Repository navigation
Conversation
|
gently ping @markshannon |
| """), PYTHON_JIT="1") | ||
| self.assertEqual(result[0].rc, 0, result) | ||
|
|
||
| def test_157875_rewound_trace_reference_leak(self): |
There was a problem hiding this comment.
Should this test assert that the rewinding happened? Or would that make the test too brittle?
| @@ -0,0 +1,2 @@ | |||
| Fix a reference leak in the JIT when a trace is rewound after encountering | |||
| unsupported bytecode. | |||
There was a problem hiding this comment.
maybe just say when tracing is rewound? I think this get's triggered for standard supported JIT traces hit natural endings such as jumps
There was a problem hiding this comment.
It seems too detailed. We can ask others?👀
| _PyUOpInstruction *curr = uop_buffer_last(trace); | ||
| while (curr->opcode != _SET_IP && uop_buffer_length(trace) > 2) { | ||
| if (_PyUop_Flags[curr->opcode] & HAS_RECORDS_VALUE_FLAG) { | ||
| Py_XDECREF((PyObject *)(uintptr_t)curr->operand0); |
There was a problem hiding this comment.
we now have a very similar pattern three times in this file, maybe worth seeing if it's worth to make it a common operation
There was a problem hiding this comment.
I think it would be better to focus on the bug fix scope of this PR. However, currently the JIT has been suspended and only bug fixes are accepted
3 participants
Uh oh!
There was an error while loading. Please reload this page.