Repository navigation
Extend dropping_{references,copy_types} lints to drop_in_place - #160229
Conversation
|
r? @mu001999 rustbot has assigned @mu001999. Use Why was this reviewer chosen?The reviewer was selected based on:
|
|
cc @rust-lang/miri |
This comment has been minimized.
This comment has been minimized.
1380cb0 to
27c8128
Compare
|
@traviscross has proposed to merge this. The next step is review by the rest of the tagged team members: No concerns currently listed. Once a majority of reviewers approve (and at most 2 approvals are outstanding), this will enter its final comment period. If you spot a major issue that hasn't been raised at any point in this process, please speak up! cc @rust-lang/lang-advisors: FCP proposed for lang, please feel free to register concerns. |
| /// The `dropping_references` lint checks for calls to `std::mem::drop` with a reference | ||
| /// instead of an owned value. | ||
| /// The `dropping_references` lint checks for calls to `std::mem::drop` | ||
| /// and `std::ptr::drop_in_place` with a reference instead of an owned value. |
There was a problem hiding this comment.
nit: might want to rephrase this because passing a reference to drop_in_place is an ordinary and expected thing to do. Maybe "where the dropped type is a reference" or something?
There was a problem hiding this comment.
Thanks. Went with your proposed wording.
|
Personally I think this doesn't need FCP. The intent of this lint is to catch trying to drop things where that doesn't do anything, and how that drop is spelled doesn't matter to me. @rfcbot reviewed |
27c8128 to
5d882cb
Compare
This comment has been minimized.
This comment has been minimized.
|
Pondering: thinking about the intent I described above made me think of |
We could, we even already have the lint for it |
|
The final comment period, with a disposition to merge, as per the review above, is now complete. As the automated representative of the governance process, I would like to thank the author for their work and everyone else who contributed. |
d7f1119 to
82a27a4
Compare
|
This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed. Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers. |
|
@bors r=mu001999 |
Rollup of 8 pull requests Successful merges: - #161558 (stdarch subtree update) - #159887 (compiletest: forward disable-minification from bootstrap) - #160229 (Extend `dropping_{references,copy_types}` lints to `drop_in_place`) - #160949 (Add floating point inline ASM support for SPARC) - #156160 (feat: add symmetric PartialEq impls for Vec, &[T], &mut [T] versus Cow<'_, [T]>) - #160914 (Derive `GenericTypeVisitable` for `RegionConstraint` _correctly_) - #161341 (be more permissive wrt overflow and and improve diagnostics) - #161530 (Add regression test for gce dependency ICE in non-gce crate)
Rollup of 8 pull requests Successful merges: - #161558 (stdarch subtree update) - #159887 (compiletest: forward disable-minification from bootstrap) - #160229 (Extend `dropping_{references,copy_types}` lints to `drop_in_place`) - #160949 (Add floating point inline ASM support for SPARC) - #156160 (feat: add symmetric PartialEq impls for Vec, &[T], &mut [T] versus Cow<'_, [T]>) - #160914 (Derive `GenericTypeVisitable` for `RegionConstraint` _correctly_) - #161341 (be more permissive wrt overflow and and improve diagnostics) - #161530 (Add regression test for gce dependency ICE in non-gce crate)
Rollup of 8 pull requests Successful merges: - #161558 (stdarch subtree update) - #159887 (compiletest: forward disable-minification from bootstrap) - #160229 (Extend `dropping_{references,copy_types}` lints to `drop_in_place`) - #160949 (Add floating point inline ASM support for SPARC) - #156160 (feat: add symmetric PartialEq impls for Vec, &[T], &mut [T] versus Cow<'_, [T]>) - #160914 (Derive `GenericTypeVisitable` for `RegionConstraint` _correctly_) - #161341 (be more permissive wrt overflow and and improve diagnostics) - #161530 (Add regression test for gce dependency ICE in non-gce crate)
