790a46a449
gh-134584: JIT: Eliminate redundant refcount ops for X_INT (GH-142765)
2025-12-15 22:22:18 +00:00
059316ade6
gh-134584: Eliminate redundant refcounting from _STORE_ATTR_INSTANCE_VALUE (GH-142759)
...
Signed-off-by: Manjusaka <me@manjusaka.me >
2025-12-15 20:17:12 +00:00
0ac4e6c6cd
gh-134584: Remove custom float decref ops (GH-142576)
2025-12-15 19:38:58 +00:00
27a2e49d18
GH-142718: JIT: Fix crash due to incorrect caching on side exits when exiting jitted code. (GH-142762)
...
JIT: Fix crash due to incorrect caching on side exits when exiting jitted code.
* Make sure that stack is in correct state at side exits with TOS cached values
* Simplify choice of cached items for side exits
2025-12-15 19:01:29 +00:00
a882ae198a
gh-142472: Clean-up _PyStackRef functions (gh-142479)
...
This combines most _PyStackRef functions and macros between the free
threaded and default builds.
- Remove Py_TAG_DEFERRED (same as Py_TAG_REFCNT)
- Remove PyStackRef_IsDeferred (same as !PyStackRef_RefcountOnObject)
2025-12-15 12:03:49 -05:00
abaaeee6a0
gh-134584: Eliminate redundant refcounting from _STORE_SUBSCR_DICT (GH-142712)
...
Co-authored-by: Ken Jin <kenjin4096@gmail.com >
2025-12-15 16:42:32 +00:00
872ab51f32
gh-134584: Eliminate redundant refcounting from _CALL_LIST_APPEND (GH-142711)
...
Signed-off-by: Manjusaka <me@manjusaka.me >
Co-authored-by: Ken Jin <kenjin4096@gmail.com >
2025-12-15 15:49:36 +00:00
cd2ca74c53
gh-142029: Raise ModuleNotFoundError instead of crashing on nonexsistent module name given to create_builtin() ( #142054 )
...
Co-authored-by: Brett Cannon <brett@python.org >
Co-authored-by: Victor Stinner <vstinner@python.org >
2025-12-15 13:20:19 +01:00
fb554ad68d
gh-140414: streamline thread state access in asyncio ( #142742 )
2025-12-15 16:57:51 +05:30
04da416e6b
gh-134584: Eliminate redundant refcounting from _STORE_SUBSCR_LIST_INT (gh-142703)
2025-12-14 12:26:00 +00:00
a2a400af1e
gh-134584: Eliminate redundant refcounting from _CALL_BUILTION_O (GH-142695)
...
Co-authored-by: Ken Jin <kenjin4096@gmail.com >
2025-12-14 10:28:02 +00:00
e90e6d155a
gh-142666: Remove unused variable package in import logic (GH-142667)
...
The variable was previously used, but became unused after 133138a284yongtaoh2022@gmail.com >
2025-12-14 11:39:45 +02:00
a154c9ed4e
gh-134584: Eliminate redundant refcounting from _CALL_STR_1 (GH-136070)
...
Signed-off-by: Manjusaka <me@manjusaka.me >
2025-12-14 09:33:05 +00:00
e02a35c365
gh-134584: Cleanups for GH-135860 (GH-142604)
2025-12-13 14:38:10 +00:00
15313dd3d7
gh-140550: Correct error message for PyModExport (PEP 793) hook (GH-142583)
2025-12-12 17:48:43 +01:00
a3a611b042
gh-134584: Revert partially GH-135860 (GH-142620)
2025-12-12 14:04:11 +00:00
e0bca091a4
gh-142627: Ignore anonymous mappings in Linux remote debugging ( #142628 )
2025-12-12 13:12:11 +00:00
2a820e2b9c
fix typos in crossinterp.c and qsbr.c ( #142612 )
2025-12-12 11:48:20 +05:30
9fe6e3ed36
gh-142571: Check for errors before calling each syscall in PyUnstable_CopyPerfMapFile() ( #142460 )
...
Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com >
Co-authored-by: Victor Stinner <vstinner@python.org >
Co-authored-by: Pablo Galindo Salgado <Pablogsal@gmail.com >
2025-12-11 21:18:52 +00:00
e38967ed60
gh-142531: Fix free-threaded GC performance regression (gh-142562)
...
If there are many untracked tuples, the GC will run too often, resulting
in poor performance. The fix is to include untracked tuples in the
"long lived" object count. The number of frozen objects is also now
included since the free-threaded GC must scan those too.
2025-12-11 12:30:56 -08:00
af185727b2
GH-65961: Stop setting __cached__ on modules (GH-142165)
2025-12-11 11:44:46 -08:00
a27538540e
gh-134584: Eliminate redundant refcounting from `_CALL_LEN` (gh-136104)
2025-12-11 15:24:34 +00:00
a78f43b001
gh-134584: Eliminate redundant refcounting from _CALL_TUPLE_1 (GH-135860)
2025-12-11 14:31:28 +00:00
4eab90f4f3
GH-140683: JIT: Improve machine code for loading smaller constants on AArch64. (GH-142511)
...
* Use movz and movk instructions for loading 16 and 32 bit operands and oparg.
* Loading of 64 bit operands is unchanged.
2025-12-11 12:33:39 +00:00
469f191a85
GH-135379: Top of stack caching for the JIT. (GH-135465)
...
Uses three registers to cache values at the top of the evaluation stack
This significantly reduces memory traffic for smaller, more common uops.
2025-12-11 10:32:52 +00:00
97e19014dd
gh-137007: Track executor before any possible deallocations (GH-137016)
2025-12-11 05:09:56 +08:00
ebf3427615
gh-141976: Protect against non-progressing specializations in tracing JIT (GH-141989)
2025-12-10 19:39:11 +00:00
46295677a1
GH-142305: JIT: Deduplicating GOT symbols in the trace ( #142316 )
2025-12-10 16:04:04 +00:00
49b1fb43f6
gh-142048: Fix lost gc allocations count on thread cleanup ( #142233 )
2025-12-10 07:29:40 +00:00
70671267c1
gh-142029: Raise ValueError instead of crashing on empty name given to create_builtin() ( #142033 )
...
Co-authored-by: Victor Stinner <vstinner@python.org >
2025-12-10 12:01:57 +05:30
97f0a1f203
gh-142276: Watch attribute loads when promoting JIT constants (GH-142303)
...
Co-authored-by: blurb-it[bot] <43283697+blurb-it[bot]@users.noreply.github.com>
Co-authored-by: Savannah Ostrowski <savannah@python.org >
2025-12-08 18:03:15 +00:00
e0451ceef8
GH-139757: JIT: Remove redundant branches to jumps in the assembly optimizer (GH-140800)
...
JIT: Remove redundant branches to jump in the assembly optimizer
* Refactor JIT assembly optimizer making instructions instances not just strings
* Remove redundant jumps and branches where legal to do so
* Modifies _BINARY_OP_SUBSCR_STR_INT to avoid excessive inlining depth
2025-12-08 17:57:11 +00:00
c4ccaf4b10
gh-141770: Annotate anonymous mmap usage if "-X dev" is used (gh-142079)
2025-12-08 14:47:19 +00:00
d6d850df89
gh-138122: Don't sample partial frame chains ( #141912 )
2025-12-07 15:53:48 +00:00
572c780aa8
gh-138122: Implement frame caching in RemoteUnwinder to reduce memory reads ( #142137 )
...
This PR implements frame caching in the RemoteUnwinder class to significantly reduce memory reads when profiling remote processes with deep call stacks.
When cache_frames=True, the unwinder stores the frame chain from each sample and reuses unchanged portions in subsequent samples. Since most profiling samples capture similar call stacks (especially the parent frames), this optimization avoids repeatedly reading the same frame data from the target process.
