Auto merge of #94468 - Amanieu:global_asm_sym, r=nagisa

Implement sym operands for global_asm!

Tracking issue: #93333

This PR is pretty much a complete rewrite of `sym` operand support for inline assembly so that the same implementation can be shared by `asm!` and `global_asm!`. The main changes are:
- At the AST level, `sym` is represented as a special `InlineAsmSym` AST node containing a path instead of an `Expr`.
- At the HIR level, `sym` is split into `SymStatic` and `SymFn` depending on whether the path resolves to a static during AST lowering (defaults to `SynFn` if `get_early_res` fails).
  - `SymFn` is just an `AnonConst`. It runs through typeck and we just collect the resulting type at the end. An error is emitted if the type is not a `FnDef`.
  - `SymStatic` directly holds a path and the `DefId` of the `static` that it is pointing to.
- The representation at the MIR level is mostly unchanged. There is a minor change to THIR where `SymFn` is a constant instead of an expression.
- At the codegen level we need to apply the target's symbol mangling to the result of `tcx.symbol_name()` depending on the target. This is done by calling the LLVM name mangler, which handles all of the details.
  - On Mach-O, all symbols have a leading underscore.
  - On x86 Windows, different mangling is used for cdecl, stdcall, fastcall and vectorcall.
  - No mangling is needed on other platforms.

r? `@nagisa`
cc `@eddyb`

Author: bors

compiler/rustc_ast/src/ast.rs

@@ -2061,6 +2061,20 @@ impl InlineAsmTemplatePiece {
     }
 }
 
+/// Inline assembly symbol operands get their own AST node that is somewhat
+/// similar to `AnonConst`.
+///
+/// The main difference is that we specifically don't assign it `DefId` in
+/// `DefCollector`. Instead this is deferred until AST lowering where we
+/// lower it to an `AnonConst` (for functions) or a `Path` (for statics)
+/// depending on what the path resolves to.
+#[derive(Clone, Encodable, Decodable, Debug)]
+pub struct InlineAsmSym {
+    pub id: NodeId,
+    pub qself: Option<QSelf>,
+    pub path: Path,
+}
+
 /// Inline assembly operand.
 ///
 /// E.g., `out("eax") result` as in `asm!("mov eax, 2", out("eax") result)`.
@@ -2090,7 +2104,7 @@ pub enum InlineAsmOperand {
         anon_const: AnonConst,
     },
     Sym {
-        expr: P<Expr>,
+        sym: InlineAsmSym,
     },
 }
 

compiler/rustc_ast/src/mut_visit.rs

@@ -280,6 +280,14 @@ pub trait MutVisitor: Sized {
     fn flat_map_pat_field(&mut self, fp: PatField) -> SmallVec<[PatField; 1]> {
         noop_flat_map_pat_field(fp, self)
     }
+
+    fn visit_inline_asm(&mut self, asm: &mut InlineAsm) {
+        noop_visit_inline_asm(asm, self)
+    }
+
+    fn visit_inline_asm_sym(&mut self, sym: &mut InlineAsmSym) {
+        noop_visit_inline_asm_sym(sym, self)
+    }
 }
 
 /// Use a map-style function (`FnOnce(T) -> T`) to overwrite a `&mut T`. Useful
@@ -1019,7 +1027,7 @@ pub fn noop_visit_item_kind<T: MutVisitor>(kind: &mut ItemKind, vis: &mut T) {
             }
         }
         ItemKind::ForeignMod(nm) => vis.visit_foreign_mod(nm),
-        ItemKind::GlobalAsm(asm) => noop_visit_inline_asm(asm, vis),
+        ItemKind::GlobalAsm(asm) => vis.visit_inline_asm(asm),
         ItemKind::TyAlias(box TyAlias {
             defaultness, generics, where_clauses, bounds, ty, ..
         }) => {
@@ -1237,25 +1245,34 @@ pub fn noop_visit_anon_const<T: MutVisitor>(AnonConst { id, value }: &mut AnonCo
     vis.visit_expr(value);
 }
 
-fn noop_visit_inline_asm<T: MutVisitor>(asm: &mut InlineAsm, vis: &mut T) {
+pub fn noop_visit_inline_asm<T: MutVisitor>(asm: &mut InlineAsm, vis: &mut T) {
     for (op, _) in &mut asm.operands {
         match op {
             InlineAsmOperand::In { expr, .. }
             | InlineAsmOperand::Out { expr: Some(expr), .. }
-            | InlineAsmOperand::InOut { expr, .. }
-            | InlineAsmOperand::Sym { expr, .. } => vis.visit_expr(expr),
+            | InlineAsmOperand::InOut { expr, .. } => vis.visit_expr(expr),
             InlineAsmOperand::Out { expr: None, .. } => {}
             InlineAsmOperand::SplitInOut { in_expr, out_expr, .. } => {
                 vis.visit_expr(in_expr);
                 if let Some(out_expr) = out_expr {
                     vis.visit_expr(out_expr);
                 }
             }
-            InlineAsmOperand::Const { anon_const, .. } => vis.visit_anon_const(anon_const),
+            InlineAsmOperand::Const { anon_const } => vis.visit_anon_const(anon_const),
+            InlineAsmOperand::Sym { sym } => vis.visit_inline_asm_sym(sym),
         }
     }
 }
 
