Bug 2513332 (CVE-2026-68237)

Summary: CVE-2026-68237 kernel: drm/amdgpu/userq: fix indefinite fence wait during GPU reset
Product: [Other] Security Response Reporter: OSIDB Bzimport <bzimport>
Component: vulnerability-draftAssignee: Product Security <prodsec-ir-bot>
Status: NEW --- QA Contact:
Severity: medium Docs Contact:
Priority: medium    
Version: unspecifiedCC: rhel-process-autobot, watson-tool-maintainers
Target Milestone: ---Keywords: Security
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: ---
Doc Text:
A flaw was found in the Linux kernel's `drm/amdgpu/userq` component. During a GPU reset, the `pre_reset` function only force-completes fences for mapped queues. If a queue is in another state, its last fence remains pending and never signals after the reset. This can cause the eviction/suspend worker and process teardown to wait indefinitely, leading to a system hang and a Denial of Service.
 Story Points: ---
Clone Of: Environment:
Last Closed: Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description OSIDB Bzimport 2026-08-10 12:31:27 UTC 
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)