GH-140058: Clear key and value if `PyTuple_New` fails in `dictiter_iternextitem` (GH-140059)
(cherry picked from commit ded59f7e8e)
Co-authored-by: Sergey Miryanov <sergey.miryanov@gmail.com>
The free threading build uses QSBR to delay the freeing of dictionary
keys and list arrays when the objects are accessed by multiple threads
in order to allow concurrent reads to proceed with holding the object
lock. The requests are processed in batches to reduce execution
overhead, but for large memory blocks this can lead to excess memory
usage.
Take into account the size of the memory block when deciding when to
process QSBR requests.
Also track the amount of memory being held by QSBR for mimalloc pages.
Advance the write sequence if this memory exceeds a limit. Advancing
the sequence will allow it to be freed more quickly.
Process the held QSBR items from the "eval breaker", rather than from
`_PyMem_FreeDelayed()`. This gives a higher chance that the global read
sequence has advanced enough so that items can be freed.
(cherry picked from commit 113de8545f)
Co-authored-by: Neil Schemenauer <nas-github@arctrix.com>
Co-authored-by: Sam Gross <colesbury@gmail.com>
Use `ma_used` instead of `ma_keys->dk_nentries` for modification check
so that we only check if the dictionary is modified, not if new keys are
added to a different dictionary that shared the same keys object.
(cherry picked from commit d8994b0a77)
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)
The function `dict_set_fromkeys()` adds elements of a set to an existing
dictionary. The size of the expanded dictionary was estimated with
`PySet_GET_SIZE(iterable)`, which did not take into account the size of the
existing dictionary.
(cherry picked from commit 421ba589d0)
Co-authored-by: Angela Liss <59097311+angela-tarantula@users.noreply.github.com>
The clearing of the key, hash, and value need to use atomic operations
to avoid a data race with concurrent read operations.
(cherry picked from commit c00ac57824)
Co-authored-by: Sam Gross <colesbury@gmail.com>
gh-130547: Fix race between dict_dealloc and split_keys_entry_added (gh-130778)
(cherry picked from commit 80e6d3ec49)
Co-authored-by: Donghee Na <donghee.na@python.org>
Found while running `test_load_attr_module` from `test_opcache` under TSan.
(cherry picked from commit 34379d0a59)
Co-authored-by: Sam Gross <colesbury@gmail.com>
Objects may be temporarily "resurrected" in destructors when calling
finalizers or watcher callbacks. We previously undid the resurrection
by decrementing the reference count using `Py_SET_REFCNT`. This was not
thread-safe because other threads might be accessing the object
(modifying its reference count) if it was exposed by the finalizer,
watcher callback, or temporarily accessed by a racy dictionary or list
access.
This adds internal-only thread-safe functions for temporary object
resurrection during destructors.
(cherry picked from commit f4f530804b)
gh-116938: Fix `dict.update` docstring and remove erraneous full stop from `dict` documentation (GH-125421)
(cherry picked from commit 5527c4051c)
Co-authored-by: Prometheus3375 <35541026+Prometheus3375@users.noreply.github.com>
Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com>
Dictionary watchers on an object's attributes dictionary
(`object.__dict__`) were not triggered when the managed dictionary used
the object's inline values.
(cherry picked from commit 5989eb7446)
Co-authored-by: Sam Gross <colesbury@gmail.com>
gh-120974: Make asyncio `swap_current_task` safe in free-threaded build (GH-122317)
* gh-120974: Make asyncio `swap_current_task` safe in free-threaded build
(cherry picked from commit b5e6fb39a2)
Co-authored-by: Sam Gross <colesbury@gmail.com>
gh-120974: Make _asyncio._leave_task atomic in the free-threaded build (GH-122139)
* gh-120974: Make _asyncio._leave_task atomic in the free-threaded build
Update `_PyDict_DelItemIf` to allow for an argument to be passed to the
predicate.
(cherry picked from commit a15feded71)
Co-authored-by: Sam Gross <colesbury@gmail.com>
gh-121266: Remove Py_ALWAYS_INLINE in dictobject.c (GH-121493)
compare_unicode_generic(), compare_unicode_unicode() and
compare_generic() are callbacks used by do_lookup(). When enabling
assertions, it's not possible to inline these functions.
