8614f86b71
gh-131525: Cache the result of tuple_hash ( #131529 )
...
* gh-131525: Cache the result of tuple_hash
* Fix debug builds
* Add blurb
* Fix formatting
* Pre-compute empty tuple singleton
* Mostly set the cache within tuple_alloc
* Fixes for TSAN
* Pre-compute empty tuple singleton
* Fix for 32-bit platforms
* Assert that op != NULL in _PyTuple_RESET_HASH_CACHE
* Use FT_ATOMIC_STORE_SSIZE_RELAXED macro
* Update Include/internal/pycore_tuple.h
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
* Fix alignment
* atomic load
* Update Objects/tupleobject.c
Co-authored-by: Chris Eibl <138194463+chris-eibl@users.noreply.github.com>
---------
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
Co-authored-by: Chris Eibl <138194463+chris-eibl@users.noreply.github.com>
2025-03-27 09:57:06 -04:00
d836d287a7
GH-131729: Consider in-memory state when merging storage and stack (GH-131773)
2025-03-27 08:32:45 +00:00
151d1bfd1b
gh-131763: Replace the redundant check with assert in remove_tools ( #131765 )
2025-03-26 18:36:04 -04:00
67fbfb42bd
gh-131586: Avoid refcount contention in some "special" calls ( #131588 )
...
In the free threaded build, the `_PyObject_LookupSpecial()` call can lead to
reference count contention on the returned function object becuase it
doesn't use stackrefs. Refactor some of the callers to use
`_PyObject_MaybeCallSpecialNoArgs`, which uses stackrefs internally.
This fixes the scaling bottleneck in the "lookup_special" microbenchmark
in `ftscalingbench.py`. However, the are still some uses of
`_PyObject_LookupSpecial()` that need to be addressed in future PRs.
2025-03-26 14:38:47 -04:00
3d4ac1a2c2
gh-123358: Use `_PyStackRef` in `LOAD_DEREF` (gh-130064)
...
Concurrent accesses from multiple threads to the same `cell` object did not
scale well in the free-threaded build. Use `_PyStackRef` and optimistically
avoid locking to improve scaling.
With the locks around cell reads gone, some of the free threading tests were
prone to starvation: the readers were able to run in a tight loop and the
writer threads weren't scheduled frequently enough to make timely progress.
Adjust the tests to avoid this.
Co-authored-by: Donghee Na <donghee.na@python.org>
2025-03-26 12:08:20 -04:00
1b8bb1ed0c
GH-131729: Code-gen better liveness analysis (GH-131732)
...
* Rename 'defined' attribute to 'in_local' to more accurately reflect how it is used
* Make death of variables explicit even for array variables.
* Convert in_memory from boolean to stack offset
* Don't apply liveness analysis to optimizer generated code
* Fix RETURN_VALUE in optimizer
2025-03-26 15:21:35 +00:00
898e6b395e
gh-130881: Handle conditionally defined annotations ( #130935 )
2025-03-26 03:48:19 +00:00
7bb41aef4b
gh-131740: Update PyUnstable_GC_VisitObjects to traverse perm gen (gh-131744)
2025-03-26 09:45:29 +09:00
929afd1d6e
gh-131670: Fix crash in `anext()` when `__anext__` is sync and raises ( #131682 )
...
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-03-24 19:00:48 +00:00
da6730caa5
gh-128421: Avoid TSAN warnings in `sys._current_frames()` (gh-131548)
...
This tells TSAN not to sanitize `PyUnstable_InterpreterFrame_GetLine()`.
There's a possible data race on the access to the frame's `instr_ptr`
if the frame is currently executing. We don't really care about the
race. In theory, we could use relaxed atomics for every access to
`instr_ptr`, but that would create more code churn and current compilers
are overly conservative with optimizations around relaxed atomic
accesses.
We also don't sanitize `_PyFrame_IsIncomplete()` because it accesses
`instr_ptr` and is called from assertions within PyFrame_GetCode().
