New submission from Antoine Pitrou <pit...@free.fr>: This is an experimental patch to optimize some operations on small ints. pystone is 5-10% faster, pybench 2-3% faster, and here are some relevant benchmarks from unladen swallow:
### nbody ### Min: 0.345136 -> 0.317502: 1.09x faster Avg: 0.346827 -> 0.319561: 1.09x faster Significant (t=79.50) Stddev: 0.00140 -> 0.00198: 1.4084x larger ### nqueens ### Min: 0.339744 -> 0.313506: 1.08x faster Avg: 0.342630 -> 0.315380: 1.09x faster Significant (t=73.41) Stddev: 0.00218 -> 0.00146: 1.4931x smaller If the principle gets accepted, we could experiment with further optimizations such as dedicated opcodes for addition-with-int-constant, subscripting-with-int-constant, etc. ---------- files: smallints.patch keywords: patch messages: 118103 nosy: Aahz, aahz, mark.dickinson, pitrou, rhettinger, stutzbach priority: normal severity: normal status: open title: small int optimization type: performance versions: Python 3.2 Added file: http://bugs.python.org/file19148/smallints.patch _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue10044> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
