50 producer and 50 consumer threads need more than 5GB of memory.
(cherry picked from commit 7a79f52d83)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
random.getrandbits() can now generate more that 2**31 bits.
random.randbytes() can now generate more that 256 MiB.
(cherry picked from commit 68784fed78)
The `textiowrapper_iternext` function called `_textiowrapper_writeflush`, but did not
use a critical section, making it racy in free-threaded builds.
(cherry picked from commit 44fb7c361c)
Co-authored-by: Duane Griffin <duaneg@dghda.com>
OpenSSL and HACL*-based hash functions constructors now support both `data` and `string` parameters.
Previously these constructor functions inconsistently supported sometimes `data` and sometimes `string`,
while the documentation expected `data` to be given in all cases.
(cherry picked from commit c6e63d9d35)
(cherry picked from commit 379d0bc956)
gh-134954: Hard-cap max file descriptors in subprocess test fd_status (GH-134955)
* Hard-cap max file descriptors in subprocess test fd_status
On some systems, `SC_OPEN_MAX` may return a very large value (i.e. 10**30), leading to the subprocess test timing out (or run forever).
Prevent this situation by applying a hard cap on how many file descriptors are checked.
* Fix typo in usage docstring
s/fd_stats/fd_status/
(cherry picked from commit f58873e4b2)
Co-authored-by: Itamar Oren <itamarost@gmail.com>
gh-133711: Fix test_readline.test_nonascii() for UTF-8 Mode (GH-134841)
Skip the test if the Python UTF-8 Mode is enabled and the LC_CTYPE
encoding is not UTF-8.
(cherry picked from commit 4635115c3f)
Co-authored-by: Victor Stinner <vstinner@python.org>
gh-133711: Fix test_regrtest for PYTHONUTF8=1 (GH-134839)
Use "backslashreplace" error handler to decode stdout and stderr.
Example:
vstinner@WIN C:\victor\python\main\build\test_python_worker_8360\x91>
"C:\victor\python\main\PCbuild\amd64\python_d.exe" -m test
--fast-ci --slow-ci --testdir
C:\Users\vstinner\AppData\Local\Temp\tmp0t59e8da
test_regrtest_noop1 test_regrtest_noop2 test_regrtest_noop3
test_regrtest_noop4
Notice the "\x91" byte at the end of the first line: it's the
non-ASCII U+00E6 character encoded to the OEM cp437 code page.
(cherry picked from commit 91618278e7)
Co-authored-by: Victor Stinner <vstinner@python.org>
* gh-132876: workaround broken ldexp() on Windows 10
ldexp() fails to round subnormal results before Windows 11,
so hide their bug.
(cherry picked from commit cf8941c603)
Co-authored-by: Tim Peters <tim.peters@gmail.com>
gh-128840: Limit the number of parts in IPv6 address parsing (GH-128841)
GH-128840: Limit the number of parts in IPv6 address parsing
Limit length of IP address string to 39
---------
(cherry picked from commit 47f1161d3a)
Co-authored-by: Seth Michael Larson <seth@python.org>
Co-authored-by: Hugo van Kemenade <1324225+hugovk@users.noreply.github.com>
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
Co-authored-by: Gregory P. Smith <greg@krypto.org>
gh-131357: Add some extra tests for empty bytes and bytearray (GH-134458)
(cherry picked from commit 7309eb60c0)
Co-authored-by: Sergey Miryanov <sergey.miryanov@gmail.com>
gh-134062: Fix hash collisions in IPv4Network and IPv6Network (GH-134063)
(cherry picked from commit f3fc0c16e0)
gh-134062: Fix hash collisions in IPv4Network and IPv6Network
gh-134062: Add hash collision regression test
Co-authored-by: Mike Salvatore <mike.s.salvatore@gmail.com>
Update `test_io` `_check_warn_on_dealloc` to use `self.` to dispatch to
different I/O implementations.
Update the `_pyio` implementation to match expected behavior, using the
same `_dealloc_warn` design as the C implementation uses to report the
topmost `__del__` object.
The FileIO one now matches all the others, so can use IOBase. There was
a missing check on closing (self._fd must be valid), add that check
(cherry picked from commit 5b0e827521)
Co-authored-by: Cody Maloney <cmaloney@users.noreply.github.com>
Test was only testing _io, expanded to cover _pyio.
(cherry picked from commit 06eaf4055c)
Co-authored-by: Cody Maloney <cmaloney@users.noreply.github.com>
Co-authored-by: Peter Bierma <zintensitydev@gmail.com>
If the error handler is used, a new bytes object is created to set as
the object attribute of UnicodeDecodeError, and that bytes object then
replaces the original data. A pointer to the decoded data will became invalid
after destroying that temporary bytes object. So we need other way to return
the first invalid escape from _PyUnicode_DecodeUnicodeEscapeInternal().
_PyBytes_DecodeEscape() does not have such issue, because it does not
use the error handlers registry, but it should be changed for compatibility
with _PyUnicode_DecodeUnicodeEscapeInternal().
(cherry picked from commit 9f69a58623)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
* Ensure that created files and dirs are always removed after test.
Now addCleanup() does not conflict with tearDown().
* Use os_helper.unlink() and os_helper.rmdir().
* Import TESTFN from os_helper.
(cherry picked from commit e29171bf8a)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
gh-133940: test_strftime incorrectly calculates expected week (GH-134281)
Let the system determine the correct tm_wday and tm_isdst.
(cherry picked from commit e3dda8f818)
Co-authored-by: Gustaf <79180496+GGyll@users.noreply.github.com>
In 3.13.3 we accidentally broke the interface for custom task factory. Factory authors added workarounds.
This PR (for 3.13.4) unbreaks task factories that haven't made a workaround yet while also supporting those that have.
NOTE: The custom task factory API will change to what we accidentally released in 3.13.3.
Co-authored-by: Kumar Aditya <kumaraditya@python.org>
Co-authored-by: Guido van Rossum <gvanrossum@gmail.com>
This fixes a crash in `_PyObject_TryGetInstanceAttribute` due to the use
of `_PyDictKeys_StringLookup` on an unlocked dictionary that may be
concurrently modified.
The underlying bug was already fixed in 3.14 and the main branch.
(partially cherry picked from commit 1b15c89a17)
gh-133744: Fix multiprocessing interrupt test: add an event (#133746)
Add an event to synchronize the parent process with the child
process: wait until the child process starts sleeping.
(cherry picked from commit c2989b7070)
T. Wouters