> ... I would expect the FastToLocals and LocalsToFast functions to require > some non-trivial adjustments ...
Thanks, Nick. I'm making precisely that change in a few places in frameobject.c. One loop for locals, another for cells & frees, a third for the stack (where the active stack is involved). So far, my second go-round is proceeding better (fingers crossed). I have added a new slot to the _frame struct (f_cellvars), initialized once at creation, then referenced elsewhere. I'm also rerunning the test suite more frequently. Once I've tweaked everything, all that will remain (in theory) to effect my change is to tweak the initialization of f_valuestack and f_cellvars. (If an extra slot is determined to be too space-expensive, a macro or inline function would suffice.) Aside: To speed up my testing it would be kinda nice if you could tell regrtest, "Just run X% of the tests at random plus any which failed on the previous run," but it's not horrible as-is. If it doesn't already exist (implying I didn't just miss it), perhaps it could be a good "easy" feature request for core dev novitiates. Skip _______________________________________________ Python-Dev mailing list -- python-dev@python.org To unsubscribe send an email to python-dev-le...@python.org https://mail.python.org/mailman3/lists/python-dev.python.org/ Message archived at https://mail.python.org/archives/list/python-dev@python.org/message/ON46TXJLMXLYTCRIQXRNNNUWU6HDB2YQ/ Code of Conduct: http://python.org/psf/codeofconduct/
