CVE-2026-68237
- Source
- https://cve.org/CVERecord?id=CVE-2026-68237
- Import Source
- https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-68237.json
- JSON Data
- https://api.osv.dev/v1/vulns/CVE-2026-68237
- Downstream
- Published
- 2026-08-10T12:01:02.144Z
- Modified
- 2026-08-12T04:19:22.140836618Z
- Summary
- drm/amdgpu/userq: fix indefinite fence wait during GPU reset
- Details
-
In the Linux kernel, the following vulnerability has been resolved:
drm/amdgpu/userq: fix indefinite fence wait during GPU reset
prereset only force-completes fences of MAPPED queues. A queue in any other state (e.g. mid-eviction) keeps its lastfence pending; after a GPU reset that fence never signals, so the eviction/suspend worker and process teardown (amdgpuevfmgrflushsuspend) wait on it forever and wedge the machine:
INFO: task kworker/6:28 blocked for more than 120 seconds. Workqueue: events amdgpuevictionfencesuspendworker [amdgpu] Call Trace: dmafencewaittimeout+0x7e/0x130 amdgpuuserqevict+0x67/0x140 [amdgpu] amdgpuevictionfencesuspendworker+0xd8/0x160 [amdgpu] processscheduled_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)
- Database specific
{ "osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/68xxx/CVE-2026-68237.json", "cna_assigner": "Linux" }- References
-
- https://git.kernel.org/stable/c/3085ae8695e025b39d208f288c6265edc75abbe8
- https://git.kernel.org/stable/c/5d75ec2e5f1736c2f10c7d6f4565bf1bf29f29a7
- https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/68xxx/CVE-2026-68237.json
- https://nvd.nist.gov/vuln/detail/CVE-2026-68237
- https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git