(cherry picked from commit c5a6b9afd8)
Co-authored-by: Victor Stinner <vstinner@python.org>
gh-121266: Change dict check_lookup() return type to int (GH-121581)
(cherry picked from commit 51da3dfbf3)
Co-authored-by: Victor Stinner <vstinner@python.org>
The object's inline values may be marked invalid if the materialized
dict was already initialized and then deleted.
(cherry picked from commit 162b41f577)
Co-authored-by: Sam Gross <colesbury@gmail.com>
The only remaining race in dictobject.c was in _PyDict_CheckConsistency
when the dictionary has shared keys.
(cherry picked from commit 3ec719fabf)
Co-authored-by: Sam Gross <colesbury@gmail.com>
Refactor the fast Unicode hash check into `_PyObject_HashFast` and use relaxed
atomic loads in the free-threaded build.
After this change, the TSAN doesn't report data races for this method.
(cherry picked from commit 294e724964)
Co-authored-by: AN Long <aisk@users.noreply.github.com>
PyDict_Next no longer locks the dictionary in the free-threaded build. Locking
around individual PyDict_Next calls is not sufficient because the function
returns borrowed references and because it allows concurrent modifications
during the iteraiton loop.
The internal locking also interferes with correct external synchronization
because it may suspend outer critical sections created by the caller.
(cherry picked from commit 375b723d58)
Co-authored-by: Sam Gross <colesbury@gmail.com>
* Add an InternalDocs file describing how interning should work and how to use it.
* Add internal functions to *explicitly* request what kind of interning is done:
- `_PyUnicode_InternMortal`
- `_PyUnicode_InternImmortal`
- `_PyUnicode_InternStatic`
* Switch uses of `PyUnicode_InternInPlace` to those.
* Disallow using `_Py_SetImmortal` on strings directly.
You should use `_PyUnicode_InternImmortal` instead:
- Strings should be interned before immortalization, otherwise you're possibly
interning a immortalizing copy.
- `_Py_SetImmortal` doesn't handle the `SSTATE_INTERNED_MORTAL` to
`SSTATE_INTERNED_IMMORTAL` update, and those flags can't be changed in
backports, as they are now part of public API and version-specific ABI.
* Add private `_only_immortal` argument for `sys.getunicodeinternedsize`, used in refleak test machinery.
* Make sure the statically allocated string singletons are unique. This means these sets are now disjoint:
- `_Py_ID`
- `_Py_STR` (including the empty string)
- one-character latin-1 singletons
Now, when you intern a singleton, that exact singleton will be interned.
* Add a `_Py_LATIN1_CHR` macro, use it instead of `_Py_ID`/`_Py_STR` for one-character latin-1 singletons everywhere (including Clinic).
* Intern `_Py_STR` singletons at startup.
* For free-threaded builds, intern `_Py_LATIN1_CHR` singletons at startup.
* Beef up the tests. Cover internal details (marked with `@cpython_only`).
* Add lots of assertions
Co-authored-by: Eric Snow <ericsnowcurrently@gmail.com>
This makes the following macros public as part of the non-limited C-API for
locking a single object or two objects at once.
* `Py_BEGIN_CRITICAL_SECTION(op)` / `Py_END_CRITICAL_SECTION()`
* `Py_BEGIN_CRITICAL_SECTION2(a, b)` / `Py_END_CRITICAL_SECTION2()`
The supporting functions and structs used by the macros are also exposed for
cases where C macros are not available.
(cherry picked from commit 8f17d69b7b)
gh-118849: Fix "code will never be executed" warning in `dictobject.c` (GH-118850)
(cherry picked from commit 82abe75e77)
Co-authored-by: Nikita Sobolev <mail@sobolevn.me>
Add _PyType_LookupRef and use incref before setting attribute on type
Makes setting an attribute on a class and signaling type modified atomic
Avoid adding re-entrancy exposing the type cache in an inconsistent state by decrefing after type is updated
This keeps track of the per-thread total reference count operations in
PyThreadState in the free-threaded builds. The count is merged into the
interpreter's total when the thread exits.
Most mutable data is protected by a striped lock that is keyed on the
referenced object's address. The weakref's hash is protected using the
weakref's per-object lock.
Note that this only affects free-threaded builds. Apart from some minor
refactoring, the added code is all either gated by `ifdef`s or is a no-op
(e.g. `Py_BEGIN_CRITICAL_SECTION`).
