Flip "region lattice" in RegionKind doc comment by khyperia · Pull Request #152968 · rust-lang/rust

khyperia · 2026-02-22T11:50:14Z

If we assume that references take their region covariantly, and use that to define the subtyping relationship of regions, 'empty is ⊤ (top) and 'static is ⊥ (bottom), and that 'a <: 'b is defined as 'a >= 'b. However, the RegionKind doc comment's "region lattice" had 'static at Top. While this is perhaps technically correct (a so-called "region lattice" is not a "type lattice"), it confused me a lot, and took me half an hour to be like "no, ok, I do understand things correctly, this region lattice is just flipped from the subtype lattice". Seems better to just have them be the same!

I also took the opportunity to rewrite some parts of the comment with notes that would have helped me when starting out.

Bikesheds welcome (as long as they don't go on forever)!

For context: the comment "'static is Top" was written in 2013, before there was a discussion in 2014 that clarified that references take their regions covariantly, and 'a <: 'b is defined as 'a >= 'b.

See also Zulip thread: https://rust-lang.zulipchat.com/#narrow/channel/131828-t-compiler/topic/RegionKind.20rustc.20doc.20comment

rustbot · 2026-02-22T11:50:18Z

r? @jieyouxu

rustbot has assigned @jieyouxu.
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, types
compiler, types expanded to 68 candidates
Random selection from 15 candidates

jieyouxu · 2026-02-22T14:45:10Z

@rustbot reroll

BoxyUwU · 2026-02-23T09:22:18Z

r? me

BoxyUwU · 2026-02-24T12:36:55Z

compiler/rustc_type_ir/src/region_kind.rs

-/// function declaration. They have relationships to one another
-/// determined based on the declared relationships from the
-/// function.
+/// Early-bound/free regions ([`RegionKind::ReEarlyParam`]) are the


I think this comment is out of date in that "free regions" here refers to a previously existing RegionKind::Free which got renamed to ReLateParam. This should probably say something like "Early/LateParam regions are the..."

alternatively we did actually mean free regions in which case this includes 'static but talking explicitly about early bound regions is confusing...

maybe this should actually be "Free regions (ReEarlyParam ReLateParam ReStatic) are the named lifetimes in scope from.." i'm not sure 😅

"Early/LateParam regions are the named lifetimes in scope from the function declaration. They have relationships to one another and 'static based on .." is probably the most useful thing to document 🤔

Early/LateParam regions are the named lifetimes in scope from the function declaration

late params can be anonymous, right? just gonna remove the word "named".

Also hm, ReLateParam can be constructed from HKT ReBound stuff (liberate_late_bound_regions), so they're not always "from the function declaration". Maybe flipping the sentence:

Lifetimes in scope from a function declaration are represented via ReEarlyParam/ReLateParam. They have relationships to one another and 'static based on the declared relationships from the function.

because all lifetimes from a function declaration are ReEarlyParam/ReLateParam, but not all ReLateParam are from a function declaration.

BoxyUwU · 2026-03-05T15:16:34Z

compiler/rustc_type_ir/src/region_kind.rs

 ///
 /// ## Bound Regions
 ///
 /// These are regions that are stored behind a binder and must be instantiated


I think the pre-existing comment here is wrong/wildly misleading. Early bound parameters are not ReBound. This comment implies that a ReBound can correspond to either an early or a late bound parameter but actually it just always corresponds to a lifetime from a for<'a, ...> binder

I would probably drop everything about early/late here and only talk about them in relation to the Binder type

just pushed an edit, how's this?

BoxyUwU

thanks yeah this looks good now :)

View changes since this review

BoxyUwU · 2026-03-13T14:59:23Z

compiler/rustc_type_ir/src/region_kind.rs

+/// The distinction between early and late exists because higher-ranked
+/// lifetimes aren't supported in all places. See [1] [2].


This is a fairly complex topic. I think we should just link to the dev guide chapter on early/late parameters and go "read here for why there are two kinds of parameters"

Agreed, I just copied this from the original text to preserve it somewhere.

Do you think it's worth it to preserve the links to the blog posts linked here, or should we just link to the dev guide and delete the links to these posts?

hmm, can you drop them but link to them from the dev guide? you should be able to edit it at src/doc/rustc-dev-guide/src/early-late-parameters.md or something like that

probably like a note at the very start of that chapter that goes "see also these blog posts from when this distinction was introduced"

rustbot · 2026-03-13T16:04:50Z

The rustc-dev-guide subtree was changed. If this PR only touches the dev guide consider submitting a PR directly to rust-lang/rustc-dev-guide otherwise thank you for updating the dev guide with your changes.

cc @BoxyUwU, @tshepang

BoxyUwU · 2026-03-17T12:23:08Z

wonderful!

@bors r+ rollup

rust-bors · 2026-03-17T12:23:12Z

📌 Commit 3eb071e has been approved by BoxyUwU

It is now in the queue for this repository.

