summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--ANNOUNCE7
-rw-r--r--CONTRIBUTORS232
-rw-r--r--ChangeLog8812
-rw-r--r--FAQ806
-rw-r--r--GNUmakefile4
-rw-r--r--Makefile5
-rw-r--r--NEWS66
-rw-r--r--PROGRESS8
-rw-r--r--README923
-rw-r--r--README.NONPORTABLE570
-rw-r--r--TODO14
-rw-r--r--cleanup.c4
-rw-r--r--create.c52
-rw-r--r--dll.c5
-rw-r--r--global.c3
-rw-r--r--implement.h21
-rw-r--r--pthread.h34
-rw-r--r--pthread_cancel.c40
-rw-r--r--pthread_cond_destroy.c1
-rw-r--r--pthread_cond_wait.c21
-rw-r--r--pthread_delay_np.c21
-rw-r--r--pthread_detach.c5
-rw-r--r--pthread_equal.c2
-rw-r--r--pthread_exit.c8
-rw-r--r--pthread_getschedparam.c2
-rw-r--r--pthread_getw32threadhandle_np.c2
-rw-r--r--pthread_join.c19
-rw-r--r--pthread_kill.c10
-rw-r--r--pthread_mutex_init.c2
-rw-r--r--pthread_mutex_unlock.c6
-rw-r--r--pthread_self.c40
-rw-r--r--pthread_setcancelstate.c23
-rw-r--r--pthread_setcanceltype.c23
-rw-r--r--pthread_setschedparam.c9
-rw-r--r--pthread_setspecific.c28
-rw-r--r--pthread_testcancel.c35
-rw-r--r--pthread_win32_attach_detach_np.c46
-rw-r--r--ptw32_InterlockedCompareExchange.c4
-rw-r--r--ptw32_callUserDestroyRoutines.c14
-rw-r--r--ptw32_new.c49
-rw-r--r--ptw32_processTerminate.c13
-rw-r--r--ptw32_reuse.c84
-rw-r--r--ptw32_semwait.c2
-rw-r--r--ptw32_threadDestroy.c13
-rw-r--r--ptw32_threadStart.c65
-rw-r--r--ptw32_throw.c8
-rw-r--r--ptw32_tkAssocCreate.c6
-rw-r--r--ptw32_tkAssocDestroy.c2
-rw-r--r--sem_timedwait.c21
-rw-r--r--sem_wait.c31
-rw-r--r--signal.c8
-rw-r--r--tests/ChangeLog1581
-rw-r--r--tests/Debug.dsp93
-rw-r--r--tests/Debug.dsw29
-rw-r--r--tests/Debug.ncbbin0 -> 66535 bytes
-rw-r--r--tests/Debug.optbin0 -> 53741 bytes
-rw-r--r--tests/Debug.plg35
-rw-r--r--tests/Debug.txt6
-rw-r--r--tests/GNUmakefile10
-rw-r--r--tests/Makefile698
-rw-r--r--tests/README14
-rw-r--r--tests/README.benchtests194
-rw-r--r--tests/cancel1.c2
-rw-r--r--tests/cancel2.c2
-rw-r--r--tests/cancel3.c2
-rw-r--r--tests/cancel4.c2
-rw-r--r--tests/cancel6a.c2
-rw-r--r--tests/cancel6d.c7
-rw-r--r--tests/cancel7.c4
-rw-r--r--tests/cancel8.c4
-rw-r--r--tests/cleanup0.c2
-rw-r--r--tests/cleanup1.c3
-rw-r--r--tests/cleanup2.c2
-rw-r--r--tests/cleanup3.c2
-rw-r--r--tests/condvar1_2.c4
-rw-r--r--tests/condvar3.c2
-rw-r--r--tests/condvar4.c2
-rw-r--r--tests/condvar5.c2
-rw-r--r--tests/condvar6.c4
-rw-r--r--tests/condvar7.c21
-rw-r--r--tests/condvar8.c4
-rw-r--r--tests/condvar9.c56
-rw-r--r--tests/context1.c2
-rw-r--r--tests/errno1.c2
-rw-r--r--tests/exception1.c2
-rw-r--r--tests/exception2.c2
-rw-r--r--tests/exception3.c2
-rw-r--r--tests/exit5.c4
-rw-r--r--tests/reuse1.c13
-rw-r--r--tests/reuse2.c71
-rw-r--r--tests/self1.c6
-rw-r--r--tests/self2.c9
-rw-r--r--tests/semaphore3.c21
-rw-r--r--tests/semaphore4.c138
-rw-r--r--tests/semaphore4t.c132
-rw-r--r--tests/sizes.c3
-rw-r--r--tests/test.h12
-rw-r--r--tests/valid2.c10
-rw-r--r--w32_CancelableWait.c25
99 files changed, 8095 insertions, 7377 deletions
diff --git a/ANNOUNCE b/ANNOUNCE
index 9784535..8528af8 100644
--- a/ANNOUNCE
+++ b/ANNOUNCE
@@ -1,4 +1,4 @@
- PTHREADS-WIN32 SNAPSHOT 2004-06-22
+ PTHREADS-WIN32 SNAPSHOT 2004-11-03
----------------------------------
Web Site: http://sources.redhat.com/pthreads-win32/
FTP Site: ftp://sources.redhat.com/pub/pthreads-win32
@@ -22,13 +22,14 @@ General Public License (LGPL).
Acknowledgements
----------------
-This library is based substantially on a Win32 pthreads
+This library is based originally on a Win32 pthreads
implementation contributed by John Bossom <John.Bossom@cognos.com>.
The implementation of Condition Variables uses algorithms developed
by Alexander Terekhov and Louis Thomas.
-The implementation of POSIX mutexes has been improved by Thomas Pfaff.
+The implementation of POSIX mutexes has been improved by Thomas Pfaff
+and later by Alexander Terekhov.
The implementation of Spinlocks and Barriers was contributed
by Ross Johnson.
diff --git a/CONTRIBUTORS b/CONTRIBUTORS
index de417df..115ce04 100644
--- a/CONTRIBUTORS
+++ b/CONTRIBUTORS
@@ -1,116 +1,116 @@
-Contributors (in approximate order of appearance)
-
-[See also the ChangeLog file where individuals are
-attributed in log entries. Likewise in the FAQ file.]
-
-Ben Elliston bje at cygnus dot com
- Initiated the project;
- setup the project infrastructure (CVS, web page, etc.);
- early prototype routines.
-Ross Johnson rpj at callisto dot canberra dot edu dot au
- early prototype routines;
- ongoing project coordination/maintenance;
- implementation of spin locks and barriers;
- bug fixes;
- documentation;
- testsuite.
-Robert Colquhoun rjc at trump dot net dot au
- Early bug fixes.
-John E. Bossom John dot Bossom at cognos dot com
- Contributed substantial original working implementation;
- bug fixes;
- ongoing guidance and standards interpretation.
-Anders Norlander anorland at hem2 dot passagen dot se
- Early enhancements and runtime checking for supported
- Win32 routines.
-Tor Lillqvist tml at iki dot fi
- General enhancements;
- early bug fixes to condition variables.
-Scott Lightner scott at curriculum dot com
- Bug fix.
-Kevin Ruland Kevin dot Ruland at anheuser-busch dot com
- Various bug fixes.
-Mike Russo miker at eai dot com
- Bug fix.
-Mark E. Armstrong avail at pacbell dot net
- Bug fixes.
-Lorin Hochstein lmh at xiphos dot ca
- general bug fixes; bug fixes to condition variables.
-Peter Slacik Peter dot Slacik at tatramed dot sk
- Bug fixes.
-Mumit Khan khan at xraylith dot wisc dot edu
- Fixes to work with Mingw32.
-Milan Gardian mg at tatramed dot sk
- Bug fixes and reports/analyses of obscure problems.
-Aurelio Medina aureliom at crt dot com
- First implementation of read-write locks.
-Graham Dumpleton Graham dot Dumpleton at ra dot pad dot otc dot telstra dot com dot au
- Bug fix in condition variables.
-Tristan Savatier tristan at mpegtv dot com
- WinCE port.
-Erik Hensema erik at hensema dot xs4all dot nl
- Bug fixes.
-Rich Peters rpeters at micro-magic dot com
-Todd Owen towen at lucidcalm dot dropbear dot id dot au
- Bug fixes to dll loading.
-Jason Nye jnye at nbnet dot nb dot ca
- Implementation of async cancelation.
-Fred Forester fforest at eticomm dot net
-Kevin D. Clark kclark at cabletron dot com
-David Baggett dmb at itasoftware dot com
- Bug fixes.
-Paul Redondo paul at matchvision dot com
-Scott McCaskill scott at 3dfx dot com
- Bug fixes.
-Jef Gearhart jgearhart at tpssys dot com
- Bug fix.
-Arthur Kantor akantor at bexusa dot com
- Mutex enhancements.
-Steven Reddie smr at essemer dot com dot au
- Bug fix.
-Alexander Terekhov TEREKHOV at de dot ibm dot com
- Re-implemented and improved read-write locks;
- (with Louis Thomas) re-implemented and improved
- condition variables;
- enhancements to semaphores;
- system clock change handling re CV timeouts;
- bug fixes.
-Thomas Pfaff tpfaff at gmx dot net
- Changes to make C version usable with C++ applications;
- re-implemented mutex routines to avoid Win32 mutexes
- and TryEnterCriticalSection;
- procedure to fix Mingw32 thread-safety issues.
-Franco Bez franco dot bez at gmx dot de
- procedure to fix Mingw32 thread-safety issues.
-Louis Thomas lthomas at arbitrade dot com
- (with Alexander Terekhov) re-implemented and improved
- condition variables.
-David Korn dgk at research dot att dot com
- Ported to UWIN.
-Phil Frisbie, Jr. phil at hawksoft dot com
- Bug fix.
-Ralf Brese Ralf dot Brese at pdb4 dot siemens dot de
- Bug fix.
-prionx at juno dot com prionx at juno dot com
- Bug fixes.
-Max Woodbury mtew at cds dot duke dot edu
- POSIX versioning conditionals;
- reduced namespace pollution;
- idea to separate routines to reduce statically
- linked image sizes.
-Rob Fanner rfanner at stonethree dot com
- Bug fix.
-Michael Johnson michaelj at maine dot rr dot com
- Bug fix.
-Nicolas Barry boozai at yahoo dot com
- Bug fixes.
-Piet van Bruggen pietvb at newbridges dot nl
- Bug fix.
-Makoto Kato raven at oldskool dot jp
- AMD64 port.
-Panagiotis E. Hadjidoukas peh at hpclab dot ceid dot upatras dot gr
- Contributed the QueueUserAPCEx package which
- makes preemptive async cancelation possible.
-
-Will Bryant will dot bryant at ecosm dot com
- Borland compiler patch and makefile.
+Contributors (in approximate order of appearance)
+
+[See also the ChangeLog file where individuals are
+attributed in log entries. Likewise in the FAQ file.]
+
+Ben Elliston bje at cygnus dot com
+ Initiated the project;
+ setup the project infrastructure (CVS, web page, etc.);
+ early prototype routines.
+Ross Johnson rpj at callisto dot canberra dot edu dot au
+ early prototype routines;
+ ongoing project coordination/maintenance;
+ implementation of spin locks and barriers;
+ bug fixes;
+ documentation;
+ testsuite.
+Robert Colquhoun rjc at trump dot net dot au
+ Early bug fixes.
+John E. Bossom John dot Bossom at cognos dot com
+ Contributed substantial original working implementation;
+ bug fixes;
+ ongoing guidance and standards interpretation.
+Anders Norlander anorland at hem2 dot passagen dot se
+ Early enhancements and runtime checking for supported
+ Win32 routines.
+Tor Lillqvist tml at iki dot fi
+ General enhancements;
+ early bug fixes to condition variables.
+Scott Lightner scott at curriculum dot com
+ Bug fix.
+Kevin Ruland Kevin dot Ruland at anheuser-busch dot com
+ Various bug fixes.
+Mike Russo miker at eai dot com
+ Bug fix.
+Mark E. Armstrong avail at pacbell dot net
+ Bug fixes.
+Lorin Hochstein lmh at xiphos dot ca
+ general bug fixes; bug fixes to condition variables.
+Peter Slacik Peter dot Slacik at tatramed dot sk
+ Bug fixes.
+Mumit Khan khan at xraylith dot wisc dot edu
+ Fixes to work with Mingw32.
+Milan Gardian mg at tatramed dot sk
+ Bug fixes and reports/analyses of obscure problems.
+Aurelio Medina aureliom at crt dot com
+ First implementation of read-write locks.
+Graham Dumpleton Graham dot Dumpleton at ra dot pad dot otc dot telstra dot com dot au
+ Bug fix in condition variables.
+Tristan Savatier tristan at mpegtv dot com
+ WinCE port.
+Erik Hensema erik at hensema dot xs4all dot nl
+ Bug fixes.
+Rich Peters rpeters at micro-magic dot com
+Todd Owen towen at lucidcalm dot dropbear dot id dot au
+ Bug fixes to dll loading.
+Jason Nye jnye at nbnet dot nb dot ca
+ Implementation of async cancelation.
+Fred Forester fforest at eticomm dot net
+Kevin D. Clark kclark at cabletron dot com
+David Baggett dmb at itasoftware dot com
+ Bug fixes.
+Paul Redondo paul at matchvision dot com
+Scott McCaskill scott at 3dfx dot com
+ Bug fixes.
+Jef Gearhart jgearhart at tpssys dot com
+ Bug fix.
+Arthur Kantor akantor at bexusa dot com
+ Mutex enhancements.
+Steven Reddie smr at essemer dot com dot au
+ Bug fix.
+Alexander Terekhov TEREKHOV at de dot ibm dot com
+ Re-implemented and improved read-write locks;
+ (with Louis Thomas) re-implemented and improved
+ condition variables;
+ enhancements to semaphores;
+ system clock change handling re CV timeouts;
+ bug fixes.
+Thomas Pfaff tpfaff at gmx dot net
+ Changes to make C version usable with C++ applications;
+ re-implemented mutex routines to avoid Win32 mutexes
+ and TryEnterCriticalSection;
+ procedure to fix Mingw32 thread-safety issues.
+Franco Bez franco dot bez at gmx dot de
+ procedure to fix Mingw32 thread-safety issues.
+Louis Thomas lthomas at arbitrade dot com
+ (with Alexander Terekhov) re-implemented and improved
+ condition variables.
+David Korn dgk at research dot att dot com
+ Ported to UWIN.
+Phil Frisbie, Jr. phil at hawksoft dot com
+ Bug fix.
+Ralf Brese Ralf dot Brese at pdb4 dot siemens dot de
+ Bug fix.
+prionx at juno dot com prionx at juno dot com
+ Bug fixes.
+Max Woodbury mtew at cds dot duke dot edu
+ POSIX versioning conditionals;
+ reduced namespace pollution;
+ idea to separate routines to reduce statically
+ linked image sizes.
+Rob Fanner rfanner at stonethree dot com
+ Bug fix.
+Michael Johnson michaelj at maine dot rr dot com
+ Bug fix.
+Nicolas Barry boozai at yahoo dot com
+ Bug fixes.
+Piet van Bruggen pietvb at newbridges dot nl
+ Bug fix.
+Makoto Kato raven at oldskool dot jp
+ AMD64 port.
+Panagiotis E. Hadjidoukas peh at hpclab dot ceid dot upatras dot gr
+ Contributed the QueueUserAPCEx package which
+ makes preemptive async cancelation possible.
+
+Will Bryant will dot bryant at ecosm dot com
+ Borland compiler patch and makefile.
diff --git a/ChangeLog b/ChangeLog
index 303cb65..eed0f6b 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,19 +1,59 @@
-2004-10-23 Ross Johnson <ross@localhost.localdomain>
+2004-11-02 Ross Johnson <rpj at callisto.canberra.edu.au>
- * sem_timedwait.c (ptw32_sem_timedwait_cleanup): Release
- Win32 semaphore.
- sem_wait.c (ptw32_sem_wait_cleanup): Likewise.
+ * pthread_cond_wait.c (ptw32_cond_wait_cleanup): Lock CV mutex at
+ start of cleanup handler rather than at the end.
+ * implement.h (PTW32_THREAD_REUSE_EMPTY): Renamed from *_BOTTOM.
+ (ptw32_threadReuseBottom): New global variable.
+ * global.c (ptw32_threadReuseBottom): Declare new variable.
+ * ptw32_reuse.c (ptw32_reuse): Change reuse LIFO stack to LILO queue
+ to more evenly distribute use of reusable thread IDs; use renamed
+ PTW32_THREAD_REUSE_EMPTY.
+ * ptw32_processTerminate.c (ptw2_processTerminate): Use renamed
+ PTW32_THREAD_REUSE_EMPTY.
-2004-10-22 Ross Johnson <rpj at callisto.canberra.edu.au>
+2004-10-31 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * implement.h (PThreadState): Add new state value
+ 'PThreadStateCancelPending'.
+ * pthread_testcancel.c (pthread_testcancel): Use new thread
+ 'PThreadStateCancelPending' state as short cut to avoid entering
+ kernel space via WaitForSingleObject() call. This was obviated
+ by user space sema acquisition in sem_wait() and sem_timedwait(),
+ which are also cancelation points. A call to pthread_testcancel()
+ was required, which introduced a kernel call, effectively nullifying
+ any gains made by the user space sem acquisition checks.
+ * pthread_cancel.c (pthread_cancel): Set new thread
+ 'PThreadStateCancelPending' state.
+
+2004-10-29 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * implement.h (pthread_t): Renamed to ptw32_thread_t; struct contains
+ all thread state.
+ * pthread.h (ptw32_handle_t): New general purpose struct to serve
+ as a handle for various reusable object IDs - currently only used
+ by pthread_t; contains a pointer to ptw32_thread_t (thread state)
+ and a general purpose uint for use as a reuse counter or flags etc.
+ (pthread_t): typedef'ed to ptw32_handle_t; the uint is the reuse
+ counter that allows the library to maintain unique POSIX thread IDs.
+ When the pthread struct reuse stack was introduced, threads would
+ often acquire an identical ID to a previously destroyed thread. The
+ same was true for the pre-reuse stack library, by virtue of pthread_t
+ being the address of the thread struct. The new pthread_t retains
+ the reuse stack but provides virtually unique thread IDs.
+ * sem_wait.c (ptw32_sem_wait_cleanup): New routine used for
+ cancelation cleanup.
+ * sem_timedwait.c (ptw32_sem_timedwait_cleanup): Likewise.
+
+2004-10-22 Ross Johnson <rpj at callisto.canberra.edu.au>
* sem_init.c (sem_init): Introduce a 'lock' element in order to
replace the interlocked operations with conventional serialisation.
- This is needed in order to be able to atomically modify the sema value
- and perform Win32 sema release operations. Win32 semaphores are used
- instead of events in order to support efficient multiple posting.
- If the whole modify/release isn't atomic, a race between sem_timedwait()
- and sem_post() could result in a release when there is no waiting
- semaphore, which would cause too many threads to proceed.
+ This is needed in order to be able to atomically modify the sema
+ value and perform Win32 sema release operations. Win32 semaphores are
+ used instead of events in order to support efficient multiple posting.
+ If the whole modify/release isn't atomic, a race between
+ sem_timedwait() and sem_post() could result in a release when there is
+ no waiting semaphore, which would cause too many threads to proceed.
* sem_wait.c (sem_wait): Use new 'lock'element.
* sem_timedwait.c (sem_timedwait): Likewise.
* sem_trywait.c (sem_trywait): Likewise.
@@ -28,17 +68,27 @@
restore sema value when cancelled.
* sem_wait.c (sem_wait): Likewise.
-2004-10-21 Ross Johnson <rpj at callisto.canberra.edu.au>
+2004-10-21 Ross Johnson <rpj at callisto.canberra.edu.au>
* pthread_mutex_unlock.c (pthread_mutex_unlock): Must use PulseEvent()
rather than SetEvent() to reset the event if there are no waiters.
-2004-10-19 Ross Johnson <rpj at callisto.canberra.edu.au>
+2004-10-19 Ross Johnson <rpj at callisto.canberra.edu.au>
* sem_init.c (sem_init): New semaphore model based on the same idea
as mutexes, i.e. user space interlocked check to avoid
unnecessarily entering kernel space. Wraps the Win32 semaphore and
- keeps it's own counter.
+ keeps it's own counter. Although the motivation to do this has existed
+ for a long time, credit goes to Alexander Terekhov for providing
+ the logic. I have deviated slightly from AT's logic to add the waiters
+ count, which has made the code more complicated by adding cancelation
+ cleanup. This also appears to have broken the VCE (C++ EH) version of
+ the library (the same problem as previously reported - see BUGS #2),
+ only apparently not fixable using the usual workaround, nor by turning
+ all optimisation off. The GCE version works fine, so it is presumed to
+ be a bug in MSVC++ 6.0. The cancelation exception is thrown and caught
+ correctly, but the cleanup class destructor is never called. The failing
+ test is tests\semaphore4.c.
* sem_wait.c (sem_wait): Implemented user space check model.
* sem_post.c (sem_post): Likewise.
* sem_trywait.c (sem_trywait): Likewise.
@@ -48,4370 +98,4370 @@
* ptw32_semwait.c (ptw32_semwait): Likewise.
* implement.h (sem_t_): Add counter element.
-2004-10-15 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * implement.h (othread_mutex_t_): Use an event in place of
- the POSIX semaphore.
- * pthread_mutex_init.c: Create the event; remove semaphore init.
- * pthread_mutex_destroy.c: Delete the event.
- * pthread_mutex_lock.c: Replace the semaphore wait with the event wait.
- * pthread_mutex_trylock.c: Likewise.
- * pthread_mutex_timedlock.c: Likewise.
- * pthread_mutex_unlock.c: Set the event.
-
-2004-10-14 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread_mutex_lock.c (pthread_mutex_lock): New algorithm using
- Terekhov's xchg based variation of Drepper's cmpxchg model.
- Theoretically, xchg uses fewer clock cycles than cmpxchg (using IA-32
- as a reference), however, in my opinion bus locking dominates the
- equation on smp systems, so the model with the least number of bus
- lock operations in the execution path should win, which is Terekhov's
- variant. On IA-32 uni-processor systems, it's faster to use the
- CMPXCHG instruction without locking the bus than to use the XCHG
- instruction, which always locks the bus. This makes the two variants
- equal for the non-contended lock (fast lane) execution path on up
- IA-32. Testing shows that the xchg variant is faster on up IA-32 as
- well if the test forces higher lock contention frequency, even though
- kernel calls should be dominating the times (on up IA-32, both
- variants used CMPXCHG instructions and neither locked the bus).
- * pthread_mutex_timedlock.c pthread_mutex_timedlock(): Similarly.
- * pthread_mutex_trylock.c (pthread_mutex_trylock): Similarly.
- * pthread_mutex_unlock.c (pthread_mutex_unlock): Similarly.
- * ptw32_InterlockedCompareExchange.c (ptw32_InterlockExchange): New
- function.
- (PTW32_INTERLOCKED_EXCHANGE): Sets up macro to use inlined
- ptw32_InterlockedExchange.
- * implement.h (PTW32_INTERLOCKED_EXCHANGE): Set default to
- InterlockedExchange().
- * Makefile: Building using /Ob2 so that asm sections within inline
- functions are inlined.
-
-2004-10-08 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread_mutex_destroy.c (pthread_mutex_destroy): Critical Section
- element is no longer required.
- * pthread_mutex_init.c (pthread_mutex_init): Likewise.
- * pthread_mutex_lock.c (pthread_mutex_lock): New algorithm following
- Drepper's paper at http://people.redhat.com/drepper/futex.pdf, but
- using the existing semaphore in place of the futex described in the
- paper. Idea suggested by Alexander Terekhov - see:
- http://sources.redhat.com/ml/pthreads-win32/2003/msg00108.html
- * pthread_mutex_timedlock.c pthread_mutex_timedlock(): Similarly.
- * pthread_mutex_trylock.c (pthread_mutex_trylock): Similarly.
- * pthread_mutex_unlock.c (pthread_mutex_unlock): Similarly.
- * pthread_barrier_wait.c (pthread_barrier_wait): Use inlined version
- of InterlockedCompareExchange() if possible - determined at
- build-time.
- * pthread_spin_destroy.c pthread_spin_destroy(): Likewise.
- * pthread_spin_lock.c pthread_spin_lock():Likewise.
- * pthread_spin_trylock.c (pthread_spin_trylock):Likewise.
- * pthread_spin_unlock.c (pthread_spin_unlock):Likewise.
- * ptw32_InterlockedCompareExchange.c: Sets up macro for inlined use.
- * implement.h (pthread_mutex_t_): Remove Critical Section element.
- (PTW32_INTERLOCKED_COMPARE_EXCHANGE): Set to default non-inlined
- version of InterlockedCompareExchange().
- * private.c: Include ptw32_InterlockedCompareExchange.c first for
- inlining.
- * GNUmakefile: Add commandline option to use inlined
- InterlockedCompareExchange().
- * Makefile: Likewise.
-
-2004-09-27 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread_mutex_lock.c (pthread_mutex_lock): Separate
- PTHREAD_MUTEX_NORMAL logic since we do not need to keep or check some
- state required by other mutex types; do not check mutex pointer arg
- for validity - leave this to the system since we are only checking
- for NULL pointers. This should improve speed of NORMAL mutexes and
- marginally improve speed of other type.
- * pthread_mutex_trylock.c (pthread_mutex_trylock): Likewise.
- * pthread_mutex_unlock.c (pthread_mutex_unlock): Likewise; also avoid
- entering the critical section for the no-waiters case, with approx.
- 30% reduction in lock/unlock overhead for this case.
- * pthread_mutex_timedlock.c (pthread_mutex_timedlock): Likewise; also
- no longer keeps mutex if post-timeout second attempt succeeds - this
- will assist applications that wish to impose strict lock deadlines,
- rather than simply to escape from frozen locks.
-
-2004-09-09 Tristan Savatier <tristan at mpegtv.com>
- * pthread.h (struct pthread_once_t_): Qualify the 'done' element
- as 'volatile'.
- * pthread_once.c: Concerned about possible race condition,
- specifically on MPU systems re concurrent access to multibyte types.
- [Maintainer's note: the race condition is harmless on SPU systems
- and only a problem on MPU systems if concurrent access results in an
- exception (presumably generated by a hardware interrupt). There are
- other instances of similar harmless race conditions that have not
- been identified as issues.]
-
-2004-09-09 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread.h: Declare additional types as volatile.
-
-2004-08-27 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread_barrier_wait.c (pthread_barrier_wait): Remove excessive code
- by substituting the internal non-cancelable version of sem_wait
- (ptw32_semwait).
-
-2004-08-25 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread_join.c (pthread_join): Rewrite and re-order the conditional
- tests in an attempt to improve efficiency and remove a race
- condition.
-
-2004-08-23 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * create.c (pthread_create): Don't create a thread if the thread
- id pointer location (first arg) is inaccessible. A memory
- protection fault will result if the thread id arg isn't an accessible
- location. This is consistent with GNU/Linux but different to
- Solaris or MKS (and possibly others), which accept NULL as meaning
- 'don't return the created thread's ID'. Applications that run
- using pthreads-win32 will run on all other POSIX threads
- implementations, at least w.r.t. this feature.
-
- It was decided not to copy the Solaris et al behaviour because,
- although it would have simplified some application porting (but only
- from Solaris to Windows), the feature is not technically necessary,
- and the alternative segfault behaviour helps avoid buggy application
- code.
-
-2004-07-01 Anuj Goyal <anuj.goyal at gmail.com>
-
- * builddmc.bat: New; Windows bat file to build the library.
- * config.h (__DMC__): Support for Digital Mars compiler.
- * create.c (__DMC__): Likewise.
- * pthread_exit.c (__DMC__): Likewise.
- * pthread_join.c (__DMC__): Likewise.
- * ptw32_threadDestroy.c (__DMC__): Likewise.
- * ptw32_threadStart.c (__DMC__): Likewise.
- * ptw32_throw.c (__DMC__): Likewise.
-
-2004-06-29 Anuj Goyal <anuj.goyal at gmail.com>
-
- * pthread.h (__DMC__): Initial support for Digital Mars compiler.
-
-2004-06-29 Will Bryant <will.bryant at ecosm.com>
-
- * README.Borland: New; description of Borland changes.
- * Bmakefile: New makefile for the Borland make utility.
- * ptw32_InterlockedCompareExchange.c:
- Add Borland compatible asm code.
-
-2004-06-26 Jason Bard <BardJA at Npt.NUWC.Navy.Mil>
-
- * pthread.h (HAVE_STRUCT_TIMESPEC): If undefined, define it
- to avoid timespec struct redefined errors elsewhere in an
- application.
-
-2004-06-21 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread.h (PTHREAD_RECURSIVE_MUTEX_INITIALIZER): Mutex
- initialiser added for compatibility with Linux threads and
- others; currently not included in SUSV3.
- * pthread.h (PTHREAD_ERRORCHECK_MUTEX_INITIALIZER): Likewise.
- * pthread.h (PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP): Likewise.
- * pthread.h (PTHREAD_ERRORCHECK_MUTEX_INITIALIZER_NP): Likewise.
-
- * ptw32_mutex_check_need_init.c (ptw32_mutex_check_need_init):
- Add new initialisers.
-
- * pthread_mutex_lock.c (pthread_mutex_lock): Check for new
- initialisers.
- * pthread_mutex_trylock.c (pthread_mutex_trylock): Likewise.
- * pthread_mutex_timedlock.c (pthread_mutex_timedlock): Likewise.
- * pthread_mutex_unlock.c (pthread_mutex_unlock): Likewise.
- * pthread_mutex_destroy.c (pthread_mutex_destroy): Likewise.
-
-2004-05-20 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * README.NONPORTABLE: Document pthread_win32_test_features_np().
- * FAQ: Update various answers.
-
-2004-05-19 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * Makefile: Don't define _WIN32_WINNT on compiler command line.
- * GNUmakefile: Likewise.
-
-2004-05-16 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread_cancel.c (pthread_cancel): Adapted to use auto-detected
- QueueUserAPCEx features at run-time.
- (ptw32_RegisterCancelation): Drop in replacement for QueueUserAPCEx()
- if it can't be used. Provides older style non-preemptive async
- cancelation.
- * pthread_win32_attach_detach_np.c (pthread_win32_attach_np):
- Auto-detect quserex.dll and the availability of alertdrv.sys;
- initialise and close on process attach/detach.
- * global.c (ptw32_register_cancelation): Pointer to either
- QueueUserAPCEx() or ptw32_RegisterCancelation() depending on
- availability. QueueUserAPCEx makes pre-emptive async cancelation
- possible.
- * implement.h: Add definitions and prototypes related to QueueUserAPC.
-
-2004-05-16 Panagiotis E. Hadjidoukas <peh at hpclab.ceid.upatras.gr>
-
- * QueueUserAPCEx (separate contributed package): Provides preemptive
- APC feature.
- * pthread_cancel.c (pthread_cancel): Initial integration of
- QueueUserAPCEx into pthreads-win32 to provide true pre-emptive
- async cancelation of threads, including blocked threads.
-
-2004-05-06 Makoto Kato <raven at oldskool.jp>
-
- * pthread.h (DWORD_PTR): Define typedef for older MSVC.
- * pthread_cancel.c (AMD64): Add architecture specific Context register.
- * ptw32_getprocessors.c: Use correct types (DWORD_PTR) for mask
- variables.
-
-2004-04-06 P. van Bruggen <pietvb at newbridges.nl>
-
- * ptw32_threadDestroy.c: Destroy threadLock mutex to
- close a memory leak.
-
-2004-02-13 Gustav Hallberg <gustav at virtutech.com>
-
- * pthread_equal.c: Remove redundant equality logic.
-
-2003-12-10 Philippe Di Cristo <philipped at voicebox.com>
-
- * sem_timedwait.c (sem_timedwait): Fix timeout calculations.
-
-2003-10-20 Alexander Terekhov <TEREKHOV at de.ibm.com>
-
- * pthread_mutex_timedlock.c (ptw32_semwait): Move to individual module.
- * ptw32_semwait.c: New module.
- * pthread_cond_wait.c (ptw32_cond_wait_cleanup): Replace cancelable
- sem_wait() call with non-cancelable ptw32_semwait() call.
- * pthread.c (private.c): Re-order for inlining. GNU C warned that
- function ptw32_semwait() was defined 'inline' after it was called.
- * pthread_cond_signal.c (ptw32_cond_unblock): Likewise.
- * pthread_delay_np.c: Disable Watcom warning with comment.
- * *.c (process.h): Remove include from .c files. This is conditionally
- included by the common project include files.
-
-2003-10-20 James Ewing <james.ewing at sveasoft.com>
-
- * ptw32_getprocessors.c: Some Win32 environments don't have
- GetProcessAffinityMask(), so always return CPU count = 1 for them.
- * config.h (NEED_PROCESSOR_AFFINITY_MASK): Define for WinCE.
-
-2003-10-15 Ross Johnson <ross at callisto.canberra.edu.au>
-
- * Re-indented all .c files using default GNU style to remove assorted
- editor ugliness (used GNU indent utility in default style).
-
-2003-10-15 Alex Blanco <Alex.Blanco at motorola.com>
-
- * sem_init.c (sem_init): Would call CreateSemaphore even if the sema
- struct calloc failed; was not freeing calloced memory if either
- CreateSemaphore or CreateEvent failed.
-
-2003-10-14 Ross Johnson <ross at callisto.canberra.edu.au>
-
- * pthread.h: Add Watcom compiler compatibility. Esssentially just add
- the cdecl attribute to all exposed function prototypes so that Watcom
- generates function call code compatible with non-Watcom built libraries.
- By default, Watcom uses registers to pass function args if possible rather
- than pushing to stack.
- * semaphore.h: Likewise.
- * sched.h: Likewise.
- * pthread_cond_wait.c (ptw32_cond_wait_cleanup): Define with cdecl attribute
- for Watcom compatibility. This routine is called via pthread_cleanup_push so
- it had to match function arg definition.
- * Wmakefile: New makefile for Watcom builds.
-
-2003-09-14 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread_setschedparam.c (pthread_setschedparam): Attempt to map
- all priority levels between max and min (as returned by
- sched_get_priority_min/max) to reasonable Win32 priority levels - i.e.
- levels between THREAD_PRIORITY_LOWEST/IDLE to THREAD_PRIORITY_LOWEST and
- between THREAD_PRIORITY_HIGHEST/TIME_CRITICAL to THREAD_PRIORITY_HIGHEST
- while others remain unchanged; record specified thread priority level
- for return by pthread_getschedparam.
-
- Note that, previously, specified levels not matching Win32 priority levels
- would silently leave the current thread priority unaltered.
-
- * pthread_getschedparam.c (pthread_getschedparam): Return the priority
- level specified by the latest pthread_setschedparam or pthread_create rather
- than the actual running thread priority as returned by GetThreadPriority - as
- required by POSIX. I.e. temporary or adjusted actual priority levels are not
- returned by this routine.
-
- * pthread_create.c (pthread_create): For priority levels specified via
- pthread attributes, attempt to map all priority levels between max and
- min (as returned by sched_get_priority_min/max) to reasonable Win32
- priority levels; record priority level given via attributes, or
- inherited from parent thread, for later return by pthread_getschedparam.
-
- * ptw32_new.c (ptw32_new): Initialise pthread_t_ sched_priority element.
-
- * pthread_self.c (pthread_self): Set newly created implicit POSIX thread
- sched_priority to Win32 thread's current actual priority. Temporarily
- altered priorities can't be avoided in this case.
-
- * implement.h (struct pthread_t_): Add new sched_priority element.
-
-2003-09-12 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * sched_get_priority_min.c (sched_get_priority_min): On error should return -1
- with errno set.
- * sched_get_priority_max.c (sched_get_priority_max): Likewise.
-
-2003-09-03 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * w32_cancelableWait.c (ptw32_cancelable_wait): Allow cancelation
- of implicit POSIX threads as well.
-
-2003-09-02 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread_win32_attach_detach_np.c (pthread_win32_thread_detach_np):
- Add comment.
-
- * pthread_exit.c (pthread_exit): Fix to recycle the POSIX thread handle in
- addition to calling user TSD destructors. Move the implicit POSIX thread exit
- handling to ptw32_throw to centralise the logic.
-
- * ptw32_throw.c (ptw32_throw): Implicit POSIX threads have no point
- to jump or throw to, so cleanup and exit the thread here in this case. For
- processes using the C runtime, the exit code will be set to the POSIX
- reason for the throw (i.e. PTHREAD_CANCEL or the value given to pthread_exit).
- Note that pthread_exit() already had similar logic, which has been moved to
- here.
-
- * ptw32_threadDestroy.c (ptw32_threadDestroy): Don't close the Win32 handle
- of implicit POSIX threads - expect this to be done by Win32?
-
-2003-09-01 Ross Johnson <rpj at callisto.canberra.edu.au>
-
- * pthread_self.c (pthread_self): The newly aquired pthread_t must be
- assigned to the reuse stack, not freed, if the routine fails somehow.
-
-2003-08-13 Ross Johnson <rpj at ise.canberra.edu.au>
-
- * pthread_getschedparam.c (pthread_getschedparam): An invalid thread ID
- parameter was returning an incorrect error value; now uses a more exhaustive
- check for validity.
-
- * pthread_setschedparam.c (pthread_setschedparam): Likewise.
-
- * pthread_join.c (pthread_join): Now uses a more exhaustive
- check for validity.
-
- * pthread_detach.c (pthread_detach): Likewise.
-
- * pthread_cancel.c (pthread_cancel): Likewise.
-
- * ptw32_threadDestroy.c (ptw32_threadDestroy): pthread_t structs are
- never freed - push them onto a stack for reuse.
-
- * ptw32_new.c (ptw32_new): Check for reusable pthread_t before dynamically
- allocating new memory for the struct.
-
- * pthread_kill.c (pthread_kill): New file; new routine; takes only a zero
- signal arg so that applications can check the thread arg for validity; checks
- that the underlying Win32 thread HANDLE is valid.
-
- * pthread.h (pthread_kill): Add prototype.
-
- * ptw32_reuse.c (ptw32_threadReusePop): New file; new routine; pop a
- pthread_t off the reuse stack. pthread_t_ structs that have been destroyed, i.e.
- have exited detached or have been joined, are cleaned up and put onto a reuse
- stack. Consequently, thread IDs are no longer freed once calloced. The library
- will attempt to get a struct off this stack before asking the system to alloc
- new memory when creating threads. The stack is guarded by a global mutex.
- (ptw32_threadReusePush): New routine; push a pthread_t onto the reuse stack.
-
- * implement.h (ptw32_threadReusePush): Add new prototype.
- (ptw32_threadReusePop): Likewise.
- (pthread_t): Add new element.
-
- * ptw32_processTerminate.c (ptw32_processTerminate): Delete the thread
- reuse lock; free all thread ID structs on the thread reuse stack.
-
- * ptw32_processInitialize.c (ptw32_processInitialize): Initialise the
- thread reuse lock.
-
-2003-07-19 Ross Johnson <rpj at ise.canberra.edu.au>
-
- * GNUmakefile: modified to work under MsysDTK environment.
- * pthread_spin_lock.c (pthread_spin_lock): Check for NULL arg.
- * pthread_spin_unlock.c (pthread_spin_unlock): Likewise.
- * pthread_spin_trylock.c (pthread_spin_trylock): Likewise;
- fix incorrect pointer value if lock is dynamically initialised by
- this function.
- * sem_init.c (sem_init): Initialise sem_t value to quell compiler warning.
- * sem_destroy.c (sem_destroy): Likewise.
- * ptw32_threadStart.c (non-MSVC code sections): Include <exception> rather
- than old-style <new.h>; fix all std:: namespace entities such as
- std::terminate_handler instances and associated methods.
- * ptw32_callUserDestroyRoutines.c (non-MSVC code sections): Likewise.
-
-2003-06-24 Piet van Bruggen <pietvb at newbridges.nl>
-
- * pthread_spin_destroy.c (pthread_spin_destroy): Was not freeing the
- spinlock struct.
-
-2003-06-22 Nicolas Barry <boozai at yahoo.com>
-
- * pthread_mutex_destroy.c (pthread_mutex_destroy): When called
- with a recursive mutex that was locked by the current thread, the
- function was failing with a success return code.
-
-2003-05-15 Steven Reddie <Steven.Reddie at ca.com>
-
- * pthread_win32_attach_detach_np.c (pthread_win32_process_detach_np):
- NULLify ptw32_selfThreadKey after the thread is destroyed, otherwise
- destructors calling pthreads routines might resurrect it again, creating
- memory leaks. Call the underlying Win32 Tls routine directly rather than
- pthread_setspecific().
- (pthread_win32_thread_detach_np): Likewise.
-
-2003-05-14 Viv <vcotirlea at hotmail.com>
-
- * pthread.dsp: Change /MT compile flag to /MD.
-
-2003-03-04 Alexander Terekhov <TEREKHOV at de.ibm.com>
-
- * pthread_mutex_timedlock.c (pthread_mutex_timedlock): Fix failure to
- set ownership of mutex on second grab after abstime timeout.
- - bug reported by Robert Strycek <strycek at posam.sk>
-
-2002-12-17 Thomas Pfaff <tpfaff at gmx.net>
-
- * pthread_mutex_lock.c (ptw32_semwait): New static routine to provide
- a non-cancelable sem_wait() function. This is consistent with the
- way that pthread_mutex_timedlock.c does it.
- (pthread_mutex_lock): Use ptw32_semwait() instead of sem_wait().
-
-2002-12-11 Thomas Pfaff <tpfaff at gmx.net>
-
- * pthread_mutex_trylock.c: Should return EBUSY rather than EDEADLK.
- * pthread_mutex_destroy.c: Remove redundant ownership test (the
- trylock call does this for us); do not destroy a recursively locked
- mutex.
-
-2002-09-20 Michael Johnson <michaelj at maine.rr.com>
-
- * pthread_cond_destroy.c (pthread_cond_destroy):
- When two different threads exist, and one is attempting to
- destroy a condition variable while the other is attempting to
- initialize a condition variable that was created with
- PTHREAD_COND_INITIALIZER, a deadlock can occur. Shrink
- the ptw32_cond_list_lock critical section to fix it.
-
-2002-07-31 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * ptw32_threadStart.c (ptw32_threadStart): Thread cancelLock
- destruction moved to ptw32_threadDestroy().
-
- * ptw32_threadDestroy.c (ptw32_threadDestroy): Destroy
- the thread's cancelLock. Moved here from ptw32_threadStart.c
- to cleanup implicit threads as well.
-
-2002-07-30 Alexander Terekhov <TEREKHOV at de.ibm.com>
-
- * pthread_cond_wait.c (ptw32_cond_wait_cleanup):
- Remove code designed to avoid/prevent spurious wakeup
- problems. It is believed that the sem_timedwait() call
- is consuming a CV signal that it shouldn't and this is
- breaking the avoidance logic.
-
-2002-07-30 Ross Johnson <rpj at ise.canberra.edu.au>
-
- * sem_timedwait.c (sem_timedwait): Tighten checks for
- unreasonable abstime values - that would result in
- unexpected timeout values.
-
- * w32_CancelableWait.c (ptw32_cancelable_wait):
- Tighten up return value checking and add comments.
-
-
-2002-06-08 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * sem_getvalue.c (sem_getvalue): Now returns a value for the
- NEED_SEM version (i.e. earlier versions of WinCE).
-
-
-2002-06-04 Rob Fanner <rfanner at stonethree.com>
-
- * sem_getvalue.c (sem_getvalue): The Johnson M. Hart
- approach didn't work - we are forced to take an
- intrusive approach. We try to decrement the sema
- and then immediately release it again to get the
- value. There is a small probability that this may
- block other threads, but only momentarily.
-
-2002-06-03 Ross Johnson <rpj at ise.canberra.edu.au>
-
- * sem_init.c (sem_init): Initialise Win32 semaphores
- to _POSIX_SEM_VALUE_MAX (which this implementation
- defines in pthread.h) so that sem_getvalue() can use
- the trick described in the comments in sem_getvalue().
- * pthread.h (_POSIX_SEM_VALUE_MAX): Defined.
- (_POSIX_SEM_NSEMS_MAX): Defined - not used but may be
- useful for source code portability.
-
-2002-06-03 Rob Fanner <rfanner at stonethree.com>
-
- * sem_getvalue.c (sem_getvalue): Did not work on NT.
- Use approach suggested by Johnson M. Hart in his book
- "Win32 System Programming".
-
-2002-02-28 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * errno.c: Compiler directive was incorrectly including code.
- * pthread.h: Conditionally added some #defines from config.h
- needed when not building the library. e.g. NEED_ERRNO, NEED_SEM.
- (PTW32_DLLPORT): Now only defined if _DLL defined.
- (_errno): Compiler directive was incorrectly including prototype.
- * sched.h: Conditionally added some #defines from config.h
- needed when not building the library.
- * semaphore.h: Replace an instance of NEED_SEM that should
- have been NEED_ERRNO. This change currently has nil effect.
-
- * GNUmakefile: Correct some recent changes.
-
- * Makefile: Add rule to generate pre-processor output.
-
-2002-02-23 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * pthread_rwlock_timedrdlock.c: New - untested.
- * pthread_rwlock_timedwrlock.c: New - untested.
-
- * Testsuite passed (except known MSVC++ problems)
-
- * pthread_cond_destroy.c: Expand the time change
- critical section to solve deadlock problem.
-
- * pthread.c: Add all remaining C modules.
- * pthread.h: Use dllexport/dllimport attributes on functions
- to avoid using pthread.def.
- * sched.h: Likewise.
- * semaphore.h: Likewise.
- * GNUmakefile: Add new targets for single translation
- unit build to maximise inlining potential; generate
- pthread.def automatically.
- * Makefile: Likewise, but no longer uses pthread.def.
-
-2002-02-20 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * pthread_cond_destroy.c (pthread_cond_destroy):
- Enter the time change critical section earlier.
-
-2002-02-17 Ross Johnson <rpj at setup1.ise.canberra.edu.au
-
- * Testsuite passed.
-
- * pthread_timechange_handler_np.c: New; following
- a suggestion from Alexander Terekhov that CVs should
- be broadcast so that they all re-evaluate their
- condition variables and reset a new timeout if
- required, whenever an application receives a
- WM_TIMECHANGE message. This message indicates that
- the system time has been changed. Therefore, CVs
- waiting for a timeout set as an abs_time will possibly
- not wake up at the expected time. Some applications
- may not be tolerant of this.
- * pthread_cond_init.c: Add CV to linked list.
- * pthread_cond_destroy.c: Remove CV from linked list.
- * global.c (ptw32_cond_list_head): New variable.
- (ptw32_cond_list_tail): New variable.
- (ptw32_cond_list_cs): New critical section.
- * ptw32_processInitialize (ptw32_cond_list_cs): Initialize.
- * ptw32_processTerminate (ptw32_cond_list_cs): Delete.
-
-
- * Reduce executable size.
- -----------------------
- When linking with the static library, only those
- routines actually called, either directly or indirectly
- should be included.
-
- [Gcc has the -ffunction-segments option to do this but MSVC
- doesn't have this feature as far as I can determine. Other
- compilers are undetermined as well. - rpj]
-
- * spin.c: Split file into function segments.
- * ptw32_spinlock_check_need_init.c: Separated routine from spin.c.
- * pthread_spin_init.c: Likewise.
- * pthread_spin_destroy.c: Likewise.
- * pthread_spin_lock.c: Likewise.
- * pthread_spin_unlock.c: Likewise.
- * pthread_spin_trylock.c: Likewise.
-
- * sync.c: Split file into function segments.
- * pthread_detach.c: Separated routine from sync.c.
- * pthread_join.c: Likewise.
-
- * tsd.c: Split file into function segments.
- * pthread_key_create.c: Separated routine from tsd.c.
- * pthread_key_delete.c: Likewise.
- * pthread_setspecific.c: Likewise.
- * pthread_getspecific.c: Likewise.
-
- * sched.c: Split file into function segments.
- * pthread_attr_setschedpolicy.c: Separated routine from sched.c.
- * pthread_attr_getschedpolicy.c: Likewise.
- * pthread_attr_setschedparam.c: Likewise.
- * pthread_attr_getschedparam.c: Likewise.
- * pthread_attr_setinheritsched.c: Likewise.
- * pthread_attr_getinheritsched.c: Likewise.
- * pthread_setschedparam.c: Likewise.
- * pthread_getschedparam.c: Likewise.
- * sched_get_priority_max.c: Likewise.
- * sched_get_priority_min.c: Likewise.
- * sched_setscheduler.c: Likewise.
- * sched_getscheduler.c: Likewise.
- * sched_yield.c: Likewise.
-
-
-2002-02-16 Ross Johnson <rpj at setup1.ise.canberra.edu.au
-
- Reduce executable size.
- -----------------------
- When linking with the static library, only those
- routines actually called, either directly or indirectly
- should be included.
-
- [Gcc has the -ffunction-segments option to do this but MSVC
- doesn't have this feature as far as I can determine. Other
- compilers are undetermined as well. - rpj]
-
- * mutex.c: Split file into function segments.