2025-03-24 09:49:39 -04:00
511a844028
gh-111178: fix UBSan failures for `Python/legacy_tracing.c` ( #131611 )
2025-03-24 11:00:32 +01:00
43fde78bef
gh-111178: fix UBSan failures for `Python/instrumentation.c` ( #131608 )
2025-03-24 10:58:33 +01:00
f65be097ed
gh-111178: fix UBSan failures for `_PyExecutorObject` ( #131610 )
2025-03-24 10:53:23 +01:00
64906bb223
gh-130080: do not fold match case constants in unoptimized AST ( #131577 )
2025-03-23 13:50:14 +00:00
9962469943
gh-131296: fix clang-cl warning in tracemalloc.c ( #131514 )
...
Always set MAX_NFRAME to UINT16_MAX.
Avoid the complicated code which emitted a compiler warning.
2025-03-22 10:38:47 +01:00
49fb75c676
gh-131238: Add missing pycore_function.h includes for JIT compiler ( #131571 )
2025-03-21 23:37:49 +00:00
d9411ae3c2
gh-130312: SET_ADD should not lock ( #130136 )
...
SET_ADD should not lock
2025-03-21 15:58:32 -07:00
7101cba6bf
gh-131238: Move _Py_VISIT_STACKREF() to pycore_stackref.h ( #131560 )
...
* Move _Py_VISIT_STACKREF() from pycore_gc.h to pycore_stackref.h.
* Remove pycore_interpframe.h include from pycore_genobject.h.
* Remove now useless includes from C files.
* Add pycore_interpframe_structs.h to Makefile.pre.in and
pythoncore.vcxproj.
2025-03-21 23:24:14 +01:00
b92ee14b80
GH-130415: Optimize constant comparison in JIT builds (GH-131489)
2025-03-21 11:23:12 -07:00
0de5e0c544
gh-117657: Fix TSAN data race in _PyEval_SetTrace assertion (gh-131561)
...
The `sys_tracing_threads` variable should be read inside `LOCK_SETUP()`.
2025-03-21 14:22:37 -04:00
61317074d4
gh-131238: Add pycore_interpframe_structs.h header ( #131553 )
...
Add an explicit include to pycore_interpframe_structs.h in
pycore_runtime_structs.h to fix a dependency cycle.
2025-03-21 17:19:47 +00:00
4f32516804
gh-128421: Add locking to most frame object functions (gh-131479)
...
This makes more operations on frame objects thread-safe in the free
threaded build, which fixes some data races that occurred when passing
exceptions between threads.
However, accessing local variables from another thread while its running
is still not thread-safe and may crash the interpreter.
2025-03-21 11:10:07 -04:00
883c2f682b
GH-131331: Rename "not" to "invert" (GH-131334)
2025-03-20 16:59:41 -07:00
a2ea417578
GH-131296: fix clang-cl warnings in sysmodule.c ( #131304 )
2025-03-20 16:29:21 +00:00
7ebd71ee14
GH-131498: Remove conditional stack effects (GH-131499)
...
* Adds some missing #includes
2025-03-20 15:39:38 +00:00
83d54fa876
GH-130296: Remove `_PyOpcode_max_stack_effect` as it is no longer used (GH-131493)
2025-03-20 11:42:54 +00:00
b69da006a4
gh-131238: Remove includes from pycore_interp.h ( #131495 )
...
Remove also now unused includes in C files.
2025-03-20 11:35:23 +00:00
34c1ea3109
gh-111178: Fix function signatures for multiple tests ( #131496 )
2025-03-20 12:27:03 +01:00
486d537065
Fix Windows build warnings ( #131487 )
...
