Auto merge of #109611 - Zoxc:query-engine-rem, r=cjgillot

Remove `QueryEngine` trait

This removes the `QueryEngine` trait and `Queries` from `rustc_query_impl` and replaced them with function pointers and fields in `QuerySystem`. As a side effect `OnDiskCache` is moved back into `rustc_middle` and the `OnDiskCache` trait is also removed.

This has a couple of benefits.
- `TyCtxt` is used in the query system instead of the removed `QueryCtxt` which is larger.
- Function pointers are more flexible to work with. A variant of https://github.com/rust-lang/rust/pull/107802 is included which avoids the double indirection. For https://github.com/rust-lang/rust/pull/108938 we can name entry point `__rust_end_short_backtrace` to avoid some overhead. For https://github.com/rust-lang/rust/pull/108062 it avoids the duplicate `QueryEngine` structs.
- `QueryContext` now implements `DepContext` which avoids many `dep_context()` calls in `rustc_query_system`.
- The `rustc_driver` size is reduced by 0.33%, hopefully that means some bootstrap improvements.
- This avoids the unsafe code around the `QueryEngine` trait.

r? `@cjgillot`

compiler/rustc_middle/Cargo.toml

@@ -11,6 +11,7 @@ chalk-ir = "0.87.0"
 derive_more = "0.99.17"
 either = "1.5.0"
 gsgdt = "0.1.2"
+measureme = "10.0.0"
 polonius-engine = "0.13.0"
 rustc_apfloat = { path = "../rustc_apfloat" }
 rustc_arena = { path = "../rustc_arena" }

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

@@ -227,7 +227,9 @@ pub fn specialized_encode_alloc_id<'tcx, E: TyEncoder<I = TyCtxt<'tcx>>>(
             // References to statics doesn't need to know about their allocations,
             // just about its `DefId`.
             AllocDiscriminant::Static.encode(encoder);
-            did.encode(encoder);
+            // Cannot use `did.encode(encoder)` because of a bug around
+            // specializations and method calls.
+            Encodable::<E>::encode(&did, encoder);
         }
     }
 }

compiler/rustc_middle/src/query/mod.rs

@@ -9,6 +9,7 @@ use rustc_span::def_id::LOCAL_CRATE;
 
 pub mod erase;
 mod keys;
+pub mod on_disk_cache;
 pub use keys::{AsLocalKey, Key, LocalCrate};
 
 // Each of these queries corresponds to a function pointer field in the

compiler/rustc_middle/src/ty/codec.rs

@@ -500,7 +500,6 @@ impl_arena_copy_decoder! {<'tcx>
 macro_rules! implement_ty_decoder {
     ($DecoderName:ident <$($typaram:tt),*>) => {
         mod __ty_decoder_impl {
-            use std::borrow::Cow;
             use rustc_serialize::Decoder;
 
             use super::$DecoderName;

compiler/rustc_middle/src/ty/context.rs

@@ -14,11 +14,14 @@ use crate::middle::resolve_bound_vars;
 use crate::middle::stability;
 use crate::mir::interpret::{self, Allocation, ConstAllocation};
 use crate::mir::{Body, Local, Place, PlaceElem, ProjectionKind, Promoted};
+use crate::query::on_disk_cache::OnDiskCache;
 use crate::query::LocalCrate;
 use crate::thir::Thir;
 use crate::traits;
 use crate::traits::solve;
 use crate::traits::solve::{ExternalConstraints, ExternalConstraintsData};
+use crate::ty::query::QuerySystem;
+use crate::ty::query::QuerySystemFns;
 use crate::ty::query::{self, TyCtxtAt};
 use crate::ty::{
     self, AdtDef, AdtDefData, AdtKind, Binder, Const, ConstData, FloatTy, FloatVar, FloatVid,
@@ -31,7 +34,6 @@ use rustc_ast::{self as ast, attr};
 use rustc_data_structures::fingerprint::Fingerprint;
 use rustc_data_structures::fx::{FxHashMap, FxHashSet};
 use rustc_data_structures::intern::Interned;
-use rustc_data_structures::memmap::Mmap;
 use rustc_data_structures::profiling::SelfProfilerRef;
 use rustc_data_structures::sharded::{IntoPointer, ShardedHashMap};
 use rustc_data_structures::stable_hasher::{HashStable, StableHasher};
@@ -61,7 +63,6 @@ use rustc_session::lint::Lint;
 use rustc_session::Limit;
 use rustc_session::Session;
 use rustc_span::def_id::{DefPathHash, StableCrateId};
-use rustc_span::source_map::SourceMap;
 use rustc_span::symbol::{kw, sym, Ident, Symbol};
 use rustc_span::{Span, DUMMY_SP};
 use rustc_target::abi::{FieldIdx, Layout, LayoutS, TargetDataLayout, VariantIdx};
@@ -84,21 +85,6 @@ use super::query::IntoQueryParam;
 
