New DDRop Attack Breaks Intel TDX and AMD SEV-SNP Confidential Computing
DDRop uses a sub-$200 DDR5 interposer to drop memory writes, breaking Intel TDX and AMD SEV-SNP confidentiality guarantees.
Researchers at KU Leuven, ETH Zurich, Durham University, and Google will present DDRop at ACM CCS 2026, a first active interposer attack on DDR5 that silently drops memory writes so processors keep reading stale encrypted data, exploiting the missing freshness guarantee in Intel TDX, Intel Scalable SGX, and AMD SEV-SNP. On Intel TDX's default logical integrity mode it enabled reading victim VM memory, toggling debug mode, and forging remote attestation; AMD SEV-SNP was limited to copying pages between VMs. TDX's stronger cryptographic integrity blocks cross-VM attacks but likely not attestation forgery. The team will release board designs, firmware, and attack code on GitHub; no evidence of real-world use exists and no simple patch is available.
- Sub-$200 interposer drops DDR5 writes; processor reads stale encrypted data undetected
- First active DDR5 interposer attack breaking up-to-date Intel TDX integrity
- TDX default mode allows victim memory read, debug mode, and attestation forgery
- Affects Intel TDX, Scalable SGX, AMD SEV-SNP; used by AWS, Azure, Google Cloud
- Requires attacker software control plus brief physical access; no simple patch
From the original article
6 min read at thehackernews.com
Researchers have disclosed a new hardware attack, called DDRop, that breaks the memory protection in Intel and AMD confidential computing by silently dropping writes to a server's memory, so the processor keeps reading old encrypted data as if it were current. The attack requires an attacker who already controls the server's software and can briefly access the machine to insert a small circuit
thehackernews.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.