about summary refs log tree commit diff
path: root/compiler/rustc_middle/src
AgeCommit message (Collapse)AuthorLines
2022-10-21Introduce deduced parameter attributes, and use them for deducing `readonly` onPatrick Walton-1/+23
indirect immutable freeze by-value function parameters. Right now, `rustc` only examines function signatures and the platform ABI when determining the LLVM attributes to apply to parameters. This results in missed optimizations, because there are some attributes that can be determined via analysis of the MIR making up the function body. In particular, `readonly` could be applied to most indirectly-passed by-value function arguments (specifically, those that are freeze and are observed not to be mutated), but it currently is not. This patch introduces the machinery that allows `rustc` to determine those attributes. It consists of a query, `deduced_param_attrs`, that, when evaluated, analyzes the MIR of the function to determine supplementary attributes. The results of this query for each function are written into the crate metadata so that the deduced parameter attributes can be applied to cross-crate functions. In this patch, we simply check the parameter for mutations to determine whether the `readonly` attribute should be applied to parameters that are indirect immutable freeze by-value. More attributes could conceivably be deduced in the future: `nocapture` and `noalias` come to mind. Adding `readonly` to indirect function parameters where applicable enables some potential optimizations in LLVM that are discussed in [issue 103103] and [PR 103070] around avoiding stack-to-stack memory copies that appear in functions like `core::fmt::Write::write_fmt` and `core::panicking::assert_failed`. These functions pass a large structure unchanged by value to a subfunction that also doesn't mutate it. Since the structure in this case is passed as an indirect parameter, it's a pointer from LLVM's perspective. As a result, the intermediate copy of the structure that our codegen emits could be optimized away by LLVM's MemCpyOptimizer if it knew that the pointer is `readonly nocapture noalias` in both the caller and callee. We already pass `nocapture noalias`, but we're missing `readonly`, as we can't determine whether a by-value parameter is mutated by examining the signature in Rust. I didn't have much success with having LLVM infer the `readonly` attribute, even with fat LTO; it seems that deducing it at the MIR level is necessary. No large benefits should be expected from this optimization *now*; LLVM needs some changes (discussed in [PR 103070]) to more aggressively use the `noalias nocapture readonly` combination in its alias analysis. I have some LLVM patches for these optimizations and have had them looked over. With all the patches applied locally, I enabled LLVM to remove all the `memcpy`s from the following code: ```rust fn main() { println!("Hello {}", 3); } ``` which is a significant codegen improvement over the status quo. I expect that if this optimization kicks in in multiple places even for such a simple program, then it will apply to Rust code all over the place. [issue 103103]: https://github.com/rust-lang/rust/issues/103103 [PR 103070]: https://github.com/rust-lang/rust/pull/103070
2022-10-20Auto merge of #103220 - compiler-errors:deny-infers, r=lcnrbors-17/+28
Deny hashing ty/re/ct inference variables cc `@cjgillot` and https://github.com/rust-lang/rust/pull/102695#issuecomment-1275706528 r? `@lcnr` best reviewed one commit at a time, mostly because the second commit that fixes `ClosureOutlivesRequirement` is mostly noise because of losing its `<'tcx>` lifetime parameter.
2022-10-20Auto merge of #102417 - oli-obk:opaque_lifetimes2, r=jackh726bors-2/+75
Require lifetime bounds for opaque types in order to allow hidden types to capture said lifetimes fixes #96996 cc `@aliemjay`
2022-10-19Deny const variables as wellMichael Goulet-4/+11
2022-10-19Don't call `own_existential_vtable_entries` on unresolved trait refMichael Goulet-2/+2
2022-10-19Make ClosureOutlivesRequirement not rely on an unresolved typeMichael Goulet-12/+14
2022-10-19Don't hash non-fresh Ty::Infer or RegionKind::InferMichael Goulet-2/+4
2022-10-19Get rid of native_library projection queriesnils-11/+0
They don't seem particularly useful as I don't expect native libraries to change frequently.
2022-10-19stop folding `UnevaluatedConst`lcnr-77/+5
2022-10-18Clean up query descriptionsnils-89/+92
Use the same tense everywhere and prefer display over debug, as these descriptions are user facing.
