Linux kernel amdgpu GEM/VM/command-submission ioctl surface (drm/amdgpu): MULTI-TENANT ISOLATION: Memory is handed

Impact

MULTI-TENANT ISOLATION: Memory is handed to a consumer without being initialised or cleared in the amdgpu GEM/VM/command-submission ioctl surface. Whatever the previous owner left behind is readable - and on a GPU node the previous owner is very often a different tenant's job. This is the classic residual-data leak between workloads sharing a card: model weights, activations, keys or tokens from the prior tenant can surface in a fresh allocation. Upstream fix: drm/amdgpu: zero-initialize GART table on allocation

Who can reach it

Local. Reachable by any process with /dev/dri/renderD* open - the render node is handed to tenant containers by every GPU device plugin, so this is unprivileged-tenant reachable. Not reachable over the network and not reachable from a container that has no GPU device node mapped in.

What to do

Kernel-side fix: this lands in mainline Linux and flows into distro kernels (RHEL/Rocky, Ubuntu HWE, SLES) and into AMD's out-of-tree DKMS amdgpu package shipped with ROCm. Patch the kernel or the DKMS module, then **reload the amdgpu module or reboot the node** - you cannot fix a running driver in place. Reloading amdgpu requires no process holding /dev/kfd or a render node, so in practice this is a cordon + drain + reboot per node. Plan it as a rolling maintenance across the fleet; there is no VBIOS flash, no SBIOS/AGESA step and no firmware update involved. Nodes running the ROCm DKMS stack often lag mainline by a release or two, so confirm the fix is actually present in the AMD driver version you deploy rather than assuming a new distro kernel covers it. Until the reboot window, the only real mitigation is to stop handing the render node to untrusted workloads - the device plugin has to be mapping /dev/dri/renderD* and /dev/kfd into the container for a tenant to reach this at all.

References