- * pthread_mutexattr_destroy.c: Separated routine from mutex.c
- * pthread_mutexattr_getpshared.c: Likewise.
- * pthread_mutexattr_gettype.c: Likewise.
- * pthread_mutexattr_init.c: Likewise.
- * pthread_mutexattr_setpshared.c: Likewise.
- * pthread_mutexattr_settype.c: Likewise.
- * ptw32_mutex_check_need_init.c: Likewise.
- * pthread_mutex_destroy.c: Likewise.
- * pthread_mutex_init.c: Likewise.
- * pthread_mutex_lock.c: Likewise.
- * pthread_mutex_timedlock.c: Likewise.
- * pthread_mutex_trylock.c: Likewise.
- * pthread_mutex_unlock.c: Likewise.
-
- * private.c: Split file into function segments.
- * ptw32_InterlockedCompareExchange.c: Separated routine from private.c
- * ptw32_callUserDestroyRoutines.c: Likewise.
- * ptw32_getprocessors.c: Likewise.
- * ptw32_processInitialize.c: Likewise.
- * ptw32_processTerminate.c: Likewise.
- * ptw32_threadDestroy.c: Likewise.
- * ptw32_threadStart.c: Likewise.
- * ptw32_throw.c: Likewise.
- * ptw32_timespec.c: Likewise.
- * ptw32_tkAssocCreate.c: Likewise.
- * ptw32_tkAssocDestroy.c: Likewise.
-
- * rwlock.c: Split file into function segments.
- * pthread_rwlockattr_destroy.c: Separated routine from rwlock.c
- * pthread_rwlockattr_getpshared.c: Likewise.
- * pthread_rwlockattr_init.c: Likewise.
- * pthread_rwlockattr_setpshared.c: Likewise.
- * ptw32_rwlock_check_need_init.c: Likewise.
- * pthread_rwlock_destroy.c: Likewise.
- * pthread_rwlock_init.c: Likewise.
- * pthread_rwlock_rdlock.c: Likewise.
- * pthread_rwlock_tryrdlock.c: Likewise.
- * pthread_rwlock_trywrlock.c: Likewise.
- * pthread_rwlock_unlock.c: Likewise.
- * pthread_rwlock_wrlock.c: Likewise.
-
-2002-02-10 Ross Johnson <rpj at setup1.ise.canberra.edu.au
-
- Reduce executable size.
- -----------------------
- When linking with the static library, only those
- routines actually called, either directly or indirectly
- should be included.
-
- [Gcc has the -ffunction-segments option to do this but MSVC
- doesn't have this feature as far as I can determine. Other
- compilers are undetermined as well. - rpj]
-
- * nonportable.c: Split file into function segments.
- * np_delay.c: Separated routine from nonportable.c
- * np_getw32threadhandle.c: Likewise.
- * np_mutexattr_setkind.c: Likewise.
- * np_mutexattr_getkind.c: Likewise.
- * np_num_processors.c: Likewise.
- * np_win32_attach_detach.c: Likewise.
-
- * misc.c: Split file into function segments.
- * pthread_equal.c: Separated routine from nonportable.c.
- * pthread_getconcurrency.c: Likewise.
- * pthread_once.c: Likewise.
- * pthread_self.c: Likewise.
- * pthread_setconcurrency.c: Likewise.
- * ptw32_calloc.c: Likewise.
- * ptw32_new.c: Likewise.
- * w32_CancelableWait.c: Likewise.
-
-2002-02-09 Ross Johnson <rpj at setup1.ise.canberra.edu.au
-
- Reduce executable size.
- -----------------------
- When linking with the static library, only those
- routines actually called, either directly or indirectly
- should be included.
-
- [Gcc has the -ffunction-segments option to do this but MSVC
- doesn't have this feature as far as I can determine. Other
- compilers are undetermined as well. - rpj]
-
- * condvar.c: Split file into function segments.
- * pthread_condattr_destroy.c: Separated routine from condvar.c.
- * pthread_condattr_getpshared.c: Likewise.
- * pthread_condattr_init.c: Likewise.
- * pthread_condattr_setpshared.c: Likewise.
- * ptw32_cond_check_need_init.c: Likewise.
- * pthread_cond_destroy.c: Likewise.
- * pthread_cond_init.c: Likewise.
- * pthread_cond_signal.c: Likewise.
- * pthread_cond_wait.c: Likewise.
-
-2002-02-07 Alexander Terekhov<TEREKHOV at de.ibm.com>
-
- * nonportable.c (pthread_delay_np): Make a true
- cancelation point. Deferred cancels will interrupt the
- wait.
-
-2002-02-07 Ross Johnson <rpj at setup1.ise.canberra.edu.au
-
- * misc.c (ptw32_new): Add creation of cancelEvent so that
- implicit POSIX threads (Win32 threads with a POSIX face)
- are cancelable; mainly so that pthread_delay_np doesn't fail
- if called from the main thread.
- * create.c (pthread_create): Remove creation of cancelEvent
- from here; now in ptw32_new().
-
- Reduce executable size.
- -----------------------
- When linking with the static library, only those
- routines actually called, either directly or indirectly
- should be included.
-
- [Gcc has the -ffunction-segments option to do this but MSVC
- doesn't have this feature as far as I can determine. Other
- compilers are undetermined as well. - rpj]
-
- * barrier.c: All routines are now in separate compilation units;
- This file is used to congregate the separate modules for
- potential inline optimisation and backward build compatibility.
- * cancel.c: Likewise.
- * pthread_barrierattr_destroy.c: Separated routine from cancel.c.
- * pthread_barrierattr_getpshared.c: Likewise.
- * pthread_barrierattr_init.c: Likewise.
- * pthread_barrierattr_setpshared.c: Likewise.
- * pthread_barrier_destroy.c: Likewise.
- * pthread_barrier_init.c: Likewise.
- * pthread_barrier_wait.c: Likewise.
- * pthread_cancel.c: Likewise.
- * pthread_setcancelstate.c: Likewise.
- * pthread_setcanceltype.c: Likewise.
- * pthread_testcancel.c: Likewise.
-
-2002-02-04 Max Woodbury <mtew at cds.duke.edu>
-
- Reduced name space pollution.
- -----------------------------
- When the appropriate symbols are defined, the headers
- will restrict the definitions of new names. In particular,
- it must be possible to NOT include the <windows.h>
- header and related definitions with some combination
- of symbol definitions. Secondly, it should be possible
- that additional definitions should be limited to POSIX
- compliant symbols by the definition of appropriate symbols.
-
- * pthread.h: POSIX conditionals.
- * sched.h: POSIX conditionals.
- * semaphore.h: POSIX conditionals.
-
- * semaphore.c: Included <limits.h>.
- (sem_init): Changed magic 0x7FFFFFFFL to INT_MAX.
- (sem_getvalue): Trial version.
-
- Reduce executable size.
- -----------------------
- When linking with the static library, only those
- routines actually called, either directly or indirectly
- should be included.
-
- [Gcc has the -ffunction-segments option to do this but MSVC
- doesn't have this feature as far as I can determine. Other
- compilers are undetermined as well. - rpj]
-
- * semaphore.c: All routines are now in separate compilation units;
- This file is used to congregate the separate modules for
- potential inline optimisation and backward build compatibility.
- * sem_close.c: Separated routine from semaphore.c.
- * ptw32_decrease_semaphore.c: Likewise.
- * sem_destroy.c: Likewise.
- * sem_getvalue.c: Likewise.
- * ptw32_increase_semaphore.c: Likewise.
- * sem_init.c: Likewise.
- * sem_open.c: Likewise.
- * sem_post.c: Likewise.
- * sem_post_multiple.c: Likewise.
- * sem_timedwait.c: Likewise.
- * sem_trywait.c: Likewise.
- * sem_unlink.c: Likewise.
- * sem_wait.c: Likewise.
-
-2002-02-04 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- The following extends the idea above to the rest of pthreads-win32 - rpj
-
- * attr.c: All routines are now in separate compilation units;
- This file is used to congregate the separate modules for
- potential inline optimisation and backward build compatibility.
- * pthread_attr_destroy.c: Separated routine from attr.c.
- * pthread_attr_getdetachstate.c: Likewise.
- * pthread_attr_getscope.c: Likewise.
- * pthread_attr_getstackaddr.c: Likewise.
- * pthread_attr_getstacksize.c: Likewise.
- * pthread_attr_init.c: Likewise.
- * pthread_attr_is_attr.c: Likewise.
- * pthread_attr_setdetachstate.c: Likewise.
- * pthread_attr_setscope.c: Likewise.
- * pthread_attr_setstackaddr.c: Likewise.
- * pthread_attr_setstacksize.c: Likewise.
-
- * pthread.c: Agregation of agregate modules for super-inlineability.
-
-2002-02-02 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * cancel.c: Rearranged some code and introduced checks
- to disable cancelation at the start of a thread's cancelation
- run to prevent double cancelation. The main problem
- arises if a thread is canceling and then receives a subsequent
- async cancel request.
- * private.c: Likewise.
- * condvar.c: Place pragmas around cleanup_push/pop to turn
- off inline optimisation (/Obn where n>0 - MSVC only). Various
- optimisation switches in MSVC turn this on, which interferes with
- the way that cleanup handlers are run in C++ EH and SEH
- code. Application code compiled with inline optimisation must
- also wrap cleanup_push/pop blocks with the pragmas, e.g.
- #pragma inline_depth(0)
- pthread_cleanup_push(...)
- ...
- pthread_cleanup_pop(...)
- #pragma inline_depth(8)
- * rwlock.c: Likewise.
- * mutex.c: Remove attempts to inline some functions.
- * signal.c: Modify misleading comment.
-
-2002-02-01 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * semaphore.c (sem_trywait): Fix missing errno return
- for systems that define NEED_SEM (e.g. early WinCE).
- * mutex.c (pthread_mutex_timedlock): Return ENOTSUP
- for systems that define NEED_SEM since they don't
- have sem_trywait().
-
-2002-01-27 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * mutex.c (pthread_mutex_timedlock): New function suggested by
- Alexander Terekhov. The logic required to implement this
- properly came from Alexander, with some collaboration
- with Thomas Pfaff.
- (pthread_mutex_unlock): Wrap the waiters check and sema
- post in a critical section to prevent a race with
- pthread_mutex_timedlock.
- (ptw32_timed_semwait): New function;
- returns a special result if the absolute timeout parameter
- represents a time already passed when called; used by
- pthread_mutex_timedwait(). Have deliberately not reused
- the name "ptw32_sem_timedwait" because they are not the same
- routine.
- * condvar.c (ptw32_cond_timedwait): Use the new sem_timedwait()
- instead of ptw32_sem_timedwait(), which now has a different
- function. See previous.
- * implement.h: Remove prototype for ptw32_sem_timedwait.
- See next.
- (pthread_mutex_t_): Add critical section element for access
- to lock_idx during mutex post-timeout processing.
- * semaphore.h (sem_timedwait): See next.
- * semaphore.c (sem_timedwait): See next.
- * private.c (ptw32_sem_timedwait): Move to semaphore.c
- and rename as sem_timedwait().
-
-2002-01-18 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * sync.c (pthread_join): Was getting the exit code from the
- calling thread rather than the joined thread if
- defined(__MINGW32__) && !defined(__MSVCRT__).
-
-2002-01-15 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * pthread.h: Unless the build explicitly defines __CLEANUP_SEH,
- __CLEANUP_CXX, or __CLEANUP_C, then the build defaults to
- __CLEANUP_C style cleanup. This style uses setjmp/longjmp
- in the cancelation and thread exit implementations and therefore
- won't do stack unwinding if linked to applications that have it
- (e.g. C++ apps). This is currently consistent with most/all
- commercial Unix POSIX threads implementations.
-
- * spin.c (pthread_spin_init): Edit renamed function call.
- * nonportable.c (pthread_num_processors_np): New.
- (pthread_getprocessors_np): Renamed to ptw32_getprocessors
- and moved to private.c.
- * private.c (pthread_getprocessors): Moved here from
- nonportable.c.
- * pthread.def (pthread_getprocessors_np): Removed
- from export list.
-
- * rwlock.c (pthread_rwlockattr_init): New.
- (pthread_rwlockattr_destroy): New.
- (pthread_rwlockattr_getpshared): New.
- (pthread_rwlockattr_setpshared): New.
-
-2002-01-14 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * attr.c (pthread_attr_setscope): Fix struct pointer
- indirection error introduced 2002-01-04.
- (pthread_attr_getscope): Likewise.
-
-2002-01-12 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * pthread.dsp (SOURCE): Add missing source files.
-
-2002-01-08 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * mutex.c (pthread_mutex_trylock): use
- ptw32_interlocked_compare_exchange function pointer
- rather than ptw32_InterlockedCompareExchange() directly
- to retain portability to non-iX86 processors,
- e.g. WinCE etc. The pointer will point to the native
- OS version of InterlockedCompareExchange() if the
- OS supports it (see ChangeLog entry of 2001-10-17).
-
-2002-01-07 Thomas Pfaff <tpfaff at gmx.net>, Alexander Terekhov <TEREKHOV at de.ibm.com>
-
- * mutex.c (pthread_mutex_init): Remove critical
- section calls.
- (pthread_mutex_destroy): Likewise.
- (pthread_mutex_unlock): Likewise.
- (pthread_mutex_trylock): Likewise; uses
- ptw32_InterlockedCompareExchange() to avoid need for
- critical section; library is no longer i386 compatible;
- recursive mutexes now increment the lock count rather
- than return EBUSY; errorcheck mutexes return EDEADLCK
- rather than EBUSY. This behaviour is consistent with the
- Solaris pthreads implementation.
- * implement.h (pthread_mutex_t_): Remove critical
- section element - no longer needed.
-
-
-2002-01-04 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * attr.c (pthread_attr_setscope): Add more error
- checking and actually store the scope value even
- though it's not really necessary.
- (pthread_attr_getscope): Return stored value.
- * implement.h (pthread_attr_t_): Add new scope element.
- * ANNOUNCE: Fix out of date comment next to
- pthread_attr_setscope in conformance section.
-
-2001-12-21 Alexander Terekhov <TEREKHOV at de.ibm.com>
-
- * mutex.c (pthread_mutex_lock): Decrementing lock_idx was
- not thread-safe.
- (pthread_mutex_trylock): Likewise.
-
-2001-10-26 prionx@juno.com
-
- * semaphore.c (sem_init): Fix typo and missing bracket
- in conditionally compiled code. Only older versions of
- WinCE require this code, hence it doesn't normally get
- tested; somehow when sem_t reverted to an opaque struct
- the calloc NULL check was left in the conditionally included
- section.
- (sem_destroy): Likewise, the calloced sem_t wasn't being freed.
-
-2001-10-25 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * GNUmakefile (libwsock32): Add to linker flags for
- WSAGetLastError() and WSASetLastError().
- * Makefile (wsock32.lib): Likewise.
- * create.c: Minor mostly inert changes.
- * implement.h (PTW32_MAX): Move into here and renamed
- from sched.h.
- (PTW32_MIN): Likewise.
- * GNUmakefile (TEST_ICE): Define if testing internal
- implementation of InterlockedCompareExchange.
- * Makefile (TEST_ICE): Likewise.
- * private.c (TEST_ICE): Likewise.
-
-2001-10-24 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * attr.c (pthread_attr_setstacksize): Quell warning
- from LCC by conditionally compiling the stacksize
- validity check. LCC correctly warns that the condition
- (stacksize < PTHREAD_STACK_MIN) is suspicious
- because STACK_MIN is 0 and stacksize is of type
- size_t (or unsigned int).
-
-2001-10-17 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * barrier.c: Move _LONG and _LPLONG defines into
- implement.h; rename to PTW32_INTERLOCKED_LONG and
- PTW32_INTERLOCKED_LPLONG respectively.
- * spin.c: Likewise; ptw32_interlocked_compare_exchange used
- in place of InterlockedCompareExchange directly.
- * global.c (ptw32_interlocked_compare_exchange): Add
- prototype for this new routine pointer to be used when
- InterlockedCompareExchange isn't supported by Windows.
- * nonportable.c (pthread_win32_process_attach_np): Check for
- support of InterlockedCompareExchange in kernel32 and assign its
- address to ptw32_interlocked_compare_exchange if it exists, or
- our own ix86 specific implementation ptw32_InterlockedCompareExchange.
- *private.c (ptw32_InterlockedCompareExchange): An
- implementation of InterlockedCompareExchange() which is
- specific to ix86; written directly in assembler for either
- MSVC or GNU C; needed because Windows 95 doesn't support
- InterlockedCompareExchange().
-
- * sched.c (sched_get_priority_min): Extend to return
- THREAD_PRIORITY_IDLE.
- (sched_get_priority_max): Extend to return
- THREAD_PRIORITY_CRITICAL.
-
-2001-10-15 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * spin.c (pthread_spin_lock): PTHREAD_SPINLOCK_INITIALIZER
- was causing a program fault.
- (pthread_spin_init): Could have alloced memory
- without freeing under some error conditions.
-
- * mutex.c (pthread_mutex_init): Move memory
- allocation of mutex struct after checking for
- PROCESS_SHARED.
-
-2001-10-12 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * spin.c (pthread_spin_unlock): Was not returning
- EPERM if the spinlock was not locked, for multi CPU
- machines.
-
-2001-10-08 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * spin.c (pthread_spin_trylock): Was not returning
- EBUSY for multi CPU machines.
-
-2001-08-24 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * condvar.c (pthread_cond_destroy): Remove cv element
- that is no longer used.
- * implement.h: Likewise.
-
-2001-08-23 Alexander Terekhov <TEREKHOV at de.ibm.com>
-
- * condvar.c (pthread_cond_destroy): fix bug with
- respect to deadlock in the case of concurrent
- _destroy/_unblock; a condition variable can be destroyed
- immediately after all the threads that are blocked on
- it are awakened.
-
-2001-08-23 Phil Frisbie, Jr. <phil at hawksoft.com>
-
- * tsd.c (pthread_getspecific): Preserve the last
- winsock error [from WSAGetLastError()].
-
-2001-07-18 Scott McCaskill <scott at magruder.org>
-
- * mutex.c (pthread_mutexattr_init): Return ENOMEM
- immediately and don't dereference the NULL pointer
- if calloc fails.
- (pthread_mutexattr_getpshared): Don't dereference
- a pointer that is possibly NULL.
- * barrier.c (pthread_barrierattr_init): Likewise
- (pthread_barrierattr_getpshared): Don't dereference
- a pointer that is possibly NULL.
- * condvar.c (pthread_condattr_getpshared): Don't dereference
- a pointer that is possibly NULL.
-
-2001-07-15 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * rwlock.c (pthread_rwlock_wrlock): Is allowed to be
- a cancelation point; re-enable deferred cancelability
- around the CV call.
-
-2001-07-10 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * barrier.c: Still more revamping. The exclusive access
- mutex isn't really needed so it has been removed and replaced
- by an InterlockedDecrement(). nSerial has been removed.
- iStep is now dual-purpose. The process shared attribute
- is now stored in the barrier struct.
- * implement.h (pthread_barrier_t_): Lost some/gained one
- elements.
- * private.c (ptw32_threadStart): Removed some comments.
-
-2001-07-10 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * barrier.c: Revamped to fix the race condition. Two alternating
- semaphores are used instead of the PulseEvent. Also improved
- overall throughput by returning PTHREAD_BARRIER_SERIAL_THREAD
- to the first waking thread.
- * implement.h (pthread_barrier_t_): Revamped.
-
-2001-07-09 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * barrier.c: Fix several bugs in all routines. Now passes
- tests/barrier5.c which is fairly rigorous. There is still
- a non-optimal work-around for a race condition between
- the barrier breeched event signal and event wait. Basically
- the last (signalling) thread to hit the barrier yields
- to allow any other threads, which may have lost the race,
- to complete.
-
-2001-07-07 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * barrier.c: Changed synchronisation mechanism to a
- Win32 manual reset Event and use PulseEvent to signal
- waiting threads. If the implementation continued to use
- a semaphore it would require a second semaphore and
- some management to use them alternately as barriers. A
- single semaphore allows threads to cascade from one barrier
- through the next, leaving some threads blocked at the first.
- * implement.h (pthread_barrier_t_): As per above.
- * general: Made a number of other routines inlinable.
-
-2001-07-07 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * spin.c: Revamped and working; included static initialiser.
- Now beta level.
- * barrier.c: Likewise.
- * condvar.c: Macro constant change; inline auto init routine.
- * mutex.c: Likewise.
- * rwlock.c: Likewise.
- * private.c: Add support for spinlock initialiser.
- * global.c: Likewise.
- * implement.h: Likewise.
- * pthread.h (PTHREAD_SPINLOCK_INITIALIZER): Fix typo.
-
-2001-07-05 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * barrier.c: Remove static initialisation - irrelevent
- for this object.
- * pthread.h (PTHREAD_BARRIER_INITIALIZER): Removed.
- * rwlock.c (pthread_rwlock_wrlock): This routine is
- not a cancelation point - disable deferred
- cancelation around call to pthread_cond_wait().
-
-2001-07-05 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * spin.c: New module implementing spin locks.
- * barrier.c: New module implementing barriers.
- * pthread.h (_POSIX_SPIN_LOCKS): defined.
- (_POSIX_BARRIERS): Defined.
- (pthread_spin_*): Defined.
- (pthread_barrier*): Defined.
- (PTHREAD_BARRIER_SERIAL_THREAD): Defined.
- * implement.h (pthread_spinlock_t_): Defined.
- (pthread_barrier_t_): Defined.
- (pthread_barrierattr_t_): Defined.
-
- * mutex.c (pthread_mutex_lock): Return with the error
- if an auto-initialiser initialisation fails.
-
- * nonportable.c (pthread_getprocessors_np): New; gets the
- number of available processors for the current process.
-
-2001-07-03 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * pthread.h (_POSIX_READER_WRITER_LOCKS): Define it
- if not already defined.
-
-2001-07-01 Alexander Terekhov <TEREKHOV at de.ibm.com>
-
- * condvar.c: Fixed lost signal bug reported by Timur Aydin
- (taydin@snet.net).
- [RPJ (me) didn't translate the original algorithm
- correctly.]
- * semaphore.c: Added sem_post_multiple; this is a useful
- routine, but it doesn't appear to be standard. For now it's
- not an exported function.
-
-2001-06-25 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * create.c (pthread_create): Add priority inheritance
- attributes.
- * mutex.c (pthread_mutex_lock): Remove some overhead for
- PTHREAD_MUTEX_NORMAL mutex types. Specifically, avoid
- calling pthread_self() and pthread_equal() to check/set
- the mutex owner. Introduce a new pseudo owner for this
- type. Test results suggest increases in speed of up to
- 90% for non-blocking locks.
- This is the default type of mutex used internally by other
- synchronising objects, ie. condition variables and
- read-write locks. The test rwlock7.c shows about a
- 30-35% speed increase over snapshot 2001-06-06. The
- price of this is that the application developer
- must ensure correct behaviour, or explicitly set the
- mutex to a safer type such as PTHREAD_MUTEX_ERRORCHECK.
- For example, PTHREAD_MUTEX_NORMAL (or PTHREAD_MUTEX_DEFAULT)
- type mutexes will not return an error if a thread which is not
- the owner calls pthread_mutex_unlock. The call will succeed
- in unlocking the mutex if it is currently locked, but a
- subsequent unlock by the true owner will then fail with EPERM.
- This is however consistent with some other implementations.
- (pthread_mutex_unlock): Likewise.
- (pthread_mutex_trylock): Likewise.
- (pthread_mutex_destroy): Likewise.
- * attr.c (pthread_attr_init): PTHREAD_EXPLICIT_SCHED is the
- default inheritance attribute; THREAD_PRIORITY_NORMAL is
- the default priority for new threads.
- * sched.c (pthread_attr_setschedpolicy): Added routine.
- (pthread_attr_getschedpolicy): Added routine.
- (pthread_attr_setinheritsched): Added routine.
- (pthread_attr_getinheritsched): Added routine.
- * pthread.h (sched_rr_set_interval): Added as a macro;
- returns -1 with errno set to ENOSYS.
-
-2001-06-23 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- *sched.c (pthread_attr_setschedparam): Add priority range
- check.
- (sched_setscheduler): New function; checks for a valid
- pid and policy; checks for permission to set information
- in the target process; expects pid to be a Win32 process ID,
- not a process handle; the only scheduler policy allowed is
- SCHED_OTHER.
- (sched_getscheduler): Likewise, but checks for permission
- to query.
- * pthread.h (SCHED_*): Moved to sched.h as defined in the
- POSIX standard.
- * sched.h (SCHED_*): Moved from pthread.h.
- (pid_t): Defined if necessary.
- (sched_setscheduler): Defined.
- (sched_getscheduler): Defined.
- * pthread.def (sched_setscheduler): Exported.
- (sched_getscheduler): Likewise.
-
-2001-06-23 Ralf Brese <Ralf.Brese at pdb4.siemens.de>
-
- * create.c (pthread_create): Set thread priority from
- thread attributes.
-
-2001-06-18 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * Made organisational-only changes to UWIN additions.
- * dll.c (dllMain): Moved UWIN process attach code
- to pthread_win32_process_attach_np(); moved
- instance of pthread_count to global.c.
- * global.c (pthread_count): Moved from dll.c.
- * nonportable.c (pthread_win32_process_attach_np):
- Moved _UWIN code to here from dll.c.
- * implement.h (pthread_count): Define extern int.
- * create.c (pthread_count): Remove extern int.
- * private.c (pthread_count): Likewise.
- * exit.c (pthread_count): Likewise.
-
-2001-06-18 David Korn <dgk at research.att.com>
-
- * dll.c: Added changes necessary to work with UWIN.
- * create.c: Likewise.
- * pthread.h: Likewise.
- * misc.c: Likewise.
- * exit.c: Likewise.
- * private.c: Likewise.
- * implement.h: Likewise.
- There is some room at the start of struct pthread_t_
- to implement the signal semantics in UWIN's posix.dll
- although this is not yet complete.
- * Nmakefile: Compatible with UWIN's Nmake utility.
- * Nmakefile.tests: Likewise - for running the tests.
-
-2001-06-08 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * semaphore.h (sem_t): Fixed for compile and test.
- * implement.h (sem_t_): Likewise.
- * semaphore.c: Likewise.
- * private.c (ptw32_sem_timedwait): Updated to use new
- opaque sem_t.
-
-2001-06-06 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * semaphore.h (sem_t): Is now an opaque pointer;
- moved actual definition to implement.h.
- * implement.h (sem_t_): Move here from semaphore.h;
- was the definition of sem_t.
- * semaphore.c: Wherever necessary, changed use of sem
- from that of a pointer to a pointer-pointer; added
- extra checks for a valid sem_t; NULL sem_t when
- it is destroyed; added extra checks when creating
- and destroying sem_t elements in the NEED_SEM
- code branches; changed from using a pthread_mutex_t
- ((*sem)->mutex) to CRITICAL_SECTION ((*sem)->sem_lock_cs)
- in NEED_SEM branches for access serialisation.
-
-2001-06-06 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * mutex.c (pthread_mutexattr_init): Remove
- ptw32_mutex_default_kind.
-
-2001-06-05 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * nonportable.c (pthread_mutex_setdefaultkind_np):
- Remove - should not have been included in the first place.
- (pthread_mutex_getdefaultkind_np): Likewise.
- * global.c (ptw32_mutex_default_kind): Likewise.
- * mutex.c (pthread_mutex_init): Remove use of
- ptw32_mutex_default_kind.
- * pthread.h (pthread_mutex_setdefaultkind_np): Likewise.
- (pthread_mutex_getdefaultkind_np): Likewise.
- * pthread.def (pthread_mutexattr_setkind_np): Added.
- (pthread_mutexattr_getkind_np): Likewise.
-
- * README: Many changes that should have gone in before
- the last snapshot.
- * README.NONPORTABLE: New - referred to by ANNOUNCE
- but never created; documents the non-portable routines
- included in the library - moved from README with new
- routines added.
- * ANNOUNCE (pthread_mutexattr_setkind_np): Added to
- compliance list.
- (pthread_mutexattr_getkind_np): Likewise.
-
-2001-06-04 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * condvar.c: Add original description of the algorithm as
- developed by Terekhov and Thomas, plus reference to
- README.CV.
-
-2001-06-03 Alexander Terekhov <TEREKHOV at de.ibm.com>, Louis Thomas <lthomas at arbitrade.com>
-
- * condvar.c (pthread_cond_init): Completely revamped.
- (pthread_cond_destroy): Likewise.
- (ptw32_cond_wait_cleanup): Likewise.
- (ptw32_cond_timedwait): Likewise.
- (ptw32_cond_unblock): New general signaling routine.
- (pthread_cond_signal): Now calls ptw32_cond_unblock.
- (pthread_cond_broadcast): Likewise.
- * implement.h (pthread_cond_t_): Revamped.
- * README.CV: New; explanation of the above changes.
-
-2001-05-30 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * pthread.h (rand_r): Fake using _seed argument to quell
- compiler warning (compiler should optimise this away later).
-
- * GNUmakefile (OPT): Leave symbolic information out of the library
- and increase optimisation level - for smaller faster prebuilt
- dlls.
-
-2001-05-29 Milan Gardian <Milan.Gardian at LEIBINGER.com>
-
- * Makefile: fix typo.
- * pthreads.h: Fix problems with stdcall/cdecl conventions, in particular
- remove the need for PT_STDCALL everywhere; remove warning supression.
- * (errno): Fix the longstanding "inconsistent dll linkage" problem
- with errno; now also works with /MD debugging libs -
- warnings emerged when compiling pthreads library with /MD (or /MDd)
- compiler switch, instead of /MT (or /MTd) (i.e. when compiling pthreads
- using Multithreaded DLL CRT instead of Multithreaded statically linked
- CRT).
- * create.c (pthread_create): Likewise; fix typo.
- * private.c (ptw32_threadStart): Eliminate use of terminate() which doesn't
- throw exceptions.
- * Remove unnecessary #includes from a number of modules -
- [I had to #include malloc.h in implement.h for gcc - rpj].
-
-2001-05-29 Thomas Pfaff <tpfaff at gmx.net>
-
- * pthread.h (PTHREAD_MUTEX_DEFAULT): New; equivalent to
- PTHREAD_MUTEX_DEFAULT_NP.
- * (PTHREAD_MUTEX_NORMAL): Similarly.
- * (PTHREAD_MUTEX_ERRORCHECK): Similarly.
- * (PTHREAD_MUTEX_RECURSIVE): Similarly.
- * (pthread_mutex_setdefaultkind_np): New; Linux compatibility stub
- for pthread_mutexattr_settype.
- * (pthread_mutexattr_getkind_np): New; Linux compatibility stub
- for pthread_mutexattr_gettype.
- * mutex.c (pthread_mutexattr_settype): New; allow
- the following types of mutex:
- PTHREAD_MUTEX_DEFAULT_NP
- PTHREAD_MUTEX_NORMAL_NP
- PTHREAD_MUTEX_ERRORCHECK_NP
- PTHREAD_MUTEX_RECURSIVE_NP
- * Note that PTHREAD_MUTEX_DEFAULT is equivalent to
- PTHREAD_MUTEX_NORMAL - ie. mutexes should no longer
- be recursive by default, and a thread will deadlock if it
- tries to relock a mutex it already owns. This is inline with
- other pthreads implementations.
- * (pthread_mutex_lock): Process the lock request
- according to the mutex type.
- * (pthread_mutex_init): Eliminate use of Win32 mutexes as the
- basis of POSIX mutexes - instead, a combination of one critical section
- and one semaphore are used in conjunction with Win32 Interlocked* routines.
- * (pthread_mutex_destroy): Likewise.
- * (pthread_mutex_lock): Likewise.
- * (pthread_mutex_trylock): Likewise.
- * (pthread_mutex_unlock): Likewise.
- * Use longjmp/setjmp to implement cancelation when building the library
- using a C compiler which doesn't support exceptions, e.g. gcc -x c (note
- that gcc -x c++ uses exceptions).
- * Also fixed some of the same typos and eliminated PT_STDCALL as
- Milan Gardian's patches above.
-
-2001-02-07 Alexander Terekhov <TEREKHOV at de.ibm.com>
-
- * rwlock.c: Revamped.
- * implement.h (pthread_rwlock_t_): Redefined.
- This implementation does not have reader/writer starvation problem.
- Rwlock attempts to behave more like a normal mutex with
- races and scheduling policy determining who is more important;
- It also supports recursive locking,
- has less synchronization overhead (no broadcasts at all,
- readers are not blocked on any condition variable) and seem to
- be faster than the current implementation [W98 appears to be
- approximately 15 percent faster at least - on top of speed increase
- from Thomas Pfaff's changes to mutex.c - rpj].
-
-2000-12-29 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * Makefile: Back-out "for" loops which don't work.
-
- * GNUmakefile: Remove the fake.a target; add the "realclean"
- target; don't remove built libs under the "clean" target.
-
- * config.h: Add a guard against multiple inclusion.
-
- * semaphore.h: Add some defines from config.h to make
- semaphore.h independent of config.h when building apps.
-
- * pthread.h (_errno): Back-out previous fix until we know how to
- fix it properly.
-
- * implement.h (lockCount): Add missing element to pthread_mutex_t_.
-
- * sync.c (pthread_join): Spelling fix in comment.
-
- * private.c (ptw32_threadStart): Reset original termination
- function (C++).
- (ptw32_threadStart): Cleanup detached threads early in case
- the library is statically linked.
- (ptw32_callUserDestroyRoutines): Remove [SEH] __try block from
- destructor call so that unhandled exceptions will be passed through
- to the system; call terminate() from [C++] try block for the same
- reason.
-
- * tsd.c (pthread_getspecific): Add comment.
-
- * mutex.c (pthread_mutex_init): Initialise new elements in
- pthread_mutex_t.
- (pthread_mutex_unlock): Invert "pthread_equal()" test.
-
-2000-12-28 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * semaphore.c (mode_t): Use ifndef HAVE_MODE_T to include definition.
-
- * config.h.in (HAVE_MODE_T): Added.
- (_UWIN): Start adding defines for the UWIN package.
-
- * private.c (ptw32_threadStart): Unhandled exceptions are
- now passed through to the system to deal with. This is consistent
- with normal Windows behaviour. C++ applications may use
- set_terminate() to override the default behaviour which is
- to call ptw32_terminate(). Ptw32_terminate() cleans up some
- POSIX thread stuff before calling the system default function
- which calls abort(). The users termination function should conform
- to standard C++ semantics which is to not return. It should
- exit the thread (call pthread_exit()) or exit the application.
- * private.c (ptw32_terminate): Added as the default set_terminate()
- function. It calls the system default function after cleaning up
- some POSIX thread stuff.
-
- * implement.h (ptw32_try_enter_critical_section): Move
- declaration.
- * global.c (ptw32_try_enter_critical_section): Moved
- from dll.c.
- * dll.c: Move process and thread attach/detach code into
- functions in nonportable.c.
- * nonportable.c (pthread_win32_process_attach_np): Process
- attach code from dll.c is now available to static linked
- applications.
- * nonportable.c (pthread_win32_process_detach_np): Likewise.
- * nonportable.c (pthread_win32_thread_attach_np): Likewise.
- * nonportable.c (pthread_win32_thread_detach_np): Likewise.
-
- * pthread.h: Add new non-portable prototypes for static
- linked applications.
-
- * GNUmakefile (OPT): Increase optimisation flag and remove
- debug info flag.
-
- * pthread.def: Add new non-portable exports for static
- linked applications.
-
-2000-12-11 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * FAQ: Update Answer 6 re getting a fully working
- Mingw32 built library.
-
-2000-10-10 Steven Reddie <smr at essemer.com.au>
-
- * misc.c (pthread_self): Restore Win32 "last error"
- cleared by TlsGetValue() call in
- pthread_getspecific()
-
-2000-09-20 Arthur Kantor <akantor at bexusa.com>
-
- * mutex.c (pthread_mutex_lock): Record the owner
- of the mutex. This requires also keeping count of
- recursive locks ourselves rather than leaving it
- to Win32 since we need to know when to NULL the
- thread owner when the mutex is unlocked.
- (pthread_mutex_trylock): Likewise.
- (pthread_mutex_unlock): Check that the calling
- thread owns the mutex, decrement the recursive
- lock count, and NULL the owner if zero. Return
- EPERM if the mutex is owned by another thread.
- * implement.h (pthread_mutex_t_): Add ownerThread
- and lockCount members.
-
-2000-09-13 Jef Gearhart <jgearhart at tpssys.com>
-
- * mutex.c (pthread_mutex_init): Call
- TryEnterCriticalSection through the pointer
- rather than directly so that the dll can load
- on Windows versions that can't resolve the
- function, eg. Windows 95
-
-2000-09-09 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * pthread.h (ctime_r): Fix arg.
-
-2000-09-08 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * GNUmakefile(_WIN32_WINNT=0x400): Define in CFLAGS;
- doesn't seem to be needed though.
-
- * cancel.c (pthread_cancel): Must get "self" through
- calling pthread_self() which will ensure a POSIX thread
- struct is built for non-POSIX threads; return an error
- if this fails
- - Ollie Leahy <ollie at mpt.ie>
- (pthread_setcancelstate): Likewise.
- (pthread_setcanceltype): Likewise.
- * misc.c (ptw32_cancelable_wait): Likewise.
-
- * private.c (ptw32_tkAssocCreate): Remove unused #if 0
- wrapped code.
-
- * pthread.h (ptw32_get_exception_services_code):
- Needed to be forward declared unconditionally.
-
-2000-09-06 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * cancel.c (pthread_cancel): If called from the main
- thread "self" would be NULL; get "self" via pthread_self()
- instead of directly from TLS so that an implicit
- pthread object is created.
-
- * misc.c (pthread_equal): Strengthen test for NULLs.
-
-2000-09-02 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * condvar.c (ptw32_cond_wait_cleanup): Ensure that all
- waking threads check if they are the last, and notify
- the broadcaster if so - even if an error occurs in the
- waiter.
-
- * semaphore.c (_decrease_semaphore): Should be
- a call to ptw32_decrease_semaphore.
- (_increase_semaphore): Should be a call to
- ptw32_increase_semaphore.
-
- * misc.c (ptw32_cancelable_wait): Renamed from
- CancelableWait.
- * rwlock.c (_rwlock_check*): Renamed to
- ptw32_rwlock_check*.
- * mutex.c (_mutex_check*): Renamed to ptw32_mutex_check*.
- * condvar.c (cond_timed*): Renamed to ptw32_cond_timed*.
- (_cond_check*): Renamed to ptw32_cond_check*.
- (cond_wait_cleanup*): Rename to ptw32_cond_wait_cleanup*.
- (ptw32_cond_timedwait): Add comments.
-
-2000-08-22 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * private.c (ptw32_throw): Fix exception test;
- move exceptionInformation declaration.
-
- * tsd.c (pthread_key_create): newkey wrongly declared.
-
- * pthread.h: Fix comment block.
-
-2000-08-18 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * mutex.c (pthread_mutex_destroy): Check that the mutex isn't
- held; invalidate the mutex as early as possible to avoid
- contention; not perfect - FIXME!
-
- * rwlock.c (pthread_rwlock_init): Remove redundant assignment
- to "rw".
- (pthread_rwlock_destroy): Invalidate the rwlock before
- freeing up any of it's resources - to avoid contention.
-
- * private.c (ptw32_tkAssocCreate): Change assoc->lock
- to use a dynamically initialised mutex - only consumes
- a W32 mutex or critical section when first used,
- not before.
-
- * mutex.c (pthread_mutex_init): Remove redundant assignment
- to "mx".
- (pthread_mutexattr_destroy): Set attribute to NULL
- before freeing it's memory - to avoid contention.
-
- * implement.h (PTW32_EPS_CANCEL/PTW32_EPS_EXIT):
- Must be defined for all compilers - used as generic
- exception selectors by ptw32_throw().
-
- * Several: Fix typos from scripted edit session
- yesterday.
-
- * nonportable.c (pthread_mutexattr_setforcecs_np):
- Moved this function from mutex.c.
- (pthread_getw32threadhandle_np): New function to
- return the win32 thread handle that the POSIX
- thread is using.
- * mutex.c (pthread_mutexattr_setforcecs_np):
- Moved to new file "nonportable.c".
-
- * pthread.h (PTW32_BUILD): Only redefine __except
- and catch compiler keywords if we aren't building
- the library (ie. PTW32_BUILD is not defined) -
- this is safer than defining and then undefining
- if not building the library.
- * implement.h: Remove __except and catch undefines.
- * Makefile (CFLAGS): Define PTW32_BUILD.
- * GNUmakefile (CFLAGS): Define PTW32_BUILD.
-
- * All appropriate: Change Pthread_exception* to
- ptw32_exception* to be consistent with internal
- identifier naming.
-
- * private.c (ptw32_throw): New function to provide
- a generic exception throw for all internal
- exceptions and EH schemes.
- (ptw32_threadStart): pthread_exit() value is now
- returned via the thread structure exitStatus
- element.
- * exit.c (pthread_exit): pthread_exit() value is now
- returned via the thread structure exitStatus
- element.
- * cancel.c (ptw32_cancel_self): Now uses ptw32_throw.
- (pthread_setcancelstate): Ditto.
- (pthread_setcanceltype): Ditto.
- (pthread_testcancel): Ditto.
- (pthread_cancel): Ditto.
- * misc.c (CancelableWait): Ditto.
- * exit.c (pthread_exit): Ditto.
- * All applicable: Change PTW32_ prefix to
- PTW32_ prefix to remove leading underscores
- from private library identifiers.
-
-2000-08-17 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * All applicable: Change _pthread_ prefix to
- ptw32_ prefix to remove leading underscores
- from private library identifiers (single
- and double leading underscores are reserved in the
- ANSI C standard for compiler implementations).
-
- * tsd.c (pthread_create_key): Initialise temporary
- key before returning it's address to avoid race
- conditions.
-
-2000-08-13 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * errno.c: Add _MD precompile condition; thus far
- had no effect when using /MD compile option but I
- thnk it should be there.
-
- * exit.c: Add __cplusplus to various #if lines;
- was compiling SEH code even when VC++ had
- C++ compile options.
-
- * private.c: ditto.
-
- * create.c (pthread_create): Add PT_STDCALL macro to
- function pointer arg in _beginthread().
-
- * pthread.h: PT_STDCALL really does need to be defined
- in both this and impliment.h; don't set it to __cdecl
- - this macro is only used to extend function pointer
- casting for functions that will be passed as parameters.
- (~PThreadCleanup): add cast and group expression.
- (_errno): Add _MD compile conditional.
- (PtW32NoCatchWarn): Change pragma message.
-
- * implement.h: Move and change PT_STDCALL define.
-
- * need_errno.h: Add _MD to compilation conditional.
-
- * GNUmakefile: Substantial rewrite for new naming
- convention; set for nil optimisation (turn it up
- when we have a working library build; add target
- "fake.a" to build a libpthreadw32.a from the VC++
- built DLL pthreadVCE.dll.
-
- * pthread.def (LIBRARY): Don't specify in the .def
- file - it is specified on the linker command line
- since we now use the same .def file for variously
- named .dlls.
-
- * Makefile: Substantial rewrite for new naming
- convention; default nmake target only issues a
- help message; run nmake with specific target
- corresponding to the EH scheme being used.
-
- * README: Update information; add naming convention
- explanation.
-
- * ANNOUNCE: Update information.
-
-2000-08-12 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * pthread.h: Add compile-time message when using
- MSC_VER compiler and C++ EH to warn application
- programmers to use PtW32Catch instead of catch(...)
- if they want cancelation and pthread_exit to work.
-
- * implement.h: Remove #include <semaphore.h>; we
- use our own local semaphore.h.
-
-2000-08-10 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * cleanup.c (pthread_pop_cleanup): Remove _pthread
- prefix from __except and catch keywords; implement.h
- now simply undefines ptw32__except and
- ptw32_catch if defined; VC++ was not textually
- substituting ptw32_catch etc back to catch as
- it was redefined; the reason for using the prefixed
- version was to make it clear that it was not using
- the pthread.h redefined catch keyword.
-
- * private.c (ptw32_threadStart): Ditto.
- (ptw32_callUserDestroyRoutines): Ditto.
-
- * implement.h (ptw32__except): Remove #define.
- (ptw32_catch): Remove #define.
-
- * GNUmakefile (pthread.a): New target to build
- libpthread32.a from pthread.dll using dlltool.
-
- * buildlib.bat: Duplicate cl commands with args to
- build C++ EH version of pthread.dll; use of .bat
- files is redundant now that nmake compatible
- Makefile is included; used as a kludge only now.
-
- * Makefile: Localise some macros and fix up the clean:
- target to extend it and work properly.
-
- * CONTRIBUTORS: Add contributors.
-
- * ANNOUNCE: Updated.
-
- * README: Updated.
-
-2000-08-06 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * pthread.h: Remove #warning - VC++ doesn't accept it.
-
-2000-08-05 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * pthread.h (PtW32CatchAll): Add macro. When compiling
- applications using VC++ with C++ EH rather than SEH
- 'PtW32CatchAll' must be used in place of any 'catch( ... )'
- if the application wants pthread cancelation or
- pthread_exit() to work.
-
-2000-08-03 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * pthread.h: Add a base class ptw32_exception for
- library internal exceptions and change the "catch"
- re-define macro to use it.
-
-2000-08-02 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * GNUmakefile (CFLAGS): Add -mthreads.
- Add new targets to generate cpp and asm output.
-
- * sync.c (pthread_join): Remove dead code.
-
-2000-07-25 Tristan Savatier <tristan at mpegtv.com>
-
- * sched.c (sched_get_priority_max): Handle different WinCE and
- Win32 priority values together.
- (sched_get_priority_min): Ditto.
-
-2000-07-25 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * create.c (pthread_create): Force new threads to wait until
- pthread_create has the new thread's handle; we also retain
- a local copy of the handle for internal use until
- pthread_create returns.
-
- * private.c (ptw32_threadStart): Initialise ei[].
- (ptw32_threadStart): When beginthread is used to start the
- thread, force waiting until the creator thread had the
- thread handle.
-
- * cancel.c (ptw32_cancel_thread): Include context switch
- code for defined(_X86_) environments in addition to _M_IX86.
-
- * rwlock.c (pthread_rwlock_destroy): Assignment changed
- to avoid compiler warning.
-
- * private.c (ptw32_get_exception_services_code): Cast
- NULL return value to avoid compiler warning.
-
- * cleanup.c (pthread_pop_cleanup): Initialise "cleanup" variable
- to avoid compiler warnings.
-
- * misc.c (ptw32_new): Change "new" variable to "t" to avoid
- confusion with the C++ keyword of the same name.
-
- * condvar.c (cond_wait_cleanup): Initialise lastWaiter variable.
- (cond_timedwait): Remove unused local variables. to avoid
- compiler warnings.
-
- * dll.c (dllMain): Remove 2000-07-21 change - problem
- appears to be in pthread_create().
-
-2000-07-22 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * tsd.c (pthread_key_create): If a destructor was given
- and the pthread_mutex_init failed, then would try to
- reference a NULL pointer (*key); eliminate this section of
- code by using a dynamically initialised mutex
- (PTHREAD_MUTEX_INITIALIZER).
-
- * tsd.c (pthread_setspecific): Return an error if
- unable to set the value; simplify cryptic conditional.
-
- * tsd.c (pthread_key_delete): Locking threadsLock relied
- on mutex_lock returning an error if the key has no destructor.
- ThreadsLock is only initialised if the key has a destructor.
- Making this mutex a static could reduce the number of mutexes
- used by an application since it is actually created only at
- first use and it's often destroyed soon after.
-
-2000-07-22 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * FAQ: Added Q5 and Q6.
-
-2000-07-21 David Baggett <dmb at itasoftware.com>
-
- * dll.c: Include resource leakage work-around. This is a
- partial FIXME which doesn't stop all leakage. The real
- problem needs to be found and fixed.
-
-2000-07-21 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
-
- * create.c (pthread_create): Set threadH to 0 (zero)
- everywhere. Some assignments were using NULL. Maybe
- it should be NULL everywhere - need to check. (I know
- they are nearly always the same thing - but not by
- definition.)
-
- * misc.c (pthread_self): Try to catch NULL thread handles
- at the point where they might be generated, even though
- they should always be valid at this point.
-
- * tsd.c (pthread_setspecific): return an error value if
- pthread_self() returns NULL.
-
- * sync.c (pthread_join): return an error value if
- pthread_self() returns NULL.
-
- * signal.c (pthread_sigmask): return an error value if
- pthread_self() returns NULL.
-
-2000-03-02 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * attr.c (pthread_attr_init): Set default stacksize to zero (0)
- rather than PTHREAD_STACK_MIN even though these are now the same.
-
- * pthread.h (PTHREAD_STACK_MIN): Lowered to 0.
-
-2000-01-28 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * mutex.c (pthread_mutex_init): Free mutex if it has been alloced;
- if critical sections can be used instead of Win32 mutexes, test
- that the critical section works and return an error if not.
