Merge branch 'main' into fix/ubsan/exceptions-111178

Author: picnixz

.github/workflows/build.yml

@@ -247,13 +247,13 @@ jobs:
         - true
         os:
         - ubuntu-24.04
-        - ubuntu-24.04-arm
+        - ubuntu-22.04-arm
         exclude:
         # Do not test BOLT with free-threading, to conserve resources
         - bolt: true
           free-threading: true
         # BOLT currently crashes during instrumentation on aarch64
-        - os: ubuntu-24.04-arm
+        - os: ubuntu-22.04-arm
           bolt: true
     uses: ./.github/workflows/reusable-ubuntu.yml
     with:

.github/workflows/jit.yml

@@ -86,7 +86,7 @@ jobs:
             runner: ubuntu-24.04
           - target: aarch64-unknown-linux-gnu/gcc
             architecture: aarch64
-            runner: ubuntu-24.04-arm
+            runner: ubuntu-22.04-arm
     steps:
       - uses: actions/checkout@v4
         with:

.github/workflows/tail-call.yml

@@ -62,7 +62,7 @@ jobs:
             runner: ubuntu-24.04
           - target: aarch64-unknown-linux-gnu/gcc
             architecture: aarch64
-            runner: ubuntu-24.04-arm
+            runner: ubuntu-22.04-arm
     steps:
       - uses: actions/checkout@v4
         with:

Doc/library/ctypes.rst

@@ -1406,6 +1406,28 @@ the shared library name at development time, and hardcode that into the wrapper
 module instead of using :func:`~ctypes.util.find_library` to locate the library at runtime.
 
 
+.. _ctypes-listing-loaded-shared-libraries:
+
+Listing loaded shared libraries
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+When writing code that relies on code loaded from shared libraries, it can be
+useful to know which shared libraries have already been loaded into the current
+process.
+
+The :mod:`!ctypes.util` module provides the :func:`~ctypes.util.dllist` function,
+which calls the different APIs provided by the various platforms to help determine
+which shared libraries have already been loaded into the current process.
+
+The exact output of this function will be system dependent. On most platforms,
+the first entry of this list represents the current process itself, which may
+be an empty string.
+For example, on glibc-based Linux, the return may look like::
+
+   >>> from ctypes.util import dllist
+   >>> dllist()
+   ['', 'linux-vdso.so.1', '/lib/x86_64-linux-gnu/libm.so.6', '/lib/x86_64-linux-gnu/libc.so.6', ... ]
+
 .. _ctypes-loading-shared-libraries:
 
 Loading shared libraries
@@ -2083,6 +2105,20 @@ Utility functions
    .. availability:: Windows
 
 
+.. function:: dllist()
+   :module: ctypes.util
+
+   Try to provide a list of paths of the shared libraries loaded into the current
+   process.  These paths are not normalized or processed in any way.  The function
+   can raise :exc:`OSError` if the underlying platform APIs fail.
+   The exact functionality is system dependent.
+
+   On most platforms, the first element of the list represents the current
+   executable file. It may be an empty string.
+
+   .. availability:: Windows, macOS, iOS, glibc, BSD libc, musl
+   .. versionadded:: next
+
 .. function:: FormatError([code])
 
    Returns a textual description of the error code *code*.  If no error code is

Doc/library/dis.rst

@@ -703,15 +703,8 @@ not have to be) the original ``STACK[-2]``.
       STACK.append(lhs op rhs)
 
    .. versionadded:: 3.11
-
-
-.. opcode:: BINARY_SUBSCR
-
-   Implements::
-
-      key = STACK.pop()
-      container = STACK.pop()
-      STACK.append(container[key])
+   .. versionchanged:: 3.14
+      With oparg :``NB_SUBSCR``, implements binary subscript (replaces opcode ``BINARY_SUBSCR``)
 
