Skip to content

Remap local def ids for deterministic metadata encoding #162809

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
rust-bors merged 1 commit into from
Oct 17, 2026
Merged

Remap local def ids for deterministic metadata encoding #162809

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 14 additions & 5 deletions compiler/rustc_hir_id/src/definitions.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,6 +98,9 @@ pub struct Definitions {
def_id_to_key: IndexVec<LocalDefId, DefKey>,
// We do only store the local hash, as all the definitions are from the current crate.
def_path_hashes: IndexVec<LocalDefId, Hash64>,
// The first non deterministic def index, may correspond to non-allocated
// def id, in this case `first_non_det_index.as_usize() == num_definitions`.
first_non_det_index: Option<DefIndex>,
Comment thread
aerooneqq marked this conversation as resolved.
def_path_hash_to_index: DefPathToIndexMap,
}

Expand Down Expand Up @@ -302,13 +305,18 @@ pub enum DefPathData {
}

impl Definitions {
/// This function indicates that the order of def id allocations
/// This function indicates that the def id allocations
/// may be non-deterministic after it was called.
pub fn commit_end_of_determinism(&mut self) {
assert!(
self.def_path_hash_to_index.after_parallel_alloc.replace(Default::default()).is_none(),
"this function should be called only once"
)
let prev = self.def_path_hash_to_index.after_parallel_alloc.replace(Default::default());
assert!(prev.is_none(), "this function should be called only once");

let def_index = DefIndex::from_usize(self.def_id_to_key.len());
self.first_non_det_index.replace(def_index);
}

pub fn first_non_det_index(&self) -> DefIndex {
self.first_non_det_index.expect("must contain index")
}

#[inline(always)]
Expand Down Expand Up @@ -362,6 +370,7 @@ impl Definitions {
def_path_hashes: Default::default(),
def_id_to_key: Default::default(),
def_path_hash_to_index: Default::default(),
first_non_det_index: Default::default(),
};

// Create the root definition.
Expand Down
7 changes: 4 additions & 3 deletions compiler/rustc_metadata/src/rmeta/decoder.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,6 +41,7 @@ use tracing::debug;

use crate::creader::CStore;
use crate::eii::EiiMapEncodedKeyValue;
use crate::rmeta::remapped_def_id::RemappedRawDefId;
use crate::rmeta::table::IsDefault;
use crate::rmeta::*;

Expand Down Expand Up @@ -103,7 +104,7 @@ pub(crate) struct CrateMetadata {
/// FIXME: make type of the value `LazyArray<(DefIndex, Option<SimplifiedType>)>`
/// through transmuting or somehow else, as `LocalDefId` here should be `DefIndex`,
/// as it is from external crate.
trait_impls: FxIndexMap<RawDefId, LazyArray<(LocalDefId, Option<SimplifiedType>)>>,
trait_impls: FxIndexMap<RemappedRawDefId, LazyArray<(LocalDefId, Option<SimplifiedType>)>>,
/// Inherent impls which do not follow the normal coherence rules.
///
/// These can be introduced using either `#![rustc_coherence_is_core]`
Expand Down Expand Up @@ -2163,10 +2164,10 @@ impl CrateMetadata {

// Translate a DefId from the current compilation environment to a DefId
// for an external crate.
fn reverse_translate_def_id(&self, did: DefId) -> Option<RawDefId> {
fn reverse_translate_def_id(&self, did: DefId) -> Option<RemappedRawDefId> {
for (local, &global) in self.cnum_map.iter_enumerated() {
if global == did.krate {
return Some(RawDefId { krate: local.as_u32(), index: did.index.as_u32() });
return Some(RemappedRawDefId::new_unchecked(local.as_u32(), did.index.as_u32()));
}
}

