Protecting from MaliciousHypervisor Using AMD SEV-SNP
summary
AMD SEV‑SNP aims to protect a VM from a malicious hypervisor by adding integrity and replay protection on top of SEV/SEV‑ES, enforced with a system‑wide reverse map table indexed per 4 KiB page and new RMP update instructions and NPF bits. Page validation is a two‑step flow using a PSC VMGEXIT then PVALIDATE, with 253‑entry batch ops covering ~500 MB. Attestation is available anytime, CPUID values go through CPUID filtering, and kernel/OVMF/QEMU patches are posted upstream; live migration and lazy validation are not yet supported.
AI-written summary · about
chapters (11)
- 0:00AMD SEV-SNP introduction
- 1:06SEV-SNP threat model
- 3:33SEV-SNP integrity with RMP
- 6:19RMP faults handling
- 9:30Guest-Host Communication Block
- 12:02Page validation in SEV-SNP
- 15:32Launch process and guest initialization
- 17:21Host commands and attestation
- 19:06SEV-SNP support status
- 20:11Unsupported features and future work
- 24:24Q&A
related talks
40mUpcoming x86 Technologies for Malicious Hypervisor Protection
34mSEV-SNP Development Status Update
15mRemote Attestation in AMD SEV-SNP Confidential VMs
11mSecure TSC for AMD SEVSNP guests
30mEnabling AMD SEV in Xen
28mProviding Confidential Guest Services with a Secure VM Service Module on AMD
33mAMD Encrypted Virtualization Update
38mInterrupt Security for AMD SEV-SNP