proc_macro: use fxhash within the proc_macro crate

Unfortunately, as it is difficult to depend on crates from within proc_macro,
this is done by vendoring a copy of the hasher as a module rather than
depending on the rustc_hash crate.

This probably doesn't have a substantial impact up-front, however will be more
relevant once symbols are interned within the proc_macro client.

Author: mystor

library/proc_macro/src/bridge/fxhash.rs

@@ -0,0 +1,117 @@
+//! This is a copy of the `rustc_hash` crate, adapted to work as a module.
+//!
+//! If in the future it becomes more reasonable to add dependencies to
+//! `proc_macro`, this module should be removed and replaced with a dependency
+//! on the `rustc_hash` crate.
+
+use std::collections::HashMap;
+use std::convert::TryInto;
+use std::default::Default;
+use std::hash::BuildHasherDefault;
+use std::hash::Hasher;
+use std::mem::size_of;
+use std::ops::BitXor;
+
+/// Type alias for a hashmap using the `fx` hash algorithm.
+pub type FxHashMap<K, V> = HashMap<K, V, BuildHasherDefault<FxHasher>>;
+
+/// A speedy hash algorithm for use within rustc. The hashmap in liballoc
+/// by default uses SipHash which isn't quite as speedy as we want. In the
+/// compiler we're not really worried about DOS attempts, so we use a fast
+/// non-cryptographic hash.
+///
+/// This is the same as the algorithm used by Firefox -- which is a homespun
+/// one not based on any widely-known algorithm -- though modified to produce
+/// 64-bit hash values instead of 32-bit hash values. It consistently
+/// out-performs an FNV-based hash within rustc itself -- the collision rate is
+/// similar or slightly worse than FNV, but the speed of the hash function
+/// itself is much higher because it works on up to 8 bytes at a time.
+pub struct FxHasher {
+    hash: usize,
+}
+
+#[cfg(target_pointer_width = "32")]
+const K: usize = 0x9e3779b9;
+#[cfg(target_pointer_width = "64")]
+const K: usize = 0x517cc1b727220a95;
+
+impl Default for FxHasher {
+    #[inline]
+    fn default() -> FxHasher {
+        FxHasher { hash: 0 }
+    }
+}
+
+impl FxHasher {
+    #[inline]
+    fn add_to_hash(&mut self, i: usize) {
+        self.hash = self.hash.rotate_left(5).bitxor(i).wrapping_mul(K);
+    }
+}
+
+impl Hasher for FxHasher {
+    #[inline]
+    fn write(&mut self, mut bytes: &[u8]) {
+        #[cfg(target_pointer_width = "32")]
+        let read_usize = |bytes: &[u8]| u32::from_ne_bytes(bytes[..4].try_into().unwrap());
+        #[cfg(target_pointer_width = "64")]
+        let read_usize = |bytes: &[u8]| u64::from_ne_bytes(bytes[..8].try_into().unwrap());
+
+        let mut hash = FxHasher { hash: self.hash };
+        assert!(size_of::<usize>() <= 8);
+        while bytes.len() >= size_of::<usize>() {
+            hash.add_to_hash(read_usize(bytes) as usize);
+            bytes = &bytes[size_of::<usize>()..];
+        }
+        if (size_of::<usize>() > 4) && (bytes.len() >= 4) {
+            hash.add_to_hash(u32::from_ne_bytes(bytes[..4].try_into().unwrap()) as usize);
+            bytes = &bytes[4..];
+        }
+        if (size_of::<usize>() > 2) && bytes.len() >= 2 {
+            hash.add_to_hash(u16::from_ne_bytes(bytes[..2].try_into().unwrap()) as usize);
+            bytes = &bytes[2..];
+        }
+        if (size_of::<usize>() > 1) && bytes.len() >= 1 {
+            hash.add_to_hash(bytes[0] as usize);
+        }
+        self.hash = hash.hash;
+    }
+
+    #[inline]
+    fn write_u8(&mut self, i: u8) {
+        self.add_to_hash(i as usize);
+    }
+
+    #[inline]
+    fn write_u16(&mut self, i: u16) {
+        self.add_to_hash(i as usize);
+    }
+
+    #[inline]
+    fn write_u32(&mut self, i: u32) {
+        self.add_to_hash(i as usize);
+    }
+
+    #[cfg(target_pointer_width = "32")]
+    #[inline]
+    fn write_u64(&mut self, i: u64) {
+        self.add_to_hash(i as usize);
+        self.add_to_hash((i >> 32) as usize);
+    }
+
+    #[cfg(target_pointer_width = "64")]
+    #[inline]
+    fn write_u64(&mut self, i: u64) {
+        self.add_to_hash(i as usize);
+    }
+
+    #[inline]
+    fn write_usize(&mut self, i: usize) {
+        self.add_to_hash(i);
+    }
+
+    #[inline]
+    fn finish(&self) -> u64 {
+        self.hash as u64
+    }
+}

library/proc_macro/src/bridge/handle.rs

@@ -1,11 +1,13 @@
 //! Server-side handles and storage for per-handle data.
 
-use std::collections::{BTreeMap, HashMap};
-use std::hash::{BuildHasher, Hash};
+use std::collections::BTreeMap;
+use std::hash::Hash;
 use std::num::NonZeroU32;
 use std::ops::{Index, IndexMut};
 use std::sync::atomic::{AtomicUsize, Ordering};
 
+use super::fxhash::FxHashMap;
+
 pub(super) type Handle = NonZeroU32;
 
 /// A store that associates values of type `T` with numeric handles. A value can
@@ -51,31 +53,15 @@ impl<T> IndexMut<Handle> for OwnedStore<T> {
     }
 }
 
-// HACK(eddyb) deterministic `std::collections::hash_map::RandomState` replacement
-// that doesn't require adding any dependencies to `proc_macro` (like `rustc-hash`).
-#[derive(Clone)]
-struct NonRandomState;
-
-impl BuildHasher for NonRandomState {
-    type Hasher = std::collections::hash_map::DefaultHasher;
-    #[inline]
-    fn build_hasher(&self) -> Self::Hasher {
-        Self::Hasher::new()
-    }
-}
-
 /// Like `OwnedStore`, but avoids storing any value more than once.
 pub(super) struct InternedStore<T: 'static> {
     owned: OwnedStore<T>,
-    interner: HashMap<T, Handle, NonRandomState>,
+    interner: FxHashMap<T, Handle>,
 }
 
 impl<T: Copy + Eq + Hash> InternedStore<T> {
     pub(super) fn new(counter: &'static AtomicUsize) -> Self {
-        InternedStore {
-            owned: OwnedStore::new(counter),
-            interner: HashMap::with_hasher(NonRandomState),
-        }
+        InternedStore { owned: OwnedStore::new(counter), interner: FxHashMap::default() }
     }
 
     pub(super) fn alloc(&mut self, x: T) -> Handle {

library/proc_macro/src/bridge/mod.rs

@@ -177,6 +177,8 @@ pub mod client;
 #[allow(unsafe_code)]
 mod closure;
 #[forbid(unsafe_code)]
+mod fxhash;
+#[forbid(unsafe_code)]
 mod handle;
 #[macro_use]
 #[forbid(unsafe_code)]