Commit Graph

264 Commits

Author SHA1 Message Date
Nicholas Nethercote
688792715b Add warn(unreachable_pub) to rustc_infer. 2024-08-27 14:47:56 +10:00
Nicholas Nethercote
84ac80f192 Reformat use declarations.
The previous commit updated `rustfmt.toml` appropriately. This commit is
the outcome of running `x fmt --all` with the new formatting options.
2024-07-29 08:26:52 +10:00
Michael Goulet
ce8a625092 Move all error reporting into rustc_trait_selection 2024-07-21 22:34:35 -04:00
Michael Goulet
fc1e7ceb4b Move some stuff to TypeErrCtxt 2024-07-21 22:33:15 -04:00
Michael Goulet
e0ba1931f4 Revert "sort suggestions for object diagnostic"
This reverts commit 540be28f6c.
2024-07-18 16:51:06 -04:00
Michael Goulet
90423a7abb Uplift elaboration 2024-07-07 11:28:01 -04:00
Michael Goulet
dfc02d28c7 Rollup merge of #127386 - compiler-errors:uplift-outlives-components, r=lcnr
Uplift outlives components to `rustc_type_ir`

We need this to uplift `push_outlives_components`, since the elaborator uses `push_outlives_components` to elaborate type outlives obligations and I want to uplift elaboration.

This ends up reworking and inlining a fair portion of the `GenericArg::walk_shallow` function, whose only callsite was this one. I believe I got the logic correct, but may be worthwhile to look at it closely just in case. Unfortunately github was too dumb to understand that this is a rename + change -- I could also rework the git history to split the "copy the file over" part from the actual logical changes if that makes this easier to review.

r? lcnr
2024-07-06 14:55:23 -04:00
Michael Goulet
cc6c5de39d Import via rustc_type_ir::outlives
We could use rustc_middle::ty::outlives I guess?
2024-07-06 10:47:46 -04:00
Michael Goulet
99429a6a18 Split out transitive_bounds_that_define_assoc_item 2024-07-05 20:21:16 -04:00
Michael Goulet
e5412da723 Supertrait elaboration doesn't need to use Predicates 2024-07-05 15:51:53 -04:00
bors
7b21c18fe4 Auto merge of #126996 - oli-obk:do_not_count_errors, r=nnethercote
Automatically taint InferCtxt when errors are emitted

r? `@nnethercote`

Basically `InferCtxt::dcx` now returns a `DiagCtxt` that refers back to the `Cell<Option<ErrorGuaranteed>>` of the `InferCtxt` and thus when invoking `Diag::emit`, and the diagnostic is an error, we taint the `InferCtxt` directly.

That change on its own has no effect at all, because `InferCtxt` already tracks whether errors have been emitted by recording the global error count when it gets opened, and checking at the end whether the count changed. So I removed that error count check, which had a bit of fallout that I immediately fixed by invoking `InferCtxt::dcx` instead of `TyCtxt::dcx` in a bunch of places.

The remaining new errors are because an error was reported in another query, and never bubbled up. I think they are minor enough for this to be ok, and sometimes it actually improves diagnostics, by not silencing useful diagnostics anymore.

