Fix PostBorrowckAnalysis for old solver
#135899
Open
+184
−188
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
rustbot
added
S-waiting-on-review
compiler-errors
commented
Jan 22, 2025
| @@ -36,5 +36,21 @@ note: this opaque type is in the signature | |||
| LL | type Bar = impl std::fmt::Display; | |||
| | ^^^^^^^^^^^^^^^^^^^^^^ | |||
|
|
|||
| error: aborting due to 3 previous errors | |||
| error: the constant `N` is not of type `{type error}` | |||
There was a problem hiding this comment.
See the comment:
// FIXME: We should reveal the TAITs that end up in where clauses here, otherwise we
// will not be able to match param-env candidates in the old solver, since we don't
// have eq-modulo-normalization. This is less of a problem than it seems, since this
// only matters if we have TAITs in where where clauses, which isn't achievable with
// RPIT anyways.
| @@ -6,5 +6,21 @@ LL | async fn test<const N: crate::Bar>() { | |||
| | | |||
| = note: the only supported types are integers, `bool`, and `char` | |||
|
|
|||
| error: aborting due to 1 previous error | |||
| error: the constant `N` is not of type `u32` | |||
There was a problem hiding this comment.
See the comment:
See the comment:
// FIXME: We should reveal the TAITs that end up in where clauses here, otherwise we
// will not be able to match param-env candidates in the old solver, since we don't
// have eq-modulo-normalization. This is less of a problem than it seems, since this
// only matters if we have TAITs in where where clauses, which isn't achievable with
// RPIT anyways.
| @@ -1,7 +1,5 @@ | |||
| // ICE failed to resolve instance for ... | |||
| // issue: rust-lang/rust#123145 | |||
| //@ build-fail | |||
There was a problem hiding this comment.
This now cycles in check_opaque_well_formed, since now we actually reveal impl Handler when proving that it satisfies its own item bounds. This seems desirable to me.
| let generic_ty = self.cx().type_of(data.def_id); | ||
| let mut concrete_ty = generic_ty.instantiate(self.cx(), args); | ||
|
|
||
| if concrete_ty == ty { |
There was a problem hiding this comment.
This recursiveness check sucks, but it prevents bad error messages.
| Ty::new_error_with_message(self.cx(), self.cause.span, "recursive opaque type"); | ||
| } | ||
|
|
||
| let concrete_ty = fold_regions(self.cx(), concrete_ty, |re, _dbi| match re.kind() { |
There was a problem hiding this comment.
I wonder if we should consolidate all of these "replace erasing with infer" so that we only need to change it in one place when we start using a real binder...
|
The job Click to see the possible cause of the failure (guessed by this bot) |
|
💀 I guess I'll minimize that ^ |
compiler-errors
added
S-waiting-on-author
lcnr
reviewed
Jan 28, 2025
| @@ -128,10 +129,10 @@ pub(super) fn needs_normalization<'tcx, T: TypeVisitable<TyCtxt<'tcx>>>( | |||
| // so we can ignore those. | |||
| match infcx.typing_mode() { | |||
| // FIXME(#132279): We likely want to reveal opaques during post borrowck analysis | |||
lcnr
reviewed
Jan 28, 2025
| @@ -224,28 +258,20 @@ impl<'a, 'b, 'tcx> TypeFolder<TyCtxt<'tcx>> for AssocTypeNormalizer<'a, 'b, 'tcx | |||
| // Only normalize `impl Trait` outside of type inference, usually in codegen. | |||
| match self.selcx.infcx.typing_mode() { | |||
| // FIXME(#132279): We likely want to reveal opaques during post borrowck analysis | |||
lcnr
reviewed
Jan 28, 2025
| @@ -973,9 +973,6 @@ impl<'tcx> InferCtxt<'tcx> { | |||
| TypingMode::Analysis { defining_opaque_types } => { | |||
| id.into().as_local().is_some_and(|def_id| defining_opaque_types.contains(&def_id)) | |||
| } | |||
| // FIXME(#132279): This function is quite weird in post-analysis | |||
| // and post-borrowck analysis mode. We may need to modify its uses | |||
| // to support PostBorrowckAnalysis in the old solver as well. | |||
There was a problem hiding this comment.
why remove this FIXME 🤔 I feel like ideally we just never call this function outside of TypingMode::Analysis, not sure whether that's possible, might be kinda ugly to implement
lcnr
reviewed
Jan 28, 2025
|
☔ The latest upstream changes (presumably #134824) made this pull request unmergeable. Please resolve the merge conflicts. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
branch name is typo'dThis "fixes" the
PostBorrowckAnalysismode for the old solver, and begins to use it incheck_opaque_well_formedandcheck_coroutine_obligations.There are several curiosities here that result.
r? lcnr