-
-2000-01-07 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * cleanup.c (pthread_pop_cleanup): Include SEH code only if MSC is not
- compiling as C++.
- (pthread_push_cleanup): Include SEH code only if MSC is not
- compiling as C++.
-
- * pthread.h: Include SEH code only if MSC is not
- compiling as C++.
-
- * implement.h: Include SEH code only if MSC is not
- compiling as C++.
-
- * cancel.c (ptw32_cancel_thread): Add _M_IX86 check.
- (pthread_testcancel): Include SEH code only if MSC is not
- compiling as C++.
- (ptw32_cancel_self): Include SEH code only if MSC is not
- compiling as C++.
-
-2000-01-06 Erik Hensema <erik.hensema at group2000.nl>
-
- * Makefile: Remove inconsistencies in 'cl' args
-
-2000-01-04 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * private.c (ptw32_get_exception_services_code): New; returns
- value of EXCEPTION_PTW32_SERVICES.
- (ptw32_processInitialize): Remove initialisation of
- ptw32_exception_services which is no longer needed.
-
- * pthread.h (ptw32_exception_services): Remove extern.
- (ptw32_get_exception_services_code): Add function prototype;
- use this to return EXCEPTION_PTW32_SERVICES value instead of
- using the ptw32_exception_services variable which I had
- trouble exporting through pthread.def.
-
- * global.c (ptw32_exception_services): Remove declaration.
-
-1999-11-22 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * implement.h: Forward declare ptw32_new();
-
- * misc.c (ptw32_new): New; alloc and initialise a new pthread_t.
- (pthread_self): New thread struct is generated by new routine
- ptw32_new().
-
- * create.c (pthread_create): New thread struct is generated
- by new routine ptw32_new().
-
-1999-11-21 Ross Johnson <rpj at special.ise.canberra.edu.au>
-
- * global.c (ptw32_exception_services): Declare new variable.
-
- * private.c (ptw32_threadStart): Destroy thread's
- cancelLock mutex; make 'catch' and '__except' usageimmune to
- redfinitions in pthread.h.
- (ptw32_processInitialize): Init new constant ptw32_exception_services.
-
- * create.c (pthread_create): Initialise thread's cancelLock
- mutex.
-
- * cleanup.c (pthread_pop_cleanup): Make 'catch' and '__except'
- usage immune to redfinition s in pthread.h.
-
- * private.c: Ditto.
-
- * pthread.h (catch): Redefine 'catch' so that C++ applications
- won't catch our internal exceptions.
- (__except): ditto for __except.
-
- * implement.h (ptw32_catch): Define internal version
- of 'catch' because 'catch' is redefined by pthread.h.
- (__except): ditto for __except.
- (struct pthread_t_): Add cancelLock mutex for async cancel
- safety.
-
-1999-11-21 Jason Nye <jnye at nbnet.nb.ca>, Erik Hensema <erik.hensema at group2000.nl>
-
- * cancel.c (ptw32_cancel_self): New; part of the async
- cancellation implementation.
- (ptw32_cancel_thread): Ditto; this function is X86
- processor specific.
- (pthread_setcancelstate): Add check for pending async
- cancel request and cancel the calling thread if
- required; add async-cancel safety lock.
- (pthread_setcanceltype): Ditto.
-
-1999-11-13 Erik Hensema <erik.hensema at group2000.nl>
-
- * configure.in (AC_OUTPUT): Put generated output into GNUmakefile
- rather than Makefile. Makefile will become the MSC nmake compatible
- version
-
-1999-11-13 John Bossom (John.Bossom@cognos.com>
-
- * misc.c (pthread_self): Add a note about GetCurrentThread
- returning a pseudo-handle
-
-1999-11-10 Todd Owen <towen at lucidcalm.dropbear.id.au>
-
- * dll.c (dllMain): Free kernel32 ASAP.
- If TryEnterCriticalSection is not being used, then free
- the kernel32.dll handle now, rather than leaving it until
- DLL_PROCESS_DETACH.
-
- Note: this is not a pedantic exercise in freeing unused
- resources! It is a work-around for a bug in Windows 95
- (see microsoft knowledge base article, Q187684) which
- does Bad Things when FreeLibrary is called within
- the DLL_PROCESS_DETACH code, in certain situations.
- Since w95 just happens to be a platform which does not
- provide TryEnterCriticalSection, the bug will be
- effortlessly avoided.
-
-1999-11-10 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * sync.c (pthread_join): Make it a deferred cancelation point.
-
- * misc.c (pthread_self): Explicitly initialise implicitly
- created thread state to default values.
-
-1999-11-05 Tristan Savatier <tristan at mpegtv.com>
-
- * pthread.h (winsock.h): Include unconditionally.
- (ETIMEDOUT): Change fallback value to that defined by winsock.h.
-
- * general: Patched for portability to WinCE. The details are
- described in the file WinCE-PORT. Follow the instructions
- in README.WinCE to make the appropriate changes in config.h.
-
-1999-10-30 Erik Hensema <erik.hensema at group2000.nl>
-
- * create.c (pthread_create): Explicitly initialise thread state to
- default values.
-
- * cancel.c (pthread_setcancelstate): Check for NULL 'oldstate'
- for compatibility with Solaris pthreads;
- (pthread_setcanceltype): ditto:
-
-1999-10-23 Erik Hensema <erik.hensema at group2000.nl>
-
- * pthread.h (ctime_r): Fix incorrect argument "_tm"
-
-1999-10-21 Aurelio Medina <aureliom at crt.com>
-
- * pthread.h (_POSIX_THREADS): Only define it if it isn't
- already defined. Projects may need to define this on
- the CC command line under Win32 as it doesn't have unistd.h
-
-1999-10-17 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * rwlock.c (pthread_rwlock_destroy): Add cast to remove compile
- warning.
-
- * condvar.c (pthread_cond_broadcast): Only release semaphores
- if there are waiting threads.
-
-1999-10-15 Lorin Hochstein <lmh at xiphos.ca>, Peter Slacik <Peter.Slacik at tatramed.sk>
-
- * condvar.c (cond_wait_cleanup): New static cleanup handler for
- cond_timedwait;
- (cond_timedwait): pthread_cleanup_push args changed;
- canceling a thread while it's in pthread_cond_wait
- will now decrement the waiters count and cleanup if it's the
- last waiter.
-
-1999-10-15 Graham Dumpleton <Graham.Dumpleton at ra.pad.otc.telstra.com.au>
-
- * condvar.c (cond_wait_cleanup): the last waiter will now reset the CV's
- wasBroadcast flag
-
-Thu Sep 16 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * rwlock.c (pthread_rwlock_destroy): Add serialisation.
- (_rwlock_check_need_init): Check for detroyed rwlock.
- * rwlock.c: Check return codes from _rwlock_check_need_init();
- modify comments; serialise access to rwlock objects during
- operations; rename rw_mutex to rw_lock.
- * implement.h: Rename rw_mutex to rw_lock.
- * mutex.c (pthread_mutex_destroy): Add serialisation.
- (_mutex_check_need_init): Check for detroyed mutex.
- * condvar.c (pthread_cond_destroy): Add serialisation.
- (_cond_check_need_init): Check for detroyed condvar.
- * mutex.c: Modify comments.
- * condvar.c: Modify comments.
-
-1999-08-10 Aurelio Medina <aureliom at crt.com>
-
- * implement.h (pthread_rwlock_t_): Add.
- * pthread.h (pthread_rwlock_t): Add.
- (PTHREAD_RWLOCK_INITIALIZER): Add.
- Add rwlock function prototypes.
- * rwlock.c: New module.
- * pthread.def: Add new rwlock functions.
- * private.c (ptw32_processInitialize): initialise
- ptw32_rwlock_test_init_lock critical section.
- * global.c (ptw32_rwlock_test_init_lock): Add.
-
- * mutex.c (pthread_mutex_destroy): Don't free mutex memory
- if mutex is PTHREAD_MUTEX_INITIALIZER and has not been
- initialised yet.
-
-1999-08-08 Milan Gardian <mg at tatramed.sk>
-
- * mutex.c (pthread_mutex_destroy): Free mutex memory.
-
-1999-08-22 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * exit.c (pthread_exit): Fix reference to potentially
- uninitialised pointer.
-
-1999-08-21 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * private.c (ptw32_threadStart): Apply fix of 1999-08-19
- this time to C++ and non-trapped C versions. Ommitted to
- do this the first time through.
-
-1999-08-19 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * private.c (ptw32_threadStart): Return exit status from
- the application thread startup routine.
- - Milan Gardian <mg at tatramed.sk>
-
-1999-08-18 John Bossom <john.Bossom at cognos.com>
-
- * exit.c (pthread_exit): Put status into pthread_t->exitStatus
- * private.c (ptw32_threadStart): Set pthread->exitStatus
- on exit of try{} block.
- * sync.c (pthread_join): use pthread_exitStatus value if the
- thread exit doesn't return a value (for Mingw32 CRTDLL
- which uses endthread instead of _endthreadex).
-
-Tue Aug 17 20:17:58 CDT 1999 Mumit Khan <khan at xraylith.wisc.edu>
-
- * create.c (pthread_create): Add CRTDLL suppport.
- * exit.c (pthread_exit): Likewise.
- * private.c (ptw32_threadStart): Likewise.
- (ptw32_threadDestroy): Likewise.
- * sync.c (pthread_join): Likewise.
- * tests/join1.c (main): Warn about partial support for CRTDLL.
-
-Tue Aug 17 20:00:08 1999 Mumit Khan <khan at xraylith.wisc.edu>
-
- * Makefile.in (LD): Delete entry point.
- * acconfig.h (STDCALL): Delete unused macro.
- * configure.in: Remove test for STDCALL.
- * config.h.in: Regenerate.
- * errno.c (_errno): Fix self type.
- * pthread.h (PT_STDCALL): Move from here to
- * implement.h (PT_STDCALL): here.
- (ptw32_threadStart): Fix prototype.
- * private.c (ptw32_threadStart): Likewise.
-
-1999-08-14 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * exit.c (pthread_exit): Don't call pthread_self() but
- get thread handle directly from TSD for efficiency.
-
-1999-08-12 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * private.c (ptw32_threadStart): ei[] only declared if _MSC_VER.
-
- * exit.c (pthread_exit): Check for implicitly created threads
- to avoid raising an unhandled exception.
-
-1999-07-12 Peter Slacik <Peter.Slacik at tatramed.sk>
-
- * condvar.c (pthread_cond_destroy): Add critical section.
- (cond_timedwait): Add critical section; check for timeout
- waiting on semaphore.
- (pthread_cond_broadcast): Add critical section.
-
-1999-07-09 Lorin Hochstein <lmh at xiphos.ca>, John Bossom <John.Bossom at Cognos.COM>
-
- The problem was that cleanup handlers were not executed when
- pthread_exit() was called.
-
- * implement.h (pthread_t_): Add exceptionInformation element for
- C++ per-thread exception information.
- (general): Define and rename exceptions.
-
-1999-07-09 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * misc.c (CancelableWait): PTW32_EPS_CANCEL (SEH) and
- ptw32_exception_cancel (C++) used to identify the exception.
-
- * cancel.c (pthread_testcancel): PTW32_EPS_CANCEL (SEH) and
- ptw32_exception_cancel (C++) used to identify the exception.
-
- * exit.c (pthread_exit): throw/raise an exception to return to
- ptw32_threadStart() to exit the thread. PTW32_EPS_EXIT (SEH)
- and ptw32_exception_exit (C++) used to identify the exception.
-
- * private.c (ptw32_threadStart): Add pthread_exit exception trap;
- clean up and exit the thread directly rather than via pthread_exit().
-
-Sun May 30 00:25:02 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * semaphore.h (mode_t): Conditionally typedef it.
-
-Fri May 28 13:33:05 1999 Mark E. Armstrong <avail at pacbell.net>
-
- * condvar.c (pthread_cond_broadcast): Fix possible memory fault
-
-Thu May 27 13:08:46 1999 Peter Slacik <Peter.Slacik at tatramed.sk>
-
- * condvar.c (pthread_cond_broadcast): Fix logic bug
-
-Thu May 27 13:08:46 1999 Bossom, John <John.Bossom at Cognos.COM>
-
- * condvar.c (pthread_cond_broadcast): optimise sem_post loop
-
-Fri May 14 12:13:18 1999 Mike Russo <miker at eai.com>
-
- * attr.c (pthread_attr_setdetachstate): Fix logic bug
-
-Sat May 8 09:42:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.def (sem_open): Add.
- (sem_close): Add.
- (sem_unlink): Add.
- (sem_getvalue): Add.
-
- * FAQ (Question 3): Add.
-
-Thu Apr 8 01:16:23 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * semaphore.c (sem_open): New function; returns an error (ENOSYS).
- (sem_close): ditto.
- (sem_unlink): ditto.
- (sem_getvalue): ditto.
-
- * semaphore.h (_POSIX_SEMAPHORES): define.
-
-Wed Apr 7 14:09:52 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * errno.c (_REENTRANT || _MT): Invert condition.
-
- * pthread.h (_errno): Conditionally include prototype.
-
-Wed Apr 7 09:37:00 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * *.c (comments): Remove individual attributions - these are
- documented sufficiently elsewhere.
-
- * implement.h (pthread.h): Remove extraneous include.
-
-Sun Apr 4 11:05:57 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * sched.c (sched.h): Include.
-
- * sched.h: New file for POSIX 1b scheduling.
-
- * pthread.h: Move opaque structures to implement.h; move sched_*
- prototypes out and into sched.h.
-
- * implement.h: Add opaque structures from pthread.h.
-
- * sched.c (sched_yield): New function.
-
- * condvar.c (ptw32_sem_*): Rename to sem_*; except for
- ptw32_sem_timedwait which is an private function.
-
-Sat Apr 3 23:28:00 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * Makefile.in (OBJS): Add errno.o.
-
-Fri Apr 2 11:08:50 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h (ptw32_sem_*): Remove prototypes now defined in
- semaphore.h.
-
- * pthread.h (sempahore.h): Include.
-
- * semaphore.h: New file for POSIX 1b semaphores.
-
- * semaphore.c (ptw32_sem_timedwait): Moved to private.c.
-
- * pthread.h (ptw32_sem_t): Change to sem_t.
-
- * private.c (ptw32_sem_timedwait): Moved from semaphore.c;
- set errno on error.
-
- * pthread.h (pthread_t_): Add per-thread errno element.
-
-Fri Apr 2 11:08:50 1999 John Bossom <jebossom at cognos.com>
-
- * semaphore.c (ptw32_sem_*): Change to sem_*; these functions
- will be exported from the library; set errno on error.
-
- * errno.c (_errno): New file. New function.
-
-Fri Mar 26 14:11:45 1999 Tor Lillqvist <tml at iki.fi>
-
- * semaphore.c (ptw32_sem_timedwait): Check for negative
- milliseconds.
-
-Wed Mar 24 11:32:07 1999 John Bossom <jebossom at cognos.com>
-
- * misc.c (CancelableWait): Initialise exceptionInformation[2].
- (pthread_self): Get a real Win32 thread handle for implicit threads.
-
- * cancel.c (pthread_testcancel): Initialise exceptionInformation[2].
-
- * implement.h (SE_INFORMATION): Fix values.
-
- * private.c (ptw32_threadDestroy): Close the thread handle.
-
-Fri Mar 19 12:57:27 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * cancel.c (comments): Update and cleanup.
-
-Fri Mar 19 09:12:59 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * private.c (ptw32_threadStart): status returns PTHREAD_CANCELED.
-
- * pthread.h (PTHREAD_CANCELED): defined.
-
-Tue Mar 16 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * all: Add GNU LGPL and Copyright and Warranty.
-
-Mon Mar 15 00:20:13 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * condvar.c (pthread_cond_init): fix possible uninitialised use
- of cv.
-
-Sun Mar 14 21:01:59 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * condvar.c (pthread_cond_destroy): don't do full cleanup if
- static initialised cv has never been used.
- (cond_timedwait): check result of auto-initialisation.
-
-Thu Mar 11 09:01:48 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.h (pthread_mutex_t): revert to (pthread_mutex_t *);
- define a value to serve as PTHREAD_MUTEX_INITIALIZER.
- (pthread_mutex_t_): remove staticinit and valid elements.
- (pthread_cond_t): revert to (pthread_cond_t_ *);
- define a value to serve as PTHREAD_COND_INITIALIZER.
- (pthread_cond_t_): remove staticinit and valid elements.
-
- * mutex.c (pthread_mutex_t args): adjust indirection of references.
- (all functions): check for PTHREAD_MUTEX_INITIALIZER value;
- check for NULL (invalid).
-
- * condvar.c (pthread_cond_t args): adjust indirection of references.
- (all functions): check for PTHREAD_COND_INITIALIZER value;
- check for NULL (invalid).
-
-Wed Mar 10 17:18:12 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * misc.c (CancelableWait): Undo changes from Mar 8 and 7.
-
-Mon Mar 8 11:18:59 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * misc.c (CancelableWait): Ensure cancelEvent handle is the lowest
- indexed element in the handles array. Enhance test for abandoned
- objects.
-
- * pthread.h (PTHREAD_MUTEX_INITIALIZER): Trailing elements not
- initialised are set to zero by the compiler. This avoids the
- problem of initialising the opaque critical section element in it.
- (PTHREAD_COND_INITIALIZER): Ditto.
-
- * semaphore.c (ptw32_sem_timedwait): Check sem == NULL earlier.
-
-Sun Mar 7 12:31:14 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * condvar.c (pthread_cond_init): set semaphore initial value
- to 0, not 1. cond_timedwait was returning signaled immediately.
-
- * misc.c (CancelableWait): Place the cancel event handle first
- in the handle table for WaitForMultipleObjects. This ensures that
- the cancel event is recognised and acted apon if both objects
- happen to be signaled together.
-
- * private.c (ptw32_cond_test_init_lock): Initialise and destroy.
-
- * implement.h (ptw32_cond_test_init_lock): Add extern.
-
- * global.c (ptw32_cond_test_init_lock): Add declaration.
-
- * condvar.c (pthread_cond_destroy): check for valid initialised CV;
- flag destroyed CVs as invalid.
- (pthread_cond_init): pthread_cond_t is no longer just a pointer.
- This is because PTHREAD_COND_INITIALIZER needs state info to reside
- in pthread_cond_t so that it can initialise on first use. Will work on
- making pthread_cond_t (and other objects like it) opaque again, if
- possible, later.
- (cond_timedwait): add check for statically initialisation of
- CV; initialise on first use.
- (pthread_cond_signal): check for valid CV.
- (pthread_cond_broadcast): check for valid CV.
- (_cond_check_need_init): Add.
-
- * pthread.h (PTHREAD_COND_INITIALIZER): Fix.
- (pthread_cond_t): no longer a pointer to pthread_cond_t_.
- (pthread_cond_t_): add 'staticinit' and 'valid' elements.
-
-Sat Mar 6 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h: Undate comments.
-
-Sun Feb 21 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.h (PTHREAD_MUTEX_INITIALIZER): missing braces around
- cs element initialiser.
-
-1999-02-21 Ben Elliston <bje at cygnus.com>
-
- * pthread.h (pthread_exit): The return type of this function is
- void, not int.
-
- * exit.c (pthread_exit): Do not return 0.
-
-Sat Feb 20 16:03:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * dll.c (DLLMain): Expand TryEnterCriticalSection support test.
-
- * mutex.c (pthread_mutex_trylock): The check for
- ptw32_try_enter_critical_section == NULL should have been
- removed long ago.
-
-Fri Feb 19 16:03:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * sync.c (pthread_join): Fix pthread_equal() test.
-
- * mutex.c (pthread_mutex_trylock): Check mutex != NULL before
- using it.
-
-Thu Feb 18 16:17:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * misc.c (pthread_equal): Fix inverted result.
-
- * Makefile.in: Use libpthread32.a as the name of the DLL export
- library instead of pthread.lib.
-
- * condvar.c (pthread_cond_init): cv could have been used unitialised;
- initialise.
-
- * create.c (pthread_create): parms could have been used unitialised;
- initialise.
-
- * pthread.h (struct pthread_once_t_): Remove redefinition.
-
-Sat Feb 13 03:03:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.h (struct pthread_once_t_): Replaced.
-
- * misc.c (pthread_once): Replace with John Bossom's version;
- has lighter weight serialisation; fixes problem of not holding
- competing threads until after the init_routine completes.
-
-Thu Feb 11 13:34:14 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * misc.c (CancelableWait): Change C++ exception throw.
-
- * sync.c (pthread_join): Change FIXME comment - issue resolved.
-
-Wed Feb 10 12:49:11 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * configure: Various temporary changes.
- - Kevin Ruland <Kevin.Ruland at anheuser-busch.com>
-
- * README: Update.
-
- * pthread.def (pthread_attr_getstackaddr): uncomment
- (pthread_attr_setstackaddr): uncomment
-
-Fri Feb 5 13:42:30 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * semaphore.c: Comment format changes.
-
-Thu Feb 4 10:07:28 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * global.c: Remove ptw32_exception instantiation.
-
- * cancel.c (pthread_testcancel): Change C++ exception throw.
-
- * implement.h: Remove extern declaration.
-
-Wed Feb 3 13:04:44 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * cleanup.c: Rename ptw32_*_cleanup() to pthread_*_cleanup().
-
- * pthread.def: Ditto.
-
- * pthread.h: Ditto.
-
- * pthread.def (pthread_cleanup_push): Remove from export list;
- the function is defined as a macro under all compilers.
- (pthread_cleanup_pop): Ditto.
-
- * pthread.h: Remove #if defined().
-
-Wed Feb 3 10:13:48 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * sync.c (pthread_join): Check for NULL value_ptr arg;
- check for detached threads.
-
-Tue Feb 2 18:07:43 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * implement.h: Add #include <pthread.h>.
- Change sem_t to ptw32_sem_t.
-
-Tue Feb 2 18:07:43 1999 Kevin Ruland <Kevin.Ruland at anheuser-busch.com>
-
- * signal.c (pthread_sigmask): Add and modify casts.
- Reverse LHS/RHS bitwise assignments.
-
- * pthread.h: Remove #include <semaphore.h>.
- (PTW32_ATTR_VALID): Add cast.
- (struct pthread_t_): Add sigmask element.
-
- * dll.c: Add "extern C" for DLLMain.
- (DllMain): Add cast.
-
- * create.c (pthread_create): Set sigmask in thread.
-
- * condvar.c: Remove #include. Change sem_* to ptw32_sem_*.
-
- * attr.c: Changed #include.
-
- * Makefile.in: Additional targets and changes to build the library
- as a DLL.
-
-Fri Jan 29 11:56:28 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * Makefile.in (OBJS): Add semaphore.o to list.
-
- * semaphore.c (ptw32_sem_timedwait): Move from private.c.
- Rename sem_* to ptw32_sem_*.
-
- * pthread.h (pthread_cond_t): Change type of sem_t.
- _POSIX_SEMAPHORES no longer defined.
-
- * semaphore.h: Contents moved to implement.h.
- Removed from source tree.
-
- * implement.h: Add semaphore function prototypes and rename all
- functions to prepend 'ptw32_'. They are
- now private to the pthreads-win32 implementation.
-
- * private.c: Change #warning.
- Move ptw32_sem_timedwait() to semaphore.c.
-
- * cleanup.c: Change #warning.
-
- * misc.c: Remove #include <errno.h>
-
- * pthread.def: Cleanup CVS merge conflicts.
-
- * global.c: Ditto.
-
- * ChangeLog: Ditto.
-
- * cleanup.c: Ditto.
-
-Sun Jan 24 01:34:52 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * semaphore.c (sem_wait): Remove second arg to
- pthreadCancelableWait() call.
-
-Sat Jan 23 17:36:40 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.def: Add new functions to export list.
-
- * pthread.h (PTHREAD_MUTEX_AUTO_CS_NP): New.
- (PTHREAD_MUTEX_FORCE_CS_NP): New.
-
- * README: Updated.
-
-Fri Jan 22 14:31:59 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * Makefile.in (CFLAGS): Remove -fhandle-exceptions. Not needed
- with egcs. Add -g for debugging.
-
- * create.c (pthread_create): Replace __stdcall with PT_STDCALL
- macro. This is a hack and must be fixed.
-
- * misc.c (CancelableWait): Remove redundant statement.
-
- * mutex.c (pthread_mutexattr_init): Cast calloc return value.
-
- * misc.c (CancelableWait): Add cast.
- (pthread_self): Add cast.
-
- * exit.c (pthread_exit): Add cast.
-
- * condvar.c (pthread_condattr_init): Cast calloc return value.
-
- * cleanup.c: Reorganise conditional compilation.
-
- * attr.c (pthread_attr_init): Remove unused 'result'.
- Cast malloc return value.
-
- * private.c (ptw32_callUserDestroyRoutines): Redo conditional
- compilation.
-
- * misc.c (CancelableWait): C++ version uses 'throw'.
-
- * cancel.c (pthread_testcancel): Ditto.
-
- * implement.h (class ptw32_exception): Define for C++.
-
- * pthread.h: Fix C, C++, and Win32 SEH condition compilation
- mayhem around pthread_cleanup_* defines. C++ version now uses John
- Bossom's cleanup handlers.
- (pthread_attr_t): Make 'valid' unsigned.
- Define '_timeb' as 'timeb' for Ming32.
- Define PT_STDCALL as nothing for Mingw32. May be temporary.
-
- * cancel.c (pthread_testcancel): Cast return value.
-
-Wed Jan 20 09:31:28 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.h (pthread_mutexattr_t): Changed to a pointer.
-
- * mutex.c (pthread_mutex_init): Conditionally create Win32 mutex
- - from John Bossom's implementation.
- (pthread_mutex_destroy): Conditionally close Win32 mutex
- - from John Bossom's implementation.
- (pthread_mutexattr_init): Replaced by John Bossom's version.
- (pthread_mutexattr_destroy): Ditto.
- (pthread_mutexattr_getpshared): New function from John Bossom's
- implementation.
- (pthread_mutexattr_setpshared): New function from John Bossom's
- implementation.
-
-Tue Jan 19 18:27:42 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * pthread.h (pthreadCancelableTimedWait): New prototype.
- (pthreadCancelableWait): Remove second argument.
-
- * misc.c (CancelableWait): New static function is
- pthreadCancelableWait() renamed.
- (pthreadCancelableWait): Now just calls CancelableWait() with
- INFINITE timeout.
- (pthreadCancelableTimedWait): Just calls CancelableWait()
- with passed in timeout.
-
-Tue Jan 19 18:27:42 1999 Scott Lightner <scott at curriculum.com>
-
- * private.c (ptw32_sem_timedwait): 'abstime' arg really is
- absolute time. Calculate relative time to wait from current
- time before passing timeout to new routine
- pthreadCancelableTimedWait().
-
-Tue Jan 19 10:27:39 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.h (pthread_mutexattr_setforcecs_np): New prototype.
-
- * mutex.c (pthread_mutexattr_init): Init 'pshared' and 'forcecs'
- attributes to 0.
- (pthread_mutexattr_setforcecs_np): New function (not portable).
-
- * pthread.h (pthread_mutex_t):
- Add 'mutex' element. Set to NULL in PTHREAD_MUTEX_INITIALIZER.
- The pthread_mutex_*() routines will try to optimise performance
- by choosing either mutexes or critical sections as the basis
- for pthread mutexes for each indevidual mutex.
- (pthread_mutexattr_t_): Add 'forcecs' element.
- Some applications may choose to force use of critical sections
- if they know that:-
- the mutex is PROCESS_PRIVATE and,
- either the OS supports TryEnterCriticalSection() or
- pthread_mutex_trylock() will never be called on the mutex.
- This attribute will be setable via a non-portable routine.
-
- Note: We don't yet support PROCESS_SHARED mutexes, so the
- implementation as it stands will default to Win32 mutexes only if
- the OS doesn't support TryEnterCriticalSection. On Win9x, and early
- versions of NT 'forcecs' will need to be set in order to get
- critical section based mutexes.
-
-Sun Jan 17 12:01:26 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.h (PTHREAD_MUTEX_INITIALIZER): Init new 'staticinit'
- value to '1' and existing 'valid' value to '1'.
-
- * global.c (ptw32_mutex_test_init_lock): Add.
-
- * implement.h (ptw32_mutex_test_init_lock.): Add extern.
-
- * private.c (ptw32_processInitialize): Init critical section for
- global lock used by _mutex_check_need_init().
- (ptw32_processTerminate): Ditto (:s/Init/Destroy/).
-
- * dll.c (dllMain): Move call to FreeLibrary() so that it is only
- called once when the process detaches.
-
- * mutex.c (_mutex_check_need_init): New static function to test
- and init PTHREAD_MUTEX_INITIALIZER mutexes. Provides serialised
- access to the internal state of the uninitialised static mutex.
- Called from pthread_mutex_trylock() and pthread_mutex_lock() which
- do a quick unguarded test to check if _mutex_check_need_init()
- needs to be called. This is safe as the test is conservative
- and is repeated inside the guarded section of
- _mutex_check_need_init(). Thus in all calls except the first
- calls to lock static mutexes, the additional overhead to lock any
- mutex is a single memory fetch and test for zero.
-
- * pthread.h (pthread_mutex_t_): Add 'staticinit' member. Mutexes
- initialised by PTHREAD_MUTEX_INITIALIZER aren't really initialised
- until the first attempt to lock it. Using the 'valid'
- flag (which flags the mutex as destroyed or not) to record this
- information would be messy. It is possible for a statically
- initialised mutex such as this to be destroyed before ever being
- used.
-
- * mutex.c (pthread_mutex_trylock): Call _mutex_check_need_init()
- to test/init PTHREAD_MUTEX_INITIALIZER mutexes.
- (pthread_mutex_lock): Ditto.
- (pthread_mutex_unlock): Add check to ensure we don't try to unlock
- an unitialised static mutex.
- (pthread_mutex_destroy): Add check to ensure we don't try to delete
- a critical section that we never created. Allows us to destroy
- a static mutex that has never been locked (and hence initialised).
- (pthread_mutex_init): Set 'staticinit' flag to 0 for the new mutex.
-
-Sun Jan 17 12:01:26 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * private.c (ptw32_sem_timedwait): Move from semaphore.c.
-
- * semaphore.c : Remove redundant #includes.
- (ptw32_sem_timedwait): Move to private.c.
- (sem_wait): Add missing abstime arg to pthreadCancelableWait() call.
-
-Fri Jan 15 23:38:05 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * condvar.c (cond_timedwait): Remove comment.
-
-Fri Jan 15 15:41:28 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * pthread.h: Add new 'abstime' arg to pthreadCancelableWait()
- prototype.
-
- * condvar.c (cond_timedwait): New generalised function called by
- both pthread_cond_wait() and pthread_cond_timedwait(). This is
- essentially pthread_cond_wait() renamed and modified to add the
- 'abstime' arg and call the new ptw32_sem_timedwait() instead of
- sem_wait().
- (pthread_cond_wait): Now just calls the internal static
- function cond_timedwait() with an INFINITE wait.
- (pthread_cond_timedwait): Now implemented. Calls the internal
- static function cond_timedwait().
-
- * implement.h (ptw32_sem_timedwait): New internal function
- prototype.
-
- * misc.c (pthreadCancelableWait): Added new 'abstime' argument
- to allow shorter than INFINITE wait.
-
- * semaphore.c (ptw32_sem_timedwait): New function for internal
- use. This is essentially sem_wait() modified to add the
- 'abstime' arg and call the modified (see above)
- pthreadCancelableWait().
-
-Thu Jan 14 14:27:13 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * cleanup.c: Correct _cplusplus to __cplusplus wherever used.
-
- * Makefile.in: Add CC=g++ and add -fhandle-exceptions to CFLAGS.
- The derived Makefile will compile all units of the package as C++
- so that those which include try/catch exception handling should work
- properly. The package should compile ok if CC=gcc, however, exception
- handling will not be included and thus thread cancellation, for
- example, will not work.
-
- * cleanup.c (ptw32_pop_cleanup): Add #warning to compile this
- file as C++ if using a cygwin32 environment. Perhaps the whole package
- should be compiled using g++ under cygwin.
-
- * private.c (ptw32_threadStart): Change #error directive
- into #warning and bracket for __CYGWIN__ and derivative compilers.
-
-Wed Jan 13 09:34:52 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * build.bat: Delete old binaries before compiling/linking.
-
-Tue Jan 12 09:58:38 1999 Tor Lillqvist <tml at iki.fi>
-
- * dll.c: The Microsoft compiler pragmas probably are more
- appropriately protected by _MSC_VER than by _WIN32.
-
- * pthread.h: Define ETIMEDOUT. This should be returned by
- pthread_cond_timedwait which is not implemented yet as of
- snapshot-1999-01-04-1305. It was implemented in the older version.
- The Microsoft compiler pragmas probably are more appropriately
- protected by _MSC_VER than by _WIN32.
-
- * pthread.def: pthread_mutex_destroy was missing from the def file
-
- * condvar.c (pthread_cond_broadcast): Ensure we only wait on threads
- if there were any waiting on the condition.
- I think pthread_cond_broadcast should do the WaitForSingleObject
- only if cv->waiters > 0? Otherwise it seems to hang, at least in the
- testg thread program from glib.
-
-Tue Jan 12 09:58:38 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * condvar.c (pthread_cond_timedwait): Fix function description
- comments.
-
- * semaphore.c (sem_post): Correct typo in comment.
-
-Mon Jan 11 20:33:19 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.h: Re-arrange conditional compile of pthread_cleanup-*
- macros.
-
- * cleanup.c (ptw32_push_cleanup): Provide conditional
- compile of cleanup->prev.
-
-1999-01-11 Tor Lillqvist <tml at iki.fi>
-
- * condvar.c (pthread_cond_init): Invert logic when testing the
- return value from calloc().
-
-Sat Jan 9 14:32:08 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h: Compile-time switch for CYGWIN derived environments
- to use CreateThread instead of _beginthreadex. Ditto for ExitThread.
- Patch provided by Anders Norlander <anorland at hem2.passagen.se>.
-
-Tue Jan 5 16:33:04 1999 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * cleanup.c (ptw32_pop_cleanup): Add C++ version of __try/__except
- block. Move trailing "}" out of #ifdef _WIN32 block left there by
- (rpj's) mistake.
-
- * private.c: Remove #include <errno.h> which is included by pthread.h.
-
-1998-12-11 Ben Elliston <bje at toilet.to.cygnus.com>
-
- * README: Update info about subscribing to the mailing list.
-
-Mon Jan 4 11:23:40 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * all: No code changes, just cleanup.
- - remove #if 0 /* Pre Bossom */ enclosed code.
- - Remove some redundant #includes.
- * pthread.h: Update implemented/unimplemented routines list.
- * Tag the bossom merge branch getting ready to merge back to main
- trunk.
-
-Tue Dec 29 13:11:16 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h: Move the following struct definitions to pthread.h:
- pthread_t_, pthread_attr_t_, pthread_mutex_t_, pthread_mutex_t_,
- pthread_mutexattr_t_, pthread_key_t_, pthread_cond_t_,
- pthread_condattr_t_, pthread_once_t_.
-
- * pthread.h: Add "_" prefix to pthread_push_cleanup and
- pthread_pop_cleanup internal routines, and associated struct and
- typedefs.
-
- * buildlib.bat: Add compile command for semaphore.c
-
- * pthread.def: Comment out pthread_atfork routine name.
- Now unimplemented.
-
- * tsd.c (pthread_setspecific): Rename tkAssocCreate to
- ptw32_tkAssocCreate.
- (pthread_key_delete): Rename tkAssocDestroy to
- ptw32_tkAssocDestroy.
-
- * sync.c (pthread_join): Rename threadDestroy to ptw32_threadDestroy
-
- * sched.c (is_attr): attr is now **attr (was *attr), so add extra
- NULL pointer test.
- (pthread_attr_setschedparam): Increase redirection for attr which is
- now a **.
- (pthread_attr_getschedparam): Ditto.
- (pthread_setschedparam): Change thread validation and rename "thread"
- Win32 thread Handle element name to match John Bossom's version.
- (pthread_getschedparam): Ditto.
-
- * private.c (ptw32_threadDestroy): Rename call to
- callUserDestroyRoutines() as ptw32_callUserDestroyRoutines()
-
- * misc.c: Add #include "implement.h".
-
- * dll.c: Remove defined(KLUDGE) wrapped code.
-
- * fork.c: Remove redefinition of ENOMEM.
- Remove pthread_atfork() and fork() with #if 0/#endif.
-
- * create.c (pthread_create): Rename threadStart and threadDestroy calls
- to ptw32_threadStart and ptw32_threadDestroy.
-
- * implement.h: Rename "detachedstate" to "detachstate".
-
- * attr.c: Rename "detachedstate" to "detachstate".
-
-Mon Dec 28 09:54:39 1998 John Bossom
-
- * semaphore.c: Initial version.
- * semaphore.h: Initial version.
-
-Mon Dec 28 09:54:39 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.h (pthread_attr_t_): Change to *pthread_attr_t.
-
-Mon Dec 28 09:54:39 1998 John Bossom, Ben Elliston
-
- * attr.c (pthread_attr_setstacksize): Merge with John's version.
- (pthread_attr_getstacksize): Merge with John's version.
- (pthread_attr_setstackaddr): Merge with John's version.
- (pthread_attr_getstackaddr): Merge with John's version.
- (pthread_attr_init): Merge with John's version.
- (pthread_attr_destroy): Merge with John's version.
- (pthread_attr_getdetachstate): Merge with John's version.
- (pthread_attr_setdetachstate): Merge with John's version.
- (is_attr): attr is now **attr (was *attr), so add extra NULL pointer
- test.
-
-Mon Dec 28 09:54:39 1998 Ross Johnson
-
- * implement.h (pthread_attr_t_): Add and rename elements in JEB's
- version to correspond to original, so that it can be used with
- original attr routines.
-
- * pthread.h: Add #endif at end which was truncated in merging.
-
-Sun Dec 20 14:51:58 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * misc.c (pthreadCancelableWait): New function by John Bossom. Non-standard
- but provides a hook that can be used to implement cancellation points in
- applications that use this library.
-
- * pthread.h (pthread_cleanup_pop): C++ (non-WIN32) version uses
- try/catch to emulate John Bossom's WIN32 __try/__finally behaviour.
- In the WIN32 version __finally block, add a test for AbnormalTermination otherwise
- cleanup is only run if the cleanup_pop execute arg is non-zero. Cancellation
- should cause the cleanup to run irrespective of the execute arg.
-
- * condvar.c (pthread_condattr_init): Replaced by John Bossom's version.
- (pthread_condattr_destroy): Replaced by John Bossom's version.
- (pthread_condattr_getpshared): Replaced by John Bossom's version.
- (pthread_condattr_setpshared): Replaced by John Bossom's version.
- (pthread_cond_init): Replaced by John Bossom's version.
- Fix comment (refered to mutex rather than condition variable).
- (pthread_cond_destroy): Replaced by John Bossom's version.
- (pthread_cond_wait): Replaced by John Bossom's version.
- (pthread_cond_timedwait): Replaced by John Bossom's version.
- (pthread_cond_signal): Replaced by John Bossom's version.
- (pthread_cond_broadcast): Replaced by John Bossom's version.
-
-Thu Dec 17 19:10:46 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * tsd.c (pthread_key_create): Replaced by John Bossom's version.
- (pthread_key_delete): Replaced by John Bossom's version.
- (pthread_setspecific): Replaced by John Bossom's version.
- (pthread_getspecific): Replaced by John Bossom's version.
-
-Mon Dec 7 09:44:40 1998 John Bossom
-
- * cancel.c (pthread_setcancelstate): Replaced.
- (pthread_setcanceltype): Replaced.
- (pthread_testcancel): Replaced.
- (pthread_cancel): Replaced.
-
- * exit.c (pthread_exit): Replaced.
-
- * misc.c (pthread_self): Replaced.
- (pthread_equal): Replaced.
-
- * sync.c (pthread_detach): Replaced.
- (pthread_join): Replaced.
-
- * create.c (pthread_create): Replaced.
-
- * private.c (ptw32_processInitialize): New.
- (ptw32_processTerminate): New.
- (ptw32_threadStart): New.
- (ptw32_threadDestroy): New.
- (ptw32_cleanupStack): New.
- (ptw32_tkAssocCreate): New.
- (ptw32_tkAssocDestroy): New.
- (ptw32_callUserDestroyRoutines): New.
-
- * implement.h: Added non-API structures and declarations.
-
- * dll.c (PthreadsEntryPoint): Cast return value of GetProcAddress
- to resolve compile warning from MSVC.
-
- * dll.c (DLLmain): Replaced.
- * dll.c (PthreadsEntryPoint):
- Re-applied Anders Norlander's patch:-
- Initialize ptw32_try_enter_critical_section at startup
- and release kernel32 handle when DLL is being unloaded.
-
-Sun Dec 6 21:54:35 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * buildlib.bat: Fix args to CL when building the .DLL
-
- * cleanup.c (ptw32_destructor_run_all): Fix TSD key management.
- This is a tidy-up before TSD and Thread management is completely
- replaced by John Bossom's code.
-
- * tsd.c (pthread_key_create): Fix TSD key management.
-
- * global.c (ptw32_key_virgin_next): Initialise.
-
- * build.bat: New DOS script to compile and link a pthreads app
- using Microsoft's CL compiler linker.
- * buildlib.bat: New DOS script to compile all the object files
- and create pthread.lib and pthread.dll using Microsoft's CL
- compiler linker.
-
-1998-12-05 Anders Norlander <anorland at hem2.passagen.se>
-
- * implement.h (ptw32_try_enter_critical_section): New extern
- * dll.c (ptw32_try_enter_critical_section): New pointer to
- TryEnterCriticalSection if it exists; otherwise NULL.
- * dll.c (PthreadsEntryPoint):
- Initialize ptw32_try_enter_critical_section at startup
- and release kernel32 handle when DLL is being unloaded.
- * mutex.c (pthread_mutex_trylock): Replaced check for NT with
- a check if ptw32_try_enter_critical_section is valid
- pointer to a function. Call ptw32_try_enter_critical_section
- instead of TryEnterCriticalSection to avoid errors on Win95.
-
-Thu Dec 3 13:32:00 1998 Ross Johnson <rpj at ise.canberra.edu.au>
-
- * README: Correct cygwin32 compatibility statement.
-
-Sun Nov 15 21:24:06 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * cleanup.c (ptw32_destructor_run_all): Declare missing void * arg.
- Fixup CVS merge conflicts.
-
-1998-10-30 Ben Elliston <bje at cygnus.com>
-
- * condvar.c (cond_wait): Fix semantic error. Test for equality
- instead of making an assignment.
-
-Fri Oct 30 15:15:50 1998 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * cleanup.c (ptw32_handler_push): Fixed bug appending new
- handler to list reported by Peter Slacik
- <Peter.Slacik at leibinger.freinet.de>.
- (new_thread): Rename poorly named local variable to
- "new_handler".
-
-Sat Oct 24 18:34:59 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * global.c: Add TSD key management array and index declarations.
-
- * implement.h: Ditto for externs.
-
-Fri Oct 23 00:08:09 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h (PTW32_TSD_KEY_REUSE): Add enum.
-
- * private.c (ptw32_delete_thread): Add call to
- ptw32_destructor_run_all() to clean up the threads keys.
-
- * cleanup.c (ptw32_destructor_run_all): Check for no more dirty
- keys to run destructors on. Assume that the destructor call always
- succeeds and set the key value to NULL.
-
-Thu Oct 22 21:44:44 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * tsd.c (pthread_setspecific): Add key management code.
- (pthread_key_create): Ditto.
- (pthread_key_delete): Ditto.
-
- * implement.h (struct ptw32_tsd_key): Add status member.
-
- * tsd.c: Add description of pthread_key_delete() from the
- standard as a comment.
-
-Fri Oct 16 17:38:47 1998 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * cleanup.c (ptw32_destructor_run_all): Fix and improve
- stepping through the key table.
-
-Thu Oct 15 14:05:01 1998 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * private.c (ptw32_new_thread): Remove init of destructorstack.
- No longer an element of pthread_t.
-
- * tsd.c (pthread_setspecific): Fix type declaration and cast.
- (pthread_getspecific): Ditto.
- (pthread_getspecific): Change error return value to NULL if key
- is not in use.
-
-Thu Oct 15 11:53:21 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * global.c (ptw32_tsd_key_table): Fix declaration.
-
- * implement.h(ptw32_TSD_keys_TlsIndex): Add missing extern.
- (ptw32_tsd_mutex): Ditto.
-
- * create.c (ptw32_start_call): Fix "keys" array declaration.
- Add comment.
-
- * tsd.c (pthread_setspecific): Fix type declaration and cast.
- (pthread_getspecific): Ditto.
-
- * cleanup.c (ptw32_destructor_run_all): Declare missing loop
- counter.
-
-Wed Oct 14 21:09:24 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * private.c (ptw32_new_thread): Increment ptw32_threads_count.
- (ptw32_delete_thread): Decrement ptw32_threads_count.
- Remove some comments.
-
- * exit.c (ptw32_exit): : Fix two pthread_mutex_lock() calls that
- should have been pthread_mutex_unlock() calls.
- (ptw32_vacuum): Remove call to ptw32_destructor_pop_all().
-
- * create.c (pthread_create): Fix two pthread_mutex_lock() calls that
- should have been pthread_mutex_unlock() calls.
-
- * global.c (ptw32_tsd_mutex): Add mutex for TSD operations.
-
- * tsd.c (pthread_key_create): Add critical section.
- (pthread_setspecific): Ditto.
- (pthread_getspecific): Ditto.
- (pthread_key_delete): Ditto.
-
- * sync.c (pthread_join): Fix two pthread_mutex_lock() calls that
- should have been pthread_mutex_unlock() calls.
-
-Mon Oct 12 00:00:44 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h (ptw32_tsd_key_table): New.
-
- * create.c (ptw32_start_call): Initialise per-thread TSD keys
- to NULL.
-
- * misc.c (pthread_once): Correct typo in comment.
-
- * implement.h (ptw32_destructor_push): Remove.
- (ptw32_destructor_pop): Remove.
- (ptw32_destructor_run_all): Rename from ptw32_destructor_pop_all.
- (PTW32_TSD_KEY_DELETED): Add enum.
- (PTW32_TSD_KEY_INUSE): Add enum.
-
- * cleanup.c (ptw32_destructor_push): Remove.
- (ptw32_destructor_pop): Remove.
- (ptw32_destructor_run_all): Totally revamped TSD.
-
- * dll.c (ptw32_TSD_keys_TlsIndex): Initialise.
-
- * tsd.c (pthread_setspecific): Totally revamped TSD.
- (pthread_getspecific): Ditto.
- (pthread_create): Ditto.
- (pthread_delete): Ditto.
-
-Sun Oct 11 22:44:55 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * global.c (ptw32_tsd_key_table): Add new global.
-
- * implement.h (ptw32_tsd_key_t and struct ptw32_tsd_key):
- Add.
- (struct _pthread): Remove destructorstack.
-
- * cleanup.c (ptw32_destructor_run_all): Rename from
- ptw32_destructor_pop_all. The key destructor stack was made
- global rather than per-thread. No longer removes destructor nodes
- from the stack. Comments updated.
-
-1998-10-06 Ben Elliston <bje at cygnus.com>
-
- * condvar.c (cond_wait): Use POSIX, not Win32 mutex calls.
- (pthread_cond_broadcast): Likewise.
- (pthread_cond_signal): Likewise.
-
-1998-10-05 Ben Elliston <bje at cygnus.com>
-
- * pthread.def: Update. Some functions aren't available yet, others
- are macros in <pthread.h>.
-
- * tests/join.c: Remove; useless.
-
-Mon Oct 5 14:25:08 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * pthread.def: New file for building the DLL.
-
-1998-10-05 Ben Elliston <bje at cygnus.com>
-
- * misc.c (pthread_equal): Correct inverted logic bug.
- (pthread_once): Use the POSIX mutex primitives, not Win32. Remove
- irrelevant FIXME comment.
-
- * global.c (PTHREAD_MUTEX_INITIALIZER): Move to pthread.h.
-
- * pthread.h (PTHREAD_MUTEX_INITIALIZER): Define.
- (pthread_mutex_t): Reimplement as a struct containing a valid
- flag. If the flag is ever down upon entry to a mutex operation,
- we call pthread_mutex_create() to initialise the object. This
- fixes the problem of how to handle statically initialised objects
- that can't call InitializeCriticalSection() due to their context.
- (PTHREAD_ONCE_INIT): Define.
-
- * mutex.c (pthread_mutex_init): Set valid flag.
- (pthread_mutex_destroy): Clear valid flag.
- (pthread_mutex_lock): Check and handle the valid flag.
- (pthread_mutex_unlock): Likewise.
- (pthread_mutex_trylock): Likewise.
-
- * tests/mutex3.c: New file; test for the static initialisation
- macro. Passes.
-
- * tests/create1.c: New file; test pthread_create(). Passes.
-
- * tests/equal.c: Poor test; remove.
-
- * tests/equal1.c New file; test pthread_equal(). Passes.
