If we have a chain of generators/coroutines that are 'yield from'ing
each other, then resuming the stack works like:
- call send() on the outermost generator
- this enters _PyEval_EvalFrameDefault, which re-executes the
YIELD_FROM opcode
- which calls send() on the next generator
- which enters _PyEval_EvalFrameDefault, which re-executes the
YIELD_FROM opcode
- ...etc.
However, every time we enter _PyEval_EvalFrameDefault, the first thing
we do is to check for pending signals, and if there are any then we
run the signal handler. And if it raises an exception, then we
immediately propagate that exception *instead* of starting to execute
bytecode. This means that e.g. a SIGINT at the wrong moment can "break
the chain" – it can be raised in the middle of our yield from chain,
with the bottom part of the stack abandoned for the garbage collector.
The fix is pretty simple: there's already a special case in
_PyEval_EvalFrameEx where it skips running signal handlers if the next
opcode is SETUP_FINALLY. (I don't see how this accomplishes anything
useful, but that's another story.) If we extend this check to also
skip running signal handlers when the next opcode is YIELD_FROM, then
that closes the hole – now the exception can only be raised at the
innermost stack frame.
This shouldn't have any performance implications, because the opcode
check happens inside the "slow path" after we've already determined
that there's a pending signal or something similar for us to process;
the vast majority of the time this isn't true and the new check
doesn't run at all.
* bpo-6532: Make the thread id an unsigned integer.
From C API side the type of results of PyThread_start_new_thread() and
PyThread_get_thread_ident(), the id parameter of
PyThreadState_SetAsyncExc(), and the thread_id field of PyThreadState
changed from "long" to "unsigned long".
* Restore a check in thread_get_ident().
When LOAD_METHOD is used for calling C mehtod, PyMethodDescrObject
was passed to profilefunc from 5566bbb.
But lsprof traces only PyCFunctionObject. Additionally, there can be
some third party extension which assumes passed arg is
PyCFunctionObject without calling PyCFunction_Check().
So make PyCFunctionObject from PyMethodDescrObject when
tstate->c_profilefunc is set.
When you use `'%s' % SubClassOfStr()`, where `SubClassOfStr.__rmod__` exists, the reverse operation is ignored as normally such string formatting operations use the `PyUnicode_Format()` fast path. This patch tests for subclasses of `str` first and picks the slow path in that case.
Patch by Martijn Pieters.
* Move all functions to call objects in a new Objects/call.c file.
* Rename fast_function() to _PyFunction_FastCallKeywords().
* Copy null_error() from Objects/abstract.c
* Inline type_error() in call.c to not have to copy it, it was only
called once.
* Export _PyEval_EvalCodeWithName() since it is now called
from call.c.
* Move all functions to call objects in a new Objects/call.c file.
* Rename fast_function() to _PyFunction_FastCallKeywords().
* Copy null_error() from Objects/abstract.c
* Inline type_error() in call.c to not have to copy it, it was only
called once.
* Export _PyEval_EvalCodeWithName() since it is now called
from call.c.
Issue #29227: Inline call_function() into _PyEval_EvalFrameDefault() using
Py_LOCAL_INLINE to reduce the stack consumption.
It reduces the stack consumption, bytes per call, before => after:
test_python_call: 1152 => 1040 (-112 B)
test_python_getitem: 1008 => 976 (-32 B)
test_python_iterator: 1232 => 1120 (-112 B)
=> total: 3392 => 3136 (- 256 B)
Special thanks to INADA Naoki for pushing the patch through
the last mile, Serhiy Storchaka for reviewing the code, and to
Victor Stinner for suggesting the idea (originally implemented
in the PyPy project).
The PEP 523 modified PyEval_EvalFrameEx(): it's now an indirection to
interp->eval_frame().
Inline the call in performance critical code. Leave PyEval_EvalFrame()
unchanged, this function is only kept for backward compatibility.
Issue #28838: Rename parameters of the "calls" functions of the Python C API.
* Rename 'callable_object' and 'func' to 'callable': any Python callable object
is accepted, not only Python functions
* Rename 'method' and 'nameid' to 'name' (method name)
* Rename 'o' to 'obj'
* Move, fix and update documentation of PyObject_CallXXX() functions
in abstract.h
* Update also the documentaton of the C API (update parameter names)
Issue #28858: The change b9c9691c72c5 introduced a regression. It seems like
_PyObject_CallArg1() uses more stack memory than
PyObject_CallFunctionObjArgs().
