trait_selection: fix assumptions-on-binders diagnostics by Dnreikronos · Pull Request #158588 · rust-lang/rust

Dnreikronos · 2026-06-29T23:02:37Z

-Zassumptions-on-binders was losing the origin of solver region constraints. By the time regionck or borrowck reported them, every constraint received the containing item span, so diagnostics pointed at an entire function or const. Some paths were also still emitting placeholder text (:3 and meoow :c).

This keeps canonical solver responses and ExternalConstraintsData span-free, so source locations do not participate in candidate equality or caching. When a response is applied, EvalCtxt::origin_span is attached to each atomic solver RegionConstraint stored in InferCtxt. Late region conversion then uses the span attached to each constraint, including for ambiguity diagnostics.

The affected paths now report higher-ranked lifetime bound could not be satisfied at the type use that introduced the failing constraint. UI coverage includes two failures from one InferCtxt to ensure their origins remain distinct. Unit coverage checks that the spanned evaluator stays in semantic parity with the type-ir evaluator and preserves the first ambiguity origin span.

Future work: carry the failed outlives predicate or originating binder far enough through this path to name the exact bound that failed, rather than only pointing at its origin.

rustbot · 2026-06-29T23:02:41Z

Some changes occurred to the core trait solver

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

rustbot · 2026-06-29T23:02:43Z

r? @folkertdev

rustbot has assigned @folkertdev.
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
compiler expanded to 73 candidates
Random selection from 17 candidates

folkertdev · 2026-06-30T07:59:18Z

r? types

lcnr · 2026-06-30T08:05:27Z

r? BoxyUwU

We intend to fully rework the way these diagnostics work and the errors are currently intentionally bad as they don't even have span information (they point to the whole containing function).

I think keeping the status quo is preferable as a very explicit "this is still very WIP" and would personally prefer to close this PR

rustbot · 2026-06-30T08:05:32Z

BoxyUwU is currently at their maximum review capacity.
They may take a while to respond.

Dnreikronos · 2026-06-30T11:37:59Z

r? BoxyUwU

We intend to fully rework the way these diagnostics work and the errors are currently intentionally bad as they don't even have span information (they point to the whole containing function).

I think keeping the status quo is preferable as a very explicit "this is still very WIP" and would personally prefer to close this PR

thanks, that makes sense. i was mostly looking at this as remove the placeholder text, but imo you're right that the nicer message is a bit misleading if the span is still basically wrong.

i'll wait for boxyuwu's opinion before closing the pr. if i poke at this again later, i think i'd rather start by carrying more of the failed outlives predicate / binder info through first, then make the diagnostic nicer once it has enough data to not lie.

BoxyUwU · 2026-07-14T19:16:19Z

I think I agree with lcnr here 👍 Would you be interested in trying to do a more involved refactor here and figure out properly spanned diagnostics for -Zassumptions-on-binders? i.e. make it so we aren't just pointing to the whole item which the constraint originates from.

I think this would be really useful to have done and I won't get the time to do it for a while because of other priorities (getting std/core to compile and then custom test suite).

Dnreikronos · 2026-07-14T19:18:28Z

I think I agree with lcnr here 👍 Would you be interested in trying to do a more involved refactor here and figure out properly spanned diagnostics for -Zassumptions-on-binders? i.e. make it so we aren't just pointing to the whole item which the constraint originates from.

I think this would be really useful to have done and I won't get the time to do it for a while because of other priorities (getting std/core to compile and then custom test suite).

Of course!
I will do my best!

BoxyUwU · 2026-07-14T19:27:07Z

sick :3 it's probably a more involved change than your previous PRs so it's probably worth us chatting on zulip about it as you're working on it ✨ I would probably start by looking into where the spans for normal region errors come from (i.e. without -Zassumptions-on-binders) and then look into how to replicate that for -Zassumptions-on-binders

BoxyUwU · 2026-07-22T15:38:16Z

@rustbot author

rustbot · 2026-07-22T15:38:21Z

Reminder, once the PR becomes ready for a review, use @rustbot ready.

Dnreikronos · 2026-07-23T18:51:10Z

@rustbot ready

BoxyUwU

yeah this looks along the right lines for having actual diagnostics :3 sick

View changes since this review

BoxyUwU · 2026-07-24T11:36:13Z

@@ -66,6 +69,9 @@ pub(super) fn fulfillment_error_for_no_solution<'tcx>(
            let expected_found = ExpectedFound::new(b, a);
            FulfillmentErrorCode::Subtype(expected_found, TypeError::Sorts(expected_found))
        }
+        ty::PredicateKind::Clause(
+            ty::ClauseKind::RegionOutlives(_) | ty::ClauseKind::TypeOutlives(_),


Do you have test cases where this change was necessary?

yep, higher-ranked-outlives-issue-157732.rs covers this. fyi, I tried removing this arm and the test ICEs at fulfillment_errors.rs:667 instead of emitting the diagnostic. imo keeping this arm is the right call. idk if it's worth adding a note to the test, but ltm if you'd prefer that.

this is ok I just wanted to check that this gets hit in practice (and it makes sense to me that it would be)

BoxyUwU · 2026-07-24T11:38:00Z

+/// participate in candidate equality or caching. Spans are attached when those responses are
+/// applied to an inference context.
+#[derive(Clone, Debug)]
+pub(crate) enum SolverRegionConstraint<'tcx> {


Instead of duplicating the definition of RegionConstraint could you look into making RegionConstraint be generic over some S for the Span. Then we can do type SpannedRegionConstraint<I> = RegionConstraint<I, I::Span> and type RegionConstraint<I> = RegionConstraint<I, ()>

done :) RegionConstraint is generic over S = () now, with SpannedRegionConstraint<I> using I::Span. btw this also got rid of the duplicate infer-side enum and lets both paths use the type-ir evaluator. imo this is much cleaner.