-
- * tests/once1.c: New file; test for pthread_once(). Passes.
-
- * tests/self.c: Remove; rename to self1.c.
-
- * tests/self1.c: This is the old self.c.
-
- * tests/self2.c: New file. Test pthread_self() with a single
- thread. Passes.
-
- * tests/self3.c: New file. Test pthread_self() with a couple of
- threads to ensure their thread IDs differ. Passes.
-
-1998-10-04 Ben Elliston <bje at cygnus.com>
-
- * tests/mutex2.c: Test pthread_mutex_trylock(). Passes.
-
- * tests/mutex1.c: New basic test for mutex functions (it passes).
- (main): Eliminate warning.
-
- * configure.in: Test for __stdcall, not _stdcall. Typo.
-
- * configure: Regenerate.
-
- * attr.c (pthread_attr_setstackaddr): Remove FIXME comment. Win32
- does know about ENOSYS after all.
- (pthread_attr_setstackaddr): Likewise.
-
-1998-10-03 Ben Elliston <bje at cygnus.com>
-
- * configure.in: Test for the `_stdcall' keyword. Define `STDCALL'
- to `_stdcall' if we have it, null otherwise.
-
- * configure: Regenerate.
-
- * acconfig.h (STDCALL): New define.
-
- * config.h.in: Regenerate.
-
- * create.c (ptw32_start_call): Add STDCALL prefix.
-
- * mutex.c (pthread_mutex_init): Correct function signature.
-
- * attr.c (pthread_attr_init): Only zero out the `sigmask' member
- if we have the sigset_t type.
-
- * pthread.h: No need to include <unistd.h>. It doesn't even exist
- on Win32! Again, an artifact of cross-compilation.
- (pthread_sigmask): Only provide if we have the sigset_t type.
-
- * process.h: Remove. This was a stand-in before we started doing
- native compilation under Win32.
-
- * pthread.h (pthread_mutex_init): Make `attr' argument const.
-
-1998-10-02 Ben Elliston <bje at cygnus.com>
-
- * COPYING: Remove.
-
- * COPYING.LIB: Add. This library is under the LGPL.
-
-1998-09-13 Ben Elliston <bje at cygnus.com>
-
- * configure.in: Test for required system features.
-
- * configure: Generate.
-
- * acconfig.h: New file.
-
- * config.h.in: Generate.
-
- * Makefile.in: Renamed from Makefile.
-
- * COPYING: Import from a recent GNU package.
-
- * config.guess: Likewise.
-
- * config.sub: Likewise.
-
- * install-sh: Likewise.
-
- * config.h: Remove.
-
- * Makefile: Likewise.
-
-1998-09-12 Ben Elliston <bje at cygnus.com>
-
- * windows.h: No longer needed; remove.
-
- * windows.c: Likewise.
-
-Sat Sep 12 20:09:24 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * windows.h: Remove error number definitions. These are in <errno.h>
-
- * tsd.c: Add comment explaining rationale for not building
- POSIX TSD on top of Win32 TLS.
-
-1998-09-12 Ben Elliston <bje at cygnus.com>
-
- * {most}.c: Include <errno.h> to get POSIX error values.
-
- * signal.c (pthread_sigmask): Only provide if HAVE_SIGSET_T is
- defined.
-
- * config.h: #undef features, don't #define them. This will be
- generated by autoconf very soon.
-
-1998-08-11 Ben Elliston <bje at cygnus.com>
-
- * Makefile (LIB): Define.
- (clean): Define target.
- (all): Build a library not just the object files.
-
- * pthread.h: Provide a definition for struct timespec if we don't
- already have one.
-
- * windows.c (TlsGetValue): Bug fix.
-
-Thu Aug 6 15:19:22 1998 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * misc.c (pthread_once): Fix arg 1 of EnterCriticalSection()
- and LeaveCriticalSection() calls to pass address-of lock.
-
- * fork.c (pthread_atfork): Typecast (void (*)(void *)) funcptr
- in each ptw32_handler_push() call.
-
- * exit.c (ptw32_exit): Fix attr arg in
- pthread_attr_getdetachstate() call.
-
- * private.c (ptw32_new_thread): Typecast (HANDLE) NULL.
- (ptw32_delete_thread): Ditto.
-
- * implement.h: (PTW32_MAX_THREADS): Add define. This keeps
- changing in an attempt to make thread administration data types
- opaque and cleanup DLL startup.
-
- * dll.c (PthreadsEntryPoint):
- (ptw32_virgins): Remove malloc() and free() calls.
- (ptw32_reuse): Ditto.
- (ptw32_win32handle_map): Ditto.
- (ptw32_threads_mutex_table): Ditto.
-
- * global.c (_POSIX_THREAD_THREADS_MAX): Initialise with
- PTW32_MAX_THREADS.
- (ptw32_virgins): Ditto.
- (ptw32_reuse): Ditto.
- (ptw32_win32handle_map): Ditto.
- (ptw32_threads_mutex_table): Ditto.
-
- * create.c (pthread_create): Typecast (HANDLE) NULL.
- Typecast (unsigned (*)(void *)) start_routine.
-
- * condvar.c (pthread_cond_init): Add address-of operator & to
- arg 1 of pthread_mutex_init() call.
- (pthread_cond_destroy): Add address-of operator & to
- arg 1 of pthread_mutex_destroy() call.
-
- * cleanup.c (ptw32_destructor_pop_all): Add (int) cast to
- pthread_getspecific() arg.
- (ptw32_destructor_pop): Add (void *) cast to "if" conditional.
- (ptw32_destructor_push): Add (void *) cast to
- ptw32_handler_push() "key" arg.
- (malloc.h): Add include.
-
- * implement.h (ptw32_destructor_pop): Add prototype.
-
- * tsd.c (implement.h): Add include.
-
- * sync.c (pthread_join): Remove target_thread_mutex and it's
- initialisation. Rename getdetachedstate to getdetachstate.
- Remove unused variable "exitcode".
- (pthread_detach): Remove target_thread_mutex and it's
- initialisation. Rename getdetachedstate to getdetachstate.
- Rename setdetachedstate to setdetachstate.
-
- * signal.c (pthread_sigmask): Rename SIG_SET to SIG_SETMASK.
- Cast "set" to (long *) in assignment to passify compiler warning.
- Add address-of operator & to thread->attr.sigmask in memcpy() call
- and assignment.
- (pthread_sigmask): Add address-of operator & to thread->attr.sigmask
- in memcpy() call and assignment.
-
- * windows.h (THREAD_PRIORITY_ERROR_RETURN): Add.
- (THREAD_PRIORITY_LOWEST): Add.
- (THREAD_PRIORITY_HIGHEST): Add.
-
- * sched.c (is_attr): Add function.
- (implement.h): Add include.
- (pthread_setschedparam): Rename all instances of "sched_policy"
- to "sched_priority".
- (pthread_getschedparam): Ditto.
-
-Tue Aug 4 16:57:58 1998 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * private.c (ptw32_delete_thread): Fix typo. Add missing ';'.
-
- * global.c (ptw32_virgins): Change types from pointer to
- array pointer.
- (ptw32_reuse): Ditto.
- (ptw32_win32handle_map): Ditto.
- (ptw32_threads_mutex_table): Ditto.
-
- * implement.h(ptw32_virgins): Change types from pointer to
- array pointer.
- (ptw32_reuse): Ditto.
- (ptw32_win32handle_map): Ditto.
- (ptw32_threads_mutex_table): Ditto.
-
- * private.c (ptw32_delete_thread): Fix "entry" should be "thread".
-
- * misc.c (pthread_self): Add extern for ptw32_threadID_TlsIndex.
-
- * global.c: Add comment.
-
- * misc.c (pthread_once): Fix member -> dereferences.
- Change ptw32_once_flag to once_control->flag in "if" test.
-
-Tue Aug 4 00:09:30 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h(ptw32_virgins): Add extern.
- (ptw32_virgin_next): Ditto.
- (ptw32_reuse): Ditto.
- (ptw32_reuse_top): Ditto.
- (ptw32_win32handle_map): Ditto.
- (ptw32_threads_mutex_table): Ditto.
-
- * global.c (ptw32_virgins): Changed from array to pointer.
- Storage allocation for the array moved into dll.c.
- (ptw32_reuse): Ditto.
- (ptw32_win32handle_map): Ditto.
- (ptw32_threads_mutex_table): Ditto.
-
- * dll.c (PthreadsEntryPoint): Set up thread admin storage when
- DLL is loaded.
-
- * fork.c (pthread_atfork): Fix function pointer arg to all
- ptw32_handler_push() calls. Change "arg" arg to NULL in child push.
-
- * exit.c: Add windows.h and process.h includes.
- (ptw32_exit): Add local detachstate declaration.
- (ptw32_exit): Fix incorrect name for pthread_attr_getdetachstate().
-
- * pthread.h (_POSIX_THREAD_ATTR_STACKSIZE): Move from global.c
- (_POSIX_THREAD_ATTR_STACKADDR): Ditto.
-
- * create.c (pthread_create): Fix #if should be #ifdef.
- (ptw32_start_call): Remove usused variables.
-
- * process.h: Create.
-
- * windows.h: Move _beginthreadex and _endthreadex into
- process.h
-
-Mon Aug 3 21:19:57 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * condvar.c (pthread_cond_init): Add NULL attr to
- pthread_mutex_init() call - default attributes will be used.
- (cond_wait): Fix typo.
- (cond_wait): Fix typo - cv was ev.
- (pthread_cond_broadcast): Fix two identical typos.
-
- * cleanup.c (ptw32_destructor_pop_all): Remove _ prefix from
- PTHREAD_DESTRUCTOR_ITERATIONS.
-
- * pthread.h: Move _POSIX_* values into posix.h
-
- * pthread.h: Fix typo in pthread_mutex_init() prototype.
-
- * attr.c (pthread_attr_init): Fix error in priority member init.
-
- * windows.h (THREAD_PRIORITY_NORMAL): Add.
-
- * pthread.h (sched_param): Add missing ';' to struct definition.
-
- * attr.c (pthread_attr_init): Remove obsolete pthread_attr_t
- member initialisation - cancelstate, canceltype, cancel_pending.
- (is_attr): Make arg "attr" a const.
-
- * implement.h (PTW32_HANDLER_POP_LIFO): Remove definition.
- (PTW32_HANDLER_POP_FIFO): Ditto.
- (PTW32_VALID): Add missing newline escape (\).
- (ptw32_handler_node): Make element "next" a pointer.
-
-1998-08-02 Ben Elliston <bje at cygnus.com>
-
- * windows.h: Remove duplicate TlsSetValue() prototype. Add
- TlsGetValue() prototype.
- (FALSE): Define.
- (TRUE): Likewise.
- Add forgotten errno values. Guard against multiple #includes.
-
- * windows.c: New file. Implement stubs for Win32 functions.
-
- * Makefile (SRCS): Remove. Not explicitly needed.
- (CFLAGS): Add -Wall for all warnings with GCC.
-
-Sun Aug 2 19:03:42 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * config.h: Create. This is a temporary stand-in for autoconf yet
- to be done.
- (HAVE_SIGNAL_H): Add.
-
- * pthread.h: Minor rearrangement for temporary config.h.
-
-Fri Jul 31 14:00:29 1998 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * cleanup.c (ptw32_destructor_pop): Implement. Removes
- destructors associated with a key without executing them.
- (ptw32_destructor_pop_all): Add FIXME comment.
-
- * tsd.c (pthread_key_delete): Add call to ptw32_destructor_pop().
-
-Fri Jul 31 00:05:45 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * tsd.c (pthread_key_create): Update to properly associate
- the destructor routine with the key.
- (pthread_key_delete): Add FIXME comment.
-
- * exit.c (ptw32_vacuum): Add call to
- ptw32_destructor_pop_all().
-
- * implement.h (ptw32_handler_pop_all): Add prototype.
- (ptw32_destructor_pop_all): Ditto.
-
- * cleanup.c (ptw32_destructor_push): Implement. This is just a
- call to ptw32_handler_push().
- (ptw32_destructor_pop_all): Implement. This is significantly
- different to ptw32_handler_pop_all().
-
- * Makefile (SRCS): Create. Preliminary.
-
- * windows.h: Create. Contains Win32 definitions for compile
- testing. This is just a standin for the real one.
-
- * pthread.h (SIG_UNBLOCK): Fix typo. Was SIG_BLOCK.
- (windows.h): Add include. Required for CRITICAL_SECTION.
- (pthread_cond_t): Move enum declaration outside of struct
- definition.
- (unistd.h): Add include - may be temporary.
-
- * condvar.c (windows.h): Add include.
-
- * implement.h (PTW32_THIS): Remove - no longer required.
- (PTW32_STACK): Use pthread_self() instead of PTW32_THIS.
-
-Thu Jul 30 23:12:45 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h: Remove ptw32_find_entry() prototype.
-
- * private.c: Extend comments.
- Remove ptw32_find_entry() - no longer needed.
-
- * create.c (ptw32_start_call): Add call to TlsSetValue() to
- store the thread ID.
-
- * dll.c (PthreadsEntryPoint): Implement. This is called
- whenever a process loads the DLL. Used to initialise thread
- local storage.
-
- * implement.h: Add ptw32_threadID_TlsIndex.
- Add ()s around PTW32_VALID expression.
-
- * misc.c (pthread_self): Re-implement using Win32 TLS to store
- the threads own ID.
-
-Wed Jul 29 11:39:03 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * private.c: Corrections in comments.
- (ptw32_new_thread): Alter "if" flow to be more natural.
-
- * cleanup.c (ptw32_handler_push): Same as below.
-
- * create.c (pthread_create): Same as below.
-
- * private.c (ptw32_new_thread): Rename "new" to "new_thread".
- Since when has a C programmer been required to know C++?
-
-Tue Jul 28 14:04:29 1998 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * implement.h: Add PTW32_VALID macro.
-
- * sync.c (pthread_join): Modify to use the new thread
- type and ptw32_delete_thread(). Rename "target" to "thread".
- Remove extra local variable "target".
- (pthread_detach): Ditto.
-
- * signal.c (pthread_sigmask): Move init of "us" out of inner block.
- Fix instance of "this" should have been "us". Rename "us" to "thread".
-
- * sched.c (pthread_setschedparam): Modify to use the new thread
- type.
- (pthread_getschedparam): Ditto.
-
- * private.c (ptw32_find_thread): Fix return type and arg.
-
- * implement.h: Remove PTW32_YES and PTW32_NO.
- (ptw32_new_thread): Add prototype.
- (ptw32_find_thread): Ditto.
- (ptw32_delete_thread): Ditto.
- (ptw32_new_thread_entry): Remove prototype.
- (ptw32_find_thread_entry): Ditto.
- (ptw32_delete_thread_entry): Ditto.
- ( PTW32_NEW, PTW32_INUSE, PTW32_EXITED, PTW32_REUSE):
- Add.
-
-
- * create.c (pthread_create): Minor rename "us" to "new" (I need
- these cues but it doesn't stop me coming out with some major bugs
- at times).
- Load start_routine and arg into the thread so the wrapper can
- call it.
-
- * exit.c (pthread_exit): Fix pthread_this should be pthread_self.
-
- * cancel.c (pthread_setcancelstate): Change
- ptw32_threads_thread_t * to pthread_t and init with
- pthread_this().
- (pthread_setcanceltype): Ditto.
-
- * exit.c (ptw32_exit): Add new pthread_t arg.
- Rename ptw32_delete_thread_entry to ptw32_delete_thread.
- Rename "us" to "thread".
- (pthread_exit): Call ptw32_exit with added thread arg.
-
- * create.c (ptw32_start_call): Insert missing ")".
- Add "us" arg to ptw32_exit() call.
- (pthread_create): Modify to use new thread allocation scheme.
-
- * private.c: Added detailed explanation of the new thread
- allocation scheme.
- (ptw32_new_thread): Totally rewritten to use
- new thread allocation scheme.
- (ptw32_delete_thread): Ditto.
- (ptw32_find_thread): Obsolete.
-
-Mon Jul 27 17:46:37 1998 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * create.c (pthread_create): Start of rewrite. Not completed yet.
-
- * private.c (ptw32_new_thread_entry): Start of rewrite. Not
- complete.
-
- * implement.h (ptw32_threads_thread): Rename, remove thread
- member, add win32handle and ptstatus members.
- (ptw32_t): Add.
-
- * pthread.h: pthread_t is no longer mapped directly to a Win32
- HANDLE type. This is so we can let the Win32 thread terminate and
- reuse the HANDLE while pthreads holds it's own thread ID until
- the last waiting join exits.
-
-Mon Jul 27 00:20:37 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * private.c (ptw32_delete_thread_entry): Destroy the thread
- entry attribute object before deleting the thread entry itself.
-
- * attr.c (pthread_attr_init): Initialise cancel_pending = FALSE.
- (pthread_attr_setdetachstate): Rename "detached" to "detachedstate".
- (pthread_attr_getdetachstate): Ditto.
-
- * exit.c (ptw32_exit): Fix incorrect check for detachedstate.
-
- * implement.h (ptw32_call_t): Remove env member.
-
-Sun Jul 26 13:06:12 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h (ptw32_new_thread_entry): Fix prototype.
- (ptw32_find_thread_entry): Ditto.
- (ptw32_delete_thread_entry): Ditto.
- (ptw32_exit): Add prototype.
-
- * exit.c (ptw32_exit): New function. Called from pthread_exit()
- and ptw32_start_call() to exit the thread. It allows an extra
- argument which is the return code passed to _endthreadex().
- (ptw32_exit): Move thread entry delete call from ptw32_vacuum()
- into here. Add more explanation of thread entry deletion.
- (ptw32_exit): Clarify comment.
-
- * create.c (ptw32_start_call): Change pthread_exit() call to
- ptw32_exit() call.
-
- * exit.c (ptw32_vacuum): Add thread entry deletion code
- moved from ptw32_start_call(). See next item.
- (pthread_exit): Remove longjmp(). Add mutex lock around thread table
- manipulation code. This routine now calls _enthreadex().
-
- * create.c (ptw32_start_call): Remove setjmp() call and move
- cleanup code out. Call pthread_exit(NULL) to terminate the thread.
-
-1998-07-26 Ben Elliston <bje at cygnus.com>
-
- * tsd.c (pthread_getspecific): Update comments.
-
- * mutex.c (pthread_mutexattr_setpshared): Not supported; remove.
- (pthread_mutexattr_getpshared): Likewise.
-
- * pthread.h (pthread_mutexattr_setpshared): Remove prototype.
- (pthread_mutexattr_getpshared): Likewise.
-
-Sun Jul 26 00:09:59 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * sync.c: Rename all instances of ptw32_count_mutex to
- ptw32_table_mutex.
-
- * implement.h: Rename ptw32_count_mutex to
- ptw32_table_mutex.
-
- * global.c: Rename ptw32_count_mutex to
- ptw32_table_mutex.
-
- * create.c (pthread_create): Add critical sections.
- (ptw32_start_call): Rename ptw32_count_mutex to
- ptw32_table_mutex.
-
- * cancel.c (pthread_setcancelstate): Fix indirection bug and rename
- "this" to "us".
-
- * signal.c (pthread_sigmask): Rename "this" to "us" and fix some
- minor syntax errors. Declare "us" and initialise it.
-
- * sync.c (pthread_detach): Rename "this" to "target".
-
- * pthread.h: Converting PTHREAD_* defines to alias the (const int)
- values in global.c.
-
- * global.c: Started converting PTHREAD_* defines to (const int) as
- a part of making the eventual pthreads DLL binary compatible
- through version changes.
-
- * condvar.c (cond_wait): Add cancelation point. This applies the
- point to both pthread_cond_wait() and pthread_cond_timedwait().
-
- * exit.c (pthread_exit): Rename "this" to "us".
-
- * implement.h: Add comment.
-
- * sync.c (pthread_join): I've satisfied myself that pthread_detach()
- does set the detached attribute in the thread entry attributes
- to PTHREAD_CREATE_DETACHED. "if" conditions were changed to test
- that attribute instead of a separate flag.
-
- * create.c (pthread_create): Rename "this" to "us".
- (pthread_create): cancelstate and canceltype are not attributes
- so the copy to thread entry attribute storage was removed.
- Only the thread itself can change it's cancelstate or canceltype,
- ie. the thread must exist already.
-
- * private.c (ptw32_delete_thread_entry): Mutex locks removed.
- Mutexes must be applied at the caller level.
- (ptw32_new_thread_entry): Ditto.
- (ptw32_new_thread_entry): Init cancelstate, canceltype, and
- cancel_pending to default values.
- (ptw32_new_thread_entry): Rename "this" to "new".
- (ptw32_find_thread_entry): Rename "this" to "entry".
- (ptw32_delete_thread_entry): Rename "thread_entry" to "entry".
-
- * create.c (ptw32_start_call): Mutexes changed to
- ptw32_count_mutex. All access to the threads table entries is
- under the one mutex. Otherwise chaos reigns.
-
-Sat Jul 25 23:16:51 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h (ptw32_threads_thread): Move cancelstate and
- canceltype members out of pthread_attr_t into here.
-
- * fork.c (fork): Add comment.
-
-1998-07-25 Ben Elliston <bje at cygnus.com>
-
- * fork.c (fork): Autoconfiscate.
-
-Sat Jul 25 00:00:13 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * create.c (ptw32_start_call): Set thread priority. Ensure our
- thread entry is removed from the thread table but only if
- pthread_detach() was called and there are no waiting joins.
- (pthread_create): Set detach flag in thread entry if the
- thread is created PTHREAD_CREATE_DETACHED.
-
- * pthread.h (pthread_attr_t): Rename member "detachedstate".
-
- * attr.c (pthread_attr_init): Rename attr members.
-
- * exit.c (pthread_exit): Fix indirection mistake.
-
- * implement.h (PTW32_THREADS_TABLE_INDEX): Add.
-
- * exit.c (ptw32_vacuum): Fix incorrect args to
- ptw32_handler_pop_all() calls.
- Make thread entry removal conditional.
-
- * sync.c (pthread_join): Add multiple join and async detach handling.
-
- * implement.h (PTW32_THREADS_TABLE_INDEX): Add.
-
- * global.c (ptw32_threads_mutex_table): Add.
-
- * implement.h (ptw32_once_flag): Remove.
- (ptw32_once_lock): Ditto.
- (ptw32_threads_mutex_table): Add.
-
- * global.c (ptw32_once_flag): Remove.
- (ptw32_once_lock): Ditto.
-
- * sync.c (pthread_join): Fix tests involving new return value
- from ptw32_find_thread_entry().
- (pthread_detach): Ditto.
-
- * private.c (ptw32_find_thread_entry): Failure return code
- changed from -1 to NULL.
-
-Fri Jul 24 23:09:33 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * create.c (pthread_create): Change . to -> in sigmask memcpy() args.
-
- * pthread.h: (pthread_cancel): Add function prototype.
- (pthread_testcancel): Ditto.
-
-1998-07-24 Ben Elliston <bje at cygnus.com>
-
- * pthread.h (pthread_condattr_t): Rename dummy structure member.
- (pthread_mutexattr_t): Likewise.
-
-Fri Jul 24 21:13:55 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * cancel.c (pthread_cancel): Implement.
- (pthread_testcancel): Implement.
-
- * exit.c (pthread_exit): Add comment explaining the longjmp().
-
- * implement.h (ptw32_threads_thread_t): New member cancelthread.
- (PTW32_YES): Define.
- (PTW32_NO): Define.
- (RND_SIZEOF): Remove.
-
- * create.c (pthread_create): Rename cancelability to cancelstate.
-
- * pthread.h (pthread_attr_t): Rename cancelability to cancelstate.
- (PTHREAD_CANCELED): Define.
-
-1998-07-24 Ben Elliston <bje at cygnus.com>
-
- * pthread.h (SIG_BLOCK): Define if not already defined.
- (SIG_UNBLOCK): Likewise.
- (SIG_SETMASK): Likewise.
- (pthread_attr_t): Add signal mask member.
- (pthread_sigmask): Add function prototype.
-
- * signal.c (pthread_sigmask): Implement.
-
- * create.c: #include <string.h> to get a prototype for memcpy().
- (pthread_create): New threads inherit their creator's signal
- mask. Copy the signal mask to the new thread structure if we know
- about signals.
-
-Fri Jul 24 16:33:17 1998 Ross Johnson <rpj at swan.canberra.edu.au>
-
- * fork.c (pthread_atfork): Add all the necessary push calls.
- Local implementation semantics:
- If we get an ENOMEM at any time then ALL handlers
- (including those from previous pthread_atfork() calls) will be
- popped off each of the three atfork stacks before we return.
- (fork): Add all the necessary pop calls. Add the thread cancellation
- and join calls to the child fork.
- Add #includes.
-
- * implement.h: (ptw32_handler_push): Fix return type and stack arg
- type in prototype.
- (ptw32_handler_pop): Fix stack arg type in prototype.
- (ptw32_handler_pop_all): Fix stack arg type in prototype.
-
- * cleanup.c (ptw32_handler_push): Change return type to int and
- return ENOMEM if malloc() fails.
-
- * sync.c (pthread_detach): Use equality test, not assignment.
-
- * create.c (ptw32_start_call): Add call to Win32 CloseHandle()
- if thread is detached.
-
-1998-07-24 Ben Elliston <bje at cygnus.com>
-
- * sync.c (pthread_detach): Close the Win32 thread handle to
- emulate detached (or daemon) threads.
-
-Fri Jul 24 03:00:25 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * sync.c (pthread_join): Save valueptr arg in joinvalueptr for
- pthread_exit() to use.
-
- * private.c (ptw32_new_thread_entry): Initialise joinvalueptr to
- NULL.
-
- * create.c (ptw32_start_call): Rewrite to facilitate joins.
- pthread_exit() will do a longjmp() back to here. Does appropriate
- cleanup and exit/return from the thread.
- (pthread_create): _beginthreadex() now passes a pointer to our
- thread table entry instead of just the call member of that entry.
-
- * implement.h (ptw32_threads_thread): New member
- void ** joinvalueptr.
- (ptw32_call_t): New member jmpbuf env.
-
- * exit.c (pthread_exit): Major rewrite to handle joins and handing
- value pointer to joining thread. Uses longjmp() back to
- ptw32_start_call().
-
- * create.c (pthread_create): Ensure values of new attribute members
- are copied to the thread attribute object.
-
- * attr.c (pthread_attr_destroy): Fix merge conflicts.
- (pthread_attr_getdetachstate): Fix merge conflicts.
- (pthread_attr_setdetachstate): Fix merge conflicts.
-
- * pthread.h: Fix merge conflicts.
-
- * sync.c (pthread_join): Fix merge conflicts.
-
-Fri Jul 24 00:21:21 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * sync.c (pthread_join): Add check for valid and joinable
- thread.
- (pthread_detach): Implement. After checking for a valid and joinable
- thread, it's still a no-op.
-
- * private.c (ptw32_find_thread_entry): Bug prevented returning
- an error value in some cases.
-
- * attr.c (pthread_attr_setdetachedstate): Implement.
- (pthread_attr_getdetachedstate): Implement.
-
- * implement.h: Move more hidden definitions into here from
- pthread.h.
-
-1998-07-24 Ben Elliston <bje at cygnus.com>
-
- * pthread.h (PTHREAD_CREATE_JOINABLE): Define.
- (PTHREAD_CREATE_DETACHED): Likewise.
- (pthread_attr_t): Add new structure member `detached'.
- (pthread_attr_getdetachstate): Add function prototype.
- (pthread_attr_setdetachstate): Likewise.
-
- * sync.c (pthread_join): Return if the target thread is detached.
-
- * attr.c (pthread_attr_init): Initialise cancelability and
- canceltype structure members.
- (pthread_attr_getdetachstate): Implement.
- (pthread_attr_setdetachstate): Likewise.
-
- * implement.h (PTW32_CANCEL_DEFAULTS): Remove. Bit fields
- proved to be too cumbersome. Set the defaults in attr.c using the
- public PTHREAD_CANCEL_* constants.
-
- * cancel.c: New file.
-
- * pthread.h (sched_param): Define this type.
- (pthread_attr_getschedparam): Add function prototype.
- (pthread_attr_setschedparam): Likewise.
- (pthread_setcancelstate): Likewise.
- (pthread_setcanceltype): Likewise.
- (sched_get_priority_min): Likewise.
- (sched_get_priority_max): Likewise.
- (pthread_mutexattr_setprotocol): Remove; not supported.
- (pthread_mutexattr_getprotocol): Likewise.
- (pthread_mutexattr_setprioceiling): Likewise.
- (pthread_mutexattr_getprioceiling): Likewise.
- (pthread_attr_t): Add canceltype member. Update comments.
- (SCHED_OTHER): Define this scheduling policy constant.
- (SCHED_FIFO): Likewise.
- (SCHED_RR): Likewise.
- (SCHED_MIN): Define the lowest possible value for this constant.
- (SCHED_MAX): Likewise, the maximum possible value.
- (PTHREAD_CANCEL_ASYNCHRONOUS): Redefine.
- (PTHREAD_CANCEL_DEFERRED): Likewise.
-
- * sched.c: New file.
- (pthread_setschedparam): Implement.
- (pthread_getschedparam): Implement.
- (sched_get_priority_max): Validate policy argument.
- (sched_get_priority_min): Likewise.
-
- * mutex.c (pthread_mutexattr_setprotocol): Remove; not supported.
- (pthread_mutexattr_getprotocol): Likewise.
- (pthread_mutexattr_setprioceiling): Likewise.
- (pthread_mutexattr_getprioceiling): Likewise.
-
-Fri Jul 24 00:21:21 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * create.c (pthread_create): Arg to ptw32_new_thread_entry()
- changed. See next entry. Move mutex locks out. Changes made yesterday
- and today allow us to start the new thread running rather than
- temporarily suspended.
-
- * private.c (ptw32_new_thread_entry): ptw32_thread_table
- was changed back to a table of thread structures rather than pointers.
- As such we're trading storage for increaded speed. This routine
- was modified to work with the new table. Mutex lock put in around
- global data accesses.
- (ptw32_find_thread_entry): Ditto
- (ptw32_delete_thread_entry): Ditto
-
-Thu Jul 23 23:25:30 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * global.c: New. Global data objects declared here. These moved from
- pthread.h.
-
- * pthread.h: Move implementation hidden definitions into
- implement.h.
-
- * implement.h: Move implementation hidden definitions from
- pthread.h. Add constants to index into the different handler stacks.
-
- * cleanup.c (ptw32_handler_push): Simplify args. Restructure.
- (ptw32_handler_pop): Simplify args. Restructure.
- (ptw32_handler_pop_all): Simplify args. Restructure.
-
-Wed Jul 22 00:16:22 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * attr.c, implement.h, pthread.h, ChangeLog: Resolve CVS merge
- conflicts.
-
- * private.c (ptw32_find_thread_entry): Changes to return type
- to support leaner ptw32_threads_table[] which now only stores
- ptw32_thread_thread_t *.
- (ptw32_new_thread_entry): Internal changes.
- (ptw32_delete_thread_entry): Internal changes to avoid contention.
- Calling routines changed accordingly.
-
- * pthread.h: Modified cleanup macros to use new generic push and pop.
- Added destructor and atfork stacks to ptw32_threads_thread_t.
-
- * cleanup.c (ptw32_handler_push, ptw32_handler_pop,
- ptw32_handler_pop_all): Renamed cleanup push and pop routines
- and made generic to handle destructors and atfork handlers as
- well.
-
- * create.c (ptw32_start_call): New function is a wrapper for
- all new threads. It allows us to do some cleanup when the thread
- returns, ie. that is otherwise only done if the thread is cancelled.
-
- * exit.c (ptw32_vacuum): New function contains code from
- pthread_exit() that we need in the new ptw32_start_call()
- as well.
-
- * implement.h: Various additions and minor changes.
-
- * pthread.h: Various additions and minor changes.
- Change cleanup handler macros to use generic handler push and pop
- functions.
-
- * attr.c: Minor mods to all functions.
- (is_attr): Implemented missing function.
-
- * create.c (pthread_create): More clean up.
-
- * private.c (ptw32_find_thread_entry): Implement.
- (ptw32_delete_thread_entry): Implement.
- (ptw32_new_thread_entry): Implement.
- These functions manipulate the implementations internal thread
- table and are part of general code cleanup and modularisation.
- They replace ptw32_getthreadindex() which was removed.
-
- * exit.c (pthread_exit): Changed to use the new code above.
-
- * pthread.h: Add cancelability constants. Update comments.
-
-1998-07-22 Ben Elliston <bje at cygnus.com>
-
- * attr.c (pthread_setstacksize): Update test of attr argument.
- (pthread_getstacksize): Likewise.
- (pthread_setstackaddr): Likewise.
- (pthread_getstackaddr): Likewise.
- (pthread_attr_init): No need to allocate any storage.
- (pthread_attr_destroy): No need to free any storage.
-
- * mutex.c (is_attr): Not likely to be needed; remove.
- (remove_attr): Likewise.
- (insert_attr): Likewise.
-
- * implement.h (ptw32_mutexattr_t): Moved to a public definition
- in pthread.h. There was little gain in hiding these details.
- (ptw32_condattr_t): Likewise.
- (ptw32_attr_t): Likewise.
-
- * pthread.h (pthread_atfork): Add function prototype.
- (pthread_attr_t): Moved here from implement.h.
-
- * fork.c (pthread_atfork): Preliminary implementation.
- (ptw32_fork): Likewise.
-
-Wed Jul 22 00:16:22 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * cleanup.c (ptw32_cleanup_push): Implement.
- (ptw32_cleanup_pop): Implement.
- (ptw32_do_cancellation): Implement.
- These are private to the implementation. The real cleanup functions
- are macros. See below.
-
- * pthread.h (pthread_cleanup_push): Implement as a macro.
- (pthread_cleanup_pop): Implement as a macro.
- Because these are macros which start and end a block, the POSIX scoping
- requirement is observed. See the comment in the file.
-
- * exit.c (pthread_exit): Refine the code.
-
- * create.c (pthread_create): Code cleanup.
-
- * implement.h (RND_SIZEOF): Add RND_SIZEOF(T) to round sizeof(T)
- up to multiple of DWORD.
- Add function prototypes.
-
- * private.c (ptw32_getthreadindex): "*thread" should have been
- "thread". Detect empty slot fail condition.
-
-1998-07-20 Ben Elliston <bje at cygnus.com>
-
- * misc.c (pthread_once): Implement. Don't use a per-application
- flag and mutex--make `pthread_once_t' contain these elements in
- their structure. The earlier version had incorrect semantics.
-
- * pthread.h (ptw32_once_flag): Add new variable. Remove.
- (ptw32_once_lock): Add new mutex lock to ensure integrity of
- access to ptw32_once_flag. Remove.
- (pthread_once): Add function prototype.
- (pthread_once_t): Define this type.
-
-Mon Jul 20 02:31:05 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * private.c (ptw32_getthreadindex): Implement.
-
- * pthread.h: Add application static data dependent on
- _PTHREADS_BUILD_DLL define. This is needed to avoid allocating
- non-sharable static data within the pthread DLL.
-
- * implement.h: Add ptw32_cleanup_stack_t, ptw32_cleanup_node_t
- and PTW32_HASH_INDEX.
-
- * exit.c (pthread_exit): Begin work on cleanup and de-allocate
- thread-private storage.
-
- * create.c (pthread_create): Add thread to thread table.
- Keep a thread-private copy of the attributes with default values
- filled in when necessary. Same for the cleanup stack. Make
- pthread_create C run-time library friendly by using _beginthreadex()
- instead of CreateThread(). Fix error returns.
-
-Sun Jul 19 16:26:23 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h: Rename pthreads_thread_count to ptw32_threads_count.
- Create ptw32_threads_thread_t struct to keep thread specific data.
-
- * create.c: Rename pthreads_thread_count to ptw32_threads_count.
- (pthread_create): Handle errors from CreateThread().
-
-1998-07-19 Ben Elliston <bje at cygnus.com>
-
- * condvar.c (pthread_cond_wait): Generalise. Moved from here ..
- (cond_wait): To here.
- (pthread_cond_timedwait): Implement; use generalised cond_wait().
-
- * pthread.h (pthread_key_t): Define this type.
- (pthread_key_create): Add function prototype.
- (pthread_setspecific): Likewise.
- (pthread_getspecific): Likwise.
- (pthread_key_delete): Likewise.
-
- * tsd.c (pthread_key_create): Implement.
- (pthread_setspecific): Likewise.
- (pthread_getspecific): Likewise.
- (pthread_key_delete): Likewise.
-
- * mutex.c (pthread_mutex_trylock): Return ENOSYS if this function
- is called on a Win32 platform which is not Windows NT.
-
-1998-07-18 Ben Elliston <bje at cygnus.com>
-
- * condvar.c (pthread_condattr_init): Do not attempt to malloc any
- storage; none is needed now that condattr_t is an empty struct.
- (pthread_condattr_destory): Likewise; do not free storage.
- (pthread_condattr_setpshared): No longer supported; return ENOSYS.
- (pthread_condattr_getpshared): Likewise.
- (pthread_cond_init): Implement with help from Douglas Schmidt.
- Remember to initialise the cv's internal mutex.
- (pthread_cond_wait): Likewise.
- (pthread_cond_signal): Likewise.
- (pthread_cond_broadcast): Likewise.
- (pthread_cond_timedwait): Preliminary implementation, but I need
- to see some API documentation for `WaitForMultipleObject'.
- (pthread_destory): Implement.
-
- * pthread.h (pthread_cond_init): Add function protoype.
- (pthread_cond_broadcast): Likewise.
- (pthread_cond_signal): Likewise.
- (pthread_cond_timedwait): Likewise.
- (pthread_cond_wait): Likewise.
- (pthread_cond_destroy): Likewise.
- (pthread_cond_t): Define this type. Fix for u_int. Do not assume
- that the mutex contained withing the pthread_cond_t structure will
- be a critical section. Use our new POSIX type!
-
- * implement.h (ptw32_condattr_t): Remove shared attribute.
-
-1998-07-17 Ben Elliston <bje at cygnus.com>
-
- * pthread.h (PTHREADS_PROCESS_PRIVATE): Remove.
- (PTHREAD_PROCESS_SHARED): Likewise. No support for mutexes shared
- across processes for now.
- (pthread_mutex_t): Use a Win32 CRITICAL_SECTION type for better
- performance.
-
- * implement.h (ptw32_mutexattr_t): Remove shared attribute.
-
- * mutex.c (pthread_mutexattr_setpshared): This optional function
- is no longer supported, since we want to implement POSIX mutex
- variables using the much more efficient Win32 critical section
- primitives. Critical section objects in Win32 cannot be shared
- between processes.
- (pthread_mutexattr_getpshared): Likewise.
- (pthread_mutexattr_init): No need to malloc any storage; the
- attributes structure is now empty.
- (pthread_mutexattr_destroy): This is now a nop.
- (pthread_mutex_init): Use InitializeCriticalSection().
- (pthread_mutex_destroy): Use DeleteCriticalSection().
- (pthread_mutex_lock): Use EnterCriticalSection().
- (pthread_mutex_trylock): Use TryEnterCriticalSection(). This is
- not supported by Windows 9x, but trylock is a hack anyway, IMHO.
- (pthread_mutex_unlock): Use LeaveCriticalSection().
-
-1998-07-14 Ben Elliston <bje at cygnus.com>
-
- * attr.c (pthread_attr_setstacksize): Implement.
- (pthread_attr_getstacksize): Likewise.
- (pthread_attr_setstackaddr): Likewise.
- (pthread_attr_getstackaddr): Likewise.
- (pthread_attr_init): Likewise.
- (pthread_attr_destroy): Likewise.
-
- * condvar.c (pthread_condattr_init): Add `_cond' to function name.
-
- * mutex.c (pthread_mutex_lock): Add `_mutex' to function name.
- (pthread_mutex_trylock): Likewise.
- (pthread_mutex_unlock): Likewise.
-
- * pthread.h (pthread_condattr_setpshared): Fix typo.
- (pthread_attr_init): Add function prototype.
- (pthread_attr_destroy): Likewise.
- (pthread_attr_setstacksize): Likewise.
- (pthread_attr_getstacksize): Likewise.
- (pthread_attr_setstackaddr): Likewise.
- (pthread_attr_getstackaddr): Likewise.
-
-Mon Jul 13 01:09:55 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h: Wrap in #ifndef _IMPLEMENT_H
-
- * create.c (pthread_create): Map stacksize attr to Win32.
-
- * mutex.c: Include implement.h
-
-1998-07-13 Ben Elliston <bje at cygnus.com>
-
- * condvar.c (pthread_condattr_init): Implement.
- (pthread_condattr_destroy): Likewise.
- (pthread_condattr_setpshared): Likewise.
- (pthread_condattr_getpshared): Likewise.
-
- * implement.h (PTHREAD_THREADS_MAX): Remove trailing semicolon.
- (PTHREAD_STACK_MIN): Specify; needs confirming.
- (ptw32_attr_t): Define this type.
- (ptw32_condattr_t): Likewise.
-
- * pthread.h (pthread_mutex_t): Define this type.
- (pthread_condattr_t): Likewise.
- (pthread_mutex_destroy): Add function prototype.
- (pthread_lock): Likewise.
- (pthread_trylock): Likewise.
- (pthread_unlock): Likewise.
- (pthread_condattr_init): Likewise.
- (pthread_condattr_destroy): Likewise.
- (pthread_condattr_setpshared): Likewise.
- (pthread_condattr_getpshared): Likewise.
-
- * mutex.c (pthread_mutex_init): Implement.
- (pthread_mutex_destroy): Likewise.
- (pthread_lock): Likewise.
- (pthread_trylock): Likewise.
- (pthread_unlock): Likewise.
-
-1998-07-12 Ben Elliston <bje at cygnus.com>
-
- * implement.h (ptw32_mutexattr_t): Define this implementation
- internal type. Application programmers only see a mutex attribute
- object as a void pointer.
-
- * pthread.h (pthread_mutexattr_t): Define this type.
- (pthread_mutexattr_init): Add function prototype.
- (pthread_mutexattr_destroy): Likewise.
- (pthread_mutexattr_setpshared): Likewise.
- (pthread_mutexattr_getpshared): Likewise.
- (pthread_mutexattr_setprotocol): Likewise.
- (pthread_mutexattr_getprotocol): Likewise.
- (pthread_mutexattr_setprioceiling): Likewise.
- (pthread_mutexattr_getprioceiling): Likewise.
- (PTHREAD_PROCESS_PRIVATE): Define.
- (PTHREAD_PROCESS_SHARED): Define.
-
- * mutex.c (pthread_mutexattr_init): Implement.
- (pthread_mutexattr_destroy): Implement.
- (pthread_mutexattr_setprotocol): Implement.
- (pthread_mutexattr_getprotocol): Likewise.
- (pthread_mutexattr_setprioceiling): Likewise.
- (pthread_mutexattr_getprioceiling): Likewise.
- (pthread_mutexattr_setpshared): Likewise.
- (pthread_mutexattr_getpshared): Likewise.
- (insert_attr): New function; very preliminary implementation!
- (is_attr): Likewise.
- (remove_attr): Likewise.
-
-Sat Jul 11 14:48:54 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
-
- * implement.h: Preliminary implementation specific defines.
-
- * create.c (pthread_create): Preliminary implementation.
-
-1998-07-11 Ben Elliston <bje at cygnus.com>
-
- * sync.c (pthread_join): Implement.
-
- * misc.c (pthread_equal): Likewise.
-
- * pthread.h (pthread_join): Add function prototype.
- (pthread_equal): Likewise.
-
-1998-07-10 Ben Elliston <bje at cygnus.com>
-
- * misc.c (pthread_self): Implement.
-
- * exit.c (pthread_exit): Implement.
-
- * pthread.h (pthread_exit): Add function prototype.
- (pthread_self): Likewise.
- (pthread_t): Define this type.
-
-1998-07-09 Ben Elliston <bje at cygnus.com>
-
- * create.c (pthread_create): A dummy stub right now.
-
- * pthread.h (pthread_create): Add function prototype.
+2004-10-15 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * implement.h (pthread_mutex_t_): Use an event in place of
+ the POSIX semaphore.
+ * pthread_mutex_init.c: Create the event; remove semaphore init.
+ * pthread_mutex_destroy.c: Delete the event.
+ * pthread_mutex_lock.c: Replace the semaphore wait with the event wait.
+ * pthread_mutex_trylock.c: Likewise.
+ * pthread_mutex_timedlock.c: Likewise.
+ * pthread_mutex_unlock.c: Set the event.
+
+2004-10-14 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread_mutex_lock.c (pthread_mutex_lock): New algorithm using
+ Terekhov's xchg based variation of Drepper's cmpxchg model.
+ Theoretically, xchg uses fewer clock cycles than cmpxchg (using IA-32
+ as a reference), however, in my opinion bus locking dominates the
+ equation on smp systems, so the model with the least number of bus
+ lock operations in the execution path should win, which is Terekhov's
+ variant. On IA-32 uni-processor systems, it's faster to use the
+ CMPXCHG instruction without locking the bus than to use the XCHG
+ instruction, which always locks the bus. This makes the two variants
+ equal for the non-contended lock (fast lane) execution path on up
+ IA-32. Testing shows that the xchg variant is faster on up IA-32 as
+ well if the test forces higher lock contention frequency, even though
+ kernel calls should be dominating the times (on up IA-32, both
+ variants used CMPXCHG instructions and neither locked the bus).
+ * pthread_mutex_timedlock.c pthread_mutex_timedlock(): Similarly.
+ * pthread_mutex_trylock.c (pthread_mutex_trylock): Similarly.
+ * pthread_mutex_unlock.c (pthread_mutex_unlock): Similarly.
+ * ptw32_InterlockedCompareExchange.c (ptw32_InterlockExchange): New
+ function.
+ (PTW32_INTERLOCKED_EXCHANGE): Sets up macro to use inlined
+ ptw32_InterlockedExchange.
+ * implement.h (PTW32_INTERLOCKED_EXCHANGE): Set default to
+ InterlockedExchange().
+ * Makefile: Building using /Ob2 so that asm sections within inline
+ functions are inlined.
+
+2004-10-08 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread_mutex_destroy.c (pthread_mutex_destroy): Critical Section
+ element is no longer required.
+ * pthread_mutex_init.c (pthread_mutex_init): Likewise.
+ * pthread_mutex_lock.c (pthread_mutex_lock): New algorithm following
+ Drepper's paper at http://people.redhat.com/drepper/futex.pdf, but
+ using the existing semaphore in place of the futex described in the
+ paper. Idea suggested by Alexander Terekhov - see:
+ http://sources.redhat.com/ml/pthreads-win32/2003/msg00108.html
+ * pthread_mutex_timedlock.c pthread_mutex_timedlock(): Similarly.
+ * pthread_mutex_trylock.c (pthread_mutex_trylock): Similarly.
+ * pthread_mutex_unlock.c (pthread_mutex_unlock): Similarly.
+ * pthread_barrier_wait.c (pthread_barrier_wait): Use inlined version
+ of InterlockedCompareExchange() if possible - determined at
+ build-time.
+ * pthread_spin_destroy.c pthread_spin_destroy(): Likewise.
+ * pthread_spin_lock.c pthread_spin_lock():Likewise.
+ * pthread_spin_trylock.c (pthread_spin_trylock):Likewise.
+ * pthread_spin_unlock.c (pthread_spin_unlock):Likewise.
+ * ptw32_InterlockedCompareExchange.c: Sets up macro for inlined use.
+ * implement.h (pthread_mutex_t_): Remove Critical Section element.
+ (PTW32_INTERLOCKED_COMPARE_EXCHANGE): Set to default non-inlined
+ version of InterlockedCompareExchange().
+ * private.c: Include ptw32_InterlockedCompareExchange.c first for
+ inlining.
+ * GNUmakefile: Add commandline option to use inlined
+ InterlockedCompareExchange().
+ * Makefile: Likewise.
+
+2004-09-27 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread_mutex_lock.c (pthread_mutex_lock): Separate
+ PTHREAD_MUTEX_NORMAL logic since we do not need to keep or check some
+ state required by other mutex types; do not check mutex pointer arg
+ for validity - leave this to the system since we are only checking
+ for NULL pointers. This should improve speed of NORMAL mutexes and
+ marginally improve speed of other type.
+ * pthread_mutex_trylock.c (pthread_mutex_trylock): Likewise.
+ * pthread_mutex_unlock.c (pthread_mutex_unlock): Likewise; also avoid
+ entering the critical section for the no-waiters case, with approx.
+ 30% reduction in lock/unlock overhead for this case.
+ * pthread_mutex_timedlock.c (pthread_mutex_timedlock): Likewise; also
+ no longer keeps mutex if post-timeout second attempt succeeds - this
+ will assist applications that wish to impose strict lock deadlines,
+ rather than simply to escape from frozen locks.
+
+2004-09-09 Tristan Savatier <tristan at mpegtv.com>
+ * pthread.h (struct pthread_once_t_): Qualify the 'done' element
+ as 'volatile'.
+ * pthread_once.c: Concerned about possible race condition,
+ specifically on MPU systems re concurrent access to multibyte types.
