tests: fix: parallel frontend test failures: different alloc ids by heinwol · Pull Request #156716 · rust-lang/rust

heinwol · 2026-05-18T14:31:26Z

Removed the //@ ignore-parallel-frontend different alloc ids directive from all the ui tests applicable and replaced it with //@ normalize-stderr so that all alloc ids mentions are converted into a single placeholder. Thus the tests are passed

Part of #154314

r? @petrochenkov

petrochenkov · 2026-05-18T14:50:04Z

--bless needs to be run in single-threaded mode to avoid reordering diagnostics in .stderr files.
@rustbot author

rustbot · 2026-05-18T14:50:09Z

Reminder, once the PR becomes ready for a review, use @rustbot ready.

heinwol · 2026-05-18T15:15:21Z

@rustbot ready

petrochenkov · 2026-05-18T15:50:48Z

r? @RalfJung
Is this a reasonable normalization?

My assumptions were that

specific numbers of allocation IDs in const evaluation are not that important
it's hard to make the ID assignment order reproducible if various parts of const evaluator can run in parallel

So it's better to just normalize the IDs away and enable these tests during parallel frontend testing, instead of ignoring them or waiting for some potential fix in the compiler.

rustbot · 2026-05-18T15:50:52Z

RalfJung is not on the review rotation at the moment.
They may take a while to respond.

petrochenkov · 2026-05-18T16:31:20Z

tests/rustdoc-gui/toggle-click-deadspace.goml

Spurious failure.

RalfJung · 2026-05-24T11:44:39Z

A while back @oli-obk wrote some code somewhere to renumber AllocId while ensuring different ID remain differentiable. Those are the IDs that you are currently seeing in the test output (ALLOC0 etc). This code now applies a second round of normalization on top of the already existing normalization, which seems a bit silly. So at the very least, if we anyway remove all alloc IDs from the result, we should remove that code (or rather, replace it by a simple normalization that we apply everywhere so that one doesn't have to add so many //@ normalize-stderr).

On the one hand it is a bit unfortunate if we have no way of testing whether we actually print consistent IDs. On the other hand these IDs are mostly meaningless and using them in diagnostics isn't great to begin with.
@rust-lang/wg-const-eval what do you think?

oli-obk · 2026-05-25T06:06:38Z

Seems fine as a short term solution. Having it as a global filter does seem better than repeating it.

There's basically two orthogonal things that we should probably do:

parent the AllocIds properly (like disambiguators for many things defined in one parent)
add spanned notes to errors pointing to where the alloc was created instead of rendering their name (this may be enough on its own)

petrochenkov · 2026-05-25T06:41:33Z

add spanned notes to errors pointing to where the alloc was created instead of rendering their name (this may be enough on its own)

@heinwol could you try this today?
Hopefully, obtaining the "where the alloc was created" span is already possible in the current infra.

oli-obk · 2026-05-25T08:25:34Z

Hopefully, obtaining the "where the alloc was created" span is already possible in the current infra.

Unfortunately not. There are obvious ones like the ones pointing to statics or fns, but we don't have spans in AllocExtra of the compile_time_machine macro decl. I don't know about the perf impact of doing that, but with relative spans it could just be zero

RalfJung · 2026-05-26T06:04:12Z

Yeah getting span info is a Project on its own. It'd be nice to have but shouldn't block the parallel frontend IMO.

Personally I could live with normalizing all AllocId away.

in `tests/ui/consts/const-eval/raw-bytes.rs` as we now handle this in our modified regex in `runtest`

Specifically, all `//@ ignore-parallel-frontend different alloc ids`

across other tests because they fail with our normalization strategy: `ALLOC0` -> `ALLOC$ID`

heinwol · 2026-07-01T13:54:57Z

@rustbot ready

petrochenkov · 2026-07-01T14:01:24Z

r=me after squashing commits.
@rustbot author

petrochenkov · 2026-07-01T14:05:10Z

r=me after squashing commits. @rustbot author

Or actually not, there're no back and forth changes, just multiple steps.
@bors r+

rust-bors · 2026-07-01T14:05:14Z

📌 Commit c7a0991 has been approved by petrochenkov

It is now in the queue for this repository.

@petrochenkov

…llel-frontend, r=petrochenkov tests: fix: parallel frontend test failures: different alloc ids Removed the `//@ ignore-parallel-frontend different alloc ids` directive from all the ui tests applicable and replaced it with `//@ normalize-stderr` so that all alloc ids mentions are converted into a single placeholder. Thus the tests are passed Part of [rust-lang#154314](rust-lang#154314) r? @petrochenkov

…uwer Rollup of 8 pull requests Successful merges: - #150075 (Implement clamp_to) - #156716 (tests: fix: parallel frontend test failures: different alloc ids) - #158397 (delegation: support simplest output `Self` mapping) - #158613 (Fix getrandom fallback test on platforms with `panic=abort`) - #158620 (Remove skip_norm_w/i/p().def_id with a helper) - #158633 (Remove unnecessary `Clone` derives on resolver types) - #158634 (Add missing `needs_drop` check to `DroplessArena`.) - #158647 (Document `strip_circumfix` behavior on overlapping prefix and suffix.)

