change HIR typeck region uniquification handling approach #145706

lcnr · 2025-08-21T10:16:46Z

#144405 causes structural lookup of opaque types to not work during HIR typeck, so instead avoid uniquifying goals and instead only reprove them if MIR borrowck actually encounters an error.

This doesn't perfectly maintain the property that HIR typeck succeeding implies that MIR typeck succeeds, instead weakening this check to only guarantee that HIR typeck implies that MIR typeck succeeds modulo region uniquification. This means we still get the actually desirable ICEs if we MIR building is broken or we forget to check some property in HIR typeck, without having to deal with the fallout of uniquification in HIR typeck itself.

We report errors using the original obligation sources of HIR typeck so diagnostics aren't that negatively impacted either.

Here's the history of region uniquification while working on the new trait solver:

r? @BoxyUwU

rustbot · 2025-08-21T10:16:53Z

Some changes occurred to the core trait solver

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

lcnr · 2025-08-21T10:16:55Z

@bors try @rust-timer queue

change HIR typeck unification handling approach

lcnr · 2025-08-21T10:19:40Z

compiler/rustc_trait_selection/src/traits/query/type_op/custom.rs

@@ -94,6 +97,8 @@ where
        let errors = ocx.select_all_or_error();
        if errors.is_empty() {
            Ok(value)
+        } else if let Err(guar) = infcx.tcx.check_potentially_region_dependent_goals(root_def_id) {
+            Err(guar)


this is the relevant change.

We store potentially region dependent goals in the TypeckResults and then reprove them in case MIR type check failed to prove something

rust-bors · 2025-08-21T12:31:52Z

☀️ Try build successful (CI)
Build commit: 792614a (792614acc44e937dc26ebf049538b29ac45fabe3, parent: 57e620e56b718abaf47f18e1e20a3cc3480b258b)

rust-timer · 2025-08-21T13:42:38Z

Finished benchmarking commit (792614a): comparison URL.

Overall result: ✅ improvements - no action needed

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

@bors rollup=never
@rustbot label: -S-waiting-on-perf -perf-regression

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.7%	[-2.5%, -1.0%]	6
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 3.2%, secondary 3.1%)

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

	mean	range	count
Regressions ❌ (primary)	3.2%	[1.2%, 5.4%]	3
Regressions ❌ (secondary)	3.1%	[2.3%, 3.8%]	7
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	3.2%	[1.2%, 5.4%]	3

Cycles

Results (secondary 2.1%)

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

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	2.1%	[2.1%, 2.1%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 469.9s -> 472.215s (0.49%)
Artifact size: 378.22 MiB -> 378.38 MiB (0.04%)

BoxyUwU

the code generally seems reasonable to me. I want to think more about whether this is what we actually want to do big picture wise though

View changes since this review

cjgillot · 2025-08-22T00:27:05Z

compiler/rustc_borrowck/src/type_check/canonical.rs

@@ -23,6 +24,7 @@ use crate::universal_regions::UniversalRegions;
 #[instrument(skip(infcx, constraints, op), level = "trace")]
 pub(crate) fn fully_perform_op_raw<'tcx, R: fmt::Debug, Op>(
    infcx: &BorrowckInferCtxt<'tcx>,
+    root_def_id: LocalDefId,


Do we need to thread root_def_id everywhere? Could we use tcx.typeck_root_def_id(body.source.def_id()) where we use it?

yes, that feels like it would slightly worsen perf though/just feels weird given that we've already got that value anyways 🤔 we use fully_perform_op_raw frequently. I personally don't mind passing this additional arg everywhere, so I would keep it as is for now

rustbot · 2025-08-22T11:39:47Z

This PR was rebased onto a different master 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.

BoxyUwU · 2025-08-22T13:17:43Z

@bors r+ rollup=never

I don't feel too strongly about the root_def_id thing, we can always change it after the fact

bors · 2025-08-22T13:17:46Z

📌 Commit 17ac2fc has been approved by BoxyUwU

It is now in the queue for this repository.

jhpratt · 2025-08-23T16:52:14Z

@bors p=1

putting ahead of rollup=iffy PR

bors · 2025-08-23T20:17:02Z

⌛ Testing commit 17ac2fc with merge 69b76df...

bors · 2025-08-23T23:29:36Z

☀️ Test successful - checks-actions
Approved by: BoxyUwU
Pushing 69b76df to master...

github-actions · 2025-08-23T23:32:22Z

What is this?

This is an experimental post-merge analysis report that shows differences in test outcomes between the merged PR and its parent PR.

Comparing c5a6a7b (parent) -> 69b76df (this PR)

Test differences

Show 94 test diffs

94 doctest diffs were found. These are ignored, as they are noisy.

