Implement free-list for single-digit longs

[1st1]

BPO	26341
Nosy	@mdickinson, @vstinner, @serhiy-storchaka, @1st1
Superseder	bpo-24165: Free list for single-digits ints
Files	long_fl.patch

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = 'https://github.com/1st1'
closed_at = <Date 2016-02-11.20:09:40.568>
created_at = <Date 2016-02-11.19:22:36.712>
labels = ['interpreter-core', 'performance']
title = 'Implement free-list for single-digit longs'
updated_at = <Date 2016-02-11.20:09:40.567>
user = 'https://github.com/1st1'

bugs.python.org fields:

activity = <Date 2016-02-11.20:09:40.567>
actor = 'yselivanov'
assignee = 'yselivanov'
closed = True
closed_date = <Date 2016-02-11.20:09:40.568>
closer = 'yselivanov'
components = ['Interpreter Core']
creation = <Date 2016-02-11.19:22:36.712>
creator = 'yselivanov'
dependencies = []
files = ['41900']
hgrepos = []
issue_num = 26341
keywords = ['patch']
message_count = 3.0
messages = ['260124', '260125', '260127']
nosy_count = 4.0
nosy_names = ['mark.dickinson', 'vstinner', 'serhiy.storchaka', 'yselivanov']
pr_nums = []
priority = 'normal'
resolution = 'duplicate'
stage = 'resolved'
status = 'closed'
superseder = '24165'
type = 'performance'
url = 'https://bugs.python.org/issue26341'
versions = ['Python 3.6']

[1st1]

The attached patch implements a free-list for single-digit longs. We already have free lists for many fundamental types, such as floats & unicode.

The patch improves performance in micro-benchmarks by 10-20%. It'll also lessen memory fragmentation issues.

== Benchmarks ==

### spectral_norm ###
Min: 0.268018 -> 0.245042: 1.09x faster
Avg: 0.289548 -> 0.257861: 1.12x faster
Significant (t=18.82)
Stddev: 0.01004 -> 0.00640: 1.5680x smaller

-m timeit -s "loops=tuple(range(1000))" "for x in loops: x+x"
with patch: 34.5 usec without patch: 45.9 usec

== Why only single-digit? ==

I've also a patch that implements free-lists for 1-digit, 2-digits and 3-digits longs, and collects statistics on them. It looks like we only want to optimize 1-digit longs:

• 2to3 benchmark (the first number is the number of all N-digit longs, the second number is the number of longs created via free-list):

===> d1_longs = 142384 124759
===> d2_longs = 6872 6264
===> d3_longs = 2907 2834

• richards:
  ===> d1_longs = 219630 219033
  ===> d2_longs = 1455 1096
  ===> d3_longs = 630 627

• spectral_norm:
  ===> d1_longs = 133928432 133927838
  ===> d2_longs = 1471 1113
  ===> d3_longs = 630 627

[serhiy-storchaka]

This looks as a duplicate of bpo-24165.

[1st1]

You're right Serhiy, closing this one.