The implementation adds a last_profiled_frame field to the thread state that tracks where the previous sample stopped. On the next sample, if the current frame chain reaches this marker, the cached frames from that point onward are reused instead of being re-read from remote memory.
The sampling profiler now enables frame caching by default.
2025-12-06 22:37:34 +00:00
35142b18ae
gh-142168: explicitly initialize stack_array in _PyEval_Vector and _PyEvalFramePushAndInit_Ex ( #142192 )
...
Co-authored-by: Kir Chou <note351@hotmail.com >
2025-12-06 19:59:52 +01:00
706fdda8b3
gh-141370: Fix undefined behavior when using Py_ABS() (GH-141548)
...
Co-authored-by: Sergey B Kirpichev <skirpichev@gmail.com >
2025-12-05 16:24:35 +02:00
b3bf212898
gh-141976: Check stack bounds in JIT optimizer (GH-142201)
2025-12-04 20:28:08 +00:00
6825d5c11d
GH-139757: Fix reference leaks introduced in GH-140800 (GH-142257)
2025-12-04 12:27:15 +00:00
62423c9c36
GH-141794: Limit size of generated machine code. (GH-142228)
...
* Factor out bodies of the largest uops, to reduce jit code size.
* Factor out common assert, also reducing jit code size.
* Limit size of jitted code for a single executor to 1MB.
2025-12-03 17:43:35 +00:00
7e5fcae09b
gh-142217: Remove internal _Py_Identifier functions ( #142219 )
...
Remove internal functions:
* _PyDict_ContainsId()
* _PyDict_DelItemId()
* _PyDict_GetItemIdWithError()
* _PyDict_SetItemId()
* _PyEval_GetBuiltinId()
* _PyObject_CallMethodIdNoArgs()
* _PyObject_CallMethodIdObjArgs()
* _PyObject_CallMethodIdOneArg()
* _PyObject_VectorcallMethodId()
* _PyUnicode_EqualToASCIIId()
These functions were not exported and so no usable outside CPython.
2025-12-03 14:33:32 +01:00
eb892868b3
gh-142048: Fix quadratically increasing GC delays (gh-142051)
...
The GC for the free threaded build would get slower with each collection due
to effectively double counting objects freed by the GC.
2025-12-01 19:04:47 -05:00
d5d9e89dde
gh-116008: Detect freed thread state in faulthandler ( #141988 )
...
Add _PyMem_IsULongFreed() function.
2025-11-27 12:35:00 +01:00
83d8134c5b
gh-127635: Use flexible array in tracemalloc ( #141991 )
...
Replace frames[1] with frames[] in tracemalloc_traceback structure.
2025-11-27 12:32:31 +01:00
2ea67caf31
GH-141861: Fix TRACE_RECORD if full (GH-141959)
2025-11-26 14:32:30 +00:00
27f62eb711
gh-140011: Delete importdl assertion that prevents importing embedded modules from packages (GH-141605)
2025-11-26 14:12:49 +01:00
d07d3a3c57
gh-138122: Split Modules/_remote_debugging_module.c into multiple files ( #141934 )
...
gh-1381228: Split Modules/_remote_debugging_module.c into multiple files
2025-11-25 12:51:24 +00:00
dc62b62252
GH-141861: Fix invalid memory read in the ENTER_EXECUTOR (GH-141921)
2025-11-24 22:07:45 +00:00
bf66bce4ee
gh-141780: Make PyModule_FromSlotsAndSpec enable GIL if needed (GH-141785)
2025-11-24 13:26:35 +01:00
e457d60daa
gh-120158: Fix inconsistent monitoring state when setting events too frequently (gh-141845)
...
If we overflowed the global version counter (i.e., after 2*24 calls to
`_PyMonitoring_SetEvents`), we bailed out after setting global monitoring
events but before instrumenting code objects, which led to assertion errors
later on.
Also add a `time.sleep()` to `test_free_threading.test_monitoring` to avoid
overflowing the global version counter.
2025-11-23 10:07:17 -05:00
Ken Jin