+pub fn noop_visit_inline_asm_sym<T: MutVisitor>(
+    InlineAsmSym { id, qself, path }: &mut InlineAsmSym,
+    vis: &mut T,
+) {
+    vis.visit_id(id);
+    vis.visit_qself(qself);
+    vis.visit_path(path);
+}
+
 pub fn noop_visit_expr<T: MutVisitor>(
     Expr { kind, id, span, attrs, tokens }: &mut Expr,
     vis: &mut T,
@@ -1374,7 +1391,7 @@ pub fn noop_visit_expr<T: MutVisitor>(
         ExprKind::Ret(expr) => {
             visit_opt(expr, |expr| vis.visit_expr(expr));
         }
-        ExprKind::InlineAsm(asm) => noop_visit_inline_asm(asm, vis),
+        ExprKind::InlineAsm(asm) => vis.visit_inline_asm(asm),
         ExprKind::MacCall(mac) => vis.visit_mac_call(mac),
         ExprKind::Struct(se) => {
             let StructExpr { qself, path, fields, rest } = se.deref_mut();

compiler/rustc_ast/src/visit.rs

@@ -214,6 +214,12 @@ pub trait Visitor<'ast>: Sized {
     fn visit_crate(&mut self, krate: &'ast Crate) {
         walk_crate(self, krate)
     }
+    fn visit_inline_asm(&mut self, asm: &'ast InlineAsm) {
+        walk_inline_asm(self, asm)
+    }
+    fn visit_inline_asm_sym(&mut self, sym: &'ast InlineAsmSym) {
+        walk_inline_asm_sym(self, sym)
+    }
 }
 
 #[macro_export]
@@ -717,13 +723,12 @@ pub fn walk_anon_const<'a, V: Visitor<'a>>(visitor: &mut V, constant: &'a AnonCo
     visitor.visit_expr(&constant.value);
 }
 
-fn walk_inline_asm<'a, V: Visitor<'a>>(visitor: &mut V, asm: &'a InlineAsm) {
+pub fn walk_inline_asm<'a, V: Visitor<'a>>(visitor: &mut V, asm: &'a InlineAsm) {
     for (op, _) in &asm.operands {
         match op {
             InlineAsmOperand::In { expr, .. }
             | InlineAsmOperand::Out { expr: Some(expr), .. }
-            | InlineAsmOperand::InOut { expr, .. }
-            | InlineAsmOperand::Sym { expr, .. } => visitor.visit_expr(expr),
+            | InlineAsmOperand::InOut { expr, .. } => visitor.visit_expr(expr),
             InlineAsmOperand::Out { expr: None, .. } => {}
             InlineAsmOperand::SplitInOut { in_expr, out_expr, .. } => {
                 visitor.visit_expr(in_expr);
@@ -732,10 +737,18 @@ fn walk_inline_asm<'a, V: Visitor<'a>>(visitor: &mut V, asm: &'a InlineAsm) {
                 }
             }
             InlineAsmOperand::Const { anon_const, .. } => visitor.visit_anon_const(anon_const),
+            InlineAsmOperand::Sym { sym } => visitor.visit_inline_asm_sym(sym),
         }
     }
 }
 
+pub fn walk_inline_asm_sym<'a, V: Visitor<'a>>(visitor: &mut V, sym: &'a InlineAsmSym) {
+    if let Some(ref qself) = sym.qself {
+        visitor.visit_ty(&qself.ty);
+    }
+    visitor.visit_path(&sym.path, sym.id);
+}
+
 pub fn walk_expr<'a, V: Visitor<'a>>(visitor: &mut V, expression: &'a Expr) {
     walk_list!(visitor, visit_attribute, expression.attrs.iter());
 

compiler/rustc_ast_lowering/src/asm.rs

@@ -1,12 +1,17 @@
+use crate::{ImplTraitContext, ImplTraitPosition, ParamMode};
+
 use super::LoweringContext;
 
