Implement CALL_FUNCTION adaptive interpreter optimizations

[Fidget-Spinner]

BPO	44525
Nosy	@markshannon, @pablogsal, @sobolevn, @Fidget-Spinner, @thesamesam
PRs	bpo-44525: Specialize CALL_FUNCTION for C function calls #26934 bpo-44525: Specialize simple Python calls. #29033 bpo-44525: Add recursive checks for CALL_FUNCTION_BUILTIN_O #29271 bpo-44525: Copy freevars in bytecode to allow calls to inner functions to be specialized #29595 bpo-44525: Specialize for calls to type and other builtin classes with 1 argument. #29942 bpo-44525: Split calls into PRECALL and CALL #30011 bpo-44525: Document new call opcodes for 3.11 #30107 bpo-44525: Specialize calls to normal python classes #30415 bpo-46709: fix race conditions in unittest/test_brake #31273

^{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 = None
closed_at = None
created_at = <Date 2021-06-28.14:22:18.192>
labels = ['3.11', 'performance']
title = 'Implement CALL_FUNCTION adaptive interpreter optimizations'
updated_at = <Date 2022-02-11.11:31:07.338>
user = 'https://github.com/Fidget-Spinner'

bugs.python.org fields:

activity = <Date 2022-02-11.11:31:07.338>
actor = 'sobolevn'
assignee = 'none'
closed = False
closed_date = None
closer = None
components = []
creation = <Date 2021-06-28.14:22:18.192>
creator = 'kj'
dependencies = []
files = []
hgrepos = []
issue_num = 44525
keywords = ['patch']
message_count = 9.0
messages = ['396639', '404376', '404470', '405108', '405192', '406831', '408554', '408609', '409779']
nosy_count = 5.0
nosy_names = ['Mark.Shannon', 'pablogsal', 'sobolevn', 'kj', 'thesamesam']
pr_nums = ['26934', '29033', '29271', '29595', '29942', '30011', '30107', '30415', '31273']
priority = 'normal'
resolution = None
stage = 'patch review'
status = 'open'
superseder = None
type = 'performance'
url = 'https://bugs.python.org/issue44525'
versions = ['Python 3.11']

Linked PRs

• GH-88691: Shrink the CALL caches #103230

[Fidget-Spinner]

CALL_FUNCTION can be specialized. Copying from Mark's comments faster-cpython/ideas#54 (comment)

There are a number of specializations of CALL_FUNCTION that make sense:

1. Calls to a Python function where the arguments and parameters match exactly.
2. Calls to a Python function with keyword arguments, or defaults, when the argument shuffle can be pre-computed.
3. Calls to builtin functions.
4. Calls to specific builtin functions, specifically len and instance.
5. Calls to type with a single argument.
6. Instantiation of a "normal" class, that is a class that does not redefine __new__ and whose metaclass is type.

I've prepared a PR that should speed up non-keyword calls to PyCFunction. It covers specializations 3 and 4. Stable pyperf microbenchmarks show 5-15% less call overhead for some PyCFunctions. Please see faster-cpython/ideas#54 (comment) for the benchmark script and results.

This issue is also tied to bpo-44207 (Add a version number to Python functions) which will be required for specializations 1 and 2.

[markshannon]

New changeset 3163e68 by Ken Jin in branch 'main':
bpo-44525: Specialize CALL_FUNCTION for C function calls (GH-26934)
3163e68

[markshannon]

New changeset 8863a0f by Mark Shannon in branch 'main':
bpo-44525: Specialize simple Python calls. (GH-29033)
8863a0f

[pablogsal]

Unfortunately, PR 26934 has broken thes 390x RHEL7 LTO 3.x buildbot as you can see before. As per the buildbot maintenance procedures, we will need to revert this PR unless is fixed in 24 hours.

@markshannon @Fidget-Spinner

[markshannon]

New changeset 0a1a36b by Ken Jin in branch 'main':
bpo-44525: Add recursive checks for CALL_FUNCTION_BUILTIN_O (GH-29271)
0a1a36b

[markshannon]

New changeset 135cabd by Mark Shannon in branch 'main':
bpo-44525: Copy free variables in bytecode to allow calls to inner functions to be specialized (GH-29595)
135cabd

[markshannon]

New changeset 9f8f451 by Mark Shannon in branch 'main':
bpo-44525: Split calls into PRECALL and CALL (GH-30011)
9f8f451

[markshannon]

New changeset 3a60bfe by Mark Shannon in branch 'main':
bpo-44525: Specialize for calls to type and other builtin classes with 1 argument. (GH-29942)
3a60bfe

[markshannon]

See faster-cpython/ideas#210

[iritkatriel]

Is there anything left to do on this?

[pablogsal]

Closing. Please feel free to re-open if we missed anything.