Commit Graph

38 Commits

Author SHA1 Message Date
Victor Stinner 7be4e350aa
bpo-40513: Per-interpreter GIL (GH-19943)
In the experimental isolated subinterpreters build mode, the GIL is
now per-interpreter.

Move gil from _PyRuntimeState.ceval to PyInterpreterState.ceval.

new_interpreter() always get the config from the main interpreter.
2020-05-05 20:27:47 +02:00
Victor Stinner 299b8c61e9
Revert "bpo-40513: Per-interpreter signals pending (GH-19924)" (GH-19932)
This reverts commit 4e01946caf.
2020-05-05 17:40:18 +02:00
Victor Stinner 0b1e3307e2
bpo-40513: Per-interpreter gil_drop_request (GH-19927)
Move gil_drop_request member from _PyRuntimeState.ceval to
PyInterpreterState.ceval.
2020-05-05 16:14:31 +02:00
Victor Stinner b54a99d643
bpo-40082: trip_signal() uses the main interpreter (GH-19441)
Fix the signal handler: it now always uses the main interpreter,
rather than trying to get the current Python thread state.

The following function now accepts an interpreter, instead of a
Python thread state:

* _PyEval_SignalReceived()
* _Py_ThreadCanHandleSignals()
* _PyEval_AddPendingCall()
* COMPUTE_EVAL_BREAKER()
* SET_GIL_DROP_REQUEST(), RESET_GIL_DROP_REQUEST()
* SIGNAL_PENDING_CALLS(), UNSIGNAL_PENDING_CALLS()
* SIGNAL_PENDING_SIGNALS(), UNSIGNAL_PENDING_SIGNALS()
* SIGNAL_ASYNC_EXC(), UNSIGNAL_ASYNC_EXC()

Py_AddPendingCall() now uses the main interpreter if it fails to the
current Python thread state.

Convert _PyThreadState_GET() and PyInterpreterState_GET_UNSAFE()
macros to static inline functions.
2020-04-08 23:35:05 +02:00
Victor Stinner da2914db4b
bpo-40010: Pass tstate to ceval GIL functions (GH-19077)
* Add eval_frame_handle_pending() helper function: cold code path.
* Fix PyEval_ReleaseLock(): don't dereference tstate if it's NULL.
2020-03-20 09:29:08 +01:00
Victor Stinner a36adfa6bb
bpo-39877: 4th take_gil() fix for daemon threads (GH-19080)
bpo-39877, bpo-40010: Add a third tstate_must_exit() check in
take_gil() to prevent using tstate which has been freed.
2020-03-19 19:48:25 +01:00
Victor Stinner 5a3a71dddb
bpo-40010: Optimize signal handling in multithreaded applications (GH-19067)
If a thread different than the main thread gets a signal, the
bytecode evaluation loop is no longer interrupted at each bytecode
instruction to check for pending signals which cannot be handled.
Only the main thread of the main interpreter can handle signals.

Previously, the bytecode evaluation loop was interrupted at each
instruction until the main thread handles signals.

Changes:

* COMPUTE_EVAL_BREAKER() and SIGNAL_PENDING_SIGNALS() no longer set
  eval_breaker to 1 if the current thread cannot handle signals.
* take_gil() now always recomputes eval_breaker.
2020-03-19 17:40:12 +01:00
Victor Stinner 50e6e99178
bpo-39984: Move pending calls to PyInterpreterState (GH-19066)
If Py_AddPendingCall() is called in a subinterpreter, the function is
now scheduled to be called from the subinterpreter, rather than being
called from the main interpreter.

Each subinterpreter now has its own list of scheduled calls.

