Commit Graph

91 Commits

Author SHA1 Message Date
Antoine Pitrou 99c160b559 Issue #7264: Fix a possible deadlock when deallocating thread-local objects
which are part of a reference cycle.
2009-11-05 13:42:29 +00:00
Georg Brandl e1254d748d #7125: fix typo. 2009-10-14 15:51:48 +00:00
Benjamin Peterson d906ea62c8 fix Thread.ident when it is the main thread or a dummy thread #5632 2009-03-31 21:34:42 +00:00
Benjamin Peterson bd9dd31abd take the usual lock precautions around _active_limbo_lock 2009-03-31 21:06:30 +00:00
Georg Brandl ef660e8e50 #1674032: return value of flag from Event.wait(). OKed by Guido. 2009-03-31 20:41:08 +00:00
Antoine Pitrou 9fb1aca5d8 Backport relevant part of r66274 (in issue #874900). 2008-09-06 23:04:32 +00:00
Benjamin Peterson 973e6c2cf3 remove py3k warnings about the threading api; update docs
Reviewer: Benjamin Peterson
2008-09-01 23:12:58 +00:00
Benjamin Peterson b6a955672f fix a few get_name() calls and turn then to .name
Reviewer: Christian Heimes
2008-08-22 20:43:48 +00:00
Benjamin Peterson 6ee1a31e9b add py3k warnings for old threading APIs
they will still live in 3.0 but it can't hurt
2008-08-18 21:53:29 +00:00
Benjamin Peterson d810626f99 bring back the old API 2008-08-18 18:13:17 +00:00
Benjamin Peterson cbae869759 backport threading property changes 2008-08-18 17:45:09 +00:00
Benjamin Peterson d8a8972ca9 change threading.getIdent to a property
This is new in 2.6 so now need to worry about backwards compatibility :)
2008-08-18 16:40:03 +00:00
Brett Cannon 2005050152 Remove a tuple unpacking in a parameter list to suppress the SyntaxWarning with
-3.
2008-08-02 03:13:46 +00:00
Jesse Noller 5e62ca4fea Apply patch for 874900: threading module can deadlock after fork 2008-07-16 20:03:47 +00:00
Benjamin Peterson 13f7382e0d add old names back into __all__ 2008-06-11 18:02:31 +00:00
Benjamin Peterson f439560265 add aliases to threading module 2008-06-11 17:50:00 +00:00
Benjamin Peterson 0fbcf69455 give the threading API PEP 8 names 2008-06-11 17:27:50 +00:00
Gregory P. Smith 8856ddae25 Adds a Thread.getIdent() method to provide the _get_ident() value for
any given threading.Thread object.  feature request issue 2871.
2008-06-01 23:48:47 +00:00
Amaury Forgeot d'Arc d7a265129c #1733757: the interpreter would hang on shutdown, if the function set by sys.settrace
calls threading.currentThread.

The correction somewhat improves the code, but it was close.
Many thanks to the "with" construct, which turns python code into C calls.

I wonder if it is not better to sys.settrace(None) just after
running the __main__ module and before finalization.
2008-04-03 23:07:55 +00:00
Jeffrey Yasskin 105f3d4fdc Block the sys.exc_clear -3 warning from threading.py. 2008-03-31 00:35:53 +00:00
Amaury Forgeot d'Arc 504a48f90a Revert my experiment. I found one reason of failures in test_logging. 2008-03-29 01:41:08 +00:00
Amaury Forgeot d'Arc 554d4f0c13 At least let the module compile 2008-03-29 00:49:07 +00:00
Amaury Forgeot d'Arc 8a69707b80 Try to understand why most buildbots suddenly turned to red.
Undo the only change that might have unexpected effects.

To be followed.
2008-03-29 00:44:58 +00:00
Jeffrey Yasskin 8b9091fba0 Kill a race in test_threading in which the exception info in a thread finishing
up after it was joined had a traceback pointing to that thread's (deleted)
target attribute, while the test was trying to check that the target was
destroyed. Big thanks to Antoine Pitrou for diagnosing the race and pointing
out sys.exc_clear() to kill the exception early. This fixes issue 2496.
2008-03-28 04:11:18 +00:00
Jeffrey Yasskin 69e1309fd4 Thread.start() used sleep(0.000001) to make sure it didn't return before the
new thread had started. At least on my MacBook Pro, that wound up sleeping for
a full 10ms (probably 1 jiffy). By using an Event instead, we can be absolutely
certain that the thread has started, and return more quickly (217us).

