GH-132554: Add stats for GET_ITER (GH-132592)

markshannon · web-flow · commit 622300bdfa62 · 2025-04-29T09:00:14.000+01:00
* Add stats for GET_ITER

* Look for common iterable types, not iterator types

* Add stats for self iter and fix naming in summary
diff --git a/Include/internal/pycore_code.h b/Include/internal/pycore_code.h
@@ -317,6 +317,7 @@ extern void _Py_Specialize_ForIter(_PyStackRef iter, _Py_CODEUNIT *instr, int op
 extern void _Py_Specialize_Send(_PyStackRef receiver, _Py_CODEUNIT *instr);
 extern void _Py_Specialize_ToBool(_PyStackRef value, _Py_CODEUNIT *instr);
 extern void _Py_Specialize_ContainsOp(_PyStackRef value, _Py_CODEUNIT *instr);
+extern void _Py_GatherStats_GetIter(_PyStackRef iterable);
 
 // Utility functions for reading/writing 32/64-bit values in the inline caches.
 // Great care should be taken to ensure that these functions remain correct and
diff --git a/Python/bytecodes.c b/Python/bytecodes.c
@@ -3029,6 +3029,9 @@ dummy_func(
         }
 
         inst(GET_ITER, (iterable -- iter)) {
+            #ifdef Py_STATS
+            _Py_GatherStats_GetIter(iterable);
+            #endif
             /* before: [obj]; after [getiter(obj)] */
             PyObject *iter_o = PyObject_GetIter(PyStackRef_AsPyObjectBorrow(iterable));
             PyStackRef_CLOSE(iterable);
diff --git a/Python/executor_cases.c.h b/Python/executor_cases.c.h
diff --git a/Python/generated_cases.c.h b/Python/generated_cases.c.h
diff --git a/Python/specialize.c b/Python/specialize.c
@@ -146,6 +146,7 @@ print_spec_stats(FILE *out, OpcodeStats *stats)
      * even though we don't specialize them yet. */
     fprintf(out, "opcode[BINARY_SLICE].specializable : 1\n");
     fprintf(out, "opcode[STORE_SLICE].specializable : 1\n");
+    fprintf(out, "opcode[GET_ITER].specializable : 1\n");
     for (int i = 0; i < 256; i++) {
         if (_PyOpcode_Caches[i]) {
             /* Ignore jumps as they cannot be specialized */
@@ -668,6 +669,7 @@ _PyCode_Quicken(_Py_CODEUNIT *instructions, Py_ssize_t size, int enable_counters
 #define SPEC_FAIL_ITER_CALLABLE 28
 #define SPEC_FAIL_ITER_ASCII_STRING 29
 #define SPEC_FAIL_ITER_ASYNC_GENERATOR_SEND 30
+#define SPEC_FAIL_ITER_SELF 31
 
 // UNPACK_SEQUENCE
 
@@ -3115,6 +3117,53 @@ _Py_Specialize_ContainsOp(_PyStackRef value_st, _Py_CODEUNIT *instr)
     return;
 }
 
+#ifdef Py_STATS
+void
+_Py_GatherStats_GetIter(_PyStackRef iterable)
+{
+    PyTypeObject *tp = PyStackRef_TYPE(iterable);
+    int kind = SPEC_FAIL_OTHER;
+    if (tp == &PyTuple_Type) {
+        kind = SPEC_FAIL_ITER_TUPLE;
+    }
+    else if (tp == &PyList_Type) {
+        kind = SPEC_FAIL_ITER_LIST;
+    }
+    else if (tp == &PyDict_Type) {
+        kind = SPEC_FAIL_ITER_DICT_KEYS;
+    }
+    else if (tp == &PySet_Type) {
+        kind = SPEC_FAIL_ITER_SET;
+    }
+    else if (tp == &PyBytes_Type) {
+        kind = SPEC_FAIL_ITER_BYTES;
+    }
+    else if (tp == &PyEnum_Type) {
+        kind = SPEC_FAIL_ITER_ENUMERATE;
+    }
+    else if (tp == &PyUnicode_Type) {
+        kind = SPEC_FAIL_ITER_STRING;
+    }
+    else if (tp == &PyGen_Type) {
+        kind = SPEC_FAIL_ITER_GENERATOR;
+    }
+    else if (tp == &PyCoro_Type) {
+        kind = SPEC_FAIL_ITER_COROUTINE;
+    }
+    else if (tp == &PyAsyncGen_Type) {
+        kind = SPEC_FAIL_ITER_ASYNC_GENERATOR;
+    }
+    else if (tp == &_PyAsyncGenASend_Type) {
+        kind = SPEC_FAIL_ITER_ASYNC_GENERATOR_SEND;
+    }
+    else if (tp->tp_iter == PyObject_SelfIter) {
+        kind = SPEC_FAIL_ITER_SELF;
+    }
+    SPECIALIZATION_FAIL(GET_ITER, kind);
+}
+#endif
+
+
 /* Code init cleanup.
  * CALL_ALLOC_AND_ENTER_INIT will set up
  * the frame to execute the EXIT_INIT_CHECK
diff --git a/Tools/scripts/summarize_stats.py b/Tools/scripts/summarize_stats.py
@@ -288,7 +288,7 @@ def kind_to_text(kind: int, opcode: str):
                 opcode = "SUPER"
             elif opcode.endswith("ATTR"):
                 opcode = "ATTR"
-            elif opcode in ("FOR_ITER", "SEND"):
+            elif opcode in ("FOR_ITER", "GET_ITER", "SEND"):
                 opcode = "ITER"
             elif opcode.endswith("SUBSCR"):
                 opcode = "SUBSCR"

Original file line number	Diff line number	Diff line change
`@@ -3029,6 +3029,9 @@ dummy_func(`
`3029`	`3029`	`}`
`3030`	`3030`
`3031`	`3031`	`inst(GET_ITER, (iterable -- iter)) {`
	`3032`	`+ #ifdef Py_STATS`
	`3033`	`+ _Py_GatherStats_GetIter(iterable);`
	`3034`	`+ #endif`
`3032`	`3035`	`/* before: [obj]; after [getiter(obj)] */`
`3033`	`3036`	`PyObject *iter_o = PyObject_GetIter(PyStackRef_AsPyObjectBorrow(iterable));`
`3034`	`3037`	`PyStackRef_CLOSE(iterable);`