news

windows多线程开发包

RELEASE 2.8.0
-------------
(2006-12-22)

General
-------
New bug fixes in this release since 2.7.0 have not been applied to the
version 1.x.x series. It is probably time to drop version 1.

Testing and verification
------------------------
This release has not yet been tested on SMP architechtures. All tests pass
on a uni-processor system.

Bug fixes
---------
Sem_destroy could return EBUSY even though no threads were waiting on the 
semaphore. Other races around invalidating semaphore structs (internally)
have been removed as well.

New tests
---------
semaphore5.c - tests the bug fix referred to above.


RELEASE 2.7.0
-------------
(2005-06-04)

General
-------
All new features in this release have been back-ported in release 1.11.0,
including the incorporation of MCS locks in pthread_once, however, versions
1 and 2 remain incompatible even though they are now identical in
performance and functionality.

Testing and verification
------------------------
This release has been tested (passed the test suite) on both uni-processor
and multi-processor systems.
- Tim Theisen

Bug fixes
---------
Pthread_once has been re-implemented to remove priority boosting and other
complexity to improve robustness. Races for Win32 handles that are not
recycle-unique have been removed. The general form of pthread_once is now
the same as that suggested earlier by Alexander Terekhov, but instead of the
'named mutex', a queue-based lock has been implemented which has the required
properties of dynamic self initialisation and destruction. This lock is also
efficient. The ABI is unaffected in as much as the size of pthread_once_t has
not changed and PTHREAD_ONCE_INIT has not changed, however, applications that
peek inside pthread_once_t, which is supposed to be opaque, will break.
- Vladimir Kliatchko

New features
------------
* Support for Mingw cross development tools added to GNUmakefile.
Mingw cross tools allow building the libraries on Linux.
- Mikael Magnusson


RELEASE 2.6.0
-------------
(2005-05-19)

General
-------
All of the bug fixes and new features in this release have been
back-ported in release 1.10.0.

Testing and verification
------------------------
This release has been tested (passed the test suite) on both uni-processor
and multi-processor systems. Thanks to Tim Theisen at TomoTherapy for
exhaustively running the MP tests and for providing crutial observations
and data when faults are detected.

Bugs fixed
----------

* pthread_detach() now reclaims remaining thread resources if called after
the target thread has terminated. Previously, this routine did nothing in
this case.

New tests
---------

* detach1.c - tests that pthread_detach properly invalidates the target
thread, which indicates that the thread resources have been reclaimed.


RELEASE 2.5.0
-------------
(2005-05-09)

General
-------

The package now includes a reference documentation set consisting of
HTML formatted Unix-style manual pages that have been edited for
consistency with Pthreads-w32. The set can also be read online at:
http://sources.redhat.com/pthreads-win32/manual/index.html

Thanks again to Tim Theisen for running the test suite pre-release
on an MP system.

All of the bug fixes and new features in this release have been
back-ported in release 1.9.0.

Bugs fixed
----------

* Thread Specific Data (TSD) key management has been ammended to
eliminate a source of (what was effectively) resource leakage (a HANDLE
plus memory for each key destruct routine/thread association). This was
not a true leak because these resources were eventually reclaimed when
pthread_key_delete was run AND each thread referencing the key had exited.
The problem was that these two conditions are often not met until very
late, and often not until the process is about to exit.

The ammended implementation avoids the need for the problematic HANDLE
and reclaims the memory as soon as either the key is deleted OR the
thread exits, whichever is first.

Thanks to Richard Hughes at Aculab for identifying and locating the leak.

* TSD key destructors are now processed up to PTHREAD_DESTRUCTOR_ITERATIONS
times instead of just once. PTHREAD_DESTRUCTOR_ITERATIONS has been
defined in pthread.h for some time but not used.

* Fix a semaphore accounting race between sem_post/sem_post_multiple
and sem_wait cancellation. This is the same issue as with
sem_timedwait that was fixed in the last release.

* sem_init, sem_post, and sem_post_multiple now check that the
semaphore count never exceeds _POSIX_SEM_VALUE_MAX.

* Although sigwait() is nothing more than a no-op, it should at least
be a cancellation point to be consistent with the standard.

New tests
---------

* stress1.c - attempts to expose problems in condition variable
and semaphore timed wait logic. This test was inspired by Stephan
Mueller's sample test code used to identify the sem_timedwait bug
from the last release. It's not a part of the regular test suite
because it can take awhile to run. To run it:
nmake clean VC-stress

* tsd2.c - tests that key destructors are re-run if the tsd key value is
not NULL after the destructor routine has run. Also tests that
pthread_setspecific() and pthread_getspecific() are callable from
destructors.


RELEASE 2.4.0
-------------
(2005-04-26)

General
-------

There is now no plan to release a version 3.0.0 to fix problems in
pthread_once(). Other possible implementations of pthread_once
will still be investigated for a possible future release in an attempt
to reduce the current implementation's complexity.

All of the bug fixes and new features in this release have been
back-ported for release 1.8.0.