 const TINY_CONST_EVAL_LIMIT: Limit = Limit(20);
 
-pub trait OnDiskCache<'tcx>: rustc_data_structures::sync::Sync {
-    /// Creates a new `OnDiskCache` instance from the serialized data in `data`.
-    fn new(sess: &'tcx Session, data: Mmap, start_pos: usize) -> Self
-    where
-        Self: Sized;
-
-    fn new_empty(source_map: &'tcx SourceMap) -> Self
-    where
-        Self: Sized;
-
-    fn drop_serialized_data(&self, tcx: TyCtxt<'tcx>);
-
-    fn serialize(&self, tcx: TyCtxt<'tcx>, encoder: FileEncoder) -> FileEncodeResult;
-}
-
 #[allow(rustc::usage_of_ty_tykind)]
 impl<'tcx> Interner for TyCtxt<'tcx> {
     type AdtDef = ty::AdtDef<'tcx>;
@@ -527,13 +513,6 @@ pub struct GlobalCtxt<'tcx> {
 
     untracked: Untracked,
 
-    /// This provides access to the incremental compilation on-disk cache for query results.
-    /// Do not access this directly. It is only meant to be used by
-    /// `DepGraph::try_mark_green()` and the query infrastructure.
-    /// This is `None` if we are not incremental compilation mode
-    pub on_disk_cache: Option<&'tcx dyn OnDiskCache<'tcx>>,
-
-    pub queries: &'tcx dyn query::QueryEngine<'tcx>,
     pub query_system: query::QuerySystem<'tcx>,
     pub(crate) query_kinds: &'tcx [DepKindStruct<'tcx>],
 
@@ -674,9 +653,9 @@ impl<'tcx> TyCtxt<'tcx> {
         hir_arena: &'tcx WorkerLocal<hir::Arena<'tcx>>,
         untracked: Untracked,
         dep_graph: DepGraph,
-        on_disk_cache: Option<&'tcx dyn OnDiskCache<'tcx>>,
-        queries: &'tcx dyn query::QueryEngine<'tcx>,
+        on_disk_cache: Option<OnDiskCache<'tcx>>,
         query_kinds: &'tcx [DepKindStruct<'tcx>],
+        query_system_fns: QuerySystemFns<'tcx>,
     ) -> GlobalCtxt<'tcx> {
         let data_layout = s.target.parse_data_layout().unwrap_or_else(|err| {
             s.emit_fatal(err);
@@ -698,9 +677,7 @@ impl<'tcx> TyCtxt<'tcx> {
             lifetimes: common_lifetimes,
             consts: common_consts,
             untracked,
-            on_disk_cache,
-            queries,
-            query_system: Default::default(),
+            query_system: QuerySystem::new(query_system_fns, on_disk_cache),
             query_kinds,
             ty_rcache: Default::default(),
             pred_rcache: Default::default(),
@@ -1039,7 +1016,7 @@ impl<'tcx> TyCtxt<'tcx> {
     }
 
     pub fn serialize_query_result_cache(self, encoder: FileEncoder) -> FileEncodeResult {
-        self.on_disk_cache.as_ref().map_or(Ok(0), |c| c.serialize(self, encoder))
+        self.query_system.on_disk_cache.as_ref().map_or(Ok(0), |c| c.serialize(self, encoder))
     }
 
