Fixes for the generator transform
* Moves cleanup annotations in pretty printed MIR so that they can be tested
* Correctly determines which drops are in cleanup blocks when elaborating generator drops
* Use the correct state for poisoning a generator
Closes#58892
This commit refactors the `UserTypeAnnotation` type to be referred to by
an index within `UserTypeProjection`. `UserTypeAnnotation` is instead
kept in an `IndexVec` within the `Mir` struct.
Further, instead of `UserTypeAnnotation` containing canonicalized types,
it now contains normal types and the entire `UserTypeAnnotation` is
canonicalized. To support this, the type was moved from the `rustc::mir`
module to `rustc::ty` module.
For some weird reason this fixes `intrinsic-move-val`. It also affects
various test heuristics. I removed one test (`reborrow_basic`) that
didn't seem to really be testing anything in particular anymore,
compared to all the other tests we've got.
This simplifies analysis and borrow-checking because liveness at the
resume point can always be simply propagated.
Later on, the "dead" Resumes are removed.
Removing nops can allow more basic blocks to be merged, but merging
basic blocks can't allow for more nops to be removed, so we should
remove nops first.
This doesn't matter *that* much, because normally we run SimplifyCfg
several times, but there's no reason not to do it.
Overall goal: reduce the amount of context a mir pass needs so that it
resembles a query.
- The hooks are no longer "threaded down" to the pass, but rather run
automatically from the top-level (we also thread down the current pass
number, so that the files are sorted better).
- The hook now receives a *single* callback, rather than a callback per-MIR.
- The traits are no longer lifetime parameters, which moved to the
methods -- given that we required
`for<'tcx>` objecs, there wasn't much point to that.
- Several passes now store a `String` instead of a `&'l str` (again, no
point).
In MIR construction, operands need to live exactly until they are used,
which is during the (sub)expression that made the call to `as_operand`.
Before this PR, operands lived until the end of the temporary scope,
which was sometimes unnecessarily longer and sometimes too short.
Fixes#38669.
