632ed10838
Revert "eagerly normalize during generalization"
...
This reverts commit 5d0863b147
2026-03-23 12:09:34 +01:00
4af62bdbac
Revert "implement eager normalization in a fresh context during typeck"
...
This reverts commit 04f2c0191e
2026-03-23 12:09:34 +01:00
e190973e4f
Revert "merge generalizer state and structurally relate aliases"
...
This reverts commit 917713b4db
2026-03-23 12:09:34 +01:00
df0f627bfc
Revert "explicitly provide type in transmute"
...
This reverts commit 7a123e87b2
2026-03-23 12:09:34 +01:00
c838deca23
Revert "normalize at the start of generalize if we can"
...
This reverts commit f50591fc10
2026-03-23 12:09:34 +01:00
38cb40091c
Revert "try generalizing if normalization isn't a tyvar"
...
This reverts commit e8e3544a71
2026-03-23 12:09:34 +01:00
bd781549f2
Revert "inline into"
...
This reverts commit 9aa065bc49
2026-03-23 12:09:34 +01:00
1fa1611e9f
Revert "address review"
...
This reverts commit d665c0b371
2026-03-23 12:09:34 +01:00
ac7f9ec7da
Auto merge of #151746 - jdonszelmann:eagerly-normalize-in-generalize, r=lcnr
...
Eagerly normalize in generalize
*[View all comments](https://triagebot.infra.rust-lang.org/gh-comments/rust-lang/rust/pull/151746 )*
r? @lcnr
cc: https://github.com/rust-lang/trait-system-refactor-initiative/issues/262
2026-03-20 18:22:43 +00:00
d665c0b371
address review
2026-03-20 14:02:24 +01:00
b77739ad81
Do not shallow resolve to root var while fudging
2026-03-19 00:26:19 +09:00
985b41d387
Remove the rustc_data_structures::assert_matches! re-exports
2026-03-08 22:02:23 +11:00
32bae1353e
Update cfg(bootstrap)
2026-03-07 10:42:02 -08:00
9aa065bc49
inline into
2026-02-26 14:53:57 +01:00
e8e3544a71
try generalizing if normalization isn't a tyvar
2026-02-26 14:53:57 +01:00
f50591fc10
normalize at the start of generalize if we can
2026-02-26 14:53:57 +01:00
7a123e87b2
explicitly provide type in transmute
2026-02-26 14:53:57 +01:00
917713b4db
merge generalizer state and structurally relate aliases
2026-02-26 14:53:57 +01:00
04f2c0191e
implement eager normalization in a fresh context during typeck
2026-02-26 14:53:57 +01:00
5d0863b147
eagerly normalize during generalization
2026-02-26 11:44:24 +01:00
4b7a05eeee
Stop using LinkedGraph in lexical_region_resolve
2026-02-26 13:00:28 +11:00
75b9d89c68
Auto merge of #151380 - ShoyuVanilla:shallow-resolve-to-root-var, r=lcnr
...
Shallow resolve ty and const vars to their root vars
Continuation of https://github.com/rust-lang/rust/pull/147193
2026-02-15 03:04:28 +00:00
2f3b952349
Stabilize assert_matches
2026-02-11 14:13:44 +01:00
955f7750c1
Remove redundant shallow_resolve call
2026-02-11 18:08:14 +09:00
13a83cb639
Shallow resolve ty and const vars to their root vars
2026-02-11 17:38:34 +09:00
1722b8e06b
Convert to inline diagnostics in rustc_infer
2026-02-03 15:14:49 +01:00
40d7cb890f
Rollup merge of #150271 - Jamesbarford:chore/refactor-struct-placeholder-pt2, r=lcnr
...
Move struct placeholder pt2
r? ghost
2026-01-29 17:47:29 +01:00
25c1365507
Part 2 refactoring of moving placeholder types to rustc_type_ir
2026-01-29 11:11:40 +00:00
f6efe7e1d5
don't return incorrectly constrained opaques in method_autoderef_steps
2026-01-26 17:30:18 +08:00
7ec34defe9
Temporarily re-export assert_matches! to reduce stabilization churn
2026-01-19 18:26:53 +11:00
4ae3c85a5e
Use the name var_kinds more.
...
Variables that are collections of `CanonicalVarKind` are sometimes
called `var_kinds` and sometimes called `variables`. The former is much
better, because `variables` is (a) non-descript, and (b) often used
nearby for collections of `I::GenericArg`. I found the inconsistency
made the canonicalization code harder to understand.
This commit renames various `variables` things as `var_kinds`.
2026-01-08 13:37:34 +11:00
ec5e96e58b
Remove Option from fold_infer_ty return type.