     /// If `true`, we should use lazy normalization for constants, otherwise

compiler/rustc_middle/src/ty/mod.rs

@@ -84,8 +84,7 @@ pub use self::consts::{
     Const, ConstData, ConstInt, ConstKind, Expr, InferConst, ScalarInt, UnevaluatedConst, ValTree,
 };
 pub use self::context::{
-    tls, CtxtInterners, DeducedParamAttrs, FreeRegionInfo, GlobalCtxt, Lift, OnDiskCache, TyCtxt,
-    TyCtxtFeed,
+    tls, CtxtInterners, DeducedParamAttrs, FreeRegionInfo, GlobalCtxt, Lift, TyCtxt, TyCtxtFeed,
 };
 pub use self::instance::{Instance, InstanceDef, ShortInstance, UnusedGenericParams};
 pub use self::list::List;

compiler/rustc_middle/src/ty/query.rs

@@ -1,6 +1,7 @@
 #![allow(unused_parens)]
 
 use crate::dep_graph;
+use crate::dep_graph::DepKind;
 use crate::infer::canonical::{self, Canonical};
 use crate::lint::LintExpectation;
 use crate::metadata::ModChild;
@@ -17,7 +18,11 @@ use crate::mir::interpret::{
 };
 use crate::mir::interpret::{LitToConstError, LitToConstInput};
 use crate::mir::mono::CodegenUnit;
+
 use crate::query::erase::{erase, restore, Erase};
+use crate::query::on_disk_cache::CacheEncoder;
+use crate::query::on_disk_cache::EncodedDepNodeIndex;
+use crate::query::on_disk_cache::OnDiskCache;
 use crate::query::{AsLocalKey, Key};
 use crate::thir;
 use crate::traits::query::{
@@ -38,13 +43,16 @@ use crate::ty::subst::{GenericArg, SubstsRef};
 use crate::ty::util::AlwaysRequiresDrop;
 use crate::ty::GeneratorDiagnosticData;
 use crate::ty::{self, CrateInherentImpls, ParamEnvAnd, Ty, TyCtxt, UnusedGenericParams};
+use measureme::StringId;
 use rustc_arena::TypedArena;
 use rustc_ast as ast;
 use rustc_ast::expand::allocator::AllocatorKind;
 use rustc_attr as attr;
+use rustc_data_structures::fingerprint::Fingerprint;
 use rustc_data_structures::fx::{FxHashMap, FxIndexMap, FxIndexSet};
 use rustc_data_structures::steal::Steal;
 use rustc_data_structures::svh::Svh;
+use rustc_data_structures::sync::AtomicU64;
 use rustc_data_structures::sync::Lrc;
 use rustc_data_structures::sync::WorkerLocal;
 use rustc_data_structures::unord::UnordSet;
@@ -58,6 +66,7 @@ use rustc_hir::hir_id::OwnerId;
 use rustc_hir::lang_items::{LangItem, LanguageItems};
 use rustc_hir::{Crate, ItemLocalId, TraitCandidate};
 use rustc_index::IndexVec;
+use rustc_query_system::ich::StableHashingContext;
 pub(crate) use rustc_query_system::query::QueryJobId;
 use rustc_query_system::query::*;
 use rustc_session::config::{EntryFnType, OptLevel, OutputFilenames, SymbolManglingVersion};
@@ -76,17 +85,70 @@ use std::ops::Deref;
 use std::path::PathBuf;
 use std::sync::Arc;
 
