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

drm/amdgpu: check ras and obj before dereference

nbio v7 9 handle ras controller intr no bifring() dereferences ras and obj without checking either for NULL. Both amdgpu ras get context() and amdgpu ras find obj() can return NULL, e.g. during the window between adev->nbio.ras being set (early in amdgpu ras init(), by design, to enable the fatal-error interrupt as soon as possible) and the PCIE BIF ras object actually being created in RAS late init. Any interrupt in that window crashes in hard-IRQ context.

This is analogous to commit d190b459b2a4 ("drm/amdgpu: the warning dereferencing obj for nbio v7 4"), which fixed the same issue in the nbio v7 4 handler.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

(cherry picked from commit c7071767a50a32ed727cf800ac84372429e3b4b3)