…uwer Rollup of 8 pull requests Successful merges: - #158294 (Use .drectve for MSVC DLL exports) - #156716 (tests: fix: parallel frontend test failures: different alloc ids) - #158397 (delegation: support simplest output `Self` mapping) - #158613 (Fix getrandom fallback test on platforms with `panic=abort`) - #158620 (Remove skip_norm_w/i/p().def_id with a helper) - #158633 (Remove unnecessary `Clone` derives on resolver types) - #158634 (Add missing `needs_drop` check to `DroplessArena`.) - #158647 (Document `strip_circumfix` behavior on overlapping prefix and suffix.)

@petrochenkov

Rollup merge of #156716 - heinwol:fix-ui-tests-alloc-id-parallel-frontend, r=petrochenkov tests: fix: parallel frontend test failures: different alloc ids Removed the `//@ ignore-parallel-frontend different alloc ids` directive from all the ui tests applicable and replaced it with `//@ normalize-stderr` so that all alloc ids mentions are converted into a single placeholder. Thus the tests are passed Part of [#154314](#154314) r? @petrochenkov

rustbot assigned petrochenkov May 18, 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 May 18, 2026

This comment has been minimized.

Sign in to view

heinwol force-pushed the fix-ui-tests-alloc-id-parallel-frontend branch 2 times, most recently from f681281 to 5abbf45 Compare May 18, 2026 14:41

heinwol changed the title ~~fix: parallel frontend test failures: different alloc ids~~ tests: fix: parallel frontend test failures: different alloc ids May 18, 2026

petrochenkov reviewed May 18, 2026

View reviewed changes

Comment thread tests/ui/const-generics/issues/issue-100313.rs Outdated

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 May 18, 2026

heinwol force-pushed the fix-ui-tests-alloc-id-parallel-frontend branch from 5abbf45 to bb760c1 Compare May 18, 2026 15:14

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels May 18, 2026

rustbot assigned RalfJung and unassigned petrochenkov May 18, 2026

This comment has been minimized.

Sign in to view

RalfJung reviewed May 24, 2026

View reviewed changes

Comment thread tests/ui/const-ptr/out_of_bounds_read.rs Outdated

This comment has been minimized.

Sign in to view

petrochenkov mentioned this pull request Jun 3, 2026

Make parallel frontend alloc-id numbering deterministic and re-enable the tests #157333

Closed

petrochenkov 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 Jun 27, 2026

heinwol added 9 commits June 29, 2026 12:44

Actual fix inside fn normalize_output

800fdb7

additional: remove unnecessary //@ normalize-stderr

15ceff0

in `tests/ui/consts/const-eval/raw-bytes.rs` as we now handle this in our modified regex in `runtest`

Remove ignore directives

5b76fe0

Specifically, all `//@ ignore-parallel-frontend different alloc ids`

bless stderr

d06599e

Rename several //~ ERROR <...> ALLOC0 directives

9803c28

across other tests because they fail with our normalization strategy: `ALLOC0` -> `ALLOC$ID`

fix regexes in problematic files manually

7026b5b

bless for 32bit

5eb0a5a

additional failures normalization

8a64598

bless from previous

c7a0991

heinwol force-pushed the fix-ui-tests-alloc-id-parallel-frontend branch from f63e293 to c7a0991 Compare July 1, 2026 12:09

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Jul 1, 2026

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 Jul 1, 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-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Jul 1, 2026

JonathanBrouwer mentioned this pull request Jul 1, 2026

Rollup of 8 pull requests #158661

Closed

JonathanBrouwer mentioned this pull request Jul 1, 2026

Rollup of 8 pull requests #158663

Merged

rust-bors Bot merged commit 9b11acc into rust-lang:main Jul 1, 2026
13 checks passed

rustbot added this to the 1.98.0 milestone Jul 1, 2026

Uh oh!

Uh oh!

Conversation

heinwol commented May 18, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment has been minimized.

Uh oh!

petrochenkov commented May 18, 2026

Uh oh!

rustbot commented May 18, 2026

Uh oh!

heinwol commented May 18, 2026

Uh oh!

petrochenkov commented May 18, 2026

Uh oh!

rustbot commented May 18, 2026

Uh oh!

This comment has been minimized.

petrochenkov commented May 18, 2026

Uh oh!

RalfJung commented May 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

oli-obk commented May 25, 2026

Uh oh!

petrochenkov commented May 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

oli-obk commented May 25, 2026

Uh oh!

This comment has been minimized.

RalfJung commented May 26, 2026

Uh oh!

heinwol commented Jul 1, 2026

Uh oh!

petrochenkov commented Jul 1, 2026

Uh oh!

petrochenkov commented Jul 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rust-bors Bot commented Jul 1, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

heinwol commented May 18, 2026 •

edited by rustbot

Loading

RalfJung commented May 24, 2026 •

edited

Loading

petrochenkov commented May 25, 2026 •

edited

Loading

petrochenkov commented Jul 1, 2026 •

edited

Loading