diagnostics: add note when param-env shadows global impl#150424

xonx4l · 2025-12-27T09:22:30Z

This PR adds a diagnostics note when param-env shadows global impl as discussed in #149910

It adds a note explaining that the definition is hidden by the generic bound.

r?lcnr

lcnr · 2026-01-08T12:21:53Z

compiler/rustc_trait_selection/src/error_reporting/infer/mod.rs

+        let all_impls =
+            impls.blanket_impls().iter().chain(impls.non_blanket_impls().values().flatten());
+
+        for &impl_def_id in all_impls {


use compute_applicable_impls_for_diagnostics instead of all_impls here

and specialization_graph::assoc_def to get the DefId of the assoc item

lcnr · 2026-01-08T12:27:24Z

compiler/rustc_trait_selection/src/error_reporting/infer/mod.rs

+                    "the associated type `{}` is defined as `{}` in the implementation, \
+                    but the generic bound `{}` hides this definition",


cc @rust-lang/types unsure about this note. I think it's certainly useful. Ideally we'd link to some reference section or github issue explaining impl shadowing.

I think it's easiest to write a special github issue for this quickly explaining why this happens, what they can do about it, and how this may change in the future

Im not sure a github issue is the best place but also I don't know where would be better... If something reference-like was t-types controlled then certainly there 🤔 github issue is probably good enough :)

should I open a issue then ?

going to do so myself as I have strong opinions on the way we frame this

opened #152409. It took me a while to get to this 😅 sorry

can you refer to this issue in the note (and also make sure its wording matches the wording used in that issue)

r=me after

xonx4l · 2026-01-14T06:14:43Z

@lcnr I made the suggested changes and CI is also green now .

compiler/rustc_trait_selection/src/error_reporting/infer/mod.rs

xonx4l · 2026-01-21T12:11:20Z

@lcnr is anything more required other than squashing commits??

Co-authored-by: lcnr <rust@lcnr.de>

rustbot · 2026-02-10T18:43:11Z

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.

lcnr · 2026-02-10T23:49:03Z

compiler/rustc_trait_selection/src/error_reporting/infer/mod.rs

+                };
+
+                let impl_item_def_id = leaf_def.item.def_id;
+                let impl_assoc_ty = tcx.type_of(impl_item_def_id).instantiate(tcx, impl_substs);


this currently incorrectly emits the note for the following I think

trait Trait { type Assoc; } impl<T> Trait for T { type Assoc = T; } fn foo<T: Trait>(x: T::Assoc) -> u32 { x }

This also doesn't handle GATs and the following probably ICEs

trait Trait { type Assoc<T>; } impl<T> Trait for T { type Assoc<U> = U; } fn foo<T: Trait>(x: T::Assoc<T>) -> u32 { x }

to handle this, consider the relevant code in fn translate_args

goal.predicate.alias.args.rebase_onto(cx, impl_trait_ref.def_id.into(), impl_args)

for this, you actually do have to equate the alias trait_ref with the impl trait_ref (and you can unwrap this equate as it should never fail)

I've made changes to the logic to handle GATs as suggested. Thanks for the direction.

please add the aboe snippets as tests

Done. Added the required tests.

lcnr · 2026-02-13T13:57:29Z

tests/ui/associated-types/param-env-shadowing-gat.rs

@@ -0,0 +1,13 @@
+trait Trait {


can you add a comment to each test explaining what it is testing?

Added all the comments for the tests.

lcnr · 2026-02-14T18:38:07Z

compiler/rustc_trait_selection/src/error_reporting/infer/mod.rs

+                let trait_def_id = alias.trait_def_id(tcx);
+                let rebased_args = alias.args.rebase_onto(tcx, trait_def_id, impl_substs);


move this below the match, they are only necessary for the let impl_assoc_ty

lcnr · 2026-02-14T18:38:35Z

tests/ui/associated-types/param-env-shadowing-issue-149910.rs

@@ -0,0 +1,14 @@
+trait Trait {


also mention that we want to tell the user about param_env shadowing in a comment here

lcnr · 2026-02-14T18:39:30Z

@bors delegate+

rust-bors · 2026-02-14T18:39:32Z

✌️ @xonx4l, you can now approve this pull request!

If @lcnr told you to "r=me" after making some further change, then please make that change and post @bors r=lcnr.

xonx4l · 2026-02-14T22:32:14Z

@bors r=lcnr

rust-bors · 2026-02-14T22:32:17Z

📌 Commit 1be4164 has been approved by lcnr

It is now in the queue for this repository.

