expand: Keep track of semicoloned builtin macros

This is quite a rough fix (like a lot of the macro expansion code...) but
it allows built-in macros to be treated as statements. I *think* asm!()
might be the only one where it really matters, but also doing something
like

{
    line!();
}

will now work, whereas before the macro invocation would not get expanded
properly and would be ignored.

gcc/rust/ChangeLog:

	* ast/rust-ast-fragment.h: Pass `is_semicolon` information to builtin
	macro transcribers.
	* ast/rust-macro.h: Pass semicolon information to transcriber..
	* expand/rust-macro-builtins-asm.cc (MacroBuiltin::asm_handler): Likewise.
	(parse_asm): Likewise.
	* expand/rust-macro-builtins-asm.h (parse_asm): Likewise.
	* expand/rust-macro-builtins-format-args.cc (MacroBuiltin::format_args_handler): Likewise.
	* expand/rust-macro-builtins-include.cc (MacroBuiltin::include_bytes_handler): Likewise.
	(MacroBuiltin::include_str_handler): Likewise.
	(MacroBuiltin::include_handler): Likewise.
	* expand/rust-macro-builtins-location.cc (MacroBuiltin::file_handler): Likewise.
	(MacroBuiltin::column_handler): Likewise.
	(MacroBuiltin::line_handler): Likewise.
	* expand/rust-macro-builtins-log-debug.cc (MacroBuiltin::assert_handler): Likewise.
	* expand/rust-macro-builtins-utility.cc (MacroBuiltin::compile_error_handler): Likewise.
	(MacroBuiltin::concat_handler): Likewise.
	(MacroBuiltin::env_handler): Likewise.
	(MacroBuiltin::cfg_handler): Likewise.
	(MacroBuiltin::stringify_handler): Likewise.
	* expand/rust-macro-builtins.cc (format_args_maker): Likewise.
	(inline_asm_maker): Likewise.
	(MacroBuiltin::sorry): Likewise.
	(MacroBuiltin::proc_macro_builtin): Likewise.
	* expand/rust-macro-builtins.h: Likewise.
	* expand/rust-macro-expand.cc (MacroExpander::expand_decl_macro): Likewise.
	(MacroExpander::expand_invoc): Likewise.
	* ast/rust-ast-visitor.cc (DefaultASTVisitor::visit): Remove rust_unreachable.
	* ast/rust-ast.cc (InlineAsm::accept_vis): Likewise.
	* hir/tree/rust-hir.cc (InlineAsm::accept_vis): Likewise.

gcc/rust/ast/rust-ast-fragment.h

@@ -128,7 +128,8 @@ class Fragment
  * rust-macro-builtins.{h,cc}.
  */
 using MacroTranscriberFunc
-  = std::function<tl::optional<Fragment> (location_t, MacroInvocData &)>;
+  = std::function<tl::optional<Fragment> (location_t, MacroInvocData &,
+					  bool semicolon)>;
 
 } // namespace AST
 } // namespace Rust

gcc/rust/ast/rust-ast-visitor.cc

@@ -664,9 +664,7 @@ DefaultASTVisitor::visit (AST::AsyncBlockExpr &expr)
 
 void
 DefaultASTVisitor::visit (AST::InlineAsm &expr)
-{
-  rust_unreachable ();
-}
+{}
 
 void
 DefaultASTVisitor::visit (AST::TypeParam &param)

gcc/rust/ast/rust-ast.cc

@@ -4659,7 +4659,6 @@ AsyncBlockExpr::accept_vis (ASTVisitor &vis)
 void
 InlineAsm::accept_vis (ASTVisitor &vis)
 {
-  rust_unreachable ();
   vis.visit (*this);
 }
 

gcc/rust/ast/rust-macro.h

@@ -482,7 +482,7 @@ class MacroRulesDefinition : public VisItem
    * should make use of the actual rules. If the macro is builtin, then another
    * associated transcriber should be used
    */
-  static Fragment dummy_builtin (location_t, MacroInvocData &)
+  static Fragment dummy_builtin (location_t, MacroInvocData &, bool)
   {
     rust_unreachable ();
     return Fragment::create_error ();

gcc/rust/expand/rust-macro-builtins-asm.cc

@@ -17,6 +17,8 @@
 // <http://www.gnu.org/licenses/>.
 