Rollup merge of #160229 - Urgau:extend-drop_useless-drop-in-place, r=mu001999 Extend `dropping_{references,copy_types}` lints to `drop_in_place` This PR extends the `dropping_{references,copy_types}` lints to also check for calls to `std::ptr::drop_in_place` and `<*mut _>::drop_in_place`. It also extends the `undropped_manually_drops` lint. Fixes #160127 cc @theemathas
Rollup of 8 pull requests Successful merges: - rust-lang/rust#161558 (stdarch subtree update) - rust-lang/rust#159887 (compiletest: forward disable-minification from bootstrap) - rust-lang/rust#160229 (Extend `dropping_{references,copy_types}` lints to `drop_in_place`) - rust-lang/rust#160949 (Add floating point inline ASM support for SPARC) - rust-lang/rust#156160 (feat: add symmetric PartialEq impls for Vec, &[T], &mut [T] versus Cow<'_, [T]>) - rust-lang/rust#160914 (Derive `GenericTypeVisitable` for `RegionConstraint` _correctly_) - rust-lang/rust#161341 (be more permissive wrt overflow and and improve diagnostics) - rust-lang/rust#161530 (Add regression test for gce dependency ICE in non-gce crate)
Rollup of 8 pull requests Successful merges: - rust-lang/rust#161558 (stdarch subtree update) - rust-lang/rust#159887 (compiletest: forward disable-minification from bootstrap) - rust-lang/rust#160229 (Extend `dropping_{references,copy_types}` lints to `drop_in_place`) - rust-lang/rust#160949 (Add floating point inline ASM support for SPARC) - rust-lang/rust#156160 (feat: add symmetric PartialEq impls for Vec, &[T], &mut [T] versus Cow<'_, [T]>) - rust-lang/rust#160914 (Derive `GenericTypeVisitable` for `RegionConstraint` _correctly_) - rust-lang/rust#161341 (be more permissive wrt overflow and and improve diagnostics) - rust-lang/rust#161530 (Add regression test for gce dependency ICE in non-gce crate)
…ight, Infallible = ! (#2093) * hir-ty-mir: a typed hole does not lower, so `should_not_reach()` stops the program again `Expr::Underscore` in a value's place was lowered to nothing, so a body with `_` ran, with whatever was in its place: the 251 calls of the evaluator tests' `fn should_not_reach() { _ }` returned, and asserted nothing. It is a TypeError of the lowering now (as the left of an assignment `_` is a wildcard pattern and never reaches it). The two tests this uncovers were wrong themselves, not the front end: drop_struct_field expected the old wrong count (2 for 3, under its own FIXME), and drop_glue_of_a_long_list_of_boxes' fixture Box had no Drop impl, so it had no drop glue and the 100,000 nodes were never dropped or counted (now 10,000, dropped and counted: 0.4 ms a node in a test build). tests/assertion_helpers.rs makes the class a failure: before a fixture's program runs (eval_main, eval_goal) a helper named `should_not_..`/`unreachable..` is called and must not return, and one named `check`/`assert..` must hold a call of type `!`. Its control: the_oracle_catches_a_helper_that_returns. hir-ty-mir: 1402 passed, 22 failing before; 1404 passed, 22 failing after (the same 22, failing on main). * blaze-lint-defs: rustc's lints are generated from its source at the pinned commit, and the gate holds them to it scripts/src/bin/lint_table.rs reads rust-lang/rust at library/UPSTREAM's commit as register_builtins and register_internals do (every declare_lint! and declare_tool_lint! with its level, text, future-incompatibility, edition level, feature gate and flags, in the order the lint passes register them; the lints never registered; the groups, the alias, the renamed, removed and ignored names; rustc::internal) and writes crates/blaze-lint-defs/src/rustc.rs. lints.rs and names.rs keep blaze's own lints, groups and renames. - gate: `lint_table check` (blaze.yaml) fails when the file was written from another commit than library/UPSTREAM's, or its text is not its digest's (a hand edit); with a checkout it compares every byte. - bump: harness/toolchain_bump.py writes it again from the new commit (step 4a). Against the hand copy this changes the order (the pin's passes are alphabetical), rustc::internal (15 lints: span_use_eq_ctxt is gone, missing_generic_type_visitable_derive is new) and the three differences left out before, now with rustc's tests at the pin: - linker_info, linker_messages are crate-level: below the root the levels say they are ignored, at a library's root they are unused at the lint's name, as is dead_code_pub_in_binary (rustc's check_lint_check); the outer-attribute path of blaze's own is gone; - the lexer's edition lints are rustc's one ReservedPrefixLint ("`z` is parsed as a prefix in Rust 2021 and onward"), rust_2021_prefixes_incompatible_syntax a change of meaning; reserved-prefixes-migration and reserved-guarded-strings-migration byte for byte; - semicolon_in_expressions_from_non_local_macros is not in the dependency report. blaze-diagnostics: 1341 passed, 63 failing, the same 63 as on main. * blaze-diagnostics: repeated_reprs, a representation the #[repr]s of an item give more than once Rust 1.100 (rust-lang/rust#157036), as rustc's check_repr: every place of a representation given twice or more, sorted as rustc sorts them (ReprAttr's order, then the place), in one diagnostic. The nine of rustc's repr-repeated-attrs test, each whole. * blaze-diagnostics: the drop lints look at drop_in_place Rust 1.100 (rust-lang/rust#160229): `std::ptr::drop_in_place` and a pointer's `drop_in_place` with a pointer to a reference (dropping_references), to a value of a Copy type (dropping_copy_types) or to a ManuallyDrop (undropped_manually_drops, with ManuallyDrop::drop suggested), as rustc's DropForgetUseless. rustc's three tests at the pin, byte for byte. * blaze-diagnostics: a `box` pattern is removed syntax Rust 1.100 (rust-lang/rust#156749): rustc's parser error at each `box` of a pattern and of a struct pattern's field shorthand, with its two suggestions (remove the keyword, or `deref!`). a_box_pattern_is_removed_syntax, which failed on main, is rustc's test at the pin byte for byte: its std's Box is the lang item now (the fixture's crate had no `#![feature(lang_items)]`, so the patterns under the `box`es had no type and E0282 followed). blaze-diagnostics: 1343 passed, 62 failing (main: 63, this one among them). * minicore: `Infallible = !` and `Future`/`IntoFuture` in the 2024 prelude, as the pinned core has them library/core/src/convert/mod.rs at the pin has `pub type Infallible = !;` (never stabilized in 1.100) and library/core/src/prelude/mod.rs's rust_2024 re-exports `future::{Future, IntoFuture}`; minicore had an empty enum and no `Future` in the prelude. auto_sized_async_block's #11815 fixture passed only because `dyn Future` did not resolve in edition 2024. With `Future` resolved it is refused, and rightly: `Future::poll` takes `self: Pin<&mut Self>`, which dispatches only where `Pin<&mut T>: DispatchFromDyn` (rustc's `receiver_is_dispatchable` returns false with no `dispatch_from_dyn` lang item), and the fixture had its own `Sized`/`Unsize`/`CoerceUnsized` and no `DispatchFromDyn`. No front-end bug: the fixture now takes minicore's lang items with `dispatch_from_dyn` (both the `+ Send` and the plain box coerce), and a control holds that without `DispatchFromDyn` neither an async block nor a struct future coerces to `dyn Future`. * hir-ty-mir: the evaluator normalizes a place's type, so a match on a field of an alias type reads its variant `projected_ty` returned a field's declared type as written. For `struct W<M: Members>(M::Repr)` that is the alias `<M as Members>::Repr`, and `compute_discriminant` answered 0 for any type that is no enum ADT, so `match w.0 { .. }` in place took the first variant's arm whatever `w.0` held (moved into a local first, it was right). Every error union is such a struct (core's `ErrorUnion<M>(M::Repr)`), so every match of one by member read member 0. `projected_ty` now normalizes each projected type (as rustc's codegen monomorphizes every place type), with an error naming the place when it does not normalize, and `compute_discriminant` normalizes the type it is given rather than taking an alias for a non-enum. Hidden by `should_not_reach()` returning: the three error-union tests that match a member (question_mark_widens_two_bstd_errors_into_one_set_and_runs, a_subset_widens_and_an_enum_member_is_matched_by_its_variants, a_member_is_matched_by_reference_bound_whole_and_in_let_else) failed once a typed hole stopped the program, and pass now. match_in_place_on_a_field_of_an_alias_type is the plain-Rust case; it fails with the fix removed. * agents: lang1100-1 names the evaluator bug the helper exposed on main's tree