+use rustc_ast::ptr::P;
 use rustc_ast::*;
 use rustc_data_structures::fx::FxHashMap;
 use rustc_data_structures::stable_set::FxHashSet;
 use rustc_errors::struct_span_err;
 use rustc_hir as hir;
+use rustc_hir::def::{DefKind, Res};
+use rustc_hir::definitions::DefPathData;
 use rustc_session::parse::feature_err;
-use rustc_span::{sym, Span};
+use rustc_span::{sym, ExpnId, Span};
 use rustc_target::asm;
 use std::collections::hash_map::Entry;
 use std::fmt::Write;
@@ -188,7 +193,7 @@ impl<'a, 'hir> LoweringContext<'a, 'hir> {
                             anon_const: self.lower_anon_const(anon_const),
                         }
                     }
-                    InlineAsmOperand::Sym { ref expr } => {
+                    InlineAsmOperand::Sym { ref sym } => {
                         if !self.sess.features_untracked().asm_sym {
                             feature_err(
                                 &self.sess.parse_sess,
@@ -198,7 +203,54 @@ impl<'a, 'hir> LoweringContext<'a, 'hir> {
                             )
                             .emit();
                         }
-                        hir::InlineAsmOperand::Sym { expr: self.lower_expr_mut(expr) }
+
+                        let static_def_id = self
+                            .resolver
+                            .get_partial_res(sym.id)
+                            .filter(|res| res.unresolved_segments() == 0)
+                            .and_then(|res| {
+                                if let Res::Def(DefKind::Static(_), def_id) = res.base_res() {
+                                    Some(def_id)
+                                } else {
+                                    None
+                                }
+                            });
+
+                        if let Some(def_id) = static_def_id {
+                            let path = self.lower_qpath(
+                                sym.id,
+                                &sym.qself,
+                                &sym.path,
+                                ParamMode::Optional,
+                                ImplTraitContext::Disallowed(ImplTraitPosition::Path),
+                            );
+                            hir::InlineAsmOperand::SymStatic { path, def_id }
+                        } else {
+                            // Replace the InlineAsmSym AST node with an
+                            // Expr using the name node id.
+                            let expr = Expr {
+                                id: sym.id,
+                                kind: ExprKind::Path(sym.qself.clone(), sym.path.clone()),
+                                span: *op_sp,
+                                attrs: AttrVec::new(),
+                                tokens: None,
+                            };
+
+                            // Wrap the expression in an AnonConst.
+                            let parent_def_id = self.current_hir_id_owner;
+                            let node_id = self.resolver.next_node_id();
+                            self.resolver.create_def(
+                                parent_def_id,
+                                node_id,
+                                DefPathData::AnonConst,
+                                ExpnId::root(),
+                                *op_sp,
+                            );
+                            let anon_const = AnonConst { id: node_id, value: P(expr) };
+                            hir::InlineAsmOperand::SymFn {
+                                anon_const: self.lower_anon_const(&anon_const),
+                            }
+                        }
                     }
                 };
                 (op, self.lower_span(*op_sp))
@@ -260,7 +312,8 @@ impl<'a, 'hir> LoweringContext<'a, 'hir> {
                         err.span_label(op_sp, "argument");
                         err.emit();
                     }
-                    hir::InlineAsmOperand::Sym { .. } => {
+                    hir::InlineAsmOperand::SymFn { .. }
+                    | hir::InlineAsmOperand::SymStatic { .. } => {
                         let mut err = sess.struct_span_err(
                             placeholder_span,
                             "asm template modifiers are not allowed for `sym` arguments",
@@ -308,7 +361,9 @@ impl<'a, 'hir> LoweringContext<'a, 'hir> {
                         hir::InlineAsmOperand::InOut { .. }
                         | hir::InlineAsmOperand::SplitInOut { .. } => (true, true),
 
-                        hir::InlineAsmOperand::Const { .. } | hir::InlineAsmOperand::Sym { .. } => {
+                        hir::InlineAsmOperand::Const { .. }
+                        | hir::InlineAsmOperand::SymFn { .. }
+                        | hir::InlineAsmOperand::SymStatic { .. } => {
                             unreachable!()
                         }
                     };

compiler/rustc_ast_pretty/src/pprust/state.rs

@@ -1266,10 +1266,14 @@ impl<'a> State<'a> {
                         s.space();
                         s.print_expr(&anon_const.value);
                     }
-                    InlineAsmOperand::Sym { expr } => {
+                    InlineAsmOperand::Sym { sym } => {
                         s.word("sym");
                         s.space();
-                        s.print_expr(expr);
+                        if let Some(qself) = &sym.qself {
+                            s.print_qpath(&sym.path, qself, true);
+                        } else {
+                            s.print_path(&sym.path, true, 0);
+                        }
                     }
                 }
             }

