Fix free-threading scaling bottleneck in sys.intern and `PyObject_SetAttr` by
avoiding the interpreter-wide lock when the string is already interned and
immortalized.
* We don't specify what happens on non-IEEE platforms.
* Use rather PY_LITTLE_ENDIAN to get native endianness.
* Mention that unpack functions don't fail in CPython.
* Mention that PyFloat_Pack8 doesn't fail in CPython.
Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com>
Co-authored-by: Hugo van Kemenade <1324225+hugovk@users.noreply.github.com>
Co-authored-by: Victor Stinner <vstinner@python.org>
Add special cases for classmethod and staticmethod descriptors in
_PyObject_GetMethodStackRef() to avoid calling tp_descr_get, which
avoids reference count contention on the bound method and underlying
callable. This improves scaling when calling classmethods and
staticmethods from multiple threads.
Also refactor method_vectorcall in classobject.c into a new _PyObject_VectorcallPrepend() helper so that it can be used by
PyObject_VectorcallMethod as well.
* Add the c_init_default attribute which is used to initialize the C variable
if the default is not explicitly provided.
* Add the c_default_init() method which is used to derive c_default from
default if c_default is not explicitly provided.
* Explicit c_default and py_default are now almost always have precedence
over the generated value.
* Add support for bytes literals as default values.
* Improve support for str literals as default values (support non-ASCII
and non-printable characters and special characters like backslash or quotes).
* Fix support for str and bytes literals containing trigraphs, "/*" and "*/".
* Improve support for default values in converters "char" and "int(accept={str})".
* Converter "int(accept={str})" now requires 1-character string instead of
integer as default value.
* Add support for non-None default values in converter "Py_buffer": NULL,
str and bytes literals.
* Improve error handling for invalid default values.
* Rename Null to NullType for consistency.
Modifies the Emscripten build script to allow for caching of dependencies, and
for automated installation of new EMSDK versions.
Co-authored-by: Russell Keith-Magee <russell@keith-magee.com>
Modifies the Emscripten build script to allow for custom cross-build directory
names, and to only clean Emscripten-specific paths (optionally including the
build python).
Co-authored-by: Russell Keith-Magee <russell@keith-magee.com>
Adds an `--emsdk-cache` argument to the Emscripten build script and an
emscripten_version.txt file. If the `--emsdk-cache` argument is passed, the build script
will look in `emscripten_version.txt` to get the expected emsdk version is installed in
a folder called e.g., 4.0.12 in the directory indicated by the `--emsdk-cache` argument,
and run the build with that Emscripten tooling activated.
Remove PyThread_type_lock (now uses PyMutex internally).
Add new benchmark options:
- work_inside/work_outside: control work inside and outside the critical section to vary contention levels
- num_locks: use multiple independent locks with threads assigned round-robin
- total_iters: fixed iteration count per thread instead of time-based, useful for measuring fairness
- num_acquisitions: lock acquisitions per loop iteration
- random_locks: acquire random lock each iteration
Also return elapsed time from benchmark_locks() and switch lockbench.py to use argparse.
