Auto merge of #114308 - matthiaskrgr:rollup-m64bkm7, r=matthiaskrgr

Rollup of 7 pull requests

Successful merges:

 - #109318 (Make `Debug` representations of `[Lazy, Once]*[Cell, Lock]` consistent with `Mutex` and `RwLock`)
 - #113701 (Re-export core::ffi::FromBytesUntilNulError in std::ffi)
 - #113804 (Resolve correct archive version name in `opt-dist`)
 - #114165 (Add missing rvalues to smir)
 - #114182 (clean up after 113312)
 - #114193 (Update lexer emoji diagnostics to Unicode 15.0)
 - #114200 (Detect trait upcasting through struct tail unsizing in new solver select)

r? `@ghost`
`@rustbot` modify labels: rollup

Cargo.lock

@@ -3786,7 +3786,7 @@ name = "rustc_lexer"
 version = "0.1.0"
 dependencies = [
  "expect-test",
- "unic-emoji-char",
+ "unicode-properties",
  "unicode-xid",
 ]
 
@@ -5446,38 +5446,6 @@ dependencies = [
  "tempfile",
 ]
 
-[[package]]
-name = "unic-char-property"
-version = "0.9.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a8c57a407d9b6fa02b4795eb81c5b6652060a15a7903ea981f3d723e6c0be221"
-dependencies = [
- "unic-char-range",
-]
-
-[[package]]
-name = "unic-char-range"
-version = "0.9.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0398022d5f700414f6b899e10b8348231abf9173fa93144cbc1a43b9793c1fbc"
-
-[[package]]
-name = "unic-common"
-version = "0.9.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "80d7ff825a6a654ee85a63e80f92f054f904f21e7d12da4e22f9834a4aaa35bc"
-
-[[package]]
-name = "unic-emoji-char"
-version = "0.9.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0b07221e68897210270a38bde4babb655869637af0f69407f96053a34f76494d"
-dependencies = [
- "unic-char-property",
- "unic-char-range",
- "unic-ucd-version",
-]
-
 [[package]]
 name = "unic-langid"
 version = "0.9.1"
@@ -5521,15 +5489,6 @@ dependencies = [
  "unic-langid-impl",
 ]
 
-[[package]]
-name = "unic-ucd-version"
-version = "0.9.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "96bd2f2237fe450fcd0a1d2f5f4e91711124f7857ba2e964247776ebeeb7b0c4"
-dependencies = [
- "unic-common",
-]
-
 [[package]]
 name = "unicase"
 version = "2.6.0"
@@ -5567,6 +5526,12 @@ dependencies = [
  "tinyvec",
 ]
 
+[[package]]
+name = "unicode-properties"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c7f91c8b21fbbaa18853c3d0801c78f4fc94cdb976699bb03e832e75f7fd22f0"
+
 [[package]]
 name = "unicode-script"
 version = "0.5.5"

compiler/rustc_lexer/Cargo.toml

@@ -16,7 +16,11 @@ Rust lexer used by rustc. No stability guarantees are provided.
 # Note that this crate purposefully does not depend on other rustc crates
 [dependencies]
 unicode-xid = "0.2.0"
-unic-emoji-char = "0.9.0"
+
+[dependencies.unicode-properties]
+version = "0.1.0"
+default-features = false
+features = ["emoji"]
 
 [dev-dependencies]
 expect-test = "1.4.0"

compiler/rustc_lexer/src/lib.rs

@@ -34,6 +34,7 @@ pub use crate::cursor::Cursor;
 use self::LiteralKind::*;
 use self::TokenKind::*;
 use crate::cursor::EOF_CHAR;
+use unicode_properties::UnicodeEmoji;
 
 /// Parsed token.
 /// It doesn't contain information about data that has been parsed,
@@ -428,9 +429,7 @@ impl Cursor<'_> {
                 Literal { kind, suffix_start }
             }
             // Identifier starting with an emoji. Only lexed for graceful error recovery.
-            c if !c.is_ascii() && unic_emoji_char::is_emoji(c) => {
-                self.fake_ident_or_unknown_prefix()
-            }
+            c if !c.is_ascii() && c.is_emoji_char() => self.fake_ident_or_unknown_prefix(),
             _ => Unknown,
         };
         let res = Token::new(token_kind, self.pos_within_token());
@@ -514,9 +513,7 @@ impl Cursor<'_> {
         // we see a prefix here, it is definitely an unknown prefix.
         match self.first() {
             '#' | '"' | '\'' => UnknownPrefix,
-            c if !c.is_ascii() && unic_emoji_char::is_emoji(c) => {
-                self.fake_ident_or_unknown_prefix()
-            }
+            c if !c.is_ascii() && c.is_emoji_char() => self.fake_ident_or_unknown_prefix(),
             _ => Ident,
         }
     }
@@ -525,7 +522,7 @@ impl Cursor<'_> {
         // Start is already eaten, eat the rest of identifier.
         self.eat_while(|c| {
             unicode_xid::UnicodeXID::is_xid_continue(c)
-                || (!c.is_ascii() && unic_emoji_char::is_emoji(c))
+                || (!c.is_ascii() && c.is_emoji_char())
                 || c == '\u{200d}'
         });
         // Known prefixes must have been handled earlier. So if