+ [Maintainer's note: the race condition is harmless on SPU systems
+ and only a problem on MPU systems if concurrent access results in an
+ exception (presumably generated by a hardware interrupt). There are
+ other instances of similar harmless race conditions that have not
+ been identified as issues.]
+
+2004-09-09 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread.h: Declare additional types as volatile.
+
+2004-08-27 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread_barrier_wait.c (pthread_barrier_wait): Remove excessive code
+ by substituting the internal non-cancelable version of sem_wait
+ (ptw32_semwait).
+
+2004-08-25 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread_join.c (pthread_join): Rewrite and re-order the conditional
+ tests in an attempt to improve efficiency and remove a race
+ condition.
+
+2004-08-23 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * create.c (pthread_create): Don't create a thread if the thread
+ id pointer location (first arg) is inaccessible. A memory
+ protection fault will result if the thread id arg isn't an accessible
+ location. This is consistent with GNU/Linux but different to
+ Solaris or MKS (and possibly others), which accept NULL as meaning
+ 'don't return the created thread's ID'. Applications that run
+ using pthreads-win32 will run on all other POSIX threads
+ implementations, at least w.r.t. this feature.
+
+ It was decided not to copy the Solaris et al behaviour because,
+ although it would have simplified some application porting (but only
+ from Solaris to Windows), the feature is not technically necessary,
+ and the alternative segfault behaviour helps avoid buggy application
+ code.
+
+2004-07-01 Anuj Goyal <anuj.goyal at gmail.com>
+
+ * builddmc.bat: New; Windows bat file to build the library.
+ * config.h (__DMC__): Support for Digital Mars compiler.
+ * create.c (__DMC__): Likewise.
+ * pthread_exit.c (__DMC__): Likewise.
+ * pthread_join.c (__DMC__): Likewise.
+ * ptw32_threadDestroy.c (__DMC__): Likewise.
+ * ptw32_threadStart.c (__DMC__): Likewise.
+ * ptw32_throw.c (__DMC__): Likewise.
+
+2004-06-29 Anuj Goyal <anuj.goyal at gmail.com>
+
+ * pthread.h (__DMC__): Initial support for Digital Mars compiler.
+
+2004-06-29 Will Bryant <will.bryant at ecosm.com>
+
+ * README.Borland: New; description of Borland changes.
+ * Bmakefile: New makefile for the Borland make utility.
+ * ptw32_InterlockedCompareExchange.c:
+ Add Borland compatible asm code.
+
+2004-06-26 Jason Bard <BardJA at Npt.NUWC.Navy.Mil>
+
+ * pthread.h (HAVE_STRUCT_TIMESPEC): If undefined, define it
+ to avoid timespec struct redefined errors elsewhere in an
+ application.
+
+2004-06-21 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread.h (PTHREAD_RECURSIVE_MUTEX_INITIALIZER): Mutex
+ initialiser added for compatibility with Linux threads and
+ others; currently not included in SUSV3.
+ * pthread.h (PTHREAD_ERRORCHECK_MUTEX_INITIALIZER): Likewise.
+ * pthread.h (PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP): Likewise.
+ * pthread.h (PTHREAD_ERRORCHECK_MUTEX_INITIALIZER_NP): Likewise.
+
+ * ptw32_mutex_check_need_init.c (ptw32_mutex_check_need_init):
+ Add new initialisers.
+
+ * pthread_mutex_lock.c (pthread_mutex_lock): Check for new
+ initialisers.
+ * pthread_mutex_trylock.c (pthread_mutex_trylock): Likewise.
+ * pthread_mutex_timedlock.c (pthread_mutex_timedlock): Likewise.
+ * pthread_mutex_unlock.c (pthread_mutex_unlock): Likewise.
+ * pthread_mutex_destroy.c (pthread_mutex_destroy): Likewise.
+
+2004-05-20 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * README.NONPORTABLE: Document pthread_win32_test_features_np().
+ * FAQ: Update various answers.
+
+2004-05-19 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * Makefile: Don't define _WIN32_WINNT on compiler command line.
+ * GNUmakefile: Likewise.
+
+2004-05-16 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread_cancel.c (pthread_cancel): Adapted to use auto-detected
+ QueueUserAPCEx features at run-time.
+ (ptw32_RegisterCancelation): Drop in replacement for QueueUserAPCEx()
+ if it can't be used. Provides older style non-preemptive async
+ cancelation.
+ * pthread_win32_attach_detach_np.c (pthread_win32_attach_np):
+ Auto-detect quserex.dll and the availability of alertdrv.sys;
+ initialise and close on process attach/detach.
+ * global.c (ptw32_register_cancelation): Pointer to either
+ QueueUserAPCEx() or ptw32_RegisterCancelation() depending on
+ availability. QueueUserAPCEx makes pre-emptive async cancelation
+ possible.
+ * implement.h: Add definitions and prototypes related to QueueUserAPC.
+
+2004-05-16 Panagiotis E. Hadjidoukas <peh at hpclab.ceid.upatras.gr>
+
+ * QueueUserAPCEx (separate contributed package): Provides preemptive
+ APC feature.
+ * pthread_cancel.c (pthread_cancel): Initial integration of
+ QueueUserAPCEx into pthreads-win32 to provide true pre-emptive
+ async cancelation of threads, including blocked threads.
+
+2004-05-06 Makoto Kato <raven at oldskool.jp>
+
+ * pthread.h (DWORD_PTR): Define typedef for older MSVC.
+ * pthread_cancel.c (AMD64): Add architecture specific Context register.
+ * ptw32_getprocessors.c: Use correct types (DWORD_PTR) for mask
+ variables.
+
+2004-04-06 P. van Bruggen <pietvb at newbridges.nl>
+
+ * ptw32_threadDestroy.c: Destroy threadLock mutex to
+ close a memory leak.
+
+2004-02-13 Gustav Hallberg <gustav at virtutech.com>
+
+ * pthread_equal.c: Remove redundant equality logic.
+
+2003-12-10 Philippe Di Cristo <philipped at voicebox.com>
+
+ * sem_timedwait.c (sem_timedwait): Fix timeout calculations.
+
+2003-10-20 Alexander Terekhov <TEREKHOV at de.ibm.com>
+
+ * pthread_mutex_timedlock.c (ptw32_semwait): Move to individual module.
+ * ptw32_semwait.c: New module.
+ * pthread_cond_wait.c (ptw32_cond_wait_cleanup): Replace cancelable
+ sem_wait() call with non-cancelable ptw32_semwait() call.
+ * pthread.c (private.c): Re-order for inlining. GNU C warned that
+ function ptw32_semwait() was defined 'inline' after it was called.
+ * pthread_cond_signal.c (ptw32_cond_unblock): Likewise.
+ * pthread_delay_np.c: Disable Watcom warning with comment.
+ * *.c (process.h): Remove include from .c files. This is conditionally
+ included by the common project include files.
+
+2003-10-20 James Ewing <james.ewing at sveasoft.com>
+
+ * ptw32_getprocessors.c: Some Win32 environments don't have
+ GetProcessAffinityMask(), so always return CPU count = 1 for them.
+ * config.h (NEED_PROCESSOR_AFFINITY_MASK): Define for WinCE.
+
+2003-10-15 Ross Johnson <ross at callisto.canberra.edu.au>
+
+ * Re-indented all .c files using default GNU style to remove assorted
+ editor ugliness (used GNU indent utility in default style).
+
+2003-10-15 Alex Blanco <Alex.Blanco at motorola.com>
+
+ * sem_init.c (sem_init): Would call CreateSemaphore even if the sema
+ struct calloc failed; was not freeing calloced memory if either
+ CreateSemaphore or CreateEvent failed.
+
+2003-10-14 Ross Johnson <ross at callisto.canberra.edu.au>
+
+ * pthread.h: Add Watcom compiler compatibility. Esssentially just add
+ the cdecl attribute to all exposed function prototypes so that Watcom
+ generates function call code compatible with non-Watcom built libraries.
+ By default, Watcom uses registers to pass function args if possible rather
+ than pushing to stack.
+ * semaphore.h: Likewise.
+ * sched.h: Likewise.
+ * pthread_cond_wait.c (ptw32_cond_wait_cleanup): Define with cdecl attribute
+ for Watcom compatibility. This routine is called via pthread_cleanup_push so
+ it had to match function arg definition.
+ * Wmakefile: New makefile for Watcom builds.
+
+2003-09-14 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread_setschedparam.c (pthread_setschedparam): Attempt to map
+ all priority levels between max and min (as returned by
+ sched_get_priority_min/max) to reasonable Win32 priority levels - i.e.
+ levels between THREAD_PRIORITY_LOWEST/IDLE to THREAD_PRIORITY_LOWEST and
+ between THREAD_PRIORITY_HIGHEST/TIME_CRITICAL to THREAD_PRIORITY_HIGHEST
+ while others remain unchanged; record specified thread priority level
+ for return by pthread_getschedparam.
+
+ Note that, previously, specified levels not matching Win32 priority levels
+ would silently leave the current thread priority unaltered.
+
+ * pthread_getschedparam.c (pthread_getschedparam): Return the priority
+ level specified by the latest pthread_setschedparam or pthread_create rather
+ than the actual running thread priority as returned by GetThreadPriority - as
+ required by POSIX. I.e. temporary or adjusted actual priority levels are not
+ returned by this routine.
+
+ * pthread_create.c (pthread_create): For priority levels specified via
+ pthread attributes, attempt to map all priority levels between max and
+ min (as returned by sched_get_priority_min/max) to reasonable Win32
+ priority levels; record priority level given via attributes, or
+ inherited from parent thread, for later return by pthread_getschedparam.
+
+ * ptw32_new.c (ptw32_new): Initialise pthread_t_ sched_priority element.
+
+ * pthread_self.c (pthread_self): Set newly created implicit POSIX thread
+ sched_priority to Win32 thread's current actual priority. Temporarily
+ altered priorities can't be avoided in this case.
+
+ * implement.h (struct pthread_t_): Add new sched_priority element.
+
+2003-09-12 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * sched_get_priority_min.c (sched_get_priority_min): On error should return -1
+ with errno set.
+ * sched_get_priority_max.c (sched_get_priority_max): Likewise.
+
+2003-09-03 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * w32_cancelableWait.c (ptw32_cancelable_wait): Allow cancelation
+ of implicit POSIX threads as well.
+
+2003-09-02 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread_win32_attach_detach_np.c (pthread_win32_thread_detach_np):
+ Add comment.
+
+ * pthread_exit.c (pthread_exit): Fix to recycle the POSIX thread handle in
+ addition to calling user TSD destructors. Move the implicit POSIX thread exit
+ handling to ptw32_throw to centralise the logic.
+
+ * ptw32_throw.c (ptw32_throw): Implicit POSIX threads have no point
+ to jump or throw to, so cleanup and exit the thread here in this case. For
+ processes using the C runtime, the exit code will be set to the POSIX
+ reason for the throw (i.e. PTHREAD_CANCEL or the value given to pthread_exit).
+ Note that pthread_exit() already had similar logic, which has been moved to
+ here.
+
+ * ptw32_threadDestroy.c (ptw32_threadDestroy): Don't close the Win32 handle
+ of implicit POSIX threads - expect this to be done by Win32?
+
+2003-09-01 Ross Johnson <rpj at callisto.canberra.edu.au>
+
+ * pthread_self.c (pthread_self): The newly aquired pthread_t must be
+ assigned to the reuse stack, not freed, if the routine fails somehow.
+
+2003-08-13 Ross Johnson <rpj at ise.canberra.edu.au>
+
+ * pthread_getschedparam.c (pthread_getschedparam): An invalid thread ID
+ parameter was returning an incorrect error value; now uses a more exhaustive
+ check for validity.
+
+ * pthread_setschedparam.c (pthread_setschedparam): Likewise.
+
+ * pthread_join.c (pthread_join): Now uses a more exhaustive
+ check for validity.
+
+ * pthread_detach.c (pthread_detach): Likewise.
+
+ * pthread_cancel.c (pthread_cancel): Likewise.
+
+ * ptw32_threadDestroy.c (ptw32_threadDestroy): pthread_t structs are
+ never freed - push them onto a stack for reuse.
+
+ * ptw32_new.c (ptw32_new): Check for reusable pthread_t before dynamically
+ allocating new memory for the struct.
+
+ * pthread_kill.c (pthread_kill): New file; new routine; takes only a zero
+ signal arg so that applications can check the thread arg for validity; checks
+ that the underlying Win32 thread HANDLE is valid.
+
+ * pthread.h (pthread_kill): Add prototype.
+
+ * ptw32_reuse.c (ptw32_threadReusePop): New file; new routine; pop a
+ pthread_t off the reuse stack. pthread_t_ structs that have been destroyed, i.e.
+ have exited detached or have been joined, are cleaned up and put onto a reuse
+ stack. Consequently, thread IDs are no longer freed once calloced. The library
+ will attempt to get a struct off this stack before asking the system to alloc
+ new memory when creating threads. The stack is guarded by a global mutex.
+ (ptw32_threadReusePush): New routine; push a pthread_t onto the reuse stack.
+
+ * implement.h (ptw32_threadReusePush): Add new prototype.
+ (ptw32_threadReusePop): Likewise.
+ (pthread_t): Add new element.
+
+ * ptw32_processTerminate.c (ptw32_processTerminate): Delete the thread
+ reuse lock; free all thread ID structs on the thread reuse stack.
+
+ * ptw32_processInitialize.c (ptw32_processInitialize): Initialise the
+ thread reuse lock.
+
+2003-07-19 Ross Johnson <rpj at ise.canberra.edu.au>
+
+ * GNUmakefile: modified to work under MsysDTK environment.
+ * pthread_spin_lock.c (pthread_spin_lock): Check for NULL arg.
+ * pthread_spin_unlock.c (pthread_spin_unlock): Likewise.
+ * pthread_spin_trylock.c (pthread_spin_trylock): Likewise;
+ fix incorrect pointer value if lock is dynamically initialised by
+ this function.
+ * sem_init.c (sem_init): Initialise sem_t value to quell compiler warning.
+ * sem_destroy.c (sem_destroy): Likewise.
+ * ptw32_threadStart.c (non-MSVC code sections): Include <exception> rather
+ than old-style <new.h>; fix all std:: namespace entities such as
+ std::terminate_handler instances and associated methods.
+ * ptw32_callUserDestroyRoutines.c (non-MSVC code sections): Likewise.
+
+2003-06-24 Piet van Bruggen <pietvb at newbridges.nl>
+
+ * pthread_spin_destroy.c (pthread_spin_destroy): Was not freeing the
+ spinlock struct.
+
+2003-06-22 Nicolas Barry <boozai at yahoo.com>
+
+ * pthread_mutex_destroy.c (pthread_mutex_destroy): When called
+ with a recursive mutex that was locked by the current thread, the
+ function was failing with a success return code.
+
+2003-05-15 Steven Reddie <Steven.Reddie at ca.com>
+
+ * pthread_win32_attach_detach_np.c (pthread_win32_process_detach_np):
+ NULLify ptw32_selfThreadKey after the thread is destroyed, otherwise
+ destructors calling pthreads routines might resurrect it again, creating
+ memory leaks. Call the underlying Win32 Tls routine directly rather than
+ pthread_setspecific().
+ (pthread_win32_thread_detach_np): Likewise.
+
+2003-05-14 Viv <vcotirlea at hotmail.com>
+
+ * pthread.dsp: Change /MT compile flag to /MD.
+
+2003-03-04 Alexander Terekhov <TEREKHOV at de.ibm.com>
+
+ * pthread_mutex_timedlock.c (pthread_mutex_timedlock): Fix failure to
+ set ownership of mutex on second grab after abstime timeout.
+ - bug reported by Robert Strycek <strycek at posam.sk>
+
+2002-12-17 Thomas Pfaff <tpfaff at gmx.net>
+
+ * pthread_mutex_lock.c (ptw32_semwait): New static routine to provide
+ a non-cancelable sem_wait() function. This is consistent with the
+ way that pthread_mutex_timedlock.c does it.
+ (pthread_mutex_lock): Use ptw32_semwait() instead of sem_wait().
+
+2002-12-11 Thomas Pfaff <tpfaff at gmx.net>
+
+ * pthread_mutex_trylock.c: Should return EBUSY rather than EDEADLK.
+ * pthread_mutex_destroy.c: Remove redundant ownership test (the
+ trylock call does this for us); do not destroy a recursively locked
+ mutex.
+
+2002-09-20 Michael Johnson <michaelj at maine.rr.com>
+
+ * pthread_cond_destroy.c (pthread_cond_destroy):
+ When two different threads exist, and one is attempting to
+ destroy a condition variable while the other is attempting to
+ initialize a condition variable that was created with
+ PTHREAD_COND_INITIALIZER, a deadlock can occur. Shrink
+ the ptw32_cond_list_lock critical section to fix it.
+
+2002-07-31 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * ptw32_threadStart.c (ptw32_threadStart): Thread cancelLock
+ destruction moved to ptw32_threadDestroy().
+
+ * ptw32_threadDestroy.c (ptw32_threadDestroy): Destroy
+ the thread's cancelLock. Moved here from ptw32_threadStart.c
+ to cleanup implicit threads as well.
+
+2002-07-30 Alexander Terekhov <TEREKHOV at de.ibm.com>
+
+ * pthread_cond_wait.c (ptw32_cond_wait_cleanup):
+ Remove code designed to avoid/prevent spurious wakeup
+ problems. It is believed that the sem_timedwait() call
+ is consuming a CV signal that it shouldn't and this is
+ breaking the avoidance logic.
+
+2002-07-30 Ross Johnson <rpj at ise.canberra.edu.au>
+
+ * sem_timedwait.c (sem_timedwait): Tighten checks for
+ unreasonable abstime values - that would result in
+ unexpected timeout values.
+
+ * w32_CancelableWait.c (ptw32_cancelable_wait):
+ Tighten up return value checking and add comments.
+
+
+2002-06-08 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * sem_getvalue.c (sem_getvalue): Now returns a value for the
+ NEED_SEM version (i.e. earlier versions of WinCE).
+
+
+2002-06-04 Rob Fanner <rfanner at stonethree.com>
+
+ * sem_getvalue.c (sem_getvalue): The Johnson M. Hart
+ approach didn't work - we are forced to take an
+ intrusive approach. We try to decrement the sema
+ and then immediately release it again to get the
+ value. There is a small probability that this may
+ block other threads, but only momentarily.
+
+2002-06-03 Ross Johnson <rpj at ise.canberra.edu.au>
+
+ * sem_init.c (sem_init): Initialise Win32 semaphores
+ to _POSIX_SEM_VALUE_MAX (which this implementation
+ defines in pthread.h) so that sem_getvalue() can use
+ the trick described in the comments in sem_getvalue().
+ * pthread.h (_POSIX_SEM_VALUE_MAX): Defined.
+ (_POSIX_SEM_NSEMS_MAX): Defined - not used but may be
+ useful for source code portability.
+
+2002-06-03 Rob Fanner <rfanner at stonethree.com>
+
+ * sem_getvalue.c (sem_getvalue): Did not work on NT.
+ Use approach suggested by Johnson M. Hart in his book
+ "Win32 System Programming".
+
+2002-02-28 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * errno.c: Compiler directive was incorrectly including code.
+ * pthread.h: Conditionally added some #defines from config.h
+ needed when not building the library. e.g. NEED_ERRNO, NEED_SEM.
+ (PTW32_DLLPORT): Now only defined if _DLL defined.
+ (_errno): Compiler directive was incorrectly including prototype.
+ * sched.h: Conditionally added some #defines from config.h
+ needed when not building the library.
+ * semaphore.h: Replace an instance of NEED_SEM that should
+ have been NEED_ERRNO. This change currently has nil effect.
+
+ * GNUmakefile: Correct some recent changes.
+
+ * Makefile: Add rule to generate pre-processor output.
+
+2002-02-23 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * pthread_rwlock_timedrdlock.c: New - untested.
+ * pthread_rwlock_timedwrlock.c: New - untested.
+
+ * Testsuite passed (except known MSVC++ problems)
+
+ * pthread_cond_destroy.c: Expand the time change
+ critical section to solve deadlock problem.
+
+ * pthread.c: Add all remaining C modules.
+ * pthread.h: Use dllexport/dllimport attributes on functions
+ to avoid using pthread.def.
+ * sched.h: Likewise.
+ * semaphore.h: Likewise.
+ * GNUmakefile: Add new targets for single translation
+ unit build to maximise inlining potential; generate
+ pthread.def automatically.
+ * Makefile: Likewise, but no longer uses pthread.def.
+
+2002-02-20 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * pthread_cond_destroy.c (pthread_cond_destroy):
+ Enter the time change critical section earlier.
+
+2002-02-17 Ross Johnson <rpj at setup1.ise.canberra.edu.au
+
+ * Testsuite passed.
+
+ * pthread_timechange_handler_np.c: New; following
+ a suggestion from Alexander Terekhov that CVs should
+ be broadcast so that they all re-evaluate their
+ condition variables and reset a new timeout if
+ required, whenever an application receives a
+ WM_TIMECHANGE message. This message indicates that
+ the system time has been changed. Therefore, CVs
+ waiting for a timeout set as an abs_time will possibly
+ not wake up at the expected time. Some applications
+ may not be tolerant of this.
+ * pthread_cond_init.c: Add CV to linked list.
+ * pthread_cond_destroy.c: Remove CV from linked list.
+ * global.c (ptw32_cond_list_head): New variable.
+ (ptw32_cond_list_tail): New variable.
+ (ptw32_cond_list_cs): New critical section.
+ * ptw32_processInitialize (ptw32_cond_list_cs): Initialize.
+ * ptw32_processTerminate (ptw32_cond_list_cs): Delete.
+
+
+ * Reduce executable size.
+ -----------------------
+ When linking with the static library, only those
+ routines actually called, either directly or indirectly
+ should be included.
+
+ [Gcc has the -ffunction-segments option to do this but MSVC
+ doesn't have this feature as far as I can determine. Other
+ compilers are undetermined as well. - rpj]
+
+ * spin.c: Split file into function segments.
+ * ptw32_spinlock_check_need_init.c: Separated routine from spin.c.
+ * pthread_spin_init.c: Likewise.
+ * pthread_spin_destroy.c: Likewise.
+ * pthread_spin_lock.c: Likewise.
+ * pthread_spin_unlock.c: Likewise.
+ * pthread_spin_trylock.c: Likewise.
+
+ * sync.c: Split file into function segments.
+ * pthread_detach.c: Separated routine from sync.c.
+ * pthread_join.c: Likewise.
+
+ * tsd.c: Split file into function segments.
+ * pthread_key_create.c: Separated routine from tsd.c.
+ * pthread_key_delete.c: Likewise.
+ * pthread_setspecific.c: Likewise.
+ * pthread_getspecific.c: Likewise.
+
+ * sched.c: Split file into function segments.
+ * pthread_attr_setschedpolicy.c: Separated routine from sched.c.
+ * pthread_attr_getschedpolicy.c: Likewise.
+ * pthread_attr_setschedparam.c: Likewise.
+ * pthread_attr_getschedparam.c: Likewise.
+ * pthread_attr_setinheritsched.c: Likewise.
+ * pthread_attr_getinheritsched.c: Likewise.
+ * pthread_setschedparam.c: Likewise.
+ * pthread_getschedparam.c: Likewise.
+ * sched_get_priority_max.c: Likewise.
+ * sched_get_priority_min.c: Likewise.
+ * sched_setscheduler.c: Likewise.
+ * sched_getscheduler.c: Likewise.
+ * sched_yield.c: Likewise.
+
+
+2002-02-16 Ross Johnson <rpj at setup1.ise.canberra.edu.au
+
+ Reduce executable size.
+ -----------------------
+ When linking with the static library, only those
+ routines actually called, either directly or indirectly
+ should be included.
+
+ [Gcc has the -ffunction-segments option to do this but MSVC
+ doesn't have this feature as far as I can determine. Other
+ compilers are undetermined as well. - rpj]
+
+ * mutex.c: Split file into function segments.
+ * pthread_mutexattr_destroy.c: Separated routine from mutex.c
+ * pthread_mutexattr_getpshared.c: Likewise.
+ * pthread_mutexattr_gettype.c: Likewise.
+ * pthread_mutexattr_init.c: Likewise.
+ * pthread_mutexattr_setpshared.c: Likewise.
+ * pthread_mutexattr_settype.c: Likewise.
+ * ptw32_mutex_check_need_init.c: Likewise.
+ * pthread_mutex_destroy.c: Likewise.
+ * pthread_mutex_init.c: Likewise.
+ * pthread_mutex_lock.c: Likewise.
+ * pthread_mutex_timedlock.c: Likewise.
+ * pthread_mutex_trylock.c: Likewise.
+ * pthread_mutex_unlock.c: Likewise.
+
+ * private.c: Split file into function segments.
+ * ptw32_InterlockedCompareExchange.c: Separated routine from private.c
+ * ptw32_callUserDestroyRoutines.c: Likewise.
+ * ptw32_getprocessors.c: Likewise.
+ * ptw32_processInitialize.c: Likewise.
+ * ptw32_processTerminate.c: Likewise.
+ * ptw32_threadDestroy.c: Likewise.
+ * ptw32_threadStart.c: Likewise.
+ * ptw32_throw.c: Likewise.
+ * ptw32_timespec.c: Likewise.
+ * ptw32_tkAssocCreate.c: Likewise.
+ * ptw32_tkAssocDestroy.c: Likewise.
+
+ * rwlock.c: Split file into function segments.
+ * pthread_rwlockattr_destroy.c: Separated routine from rwlock.c
+ * pthread_rwlockattr_getpshared.c: Likewise.
+ * pthread_rwlockattr_init.c: Likewise.
+ * pthread_rwlockattr_setpshared.c: Likewise.
+ * ptw32_rwlock_check_need_init.c: Likewise.
+ * pthread_rwlock_destroy.c: Likewise.
+ * pthread_rwlock_init.c: Likewise.
+ * pthread_rwlock_rdlock.c: Likewise.
+ * pthread_rwlock_tryrdlock.c: Likewise.
+ * pthread_rwlock_trywrlock.c: Likewise.
+ * pthread_rwlock_unlock.c: Likewise.
+ * pthread_rwlock_wrlock.c: Likewise.
+
+2002-02-10 Ross Johnson <rpj at setup1.ise.canberra.edu.au
+
+ Reduce executable size.
+ -----------------------
+ When linking with the static library, only those
+ routines actually called, either directly or indirectly
+ should be included.
+
+ [Gcc has the -ffunction-segments option to do this but MSVC
+ doesn't have this feature as far as I can determine. Other
+ compilers are undetermined as well. - rpj]
+
+ * nonportable.c: Split file into function segments.
+ * np_delay.c: Separated routine from nonportable.c
+ * np_getw32threadhandle.c: Likewise.
+ * np_mutexattr_setkind.c: Likewise.
+ * np_mutexattr_getkind.c: Likewise.
+ * np_num_processors.c: Likewise.
+ * np_win32_attach_detach.c: Likewise.
+
+ * misc.c: Split file into function segments.
+ * pthread_equal.c: Separated routine from nonportable.c.
+ * pthread_getconcurrency.c: Likewise.
+ * pthread_once.c: Likewise.
+ * pthread_self.c: Likewise.
+ * pthread_setconcurrency.c: Likewise.
+ * ptw32_calloc.c: Likewise.
+ * ptw32_new.c: Likewise.
+ * w32_CancelableWait.c: Likewise.
+
+2002-02-09 Ross Johnson <rpj at setup1.ise.canberra.edu.au
+
+ Reduce executable size.
+ -----------------------
+ When linking with the static library, only those
+ routines actually called, either directly or indirectly
+ should be included.
+
+ [Gcc has the -ffunction-segments option to do this but MSVC
+ doesn't have this feature as far as I can determine. Other
+ compilers are undetermined as well. - rpj]
+
+ * condvar.c: Split file into function segments.
+ * pthread_condattr_destroy.c: Separated routine from condvar.c.
+ * pthread_condattr_getpshared.c: Likewise.
+ * pthread_condattr_init.c: Likewise.
+ * pthread_condattr_setpshared.c: Likewise.
+ * ptw32_cond_check_need_init.c: Likewise.
+ * pthread_cond_destroy.c: Likewise.
+ * pthread_cond_init.c: Likewise.
+ * pthread_cond_signal.c: Likewise.
+ * pthread_cond_wait.c: Likewise.
+
+2002-02-07 Alexander Terekhov<TEREKHOV at de.ibm.com>
+
+ * nonportable.c (pthread_delay_np): Make a true
+ cancelation point. Deferred cancels will interrupt the
+ wait.
+
+2002-02-07 Ross Johnson <rpj at setup1.ise.canberra.edu.au
+
+ * misc.c (ptw32_new): Add creation of cancelEvent so that
+ implicit POSIX threads (Win32 threads with a POSIX face)
+ are cancelable; mainly so that pthread_delay_np doesn't fail
+ if called from the main thread.
+ * create.c (pthread_create): Remove creation of cancelEvent
+ from here; now in ptw32_new().
+
+ Reduce executable size.
+ -----------------------
+ When linking with the static library, only those
+ routines actually called, either directly or indirectly
+ should be included.
+
+ [Gcc has the -ffunction-segments option to do this but MSVC
+ doesn't have this feature as far as I can determine. Other
+ compilers are undetermined as well. - rpj]
+
+ * barrier.c: All routines are now in separate compilation units;
+ This file is used to congregate the separate modules for
+ potential inline optimisation and backward build compatibility.
+ * cancel.c: Likewise.
+ * pthread_barrierattr_destroy.c: Separated routine from cancel.c.
+ * pthread_barrierattr_getpshared.c: Likewise.
+ * pthread_barrierattr_init.c: Likewise.
+ * pthread_barrierattr_setpshared.c: Likewise.
+ * pthread_barrier_destroy.c: Likewise.
+ * pthread_barrier_init.c: Likewise.
+ * pthread_barrier_wait.c: Likewise.
+ * pthread_cancel.c: Likewise.
+ * pthread_setcancelstate.c: Likewise.
+ * pthread_setcanceltype.c: Likewise.
+ * pthread_testcancel.c: Likewise.
+
+2002-02-04 Max Woodbury <mtew at cds.duke.edu>
+
+ Reduced name space pollution.
+ -----------------------------
+ When the appropriate symbols are defined, the headers
+ will restrict the definitions of new names. In particular,
+ it must be possible to NOT include the <windows.h>
+ header and related definitions with some combination
+ of symbol definitions. Secondly, it should be possible
+ that additional definitions should be limited to POSIX
+ compliant symbols by the definition of appropriate symbols.
+
+ * pthread.h: POSIX conditionals.
+ * sched.h: POSIX conditionals.
+ * semaphore.h: POSIX conditionals.
+
+ * semaphore.c: Included <limits.h>.
+ (sem_init): Changed magic 0x7FFFFFFFL to INT_MAX.
+ (sem_getvalue): Trial version.
+
+ Reduce executable size.
+ -----------------------
+ When linking with the static library, only those
+ routines actually called, either directly or indirectly
+ should be included.
+
+ [Gcc has the -ffunction-segments option to do this but MSVC
+ doesn't have this feature as far as I can determine. Other
+ compilers are undetermined as well. - rpj]
+
+ * semaphore.c: All routines are now in separate compilation units;
+ This file is used to congregate the separate modules for
+ potential inline optimisation and backward build compatibility.
+ * sem_close.c: Separated routine from semaphore.c.
+ * ptw32_decrease_semaphore.c: Likewise.
+ * sem_destroy.c: Likewise.
+ * sem_getvalue.c: Likewise.
+ * ptw32_increase_semaphore.c: Likewise.
+ * sem_init.c: Likewise.
+ * sem_open.c: Likewise.
+ * sem_post.c: Likewise.
+ * sem_post_multiple.c: Likewise.
+ * sem_timedwait.c: Likewise.
+ * sem_trywait.c: Likewise.
+ * sem_unlink.c: Likewise.
+ * sem_wait.c: Likewise.
+
+2002-02-04 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ The following extends the idea above to the rest of pthreads-win32 - rpj
+
+ * attr.c: All routines are now in separate compilation units;
+ This file is used to congregate the separate modules for
+ potential inline optimisation and backward build compatibility.
+ * pthread_attr_destroy.c: Separated routine from attr.c.
+ * pthread_attr_getdetachstate.c: Likewise.
+ * pthread_attr_getscope.c: Likewise.
+ * pthread_attr_getstackaddr.c: Likewise.
+ * pthread_attr_getstacksize.c: Likewise.
+ * pthread_attr_init.c: Likewise.
+ * pthread_attr_is_attr.c: Likewise.
+ * pthread_attr_setdetachstate.c: Likewise.
+ * pthread_attr_setscope.c: Likewise.
+ * pthread_attr_setstackaddr.c: Likewise.
+ * pthread_attr_setstacksize.c: Likewise.
+
+ * pthread.c: Agregation of agregate modules for super-inlineability.
+
+2002-02-02 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * cancel.c: Rearranged some code and introduced checks
+ to disable cancelation at the start of a thread's cancelation
+ run to prevent double cancelation. The main problem
+ arises if a thread is canceling and then receives a subsequent
+ async cancel request.
+ * private.c: Likewise.
+ * condvar.c: Place pragmas around cleanup_push/pop to turn
+ off inline optimisation (/Obn where n>0 - MSVC only). Various
+ optimisation switches in MSVC turn this on, which interferes with
+ the way that cleanup handlers are run in C++ EH and SEH
+ code. Application code compiled with inline optimisation must
+ also wrap cleanup_push/pop blocks with the pragmas, e.g.
+ #pragma inline_depth(0)
+ pthread_cleanup_push(...)
+ ...
+ pthread_cleanup_pop(...)
+ #pragma inline_depth(8)
+ * rwlock.c: Likewise.
+ * mutex.c: Remove attempts to inline some functions.
+ * signal.c: Modify misleading comment.
+
+2002-02-01 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * semaphore.c (sem_trywait): Fix missing errno return
+ for systems that define NEED_SEM (e.g. early WinCE).
+ * mutex.c (pthread_mutex_timedlock): Return ENOTSUP
+ for systems that define NEED_SEM since they don't
+ have sem_trywait().
+
+2002-01-27 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * mutex.c (pthread_mutex_timedlock): New function suggested by
+ Alexander Terekhov. The logic required to implement this
+ properly came from Alexander, with some collaboration
+ with Thomas Pfaff.
+ (pthread_mutex_unlock): Wrap the waiters check and sema
+ post in a critical section to prevent a race with
+ pthread_mutex_timedlock.
+ (ptw32_timed_semwait): New function;
+ returns a special result if the absolute timeout parameter
+ represents a time already passed when called; used by
+ pthread_mutex_timedwait(). Have deliberately not reused
+ the name "ptw32_sem_timedwait" because they are not the same
+ routine.
+ * condvar.c (ptw32_cond_timedwait): Use the new sem_timedwait()
+ instead of ptw32_sem_timedwait(), which now has a different
+ function. See previous.
+ * implement.h: Remove prototype for ptw32_sem_timedwait.
+ See next.
+ (pthread_mutex_t_): Add critical section element for access
+ to lock_idx during mutex post-timeout processing.
+ * semaphore.h (sem_timedwait): See next.
+ * semaphore.c (sem_timedwait): See next.
+ * private.c (ptw32_sem_timedwait): Move to semaphore.c
+ and rename as sem_timedwait().
+
+2002-01-18 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * sync.c (pthread_join): Was getting the exit code from the
+ calling thread rather than the joined thread if
+ defined(__MINGW32__) && !defined(__MSVCRT__).
+
+2002-01-15 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * pthread.h: Unless the build explicitly defines __CLEANUP_SEH,
+ __CLEANUP_CXX, or __CLEANUP_C, then the build defaults to
+ __CLEANUP_C style cleanup. This style uses setjmp/longjmp
+ in the cancelation and thread exit implementations and therefore
+ won't do stack unwinding if linked to applications that have it
+ (e.g. C++ apps). This is currently consistent with most/all
+ commercial Unix POSIX threads implementations.
+
+ * spin.c (pthread_spin_init): Edit renamed function call.
+ * nonportable.c (pthread_num_processors_np): New.
+ (pthread_getprocessors_np): Renamed to ptw32_getprocessors
+ and moved to private.c.
+ * private.c (pthread_getprocessors): Moved here from
+ nonportable.c.
+ * pthread.def (pthread_getprocessors_np): Removed
+ from export list.
+
+ * rwlock.c (pthread_rwlockattr_init): New.
+ (pthread_rwlockattr_destroy): New.
+ (pthread_rwlockattr_getpshared): New.
+ (pthread_rwlockattr_setpshared): New.
+
+2002-01-14 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * attr.c (pthread_attr_setscope): Fix struct pointer
+ indirection error introduced 2002-01-04.
+ (pthread_attr_getscope): Likewise.
+
+2002-01-12 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * pthread.dsp (SOURCE): Add missing source files.
+
+2002-01-08 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * mutex.c (pthread_mutex_trylock): use
+ ptw32_interlocked_compare_exchange function pointer
+ rather than ptw32_InterlockedCompareExchange() directly
+ to retain portability to non-iX86 processors,
+ e.g. WinCE etc. The pointer will point to the native
+ OS version of InterlockedCompareExchange() if the
+ OS supports it (see ChangeLog entry of 2001-10-17).
+
+2002-01-07 Thomas Pfaff <tpfaff at gmx.net>, Alexander Terekhov <TEREKHOV at de.ibm.com>
+
+ * mutex.c (pthread_mutex_init): Remove critical
+ section calls.
+ (pthread_mutex_destroy): Likewise.
+ (pthread_mutex_unlock): Likewise.
+ (pthread_mutex_trylock): Likewise; uses
+ ptw32_InterlockedCompareExchange() to avoid need for
+ critical section; library is no longer i386 compatible;
+ recursive mutexes now increment the lock count rather
+ than return EBUSY; errorcheck mutexes return EDEADLCK
+ rather than EBUSY. This behaviour is consistent with the
+ Solaris pthreads implementation.
+ * implement.h (pthread_mutex_t_): Remove critical
+ section element - no longer needed.
+
+
+2002-01-04 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * attr.c (pthread_attr_setscope): Add more error
+ checking and actually store the scope value even
+ though it's not really necessary.
+ (pthread_attr_getscope): Return stored value.
+ * implement.h (pthread_attr_t_): Add new scope element.
+ * ANNOUNCE: Fix out of date comment next to
+ pthread_attr_setscope in conformance section.
+
+2001-12-21 Alexander Terekhov <TEREKHOV at de.ibm.com>
+
+ * mutex.c (pthread_mutex_lock): Decrementing lock_idx was
+ not thread-safe.
+ (pthread_mutex_trylock): Likewise.
+
+2001-10-26 prionx@juno.com
+
+ * semaphore.c (sem_init): Fix typo and missing bracket
+ in conditionally compiled code. Only older versions of
+ WinCE require this code, hence it doesn't normally get
+ tested; somehow when sem_t reverted to an opaque struct
+ the calloc NULL check was left in the conditionally included
+ section.
+ (sem_destroy): Likewise, the calloced sem_t wasn't being freed.
+
+2001-10-25 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * GNUmakefile (libwsock32): Add to linker flags for
+ WSAGetLastError() and WSASetLastError().
+ * Makefile (wsock32.lib): Likewise.
+ * create.c: Minor mostly inert changes.
+ * implement.h (PTW32_MAX): Move into here and renamed
+ from sched.h.
+ (PTW32_MIN): Likewise.
+ * GNUmakefile (TEST_ICE): Define if testing internal
+ implementation of InterlockedCompareExchange.
+ * Makefile (TEST_ICE): Likewise.
+ * private.c (TEST_ICE): Likewise.
+
+2001-10-24 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * attr.c (pthread_attr_setstacksize): Quell warning
+ from LCC by conditionally compiling the stacksize
+ validity check. LCC correctly warns that the condition
+ (stacksize < PTHREAD_STACK_MIN) is suspicious
+ because STACK_MIN is 0 and stacksize is of type
+ size_t (or unsigned int).
+
+2001-10-17 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * barrier.c: Move _LONG and _LPLONG defines into
+ implement.h; rename to PTW32_INTERLOCKED_LONG and
+ PTW32_INTERLOCKED_LPLONG respectively.
+ * spin.c: Likewise; ptw32_interlocked_compare_exchange used
+ in place of InterlockedCompareExchange directly.
+ * global.c (ptw32_interlocked_compare_exchange): Add
+ prototype for this new routine pointer to be used when
+ InterlockedCompareExchange isn't supported by Windows.
+ * nonportable.c (pthread_win32_process_attach_np): Check for
+ support of InterlockedCompareExchange in kernel32 and assign its
+ address to ptw32_interlocked_compare_exchange if it exists, or
+ our own ix86 specific implementation ptw32_InterlockedCompareExchange.
+ *private.c (ptw32_InterlockedCompareExchange): An
+ implementation of InterlockedCompareExchange() which is
+ specific to ix86; written directly in assembler for either
+ MSVC or GNU C; needed because Windows 95 doesn't support
+ InterlockedCompareExchange().
+
+ * sched.c (sched_get_priority_min): Extend to return
+ THREAD_PRIORITY_IDLE.
+ (sched_get_priority_max): Extend to return
+ THREAD_PRIORITY_CRITICAL.
+
+2001-10-15 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * spin.c (pthread_spin_lock): PTHREAD_SPINLOCK_INITIALIZER
+ was causing a program fault.
+ (pthread_spin_init): Could have alloced memory
+ without freeing under some error conditions.
+
+ * mutex.c (pthread_mutex_init): Move memory
+ allocation of mutex struct after checking for
+ PROCESS_SHARED.
+
+2001-10-12 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * spin.c (pthread_spin_unlock): Was not returning
+ EPERM if the spinlock was not locked, for multi CPU
+ machines.
+
+2001-10-08 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * spin.c (pthread_spin_trylock): Was not returning
+ EBUSY for multi CPU machines.
+
+2001-08-24 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * condvar.c (pthread_cond_destroy): Remove cv element
+ that is no longer used.
+ * implement.h: Likewise.
+
+2001-08-23 Alexander Terekhov <TEREKHOV at de.ibm.com>
+
+ * condvar.c (pthread_cond_destroy): fix bug with
+ respect to deadlock in the case of concurrent
+ _destroy/_unblock; a condition variable can be destroyed
+ immediately after all the threads that are blocked on
+ it are awakened.
+
+2001-08-23 Phil Frisbie, Jr. <phil at hawksoft.com>
+
+ * tsd.c (pthread_getspecific): Preserve the last
+ winsock error [from WSAGetLastError()].
+
+2001-07-18 Scott McCaskill <scott at magruder.org>
+
+ * mutex.c (pthread_mutexattr_init): Return ENOMEM
+ immediately and don't dereference the NULL pointer
+ if calloc fails.
+ (pthread_mutexattr_getpshared): Don't dereference
+ a pointer that is possibly NULL.
+ * barrier.c (pthread_barrierattr_init): Likewise
+ (pthread_barrierattr_getpshared): Don't dereference
+ a pointer that is possibly NULL.
+ * condvar.c (pthread_condattr_getpshared): Don't dereference
+ a pointer that is possibly NULL.
+
+2001-07-15 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * rwlock.c (pthread_rwlock_wrlock): Is allowed to be
+ a cancelation point; re-enable deferred cancelability
+ around the CV call.
+
+2001-07-10 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * barrier.c: Still more revamping. The exclusive access
+ mutex isn't really needed so it has been removed and replaced
+ by an InterlockedDecrement(). nSerial has been removed.
+ iStep is now dual-purpose. The process shared attribute
+ is now stored in the barrier struct.
+ * implement.h (pthread_barrier_t_): Lost some/gained one
+ elements.
+ * private.c (ptw32_threadStart): Removed some comments.
+
+2001-07-10 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * barrier.c: Revamped to fix the race condition. Two alternating
+ semaphores are used instead of the PulseEvent. Also improved
+ overall throughput by returning PTHREAD_BARRIER_SERIAL_THREAD
+ to the first waking thread.
+ * implement.h (pthread_barrier_t_): Revamped.
+
+2001-07-09 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * barrier.c: Fix several bugs in all routines. Now passes
+ tests/barrier5.c which is fairly rigorous. There is still
+ a non-optimal work-around for a race condition between
+ the barrier breeched event signal and event wait. Basically
+ the last (signalling) thread to hit the barrier yields
+ to allow any other threads, which may have lost the race,
+ to complete.
+
+2001-07-07 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * barrier.c: Changed synchronisation mechanism to a
+ Win32 manual reset Event and use PulseEvent to signal
+ waiting threads. If the implementation continued to use
+ a semaphore it would require a second semaphore and
+ some management to use them alternately as barriers. A
+ single semaphore allows threads to cascade from one barrier
+ through the next, leaving some threads blocked at the first.
+ * implement.h (pthread_barrier_t_): As per above.
+ * general: Made a number of other routines inlinable.
+
+2001-07-07 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * spin.c: Revamped and working; included static initialiser.
+ Now beta level.
+ * barrier.c: Likewise.
+ * condvar.c: Macro constant change; inline auto init routine.
+ * mutex.c: Likewise.
+ * rwlock.c: Likewise.
+ * private.c: Add support for spinlock initialiser.
+ * global.c: Likewise.
+ * implement.h: Likewise.
+ * pthread.h (PTHREAD_SPINLOCK_INITIALIZER): Fix typo.
+
+2001-07-05 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * barrier.c: Remove static initialisation - irrelevent
+ for this object.
+ * pthread.h (PTHREAD_BARRIER_INITIALIZER): Removed.
+ * rwlock.c (pthread_rwlock_wrlock): This routine is
+ not a cancelation point - disable deferred
+ cancelation around call to pthread_cond_wait().
+
+2001-07-05 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * spin.c: New module implementing spin locks.
+ * barrier.c: New module implementing barriers.
+ * pthread.h (_POSIX_SPIN_LOCKS): defined.
+ (_POSIX_BARRIERS): Defined.
+ (pthread_spin_*): Defined.
+ (pthread_barrier*): Defined.
+ (PTHREAD_BARRIER_SERIAL_THREAD): Defined.
+ * implement.h (pthread_spinlock_t_): Defined.
+ (pthread_barrier_t_): Defined.
+ (pthread_barrierattr_t_): Defined.
+
+ * mutex.c (pthread_mutex_lock): Return with the error
+ if an auto-initialiser initialisation fails.
+
+ * nonportable.c (pthread_getprocessors_np): New; gets the
+ number of available processors for the current process.
+
+2001-07-03 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * pthread.h (_POSIX_READER_WRITER_LOCKS): Define it
+ if not already defined.
+
+2001-07-01 Alexander Terekhov <TEREKHOV at de.ibm.com>
+
+ * condvar.c: Fixed lost signal bug reported by Timur Aydin
+ (taydin@snet.net).
+ [RPJ (me) didn't translate the original algorithm
+ correctly.]
+ * semaphore.c: Added sem_post_multiple; this is a useful
+ routine, but it doesn't appear to be standard. For now it's
+ not an exported function.
+
+2001-06-25 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * create.c (pthread_create): Add priority inheritance
+ attributes.
+ * mutex.c (pthread_mutex_lock): Remove some overhead for
+ PTHREAD_MUTEX_NORMAL mutex types. Specifically, avoid
+ calling pthread_self() and pthread_equal() to check/set
+ the mutex owner. Introduce a new pseudo owner for this
+ type. Test results suggest increases in speed of up to
+ 90% for non-blocking locks.
+ This is the default type of mutex used internally by other
+ synchronising objects, ie. condition variables and
+ read-write locks. The test rwlock7.c shows about a
+ 30-35% speed increase over snapshot 2001-06-06. The
+ price of this is that the application developer
+ must ensure correct behaviour, or explicitly set the
+ mutex to a safer type such as PTHREAD_MUTEX_ERRORCHECK.
+ For example, PTHREAD_MUTEX_NORMAL (or PTHREAD_MUTEX_DEFAULT)
+ type mutexes will not return an error if a thread which is not
+ the owner calls pthread_mutex_unlock. The call will succeed
+ in unlocking the mutex if it is currently locked, but a
+ subsequent unlock by the true owner will then fail with EPERM.
+ This is however consistent with some other implementations.
+ (pthread_mutex_unlock): Likewise.
+ (pthread_mutex_trylock): Likewise.
+ (pthread_mutex_destroy): Likewise.
+ * attr.c (pthread_attr_init): PTHREAD_EXPLICIT_SCHED is the
+ default inheritance attribute; THREAD_PRIORITY_NORMAL is
+ the default priority for new threads.
+ * sched.c (pthread_attr_setschedpolicy): Added routine.
+ (pthread_attr_getschedpolicy): Added routine.
+ (pthread_attr_setinheritsched): Added routine.
+ (pthread_attr_getinheritsched): Added routine.
+ * pthread.h (sched_rr_set_interval): Added as a macro;
+ returns -1 with errno set to ENOSYS.
+
+2001-06-23 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ *sched.c (pthread_attr_setschedparam): Add priority range
+ check.
+ (sched_setscheduler): New function; checks for a valid
+ pid and policy; checks for permission to set information
+ in the target process; expects pid to be a Win32 process ID,
+ not a process handle; the only scheduler policy allowed is
+ SCHED_OTHER.
