Q: Do you document the verification boundary of your toolchain with any formalism?
A: Difficult to achieve this; described in english in various papers
#realworldcrypto
The models are hand-written, there could be a bug in there, try to address with extensive testing; but is it the 'best' level of abstraction?
#realworldcrypto
Q: Why are there so many Fiat-Shamir instantiation mistakes out there?
A: We lack to right tooling to auto catch these issues; must be secure in the spec first
#realworldcrypto
Next up, 'TEE.fail: Breaking Trusted Execution Environments via Memory Bus Interposition', presented by Christina Garman and Daniel Genkin
#realworldcrypto
Q: You believe these protections only got removed bc of performance?
A: Yes I do, we talked to the engineers, they were so upset and repeatedly saying "I told you so" etc
#realworldcrypto