Open
Description
Looking at some code MIR-verifier, I began to wonder if it would make sense to have an operation to promote a crucible GlobalVar
value to a RefCell
. This would allow mux-tree style muxing of ref cells and global variables, and allow passing global variables dynamically to functions, etc.
We would lose the property that global variables are have statically known read/write locations in crucible CFG programs; however, we already can't rely on knowing where all the access points for global variables are because they can be accessed by extension statements, overrides, etc., so I don't think that's any great loss.