Skip to main content

Linux Kernel EUVDEUVD-2026-55570

| CVE-2026-68384 HIGH
2026-08-10 Linux GHSA-hr8j-ghcq-qgrr
7.8
CVSS 3.1 · Vendor: Linux
Share

Severity by source

Vendor (Linux) PRIMARY
7.8 HIGH
AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
vuln.today AI
7.0 HIGH

Race condition requires VF migration to coincide with active BO copy, raising AC to H versus the NVD AC:L; all other metrics match the provided vector.

3.1 AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
4.0 AV:L/AC:H/AT:P/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
SUSE
5.5 MEDIUM
AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
Red Hat
5.5 MEDIUM
qualitative

Primary rating from Vendor (Linux).

CVSS VectorVendor: Linux

Attack Vector
Local
Attack Complexity
Low
Privileges Required
Low
User Interaction
None
Scope
Unchanged
Confidentiality
High
Integrity
High
Availability
High

Lifecycle Timeline

5
Analysis Generated
Aug 14, 2026 - 01:42 vuln.today
CVSS changed
Aug 13, 2026 - 23:37 NVD
7.8 (HIGH)
Patch available
Aug 10, 2026 - 14:18 EUVD
CVE Published
Aug 10, 2026 - 12:04 cve.org
HIGH 7.8
CVE Published
Aug 10, 2026 - 12:04 cve.org
UNKNOWN (no severity yet)

DescriptionCVE.org

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

drm/xe/vf: Fix VF CCS attach/detach race with in-flight BO moves

xe_bo_move() attaches VF CCS read/write batch buffers (BBs) to a BO after it transitions NULL/SYSTEM -> TT, and detaches them after it transitions TT -> SYSTEM. Both operations were done synchronously on the CPU immediately after building the move's copy/clear fence, without waiting for that fence to signal. This creates two races with VF migration:

  • Attach happens too late relative to the copy job it is meant to

protect. If the copy job is submitted before the CCS BBs are attached, a VF migration event that pauses execution mid-copy can observe partially copied CCS metadata without the attach state needed to correctly save/restore it.

  • Detach happens too early relative to the copy job that moves data

out of TT. The CCS BBs are torn down right after the copy fence is obtained, while the actual blit may still be in flight. A VF migration event that pauses execution mid-copy can then race the save/restore path against the still-running blit, and the CCS BBs it would need to make sense of the paused state have already been removed.

Fix both races:

  • Move the attach call to before the copy/clear job is submitted, so

the CCS BBs are already registered by the time the copy runs. On attach failure, unwind and bail out of the move. xe_migrate_ccs_rw_copy() now takes the destination resource explicitly, since bo->ttm.resource is not updated to the new resource until after the move commits.

  • Detach only after explicitly waiting for the copy fence to signal,

instead of tearing down the CCS BBs immediately after obtaining it.

While here, also fix xe_sriov_vf_ccs_attach_bo() to properly unwind and propagate errors: the per-context loop previously never broke out on error, silently discarding earlier failures. Unwind by clearing each attached context directly via xe_migrate_ccs_rw_copy_clear() instead of reusing xe_sriov_vf_ccs_detach_bo(), which requires both contexts to be attached before it will clean up either one.

(cherry picked from commit d45ad0aa7a1eb5d7288b5ed948b05695611dc39e)

AnalysisAI

Race conditions in the Linux kernel's drm/xe Virtual Function CCS state management expose Intel Xe GPU SR-IOV environments to local privilege escalation, information disclosure, and memory corruption. The driver's buffer object move operations improperly sequence CCS batch buffer attach and detach relative to in-flight copy jobs, creating two distinct timing windows where a VF migration event can observe partially-copied or already-torn-down CCS metadata. No CISA KEV listing or public proof-of-concept exists at time of analysis, and EPSS of 0.18% (8th percentile) reflects negligible observed exploitation activity; the primary exposure surface is cloud and VDI infrastructure using Intel Xe GPUs with SR-IOV enabled.

Technical ContextAI

The vulnerability resides in the drm/xe kernel driver, which manages Intel Xe GPU hardware (Arc discrete, Data Center GPU Flex/Max series, and integrated Xe graphics in Lunar Lake/Arrow Lake). SR-IOV Virtual Functions (VFs) share physical GPU resources across guest VMs, and Color Compression Surface (CCS) batch buffers (BBs) track metadata about compressed GPU memory regions. During buffer object (BO) migrations between memory tiers (NULL/SYSTEM → TT and TT → SYSTEM), these CCS BBs must be registered before the copy job is submitted and unregistered only after the copy fence signals completion. The reported bug is a synchronization defect analogous to CWE-362 (Concurrent Execution Using Shared Resource with Improper Synchronization): the attach call was placed after xe_migrate_ccs_rw_copy() submitted its job, and the detach call was placed immediately after obtaining the fence, before the DMA blit finished. A VF migration event landing in either window can observe inconsistent CCS state - partially-copied metadata with no attach record, or a live blit racing against a save/restore path that already lost its CCS BB references. A secondary bug in xe_sriov_vf_ccs_attach_bo() also silently discarded per-context errors due to a missing break on the error path. The affected CPE is cpe:2.3:a:linux:linux:*:*:*:*:*:*:*:*.

RemediationAI

Upgrade to Linux 6.18.42, 7.1.6, or 7.2-rc4 or later, which contain the corrected xe_bo_move() synchronization logic and the xe_sriov_vf_ccs_attach_bo() error-unwind fix. Stable-tree patches are available at https://git.kernel.org/stable/c/35ba43b541117bfb595e4b807ba447cf4335cc7d (6.18 branch), https://git.kernel.org/stable/c/56441f9e08ad68697295b8835266d2backend447cf4335cc7d (7.1 branch), and https://git.kernel.org/stable/c/f2ebfd5cc87f1393a30c8b8b0a6c20cb22cffa97 (mainline/7.2 branch); the NVD advisory is at https://nvd.nist.gov/vuln/detail/CVE-2026-68384. For environments where a live kernel upgrade is operationally blocked, disabling SR-IOV VF support for Intel Xe GPUs (removing the xe.enable_sriov kernel parameter or equivalent module option) eliminates the vulnerable code path entirely, at the cost of GPU pass-through and virtualization functionality for guest workloads. Restricting GPU device access via cgroups v2 device controllers or SELinux/AppArmor policy to only trusted, privileged processes reduces the pool of callers that can initiate BO move operations, shrinking the attack surface without disabling SR-IOV. Distribution vendors (Red Hat, SUSE, Ubuntu, Debian) shipping kernels in the affected range should be monitored for backport availability.

Vendor StatusVendor

SUSE

Severity: Moderate
Product Status
SUSE Linux Enterprise Desktop 15 SP7 Not-Affected
SUSE Linux Enterprise Desktop 15 SP7 Not-Affected
SUSE Linux Enterprise High Availability Extension 15 SP7 Not-Affected
SUSE Linux Enterprise High Availability Extension 15 SP7 Not-Affected
SUSE Linux Enterprise High Availability Extension 16.0 Not-Affected

Share

EUVD-2026-55570 vulnerability details – vuln.today

This site uses cookies essential for authentication and security. No tracking or analytics cookies are used. Privacy Policy