Some systems have the definitions of the mask bits without having the
corresponding members in struct statx. Add configure checks for members
added after Linux 4.11 (when statx itself was added).
stx_atomic_write_unit_max_opt was added in Linux 6.16, but is controlled
by the STATX_WRITE_ATOMIC mask bit added in Linux 6.11. That's safe at
runtime because all kernels clear the reserved space in struct statx and
zero is a valid value for stx_atomic_write_unit_max_opt, and it avoids
allocating another mask bit, which are a limited resource. But it also
means the kernel headers don't provide a way to check whether
stx_atomic_write_unit_max_opt exists, so add a configure check.
On modern systems, the result of wcsxfrm() is much larger the size of
the input string (from 4+2*n on Windows to 4+5*n on Linux for simple
ASCII strings), so optimistic allocation of the buffer of the same size
never works.
The exception is if the locale is "C" (or unset), but in that case the `wcsxfrm`
call should be fast (and calling `locale.strxfrm()` doesn't make too much
sense in the first place).
This fixes a regression introduced by GH-136004, in which finalization would hang while executing atexit handlers if the system was out of memory.
---------
Signed-off-by: yihong0618 <zouzou0208@gmail.com>
Co-authored-by: Peter Bierma <zintensitydev@gmail.com>
Co-authored-by: Victor Stinner <vstinner@python.org>
The previous `Py_REFCNT(x) == 1` checks can have data races in the free
threaded build. `_PyObject_IsUniquelyReferenced(x)` is a more conservative
check that is safe in the free threaded build and is identical to
`Py_REFCNT(x) == 1` in the default GIL-enabled build.
Previously, the _BlocksOutputBuffer code creates a list of bytes objects to handle the output data from compression libraries. This ends up being slow due to the output buffer code needing to copy each bytes element of the list into the final bytes object buffer at the end of compression.
The new PyBytesWriter API introduced in PEP 782 is an ergonomic and fast method of writing data into a buffer that will later turn into a bytes object. Benchmarks show that using the PyBytesWriter API is 10-30% faster for decompression across a variety of settings. The performance gains are greatest when the decompressor is very performant, such as for Zstandard (and likely zlib-ng). Otherwise the decompressor can bottleneck decompression and the gains are more modest, but still sizable (e.g. 10% faster for zlib)!
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
Fix incorrect sharing of current task with the forked child process by clearing thread state's current task and current loop in `PyOS_AfterFork_Child`.
* Modules/pyexpat.c: Disallow collection of in-use parent parsers.
Within libexpat, a parser created via `XML_ExternalEntityParserCreate`
is relying on its parent parser throughout its entire lifetime.
Prior to this fix, is was possible for the parent parser to be
garbage-collected too early.
Fix a compiler warning `-Wunused-function` after f04bea44c3.
The `set_invalid_arg` function in `Modules/pyexpat.c` may be unused if the underlying Expat
version is less than 2.4.0.
Passing a negative or zero size to `cursor.fetchmany()` made it fetch all rows
instead of none.
While this could be considered a security vulnerability, it was decided to treat
this issue as a regular bug as passing a non-sanitized *size* value in the first
place is not recommended.
Expose the XML Expat 2.7.2 APIs to tune protections against
"billion laughs" [1] attacks.
The exposed APIs are available on Expat parsers, that is,
parsers created by `xml.parsers.expat.ParserCreate()`, as:
- `parser.SetBillionLaughsAttackProtectionActivationThreshold(threshold)`, and
- `parser.SetBillionLaughsAttackProtectionMaximumAmplification(max_factor)`.
This completes the work in f04bea44c3,
and improves the existing related documentation.
[1]: https://en.wikipedia.org/wiki/Billion_laughs_attack
