CVE-2025-37854

Summary

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.

Affected Software

VendorProductVersion RangeStatus
LinuxLinuxa4c63cafa58b4bd9e15511bab77a4752b93d3aa0 < 57c9dabda80ac167de8cd71231baae37cc2f442daffected
LinuxLinuxa4c63cafa58b4bd9e15511bab77a4752b93d3aa0 < 89af6b39f028c130d4362f57042927f005423e6aaffected
LinuxLinuxa4c63cafa58b4bd9e15511bab77a4752b93d3aa0 < ffd37d7d44d7e0b6e769d4fe6590e327f8cc3951affected
LinuxLinuxa4c63cafa58b4bd9e15511bab77a4752b93d3aa0 < 6f30a847432cae84c7428e9b684b3e3fa49b2391affected
LinuxLinuxa4c63cafa58b4bd9e15511bab77a4752b93d3aa0 < 9c4bcdf4068aae3e17e31c144300be405cfa03ffaffected
LinuxLinuxa4c63cafa58b4bd9e15511bab77a4752b93d3aa0 < f0b4440cdc1807bb6ec3dce0d6de81170803569baffected
LinuxLinux5.18affected
LinuxLinux0 < 5.18unaffected
LinuxLinux6.1.135 <= 6.1.*unaffected
LinuxLinux6.6.88 <= 6.6.*unaffected
LinuxLinux6.12.24 <= 6.12.*unaffected
LinuxLinux6.13.12 <= 6.13.*unaffected
LinuxLinux6.14.3 <= 6.14.*unaffected
LinuxLinux6.15 <= *unaffected

Weaknesses

ADP Enrichment

CVE Program Container

Additional References

References