First published: Fri May 09 2025(Updated: )
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.
Credit: 416baaa9-dc9f-4396-8d5f-8c081fb06d67
Affected Software | Affected Version | How to fix |
---|---|---|
Linux kernel |
Sign up to SecAlerts for real-time vulnerability data matched to your software, aggregated from hundreds of sources.
CVE-2025-37854 has been evaluated and categorized based on its potential impact on system security.
To mitigate CVE-2025-37854, users should apply the latest updates and patches provided by their Linux distribution.
CVE-2025-37854 affects the Linux kernel, specifically those versions that utilize the drm/amdkfd module.
CVE-2025-37854 could lead to instability and potential crashes in graphics processing related to kernel functionalities.
Yes, CVE-2025-37854 has been resolved through updates to the Linux kernel to address the identified issues.