BoxyUwU · 2026-07-24T11:40:30Z

        let constraint = self.inner.borrow().solver_region_constraint_storage.get_constraint();
        debug!(?constraint);
-        let constraint =
+        let constraint = constraint.map_atomic_constraints(&mut |constraint| {


If you make RegionConstraint generic over whether it's spanned or not then we can get rid of map_atomic_constraints and make destructure_type_outlives_constraints_in_root be generic over S too. That would be ideal I think

done too. destructure_type_outlives_constraints_in_root is generic over S and carries the same span into the constraints it expands. map_atomic_constraints is gone. The shared evaluator still keeps the first ambiguity span. ltm if you had a different shape in mind.

BoxyUwU

ok sick this is what I had in mind :) thanks, I think this will merge conflict with your other PR so lets wait for that to merge first

View changes since this review

BoxyUwU · 2026-07-24T14:12:29Z

+    Or(Box<[RegionConstraint<I, S>]>),
 }

+/// A solver region constraint together with the span that caused each atomic constraint.


Suggested change

/// A solver region constraint together with the span that caused each atomic constraint.

/// A solver region constraint together with the span that caused each leaf constraint.

Just adjusted, Boxy :)

rustbot · 2026-07-27T16:47:10Z

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

rustbot added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Jun 29, 2026

rustbot assigned folkertdev Jun 29, 2026

rustbot added T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. WG-trait-system-refactor The Rustc Trait System Refactor Initiative (-Znext-solver) labels Jun 29, 2026

This comment has been minimized.

Sign in to view

Dnreikronos force-pushed the trait_selection/assumptions_binders_diagnostics branch from 6d112e4 to 36f4d5a Compare June 29, 2026 23:09

This comment has been minimized.

Sign in to view

rustbot added the T-types Relevant to the types team, which will review and decide on the PR/issue. label Jun 30, 2026

rustbot assigned jackh726 and unassigned folkertdev Jun 30, 2026

rustbot assigned BoxyUwU and unassigned jackh726 Jun 30, 2026

BoxyUwU mentioned this pull request Jul 13, 2026

Assumptions on Binders rust-lang/rust-project-goals#621

Open

rustbot added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Jul 22, 2026

Dnreikronos force-pushed the trait_selection/assumptions_binders_diagnostics branch from 36f4d5a to 269fb71 Compare July 23, 2026 17:10

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Jul 23, 2026

BoxyUwU reviewed Jul 24, 2026

View reviewed changes

This comment has been minimized.

Sign in to view

rustbot added has-merge-commits PR has merge commits, merge with caution. S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Jul 24, 2026

Dnreikronos force-pushed the trait_selection/assumptions_binders_diagnostics branch from 3acffb6 to b3495e0 Compare July 24, 2026 13:19

rustbot removed has-merge-commits PR has merge commits, merge with caution. S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Jul 24, 2026

BoxyUwU reviewed Jul 24, 2026

View reviewed changes

This comment has been minimized.

Sign in to view

Dnreikronos added 3 commits July 27, 2026 13:38

Fix assumptions-on-binders outlives diagnostics

b740e13

Generalize solver region constraint spans

c86b20d

Clarify region constraint span documentation

09ce023

Dnreikronos force-pushed the trait_selection/assumptions_binders_diagnostics branch from c5beff5 to 09ce023 Compare July 27, 2026 16:47

	/// A solver region constraint together with the span that caused each atomic constraint.
	/// A solver region constraint together with the span that caused each leaf constraint.

Uh oh!

Uh oh!

Conversation

Dnreikronos commented Jun 29, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Jun 29, 2026

Uh oh!

rustbot commented Jun 29, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

folkertdev commented Jun 30, 2026

Uh oh!

lcnr commented Jun 30, 2026

Uh oh!

rustbot commented Jun 30, 2026

Uh oh!

Dnreikronos commented Jun 30, 2026

Uh oh!

BoxyUwU commented Jul 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Dnreikronos commented Jul 14, 2026

Uh oh!

BoxyUwU commented Jul 14, 2026

Uh oh!

BoxyUwU commented Jul 22, 2026

Uh oh!

rustbot commented Jul 22, 2026

Uh oh!

Dnreikronos commented Jul 23, 2026

Uh oh!

BoxyUwU left a comment • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

This comment has been minimized.

BoxyUwU left a comment • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

rustbot commented Jul 27, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

Dnreikronos commented Jun 29, 2026 •

edited by rustbot

Loading

BoxyUwU commented Jul 14, 2026 •

edited

Loading

BoxyUwU left a comment •

edited by rustbot

Loading

BoxyUwU left a comment •

edited by rustbot

Loading