test_builtin and test_socketserver no longer use signal.alarm() to
implement a watchdog with a hardcoded timeout (2 and 60 seconds).
Python test runner regrtest has two watchdogs: faulthandler and
timeout on running worker processes. Tests using short hardcoded
timeout can fail on slowest buildbots just because the timeout is too
short.
* Use `FindFirstFile` Win32 API to fix a bug where `ntpath.realpath()`
breaks out of traversing a series of paths where a (handled)
`ERROR_ACCESS_DENIED` or `ERROR_SHARING_VIOLATION` occurs.
* Update docs to reflect that `ntpath.realpath()` eliminates MS-DOS
style names.
When using worker processes (-jN) with --verbose3 option, regrtest
can now display the worker output even if a worker process does
crash. Previously, sys.stdout and sys.stderr were replaced and so
the worker output was lost on a crash.
CFunctionFullTests now also runs "bt" command before "py-bt-full",
similar to CFunctionTests which also runs "bt" command before
"py-bt". So test_gdb can skip the test if patterns like "?? ()" are
found in the gdb output.
dataclasses.replace() now raises TypeError instead of ValueError if
specify keyword argument for a field declared with init=False or miss keyword
argument for required InitVar field.
The test had an instability issue due to the ordering of the dummy
queue operation and the real wakeup pipe operations. Both primitives
are thread safe but not done atomically as a single update and may
interleave arbitrarily. With the old order of operations this can lead
to an incorrect state where the dummy queue is full but the wakeup
pipe is empty. By swapping the order in clear() I think this can no
longer happen in any possible operation interleaving (famous last
words).
In Python/bytecodes.c, you now write
```
DEOPT_IF(condition);
```
The code generator expands this to
```
DEOPT_IF(condition, opcode);
```
where `opcode` is the name of the unspecialized instruction.
This works inside macro expansions too.
**CAVEAT:** The entire `DEOPT_IF(condition)` statement must be on a single line.
If it isn't, the substitution will fail; an error will be printed by the code generator
and the C compiler will report some errors.
In a few places we switch to another interpreter without knowing if it has a thread state associated with the current thread. For the main interpreter there wasn't much of a problem, but for subinterpreters we were *mostly* okay re-using the tstate created with the interpreter (located via PyInterpreterState_ThreadHead()). There was a good chance that tstate wasn't actually in use by another thread.
However, there are no guarantees of that. Furthermore, re-using an already used tstate is currently fragile. To address this, now we create a new thread state in each of those places and use it.
One consequence of this change is that PyInterpreterState_ThreadHead() may not return NULL (though that won't happen for the main interpreter).
The existence of background threads running on a subinterpreter was preventing interpreters from getting properly destroyed, as well as impacting the ability to run the interpreter again. It also affected how we wait for non-daemon threads to finish.
We add PyInterpreterState.threads.main, with some internal C-API functions.
This change makes sure sys.path[0] is set properly for subinterpreters. Before, it wasn't getting set at all. This PR does not address the broader concerns from gh-109853.
If the timeout is greater than PY_TIMEOUT_MAX,
PyThread_acquire_lock_timed() uses a timeout of PY_TIMEOUT_MAX
microseconds, which is around 280.6 years. This case is unlikely and
limiting a timeout to 280.6 years sounds like a reasonable trade-off.
The constant PY_TIMEOUT_MAX is not used in PyPI top 5,000 projects.
This method supports file URIs (including variants) as described in RFC 8089, such as URIs generated by `pathlib.Path.as_uri()` and `urllib.request.pathname2url()`.
The method is added to `Path` rather than `PurePath` because it uses `os.fsdecode()`, and so its results vary from system to system. I intend to deprecate `PurePath.as_uri()` and move it to `Path` for the same reason.
Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com>
* _add_python_opts() now handles cross compilation and HOSTRUNNER.
* display_header() now tells if Python is cross-compiled, display
HOSTRUNNER, and get the host platform.
