[Bryanskiy]

Partial implementation of build-std: Standard library crate stability. More specifically:

• standard_library_support field is added to the target spec, but metadata.std isn't removed yet.
• unstable option --print target-standard-library-support is added to query standard_library_support (output only in JSON, since Extend --print requests with output format suffixes compiler-team#860 is apparently not implemented).

How was standard_library_support set?

• metadata.std = Some(true) -> standard_library_support.{supported,default} = "core, alloc, std"
• metadata.std = None -> standard_library_support = None
• metadata.std = Some(false) ->
  • standard_library_support.{supported,default} = "core, alloc" if the docs say that the alloc crate is supported.
  • standard_library_support.{supported,default} = "core" if the docs explicitly say it's not supported or say nothing.
  • For aarch64_unknown_linux_pauthtest the spec says full std is supported while metadata.std was false, so I changed it.

Currently I set default equal to supported, as I think it's a reasonable start and target maintainers can change it in the future. Also, there might be other discrepancies with the spec (like for aarch64_unknown_linux_pauthtest), but I didn't check all the targets.

Questions

• The RFC first suggests using strings for both supported and default, but then in the examples it uses an array of crate names for supported and a single crate name for default. In the PR I used the first option.
• Should it be part of the metadata field?

These are minor issues and I think they can be changed in the future anyway.

Future steps

• Update Cargo to use --print target-standard-library-support instead of metadata.std
• Update  docs
• remove metadata.std

LLM disclosure: no LLM was used.

[rustbot]

Some changes occurred in compiler/rustc_target/src/spec/targets/amdgcn_amd_amdhsa.rs

cc @Flakebi

Target consistency checks are being changed

cc @RalfJung

These commits modify compiler targets.
(See the Target Tier Policy.)

[rustbot]

r? @oli-obk

rustbot has assigned @oli-obk.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler
• compiler expanded to 77 candidates
• Random selection from 22 candidates

[Bryanskiy]

cc @adamgemmell

[rust-bors]

☔ The latest upstream changes made this pull request unmergeable. Please resolve the merge conflicts by rebasing.