-use rustc_data_structures::fingerprint::Fingerprint;
-use rustc_query_system::ich::StableHashingContext;
+pub struct QueryKeyStringCache {
+    pub def_id_cache: FxHashMap<DefId, StringId>,
+}
+
+impl QueryKeyStringCache {
+    pub fn new() -> QueryKeyStringCache {
+        QueryKeyStringCache { def_id_cache: Default::default() }
+    }
+}
+
+#[derive(Clone, Copy)]
+pub struct QueryStruct<'tcx> {
+    pub try_collect_active_jobs: fn(TyCtxt<'tcx>, &mut QueryMap<DepKind>) -> Option<()>,
+    pub alloc_self_profile_query_strings: fn(TyCtxt<'tcx>, &mut QueryKeyStringCache),
+    pub encode_query_results:
+        Option<fn(TyCtxt<'tcx>, &mut CacheEncoder<'_, 'tcx>, &mut EncodedDepNodeIndex)>,
+}
+
+pub struct QuerySystemFns<'tcx> {
+    pub engine: QueryEngine,
+    pub local_providers: Providers,
+    pub extern_providers: ExternProviders,
+    pub query_structs: Vec<QueryStruct<'tcx>>,
+    pub encode_query_results: fn(
+        tcx: TyCtxt<'tcx>,
+        encoder: &mut CacheEncoder<'_, 'tcx>,
+        query_result_index: &mut EncodedDepNodeIndex,
+    ),
+    pub try_mark_green: fn(tcx: TyCtxt<'tcx>, dep_node: &dep_graph::DepNode) -> bool,
+}
 
-#[derive(Default)]
 pub struct QuerySystem<'tcx> {
+    pub states: QueryStates<'tcx>,
     pub arenas: QueryArenas<'tcx>,
     pub caches: QueryCaches<'tcx>,
+
+    /// This provides access to the incremental compilation on-disk cache for query results.
+    /// Do not access this directly. It is only meant to be used by
+    /// `DepGraph::try_mark_green()` and the query infrastructure.
+    /// This is `None` if we are not incremental compilation mode
+    pub on_disk_cache: Option<OnDiskCache<'tcx>>,
+
+    pub fns: QuerySystemFns<'tcx>,
+
+    pub jobs: AtomicU64,
+
     // Since we erase query value types we tell the typesystem about them with `PhantomData`.
     _phantom_values: QueryPhantomValues<'tcx>,
 }
 
+impl<'tcx> QuerySystem<'tcx> {
+    pub fn new(fns: QuerySystemFns<'tcx>, on_disk_cache: Option<OnDiskCache<'tcx>>) -> Self {
+        QuerySystem {
+            states: Default::default(),
+            arenas: Default::default(),
+            caches: Default::default(),
+            on_disk_cache,
+            fns,
+            jobs: AtomicU64::new(1),
+            _phantom_values: Default::default(),
+        }
+    }
+}
+
 #[derive(Copy, Clone)]
 pub struct TyCtxtAt<'tcx> {
     pub tcx: TyCtxt<'tcx>,
@@ -136,7 +198,41 @@ impl<'tcx> TyCtxt<'tcx> {
     }
 
     pub fn try_mark_green(self, dep_node: &dep_graph::DepNode) -> bool {
-        self.queries.try_mark_green(self, dep_node)
+        (self.query_system.fns.try_mark_green)(self, dep_node)
+    }
+}
+
+#[inline]
+fn query_get_at<'tcx, Cache>(
+    tcx: TyCtxt<'tcx>,
+    execute_query: fn(TyCtxt<'tcx>, Span, Cache::Key, QueryMode) -> Option<Cache::Value>,
+    query_cache: &Cache,
+    span: Span,
+    key: Cache::Key,
+) -> Cache::Value
+where
+    Cache: QueryCache,
+{
+    let key = key.into_query_param();
+    match try_get_cached(tcx, query_cache, &key) {
+        Some(value) => value,
+        None => execute_query(tcx, span, key, QueryMode::Get).unwrap(),
+    }
+}
+
+#[inline]
+fn query_ensure<'tcx, Cache>(
+    tcx: TyCtxt<'tcx>,
+    execute_query: fn(TyCtxt<'tcx>, Span, Cache::Key, QueryMode) -> Option<Cache::Value>,
+    query_cache: &Cache,
+    key: Cache::Key,
+    check_cache: bool,
+) where
+    Cache: QueryCache,
+{
+    let key = key.into_query_param();
+    if try_get_cached(tcx, query_cache, &key).is_none() {
+        execute_query(tcx, DUMMY_SP, key, QueryMode::Ensure { check_cache });
     }
 }
 
