Linux kernel CTR_DRBG: generate can report success while leaving the output buffer uninitialized

Impact

drbg_ctr_generate() can return success on a path where it actually failed, so the caller consumes an output buffer that was never filled. Any kernel consumer of the CTR_DRBG - including FIPS-mode configurations where the DRBG is the sanctioned RNG - can therefore receive data it believes is random but is not, with no error to key off. The record does not identify a specific caller or quantify predictability, so treat this as a correctness and assurance problem for kernel-generated key material rather than a demonstrated key-recovery path. It matters on a GPU fleet because the same host kernel image is usually rolled across every node, so a build with this defect is fleet-wide.

Who can reach it

No remote attack path is described in the record. The failure is triggered by the DRBG's own error handling inside the kernel; a local user can reach the CTR_DRBG through the kernel crypto API, but the record does not establish attacker control over the failure condition.

What to do

Take the stable kernel update carrying the fix and reboot each node; there is no runtime toggle for this. On a GPU fleet that means the usual rolling drain-and-reboot, one node at a time, and hosts in FIPS mode should be prioritised because the CTR_DRBG is the approved generator there. Distribution kernel-livepatch coverage for this commit is not stated in the record.

References

Related entries

• Linux kernel chacha20poly1305 template: missing argument check dereferences an error pointerCVE-2026-64314 · Linux kernel crypto chacha20poly1305 template (chachapoly_create)Medium
• Linux kernel BPF verifier: map-in-map lookup nullness elided using the wrong max_entriesCVE-2026-64353 · Linux kernel BPF verifier (ARRAY_OF_MAPS with BPF_F_INNER_MAP inner arrays)Medium
• Linux kernel (drivers/vfio/pci): The disable_idle_d3 power-management flag was a module-wide global that could changeCVE-2026-64476 · Linux kernel (drivers/vfio/pci)Medium
• Linux kernel (drivers/iommu/intel): SVA bind and unbind are asymmetric on VT-d hardware without PCI/PRI - bind skipsCVE-2026-64591 · Linux kernel (drivers/iommu/intel)Medium
• Linux kernel (arch/x86/kvm/vmx): When a nested VM-Enter fails on invalid guest state, KVM took an open-coded exit pathCVE-2026-68081 · Linux kernel (arch/x86/kvm/vmx)Medium
• AMD SEV-ES (CacheWarp): CacheWarp: INVD lets a malicious hypervisor revert SEV-ES guest memory writes, breaking guestCVE-2023-20592 · AMD SEV-ES (CacheWarp)Medium