 #include "rust-macro-builtins-asm.h"
+#include "rust-ast.h"
+#include "rust-stmt.h"
 
 namespace Rust {
 std::map<AST::InlineAsmOption, std::string> InlineAsmOptionMap{
@@ -535,9 +537,9 @@ parse_format_string (InlineAsmContext &inline_asm_ctx)
 
 tl::optional<AST::Fragment>
 MacroBuiltin::asm_handler (location_t invoc_locus, AST::MacroInvocData &invoc,
-			   bool is_global_asm)
+			   bool semicolon, bool is_global_asm)
 {
-  return parse_asm (invoc_locus, invoc, is_global_asm);
+  return parse_asm (invoc_locus, invoc, is_global_asm, semicolon);
 }
 
 tl::expected<InlineAsmContext, std::string>
@@ -607,7 +609,7 @@ parse_asm_arg (InlineAsmContext inline_asm_ctx)
 
 tl::optional<AST::Fragment>
 parse_asm (location_t invoc_locus, AST::MacroInvocData &invoc,
-	   bool is_global_asm)
+	   bool is_global_asm, bool semicolon)
 {
   // From the rule of asm.
   // We first peek and see if it is a format string or not.
@@ -646,9 +648,18 @@ parse_asm (location_t invoc_locus, AST::MacroInvocData &invoc,
 
   if (is_valid)
     {
-      AST::SingleASTNode single = AST::SingleASTNode (
-	inline_asm_ctx.inline_asm.clone_expr_without_block ());
-      std::vector<AST::SingleASTNode> single_vec = {single};
+      auto node = inline_asm_ctx.inline_asm.clone_expr_without_block ();
+
+      std::vector<AST::SingleASTNode> single_vec = {};
+
+      // If the macro invocation has a semicolon (`asm!("...");`), then we need
+      // to make it a statement. This way, it will be expanded properly.
+      if (semicolon)
+	single_vec.emplace_back (
+	  AST::SingleASTNode (std::unique_ptr<AST::Stmt> (
+	    new AST::ExprStmt (std::move (node), invoc_locus, semicolon))));
+      else
+	single_vec.emplace_back (AST::SingleASTNode (std::move (node)));
 
       AST::Fragment fragment_ast
 	= AST::Fragment (single_vec,

gcc/rust/expand/rust-macro-builtins-asm.h

@@ -59,7 +59,7 @@ parse_reg_operand (InlineAsmContext inline_asm_ctx);
 
 tl::optional<AST::Fragment>
 parse_asm (location_t invoc_locus, AST::MacroInvocData &invoc,
-	   bool is_global_asm);
+	   bool is_global_asm, bool semicolon);
 
 bool
 check_identifier (Parser<MacroInvocLexer> &parser, std::string ident);

gcc/rust/expand/rust-macro-builtins-format-args.cc

@@ -115,7 +115,7 @@ format_args_parse_arguments (AST::MacroInvocData &invoc)
 
 tl::optional<AST::Fragment>
 MacroBuiltin::format_args_handler (location_t invoc_locus,
-				   AST::MacroInvocData &invoc,
+				   AST::MacroInvocData &invoc, bool semicolon,
 				   AST::FormatArgs::Newline nl)
 {
   auto input = format_args_parse_arguments (invoc);

gcc/rust/expand/rust-macro-builtins-include.cc

@@ -27,7 +27,7 @@ of the given file as reference to a byte array. Yields an expression of type
 
 tl::optional<AST::Fragment>
 MacroBuiltin::include_bytes_handler (location_t invoc_locus,
-				     AST::MacroInvocData &invoc)
+				     AST::MacroInvocData &invoc, bool semicolon)
 {
   /* Get target filename from the macro invocation, which is treated as a path
      relative to the include!-ing file (currently being compiled).  */
@@ -93,7 +93,7 @@ MacroBuiltin::include_bytes_handler (location_t invoc_locus,
 
 tl::optional<AST::Fragment>
 MacroBuiltin::include_str_handler (location_t invoc_locus,
-				   AST::MacroInvocData &invoc)
+				   AST::MacroInvocData &invoc, bool semicolon)
 {
   /* Get target filename from the macro invocation, which is treated as a path
      relative to the include!-ing file (currently being compiled).  */
@@ -182,7 +182,7 @@ scope compile time. */
 
 tl::optional<AST::Fragment>
 MacroBuiltin::include_handler (location_t invoc_locus,
-			       AST::MacroInvocData &invoc)
+			       AST::MacroInvocData &invoc, bool semicolon)
 {
   /* Get target filename from the macro invocation, which is treated as a path
      relative to the include!-ing file (currently being compiled).  */

