You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Milestone 4 requires us to build a static analysis tool operating over Covenant's serial form. In order for this to be meaningful, it must meet three criteria:
Static: we cannot evaluate, or simulate evaluating, either Covenant itself or UPLC as part of the analysis.
Meaningful: should be something that would be of genuine use.
Novel: not something that can be done against UPLC.
To give some examples of what wouldn't be suitable:
CTE (compile-time evaluation) wouldn't be static or novel: not only can UPLC do this, it requires simulation in basically any non-trivial case.
Outputting the node count of the Covenant IR graph wouldn't be meaningful: nobody is really interested in that.
DCE (dead code elimination) wouldn't be novel: you can do that with UPLC just fine.
Ideally we want to know what kind of analysis tool we're building before we decide Covenant's exact form, as this could make our lives much easier. A security-oriented analysis tool would also be quite good to have, assuming we can find such a case.
The text was updated successfully, but these errors were encountered:
Milestone 4 requires us to build a static analysis tool operating over Covenant's serial form. In order for this to be meaningful, it must meet three criteria:
To give some examples of what wouldn't be suitable:
Ideally we want to know what kind of analysis tool we're building before we decide Covenant's exact form, as this could make our lives much easier. A security-oriented analysis tool would also be quite good to have, assuming we can find such a case.
The text was updated successfully, but these errors were encountered: