Make DefIdForest cheaper to clone

Since `DefIdForest` contains 0 or 1 elements the large majority of the
time, by allocating only in the >1 case we avoid almost all allocations,
compared to `Arc<SmallVec<[DefId;1]>>`. This shaves off 0.2% on the
benchmark that stresses uninhabitedness checking.
This commit is contained in:
Nadrieril 2020-12-03 01:52:24 +00:00
parent 8598c9f6e5
commit e608d8f4e5
3 changed files with 80 additions and 58 deletions

View file

@ -1314,7 +1314,7 @@ rustc_queries! {
/// check whether the forest is empty.
query type_uninhabited_from(
key: ty::ParamEnvAnd<'tcx, Ty<'tcx>>
) -> Arc<ty::inhabitedness::DefIdForest> {
) -> ty::inhabitedness::DefIdForest {
desc { "computing the inhabitedness of `{:?}`", key }
}
}

View file

@ -3,6 +3,9 @@ use crate::ty::{DefId, DefIdTree};
use rustc_hir::CRATE_HIR_ID;
use smallvec::SmallVec;
use std::mem;
use std::sync::Arc;
use DefIdForest::*;
/// Represents a forest of `DefId`s closed under the ancestor relation. That is,
/// if a `DefId` representing a module is contained in the forest then all
@ -11,45 +14,77 @@ use std::mem;
///
/// This is used to represent a set of modules in which a type is visibly
/// uninhabited.
///
/// We store the minimal set of `DefId`s required to represent the whole set. If A and B are
/// `DefId`s in the `DefIdForest`, and A is a parent of B, then only A will be stored. When this is
/// used with `type_uninhabited_from`, there will very rarely be more than one `DefId` stored.
#[derive(Clone, HashStable)]
pub struct DefIdForest {
/// The minimal set of `DefId`s required to represent the whole set.
/// If A and B are DefIds in the `DefIdForest`, and A is a descendant
/// of B, then only B will be in `root_ids`.
/// We use a `SmallVec` here because (for its use for caching inhabitedness)
/// it's rare that this will contain even two IDs.
root_ids: SmallVec<[DefId; 1]>,
pub enum DefIdForest {
Empty,
Single(DefId),
/// This variant is very rare.
/// Invariant: >1 elements
/// We use `Arc` because this is used in the output of a query.
Multiple(Arc<[DefId]>),
}
/// Tests whether a slice of roots contains a given DefId.
#[inline]
fn slice_contains(tcx: TyCtxt<'tcx>, slice: &[DefId], id: DefId) -> bool {
slice.iter().any(|root_id| tcx.is_descendant_of(id, *root_id))
}
impl<'tcx> DefIdForest {
/// Creates an empty forest.
pub fn empty() -> DefIdForest {
DefIdForest { root_ids: SmallVec::new() }
DefIdForest::Empty
}
/// Creates a forest consisting of a single tree representing the entire
/// crate.
#[inline]
pub fn full(tcx: TyCtxt<'tcx>) -> DefIdForest {
let crate_id = tcx.hir().local_def_id(CRATE_HIR_ID);
DefIdForest::from_id(crate_id.to_def_id())
DefIdForest::from_id(tcx.hir().local_def_id(CRATE_HIR_ID).to_def_id())
}
/// Creates a forest containing a `DefId` and all its descendants.
pub fn from_id(id: DefId) -> DefIdForest {
let mut root_ids = SmallVec::new();
root_ids.push(id);
DefIdForest { root_ids }
DefIdForest::Single(id)
}
fn as_slice(&self) -> &[DefId] {
match self {
Empty => &[],
Single(id) => std::slice::from_ref(id),
Multiple(root_ids) => root_ids,
}
}
// Only allocates in the rare `Multiple` case.
fn from_slice(root_ids: &[DefId]) -> DefIdForest {
match root_ids {
[] => Empty,
[id] => Single(*id),
_ => DefIdForest::Multiple(root_ids.into()),
}
}
/// Tests whether the forest is empty.
pub fn is_empty(&self) -> bool {
self.root_ids.is_empty()
match self {
Empty => true,
Single(..) | Multiple(..) => false,
}
}
/// Iterate over the set of roots.
fn iter(&self) -> impl Iterator<Item = DefId> + '_ {
self.as_slice().iter().copied()
}
/// Tests whether the forest contains a given DefId.
pub fn contains(&self, tcx: TyCtxt<'tcx>, id: DefId) -> bool {
self.root_ids.iter().any(|root_id| tcx.is_descendant_of(id, *root_id))
slice_contains(tcx, self.as_slice(), id)
}
/// Calculate the intersection of a collection of forests.
@ -58,44 +93,28 @@ impl<'tcx> DefIdForest {
I: IntoIterator<Item = DefIdForest>,
{
let mut iter = iter.into_iter();
let mut ret = if let Some(first) = iter.next() {
first
let mut ret: SmallVec<[_; 1]> = if let Some(first) = iter.next() {
SmallVec::from_slice(first.as_slice())
} else {
return DefIdForest::full(tcx);
};
let mut next_ret = SmallVec::new();
let mut old_ret: SmallVec<[DefId; 1]> = SmallVec::new();
let mut next_ret: SmallVec<[_; 1]> = SmallVec::new();
for next_forest in iter {
// No need to continue if the intersection is already empty.
if ret.is_empty() {
break;
if ret.is_empty() || next_forest.is_empty() {
return DefIdForest::empty();
}
// `next_ret` and `old_ret` are empty here.
// We keep the elements in `ret` that are also in `next_forest`. Those that aren't are
// put back in `ret` via `old_ret`.
for id in ret.root_ids.drain(..) {
if next_forest.contains(tcx, id) {
next_ret.push(id);
} else {
old_ret.push(id);
}
}
ret.root_ids.extend(old_ret.drain(..));
// We keep the elements in `ret` that are also in `next_forest`.
next_ret.extend(ret.iter().copied().filter(|&id| next_forest.contains(tcx, id)));
// We keep the elements in `next_forest` that are also in `ret`.
// You'd think this is not needed because `next_ret` already contains `ret \inter
// next_forest`. But those aren't just sets of things. If `ret = [a]`, `next_forest =
// [b]` and `b` is a submodule of `a`, then `b` belongs in the intersection but we
// didn't catch it in the loop above.
next_ret.extend(next_forest.root_ids.into_iter().filter(|&id| ret.contains(tcx, id)));
// `next_ret` now contains the intersection of the original `ret` and `next_forest`.
next_ret.extend(next_forest.iter().filter(|&id| slice_contains(tcx, &ret, id)));
mem::swap(&mut next_ret, &mut ret.root_ids);
next_ret.drain(..);
mem::swap(&mut next_ret, &mut ret);
next_ret.clear();
}
ret
DefIdForest::from_slice(&ret)
}
/// Calculate the union of a collection of forests.
@ -103,20 +122,26 @@ impl<'tcx> DefIdForest {
where
I: IntoIterator<Item = DefIdForest>,
{
let mut ret = DefIdForest::empty();
let mut next_ret = SmallVec::new();
let mut ret: SmallVec<[_; 1]> = SmallVec::new();
let mut next_ret: SmallVec<[_; 1]> = SmallVec::new();
for next_forest in iter {
next_ret.extend(ret.root_ids.drain(..).filter(|&id| !next_forest.contains(tcx, id)));
// Union with the empty set is a no-op.
if next_forest.is_empty() {
continue;
}
for id in next_forest.root_ids {
if !next_ret.contains(&id) {
// We add everything in `ret` that is not in `next_forest`.
next_ret.extend(ret.iter().copied().filter(|&id| !next_forest.contains(tcx, id)));
// We add everything in `next_forest` that we haven't added yet.
for id in next_forest.iter() {
if !slice_contains(tcx, &next_ret, id) {
next_ret.push(id);
}
}
mem::swap(&mut next_ret, &mut ret.root_ids);
next_ret.drain(..);
mem::swap(&mut next_ret, &mut ret);
next_ret.clear();
}
ret
DefIdForest::from_slice(&ret)
}
}

