drm/amdgpu: disallow multiple FENCE chunks in one submit
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
≥ d38ceaf99ed015f2a0b9af3499791bd3a3daae21 · < 931cd1d1baeae68e8eb2c23bc1f3d8934dca6241| No fix yet |
| Linux Linux |
|
|
In the Linux kernel, the following vulnerability has been resolved:
drm/amdgpu: disallow multiple FENCE chunks in one submit
amdgpu_cs_pass1() dispatches on chunk_id once per chunk without rejecting repeated ids. p->uf_bo is a single-slot field, so a submission carrying two AMDGPU_CHUNK_ID_FENCE chunks runs amdgpu_cs_p1_user_fence() twice, and the second run overwrites p->uf_bo with a freshly referenced BO without dropping the reference taken by the first.
amdgpu_cs_parser_fini() only unrefs the final p->uf_bo, so every FENCE chunk but the last leaks a BO reference. The leaked BO outlives handle close and process exit.
Reject duplicate FENCE chunks the same way commit fec5f8e8c6bc ("drm/amdgpu: disallow multiple BO_HANDLES chunks in one submit") did for p->bo_list.
(cherry picked from commit 665b1fc2a1845206408f9a2c6da67101789edb82)
From the CVE record · Linux
The Linux kernel’s amdgpu command-submission parser accepts multiple FENCE chunks in one submission. A local user able to submit GPU commands can cause buffer-object references to leak; those references persist after handle close and process exit. No separate user interaction is described.
Engineers running Linux systems with amdgpu should check their kernel versions; affected versions are not specified.