* Remove Tools/scripts/run_tests.py script.
* Remove "make hostrunnertest": use "make buildbottest"
or "make test" instead.
* Refactor os_sched_getaffinity_impl(): move variable definitions to
their first assignment.
* Fix test_posix.test_sched_getaffinity(): restore the old CPU mask
when the test completes!
* Doc: Specify that os.cpu_count() counts *logicial* CPUs.
* Doc: Specify that os.sched_getaffinity(0) is related to the calling
thread.
Rename WORKER_ERROR to WORKER_BUG. Add WORKER_FAILED state: it does
not stop the manager, whereas WORKER_BUG does.
Change also TestResults.display_result() order: display failed tests
at the end, the important important information.
WorkerThread now tries to get the signal name for negative exit code.
Call also copy_python_src_ignore() on listdir() names.
shutil.copytree(): replace set() with an empty tuple. An empty tuple
becomes a constant in the compiler and checking if an item is in an
empty tuple is cheap.
* pycore_pythread.h is now the central place to make sure that
_POSIX_THREADS and _POSIX_SEMAPHORES macros are defined if
available.
* Make sure that pycore_pythread.h is included when _POSIX_THREADS
and _POSIX_SEMAPHORES macros are tested.
* PY_TIMEOUT_MAX is now defined as a constant, since its value
depends on _POSIX_THREADS, instead of being defined as a macro.
* Prevent integer overflow in the preprocessor when computing
PY_TIMEOUT_MAX_VALUE on Windows:
replace "0xFFFFFFFELL * 1000 < LLONG_MAX"
with "0xFFFFFFFELL < LLONG_MAX / 1000".
* Document the change and give hints how to fix affected code.
* Add an exception for PY_TIMEOUT_MAX name to smelly.py
* Add PY_TIMEOUT_MAX to the stable ABI
Add private `pathlib._PathBase` class. This will be used by an experimental PyPI package to incubate a `tarfile.TarPath` class.
Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com>
concurrent.futures: The *executor manager thread* now catches
exceptions when adding an item to the *call queue*. During Python
finalization, creating a new thread can now raise RuntimeError. Catch
the exception and call terminate_broken() in this case.
Add test_python_finalization_error() to test_concurrent_futures.
concurrent.futures._ExecutorManagerThread changes:
* terminate_broken() no longer calls shutdown_workers() since the
call queue is no longer working anymore (read and write ends of
the queue pipe are closed).
* terminate_broken() now terminates child processes, not only
wait until they complete.
* _ExecutorManagerThread.terminate_broken() now holds shutdown_lock
to prevent race conditons with ProcessPoolExecutor.submit().
multiprocessing.Queue changes:
* Add _terminate_broken() method.
* _start_thread() sets _thread to None on exception to prevent
leaking "dangling threads" even if the thread was not started
yet.
* Add context manager on Bunch class.
* Bunch now catchs exceptions on executed functions and re-raise them
at __exit__() as an ExceptionGroup.
* Rewrite BarrierProxy.test_default_timeout(). Use a single thread.
Only check that barrier.wait() blocks for at least default timeout
seconds.
* test_with(): inline _with() function.
Fix test_asyncio timeouts: don't measure the maximum duration, a test
should not measure a CI performance. Only measure the minimum
duration when a task has a timeout or delay. Add CLOCK_RES to
test_asyncio.utils.
In the subprocess code path, wait until the child process completes
with a timeout of EXIT_TIMEOUT seconds.
Fix create_worker_process() regression: use start_new_session=True if
USE_PROCESS_GROUP is true.
WorkerThread.wait_stopped() uses a timeout of 60 seconds, instead of
30 seconds.
On Windows, multiprocessing Popen.terminate() now catchs
PermissionError and get the process exit code. If the process is
still running, raise again the PermissionError. Otherwise, the
process terminated as expected: store its exit code.
