Auto merge of rust-lang#113772 - nnethercote:streamline-size-estimates-2, r=wesleywiser

Streamline size estimates (take 2)

This was merged in rust-lang#113684 but then [something happened](rust-lang#113684 (comment)):

> There has been a bors issue that lead to the merge commit of this PR getting purged from master.
> You'll have to make a new PR to reapply it.

So this is exactly the same changes.

`@bors` r=wesleywiser

Author: bors

compiler/rustc_codegen_cranelift/src/driver/mod.rs

@@ -5,7 +5,7 @@
 //! [`codegen_static`]: crate::constant::codegen_static
 
 use rustc_data_structures::profiling::SelfProfilerRef;
-use rustc_middle::mir::mono::{Linkage as RLinkage, MonoItem, Visibility};
+use rustc_middle::mir::mono::{MonoItem, MonoItemData};
 
 use crate::prelude::*;
 
@@ -16,11 +16,11 @@ pub(crate) mod jit;
 fn predefine_mono_items<'tcx>(
     tcx: TyCtxt<'tcx>,
     module: &mut dyn Module,
-    mono_items: &[(MonoItem<'tcx>, (RLinkage, Visibility))],
+    mono_items: &[(MonoItem<'tcx>, MonoItemData)],
 ) {
     tcx.prof.generic_activity("predefine functions").run(|| {
         let is_compiler_builtins = tcx.is_compiler_builtins(LOCAL_CRATE);
-        for &(mono_item, (linkage, visibility)) in mono_items {
+        for &(mono_item, data) in mono_items {
             match mono_item {
                 MonoItem::Fn(instance) => {
                     let name = tcx.symbol_name(instance).name;
@@ -29,8 +29,8 @@ fn predefine_mono_items<'tcx>(
                         get_function_sig(tcx, module.target_config().default_call_conv, instance);
                     let linkage = crate::linkage::get_clif_linkage(
                         mono_item,
-                        linkage,
-                        visibility,
+                        data.linkage,
+                        data.visibility,
                         is_compiler_builtins,
                     );
                     module.declare_function(name, linkage, &sig).unwrap();

compiler/rustc_codegen_gcc/src/base.rs

@@ -159,8 +159,8 @@ pub fn compile_codegen_unit(tcx: TyCtxt<'_>, cgu_name: Symbol, supports_128bit_i
             let cx = CodegenCx::new(&context, cgu, tcx, supports_128bit_integers);
 
             let mono_items = cgu.items_in_deterministic_order(tcx);
-            for &(mono_item, (linkage, visibility)) in &mono_items {
-                mono_item.predefine::<Builder<'_, '_, '_>>(&cx, linkage, visibility);
+            for &(mono_item, data) in &mono_items {
+                mono_item.predefine::<Builder<'_, '_, '_>>(&cx, data.linkage, data.visibility);
             }
 
             // ... and now that we have everything pre-defined, fill out those definitions.

compiler/rustc_codegen_llvm/src/base.rs

@@ -86,8 +86,8 @@ pub fn compile_codegen_unit(tcx: TyCtxt<'_>, cgu_name: Symbol) -> (ModuleCodegen
         {
             let cx = CodegenCx::new(tcx, cgu, &llvm_module);
             let mono_items = cx.codegen_unit.items_in_deterministic_order(cx.tcx);
-            for &(mono_item, (linkage, visibility)) in &mono_items {
-                mono_item.predefine::<Builder<'_, '_, '_>>(&cx, linkage, visibility);
+            for &(mono_item, data) in &mono_items {
+                mono_item.predefine::<Builder<'_, '_, '_>>(&cx, data.linkage, data.visibility);
             }
 
             // ... and now that we have everything pre-defined, fill out those definitions.

compiler/rustc_codegen_ssa/src/back/symbol_export.rs

@@ -328,14 +328,14 @@ fn exported_symbols_provider_local(
 
         let (_, cgus) = tcx.collect_and_partition_mono_items(());
 
-        for (mono_item, &(linkage, visibility)) in cgus.iter().flat_map(|cgu| cgu.items().iter()) {
-            if linkage != Linkage::External {
+        for (mono_item, data) in cgus.iter().flat_map(|cgu| cgu.items().iter()) {
+            if data.linkage != Linkage::External {
                 // We can only re-use things with external linkage, otherwise
                 // we'll get a linker error
                 continue;
             }
 