...
It's no longer needed.
2025-12-23 07:46:12 +11:00
2d10f47693
Disallow freshening an already-freshened type/const.
...
It never happens in practice.
2025-12-22 23:46:39 +11:00
5c49aee9ec
Remove InferCtxt::freshen.
...
Sometimes we freshen using a new `TypeFreshener`, and sometimes we
freshen with an existing `TypeFreshener`. For the former we have the
method `InferCtxt::freshen`. For the latter we just call `fold_with`.
This asymmetry has been confusing to me.
This commit removes `InferCtxt::freshen` so that all the freshening
sites consistently use `fold_with` and it's obvious if each one is using
a new or existing `TypeFreshener`.
2025-12-22 23:32:11 +11:00
1bb9ff05c0
Remove InferCtxt::freshener.
...
I have always found this confusingly named, because it creates a new
freshener rather than returning an existing one. We can remove it and
just use `TypeFreshener::new()` at the two call sites, avoiding this
confusion.
2025-12-22 17:48:37 +11:00
4652295a3e
Simplify TypeFreshener methods.
...
`freshen_{ty,const}` take a `Result` and just do a fold if the input is
`Ok`. It's simpler to do those folds at the call site, and only call
`freshen_{ty,const}` in the `Err` case. That way we can also avoid
useless fold operations on the results of `new_{int,uint,float}`.
Also, make some `bug!` calls more concise.
2025-12-22 17:22:28 +11:00
1986be2bcd
Moved struct Placeholder<T>
2025-12-09 13:40:18 +00:00
1b71459736
emit WF goals in relate, not in generalize
2025-12-04 15:43:05 +01:00
b111aed11f
fudge infer vars in cause code intentionally
2025-12-01 12:40:49 +08:00
973c7527b4
Unify the configuration of the compiler docs
...
Previously it was rather inconsistent which crates got the rust logo and
which didn't and setting html_root_url was forgotten in many cases.
2025-11-05 11:25:27 +00:00
5f312079f2
add logging to fudge_inference_if_ok
2025-11-04 12:03:19 +01:00
844a41903e
Rollup merge of #148290 - oli-obk:push-qwxvxyopypry, r=nnethercote
...
Do not emit solver errors that contain error types
any follow-up errors are going to either be duplicates or often disappear if the error itself is fixed.
in this PR it mostly silences dyn-compat errors as all the other errors are already deduplicated outside of the test suite. The dyn compat errors are independent errors and I think if the dyn compatiblity depended on an error type it would not actually show, so this is PR is actually silencing independent errors, too.
I am opening this PR because I am seeing lots of `{type error}: const Trait` errors when adding more const checking. So instead of targetting just those specific errors, I wanted to try out fully avoiding such errors near the trait solver.
cc ````@rust-lang/types```` for thoughts
2025-11-02 09:10:38 +01:00
5cbb5d0ac6
rename OpaqueHiddenType
2025-10-31 14:53:03 +01:00
cac6f8760a
Do not emit solver errors that contain error types
2025-10-31 08:17:44 +00:00
8e9b0c4ca9
rename select_where_possible and select_all_or_error
2025-10-07 23:02:23 +01:00
4b9c62b4da
Auto merge of #147138 - jackh726:split-canonical-bound, r=lcnr
...
Split Bound index into Canonical and Bound
See [#t-types/trait-system-refactor > perf `async-closures/post-mono-higher-ranked-hang.rs`](https://rust-lang.zulipchat.com/#narrow/channel/364551-t-types.2Ftrait-system-refactor/topic/perf.20.60async-closures.2Fpost-mono-higher-ranked-hang.2Ers.60/with/541535613 ) for context
Things compile and tests pass, but not sure if this actually solves the perf issue (edit: it does). Opening up this to do a perf (and maybe crater) run.
r? lcnr
2025-10-02 08:09:33 +00:00
d4a0f21290
Rollup merge of #147199 - jdonszelmann:outdated-comment-infctx, r=lcnr
...
remove outdated comment in (inner) `InferCtxt`
This comment seems to have stopped being relevant around 3 years ago after https://github.com/rust-lang/rust/commit/9f95c605f83bcf4c158ea4b3fd5a7abb825a4178 . A map? what map? :P
r? `@lcnr`
2025-10-01 22:15:01 +10:00
bdebd479ac
remove outdated context (inner) infctx
2025-09-30 21:56:06 +02:00
80e598bb12
clone region obligations instead of taking in implied bounds hack
2025-09-30 21:48:55 +02:00
d1bbd39c59
Split Bound into Canonical and Bound
2025-09-30 12:58:28 -04:00
Jana Dönszelmann