[issue18771] Reduce the cost of hash collisions for set objects

[Tal Einat]

Tal Einat added the comment:

Because of the unusually low differences, I tried running the benchmarks again on the MacBook (OSX 10.8). The first time was actually with so.diff, this time was with so2.diff.

0.69% less time for first benchmark (7.30 before, 7.25 after)
8.08% less time for second benchmark (10.85 before, 9.97 after)

Strange...

----------

_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue18771>
_______________________________________