Make `validate_mir` ensure the final MIR for all bodies by compiler-errors · Pull Request #128612 · rust-lang/rust

compiler-errors · 2024-08-03T18:30:24Z

A lot of the crashes tests use -Zpolymorphize or -Zdump-mir for their side effect of computing the optimized_mir for all bodies, which will uncover bugs with late MIR passes like the inliner. I don't like having all these tests depend on -Zpolymorphize (or other hacky ways) for no reason, so this PR extends the -Zvalidate-mir flag to ensure optimized_mir/mir_for_ctfe for all body owners during the analysis phase.

Two thoughts:

This could be moved later in the compilation pipeline I guess? I don't really think it matters, though.
This could alternatively be expressed using a new flag, though I don't necessarily see much value in separating these.

For example, #128171 could have used this flag, in the tests/ui/polymorphization/inline-incorrect-early-bound.rs.

r? mir

compiler-errors · 2024-08-03T18:30:48Z

hi @rust-lang/wg-mir-opt anyone got strong feelings abt this?

compiler-errors · 2024-08-03T18:35:22Z

For the record, I have a fix for #128601 but want to put up better test that doesn't need the polymorphize flag to trigger it.

compiler-errors · 2024-08-03T18:53:26Z

Ah, let me just use the instance mir instead of opt mir, since we have some const body owners.

matthiaskrgr · 2024-08-03T19:01:50Z

this is like -Clink-dead-code but unused/unrechable mir?

davidtwco

LGTM, r=me unless you want to wait for others to chime in with an opinion

compiler-errors · 2024-08-08T14:56:37Z

@bors r=davidtwco rollup

bors · 2024-08-08T14:56:40Z

📌 Commit 470ada2 has been approved by davidtwco

It is now in the queue for this repository.

…ir, r=davidtwco Make `validate_mir` ensure the final MIR for all bodies A lot of the crashes tests use `-Zpolymorphize` or `-Zdump-mir` for their side effect of computing the `optimized_mir` for all bodies, which will uncover bugs with late MIR passes like the inliner. I don't like having all these tests depend on `-Zpolymorphize` (or other hacky ways) for no reason, so this PR extends the `-Zvalidate-mir` flag to ensure `optimized_mir`/`mir_for_ctfe` for all body owners during the analysis phase. Two thoughts: 1. This could be moved later in the compilation pipeline I guess? I don't really think it matters, though. 1. This could alternatively be expressed using a new flag, though I don't necessarily see much value in separating these. For example, rust-lang#128171 could have used this flag, in the `tests/ui/polymorphization/inline-incorrect-early-bound.rs`. r? mir

…iaskrgr Rollup of 7 pull requests Successful merges: - rust-lang#128306 (Update NonNull::align_offset quarantees) - rust-lang#128612 (Make `validate_mir` ensure the final MIR for all bodies) - rust-lang#128648 (Add regression test) - rust-lang#128791 (Don't implement `AsyncFn` for `FnDef`/`FnPtr` that wouldnt implement `Fn`) - rust-lang#128795 (Update E0517 message to reflect RFC 2195.) - rust-lang#128825 (rm `declared_features` field in resolver) - rust-lang#128826 (Only suggest `#[allow]` for `--warn` and `--deny` lint level flags) r? `@ghost` `@rustbot` modify labels: rollup

…ir, r=davidtwco Make `validate_mir` ensure the final MIR for all bodies A lot of the crashes tests use `-Zpolymorphize` or `-Zdump-mir` for their side effect of computing the `optimized_mir` for all bodies, which will uncover bugs with late MIR passes like the inliner. I don't like having all these tests depend on `-Zpolymorphize` (or other hacky ways) for no reason, so this PR extends the `-Zvalidate-mir` flag to ensure `optimized_mir`/`mir_for_ctfe` for all body owners during the analysis phase. Two thoughts: 1. This could be moved later in the compilation pipeline I guess? I don't really think it matters, though. 1. This could alternatively be expressed using a new flag, though I don't necessarily see much value in separating these. For example, rust-lang#128171 could have used this flag, in the `tests/ui/polymorphization/inline-incorrect-early-bound.rs`. r? mir

…iaskrgr Rollup of 7 pull requests Successful merges: - rust-lang#128306 (Update NonNull::align_offset quarantees) - rust-lang#128612 (Make `validate_mir` ensure the final MIR for all bodies) - rust-lang#128648 (Add regression test) - rust-lang#128749 (Mark `{f32,f64}::{next_up,next_down,midpoint}` inline) - rust-lang#128795 (Update E0517 message to reflect RFC 2195.) - rust-lang#128825 (rm `declared_features` field in resolver) - rust-lang#128826 (Only suggest `#[allow]` for `--warn` and `--deny` lint level flags) r? `@ghost` `@rustbot` modify labels: rollup

