Alexey Vishnyakov

@derekbruening, thank you! Workaround in #5902 resolved my issue.

Yeah, that's a nice idea! Some our student may add it one day. And PRs are always welcomed ;)

We could also add clang-format checks to CI

Won't this break the dynamic symbolic execution approach where we make the constraints based on the path that was actually executed?

Anyway, if we decide to proceed with the PR, we have local DSE benchmarks to measure symbolic execution accuracy and speed.

> > Won't this break the dynamic symbolic execution approach where we make the constraints based on the path that was actually executed? > > @SweetVishnya Can you give a...

Maybe, we should split this PR by instruction groups. I can benchmark symbolic execution accuracy on our side. And we can merge succeeding PRs one by one.

@Colton1skees, I do not see the problem here. But you never know what break symbolic reasoning in real world. We may try to benchmark all string instructions in a separate...

> So far for this `stosb` instruction, I see no problem. What I suggest is to make one PR per each instruction with a bunch of unit tests to be...

Moreover, running one benchmark requires one night of a time ;)