fixes #126485 (cc `@olafes)`

There are more improvements we can do (like tainting in hir ty lowering), but I would rather do that in follow up PRs, because it requires some refactorings.
2024-07-01 06:35:58 +00:00
Michael Goulet
81c2c57519 Make queries more explicit 2024-06-27 12:03:57 -04:00
Oli Scherer
86c8eae774 Automatically taint InferCtxt when errors are emitted 2024-06-26 16:01:45 +00:00
Michael Goulet
2c0348a0d8 Stop passing traitref/traitpredicate by ref 2024-06-12 20:57:24 -04:00
Michael Goulet
4038010436 Get rid of PredicateObligations 2024-06-11 13:52:51 -04:00
Michael Goulet
a41c44f21c Nits and formatting 2024-06-03 10:02:08 -04:00
Michael Goulet
27f5eccd1f Move FulfillmentErrorCode to rustc_trait_selection too 2024-06-03 09:27:52 -04:00
Michael Goulet
54b2b7d460 Make TraitEngines generic over error 2024-06-03 09:27:52 -04:00
Michael Goulet
084ccd2390 Remove unnecessary extension trait 2024-06-03 09:27:52 -04:00
Michael Goulet
05e0f8740a Uplift GenericArgKind, CanonicalVarValues, QueryInput
and make NestedGoals generic
2024-05-18 16:21:43 -04:00
Michael Goulet
2025e44ef8 to_opt_poly_X_pred -> as_X_clause 2024-05-17 12:58:33 -04:00
Michael Goulet
7b3d6dad20 Remove trivial Binder::dummy calls 2024-05-16 14:24:23 -04:00
Michael Goulet
138881b315 Uplift Goal to rustc_type_ir 2024-05-16 14:24:22 -04:00
Michael Goulet
412dc28d6a Make P parameter explicit 2024-05-16 14:23:47 -04:00
Michael Goulet
11ec3eca74 Rename ToPredicate for Upcast 2024-05-16 14:23:47 -04:00
Michael Goulet
fa84018c2e Apply nits 2024-05-13 16:55:58 -04:00
Michael Goulet
3bcdf3058e split out AliasTy -> AliasTerm 2024-05-13 11:59:42 -04:00
Michael Goulet
9108294a6c More rename fallout 2024-05-10 10:42:57 -04:00
Michael Goulet
534e267d48 Rename some ObligationCauseCode variants 2024-05-10 10:42:57 -04:00
Michael Goulet
4bde8a8f4b Remove glob imports for ObligationCauseCode 2024-05-10 10:42:56 -04:00
Michael Goulet
04c049498d rename some variants in FulfillmentErrorCode 2024-05-10 10:42:14 -04:00
Michael Goulet
d9eb5232b6 Use ObligationCtxt in favor of TraitEngine in many places 2024-05-02 22:03:01 -04:00
Nicholas Nethercote
7418aa1a07 Remove extern crate rustc_data_structures from numerous crates. 2024-04-29 18:45:14 +10:00
Nicholas Nethercote
4814fd0a4b Remove extern crate rustc_macros from numerous crates. 2024-04-29 10:21:54 +10:00
Lukas Markeffsky
88b10c1162 include ParamEnv in projection cache key 2024-04-20 16:42:18 +02:00
Nicholas Nethercote
0d97669a17 Simplify static_assert_sizes.
We want to run them on all 64-bit platforms.
2024-04-18 15:36:25 +10:00
Zalathar
2d47cd77ac Check x86_64 size assertions on aarch64, too
This makes it easier for contributors on aarch64 workstations (e.g. Macs) to
notice when these assertions have been violated.
2024-04-03 16:53:03 +11:00
Michael Goulet
127e42d33b Use != Positive rather than == Negative
Feels more complete, and for ImplPolarity has the side-effect of making
sure we also handle reservation impls correctly
2024-03-22 11:16:57 -04:00
Michael Goulet
4b87c0b9c9 Split out ImplPolarity and PredicatePolarity 2024-03-22 11:16:56 -04:00
lcnr
2339317cfb move snapshot handling into mod 2024-03-07 15:39:30 +01:00
Jason Newcomb
be9b125d41 Convert TypeVisitor and DefIdVisitor to use VisitorResult 2024-03-05 13:28:15 -05:00
lcnr
5ec9b8d778 distinguish recursion limit based overflow for diagnostics
also change the number of allowed fixpoint steps to be fixed instead
of using the `log` of the total recursion depth.
2024-02-29 10:14:02 +01:00
Nicholas Nethercote
899cb40809 Rename DiagnosticBuilder as Diag.
Much better!

Note that this involves renaming (and updating the value of)
`DIAGNOSTIC_BUILDER` in clippy.
2024-02-28 08:55:35 +11:00
Michael Goulet
f624d55ea7 Nits 2024-02-16 15:07:37 +00:00
Michael Goulet
a9dbf63087 Move trait into attr so it's greppable 2024-02-16 15:07:37 +00:00
Michael Goulet
9c25823bb4 Use extension trait derive 2024-02-16 15:07:37 +00:00
Shoyu Vanilla
3856df059e Dejargnonize subst 2024-02-12 15:46:35 +09:00
Matthias Krüger
4c154a1a48 Rollup merge of #120872 - petrochenkov:opthirpar, r=cjgillot
hir: Refactor getters for HIR parents

See individual commits.

I ended up removing on of the FIXMEs from https://github.com/rust-lang/rust/pull/120206 instead of addressing it.
2024-02-11 23:19:08 +01:00
Vadim Petrochenkov
b07283815b hir: Remove hir::Map::{opt_parent_id,parent_id,get_parent,find_parent} 2024-02-10 12:24:46 +03:00
bors
757b8efed4 Auto merge of #120712 - compiler-errors:async-closures-harmonize, r=oli-obk
Harmonize `AsyncFn` implementations, make async closures conditionally impl `Fn*` traits

This PR implements several changes to the built-in and libcore-provided implementations of `Fn*` and `AsyncFn*` to address two problems:
1. async closures do not implement the `Fn*` family traits, leading to breakage: https://crater-reports.s3.amazonaws.com/pr-120361/index.html
2. *references* to async closures do not implement `AsyncFn*`, as a consequence of the existing blanket impls of the shape `AsyncFn for F where F: Fn, F::Output: Future`.

In order to fix (1.), we implement `Fn` traits appropriately for async closures. It turns out that async closures can:
* always implement `FnOnce`, meaning that they're drop-in compatible with `FnOnce`-bound combinators like `Option::map`.
* conditionally implement `Fn`/`FnMut` if they have no captures, which means that existing usages of async closures should *probably* work without breakage (crater checking this: https://github.com/rust-lang/rust/pull/120712#issuecomment-1930587805).

In order to fix (2.), we make all of the built-in callables implement `AsyncFn*` via built-in impls, and instead adjust the blanket impls for `AsyncFn*` provided by libcore to match the blanket impls for `Fn*`.
2024-02-10 07:15:15 +00:00