Rollup merge of rust-lang#128612 - compiler-errors:validate-mir-opt-mir, r=davidtwco Make `validate_mir` ensure the final MIR for all bodies A lot of the crashes tests use `-Zpolymorphize` or `-Zdump-mir` for their side effect of computing the `optimized_mir` for all bodies, which will uncover bugs with late MIR passes like the inliner. I don't like having all these tests depend on `-Zpolymorphize` (or other hacky ways) for no reason, so this PR extends the `-Zvalidate-mir` flag to ensure `optimized_mir`/`mir_for_ctfe` for all body owners during the analysis phase. Two thoughts: 1. This could be moved later in the compilation pipeline I guess? I don't really think it matters, though. 1. This could alternatively be expressed using a new flag, though I don't necessarily see much value in separating these. For example, rust-lang#128171 could have used this flag, in the `tests/ui/polymorphization/inline-incorrect-early-bound.rs`. r? mir

matthiaskrgr · 2024-08-08T21:54:37Z

somehow github does not recognize this as merged, closing manually..

compiler-errors · 2024-08-08T21:56:27Z

@matthiaskrgr: it's because bors merged an old version :(

Since you merge a lot of PRs and are likely to catch this: When you close PRs that GH doesn't mention, can you check that the version that is pushed is the one that shows up in the bors approval? #128612 (comment)

matthiaskrgr · 2024-08-08T22:00:11Z

oh wut? :O

usually it would show the commits being added after the rollup being mentioned in the pr history?

matthiaskrgr · 2024-08-08T22:02:07Z

oh I see, there are some comments in this pr that were not in the rollup...

…ir, r=matthiaskrgr Add comment that bors did not see pushed before it merged In rust-lang#128612, bors merged 470ada2 instead of 1e07c19. This means it dropped a useful comment I added, and a stage rename that is more descriptive.

bors · 2024-08-08T22:37:46Z

☔ The latest upstream changes (presumably #128835) made this pull request unmergeable. Please resolve the merge conflicts.

Rollup merge of rust-lang#128851 - compiler-errors:validate-mir-opt-mir, r=matthiaskrgr Add comment that bors did not see pushed before it merged In rust-lang#128612, bors merged 470ada2 instead of 1e07c19. This means it dropped a useful comment I added, and a stage rename that is more descriptive.

…iler-errors Do not compute optimized MIR if code does not type-check. Since rust-lang#128612, we compute optimized MIR when `-Zvalidate-mir` is present. This is done as part of required analyses, even if type-checking fails. This causes ICEs, as most of the mir-opt pipeline expects well-formed code. Fixes rust-lang#129095 Fixes rust-lang#134174 Fixes rust-lang#134654 Fixes rust-lang#136381 Fixes rust-lang#137468 Fixes rust-lang#144491 Fixes rust-lang#147011 This does not fix issue rust-lang#137190, as it ICEs without `-Zvalidate-mir`. r? `@compiler-errors`

…iler-errors Do not compute optimized MIR if code does not type-check. Since rust-lang#128612, we compute optimized MIR when `-Zvalidate-mir` is present. This is done as part of required analyses, even if type-checking fails. This causes ICEs, as most of the mir-opt pipeline expects well-formed code. Fixes rust-lang#129095 Fixes rust-lang#134174 Fixes rust-lang#134654 Fixes rust-lang#135570 Fixes rust-lang#136381 Fixes rust-lang#137468 Fixes rust-lang#144491 Fixes rust-lang#147011 This does not fix issue rust-lang#137190, as it ICEs without `-Zvalidate-mir`. r? `@compiler-errors`

…iler-errors Do not compute optimized MIR if code does not type-check. Since rust-lang#128612, we compute optimized MIR when `-Zvalidate-mir` is present. This is done as part of required analyses, even if type-checking fails. This causes ICEs, as most of the mir-opt pipeline expects well-formed code. Fixes rust-lang#129095 Fixes rust-lang#134174 Fixes rust-lang#134654 Fixes rust-lang#135570 Fixes rust-lang#136381 Fixes rust-lang#137468 Fixes rust-lang#144491 Fixes rust-lang#147011 This does not fix issue rust-lang#137190, as it ICEs without `-Zvalidate-mir`. r? ``@compiler-errors``

