Infinite loop uninterruptable on Windows in 3.10

[gvanrossum]

BPO	42296
Nosy	@gvanrossum, @pfmoore, @vstinner, @tjguk, @markshannon, @zware, @serhiy-storchaka, @eryksun, @zooba, @miss-islington
PRs	bpo-42296: On Windows, fix CTRL+C regression #23257 [3.9] bpo-42296: On Windows, fix CTRL+C regression (GH-23257) #23259

^{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 2020-11-13.19:22:23.856>
created_at = <Date 2020-11-09.17:25:49.138>
labels = ['3.10', 'type-bug', '3.9', 'OS-windows']
title = 'Infinite loop uninterruptable on Windows in 3.10'
updated_at = <Date 2020-11-14.00:44:58.378>
user = 'https://github.com/gvanrossum'

bugs.python.org fields:

activity = <Date 2020-11-14.00:44:58.378>
actor = 'gvanrossum'
assignee = 'none'
closed = True
closed_date = <Date 2020-11-13.19:22:23.856>
closer = 'eryksun'
components = ['Windows']
creation = <Date 2020-11-09.17:25:49.138>
creator = 'gvanrossum'
dependencies = []
files = []
hgrepos = []
issue_num = 42296
keywords = ['patch']
message_count = 16.0
messages = ['380597', '380600', '380606', '380608', '380609', '380853', '380864', '380869', '380878', '380884', '380886', '380887', '380892', '380893', '380927', '380947']
nosy_count = 10.0
nosy_names = ['gvanrossum', 'paul.moore', 'vstinner', 'tim.golden', 'Mark.Shannon', 'zach.ware', 'serhiy.storchaka', 'eryksun', 'steve.dower', 'miss-islington']
pr_nums = ['23257', '23259']
priority = None
resolution = None
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue42296'
versions = ['Python 3.9', 'Python 3.10']

[gvanrossum]

This code cannot be interrupted with ^C on Windows (e.g. in the REPL)

while True:
    pass

This seems to be a regression, it works in earlier versions.

[eryksun]

It also cannot be interrupted in 3.9. But 3.8 and earlier work correctly.

[eryksun]

See bpo-40010. COMPUTE_EVAL_BREAKER() in Python/ceval.c -- or _Py_ThreadCanHandleSignals in Include/internal/pycore_pystate.h -- needs to take into account that SIGNAL_PENDING_SIGNALS() gets called on a completely new thread in Windows.

[gvanrossum]

So you're saying this war broken by #19087 ?

[eryksun]

Yes, if I force the return value of _Py_ThreadCanHandleSignals to 1, the loop is broken by a KeyboardInterrupt.