Computing crate_hash from metadata encoding instead of HIR (implements #94878) (very draft)

compiler/rustc_ast_lowering/src/lib.rs

@@ -45,7 +45,6 @@ use rustc_attr_parsing::{AttributeParser, OmitDoc, Recovery, ShouldEmit};
 use rustc_data_structures::fingerprint::Fingerprint;
 use rustc_data_structures::fx::FxIndexSet;
 use rustc_data_structures::sorted_map::SortedMap;
-use rustc_data_structures::stable_hasher::{StableHash, StableHasher};
 use rustc_data_structures::steal::Steal;
 use rustc_data_structures::tagged_ptr::TaggedRef;
 use rustc_errors::{DiagArgFromDisplay, DiagCtxtHandle};
@@ -506,25 +505,15 @@ fn index_crate<'a, 'b>(
 
 /// Compute the hash for the HIR of the full crate.
 /// This hash will then be part of the crate_hash which is stored in the metadata.
-fn compute_hir_hash(
-    tcx: TyCtxt<'_>,
-    owners: &IndexSlice<LocalDefId, hir::MaybeOwner<'_>>,
-) -> Fingerprint {
-    let mut hir_body_nodes: Vec<_> = owners
+fn compute_hir_hash(owners: &IndexSlice<LocalDefId, hir::MaybeOwner<'_>>) -> Fingerprint {
+    owners
         .iter_enumerated()
-        .filter_map(|(def_id, info)| {
+        .filter_map(|(_, info)| {
             let info = info.as_owner()?;
-            let def_path_hash = tcx.hir_def_path_hash(def_id);
-            Some((def_path_hash, info))
+            Some(info.fingerprint())
         })
-        .collect();
-    hir_body_nodes.sort_unstable_by_key(|bn| bn.0);
-
-    tcx.with_stable_hashing_context(|mut hcx| {
-        let mut stable_hasher = StableHasher::new();
-        hir_body_nodes.stable_hash(&mut hcx, &mut stable_hasher);
-        stable_hasher.finish()
-    })
+        .reduce(Fingerprint::combine_commutative)
+        .expect("HIR hash requested without any content")
 }
 
 pub fn lower_to_hir(tcx: TyCtxt<'_>, (): ()) -> mid_hir::Crate<'_> {
@@ -561,8 +550,7 @@ pub fn lower_to_hir(tcx: TyCtxt<'_>, (): ()) -> mid_hir::Crate<'_> {
     }
 
     // Don't hash unless necessary, because it's expensive.
-    let opt_hir_hash =
-        if tcx.needs_hir_hash() { Some(compute_hir_hash(tcx, &owners)) } else { None };
+    let opt_hir_hash = if tcx.needs_hir_hash() { Some(compute_hir_hash(&owners)) } else { None };
 
     let delayed_resolver = Steal::new((resolver, krate));
     mid_hir::Crate::new(owners, delayed_ids, delayed_resolver, opt_hir_hash)

compiler/rustc_driver_impl/src/lib.rs

@@ -324,10 +324,6 @@ pub fn run_compiler(at_args: &[String], callbacks: &mut (dyn Callbacks + Send))
 
             tcx.ensure_ok().analysis(());
 
-            if let Some(metrics_dir) = &sess.opts.unstable_opts.metrics_dir {
-                dump_feature_usage_metrics(tcx, metrics_dir);
-            }
-
             if callbacks.after_analysis(compiler, tcx) == Compilation::Stop {
                 return early_exit();
             }
@@ -340,6 +336,10 @@ pub fn run_compiler(at_args: &[String], callbacks: &mut (dyn Callbacks + Send))
 
             let linker = Linker::codegen_and_build_linker(tcx, &*compiler.codegen_backend);
 
+            if let Some(metrics_dir) = &sess.opts.unstable_opts.metrics_dir {
+                dump_feature_usage_metrics(tcx, metrics_dir);
+            }
+
             tcx.report_unused_features();
 
             Some(linker)

compiler/rustc_hir/src/hir.rs

@@ -1645,6 +1645,19 @@ impl<'tcx> OwnerInfo<'tcx> {
     pub fn node(&self) -> OwnerNode<'tcx> {
         self.nodes.node()
     }
+
+    // A fingerprint that identifies the contents of the OwnerInfo.
+    // It only depends on `nodes` and `attrs` because `parenting` and `trait_map` are
+    // deterministically calculated from `nodes` and `attrs`.
+    #[inline]
+    pub fn fingerprint(&self) -> Fingerprint {
+        let body = self
+            .nodes
+            .opt_hash_including_bodies
+            .expect("HIR hash requested without needs_hir_hash");
+        let attrs = self.attrs.opt_hash.expect("HIR hash requested without needs_hir_hash");
+        body.combine(attrs)
+    }
 }
 
