CVE-2025-38521

In the Linux kernel, the following vulnerability has been resolved:

drm/imagination: Fix kernel crash when hard resetting the GPU

The GPU hard reset sequence calls pmruntimeforcesuspend() and pmruntimeforceresume(), which according to their documentation should only be used during system-wide PM transitions to sleep states.

The main issue though is that depending on some internal runtime PM state as seen by pmruntimeforcesuspend() (whether the usage count is <= 1), pmruntimeforceresume() might not resume the device unless needed. If that happens, the runtime PM resume callback pvrpowerdevice_resume() is not called, the GPU clocks are not re-enabled, and the kernel crashes on the next attempt to access GPU registers as part of the power-on sequence.

Replace calls to pmruntimeforcesuspend() and pmruntimeforceresume() with direct calls to the driver's runtime PM callbacks, pvrpowerdevicesuspend() and pvrpowerdeviceresume(), to ensure clocks are re-enabled and avoid the kernel crash.