2022-10-18`const_evaluatable_unchecked` to const evallcnr-1/+26
2022-10-18change `ConstEvaluatable` to use `ty::Const`lcnr-6/+34
2022-10-17Auto merge of #102355 - lcnr:bye-bye-type-traversal, r=oli-obkbors-185/+21
remove type traversal for mir constants r? `@oli-obk` cc `@b-naber`
2022-10-17add inline to `TrivialTypeTraversalImpls`lcnr-14/+17
2022-10-17rm `try_normalize_mir_const_after_erasing_regions`lcnr-8/+0
2022-10-17mir constants: type traversing bye byelcnr-169/+10
2022-10-17Duplicate comment in mod.rsSamuel Moelius-5/+0
2022-10-16Auto merge of #102026 - Bryanskiy:resolve_update, r=petrochenkovbors-32/+104
Populate effective visibilities in 'rustc_resolve' Next part of RFC https://github.com/rust-lang/rust/issues/48054. previous: https://github.com/rust-lang/rust/pull/101713 `@rustbot` author r? `@petrochenkov`
2022-10-16Populate effective visibilities in 'rustc_resolve'Bryanskiy-32/+104
2022-10-16Rollup merge of #102953 - WaffleLapkin:better_docs_for_decorate_param, ↵Matthias Krüger-0/+43
r=RalfJung Improve docs for `struct_lint_level` function. r? ``@RalfJung`` Does this answer your questions?
2022-10-16fix typoRalf Jung-1/+1
2022-10-15Fix subst issues with RPITITMichael Goulet-3/+16
2022-10-15Auto merge of #102895 - Nilstrieb:query-cleanups, r=cjgillotbors-2/+14
Get rid of `rustc_query_description!` **I am not entirely sure whether this is an improvement and would like to get your feedback on it.** Helps with #96524. Queries can provide an arbitrary expression for their description and their caching behavior. Before, these expressions where stored in a `rustc_query_description` macro emitted by the `rustc_queries` macro, and then used in `rustc_query_impl` to fill out the methods for the `QueryDescription` trait. Instead, we now emit two new modules from `rustc_queries` containing the functions with the expressions. `rustc_query_impl` calls these functions now instead of invoking the macro. Since we are now defining some of the functions in `rustc_middle::query`, we now need all the imports for the key types mthere as well. r? `@cjgillot`
2022-10-15Auto merge of #101832 - compiler-errors:dyn-star-plus, r=eholkbors-2/+5
Make `dyn*` casts into a coercion, allow `dyn*` upcasting I know that `dyn*` is likely not going to be a feature exposed to surface Rust, but this makes it slightly more ergonomic to write tests for these types anyways. ... and this was just fun to implement anyways. 1. Make `dyn*` into a coercion instead of a cast 2. Enable `dyn*` upcasting since we basically get it for free 3. Simplify some of the cast checking code since we're using the coercion path now r? `@eholk` but feel free to reassign cc `@nikomatsakis` and `@tmandry` who might care about making `dyn*` casts into a coercion
2022-10-14Fix line numbers for MIR inlined codeWesley Wiser-1/+3
`should_collapse_debuginfo` detects if the specified span is part of a macro expansion however it does this by checking if the span is anything other than a normal (non-expanded) kind, then the span sequence is walked backwards to the root span. This doesn't work when the MIR inliner inlines code as it creates spans with expansion information set to `ExprKind::Inlined` and results in the line number being attributed to the inline callsite rather than the normal line number of the inlined code.
2022-10-14Get rid of `rustc_query_description!`Nilstrieb-2/+14
Queries can provide an arbitrary expression for their description and their caching behavior. Before, these expressions where stored in a `rustc_query_description` macro emitted by the `rustc_queries` macro, and then used in `rustc_query_impl` to fill out the methods for the `QueryDescription` trait. Instead, we now emit two new modules from `rustc_queries` containing the functions with the expressions. `rustc_query_impl` calls these functions now instead of invoking the macro. Since we are now defining some of the functions in `rustc_middle::query`, we now need all the imports for the key types there as well.
