Avoid manually instantiating some binders in error reporting with `-Znext-solver` by sjwang05 · Pull Request #162262 · rust-lang/rust

sjwang05 · 2026-09-03T21:27:26Z

If we call enter_forall on a binder before passing it to the new solver, the placeholders that come from instantiation get canonicalized before solving happens. This can be a problem if we end up equating one variable whose universe can't name another variable's, since the equate will succeed in the solver since canonicalization sets all universe indices to the root universe but fail when we actually construct the response with fresh vars at the correct universes, which causes an ICE. We now just hand the unopened binder to the trait solver directly, since (I believe) both solvers are able to handle that just fine.

This came up while I was looking at https://rust-lang.zulipchat.com/#narrow/channel/618216-t-types.2Fcall-for-participation/topic/next-solver.20UI.20test.20triage/with/612965995, and the universe-error-host-effect.rs/poly_select change was basically the same type of ICE that I found while investigating the first one.

rustbot · 2026-09-03T21:27:30Z

Some changes occurred to the core trait solver

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

rustbot · 2026-09-03T21:27:32Z

r? @fee1-dead

rustbot has assigned @fee1-dead.
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 75 candidates
Random selection from 22 candidates

fee1-dead · 2026-09-04T01:50:19Z

@rustbot reroll

sjwang05 · 2026-09-04T03:33:25Z

-   |         -----^-----
-   |         |    |
-   |         |    the trait solver is unable to infer the generic types that should be inferred from this argument
-   |         add turbofish arguments to this call to specify the types manually, even if it's redundant


This is kind of a diagnostics regression, since before, we called ocx.normalize on the obligation before registering it, whereas now it doesn't seem like we can handle the unnormalized <Self as ATC<'a>>::Type inside the binder if it's not normalized beforehand.

View changes since the review

Enselic · 2026-09-04T03:38:17Z

@rustbot reroll

ShoyuVanilla · 2026-09-04T06:08:26Z

We used to normalize the predicate here. Don't we need it anymore?

View changes since the review

mati865 · 2026-09-04T07:54:44Z

@rustbot reroll

BoxyUwU · 2026-09-04T16:49:52Z

reminds me a bit of #160955. tbh in general im not convinced that erasing universes in the inputs makes sense if we have placeholder types... 🤷‍♀️ I don't have the capacity to properly engage with this though 😅 seems like a nice enough cleanup regardless so...

r? me

@bors r+

rust-bors · 2026-09-04T16:49:56Z

📌 Commit 368791e has been approved by BoxyUwU

It is now in the queue for this repository.

Avoid manually instantiating some binders in error reporting with `-Znext-solver` If we call `enter_forall` on a binder before passing it to the new solver, the placeholders that come from instantiation get canonicalized before solving happens. This can be a problem if we end up equating one variable whose universe can't name another variable's, since the equate will succeed in the solver since canonicalization sets all universe indices to the root universe but fail when we actually construct the response with fresh vars at the correct universes, which causes an ICE. We now just hand the unopened binder to the trait solver directly, since (I believe) both solvers are able to handle that just fine. This came up while I was looking at https://rust-lang.zulipchat.com/#narrow/channel/618216-t-types.2Fcall-for-participation/topic/next-solver.20UI.20test.20triage/with/612965995, and the `universe-error-host-effect.rs`/`poly_select` change was basically the same type of ICE that I found while investigating the first one.

