create mir for direct-rhs consts by khyperia · Pull Request #165070 · rust-lang/rust

khyperia · 2026-10-28T09:06:40Z

Alternative to #164982 - see that PR for context.

When referring to a direct constant via a mir::Const::Unevaluated, synthesize a MIR body, whose contents are simply returning a mir::Const::Ty of itself. This effectively provides both a MIR view, as well as a typesystem view, of the same constant, without risking representing the same constant in different ways in the two different representations, because one is just referring to the other.

r? @BoxyUwU

fyi @oli-obk @lcnr

BoxyUwU · 2026-10-28T10:18:11Z


    let root = tcx.typeck_root_def_id_local(def);
-    if let Err(e) = tcx.check_transmutes(root) {
+    if !tcx.is_synthetic_mir(root)


why this :3

View changes since the review

check_transmutes calls typeck, typeck requires HIR, synthetic MIR does not have HIR.

previously, the only synthetic MIR was DefKind::SyntheticCoroutineBody, which is a typeck child, so this call to check_transmutes (which must be called on typeck roots) didn't care and we just happened to be correct here

rustbot · 2026-10-28T13:57:27Z

BoxyUwU is currently at their maximum review capacity.
They may take a while to respond.

khyperia · 2026-10-28T13:57:43Z

The test differences are generally due to, if the direct const item being referenced by a mir::Const is not WF:

Const::Unevaluated bails immediately due to the referenced item's synthesized MIR Body::tainted_by_errors and does not emit additional errors. The taint comes from a tcx.ensure_result().check_well_formed().
Const::Ty does not, we just call the try_normalize_generic_arg_after_erasing_regions query and it returns NoSolution and we're like "idk why that is lol, gonna report another error just in case"
- I think this could match behavior to the mir-wrapper version by returning ErrorHandled::Reported instead of ErrorHandled::TooGeneric here if the trait solver could tell us it resolved to a not-wf impl, but not in this PR IMO
  
  rust/compiler/rustc_const_eval/src/interpret/eval_context.rs
  
  Line 356 in 41089d0
  
  .map_err(|_| ErrorHandled::TooGeneric(self.cur_span()))

rustbot · 2026-10-28T14:01:47Z

Some changes occurred to the CTFE machinery

cc @RalfJung, @oli-obk, @lcnr

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

rustbot · 2026-11-02T10:13:55Z

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.

BoxyUwU · 2026-11-02T14:18:20Z

+            } else {
+                let args = ty::GenericArgs::identity_for_item(tcx, def_id);
+                let kind = ty::AliasConstKind::InherentImpl { def_id: def_id.into() };
+                ty::AliasConst::new(tcx, kind, args)


can you add a comment for why we don't use mir::Const::Ty(const_of_item(...)), then r=me

View changes since the review

BoxyUwU · 2026-11-02T14:19:28Z

@rustbot author

this is a much smaller hack and doesn't touch const eval so im comfortable approving this without talking to const eval folks

khyperia · 2026-11-02T14:38:49Z

@bors r=BoxyUwU rollup

rust-bors · 2026-11-02T14:38:52Z

📌 Commit f29e22f has been tentatively approved by BoxyUwU

It will be put into the queue for this repository once PR CI succeeds.

…uwer Rollup of 6 pull requests Successful merges: - #163764 (Tweak suggestion of "unreachable due to never pattern") - #164771 (std: fix platform checks for hermit) - #165070 (create mir for direct-rhs consts) - #165121 (test DSL evaluate sequentially) - #165288 (Add assembly test for masked index without redundant shifts) - #165307 (test(lexer): Add test that ZWJ/ZWNJ are accepted in identifiers)

@BoxyUwU

Rollup merge of #165070 - khyperia:mir_for_direct_const, r=BoxyUwU create mir for direct-rhs consts Alternative to #164982 - see that PR for context. When referring to a direct constant via a `mir::Const::Unevaluated`, synthesize a MIR body, whose contents are simply returning a `mir::Const::Ty` of itself. This effectively provides both a MIR view, as well as a typesystem view, of the same constant, without risking representing the same constant in different ways in the two different representations, because one is just referring to the other. r? @BoxyUwU fyi @oli-obk @lcnr

rust-timer · 2026-11-03T07:18:49Z

Note

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

Finished benchmarking commit (e7363cb): comparison URL.

Overall result: ❌✅ regressions and improvements - please read:

Our benchmarks found a performance regression caused by this PR.
This might be an actual regression, but it can also be just noise.

