* Never change types' cached keys. It could invalidate inline attribute objects.
* Lazily create object dictionaries.
* Update specialization of LOAD/STORE_ATTR.
* Don't update shared keys version for deletion of value.
* Update gdb support to handle instance values.
* Rename SPLIT_KEYS opcodes to INSTANCE_VALUE.
Places the locals between the specials and stack. This is the more "natural" layout for a C struct, makes the code simpler and gives a slight speedup (~1%)
* Convert "specials" array to InterpreterFrame struct, adding f_lasti, f_state and other non-debug FrameObject fields to it.
* Refactor, calls pushing the call to the interpreter upward toward _PyEval_Vector.
* Compute f_back when on thread stack, only filling in value when frame object outlives stack invocation.
* Move ownership of InterpreterFrame in generator from frame object to generator object.
* Do not create frame objects for Python calls.
* Do not create frame objects for generators.
These were reverted in gh-26530 (commit 17c4edc) due to refleaks.
* 2c1e258 - Compute deref offsets in compiler (gh-25152)
* b2bf2bc - Add new internal code objects fields: co_fastlocalnames and co_fastlocalkinds. (gh-26388)
This change fixes the refleaks.
https://bugs.python.org/issue43693
* Revert "bpo-43693: Compute deref offsets in compiler (gh-25152)"
This reverts commit b2bf2bc1ec.
* Revert "bpo-43693: Add new internal code objects fields: co_fastlocalnames and co_fastlocalkinds. (gh-26388)"
This reverts commit 2c1e2583fd.
These two commits are breaking the refleak buildbots.
A number of places in the code base (notably ceval.c and frameobject.c) rely on mapping variable names to indices in the frame "locals plus" array (AKA fast locals), and thus opargs. Currently the compiler indirectly encodes that information on the code object as the tuples co_varnames, co_cellvars, and co_freevars. At runtime the dependent code must calculate the proper mapping from those, which isn't ideal and impacts performance-sensitive sections. This is something we can easily address in the compiler instead.
This change addresses the situation by replacing internal use of co_varnames, etc. with a single combined tuple of names in locals-plus order, along with a minimal array mapping each to its kind (local vs. cell vs. free). These two new PyCodeObject fields, co_fastlocalnames and co_fastllocalkinds, are not exposed to Python code for now, but co_varnames, etc. are still available with the same values as before (though computed lazily).
Aside from the (mild) performance impact, there are a number of other benefits:
* there's now a clear, direct relationship between locals-plus and variables
* code that relies on the locals-plus-to-name mapping is simpler
* marshaled code objects are smaller and serialize/de-serialize faster
Also note that we can take this approach further by expanding the possible values in co_fastlocalkinds to include specific argument types (e.g. positional-only, kwargs). Doing so would allow further speed-ups in _PyEval_MakeFrameVector(), which is where args get unpacked into the locals-plus array. It would also allow us to shrink marshaled code objects even further.
https://bugs.python.org/issue43693
* Remove 'zombie' frames. We won't need them once we are allocating fixed-size frames.
* Add co_nlocalplus field to code object to avoid recomputing size of locals + frees + cells.
* Move locals, cells and freevars out of frame object into separate memory buffer.
* Use per-threadstate allocated memory chunks for local variables.
* Move globals and builtins from frame object to per-thread stack.
* Move (slow) locals frame object to per-thread stack.
* Move internal frame functions to internal header.
* Use instruction offset, rather than bytecode offset. Streamlines interpreter dispatch a bit, and removes most EXTENDED_ARGs for jumps.
* Change some uses of PyCode_Addr2Line to PyFrame_GetLineNumber
On Fedora 31 gdb is using python 3.7.9, calling `proxyval` on an instance with a dictionary fails because of the `dict.iteritems` usage. This PR changes the code to be compatible with py2 and py3.
This changed seemed small enough to not need an issue and news blurb, if one is required please let me know.
Automerge-Triggered-By: GH:benjaminp
python-gdb.py now checks for "take_gil" function name to check if a
frame tries to acquire the GIL, instead of checking for
"pthread_cond_timedwait" which is specific to Linux and can be a
different condition than the GIL.
This fixes the exception '`ValueError: invalid literal for int() with base 10`
if `str(gdbval)` returns a hexadecimal value (e.g. '0xa0'). This is the case if
the output-radix is set to 16 in gdb. See
https://sourceware.org/gdb/onlinedocs/gdb/Numbers.html for more information.
bpo-37151: remove special case for PyCFunction from PyObject_Call
Alse, make the undocumented function PyCFunction_Call an alias
of PyObject_Call and deprecate it.
As it changes the way functions are called, the PEP 590 implementation
skipped the functions that the GDB integration is looking for
(by name) to find function calls.
Looking for the new helper `cfunction_call_varargs` hopefully fixes the
tests, and thus buildbots.
The changed frame nuber in test_gdb is due to there being fewer
C calls when calling a built-in method.
python-gdb.py now handles errors on computing the line number
of a Python frame.
Changes:
* PyFrameObjectPtr.current_line_num() now catchs any Exception on
calling addr2line(), instead of failing with a surprising "<class
'TypeError'> 'FakeRepr' object is not subscriptable" error.
* All callers of current_line_num() now handle current_line_num()
returning None.
* PyFrameObjectPtr.current_line() now also catchs IndexError on
getting a line from the Python source file.
PEP 523 introduced _PyEval_EvalFrameDefault which inlines PyEval_EvalFrameEx on
non-debug shared builds. This breaks the ability to use py-bt, py-up, and
a few other Python-specific gdb integrations.
This patch fixes the problem by only looking for _PyEval_EvalFrameDefault
frames.
test_gdb passes on both a debug and a non-debug build.
Original patch by Bruno "Polaco" Penteado.
* Revert "Add Bruno Penteado to ACKS (#3091)"
This reverts commit f978405b3f.
* Revert "bpo-30983: eval frame rename in pep 0523 broke gdb's python extension (#2803)"
This reverts commit 2e0f4db114.
pep 0523 renames PyEval_EvalFrameEx to _PyEval_EvalFrameDefault while the gdb python extension only looks for PyEval_EvalFrameEx to understand if it is dealing with a frame.
Final effect is that attaching gdb to a python3.6 process doesnt resolve python objects. Eg. py-list and py-bt dont work properly.
This patch fixes that. Tested locally on python3.6
Issue #29259:
* Detect PyCFunction is the current frame, not only in the older frame
* Ignore PyCFunction_Call() since it now calls _PyCFunction_FastCallDict(), and
_PyCFunction_FastCallDict() is already detected
Issue #29259: Write fast path in _PyCFunction_FastCallKeywords() for
METH_FASTCALL, avoid the creation of a temporary dictionary for keyword
arguments.
Cleanup also _PyCFunction_FastCallDict():
* Don't dereference func before checking that it's not NULL
* Move code to raise the "no keyword argument" exception into a new
no_keyword_error label.
Update python-gdb.py for the change.
Frame.is_other_python_frame() now also handles _PyCFunction_FastCallDict()
frames.
Thanks to the new code to handle fast calls, python-gdb.py is now also able to
detect the <built-in id method of module ...> frame.