compiler/rustc_builtin_macros/src/asm.rs

@@ -154,17 +154,19 @@ pub fn parse_asm_args<'a>(
         } else if p.eat_keyword(kw::Const) {
             let anon_const = p.parse_anon_const_expr()?;
             ast::InlineAsmOperand::Const { anon_const }
-        } else if !is_global_asm && p.eat_keyword(sym::sym) {
+        } else if p.eat_keyword(sym::sym) {
             let expr = p.parse_expr()?;
-            match expr.kind {
-                ast::ExprKind::Path(..) => {}
-                _ => {
-                    let err = diag
-                        .struct_span_err(expr.span, "argument to `sym` must be a path expression");
-                    return Err(err);
-                }
-            }
-            ast::InlineAsmOperand::Sym { expr }
+            let ast::ExprKind::Path(qself, path) = &expr.kind else {
+                let err = diag
+                    .struct_span_err(expr.span, "expected a path for argument to `sym`");
+                return Err(err);
+            };
+            let sym = ast::InlineAsmSym {
+                id: ast::DUMMY_NODE_ID,
+                qself: qself.clone(),
+                path: path.clone(),
+            };
+            ast::InlineAsmOperand::Sym { sym }
         } else if allow_templates {
             let template = p.parse_expr()?;
             // If it can't possibly expand to a string, provide diagnostics here to include other

compiler/rustc_codegen_gcc/src/asm.rs

@@ -258,9 +258,14 @@ impl<'a, 'gcc, 'tcx> AsmBuilderMethods<'tcx> for Builder<'a, 'gcc, 'tcx> {
                 }
 
                 InlineAsmOperandRef::SymFn { instance } => {
+                    // TODO(@Amanieu): Additional mangling is needed on
+                    // some targets to add a leading underscore (Mach-O)
+                    // or byte count suffixes (x86 Windows).
                     constants_len += self.tcx.symbol_name(instance).name.len();
                 }
                 InlineAsmOperandRef::SymStatic { def_id } => {
+                    // TODO(@Amanieu): Additional mangling is needed on
+                    // some targets to add a leading underscore (Mach-O).
                     constants_len += self.tcx.symbol_name(Instance::mono(self.tcx, def_id)).name.len();
                 }
             }
@@ -412,13 +417,16 @@ impl<'a, 'gcc, 'tcx> AsmBuilderMethods<'tcx> for Builder<'a, 'gcc, 'tcx> {
                         }
 
                         InlineAsmOperandRef::SymFn { instance } => {
+                            // TODO(@Amanieu): Additional mangling is needed on
+                            // some targets to add a leading underscore (Mach-O)
+                            // or byte count suffixes (x86 Windows).
                             let name = self.tcx.symbol_name(instance).name;
                             template_str.push_str(name);
                         }
 
                         InlineAsmOperandRef::SymStatic { def_id } => {
-                            // TODO(@Commeownist): This may not be sufficient for all kinds of statics.
-                            // Some statics may need the `@plt` suffix, like thread-local vars.
+                            // TODO(@Amanieu): Additional mangling is needed on
+                            // some targets to add a leading underscore (Mach-O).
                             let instance = Instance::mono(self.tcx, def_id);
                             let name = self.tcx.symbol_name(instance).name;
                             template_str.push_str(name);
@@ -656,8 +664,8 @@ fn dummy_output_type<'gcc, 'tcx>(cx: &CodegenCx<'gcc, 'tcx>, reg: InlineAsmRegCl
     }
 }
 
-impl<'gcc, 'tcx> AsmMethods for CodegenCx<'gcc, 'tcx> {
-    fn codegen_global_asm(&self, template: &[InlineAsmTemplatePiece], operands: &[GlobalAsmOperandRef], options: InlineAsmOptions, _line_spans: &[Span]) {
+impl<'gcc, 'tcx> AsmMethods<'tcx> for CodegenCx<'gcc, 'tcx> {
+    fn codegen_global_asm(&self, template: &[InlineAsmTemplatePiece], operands: &[GlobalAsmOperandRef<'tcx>], options: InlineAsmOptions, _line_spans: &[Span]) {
         let asm_arch = self.tcx.sess.asm_arch.unwrap();
 
         // Default to Intel syntax on x86
@@ -690,6 +698,22 @@ impl<'gcc, 'tcx> AsmMethods for CodegenCx<'gcc, 'tcx> {
                             // here unlike normal inline assembly.
                             template_str.push_str(string);
                         }
+
+                        GlobalAsmOperandRef::SymFn { instance } => {
+                            // TODO(@Amanieu): Additional mangling is needed on
+                            // some targets to add a leading underscore (Mach-O)
+                            // or byte count suffixes (x86 Windows).
+                            let name = self.tcx.symbol_name(instance).name;
+                            template_str.push_str(name);
+                        }
+
+                        GlobalAsmOperandRef::SymStatic { def_id } => {
+                            // TODO(@Amanieu): Additional mangling is needed on
+                            // some targets to add a leading underscore (Mach-O).
+                            let instance = Instance::mono(self.tcx, def_id);
+                            let name = self.tcx.symbol_name(instance).name;
+                            template_str.push_str(name);
+                        }
                     }
                 }
             }