View file

@ -7,8 +7,6 @@ use crate::ty::{AdtDef, FieldDef, Ty, TyS, VariantDef};
use crate::ty::{AdtKind, Visibility};
use crate::ty::{DefId, SubstsRef};
use std::sync::Arc;
mod def_id_forest;
// The methods in this module calculate `DefIdForest`s of modules in which a
@ -193,7 +191,7 @@ impl<'tcx> TyS<'tcx> {
tcx: TyCtxt<'tcx>,
param_env: ty::ParamEnv<'tcx>,
) -> DefIdForest {
tcx.type_uninhabited_from(param_env.and(self)).as_ref().clone()
tcx.type_uninhabited_from(param_env.and(self))
}
}
@ -201,10 +199,10 @@ impl<'tcx> TyS<'tcx> {
pub(crate) fn type_uninhabited_from<'tcx>(
tcx: TyCtxt<'tcx>,
key: ty::ParamEnvAnd<'tcx, Ty<'tcx>>,
) -> Arc<DefIdForest> {
) -> DefIdForest {
let ty = key.value;
let param_env = key.param_env;
let forest = match *ty.kind() {
match *ty.kind() {
Adt(def, substs) => def.uninhabited_from(tcx, substs, param_env),
Never => DefIdForest::full(tcx),
@ -229,6 +227,5 @@ pub(crate) fn type_uninhabited_from<'tcx>(
Ref(..) => DefIdForest::empty(),
_ => DefIdForest::empty(),
};
Arc::new(forest)
}
}