diff options
Diffstat (limited to 'compiler/rustc_middle/src/mir/syntax.rs')
| -rw-r--r-- | compiler/rustc_middle/src/mir/syntax.rs | 207 |
1 files changed, 141 insertions, 66 deletions
diff --git a/compiler/rustc_middle/src/mir/syntax.rs b/compiler/rustc_middle/src/mir/syntax.rs index eb90169d0..c7d0283aa 100644 --- a/compiler/rustc_middle/src/mir/syntax.rs +++ b/compiler/rustc_middle/src/mir/syntax.rs @@ -23,75 +23,110 @@ use rustc_span::symbol::Symbol; use rustc_span::Span; use rustc_target::asm::InlineAsmRegOrRegClass; -/// The various "big phases" that MIR goes through. +/// Represents the "flavors" of MIR. /// -/// These phases all describe dialects of MIR. Since all MIR uses the same datastructures, the -/// dialects forbid certain variants or values in certain phases. The sections below summarize the -/// changes, but do not document them thoroughly. The full documentation is found in the appropriate -/// documentation for the thing the change is affecting. +/// All flavors of MIR use the same data structure, but there are some important differences. These +/// differences come in two forms: Dialects and phases. /// -/// Warning: ordering of variants is significant. +/// Dialects represent a stronger distinction than phases. This is because the transitions between +/// dialects are semantic changes, and therefore technically *lowerings* between distinct IRs. In +/// other words, the same [`Body`](crate::mir::Body) might be well-formed for multiple dialects, but +/// have different semantic meaning and different behavior at runtime. +/// +/// Each dialect additionally has a number of phases. However, phase changes never involve semantic +/// changes. If some MIR is well-formed both before and after a phase change, it is also guaranteed +/// that it has the same semantic meaning. In this sense, phase changes can only add additional +/// restrictions on what MIR is well-formed. +/// +/// When adding phases, remember to update [`MirPhase::phase_index`]. #[derive(Copy, Clone, TyEncodable, TyDecodable, Debug, PartialEq, Eq, PartialOrd, Ord)] #[derive(HashStable)] pub enum MirPhase { - /// The dialect of MIR used during all phases before `DropsLowered` is the same. This is also - /// the MIR that analysis such as borrowck uses. - /// - /// One important thing to remember about the behavior of this section of MIR is that drop terminators - /// (including drop and replace) are *conditional*. The elaborate drops pass will then replace each - /// instance of a drop terminator with a nop, an unconditional drop, or a drop conditioned on a drop - /// flag. Of course, this means that it is important that the drop elaboration can accurately recognize - /// when things are initialized and when things are de-initialized. That means any code running on this - /// version of MIR must be sure to produce output that drop elaboration can reason about. See the - /// section on the drop terminatorss for more details. - Built = 0, - // FIXME(oli-obk): it's unclear whether we still need this phase (and its corresponding query). - // We used to have this for pre-miri MIR based const eval. - Const = 1, - /// This phase checks the MIR for promotable elements and takes them out of the main MIR body - /// by creating a new MIR body per promoted element. After this phase (and thus the termination - /// of the `mir_promoted` query), these promoted elements are available in the `promoted_mir` - /// query. - ConstsPromoted = 2, - /// After this projections may only contain deref projections as the first element. - Derefered = 3, - /// Beginning with this phase, the following variants are disallowed: - /// * [`TerminatorKind::DropAndReplace`] + /// The MIR that is generated by MIR building. + /// + /// The only things that operate on this dialect are unsafeck, the various MIR lints, and const + /// qualifs. + /// + /// This has no distinct phases. + Built, + /// The MIR used for most analysis. + /// + /// The only semantic change between analysis and built MIR is constant promotion. In built