Fix race conditions in test_threading lock tests. Wait until a
condition is met rather than using time.sleep() with a hardcoded
number of seconds.
* Replace sleeping loops with support.sleeping_retry() which raises
an exception on timeout.
* Add wait_threads_blocked(nthread) which computes a sleep depending
on the number of threads. Remove _wait() function.
* test_set_and_clear(): use a way longer Event.wait() timeout.
* BarrierTests.test_repr(): wait until the 2 threads are waiting for
the barrier. Use a way longer timeout for Barrier.wait() timeout.
* test_thread_leak() no longer needs to count
len(threading.enumerate()): Bunch uses
threading_helper.wait_threads_exit() internally which does it in
wait_for_finished().
* Add BaseLockTests.wait_phase() which implements a timeout.
test_reacquire() and test_recursion_count() use wait_phase().
In 2003, test_pty got a hardcoded timeout of 10 seconds to prevent
hanging on AIX & HPUX "if run after test_openpty":
commit 7d8145268e. Since 2003, test_pty
was no longer reported to hang on AIX. But today, the test can fail
simply because a CI is busy running other tests in parallel.
The timeout of 10 seconds is no longer needed, just remove it.
Moreover, regrtest now has multiple built-in generic timeout
mecanisms.
Fix test_interprocess_signal() of test_signal. Make sure that the
subprocess.Popen object is deleted before the test raising an
exception in a signal handler. Otherwise, Popen.__del__() can get the
exception which is logged as "Exception ignored in: ...." and the
test fails.
* Split test_pycfunction.py: add test_cfunction_full.py.
Split the function into the following 6 functions. In verbose
mode, these "pycfunction" tests now log each tested call.
* test_pycfunction_noargs()
* test_pycfunction_o()
* test_pycfunction_varargs()
* test_pycfunction_varargs_keywords()
* test_pycfunction_fastcall()
* test_pycfunction_fastcall_keywords()
* Move get_gdb_repr() to PrettyPrintTests.
* Replace DebuggerTests.get_sample_script() with SAMPLE_SCRIPT.
* Rename checkout_hook_path to CHECKOUT_HOOK_PATH.
* Rename gdb_version to GDB_VERSION_TEXT.
* Replace (gdb_major_version, gdb_minor_version) with GDB_VERSION.
* run_gdb() uses "backslashreplace" error handler instead of "replace".
* Add check_gdb() function to util.py.
* Enhance support.check_cflags_pgo(): check also for sysconfig
PGO_PROF_USE_FLAG (if available) in compiler flags.
* Move some SkipTest checks to test_gdb/__init__.py.
* Elaborate why gdb cannot be tested on Windows: gdb doesn't support
PDB debug symbol files.
Fix test_timeout() of test_concurrent_futures.test_wait. Remove the
future which may or may not complete depending if it takes longer
than the timeout ot not. Keep the second future which does not
complete before wait(). Make also the test faster: 0.5 second instead
of 6 seconds, so remove @support.requires_resource('walltime')
decorator.
Split test_gdb.py file into a test_gdb package made of multiple
tests, so tests can now be run in parallel.
* Create Lib/test/test_gdb/ directory.
* Split test_gdb.py into multiple files in Lib/test/test_gdb/
directory.
* Move Lib/test/gdb_sample.py to Lib/test/test_gdb/ directory.
Update get_sample_script(): use __file__ to locate gdb_sample.py.
* Move gdb_has_frame_select() and HAS_PYUP_PYDOWN to test_misc.py.
* Explicitly skip test_gdb on Windows. Previously, test_gdb was
skipped even if gdb was available because of
gdb_has_frame_select().
Fix the test when run on an installed Python: use "abs_srcdir" of
sysconfig, and skip the test if the Python source code cannot be
found.
* Tools/patchcheck/patchcheck.py, Tools/freeze/test/freeze.py and
Lib/test/libregrtest/utils.py now first try to get "abs_srcdir"
from sysconfig, before getting "srcdir" from sysconfig.
