Rollup merge of #109953 - thomcc:thomcc/typeid128, r=WaffleLapkin

Use 128 bits for TypeId hash

Preliminary/Draft impl of https://github.com/rust-lang/compiler-team/issues/608

Prior art (probably incomplete list)
- https://github.com/rust-lang/rust/pull/75923
- https://github.com/rust-lang/rust/pull/95845

compiler/rustc_const_eval/src/interpret/intrinsics.rs

@@ -77,7 +77,7 @@ pub(crate) fn eval_nullary_intrinsic<'tcx>(
         }
         sym::type_id => {
             ensure_monomorphic_enough(tcx, tp_ty)?;
-            ConstValue::from_u64(tcx.type_id_hash(tp_ty).as_u64())
+            ConstValue::from_u128(tcx.type_id_hash(tp_ty).as_u128())
         }
         sym::variant_count => match tp_ty.kind() {
             // Correctly handles non-monomorphic calls, so there is no need for ensure_monomorphic_enough.
@@ -169,7 +169,7 @@ impl<'mir, 'tcx: 'mir, M: Machine<'mir, 'tcx>> InterpCx<'mir, 'tcx, M> {
                 let ty = match intrinsic_name {
                     sym::pref_align_of | sym::variant_count => self.tcx.types.usize,
                     sym::needs_drop => self.tcx.types.bool,
-                    sym::type_id => self.tcx.types.u64,
+                    sym::type_id => self.tcx.types.u128,
                     sym::type_name => self.tcx.mk_static_str(),
                     _ => bug!(),
                 };

compiler/rustc_hir_analysis/src/check/intrinsic.rs

@@ -217,7 +217,7 @@ pub fn check_intrinsic_type(tcx: TyCtxt<'_>, it: &hir::ForeignItem<'_>) {
             sym::needs_drop => (1, Vec::new(), tcx.types.bool),
 
             sym::type_name => (1, Vec::new(), tcx.mk_static_str()),
-            sym::type_id => (1, Vec::new(), tcx.types.u64),
+            sym::type_id => (1, Vec::new(), tcx.types.u128),
             sym::offset => (2, vec![param(0), param(1)], param(0)),
             sym::arith_offset => (
                 1,

compiler/rustc_middle/src/mir/interpret/value.rs

@@ -97,6 +97,10 @@ impl<'tcx> ConstValue<'tcx> {
         ConstValue::Scalar(Scalar::from_u64(i))
     }
 
+    pub fn from_u128(i: u128) -> Self {
+        ConstValue::Scalar(Scalar::from_u128(i))
+    }
+
     pub fn from_target_usize(i: u64, cx: &impl HasDataLayout) -> Self {
         ConstValue::Scalar(Scalar::from_target_usize(i, cx))
     }
@@ -240,6 +244,11 @@ impl<Prov> Scalar<Prov> {
         Scalar::Int(i.into())
     }
 
+    #[inline]
+    pub fn from_u128(i: u128) -> Self {
+        Scalar::Int(i.into())
+    }
+
     #[inline]
     pub fn from_target_usize(i: u64, cx: &impl HasDataLayout) -> Self {
         Self::from_uint(i, cx.data_layout().pointer_size)

compiler/rustc_middle/src/ty/util.rs

@@ -11,7 +11,7 @@ use crate::ty::{
 use crate::ty::{GenericArgKind, SubstsRef};
 use rustc_apfloat::Float as _;
 use rustc_data_structures::fx::{FxHashMap, FxHashSet};
-use rustc_data_structures::stable_hasher::{Hash64, HashStable, StableHasher};
+use rustc_data_structures::stable_hasher::{Hash128, HashStable, StableHasher};
 use rustc_errors::ErrorGuaranteed;
 use rustc_hir as hir;
 use rustc_hir::def::{CtorOf, DefKind, Res};
@@ -129,7 +129,7 @@ impl IntTypeExt for IntegerType {
 impl<'tcx> TyCtxt<'tcx> {
     /// Creates a hash of the type `Ty` which will be the same no matter what crate
     /// context it's calculated within. This is used by the `type_id` intrinsic.
-    pub fn type_id_hash(self, ty: Ty<'tcx>) -> Hash64 {
+    pub fn type_id_hash(self, ty: Ty<'tcx>) -> Hash128 {
         // We want the type_id be independent of the types free regions, so we
         // erase them. The erase_regions() call will also anonymize bound
         // regions, which is desirable too.

library/core/src/any.rs

@@ -153,6 +153,7 @@
 #![stable(feature = "rust1", since = "1.0.0")]
 
 use crate::fmt;
+use crate::hash;
 use crate::intrinsics;
 
 ///////////////////////////////////////////////////////////////////////////////
@@ -662,10 +663,10 @@ impl dyn Any + Send + Sync {
 /// While `TypeId` implements `Hash`, `PartialOrd`, and `Ord`, it is worth
 /// noting that the hashes and ordering will vary between Rust releases. Beware
 /// of relying on them inside of your code!
-#[derive(Clone, Copy, Debug, Hash, Eq, PartialOrd, Ord)]
+#[derive(Clone, Copy, Debug, Eq, PartialOrd, Ord)]
 #[stable(feature = "rust1", since = "1.0.0")]
 pub struct TypeId {
-    t: u64,
+    t: u128,
 }
 
 #[stable(feature = "rust1", since = "1.0.0")]
@@ -696,7 +697,31 @@ impl TypeId {
     #[stable(feature = "rust1", since = "1.0.0")]
     #[rustc_const_unstable(feature = "const_type_id", issue = "77125")]
     pub const fn of<T: ?Sized + 'static>() -> TypeId {
-        TypeId { t: intrinsics::type_id::<T>() }
+        #[cfg(bootstrap)]
+        let t = intrinsics::type_id::<T>() as u128;
+        #[cfg(not(bootstrap))]
+        let t: u128 = intrinsics::type_id::<T>();
+        TypeId { t }
+    }
+}
+
+#[stable(feature = "rust1", since = "1.0.0")]
+impl hash::Hash for TypeId {
+    #[inline]
+    fn hash<H: hash::Hasher>(&self, state: &mut H) {
+        // We only hash the lower 64 bits of our (128 bit) internal numeric ID,
+        // because:
+        // - The hashing algorithm which backs `TypeId` is expected to be
+        //   unbiased and high quality, meaning further mixing would be somewhat
+        //   redundant compared to choosing (the lower) 64 bits arbitrarily.
+        // - `Hasher::finish` returns a u64 anyway, so the extra entropy we'd
+        //   get from hashing the full value would probably not be useful
+        //   (especially given the previous point about the lower 64 bits being
+        //   high quality on their own).
+        // - It is correct to do so -- only hashing a subset of `self` is still
+        //   with an `Eq` implementation that considers the entire value, as
+        //   ours does.
+        (self.t as u64).hash(state);
     }
 }
 

library/core/src/intrinsics.rs

@@ -1057,8 +1057,25 @@ extern "rust-intrinsic" {
     #[rustc_const_unstable(feature = "const_type_id", issue = "77125")]
     #[rustc_safe_intrinsic]
     #[rustc_nounwind]
+    #[cfg(bootstrap)]
     pub fn type_id<T: ?Sized + 'static>() -> u64;
 
+    /// Gets an identifier which is globally unique to the specified type. This
+    /// function will return the same value for a type regardless of whichever
+    /// crate it is invoked in.
+    ///
+    /// Note that, unlike most intrinsics, this is safe to call;
+    /// it does not require an `unsafe` block.
+    /// Therefore, implementations must not require the user to uphold
+    /// any safety invariants.
+    ///
+    /// The stabilized version of this intrinsic is [`core::any::TypeId::of`].
+    #[rustc_const_unstable(feature = "const_type_id", issue = "77125")]
+    #[rustc_safe_intrinsic]
+    #[rustc_nounwind]
+    #[cfg(not(bootstrap))]
+    pub fn type_id<T: ?Sized + 'static>() -> u128;
+
     /// A guard for unsafe functions that cannot ever be executed if `T` is uninhabited:
     /// This will statically either panic, or do nothing.
     ///