CVE-2025-37854
drm/amdkfd: Fix mode1 reset crash issue
We use these services and cookies to improve your user experience. You may opt out if you wish, however, this may limit some features on this site.
Please see our statement on Data Privacy.
drm/amdkfd: Fix mode1 reset crash issue
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.
Reserved 2025-04-16 | Published 2025-05-09 | Updated 2025-05-09 | Assigner Linuxgit.kernel.org/...c/57c9dabda80ac167de8cd71231baae37cc2f442d
git.kernel.org/...c/89af6b39f028c130d4362f57042927f005423e6a
git.kernel.org/...c/ffd37d7d44d7e0b6e769d4fe6590e327f8cc3951
git.kernel.org/...c/6f30a847432cae84c7428e9b684b3e3fa49b2391
git.kernel.org/...c/9c4bcdf4068aae3e17e31c144300be405cfa03ff
git.kernel.org/...c/f0b4440cdc1807bb6ec3dce0d6de81170803569b
Support options
