Skip to main content

Linux Kernel CVE-2026-68305

| EUVDEUVD-2026-55406 HIGH
2026-08-10 Linux GHSA-qfm9-4w8r-pq56
High
Disputed · 7.8 Vendor: Linux
Share

Severity by source

Sources disagree (Low–High)
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

AC:H reflects the timing-dependent race between device unbind and BO destruction; PR:L for required local GPU device access; C/I/A:H for kernel-level NULL deref potential.

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 LOW
qualitative

vuln.today treats the vendor’s rating as authoritative. A higher third-party CVSS (e.g. CISA-ADP) is shown for transparency but does not drive the headline severity.

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 - 03:13 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 - 12:02 cve.org
HIGH 7.8
CVE Published
Aug 10, 2026 - 12:02 cve.org
UNKNOWN (no severity yet)

DescriptionCVE.org

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

drm/xe/vf: Add drm_dev guards when detaching CCS read/write buffers

CCS read/write buffers are freed during BO destruction. In some cases, BOs may be destroyed after the device is unbound but while the DRM structure remains valid, leading to NULL pointer dereferences when accessing device resources.

BUG: kernel NULL pointer dereference, address: 0000000000000000 PGD 0 P4D 0 Oops: Oops: 0000 [#1] SMP NOPTI CPU: 0 UID: 0 PID: 9376 Comm: xe_pat Not tainted 7.2.0-rc2+ #1 PREEMPT(lazy) RIP: 0010:xe_sriov_vf_ccs_rw_update_bb_addr+0x4d/0xa0 [xe] RSP: 0018:ffffcf304110b9c8 EFLAGS: 00010246 RAX: ffff8a85c38a0a00 RBX: 00000000810ef000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff8a85c39c1888 RBP: ffffcf304110b9e8 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: ffff8a85c39c1888 R13: 0000000000000000 R14: ffff8a85c39b4f28 R15: ffff8a85c3885000 FS: 0000000000000000(0000) GS:ffff8a878b809000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 000000010314a002 CR4: 0000000000772ef0 PKRU: 55555554 Call Trace: <TASK> xe_migrate_ccs_rw_copy_clear+0x98/0x120 [xe] xe_sriov_vf_ccs_detach_bo+0x2c/0x60 [xe] xe_ttm_bo_delete_mem_notify+0xc8/0xe0 [xe] ttm_bo_cleanup_memtype_use+0x26/0x80 [ttm] ttm_bo_release+0x29e/0x2d0 [ttm] ttm_bo_fini+0x39/0x70 [ttm] xe_gem_object_free+0x1f/0x30 [xe] drm_gem_object_free+0x1d/0x40 ttm_bo_vm_close+0x5f/0x90 [ttm] remove_vma+0x2c/0x70 tear_down_vmas+0x63/0xf0 exit_mmap+0x20d/0x3f0 __mmput+0x45/0x170 mmput+0x31/0x40 do_exit+0x2ba/0xac0 do_group_exit+0x2d/0xb0 __x64_sys_exit_group+0x18/0x20 x64_sys_call+0x14a0/0x2390 do_syscall_64+0xdd/0x640 ? count_memcg_events+0xea/0x240 ? handle_mm_fault+0x1ec/0x2f0

(cherry picked from commit 1ae415a6eefe5004954a1d352b1718faca8844ef)

AnalysisAI

NULL pointer dereference in the Linux kernel's Intel Xe GPU SR-IOV virtual function (VF) driver allows a local user to crash the kernel during buffer object (BO) teardown. The flaw is triggered when CCS (Color Compression Surface) read/write buffers are freed after the device is unbound but while the DRM structure remains live, causing xe_sriov_vf_ccs_rw_update_bb_addr to dereference a NULL device pointer during process exit. CVSS rates this 7.8 (C:H/I:H/A:H) reflecting worst-case kernel-level impact, though practical exploitation is constrained by hardware and configuration prerequisites; EPSS is 0.16% (6th percentile) and no active exploitation is confirmed.

Technical ContextAI

The vulnerability resides in the drm/xe subsystem - the Linux upstream driver for Intel Xe architecture GPUs (Arc and Data Center GPU Flex/Max series). SR-IOV (Single Root I/O Virtualization) allows a physical GPU to present multiple virtual functions to VMs or containers. CCS (Color Compression Surface) buffers are GPU memory objects managed via TTM (Translation Table Maps), the kernel's GPU memory manager. During process exit, the kernel tears down VMAs, which triggers TTM BO release via ttm_bo_release → ttm_bo_fini → xe_gem_object_free, ultimately calling xe_sriov_vf_ccs_detach_bo. If the physical device has already been unbound at this point, the device pointer accessed inside xe_sriov_vf_ccs_rw_update_bb_addr is NULL, causing an Oops. The fix adds drm_dev guards to gate the detach path. The root cause class is CWE-476 (NULL Pointer Dereference), though NVD has not formally assigned a CWE. The CPE is cpe:2.3:a:linux:linux:*:*:*:*:*:*:*:*.

RemediationAI

Apply the upstream stable kernel patches available at https://git.kernel.org/stable/c/523ed2831ee55b2a1edabdea96781651f9df9685 (targeting the 7.2 stable branch) and https://git.kernel.org/stable/c/4c92afb4c143526d340545ca581e88e6952ea511 (targeting the 7.1.x branch, confirmed fix for 7.1.6). Distributions shipping kernel 6.18 or 7.x with the Xe driver should await or request backport patches. As a compensating control where patching is not immediately feasible, disabling SR-IOV VF allocation on Intel Xe GPU hardware eliminates the vulnerable code path entirely - note that this removes VM/container GPU sharing capability. Systems without Intel Xe GPU hardware require no action.

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

CVE-2026-68305 vulnerability details – vuln.today

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