International Association
for Cryptologic Research

Cryptographic software aiming to be constant-time must avoid overwriting a memory location with a value that, depending on a secret, is either all-zero or not all-zero. The reason is silent zero store suppression, a microarchitectural optimization that allows evicted all-zero cachelines that are dirty but unchanged to be dropped instead of written back. As discovered by Travis Downs, recent Intel processors implement silent zero store suppression for some evictions from the L2 cache. We describe an adaptive chosen-ciphertext attack strategy against SIKE (a popular post-quantum key-encapsulation primitive) in which a correct key-bit guess triggers thousands of suppressed silent zero stores. We show that our attack strategy renders both the Cloudflare CIRCL implementation of SIKE (written in Go) and the Microsoft PQCrypto-SIKE implementation (written in C) vulnerable to a remote timing attack when running on an Intel Ice Lake CPU, despite having been written to be side-channel resistant. Our attack recovers the complete 378-bit SIKE-751 secret key from a CIRCL server in 39 hours and from a PQCrypto-SIKE server in 72 hours.

In a proof-of-retrievability system, a data storage center must prove to a verifier that he is actually storing all of a client’s data. The central challenge is to build systems that are both efficient and provably secure—that is, it should be possible to extract the client’s data from any prover that passes a verification check. In this paper, we give the first proof-of-retrievability schemes with full proofs of security against arbitrary adversaries in the strongest model, that of Juels and Kaliski.Our first scheme, built from BLS signatures and secure in the random oracle model, features a proof-of-retrievability protocol in which the client’s query and server’s response are both extremely short. This scheme allows public verifiability: anyone can act as a verifier, not just the file owner. Our second scheme, which builds on pseudorandom functions (PRFs) and is secure in the standard model, allows only private verification. It features a proof-of-retrievability protocol with an even shorter server’s response than our first scheme, but the client’s query is long. Both schemes rely on homomorphic properties to aggregate a proof into one small authenticator value.