gcc/rust/expand/rust-macro-builtins-location.cc

@@ -21,7 +21,7 @@
 
 namespace Rust {
 tl::optional<AST::Fragment>
-MacroBuiltin::file_handler (location_t invoc_locus, AST::MacroInvocData &)
+MacroBuiltin::file_handler (location_t invoc_locus, AST::MacroInvocData &, bool)
 {
   auto current_file = LOCATION_FILE (invoc_locus);
   auto file_str = AST::SingleASTNode (make_string (invoc_locus, current_file));
@@ -32,7 +32,8 @@ MacroBuiltin::file_handler (location_t invoc_locus, AST::MacroInvocData &)
 }
 
 tl::optional<AST::Fragment>
-MacroBuiltin::column_handler (location_t invoc_locus, AST::MacroInvocData &)
+MacroBuiltin::column_handler (location_t invoc_locus, AST::MacroInvocData &,
+			      bool)
 {
   auto current_column = LOCATION_COLUMN (invoc_locus);
 
@@ -46,7 +47,7 @@ MacroBuiltin::column_handler (location_t invoc_locus, AST::MacroInvocData &)
 }
 
 tl::optional<AST::Fragment>
-MacroBuiltin::line_handler (location_t invoc_locus, AST::MacroInvocData &)
+MacroBuiltin::line_handler (location_t invoc_locus, AST::MacroInvocData &, bool)
 {
   auto current_line = LOCATION_LINE (invoc_locus);
 

gcc/rust/expand/rust-macro-builtins-log-debug.cc

@@ -22,7 +22,7 @@
 namespace Rust {
 tl::optional<AST::Fragment>
 MacroBuiltin::assert_handler (location_t invoc_locus,
-			      AST::MacroInvocData &invoc)
+			      AST::MacroInvocData &invoc, bool semicolon)
 {
   rust_debug ("assert!() called");
 

gcc/rust/expand/rust-macro-builtins-utility.cc

@@ -26,7 +26,7 @@ namespace Rust {
    during the compile time. */
 tl::optional<AST::Fragment>
 MacroBuiltin::compile_error_handler (location_t invoc_locus,
-				     AST::MacroInvocData &invoc)
+				     AST::MacroInvocData &invoc, bool semicolon)
 {
   auto lit_expr
     = parse_single_string_literal (BuiltinMacro::CompileError,
@@ -87,7 +87,7 @@ MacroBuiltin::compile_error_handler (location_t invoc_locus,
 // Can we do that easily?
 tl::optional<AST::Fragment>
 MacroBuiltin::concat_handler (location_t invoc_locus,
-			      AST::MacroInvocData &invoc)
+			      AST::MacroInvocData &invoc, bool semicolon)
 {
   auto invoc_token_tree = invoc.get_delim_tok_tree ();
   MacroInvocLexer lex (invoc_token_tree.to_token_stream ());
@@ -151,7 +151,8 @@ MacroBuiltin::concat_handler (location_t invoc_locus,
 /* Expand builtin macro env!(), which inspects an environment variable at
    compile time. */
 tl::optional<AST::Fragment>
-MacroBuiltin::env_handler (location_t invoc_locus, AST::MacroInvocData &invoc)
+MacroBuiltin::env_handler (location_t invoc_locus, AST::MacroInvocData &invoc,
+			   bool semicolon)
 {
   auto invoc_token_tree = invoc.get_delim_tok_tree ();
   MacroInvocLexer lex (invoc_token_tree.to_token_stream ());
@@ -224,7 +225,8 @@ MacroBuiltin::env_handler (location_t invoc_locus, AST::MacroInvocData &invoc)
 }
 
 tl::optional<AST::Fragment>
-MacroBuiltin::cfg_handler (location_t invoc_locus, AST::MacroInvocData &invoc)
+MacroBuiltin::cfg_handler (location_t invoc_locus, AST::MacroInvocData &invoc,
+			   bool semicolon)
 {
   // only parse if not already parsed
   if (!invoc.is_parsed ())
@@ -263,7 +265,7 @@ MacroBuiltin::cfg_handler (location_t invoc_locus, AST::MacroInvocData &invoc)
 
 tl::optional<AST::Fragment>
 MacroBuiltin::stringify_handler (location_t invoc_locus,
-				 AST::MacroInvocData &invoc)
+				 AST::MacroInvocData &invoc, bool semicolon)
 {
   std::string content;
   auto invoc_token_tree = invoc.get_delim_tok_tree ();

gcc/rust/expand/rust-macro-builtins.cc

@@ -87,8 +87,8 @@ const BiMap<std::string, BuiltinMacro> MacroBuiltin::builtins = {{
 AST::MacroTranscriberFunc
 format_args_maker (AST::FormatArgs::Newline nl)
 {
-  return [nl] (location_t loc, AST::MacroInvocData &invoc) {
-    return MacroBuiltin::format_args_handler (loc, invoc, nl);
+  return [nl] (location_t loc, AST::MacroInvocData &invoc, bool semicolon) {
+    return MacroBuiltin::format_args_handler (loc, invoc, semicolon, nl);
   };
 }
 