 #[derive(Copy, Clone, Debug, StableHash)]

compiler/rustc_interface/src/passes.rs

@@ -948,8 +948,13 @@ pub fn create_and_enter_global_ctxt<T, F: for<'tcx> FnOnce(TyCtxt<'tcx>) -> T>(
     let definitions = FreezeLock::new(Definitions::new(stable_crate_id));
 
     let stable_crate_ids = FreezeLock::new(StableCrateIdMap::default());
-    let untracked =
-        Untracked { cstore, source_span: AppendOnlyIndexVec::new(), definitions, stable_crate_ids };
+    let untracked = Untracked {
+        cstore,
+        source_span: AppendOnlyIndexVec::new(),
+        definitions,
+        stable_crate_ids,
+        local_crate_hash: OnceLock::new(),
+    };
 
     // We're constructing the HIR here; we don't care what we will
     // read, since we haven't even constructed the *input* to

compiler/rustc_metadata/src/lib.rs

@@ -1,5 +1,6 @@
 // tidy-alphabetical-start
 #![allow(internal_features)]
+#![feature(core_intrinsics)]
 #![feature(error_iter)]
 #![feature(file_buffered)]
 #![feature(gen_blocks)]

compiler/rustc_metadata/src/rmeta/encoder.rs

@@ -2,11 +2,14 @@ use std::borrow::Borrow;
 use std::collections::hash_map::Entry;
 use std::fs::File;
 use std::io::{Read, Seek, Write};
+use std::ops::Deref;
 use std::path::{Path, PathBuf};
 use std::sync::Arc;
 
 use rustc_data_structures::fx::{FxIndexMap, FxIndexSet};
 use rustc_data_structures::memmap::{Mmap, MmapMut};
+use rustc_data_structures::owned_slice::slice_owned;
+use rustc_data_structures::stable_hasher::{StableHash, StableHasher};
 use rustc_data_structures::sync::{par_for_each_in, par_join};
 use rustc_data_structures::temp_dir::MaybeTempDir;
 use rustc_data_structures::thousands::usize_with_underscores;
@@ -25,7 +28,7 @@ use rustc_middle::ty::AssocContainer;
 use rustc_middle::ty::codec::TyEncoder;
 use rustc_middle::ty::fast_reject::{self, TreatParams};
 use rustc_middle::{bug, span_bug};
-use rustc_serialize::{Decodable, Decoder, Encodable, Encoder, opaque};
+use rustc_serialize::{Decodable, Decoder, Encodable, Encoder};
 use rustc_session::config::mitigation_coverage::DeniedPartialMitigation;
 use rustc_session::config::{CrateType, OptLevel, TargetModifier};
 use rustc_span::hygiene::HygieneEncodeContext;
@@ -40,7 +43,7 @@ use crate::errors::{FailCreateFileEncoder, FailWriteFile};
 use crate::rmeta::*;
 
 pub(super) struct EncodeContext<'a, 'tcx> {
-    opaque: opaque::FileEncoder,
+    opaque: opaque::FileEncoder<'a>,
     tcx: TyCtxt<'tcx>,
     feat: &'tcx rustc_feature::Features,
     tables: TableBuilders,
@@ -718,13 +721,16 @@ impl<'a, 'tcx> EncodeContext<'a, 'tcx> {
         let denied_partial_mitigations = stat!("denied-partial-mitigations", || self
             .encode_enabled_denied_partial_mitigations());
 
+        let hash = Svh::new(self.opaque.hash());
+        tcx.untracked().local_crate_hash.set(hash).expect("local_crate_hash set twice");
+
         let root = stat!("final", || {
             let attrs = tcx.hir_krate_attrs();
             self.lazy(CrateRoot {
                 header: CrateHeader {
                     name: tcx.crate_name(LOCAL_CRATE),
                     triple: tcx.sess.opts.target_triple.clone(),
-                    hash: tcx.crate_hash(LOCAL_CRATE),
+                    hash,
                     is_proc_macro_crate: proc_macro_data.is_some(),
                     is_stub: false,
                 },
@@ -2428,22 +2434,6 @@ pub fn encode_metadata(tcx: TyCtxt<'_>, path: &Path, ref_path: Option<&Path>) {
     // there's no need to do dep-graph tracking for any of it.
     tcx.dep_graph.assert_ignored();
 
