[issue41972] bytes.find consistently hangs in a particular scenario

[Dennis Sweeney]

Dennis Sweeney <sweeney.dennis650 at gmail.com> added the comment:

I am attempting to better understand the performance characteristics to determine where a cutoff should go.

Attached is a colorful table of benchmarks of the existing algorithm to the PR with the cutoff changed to `if (1)` (always two-way) or `if (0)` (always status quo), and tested on a variety of needle lengths and haystack lengths.

----------
Added file: https://bugs.python.org/file49577/Table of benchmarks on lengths.jpg

_______________________________________
Python tracker <report at bugs.python.org>
<https://bugs.python.org/issue41972>
_______________________________________