This is needed to even the run the test suite on buildbots for affected platforms; e.g.:

```
./python.exe  ./Tools/scripts/run_tests.py -j 1 -u all -W --slowest --fail-env-changed --timeout=11700 -j2
/home/embray/src/python/test-worker/3.x.test-worker/build/python -u -W default -bb -E -W error::BytesWarning -m test -r -w -j 1 -u all -W --slowest --fail-env-changed --timeout=11700 -j2
Traceback (most recent call last):
  File "./Tools/scripts/run_tests.py", line 56, in <module>
    main(sys.argv[1:])
  File "./Tools/scripts/run_tests.py", line 52, in main
    os.execv(sys.executable, args)
PermissionError: [Errno 13] Permission denied
make: *** [Makefile:1073: buildbottest] Error 1
```

https://bugs.python.org/issue34871

`arg` is misspelled as `agr`. I noticed this when playing with https://bugs.python.org/issue34871


https://bugs.python.org/issue34871

Use tp_members and tp_getset instead.

Use tp_members and tp_getset instead.

The C implementation of asyncio.Task currently fails to perform the
cancellation cleanup correctly in the following scenario.

    async def task1():
        async def task2():
            await task3     # task3 is never cancelled

        asyncio.current_task().cancel()
        await asyncio.create_task(task2())

The actuall error is a hardcoded call to `future_cancel()` instead of
calling the `cancel()` method of a future-like object.

Thanks to Vladimir Matveev for noticing the code discrepancy and to
Yury Selivanov for coming up with a pathological scenario.

formatfloat() was not checking if PyBytes_FromStringAndSize()
failed, which could lead to a null pointer dereference in
_PyBytes_FormatEx().

Make it clear that the msg argument to assertWarns/assertWarnsRegex/assertRaisesRegex is keyword-only. (GH-9680)

A follow up to be4e5b89.

Fix a reference issue inside multiprocessing.Pool that caused the pool to remain alive if it was deleted without being closed or terminated explicitly.

Fixes broken build on OpenBSD-current.

closes bpo-34868: Improve error message with '_' is combined with an invalid type specifier. (GH-9666)

property_descr_get() uses a "cached" tuple to optimize function
calls. But this tuple can be discovered in debug mode with
sys.getobjects(). Remove the optimization, it's not really worth it
and it causes 3 different crashes last years.

Microbenchmark:

./python -m perf timeit -v \
    -s "from collections import namedtuple; P = namedtuple('P', 'x y'); p = P(1, 2)" \
    --duplicate 1024 "p.x"

Result:

Mean +- std dev: [ref] 32.8 ns +- 0.8 ns -> [patch] 40.4 ns +- 1.3 ns: 1.23x slower (+23%)

Patch by Serhiy Storchaka (in PR #9642).

* Compiling a string annotation containing a lambda with keyword-only
argument without default value caused a crash.

* Remove the final "*" (it is incorrect syntax) in the representation of
lambda without *args and keyword-only arguments when compile from AST.

* Improve the representation of lambda without arguments.

The waiting is pretty normal for any asyncio program, logging its time just adds
a noise to logs without any useful information provided.



https://bugs.python.org/issue34849

Switch "list" with "iterable" to match with the implementation.

Support for threadless builds was removed in a6a4dc81.

_pickle.Unpickler.__init__() should return -1 if Pdata_New() fails, not 1.

Since ProactorEventLoop is now the default in 3.8, remove examples
using it explicitly on Windows.





https://bugs.python.org/issue34687

After some failures in AMD64 FreeBSD CURRENT Debug 3.x buildbots
regarding tests in test_multiprocessing_spawn and after examining
similar failures in test_socket, some errors in the calculation of
ancillary data buffers were found in multiprocessing.reduction.

CMSG_LEN() can often be used as the buffer size for recvmsg() to
receive a single item of ancillary data, but RFC 3542 requires portable
applications to use CMSG_SPACE() and thus include space for padding,
even when the item will be the last in the buffer.

The failures we experience are due to the usage of CMSG_LEN() instead of
CMSG_SPACE().

Improvements:
1. Include the number of valid data characters in the error message.
2. Mention "number of data characters" rather than "length".


https://bugs.python.org/issue34736

Report the filename to the exception when raising {gdbm,dbm.ndbm}.error in
dbm.gnu.open() and dbm.ndbm.open() functions, so it gets printed when the
exception is raised, and can also be obtained by the filename attribute of the
exception object.

This fixes various compiler warnings.

Use a monotonic clock to compute timeouts in :meth:`Executor.map` and :func:`as_completed`, in order to prevent timeouts from deviating when the system clock is adjusted.

This may not be sufficient on all systems. On POSIX for example, the actual waiting (e.g. in ``sem_timedwait``) is specified to rely on the CLOCK_REALTIME clock.

After some failures in AMD64 FreeBSD CURRENT Debug 3.x buildbots
regarding tests in test_socket that are using
testFDPassSeparateMinSpace(), FreeBDS revision 337423 was pointed
out to be the reason the test started to fail.

A close examination of the manpage for cmsg_space(3) reveals that
the number of file descriptors needs to be taken into account when
using CMSG_LEN().

This commit fixes tests in test_socket to use correctly CMSG_LEN, taking
into account the number of FDs.