@@ -103,9 +103,10 @@ inline_asm_maker (isGlobalAsm is_global_asm)
 {
   bool global_asm = is_global_asm == isGlobalAsm::Global ? true : false;
 
-  return [global_asm] (location_t loc, AST::MacroInvocData &invoc) {
-    return MacroBuiltin::asm_handler (loc, invoc, global_asm);
-  };
+  return
+    [global_asm] (location_t loc, AST::MacroInvocData &invoc, bool semicolon) {
+      return MacroBuiltin::asm_handler (loc, invoc, semicolon, global_asm);
+    };
 }
 
 std::unordered_map<std::string, AST::MacroTranscriberFunc>
@@ -167,7 +168,8 @@ builtin_macro_from_string (const std::string &identifier)
 }
 
 tl::optional<AST::Fragment>
-MacroBuiltin::sorry (location_t invoc_locus, AST::MacroInvocData &invoc)
+MacroBuiltin::sorry (location_t invoc_locus, AST::MacroInvocData &invoc,
+		     bool semicolon)
 {
   rust_sorry_at (invoc_locus, "unimplemented builtin macro: %qs",
 		 invoc.get_path ().as_string ().c_str ());
@@ -177,7 +179,7 @@ MacroBuiltin::sorry (location_t invoc_locus, AST::MacroInvocData &invoc)
 
 tl::optional<AST::Fragment>
 MacroBuiltin::proc_macro_builtin (location_t invoc_locus,
-				  AST::MacroInvocData &invoc)
+				  AST::MacroInvocData &invoc, bool semicolon)
 {
   rust_error_at (invoc_locus, "cannot invoke derive macro: %qs",
 		 invoc.get_path ().as_string ().c_str ());

gcc/rust/expand/rust-macro-builtins.h

@@ -123,57 +123,70 @@ class MacroBuiltin
   static std::unordered_map<std::string, AST::MacroTranscriberFunc>
     builtin_transcribers;
 
-  static tl::optional<AST::Fragment>
-  assert_handler (location_t invoc_locus, AST::MacroInvocData &invoc);
+  static tl::optional<AST::Fragment> assert_handler (location_t invoc_locus,
+						     AST::MacroInvocData &invoc,
+						     bool semicolon);
 
   static tl::optional<AST::Fragment> file_handler (location_t invoc_locus,
-						   AST::MacroInvocData &invoc);
+						   AST::MacroInvocData &invoc,
+						   bool semicolon);
 
-  static tl::optional<AST::Fragment>
-  column_handler (location_t invoc_locus, AST::MacroInvocData &invoc);
+  static tl::optional<AST::Fragment> column_handler (location_t invoc_locus,
+						     AST::MacroInvocData &invoc,
+						     bool semicolon);
 
   static tl::optional<AST::Fragment>
-  include_bytes_handler (location_t invoc_locus, AST::MacroInvocData &invoc);
+  include_bytes_handler (location_t invoc_locus, AST::MacroInvocData &invoc,
+			 bool semicolon);
 
   static tl::optional<AST::Fragment>
-  include_str_handler (location_t invoc_locus, AST::MacroInvocData &invoc);
+  include_str_handler (location_t invoc_locus, AST::MacroInvocData &invoc,
+		       bool semicolon);
 
   static tl::optional<AST::Fragment>
-  stringify_handler (location_t invoc_locus, AST::MacroInvocData &invoc);
+  stringify_handler (location_t invoc_locus, AST::MacroInvocData &invoc,
+		     bool semicolon);
 
   static tl::optional<AST::Fragment>
-  compile_error_handler (location_t invoc_locus, AST::MacroInvocData &invoc);
+  compile_error_handler (location_t invoc_locus, AST::MacroInvocData &invoc,
+			 bool semicolon);
 