Pkgsrc changes: * Version & checksum changes. * Adapt patches to match. Upstream changes: Version 1.100.0 (2026-11-12) ========================== Language -------- - [Stabilize never type] (rust-lang/rust#155499) - [Lint against repeated `#[repr(..)]` attributes] (rust-lang/rust#157036) - [Stabilize `extern "custom"`] (rust-lang/rust#158504) - [Extend `dropping_references`, `dropping_copy_types` and `undropped_manually_drops` lints to `drop_in_place`] (rust-lang/rust#160229) - [`unfulfilled_lint_expectations` no longer fires on `#[expect(dead_code)]` inside code that is covered by a dead parent's diagnostic] (rust-lang/rust#161005) - [Turn the error for `repr(align)` fields in a `repr(packed)` type into a deny-by-default lint `aligned_fields_in_packed`] (rust-lang/rust#162160) - [volatile: allow accesses to non-AM memory to trap] (rust-lang/rust#160564) Compiler -------- - [Stabilize `-Cprofile-sample-use`] (rust-lang/rust#155942) - [Enable next solver by default in orphanck] (rust-lang/rust#160668) Platform Support ---------------- - [`i686-pc-windows-gnu` is now a tier 2 without host tools target] (rust-lang/rust#161681) - [Promote `wasm32-wasip3` to Tier 2] (rust-lang/rust#161940) - [Add new Tier 3 target `powerpc64-sony-ps3`] (rust-lang/rust#162072) Refer to Rust's [platform support page][platform-support-doc] for more information on Rust's tiered platform support. [platform-support-doc]: https://doc.rust-lang.org/rustc/platform-support.html Libraries --------- - [Stabilize custom allocators] (rust-lang/rust#156882) - [Map `ENOTSUP` to `ErrorKind::Unsupported`] (rust-lang/rust#158580) - [Fix soundness issues in `std::sys::pal::sgx::waitqueue::unsafe_list`] (rust-lang/rust#161334) Stabilized APIs --------------- - [`impl PartialEq<Cow<'_, [U]>> for Vec<T, A>`] (https://doc.rust-lang.org/stable/std/vec/struct.Vec.html#impl-PartialEq%3CCow%3C'_,+%5BU%5D%3E%3E-for-Vec%3CT,+A%3E) - [`impl PartialEq<Cow<'_, [U]>> for &[T]`] (https://doc.rust-lang.org/stable/std/primitive.slice.html#impl-PartialEq%3CCow%3C'_,+%5BU%5D%3E%3E-for-%26%5BT%5D) - [`impl PartialEq<Cow<'_, [U]>> for &mut [T]`] (https://doc.rust-lang.org/stable/std/primitive.slice.html#impl-PartialEq%3CCow%3C'_,+%5BU%5D%3E%3E-for-%26mut+%5BT%5D) - [`std::alloc::Allocator`] (https://doc.rust-lang.org/stable/std/alloc/trait.Allocator.html) - [`std::alloc::Global`] (https://doc.rust-lang.org/stable/std/alloc/struct.Global.html) - [`std::alloc::System`] (https://doc.rust-lang.org/stable/std/alloc/struct.System.html) - [`impl Allocator for Global`] (https://doc.rust-lang.org/stable/std/alloc/trait.Allocator.html#impl-Allocator-for-Global) - [`impl Allocator for System`] (https://doc.rust-lang.org/stable/std/alloc/trait.Allocator.html#impl-Allocator-for-System) - [`impl<A: Allocator> Allocator for &A`] (https://doc.rust-lang.org/stable/std/alloc/trait.Allocator.html#impl-Allocator-for-%26A) - [`impl<T: Allocator, A: Allocator> Allocator for Box<T, A>`] (https://doc.rust-lang.org/stable/std/alloc/trait.Allocator.html#impl-Allocator-for-Box%3CT,+A%3E) - [`impl<A: Allocator> Allocator for &mut A`] (https://doc.rust-lang.org/stable/std/alloc/trait.Allocator.html#impl-Allocator-for-%26mut+A) - [`impl<T: Allocator, A: Allocator> Allocator for Arc<T, A>`] (https://doc.rust-lang.org/stable/std/alloc/trait.Allocator.html#impl-Allocator-for-Arc%3CT,+A%3E) - [`impl<T: Allocator, A: Allocator> Allocator for