Introduce -Zmacro-stats.

It collects data about macro expansions and prints them in a table after
expansion finishes. It's very useful for detecting macro bloat,
especially for proc macros.

Details:
- It measures code snippets by pretty-printing them and then measuring
  lines and bytes. This required a bunch of additional pretty-printing
  plumbing, in `rustc_ast_pretty` and `rustc_expand`.
- The measurement is done in `MacroExpander::expand_invoc`.
- The measurements are stored in `ExtCtxt::macro_stats`.

Author: nnethercote

compiler/rustc_ast_pretty/src/pprust/mod.rs

@@ -53,6 +53,18 @@ pub fn item_to_string(i: &ast::Item) -> String {
     State::new().item_to_string(i)
 }
 
+pub fn assoc_item_to_string(i: &ast::AssocItem) -> String {
+    State::new().assoc_item_to_string(i)
+}
+
+pub fn foreign_item_to_string(i: &ast::ForeignItem) -> String {
+    State::new().foreign_item_to_string(i)
+}
+
+pub fn stmt_to_string(s: &ast::Stmt) -> String {
+    State::new().stmt_to_string(s)
+}
+
 pub fn path_to_string(p: &ast::Path) -> String {
     State::new().path_to_string(p)
 }

compiler/rustc_ast_pretty/src/pprust/state.rs

@@ -1063,6 +1063,14 @@ pub trait PrintState<'a>: std::ops::Deref<Target = pp::Printer> + std::ops::Dere
         Self::to_string(|s| s.print_item(i))
     }
 
+    fn assoc_item_to_string(&self, i: &ast::AssocItem) -> String {
+        Self::to_string(|s| s.print_assoc_item(i))
+    }
+
+    fn foreign_item_to_string(&self, i: &ast::ForeignItem) -> String {
+        Self::to_string(|s| s.print_foreign_item(i))
+    }
+
     fn path_to_string(&self, p: &ast::Path) -> String {
         Self::to_string(|s| s.print_path(p, false, 0))
     }

compiler/rustc_ast_pretty/src/pprust/state/item.rs

@@ -28,7 +28,7 @@ impl<'a> State<'a> {
         }
     }
 
-    fn print_foreign_item(&mut self, item: &ast::ForeignItem) {
+    pub(crate) fn print_foreign_item(&mut self, item: &ast::ForeignItem) {
         let ast::Item { id, span, ref attrs, ref kind, ref vis, tokens: _ } = *item;
         self.ann.pre(self, AnnNode::SubItem(id));
         self.hardbreak_if_not_bol();
@@ -548,7 +548,7 @@ impl<'a> State<'a> {
         }
     }
 
-    fn print_assoc_item(&mut self, item: &ast::AssocItem) {
+    pub(crate) fn print_assoc_item(&mut self, item: &ast::AssocItem) {
         let ast::Item { id, span, ref attrs, ref kind, ref vis, tokens: _ } = *item;
         self.ann.pre(self, AnnNode::SubItem(id));
         self.hardbreak_if_not_bol();

compiler/rustc_expand/src/base.rs

@@ -12,7 +12,7 @@ use rustc_ast::tokenstream::TokenStream;
 use rustc_ast::visit::{AssocCtxt, Visitor};
 use rustc_ast::{self as ast, AttrVec, Attribute, HasAttrs, Item, NodeId, PatKind};
 use rustc_attr_data_structures::{AttributeKind, Deprecation, Stability, find_attr};
-use rustc_data_structures::fx::FxIndexMap;
+use rustc_data_structures::fx::{FxHashMap, FxIndexMap};
 use rustc_data_structures::sync;
 use rustc_errors::{DiagCtxtHandle, ErrorGuaranteed, PResult};
 use rustc_feature::Features;
@@ -727,6 +727,7 @@ pub enum SyntaxExtensionKind {
     /// A trivial attribute "macro" that does nothing,
     /// only keeps the attribute and marks it as inert,
     /// thus making it ineligible for further expansion.
+    /// E.g. `#[default]`, `#[rustfmt::skip]`.
     NonMacroAttr,
 
     /// A token-based derive macro.
@@ -1166,6 +1167,49 @@ pub struct ExpansionData {
     pub is_trailing_mac: bool,
 }
 
+#[derive(Default)]
+pub struct MacroStat {
+    /// Number of invocations of the macro.
+    pub count: usize,
+
+    /// Net increase in number of lines of code (when pretty-printed), i.e.
+    /// `lines(output) - lines(invocation)`. Can be negative because a macro
+    /// output may be smaller than the invocation.
+    pub lines: isize,
+
+    /// Net increase in number of lines of code (when pretty-printed), i.e.
+    /// `bytes(output) - bytes(invocation)`. Can be negative because a macro
+    /// output may be smaller than the invocation.
+    pub bytes: isize,
+}
+
+#[derive(Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash)]
+pub enum MacroStatKind {
+    Attr,
+    Derive,
+    // This covers declarative macros and fn-like proc macros.
+    FnLike,
+}
+
+impl MacroStatKind {
+    pub fn label(&self) -> String {
+        match self {
+            MacroStatKind::Attr => "attr".to_string(),
+            MacroStatKind::Derive => "derive".to_string(),
+            MacroStatKind::FnLike => "fn-like".to_string(),
+        }
+    }
+
+    // Decorate the name to look like the macro invocation.
+    pub fn decorated_name(&self, name: &str) -> String {
+        match self {
+            MacroStatKind::Attr => format!("#[{name}]"),
+            MacroStatKind::Derive => name.to_string(),
+            MacroStatKind::FnLike => format!("{name}!"),
+        }
+    }
+}
+
 /// One of these is made during expansion and incrementally updated as we go;
 /// when a macro expansion occurs, the resulting nodes have the `backtrace()
 /// -> expn_data` of their expansion context stored into their span.
@@ -1189,6 +1233,8 @@ pub struct ExtCtxt<'a> {
     /// in the AST, but insert it here so that we know
     /// not to expand it again.
     pub(super) expanded_inert_attrs: MarkedAttrs,
+    /// `-Zmacro-stats` data.
+    pub macro_stats: FxHashMap<(String, MacroStatKind), MacroStat>,
 }
 
 impl<'a> ExtCtxt<'a> {
@@ -1218,6 +1264,7 @@ impl<'a> ExtCtxt<'a> {
             expansions: FxIndexMap::default(),
             expanded_inert_attrs: MarkedAttrs::new(),
             buffered_early_lint: vec![],
+            macro_stats: Default::default(),
         }
     }