Rust Cryptography Should Be Written in Rust

Re:Rust Cryptography Should Be Written in Rust by charsii on Saturday August 26, 2023 @02:15PM (#37275377)

Crypto code should be written in assembly. Zero ambiguity, zero undefined behavior, 100% verifiable.

[ Parent | on HN ]

Re:Rust Cryptography Should Be Written in Rust by woodruffw on Saturday August 26, 2023 @02:39PM (#37275627)

ISAs regularly leave all kinds of behavior undefined when they think it doesn’t matter (such as the state of the arithmetic flags after operations that shouldn’t need to the tested).

(But this is also irrelevant: assembly can be completely wrong and exploitable while also being perfectly well defined.)

[ Parent | on HN ]

Re:Rust Cryptography Should Be Written in Rust by otikik on Saturday August 26, 2023 @02:40PM (#37275638)

Wouldn’t the low readability make it easier to slip a vulnerability in by splicing it out into several changes?

[ Parent | on HN ]

Re:Rust Cryptography Should Be Written in Rust by tromp on Saturday August 26, 2023 @02:18PM (#37275411)

Readability suffers though, with negative impacts on maintainability and even verifiability (fewer people able/willing to examine the source code).

[ Parent | on HN ]

Re:Rust Cryptography Should Be Written in Rust by torstenvl on Saturday August 26, 2023 @02:20PM (#37275430)

goto fail wasn't caused by ambiguity or undefined behavior. C's rules here are crystal clear, and conditional branches in assembly also do not make the following instructions condition (unless you're using delay slots à la SPARC).

Heartbleed also wasn't caused by ambiguity or undefined behavior, if you believe compiler writers.

[ Parent | on HN ]