Rc<T, A>`] (https://doc.rust-lang.org/stable/std/alloc/trait.Allocator.html#impl-Allocator-for-Rc%3CT,+A%3E) - [`Box::new_in`] (https://doc.rust-lang.org/stable/std/boxed/struct.Box.html#method.new_in) - [`Box::from_raw_in`] (https://doc.rust-lang.org/stable/std/boxed/struct.Box.html#method.from_raw_in) - [`Box::from_non_null_in`] (https://doc.rust-lang.org/stable/std/boxed/struct.Box.html#method.from_non_null_in) - [`Box::into_raw_with_allocator`] (https://doc.rust-lang.org/std/std/boxed/struct.Box.html#method.into_raw_with_allocator) - [`Box::into_non_null_with_allocator`] (https://doc.rust-lang.org/stable/std/boxed/struct.Box.html#method.into_non_null_with_allocator) - [`Box::allocator`] (https://doc.rust-lang.org/stable/std/boxed/struct.Box.html#method.allocator) - [`Vec::new_in`] (https://doc.rust-lang.org/stable/std/vec/struct.Vec.html#method.new_in) - [`Vec::with_capacity_in`] (https://doc.rust-lang.org/stable/std/vec/struct.Vec.html#method.with_capacity_in) - [`Vec::from_raw_parts_in`] (https://doc.rust-lang.org/stable/std/vec/struct.Vec.html#method.from_raw_parts_in) - [`Vec::from_parts_in`] (https://doc.rust-lang.org/stable/stable/vec/struct.Vec.html#method.from_parts_in) - [`Vec::into_raw_parts_with_allocator`] (https://doc.rust-lang.org/stable/std/vec/struct.Vec.html#method.into_raw_parts_with_allocator) - [`Vec::into_parts_with_allocator`] (https://doc.rust-lang.org/stable/std/vec/struct.Vec.html#method.into_parts_with_allocator) - [`Vec::allocator`] (https://doc.rust-lang.org/stable/std/vec/struct.Vec.html#method.allocator) - [`Child::main_thread_handle`] (https://doc.rust-lang.org/stable/std/os/windows/process/trait.ChildExt.html#tymethod.main_thread_handle) - [`impl Default for std::sync::Once`] (https://doc.rust-lang.org/stable/std/sync/struct.Once.html#impl-Default-for-Once) - [`bool::toggle`] (https://doc.rust-lang.org/stable/std/primitive.bool.html#method.toggle) - [`Box::take`] (https://doc.rust-lang.org/stable/std/boxed/struct.Box.html#method.take) - [`Box::map`] (https://doc.rust-lang.org/stable/std/boxed/struct.Box.html#method.map) - [`Rc::map`] (https://doc.rust-lang.org/stable/std/rc/struct.Rc.html#method.map) - [`Arc::map`] (https://doc.rust-lang.org/stable/std/sync/struct.Arc.html#method.map) - [`impl AsRef<T> for !`] (https://doc.rust-lang.org/stable/std/primitive.never.html#impl-AsRef<T>-for-!) - [`impl AsMut<T> for !`] (https://doc.rust-lang.org/stable/std/primitive.never.html#impl-AsMut<T>-for-!) - [`core::mem::DropGuard`] (https://doc.rust-lang.org/stable/core/mem/struct.DropGuard.html) - [`core::mem::conjure_zst`] (https://doc.rust-lang.org/stable/core/mem/fn.conjure_zst.html) - [`UnsafeCell::replace`] (https://doc.rust-lang.org/stable/std/cell/struct.UnsafeCell.html#method.replace) - [`UnsafeCell::as_ref_unchecked`] (https://doc.rust-lang.org/stable/std/cell/struct.UnsafeCell.html#method.as_ref_unchecked) - [`UnsafeCell::as_mut_unchecked`] (https://doc.rust-lang.org/stable/std/cell/struct.UnsafeCell.html#method.as_mut_unchecked) - [`str::trim_prefix`] (https://doc.rust-lang.org/stable/std/primitive.str.html#method.trim_prefix) - [`str::trim_suffix`] (https://doc.rust-lang.org/stable/std/primitive.str.html#method.trim_suffix) - [`[T]::trim_prefix`] (https://doc.rust-lang.org/stable/std/primitive.slice.html#method.trim_prefix) - [`[T]::trim_suffix`] (https://doc.rust-lang.org/stable/std/primitive.slice.html#method.trim_suffix) - [`Path::trim_prefix`] (https://doc.rust-lang.org/stable/std/path/struct.Path.html#method.trim_prefix) - [`Vec::from_fn`] (https://doc.rust-lang.org/stable/std/vec/struct.Vec.html#method.from_fn) - [`os::windows::process::CommandExt::show_window`] (https://doc.rust-lang.org/stable/std/os/windows/process/trait.CommandExt.html#tymethod.show_window) Cargo ----- - Introduced Cargo's own linting system: some Cargo warnings are now emitted as lints that users can control in the `[lints.cargo]` table. This includes a new built-in unused dependency check via the `cargo::unused_dependencies` lint. [docs](https://doc.rust-lang.org/stable/cargo/reference/lints.html) [#17298](rust-lang/cargo#17298) - Introduced minimum publish age configurations, also known