Bugs fixed
----------

* Fixed pthread_once race (failures on an MP system). Thanks to
Tim Theisen for running exhaustive pre-release testing on his MP system
using a range of compilers:
  VC++ 6
  VC++ 7.1
  Intel C++ version 8.0
All tests passed.
Some minor speed improvements were also done.

* Fix integer overrun error in pthread_mutex_timedlock() - missed when
sem_timedwait() was fixed in release 2.2.0. This routine no longer returns
ENOTSUP when NEED_SEM is defined - it is supported (NEED_SEM is only
required for WinCE versions prior to 3.0).

* Fix timeout bug in sem_timedwait().
- Thanks to Stephan Mueller for reporting, providing diagnostic output
and test code.

* Fix several problems in the NEED_SEM conditionally included code.
NEED_SEM included code is provided for systems that don't implement W32
semaphores, such as WinCE prior to version 3.0. An alternate implementation
of POSIX semaphores is built using W32 events for these systems when
NEED_SEM is defined. This code has been completely rewritten in this
release to reuse most of the default POSIX semaphore code, and particularly,
to implement all of the sem_* routines supported by pthreads-win32. Tim
Theisen also run the test suite over the NEED_SEM code on his MP system. All
tests passed.

* The library now builds without errors for the Borland Builder 5.5 compiler.

New features
------------

* pthread_mutex_timedlock() and all sem_* routines provided by
pthreads-win32 are now implemented for WinCE versions prior to 3.0. Those
versions did not implement W32 semaphores. Define NEED_SEM in config.h when
building the library for these systems.

Known issues in this release
----------------------------

* pthread_once is too complicated - but it works as far as testing can
determine..

* The Borland version of the dll fails some of the tests with a memory read
exception. The cause is not yet known but a compiler bug has not been ruled
out.


RELEASE 2.3.0
-------------
(2005-04-12)

General
-------

Release 1.7.0 is a backport of features and bug fixes new in
this release. See earlier notes under Release 2.0.0/General.

Bugs fixed
----------

* Fixed pthread_once potential for post once_routine cancellation
hanging due to starvation. See comments in pthread_once.c.
Momentary priority boosting is used to ensure that, after a
once_routine is cancelled, the thread that will run the
once_routine is not starved by higher priority waiting threads at
critical times. Priority boosting occurs only AFTER a once_routine 
cancellation, and is applied only to that once_control. The
once_routine is run at the thread's normal base priority.

New tests
---------

* once4.c: Aggressively tests pthread_once() under realtime
conditions using threads with varying priorities. Windows'
random priority boosting does not occur for threads with realtime
priority levels.


RELEASE 2.2.0
-------------
(2005-04-04)

General
-------

* Added makefile targets to build static link versions of the library.
Both MinGW and MSVC. Please note that this does not imply any change
to the LGPL licensing, which still imposes psecific conditions on
distributing software that has been statically linked with this library.

* There is a known bug in pthread_once(). Cancellation of the init_routine
exposes a potential starvation (i.e. deadlock) problem if a waiting thread
has a higher priority than the initting thread. This problem will be fixed
in version 3.0.0 of the library.

Bugs fixed
----------

* Fix integer overrun error in sem_timedwait().
Kevin Lussier

* Fix preprocessor directives for static linking.
Dimitar Panayotov


RELEASE 2.1.0
-------------
(2005-03-16)

Bugs fixed
----------

* Reverse change to pthread_setcancelstate() in 2.0.0.


RELEASE 2.0.0
-------------
(2005-03-16)

General
-------

This release represents an ABI change and the DLL version naming has
incremented from 1 to 2, e.g. pthreadVC2.dll.

Version 1.4.0 back-ports the new functionality included in this
release. Please distribute DLLs built from that version with updates
to applications built on pthreads-win32 version 1.x.x.

The package naming has changed, replacing the snapshot date with 
the version number + descriptive information. E.g. this
release is "pthreads-w32-2-0-0-release".

Bugs fixed
----------

* pthread_setcancelstate() no longer checks for a pending
async cancel event if the library is using alertable async
cancel. See the README file (Prerequisites section) for info
on adding alertable async cancelation.

New features
------------

* pthread_once() now supports init_routine cancellability.

New tests
---------

* Agressively test pthread_once() init_routine cancellability.


SNAPSHOT 2005-03-08
-------------------
Version 1.3.0

Bug reports (fixed)
-------------------

* Implicitly created threads leave Win32 handles behind after exiting.
- Dmitrii Semii

* pthread_once() starvation problem.
- Gottlob Frege

New tests
---------

* More intense testing of pthread_once().


SNAPSHOT 2005-01-25
-------------------
Version 1.2.0

Bug fixes
---------

* Attempted acquisition of a recursive mutex could cause waiting threads
to not be woken when the mutex was released.
- Ralf Kubis  <RKubis at mc.com>

* Various package omissions have been fixed.


SNAPSHOT 2005-01-03
-------------------
Version 1.1.0

Bug fixes
---------

* Unlocking recursive or errorcheck mutexes would sometimes
unexpectedly return an EPERM error (bug introduced in