Normalize field type for InhabitedPredicate

[camsteffen]

Normalizing the field type can sometimes lead to a simpler InhabitedPredicate. For instance if a projection alias can be resolved.

[rustbot]

r? @wesleywiser

rustbot has assigned @wesleywiser.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler, types
• compiler, types expanded to 75 candidates
• Random selection from 18 candidates

[rust-log-analyzer]

The job aarch64-gnu-llvm-21-1 failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

---- [ui] tests/ui/associated-types/normalization-ice-issue-149746.rs stdout ----

error: Error: expected failure status (Some(1)) but received status Some(101).
status: exit status: 101
command: env -u RUSTC_LOG_COLOR RUSTC_ICE="0" RUST_BACKTRACE="short" "/checkout/obj/build/aarch64-unknown-linux-gnu/stage2/bin/rustc" "/checkout/tests/ui/associated-types/normalization-ice-issue-149746.rs" "-Zthreads=1" "-Zsimulate-remapped-rust-src-base=/rustc/FAKE_PREFIX" "-Ztranslate-remapped-path-to-local-path=no" "-Z" "ignore-directory-in-diagnostics-source-blocks=/cargo" "-Z" "ignore-directory-in-diagnostics-source-blocks=/checkout/vendor" "--sysroot" "/checkout/obj/build/aarch64-unknown-linux-gnu/stage2" "--target=aarch64-unknown-linux-gnu" "--check-cfg" "cfg(test,FALSE)" "--error-format" "json" "--json" "future-incompat" "-Ccodegen-units=1" "-Zui-testing" "-Zdeduplicate-diagnostics=no" "-Zwrite-long-types-to-disk=no" "-Cstrip=debuginfo" "--emit" "metadata" "-C" "prefer-dynamic" "--out-dir" "/checkout/obj/build/aarch64-unknown-linux-gnu/test/ui/associated-types/normalization-ice-issue-149746" "-A" "unused" "-W" "unused_attributes" "-A" "internal_features" "-A" "incomplete_features" "-A" "unused_parens" "-A" "unused_braces" "-Crpath" "-Cdebuginfo=0" "-Lnative=/checkout/obj/build/aarch64-unknown-linux-gnu/native/rust-test-helpers" "--edition=2015"
stdout: none
--- stderr -------------------------------
error[E0277]: expected an `FnMut()` closure, found `T`
##[error]  --> /checkout/tests/ui/associated-types/normalization-ice-issue-149746.rs:6:24
   |
LL |     _significant_drop: <() as Owner>::Ty<T>,
   |                        ^^^^^^^^^^^^^^^^^^^^ expected an `FnMut()` closure, found `T`
   |
   = note: wrap the `T` in a closure with no arguments: `|| { /* code */ }`
note: required by a bound in `Owner::Ty`
  --> /checkout/tests/ui/associated-types/normalization-ice-issue-149746.rs:3:26
   |
LL | trait Owner { type Ty<T: FnMut()>; }
   |                          ^^^^^^^ required by this bound in `Owner::Ty`
help: consider restricting type parameter `T` with trait `FnMut`
   |
LL | pub struct Warns<T: FnMut()> {
   |                   +++++++++

warning: changes to closure capture in Rust 2021 will affect drop order
##[warning]  --> /checkout/tests/ui/associated-types/normalization-ice-issue-149746.rs:12:9
   |
LL |     _ = || w.field
   |         ^^ -------
   |            |     |
   |            |     in Rust 2018, `w` is dropped here, but in Rust 2021, only `w.field` will be dropped here as part of the closure
   |            in Rust 2018, this closure captures all of `w`, but in Rust 2021, it will only capture `w.field`
   |
   = note: for more information, see <https://doc.rust-lang.org/edition-guide/rust-2021/disjoint-capture-in-closures.html>
note: the lint level is defined here
  --> /checkout/tests/ui/associated-types/normalization-ice-issue-149746.rs:2:9
   |
LL | #![warn(rust_2021_incompatible_closure_captures)]
   |         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
help: add a dummy let to cause `w` to be fully captured
   |
LL |     _ = || { let _ = &w; w.field }
   |            +++++++++++++         +

[wesleywiser]

I don't really understand if this is a reasonable thing to do or not.

r? types

[camsteffen]

The tests seem to be fixable. I would also like to hear a types expert's opinion on whether this looks like a sensible idea. If yes, maybe a perf run. Also I wonder if it might make sense to only normalize if an Alias is encountered.

[rust-bors]

☔ The latest upstream changes (presumably #159541) made this pull request unmergeable. Please resolve the merge conflicts by rebasing.