MIR, + /// sequences of statements that would generally be subject to constant promotion are + /// semantically constants, while in analysis MIR all constants are explicit. + /// + /// The result of const promotion is available from the `mir_promoted` and `promoted_mir` queries. + /// + /// This is the version of MIR used by borrowck and friends. + Analysis(AnalysisPhase), + /// The MIR used for CTFE, optimizations, and codegen. + /// + /// The semantic changes that occur in the lowering from analysis to runtime MIR are as follows: + /// + /// - Drops: In analysis MIR, `Drop` terminators represent *conditional* drops; roughly speaking, + /// if dataflow analysis determines that the place being dropped is uninitialized, the drop will + /// not be executed. The exact semantics of this aren't written down anywhere, which means they + /// are essentially "what drop elaboration does." In runtime MIR, the drops are unconditional; + /// when a `Drop` terminator is reached, if the type has drop glue that drop glue is always + /// executed. This may be UB if the underlying place is not initialized. + /// - Packed drops: Places might in general be misaligned - in most cases this is UB, the exception + /// is fields of packed structs. In analysis MIR, `Drop(P)` for a `P` that might be misaligned + /// for this reason implicitly moves `P` to a temporary before dropping. Runtime MIR has no such + /// rules, and dropping a misaligned place is simply UB. + /// - Unwinding: in analysis MIR, unwinding from a function which may not unwind aborts. In runtime + /// MIR, this is UB. + /// - Retags: If `-Zmir-emit-retag` is enabled, analysis MIR has "implicit" retags in the same way + /// that Rust itself has them. Where exactly these are is generally subject to change, and so we + /// don't document this here. Runtime MIR has all retags explicit. + /// - Generator bodies: In analysis MIR, locals may actually be behind a pointer that user code has + /// access to. This occurs in generator bodies. Such locals do not behave like other locals, + /// because they eg may be aliased in surprising ways. Runtime MIR has no such special locals - + /// all generator bodies are lowered and so all places that look like locals really are locals. + /// - Const prop lints: The lint pass which reports eg `200_u8 + 200_u8` as an error is run as a + /// part of analysis to runtime MIR lowering. This means that transformations which may supress + /// such errors may not run on analysis MIR. + Runtime(RuntimePhase), +} + +/// See [`MirPhase::Analysis`]. +#[derive(Copy, Clone, TyEncodable, TyDecodable, Debug, PartialEq, Eq, PartialOrd, Ord)] +#[derive(HashStable)] +pub enum AnalysisPhase { + Initial = 0, + /// Beginning in this phase, the following variants are disallowed: /// * [`TerminatorKind::FalseUnwind`] /// * [`TerminatorKind::FalseEdge`] /// * [`StatementKind::FakeRead`] /// * [`StatementKind::AscribeUserType`] /// * [`Rvalue::Ref`] with `BorrowKind::Shallow` /// - /// And the following variant is allowed: - /// * [`StatementKind::Retag`] - /// - /// Furthermore, `Drop` now uses explicit drop flags visible in the MIR and reaching a `Drop` - /// terminator means that the auto-generated drop glue will be invoked. Also, `Copy` operands - /// are allowed for non-`Copy` types. - DropsLowered = 4, - /// Beginning with this phase, the following variant is disallowed: + /// Furthermore, `Deref` projections must be the first projection within any place (if they + /// appear at all) + PostCleanup = 1, +} + +/// See [`MirPhase::Runtime`]. +#[derive(Copy, Clone, TyEncodable, TyDecodable, Debug, PartialEq, Eq, PartialOrd, Ord)] +#[derive(HashStable)] +pub enum RuntimePhase { + /// In addition to the semantic changes, beginning with this phase, the following variants are + /// disallowed: + /// * [`TerminatorKind::DropAndReplace`] + /// * [`TerminatorKind::Yield`] + /// * [`TerminatorKind::GeneratorDrop`] /// * [`Rvalue::Aggregate`] for any `AggregateKind` except `Array` /// - /// And the following variant is allowed: + /// And the following variants are allowed: + /// * [`StatementKind::Retag`] /// * [`StatementKind::SetDiscriminant`] - Deaggregated = 5, - /// Before this phase, generators are in the "source code" form, featuring `yield` statements - /// and such. With this phase change, they are transformed into a proper state machine. Running - /// optimizations before this change can be potentially dangerous because the source code is to - /// some extent a "lie." In particular, `yield` terminators effectively make the value of all - /// locals visible to the caller. This means that dead store elimination before them, or code - /// motion across them, is not correct in general. This is also exasperated by type checking - /// having pre-computed a list of the types that it thinks are ok to be live across a yield - /// point - this is necessary to decide eg whether autotraits are implemented. Introducing new - /// types across a yield point will lead to ICEs becaues of this. - /// - /// Beginning with this phase, the following variants are disallowed: - /// * [`TerminatorKind::Yield`] - /// * [`TerminatorKind::GeneratorDrop`] + /// * [`StatementKind::Deinit`] + /// + /// Furthermore, `Copy` operands are allowed for non-`Copy` types. + Initial = 0, + /// Beginning with this phase, the following variant is disallowed: /// * [`ProjectionElem::Deref`] of `Box` - GeneratorsLowered = 6, - Optimized = 7, + PostCleanup = 1, + Optimized = 2, } /////////////////////////////////////////////////////////////////////////// @@ -292,12 +327,40 @@ pub enum StatementKind<'tcx> { /// executed. Coverage(Box<Coverage>), + /// Denotes a call to an intrinsic that does not require an unwind path and always returns. + /// This avoids adding a new block and a terminator for simple intrinsics. + Intrinsic(Box<NonDivergingIntrinsic<'tcx>>), + + /// No-op. Useful for deleting instructions without affecting statement indices. + Nop, +} + +#[derive( + Clone, + TyEncodable, + TyDecodable, + Debug, + PartialEq, + Hash, + HashStable, + TypeFoldable, + TypeVisitable +)] +pub enum NonDivergingIntrinsic<'tcx> { + /// Denotes a call to the intrinsic function `assume`. + /// + /// The operand must be a boolean. Optimizers may use the value of the boolean to backtrack its + /// computation to infer information about other variables. So if the boolean came from a + /// `x < y` operation, subsequent operations on `x` and `y` could elide various bound checks. + /// If the argument is `false`, this operation is equivalent to `TerminatorKind::Unreachable`. + Assume(Operand<'tcx>), + /// Denotes a call to the intrinsic function `copy_nonoverlapping`. /// /// First, all three operands are evaluated. `src` and `dest` must each be a reference, pointer, /// or `Box` pointing to the same type `T`. `count` must evaluate to a `usize`. Then, `src` and /// `dest` are dereferenced, and `count * size_of::<T>()` bytes beginning with the first byte of - /// the `src` place are copied to the continguous range of bytes beginning with the first byte + /// the `src` place are copied to the contiguous range of bytes beginning with the first byte /// of `dest`. /// /// **Needs clarification**: In what order are operands computed and dereferenced? It should @@ -305,10 +368,18 @@ pub enum StatementKind<'tcx> { /// /// **Needs clarification**: Is this typed or not, ie is there a typed load and store involved? /// I vaguely remember Ralf saying somewhere that he thought it should not be. - CopyNonOverlapping(Box<CopyNonOverlapping<'tcx>>), + CopyNonOverlapping(CopyNonOverlapping<'tcx>), +} - /// No-op. Useful for deleting instructions without affecting statement indices. - Nop, +impl std::fmt::Display for NonDivergingIntrinsic<'_> { + fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { + match self { + Self::Assume(op) => write!(f, "assume({op:?})"), + Self::CopyNonOverlapping(CopyNonOverlapping { src, dst, count }) => { + write!(f, "copy_nonoverlapping(dst = {dst:?}, src = {src:?}, count = {count:?