Fix the following warnings:
* Modules\_io\fileio.c(1296,13): unused variable 'self'
* Modules\_io\winconsoleio.c(334,9): unused variable 'fd_is_own'
* Modules\faulthandler.c(409,11): unused variable 'flags'
* Modules\posixmodule.c(5699,9): unused variable 'pathError'
* PC\winreg.c(2077,5): suggest braces around initialization of
subobject
* PC\winreg.c(34,13): unused variable 'errNotAHandle'
* Python\fileutils.c(132,12): result of comparison of constant
1114111 with expression of type 'wchar_t' (aka 'unsigned short') is
always false
* Python\fileutils.c(58,21): unused variable 'INCOMPLETE_CHARACTER'
* Python\sysmodule.c(2534,21): unused variable 'perf_map_state'
2025-03-20 12:03:20 +01:00
39b37b0110
gh-128421: add critical section around `traceback.tb_next` ( #131322 )
2025-03-20 12:03:54 +05:30
b12af0a48f
gh-131401: fix data races in exception handling ( #131447 )
2025-03-20 12:02:05 +05:30
20c5f969dd
gh-131238: Remove more includes from pycore_interp.h ( #131480 )
2025-03-19 23:01:32 +01:00
75103d975c
gh-126835: Move constant tuple folding from ast_opt to CFG ( #130769 )
2025-03-19 20:59:55 +00:00
8a33034d82
GH-130956: Only emit AArch64 trampolines for long jumps (GH-131041)
2025-03-19 12:19:21 -07:00
22706843e0
gh-131238: Remove many includes from pycore_interp.h ( #131472 )
2025-03-19 17:46:24 +00:00
5c44d7d99c
gh-130931: Add pycore_interpframe.h internal header ( #131249 )
...
Move _PyInterpreterFrame and associated functions
to a new pycore_interpframe.h header.
2025-03-19 18:17:44 +01:00
4b54031323
gh-131238: Remove pycore_runtime.h from pycore_pystate.h ( #131356 )
...
* Remove includes from pycore_pystate.h:
* pycore_runtime_structs.h
* pycore_runtime.h
* pycore_tstate.h
* pycore_interp.h
* Reorganize internal headers. Move _gc_thread_state from
pycore_interp_structs.h to pycore_tstate.h.
* Add 3 new header files to PCbuild/pythoncore.vcxproj.
2025-03-19 17:33:24 +01:00
0a54bd6dd7
gh-111178: Fix function signatures in legacy_tracing.c ( #131464 )
2025-03-19 16:54:41 +01:00
bbe5baad6c
gh-111178: Fix function signatures for test_types ( #131455 )
2025-03-19 13:46:17 +00:00
74b87515a7
gh-131401: fix data race in `_PyErr_Restore` ( #131406 )
2025-03-19 14:35:47 +05:30
ffc2f1dd1c
gh-130080: implement PEP 765 ( #130087 )
2025-03-17 20:48:54 +00:00
b2ed7a6d6a
gh-131281: Add include for pystats builds ( #131369 )
...
Add include to for pystats builds
2025-03-18 00:36:06 +08:00
a45f25361d
GH-131238: More refactoring of core header files (GH-131351)
...
Adds new pycore_stats.h header file to help break dependencies involving the pycore_code.h header.
2025-03-17 14:41:05 +00:00
978e37bb5f
gh-131238: Add explicit includes to pycore headers ( #131257 )
2025-03-17 12:32:43 +01:00
0453e494b6
gh-131238: Convert pycore_pystate.h static inline to functions ( #131352 )
...
Convert static inline functions to functions:
* _Py_IsMainThread()
* _PyInterpreterState_Main()
* _Py_IsMainInterpreterFinalizing()
* _Py_GetMainConfig()
2025-03-17 12:31:55 +01:00
a1aeec61c4
GH-131238: Core header refactor (GH-131250)
...
* Moves most structs in pycore_ header files into pycore_structs.h and pycore_runtime_structs.h
* Removes many cross-header dependencies
2025-03-17 09:19:04 +00:00
f104c19a94
GH-131296: fix clang-cl warning on Windows in pytime.c (GH-131297)
...
fix warning : integer literal is too large
to be represented in a signed integer type,
interpreting as unsigned [-Wimplicitly-unsigned-literal]
2025-03-15 11:41:34 -07:00
1821f8f10c
gh-131281: fix compile error due to `BINARY_SUBSCR` (GH-131283)
...
* fix compile error due to `BINARY_SUBSCR`
* replace stat_inc with `BINARY_OP`
2025-03-15 23:38:46 +08:00
55815a6474
gh-130480: Move duplicate `LOAD_SMALL_INT` optimization from codegen to CFG ( #130481 )
2025-03-14 21:23:27 +00:00
a5776639c8
gh-111178: Fix function signatures to fix undefined behavior ( #131191 )
2025-03-14 09:52:15 +00:00
Michael Droettboom