implement Aarch64 UMOV instruction semantics

JonathanSalwan / Triton

Triton is a dynamic binary analysis library. Build your own program analysis tools, automate your reverse engineering, perform software verification or just emulate code.

https://triton-library.github.io

Apache License 2.0

3.39k stars 524 forks source link

implement Aarch64 UMOV instruction semantics #1331

Closed psi-func closed 1 month ago

psi-func commented 1 month ago

add more VAS types to correctly handle different cases which part of vector register should move to GPR register add umov semantics in aarch64 instruction reference here

psi-func commented 1 month ago

Capstone v4.0.1 missing important feature with parsing extended VAS types build fails in CI

gulldan commented 1 month ago

also need it ASAP

JonathanSalwan commented 1 month ago

Lol, I think this is the first ever MR that involves so many reactions ^^.

Regarding Capstone 5: As 5.0.1 is now tagged as release, I think we can drop support for 4.x.
Regarding unit tests. Indeed, it could be great to add some in https://github.com/JonathanSalwan/Triton/blob/master/src/testers/aarch64/unicorn_test_aarch64.py

psi-func commented 1 month ago

Found problem in this code via bug in capstone Tryed to add tests in python "\x43\x3c\x1c\x0e", "umov w3, v2.S[3]" Parsed by capstone as MOV instruction, because of this buildSemantics is performed incorrectly Left issue here They advise using the next branch