…yUwU Flip "region lattice" in RegionKind doc comment If we assume that references take their region covariantly, and use that to define the subtyping relationship of regions, `'empty` is ⊤ (top) and `'static` is ⊥ (bottom), and that `'a <: 'b` is defined as `'a >= 'b`. However, the RegionKind doc comment's "region lattice" had `'static` at Top. While this is perhaps technically correct (a so-called "region lattice" is not a "type lattice"), it confused me a lot, and took me half an hour to be like "no, ok, I *do* understand things correctly, this region lattice is just flipped from the subtype lattice". Seems better to just have them be the same! I also took the opportunity to rewrite some parts of the comment with notes that would have helped me when starting out. Bikesheds welcome (as long as they don't go on forever)! For context: the comment "`'static` is Top" was [written in 2013](rust-lang@a896440#diff-8780054cdf09361c4ac2540c7f544ecfdc52a9e610822aabb8bcd2964affcb1cR430), before there was a [discussion in 2014](rust-lang#15699) that clarified that references take their regions covariantly, and `'a <: 'b` is defined as `'a >= 'b`. See also Zulip thread: https://rust-lang.zulipchat.com/#narrow/channel/131828-t-compiler/topic/RegionKind.20rustc.20doc.20comment

Rollup of 7 pull requests Successful merges: - #153801 (Add the option to run UI tests with the parallel frontend) - #153967 (Tweak wording of failed predicate in inference error) - #152968 (Flip "region lattice" in RegionKind doc comment) - #153531 (Fix LegacyKeyValueFormat report from docker build: various) - #153709 (Fix hypothetical ICE in `variances_of`) - #153884 (test `classify-runtime-const` for `f16`) - #153946 (dissolve `tests/ui/cross`)

…uwer Rollup of 14 pull requests Successful merges: - #153972 (stdarch subtree update) - #153801 (Add the option to run UI tests with the parallel frontend) - #153959 (Fix non-module `parent_module` in stripped cfg diagnostics) - #153967 (Tweak wording of failed predicate in inference error) - #152968 (Flip "region lattice" in RegionKind doc comment) - #153531 (Fix LegacyKeyValueFormat report from docker build: various) - #153622 (remove concept of soft-unstable features) - #153709 (Fix hypothetical ICE in `variances_of`) - #153884 (test `classify-runtime-const` for `f16`) - #153894 (Point at unit structs on foreign crates in type errors when they are the pattern of a binding) - #153920 (improve `#[track_caller]` invalid ABI error) - #153946 (dissolve `tests/ui/cross`) - #153965 (Fix minor kasan bugs) - #153991 (Small report_cycle refactor)

…uwer Rollup of 14 pull requests Successful merges: - rust-lang/rust#153972 (stdarch subtree update) - rust-lang/rust#153801 (Add the option to run UI tests with the parallel frontend) - rust-lang/rust#153959 (Fix non-module `parent_module` in stripped cfg diagnostics) - rust-lang/rust#153967 (Tweak wording of failed predicate in inference error) - rust-lang/rust#152968 (Flip "region lattice" in RegionKind doc comment) - rust-lang/rust#153531 (Fix LegacyKeyValueFormat report from docker build: various) - rust-lang/rust#153622 (remove concept of soft-unstable features) - rust-lang/rust#153709 (Fix hypothetical ICE in `variances_of`) - rust-lang/rust#153884 (test `classify-runtime-const` for `f16`) - rust-lang/rust#153894 (Point at unit structs on foreign crates in type errors when they are the pattern of a binding) - rust-lang/rust#153920 (improve `#[track_caller]` invalid ABI error) - rust-lang/rust#153946 (dissolve `tests/ui/cross`) - rust-lang/rust#153965 (Fix minor kasan bugs) - rust-lang/rust#153991 (Small report_cycle refactor)

rustbot assigned jieyouxu Feb 22, 2026

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 Feb 22, 2026

rustbot assigned wesleywiser and unassigned jieyouxu Feb 22, 2026

rustbot assigned BoxyUwU and unassigned wesleywiser Feb 23, 2026

BoxyUwU reviewed Feb 24, 2026

View reviewed changes

khyperia force-pushed the regionkind-comment branch from bbbbc92 to 8cca904 Compare February 25, 2026 07:49

BoxyUwU reviewed Mar 5, 2026

View reviewed changes

khyperia force-pushed the regionkind-comment branch 2 times, most recently from d0ac5b1 to 40374b4 Compare March 8, 2026 12:20

BoxyUwU reviewed Mar 13, 2026

View reviewed changes

Flip "region lattice" in RegionKind doc comment

3eb071e

khyperia force-pushed the regionkind-comment branch from 40374b4 to 3eb071e Compare March 13, 2026 16:04

rustbot added the A-rustc-dev-guide Area: rustc-dev-guide label Mar 13, 2026

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 Mar 17, 2026

Zalathar mentioned this pull request Mar 17, 2026

Rollup of 7 pull requests #154001

Closed

JonathanBrouwer mentioned this pull request Mar 17, 2026

Rollup of 14 pull requests #154008

Merged

rust-bors bot merged commit f7c9081 into rust-lang:main Mar 17, 2026
11 checks passed

khyperia deleted the regionkind-comment branch March 17, 2026 21:39

		/// The distinction between early and late exists because higher-ranked
		/// lifetimes aren't supported in all places. See [1] [2].

Uh oh!

Conversation

khyperia commented Feb 22, 2026

Uh oh!

rustbot commented Feb 22, 2026

Uh oh!

jieyouxu commented Feb 22, 2026

Uh oh!

BoxyUwU commented Feb 23, 2026

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!

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!

rustbot commented Mar 13, 2026

Uh oh!

BoxyUwU commented Mar 17, 2026

Uh oh!

rust-bors bot commented Mar 17, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

BoxyUwU left a comment •

edited by rustbot

Loading