asyncio debug mode is very slow

[pitrou]

BPO	31970
Nosy	@pitrou, @vstinner, @giampaolo, @asvetlov, @1st1
PRs	bpo-31970: Reduce performance overhead of asyncio debug mode. #4314 [3.6] bpo-31970: Reduce performance overhead of asyncio debug mode. (GH-4314) #4322

^{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 = <Date 2017-11-07.16:51:22.516>
created_at = <Date 2017-11-07.13:15:04.297>
labels = ['3.7', 'type-bug', 'library', 'expert-asyncio']
title = 'asyncio debug mode is very slow'
updated_at = <Date 2017-11-08.13:45:41.420>
user = 'https://github.com/pitrou'

bugs.python.org fields:

activity = <Date 2017-11-08.13:45:41.420>
actor = 'pitrou'
assignee = 'none'
closed = True
closed_date = <Date 2017-11-07.16:51:22.516>
closer = 'pitrou'
components = ['Library (Lib)', 'asyncio']
creation = <Date 2017-11-07.13:15:04.297>
creator = 'pitrou'
dependencies = []
files = []
hgrepos = []
issue_num = 31970
keywords = ['patch']
message_count = 13.0
messages = ['305752', '305757', '305777', '305780', '305781', '305784', '305834', '305836', '305837', '305838', '305839', '305840', '305841']
nosy_count = 5.0
nosy_names = ['pitrou', 'vstinner', 'giampaolo.rodola', 'asvetlov', 'yselivanov']
pr_nums = ['4314', '4322']
priority = 'normal'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue31970'
versions = ['Python 3.6', 'Python 3.7']

[pitrou]

Each time a new future, handle or task is created, asyncio with debug mode enabled will parse the whole call stack and create a StackSummary object for it. Imagine a recursive coroutine: with N nested calls, you get O(N**2) performance.

Ideally debug mode wouldn't slow things too much, at least for development and testing setups.

[pitrou]

If this is accepted on the principle, it would be great to also patch 3.6.

[pitrou]

New changeset 921e943 by Antoine Pitrou in branch 'master':
bpo-31970: Reduce performance overhead of asyncio debug mode. (bpo-4314)
921e943

[pitrou]

New changeset d8d218f by Antoine Pitrou in branch '3.6':
[3.6] bpo-31970: Reduce performance overhead of asyncio debug mode. (GH-4314) (bpo-4322)
d8d218f

[pitrou]

Thank you for the quick reviews!

[1st1]

Thank you for helping with asyncio! I'll try to get to that transport performance issues you found sometime this week. I've a few ideas how to add 0-copy support to protocols.

[pitrou]

I've a few ideas how to add 0-copy support to protocols.

I'd be interesting to hear about them. The main challenge IMHO is to find a way to allow a readinto()-like functionality.

[giampaolo]

> I've a few ideas how to add 0-copy support to protocols.

I'd be interesting to hear about them. The main challenge IMHO is to find a way to allow a readinto()-like functionality.

Exposing sendfile() should be straightforward. I started implementing it years ago but I gave up pretty soon because asyncio had no solid test framework for testing an actual data transfer between two sockets and that basically represented a blocker. Basically back then all recv() / send() related tests were mocks. Not sure about the current situation but if that has changed I would be happy to contribute a PR (I was the one who contributed socket.sendfile()).

As for a readinto()-like functionality: the only thing I'm aware of is splice() syscall but it's Linux only. It must be noted that also sendfile() is UNIX only. Windows achieve the same via TransmitFile.

[pitrou]

Le 08/11/2017 à 14:40, Giampaolo Rodola' a écrit :

Exposing sendfile() should be straightforward.

sendfile() is not useful for data that's read from (or written to) memory.

As for a readinto()-like functionality: the only thing I'm aware of is splice() syscall but it's Linux only.

This is not the abstraction level we are talking about. The problem is
the Protocol API imposes copies by default (data_received() gives you an
arbitrarily-sized bytes object that doesn't match how your protocol
chunks data).

[giampaolo]

Sorry. It looks like I completely misunderstood. =)