Before:
$  ./python.exe -m timeit -s 'from threading import Thread'  't = Thread(); t.start(); t.join()'
100 loops, best of 3: 10.3 msec per loop
$  ./python.exe -m timeit -s 'from threading import Thread; t = Thread()'  't.isAlive()'
1000000 loops, best of 3: 0.47 usec per loop

After:
$  ./python.exe -m timeit -s 'from threading import Thread'  't = Thread(); t.start(); t.join()'
1000 loops, best of 3: 217 usec per loop
$  ./python.exe -m timeit -s 'from threading import Thread; t = Thread()'  't.isAlive()'
1000000 loops, best of 3: 0.86 usec per loop

To be fair, the 10ms isn't CPU time, and other threads including the spawned
one get to run during it. There are also some slightly more complicated ways to
get back the .4us in isAlive() if we want.
2008-02-28 06:09:19 +00:00
Jeffrey Yasskin a885c1521a Followup to r61011: Also avoid the reference cycle when the Thread's target
raises an exception.
2008-02-23 20:40:35 +00:00
Jeffrey Yasskin 3414ea9ed9 Prevent classes like:
class RunSelfFunction(object):
        def __init__(self):
            self.thread = threading.Thread(target=self._run)
            self.thread.start()
        def _run(self):
            pass
from creating a permanent cycle between the object and the thread by having the
Thread delete its references to the object when it completes.

As an example of the effect of this bug, paramiko.Transport inherits from
Thread to avoid it.
2008-02-23 19:40:54 +00:00
Raymond Hettinger 70ec29d0f4 Revert 60189 and restore performance. 2008-01-24 18:12:23 +00:00
Gregory P. Smith 95cd5c0b72 - Fix Issue #1703448: A joined thread could show up in the
threading.enumerate() list after the join() for a brief period until
  it actually exited.
2008-01-22 01:20:42 +00:00
Gregory P. Smith 64c5677de4 Replace spam.acquire() try: ... finally: spam.release() with "with spam:" 2008-01-22 01:12:02 +00:00
Guido van Rossum 54ec61ea6e Add a hack (originally devised in a slightly different form by Thomas Wouters)
to prevent spurious tracebacks when a daemon thread's cleanup happens to wake
up when the world around it has already been destroyed.
2007-08-20 15:18:04 +00:00
Nick Coghlan f8bbaa962f Eliminate RLock race condition reported in SF bug #1764059 2007-07-31 13:38:01 +00:00
Collin Winter 50b79ce8e6 Patch #1731049: make threading.py use a proper "raise" when checking internal state, rather than assert statements (which get stripped out by -O). 2007-06-06 00:17:35 +00:00
Martin v. Löwis 7b7c9d4208 Bug #1566280: Explicitly invoke threading._shutdown from Py_Main,
to avoid relying on atexit.
Will backport to 2.5.
2007-01-04 21:06:12 +00:00
Andrew MacIntyre 9291332de1 Patch #1454481: Make thread stack size runtime tunable.
Heavily revised, comprising revisions:
46640 - original trunk revision (backed out in r46655)
46647 - markup fix (backed out in r46655)
46692:46918 merged from branch aimacintyre-sf1454481

branch tested on buildbots (Windows buildbots had problems
not related to these changes).
2006-06-13 15:04:24 +00:00
Tim Peters 28eeefe566 Revert revisions:
46640 Patch #1454481:  Make thread stack size runtime tunable.
46647 Markup fix

The first is causing many buildbots to fail test runs, and there
are multiple causes with seemingly no immediate prospects for
repairing them.  See python-dev discussion.