 
 .. opcode:: STORE_SUBSCR

Doc/library/pathlib.rst

@@ -1177,6 +1177,38 @@ Querying file type and status
    .. versionadded:: 3.5
 
 
+.. attribute:: Path.info
+
+   A :class:`~pathlib.types.PathInfo` object that supports querying file type
+   information. The object exposes methods that cache their results, which can
+   help reduce the number of system calls needed when switching on file type.
+   For example::
+
+      >>> p = Path('src')
+      >>> if p.info.is_symlink():
+      ...     print('symlink')
+      ... elif p.info.is_dir():
+      ...     print('directory')
+      ... elif p.info.exists():
+      ...     print('something else')
+      ... else:
+      ...     print('not found')
+      ...
+      directory
+
+   If the path was generated from :meth:`Path.iterdir` then this attribute is
+   initialized with some information about the file type gleaned from scanning
+   the parent directory. Merely accessing :attr:`Path.info` does not perform
+   any filesystem queries.
+
+   To fetch up-to-date information, it's best to call :meth:`Path.is_dir`,
+   :meth:`~Path.is_file` and :meth:`~Path.is_symlink` rather than methods of
+   this attribute. There is no way to reset the cache; instead you can create
+   a new path object with an empty info cache via ``p = Path(p)``.
+
+   .. versionadded:: 3.14
+
+
 Reading and writing files
 ^^^^^^^^^^^^^^^^^^^^^^^^^
 
@@ -1903,3 +1935,56 @@ Below is a table mapping various :mod:`os` functions to their corresponding
 .. [4] :func:`os.walk` always follows symlinks when categorizing paths into
    *dirnames* and *filenames*, whereas :meth:`Path.walk` categorizes all
    symlinks into *filenames* when *follow_symlinks* is false (the default.)
+
+
+Protocols
+---------
+
+.. module:: pathlib.types
+   :synopsis: pathlib types for static type checking
+
+
+The :mod:`pathlib.types` module provides types for static type checking.
+
+.. versionadded:: 3.14
+
+
+.. class:: PathInfo()
+
+   A :class:`typing.Protocol` describing the
+   :attr:`Path.info <pathlib.Path.info>` attribute. Implementations may
+   return cached results from their methods.
+
+   .. method:: exists(*, follow_symlinks=True)
+
+      Return ``True`` if the path is an existing file or directory, or any
+      other kind of file; return ``False`` if the path doesn't exist.
+
+      If *follow_symlinks* is ``False``, return ``True`` for symlinks without
+      checking if their targets exist.
+
+   .. method:: is_dir(*, follow_symlinks=True)
+
+      Return ``True`` if the path is a directory, or a symbolic link pointing
+      to a directory; return ``False`` if the path is (or points to) any other
+      kind of file, or if it doesn't exist.
+
+      If *follow_symlinks* is ``False``, return ``True`` only if the path
+      is a directory (without following symlinks); return ``False`` if the
+      path is any other kind of file, or if it doesn't exist.
+
+   .. method:: is_file(*, follow_symlinks=True)
+
+      Return ``True`` if the path is a file, or a symbolic link pointing to
+      a file; return ``False`` if the path is (or points to) a directory or
+      other non-file, or if it doesn't exist.
+
+      If *follow_symlinks* is ``False``, return ``True`` only if the path
+      is a file (without following symlinks); return ``False`` if the path
+      is a directory or other other non-file, or if it doesn't exist.
+
+   .. method:: is_symlink()
+
+      Return ``True`` if the path is a symbolic link (even if broken); return
+      ``False`` if the path is a directory or any kind of file, or if it
+      doesn't exist.

Doc/library/pdb.rst

@@ -697,6 +697,17 @@ can be overridden by the local file.
 .. pdbcommand:: q(uit)
 
    Quit from the debugger.  The program being executed is aborted.
+   An end-of-file input is equivalent to :pdbcmd:`quit`.
+
+   A confirmation prompt will be shown if the debugger is invoked in
+   ``'inline'`` mode. Either ``y``, ``Y``, ``<Enter>`` or ``EOF``
+   will confirm the quit.
+
+   .. versionchanged:: 3.14
+      A confirmation prompt will be shown if the debugger is invoked in
+      ``'inline'`` mode. After the confirmation, the debugger will call
+      :func:`sys.exit` immediately, instead of raising :exc:`bdb.BdbQuit`
+      in the next trace event.
 
 .. pdbcommand:: debug code
 

Doc/using/configure.rst

@@ -618,6 +618,16 @@ also be used to improve performance.
    Enable computed gotos in evaluation loop (enabled by default on supported
    compilers).
 
