A critical security vulnerability has been uncovered in AMD’s Secure Encrypted Virtualization (SEV) that could potentially allow an attacker to inject malicious CPU microcode in certain situations. Designated as CVE-2024-56161, this flaw has been rated with a CVSS score of 7.2 out of 10.0, indicating its high severity. AMD has acknowledged the issue, stating that improper signature verification in the AMD CPU ROM microcode patch loader could lead to a breach of confidentiality and integrity for a confidential guest operating under AMD SEV-SNP.
The discovery and reporting of this vulnerability were credited to Google security researchers Josh Eads, Kristoffer Janke, Eduardo Vela, Tavis Ormandy, and Matteo Rizzo on September 25, 2024. SEV is a security feature that employs a unique key per virtual machine to isolate VMs and the hypervisor from each other. SNP, or Secure Nested Paging, provides additional memory integrity protections to create a secure execution environment and defend against hypervisor-based attacks.
According to AMD, SEV-SNP introduces optional security enhancements to support different VM use cases, strengthen protection around interrupt behavior, and enhance defense against recently disclosed side channel attacks. Google has highlighted that the vulnerability in question is the result of an insecure hash function in the signature validation for microcode updates, which could potentially compromise confidential computing workloads. To demonstrate the vulnerability, Google has released a test payload, with additional technical details slated to be revealed after a month to allow for the fix to propagate throughout the supply chain.
If you found this article intriguing, you can follow us on Twitter and LinkedIn for more exclusive content.
Source link