Rollup of 9 pull requests Successful merges: - #150424 (diagnostics: add note when param-env shadows global impl) - #152132 (implement `carryless_mul`) - #152508 (Improve write! and writeln! error when called without destination) - #152534 (Test(lib/win/net): Skip UDS tests when under Win7) - #152578 (ci: Lock cross toolchain version and update docs) - #152188 (Include `library/stdarch` for `CURRENT_RUSTC_VERSION` updates) - #152402 (Add regression test for #141738) - #152472 (unwind/wasm: fix compile error by wrapping wasm_throw in unsafe block) - #152610 (Exchange js_lint message between bless and non-bless)

Rollup merge of #150424 - xonx4l:suggest_param_env_shadowing, r=lcnr diagnostics: add note when param-env shadows global impl This PR adds a diagnostics note when param-env shadows global impl as discussed in #149910 It adds a note explaining that the definition is hidden by the generic bound. r?lcnr

Rollup of 9 pull requests Successful merges: - rust-lang/rust#150424 (diagnostics: add note when param-env shadows global impl) - rust-lang/rust#152132 (implement `carryless_mul`) - rust-lang/rust#152508 (Improve write! and writeln! error when called without destination) - rust-lang/rust#152534 (Test(lib/win/net): Skip UDS tests when under Win7) - rust-lang/rust#152578 (ci: Lock cross toolchain version and update docs) - rust-lang/rust#152188 (Include `library/stdarch` for `CURRENT_RUSTC_VERSION` updates) - rust-lang/rust#152402 (Add regression test for rust-lang/rust#141738) - rust-lang/rust#152472 (unwind/wasm: fix compile error by wrapping wasm_throw in unsafe block) - rust-lang/rust#152610 (Exchange js_lint message between bless and non-bless)

rustbot assigned lcnr Dec 27, 2025

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Dec 27, 2025

This comment has been minimized.

Sign in to view

lcnr reviewed Jan 8, 2026

View reviewed changes

This comment has been minimized.

Sign in to view

xonx4l force-pushed the suggest_param_env_shadowing branch from 22a4a6b to f616235 Compare January 14, 2026 04:45

lcnr reviewed Jan 15, 2026

View reviewed changes

compiler/rustc_trait_selection/src/error_reporting/infer/mod.rs Show resolved Hide resolved

This comment has been minimized.

Sign in to view

xonx4l and others added 4 commits February 10, 2026 18:42

Add note for param-env shadowing

5b1070b

add suggested changes

a873a3f

delete middle check.

c576dc5

Co-authored-by: lcnr <rust@lcnr.de>

rewording note and refer issue

071d0cd

xonx4l force-pushed the suggest_param_env_shadowing branch from 408b4a3 to 071d0cd Compare February 10, 2026 18:43

This comment has been minimized.

Sign in to view

update tests

6b3c425

lcnr reviewed Feb 10, 2026

View reviewed changes

xonx4l added 3 commits February 12, 2026 12:10

fix GATs logic

1eddc17

fix tidy

7a2f2fe

add tests

7a7d48e

lcnr reviewed Feb 13, 2026

View reviewed changes

add comments to tests

34981bf

lcnr reviewed Feb 14, 2026

View reviewed changes

move and add comment

1be4164

rust-bors bot added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Feb 14, 2026

jhpratt mentioned this pull request Feb 15, 2026

Rollup of 9 pull requests #152639

Merged

rust-bors bot merged commit 79adcd1 into rust-lang:main Feb 15, 2026
11 checks passed

rustbot added this to the 1.95.0 milestone Feb 15, 2026

This was referenced Feb 15, 2026

[ICE]: associated type missing default #152663

Open

[ICE]: DefId does not have a "type_of" #152668

Open

[ICE]: type parameter .. out of range when instantiating #152684

Open

		"the associated type `{}` is defined as `{}` in the implementation, \
		but the generic bound `{}` hides this definition",

		let trait_def_id = alias.trait_def_id(tcx);
		let rebased_args = alias.args.rebase_onto(tcx, trait_def_id, impl_substs);

Uh oh!

Conversation

xonx4l commented Dec 27, 2025

Uh oh!

This comment has been minimized.

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.

xonx4l commented Jan 14, 2026

Uh oh!

Uh oh!

xonx4l commented Jan 21, 2026

Uh oh!

This comment has been minimized.

rustbot commented Feb 10, 2026

Uh oh!

This comment has been minimized.

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!

Choose a reason for hiding this comment

Uh oh!

lcnr commented Feb 14, 2026

Uh oh!

rust-bors bot commented Feb 14, 2026

Uh oh!

xonx4l commented Feb 14, 2026

Uh oh!

rust-bors bot commented Feb 14, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants