More robustly deal with relaxed bounds and improve their diagnostics #142693

fmease · 2025-06-18T20:23:53Z

Scaffolding for #135229 (CC #135331)

Fixes #136944 (6th commit).
Fixes #142718 (8th commit).

fmease · 2025-06-20T21:20:38Z

@bors2 try @rust-timer queue

rust-bors · 2025-06-20T21:20:42Z

⌛ Trying commit 04dad4e with merge bfa10de…

To cancel the try build, run the command @bors2 try cancel.

More robustly deal with relaxed bounds and improve their diagnostics Scaffolding for #135229 (CC #135331) Fixes #136944. Fixes #142718.

rust-bors · 2025-06-20T23:49:40Z

☀️ Try build successful (CI)
Build commit: bfa10de (bfa10de6407128926b18c60d51c9f19df73e336c, parent: 9c4ff566babe632af5e30281a822d1ae9972873b)

rust-timer · 2025-06-21T02:22:46Z

Finished benchmarking commit (bfa10de): comparison URL.

Overall result: ❌ regressions - please read the text below

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.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

@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)	3.0%	[3.0%, 3.0%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	3.0%	[3.0%, 3.0%]	1

Max RSS (memory usage)

Results (secondary -2.7%)

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)	2.4%	[2.4%, 2.4%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.5%	[-5.8%, -1.8%]	6
All ❌✅ (primary)	-	-	0

Cycles

Results (primary 2.9%, secondary 4.4%)

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

	mean	range	count
Regressions ❌ (primary)	2.9%	[2.9%, 2.9%]	1
Regressions ❌ (secondary)	4.4%	[2.8%, 7.6%]	10
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.9%	[2.9%, 2.9%]	1

Binary size

Results (primary 1.1%)

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

	mean	range	count
Regressions ❌ (primary)	1.1%	[1.1%, 1.1%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.1%	[1.1%, 1.1%]	1

Bootstrap: 692.606s -> 692.191s (-0.06%)
Artifact size: 372.01 MiB -> 371.99 MiB (-0.01%)

bors · 2025-06-27T20:17:23Z

☔ The latest upstream changes (presumably #143091) made this pull request unmergeable. Please resolve the merge conflicts.

fmease · 2025-06-27T22:56:30Z

#142693 (comment)

Looks spurious?

Only relevant to the internal feature `more_maybe_bounds`.

We no longer move trait predicates where the self ty is a ty param to "the bounds of a ty param".

* The phrasing "only does something for" made sense back when this diagnostic was a (hard) warning. Now however, it's simply a hard error and thus completely rules out "doing something". * The primary message was way too long * The new wording more closely mirrors the wording we use for applying other bound modifiers (like `const` and `async`) to incompatible traits. * "all other traits are not bound by default" is no longer accurate under Sized Hierarchy. E.g., traits and assoc tys are (currently) bounded by `MetaSized` by default but can't be relaxed using `?MetaSized` (instead, you relax it by adding `PointeeSized`). * I've decided against adding any diagnositic notes or suggestions for now like "trait `Trait` can't be relaxed as it's not bound by default" which would be incorrect for `MetaSized` and assoc tys as mentioned above) or "consider changing `?MetaSized` to `PointeeSized`" as the Sized Hierarchy impl is still WIP)

Having multiple relaxed bounds like `?Sized + ?Iterator` is actually *fine*. We actually want to reject *duplicate* relaxed bounds like `?Sized + ?Sized` because these most certainly represent a user error. Note that this doesn't mean that we accept more code because a bound like `?Iterator` is still invalid as it's not relaxing a *default* trait and the only way to define / use more default bounds is under the experimental and internal feature `more_maybe_bounds` plus `lang_items` plus unstable flag `-Zexperimental-default-bounds` (historical context: for the longest time, bounds like `?Iterator` were actually allowed and lead to a hard warning). Ultimately, this simply *reframes* the diagnostic. The scope of `more_maybe_bounds` / `-Zexperimental-default-bounds` remains unchanged as well.

fmease · 2025-07-18T10:30:22Z

(lol I should've just built the compiler for the rebase instead of doing a CI-directed one, that was painful, thanks tired me doing a night shift)

Rebased, added some more (doc) comments and got rid of the more_maybe_bounds advertisement (didn't do it as a follow-up so I can already use this as a good base branch for #135331.

@bors r=compiler-errors

bors · 2025-07-18T10:30:25Z

📌 Commit 82a02ae has been approved by compiler-errors

It is now in the queue for this repository.

fmease · 2025-07-18T10:32:02Z

@bors rollup- alleviate pressure

…er-errors More robustly deal with relaxed bounds and improve their diagnostics Scaffolding for rust-lang#135229 (CC rust-lang#135331) Fixes rust-lang#136944 (6th commit). Fixes rust-lang#142718 (8th commit).

Rollup of 9 pull requests Successful merges: - #138554 (Distinguish delim kind to decide whether to emit unexpected closing delimiter) - #142673 (Show the offset, length and memory of uninit read errors) - #142693 (More robustly deal with relaxed bounds and improve their diagnostics) - #143382 (stabilize `const_slice_reverse`) - #143928 (opt-dist: make llvm builds optional) - #143961 (Correct which exploit mitigations are enabled by default) - #144050 (Fix encoding of link_section and no_mangle cross crate) - #144059 (Refactor `CrateLoader` into the `CStore`) - #144123 (Generalize `unsize` and `unsize_into` destinations) r? `@ghost` `@rustbot` modify labels: rollup

rustbot added the T-compiler label Jun 18, 2025

fmease force-pushed the unbound-bettering branch from 8d31700 to 4634541 Compare June 18, 2025 20:27

fmease changed the title ~~[WIP] unbound bettering (refactoring)~~ [WIP] unbound bettering Jun 18, 2025

fmease force-pushed the unbound-bettering branch 2 times, most recently from 69e237d to 03d01fa Compare June 19, 2025 02:46

fmease added the rla-silenced label Jun 19, 2025

rust-lang deleted a comment from rust-log-analyzer Jun 19, 2025

fmease mentioned this pull request Jun 19, 2025

Sized hierarchy: PointeeSized bounds aren't fully validated during HIR ty lowering #142718

Open

fmease force-pushed the unbound-bettering branch 3 times, most recently from 8a6f65d to 04dad4e Compare June 20, 2025 21:17

fmease changed the title ~~[WIP] unbound bettering~~ More robustly deal with relaxed bounds and improve their diagnostics Jun 20, 2025

rust-lang deleted a comment from rust-bors bot Jun 20, 2025

This comment has been minimized.

Sign in to view

rust-bors bot added a commit that referenced this pull request Jun 20, 2025

Auto merge of #142693 - fmease:unbound-bettering, r=<try>

bfa10de

More robustly deal with relaxed bounds and improve their diagnostics Scaffolding for #135229 (CC #135331) Fixes #136944. Fixes #142718.

rustbot added the S-waiting-on-perf label Jun 20, 2025

fmease mentioned this pull request Jun 20, 2025

Rehome tests in tests/ui/ to accurate subdirectories #142808

Closed

This comment has been minimized.

Sign in to view

rustbot added perf-regression and removed S-waiting-on-perf labels Jun 21, 2025

bors added the S-waiting-on-author label Jun 27, 2025

fmease mentioned this pull request Jun 27, 2025

We don't fully validate relaxed trait bounds on the RHS of trait aliases (should we actually reject them outright?) #143122

Open

rustbot added the S-waiting-on-author label Jul 16, 2025

fmease force-pushed the unbound-bettering branch from 38ac3ca to d39def2 Compare July 17, 2025 23:51

fmease added the rla-silenced label Jul 17, 2025

fmease force-pushed the unbound-bettering branch from d39def2 to 505e8c1 Compare July 18, 2025 00:19

fmease added 4 commits July 18, 2025 03:13

HIR ty lowering: Simplify signature of lower_poly_trait_ref

2ce0b66

HIR ty lowering: Validate relaxed bounds in trait object types

10d7e5f

Only relevant to the internal feature `more_maybe_bounds`.

Update comment about where Ty:

9788f59

We no longer move trait predicates where the self ty is a ty param to "the bounds of a ty param".

AST lowering: More robustly deal with relaxed bounds

1df99f2

fmease force-pushed the unbound-bettering branch 3 times, most recently from eb7b67e to 94a6965 Compare July 18, 2025 03:07

fmease added 4 commits July 18, 2025 12:13

Reword diagnostics about relaxed bounds in invalid contexts

84ed70b

HIR ty lowering: Validate PointeeSized bounds

82a02ae

fmease force-pushed the unbound-bettering branch from 94a6965 to 82a02ae Compare July 18, 2025 10:25

bors added S-waiting-on-bors and removed S-waiting-on-author labels Jul 18, 2025

fmease mentioned this pull request Jul 18, 2025

[STACKED] Reject relaxed bounds inside associated type bounds (ATB) #135331

Open

fmease removed the rla-silenced label Jul 18, 2025

matthiaskrgr mentioned this pull request Jul 18, 2025

Rollup of 10 pull requests #144138

Closed

matthiaskrgr mentioned this pull request Jul 18, 2025

Rollup of 9 pull requests #144145

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

More robustly deal with relaxed bounds and improve their diagnostics #142693

More robustly deal with relaxed bounds and improve their diagnostics #142693

fmease commented Jun 18, 2025 •

edited

Loading

Uh oh!

fmease commented Jun 20, 2025

Uh oh!

This comment has been minimized.

rust-bors bot commented Jun 20, 2025

Uh oh!

rust-bors bot commented Jun 20, 2025

Uh oh!

This comment has been minimized.

rust-timer commented Jun 21, 2025

Uh oh!

bors commented Jun 27, 2025

Uh oh!

fmease commented Jun 27, 2025

Uh oh!

fmease commented Jul 18, 2025

Uh oh!

bors commented Jul 18, 2025

Uh oh!

fmease commented Jul 18, 2025

Uh oh!

Uh oh!

More robustly deal with relaxed bounds and improve their diagnostics #142693

Are you sure you want to change the base?

More robustly deal with relaxed bounds and improve their diagnostics #142693

Conversation

fmease commented Jun 18, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

fmease commented Jun 20, 2025

Uh oh!

This comment has been minimized.

rust-bors bot commented Jun 20, 2025

Uh oh!

rust-bors bot commented Jun 20, 2025

Uh oh!

This comment has been minimized.

rust-timer commented Jun 21, 2025

Overall result: ❌ regressions - please read the text below

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

bors commented Jun 27, 2025

Uh oh!

fmease commented Jun 27, 2025

Uh oh!

fmease commented Jul 18, 2025

Uh oh!

bors commented Jul 18, 2025

Uh oh!

fmease commented Jul 18, 2025

Uh oh!

Uh oh!

fmease commented Jun 18, 2025 •

edited

Loading