[issue45767] Fix types for dev_t processing in posix module

2022-01-20 Thread Dmitry Marakasov
Dmitry Marakasov added the comment: > Is device number -1 used in any context (for example as "unknown device > number")? Yes, there's NODEV macro in both Linux and FreeBSD which expands to ((dev_t)-1). -- ___ Python

[issue45767] Fix types for dev_t processing in posix module

2021-11-09 Thread Dmitry Marakasov
Dmitry Marakasov added the comment: In case you're curious, here are some st_dev values which are encountered on my FreeBSD: - 0xac2308de99d1f699 - ZFS - 0x7100ff00 - devfs - 0x8700ff01 - tmpfs - 0x2900ff4e - nullfs I suspect Linux uses smaller numbers which do not cause overflows so

[issue45767] Fix types for dev_t processing in posix module

2021-11-09 Thread Dmitry Marakasov
Change by Dmitry Marakasov : -- pull_requests: +27743 stage: -> patch review pull_request: https://github.com/python/cpython/pull/29494 ___ Python tracker <https://bugs.python.org/issu

[issue45767] Fix types for dev_t processing in posix module

2021-11-09 Thread Dmitry Marakasov
New submission from Dmitry Marakasov : So, I was investigating a test failure of python 3.11 and 3.10 on FreeBSD (but it likely applies to all python versions): == FAIL: test_makedev (test.test_posix.PosixTester

[issue45326] Unexpected TypeError with type alias+issubclass+ABC

2021-09-29 Thread Dmitry Marakasov
New submission from Dmitry Marakasov : Here's a curious problem. issubclass() check of a type against an ABC-derived class raises TypeError claiming that type is not a class, however inspect.isclass() says it's a class, and issubclass() check against a simple class works fine: ``` from abc

[issue43792] Cython is not compatible with Python 3.10 yet: "SystemError: bad argument to internal function"

2021-04-11 Thread Dmitry Marakasov
Dmitry Marakasov added the comment: > This issue is a bug in Cython, not in Python. I close it. This is correct, thanks! The problem is gone with git master version of cython. -- ___ Python tracker <https://bugs.python.org/issu

[issue43792] "bad argument to internal function" in cython

2021-04-09 Thread Dmitry Marakasov
New submission from Dmitry Marakasov : I'm playing with adding python3.10a7 port to FreeBSD ports collection, and have run into similar problem with building multiple third party modules (for instance, yaml, yarl and pystemmer). Here's log from yaml: running build_ext cythoning yaml

[issue38744] python 3.8 hang in multiprocessing.Pool() locking on FreeBSD

2019-11-13 Thread Dmitry Marakasov
Dmitry Marakasov added the comment: > What happens if you instead write: It survived more than 4000 runs, it looks like it doesn't experience the problem. -- ___ Python tracker <https://bugs.python.org/issu

[issue38744] python 3.8 hang in multiprocessing.Pool() locking on FreeBSD

2019-11-08 Thread Dmitry Marakasov
New submission from Dmitry Marakasov : System: FreeBSD 12.0-RELEASE, amd64. This simple program from multiprocessing import Pool from time import sleep Pool().map(sleep, [0.01] * 10) works fine with python 3.7, but is likely (about 20-50% probability on my 4 core box) to hang

[issue36527] unused parameter warnings in Include/object.h (affecting building third party code)

2019-04-04 Thread Dmitry Marakasov
New submission from Dmitry Marakasov : Python 3.8 and nightly introduces unused (in some cases) parameters in object.h header. This makes compilation of third party code which includes the header fail if it's built with -Werror. Build log excerpt: --- g++ -Wno-unused-result -Wsign-compare