+ (sched_getscheduler): Likewise, but checks for permission
+ to query.
+ * pthread.h (SCHED_*): Moved to sched.h as defined in the
+ POSIX standard.
+ * sched.h (SCHED_*): Moved from pthread.h.
+ (pid_t): Defined if necessary.
+ (sched_setscheduler): Defined.
+ (sched_getscheduler): Defined.
+ * pthread.def (sched_setscheduler): Exported.
+ (sched_getscheduler): Likewise.
+
+2001-06-23 Ralf Brese <Ralf.Brese at pdb4.siemens.de>
+
+ * create.c (pthread_create): Set thread priority from
+ thread attributes.
+
+2001-06-18 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * Made organisational-only changes to UWIN additions.
+ * dll.c (dllMain): Moved UWIN process attach code
+ to pthread_win32_process_attach_np(); moved
+ instance of pthread_count to global.c.
+ * global.c (pthread_count): Moved from dll.c.
+ * nonportable.c (pthread_win32_process_attach_np):
+ Moved _UWIN code to here from dll.c.
+ * implement.h (pthread_count): Define extern int.
+ * create.c (pthread_count): Remove extern int.
+ * private.c (pthread_count): Likewise.
+ * exit.c (pthread_count): Likewise.
+
+2001-06-18 David Korn <dgk at research.att.com>
+
+ * dll.c: Added changes necessary to work with UWIN.
+ * create.c: Likewise.
+ * pthread.h: Likewise.
+ * misc.c: Likewise.
+ * exit.c: Likewise.
+ * private.c: Likewise.
+ * implement.h: Likewise.
+ There is some room at the start of struct pthread_t_
+ to implement the signal semantics in UWIN's posix.dll
+ although this is not yet complete.
+ * Nmakefile: Compatible with UWIN's Nmake utility.
+ * Nmakefile.tests: Likewise - for running the tests.
+
+2001-06-08 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * semaphore.h (sem_t): Fixed for compile and test.
+ * implement.h (sem_t_): Likewise.
+ * semaphore.c: Likewise.
+ * private.c (ptw32_sem_timedwait): Updated to use new
+ opaque sem_t.
+
+2001-06-06 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * semaphore.h (sem_t): Is now an opaque pointer;
+ moved actual definition to implement.h.
+ * implement.h (sem_t_): Move here from semaphore.h;
+ was the definition of sem_t.
+ * semaphore.c: Wherever necessary, changed use of sem
+ from that of a pointer to a pointer-pointer; added
+ extra checks for a valid sem_t; NULL sem_t when
+ it is destroyed; added extra checks when creating
+ and destroying sem_t elements in the NEED_SEM
+ code branches; changed from using a pthread_mutex_t
+ ((*sem)->mutex) to CRITICAL_SECTION ((*sem)->sem_lock_cs)
+ in NEED_SEM branches for access serialisation.
+
+2001-06-06 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * mutex.c (pthread_mutexattr_init): Remove
+ ptw32_mutex_default_kind.
+
+2001-06-05 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * nonportable.c (pthread_mutex_setdefaultkind_np):
+ Remove - should not have been included in the first place.
+ (pthread_mutex_getdefaultkind_np): Likewise.
+ * global.c (ptw32_mutex_default_kind): Likewise.
+ * mutex.c (pthread_mutex_init): Remove use of
+ ptw32_mutex_default_kind.
+ * pthread.h (pthread_mutex_setdefaultkind_np): Likewise.
+ (pthread_mutex_getdefaultkind_np): Likewise.
+ * pthread.def (pthread_mutexattr_setkind_np): Added.
+ (pthread_mutexattr_getkind_np): Likewise.
+
+ * README: Many changes that should have gone in before
+ the last snapshot.
+ * README.NONPORTABLE: New - referred to by ANNOUNCE
+ but never created; documents the non-portable routines
+ included in the library - moved from README with new
+ routines added.
+ * ANNOUNCE (pthread_mutexattr_setkind_np): Added to
+ compliance list.
+ (pthread_mutexattr_getkind_np): Likewise.
+
+2001-06-04 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * condvar.c: Add original description of the algorithm as
+ developed by Terekhov and Thomas, plus reference to
+ README.CV.
+
+2001-06-03 Alexander Terekhov <TEREKHOV at de.ibm.com>, Louis Thomas <lthomas at arbitrade.com>
+
+ * condvar.c (pthread_cond_init): Completely revamped.
+ (pthread_cond_destroy): Likewise.
+ (ptw32_cond_wait_cleanup): Likewise.
+ (ptw32_cond_timedwait): Likewise.
+ (ptw32_cond_unblock): New general signaling routine.
+ (pthread_cond_signal): Now calls ptw32_cond_unblock.
+ (pthread_cond_broadcast): Likewise.
+ * implement.h (pthread_cond_t_): Revamped.
+ * README.CV: New; explanation of the above changes.
+
+2001-05-30 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * pthread.h (rand_r): Fake using _seed argument to quell
+ compiler warning (compiler should optimise this away later).
+
+ * GNUmakefile (OPT): Leave symbolic information out of the library
+ and increase optimisation level - for smaller faster prebuilt
+ dlls.
+
+2001-05-29 Milan Gardian <Milan.Gardian at LEIBINGER.com>
+
+ * Makefile: fix typo.
+ * pthreads.h: Fix problems with stdcall/cdecl conventions, in particular
+ remove the need for PT_STDCALL everywhere; remove warning supression.
+ * (errno): Fix the longstanding "inconsistent dll linkage" problem
+ with errno; now also works with /MD debugging libs -
+ warnings emerged when compiling pthreads library with /MD (or /MDd)
+ compiler switch, instead of /MT (or /MTd) (i.e. when compiling pthreads
+ using Multithreaded DLL CRT instead of Multithreaded statically linked
+ CRT).
+ * create.c (pthread_create): Likewise; fix typo.
+ * private.c (ptw32_threadStart): Eliminate use of terminate() which doesn't
+ throw exceptions.
+ * Remove unnecessary #includes from a number of modules -
+ [I had to #include malloc.h in implement.h for gcc - rpj].
+
+2001-05-29 Thomas Pfaff <tpfaff at gmx.net>
+
+ * pthread.h (PTHREAD_MUTEX_DEFAULT): New; equivalent to
+ PTHREAD_MUTEX_DEFAULT_NP.
+ * (PTHREAD_MUTEX_NORMAL): Similarly.
+ * (PTHREAD_MUTEX_ERRORCHECK): Similarly.
+ * (PTHREAD_MUTEX_RECURSIVE): Similarly.
+ * (pthread_mutex_setdefaultkind_np): New; Linux compatibility stub
+ for pthread_mutexattr_settype.
+ * (pthread_mutexattr_getkind_np): New; Linux compatibility stub
+ for pthread_mutexattr_gettype.
+ * mutex.c (pthread_mutexattr_settype): New; allow
+ the following types of mutex:
+ PTHREAD_MUTEX_DEFAULT_NP
+ PTHREAD_MUTEX_NORMAL_NP
+ PTHREAD_MUTEX_ERRORCHECK_NP
+ PTHREAD_MUTEX_RECURSIVE_NP
+ * Note that PTHREAD_MUTEX_DEFAULT is equivalent to
+ PTHREAD_MUTEX_NORMAL - ie. mutexes should no longer
+ be recursive by default, and a thread will deadlock if it
+ tries to relock a mutex it already owns. This is inline with
+ other pthreads implementations.
+ * (pthread_mutex_lock): Process the lock request
+ according to the mutex type.
+ * (pthread_mutex_init): Eliminate use of Win32 mutexes as the
+ basis of POSIX mutexes - instead, a combination of one critical section
+ and one semaphore are used in conjunction with Win32 Interlocked* routines.
+ * (pthread_mutex_destroy): Likewise.
+ * (pthread_mutex_lock): Likewise.
+ * (pthread_mutex_trylock): Likewise.
+ * (pthread_mutex_unlock): Likewise.
+ * Use longjmp/setjmp to implement cancelation when building the library
+ using a C compiler which doesn't support exceptions, e.g. gcc -x c (note
+ that gcc -x c++ uses exceptions).
+ * Also fixed some of the same typos and eliminated PT_STDCALL as
+ Milan Gardian's patches above.
+
+2001-02-07 Alexander Terekhov <TEREKHOV at de.ibm.com>
+
+ * rwlock.c: Revamped.
+ * implement.h (pthread_rwlock_t_): Redefined.
+ This implementation does not have reader/writer starvation problem.
+ Rwlock attempts to behave more like a normal mutex with
+ races and scheduling policy determining who is more important;
+ It also supports recursive locking,
+ has less synchronization overhead (no broadcasts at all,
+ readers are not blocked on any condition variable) and seem to
+ be faster than the current implementation [W98 appears to be
+ approximately 15 percent faster at least - on top of speed increase
+ from Thomas Pfaff's changes to mutex.c - rpj].
+
+2000-12-29 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * Makefile: Back-out "for" loops which don't work.
+
+ * GNUmakefile: Remove the fake.a target; add the "realclean"
+ target; don't remove built libs under the "clean" target.
+
+ * config.h: Add a guard against multiple inclusion.
+
+ * semaphore.h: Add some defines from config.h to make
+ semaphore.h independent of config.h when building apps.
+
+ * pthread.h (_errno): Back-out previous fix until we know how to
+ fix it properly.
+
+ * implement.h (lockCount): Add missing element to pthread_mutex_t_.
+
+ * sync.c (pthread_join): Spelling fix in comment.
+
+ * private.c (ptw32_threadStart): Reset original termination
+ function (C++).
+ (ptw32_threadStart): Cleanup detached threads early in case
+ the library is statically linked.
+ (ptw32_callUserDestroyRoutines): Remove [SEH] __try block from
+ destructor call so that unhandled exceptions will be passed through
+ to the system; call terminate() from [C++] try block for the same
+ reason.
+
+ * tsd.c (pthread_getspecific): Add comment.
+
+ * mutex.c (pthread_mutex_init): Initialise new elements in
+ pthread_mutex_t.
+ (pthread_mutex_unlock): Invert "pthread_equal()" test.
+
+2000-12-28 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * semaphore.c (mode_t): Use ifndef HAVE_MODE_T to include definition.
+
+ * config.h.in (HAVE_MODE_T): Added.
+ (_UWIN): Start adding defines for the UWIN package.
+
+ * private.c (ptw32_threadStart): Unhandled exceptions are
+ now passed through to the system to deal with. This is consistent
+ with normal Windows behaviour. C++ applications may use
+ set_terminate() to override the default behaviour which is
+ to call ptw32_terminate(). Ptw32_terminate() cleans up some
+ POSIX thread stuff before calling the system default function
+ which calls abort(). The users termination function should conform
+ to standard C++ semantics which is to not return. It should
+ exit the thread (call pthread_exit()) or exit the application.
+ * private.c (ptw32_terminate): Added as the default set_terminate()
+ function. It calls the system default function after cleaning up
+ some POSIX thread stuff.
+
+ * implement.h (ptw32_try_enter_critical_section): Move
+ declaration.
+ * global.c (ptw32_try_enter_critical_section): Moved
+ from dll.c.
+ * dll.c: Move process and thread attach/detach code into
+ functions in nonportable.c.
+ * nonportable.c (pthread_win32_process_attach_np): Process
+ attach code from dll.c is now available to static linked
+ applications.
+ * nonportable.c (pthread_win32_process_detach_np): Likewise.
+ * nonportable.c (pthread_win32_thread_attach_np): Likewise.
+ * nonportable.c (pthread_win32_thread_detach_np): Likewise.
+
+ * pthread.h: Add new non-portable prototypes for static
+ linked applications.
+
+ * GNUmakefile (OPT): Increase optimisation flag and remove
+ debug info flag.
+
+ * pthread.def: Add new non-portable exports for static
+ linked applications.
+
+2000-12-11 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * FAQ: Update Answer 6 re getting a fully working
+ Mingw32 built library.
+
+2000-10-10 Steven Reddie <smr at essemer.com.au>
+
+ * misc.c (pthread_self): Restore Win32 "last error"
+ cleared by TlsGetValue() call in
+ pthread_getspecific()
+
+2000-09-20 Arthur Kantor <akantor at bexusa.com>
+
+ * mutex.c (pthread_mutex_lock): Record the owner
+ of the mutex. This requires also keeping count of
+ recursive locks ourselves rather than leaving it
+ to Win32 since we need to know when to NULL the
+ thread owner when the mutex is unlocked.
+ (pthread_mutex_trylock): Likewise.
+ (pthread_mutex_unlock): Check that the calling
+ thread owns the mutex, decrement the recursive
+ lock count, and NULL the owner if zero. Return
+ EPERM if the mutex is owned by another thread.
+ * implement.h (pthread_mutex_t_): Add ownerThread
+ and lockCount members.
+
+2000-09-13 Jef Gearhart <jgearhart at tpssys.com>
+
+ * mutex.c (pthread_mutex_init): Call
+ TryEnterCriticalSection through the pointer
+ rather than directly so that the dll can load
+ on Windows versions that can't resolve the
+ function, eg. Windows 95
+
+2000-09-09 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * pthread.h (ctime_r): Fix arg.
+
+2000-09-08 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * GNUmakefile(_WIN32_WINNT=0x400): Define in CFLAGS;
+ doesn't seem to be needed though.
+
+ * cancel.c (pthread_cancel): Must get "self" through
+ calling pthread_self() which will ensure a POSIX thread
+ struct is built for non-POSIX threads; return an error
+ if this fails
+ - Ollie Leahy <ollie at mpt.ie>
+ (pthread_setcancelstate): Likewise.
+ (pthread_setcanceltype): Likewise.
+ * misc.c (ptw32_cancelable_wait): Likewise.
+
+ * private.c (ptw32_tkAssocCreate): Remove unused #if 0
+ wrapped code.
+
+ * pthread.h (ptw32_get_exception_services_code):
+ Needed to be forward declared unconditionally.
+
+2000-09-06 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * cancel.c (pthread_cancel): If called from the main
+ thread "self" would be NULL; get "self" via pthread_self()
+ instead of directly from TLS so that an implicit
+ pthread object is created.
+
+ * misc.c (pthread_equal): Strengthen test for NULLs.
+
+2000-09-02 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * condvar.c (ptw32_cond_wait_cleanup): Ensure that all
+ waking threads check if they are the last, and notify
+ the broadcaster if so - even if an error occurs in the
+ waiter.
+
+ * semaphore.c (_decrease_semaphore): Should be
+ a call to ptw32_decrease_semaphore.
+ (_increase_semaphore): Should be a call to
+ ptw32_increase_semaphore.
+
+ * misc.c (ptw32_cancelable_wait): Renamed from
+ CancelableWait.
+ * rwlock.c (_rwlock_check*): Renamed to
+ ptw32_rwlock_check*.
+ * mutex.c (_mutex_check*): Renamed to ptw32_mutex_check*.
+ * condvar.c (cond_timed*): Renamed to ptw32_cond_timed*.
+ (_cond_check*): Renamed to ptw32_cond_check*.
+ (cond_wait_cleanup*): Rename to ptw32_cond_wait_cleanup*.
+ (ptw32_cond_timedwait): Add comments.
+
+2000-08-22 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * private.c (ptw32_throw): Fix exception test;
+ move exceptionInformation declaration.
+
+ * tsd.c (pthread_key_create): newkey wrongly declared.
+
+ * pthread.h: Fix comment block.
+
+2000-08-18 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * mutex.c (pthread_mutex_destroy): Check that the mutex isn't
+ held; invalidate the mutex as early as possible to avoid
+ contention; not perfect - FIXME!
+
+ * rwlock.c (pthread_rwlock_init): Remove redundant assignment
+ to "rw".
+ (pthread_rwlock_destroy): Invalidate the rwlock before
+ freeing up any of it's resources - to avoid contention.
+
+ * private.c (ptw32_tkAssocCreate): Change assoc->lock
+ to use a dynamically initialised mutex - only consumes
+ a W32 mutex or critical section when first used,
+ not before.
+
+ * mutex.c (pthread_mutex_init): Remove redundant assignment
+ to "mx".
+ (pthread_mutexattr_destroy): Set attribute to NULL
+ before freeing it's memory - to avoid contention.
+
+ * implement.h (PTW32_EPS_CANCEL/PTW32_EPS_EXIT):
+ Must be defined for all compilers - used as generic
+ exception selectors by ptw32_throw().
+
+ * Several: Fix typos from scripted edit session
+ yesterday.
+
+ * nonportable.c (pthread_mutexattr_setforcecs_np):
+ Moved this function from mutex.c.
+ (pthread_getw32threadhandle_np): New function to
+ return the win32 thread handle that the POSIX
+ thread is using.
+ * mutex.c (pthread_mutexattr_setforcecs_np):
+ Moved to new file "nonportable.c".
+
+ * pthread.h (PTW32_BUILD): Only redefine __except
+ and catch compiler keywords if we aren't building
+ the library (ie. PTW32_BUILD is not defined) -
+ this is safer than defining and then undefining
+ if not building the library.
+ * implement.h: Remove __except and catch undefines.
+ * Makefile (CFLAGS): Define PTW32_BUILD.
+ * GNUmakefile (CFLAGS): Define PTW32_BUILD.
+
+ * All appropriate: Change Pthread_exception* to
+ ptw32_exception* to be consistent with internal
+ identifier naming.
+
+ * private.c (ptw32_throw): New function to provide
+ a generic exception throw for all internal
+ exceptions and EH schemes.
+ (ptw32_threadStart): pthread_exit() value is now
+ returned via the thread structure exitStatus
+ element.
+ * exit.c (pthread_exit): pthread_exit() value is now
+ returned via the thread structure exitStatus
+ element.
+ * cancel.c (ptw32_cancel_self): Now uses ptw32_throw.
+ (pthread_setcancelstate): Ditto.
+ (pthread_setcanceltype): Ditto.
+ (pthread_testcancel): Ditto.
+ (pthread_cancel): Ditto.
+ * misc.c (CancelableWait): Ditto.
+ * exit.c (pthread_exit): Ditto.
+ * All applicable: Change PTW32_ prefix to
+ PTW32_ prefix to remove leading underscores
+ from private library identifiers.
+
+2000-08-17 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * All applicable: Change _pthread_ prefix to
+ ptw32_ prefix to remove leading underscores
+ from private library identifiers (single
+ and double leading underscores are reserved in the
+ ANSI C standard for compiler implementations).
+
+ * tsd.c (pthread_create_key): Initialise temporary
+ key before returning it's address to avoid race
+ conditions.
+
+2000-08-13 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * errno.c: Add _MD precompile condition; thus far
+ had no effect when using /MD compile option but I
+ thnk it should be there.
+
+ * exit.c: Add __cplusplus to various #if lines;
+ was compiling SEH code even when VC++ had
+ C++ compile options.
+
+ * private.c: ditto.
+
+ * create.c (pthread_create): Add PT_STDCALL macro to
+ function pointer arg in _beginthread().
+
+ * pthread.h: PT_STDCALL really does need to be defined
+ in both this and impliment.h; don't set it to __cdecl
+ - this macro is only used to extend function pointer
+ casting for functions that will be passed as parameters.
+ (~PThreadCleanup): add cast and group expression.
+ (_errno): Add _MD compile conditional.
+ (PtW32NoCatchWarn): Change pragma message.
+
+ * implement.h: Move and change PT_STDCALL define.
+
+ * need_errno.h: Add _MD to compilation conditional.
+
+ * GNUmakefile: Substantial rewrite for new naming
+ convention; set for nil optimisation (turn it up
+ when we have a working library build; add target
+ "fake.a" to build a libpthreadw32.a from the VC++
+ built DLL pthreadVCE.dll.
+
+ * pthread.def (LIBRARY): Don't specify in the .def
+ file - it is specified on the linker command line
+ since we now use the same .def file for variously
+ named .dlls.
+
+ * Makefile: Substantial rewrite for new naming
+ convention; default nmake target only issues a
+ help message; run nmake with specific target
+ corresponding to the EH scheme being used.
+
+ * README: Update information; add naming convention
+ explanation.
+
+ * ANNOUNCE: Update information.
+
+2000-08-12 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * pthread.h: Add compile-time message when using
+ MSC_VER compiler and C++ EH to warn application
+ programmers to use PtW32Catch instead of catch(...)
+ if they want cancelation and pthread_exit to work.
+
+ * implement.h: Remove #include <semaphore.h>; we
+ use our own local semaphore.h.
+
+2000-08-10 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * cleanup.c (pthread_pop_cleanup): Remove _pthread
+ prefix from __except and catch keywords; implement.h
+ now simply undefines ptw32__except and
+ ptw32_catch if defined; VC++ was not textually
+ substituting ptw32_catch etc back to catch as
+ it was redefined; the reason for using the prefixed
+ version was to make it clear that it was not using
+ the pthread.h redefined catch keyword.
+
+ * private.c (ptw32_threadStart): Ditto.
+ (ptw32_callUserDestroyRoutines): Ditto.
+
+ * implement.h (ptw32__except): Remove #define.
+ (ptw32_catch): Remove #define.
+
+ * GNUmakefile (pthread.a): New target to build
+ libpthread32.a from pthread.dll using dlltool.
+
+ * buildlib.bat: Duplicate cl commands with args to
+ build C++ EH version of pthread.dll; use of .bat
+ files is redundant now that nmake compatible
+ Makefile is included; used as a kludge only now.
+
+ * Makefile: Localise some macros and fix up the clean:
+ target to extend it and work properly.
+
+ * CONTRIBUTORS: Add contributors.
+
+ * ANNOUNCE: Updated.
+
+ * README: Updated.
+
+2000-08-06 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * pthread.h: Remove #warning - VC++ doesn't accept it.
+
+2000-08-05 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * pthread.h (PtW32CatchAll): Add macro. When compiling
+ applications using VC++ with C++ EH rather than SEH
+ 'PtW32CatchAll' must be used in place of any 'catch( ... )'
+ if the application wants pthread cancelation or
+ pthread_exit() to work.
+
+2000-08-03 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * pthread.h: Add a base class ptw32_exception for
+ library internal exceptions and change the "catch"
+ re-define macro to use it.
+
+2000-08-02 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * GNUmakefile (CFLAGS): Add -mthreads.
+ Add new targets to generate cpp and asm output.
+
+ * sync.c (pthread_join): Remove dead code.
+
+2000-07-25 Tristan Savatier <tristan at mpegtv.com>
+
+ * sched.c (sched_get_priority_max): Handle different WinCE and
+ Win32 priority values together.
+ (sched_get_priority_min): Ditto.
+
+2000-07-25 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * create.c (pthread_create): Force new threads to wait until
+ pthread_create has the new thread's handle; we also retain
+ a local copy of the handle for internal use until
+ pthread_create returns.
+
+ * private.c (ptw32_threadStart): Initialise ei[].
+ (ptw32_threadStart): When beginthread is used to start the
+ thread, force waiting until the creator thread had the
+ thread handle.
+
+ * cancel.c (ptw32_cancel_thread): Include context switch
+ code for defined(_X86_) environments in addition to _M_IX86.
+
+ * rwlock.c (pthread_rwlock_destroy): Assignment changed
+ to avoid compiler warning.
+
+ * private.c (ptw32_get_exception_services_code): Cast
+ NULL return value to avoid compiler warning.
+
+ * cleanup.c (pthread_pop_cleanup): Initialise "cleanup" variable
+ to avoid compiler warnings.
+
+ * misc.c (ptw32_new): Change "new" variable to "t" to avoid
+ confusion with the C++ keyword of the same name.
+
+ * condvar.c (cond_wait_cleanup): Initialise lastWaiter variable.
+ (cond_timedwait): Remove unused local variables. to avoid
+ compiler warnings.
+
+ * dll.c (dllMain): Remove 2000-07-21 change - problem
+ appears to be in pthread_create().
+
+2000-07-22 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * tsd.c (pthread_key_create): If a destructor was given
+ and the pthread_mutex_init failed, then would try to
+ reference a NULL pointer (*key); eliminate this section of
+ code by using a dynamically initialised mutex
+ (PTHREAD_MUTEX_INITIALIZER).
+
+ * tsd.c (pthread_setspecific): Return an error if
+ unable to set the value; simplify cryptic conditional.
+
+ * tsd.c (pthread_key_delete): Locking threadsLock relied
+ on mutex_lock returning an error if the key has no destructor.
+ ThreadsLock is only initialised if the key has a destructor.
+ Making this mutex a static could reduce the number of mutexes
+ used by an application since it is actually created only at
+ first use and it's often destroyed soon after.
+
+2000-07-22 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * FAQ: Added Q5 and Q6.
+
+2000-07-21 David Baggett <dmb at itasoftware.com>
+
+ * dll.c: Include resource leakage work-around. This is a
+ partial FIXME which doesn't stop all leakage. The real
+ problem needs to be found and fixed.
+
+2000-07-21 Ross Johnson <rpj at setup1.ise.canberra.edu.au>
+
+ * create.c (pthread_create): Set threadH to 0 (zero)
+ everywhere. Some assignments were using NULL. Maybe
+ it should be NULL everywhere - need to check. (I know
+ they are nearly always the same thing - but not by
+ definition.)
+
+ * misc.c (pthread_self): Try to catch NULL thread handles
+ at the point where they might be generated, even though
+ they should always be valid at this point.
+
+ * tsd.c (pthread_setspecific): return an error value if
+ pthread_self() returns NULL.
+
+ * sync.c (pthread_join): return an error value if
+ pthread_self() returns NULL.
+
+ * signal.c (pthread_sigmask): return an error value if
+ pthread_self() returns NULL.
+
+2000-03-02 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * attr.c (pthread_attr_init): Set default stacksize to zero (0)
+ rather than PTHREAD_STACK_MIN even though these are now the same.
+
+ * pthread.h (PTHREAD_STACK_MIN): Lowered to 0.
+
+2000-01-28 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * mutex.c (pthread_mutex_init): Free mutex if it has been alloced;
+ if critical sections can be used instead of Win32 mutexes, test
+ that the critical section works and return an error if not.
+
+2000-01-07 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * cleanup.c (pthread_pop_cleanup): Include SEH code only if MSC is not
+ compiling as C++.
+ (pthread_push_cleanup): Include SEH code only if MSC is not
+ compiling as C++.
+
+ * pthread.h: Include SEH code only if MSC is not
+ compiling as C++.
+
+ * implement.h: Include SEH code only if MSC is not
+ compiling as C++.
+
+ * cancel.c (ptw32_cancel_thread): Add _M_IX86 check.
+ (pthread_testcancel): Include SEH code only if MSC is not
+ compiling as C++.
+ (ptw32_cancel_self): Include SEH code only if MSC is not
+ compiling as C++.
+
+2000-01-06 Erik Hensema <erik.hensema at group2000.nl>
+
+ * Makefile: Remove inconsistencies in 'cl' args
+
+2000-01-04 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * private.c (ptw32_get_exception_services_code): New; returns
+ value of EXCEPTION_PTW32_SERVICES.
+ (ptw32_processInitialize): Remove initialisation of
+ ptw32_exception_services which is no longer needed.
+
+ * pthread.h (ptw32_exception_services): Remove extern.
+ (ptw32_get_exception_services_code): Add function prototype;
+ use this to return EXCEPTION_PTW32_SERVICES value instead of
+ using the ptw32_exception_services variable which I had
+ trouble exporting through pthread.def.
+
+ * global.c (ptw32_exception_services): Remove declaration.
+
+1999-11-22 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * implement.h: Forward declare ptw32_new();
+
+ * misc.c (ptw32_new): New; alloc and initialise a new pthread_t.
+ (pthread_self): New thread struct is generated by new routine
+ ptw32_new().
+
+ * create.c (pthread_create): New thread struct is generated
+ by new routine ptw32_new().
+
+1999-11-21 Ross Johnson <rpj at special.ise.canberra.edu.au>
+
+ * global.c (ptw32_exception_services): Declare new variable.
+
+ * private.c (ptw32_threadStart): Destroy thread's
+ cancelLock mutex; make 'catch' and '__except' usageimmune to
+ redfinitions in pthread.h.
+ (ptw32_processInitialize): Init new constant ptw32_exception_services.
+
+ * create.c (pthread_create): Initialise thread's cancelLock
+ mutex.
+
+ * cleanup.c (pthread_pop_cleanup): Make 'catch' and '__except'
+ usage immune to redfinition s in pthread.h.
+
+ * private.c: Ditto.
+
+ * pthread.h (catch): Redefine 'catch' so that C++ applications
+ won't catch our internal exceptions.
+ (__except): ditto for __except.
+
+ * implement.h (ptw32_catch): Define internal version
+ of 'catch' because 'catch' is redefined by pthread.h.
+ (__except): ditto for __except.
+ (struct pthread_t_): Add cancelLock mutex for async cancel
+ safety.
+
+1999-11-21 Jason Nye <jnye at nbnet.nb.ca>, Erik Hensema <erik.hensema at group2000.nl>
+
+ * cancel.c (ptw32_cancel_self): New; part of the async
+ cancellation implementation.
+ (ptw32_cancel_thread): Ditto; this function is X86
+ processor specific.
+ (pthread_setcancelstate): Add check for pending async
+ cancel request and cancel the calling thread if
+ required; add async-cancel safety lock.
+ (pthread_setcanceltype): Ditto.
+
+1999-11-13 Erik Hensema <erik.hensema at group2000.nl>
+
+ * configure.in (AC_OUTPUT): Put generated output into GNUmakefile
+ rather than Makefile. Makefile will become the MSC nmake compatible
+ version
+
+1999-11-13 John Bossom (John.Bossom@cognos.com>
+
+ * misc.c (pthread_self): Add a note about GetCurrentThread
+ returning a pseudo-handle
+
+1999-11-10 Todd Owen <towen at lucidcalm.dropbear.id.au>
+
+ * dll.c (dllMain): Free kernel32 ASAP.
+ If TryEnterCriticalSection is not being used, then free
+ the kernel32.dll handle now, rather than leaving it until
+ DLL_PROCESS_DETACH.
+
+ Note: this is not a pedantic exercise in freeing unused
+ resources! It is a work-around for a bug in Windows 95
+ (see microsoft knowledge base article, Q187684) which
+ does Bad Things when FreeLibrary is called within
+ the DLL_PROCESS_DETACH code, in certain situations.
+ Since w95 just happens to be a platform which does not
+ provide TryEnterCriticalSection, the bug will be
+ effortlessly avoided.
+
+1999-11-10 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * sync.c (pthread_join): Make it a deferred cancelation point.
+
+ * misc.c (pthread_self): Explicitly initialise implicitly
+ created thread state to default values.
+
+1999-11-05 Tristan Savatier <tristan at mpegtv.com>
+
+ * pthread.h (winsock.h): Include unconditionally.
+ (ETIMEDOUT): Change fallback value to that defined by winsock.h.
+
+ * general: Patched for portability to WinCE. The details are
+ described in the file WinCE-PORT. Follow the instructions
+ in README.WinCE to make the appropriate changes in config.h.
+
+1999-10-30 Erik Hensema <erik.hensema at group2000.nl>
+
+ * create.c (pthread_create): Explicitly initialise thread state to
+ default values.
+
+ * cancel.c (pthread_setcancelstate): Check for NULL 'oldstate'
+ for compatibility with Solaris pthreads;
+ (pthread_setcanceltype): ditto:
+
+1999-10-23 Erik Hensema <erik.hensema at group2000.nl>
+
+ * pthread.h (ctime_r): Fix incorrect argument "_tm"
+
+1999-10-21 Aurelio Medina <aureliom at crt.com>
+
+ * pthread.h (_POSIX_THREADS): Only define it if it isn't
+ already defined. Projects may need to define this on
+ the CC command line under Win32 as it doesn't have unistd.h
+
+1999-10-17 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * rwlock.c (pthread_rwlock_destroy): Add cast to remove compile
+ warning.
+
+ * condvar.c (pthread_cond_broadcast): Only release semaphores
+ if there are waiting threads.
+
+1999-10-15 Lorin Hochstein <lmh at xiphos.ca>, Peter Slacik <Peter.Slacik at tatramed.sk>
+
+ * condvar.c (cond_wait_cleanup): New static cleanup handler for
+ cond_timedwait;
+ (cond_timedwait): pthread_cleanup_push args changed;
+ canceling a thread while it's in pthread_cond_wait
+ will now decrement the waiters count and cleanup if it's the
+ last waiter.
+
+1999-10-15 Graham Dumpleton <Graham.Dumpleton at ra.pad.otc.telstra.com.au>
+
+ * condvar.c (cond_wait_cleanup): the last waiter will now reset the CV's
+ wasBroadcast flag
+
+Thu Sep 16 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * rwlock.c (pthread_rwlock_destroy): Add serialisation.
+ (_rwlock_check_need_init): Check for detroyed rwlock.
+ * rwlock.c: Check return codes from _rwlock_check_need_init();
+ modify comments; serialise access to rwlock objects during
+ operations; rename rw_mutex to rw_lock.
+ * implement.h: Rename rw_mutex to rw_lock.
+ * mutex.c (pthread_mutex_destroy): Add serialisation.
+ (_mutex_check_need_init): Check for detroyed mutex.
+ * condvar.c (pthread_cond_destroy): Add serialisation.
+ (_cond_check_need_init): Check for detroyed condvar.
+ * mutex.c: Modify comments.
+ * condvar.c: Modify comments.
+
+1999-08-10 Aurelio Medina <aureliom at crt.com>
+
+ * implement.h (pthread_rwlock_t_): Add.
+ * pthread.h (pthread_rwlock_t): Add.
+ (PTHREAD_RWLOCK_INITIALIZER): Add.
+ Add rwlock function prototypes.
+ * rwlock.c: New module.
+ * pthread.def: Add new rwlock functions.
+ * private.c (ptw32_processInitialize): initialise
+ ptw32_rwlock_test_init_lock critical section.
+ * global.c (ptw32_rwlock_test_init_lock): Add.
+
+ * mutex.c (pthread_mutex_destroy): Don't free mutex memory
+ if mutex is PTHREAD_MUTEX_INITIALIZER and has not been
+ initialised yet.
+
+1999-08-08 Milan Gardian <mg at tatramed.sk>
+
+ * mutex.c (pthread_mutex_destroy): Free mutex memory.
+
+1999-08-22 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * exit.c (pthread_exit): Fix reference to potentially
+ uninitialised pointer.
+
+1999-08-21 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * private.c (ptw32_threadStart): Apply fix of 1999-08-19
+ this time to C++ and non-trapped C versions. Ommitted to
+ do this the first time through.
+
+1999-08-19 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * private.c (ptw32_threadStart): Return exit status from
+ the application thread startup routine.
+ - Milan Gardian <mg at tatramed.sk>
+
+1999-08-18 John Bossom <john.Bossom at cognos.com>
+
+ * exit.c (pthread_exit): Put status into pthread_t->exitStatus
+ * private.c (ptw32_threadStart): Set pthread->exitStatus
+ on exit of try{} block.
+ * sync.c (pthread_join): use pthread_exitStatus value if the
+ thread exit doesn't return a value (for Mingw32 CRTDLL
+ which uses endthread instead of _endthreadex).
+
+Tue Aug 17 20:17:58 CDT 1999 Mumit Khan <khan at xraylith.wisc.edu>
+
+ * create.c (pthread_create): Add CRTDLL suppport.
+ * exit.c (pthread_exit): Likewise.
+ * private.c (ptw32_threadStart): Likewise.
+ (ptw32_threadDestroy): Likewise.
+ * sync.c (pthread_join): Likewise.
+ * tests/join1.c (main): Warn about partial support for CRTDLL.
+
+Tue Aug 17 20:00:08 1999 Mumit Khan <khan at xraylith.wisc.edu>
+
+ * Makefile.in (LD): Delete entry point.
+ * acconfig.h (STDCALL): Delete unused macro.
+ * configure.in: Remove test for STDCALL.
+ * config.h.in: Regenerate.
+ * errno.c (_errno): Fix self type.
+ * pthread.h (PT_STDCALL): Move from here to
+ * implement.h (PT_STDCALL): here.
+ (ptw32_threadStart): Fix prototype.
+ * private.c (ptw32_threadStart): Likewise.
+
+1999-08-14 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * exit.c (pthread_exit): Don't call pthread_self() but
+ get thread handle directly from TSD for efficiency.
+
+1999-08-12 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * private.c (ptw32_threadStart): ei[] only declared if _MSC_VER.
+
+ * exit.c (pthread_exit): Check for implicitly created threads
+ to avoid raising an unhandled exception.
+
+1999-07-12 Peter Slacik <Peter.Slacik at tatramed.sk>
+
+ * condvar.c (pthread_cond_destroy): Add critical section.
+ (cond_timedwait): Add critical section; check for timeout
+ waiting on semaphore.
+ (pthread_cond_broadcast): Add critical section.
+
+1999-07-09 Lorin Hochstein <lmh at xiphos.ca>, John Bossom <John.Bossom at Cognos.COM>
+
+ The problem was that cleanup handlers were not executed when
+ pthread_exit() was called.
+
+ * implement.h (pthread_t_): Add exceptionInformation element for
+ C++ per-thread exception information.
+ (general): Define and rename exceptions.
+
+1999-07-09 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * misc.c (CancelableWait): PTW32_EPS_CANCEL (SEH) and
+ ptw32_exception_cancel (C++) used to identify the exception.
+
+ * cancel.c (pthread_testcancel): PTW32_EPS_CANCEL (SEH) and
+ ptw32_exception_cancel (C++) used to identify the exception.
+
+ * exit.c (pthread_exit): throw/raise an exception to return to
+ ptw32_threadStart() to exit the thread. PTW32_EPS_EXIT (SEH)
+ and ptw32_exception_exit (C++) used to identify the exception.
+
+ * private.c (ptw32_threadStart): Add pthread_exit exception trap;
+ clean up and exit the thread directly rather than via pthread_exit().
+
+Sun May 30 00:25:02 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * semaphore.h (mode_t): Conditionally typedef it.
+
+Fri May 28 13:33:05 1999 Mark E. Armstrong <avail at pacbell.net>
+
+ * condvar.c (pthread_cond_broadcast): Fix possible memory fault
+
+Thu May 27 13:08:46 1999 Peter Slacik <Peter.Slacik at tatramed.sk>
+
+ * condvar.c (pthread_cond_broadcast): Fix logic bug
+
+Thu May 27 13:08:46 1999 Bossom, John <John.Bossom at Cognos.COM>
+
+ * condvar.c (pthread_cond_broadcast): optimise sem_post loop
+
+Fri May 14 12:13:18 1999 Mike Russo <miker at eai.com>
+
+ * attr.c (pthread_attr_setdetachstate): Fix logic bug
+
+Sat May 8 09:42:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.def (sem_open): Add.
+ (sem_close): Add.
+ (sem_unlink): Add.
+ (sem_getvalue): Add.
+
+ * FAQ (Question 3): Add.
+
+Thu Apr 8 01:16:23 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * semaphore.c (sem_open): New function; returns an error (ENOSYS).
+ (sem_close): ditto.
+ (sem_unlink): ditto.
+ (sem_getvalue): ditto.
+
+ * semaphore.h (_POSIX_SEMAPHORES): define.
+
+Wed Apr 7 14:09:52 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * errno.c (_REENTRANT || _MT): Invert condition.
+
+ * pthread.h (_errno): Conditionally include prototype.
+
+Wed Apr 7 09:37:00 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * *.c (comments): Remove individual attributions - these are
+ documented sufficiently elsewhere.
+
+ * implement.h (pthread.h): Remove extraneous include.
+
+Sun Apr 4 11:05:57 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * sched.c (sched.h): Include.
+
+ * sched.h: New file for POSIX 1b scheduling.
+
+ * pthread.h: Move opaque structures to implement.h; move sched_*
+ prototypes out and into sched.h.
+
+ * implement.h: Add opaque structures from pthread.h.
+
+ * sched.c (sched_yield): New function.
+
+ * condvar.c (ptw32_sem_*): Rename to sem_*; except for
+ ptw32_sem_timedwait which is an private function.
+
+Sat Apr 3 23:28:00 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * Makefile.in (OBJS): Add errno.o.
+
+Fri Apr 2 11:08:50 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h (ptw32_sem_*): Remove prototypes now defined in
+ semaphore.h.
+
+ * pthread.h (sempahore.h): Include.
+
+ * semaphore.h: New file for POSIX 1b semaphores.
+
+ * semaphore.c (ptw32_sem_timedwait): Moved to private.c.
+
+ * pthread.h (ptw32_sem_t): Change to sem_t.
+
+ * private.c (ptw32_sem_timedwait): Moved from semaphore.c;
+ set errno on error.
+
+ * pthread.h (pthread_t_): Add per-thread errno element.
+
+Fri Apr 2 11:08:50 1999 John Bossom <jebossom at cognos.com>
+
+ * semaphore.c (ptw32_sem_*): Change to sem_*; these functions
+ will be exported from the library; set errno on error.
+
+ * errno.c (_errno): New file. New function.
+
+Fri Mar 26 14:11:45 1999 Tor Lillqvist <tml at iki.fi>
+
+ * semaphore.c (ptw32_sem_timedwait): Check for negative
+ milliseconds.
+
+Wed Mar 24 11:32:07 1999 John Bossom <jebossom at cognos.com>
+
+ * misc.c (CancelableWait): Initialise exceptionInformation[2].
+ (pthread_self): Get a real Win32 thread handle for implicit threads.
+
+ * cancel.c (pthread_testcancel): Initialise exceptionInformation[2].
+
+ * implement.h (SE_INFORMATION): Fix values.
+
+ * private.c (ptw32_threadDestroy): Close the thread handle.
+
+Fri Mar 19 12:57:27 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * cancel.c (comments): Update and cleanup.
+
+Fri Mar 19 09:12:59 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * private.c (ptw32_threadStart): status returns PTHREAD_CANCELED.
+
+ * pthread.h (PTHREAD_CANCELED): defined.
+
+Tue Mar 16 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * all: Add GNU LGPL and Copyright and Warranty.
+
+Mon Mar 15 00:20:13 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * condvar.c (pthread_cond_init): fix possible uninitialised use
+ of cv.
+
+Sun Mar 14 21:01:59 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * condvar.c (pthread_cond_destroy): don't do full cleanup if
+ static initialised cv has never been used.
+ (cond_timedwait): check result of auto-initialisation.
+
+Thu Mar 11 09:01:48 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.h (pthread_mutex_t): revert to (pthread_mutex_t *);
+ define a value to serve as PTHREAD_MUTEX_INITIALIZER.
+ (pthread_mutex_t_): remove staticinit and valid elements.
+ (pthread_cond_t): revert to (pthread_cond_t_ *);
+ define a value to serve as PTHREAD_COND_INITIALIZER.
+ (pthread_cond_t_): remove staticinit and valid elements.
+
+ * mutex.c (pthread_mutex_t args): adjust indirection of references.
+ (all functions): check for PTHREAD_MUTEX_INITIALIZER value;
+ check for NULL (invalid).
+
+ * condvar.c (pthread_cond_t args): adjust indirection of references.
+ (all functions): check for PTHREAD_COND_INITIALIZER value;
+ check for NULL (invalid).
+
+Wed Mar 10 17:18:12 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * misc.c (CancelableWait): Undo changes from Mar 8 and 7.
+
+Mon Mar 8 11:18:59 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * misc.c (CancelableWait): Ensure cancelEvent handle is the lowest
+ indexed element in the handles array. Enhance test for abandoned
+ objects.
+
+ * pthread.h (PTHREAD_MUTEX_INITIALIZER): Trailing elements not
+ initialised are set to zero by the compiler. This avoids the
+ problem of initialising the opaque critical section element in it.
+ (PTHREAD_COND_INITIALIZER): Ditto.
+
+ * semaphore.c (ptw32_sem_timedwait): Check sem == NULL earlier.
+
+Sun Mar 7 12:31:14 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * condvar.c (pthread_cond_init): set semaphore initial value
+ to 0, not 1. cond_timedwait was returning signaled immediately.
+
+ * misc.c (CancelableWait): Place the cancel event handle first
+ in the handle table for WaitForMultipleObjects. This ensures that
+ the cancel event is recognised and acted apon if both objects
+ happen to be signaled together.
+
+ * private.c (ptw32_cond_test_init_lock): Initialise and destroy.
+
+ * implement.h (ptw32_cond_test_init_lock): Add extern.
+
+ * global.c (ptw32_cond_test_init_lock): Add declaration.
+
+ * condvar.c (pthread_cond_destroy): check for valid initialised CV;
+ flag destroyed CVs as invalid.
+ (pthread_cond_init): pthread_cond_t is no longer just a pointer.
+ This is because PTHREAD_COND_INITIALIZER needs state info to reside
+ in pthread_cond_t so that it can initialise on first use. Will work on
+ making pthread_cond_t (and other objects like it) opaque again, if
+ possible, later.
+ (cond_timedwait): add check for statically initialisation of
+ CV; initialise on first use.
+ (pthread_cond_signal): check for valid CV.
+ (pthread_cond_broadcast): check for valid CV.
+ (_cond_check_need_init): Add.
+
+ * pthread.h (PTHREAD_COND_INITIALIZER): Fix.
+ (pthread_cond_t): no longer a pointer to pthread_cond_t_.
+ (pthread_cond_t_): add 'staticinit' and 'valid' elements.
+
+Sat Mar 6 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h: Undate comments.
+
+Sun Feb 21 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.h (PTHREAD_MUTEX_INITIALIZER): missing braces around
+ cs element initialiser.
+
+1999-02-21 Ben Elliston <bje at cygnus.com>
+
+ * pthread.h (pthread_exit): The return type of this function is
+ void, not int.
+
+ * exit.c (pthread_exit): Do not return 0.
+
+Sat Feb 20 16:03:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * dll.c (DLLMain): Expand TryEnterCriticalSection support test.
+
+ * mutex.c (pthread_mutex_trylock): The check for
+ ptw32_try_enter_critical_section == NULL should have been
+ removed long ago.
+
+Fri Feb 19 16:03:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * sync.c (pthread_join): Fix pthread_equal() test.
+
+ * mutex.c (pthread_mutex_trylock): Check mutex != NULL before
+ using it.
+
+Thu Feb 18 16:17:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * misc.c (pthread_equal): Fix inverted result.
+
+ * Makefile.in: Use libpthread32.a as the name of the DLL export
+ library instead of pthread.lib.
+
+ * condvar.c (pthread_cond_init): cv could have been used unitialised;
+ initialise.
+
+ * create.c (pthread_create): parms could have been used unitialised;
+ initialise.
+
+ * pthread.h (struct pthread_once_t_): Remove redefinition.
+
+Sat Feb 13 03:03:30 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.h (struct pthread_once_t_): Replaced.
+
+ * misc.c (pthread_once): Replace with John Bossom's version;
+ has lighter weight serialisation; fixes problem of not holding
+ competing threads until after the init_routine completes.
+
+Thu Feb 11 13:34:14 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * misc.c (CancelableWait): Change C++ exception throw.
+
+ * sync.c (pthread_join): Change FIXME comment - issue resolved.
+
+Wed Feb 10 12:49:11 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * configure: Various temporary changes.
+ - Kevin Ruland <Kevin.Ruland at anheuser-busch.com>
+
+ * README: Update.
+
+ * pthread.def (pthread_attr_getstackaddr): uncomment
+ (pthread_attr_setstackaddr): uncomment
+
+Fri Feb 5 13:42:30 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * semaphore.c: Comment format changes.
+
+Thu Feb 4 10:07:28 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * global.c: Remove ptw32_exception instantiation.
+
+ * cancel.c (pthread_testcancel): Change C++ exception throw.
+
+ * implement.h: Remove extern declaration.
+
+Wed Feb 3 13:04:44 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * cleanup.c: Rename ptw32_*_cleanup() to pthread_*_cleanup().
+
+ * pthread.def: Ditto.
+
+ * pthread.h: Ditto.
+
+ * pthread.def (pthread_cleanup_push): Remove from export list;
+ the function is defined as a macro under all compilers.
+ (pthread_cleanup_pop): Ditto.
+
+ * pthread.h: Remove #if defined().
+
+Wed Feb 3 10:13:48 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * sync.c (pthread_join): Check for NULL value_ptr arg;
+ check for detached threads.
+
+Tue Feb 2 18:07:43 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * implement.h: Add #include <pthread.h>.
+ Change sem_t to ptw32_sem_t.
+
+Tue Feb 2 18:07:43 1999 Kevin Ruland <Kevin.Ruland at anheuser-busch.com>
+
+ * signal.c (pthread_sigmask): Add and modify casts.
+ Reverse LHS/RHS bitwise assignments.
+
+ * pthread.h: Remove #include <semaphore.h>.
+ (PTW32_ATTR_VALID): Add cast.
+ (struct pthread_t_): Add sigmask element.
+
+ * dll.c: Add "extern C" for DLLMain.
+ (DllMain): Add cast.
+
+ * create.c (pthread_create): Set sigmask in thread.
+
+ * condvar.c: Remove #include. Change sem_* to ptw32_sem_*.
+
+ * attr.c: Changed #include.