* PyObject_CallFunctionObjArgs(func, NULL) => _PyObject_CallNoArg(func)
* PyObject_CallFunctionObjArgs(func, arg, NULL) => _PyObject_CallArg1(func, arg)
PyObject_CallFunctionObjArgs() allocates 40 bytes on the C stack and requires
extra work to "parse" C arguments to build a C array of PyObject*.
_PyObject_CallNoArg() and _PyObject_CallArg1() are simpler and don't allocate
memory on the C stack.
This change is part of the fastcall project. The change on listsort() is
related to the issue #23507.
Issue #28799:
* Remove the PyEval_GetCallStats() function.
* Deprecate the untested and undocumented sys.callstats() function.
* Remove the CALL_PROFILE special build
Use the sys.setprofile() function, cProfile or profile module to profile
function calls.
Issue #28782: Fix a bug in the implementation ``yield from`` when checking
if the next instruction is YIELD_FROM. Regression introduced by WORDCODE
(issue #26647).
Reviewed by Serhiy Storchaka and Yury Selivanov.
When Python is not compiled with PGO, the performance of Python on call_simple
and call_method microbenchmarks depend highly on the code placement. In the
worst case, the performance slowdown can be up to 70%.
The GCC __attribute__((hot)) attribute helps to keep hot code close to reduce
the risk of such major slowdown. This attribute is ignored when Python is
compiled with PGO.
The following functions are considered as hot according to statistics collected
by perf record/perf report:
* _PyEval_EvalFrameDefault()
* call_function()
* _PyFunction_FastCall()
* PyFrame_New()
* frame_dealloc()
* PyErr_Occurred()
* BUILD_TUPLE_UNPACK and BUILD_MAP_UNPACK_WITH_CALL no longer generated with
single tuple or dict.
* Restored more informative error messages for incorrect var-positional and
var-keyword arguments.
* Removed code duplications in _PyEval_EvalCodeWithName().
* Removed redundant runtime checks and parameters in _PyStack_AsDict().
* Added a workaround and enabled previously disabled test in test_traceback.
* Removed dead code from the dis module.
Tested on macOS 10.11 dtrace, Ubuntu 16.04 SystemTap, and libbcc.
Largely based by an initial patch by Jesús Cea Avión, with some
influence from Dave Malcolm's SystemTap patch and Nikhil Benesch's
unification patch.
Things deliberately left out for simplicity:
- ustack helpers, I have no way of testing them at this point since
they are Solaris-specific
- PyFrameObject * in function__entry/function__return, this is
SystemTap-specific
- SPARC support
- dynamic tracing
- sys module dtrace facility introspection
All of those might be added later.
Issue #27830: Add _PyObject_FastCallKeywords(): avoid the creation of a
temporary dictionary for keyword arguments.
Other changes:
* Cleanup call_function() and fast_function() (ex: rename nk to nkwargs)
* Remove now useless do_call(), replaced with _PyObject_FastCallKeywords()
Issue #27213: Rework CALL_FUNCTION* opcodes to produce shorter and more
efficient bytecode:
* CALL_FUNCTION now only accepts position arguments
* CALL_FUNCTION_KW accepts position arguments and keyword arguments, but keys
of keyword arguments are packed into a constant tuple.
* CALL_FUNCTION_EX is the most generic, it expects a tuple and a dict for
positional and keyword arguments.
CALL_FUNCTION_VAR and CALL_FUNCTION_VAR_KW opcodes have been removed.
2 tests of test_traceback are currently broken: skip test, the issue #28050 was
created to track the issue.
Patch by Demur Rumed, design by Serhiy Storchaka, reviewed by Serhiy Storchaka
and Victor Stinner.
Issue #27830: Similar to _PyObject_FastCallDict(), but keyword arguments are
also passed in the same C array than positional arguments, rather than being
passed as a Python dict.
Issue #27809: PyEval_CallObjectWithKeywords() doesn't increment temporary the
reference counter of the args tuple (positional arguments). The caller already
holds a strong reference to it.
