CVE-2025-37854
drm/amdkfd: Fix mode1 reset crash issue
- State
- Published
- Severity
- 7.8 High
- Published
- May 9, 2025
- Updated
- Aug 5, 2026
- Reserved
- Apr 16, 2025
Description
In the Linux kernel, the following vulnerability has been resolved:
drm/amdkfd: Fix mode1 reset crash issue
If HW scheduler hangs and mode1 reset is used to recover GPU, KFD signal user space to abort the processes. After process abort exit, user queues still use the GPU to access system memory before h/w is reset while KFD cleanup worker free system memory and free VRAM.
There is use-after-free race bug that KFD allocate and reuse the freed system memory, and user queue write to the same system memory to corrupt the data structure and cause driver crash.
To fix this race, KFD cleanup worker terminate user queues, then flush reset_domain wq to wait for any GPU ongoing reset complete, and then free outstanding BOs.
Scores
| CVSS | Score | Vector | Source |
|---|---|---|---|
| 3.1 | 7.8 High | CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H | Linux |
Weaknesses
The record names no weakness.
Affected Products
| Product | Versions |
|---|---|
LinuxLinux |
|
LinuxLinux |
|
References
- https://git.kernel.org/stable/c/57c9dabda80ac167de8cd71231baae37cc2f442d
- https://git.kernel.org/stable/c/89af6b39f028c130d4362f57042927f005423e6a
- https://git.kernel.org/stable/c/ffd37d7d44d7e0b6e769d4fe6590e327f8cc3951
- https://git.kernel.org/stable/c/6f30a847432cae84c7428e9b684b3e3fa49b2391
- https://git.kernel.org/stable/c/9c4bcdf4068aae3e17e31c144300be405cfa03ff
- https://git.kernel.org/stable/c/f0b4440cdc1807bb6ec3dce0d6de81170803569b
- https://lists.debian.org/debian-lts-announce/2025/05/msg00045.html
Every record here comes from the CVE Program’s CVE List. © 1999–2026 The MITRE Corporation. CVE is a registered trademark of The MITRE Corporation. The records are reproduced under the CVE Program terms of use.