+
+ * Makefile.in: Additional targets and changes to build the library
+ as a DLL.
+
+Fri Jan 29 11:56:28 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * Makefile.in (OBJS): Add semaphore.o to list.
+
+ * semaphore.c (ptw32_sem_timedwait): Move from private.c.
+ Rename sem_* to ptw32_sem_*.
+
+ * pthread.h (pthread_cond_t): Change type of sem_t.
+ _POSIX_SEMAPHORES no longer defined.
+
+ * semaphore.h: Contents moved to implement.h.
+ Removed from source tree.
+
+ * implement.h: Add semaphore function prototypes and rename all
+ functions to prepend 'ptw32_'. They are
+ now private to the pthreads-win32 implementation.
+
+ * private.c: Change #warning.
+ Move ptw32_sem_timedwait() to semaphore.c.
+
+ * cleanup.c: Change #warning.
+
+ * misc.c: Remove #include <errno.h>
+
+ * pthread.def: Cleanup CVS merge conflicts.
+
+ * global.c: Ditto.
+
+ * ChangeLog: Ditto.
+
+ * cleanup.c: Ditto.
+
+Sun Jan 24 01:34:52 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * semaphore.c (sem_wait): Remove second arg to
+ pthreadCancelableWait() call.
+
+Sat Jan 23 17:36:40 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.def: Add new functions to export list.
+
+ * pthread.h (PTHREAD_MUTEX_AUTO_CS_NP): New.
+ (PTHREAD_MUTEX_FORCE_CS_NP): New.
+
+ * README: Updated.
+
+Fri Jan 22 14:31:59 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * Makefile.in (CFLAGS): Remove -fhandle-exceptions. Not needed
+ with egcs. Add -g for debugging.
+
+ * create.c (pthread_create): Replace __stdcall with PT_STDCALL
+ macro. This is a hack and must be fixed.
+
+ * misc.c (CancelableWait): Remove redundant statement.
+
+ * mutex.c (pthread_mutexattr_init): Cast calloc return value.
+
+ * misc.c (CancelableWait): Add cast.
+ (pthread_self): Add cast.
+
+ * exit.c (pthread_exit): Add cast.
+
+ * condvar.c (pthread_condattr_init): Cast calloc return value.
+
+ * cleanup.c: Reorganise conditional compilation.
+
+ * attr.c (pthread_attr_init): Remove unused 'result'.
+ Cast malloc return value.
+
+ * private.c (ptw32_callUserDestroyRoutines): Redo conditional
+ compilation.
+
+ * misc.c (CancelableWait): C++ version uses 'throw'.
+
+ * cancel.c (pthread_testcancel): Ditto.
+
+ * implement.h (class ptw32_exception): Define for C++.
+
+ * pthread.h: Fix C, C++, and Win32 SEH condition compilation
+ mayhem around pthread_cleanup_* defines. C++ version now uses John
+ Bossom's cleanup handlers.
+ (pthread_attr_t): Make 'valid' unsigned.
+ Define '_timeb' as 'timeb' for Ming32.
+ Define PT_STDCALL as nothing for Mingw32. May be temporary.
+
+ * cancel.c (pthread_testcancel): Cast return value.
+
+Wed Jan 20 09:31:28 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.h (pthread_mutexattr_t): Changed to a pointer.
+
+ * mutex.c (pthread_mutex_init): Conditionally create Win32 mutex
+ - from John Bossom's implementation.
+ (pthread_mutex_destroy): Conditionally close Win32 mutex
+ - from John Bossom's implementation.
+ (pthread_mutexattr_init): Replaced by John Bossom's version.
+ (pthread_mutexattr_destroy): Ditto.
+ (pthread_mutexattr_getpshared): New function from John Bossom's
+ implementation.
+ (pthread_mutexattr_setpshared): New function from John Bossom's
+ implementation.
+
+Tue Jan 19 18:27:42 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * pthread.h (pthreadCancelableTimedWait): New prototype.
+ (pthreadCancelableWait): Remove second argument.
+
+ * misc.c (CancelableWait): New static function is
+ pthreadCancelableWait() renamed.
+ (pthreadCancelableWait): Now just calls CancelableWait() with
+ INFINITE timeout.
+ (pthreadCancelableTimedWait): Just calls CancelableWait()
+ with passed in timeout.
+
+Tue Jan 19 18:27:42 1999 Scott Lightner <scott at curriculum.com>
+
+ * private.c (ptw32_sem_timedwait): 'abstime' arg really is
+ absolute time. Calculate relative time to wait from current
+ time before passing timeout to new routine
+ pthreadCancelableTimedWait().
+
+Tue Jan 19 10:27:39 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.h (pthread_mutexattr_setforcecs_np): New prototype.
+
+ * mutex.c (pthread_mutexattr_init): Init 'pshared' and 'forcecs'
+ attributes to 0.
+ (pthread_mutexattr_setforcecs_np): New function (not portable).
+
+ * pthread.h (pthread_mutex_t):
+ Add 'mutex' element. Set to NULL in PTHREAD_MUTEX_INITIALIZER.
+ The pthread_mutex_*() routines will try to optimise performance
+ by choosing either mutexes or critical sections as the basis
+ for pthread mutexes for each indevidual mutex.
+ (pthread_mutexattr_t_): Add 'forcecs' element.
+ Some applications may choose to force use of critical sections
+ if they know that:-
+ the mutex is PROCESS_PRIVATE and,
+ either the OS supports TryEnterCriticalSection() or
+ pthread_mutex_trylock() will never be called on the mutex.
+ This attribute will be setable via a non-portable routine.
+
+ Note: We don't yet support PROCESS_SHARED mutexes, so the
+ implementation as it stands will default to Win32 mutexes only if
+ the OS doesn't support TryEnterCriticalSection. On Win9x, and early
+ versions of NT 'forcecs' will need to be set in order to get
+ critical section based mutexes.
+
+Sun Jan 17 12:01:26 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.h (PTHREAD_MUTEX_INITIALIZER): Init new 'staticinit'
+ value to '1' and existing 'valid' value to '1'.
+
+ * global.c (ptw32_mutex_test_init_lock): Add.
+
+ * implement.h (ptw32_mutex_test_init_lock.): Add extern.
+
+ * private.c (ptw32_processInitialize): Init critical section for
+ global lock used by _mutex_check_need_init().
+ (ptw32_processTerminate): Ditto (:s/Init/Destroy/).
+
+ * dll.c (dllMain): Move call to FreeLibrary() so that it is only
+ called once when the process detaches.
+
+ * mutex.c (_mutex_check_need_init): New static function to test
+ and init PTHREAD_MUTEX_INITIALIZER mutexes. Provides serialised
+ access to the internal state of the uninitialised static mutex.
+ Called from pthread_mutex_trylock() and pthread_mutex_lock() which
+ do a quick unguarded test to check if _mutex_check_need_init()
+ needs to be called. This is safe as the test is conservative
+ and is repeated inside the guarded section of
+ _mutex_check_need_init(). Thus in all calls except the first
+ calls to lock static mutexes, the additional overhead to lock any
+ mutex is a single memory fetch and test for zero.
+
+ * pthread.h (pthread_mutex_t_): Add 'staticinit' member. Mutexes
+ initialised by PTHREAD_MUTEX_INITIALIZER aren't really initialised
+ until the first attempt to lock it. Using the 'valid'
+ flag (which flags the mutex as destroyed or not) to record this
+ information would be messy. It is possible for a statically
+ initialised mutex such as this to be destroyed before ever being
+ used.
+
+ * mutex.c (pthread_mutex_trylock): Call _mutex_check_need_init()
+ to test/init PTHREAD_MUTEX_INITIALIZER mutexes.
+ (pthread_mutex_lock): Ditto.
+ (pthread_mutex_unlock): Add check to ensure we don't try to unlock
+ an unitialised static mutex.
+ (pthread_mutex_destroy): Add check to ensure we don't try to delete
+ a critical section that we never created. Allows us to destroy
+ a static mutex that has never been locked (and hence initialised).
+ (pthread_mutex_init): Set 'staticinit' flag to 0 for the new mutex.
+
+Sun Jan 17 12:01:26 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * private.c (ptw32_sem_timedwait): Move from semaphore.c.
+
+ * semaphore.c : Remove redundant #includes.
+ (ptw32_sem_timedwait): Move to private.c.
+ (sem_wait): Add missing abstime arg to pthreadCancelableWait() call.
+
+Fri Jan 15 23:38:05 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * condvar.c (cond_timedwait): Remove comment.
+
+Fri Jan 15 15:41:28 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * pthread.h: Add new 'abstime' arg to pthreadCancelableWait()
+ prototype.
+
+ * condvar.c (cond_timedwait): New generalised function called by
+ both pthread_cond_wait() and pthread_cond_timedwait(). This is
+ essentially pthread_cond_wait() renamed and modified to add the
+ 'abstime' arg and call the new ptw32_sem_timedwait() instead of
+ sem_wait().
+ (pthread_cond_wait): Now just calls the internal static
+ function cond_timedwait() with an INFINITE wait.
+ (pthread_cond_timedwait): Now implemented. Calls the internal
+ static function cond_timedwait().
+
+ * implement.h (ptw32_sem_timedwait): New internal function
+ prototype.
+
+ * misc.c (pthreadCancelableWait): Added new 'abstime' argument
+ to allow shorter than INFINITE wait.
+
+ * semaphore.c (ptw32_sem_timedwait): New function for internal
+ use. This is essentially sem_wait() modified to add the
+ 'abstime' arg and call the modified (see above)
+ pthreadCancelableWait().
+
+Thu Jan 14 14:27:13 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * cleanup.c: Correct _cplusplus to __cplusplus wherever used.
+
+ * Makefile.in: Add CC=g++ and add -fhandle-exceptions to CFLAGS.
+ The derived Makefile will compile all units of the package as C++
+ so that those which include try/catch exception handling should work
+ properly. The package should compile ok if CC=gcc, however, exception
+ handling will not be included and thus thread cancellation, for
+ example, will not work.
+
+ * cleanup.c (ptw32_pop_cleanup): Add #warning to compile this
+ file as C++ if using a cygwin32 environment. Perhaps the whole package
+ should be compiled using g++ under cygwin.
+
+ * private.c (ptw32_threadStart): Change #error directive
+ into #warning and bracket for __CYGWIN__ and derivative compilers.
+
+Wed Jan 13 09:34:52 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * build.bat: Delete old binaries before compiling/linking.
+
+Tue Jan 12 09:58:38 1999 Tor Lillqvist <tml at iki.fi>
+
+ * dll.c: The Microsoft compiler pragmas probably are more
+ appropriately protected by _MSC_VER than by _WIN32.
+
+ * pthread.h: Define ETIMEDOUT. This should be returned by
+ pthread_cond_timedwait which is not implemented yet as of
+ snapshot-1999-01-04-1305. It was implemented in the older version.
+ The Microsoft compiler pragmas probably are more appropriately
+ protected by _MSC_VER than by _WIN32.
+
+ * pthread.def: pthread_mutex_destroy was missing from the def file
+
+ * condvar.c (pthread_cond_broadcast): Ensure we only wait on threads
+ if there were any waiting on the condition.
+ I think pthread_cond_broadcast should do the WaitForSingleObject
+ only if cv->waiters > 0? Otherwise it seems to hang, at least in the
+ testg thread program from glib.
+
+Tue Jan 12 09:58:38 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * condvar.c (pthread_cond_timedwait): Fix function description
+ comments.
+
+ * semaphore.c (sem_post): Correct typo in comment.
+
+Mon Jan 11 20:33:19 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.h: Re-arrange conditional compile of pthread_cleanup-*
+ macros.
+
+ * cleanup.c (ptw32_push_cleanup): Provide conditional
+ compile of cleanup->prev.
+
+1999-01-11 Tor Lillqvist <tml at iki.fi>
+
+ * condvar.c (pthread_cond_init): Invert logic when testing the
+ return value from calloc().
+
+Sat Jan 9 14:32:08 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h: Compile-time switch for CYGWIN derived environments
+ to use CreateThread instead of _beginthreadex. Ditto for ExitThread.
+ Patch provided by Anders Norlander <anorland at hem2.passagen.se>.
+
+Tue Jan 5 16:33:04 1999 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * cleanup.c (ptw32_pop_cleanup): Add C++ version of __try/__except
+ block. Move trailing "}" out of #ifdef _WIN32 block left there by
+ (rpj's) mistake.
+
+ * private.c: Remove #include <errno.h> which is included by pthread.h.
+
+1998-12-11 Ben Elliston <bje at toilet.to.cygnus.com>
+
+ * README: Update info about subscribing to the mailing list.
+
+Mon Jan 4 11:23:40 1999 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * all: No code changes, just cleanup.
+ - remove #if 0 /* Pre Bossom */ enclosed code.
+ - Remove some redundant #includes.
+ * pthread.h: Update implemented/unimplemented routines list.
+ * Tag the bossom merge branch getting ready to merge back to main
+ trunk.
+
+Tue Dec 29 13:11:16 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h: Move the following struct definitions to pthread.h:
+ pthread_t_, pthread_attr_t_, pthread_mutex_t_, pthread_mutex_t_,
+ pthread_mutexattr_t_, pthread_key_t_, pthread_cond_t_,
+ pthread_condattr_t_, pthread_once_t_.
+
+ * pthread.h: Add "_" prefix to pthread_push_cleanup and
+ pthread_pop_cleanup internal routines, and associated struct and
+ typedefs.
+
+ * buildlib.bat: Add compile command for semaphore.c
+
+ * pthread.def: Comment out pthread_atfork routine name.
+ Now unimplemented.
+
+ * tsd.c (pthread_setspecific): Rename tkAssocCreate to
+ ptw32_tkAssocCreate.
+ (pthread_key_delete): Rename tkAssocDestroy to
+ ptw32_tkAssocDestroy.
+
+ * sync.c (pthread_join): Rename threadDestroy to ptw32_threadDestroy
+
+ * sched.c (is_attr): attr is now **attr (was *attr), so add extra
+ NULL pointer test.
+ (pthread_attr_setschedparam): Increase redirection for attr which is
+ now a **.
+ (pthread_attr_getschedparam): Ditto.
+ (pthread_setschedparam): Change thread validation and rename "thread"
+ Win32 thread Handle element name to match John Bossom's version.
+ (pthread_getschedparam): Ditto.
+
+ * private.c (ptw32_threadDestroy): Rename call to
+ callUserDestroyRoutines() as ptw32_callUserDestroyRoutines()
+
+ * misc.c: Add #include "implement.h".
+
+ * dll.c: Remove defined(KLUDGE) wrapped code.
+
+ * fork.c: Remove redefinition of ENOMEM.
+ Remove pthread_atfork() and fork() with #if 0/#endif.
+
+ * create.c (pthread_create): Rename threadStart and threadDestroy calls
+ to ptw32_threadStart and ptw32_threadDestroy.
+
+ * implement.h: Rename "detachedstate" to "detachstate".
+
+ * attr.c: Rename "detachedstate" to "detachstate".
+
+Mon Dec 28 09:54:39 1998 John Bossom
+
+ * semaphore.c: Initial version.
+ * semaphore.h: Initial version.
+
+Mon Dec 28 09:54:39 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.h (pthread_attr_t_): Change to *pthread_attr_t.
+
+Mon Dec 28 09:54:39 1998 John Bossom, Ben Elliston
+
+ * attr.c (pthread_attr_setstacksize): Merge with John's version.
+ (pthread_attr_getstacksize): Merge with John's version.
+ (pthread_attr_setstackaddr): Merge with John's version.
+ (pthread_attr_getstackaddr): Merge with John's version.
+ (pthread_attr_init): Merge with John's version.
+ (pthread_attr_destroy): Merge with John's version.
+ (pthread_attr_getdetachstate): Merge with John's version.
+ (pthread_attr_setdetachstate): Merge with John's version.
+ (is_attr): attr is now **attr (was *attr), so add extra NULL pointer
+ test.
+
+Mon Dec 28 09:54:39 1998 Ross Johnson
+
+ * implement.h (pthread_attr_t_): Add and rename elements in JEB's
+ version to correspond to original, so that it can be used with
+ original attr routines.
+
+ * pthread.h: Add #endif at end which was truncated in merging.
+
+Sun Dec 20 14:51:58 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * misc.c (pthreadCancelableWait): New function by John Bossom. Non-standard
+ but provides a hook that can be used to implement cancellation points in
+ applications that use this library.
+
+ * pthread.h (pthread_cleanup_pop): C++ (non-WIN32) version uses
+ try/catch to emulate John Bossom's WIN32 __try/__finally behaviour.
+ In the WIN32 version __finally block, add a test for AbnormalTermination otherwise
+ cleanup is only run if the cleanup_pop execute arg is non-zero. Cancellation
+ should cause the cleanup to run irrespective of the execute arg.
+
+ * condvar.c (pthread_condattr_init): Replaced by John Bossom's version.
+ (pthread_condattr_destroy): Replaced by John Bossom's version.
+ (pthread_condattr_getpshared): Replaced by John Bossom's version.
+ (pthread_condattr_setpshared): Replaced by John Bossom's version.
+ (pthread_cond_init): Replaced by John Bossom's version.
+ Fix comment (refered to mutex rather than condition variable).
+ (pthread_cond_destroy): Replaced by John Bossom's version.
+ (pthread_cond_wait): Replaced by John Bossom's version.
+ (pthread_cond_timedwait): Replaced by John Bossom's version.
+ (pthread_cond_signal): Replaced by John Bossom's version.
+ (pthread_cond_broadcast): Replaced by John Bossom's version.
+
+Thu Dec 17 19:10:46 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * tsd.c (pthread_key_create): Replaced by John Bossom's version.
+ (pthread_key_delete): Replaced by John Bossom's version.
+ (pthread_setspecific): Replaced by John Bossom's version.
+ (pthread_getspecific): Replaced by John Bossom's version.
+
+Mon Dec 7 09:44:40 1998 John Bossom
+
+ * cancel.c (pthread_setcancelstate): Replaced.
+ (pthread_setcanceltype): Replaced.
+ (pthread_testcancel): Replaced.
+ (pthread_cancel): Replaced.
+
+ * exit.c (pthread_exit): Replaced.
+
+ * misc.c (pthread_self): Replaced.
+ (pthread_equal): Replaced.
+
+ * sync.c (pthread_detach): Replaced.
+ (pthread_join): Replaced.
+
+ * create.c (pthread_create): Replaced.
+
+ * private.c (ptw32_processInitialize): New.
+ (ptw32_processTerminate): New.
+ (ptw32_threadStart): New.
+ (ptw32_threadDestroy): New.
+ (ptw32_cleanupStack): New.
+ (ptw32_tkAssocCreate): New.
+ (ptw32_tkAssocDestroy): New.
+ (ptw32_callUserDestroyRoutines): New.
+
+ * implement.h: Added non-API structures and declarations.
+
+ * dll.c (PthreadsEntryPoint): Cast return value of GetProcAddress
+ to resolve compile warning from MSVC.
+
+ * dll.c (DLLmain): Replaced.
+ * dll.c (PthreadsEntryPoint):
+ Re-applied Anders Norlander's patch:-
+ Initialize ptw32_try_enter_critical_section at startup
+ and release kernel32 handle when DLL is being unloaded.
+
+Sun Dec 6 21:54:35 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * buildlib.bat: Fix args to CL when building the .DLL
+
+ * cleanup.c (ptw32_destructor_run_all): Fix TSD key management.
+ This is a tidy-up before TSD and Thread management is completely
+ replaced by John Bossom's code.
+
+ * tsd.c (pthread_key_create): Fix TSD key management.
+
+ * global.c (ptw32_key_virgin_next): Initialise.
+
+ * build.bat: New DOS script to compile and link a pthreads app
+ using Microsoft's CL compiler linker.
+ * buildlib.bat: New DOS script to compile all the object files
+ and create pthread.lib and pthread.dll using Microsoft's CL
+ compiler linker.
+
+1998-12-05 Anders Norlander <anorland at hem2.passagen.se>
+
+ * implement.h (ptw32_try_enter_critical_section): New extern
+ * dll.c (ptw32_try_enter_critical_section): New pointer to
+ TryEnterCriticalSection if it exists; otherwise NULL.
+ * dll.c (PthreadsEntryPoint):
+ Initialize ptw32_try_enter_critical_section at startup
+ and release kernel32 handle when DLL is being unloaded.
+ * mutex.c (pthread_mutex_trylock): Replaced check for NT with
+ a check if ptw32_try_enter_critical_section is valid
+ pointer to a function. Call ptw32_try_enter_critical_section
+ instead of TryEnterCriticalSection to avoid errors on Win95.
+
+Thu Dec 3 13:32:00 1998 Ross Johnson <rpj at ise.canberra.edu.au>
+
+ * README: Correct cygwin32 compatibility statement.
+
+Sun Nov 15 21:24:06 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * cleanup.c (ptw32_destructor_run_all): Declare missing void * arg.
+ Fixup CVS merge conflicts.
+
+1998-10-30 Ben Elliston <bje at cygnus.com>
+
+ * condvar.c (cond_wait): Fix semantic error. Test for equality
+ instead of making an assignment.
+
+Fri Oct 30 15:15:50 1998 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * cleanup.c (ptw32_handler_push): Fixed bug appending new
+ handler to list reported by Peter Slacik
+ <Peter.Slacik at leibinger.freinet.de>.
+ (new_thread): Rename poorly named local variable to
+ "new_handler".
+
+Sat Oct 24 18:34:59 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * global.c: Add TSD key management array and index declarations.
+
+ * implement.h: Ditto for externs.
+
+Fri Oct 23 00:08:09 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h (PTW32_TSD_KEY_REUSE): Add enum.
+
+ * private.c (ptw32_delete_thread): Add call to
+ ptw32_destructor_run_all() to clean up the threads keys.
+
+ * cleanup.c (ptw32_destructor_run_all): Check for no more dirty
+ keys to run destructors on. Assume that the destructor call always
+ succeeds and set the key value to NULL.
+
+Thu Oct 22 21:44:44 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * tsd.c (pthread_setspecific): Add key management code.
+ (pthread_key_create): Ditto.
+ (pthread_key_delete): Ditto.
+
+ * implement.h (struct ptw32_tsd_key): Add status member.
+
+ * tsd.c: Add description of pthread_key_delete() from the
+ standard as a comment.
+
+Fri Oct 16 17:38:47 1998 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * cleanup.c (ptw32_destructor_run_all): Fix and improve
+ stepping through the key table.
+
+Thu Oct 15 14:05:01 1998 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * private.c (ptw32_new_thread): Remove init of destructorstack.
+ No longer an element of pthread_t.
+
+ * tsd.c (pthread_setspecific): Fix type declaration and cast.
+ (pthread_getspecific): Ditto.
+ (pthread_getspecific): Change error return value to NULL if key
+ is not in use.
+
+Thu Oct 15 11:53:21 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * global.c (ptw32_tsd_key_table): Fix declaration.
+
+ * implement.h(ptw32_TSD_keys_TlsIndex): Add missing extern.
+ (ptw32_tsd_mutex): Ditto.
+
+ * create.c (ptw32_start_call): Fix "keys" array declaration.
+ Add comment.
+
+ * tsd.c (pthread_setspecific): Fix type declaration and cast.
+ (pthread_getspecific): Ditto.
+
+ * cleanup.c (ptw32_destructor_run_all): Declare missing loop
+ counter.
+
+Wed Oct 14 21:09:24 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * private.c (ptw32_new_thread): Increment ptw32_threads_count.
+ (ptw32_delete_thread): Decrement ptw32_threads_count.
+ Remove some comments.
+
+ * exit.c (ptw32_exit): : Fix two pthread_mutex_lock() calls that
+ should have been pthread_mutex_unlock() calls.
+ (ptw32_vacuum): Remove call to ptw32_destructor_pop_all().
+
+ * create.c (pthread_create): Fix two pthread_mutex_lock() calls that
+ should have been pthread_mutex_unlock() calls.
+
+ * global.c (ptw32_tsd_mutex): Add mutex for TSD operations.
+
+ * tsd.c (pthread_key_create): Add critical section.
+ (pthread_setspecific): Ditto.
+ (pthread_getspecific): Ditto.
+ (pthread_key_delete): Ditto.
+
+ * sync.c (pthread_join): Fix two pthread_mutex_lock() calls that
+ should have been pthread_mutex_unlock() calls.
+
+Mon Oct 12 00:00:44 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h (ptw32_tsd_key_table): New.
+
+ * create.c (ptw32_start_call): Initialise per-thread TSD keys
+ to NULL.
+
+ * misc.c (pthread_once): Correct typo in comment.
+
+ * implement.h (ptw32_destructor_push): Remove.
+ (ptw32_destructor_pop): Remove.
+ (ptw32_destructor_run_all): Rename from ptw32_destructor_pop_all.
+ (PTW32_TSD_KEY_DELETED): Add enum.
+ (PTW32_TSD_KEY_INUSE): Add enum.
+
+ * cleanup.c (ptw32_destructor_push): Remove.
+ (ptw32_destructor_pop): Remove.
+ (ptw32_destructor_run_all): Totally revamped TSD.
+
+ * dll.c (ptw32_TSD_keys_TlsIndex): Initialise.
+
+ * tsd.c (pthread_setspecific): Totally revamped TSD.
+ (pthread_getspecific): Ditto.
+ (pthread_create): Ditto.
+ (pthread_delete): Ditto.
+
+Sun Oct 11 22:44:55 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * global.c (ptw32_tsd_key_table): Add new global.
+
+ * implement.h (ptw32_tsd_key_t and struct ptw32_tsd_key):
+ Add.
+ (struct _pthread): Remove destructorstack.
+
+ * cleanup.c (ptw32_destructor_run_all): Rename from
+ ptw32_destructor_pop_all. The key destructor stack was made
+ global rather than per-thread. No longer removes destructor nodes
+ from the stack. Comments updated.
+
+1998-10-06 Ben Elliston <bje at cygnus.com>
+
+ * condvar.c (cond_wait): Use POSIX, not Win32 mutex calls.
+ (pthread_cond_broadcast): Likewise.
+ (pthread_cond_signal): Likewise.
+
+1998-10-05 Ben Elliston <bje at cygnus.com>
+
+ * pthread.def: Update. Some functions aren't available yet, others
+ are macros in <pthread.h>.
+
+ * tests/join.c: Remove; useless.
+
+Mon Oct 5 14:25:08 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * pthread.def: New file for building the DLL.
+
+1998-10-05 Ben Elliston <bje at cygnus.com>
+
+ * misc.c (pthread_equal): Correct inverted logic bug.
+ (pthread_once): Use the POSIX mutex primitives, not Win32. Remove
+ irrelevant FIXME comment.
+
+ * global.c (PTHREAD_MUTEX_INITIALIZER): Move to pthread.h.
+
+ * pthread.h (PTHREAD_MUTEX_INITIALIZER): Define.
+ (pthread_mutex_t): Reimplement as a struct containing a valid
+ flag. If the flag is ever down upon entry to a mutex operation,
+ we call pthread_mutex_create() to initialise the object. This
+ fixes the problem of how to handle statically initialised objects
+ that can't call InitializeCriticalSection() due to their context.
+ (PTHREAD_ONCE_INIT): Define.
+
+ * mutex.c (pthread_mutex_init): Set valid flag.
+ (pthread_mutex_destroy): Clear valid flag.
+ (pthread_mutex_lock): Check and handle the valid flag.
+ (pthread_mutex_unlock): Likewise.
+ (pthread_mutex_trylock): Likewise.
+
+ * tests/mutex3.c: New file; test for the static initialisation
+ macro. Passes.
+
+ * tests/create1.c: New file; test pthread_create(). Passes.
+
+ * tests/equal.c: Poor test; remove.
+
+ * tests/equal1.c New file; test pthread_equal(). Passes.
+
+ * tests/once1.c: New file; test for pthread_once(). Passes.
+
+ * tests/self.c: Remove; rename to self1.c.
+
+ * tests/self1.c: This is the old self.c.
+
+ * tests/self2.c: New file. Test pthread_self() with a single
+ thread. Passes.
+
+ * tests/self3.c: New file. Test pthread_self() with a couple of
+ threads to ensure their thread IDs differ. Passes.
+
+1998-10-04 Ben Elliston <bje at cygnus.com>
+
+ * tests/mutex2.c: Test pthread_mutex_trylock(). Passes.
+
+ * tests/mutex1.c: New basic test for mutex functions (it passes).
+ (main): Eliminate warning.
+
+ * configure.in: Test for __stdcall, not _stdcall. Typo.
+
+ * configure: Regenerate.
+
+ * attr.c (pthread_attr_setstackaddr): Remove FIXME comment. Win32
+ does know about ENOSYS after all.
+ (pthread_attr_setstackaddr): Likewise.
+
+1998-10-03 Ben Elliston <bje at cygnus.com>
+
+ * configure.in: Test for the `_stdcall' keyword. Define `STDCALL'
+ to `_stdcall' if we have it, null otherwise.
+
+ * configure: Regenerate.
+
+ * acconfig.h (STDCALL): New define.
+
+ * config.h.in: Regenerate.
+
+ * create.c (ptw32_start_call): Add STDCALL prefix.
+
+ * mutex.c (pthread_mutex_init): Correct function signature.
+
+ * attr.c (pthread_attr_init): Only zero out the `sigmask' member
+ if we have the sigset_t type.
+
+ * pthread.h: No need to include <unistd.h>. It doesn't even exist
+ on Win32! Again, an artifact of cross-compilation.
+ (pthread_sigmask): Only provide if we have the sigset_t type.
+
+ * process.h: Remove. This was a stand-in before we started doing
+ native compilation under Win32.
+
+ * pthread.h (pthread_mutex_init): Make `attr' argument const.
+
+1998-10-02 Ben Elliston <bje at cygnus.com>
+
+ * COPYING: Remove.
+
+ * COPYING.LIB: Add. This library is under the LGPL.
+
+1998-09-13 Ben Elliston <bje at cygnus.com>
+
+ * configure.in: Test for required system features.
+
+ * configure: Generate.
+
+ * acconfig.h: New file.
+
+ * config.h.in: Generate.
+
+ * Makefile.in: Renamed from Makefile.
+
+ * COPYING: Import from a recent GNU package.
+
+ * config.guess: Likewise.
+
+ * config.sub: Likewise.
+
+ * install-sh: Likewise.
+
+ * config.h: Remove.
+
+ * Makefile: Likewise.
+
+1998-09-12 Ben Elliston <bje at cygnus.com>
+
+ * windows.h: No longer needed; remove.
+
+ * windows.c: Likewise.
+
+Sat Sep 12 20:09:24 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * windows.h: Remove error number definitions. These are in <errno.h>
+
+ * tsd.c: Add comment explaining rationale for not building
+ POSIX TSD on top of Win32 TLS.
+
+1998-09-12 Ben Elliston <bje at cygnus.com>
+
+ * {most}.c: Include <errno.h> to get POSIX error values.
+
+ * signal.c (pthread_sigmask): Only provide if HAVE_SIGSET_T is
+ defined.
+
+ * config.h: #undef features, don't #define them. This will be
+ generated by autoconf very soon.
+
+1998-08-11 Ben Elliston <bje at cygnus.com>
+
+ * Makefile (LIB): Define.
+ (clean): Define target.
+ (all): Build a library not just the object files.
+
+ * pthread.h: Provide a definition for struct timespec if we don't
+ already have one.
+
+ * windows.c (TlsGetValue): Bug fix.
+
+Thu Aug 6 15:19:22 1998 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * misc.c (pthread_once): Fix arg 1 of EnterCriticalSection()
+ and LeaveCriticalSection() calls to pass address-of lock.
+
+ * fork.c (pthread_atfork): Typecast (void (*)(void *)) funcptr
+ in each ptw32_handler_push() call.
+
+ * exit.c (ptw32_exit): Fix attr arg in
+ pthread_attr_getdetachstate() call.
+
+ * private.c (ptw32_new_thread): Typecast (HANDLE) NULL.
+ (ptw32_delete_thread): Ditto.
+
+ * implement.h: (PTW32_MAX_THREADS): Add define. This keeps
+ changing in an attempt to make thread administration data types
+ opaque and cleanup DLL startup.
+
+ * dll.c (PthreadsEntryPoint):
+ (ptw32_virgins): Remove malloc() and free() calls.
+ (ptw32_reuse): Ditto.
+ (ptw32_win32handle_map): Ditto.
+ (ptw32_threads_mutex_table): Ditto.
+
+ * global.c (_POSIX_THREAD_THREADS_MAX): Initialise with
+ PTW32_MAX_THREADS.
+ (ptw32_virgins): Ditto.
+ (ptw32_reuse): Ditto.
+ (ptw32_win32handle_map): Ditto.
+ (ptw32_threads_mutex_table): Ditto.
+
+ * create.c (pthread_create): Typecast (HANDLE) NULL.
+ Typecast (unsigned (*)(void *)) start_routine.
+
+ * condvar.c (pthread_cond_init): Add address-of operator & to
+ arg 1 of pthread_mutex_init() call.
+ (pthread_cond_destroy): Add address-of operator & to
+ arg 1 of pthread_mutex_destroy() call.
+
+ * cleanup.c (ptw32_destructor_pop_all): Add (int) cast to
+ pthread_getspecific() arg.
+ (ptw32_destructor_pop): Add (void *) cast to "if" conditional.
+ (ptw32_destructor_push): Add (void *) cast to
+ ptw32_handler_push() "key" arg.
+ (malloc.h): Add include.
+
+ * implement.h (ptw32_destructor_pop): Add prototype.
+
+ * tsd.c (implement.h): Add include.
+
+ * sync.c (pthread_join): Remove target_thread_mutex and it's
+ initialisation. Rename getdetachedstate to getdetachstate.
+ Remove unused variable "exitcode".
+ (pthread_detach): Remove target_thread_mutex and it's
+ initialisation. Rename getdetachedstate to getdetachstate.
+ Rename setdetachedstate to setdetachstate.
+
+ * signal.c (pthread_sigmask): Rename SIG_SET to SIG_SETMASK.
+ Cast "set" to (long *) in assignment to passify compiler warning.
+ Add address-of operator & to thread->attr.sigmask in memcpy() call
+ and assignment.
+ (pthread_sigmask): Add address-of operator & to thread->attr.sigmask
+ in memcpy() call and assignment.
+
+ * windows.h (THREAD_PRIORITY_ERROR_RETURN): Add.
+ (THREAD_PRIORITY_LOWEST): Add.
+ (THREAD_PRIORITY_HIGHEST): Add.
+
+ * sched.c (is_attr): Add function.
+ (implement.h): Add include.
+ (pthread_setschedparam): Rename all instances of "sched_policy"
+ to "sched_priority".
+ (pthread_getschedparam): Ditto.
+
+Tue Aug 4 16:57:58 1998 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * private.c (ptw32_delete_thread): Fix typo. Add missing ';'.
+
+ * global.c (ptw32_virgins): Change types from pointer to
+ array pointer.
+ (ptw32_reuse): Ditto.
+ (ptw32_win32handle_map): Ditto.
+ (ptw32_threads_mutex_table): Ditto.
+
+ * implement.h(ptw32_virgins): Change types from pointer to
+ array pointer.
+ (ptw32_reuse): Ditto.
+ (ptw32_win32handle_map): Ditto.
+ (ptw32_threads_mutex_table): Ditto.
+
+ * private.c (ptw32_delete_thread): Fix "entry" should be "thread".
+
+ * misc.c (pthread_self): Add extern for ptw32_threadID_TlsIndex.
+
+ * global.c: Add comment.
+
+ * misc.c (pthread_once): Fix member -> dereferences.
+ Change ptw32_once_flag to once_control->flag in "if" test.
+
+Tue Aug 4 00:09:30 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h(ptw32_virgins): Add extern.
+ (ptw32_virgin_next): Ditto.
+ (ptw32_reuse): Ditto.
+ (ptw32_reuse_top): Ditto.
+ (ptw32_win32handle_map): Ditto.
+ (ptw32_threads_mutex_table): Ditto.
+
+ * global.c (ptw32_virgins): Changed from array to pointer.
+ Storage allocation for the array moved into dll.c.
+ (ptw32_reuse): Ditto.
+ (ptw32_win32handle_map): Ditto.
+ (ptw32_threads_mutex_table): Ditto.
+
+ * dll.c (PthreadsEntryPoint): Set up thread admin storage when
+ DLL is loaded.
+
+ * fork.c (pthread_atfork): Fix function pointer arg to all
+ ptw32_handler_push() calls. Change "arg" arg to NULL in child push.
+
+ * exit.c: Add windows.h and process.h includes.
+ (ptw32_exit): Add local detachstate declaration.
+ (ptw32_exit): Fix incorrect name for pthread_attr_getdetachstate().
+
+ * pthread.h (_POSIX_THREAD_ATTR_STACKSIZE): Move from global.c
+ (_POSIX_THREAD_ATTR_STACKADDR): Ditto.
+
+ * create.c (pthread_create): Fix #if should be #ifdef.
+ (ptw32_start_call): Remove usused variables.
+
+ * process.h: Create.
+
+ * windows.h: Move _beginthreadex and _endthreadex into
+ process.h
+
+Mon Aug 3 21:19:57 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * condvar.c (pthread_cond_init): Add NULL attr to
+ pthread_mutex_init() call - default attributes will be used.
+ (cond_wait): Fix typo.
+ (cond_wait): Fix typo - cv was ev.
+ (pthread_cond_broadcast): Fix two identical typos.
+
+ * cleanup.c (ptw32_destructor_pop_all): Remove _ prefix from
+ PTHREAD_DESTRUCTOR_ITERATIONS.
+
+ * pthread.h: Move _POSIX_* values into posix.h
+
+ * pthread.h: Fix typo in pthread_mutex_init() prototype.
+
+ * attr.c (pthread_attr_init): Fix error in priority member init.
+
+ * windows.h (THREAD_PRIORITY_NORMAL): Add.
+
+ * pthread.h (sched_param): Add missing ';' to struct definition.
+
+ * attr.c (pthread_attr_init): Remove obsolete pthread_attr_t
+ member initialisation - cancelstate, canceltype, cancel_pending.
+ (is_attr): Make arg "attr" a const.
+
+ * implement.h (PTW32_HANDLER_POP_LIFO): Remove definition.
+ (PTW32_HANDLER_POP_FIFO): Ditto.
+ (PTW32_VALID): Add missing newline escape (\).
+ (ptw32_handler_node): Make element "next" a pointer.
+
+1998-08-02 Ben Elliston <bje at cygnus.com>
+
+ * windows.h: Remove duplicate TlsSetValue() prototype. Add
+ TlsGetValue() prototype.
+ (FALSE): Define.
+ (TRUE): Likewise.
+ Add forgotten errno values. Guard against multiple #includes.
+
+ * windows.c: New file. Implement stubs for Win32 functions.
+
+ * Makefile (SRCS): Remove. Not explicitly needed.
+ (CFLAGS): Add -Wall for all warnings with GCC.
+
+Sun Aug 2 19:03:42 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * config.h: Create. This is a temporary stand-in for autoconf yet
+ to be done.
+ (HAVE_SIGNAL_H): Add.
+
+ * pthread.h: Minor rearrangement for temporary config.h.
+
+Fri Jul 31 14:00:29 1998 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * cleanup.c (ptw32_destructor_pop): Implement. Removes
+ destructors associated with a key without executing them.
+ (ptw32_destructor_pop_all): Add FIXME comment.
+
+ * tsd.c (pthread_key_delete): Add call to ptw32_destructor_pop().
+
+Fri Jul 31 00:05:45 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * tsd.c (pthread_key_create): Update to properly associate
+ the destructor routine with the key.
+ (pthread_key_delete): Add FIXME comment.
+
+ * exit.c (ptw32_vacuum): Add call to
+ ptw32_destructor_pop_all().
+
+ * implement.h (ptw32_handler_pop_all): Add prototype.
+ (ptw32_destructor_pop_all): Ditto.
+
+ * cleanup.c (ptw32_destructor_push): Implement. This is just a
+ call to ptw32_handler_push().
+ (ptw32_destructor_pop_all): Implement. This is significantly
+ different to ptw32_handler_pop_all().
+
+ * Makefile (SRCS): Create. Preliminary.
+
+ * windows.h: Create. Contains Win32 definitions for compile
+ testing. This is just a standin for the real one.
+
+ * pthread.h (SIG_UNBLOCK): Fix typo. Was SIG_BLOCK.
+ (windows.h): Add include. Required for CRITICAL_SECTION.
+ (pthread_cond_t): Move enum declaration outside of struct
+ definition.
+ (unistd.h): Add include - may be temporary.
+
+ * condvar.c (windows.h): Add include.
+
+ * implement.h (PTW32_THIS): Remove - no longer required.
+ (PTW32_STACK): Use pthread_self() instead of PTW32_THIS.
+
+Thu Jul 30 23:12:45 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h: Remove ptw32_find_entry() prototype.
+
+ * private.c: Extend comments.
+ Remove ptw32_find_entry() - no longer needed.
+
+ * create.c (ptw32_start_call): Add call to TlsSetValue() to
+ store the thread ID.
+
+ * dll.c (PthreadsEntryPoint): Implement. This is called
+ whenever a process loads the DLL. Used to initialise thread
+ local storage.
+
+ * implement.h: Add ptw32_threadID_TlsIndex.
+ Add ()s around PTW32_VALID expression.
+
+ * misc.c (pthread_self): Re-implement using Win32 TLS to store
+ the threads own ID.
+
+Wed Jul 29 11:39:03 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * private.c: Corrections in comments.
+ (ptw32_new_thread): Alter "if" flow to be more natural.
+
+ * cleanup.c (ptw32_handler_push): Same as below.
+
+ * create.c (pthread_create): Same as below.
+
+ * private.c (ptw32_new_thread): Rename "new" to "new_thread".
+ Since when has a C programmer been required to know C++?
+
+Tue Jul 28 14:04:29 1998 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * implement.h: Add PTW32_VALID macro.
+
+ * sync.c (pthread_join): Modify to use the new thread
+ type and ptw32_delete_thread(). Rename "target" to "thread".
+ Remove extra local variable "target".
+ (pthread_detach): Ditto.
+
+ * signal.c (pthread_sigmask): Move init of "us" out of inner block.
+ Fix instance of "this" should have been "us". Rename "us" to "thread".
+
+ * sched.c (pthread_setschedparam): Modify to use the new thread
+ type.
+ (pthread_getschedparam): Ditto.
+
+ * private.c (ptw32_find_thread): Fix return type and arg.
+
+ * implement.h: Remove PTW32_YES and PTW32_NO.
+ (ptw32_new_thread): Add prototype.
+ (ptw32_find_thread): Ditto.
+ (ptw32_delete_thread): Ditto.
+ (ptw32_new_thread_entry): Remove prototype.
+ (ptw32_find_thread_entry): Ditto.
+ (ptw32_delete_thread_entry): Ditto.
+ ( PTW32_NEW, PTW32_INUSE, PTW32_EXITED, PTW32_REUSE):
+ Add.
+
+
+ * create.c (pthread_create): Minor rename "us" to "new" (I need
+ these cues but it doesn't stop me coming out with some major bugs
+ at times).
+ Load start_routine and arg into the thread so the wrapper can
+ call it.
+
+ * exit.c (pthread_exit): Fix pthread_this should be pthread_self.
+
+ * cancel.c (pthread_setcancelstate): Change
+ ptw32_threads_thread_t * to pthread_t and init with
+ pthread_this().
+ (pthread_setcanceltype): Ditto.
+
+ * exit.c (ptw32_exit): Add new pthread_t arg.
+ Rename ptw32_delete_thread_entry to ptw32_delete_thread.
+ Rename "us" to "thread".
+ (pthread_exit): Call ptw32_exit with added thread arg.
+
+ * create.c (ptw32_start_call): Insert missing ")".
+ Add "us" arg to ptw32_exit() call.
+ (pthread_create): Modify to use new thread allocation scheme.
+
+ * private.c: Added detailed explanation of the new thread
+ allocation scheme.
+ (ptw32_new_thread): Totally rewritten to use
+ new thread allocation scheme.
+ (ptw32_delete_thread): Ditto.
+ (ptw32_find_thread): Obsolete.
+
+Mon Jul 27 17:46:37 1998 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * create.c (pthread_create): Start of rewrite. Not completed yet.
+
+ * private.c (ptw32_new_thread_entry): Start of rewrite. Not
+ complete.
+
+ * implement.h (ptw32_threads_thread): Rename, remove thread
+ member, add win32handle and ptstatus members.
+ (ptw32_t): Add.
+
+ * pthread.h: pthread_t is no longer mapped directly to a Win32
+ HANDLE type. This is so we can let the Win32 thread terminate and
+ reuse the HANDLE while pthreads holds it's own thread ID until
+ the last waiting join exits.
+
+Mon Jul 27 00:20:37 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * private.c (ptw32_delete_thread_entry): Destroy the thread
+ entry attribute object before deleting the thread entry itself.
+
+ * attr.c (pthread_attr_init): Initialise cancel_pending = FALSE.
+ (pthread_attr_setdetachstate): Rename "detached" to "detachedstate".
+ (pthread_attr_getdetachstate): Ditto.
+
+ * exit.c (ptw32_exit): Fix incorrect check for detachedstate.
+
+ * implement.h (ptw32_call_t): Remove env member.
+
+Sun Jul 26 13:06:12 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h (ptw32_new_thread_entry): Fix prototype.
+ (ptw32_find_thread_entry): Ditto.
+ (ptw32_delete_thread_entry): Ditto.
+ (ptw32_exit): Add prototype.
+
+ * exit.c (ptw32_exit): New function. Called from pthread_exit()
+ and ptw32_start_call() to exit the thread. It allows an extra
+ argument which is the return code passed to _endthreadex().
+ (ptw32_exit): Move thread entry delete call from ptw32_vacuum()
+ into here. Add more explanation of thread entry deletion.
+ (ptw32_exit): Clarify comment.
+
+ * create.c (ptw32_start_call): Change pthread_exit() call to
+ ptw32_exit() call.
+
+ * exit.c (ptw32_vacuum): Add thread entry deletion code
+ moved from ptw32_start_call(). See next item.
+ (pthread_exit): Remove longjmp(). Add mutex lock around thread table
+ manipulation code. This routine now calls _enthreadex().
+
+ * create.c (ptw32_start_call): Remove setjmp() call and move
+ cleanup code out. Call pthread_exit(NULL) to terminate the thread.
+
+1998-07-26 Ben Elliston <bje at cygnus.com>
+
+ * tsd.c (pthread_getspecific): Update comments.
+
+ * mutex.c (pthread_mutexattr_setpshared): Not supported; remove.
+ (pthread_mutexattr_getpshared): Likewise.
+
+ * pthread.h (pthread_mutexattr_setpshared): Remove prototype.
+ (pthread_mutexattr_getpshared): Likewise.
+
+Sun Jul 26 00:09:59 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * sync.c: Rename all instances of ptw32_count_mutex to
+ ptw32_table_mutex.
+
+ * implement.h: Rename ptw32_count_mutex to
+ ptw32_table_mutex.
+
+ * global.c: Rename ptw32_count_mutex to
+ ptw32_table_mutex.
+
+ * create.c (pthread_create): Add critical sections.
+ (ptw32_start_call): Rename ptw32_count_mutex to
+ ptw32_table_mutex.
+
+ * cancel.c (pthread_setcancelstate): Fix indirection bug and rename
+ "this" to "us".
+
+ * signal.c (pthread_sigmask): Rename "this" to "us" and fix some
+ minor syntax errors. Declare "us" and initialise it.
+
+ * sync.c (pthread_detach): Rename "this" to "target".
+
+ * pthread.h: Converting PTHREAD_* defines to alias the (const int)
+ values in global.c.
+
+ * global.c: Started converting PTHREAD_* defines to (const int) as
+ a part of making the eventual pthreads DLL binary compatible
+ through version changes.
+
+ * condvar.c (cond_wait): Add cancelation point. This applies the
+ point to both pthread_cond_wait() and pthread_cond_timedwait().
+
+ * exit.c (pthread_exit): Rename "this" to "us".
+
+ * implement.h: Add comment.
+
+ * sync.c (pthread_join): I've satisfied myself that pthread_detach()
+ does set the detached attribute in the thread entry attributes
+ to PTHREAD_CREATE_DETACHED. "if" conditions were changed to test
+ that attribute instead of a separate flag.
+
+ * create.c (pthread_create): Rename "this" to "us".
+ (pthread_create): cancelstate and canceltype are not attributes
+ so the copy to thread entry attribute storage was removed.
+ Only the thread itself can change it's cancelstate or canceltype,
+ ie. the thread must exist already.
+
+ * private.c (ptw32_delete_thread_entry): Mutex locks removed.
+ Mutexes must be applied at the caller level.
+ (ptw32_new_thread_entry): Ditto.
+ (ptw32_new_thread_entry): Init cancelstate, canceltype, and
+ cancel_pending to default values.
+ (ptw32_new_thread_entry): Rename "this" to "new".
+ (ptw32_find_thread_entry): Rename "this" to "entry".
+ (ptw32_delete_thread_entry): Rename "thread_entry" to "entry".