+.. option:: --with-tail-call-interp
+
+   Enable interpreters using tail calls in CPython. If enabled, enabling PGO
+   (:option:`--enable-optimizations`) is highly recommended. This option specifically
+   requires a C compiler with proper tail call support, and the
+   `preserve_none <https://clang.llvm.org/docs/AttributeReference.html#preserve-none>`_
+   calling convention. For example, Clang 19 and newer supports this feature.
+
+   .. versionadded:: next
+
 .. option:: --without-mimalloc
 
    Disable the fast :ref:`mimalloc <mimalloc>` allocator

Doc/whatsnew/3.14.rst

@@ -68,6 +68,7 @@ Summary -- release highlights
 * :ref:`PEP 649: deferred evaluation of annotations <whatsnew314-pep649>`
 * :ref:`PEP 741: Python Configuration C API <whatsnew314-pep741>`
 * :ref:`PEP 761: Discontinuation of PGP signatures <whatsnew314-pep761>`
+* :ref:`A new type of interpreter  <whatsnew314-tail-call>`
 
 
 New features
@@ -208,6 +209,37 @@ configuration mechanisms).
 .. seealso::
    :pep:`741`.
 
+.. _whatsnew314-tail-call:
+
+A new type of interpreter
+-------------------------
+
+A new type of interpreter based on tail calls has been added to CPython.
+For certain newer compilers, this interpreter provides
+significantly better performance. Preliminary numbers on our machines suggest
+anywhere from -3% to 30% faster Python code, and a geometric mean of 9-15%
+faster on ``pyperformance`` depending on platform and architecture.
+
+This interpreter currently only works with Clang 19 and newer
+on x86-64 and AArch64 architectures. However, we expect
+that a future release of GCC will support this as well.
+
+This feature is opt-in for now. We highly recommend enabling profile-guided
+optimization with the new interpreter as it is the only configuration we have
+tested and can validate its improved performance.
+For further information on how to build Python, see
+:option:`--with-tail-call-interp`.
+
+.. note::
+
+   This is not to be confused with `tail call optimization`__ of Python
+   functions, which is currently not implemented in CPython.
+
+   __ https://en.wikipedia.org/wiki/Tail_call
+
+(Contributed by Ken Jin in :gh:`128718`, with ideas on how to implement this
+in CPython by Mark Shannon, Garrett Gu, Haoran Xu, and Josh Haberman.)
+
 
 Other language changes
 ======================
@@ -357,6 +389,9 @@ ctypes
   complex C types.
   (Contributed by Sergey B Kirpichev in :gh:`61103`).
 
+* Add :func:`ctypes.util.dllist` for listing the shared libraries
+  loaded by the current process.
+  (Contributed by Brian Ward in :gh:`119349`.)
 
 datetime
 --------
@@ -585,6 +620,15 @@ pathlib
 
   (Contributed by Barney Gale in :gh:`73991`.)
 
+* Add :attr:`pathlib.Path.info` attribute, which stores an object
+  implementing the :class:`pathlib.types.PathInfo` protocol (also new). The
+  object supports querying the file type and internally caching
+  :func:`~os.stat` results. Path objects generated by
+  :meth:`~pathlib.Path.iterdir` are initialized with file type information
+  gleaned from scanning the parent directory.
+
+  (Contributed by Barney Gale in :gh:`125413`.)
+
 
 pdb
 ---
@@ -600,6 +644,11 @@ pdb
   command when :mod:`pdb` is in ``inline`` mode.
   (Contributed by Tian Gao in :gh:`123757`.)
 
+* A confirmation prompt will be shown when the user tries to quit :mod:`pdb`
+  in ``inline`` mode. ``y``, ``Y``, ``<Enter>`` or ``EOF`` will confirm
+  the quit and call :func:`sys.exit`, instead of raising :exc:`bdb.BdbQuit`.
+  (Contributed by Tian Gao in :gh:`124704`.)
+
 
 pickle
 ------
@@ -1167,6 +1216,12 @@ Others
   :meth:`~object.__index__`. (Contributed by Mark Dickinson in :gh:`119743`.)
 
 
+CPython Bytecode Changes
+========================
+
+* Replaced the opcode ``BINARY_SUBSCR`` by :opcode:`BINARY_OP` with oparg ``NB_SUBSCR``.
+  (Contributed by Irit Katriel in :gh:`100239`.)
+
 Porting to Python 3.14
 ======================
 

Include/cpython/pystats.h

@@ -129,6 +129,7 @@ typedef struct _optimization_stats {
     uint64_t inner_loop;
     uint64_t recursive_call;
     uint64_t low_confidence;
+    uint64_t unknown_callee;
     uint64_t executors_invalidated;
     UOpStats opcode[PYSTATS_MAX_UOP_ID + 1];
     uint64_t unsupported_opcode[256];