8000 gh-117578: Fix inlining regression in PyType_GetModuleByDef() by neonene · Pull Request #123100 · python/cpython · GitHub
[go: up one dir, main page]
Skip to content

gh-117578: Fix inlining regression in PyType_GetModuleByDef() #123100

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 4 commits into from
Closed

gh-117578: Fix inlining regression in PyType_GetModuleByDef() #123100

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
01c3322
add Py_ALWAYS_INLINE
neonene Aug 17, 2024
f740a5d
benchmark setup: /Ob3, __declspec(noinline)
neonene Aug 20, 2024
6e62c38
unsafe experiment: do not respect TLS access
neonene Aug 20, 2024
89754d7
revert benchmark stuff
neonene Aug 20, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
unsafe experiment: do not respect TLS access
  • Loading branch information
@neonene
neonene committed Aug 20, 2024
commit 6e62c38ce334b1a04778097ff232bad08ae4a5e9
2 changes: 1 addition & 1 deletion Include/internal/pycore_pystate.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -135,7 +135,7 @@ static inline PyThreadState*
_PyThreadState_GET(void)
{
#if defined(HAVE_THREAD_LOCAL) && !defined(Py_BUILD_CORE_MODULE)
return _Py_tss_tstate;
return (PyThreadState*)_Py_atomic_load_ptr_relaxed(&_PyRuntime.tstate_current);
#else
return _PyThreadState_GetCurrent();
#endif
Expand Down
1 change: 1 addition & 0 deletions Include/internal/pycore_runtime.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -285,6 +285,7 @@ typedef struct pyruntimestate {
struct _pythread_runtime_state threads;
struct _signals_runtime_state signals;

PyThreadState *tstate_current;
/* Used for the thread state bound to the current thread. */
Py_tss_t autoTSSkey;

Expand Down
18 changes: 13 additions & 5 deletions Python/pystate.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -79,23 +79,31 @@ current_fast_get(void)
#endif
}

static inline PyThreadState *
current_fast_get2(void)
{
return (PyThreadState*)_Py_atomic_load_ptr_relaxed(&_PyRuntime.tstate_current);
}

static inline void
current_fast_set(_PyRuntimeState *Py_UNUSED(runtime), PyThreadState *tstate)
{
assert(tstate != NULL);
#ifdef HAVE_THREAD_LOCAL
_Py_tss_tstate = tstate;
_Py_atomic_store_ptr_relaxed(&_PyRuntime.tstate_current, tstate);
#else
// XXX Fall back to the PyThread_tss_*() API.
# error "no supported thread-local variable storage classifier"
#endif
}

static inline void
current_fast_clear(_PyRuntimeState *Py_UNUSED(runtime))
current_fast_clear(_PyRuntimeState *runtime)
{
#ifdef HAVE_THREAD_LOCAL
_Py_tss_tstate = NULL;
_Py_atomic_store_ptr_relaxed(&runtime->tstate_current, NULL);
#else
// XXX Fall back to the PyThread_tss_*() API.
# error "no supported thread-local variable storage classifier"
Expand All @@ -110,7 +118,7 @@ current_fast_clear(_PyRuntimeState *Py_UNUSED(runtime))
PyThreadState *
_PyThreadState_GetCurrent(void)
{
return current_fast_get();
return current_fast_get2();
}


Expand Down Expand Up @@ -1331,7 +1339,7 @@ _PyInterpreterState_RequireIDRef(PyInterpreterState *interp, int required)
PyInterpreterState*
PyInterpreterState_Get(void)
{
PyThreadState *tstate = current_fast_get();
PyThreadState *tstate = current_fast_get2();
_Py_EnsureTstateNotNULL(tstate);
PyInterpreterState *interp = tstate->interp;
if (interp == NULL) {
Expand Down Expand Up @@ -2412,14 +2420,14 @@ PyThreadState_SetAsyncExc(unsigned long id, PyObject *exc)
PyThreadState *
PyThreadState_GetUnchecked(void)
{
return current_fast_get();
return current_fast_get2();
}


PyThreadState *
PyThreadState_Get(void)
{
PyThreadState *tstate = current_fast_get();
PyThreadState *tstate = current_fast_get2();
_Py_EnsureTstateNotNULL(tstate);
return tstate;
}
Expand Down
0