Rollup of 27 pull requests Successful merges: - #159074 ([PAC] FnAbi, llvm.ptrauth.resign and Session API change (2/8)) - #159792 (A more readable debug map for IndexMaps) - #160745 (make closures act like MaybeDangling) - #161940 (Promote `wasm32-wasip3` to a tier 2 target) - #162030 (Prevent `--test` to be used in `rustdoc-html` testsuite) - #162072 (Add new Tier-3 target: `powerpc64-sony-ps3`) - #162179 (type system const items via direct rhs) - #162262 (Avoid manually instantiating some binders in error reporting with `-Znext-solver`) - #162277 (Introduce `rustc_middle::middel::resolve`) - #162285 (box: fixup map/try_map deallocate calls) - #162286 (string: don't unwind prematurely) - #162289 (alloc: a bunch of safety comments) - #162290 (abby test DSL: AliasTyOutlivesViaEnv) - #162292 (Update `askama` version to `0.16.1`) - #160509 (Remove `RegionExt`; move methods to `Region` in `rustc_type_ir`) - #160906 (Suggest usize instead of placeholder type for array length constants) - #160936 (traits: Represent live alias arguments as bitsets) - #161400 (Improve diagnostics for references to closures) - #161656 (Suggest mutable references for FnMut closure arguments) - #161711 (Add more splat fn type tests) - #161786 (Make `tcx.def_id_partial_cmp` public) - #161953 (sanitizers: Implicitly disable mutually exclusive sanitizers) - #162155 (add suggestion for `rustc_allowed_through_unstable_modules` attribute) - #162212 (Implement `Rng` for `Box`) - #162246 (Fix incorrect meta span) - #162266 (std: fix typo) - #162291 (Add regression test from 1.98.1)

…uwer Rollup of 25 pull requests Successful merges: - #162072 (Add new Tier-3 target: `powerpc64-sony-ps3`) - #162179 (type system const items via direct rhs) - #162255 (let people change rustc-dev-guide in-tree) - #162262 (Avoid manually instantiating some binders in error reporting with `-Znext-solver`) - #162277 (Introduce `rustc_middle::middel::resolve`) - #162285 (box: fixup map/try_map deallocate calls) - #162286 (string: don't unwind prematurely) - #162289 (alloc: a bunch of safety comments) - #162290 (abby test DSL: AliasTyOutlivesViaEnv) - #162292 (Update `askama` version to `0.16.1`) - #160509 (Remove `RegionExt`; move methods to `Region` in `rustc_type_ir`) - #160906 (Suggest usize instead of placeholder type for array length constants) - #160936 (traits: Represent live alias arguments as bitsets) - #161394 (fix `is_homogeneous_aggregate`: use unit size, not total size ) - #161400 (Improve diagnostics for references to closures) - #161656 (Suggest mutable references for FnMut closure arguments) - #161711 (Add more splat fn type tests) - #161786 (Make `tcx.def_id_partial_cmp` public) - #161953 (sanitizers: Implicitly disable mutually exclusive sanitizers) - #162155 (add suggestion for `rustc_allowed_through_unstable_modules` attribute) - #162212 (Implement `Rng` for `Box`) - #162246 (Fix incorrect meta span) - #162266 (std: fix typo) - #162291 (Add regression test from 1.98.1) - #162303 (Fix a minor spelling error in `library/std/src/sys/fs/unix.rs`)

Rollup merge of #162262 - sjwang05:forall-binder-ice, r=BoxyUwU Avoid manually instantiating some binders in error reporting with `-Znext-solver` If we call `enter_forall` on a binder before passing it to the new solver, the placeholders that come from instantiation get canonicalized before solving happens. This can be a problem if we end up equating one variable whose universe can't name another variable's, since the equate will succeed in the solver since canonicalization sets all universe indices to the root universe but fail when we actually construct the response with fresh vars at the correct universes, which causes an ICE. We now just hand the unopened binder to the trait solver directly, since (I believe) both solvers are able to handle that just fine. This came up while I was looking at https://rust-lang.zulipchat.com/#narrow/channel/618216-t-types.2Fcall-for-participation/topic/next-solver.20UI.20test.20triage/with/612965995, and the `universe-error-host-effect.rs`/`poly_select` change was basically the same type of ICE that I found while investigating the first one.

rust-timer · 2026-09-05T22:53:09Z

Note

This PR was benchmarked as part of triage of its containing rollup: triage URL.

Finished benchmarking commit (08587d4): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

This perf run didn't have relevant results for this metric.

