gh-115832: Fix instrumentation version mismatch during interpreter shutdown (#115856)

A previous commit introduced a bug to `interpreter_clear()`: it set
`interp->ceval.instrumentation_version` to 0, without making the corresponding
change to `tstate->eval_breaker` (which holds a thread-local copy of the
version). After this happens, Python code can still run due to object finalizers
during a GC, and the version check in bytecodes.c will see a different result
than the one in instrumentation.c causing an infinite loop.

The fix itself is straightforward: clear `tstate->eval_breaker` when clearing
`interp->ceval.instrumentation_version`.
This commit is contained in:
Brett Simmers 2024-03-04 08:29:39 -08:00 committed by GitHub
parent 15dc2979bc
commit 0adfa8482d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 53 additions and 2 deletions

View file

@ -795,7 +795,10 @@ interpreter_clear(PyInterpreterState *interp, PyThreadState *tstate)
Py_CLEAR(interp->audit_hooks);
// At this time, all the threads should be cleared so we don't need atomic
// operations for instrumentation_version or eval_breaker.
interp->ceval.instrumentation_version = 0;
tstate->eval_breaker = 0;
for (int i = 0; i < _PY_MONITORING_UNGROUPED_EVENTS; i++) {
interp->monitors.tools[i] = 0;