2022-10-14Rollup merge of #103018 - Rageking8:more-dupe-word-typos, r=TaKO8KiDylan DPC-6/+4
More dupe word typos I only picked those changes (from the regex search) that I am pretty certain doesn't change meaning and is just a typo fix. Do correct me if any fix is undesirable and I can revert those. Thanks.
2022-10-14Require lifetime bounds for opaque types in order to allow hidden types to ↵Oli Scherer-2/+75
capture said lifetimes
2022-10-14Auto merge of #102695 - compiler-errors:int-and-float-trivial-copy, r=lcnrbors-1/+4
Int and float inference variables are trivially copy Fixes #102645
2022-10-14float and int vars are trivially copyMichael Goulet-1/+4
2022-10-14more dupe word typosRageking8-6/+4
2022-10-14Auto merge of #102684 - ↵bors-2/+1
JhonnyBillM:delete-target-data-layout-errors-wrapper, r=davidtwco Move `IntoDiagnostic` conformance for `TargetDataLayoutErrors` into `rustc_errors` Addressed this suggestion https://github.com/rust-lang/rust/pull/101558#issuecomment-1243830009. This way we comply with the Coherence rule given that `IntoDiagnostic` trait is defined in `rustc_errors`, and almost all other crates depend on it.
2022-10-14Make dyn* cast into a coercionMichael Goulet-2/+5
2022-10-13Auto merge of #102700 - oli-obk:0xDEAD_TAIT, r=compiler-errorsbors-5/+258
Check hidden types in dead code fixes #99490 r? `@compiler-errors` best reviewed commit by commit
2022-10-13Print RPITIT verbosely if -ZverboseMichael Goulet-1/+3
2022-10-12Auto merge of #101679 - compiler-errors:rpitit-default-body, r=nikomatsakisbors-0/+1
Support default-body trait functions with return-position `impl Trait` in traits Introduce a new `Trait` candidate kind for the `ImplTraitInTrait` projection candidate, which just projects an RPITIT down to its opaque type form. This is a hack until we lower RPITITs to regular associated types, after which we will need to rework how these default bodies are type-checked, so comments are left in a few places for us to clean up later. Fixes #101665
2022-10-12ADD - IntoDiagnostic conformance for TargetDataLayoutErrors in rustc_errorsJhonny Bill Mena-2/+1
This way we comply with the Coherence rule given that IntoDiagnostic trait is defined in rustc_errors, and almost all other crates depend on it.
2022-10-12link lint function with `decorate` function param to `struct_lint_level`Maybe Waffle-0/+10
2022-10-12Apply suggestions from code reviewWaffle Maybe-9/+4
Co-authored-by: Ralf Jung <post@ralfj.de>
2022-10-12Use `tidy-alphabetical` in the compilerNilstrieb-4/+8
2022-10-12Improve docs for `struct_lint_level` function.Maybe Waffle-0/+38
2022-10-12Rollup merge of #102890 - camsteffen:adt-sized-representability, r=cjgillotDylan DPC-27/+4
Check representability in adt_sized_constraint Now that representability is a query, we can use it to preemptively avoid a cycle in `adt_sized_constraint`. I moved the representability check into `check_mod_type_wf` to avoid a scenario where rustc quits before checking all the types for representability. This also removes the check from rustdoc, which is alright AFAIK. r? ``@cjgillot``
2022-10-12Rollup merge of #102110 - CleanCut:migrate_rustc_passes_diagnostics, r=davidtwcoDylan DPC-1/+20
