upenn-acg/verified-tsan

[TODO]: prove a corollary of instrument_sim_safe2 that holds for iexec_star

stilltracy opened this issue · 2 comments

In order to prove the clause about memory consistency, I wrote down a lemma here. @mansky1 William, would you mind taking a look at it to make sure it's provable?

The corollary is almost done, except the admit here: we should have been able to apply ss_trans here, however it somehow just failed. @mansky1

Something went wrong with the type inference in ss_trans, and it was choosing a weird instantiation of VectorClocks.state that would never be applicable. I put in a type annotation that fixed it.