gh-126204: Add a sys._jit_enabled helper - #126247
Conversation
Sorry, something went wrong.
zooba
left a comment
There was a problem hiding this comment.
I'm not convinced that pyconfig.h is the right home for this - it seems like just pulling it from sysconfig is sufficient.
I'd rather not take the changes in the PC/PCbuild directories if we don't have to.
Sorry, something went wrong.
| # Temporarily enable ignore_stderr=True to ignore warnings on JIT builds | ||
| # See gh-126255 for more information | ||
| self.check_all_configs("test_initconfig_api", config, preconfig, | ||
| api=API_ISOLATED, ignore_stderr=True) |
There was a problem hiding this comment.
That's fun. Due to incorrect assumptions, where JIT is actually enabled or not, this hack is no longer needed. Just setting PYTHON_JIT=0 is enough here.
Sorry, something went wrong.
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
There was a problem hiding this comment.
A final nitpick. And a question: is there a lot of
_PyOptimizerObject* optimizer = _Py_GetOptimizer();
if (optimizer != NULL) {
Py_DECREF(optimizer);
...
}in the code? if so, we could perhaps have some function that checks if there is an optimizer.
EDIT: Since we are moving to removing the _PyOptimizer API, it's possible we don't need this.
Sorry, something went wrong.
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
In fact, there are only two uses of _Py_GetOptimizer. I'm not sure if we should remove this usage in this PR. |
Sorry, something went wrong.
📚 Documentation preview 📚: https://cpython-previews--126247.org.readthedocs.build/