* test.pythoninfo logs sysconfig "abs_srcdir".
WASM and WASI buildbots use multiple PYTHON environment variables
such as PYTHONPATH and _PYTHON_HOSTRUNNER. Don't use -E if the
--python=COMMAND option is used.
Fix copy_source_tree() function of test_tools.test_freeze:
* Don't copy SRC_DIR/build/ anymore. This directory is modified by
other tests running in parallel.
* Add test.support.copy_python_src_ignore().
* Use sysconfig to get the source directory.
* Use sysconfig.get_config_var() to get CONFIG_ARGS variable.
* On Windows, use subprocess.run() instead of os.execv().
* Only add needed options
* Rename reexec parameter to _add_python_opts.
* Rename --no-reexec option to --dont-add-python-opts.
Remove replace_stdout(): call sys.stdout.reconfigure() instead of set
the error handler to backslashreplace.
display_header() logs an empty line and flush stdout.
Remove encoding workaround in display_header() since stdout error
handler is now set to backslashreplace earlier.
When --fast-ci or --slow-ci option is used, regrtest now replaces the
current process with a new process to add "-u -W default -bb -E"
options to Python.
Changes:
* PCbuild/rt.bat and Tools/scripts/run_tests.py no longer need to add
"-u -W default -bb -E" options to Python: it's now done by
regrtest.
* Fix Tools/scripts/run_tests.py: flush stdout before replacing the
process. Previously, buffered messages were lost.
* Add --fast-ci and --slow-ci options to libregrtest:
* --fast-ci uses a default timeout of 10 minutes and "-u all,-cpu"
(skip slowest tests).
* --slow-ci uses a default timeout of 20 minues and "-u all" (run
all tests).
* regrtest header now lists test resources.
* Makefile changes:
* "make test", "make hostrunnertest" and "make coverage-report" now
use --fast-ci option and TESTTIMEOUT variable.
* "make buildbottest" now uses "--slow-ci". Remove options which
became redundant with "--slow-ci".
* "make testall" and "make testuniversal" now use --slow-ci option
and TESTTIMEOUT variable.
* "make testall" now uses "find -exec rm ..." instead of
"find ... -print|xargs rm ...", same as "make clean".
* GitHub Actions workflow:
* Ubuntu and Address Sanitizer jobs now use "make test". Remove
options which became redundant with "--fast-ci".
* Windows jobs now use --fast-ci option.
* Use -j0 to detect the number of CPUs.
* Set Makefile TESTTIMEOUT default to an empty string, since
--slow-ci and --fast-ci use different default timeout. It's now
accepted to pass "--timeout=" to regrtest: treated as not timeout.
* Tools/scripts/run_tests.py now uses --fast-ci option.
* Tools/buildbot/test.bat now uses --slow-ci option. Remove
--timeout=1200 option, redundant with --slow-ci.
test_error_at_task_unpickle() and
test_error_during_result_unpickle_in_result_handler() now restore
sys.stderr which is overriden by _raise_error_ignore_stderr().
Follow-up of gh-107219.
* Only close the connection writer on Windows.
* Also use existing constant _winapi.ERROR_OPERATION_ABORTED instead of
WSA_OPERATION_ABORTED.
AN Long
Victor Stinner
박문식
sunmy2019
Pablo Galindo Salgado
Sam Gross
Xuehai Pan
Hugo van Kemenade
Serhiy Storchaka
Nikita Sobolev
Alex Waygood
Eric Snow
elfstrom
Guido van Rossum
Jelle Zijlstra
Steve Dower
Charles Machalow
Barney Gale
Raymond Hettinger
Furkan Onder
James Hilton-Balfe
Pieter Eendebak
Irit Katriel
Davide Rizzo
Kristján Valur Jónsson
Dale Collison
Tom Gillespie
Antoine Pitrou