})") + } + } + } } /// Describes what kind of retag is to be performed. @@ -343,7 +414,7 @@ pub enum FakeReadCause { /// Some(closure_def_id). /// Otherwise, the value of the optional LocalDefId will be None. // - // We can use LocaDefId here since fake read statements are removed + // We can use LocalDefId here since fake read statements are removed // before codegen in the `CleanupNonCodegenStatements` pass. ForMatchedPlace(Option<LocalDefId>), @@ -417,7 +488,7 @@ pub struct CopyNonOverlapping<'tcx> { /// must also be `cleanup`. This is a part of the type system and checked statically, so it is /// still an error to have such an edge in the CFG even if it's known that it won't be taken at /// runtime. -#[derive(Clone, TyEncodable, TyDecodable, Hash, HashStable, PartialEq)] +#[derive(Clone, TyEncodable, TyDecodable, Hash, HashStable, PartialEq, TypeFoldable, TypeVisitable)] pub enum TerminatorKind<'tcx> { /// Block has one successor; we continue execution there. Goto { target: BasicBlock }, @@ -670,7 +741,7 @@ pub enum TerminatorKind<'tcx> { } /// Information about an assertion failure. -#[derive(Clone, TyEncodable, TyDecodable, Hash, HashStable, PartialEq, PartialOrd)] +#[derive(Clone, TyEncodable, TyDecodable, Hash, HashStable, PartialEq, TypeFoldable, TypeVisitable)] pub enum AssertKind<O> { BoundsCheck { len: O, index: O }, Overflow(BinOp, O, O), @@ -792,7 +863,7 @@ pub type AssertMessage<'tcx> = AssertKind<Operand<'tcx>>; /// /// Rust currently requires that every place obey those two rules. This is checked by MIRI and taken /// advantage of by codegen (via `gep inbounds`). That is possibly subject to change. -#[derive(Copy, Clone, PartialEq, Eq, Hash, TyEncodable, HashStable)] +#[derive(Copy, Clone, PartialEq, Eq, Hash, TyEncodable, HashStable, TypeFoldable, TypeVisitable)] pub struct Place<'tcx> { pub local: Local, @@ -801,7 +872,7 @@ pub struct Place<'tcx> { } #[derive(Copy, Clone, Debug, PartialEq, Eq, PartialOrd, Ord, Hash)] -#[derive(TyEncodable, TyDecodable, HashStable)] +#[derive(TyEncodable, TyDecodable, HashStable, TypeFoldable, TypeVisitable)] pub enum ProjectionElem<V, T> { Deref, Field(Field, T), @@ -884,7 +955,7 @@ pub type PlaceElem<'tcx> = ProjectionElem<Local, Ty<'tcx>>; /// **Needs clarifiation:** Is loading a place that has its variant index set well-formed? Miri /// currently implements it, but it seems like this may be something to check against in the /// validator. -#[derive(Clone, PartialEq, TyEncodable, TyDecodable, Hash, HashStable)] +#[derive(Clone, PartialEq, TyEncodable, TyDecodable, Hash, HashStable, TypeFoldable, TypeVisitable)] pub enum Operand<'tcx> { /// Creates a value by loading the given place. /// @@ -915,7 +986,7 @@ pub enum Operand<'tcx> { /// Computing any rvalue begins by evaluating the places and operands in some order (**Needs /// clarification**: Which order?). These are then used to produce a "value" - the same kind of /// value that an [`Operand`] produces. -#[derive(Clone, TyEncodable, TyDecodable, Hash, HashStable, PartialEq)] +#[derive(Clone, TyEncodable, TyDecodable, Hash, HashStable, PartialEq, TypeFoldable, TypeVisitable)] pub enum Rvalue<'tcx> { /// Yields the operand unchanged Use(Operand<'tcx>), @@ -1068,11 +1139,14 @@ pub enum CastKind { /// All sorts of pointer-to-pointer casts. Note that reference-to-raw-ptr casts are /// translated into `&raw mut/const *r`, i.e., they are not actually casts. Pointer(PointerCast), + /// Cast into a dyn* object. + DynStar, /// Remaining unclassified casts. Misc, } #[derive(Clone, Debug, PartialEq, Eq, TyEncodable, TyDecodable, Hash, HashStable)] +#[derive(TypeFoldable, TypeVisitable)] pub enum AggregateKind<'tcx> { /// The type is of the element Array(Ty<'tcx>), @@ -1160,7 +1234,8 @@ pub enum BinOp { mod size_asserts { use super::*; // These are in alphabetical order, which is easy to maintain. - static_assert_size!(AggregateKind<'_>, 48); + #[cfg(not(bootstrap))] + static_assert_size!(AggregateKind<'_>, 40); static_assert_size!(Operand<'_>, 24); static_assert_size!(Place<'_>, 16); static_assert_size!(PlaceElem<'_>, 24); |