diff options
| author | Matthias Krüger <matthias.krueger@famsik.de> | 2024-09-02 04:19:28 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-09-02 04:19:28 +0200 |
| commit | e0039171ffdb546a1b22012e4d6febe0634af2a9 (patch) | |
| tree | 83a9954b0555b7a4f424687fef8d81bddbed70a5 /compiler/rustc_driver_impl/src | |
| parent | c90991db1759d0da7034146755c048d8e581d73c (diff) | |
| parent | 91854453f233c2b7084e665c2cf36f5494b26b76 (diff) | |
| download | rust-e0039171ffdb546a1b22012e4d6febe0634af2a9.tar.gz rust-e0039171ffdb546a1b22012e4d6febe0634af2a9.zip | |
Rollup merge of #129678 - compiler-errors:type-ir-inherent, r=fmease
Deny imports of `rustc_type_ir::inherent` outside of type ir + new trait solver We shouldn't encourage using `rustc_type_ir::inherent` outside of the new solver[^1], though this can happen by accident due to rust-analyzer, for example. See https://github.com/rust-lang/rust/pull/127537#discussion_r1733813842 for an example in practice. r? fmease [^1]: Unless we go the fully radical approach of always using these inherent methods everywhere in favor of inherent methods, which would be a major overhaul of the compiler, IMO. I don't really want to consider that possibility right now, tho.
Diffstat (limited to 'compiler/rustc_driver_impl/src')
0 files changed, 0 insertions, 0 deletions