* Move pending and eval_breaker fields from _PyRuntimeState.ceval
  to PyInterpreterState.ceval.
* new_interpreter() now calls _PyEval_InitThreads() to create
  pending calls lock.
* Fix Py_AddPendingCall() for subinterpreters. It now calls
  _PyThreadState_GET() which works in a subinterpreter if the
  caller holds the GIL, and only falls back on
  PyGILState_GetThisThreadState() if _PyThreadState_GET()
  returns NULL.
2020-03-19 02:41:21 +01:00
Victor Stinner 56bfdebfb1
bpo-39984: Pass tstate to _PyEval_SignalAsyncExc() (GH-19049)
_PyEval_SignalAsyncExc() and _PyEval_FiniThreads() now expect tstate,
instead of ceval.
2020-03-18 09:26:25 +01:00
Victor Stinner 29356e03d4
bpo-39877: Fix take_gil() for daemon threads (GH-19054)
bpo-39877, bpo-39984: If the thread must exit, don't access tstate to
prevent a potential crash: tstate memory has been freed.
2020-03-18 03:04:33 +01:00
Victor Stinner 9229eeee10
bpo-39877: take_gil() checks tstate_must_exit() twice (GH-18890)
take_gil() now also checks tstate_must_exit() after acquiring
the GIL: exit the thread if Py_Finalize() has been called.
2020-03-09 23:10:53 +01:00
Victor Stinner 85f5a69ae1
bpo-39877: Refactor take_gil() function (GH-18885)
* Remove ceval parameter of take_gil(): get it from tstate.
* Move exit_thread_if_finalizing() call inside take_gil(). Replace
  exit_thread_if_finalizing() with tstate_must_exit(): the caller is
  now responsible to call PyThread_exit_thread().
* Move is_tstate_valid() assertion inside take_gil(). Remove
  is_tstate_valid(): inline code into take_gil().
* Move gil_created() assertion inside take_gil().
2020-03-09 22:12:04 +01:00
Victor Stinner eb4e2ae2b8
bpo-39877: Fix PyEval_RestoreThread() for daemon threads (GH-18811)
* exit_thread_if_finalizing() does now access directly _PyRuntime
  variable, rather than using tstate->interp->runtime since tstate
  can be a dangling pointer after Py_Finalize() has been called.
* exit_thread_if_finalizing() is now called *before* calling
  take_gil(). _PyRuntime.finalizing is an atomic variable,
  we don't need to hold the GIL to access it.
* Add ensure_tstate_not_null() function to check that tstate is not
  NULL at runtime. Check tstate earlier. take_gil() does not longer
  check if tstate is NULL.

Cleanup:

* PyEval_RestoreThread() no longer saves/restores errno: it's already
  done inside take_gil().
* PyEval_AcquireLock(), PyEval_AcquireThread(),
  PyEval_RestoreThread() and _PyEval_EvalFrameDefault() now check if
  tstate is valid with the new is_tstate_valid() function which uses
  _PyMem_IsPtrFreed().
2020-03-08 11:57:45 +01:00
Victor Stinner e225bebc14
Revert "bpo-33608: Factor out a private, per-interpreter _Py_AddPendingCall(). (gh-13714)" (GH-13780)
This reverts commit 6a150bcaeb.
2019-06-03 18:14:24 +02:00
Eric Snow 6a150bcaeb
bpo-33608: Factor out a private, per-interpreter _Py_AddPendingCall(). (gh-13714) 2019-06-01 15:39:46 -06:00
Victor Stinner 09532feeec
bpo-36710: Add 'ceval' local variable to ceval.c (GH-12934)
Add "struct _ceval_runtime_state *ceval = &_PyRuntime.ceval;" local
variables to function to better highlight the dependency on the
global variable _PyRuntime and to point directly to _PyRuntime.ceval
field rather than on the larger _PyRuntime.