Issue #27128. When a Python function is called with no arguments, but all
parameters have a default value: use default values as arguments for the fast
path.
Issue #27128: Modify PyEval_CallObjectWithKeywords() to use
_PyObject_FastCall() when args==NULL and kw==NULL. It avoids the creation of a
temporary empty tuple for positional arguments.
Issue #27128: Add _PyObject_FastCall(), a new calling convention avoiding a
temporary tuple to pass positional parameters in most cases, but create a
temporary tuple if needed (ex: for the tp_call slot).
The API is prepared to support keyword parameters, but the full implementation
will come later (_PyFunction_FastCall() doesn't support keyword parameters
yet).
Add also:
* _PyStack_AsTuple() helper function: convert a "stack" of parameters to
a tuple.
* _PyCFunction_FastCall(): fast call implementation for C functions
* _PyFunction_FastCall(): fast call implementation for Python functions
Issue #27558: Fix a SystemError in the implementation of "raise" statement.
In a brand new thread, raise a RuntimeError since there is no active
exception to reraise.
Patch written by Xiang Zhang.
Issue #27128, #18295: replace int type with Py_ssize_t for index variables used
for positional arguments. It should help to avoid integer overflow and help to
emit better machine code for "i++" (no trap needed for overflow).
Make also the total_args variable constant.
* Add comments
* Add empty lines for readability
* PEP 7 style for if block
* Remove useless assert(globals != NULL); (globals is tested a few lines
before)
Don't fallback to PyDict_GetItemWithError() if the hash is unknown: compute the
hash instead. Add also comments to explain the optimization a little bit.
requested name doesn't exist in globals: clear the KeyError exception before
calling PyObject_GetItem(). Fail also if the raised exception is not a
KeyError.
Summary of changes:
1. Coroutines now have a distinct, separate from generators
type at the C level: PyGen_Type, and a new typedef PyCoroObject.
PyCoroObject shares the initial segment of struct layout with
PyGenObject, making it possible to reuse existing generators
machinery. The new type is exposed as 'types.CoroutineType'.
As a consequence of having a new type, CO_GENERATOR flag is
no longer applied to coroutines.
2. Having a separate type for coroutines made it possible to add
an __await__ method to the type. Although it is not used by the
interpreter (see details on that below), it makes coroutines
naturally (without using __instancecheck__) conform to
collections.abc.Coroutine and collections.abc.Awaitable ABCs.
[The __instancecheck__ is still used for generator-based
coroutines, as we don't want to add __await__ for generators.]
3. Add new opcode: GET_YIELD_FROM_ITER. The opcode is needed to
allow passing native coroutines to the YIELD_FROM opcode.
Before this change, 'yield from o' expression was compiled to:
(o)
GET_ITER
LOAD_CONST
YIELD_FROM
Now, we use GET_YIELD_FROM_ITER instead of GET_ITER.
The reason for adding a new opcode is that GET_ITER is used
in some contexts (such as 'for .. in' loops) where passing
a coroutine object is invalid.
4. Add two new introspection functions to the inspec module:
getcoroutinestate(c) and getcoroutinelocals(c).
5. inspect.iscoroutine(o) is updated to test if 'o' is a native
coroutine object. Before this commit it used abc.Coroutine,
and it was requested to update inspect.isgenerator(o) to use
abc.Generator; it was decided, however, that inspect functions
should really be tailored for checking for native types.
6. sys.set_coroutine_wrapper(w) API is updated to work with only
native coroutines. Since types.coroutine decorator supports
any type of callables now, it would be confusing that it does
not work for all types of coroutines.
7. Exceptions logic in generators C implementation was updated
to raise clearer messages for coroutines:
Before: TypeError("generator raised StopIteration")
After: TypeError("coroutine raised StopIteration")
Antoine Pitrou
Nathaniel J. Smith
Xiang Zhang
Serhiy Storchaka
INADA Naoki
Yury Selivanov
Matthias Bussonnier
Martijn Pieters
Victor Stinner
INADA Naoki
Yury Selivanov
Raymond Hettinger
Berker Peksag
Łukasz Langa
Benjamin Peterson
Brett Cannon
Martin Panter
Yury Selivanov
Eric V. Smith