Uh oh!
There was an error while loading. Please reload this page.
Rollup of 17 pull requests - #159281
Closed
JonathanBrouwer wants to merge 41 commits into
Closed
Conversation
Because that's what it is: any type with tokens must also have attrs. This simplifies some trait bounds.
…ctions When a const function drops a value whose type is a type parameter, suggest adding a `[const] Destruct` bound so the destructor can be evaluated at compile-time.
The suggested bound requires the unstable `const_destruct` feature, so it is not actionable on a stable compiler. Add a run-make test that verifies the suggestion is only emitted on nightly.
This way codegen backends don't have to check for a magic symbol name prefix to detect LLVM intrinsic calls and thus can more easily handle them separately as necessary.
Signed-off-by: Amirhossein Akhlaghpour <m9.akhlaghpoor@gmail.com>
This optimisation is currently broken on s390x[^1]. It looks like a big-endian issue in general. but as we only know about the test fails on s390x we only disable it there. Should be reenabled when the problem is fixed and llvm backend is updated. [^1]: llvm/llvm-project#208712
This function's name is referring to `Step::make_run`, and has nothing to do with run-make tests.
This was a reminder to make the lint translatable, but the code to make lints translatable has since been removed
Comptime inherent impls A fairly trivial addition to comptime. Now that we have inherent methods that can individually be `#[rustc_comptime]`, we can just extend that to inherent impls. lots of new test coverage as the attribute (even if it errors due to being illegal there) affects how the items it's applied to behave. r? @fee1-dead
…, r=jdonszelmann rerun in original typing mode if we meet any opaques in post analysis When the rerun condition is `OpaqueInStorage` and typing mode is `PostAnalysis/Codegen`, we do rerun the evaluation in original typing mode. It's strange that we don't do the same when the rerun condition is `OpaqueInStorageOrAnyOpaqueHasInferAsHidden`. This fixes an ICE when building `cargo` with the next solver. r? @jdonszelmann cc @lcnr
…petrochenkov Make `HasTokens` a sub-trait of `HasAttrs`. Because that's what it is: any type with tokens must also have attrs. This simplifies some trait bounds. r? @petrochenkov
…RalfJung,folkertdev Introduce InstanceKind::LlvmIntrinsic This way codegen backends don't have to check for a magic symbol name prefix to detect LLVM intrinsic calls and thus can more easily handle them separately as necessary.
Bump rustc-perf submodule To merge in rust-lang/rustc-perf#2500, which is needed to unblock rust-lang#159091.
Suggest the `[const] Destruct` bound for type parameters in const functions when missing When a const function drops a value whose type is a type parameter without a `[const] Destruct` bound, suggest adding it so that the destructor can be evaluated at compile-time. Closesrust-lang#103270. Example: ```rust const fn f(_x: impl std::fmt::Debug) {} ``` Before: ``` error[E0493]: destructor of `impl std::fmt::Debug` cannot be evaluated at compile-time --> $DIR/min_const_fn.rs:1:11 | LL | const fn f(_x: impl std::fmt::Debug) {} | ^^ - value is dropped here | | | the destructor for this type cannot be evaluated in constant functions ``` After: ``` error[E0493]: destructor of `impl std::fmt::Debug` cannot be evaluated at compile-time --> $DIR/min_const_fn.rs:1:11 | LL | const fn f(_x: impl std::fmt::Debug) {} | ^^ - value is dropped here | | | the destructor for this type cannot be evaluated in constant functions | help: consider restricting opaque type `impl std::fmt::Debug` with unstable trait `Destruct` | LL | const fn f(_x: impl std::fmt::Debug + [const] Destruct) {} | ++++++++++++++++++ ``` No new tests as the existing ones had their results updated with the new suggestion.
…-unchecked, r=jhpratt slice: make swap delegate to swap_unchecked It had duplicate unsafe code that already existed in swap_unchecked. So I just made swap call swap_unchecked instead of copying the same logic twice.
…jieyouxu bootstrap: Rename `std_crates_for_run_make` to `std_crates_for_make_run` This function's name is referring to `Step::make_run`, and has nothing to do with run-make tests. (I was getting jump-scared every time I saw it!) r? jieyouxu (or bootstrap)
remove obsolete comment This was a reminder to make the lint translatable, but the code to make lints translatable has since been removed
…nszelmann Construct `tokens` for attrs made by `mk_attr_word` and other variants Fixesrust-lang#159261 `tokens` won't be constructed for generated built-in attrs before, so that after rust-lang#158854, `#[rustc_test_entrypoint_marker]` without `tokens` would be added when expanding `#[test]`. Then later expansions would cause the ICE when they wanna get the attr's `tokens`. This PR makes `tokens` constructed for such attrs, although this specific ICE only need editing `mk_attr_word`, but I think it would be good to also edit `mk_attr_nested_word` and `mk_attr_name_value_str`.
JonathanBrouwer
commented
Jul 14, 2026
MemberAuthor
Contributor
This comment has been minimized.
This comment has been minimized.
rust-borsBot
pushed a commit
that referenced
this pull request
Jul 14, 2026
Rollup of 17 pull requests try-job: dist-various-1 try-job: test-various try-job: x86_64-gnu-aux try-job: x86_64-gnu-llvm-21-3 try-job: x86_64-msvc-1 try-job: aarch64-apple try-job: x86_64-mingw-1 try-job: i686-msvc-*
JonathanBrouwer
commented
Jul 14, 2026
MemberAuthor
@bors p=7 |
Contributor
PR #159239, which is a member of this rollup, changed its commit SHA. This rollup was thus unapproved due to being closed. |
Contributor
This file contains hidden or 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
Sign up for freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
18 participants
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.
Successful merges:
HasTokensa sub-trait ofHasAttrs. #159160 (MakeHasTokensa sub-trait ofHasAttrs.)[const] Destructbound for type parameters in const functions when missing #155013 (Suggest the[const] Destructbound for type parameters in const functions when missing)_instead of...as placeholder #159255 (Replace shortened type with_instead of...as placeholder)cannot relate bound region#144033)std_crates_for_run_maketostd_crates_for_make_run#159274 (bootstrap: Renamestd_crates_for_run_maketostd_crates_for_make_run)tokensfor attrs made bymk_attr_wordand other variants #159277 (Constructtokensfor attrs made bymk_attr_wordand other variants)r? @ghost
Create a similar rollup