Threat Intelligence

Google Cloud Tells Customers to Patch and Restart VMs on AMD Turin Chips Over Three CPU Flaws

The Hacker News · 12 Nov 2026
Key Takeaway If you run workloads on affected Google Cloud AMD VMs, check your OS vendor for a fix, apply it as soon as it is available, and then stop and start the VM as Google instructs.

Google Cloud has warned that virtual machines running on AMD Turin processors (fifth-generation AMD EPYC) need a guest operating system patch and a stop and start to guard against three CPU side-channel flaws. The affected machine series are C4D, N4D, H4D, Z4D and G4, including bare-metal instances of C4D and Z4D. Google has already applied mitigations against attacks between VMs, and between VMs and the host.

Inside a VM, the flaws may let an attacker who can run code on the same processor core infer data from other processes or the kernel. All three are Transient Scheduler Attacks. Two, CVE-2024-36350 and CVE-2024-36357, were first disclosed on July 8, 2025 for earlier EPYC chips. AMD now says its Zen 5 processors are impacted differently from previously characterised processors. The third flaw, CVE-2026-0004, is new and may allow inference of data in the processor's L1 data cache. Google rates the issue Medium. AMD describes all three as local attacks with high complexity requiring only low privileges. Neither company reports exploitation, and none is in CISA's KEV catalog.

There are gaps for customers. Google lists nothing to do in the meantime for VMs whose OS vendor has not shipped a fix, and does not say how to tell when the CPU information update has reached a VM. A Linux kernel commit mitigating the issue exists, but Debian has so far fixed CVE-2026-0004 only in its unstable branch, with bookworm, trixie and forky still listed as vulnerable.

Google Cloud AMD CPU vulnerabilities Side-channel attacks Patch management

Summarised by CISO AI from The Hacker News, written with Claude Sonnet 5.5. We link back to every original so you can read it yourself.