compiler/rustc_smir/src/rustc_smir/mod.rs

@@ -132,7 +132,7 @@ impl<'tcx> Stable<'tcx> for mir::Rvalue<'tcx> {
         use mir::Rvalue::*;
         match self {
             Use(op) => stable_mir::mir::Rvalue::Use(op.stable(tables)),
-            Repeat(_, _) => todo!(),
+            Repeat(op, len) => stable_mir::mir::Rvalue::Repeat(op.stable(tables), opaque(len)),
             Ref(region, kind, place) => stable_mir::mir::Rvalue::Ref(
                 opaque(region),
                 kind.stable(tables),
@@ -145,7 +145,11 @@ impl<'tcx> Stable<'tcx> for mir::Rvalue<'tcx> {
                 stable_mir::mir::Rvalue::AddressOf(mutability.stable(tables), place.stable(tables))
             }
             Len(place) => stable_mir::mir::Rvalue::Len(place.stable(tables)),
-            Cast(_, _, _) => todo!(),
+            Cast(cast_kind, op, ty) => stable_mir::mir::Rvalue::Cast(
+                cast_kind.stable(tables),
+                op.stable(tables),
+                tables.intern_ty(*ty),
+            ),
             BinaryOp(bin_op, ops) => stable_mir::mir::Rvalue::BinaryOp(
                 bin_op.stable(tables),
                 ops.0.stable(tables),
@@ -163,8 +167,13 @@ impl<'tcx> Stable<'tcx> for mir::Rvalue<'tcx> {
                 stable_mir::mir::Rvalue::UnaryOp(un_op.stable(tables), op.stable(tables))
             }
             Discriminant(place) => stable_mir::mir::Rvalue::Discriminant(place.stable(tables)),
-            Aggregate(_, _) => todo!(),
-            ShallowInitBox(_, _) => todo!(),
+            Aggregate(agg_kind, operands) => {
+                let operands = operands.iter().map(|op| op.stable(tables)).collect();
+                stable_mir::mir::Rvalue::Aggregate(agg_kind.stable(tables), operands)
+            }
+            ShallowInitBox(op, ty) => {
+                stable_mir::mir::Rvalue::ShallowInitBox(op.stable(tables), tables.intern_ty(*ty))
+            }
             CopyForDeref(place) => stable_mir::mir::Rvalue::CopyForDeref(place.stable(tables)),
         }
     }
@@ -478,6 +487,40 @@ impl<'tcx> Stable<'tcx> for mir::UnOp {
     }
 }
 
+impl<'tcx> Stable<'tcx> for mir::AggregateKind<'tcx> {
+    type T = stable_mir::mir::AggregateKind;
+    fn stable(&self, tables: &mut Tables<'tcx>) -> Self::T {
+        match self {
+            mir::AggregateKind::Array(ty) => {
+                stable_mir::mir::AggregateKind::Array(tables.intern_ty(*ty))
+            }
+            mir::AggregateKind::Tuple => stable_mir::mir::AggregateKind::Tuple,
+            mir::AggregateKind::Adt(def_id, var_idx, generic_arg, user_ty_index, field_idx) => {
+                stable_mir::mir::AggregateKind::Adt(
+                    rustc_internal::adt_def(*def_id),
+                    var_idx.index(),
+                    generic_arg.stable(tables),
+                    user_ty_index.map(|idx| idx.index()),
+                    field_idx.map(|idx| idx.index()),
+                )
+            }
+            mir::AggregateKind::Closure(def_id, generic_arg) => {
+                stable_mir::mir::AggregateKind::Closure(
+                    rustc_internal::closure_def(*def_id),
+                    generic_arg.stable(tables),
+                )
+            }
+            mir::AggregateKind::Generator(def_id, generic_arg, movability) => {
+                stable_mir::mir::AggregateKind::Generator(
+                    rustc_internal::generator_def(*def_id),
+                    generic_arg.stable(tables),
+                    movability.stable(tables),
+                )
+            }
+        }
+    }
+}
+
 impl<'tcx> Stable<'tcx> for rustc_hir::GeneratorKind {
     type T = stable_mir::mir::GeneratorKind;
     fn stable(&self, _: &mut Tables<'tcx>) -> Self::T {