Next Steps:

If the regression was expected or you think it can be justified,
please write a comment with sufficient written justification, and add
@rustbot label: +perf-regression-triaged to it, to mark the regression as triaged.
If you think that you know of a way to resolve the regression, try to create
a new PR with a fix for the regression.
If you do not understand the regression or you think that it is just noise,
you can ask the @rust-lang/wg-compiler-performance working group for help (members of this group
were already notified of this PR).

@rustbot label: +perf-regression
cc @rust-lang/wg-compiler-performance

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)	0.0%	[0.0%, 0.0%]	1
Improvements ✅ (primary)	-0.2%	[-0.2%, -0.2%]	3
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.2%	[-0.2%, -0.2%]	3

Max RSS (memory usage)

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

Cycles

Results (secondary -3.6%)

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)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.6%	[-3.6%, -3.6%]	1
All ❌✅ (primary)	-	-	0

Binary size

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

Artifact size: 406.63 MiB -> 406.59 MiB (-0.01%)

Kobzol · 2026-11-08T19:05:48Z

Noise.

@rustbot label: +perf-regression-triaged

rustbot added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Oct 28, 2026

BoxyUwU reviewed Oct 28, 2026

View reviewed changes

rustbot assigned BoxyUwU Oct 28, 2026

khyperia force-pushed the mir_for_direct_const branch from 6bdcf68 to beca84e Compare October 28, 2026 14:01

khyperia marked this pull request as ready for review October 28, 2026 14:01

rustbot added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Oct 28, 2026

rustbot removed the S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. label Oct 28, 2026

panstromek mentioned this pull request Oct 29, 2026

Identify trivial consts without building MIR #165112

Draft

This comment has been minimized.

Sign in to view

khyperia force-pushed the mir_for_direct_const branch from beca84e to 6a9928d Compare November 2, 2026 10:13

BoxyUwU reviewed Nov 2, 2026

View reviewed changes

rustbot 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 Nov 2, 2026

create mir for direct-rhs consts

f29e22f

khyperia force-pushed the mir_for_direct_const branch from 6a9928d to f29e22f Compare November 2, 2026 14:36

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-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Nov 2, 2026

khyperia mentioned this pull request Nov 2, 2026

teach Const::Unevaluated to handle direct consts #164982

Closed

JonathanBrouwer mentioned this pull request Nov 2, 2026

Rollup of 6 pull requests #165361

Merged

rust-bors Bot merged commit eed4130 into rust-lang:main Nov 2, 2026
14 checks passed

rustbot added this to the 1.101.0 milestone Nov 2, 2026

rustbot added the perf-regression Performance regression. label Nov 3, 2026

khyperia deleted the mir_for_direct_const branch November 3, 2026 08:05

khyperia mentioned this pull request Nov 4, 2026

wf-check types of ty-level consts before evaluating them in KnownPanicslint #164294

Open

BoxyUwU mentioned this pull request Nov 6, 2026

Full Const Generics rust-lang/goals#100

Open

This was referenced Nov 6, 2026

[ICE]: erroneous constant missed by mono item collection #162337

Open

[ICE]: shouldn't have created a ValTree for pattern_type!(u32 is 0..=999999999) #162336

Open

rustc 1.100 Rust-GPU/rust-gpu#664

Open

zakrad mentioned this pull request Nov 8, 2026

Add regression test for GCA associated const with a pattern type field #165652

Open

rustbot added the perf-regression-triaged The performance regression has been triaged. label Nov 8, 2026

Conversation

khyperia commented Oct 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

BoxyUwU Oct 28, 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!

khyperia Oct 28, 2026

Choose a reason for hiding this comment

Uh oh!

rustbot commented Oct 28, 2026

Uh oh!

khyperia commented Oct 28, 2026

Uh oh!

rustbot commented Oct 28, 2026

Uh oh!

This comment has been minimized.

rustbot commented Nov 2, 2026

Uh oh!

BoxyUwU Nov 2, 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!

BoxyUwU commented Nov 2, 2026

Uh oh!

khyperia commented Nov 2, 2026

Uh oh!

rust-bors Bot commented Nov 2, 2026

Uh oh!

Uh oh!

rust-timer commented Nov 3, 2026

Overall result: ❌✅ regressions and improvements - please read:

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

Kobzol commented Nov 8, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

khyperia commented Oct 28, 2026 •

edited

Loading

BoxyUwU Oct 28, 2026 •

edited by rustbot

Loading

BoxyUwU Nov 2, 2026 •

edited by rustbot

Loading