-    // Generate the metadata stub manually, as that is a small file compared to full metadata.
-    if let Some(ref_path) = ref_path {
-        let _prof_timer = tcx.prof.verbose_generic_activity("generate_crate_metadata_stub");
-
-        with_encode_metadata_header(tcx, ref_path, |ecx| {
-            let header: LazyValue<CrateHeader> = ecx.lazy(CrateHeader {
-                name: tcx.crate_name(LOCAL_CRATE),
-                triple: tcx.sess.opts.target_triple.clone(),
-                hash: tcx.crate_hash(LOCAL_CRATE),
-                is_proc_macro_crate: false,
-                is_stub: true,
-            });
-            header.position.get()
-        })
-    }
-
     let _prof_timer = tcx.prof.verbose_generic_activity("generate_crate_metadata");
 
     let dep_node = tcx.metadata_dep_node();
@@ -2462,6 +2452,31 @@ pub fn encode_metadata(tcx: TyCtxt<'_>, path: &Path, ref_path: Option<&Path>) {
             Ok(_) => {}
             Err(err) => tcx.dcx().emit_fatal(FailCreateFileEncoder { err }),
         };
+
+        // Read the SVH from the old metadata header.
+        let file = std::fs::File::open(&source_file).unwrap();
+        let mmap = unsafe { Mmap::map(file) }.unwrap();
+        let owned = slice_owned(mmap, Deref::deref);
+        let blob = MetadataBlob::new(owned);
+        let header = blob.expect("file already created").get_header();
+        tcx.untracked().local_crate_hash.set(header.hash).expect("local_crate_hash set twice");
+
+        // Generate the metadata stub manually, as that is a small file compared to full metadata.
+        if let Some(ref_path) = ref_path {
+            let _prof_timer = tcx.prof.verbose_generic_activity("generate_crate_metadata_stub");
+
+            with_encode_metadata_header(tcx, ref_path, |ecx| {
+                let header: LazyValue<CrateHeader> = ecx.lazy(CrateHeader {
+                    name: tcx.crate_name(LOCAL_CRATE),
+                    triple: tcx.sess.opts.target_triple.clone(),
+                    hash: tcx.crate_hash(LOCAL_CRATE),
+                    is_proc_macro_crate: false,
+                    is_stub: true,
+                });
+                header.position.get()
+            })
+        }
+
         return;
     };
 
@@ -2503,14 +2518,36 @@ pub fn encode_metadata(tcx: TyCtxt<'_>, path: &Path, ref_path: Option<&Path>) {
         },
         None,
     );
+
+    // Generate the metadata stub manually, as that is a small file compared to full metadata.
+    if let Some(ref_path) = ref_path {
+        let _prof_timer = tcx.prof.verbose_generic_activity("generate_crate_metadata_stub");
+
+        with_encode_metadata_header(tcx, ref_path, |ecx| {
+            let header: LazyValue<CrateHeader> = ecx.lazy(CrateHeader {
+                name: tcx.crate_name(LOCAL_CRATE),
+                triple: tcx.sess.opts.target_triple.clone(),
+                hash: tcx.crate_hash(LOCAL_CRATE),
+                is_proc_macro_crate: false,
+                is_stub: true,
+            });
+            header.position.get()
+        })
+    }
 }
 
 fn with_encode_metadata_header(
     tcx: TyCtxt<'_>,
     path: &Path,
     f: impl FnOnce(&mut EncodeContext<'_, '_>) -> usize,
 ) {
-    let mut encoder = opaque::FileEncoder::new(path)
+    let mut stable_hasher = StableHasher::new();
+    // Add dep_tracking_hash to ensure the SVH changes when any tracked flag changes.
+    tcx.with_stable_hashing_context(|mut hcx| {
+        tcx.sess.opts.dep_tracking_hash(true).stable_hash(&mut hcx, &mut stable_hasher);
+    });
+
+    let mut encoder = opaque::FileEncoder::new(path, &mut stable_hasher)
         .unwrap_or_else(|err| tcx.dcx().emit_fatal(FailCreateFileEncoder { err }));
     encoder.emit_raw_bytes(METADATA_HEADER);
 

compiler/rustc_metadata/src/rmeta/mod.rs

@@ -35,7 +35,6 @@ use rustc_middle::mir::ConstValue;
 use rustc_middle::ty::fast_reject::SimplifiedType;
 use rustc_middle::ty::{self, Ty, TyCtxt};
 use rustc_middle::util::Providers;
-use rustc_serialize::opaque::FileEncoder;
 use rustc_session::config::mitigation_coverage::DeniedPartialMitigation;
 use rustc_session::config::{SymbolManglingVersion, TargetModifier};
 use rustc_session::cstore::{CrateDepKind, ForeignModule, LinkagePreference, NativeLib};
@@ -50,6 +49,7 @@ use crate::eii::EiiMapEncodedKeyValue;
 mod decoder;
 mod def_path_hash_map;
 mod encoder;
+mod opaque;
 mod parameterized;
 mod table;
 
@@ -364,7 +364,7 @@ macro_rules! define_tables {
         }
 
         impl TableBuilders {
-            fn encode(&self, buf: &mut FileEncoder) -> LazyTables {
+            fn encode(&self, buf: &mut opaque::FileEncoder<'_>) -> LazyTables {
                 LazyTables {
                     $($name1: self.$name1.encode(buf),)+
                     $($name2: self.$name2.encode(buf),)+