Rollup merge of #147092 - cjgillot:late-validate-mir, r=compiler-errors Do not compute optimized MIR if code does not type-check. Since #128612, we compute optimized MIR when `-Zvalidate-mir` is present. This is done as part of required analyses, even if type-checking fails. This causes ICEs, as most of the mir-opt pipeline expects well-formed code. Fixes #129095 Fixes #134174 Fixes #134654 Fixes #135570 Fixes #136381 Fixes #137468 Fixes #144491 Fixes #147011 This does not fix issue #137190, as it ICEs without `-Zvalidate-mir`. r? ``@compiler-errors``

Do not compute optimized MIR if code does not type-check. Since rust-lang/rust#128612, we compute optimized MIR when `-Zvalidate-mir` is present. This is done as part of required analyses, even if type-checking fails. This causes ICEs, as most of the mir-opt pipeline expects well-formed code. Fixes rust-lang/rust#129095 Fixes rust-lang/rust#134174 Fixes rust-lang/rust#134654 Fixes rust-lang/rust#135570 Fixes rust-lang/rust#136381 Fixes rust-lang/rust#137468 Fixes rust-lang/rust#144491 Fixes rust-lang/rust#147011 This does not fix issue rust-lang/rust#137190, as it ICEs without `-Zvalidate-mir`. r? ``@compiler-errors``

rustbot assigned davidtwco Aug 3, 2024

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 Aug 3, 2024

compiler-errors force-pushed the validate-mir-opt-mir branch from bfb8ab6 to f48fe22 Compare August 3, 2024 18:32

This comment has been minimized.

Sign in to view

compiler-errors force-pushed the validate-mir-opt-mir branch 2 times, most recently from 74dace7 to 470ada2 Compare August 3, 2024 19:18

compiler-errors changed the title ~~Make validate_mir ensure optimized MIR for all bodies~~ Make validate_mir ensure the final MIR for all bodies Aug 3, 2024

davidtwco approved these changes Aug 6, 2024

View reviewed changes

Make validate_mir pull optimized/ctfe MIR for all bodies

1e07c19

compiler-errors force-pushed the validate-mir-opt-mir branch from 470ada2 to 1e07c19 Compare August 8, 2024 14:56

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-review Status: Awaiting review from the assignee but also interested parties. labels Aug 8, 2024

matthiaskrgr mentioned this pull request Aug 8, 2024

Rollup of 7 pull requests #128829

Closed

matthiaskrgr mentioned this pull request Aug 8, 2024

Rollup of 7 pull requests #128835

Merged

matthiaskrgr closed this Aug 8, 2024

compiler-errors deleted the validate-mir-opt-mir branch August 8, 2024 21:57

compiler-errors mentioned this pull request Aug 8, 2024

Add comment that bors did not see pushed before it merged #128851

Merged

bors 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-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Aug 8, 2024

This was referenced Aug 9, 2024

ICE non-Aggregate field with matching ABI but differing alignment #128870

Closed

ICE: tcx: index out of bounds #129075

Closed

matthiaskrgr mentioned this pull request Sep 28, 2024

ICE: expected wide pointer extra data #130970

Closed

matthiaskrgr mentioned this pull request Oct 7, 2024

ICE: primitive read not possible for type #131347

Closed

workingjubilee mentioned this pull request Oct 19, 2024

compiler: Error on layout of enums with invalid reprs #131843

Merged

matthiaskrgr mentioned this pull request Nov 15, 2024

ICE: is not a pointer or reference type #133065

Closed

matthiaskrgr mentioned this pull request Nov 29, 2024

ICE: entered unreachable code: FieldsShape::offset: Primitives have no fields #133624

Closed

meithecatte mentioned this pull request Apr 28, 2025

ICE: layout.is_sized() #139872

Closed

cjgillot mentioned this pull request Sep 27, 2025

Do not compute optimized MIR if code does not type-check. #147092

Merged

Uh oh!

Conversation

compiler-errors commented Aug 3, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

compiler-errors commented Aug 3, 2024

Uh oh!

compiler-errors commented Aug 3, 2024

Uh oh!

compiler-errors commented Aug 3, 2024

Uh oh!

This comment has been minimized.

matthiaskrgr commented Aug 3, 2024

Uh oh!

davidtwco left a comment

Choose a reason for hiding this comment

Uh oh!

compiler-errors commented Aug 8, 2024

Uh oh!

bors commented Aug 8, 2024

Uh oh!

matthiaskrgr commented Aug 8, 2024

Uh oh!

compiler-errors commented Aug 8, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

matthiaskrgr commented Aug 8, 2024

Uh oh!

matthiaskrgr commented Aug 8, 2024

Uh oh!

bors commented Aug 8, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

compiler-errors commented Aug 3, 2024 •

edited

Loading

compiler-errors commented Aug 8, 2024 •

edited

Loading