[pypy-issue] [issue1067] Performance on Stack-Based VM machine

Armin Rigo tracker at bugs.pypy.org
Sat Sep 22 22:15:16 CEST 2012


Armin Rigo <armin.rigo at gmail.com> added the comment:

Finally found a simple idea (c7c74257157c): when we hit the limit
'max_unroll_loops', then instead of aborting, retry to compile by
disabling the 'unroll' optimization.

This alone fixes go2.py, which runs 235 times faster now than before,
and 8.7 times faster than CPython.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1067>
________________________________________


More information about the pypy-issue mailing list