fix(internal): closure bytecode wrapping for Python 3.11 [backport 2.…

…0] (#7052)

Backport 9bd1fb9 from #7044 to 2.0.

We fix the bytecode wrapping for closures to ensure that the frame
objects don't end up referencing missing data, ultimately resulting in a
segmentation fault in profiling tools.

Fixes #6701.

## Checklist

- [x] Change(s) are motivated and described in the PR description.
- [x] Testing strategy is described if automated tests are not included
in the PR.
- [x] Risk is outlined (performance impact, potential for breakage,
maintainability, etc).
- [x] Change is maintainable (easy to change, telemetry, documentation).
- [x] [Library release note
guidelines](https://ddtrace.readthedocs.io/en/stable/releasenotes.html)
are followed. If no release note is required, add label
`changelog/no-changelog`.
- [x] Documentation is included (in-code, generated user docs, [public
corp docs](https://github.com/DataDog/documentation/)).
- [x] Backport labels are set (if
[applicable](https://ddtrace.readthedocs.io/en/latest/contributing.html#backporting))

## Reviewer Checklist

- [x] Title is accurate.
- [x] No unnecessary changes are introduced.
- [x] Description motivates each change.
- [x] Avoids breaking
[API](https://ddtrace.readthedocs.io/en/stable/versioning.html#interfaces)
changes unless absolutely necessary.
- [x] Testing strategy adequately addresses listed risk(s).
- [x] Change is maintainable (easy to change, telemetry, documentation).
- [x] Release note makes sense to a user of the library.
- [x] Reviewer has explicitly acknowledged and discussed the performance
implications of this PR as reported in the benchmarks PR comment.
- [x] Backport labels are set in a manner that is consistent with the
[release branch maintenance
policy](https://ddtrace.readthedocs.io/en/latest/contributing.html#backporting)
- [x] If this PR touches code that signs or publishes builds or
packages, or handles credentials of any kind, I've requested a review
from `@DataDog/security-design-and-guidance`.
- [x] This PR doesn't touch any of that.

Co-authored-by: Gabriele N. Tornetta <P403n1x87@users.noreply.github.com>

ddtrace/internal/wrapping/__init__.py

@@ -130,15 +130,30 @@ def wrap_bytecode(wrapper, wrapped):
  Instr("LOAD_CONST", wrapped, lineno=lineno),
  ]
  if PY >= (3, 11):
- # THis is required to start a new frame
+ # From insert_prefix_instructions
  instrs[0:0] = [
  Instr("RESUME", 0, lineno=lineno - 1),
  Instr("PUSH_NULL", lineno=lineno),
  ]
 
+ if code.co_cellvars:
+ from bytecode import CellVar
+
+ instrs[0:0] = [Instr("MAKE_CELL", CellVar(_), lineno=lineno) for _ in code.co_cellvars]
+
+ if code.co_freevars:
+ instrs.insert(0, Instr("COPY_FREE_VARS", len(code.co_freevars), lineno=lineno))
+
  # Build the tuple of all the positional arguments
  if nargs:
- instrs.extend([Instr("LOAD_FAST", argname, lineno=lineno) for argname in argnames])
+ instrs.extend(
+ [
+ Instr("LOAD_DEREF", CellVar(argname), lineno=lineno)
+ if PY >= (3, 11) and argname in code.co_cellvars
+ else Instr("LOAD_FAST", argname, lineno=lineno)
+ for argname in argnames
+ ]
+ )
  instrs.append(Instr("BUILD_TUPLE", nargs, lineno=lineno))
  if varargs:
  instrs.extend(
@@ -213,6 +228,8 @@ def wrap(f, wrapper):
 
  code = wrap_bytecode(wrapper, wrapped)
  code.freevars = f.__code__.co_freevars
+ if PY >= (3, 11):
+ code.cellvars = f.__code__.co_cellvars
  code.name = f.__code__.co_name
  code.filename = f.__code__.co_filename
  code.flags = f.__code__.co_flags

releasenotes/notes/fix-py311-closure-wrapping-fed33b584d65a844.yaml

@@ -0,0 +1,6 @@
+---
+fixes:
+ - |
+ Fix an issue that could have caused some tracing integrations to create
+ invalid references to objects in Python frames, ultimately causing profiling
+ tools to potentially induce a segmentation fault.

tests/internal/test_wrapping.py

@@ -467,3 +467,27 @@ async def g():
  await gen.aclose()
 
  assert channel == [True] + [0, ValueError, 1] * 10 + ["GeneratorExit"]
+
+
+def test_wrap_closure():
+ channel = []
+
+ def wrapper(f, args, kwargs):
+ channel.append((args, kwargs))
+ retval = f(*args, **kwargs)
+ channel.append(retval)
+ return retval
+
+ def outer(answer=42):
+ def f(a, b, c=None):
+ return (a, b, c, answer)
+
+ return f
+
+ wrap(outer, wrapper)
+
+ closure = outer()
+ wrap(closure, wrapper)
+
+ assert closure(1, 2, 3) == (1, 2, 3, 42)
+ assert channel == [((42,), {}), closure, ((1, 2, 3), {}), (1, 2, 3, 42)]