Max RSS (memory usage)

Results (primary -2.5%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	7.1%	[7.1%, 7.1%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-7.2%	[-11.6%, -2.8%]	2
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-2.5%	[-11.6%, 7.1%]	3

Cycles

This perf run didn't have relevant results for this metric.

Binary size

This perf run didn't have relevant results for this metric.

Bootstrap: missing data
Artifact size: 403.29 MiB -> 403.36 MiB (0.02%)

sjwang05 added 2 commits September 3, 2026 13:59

don't manually instantiate binders in explain_hrtb_projection

da7f6c0

make poly_select delegate to next-solver directly

368791e

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. WG-trait-system-refactor The Rustc Trait System Refactor Initiative (-Znext-solver) labels Sep 3, 2026

rustbot assigned fee1-dead Sep 3, 2026

This comment was marked as resolved.

Sign in to view

rustbot assigned Enselic and unassigned fee1-dead Sep 4, 2026

sjwang05 commented Sep 4, 2026

View reviewed changes

rustbot assigned mati865 and unassigned Enselic Sep 4, 2026

ShoyuVanilla reviewed Sep 4, 2026

View reviewed changes

rustbot assigned ShoyuVanilla and unassigned mati865 Sep 4, 2026

rustbot assigned BoxyUwU Sep 4, 2026

rust-bors Bot added the S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. label Sep 4, 2026

rustbot unassigned ShoyuVanilla Sep 4, 2026

rust-bors Bot removed the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Sep 4, 2026

GuillaumeGomez mentioned this pull request Sep 4, 2026

Rollup of 28 pull requests #162307

Closed

GuillaumeGomez mentioned this pull request Sep 4, 2026

Rollup of 27 pull requests #162308

Closed

JonathanBrouwer mentioned this pull request Sep 4, 2026

Rollup of 25 pull requests #162310

Merged

rust-bors Bot merged commit 772a5dd into rust-lang:main Sep 5, 2026
13 checks passed

rustbot added this to the 1.100.0 milestone Sep 5, 2026

Navigation Menu

Uh oh!

Uh oh!

Avoid manually instantiating some binders in error reporting with `-Znext-solver` - #162262

Avoid manually instantiating some binders in error reporting with `-Znext-solver`#162262
rust-bors[bot] merged 2 commits into
rust-lang:mainfrom
sjwang05:forall-binder-ice

sjwang05 commented Sep 3, 2026 •

edited

Loading

Uh oh!

rustbot commented Sep 3, 2026

Uh oh!

rustbot commented Sep 3, 2026

Uh oh!

fee1-dead commented Sep 4, 2026 •

edited

Loading

Uh oh!

This comment was marked as resolved.

sjwang05 Sep 4, 2026 •

edited

Loading

Uh oh!

Enselic commented Sep 4, 2026

Uh oh!

ShoyuVanilla Sep 4, 2026 •

edited by rustbot

Loading

Uh oh!

mati865 commented Sep 4, 2026

Uh oh!

BoxyUwU commented Sep 4, 2026

Uh oh!

rust-bors Bot commented Sep 4, 2026

Uh oh!

Uh oh!

rust-timer commented Sep 5, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

Uh oh!

Uh oh!

Conversation

sjwang05 commented Sep 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Sep 3, 2026

Uh oh!

rustbot commented Sep 3, 2026

Uh oh!

fee1-dead commented Sep 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment was marked as resolved.

sjwang05 Sep 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Enselic commented Sep 4, 2026

Uh oh!

ShoyuVanilla Sep 4, 2026 • 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!

mati865 commented Sep 4, 2026

Uh oh!

BoxyUwU commented Sep 4, 2026

Uh oh!

rust-bors Bot commented Sep 4, 2026

Uh oh!

Uh oh!

rust-timer commented Sep 5, 2026

Overall result: no relevant changes - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

sjwang05 commented Sep 3, 2026 •

edited

Loading

fee1-dead commented Sep 4, 2026 •

edited

Loading

sjwang05 Sep 4, 2026 •

edited

Loading

ShoyuVanilla Sep 4, 2026 •

edited by rustbot

Loading