-            if need_visibility && visibility == Visibility::Hidden {
+            if need_visibility && data.visibility == Visibility::Hidden {
                 // If we potentially share things from Rust dylibs, they must
                 // not be hidden
                 continue;

compiler/rustc_middle/src/mir/mono.rs

@@ -59,12 +59,19 @@ impl<'tcx> MonoItem<'tcx> {
     pub fn size_estimate(&self, tcx: TyCtxt<'tcx>) -> usize {
         match *self {
             MonoItem::Fn(instance) => {
-                // Estimate the size of a function based on how many statements
-                // it contains.
-                tcx.instance_def_size_estimate(instance.def)
+                match instance.def {
+                    // "Normal" functions size estimate: the number of
+                    // statements, plus one for the terminator.
+                    InstanceDef::Item(..) | InstanceDef::DropGlue(..) => {
+                        let mir = tcx.instance_mir(instance.def);
+                        mir.basic_blocks.iter().map(|bb| bb.statements.len() + 1).sum()
+                    }
+                    // Other compiler-generated shims size estimate: 1
+                    _ => 1,
+                }
             }
-            // Conservatively estimate the size of a static declaration
-            // or assembly to be 1.
+            // Conservatively estimate the size of a static declaration or
+            // assembly item to be 1.
             MonoItem::Static(_) | MonoItem::GlobalAsm(_) => 1,
         }
     }
@@ -230,14 +237,22 @@ pub struct CodegenUnit<'tcx> {
     /// contain something unique to this crate (e.g., a module path)
     /// as well as the crate name and disambiguator.
     name: Symbol,
-    items: FxHashMap<MonoItem<'tcx>, (Linkage, Visibility)>,
+    items: FxHashMap<MonoItem<'tcx>, MonoItemData>,
     size_estimate: usize,
     primary: bool,
     /// True if this is CGU is used to hold code coverage information for dead code,
     /// false otherwise.
     is_code_coverage_dead_code_cgu: bool,
 }
 
+/// Auxiliary info about a `MonoItem`.
+#[derive(Copy, Clone, PartialEq, Debug, HashStable)]
+pub struct MonoItemData {
+    pub linkage: Linkage,
+    pub visibility: Visibility,
+    pub size_estimate: usize,
+}
+
 /// Specifies the linkage type for a `MonoItem`.
 ///
 /// See <https://llvm.org/docs/LangRef.html#linkage-types> for more details about these variants.
@@ -292,12 +307,12 @@ impl<'tcx> CodegenUnit<'tcx> {
     }
 
     /// The order of these items is non-determinstic.
-    pub fn items(&self) -> &FxHashMap<MonoItem<'tcx>, (Linkage, Visibility)> {
+    pub fn items(&self) -> &FxHashMap<MonoItem<'tcx>, MonoItemData> {
         &self.items
     }
 
     /// The order of these items is non-determinstic.
-    pub fn items_mut(&mut self) -> &mut FxHashMap<MonoItem<'tcx>, (Linkage, Visibility)> {
+    pub fn items_mut(&mut self) -> &mut FxHashMap<MonoItem<'tcx>, MonoItemData> {
         &mut self.items
     }
 
@@ -320,16 +335,16 @@ impl<'tcx> CodegenUnit<'tcx> {
         base_n::encode(hash, base_n::CASE_INSENSITIVE)
     }
 
-    pub fn compute_size_estimate(&mut self, tcx: TyCtxt<'tcx>) {
-        // Estimate the size of a codegen unit as (approximately) the number of MIR
-        // statements it corresponds to.
-        self.size_estimate = self.items.keys().map(|mi| mi.size_estimate(tcx)).sum();
+    pub fn compute_size_estimate(&mut self) {
+        // The size of a codegen unit as the sum of the sizes of the items
+        // within it.
+        self.size_estimate = self.items.values().map(|data| data.size_estimate).sum();
     }
 
-    #[inline]
     /// Should only be called if [`compute_size_estimate`] has previously been called.
     ///
     /// [`compute_size_estimate`]: Self::compute_size_estimate
+    #[inline]
     pub fn size_estimate(&self) -> usize {
         // Items are never zero-sized, so if we have items the estimate must be
         // non-zero, unless we forgot to call `compute_size_estimate` first.
@@ -355,7 +370,7 @@ impl<'tcx> CodegenUnit<'tcx> {
     pub fn items_in_deterministic_order(
         &self,
         tcx: TyCtxt<'tcx>,
-    ) -> Vec<(MonoItem<'tcx>, (Linkage, Visibility))> {
+    ) -> Vec<(MonoItem<'tcx>, MonoItemData)> {
         // The codegen tests rely on items being process in the same order as
         // they appear in the file, so for local items, we sort by node_id first
         #[derive(PartialEq, Eq, PartialOrd, Ord)]
@@ -390,7 +405,7 @@ impl<'tcx> CodegenUnit<'tcx> {
             )
         }
 
-        let mut items: Vec<_> = self.items().iter().map(|(&i, &l)| (i, l)).collect();
+        let mut items: Vec<_> = self.items().iter().map(|(&i, &data)| (i, data)).collect();
         items.sort_by_cached_key(|&(i, _)| item_sort_key(tcx, i));
         items
     }

compiler/rustc_middle/src/query/mod.rs

@@ -2080,12 +2080,6 @@ rustc_queries! {
         desc { "looking up supported target features" }
     }
 
-    /// Get an estimate of the size of an InstanceDef based on its MIR for CGU partitioning.
-    query instance_def_size_estimate(def: ty::InstanceDef<'tcx>)
-        -> usize {
-        desc { |tcx| "estimating size for `{}`", tcx.def_path_str(def.def_id()) }
-    }
-
     query features_query(_: ()) -> &'tcx rustc_feature::Features {
         feedable
         desc { "looking up enabled feature gates" }