Expand Down
60 changes: 41 additions & 19 deletions compiler/rustc_metadata/src/rmeta/encoder.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,6 +47,7 @@ use twox_hash::XxHash3_128;

use crate::diagnostics::{FailCreateFileEncoder, FailWriteFile};
use crate::eii::EiiMapEncodedKeyValue;
use crate::rmeta::remapped_def_id::RemappedRawDefId;
use crate::rmeta::*;

pub(super) struct EncodeContext<'a, 'tcx> {
Expand Down Expand Up @@ -78,6 +79,7 @@ pub(super) struct EncodeContext<'a, 'tcx> {
hygiene_ctxt: Rc<RefCell<HygieneEncodeContext>>,
// Used for both `Symbol`s and `ByteSymbol`s.
symbol_index_table: FxHashMap<u32, usize>,
local_defs_remapping: LocalDefsRemapping,
}

/// If the current crate is a proc-macro, returns early with `LazyArray::default()`.
Expand Down Expand Up @@ -164,6 +166,8 @@ impl<'a, 'tcx> SpanEncoder for EncodeContext<'a, 'tcx> {
}

fn encode_def_id(&mut self, def_id: DefId) {
let def_id = self.local_defs_remapping.map_def_id(def_id);

def_id.krate.encode(self);
self.emit_u32(def_id.index.as_u32());
}
Expand Down Expand Up @@ -235,7 +239,7 @@ impl<'a, 'tcx> SpanEncoder for EncodeContext<'a, 'tcx> {
/// `LocalDefId` as a wrapper around def index we want to encode
/// def index only, so we override the encode implementation.
fn encode_local_def_id(&mut self, def_id: LocalDefId) {
self.emit_u32(def_id.local_def_index.as_u32());
self.emit_u32(self.local_defs_remapping.map_local_def_id(def_id).index.as_u32());
}

/// We override encoding logic for `DefKey` during metadata encoding,
Expand Down Expand Up @@ -434,20 +438,23 @@ macro_rules! record_some_lazy {
($self:ident.$tables:ident.$table:ident[$def_id:expr] <- $value:expr) => {{
let value = $value;
let lazy = $self.lazy(value);
$self.$tables.$table.set_some($def_id, lazy);
let def_id = $self.local_defs_remapping.map_local_def_id($def_id);
$self.$tables.$table.set_some(def_id, lazy);
Comment thread
oli-obk marked this conversation as resolved.
}};
}

macro_rules! record_some {
($self:ident.$tables:ident.$table:ident[$def_id:expr] <- $value:expr) => {
$self.$tables.$table.set_some($def_id, $value)
};
($self:ident.$tables:ident.$table:ident[$def_id:expr] <- $value:expr) => {{
let def_id = $self.local_defs_remapping.map_local_def_id($def_id);
$self.$tables.$table.set_some(def_id, $value)
}};
}

macro_rules! record_value {
($self:ident.$tables:ident.$table:ident[$def_id:expr] <- $value:expr) => {
$self.$tables.$table.set($def_id, $value)
};
($self:ident.$tables:ident.$table:ident[$def_id:expr] <- $value:expr) => {{
let def_id = $self.local_defs_remapping.map_local_def_id($def_id);
$self.$tables.$table.set(def_id, $value)
}};
}

// Shorthand for `$self.$tables.$table.set_some($def_id.index, $self.lazy_array($value))`, which would
Expand All @@ -456,15 +463,17 @@ macro_rules! record_array {
($self:ident.$tables:ident.$table:ident[$def_id:expr] <- $value:expr) => {{
let value = $value;
let lazy = $self.lazy_array(value);
$self.$tables.$table.set_some($def_id, lazy);
let def_id = $self.local_defs_remapping.map_local_def_id($def_id);
$self.$tables.$table.set_some(def_id, lazy);
}};
}

macro_rules! record_defaulted_array {
($self:ident.$tables:ident.$table:ident[$def_id:expr] <- $value:expr) => {{
let value = $value;
let lazy = $self.lazy_array(value);
$self.$tables.$table.set($def_id, lazy);
let def_id = $self.local_defs_remapping.map_local_def_id($def_id);
$self.$tables.$table.set(def_id, lazy);
}};
}