Changes:

* Add _PyRuntimeState_GetThreadState(runtime) macro.
* Add _PyEval_AddPendingCall(ceval, ...) and
  _PyThreadState_Swap(gilstate, ...) functions.
* _PyThreadState_GET() macro now calls
  _PyRuntimeState_GetThreadState() using &_PyRuntime.
* Add 'ceval' parameter to COMPUTE_EVAL_BREAKER(),
  SIGNAL_PENDING_SIGNALS(), _PyEval_SignalAsyncExc(),
  _PyEval_SignalReceived() and _PyEval_FiniThreads() macros and
  functions.
* Add 'tstate' parameter to call_function(), do_call_core() and
  do_raise().
* Add 'runtime' parameter to _Py_CURRENTLY_FINALIZING(),
  _Py_FinishPendingCalls() and _PyThreadState_DeleteExcept()
  macros and functions.
* Declare 'runtime', 'tstate', 'ceval' and 'eval_breaker' variables
  as constant.
2019-05-10 23:39:09 +02:00
Eric Snow b75b1a3504
bpo-33608: Revert "Factor out a private, per-interpreter _Py_AddPendingCall()." (gh-12806)
This reverts commit f13c5c8b94 (gh-12360).
2019-04-12 10:20:10 -06:00
Eric Snow f13c5c8b94
bpo-33608: Factor out a private, per-interpreter _Py_AddPendingCall(). (gh-12360)
This is effectively an un-revert of #11617 and #12024 (reverted in #12159). Portions of those were merged in other PRs (with lower risk) and this represents the remainder. Note that I found 3 different bugs in the original PRs and have fixed them here.
2019-04-12 09:18:16 -06:00
Victor Stinner 4d61e6e3b8
Revert: bpo-33608: Factor out a private, per-interpreter _Py_AddPendingCall(). (GH-11617) (GH-12159)
* Revert "bpo-36097: Use only public C-API in the_xxsubinterpreters module (adding as necessary). (#12003)"

This reverts commit bcfa450f21.

* Revert "bpo-33608: Simplify ceval's DISPATCH by hoisting eval_breaker ahead of time. (gh-12062)"

This reverts commit bda918bf65.

* Revert "bpo-33608: Use _Py_AddPendingCall() in _PyCrossInterpreterData_Release(). (gh-12024)"

This reverts commit b05b711a2c.

* Revert "bpo-33608: Factor out a private, per-interpreter _Py_AddPendingCall(). (GH-11617)"

This reverts commit ef4ac967e2.
2019-03-04 14:21:28 +01:00
Eric Snow ef4ac967e2
bpo-33608: Factor out a private, per-interpreter _Py_AddPendingCall(). (GH-11617)
This involves moving the global "pending calls" state to PyInterpreterState.

https://bugs.python.org/issue33608
2019-02-24 15:40:47 -08:00
Victor Stinner 27e2d1f219
bpo-35081: Add pycore_ prefix to internal header files (GH-10263)
* Rename Include/internal/ header files:

  * pyatomic.h -> pycore_atomic.h
  * ceval.h -> pycore_ceval.h
  * condvar.h -> pycore_condvar.h
  * context.h -> pycore_context.h
  * pygetopt.h -> pycore_getopt.h
  * gil.h -> pycore_gil.h
  * hamt.h -> pycore_hamt.h
  * hash.h -> pycore_hash.h
  * mem.h -> pycore_mem.h
  * pystate.h -> pycore_state.h
  * warnings.h -> pycore_warnings.h

* PCbuild project, Makefile.pre.in, Modules/Setup: add the
  Include/internal/ directory to the search paths of header files.
* Update includes. For example, replace #include "internal/mem.h"
  with #include "pycore_mem.h".
2018-11-01 00:52:28 +01:00
Victor Stinner 31368a4f0e
bpo-35081: Move Include/pyatomic.c to Include/internal/ (GH-10239)
Add pyatomic.h to the VS project (it wasn't referenced).
2018-10-30 15:14:25 +01:00
Eric Snow 2ebc5ce42a bpo-30860: Consolidate stateful runtime globals. (#3397)
* group the (stateful) runtime globals into various topical structs
* consolidate the topical structs under a single top-level _PyRuntimeState struct
* add a check-c-globals.py script that helps identify runtime globals

