| Age | Commit message (Collapse) | Author | Lines | |
|---|---|---|---|---|
| 2023-10-24 | Auto merge of #116300 - cjgillot:split-move, r=petrochenkov | bors | -253/+125 | |
| Separate move path tracking between borrowck and drop elaboration. The primary goal of this PR is to skip creating a `MovePathIndex` for path that do not need dropping in drop elaboration. The 2 first commits are cleanups. The next 2 commits displace `move` errors from move-path builder to borrowck. Move-path builder keeps the same logic, but does not carry error information any more. The remaining commits allow to filter `MovePathIndex` creation according to types. This is used in drop elaboration, to avoid computing dataflow for paths that do not need dropping. | ||||
| 2023-10-23 | Auto merge of #116033 - bvanjoi:fix-116032, r=petrochenkov | bors | -1/+1 | |
| report `unused_import` for empty reexports even it is pub Fixes #116032 An easy fix. r? `@petrochenkov` (Discovered this issue while reviewing #115993.) | ||||
| 2023-10-22 | use visibility to check unused imports and delete some stmts | bohan | -1/+1 | |
| 2023-10-21 | Use terse form for Fn bound. | Camille GILLOT | -16/+4 | |
| 2023-10-21 | Remove on_all_drop_children_bits. | Camille GILLOT | -27/+3 | |
| As drop elaboration only tracks places that need dropping, is has become equivalent to `on_all_children_bits`. | ||||
| 2023-10-21 | Parametrize gather_moves by filter. | Camille GILLOT | -60/+84 | |
| 2023-10-21 | Stop computing error info in move path builder. | Camille GILLOT | -82/+26 | |
| 2023-10-21 | Avoid using a magic value for untracked locals. | Camille GILLOT | -25/+32 | |
| 2023-10-21 | Do not report errors from move path builder. | Camille GILLOT | -17/+8 | |
| 2023-10-21 | Remove is_terminal_path. | Camille GILLOT | -59/+1 | |
| That information is redundant. Is the path is terminal, `first_child` will already be `None`. | ||||
| 2023-10-21 | Improve naming and comments. | Camille GILLOT | -17/+45 | |
| 2023-10-21 | Implement JumpThreading pass. | Camille GILLOT | -18/+89 | |
| 2023-10-20 | s/generator/coroutine/ | Oli Scherer | -13/+13 | |
| 2023-10-20 | s/Generator/Coroutine/ | Oli Scherer | -11/+11 | |
| 2023-10-13 | Format all the let chains in compiler | Michael Goulet | -4/+8 | |
| 2023-10-06 | Preserve DebugInfo in DeadStoreElimination. | Camille GILLOT | -0/+21 | |
| 2023-10-02 | have better explanation for `relate_types` | ouz-a | -4/+3 | |
| 2023-10-02 | Add docs, remove code, change subtyper code | ouz-a | -4/+5 | |
| 2023-10-02 | subtyping_projections | ouz-a | -0/+4 | |
| 2023-09-28 | Some tracing cleanups | Oli Scherer | -4/+3 | |
| 2023-09-24 | Auto merge of #115794 - cjgillot:aggregate-no-box, r=davidtwco | bors | -1/+3 | |
| Do not create a DerefLen place for `Box<[T]>`. Fixes https://github.com/rust-lang/rust/issues/115789 | ||||
| 2023-09-23 | Remove GeneratorWitness and rename GeneratorWitnessMIR. | Camille GILLOT | -4/+2 | |
| 2023-09-22 | Rollup merge of #115770 - ouz-a:match_elem_builder, r=lcnr | Matthias Krüger | -32/+113 | |
| Match on elem first while building move paths While working on https://github.com/rust-lang/rust/pull/115025 `@lcnr` and I observed "move_paths_for" function matched on the `Ty` instead of `Projection` which seems flawed as it's the `Projection`s that cause the problem not the type. r? `@lcnr` | ||||
| 2023-09-22 | match array for constantindex and subslice | ouz-a | -8/+9 | |
| 2023-09-22 | better bug message | ouz-a | -3/+7 | |
| 2023-09-22 | remove inner match | ouz-a | -6/+4 | |
| 2023-09-21 | rename mir::Constant -> mir::ConstOperand, mir::ConstKind -> mir::Const | Ralf Jung | -5/+5 | |
| 2023-09-20 | Do not create a DerefLen place for `Box<[T]>`. | Camille GILLOT | -1/+3 | |
| 2023-09-12 | very verbose error handling | ouz-a | -20/+86 | |
| 2023-09-12 | match on elem first | ouz-a | -33/+45 | |
| 2023-09-11 | Support non-scalar constants. | Camille GILLOT | -1/+26 | |
| 2023-09-07 | Correct comment and assumption. | Camille GILLOT | -6/+5 | |
| 2023-09-06 | Support array length. | Camille GILLOT | -0/+36 | |
| 2023-09-05 | Support non-trivial scalars in ConstProp. | Camille GILLOT | -45/+29 | |
| 2023-09-02 | Take `&mut Results` in `ResultsVisitor` | Jason Newcomb | -12/+12 | |
| 2023-08-24 | when terminating during unwinding, show the reason why | Ralf Jung | -10/+10 | |
| 2023-08-20 | give some unwind-related terminators a more clear name | Ralf Jung | -10/+10 | |
| 2023-08-16 | Use Terminator::edges for backward analysis too. | Camille GILLOT | -1/+1 | |
| 2023-08-16 | Only evaluate yield place after resume in liveness. | Camille GILLOT | -15/+31 | |
| 2023-08-16 | Specify that method only applies statement effects. | Camille GILLOT | -9/+14 | |
| 2023-08-16 | Rename YieldResumeEffect. | Camille GILLOT | -6/+6 | |
| 2023-08-16 | Make TerminatorEdge plural. | Camille GILLOT | -39/+39 | |
| 2023-08-16 | Rename MaybeUnreachable. | Camille GILLOT | -47/+53 | |
| 2023-08-16 | Use TerminatorEdge for dataflow-const-prop. | Camille GILLOT | -18/+31 | |
| 2023-08-16 | Only run MaybeInitializedPlaces once for drop elaboration. | Camille GILLOT | -9/+63 | |
| 2023-08-16 | Allow apply_terminator_effect to customize edges. | Camille GILLOT | -295/+177 | |
| 2023-08-16 | Introduce MaybeUnreachable. | Camille GILLOT | -3/+136 | |
| 2023-08-16 | Move domain_size to GenKillAnalysis. | Camille GILLOT | -14/+39 | |
| 2023-08-16 | Move initialization dataflow impls into their own module. | Camille GILLOT | -749/+756 | |
| 2023-08-16 | Create bottom on-the-fly instead of cloning it. | Camille GILLOT | -3/+4 | |
