add `TypingMode::Borrowck` #138785

lcnr · 2025-03-21T13:19:00Z

First two commits are from #139191.

Introduces TypingMode::Borrowck which unlike TypingMode::Analysis, uses the hidden type computed by HIR typeck as the initial value of opaques instead of an unconstrained infer var. This is a part of rust-lang/types-team#129.

Using this new TypingMode is unfortunately a breaking change for now, see tests/ui/impl-trait/non-defining-uses/as-projection-term.rs. Using an inference variable as the initial value results in non-defining uses in the defining scope. We therefore only enable it if with -Znext-solver=globally or -Ztyping-mode-borrowck

To do that the PR contains the following changes:

TypeckResults::concrete_opaque_type are already mapped to the definition of the opaque type
- writeback now checks that the non-lifetime parameters of the opaque are universal
- for this, fn check_opaque_type_parameter_valid is moved from rustc_borrowck to rustc_trait_selection
we add a new query type_of_opaque_hir_typeck which, using the same visitors as MIR typeck, attempts to merge the hidden types from HIR typeck from all defining scopes
- done by adding a DefiningScopeKind flag to toggle between using borrowck and HIR typeck
- the visitors stop checking that the MIR type matches the HIR type. This is trivial as the HIR type are now used as the initial hidden types of the opaque. This check is useful as a safeguard when not using TypingMode::Borrowck, but adding it to the new structure is annoying and it's not soundness critical, so I intend to not add it back.
add a TypingMode::Borrowck which behaves just like TypingMode::Analysis except when normalizing opaque types
- it uses type_of_opaque_hir_typeck(opaque) as the initial value after replacing its regions with new inference vars
- it uses structural lookup in the new solver

fixes #112201, fixes #132335, fixes #137751

r? @compiler-errors @oli-obk

add `TypingMode::Borrowck` Still not quite ready Based on rust-lang#138492 and rust-lang#138719 r? `@compiler-errors` `@oli-obk`

increment depth of nested obligations properly fixes the root cause of rust-lang#109268. While we didn't get hangs here before, I ended up encountering its root cause again with rust-lang#138785. r? types

lcnr · 2025-04-01T13:22:35Z

@bors try @rust-timer queue

rust-timer · 2025-04-01T13:22:37Z

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

add `TypingMode::Borrowck` Still not quite ready Based on rust-lang#138492 r? `@compiler-errors` `@oli-obk`

bors · 2025-04-01T13:23:48Z

⌛ Trying commit d7cf428 with merge f8ad33e35a8aba1c9928e3c863abba2373e0d978...

increment depth of nested obligations properly fixes the root cause of rust-lang#109268. While we didn't get hangs here before, I ended up encountering its root cause again with rust-lang#138785. r? types

lcnr · 2025-04-01T21:41:42Z

@rustbot ready

rustbot · 2025-04-01T21:41:46Z

Some changes occurred to the core trait solver

cc @rust-lang/initiative-trait-system-refactor

This PR changes a file inside tests/crashes. If a crash was fixed, please move into the corresponding ui subdir and add 'Fixes #' to the PR description to autoclose the issue upon merge.

Some changes occurred in exhaustiveness checking

cc @Nadrieril

lcnr · 2025-04-01T21:53:46Z

tests/ui/impl-trait/issue-55872.rs

        //~^ ERROR type parameter `T` is part of concrete type but not used in parameter list for the `impl Trait` type alias
+        || ()


unfortunate side-effect from HIR typeck eagerly replacing the return type with an inference var. This will be partially fixed when removing this eager normalization call. The new solver does point to the closure instead.

However, as we're still normalizing away opaques to infer vars during generalization, there may still be span regressions where we point at the place introducing the opaque instead of the place which actually constrains it. This feels unavoidable without major changes to the way we track spans for opaques/in the type system in general.

lcnr · 2025-04-01T21:54:37Z

tests/ui/impl-trait/issue-99073-2.rs

@@ -7,8 +7,7 @@ fn main() {
 fn test<T: Display>(t: T, recurse: bool) -> impl Display {
    let f = || {
        let i: u32 = test::<i32>(-1, false);
-        //~^ ERROR concrete type differs from previous defining opaque type use
-        //~| ERROR expected generic type parameter, found `i32`
+        //~^ ERROR expected generic type parameter, found `i32`


error tainting 🎉

lcnr · 2025-04-01T21:59:31Z

@rust-timer build d7cf428

Rollup merge of rust-lang#139022 - lcnr:incr-obligation-depth, r=oli-obk increment depth of nested obligations properly fixes the root cause of rust-lang#109268. While we didn't get hangs here before, I ended up encountering its root cause again with rust-lang#138785. r? types

…errors small opaque type/borrowck cleanup pulled out of rust-lang#138785

rustbot assigned compiler-errors Mar 21, 2025

rustbot added A-tidy S-waiting-on-review T-bootstrap T-compiler T-libs WG-trait-system-refactor labels Mar 21, 2025