Other globals are excluded (see globals.txt and check-c-globals.py).
2017-09-07 23:51:28 -06:00
Eric Snow 05351c1bd8 Revert "bpo-30860: Consolidate stateful runtime globals." (#3379)
Windows buildbots started failing due to include-related errors.
2017-09-05 21:43:08 -07:00
Eric Snow 76d5abc868 bpo-30860: Consolidate stateful runtime globals. (#2594)
* group the (stateful) runtime globals into various topical structs
* consolidate the topical structs under a single top-level _PyRuntimeState struct
* add a check-c-globals.py script that helps identify runtime globals

Other globals are excluded (see globals.txt and check-c-globals.py).
2017-09-05 18:26:16 -07:00
Benjamin Peterson ca47063998 replace Py_(u)intptr_t with the c99 standard types 2016-09-06 13:47:26 -07:00
Victor Stinner b02ef715a3 Use Py_uintptr_t for atomic pointers
Issue #26161: Use Py_uintptr_t instead of void* for atomic pointers in
pyatomic.h. Use atomic_uintptr_t when <stdatomic.h> is used.

Using void* causes compilation warnings depending on which implementation of
atomic types is used.
2016-01-22 14:09:55 +01:00
Serhiy Storchaka 009b811d67 Removed unintentional trailing spaces in non-external and non-generated C files. 2015-03-18 21:53:15 +02:00
Serhiy Storchaka 53fa8b2a4b Fixed few compiler warnings. 2015-02-16 09:40:12 +02:00
Kristján Valur Jónsson e75ff35af2 Issue #15038: Optimize python Locks on Windows
Extract cross-platform condition variable support into a separate file and
provide user-mode non-recursive locks for Windows.
2012-06-18 20:30:44 +00:00
Kristján Valur Jónsson 187aa54516 Signal condition variables with the mutex held. Destroy condition variables
before their mutexes.
2012-06-05 22:17:42 +00:00
Antoine Pitrou 6a002c02da Fix the GIL with subinterpreters. Hopefully this will allow mod_wsgi to work with 3.2.
(we need some tests for this)
2011-01-15 11:37:11 +00:00
Antoine Pitrou 1df1536fb9 Issue #9828: Destroy the GIL in Py_Finalize(), so that it gets properly
re-created on a subsequent call to Py_Initialize().  The problem (a crash)
wouldn't appear in 3.1 or 2.7 where the GIL's structure is more trivial.
2010-09-13 14:16:46 +00:00
Antoine Pitrou e1dd1747e8 Issue #8411: new condition variable emulation under Windows for the new GIL,
by Kristján.  Unfortunately the 3.x Windows buildbots are in a wreck, so we'll
have to watch them when they become fit again.
2010-08-10 13:48:51 +00:00
Jeffrey Yasskin 39370830a9 Make (most of) Python's tests pass under Thread Sanitizer.
http://code.google.com/p/data-race-test/wiki/ThreadSanitizer is a dynamic data
race detector that runs on top of valgrind. With this patch, the binaries at
http://code.google.com/p/data-race-test/wiki/ThreadSanitizer#Binaries pass many
but not all of the Python tests. All of regrtest still passes outside of tsan.

I've implemented part of the C1x atomic types so that we can explicitly mark
variables that are used across threads, and get defined behavior as compilers
advance.

I've added tsan's client header and implementation to the codebase in
dynamic_annotations.{h,c} (docs at
http://code.google.com/p/data-race-test/wiki/DynamicAnnotations).
Unfortunately, I haven't been able to get helgrind and drd to give sensible
error messages, even when I use their client annotations, so I'm not supporting
them.
2010-05-03 19:29:34 +00:00
Antoine Pitrou a1d23326b1 Try to strengthen condition-waiting under Windows.
If it doesn't work (doesn't solve erratic freezes) we'll have to resort
to tougher (Windows-only) measures.
2009-11-12 22:56:02 +00:00
Antoine Pitrou cf4cabbe2a Our condition variable emulation under Windows is imperfect, which
seems to be the cause of the buildbot hangs. Try to fix it, and add
some comments.
2009-11-11 18:11:36 +00:00
Antoine Pitrou 074e5ed974 Merge in the new GIL. 2009-11-10 19:50:40 +00:00