-  static tl::optional<AST::Fragment>
-  concat_handler (location_t invoc_locus, AST::MacroInvocData &invoc);
+  static tl::optional<AST::Fragment> concat_handler (location_t invoc_locus,
+						     AST::MacroInvocData &invoc,
+						     bool semicolon);
 
   static tl::optional<AST::Fragment> env_handler (location_t invoc_locus,
-						  AST::MacroInvocData &invoc);
+						  AST::MacroInvocData &invoc,
+						  bool semicolon);
 
   static tl::optional<AST::Fragment> cfg_handler (location_t invoc_locus,
-						  AST::MacroInvocData &invoc);
+						  AST::MacroInvocData &invoc,
+						  bool semicolon);
 
   static tl::optional<AST::Fragment>
-  include_handler (location_t invoc_locus, AST::MacroInvocData &invoc);
+  include_handler (location_t invoc_locus, AST::MacroInvocData &invoc,
+		   bool semicolon);
 
   static tl::optional<AST::Fragment> line_handler (location_t invoc_locus,
-						   AST::MacroInvocData &invoc);
+						   AST::MacroInvocData &invoc,
+						   bool semicolon);
 
   static tl::optional<AST::Fragment> asm_handler (location_t invoc_locus,
 						  AST::MacroInvocData &invoc,
+						  bool semicolon,
 						  bool is_global_asm);
 
   static tl::optional<AST::Fragment>
   format_args_handler (location_t invoc_locus, AST::MacroInvocData &invoc,
-		       AST::FormatArgs::Newline nl);
+		       bool semicolon, AST::FormatArgs::Newline nl);
 
-  static tl::optional<AST::Fragment> sorry (location_t invoc_locus,
-					    AST::MacroInvocData &invoc);
+  static tl::optional<AST::Fragment>
+  sorry (location_t invoc_locus, AST::MacroInvocData &invoc, bool semicolon);
 
   /* Builtin procedural macros do not work directly on tokens, but still need a
    * builtin transcriber to be considered proper builtin macros */
-  static tl::optional<AST::Fragment> proc_macro_builtin (location_t,
-							 AST::MacroInvocData &);
+  static tl::optional<AST::Fragment>
+  proc_macro_builtin (location_t, AST::MacroInvocData &, bool);
 };
 } // namespace Rust
 

gcc/rust/expand/rust-macro-expand.cc

@@ -43,9 +43,10 @@ MacroExpander::expand_decl_macro (location_t invoc_locus,
 
   /* probably something here about parsing invoc and rules def token trees to
    * token stream. if not, how would parser handle the captures of exprs and
-   * stuff? on the other hand, token trees may be kind of useful in rules def as
-   * creating a point where recursion can occur (like having
-   * "compare_macro_match" and then it calling itself when it finds delimiters)
+   * stuff? on the other hand, token trees may be kind of useful in rules def
+   * as creating a point where recursion can occur (like having
+   * "compare_macro_match" and then it calling itself when it finds
+   * delimiters)
    */
 
   /* find matching rule to invoc token tree, based on macro rule's matcher. if
@@ -70,9 +71,10 @@ MacroExpander::expand_decl_macro (location_t invoc_locus,
 
   /* TODO: it is probably better to modify AST::Token to store a pointer to a
    * Lexer::Token (rather than being converted) - i.e. not so much have
-   * AST::Token as a Token but rather a TokenContainer (as it is another type of
-   * TokenTree). This will prevent re-conversion of Tokens between each type
-   * all the time, while still allowing the heterogenous storage of token trees.
+   * AST::Token as a Token but rather a TokenContainer (as it is another type
+   * of TokenTree). This will prevent re-conversion of Tokens between each
+   * type all the time, while still allowing the heterogenous storage of token
+   * trees.
    */
 
   AST::DelimTokenTree &invoc_token_tree = invoc.get_delim_tok_tree ();
@@ -286,8 +288,12 @@ MacroExpander::expand_invoc (AST::MacroInvocation &invoc, bool has_semicolon)
   last_invoc = *invoc.clone_macro_invocation_impl ();
   last_def = *rdef;
 
+  rust_debug ("[ARTHUR] semicolon: %s", has_semicolon ? "yes" : "no");
+
   if (rdef->is_builtin ())
-    fragment = rdef->get_builtin_transcriber () (invoc.get_locus (), invoc_data)
+    fragment = rdef
+		 ->get_builtin_transcriber () (invoc.get_locus (), invoc_data,
+					       has_semicolon)
 		 .value_or (AST::Fragment::create_empty ());
   else
     fragment = expand_decl_macro (invoc.get_locus (), invoc_data, *rdef,