb7d22a700c66a78db0cf88c6e6a649aba9d0
Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=665fb7d22a700c66a78db0cf88c6e6a649aba9d0

Changed files:
  drivers/infiniband/sw/rxe/rxe_mmap.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=665fb7d22a700c66a78db0cf88c6e6a649aba9d0

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

The original Linux kernel CVE announcement for CVE-2026-64582 can be found here:
https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64582

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: 5
  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: IMPORTANT
KPANIC detected for this report: YES
Published priority for this report (same as in Subject): IMPORTANT

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).