test_track_literals depends on a garbage collection

[serhiy-storchaka]

Bug report

_not_tracked_instantly() requires the tuple to be untracked without a collection. This holds for tuples created at run time, but tuple constants are created tracked and are only untracked by _PyTuple_MaybeUntrack(), which is called only from the collector.

$ ./python -c "import gc; print(gc.is_tracked((1,)))"
True

So test_track_literals passes only if a collection happens between importing the module and running the test. Under regrtest it does not:

$ ./python -m test test_tuple -m test_track_literals
AssertionError: True is not false : (1,)

while ./python -m unittest test.test_tuple.TupleTest.test_track_literals passes.

Untracking tuple constants when the code object is created would let the test keep the stronger check.

Linked PRs

• GH-155728: Untrack const tuples #156419
• GH-155728: Untrack tuples while unmarshalling  #156423

[serhiy-storchaka]

The failing test assertion were added in #140575. cc @vstinner, @efimov-mikhail.

[sergey-miryanov]

Do you mind I will take this?

[serhiy-storchaka]

You are welcome.

[vstinner]

Maybe (1,) can be rewritten as x=1; (x,) to avoid storing the tuple as a constant.

while ./python -m unittest test.test_tuple.TupleTest.test_track_literals passes.

I understand that this issue is more theorical since it's not possible to get a failure when running test_tuple.

[serhiy-storchaka]

You can get failure by running ./python -m test test_tuple -m test_track_literals.

It means that the test asserts wrong condition. The result depends on previously executed code, it is not invariant.

[vstinner]

Ah thanks, it's good to have a way to reproduce the issue.

[sergey-miryanov]

There are two aspects to this problem: one with flowgraph optimizations (the CFG should produce untracked tuples where possible), and the second with marshal (since optimization results are cached in __pycache__).

I'd like to open two separate PRs — one for flowgraph + tupleobject changes, one for marshal.

Does that work?

@vstinner @serhiy-storchaka