From 75b54da2f30b7ffa0c69c1659e3f3538bfdbd339 Mon Sep 17 00:00:00 2001
From: Thomas Schwinge <thomas@codesourcery.com>
Date: Tue, 4 Jul 2023 14:07:47 +0200
Subject: [PATCH] [RFC] Verify no embedded NULs in 'struct ht_identifier'

---
 gcc/analyzer/pending-diagnostic.cc | 2 +-
 gcc/c-family/c-spellcheck.h        | 2 +-
 gcc/c/c-decl.cc                    | 2 +-
 gcc/tree.h                         | 2 +-
 libcpp/include/symtab.h            | 4 ++--
 libcpp/symtab.cc                   | 2 +-
 6 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/gcc/analyzer/pending-diagnostic.cc b/gcc/analyzer/pending-diagnostic.cc
index e36ed4fd9c1..9452719a908 100644
--- a/gcc/analyzer/pending-diagnostic.cc
+++ b/gcc/analyzer/pending-diagnostic.cc
@@ -128,7 +128,7 @@ pending_diagnostic::same_tree_p (tree t1, tree t2)
 static bool
 ht_ident_eq (ht_identifier ident, const char *str)
 {
-  return (strlen (str) == ident.len
+  return (strlen (str) == HT_LEN (&ident)
 	  && 0 == strcmp (str, (const char *)ident.str));
 }
 
diff --git a/gcc/c-family/c-spellcheck.h b/gcc/c-family/c-spellcheck.h
index e3748430b18..0f0c264ef01 100644
--- a/gcc/c-family/c-spellcheck.h
+++ b/gcc/c-family/c-spellcheck.h
@@ -31,7 +31,7 @@ struct edit_distance_traits<cpp_hashnode *>
 {
   static size_t get_length (cpp_hashnode *hashnode)
   {
-    return hashnode->ident.len;
+    return HT_LEN (&hashnode->ident);
   }
 
   static const char *get_string (cpp_hashnode *hashnode)
diff --git a/gcc/c/c-decl.cc b/gcc/c/c-decl.cc
index 1af51c4acfc..c67ae5be514 100644
--- a/gcc/c/c-decl.cc
+++ b/gcc/c/c-decl.cc
@@ -4499,7 +4499,7 @@ lookup_name_fuzzy (tree name, enum lookup_name_fuzzy_kind kind, location_t loc)
     {
       const char *id = (const char *)best_macro->ident.str;
       tree macro_as_identifier
-	= get_identifier_with_length (id, best_macro->ident.len);
+	= get_identifier_with_length (id, HT_LEN (&best_macro->ident));
       bm.set_best_so_far (macro_as_identifier,
 			  bmm.get_best_distance (),
 			  bmm.get_best_candidate_length ());
diff --git a/gcc/tree.h b/gcc/tree.h
index 1854fe4a7d4..20169e7a467 100644
--- a/gcc/tree.h
+++ b/gcc/tree.h
@@ -1174,7 +1174,7 @@ extern void omp_clause_range_check_failed (const_tree, const char *, int,
 
 /* Unlike STRING_CST, in C terms this is strlen, not sizeof.  */
 #define IDENTIFIER_LENGTH(NODE) \
-  (IDENTIFIER_NODE_CHECK (NODE)->identifier.id.len)
+  (HT_LEN (&IDENTIFIER_NODE_CHECK (NODE)->identifier.id))
 #define IDENTIFIER_POINTER(NODE) \
   ((const char *) IDENTIFIER_NODE_CHECK (NODE)->identifier.id.str)
 #define IDENTIFIER_HASH_VALUE(NODE) \
diff --git a/libcpp/include/symtab.h b/libcpp/include/symtab.h
index c7ccc6db9f0..ea23e3dbedb 100644
--- a/libcpp/include/symtab.h
+++ b/libcpp/include/symtab.h
@@ -33,11 +33,11 @@ struct GTY(()) ht_identifier {
      stream to PCH correctly, if a null byte were to make its way into an
      identifier somehow.  */
   const unsigned char * GTY((string_length ("1 + %h.len"))) str;
-  unsigned int len;
+  unsigned int len_;
   unsigned int hash_value;
 };
 
-#define HT_LEN(NODE) ((NODE)->len)
+#define HT_LEN(NODE) ({ gcc_assert ((NODE)->len_ == strlen ((const char *) HT_STR (NODE))); (NODE)->len_; })
 #define HT_STR(NODE) ((NODE)->str)
 
 typedef struct ht cpp_hash_table;
diff --git a/libcpp/symtab.cc b/libcpp/symtab.cc
index ea9f26905a9..be5c1da9d67 100644
--- a/libcpp/symtab.cc
+++ b/libcpp/symtab.cc
@@ -155,7 +155,7 @@ ht_lookup_with_hash (cpp_hash_table *table, const unsigned char *str,
   node = (*table->alloc_node) (table);
   table->entries[index] = node;
 
-  HT_LEN (node) = (unsigned int) len;
+  node->len_ = (unsigned int) len;
   node->hash_value = hash;
 
   if (table->alloc_subobject)
-- 
2.34.1