Since run-make tests do not support revisions.

Without suitable `needs-llvm-components` directives, these tests that
rely on target-specific codegen can fail if used with a LLVM that is
built without the required components.

This also migrates from legacy `cargo:` directives to the newer `cargo::`
prefix.

This file was introduced in LLVM 11, but was then removed in LLVM 13.

The whole file is surrounded by `#if defined(_WIN32)`, so there's no need to
have separate logic to exclue it from non-Windows builds.

This behaviour was introduced during the upgrade to LLVM 11. Now that the list
of source files has been cleaned up, we can reasonably expect _all_ of the
listed source files to be present.

Signed-off-by: onur-ozkan <[email protected]>

opt-dist overrides the stage 0 compiler with previously compiled compilers,
which can cause confusion in bootstrap's target sanity checks. It is best to
skip that check.

Signed-off-by: onur-ozkan <[email protected]>

This keeps it up-to-date by moving from 0.5.6 to 0.5.7. While here I've
additionally updated some other wasm-related dependencies in the
workspace to keep them up-to-date and try to avoid duplicate versions as
well.

Co-authored-by: Ralf Jung <[email protected]>

…-stability

This typo looks unnecessary

…mulacrum

Add release notes for 1.81.0

cc ``@rust-lang/release``
r? ``@Mark-Simulacrum``

…Mark-Simulacrum

Add missing `needs-llvm-components` directives for run-make tests that need target-specific codegen

Without suitable `needs-llvm-components` directives, some run-make tests exercising target-specific codegen can fail if the LLVM used is built without the necessary components. Currently, the list is:

```
tests\run-make\print-target-list
tests\run-make\print-to-output
tests\run-make\print-cfg
tests\run-make\target-without-atomic-cas
```

This PR also skips tidy checks for revisions and `needs-llvm-components` for run-make tests since revisions are not supported.

Fixes rust-lang#129390.
Fixes rust-lang#127895.

cc ``@petrochenkov`` who noticed this, thanks! Would be great if you could confirm that this fixes the test errors for you locally.

…-Simulacrum

Clean up `library/profiler_builtins/build.rs`

This PR makes a series of improvements to the long-neglected build script for `profiler_builtins`.

Most notably:
- The logic that silently skips missing source files has been removed, since it is currently unnecessary and makes build errors more confusing.
- The script now emits `cargo::rerun-if-changed` directives for the `compiler-rt` source and include directories.

Compiler behaviour and user programs should be unaffected by these changes.

…eck, r=Kobzol

skip stage 0 target check if `BOOTSTRAP_SKIP_TARGET_SANITY` is set

When adding a new target to `rustc` and extending `STAGE0_MISSING_TARGETS`, there is a chance that in the merge CI bootstrap target sanity check might fail [here](https://github.com/rust-lang/rust/blob/26d27b7c8729fb61fe8321fcd2ce734a79aa695d/src/bootstrap/src/core/sanity.rs#L243-L256) because the stage 0 compiler will assume to already support the new target since `opt-dist` uses the previously compiled compiler as the stage 0 compiler.

This PR skips this check if `BOOTSTRAP_SKIP_TARGET_SANITY` is set, and makes `opt-dist` to set `BOOTSTRAP_SKIP_TARGET_SANITY` so bootstrap doesn't run this logic for opt-dist tests.

Fixes rust-lang#127021 (comment).

Zulip thread: https://rust-lang.zulipchat.com/#narrow/stream/326414-t-infra.2Fbootstrap/topic/.60STAGE0_MISSING_TARGETS.60.20seems.20to.20check.20stage1

Blocker for rust-lang#127021

… r=RalfJung

Enable Miri to pass pointers through FFI

Following rust-lang#126787, the purpose of this PR is to now enable Miri to execute native calls that make use of pointers.

> <details>
>
> <summary> Simple example </summary>
>
> ```rust
> extern "C" {
>     fn ptr_printer(ptr: *mut i32);
> }
>
> fn main() {
>     let ptr = &mut 42 as *mut i32;
>     unsafe {
>         ptr_printer(ptr);
>     }
> }
> ```
> ```c
> void ptr_printer(int *ptr) {
>   printf("printing pointer dereference from C: %d\n", *ptr);
> }
> ```
> should now show `printing pointer dereference from C: 42`.
>
> </details>

Note that this PR does not yet implement any logic involved in updating Miri's "analysis" state (byte initialization, provenance) upon such a native call.

r? ``@RalfJung``

…-ld, r=jieyouxu

Update the `wasm-component-ld` binary dependency

This keeps it up-to-date by moving from 0.5.6 to 0.5.7. While here I've additionally updated some other wasm-related dependencies in the workspace to keep them up-to-date and try to avoid duplicate versions as well.

couple more crash tests

r? ```@jieyouxu```

…ch, r=compiler-errors

tidy: say which feature gate has a stability issue mismatch

This gives some valuable context to what the error is actually about :)

…er-errors

make the const-unstable-in-stable error more clear

The default should be to add `rustc_const_unstable`, not `rustc_allow_const_fn_unstable`.

Also I discovered our check for missing const stability attributes on stable functions -- but strangely that check only kicks in for "reachable" functions. `check_missing_stability` checks for reachability since all reachable functions must have a stability attribute, but I would say if a function has `#[stable]` it should also have const-stability attributes regardless of reachability.

…s-mobile, r=notriddle

Fix code examples buttons not appearing on click on mobile

When browsing docs on mobile today, I realized that the buttons didn't appear when I tapped on the code example.

One issue: I have no idea how to add a regression test for this case...

r? ``@notriddle``

library: Fix typo in `core::mem`

This typo looks unnecessary