@@ -345,17 +441,13 @@ macro_rules! define_callbacks {
             $($(#[$attr])*
             #[inline(always)]
             pub fn $name(self, key: query_helper_param_ty!($($K)*)) {
-                let key = key.into_query_param();
-
-                match try_get_cached(self.tcx, &self.tcx.query_system.caches.$name, &key) {
-                    Some(_) => return,
-                    None => self.tcx.queries.$name(
-                        self.tcx,
-                        DUMMY_SP,
-                        key,
-                        QueryMode::Ensure { check_cache: false },
-                    ),
-                };
+                query_ensure(
+                    self.tcx,
+                    self.tcx.query_system.fns.engine.$name,
+                    &self.tcx.query_system.caches.$name,
+                    key.into_query_param(),
+                    false,
+                );
             })*
         }
 
@@ -363,17 +455,13 @@ macro_rules! define_callbacks {
             $($(#[$attr])*
             #[inline(always)]
             pub fn $name(self, key: query_helper_param_ty!($($K)*)) {
-                let key = key.into_query_param();
-
-                match try_get_cached(self.tcx, &self.tcx.query_system.caches.$name, &key) {
-                    Some(_) => return,
-                    None => self.tcx.queries.$name(
-                        self.tcx,
-                        DUMMY_SP,
-                        key,
-                        QueryMode::Ensure { check_cache: true },
-                    ),
-                };
+                query_ensure(
+                    self.tcx,
+                    self.tcx.query_system.fns.engine.$name,
+                    &self.tcx.query_system.caches.$name,
+                    key.into_query_param(),
+                    true,
+                );
             })*
         }
 
@@ -392,15 +480,23 @@ macro_rules! define_callbacks {
             #[inline(always)]
             pub fn $name(self, key: query_helper_param_ty!($($K)*)) -> $V
             {
-                let key = key.into_query_param();
-
-                restore::<$V>(match try_get_cached(self.tcx, &self.tcx.query_system.caches.$name, &key) {
-                    Some(value) => value,
-                    None => self.tcx.queries.$name(self.tcx, self.span, key, QueryMode::Get).unwrap(),
-                })
+                restore::<$V>(query_get_at(
+                    self.tcx,
+                    self.tcx.query_system.fns.engine.$name,
+                    &self.tcx.query_system.caches.$name,
+                    self.span,
+                    key.into_query_param(),
+                ))
             })*
         }
 
+        #[derive(Default)]
+        pub struct QueryStates<'tcx> {
+            $(
+                pub $name: QueryState<$($K)*, DepKind>,
+            )*
+        }
+
         pub struct Providers {
             $(pub $name: for<'tcx> fn(
                 TyCtxt<'tcx>,
@@ -446,19 +542,13 @@ macro_rules! define_callbacks {
             fn clone(&self) -> Self { *self }
         }
 
-        pub trait QueryEngine<'tcx>: rustc_data_structures::sync::Sync {
-            fn as_any(&'tcx self) -> &'tcx dyn std::any::Any;
-
-            fn try_mark_green(&'tcx self, tcx: TyCtxt<'tcx>, dep_node: &dep_graph::DepNode) -> bool;
-
-            $($(#[$attr])*
-            fn $name(
-                &'tcx self,
-                tcx: TyCtxt<'tcx>,
-                span: Span,
-                key: query_keys::$name<'tcx>,
-                mode: QueryMode,
-            ) -> Option<Erase<$V>>;)*
+        pub struct QueryEngine {
+            $(pub $name: for<'tcx> fn(
+                TyCtxt<'tcx>,
+                Span,
+                query_keys::$name<'tcx>,
+                QueryMode,
+            ) -> Option<Erase<$V>>,)*
         }
     };
 }