compiler/rustc_smir/src/stable_mir/mir/body.rs

@@ -1,4 +1,6 @@
-use crate::stable_mir::ty::Region;
+use crate::stable_mir::ty::{
+    AdtDef, ClosureDef, Const, GeneratorDef, GenericArgs, Movability, Region,
+};
 use crate::stable_mir::{self, ty::Ty};
 
 #[derive(Clone, Debug)]
@@ -137,7 +139,6 @@ pub enum Statement {
     Nop,
 }
 
-// FIXME this is incomplete
 #[derive(Clone, Debug)]
 pub enum Rvalue {
     /// Creates a pointer with the indicated mutability to the place.
@@ -146,6 +147,16 @@ pub enum Rvalue {
     /// `&raw v` or `addr_of!(v)`.
     AddressOf(Mutability, Place),
 
+    /// Creates an aggregate value, like a tuple or struct.
+    ///
+    /// This is needed because dataflow analysis needs to distinguish
+    /// `dest = Foo { x: ..., y: ... }` from `dest.x = ...; dest.y = ...;` in the case that `Foo`
+    /// has a destructor.
+    ///
+    /// Disallowed after deaggregation for all aggregate kinds except `Array` and `Generator`. After
+    /// generator lowering, `Generator` aggregate kinds are disallowed too.
+    Aggregate(AggregateKind, Vec<Operand>),
+
     /// * `Offset` has the same semantics as [`offset`](pointer::offset), except that the second
     ///   parameter may be a `usize` as well.
     /// * The comparison operations accept `bool`s, `char`s, signed or unsigned integers, floats,
@@ -198,6 +209,16 @@ pub enum Rvalue {
     /// Creates a reference to the place.
     Ref(Region, BorrowKind, Place),
 
+    /// Creates an array where each element is the value of the operand.
+    ///
+    /// This is the cause of a bug in the case where the repetition count is zero because the value
+    /// is not dropped, see [#74836].
+    ///
+    /// Corresponds to source code like `[x; 32]`.
+    ///
+    /// [#74836]: https://github.com/rust-lang/rust/issues/74836
+    Repeat(Operand, Const),
+
     /// Transmutes a `*mut u8` into shallow-initialized `Box<T>`.
     ///
     /// This is different from a normal transmute because dataflow analysis will treat the box as
@@ -232,6 +253,15 @@ pub enum Rvalue {
     Use(Operand),
 }
 
+#[derive(Clone, Debug)]
+pub enum AggregateKind {
+    Array(Ty),
+    Tuple,
+    Adt(AdtDef, VariantIdx, GenericArgs, Option<UserTypeAnnotationIndex>, Option<FieldIdx>),
+    Closure(ClosureDef, GenericArgs),
+    Generator(GeneratorDef, GenericArgs, Movability),
+}
+
 #[derive(Clone, Debug)]
 pub enum Operand {
     Copy(Place),
@@ -247,6 +277,11 @@ pub struct Place {
 
 type FieldIdx = usize;
 
+/// The source-order index of a variant in a type.
+type VariantIdx = usize;
+
+type UserTypeAnnotationIndex = usize;
+
 #[derive(Clone, Debug)]
 pub struct SwitchTarget {
     pub value: u128,

compiler/rustc_smir/src/stable_mir/ty.rs

@@ -10,7 +10,7 @@ impl Ty {
     }
 }
 
-type Const = Opaque;
+pub(crate) type Const = Opaque;
 pub(crate) type Region = Opaque;
 type Span = Opaque;
 

compiler/rustc_trait_selection/src/solve/eval_ctxt/select.rs

@@ -100,10 +100,18 @@ impl<'tcx> InferCtxtSelectExt<'tcx> for InferCtxt<'tcx> {
                 rematch_impl(self, goal, def_id, nested_obligations)
             }
 
