Select Polonius algorithm via `POLONIUS_ALGORITHM` environment variable
This pull request allows selecting the Polonius algorithm being used by providing an environment variable `POLONIUS_ALGORITHM`.
Example usage:
```
POLONIUS_ALGORITHM=compare RUST_LOG=rustc_mir::borrow_check::nll=trace ./x.py test --stage 1 --compare-mode polonius -- src/test/ui/nll/issue-47680.rs
...
stderr:
------------------------------------------
INFO 2018-05-31T17:35:31Z: rustc_mir::borrow_check::nll: Using Polonius algorithm: Compare
INFO 2018-05-31T17:35:31Z: rustc_mir::borrow_check::nll: Using Polonius algorithm: Compare
------------------------------------------
...
```
r? @nikomatsakis
Suggest not mutably borrowing a mutable reference
This PR would (hopefully) solve #45392. I deviated a bit from @estebank's instructions since the error span only included the borrowed expression (e.g. the `b` in `&mut b`). I also didn't check the mutability of the local binding, since this whole case is concerned with an immutable local.
I can see two outstanding questions:
1. `note_immutability_blame` is called in two places, but I only have one test case. I think it covers the call in `report_bckerror`, but I'm not sure how to trigger the call from `report_aliasability_violation`.
2. There is one failing test, where the local binding is `self: &mut Self`. I'm not entirely sure what the correct output should be, but I think the new message should also apply. Unfortunately, since this parameter is parsed differently, its `let_span` covers both the pattern and the type, leading to a wrong suggestion text. I'm not sure how to correctly identify this case.
Optimize layout calculations in HashMap
This now produces the same assembly code as the previous implementation.
cc #51163 @nnethercote @gnzlbg @andjo403
Do not promote union field accesses
r? @eddyb
technically a breaking change, but the code errored on the previous stable and produces UB + a warning on the current stable. I don't think we need a crater run in that case.
Remove the unused `-Z trans-time-graph` flag.
Rebase of #50783 has accidentally revived the flag (which should be renamed to `-Z codegen-time-graph` by #50615).
Generate br for all two target SwitchInts
Instead of only for booleans. This means that `if let` also becomes a br.
Apart from making the IR slightly simpler, this is supported by FastISel (#4353).
const fn integer operations
A follow up to #51171Fixes#51267
Makes a lot of the integer methods (`swap_bytes`, `count_ones` etc) `const fn`s. See #51267 for a discussion about why this is wanted and the solution used.
Remove rustdoc-specific is_import field from HIR
Fixes#47100.
I believe that there is no need to check for the name being the same, as this
part of rustdoc seems to be strictly interested in exploring "public modules."
Re-exports from the same module cannot visit another module; and, re-exports
cannot export items with a greater visibility than that item declares.
Therefore, I think this code is either sufficient, or in fact does more than
is necessary, depending on whether rustdoc cares about the re-export itself.
r? @eddyb
