Description
In the Linux kernel, the following vulnerability has been resolved:

vhost_iotlb: bound map allocation in add_range

vhost_iotlb_add_range_ctx() only retires an old entry when the table
has a non-zero limit, has exactly reached that limit and has
VHOST_IOTLB_FLAG_RETIRE set. Non-retiring tables can keep allocating
entries after reaching their configured limit.

Existing vhost devices allocate their IOTLB with max_iotlb_entries from
vhost.c, which defaults to 2048 and is tunable by module parameter. Use
the caller-provided limit at the allocation point instead of adding a
separate default in the common IOTLB helper, and reject non-positive
values in vhost paths that can report an error.

Other vhost IOTLB users should not create zero-limit tables when entries
can be populated from userspace or guest-controlled requests. Add
caller-side max_iotlb_entries parameters for mlx5 vDPA, VDUSE and
vhost-vDPA. Reject non-positive VDUSE and vhost-vDPA values, and require
at least two entries for vdpa_sim and mlx5 vDPA paths that install
full-range mappings, since those mappings are split into two IOTLB
entries.

Handle full-range mappings in the common helper by checking that the
IOTLB can hold both split entries before inserting the first half. This
avoids returning an error after leaving a half mapping behind.

When the table is full, keep the existing retire behavior for retiring
tables and return -ENOSPC for non-retiring tables. Reuse the retired map
node instead of freeing it and allocating a replacement, so a stream of
IOTLB updates cannot keep forcing GFP_ATOMIC allocations after the table
has reached its limit. If a zero-limit IOTLB still reaches the common
helper, treat it as a configuration error and return -EINVAL.

I found this bug myself, though the patch was written with AI assistance.
Published: 2026-08-22
Score: n/a
EPSS: n/a
KEV: No
Impact: n/a
Action: n/a
AI Analysis

Impact

The vhost IOTLB subsystem permits allocation of map entries beyond the configured limit when the table is not set to retire entries. The add_range function therefore allows non‑retiring tables to continue consuming kernel memory after reaching their maximum capacity. This unchecked growth can exhaust kernel resources, potentially causing a denial‑of‑service, out‑of‑memory failures, or kernel panics. The flaw is specific to the IOTLB allocator and does not provide a direct code‑execution path but can degrade system availability.

Affected Systems

All Linux kernel builds that incorporate the vhost IOTLB subsystem are affected, including standard VM and container driver modules. No version range is specified in the advisory, so any release before the official patch is vulnerable.

Risk and Exploitability

No CVSS or EPSS score is published for this entry, and it is not listed in CISA KEV. The vulnerability is restricted to contexts that load or interact with vhost IOTLB devices. An attacker with local or privileged access to the kernel runtime that can influence the configuration of a vhost device could trigger the over‑allocation. Remote exploitation would require compromise of the guest or host to manipulate vhost parameters, making the attack surface limited. The primary impact is a potential denial of service via resource exhaustion.

Generated by OpenCVE AI on August 22, 2026 at 18:01 UTC.

Remediation

No vendor fix or workaround currently provided.

OpenCVE Recommended Actions

  • Upgrade the Linux kernel to the latest stable release that includes the vhost IOTLB fix.
  • If an upgrade is not immediately possible, reduce the maximum IOTLB size by setting the module parameter max_iotlb_entries to a small positive value or disable non‑retiring tables in the driver configuration.
  • Validate all custom vhost drivers to ensure they reject non‑positive or zero limits and enforce at least two entries for full‑range mappings, monitoring kernel logs for allocation failures or crashes.

Generated by OpenCVE AI on August 22, 2026 at 18:01 UTC.

Tracking

Sign in to view the affected projects.

Advisories

No advisories yet.

History

Sat, 22 Aug 2026 18:30:00 +0000

Type Values Removed Values Added
Weaknesses CWE-400

Sat, 22 Aug 2026 15:45:00 +0000

Type Values Removed Values Added
Description In the Linux kernel, the following vulnerability has been resolved: vhost_iotlb: bound map allocation in add_range vhost_iotlb_add_range_ctx() only retires an old entry when the table has a non-zero limit, has exactly reached that limit and has VHOST_IOTLB_FLAG_RETIRE set. Non-retiring tables can keep allocating entries after reaching their configured limit. Existing vhost devices allocate their IOTLB with max_iotlb_entries from vhost.c, which defaults to 2048 and is tunable by module parameter. Use the caller-provided limit at the allocation point instead of adding a separate default in the common IOTLB helper, and reject non-positive values in vhost paths that can report an error. Other vhost IOTLB users should not create zero-limit tables when entries can be populated from userspace or guest-controlled requests. Add caller-side max_iotlb_entries parameters for mlx5 vDPA, VDUSE and vhost-vDPA. Reject non-positive VDUSE and vhost-vDPA values, and require at least two entries for vdpa_sim and mlx5 vDPA paths that install full-range mappings, since those mappings are split into two IOTLB entries. Handle full-range mappings in the common helper by checking that the IOTLB can hold both split entries before inserting the first half. This avoids returning an error after leaving a half mapping behind. When the table is full, keep the existing retire behavior for retiring tables and return -ENOSPC for non-retiring tables. Reuse the retired map node instead of freeing it and allocating a replacement, so a stream of IOTLB updates cannot keep forcing GFP_ATOMIC allocations after the table has reached its limit. If a zero-limit IOTLB still reaches the common helper, treat it as a configuration error and return -EINVAL. I found this bug myself, though the patch was written with AI assistance.
Title vhost_iotlb: bound map allocation in add_range
First Time appeared Linux
Linux linux Kernel
CPEs cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*
Vendors & Products Linux
Linux linux Kernel
References

Subscriptions

Linux Linux Kernel
cve-icon MITRE

Status: PUBLISHED

Assigner: Linux

Published:

Updated: 2026-08-22T15:33:09.028Z

Reserved: 2026-08-15T05:44:03.928Z

Link: CVE-2026-74713

cve-icon Vulnrichment

No data.

cve-icon NVD

Status : Received

Published: 2026-08-22T16:16:46.063

Modified: 2026-08-22T16:16:46.063

Link: CVE-2026-74713

cve-icon Redhat

No data.

cve-icon OpenCVE Enrichment

Updated: 2026-08-22T18:15:03Z

Weaknesses
  • CWE-400

    Uncontrolled Resource Consumption