GH-127953: Make line number lookup O(1) regardless of the size of the code object (GH-128350)

markshannon · web-flow · commit 7239da755920 · 2025-01-21T09:33:23.000Z
diff --git a/Include/cpython/code.h b/Include/cpython/code.h
@@ -35,11 +35,12 @@ typedef struct {
 } _PyCoCached;
 
 /* Ancillary data structure used for instrumentation.
-   Line instrumentation creates an array of
-   these. One entry per code unit.*/
+   Line instrumentation creates this with sufficient
+   space for one entry per code unit. The total size
+   of the data will be `bytes_per_entry * Py_SIZE(code)` */
 typedef struct {
-    uint8_t original_opcode;
-    int8_t line_delta;
+    uint8_t bytes_per_entry;
+    uint8_t data[1];
 } _PyCoLineInstrumentationData;
 
 
diff --git a/Misc/NEWS.d/next/Core_and_Builtins/2024-12-30-15-49-31.gh-issue-127953.B4_6L9.rst b/Misc/NEWS.d/next/Core_and_Builtins/2024-12-30-15-49-31.gh-issue-127953.B4_6L9.rst
@@ -0,0 +1,2 @@
+The time to handle a ``LINE`` event in sys.monitoring (and sys.settrace) is
+now independent of the number of lines in the code object.
diff --git a/Objects/codeobject.c b/Objects/codeobject.c
@@ -979,6 +979,9 @@ PyCode_Addr2Line(PyCodeObject *co, int addrq)
     if (addrq < 0) {
         return co->co_firstlineno;
     }
+    if (co->_co_monitoring && co->_co_monitoring->lines) {
+        return _Py_Instrumentation_GetLine(co, addrq/sizeof(_Py_CODEUNIT));
+    }
     assert(addrq >= 0 && addrq < _PyCode_NBYTES(co));
     PyCodeAddressRange bounds;
     _PyCode_InitAddressRange(co, &bounds);
diff --git a/Python/bytecodes.c b/Python/bytecodes.c
@@ -4816,7 +4816,8 @@ dummy_func(
             int original_opcode = 0;
             if (tstate->tracing) {
                 PyCodeObject *code = _PyFrame_GetCode(frame);
-                original_opcode = code->_co_monitoring->lines[(int)(this_instr - _PyFrame_GetBytecode(frame))].original_opcode;
+                int index = (int)(this_instr - _PyFrame_GetBytecode(frame));
+                original_opcode = code->_co_monitoring->lines->data[index*code->_co_monitoring->lines->bytes_per_entry];
                 next_instr = this_instr;
             } else {
                 original_opcode = _Py_call_instrumentation_line(
diff --git a/Python/generated_cases.c.h b/Python/generated_cases.c.h
diff --git a/Python/instrumentation.c b/Python/instrumentation.c

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,2 @@`
	`1`	+The time to handle a ``LINE`` event in sys.monitoring (and sys.settrace) is
	`2`	`+now independent of the number of lines in the code object.`