CVE-2026-68237
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.
Does this matter?
Not yet scored. NVD analysis is pending; check back once CVSS and EPSS values are published.
Description
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)
- CVSS
- Not yet scored
- EPSS
- 0.15% probability · 5th percentile
- CISA KEV
- Not listed
- Source
- 416baaa9-dc9f-4396-8d5f-8c081fb06d67
References
Source: NVD record, EPSS from FIRST.org, KEV from CISA. Refreshed daily.