+            // If an unsize goal is ambiguous, then we can manually rematch it to make
+            // selection progress for coercion during HIR typeck. If it is *not* ambiguous,
+            // but is `BuiltinImplSource::Misc`, it may have nested `Unsize` goals,
+            // and we need to rematch those to detect tuple unsizing and trait upcasting.
+            // FIXME: This will be wrong if we have param-env or where-clause bounds
+            // with the unsize goal -- we may need to mark those with different impl
+            // sources.
             (Certainty::Maybe(_), CandidateSource::BuiltinImpl(src))
+            | (Certainty::Yes, CandidateSource::BuiltinImpl(src @ BuiltinImplSource::Misc))
                 if self.tcx.lang_items().unsize_trait() == Some(goal.predicate.def_id()) =>
             {
-                rematch_unsize(self, goal, nested_obligations, src)
+                rematch_unsize(self, goal, nested_obligations, src, certainty)
             }
 
             // Technically some builtin impls have nested obligations, but if
@@ -217,6 +225,7 @@ fn rematch_unsize<'tcx>(
     goal: Goal<'tcx, ty::TraitPredicate<'tcx>>,
     mut nested: Vec<PredicateObligation<'tcx>>,
     source: BuiltinImplSource,
+    certainty: Certainty,
 ) -> SelectionResult<'tcx, Selection<'tcx>> {
     let tcx = infcx.tcx;
     let a_ty = structurally_normalize(goal.predicate.self_ty(), infcx, goal.param_env, &mut nested);
@@ -227,6 +236,12 @@ fn rematch_unsize<'tcx>(
         &mut nested,
     );
     match (a_ty.kind(), b_ty.kind()) {
+        // Stall any ambiguous upcasting goals, since we can't rematch those
+        (ty::Dynamic(_, _, ty::Dyn), ty::Dynamic(_, _, ty::Dyn)) => match certainty {
+            Certainty::Yes => Ok(Some(ImplSource::Builtin(source, nested))),
+            _ => Ok(None),
+        },
+        // `T` -> `dyn Trait` upcasting
         (_, &ty::Dynamic(data, region, ty::Dyn)) => {
             // Check that the type implements all of the predicates of the def-id.
             // (i.e. the principal, all of the associated types match, and any auto traits)
@@ -354,10 +369,10 @@ fn rematch_unsize<'tcx>(
             );
             Ok(Some(ImplSource::Builtin(source, nested)))
         }
-        // FIXME: We *could* ICE here if either:
-        // 1. the certainty is `Certainty::Yes`,
-        // 2. we're in codegen (which should mean `Certainty::Yes`).
-        _ => Ok(None),
+        _ => {
+            assert_ne!(certainty, Certainty::Yes);
+            Ok(None)
+        }
     }
 }
 

library/core/src/cell/once.rs

@@ -250,10 +250,12 @@ impl<T> Default for OnceCell<T> {
 #[stable(feature = "once_cell", since = "1.70.0")]
 impl<T: fmt::Debug> fmt::Debug for OnceCell<T> {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        let mut d = f.debug_tuple("OnceCell");
         match self.get() {
-            Some(v) => f.debug_tuple("OnceCell").field(v).finish(),
-            None => f.write_str("OnceCell(Uninit)"),
-        }
+            Some(v) => d.field(v),
+            None => d.field(&format_args!("<uninit>")),
+        };
+        d.finish()
     }
 }
 

library/core/src/fmt/mod.rs

@@ -2521,22 +2521,12 @@ impl<T: Copy + Debug> Debug for Cell<T> {
 #[stable(feature = "rust1", since = "1.0.0")]
 impl<T: ?Sized + Debug> Debug for RefCell<T> {
     fn fmt(&self, f: &mut Formatter<'_>) -> Result {
+        let mut d = f.debug_struct("RefCell");
         match self.try_borrow() {
-            Ok(borrow) => f.debug_struct("RefCell").field("value", &borrow).finish(),
-            Err(_) => {
-                // The RefCell is mutably borrowed so we can't look at its value
-                // here. Show a placeholder instead.
-                struct BorrowedPlaceholder;
-
-                impl Debug for BorrowedPlaceholder {
-                    fn fmt(&self, f: &mut Formatter<'_>) -> Result {
-                        f.write_str("<borrowed>")
-                    }
-                }
-
-                f.debug_struct("RefCell").field("value", &BorrowedPlaceholder).finish()
-            }
-        }
+            Ok(borrow) => d.field("value", &borrow),
+            Err(_) => d.field("value", &format_args!("<borrowed>")),
+        };
+        d.finish()
     }
 }