Expand Down Expand Up @@ -566,10 +575,12 @@ impl<'a, 'tcx> EncodeContext<'a, 'tcx> {
record_value!(self.tables.def_path_hashes[def_id] <- def_path_hash.local_hash().as_u64())
}
} else {
for (def_index, def_key, def_path_hash) in defs.enumerated_keys_and_path_hashes() {
let def_id = LocalDefId { local_def_index: def_index };
for def_id in self.local_defs_remapping.iter_sorted_local_ids() {
let def_key = defs.def_key(def_id);
let hash = defs.def_path_hash(def_id).local_hash().as_u64();

record_some_lazy!(self.tables.def_keys[def_id] <- def_key);
record_value!(self.tables.def_path_hashes[def_id] <- def_path_hash.local_hash().as_u64())
record_value!(self.tables.def_path_hashes[def_id] <- hash);
}
}
}
Expand Down Expand Up @@ -650,6 +661,11 @@ impl<'a, 'tcx> EncodeContext<'a, 'tcx> {

fn encode_crate_root(&mut self) -> (LazyValue<CrateRoot>, LazyValue<CrateRootUnhashed>) {
let tcx = self.tcx;

// We set it here as we need it only for metadata encoding,
// while `with_encode_metadata_header` is used in some other places too.
self.local_defs_remapping = LocalDefsRemapping::new(tcx);

let mut stats: Vec<(&'static str, usize)> = Vec::with_capacity(32);

macro_rules! stat {
Expand Down Expand Up @@ -1489,8 +1505,7 @@ impl<'a, 'tcx> EncodeContext<'a, 'tcx> {
}

let tcx = self.tcx;

for local_id in tcx.iter_local_def_id() {
for local_id in self.local_defs_remapping.iter_sorted_local_ids() {
let def_id = local_id.to_def_id();
let def_kind = tcx.def_kind(local_id);
record_some!(self.tables.def_kind[local_id] <- def_kind);
Expand Down Expand Up @@ -1630,13 +1645,18 @@ impl<'a, 'tcx> EncodeContext<'a, 'tcx> {
&& tcx.type_of(def_id).skip_binder().is_coroutine_closure()
{
let coroutine_for_closure = self.tcx.coroutine_for_closure(def_id);
record_some!(self.tables.coroutine_for_closure[local_id] <- coroutine_for_closure.into());
let remapped_id =
RemappedRawDefId::create(coroutine_for_closure, &self.local_defs_remapping);

record_some!(self.tables.coroutine_for_closure[local_id] <- remapped_id);

// If this async closure has a by-move body, record it too.
if tcx.needs_coroutine_by_move_body_def_id(coroutine_for_closure) {
let id = self.tcx.coroutine_by_move_body_def_id(coroutine_for_closure);
let id = RemappedRawDefId::create(id, &self.local_defs_remapping);

record_some!(
self.tables.coroutine_by_move_body_def_id[coroutine_for_closure.expect_local()] <- id.into()
self.tables.coroutine_by_move_body_def_id[coroutine_for_closure.expect_local()] <- id
)
}
}
Expand Down Expand Up @@ -2325,7 +2345,8 @@ impl<'a, 'tcx> EncodeContext<'a, 'tcx> {
if let Ok(mut an) = trait_def.ancestors(tcx, local_id.to_def_id())
&& let Some(specialization_graph::Node::Impl(parent)) = an.nth(1)
{
record_some!(self.tables.impl_parent[local_id] <- parent.into());
let parent = RemappedRawDefId::create(parent, &self.local_defs_remapping);
record_some!(self.tables.impl_parent[local_id] <- parent);
}

// if this is an impl of `CoerceUnsized`, create its
Expand All @@ -2340,7 +2361,7 @@ impl<'a, 'tcx> EncodeContext<'a, 'tcx> {
let trait_impls: Vec<_> = trait_impls
.into_iter()
.map(|(trait_def_id, impls)| TraitImpls {
trait_id: trait_def_id.into(),
trait_id: RemappedRawDefId::create(trait_def_id, &self.local_defs_remapping),
impls: self.lazy_array(&impls),
})
.collect();
Expand Down Expand Up @@ -2734,6 +2755,7 @@ fn with_encode_metadata_header(
is_proc_macro: tcx.crate_types().contains(&CrateType::ProcMacro),
hygiene_ctxt: Default::default(),
symbol_index_table: Default::default(),
local_defs_remapping: Default::default(),
};

// Encode the rustc version string in a predictable location.
Expand Down
Loading
Loading