Skip to content

gh-117657: Fix some races in tracebacks #121748

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 6 commits into from
Closed

gh-117657: Fix some races in tracebacks #121748

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
6cb8897
Fix races in tracebacks
Fidget-Spinner Jul 14, 2024
5bc855c
use atomics in new_reference
Fidget-Spinner Jul 14, 2024
da8415d
Add more atomics
Fidget-Spinner Jul 14, 2024
ef99cb9
Update Include/cpython/pyatomic_msc.h
Fidget-Spinner Jul 22, 2024
951a34d
revert atomics
Fidget-Spinner Jul 24, 2024
6e0e176
Use argumentclinic
Fidget-Spinner Jul 24, 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
Add more atomics
  • Loading branch information
@Fidget-Spinner
Fidget-Spinner committed Jul 14, 2024
commit da8415dee9fd519cd8fa9f2c6f6cad75c8d91ab0
4 changes: 4 additions & 0 deletions Include/cpython/pyatomic_gcc.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -516,6 +516,10 @@ static inline int
_Py_atomic_load_int_acquire(const int *obj)
{ return __atomic_load_n(obj, __ATOMIC_ACQUIRE); }

static inline void
_Py_atomic_store_uint8_release(uint8_t *obj, uint8_t value)
{ __atomic_store_n(obj, value, __ATOMIC_RELEASE); }

static inline void
_Py_atomic_store_uint32_release(uint32_t *obj, uint32_t value)
{ __atomic_store_n(obj, value, __ATOMIC_RELEASE); }
Expand Down
13 changes: 13 additions & 0 deletions Include/cpython/pyatomic_msc.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -989,6 +989,19 @@ _Py_atomic_load_int_acquire(const int *obj)
#endif
}

static inline void
_Py_atomic_store_uint8_release(uint8_t *obj, uint8_t value)
{
#if defined(_M_X64) || defined(_M_IX86)
*(uint8_t volatile *)obj = value;
#elif defined(_M_ARM64)
_Py_atomic_ASSERT_ARG_TYPE(unsigned __int8);
__stlr8((unsigned __int8 volatile *)obj, (unsigned __int8)value);
#else
# error "no implementation of _Py_atomic_store_uint32_release"
#endif
}

static inline void
_Py_atomic_store_uint32_release(uint32_t *obj, uint32_t value)
{
Expand Down
8 changes: 8 additions & 0 deletions Include/cpython/pyatomic_std.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -911,6 +911,14 @@ _Py_atomic_load_int_acquire(const int *obj)
memory_order_acquire);
}

static inline void
_Py_atomic_store_uint8_release(uint8_t *obj, uint8_t value)
{
_Py_USING_STD;
atomic_store_explicit((_Atomic(uint8_t)*)obj, value,
memory_order_release);
}

static inline void
_Py_atomic_store_uint32_release(uint32_t *obj, uint32_t value)
{
Expand Down
4 changes: 2 additions & 2 deletions Objects/object.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2360,10 +2360,10 @@ new_reference(PyObject *op)
#if !defined(Py_GIL_DISABLED)
op->ob_refcnt = 1;
#else
op->ob_tid = _Py_ThreadId();
_Py_atomic_store_uintptr_release(&op->ob_tid, _Py_ThreadId());
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't think we want to do this, at least not for now. I want to wait until we fix all the other races before we address new_reference because there are perf implications and the race is in practice benign.

The "release" memory order is also stronger (and more expensive on arm64) than necessary.

op->_padding = 0;
op->ob_mutex = (PyMutex){ 0 };
op->ob_gc_bits = 0;
_Py_atomic_store_uint8_release(&op->ob_gc_bits, 0);
_Py_atomic_store_uint32_release(&op->ob_ref_local, 1);
_Py_atomic_store_ssize_release(&op->ob_ref_shared, 0);
#endif
Expand Down
Loading
pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy