bpo-46110: Add a recursion check to avoid stack overflow in the PEG parser (GH-30177)

Co-authored-by: Batuhan Taskaya <isidentical@gmail.com>

Lib/test/test_syntax.py

@@ -1729,6 +1729,14 @@ while 1:
 """
         self._check_error(source, "too many statically nested blocks")
 
+    @support.cpython_only
+    def test_error_on_parser_stack_overflow(self):
+        source = "-" * 100000 + "4"
+        for mode in ["exec", "eval", "single"]:
+            with self.subTest(mode=mode):
+                with self.assertRaises(MemoryError):
+                    compile(source, "<string>", mode)
+
 
 def load_tests(loader, tests, pattern):
     tests.addTest(doctest.DocTestSuite())

Misc/NEWS.d/next/Core and Builtins/2021-12-18-02-37-07.bpo-46110.B6hAfu.rst

@@ -0,0 +1,2 @@
+Add a maximum recursion check to the PEG parser to avoid stack overflow.
+Patch by Pablo Galindo

Parser/pegen.c

@@ -815,6 +815,7 @@ void *
 _PyPegen_run_parser(Parser *p)
 {
     void *res = _PyPegen_parse(p);
+    assert(p->level == 0);
     if (res == NULL) {
         if (PyErr_Occurred() && !PyErr_ExceptionMatches(PyExc_SyntaxError)) {
             return NULL;

Tools/peg_generator/pegen/c_generator.py

@@ -37,6 +37,8 @@ EXTENSION_PREFIX = """\
 #  define D(x)
 #endif
 
+# define MAXSTACK 6000
+
 """
 
 
@@ -364,10 +366,14 @@ class CParserGenerator(ParserGenerator, GrammarVisitor):
         self.skip_actions = skip_actions
 
     def add_level(self) -> None:
-        self.print("D(p->level++);")
+        self.print("if (p->level++ == MAXSTACK) {")
+        with self.indent():
+            self.print("p->error_indicator = 1;")
+            self.print("PyErr_NoMemory();")
+        self.print("}")
 
     def remove_level(self) -> None:
-        self.print("D(p->level--);")
+        self.print("p->level--;")
 
     def add_return(self, ret_val: str) -> None:
         self.remove_level()
@@ -544,9 +550,10 @@ class CParserGenerator(ParserGenerator, GrammarVisitor):
                 self.print("p->in_raw_rule++;")
                 self.print(f"void *_raw = {node.name}_raw(p);")
                 self.print("p->in_raw_rule--;")
-                self.print("if (p->error_indicator)")
+                self.print("if (p->error_indicator) {")
                 with self.indent():
-                    self.print("return NULL;")
+                    self.add_return("NULL")
+                self.print("}")
                 self.print("if (_raw == NULL || p->mark <= _resmark)")
                 with self.indent():
                     self.print("break;")