Note that a branch can (and should) be created for resolving these
problems, like

svn copy svn+ssh://svn.python.org/python/trunk -r46640 svn+ssh://svn.python.org/python/branches/NEW_BRANCH

followed by merging rev 46647 to the new branch.
2006-06-04 23:52:47 +00:00
Andrew MacIntyre 6539d2d3c7 Patch #1454481: Make thread stack size runtime tunable. 2006-06-04 12:31:09 +00:00
Guido van Rossum da5b701aee Get rid of __context__, per the latest changes to PEP 343 and python-dev
discussion.
There are two places of documentation that still mention __context__:
Doc/lib/libstdtypes.tex -- I wasn't quite sure how to rewrite that without
spending a whole lot of time thinking about it; and whatsnew, which Andrew
usually likes to change himself.
2006-05-02 19:47:52 +00:00
Guido van Rossum 8f56d02309 Implement MvL's improvement on __context__ in Condition;
this can just call __context__ on the underlying lock.
(The same change for Semaphore does *not* work!)
2006-04-25 20:12:45 +00:00
Guido van Rossum f669436189 Um, I thought I'd already checked this in.
Anyway, this is the changes to the with-statement
so that __exit__ must return a true value in order
for a pending exception to be ignored.
The PEP (343) is already updated.
2006-03-10 02:28:35 +00:00
Guido van Rossum 1a5e21e033 Updates to the with-statement:
- New semantics for __exit__() -- it must re-raise the exception
  if type is not None; the with-statement itself doesn't do this.
  (See the updated PEP for motivation.)

- Added context managers to:
  - file
  - thread.LockType
  - threading.{Lock,RLock,Condition,Semaphore,BoundedSemaphore}
  - decimal.Context

- Added contextlib.py, which defines @contextmanager, nested(), closing().

- Unit tests all around; bot no docs yet.
2006-02-28 21:57:43 +00:00
Brett Cannon ad07ff2c77 Prevent threading.Thread.join() from blocking when a previous call raised an
exception (e.g., passing in an illegal argument).

Applies patch #1314396.  Thanks Eric Blossom.
2005-11-23 02:15:50 +00:00
Georg Brandl a4a8b820aa bug [ 1238170 ] threading.Thread uses {} as default argument 2005-07-15 09:13:21 +00:00
Brett Cannon 90cece7f89 Fixed typo in verbose output.
Closes bug #1110998.  Thanks Matthew Bogosian.
2005-01-27 22:48:30 +00:00
Tim Peters 711906e0c2 threading._DummyThread.__init__(): document obscure new code.
test_threading.test_foreign_thread():  new test does a basic check that
"foreign" threads can using the threading module, and that they create
a _DummyThread instance in at least one use case.  This isn't a very
good test, since a thread created by thread.start_new_thread() isn't
particularly "foreign".
2005-01-08 07:30:42 +00:00
Brett Cannon e6539c4401 In _DummyThread objects the lock stored in __block (allocated thanks to
_Thread.__init__) was never used.  This is a waste since locks use OS
primitives that are in limited supply.  So the lock is deleted in
_DummyThread.__init__ .

Closes bug #1089632.
2005-01-08 02:43:53 +00:00
Tim Peters 21429932e4 Thread.__delete: Discussion of internal obscurities belongs in comments
rather than in docstrings.  Rewrote so that _active_limbo_lock is released
no matter what happens (it could have been left locked if _sys got None'd
out).  Use "in" in preference to has_key() for dict lookup.  Don't bother
looking for 'dummy_threading' in sys.modules unless KeyError is raised.
Since the heart of the method is the del, do that in only one place.
2004-07-21 03:36:52 +00:00
Brett Cannon 8b3d92a977 Fix bug where a KeyError was raised if -O was being used for the interpreter
and Thread.__delete() was called after a Thread instance was created.  Problem
resulted from a currentThread() call in an 'assert' statement being optimized
out and dummy_thread.get_ident() always returning -1 and thus overwriting the
entry for the _MainThread() instance created in 'threading' at import time.

Closes bug #993394.
2004-07-21 02:21:58 +00:00
Jim Fulton d15dc06df0 Implemented thread-local data as proposed on python-dev:
http://mail.python.org/pipermail/python-dev/2004-June/045785.html
2004-07-14 19:11:50 +00:00
Brett Cannon cc4e935ea5 threading.Thread objects will now print a traceback for an exception raised
during interpreter shutdown instead of masking it with another traceback about
accessing a NoneType when trying to print the exception out in the first place.

Closes bug #754449 (using patch #954922).
2004-07-03 03:52:35 +00:00