as dependency cooldowns. When the relevant config is set, the resolver ignores versions published more recently than that, unless they are already in `Cargo.lock`. ([RFC 3923] (https://rust-lang.github.io/rfcs/3923-cargo-min-publish-age.html)) [docs] (https://doc.rust-lang.org/stable/cargo/reference/resolver.html#publish-age) [#17335] (rust-lang/cargo#17335) - cargo-install: now uses the existing lockfile in the `.crate` tarball by default. `--locked` in cargo install becomes a no-op. [#17388](rust-lang/cargo#17388) - Cargo now uses a new directory layout for intermediate build artifacts. Tools that inspect the build directory may need to use [Cargoâs JSON messages] (https://doc.rust-lang.org/stable/cargo/reference/external-tools.html#json-messages) to locate build artifacts instead of relying on the build directory layout. The layout is not a stable interface and is subject to change. ([build cache doc] (https://doc.rust-lang.org/stable/cargo/reference/build-cache.html)) [#17354](rust-lang/cargo#17354) Rustdoc ----- - [Add new `invalid_markdown_table` rustdoc lint] (rust-lang/rust#159583) Compatibility Notes ------------------- - [Closures now more precisely capture variables used in `if let`. This may cause new errors] (rust-lang/rust#154210) - [Check dyn-compatibility of trait objects in more places] (rust-lang/rust#154992) - [`Infallible` is now a type alias to `!`] (rust-lang/rust#155499) - [The never type fallback was changed to `!` on all editions] (rust-lang/rust#155499) - [`box` patterns are no longer syntactically allowed] (rust-lang/rust#156749) - [Report Query Cycle errors in more cases when computing the layout of enums with explicit discriminants] (rust-lang/rust#157562) - [Always escape grapheme extenders in `str::escape_debug`] (rust-lang/rust#158303) - [Make let-else respect `macro_rules` `expr` metavariable grouping] (rust-lang/rust#158515) - [`Iterator::{min,max}(_by_key)` changed which `Ord` methods they use.] (rust-lang/rust#160203) This cannot change behaviour (other than being potentially faster) for types correctly following the [`Ord` rules] (https://doc.rust-lang.org/std/cmp/trait.Ord.html), but may impact types with inconsistent trait implementations. - [Emit an error when trying to enable SSE/AVX on an x86 (32bit or 64bit) softfloat target] (rust-lang/rust#160302) - [Require windowed (and exception) for Xtensa ABI] (rust-lang/rust#160643) - [windows-gnullvm: always link libunwind statically] (rust-lang/rust#160712) - [From version 1.94 up to 1.100 a number of ARMv7-A hardfloat targets required the `VFPv3-D32` FPU as a minimum, as they erroneously had the `d32` feature in their baseline list of features. Breakage introduced as a result of correcting this can be fixed by passing `-Ctarget-feature=+d32` to Rust.] (rust-lang/rust#160911) - [Change "Rust" ABI on x86 to use SSE registers if they are always-available for the current target] (rust-lang/rust#161583) - [Make the `ambiguous_glob_imported_traits` lint fire in slightly more cases] (rust-lang/rust#161805) - [Make `+fix-cortex-a53-835769` a default feature for `aarch64-unknown-fuchsia`] (rust-lang/rust#161886) - [Some impls on `Box` have been generalised to arbitrary allocators, which may cause new type inference errors] (rust-lang/rust#161946) - [The standard library now uses Unicode version 18.0.0] (https://doc.rust-lang.org/stable/std/primitive.char.html#associatedconstant.UNICODE_VERSION) - [WASI SDK version was bumped from 33 to 34] (rust-lang/rust#163809) - [On ARM targets, `-Ctarget-feature`/`-Ctarget-cpu` values that would alter the ABI are rejected as hard errors now.] (rust-lang/rust#161280) Other ----- - [Test harness now allows passing `--test-threads` and `--color` multiple times, with later arguments overriding earlier ones] (rust-lang/rust#161312)
View all comments
This PR extends the
dropping_{references,copy_types}lints to also check for calls tostd::ptr::drop_in_placeand<*mut _>::drop_in_place.It also extends the
undropped_manually_dropslint.Fixes #160127
cc @theemathas