Migrate rustc_passes diagnostics Picks up abandoned work from https://github.com/rust-lang/rust/pull/100870 I would like to do this collaboratively, as there is a lot of work! Here's the process: - Comment below that you are willing to help and I will add you as a collaborator to my `rust` fork (that gives you write access) - Indicate which file/task you would like to work on (so we don't duplicate work) from the list below - Do the work, push up a commit, comment that you're done with that file/task - Repeat until done 😄 ### Files to Migrate (in `compiler/rustc_passes/src/`) - [x] check_attr.rs ``@CleanCut`` - [x] check_const.rs ``@CleanCut`` - [x] dead.rs ``@CleanCut`` - [x] debugger_visualizer.rs ``@CleanCut`` - [x] diagnostic_items.rs ``@CleanCut`` - [x] entry.rs ``@CleanCut`` - [x] lang_items.rs ``@CleanCut`` - [x] layout_test.rs ``@CleanCut`` - [x] lib_features.rs ``@CleanCut`` - [x] ~liveness.rs~ ``@CleanCut`` Nothing to do - [x] loops.rs ``@CleanCut`` - [x] naked_functions.rs ``@CleanCut`` - [x] stability.rs ``@CleanCut`` - [x] weak_lang_items.rs ``@CleanCut`` ### Tasks - [x] Rebase on current `master` ``@CleanCut`` - [x] Review work from [the earlier PR](https://github.com/rust-lang/rust/pull/100870) and make sure it all looks good - [x] compiler/rustc_error_messages/locales/en-US/passes.ftl ``@CleanCut`` - [x] compiler/rustc_passes/src/check_attr.rs ``@CleanCut`` - [x] compiler/rustc_passes/src/errors.rs ``@CleanCut`` - [x] compiler/rustc_passes/src/lang_items.rs ``@CleanCut`` - [x] compiler/rustc_passes/src/lib.rs ``@CleanCut`` - [x] compiler/rustc_passes/src/weak_lang_items.rs ``@CleanCut``
2022-10-11Rollup merge of #101727 - est31:stabilize_map_first_last, r=m-ou-seMatthias Krüger-1/+0
Stabilize map_first_last Stabilizes the following functions: ```Rust impl<T> BTreeSet<T> { pub fn first(&self) -> Option<&T> where T: Ord; pub fn last(&self) -> Option<&T> where T: Ord; pub fn pop_first(&mut self) -> Option<T> where T: Ord; pub fn pop_last(&mut self) -> Option<T> where T: Ord; } impl<K, V> BTreeMap<K, V> { pub fn first_key_value(&self) -> Option<(&K, &V)> where K: Ord; pub fn last_key_value(&self) -> Option<(&K, &V)> where K: Ord; pub fn first_entry(&mut self) -> Option<OccupiedEntry<'_, K, V>> where K: Ord; pub fn last_entry(&mut self) -> Option<OccupiedEntry<'_, K, V>> where K: Ord; pub fn pop_first(&mut self) -> Option<(K, V)> where K: Ord; pub fn pop_last(&mut self) -> Option<(K, V)> where K: Ord; } ``` Closes #62924 ~~Blocked on the [FCP](https://github.com/rust-lang/rust/issues/62924#issuecomment-1179489929) finishing.~~ Edit: It finished!
2022-10-10Check representability in adt_sized_constraintCameron Steffen-27/+4
2022-10-10Fix compiler docsGuillaume Gomez-3/+3
2022-10-10Rollup merge of #102786 - compiler-errors:no-tuple-candidate, r=lcnrDylan DPC-14/+2
Remove tuple candidate, nothing special about it r? `@lcnr` you mentioned this during the talk you gave i think
2022-10-10Rollup merge of #102275 - Urgau:stabilize-half_open_range_patterns, r=cjgillotYuki Okushi-1/+1
Stabilize `half_open_range_patterns` This PR stabilize `feature(half_open_range_patterns)`: ``` Allows using `..=X` as a pattern. ``` And adds a new `feature(half_open_range_patterns_in_slices)` for the slice part, https://github.com/rust-lang/rust/pull/102275#issuecomment-1267422806. The FCP was completed in https://github.com/rust-lang/rust/issues/67264.
2022-10-10Rollup merge of #102829 - compiler-errors:rename-impl-item-kind, r=TaKO8KiYuki Okushi-2/+2
rename `ImplItemKind::TyAlias` to `ImplItemKind::Type` The naming of this variant seems inconsistent given that this is not really a "type alias", and the associated type variant for `TraitItemKind` is just called `Type`.
2022-10-09ImplItemKind::TyAlias => ImplItemKind::TypeMichael Goulet-2/+2
generated by cgit-pink 1.4.1-3-g733a5 (git 2.37.1) at 2026-07-15 06:38:10 +0000