CVE Catalog

CVE-2026-68237

Low risk· EPSS 5%
Published: Updated: Translated: NVD NIST

Exploitation Probability (EPSS)

Low risk
0.16%

5th percentile - higher than 5% of all known CVEs

Summary

In the Linux kernel, the amdgpu driver has a bug where not all pending fences are force-completed during GPU reset for queues not in MAPPED state. This causes processes waiting on the fence to hang, potentially freezing the system.

Risk Assessment

The risk is a potential system hang during GPU reset, leading to service unavailability and requiring manual reboot.

Recommendation

Update the Linux kernel to a version containing the fix that force-completes fences for all queues regardless of state.

Other vulnerabilities in Linux kernel

See all
Original NVD description (English source)

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/userq: fix indefinite fence wait during GPU reset pre_reset only force-completes fences of MAPPED queues. A queue in any other state (e.g. mid-eviction) keeps its last_fence pending; after a GPU reset that fence never signals, so the eviction/suspend worker and process teardown (amdgpu_evf_mgr_flush_suspend) wait on it forever and wedge the machine: INFO: task kworker/6:28 blocked for more than 120 seconds. Workqueue: events amdgpu_eviction_fence_suspend_worker [amdgpu] Call Trace: dma_fence_wait_timeout+0x7e/0x130 amdgpu_userq_evict+0x67/0x140 [amdgpu] amdgpu_eviction_fence_suspend_worker+0xd8/0x160 [amdgpu] process_scheduled_works+0xa6/0x420 Force-complete every queue's fence regardless of state. The unmap and mark-hung step stays gated on MAPPED, since unmapping a queue that is not mapped is invalid. (cherry picked from commit 9102b39fa924dcc3dc75a3137bfa9633c40b88c0)

Vulnerability data from NVD (NIST) · CISA KEV · EPSS