diff options
| author | bors <bors@rust-lang.org> | 2024-11-20 18:51:54 +0000 |
|---|---|---|
| committer | bors <bors@rust-lang.org> | 2024-11-20 18:51:54 +0000 |
| commit | 3fee0f12e4f595948f8f54f57c8b7a7a58127124 (patch) | |
| tree | 16a282355e2224233cf1d36c193070fe6bba6c40 /compiler/rustc_middle/src/query/mod.rs | |
| parent | a1f299953656f95004c69b24ad8071d6899fa9da (diff) | |
| parent | 297b618944d4619a1990b1992d9142c6cf893dc6 (diff) | |
| download | rust-3fee0f12e4f595948f8f54f57c8b7a7a58127124.tar.gz rust-3fee0f12e4f595948f8f54f57c8b7a7a58127124.zip | |
Auto merge of #131326 - dingxiangfei2009:issue-130836-attempt-2, r=nikomatsakis
Reduce false positives of tail-expr-drop-order from consumed values (attempt #2) r? `@nikomatsakis` Tracked by #123739. Related to #129864 but not replacing, yet. Related to #130836. This is an implementation of the approach suggested in the [Zulip stream](https://rust-lang.zulipchat.com/#narrow/stream/213817-t-lang/topic/temporary.20drop.20order.20changes). A new MIR statement `BackwardsIncompatibleDrop` is added to the MIR syntax. The lint now works by inspecting possibly live move paths before at the `BackwardsIncompatibleDrop` location and the actual drop under the current edition, which should be one before Edition 2024 in practice.
Diffstat (limited to 'compiler/rustc_middle/src/query/mod.rs')
| -rw-r--r-- | compiler/rustc_middle/src/query/mod.rs | 22 |
1 files changed, 22 insertions, 0 deletions
diff --git a/compiler/rustc_middle/src/query/mod.rs b/compiler/rustc_middle/src/query/mod.rs index 9ae519dfe7e..3fdb38a433e 100644 --- a/compiler/rustc_middle/src/query/mod.rs +++ b/compiler/rustc_middle/src/query/mod.rs @@ -1448,6 +1448,28 @@ rustc_queries! { cache_on_disk_if { false } } + /// Returns a list of types which (a) have a potentially significant destructor + /// and (b) may be dropped as a result of dropping a value of some type `ty` + /// (in the given environment). + /// + /// The idea of "significant" drop is somewhat informal and is used only for + /// diagnostics and edition migrations. The idea is that a significant drop may have + /// some visible side-effect on execution; freeing memory is NOT considered a side-effect. + /// The rules are as follows: + /// * Type with no explicit drop impl do not have significant drop. + /// * Types with a drop impl are assumed to have significant drop unless they have a `#[rustc_insignificant_dtor]` annotation. + /// + /// Note that insignificant drop is a "shallow" property. A type like `Vec<LockGuard>` does not + /// have significant drop but the type `LockGuard` does, and so if `ty = Vec<LockGuard>` + /// then the return value would be `&[LockGuard]`. + /// *IMPORTANT*: *DO NOT* run this query before promoted MIR body is constructed, + /// because this query partially depends on that query. + /// Otherwise, there is a risk of query cycles. + query list_significant_drop_tys(ty: ty::ParamEnvAnd<'tcx, Ty<'tcx>>) -> &'tcx ty::List<Ty<'tcx>> { + desc { |tcx| "computing when `{}` has a significant destructor", ty.value } + cache_on_disk_if { false } + } + /// Computes the layout of a type. Note that this implicitly /// executes in "reveal all" mode, and will normalize the input type. query layout_of( |
