STINNER Victor added the comment: Results of performance 0.5.0 on speed-python:
haypo@speed-python$ python3 -m perf compare_to -G --min-speed=3 2016-11-23_19-34-default-3d660ed2a60e.json patch.json Slower (3): - spectral_norm: 265 ms +- 2 ms -> 277 ms +- 7 ms: 1.04x slower - xml_etree_iterparse: 217 ms +- 2 ms -> 226 ms +- 4 ms: 1.04x slower - nqueens: 261 ms +- 2 ms -> 269 ms +- 3 ms: 1.03x slower Faster (3): - scimark_sor: 519 ms +- 10 ms -> 496 ms +- 7 ms: 1.05x faster - mako: 43.0 ms +- 0.8 ms -> 41.5 ms +- 0.2 ms: 1.04x faster - call_method: 16.2 ms +- 0.2 ms -> 15.7 ms +- 0.3 ms: 1.03x faster Benchmark hidden because not significant (58): 2to3, call_method_slots, call_method_unknown, (...) Hum, boring result. This change alone doesn't change any significant speedup, even some slowndon. Maybe it's just a bad idea. Maybe it should be combined with other new bytecode instructions. Maybe only a full new instruction set using registers show significant speedup. I don't know :-( ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue28800> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com