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 69b76df90c7ea63b5350d1865f92902a0b27c9a2 --output-dir test-dashboard

And then open test-dashboard/index.html in your browser to see an overview of all executed tests.

Job duration changes

dist-aarch64-apple: 5957.3s -> 7095.1s (19.1%)
tidy: 212.1s -> 187.3s (-11.7%)
dist-aarch64-msvc: 5272.3s -> 5820.8s (10.4%)
aarch64-gnu-llvm-19-2: 2657.9s -> 2394.9s (-9.9%)
x86_64-gnu-llvm-20-1: 3830.3s -> 3471.0s (-9.4%)
x86_64-gnu-llvm-19: 2731.0s -> 2973.7s (8.9%)
dist-x86_64-apple: 6774.3s -> 7344.0s (8.4%)
dist-apple-various: 5106.3s -> 4740.3s (-7.2%)
dist-arm-linux-musl: 5957.4s -> 5545.5s (-6.9%)
x86_64-gnu-tools: 3997.7s -> 3735.2s (-6.6%)

How to interpret the job duration changes?

Job durations can vary a lot, based on the actual runner instance
that executed the job, system noise, invalidated caches, etc. The table above is provided
mostly for t-infra members, for simpler debugging of potential CI slow-downs.

rust-timer · 2025-08-24T00:40:40Z

Finished benchmarking commit (69b76df): comparison URL.

Overall result: ✅ improvements - no action needed

@rustbot label: -perf-regression

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.7%	[-2.5%, -0.9%]	6
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (secondary 2.2%)

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

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	2.8%	[2.4%, 3.4%]	7
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.1%	[-2.1%, -2.1%]	1
All ❌✅ (primary)	-	-	0

Cycles

Results (secondary 3.9%)

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

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.9%	[3.9%, 3.9%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 467.196s -> 466.542s (-0.14%)
Artifact size: 378.17 MiB -> 378.32 MiB (0.04%)

rustbot assigned BoxyUwU Aug 21, 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. WG-trait-system-refactor The Rustc Trait System Refactor Initiative (-Znext-solver) labels Aug 21, 2025

This comment has been minimized.

Sign in to view

rust-bors bot added a commit that referenced this pull request Aug 21, 2025

Auto merge of #145706 - lcnr:uniquification, r=<try>

792614a

change HIR typeck unification handling approach

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Aug 21, 2025

lcnr commented Aug 21, 2025

View reviewed changes

lcnr force-pushed the uniquification branch from 4b4904a to a8f4c7d Compare August 21, 2025 10:45

This comment has been minimized.

Sign in to view

lcnr changed the title ~~change HIR typeck unification handling approach~~ change HIR typeck region uniquification handling approach Aug 21, 2025

lcnr mentioned this pull request Aug 21, 2025

Support non-defining uses in HIR typeck #145711

Open

3 tasks

This comment has been minimized.

Sign in to view

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Aug 21, 2025

BoxyUwU approved these changes Aug 21, 2025

View reviewed changes

BoxyUwU 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 Aug 21, 2025

cjgillot reviewed Aug 22, 2025

View reviewed changes

change HIR typeck unification handling approach

17ac2fc

lcnr force-pushed the uniquification branch from a8f4c7d to 17ac2fc Compare August 22, 2025 11:39

bors 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-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Aug 22, 2025

bors added the merged-by-bors This PR was explicitly merged by bors. label Aug 23, 2025

bors merged commit 69b76df into rust-lang:master Aug 23, 2025
11 checks passed

rustbot added this to the 1.91.0 milestone Aug 23, 2025

bors mentioned this pull request Aug 24, 2025

When a trait isn't implemented, but another similar impl is found, point at it #145640

Open

change HIR typeck region uniquification handling approach #145706

change HIR typeck region uniquification handling approach #145706

Conversation

lcnr commented Aug 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Aug 21, 2025

Uh oh!

lcnr commented Aug 21, 2025

Uh oh!

This comment has been minimized.

This comment has been minimized.

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

rust-bors bot commented Aug 21, 2025

Uh oh!

This comment has been minimized.

rust-timer commented Aug 21, 2025

Overall result: ✅ improvements - no action needed

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!

rustbot commented Aug 22, 2025

Uh oh!

BoxyUwU commented Aug 22, 2025

Uh oh!

bors commented Aug 22, 2025

Uh oh!

jhpratt commented Aug 23, 2025

Uh oh!

bors commented Aug 23, 2025

Uh oh!

bors commented Aug 23, 2025

Uh oh!

Uh oh!

github-actions bot commented Aug 23, 2025

Test differences

Job duration changes

Uh oh!

rust-timer commented Aug 24, 2025

Overall result: ✅ improvements - no action needed

Uh oh!

Uh oh!

lcnr commented Aug 21, 2025 •

edited

Loading

BoxyUwU left a comment •

edited by rustbot

Loading