+
+ * create.c (ptw32_start_call): Mutexes changed to
+ ptw32_count_mutex. All access to the threads table entries is
+ under the one mutex. Otherwise chaos reigns.
+
+Sat Jul 25 23:16:51 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h (ptw32_threads_thread): Move cancelstate and
+ canceltype members out of pthread_attr_t into here.
+
+ * fork.c (fork): Add comment.
+
+1998-07-25 Ben Elliston <bje at cygnus.com>
+
+ * fork.c (fork): Autoconfiscate.
+
+Sat Jul 25 00:00:13 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * create.c (ptw32_start_call): Set thread priority. Ensure our
+ thread entry is removed from the thread table but only if
+ pthread_detach() was called and there are no waiting joins.
+ (pthread_create): Set detach flag in thread entry if the
+ thread is created PTHREAD_CREATE_DETACHED.
+
+ * pthread.h (pthread_attr_t): Rename member "detachedstate".
+
+ * attr.c (pthread_attr_init): Rename attr members.
+
+ * exit.c (pthread_exit): Fix indirection mistake.
+
+ * implement.h (PTW32_THREADS_TABLE_INDEX): Add.
+
+ * exit.c (ptw32_vacuum): Fix incorrect args to
+ ptw32_handler_pop_all() calls.
+ Make thread entry removal conditional.
+
+ * sync.c (pthread_join): Add multiple join and async detach handling.
+
+ * implement.h (PTW32_THREADS_TABLE_INDEX): Add.
+
+ * global.c (ptw32_threads_mutex_table): Add.
+
+ * implement.h (ptw32_once_flag): Remove.
+ (ptw32_once_lock): Ditto.
+ (ptw32_threads_mutex_table): Add.
+
+ * global.c (ptw32_once_flag): Remove.
+ (ptw32_once_lock): Ditto.
+
+ * sync.c (pthread_join): Fix tests involving new return value
+ from ptw32_find_thread_entry().
+ (pthread_detach): Ditto.
+
+ * private.c (ptw32_find_thread_entry): Failure return code
+ changed from -1 to NULL.
+
+Fri Jul 24 23:09:33 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * create.c (pthread_create): Change . to -> in sigmask memcpy() args.
+
+ * pthread.h: (pthread_cancel): Add function prototype.
+ (pthread_testcancel): Ditto.
+
+1998-07-24 Ben Elliston <bje at cygnus.com>
+
+ * pthread.h (pthread_condattr_t): Rename dummy structure member.
+ (pthread_mutexattr_t): Likewise.
+
+Fri Jul 24 21:13:55 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * cancel.c (pthread_cancel): Implement.
+ (pthread_testcancel): Implement.
+
+ * exit.c (pthread_exit): Add comment explaining the longjmp().
+
+ * implement.h (ptw32_threads_thread_t): New member cancelthread.
+ (PTW32_YES): Define.
+ (PTW32_NO): Define.
+ (RND_SIZEOF): Remove.
+
+ * create.c (pthread_create): Rename cancelability to cancelstate.
+
+ * pthread.h (pthread_attr_t): Rename cancelability to cancelstate.
+ (PTHREAD_CANCELED): Define.
+
+1998-07-24 Ben Elliston <bje at cygnus.com>
+
+ * pthread.h (SIG_BLOCK): Define if not already defined.
+ (SIG_UNBLOCK): Likewise.
+ (SIG_SETMASK): Likewise.
+ (pthread_attr_t): Add signal mask member.
+ (pthread_sigmask): Add function prototype.
+
+ * signal.c (pthread_sigmask): Implement.
+
+ * create.c: #include <string.h> to get a prototype for memcpy().
+ (pthread_create): New threads inherit their creator's signal
+ mask. Copy the signal mask to the new thread structure if we know
+ about signals.
+
+Fri Jul 24 16:33:17 1998 Ross Johnson <rpj at swan.canberra.edu.au>
+
+ * fork.c (pthread_atfork): Add all the necessary push calls.
+ Local implementation semantics:
+ If we get an ENOMEM at any time then ALL handlers
+ (including those from previous pthread_atfork() calls) will be
+ popped off each of the three atfork stacks before we return.
+ (fork): Add all the necessary pop calls. Add the thread cancellation
+ and join calls to the child fork.
+ Add #includes.
+
+ * implement.h: (ptw32_handler_push): Fix return type and stack arg
+ type in prototype.
+ (ptw32_handler_pop): Fix stack arg type in prototype.
+ (ptw32_handler_pop_all): Fix stack arg type in prototype.
+
+ * cleanup.c (ptw32_handler_push): Change return type to int and
+ return ENOMEM if malloc() fails.
+
+ * sync.c (pthread_detach): Use equality test, not assignment.
+
+ * create.c (ptw32_start_call): Add call to Win32 CloseHandle()
+ if thread is detached.
+
+1998-07-24 Ben Elliston <bje at cygnus.com>
+
+ * sync.c (pthread_detach): Close the Win32 thread handle to
+ emulate detached (or daemon) threads.
+
+Fri Jul 24 03:00:25 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * sync.c (pthread_join): Save valueptr arg in joinvalueptr for
+ pthread_exit() to use.
+
+ * private.c (ptw32_new_thread_entry): Initialise joinvalueptr to
+ NULL.
+
+ * create.c (ptw32_start_call): Rewrite to facilitate joins.
+ pthread_exit() will do a longjmp() back to here. Does appropriate
+ cleanup and exit/return from the thread.
+ (pthread_create): _beginthreadex() now passes a pointer to our
+ thread table entry instead of just the call member of that entry.
+
+ * implement.h (ptw32_threads_thread): New member
+ void ** joinvalueptr.
+ (ptw32_call_t): New member jmpbuf env.
+
+ * exit.c (pthread_exit): Major rewrite to handle joins and handing
+ value pointer to joining thread. Uses longjmp() back to
+ ptw32_start_call().
+
+ * create.c (pthread_create): Ensure values of new attribute members
+ are copied to the thread attribute object.
+
+ * attr.c (pthread_attr_destroy): Fix merge conflicts.
+ (pthread_attr_getdetachstate): Fix merge conflicts.
+ (pthread_attr_setdetachstate): Fix merge conflicts.
+
+ * pthread.h: Fix merge conflicts.
+
+ * sync.c (pthread_join): Fix merge conflicts.
+
+Fri Jul 24 00:21:21 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * sync.c (pthread_join): Add check for valid and joinable
+ thread.
+ (pthread_detach): Implement. After checking for a valid and joinable
+ thread, it's still a no-op.
+
+ * private.c (ptw32_find_thread_entry): Bug prevented returning
+ an error value in some cases.
+
+ * attr.c (pthread_attr_setdetachedstate): Implement.
+ (pthread_attr_getdetachedstate): Implement.
+
+ * implement.h: Move more hidden definitions into here from
+ pthread.h.
+
+1998-07-24 Ben Elliston <bje at cygnus.com>
+
+ * pthread.h (PTHREAD_CREATE_JOINABLE): Define.
+ (PTHREAD_CREATE_DETACHED): Likewise.
+ (pthread_attr_t): Add new structure member `detached'.
+ (pthread_attr_getdetachstate): Add function prototype.
+ (pthread_attr_setdetachstate): Likewise.
+
+ * sync.c (pthread_join): Return if the target thread is detached.
+
+ * attr.c (pthread_attr_init): Initialise cancelability and
+ canceltype structure members.
+ (pthread_attr_getdetachstate): Implement.
+ (pthread_attr_setdetachstate): Likewise.
+
+ * implement.h (PTW32_CANCEL_DEFAULTS): Remove. Bit fields
+ proved to be too cumbersome. Set the defaults in attr.c using the
+ public PTHREAD_CANCEL_* constants.
+
+ * cancel.c: New file.
+
+ * pthread.h (sched_param): Define this type.
+ (pthread_attr_getschedparam): Add function prototype.
+ (pthread_attr_setschedparam): Likewise.
+ (pthread_setcancelstate): Likewise.
+ (pthread_setcanceltype): Likewise.
+ (sched_get_priority_min): Likewise.
+ (sched_get_priority_max): Likewise.
+ (pthread_mutexattr_setprotocol): Remove; not supported.
+ (pthread_mutexattr_getprotocol): Likewise.
+ (pthread_mutexattr_setprioceiling): Likewise.
+ (pthread_mutexattr_getprioceiling): Likewise.
+ (pthread_attr_t): Add canceltype member. Update comments.
+ (SCHED_OTHER): Define this scheduling policy constant.
+ (SCHED_FIFO): Likewise.
+ (SCHED_RR): Likewise.
+ (SCHED_MIN): Define the lowest possible value for this constant.
+ (SCHED_MAX): Likewise, the maximum possible value.
+ (PTHREAD_CANCEL_ASYNCHRONOUS): Redefine.
+ (PTHREAD_CANCEL_DEFERRED): Likewise.
+
+ * sched.c: New file.
+ (pthread_setschedparam): Implement.
+ (pthread_getschedparam): Implement.
+ (sched_get_priority_max): Validate policy argument.
+ (sched_get_priority_min): Likewise.
+
+ * mutex.c (pthread_mutexattr_setprotocol): Remove; not supported.
+ (pthread_mutexattr_getprotocol): Likewise.
+ (pthread_mutexattr_setprioceiling): Likewise.
+ (pthread_mutexattr_getprioceiling): Likewise.
+
+Fri Jul 24 00:21:21 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * create.c (pthread_create): Arg to ptw32_new_thread_entry()
+ changed. See next entry. Move mutex locks out. Changes made yesterday
+ and today allow us to start the new thread running rather than
+ temporarily suspended.
+
+ * private.c (ptw32_new_thread_entry): ptw32_thread_table
+ was changed back to a table of thread structures rather than pointers.
+ As such we're trading storage for increaded speed. This routine
+ was modified to work with the new table. Mutex lock put in around
+ global data accesses.
+ (ptw32_find_thread_entry): Ditto
+ (ptw32_delete_thread_entry): Ditto
+
+Thu Jul 23 23:25:30 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * global.c: New. Global data objects declared here. These moved from
+ pthread.h.
+
+ * pthread.h: Move implementation hidden definitions into
+ implement.h.
+
+ * implement.h: Move implementation hidden definitions from
+ pthread.h. Add constants to index into the different handler stacks.
+
+ * cleanup.c (ptw32_handler_push): Simplify args. Restructure.
+ (ptw32_handler_pop): Simplify args. Restructure.
+ (ptw32_handler_pop_all): Simplify args. Restructure.
+
+Wed Jul 22 00:16:22 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * attr.c, implement.h, pthread.h, ChangeLog: Resolve CVS merge
+ conflicts.
+
+ * private.c (ptw32_find_thread_entry): Changes to return type
+ to support leaner ptw32_threads_table[] which now only stores
+ ptw32_thread_thread_t *.
+ (ptw32_new_thread_entry): Internal changes.
+ (ptw32_delete_thread_entry): Internal changes to avoid contention.
+ Calling routines changed accordingly.
+
+ * pthread.h: Modified cleanup macros to use new generic push and pop.
+ Added destructor and atfork stacks to ptw32_threads_thread_t.
+
+ * cleanup.c (ptw32_handler_push, ptw32_handler_pop,
+ ptw32_handler_pop_all): Renamed cleanup push and pop routines
+ and made generic to handle destructors and atfork handlers as
+ well.
+
+ * create.c (ptw32_start_call): New function is a wrapper for
+ all new threads. It allows us to do some cleanup when the thread
+ returns, ie. that is otherwise only done if the thread is cancelled.
+
+ * exit.c (ptw32_vacuum): New function contains code from
+ pthread_exit() that we need in the new ptw32_start_call()
+ as well.
+
+ * implement.h: Various additions and minor changes.
+
+ * pthread.h: Various additions and minor changes.
+ Change cleanup handler macros to use generic handler push and pop
+ functions.
+
+ * attr.c: Minor mods to all functions.
+ (is_attr): Implemented missing function.
+
+ * create.c (pthread_create): More clean up.
+
+ * private.c (ptw32_find_thread_entry): Implement.
+ (ptw32_delete_thread_entry): Implement.
+ (ptw32_new_thread_entry): Implement.
+ These functions manipulate the implementations internal thread
+ table and are part of general code cleanup and modularisation.
+ They replace ptw32_getthreadindex() which was removed.
+
+ * exit.c (pthread_exit): Changed to use the new code above.
+
+ * pthread.h: Add cancelability constants. Update comments.
+
+1998-07-22 Ben Elliston <bje at cygnus.com>
+
+ * attr.c (pthread_setstacksize): Update test of attr argument.
+ (pthread_getstacksize): Likewise.
+ (pthread_setstackaddr): Likewise.
+ (pthread_getstackaddr): Likewise.
+ (pthread_attr_init): No need to allocate any storage.
+ (pthread_attr_destroy): No need to free any storage.
+
+ * mutex.c (is_attr): Not likely to be needed; remove.
+ (remove_attr): Likewise.
+ (insert_attr): Likewise.
+
+ * implement.h (ptw32_mutexattr_t): Moved to a public definition
+ in pthread.h. There was little gain in hiding these details.
+ (ptw32_condattr_t): Likewise.
+ (ptw32_attr_t): Likewise.
+
+ * pthread.h (pthread_atfork): Add function prototype.
+ (pthread_attr_t): Moved here from implement.h.
+
+ * fork.c (pthread_atfork): Preliminary implementation.
+ (ptw32_fork): Likewise.
+
+Wed Jul 22 00:16:22 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * cleanup.c (ptw32_cleanup_push): Implement.
+ (ptw32_cleanup_pop): Implement.
+ (ptw32_do_cancellation): Implement.
+ These are private to the implementation. The real cleanup functions
+ are macros. See below.
+
+ * pthread.h (pthread_cleanup_push): Implement as a macro.
+ (pthread_cleanup_pop): Implement as a macro.
+ Because these are macros which start and end a block, the POSIX scoping
+ requirement is observed. See the comment in the file.
+
+ * exit.c (pthread_exit): Refine the code.
+
+ * create.c (pthread_create): Code cleanup.
+
+ * implement.h (RND_SIZEOF): Add RND_SIZEOF(T) to round sizeof(T)
+ up to multiple of DWORD.
+ Add function prototypes.
+
+ * private.c (ptw32_getthreadindex): "*thread" should have been
+ "thread". Detect empty slot fail condition.
+
+1998-07-20 Ben Elliston <bje at cygnus.com>
+
+ * misc.c (pthread_once): Implement. Don't use a per-application
+ flag and mutex--make `pthread_once_t' contain these elements in
+ their structure. The earlier version had incorrect semantics.
+
+ * pthread.h (ptw32_once_flag): Add new variable. Remove.
+ (ptw32_once_lock): Add new mutex lock to ensure integrity of
+ access to ptw32_once_flag. Remove.
+ (pthread_once): Add function prototype.
+ (pthread_once_t): Define this type.
+
+Mon Jul 20 02:31:05 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * private.c (ptw32_getthreadindex): Implement.
+
+ * pthread.h: Add application static data dependent on
+ _PTHREADS_BUILD_DLL define. This is needed to avoid allocating
+ non-sharable static data within the pthread DLL.
+
+ * implement.h: Add ptw32_cleanup_stack_t, ptw32_cleanup_node_t
+ and PTW32_HASH_INDEX.
+
+ * exit.c (pthread_exit): Begin work on cleanup and de-allocate
+ thread-private storage.
+
+ * create.c (pthread_create): Add thread to thread table.
+ Keep a thread-private copy of the attributes with default values
+ filled in when necessary. Same for the cleanup stack. Make
+ pthread_create C run-time library friendly by using _beginthreadex()
+ instead of CreateThread(). Fix error returns.
+
+Sun Jul 19 16:26:23 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h: Rename pthreads_thread_count to ptw32_threads_count.
+ Create ptw32_threads_thread_t struct to keep thread specific data.
+
+ * create.c: Rename pthreads_thread_count to ptw32_threads_count.
+ (pthread_create): Handle errors from CreateThread().
+
+1998-07-19 Ben Elliston <bje at cygnus.com>
+
+ * condvar.c (pthread_cond_wait): Generalise. Moved from here ..
+ (cond_wait): To here.
+ (pthread_cond_timedwait): Implement; use generalised cond_wait().
+
+ * pthread.h (pthread_key_t): Define this type.
+ (pthread_key_create): Add function prototype.
+ (pthread_setspecific): Likewise.
+ (pthread_getspecific): Likwise.
+ (pthread_key_delete): Likewise.
+
+ * tsd.c (pthread_key_create): Implement.
+ (pthread_setspecific): Likewise.
+ (pthread_getspecific): Likewise.
+ (pthread_key_delete): Likewise.
+
+ * mutex.c (pthread_mutex_trylock): Return ENOSYS if this function
+ is called on a Win32 platform which is not Windows NT.
+
+1998-07-18 Ben Elliston <bje at cygnus.com>
+
+ * condvar.c (pthread_condattr_init): Do not attempt to malloc any
+ storage; none is needed now that condattr_t is an empty struct.
+ (pthread_condattr_destory): Likewise; do not free storage.
+ (pthread_condattr_setpshared): No longer supported; return ENOSYS.
+ (pthread_condattr_getpshared): Likewise.
+ (pthread_cond_init): Implement with help from Douglas Schmidt.
+ Remember to initialise the cv's internal mutex.
+ (pthread_cond_wait): Likewise.
+ (pthread_cond_signal): Likewise.
+ (pthread_cond_broadcast): Likewise.
+ (pthread_cond_timedwait): Preliminary implementation, but I need
+ to see some API documentation for `WaitForMultipleObject'.
+ (pthread_destory): Implement.
+
+ * pthread.h (pthread_cond_init): Add function protoype.
+ (pthread_cond_broadcast): Likewise.
+ (pthread_cond_signal): Likewise.
+ (pthread_cond_timedwait): Likewise.
+ (pthread_cond_wait): Likewise.
+ (pthread_cond_destroy): Likewise.
+ (pthread_cond_t): Define this type. Fix for u_int. Do not assume
+ that the mutex contained withing the pthread_cond_t structure will
+ be a critical section. Use our new POSIX type!
+
+ * implement.h (ptw32_condattr_t): Remove shared attribute.
+
+1998-07-17 Ben Elliston <bje at cygnus.com>
+
+ * pthread.h (PTHREADS_PROCESS_PRIVATE): Remove.
+ (PTHREAD_PROCESS_SHARED): Likewise. No support for mutexes shared
+ across processes for now.
+ (pthread_mutex_t): Use a Win32 CRITICAL_SECTION type for better
+ performance.
+
+ * implement.h (ptw32_mutexattr_t): Remove shared attribute.
+
+ * mutex.c (pthread_mutexattr_setpshared): This optional function
+ is no longer supported, since we want to implement POSIX mutex
+ variables using the much more efficient Win32 critical section
+ primitives. Critical section objects in Win32 cannot be shared
+ between processes.
+ (pthread_mutexattr_getpshared): Likewise.
+ (pthread_mutexattr_init): No need to malloc any storage; the
+ attributes structure is now empty.
+ (pthread_mutexattr_destroy): This is now a nop.
+ (pthread_mutex_init): Use InitializeCriticalSection().
+ (pthread_mutex_destroy): Use DeleteCriticalSection().
+ (pthread_mutex_lock): Use EnterCriticalSection().
+ (pthread_mutex_trylock): Use TryEnterCriticalSection(). This is
+ not supported by Windows 9x, but trylock is a hack anyway, IMHO.
+ (pthread_mutex_unlock): Use LeaveCriticalSection().
+
+1998-07-14 Ben Elliston <bje at cygnus.com>
+
+ * attr.c (pthread_attr_setstacksize): Implement.
+ (pthread_attr_getstacksize): Likewise.
+ (pthread_attr_setstackaddr): Likewise.
+ (pthread_attr_getstackaddr): Likewise.
+ (pthread_attr_init): Likewise.
+ (pthread_attr_destroy): Likewise.
+
+ * condvar.c (pthread_condattr_init): Add `_cond' to function name.
+
+ * mutex.c (pthread_mutex_lock): Add `_mutex' to function name.
+ (pthread_mutex_trylock): Likewise.
+ (pthread_mutex_unlock): Likewise.
+
+ * pthread.h (pthread_condattr_setpshared): Fix typo.
+ (pthread_attr_init): Add function prototype.
+ (pthread_attr_destroy): Likewise.
+ (pthread_attr_setstacksize): Likewise.
+ (pthread_attr_getstacksize): Likewise.
+ (pthread_attr_setstackaddr): Likewise.
+ (pthread_attr_getstackaddr): Likewise.
+
+Mon Jul 13 01:09:55 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h: Wrap in #ifndef _IMPLEMENT_H
+
+ * create.c (pthread_create): Map stacksize attr to Win32.
+
+ * mutex.c: Include implement.h
+
+1998-07-13 Ben Elliston <bje at cygnus.com>
+
+ * condvar.c (pthread_condattr_init): Implement.
+ (pthread_condattr_destroy): Likewise.
+ (pthread_condattr_setpshared): Likewise.
+ (pthread_condattr_getpshared): Likewise.
+
+ * implement.h (PTHREAD_THREADS_MAX): Remove trailing semicolon.
+ (PTHREAD_STACK_MIN): Specify; needs confirming.
+ (ptw32_attr_t): Define this type.
+ (ptw32_condattr_t): Likewise.
+
+ * pthread.h (pthread_mutex_t): Define this type.
+ (pthread_condattr_t): Likewise.
+ (pthread_mutex_destroy): Add function prototype.
+ (pthread_lock): Likewise.
+ (pthread_trylock): Likewise.
+ (pthread_unlock): Likewise.
+ (pthread_condattr_init): Likewise.
+ (pthread_condattr_destroy): Likewise.
+ (pthread_condattr_setpshared): Likewise.
+ (pthread_condattr_getpshared): Likewise.
+
+ * mutex.c (pthread_mutex_init): Implement.
+ (pthread_mutex_destroy): Likewise.
+ (pthread_lock): Likewise.
+ (pthread_trylock): Likewise.
+ (pthread_unlock): Likewise.
+
+1998-07-12 Ben Elliston <bje at cygnus.com>
+
+ * implement.h (ptw32_mutexattr_t): Define this implementation
+ internal type. Application programmers only see a mutex attribute
+ object as a void pointer.
+
+ * pthread.h (pthread_mutexattr_t): Define this type.
+ (pthread_mutexattr_init): Add function prototype.
+ (pthread_mutexattr_destroy): Likewise.
+ (pthread_mutexattr_setpshared): Likewise.
+ (pthread_mutexattr_getpshared): Likewise.
+ (pthread_mutexattr_setprotocol): Likewise.
+ (pthread_mutexattr_getprotocol): Likewise.
+ (pthread_mutexattr_setprioceiling): Likewise.
+ (pthread_mutexattr_getprioceiling): Likewise.
+ (PTHREAD_PROCESS_PRIVATE): Define.
+ (PTHREAD_PROCESS_SHARED): Define.
+
+ * mutex.c (pthread_mutexattr_init): Implement.
+ (pthread_mutexattr_destroy): Implement.
+ (pthread_mutexattr_setprotocol): Implement.
+ (pthread_mutexattr_getprotocol): Likewise.
+ (pthread_mutexattr_setprioceiling): Likewise.
+ (pthread_mutexattr_getprioceiling): Likewise.
+ (pthread_mutexattr_setpshared): Likewise.
+ (pthread_mutexattr_getpshared): Likewise.
+ (insert_attr): New function; very preliminary implementation!
+ (is_attr): Likewise.
+ (remove_attr): Likewise.
+
+Sat Jul 11 14:48:54 1998 Ross Johnson <rpj at ixobrychus.canberra.edu.au>
+
+ * implement.h: Preliminary implementation specific defines.
+
+ * create.c (pthread_create): Preliminary implementation.
+
+1998-07-11 Ben Elliston <bje at cygnus.com>
+
+ * sync.c (pthread_join): Implement.
+
+ * misc.c (pthread_equal): Likewise.
+
+ * pthread.h (pthread_join): Add function prototype.
+ (pthread_equal): Likewise.
+
+1998-07-10 Ben Elliston <bje at cygnus.com>
+
+ * misc.c (pthread_self): Implement.
+
+ * exit.c (pthread_exit): Implement.
+
+ * pthread.h (pthread_exit): Add function prototype.
+ (pthread_self): Likewise.
+ (pthread_t): Define this type.
+
+1998-07-09 Ben Elliston <bje at cygnus.com>
+
+ * create.c (pthread_create): A dummy stub right now.
+
+ * pthread.h (pthread_create): Add function prototype.
diff --git a/FAQ b/FAQ
index ee8968b..4f2d449 100644
--- a/FAQ
+++ b/FAQ
@@ -1,403 +1,403 @@
- =========================================
- PTHREADS-WIN32 Frequently Asked Questions
- =========================================
-
-INDEX
------
-
-Q 1 What is it?
-
-Q 2 Which of the several dll versions do I use?
- or,
- What are all these pthread*.dll and pthread*.lib files?
-
-Q 3 What is the library naming convention?
-
-Q 4 Cleanup code default style or: it used to work when I built
- the library myself, but now it doesn't - why?
-
-Q 5 Why is the default library version now less exception-friendly?
-
-Q 6 Should I use Cygwin or Mingw32 as a development environment?
-
-Q 7 Now that pthreads-win32 builds under Mingw32, why do I get
- memory access violations (segfaults)?
-
-Q 8 How do I use pthread.dll for Win32 (Visual C++ 5.0)
-
-Q 9 Cancelation doesn't work for me, why?
-
-Q 10 How do I generate pthreadGCE.dll and libpthreadw32.a for use
- with Mingw32?
-
-=============================================================================
-
-Q 1 What is it?
----
-
-Pthreads-win32 is an Open Source Software implementation of the
-Threads component of the POSIX 1003.1c 1995 Standard for Microsoft's
-Win32 environment. Some functions from POSIX 1003.1b are also
-supported including semaphores. Other related functions include
-the set of read-write lock functions. The library also supports
-some of the functionality of the Open Group's Single Unix
-specification, version 2, namely mutex types.
-
-See the file "ANNOUNCE" for more information including standards
-conformance details and list of supported routines.
-
-
-------------------------------------------------------------------------------
-
-Q 2 Which of the several dll versions do I use?
---- or,
- What are all these pthread*.dll and pthread*.lib files?
-
-Simply, you only use one of them, but you need to choose carefully.
-
-The most important choice you need to make is whether to use a
-version that uses exceptions internally, or not (there are versions
-of the library that use exceptions as part of the thread
-cancelation and cleanup implementation, and one that uses
-setjmp/longjmp instead).
-
-There is some contension amongst POSIX threads experts as
-to how POSIX threads cancelation and exit should work
-with languages that include exceptions and handlers, e.g.
-C++ and even C (Microsoft's Structured Exceptions).
-
-The issue is: should cancelation of a thread in, say,
-a C++ application cause object destructors and C++ exception
-handlers to be invoked as the stack unwinds during thread
-exit, or not?
-
-There seems to be more opinion in favour of using the
-standard C version of the library (no EH) with C++ applications
-since this appears to be the assumption commercial pthreads
-implementations make. Therefore, if you use an EH version
-of pthreads-win32 then you may be under the illusion that
-your application will be portable, when in fact it is likely to
-behave very differently linked with other pthreads libraries.
-
-Now you may be asking: why have you kept the EH versions of
-the library?
-
-There are a couple of reasons:
-- there is division amongst the experts and so the code may
- be needed in the future. (Yes, it's in the repository and we
- can get it out anytime in the future, but ...)
-- pthreads-win32 is one of the few implementations, and possibly
- the only freely available one, that has EH versions. It may be
- useful to people who want to play with or study application
- behaviour under these conditions.
-
-
-------------------------------------------------------------------------------
-
-Q 3 What is the library naming convention?
----
-
-Because the library is being built using various exception
-handling schemes and compilers - and because the library
-may not work reliably if these are mixed in an application,
-each different version of the library has it's own name.
-
-Note 1: the incompatibility is really between EH implementations
-of the different compilers. It should be possible to use the
-standard C version from either compiler with C++ applications
-built with a different compiler. If you use an EH version of
-the library, then you must use the same compiler for the
-application. This is another complication and dependency that
-can be avoided by using only the standard C library version.
-
-Note 2: if you use a standard C pthread*.dll with a C++
-application, then any functions that you define that are
-intended to be called via pthread_cleanup_push() must be
-__cdecl.
-
-Note 3: the intention is to also name either the VC or GC
-version (it should be arbitrary) as pthread.dll, including
-pthread.lib and libpthread.a as appropriate.
-
-In general:
- pthread[VG]{SE,CE,C}.dll
- pthread[VG]{SE,CE,C}.lib
-
-where:
- [VG] indicates the compiler
- V - MS VC
- G - GNU C
-
- {SE,CE,C} indicates the exception handling scheme
- SE - Structured EH
- CE - C++ EH
- C - no exceptions - uses setjmp/longjmp
-
-For example:
- pthreadVSE.dll (MSVC/SEH)
- pthreadGCE.dll (GNUC/C++ EH)
- pthreadGC.dll (GNUC/not dependent on exceptions)
-
-The GNU library archive file names have changed to:
-
- libpthreadGCE.a
- libpthreadGC.a
-
-
-------------------------------------------------------------------------------
-
-Q 4 Cleanup code default style or: it used to work when I built
---- the library myself, but now it doesn't - why?
-
-Up to and including snapshot 2001-07-12, if not defined, the cleanup
-style was determined automatically from the compiler used, and one
-of the following was defined accordingly:
-
- __CLEANUP_SEH MSVC only
- __CLEANUP_CXX C++, including MSVC++, GNU G++
- __CLEANUP_C C, including GNU GCC, not MSVC
-
-These defines determine the style of cleanup (see pthread.h) and,
-most importantly, the way that cancelation and thread exit (via
-pthread_exit) is performed (see the routine ptw32_throw() in private.c).
-
-In short, the exceptions versions of the library throw an exception
-when a thread is canceled or exits (via pthread_exit()), which is
-caught by a handler in the thread startup routine, so that the
-the correct stack unwinding occurs regardless of where the thread
-is when it's canceled or exits via pthread_exit().
-
-After snapshot 2001-07-12, unless your build explicitly defines (e.g.
-via a compiler option) __CLEANUP_SEH, __CLEANUP_CXX, or __CLEANUP_C, then
-the build now ALWAYS defaults to __CLEANUP_C style cleanup. This style
-uses setjmp/longjmp in the cancelation and pthread_exit implementations,
-and therefore won't do stack unwinding even when linked to applications
-that have it (e.g. C++ apps). This is for consistency with most/all
-commercial Unix POSIX threads implementations.
-
-Although it was not clearly documented before, it is still necessary to
-build your application using the same __CLEANUP_* define as was
-used for the version of the library that you link with, so that the
-correct parts of pthread.h are included. That is, the possible
-defines require the following library versions:
-
- __CLEANUP_SEH pthreadVSE.dll
- __CLEANUP_CXX pthreadVCE.dll or pthreadGCE.dll
- __CLEANUP_C pthreadVC.dll or pthreadGC.dll
-
-THE POINT OF ALL THIS IS: if you have not been defining one of these
-explicitly, then the defaults have been set according to the compiler
-and language you are using, as described at the top of this
-section.
-
-THIS NOW CHANGES, as has been explained above. For example:
-
-If you were building your application with MSVC++ i.e. using C++
-exceptions (rather than SEH) and not explicitly defining one of
-__CLEANUP_*, then __CLEANUP_C++ was defined for you in pthread.h.
-You should have been linking with pthreadVCE.dll, which does
-stack unwinding.
-
-If you now build your application as you had before, pthread.h will now
-set __CLEANUP_C as the default style, and you will need to link
-with pthreadVC.dll. Stack unwinding will now NOT occur when a
-thread is canceled, nor when the thread calls pthread_exit().
-
-Your application will now most likely behave differently to previous
-versions, and in non-obvious ways. Most likely is that local
-objects may not be destroyed or cleaned up after a thread
-is canceled.
-
-If you want the same behaviour as before, then you must now define
-__CLEANUP_C++ explicitly using a compiler option and link with
-pthreadVCE.dll as you did before.
-
-
-------------------------------------------------------------------------------
-
-Q 5 Why is the default library version now less exception-friendly?
----
-
-Because most commercial Unix POSIX threads implementations don't allow you to
-choose to have stack unwinding. (Compaq's TRU64 Unix is possibly an exception.)
-
-Therefore, providing it in pthread-win32 as a default could be dangerous
-and non-portable. We still provide the choice but you must now consciously
-make it.
-
-WHY NOT REMOVE THE EXCEPTIONS VERSIONS OF THE LIBRARY ALTOGETHER?
-There are a few reasons:
-- because there are well respected POSIX threads people who believe
- that POSIX threads implementations should be exceptions-aware and
- do the expected thing in that context. (There are equally respected
- people who believe it should not be easily accessible, if it's there
- at all.)
-- because pthreads-win32 is one of the few implementations that has
- the choice, perhaps the only freely available one, and so offers
- a laboratory to people who may want to explore the effects;
-- although the code will always be around somewhere for anyone who
- wants it, once it's removed from the current version it will not be
- nearly as visible to people who may have a use for it.
-
-
-------------------------------------------------------------------------------
-
-Q 6 Should I use Cygwin or Mingw32 as a development environment?
----
-
-Important: see Q7 also.
-
-Use Mingw32 with the MSVCRT library to build applications that use
-the pthreads DLL.
-
-Cygwin's own internal support for POSIX threads is growing.
-Consult that project's documentation for more information.
-
-------------------------------------------------------------------------------
-
-Q 7 Now that pthreads-win32 builds under Mingw32, why do I get
---- memory access violations (segfaults)?
-
-The latest Mingw32 package has thread-safe exception handling (see Q10).
-Also, see Q6 above.
-
-------------------------------------------------------------------------------
-
-Q 8 How do I use pthread.dll for Win32 (Visual C++ 5.0)
----
-
->
-> I'm a "rookie" when it comes to your pthread implementation. I'm currently
-> desperately trying to install the prebuilt .dll file into my MSVC compiler.
-> Could you please provide me with explicit instructions on how to do this (or
-> direct me to a resource(s) where I can acquire such information)?
->
-> Thank you,
->
-
-You should have a .dll, .lib, .def, and three .h files. It is recommended
-that you use pthreadVC.dll, rather than pthreadVCE.dll or pthreadVSE.dll
-(see Q2 above).
-
-The .dll can go in any directory listed in your PATH environment
-variable, so putting it into C:\WINDOWS should work.
-
-The .lib file can go in any directory listed in your LIB environment
-variable.
-
-The .h files can go in any directory listed in your INCLUDE
-environment variable.
-
-Or you might prefer to put the .lib and .h files into a new directory
-and add its path to LIB and INCLUDE. You can probably do this easiest
-by editing the file:-
-
-C:\Program Files\DevStudio\vc\bin\vcvars32.bat
-
-The .def file isn't used by anything in the pre-compiled version but
-is included for information.
-
-Cheers.
-Ross
-
-------------------------------------------------------------------------------
-
-Q 9 Cancelation doesn't work for me, why?
----
-
-> I'm investigating a problem regarding thread cancelation. The thread I want
-> to cancel has PTHREAD_CANCEL_ASYNCHRONOUS, however, this piece of code
-> blocks on the join():
->
-> if ((retv = Pthread_cancel( recvThread )) == 0)
-> {
-> retv = Pthread_join( recvThread, 0 );
-> }
->
-> Pthread_* are just macro's; they call pthread_*.
->
-> The thread recvThread seems to block on a select() call. It doesn't get
-> cancelled.
->
-> Two questions:
->
-> 1) is this normal behaviour?
->
-> 2) if not, how does the cancel mechanism work? I'm not very familliar to
-> win32 programming, so I don't really understand how the *Event() family of
-> calls work.
-
-The answer to your first question is, normal POSIX behaviour would
-be to asynchronously cancel the thread. However, even that doesn't
-guarantee cancelation as the standard only says it should be
-cancelled as soon as possible.
-
-Snapshot 99-11-02 or earlier only partially supports asynchronous cancellation.
-Snapshots since then simulate async cancelation by poking the address of
-a cancelation routine into the PC of the threads context. This requires
-the thread to be resumed in some way for the cancelation to actually
-proceed. This is not true async cancelation, but it is as close as we've
-been able to get to it.
-
-If the thread you're trying to cancel is blocked (for instance, it could be
-waiting for data from the network), it will only get cancelled when it unblocks
-(when the data arrives). For true pre-emptive cancelation in these cases,
-pthreads-win32 from snapshot 2004-05-16 can automatically recognise and use the
-QueueUserAPCEx package by Panagiotis E. Hadjidoukas. This package is available
-from the pthreads-win32 ftp site and is included in the pthreads-win32
-self-unpacking zip from 2004-05-16 onwards.
-
-Using deferred cancelation would normally be the way to go, however,
-even though the POSIX threads standard lists a number of C library
-functions that are defined as deferred cancelation points, there is
-no hookup between those which are provided by Windows and the
-pthreads-win32 library.
-
-Incidently, it's worth noting for code portability that the older POSIX
-threads standards cancelation point lists didn't include "select" because
-(as I read in Butenhof) it wasn't part of POSIX. However, it does appear in
-the SUSV3.
-
-Effectively, the only mandatory cancelation points that pthreads-win32
-recognises are those the library implements itself, ie.
-
- pthread_testcancel
- pthread_cond_wait
- pthread_cond_timedwait
- pthread_join
- sem_wait
- sem_timedwait
- pthread_delay_np
-
-The following routines from the non-mandatory list in SUSV3 are
-cancelation points in pthreads-win32:
-
- pthread_rwlock_wrlock
- pthread_rwlock_timedwrlock
-
-The following routines from the non-mandatory list in SUSV3 are not
-cancelation points in pthreads-win32:
-
- pthread_rwlock_rdlock
- pthread_rwlock_timedrdlock
-
-Pthreads-win32 also provides two functions that allow you to create
-cancelation points within your application, but only for cases where
-a thread is going to block on a Win32 handle. These are:
-
- pthreadCancelableWait(HANDLE waitHandle) /* Infinite wait */
-
- pthreadCancelableTimedWait(HANDLE waitHandle, DWORD timeout)
-
-------------------------------------------------------------------------------
-
-
-Q 10 How do I create thread-safe applications using
----- pthreadGCE.dll, libpthreadw32.a and Mingw32?
-
-This should not be a problem with recent versions of MinGW32.
-
-For early versions, see Thomas Pfaff's email at:
-http://sources.redhat.com/ml/pthreads-win32/2002/msg00000.html
-------------------------------------------------------------------------------
-
+ =========================================
+ PTHREADS-WIN32 Frequently Asked Questions
+ =========================================
+
+INDEX
+-----
+
+Q 1 What is it?
+
+Q 2 Which of the several dll versions do I use?
+ or,
+ What are all these pthread*.dll and pthread*.lib files?
+
+Q 3 What is the library naming convention?
+
+Q 4 Cleanup code default style or: it used to work when I built
+ the library myself, but now it doesn't - why?
+
+Q 5 Why is the default library version now less exception-friendly?
+
+Q 6 Should I use Cygwin or Mingw32 as a development environment?
+
+Q 7 Now that pthreads-win32 builds under Mingw32, why do I get
+ memory access violations (segfaults)?
+
+Q 8 How do I use pthread.dll for Win32 (Visual C++ 5.0)
+
+Q 9 Cancelation doesn't work for me, why?
+
+Q 10 How do I generate pthreadGCE.dll and libpthreadw32.a for use
+ with Mingw32?
+
+=============================================================================
+
+Q 1 What is it?
+---
+
+Pthreads-win32 is an Open Source Software implementation of the
+Threads component of the POSIX 1003.1c 1995 Standard for Microsoft's
+Win32 environment. Some functions from POSIX 1003.1b are also
+supported including semaphores. Other related functions include
+the set of read-write lock functions. The library also supports
+some of the functionality of the Open Group's Single Unix
+specification, version 2, namely mutex types.
+
+See the file "ANNOUNCE" for more information including standards
+conformance details and list of supported routines.
+
+
+------------------------------------------------------------------------------
+
+Q 2 Which of the several dll versions do I use?
+--- or,
+ What are all these pthread*.dll and pthread*.lib files?
+
+Simply, you only use one of them, but you need to choose carefully.
+
+The most important choice you need to make is whether to use a
+version that uses exceptions internally, or not (there are versions
+of the library that use exceptions as part of the thread
+cancelation and cleanup implementation, and one that uses
+setjmp/longjmp instead).
+
+There is some contension amongst POSIX threads experts as
+to how POSIX threads cancelation and exit should work
+with languages that include exceptions and handlers, e.g.
+C++ and even C (Microsoft's Structured Exceptions).
+
+The issue is: should cancelation of a thread in, say,
+a C++ application cause object destructors and C++ exception
+handlers to be invoked as the stack unwinds during thread
+exit, or not?
+
+There seems to be more opinion in favour of using the
+standard C version of the library (no EH) with C++ applications
+since this appears to be the assumption commercial pthreads
+implementations make. Therefore, if you use an EH version
+of pthreads-win32 then you may be under the illusion that
+your application will be portable, when in fact it is likely to
+behave very differently linked with other pthreads libraries.
+
+Now you may be asking: why have you kept the EH versions of
+the library?
+
+There are a couple of reasons:
+- there is division amongst the experts and so the code may
+ be needed in the future. (Yes, it's in the repository and we
+ can get it out anytime in the future, but ...)
+- pthreads-win32 is one of the few implementations, and possibly
+ the only freely available one, that has EH versions. It may be
+ useful to people who want to play with or study application
+ behaviour under these conditions.
+
+
+------------------------------------------------------------------------------
+
+Q 3 What is the library naming convention?
+---
+
+Because the library is being built using various exception
+handling schemes and compilers - and because the library
+may not work reliably if these are mixed in an application,
+each different version of the library has it's own name.
+
+Note 1: the incompatibility is really between EH implementations
+of the different compilers. It should be possible to use the
+standard C version from either compiler with C++ applications
+built with a different compiler. If you use an EH version of
+the library, then you must use the same compiler for the
+application. This is another complication and dependency that
+can be avoided by using only the standard C library version.
+
+Note 2: if you use a standard C pthread*.dll with a C++
+application, then any functions that you define that are
+intended to be called via pthread_cleanup_push() must be
+__cdecl.
+
+Note 3: the intention is to also name either the VC or GC
+version (it should be arbitrary) as pthread.dll, including
+pthread.lib and libpthread.a as appropriate.
+
+In general:
+ pthread[VG]{SE,CE,C}.dll
+ pthread[VG]{SE,CE,C}.lib
+
+where:
+ [VG] indicates the compiler
+ V - MS VC
+ G - GNU C
+
+ {SE,CE,C} indicates the exception handling scheme
+ SE - Structured EH
+ CE - C++ EH
+ C - no exceptions - uses setjmp/longjmp
+
+For example:
+ pthreadVSE.dll (MSVC/SEH)
+ pthreadGCE.dll (GNUC/C++ EH)
+ pthreadGC.dll (GNUC/not dependent on exceptions)
+
+The GNU library archive file names have changed to:
+
+ libpthreadGCE.a
+ libpthreadGC.a
+
+
+------------------------------------------------------------------------------
+
+Q 4 Cleanup code default style or: it used to work when I built
+--- the library myself, but now it doesn't - why?
+
+Up to and including snapshot 2001-07-12, if not defined, the cleanup
+style was determined automatically from the compiler used, and one
+of the following was defined accordingly:
+
+ __CLEANUP_SEH MSVC only
+ __CLEANUP_CXX C++, including MSVC++, GNU G++
+ __CLEANUP_C C, including GNU GCC, not MSVC
+
+These defines determine the style of cleanup (see pthread.h) and,
+most importantly, the way that cancelation and thread exit (via
+pthread_exit) is performed (see the routine ptw32_throw() in private.c).
+
+In short, the exceptions versions of the library throw an exception
+when a thread is canceled or exits (via pthread_exit()), which is
+caught by a handler in the thread startup routine, so that the
+the correct stack unwinding occurs regardless of where the thread
+is when it's canceled or exits via pthread_exit().
+
+After snapshot 2001-07-12, unless your build explicitly defines (e.g.
+via a compiler option) __CLEANUP_SEH, __CLEANUP_CXX, or __CLEANUP_C, then
+the build now ALWAYS defaults to __CLEANUP_C style cleanup. This style
+uses setjmp/longjmp in the cancelation and pthread_exit implementations,
+and therefore won't do stack unwinding even when linked to applications
+that have it (e.g. C++ apps). This is for consistency with most/all
+commercial Unix POSIX threads implementations.
+
+Although it was not clearly documented before, it is still necessary to
+build your application using the same __CLEANUP_* define as was
+used for the version of the library that you link with, so that the
+correct parts of pthread.h are included. That is, the possible
+defines require the following library versions:
+
+ __CLEANUP_SEH pthreadVSE.dll
+ __CLEANUP_CXX pthreadVCE.dll or pthreadGCE.dll
+ __CLEANUP_C pthreadVC.dll or pthreadGC.dll
+
+THE POINT OF ALL THIS IS: if you have not been defining one of these
+explicitly, then the defaults have been set according to the compiler
+and language you are using, as described at the top of this
+section.
+
+THIS NOW CHANGES, as has been explained above. For example:
+
+If you were building your application with MSVC++ i.e. using C++
+exceptions (rather than SEH) and not explicitly defining one of
+__CLEANUP_*, then __CLEANUP_C++ was defined for you in pthread.h.
+You should have been linking with pthreadVCE.dll, which does
+stack unwinding.
+
+If you now build your application as you had before, pthread.h will now
+set __CLEANUP_C as the default style, and you will need to link
+with pthreadVC.dll. Stack unwinding will now NOT occur when a
+thread is canceled, nor when the thread calls pthread_exit().
+
+Your application will now most likely behave differently to previous
+versions, and in non-obvious ways. Most likely is that local
+objects may not be destroyed or cleaned up after a thread
+is canceled.
+
+If you want the same behaviour as before, then you must now define
+__CLEANUP_C++ explicitly using a compiler option and link with
+pthreadVCE.dll as you did before.
+
+
+------------------------------------------------------------------------------
+
+Q 5 Why is the default library version now less exception-friendly?
+---
+
+Because most commercial Unix POSIX threads implementations don't allow you to
+choose to have stack unwinding. (Compaq's TRU64 Unix is possibly an exception.)
+
+Therefore, providing it in pthread-win32 as a default could be dangerous
+and non-portable. We still provide the choice but you must now consciously
+make it.
+
+WHY NOT REMOVE THE EXCEPTIONS VERSIONS OF THE LIBRARY ALTOGETHER?
+There are a few reasons:
+- because there are well respected POSIX threads people who believe
+ that POSIX threads implementations should be exceptions-aware and
+ do the expected thing in that context. (There are equally respected
+ people who believe it should not be easily accessible, if it's there
+ at all.)
+- because pthreads-win32 is one of the few implementations that has
+ the choice, perhaps the only freely available one, and so offers
+ a laboratory to people who may want to explore the effects;
+- although the code will always be around somewhere for anyone who
+ wants it, once it's removed from the current version it will not be
+ nearly as visible to people who may have a use for it.
+
+
+------------------------------------------------------------------------------
+
+Q 6 Should I use Cygwin or Mingw32 as a development environment?
+---
+
+Important: see Q7 also.
+
+Use Mingw32 with the MSVCRT library to build applications that use
+the pthreads DLL.
+
+Cygwin's own internal support for POSIX threads is growing.
+Consult that project's documentation for more information.
+
+------------------------------------------------------------------------------
+
+Q 7 Now that pthreads-win32 builds under Mingw32, why do I get
+--- memory access violations (segfaults)?
+
+The latest Mingw32 package has thread-safe exception handling (see Q10).
+Also, see Q6 above.
+
+------------------------------------------------------------------------------
+
+Q 8 How do I use pthread.dll for Win32 (Visual C++ 5.0)
+---
+
+>
+> I'm a "rookie" when it comes to your pthread implementation. I'm currently
+> desperately trying to install the prebuilt .dll file into my MSVC compiler.
+> Could you please provide me with explicit instructions on how to do this (or
+> direct me to a resource(s) where I can acquire such information)?
+>
+> Thank you,
+>
+
+You should have a .dll, .lib, .def, and three .h files. It is recommended
+that you use pthreadVC.dll, rather than pthreadVCE.dll or pthreadVSE.dll
+(see Q2 above).
+
+The .dll can go in any directory listed in your PATH environment
+variable, so putting it into C:\WINDOWS should work.
+
+The .lib file can go in any directory listed in your LIB environment
+variable.
+
+The .h files can go in any directory listed in your INCLUDE
+environment variable.
+
+Or you might prefer to put the .lib and .h files into a new directory
+and add its path to LIB and INCLUDE. You can probably do this easiest
+by editing the file:-
+
+C:\Program Files\DevStudio\vc\bin\vcvars32.bat
+
+The .def file isn't used by anything in the pre-compiled version but
+is included for information.
+
+Cheers.
+Ross
+
+------------------------------------------------------------------------------
+
+Q 9 Cancelation doesn't work for me, why?
+---
+
+> I'm investigating a problem regarding thread cancelation. The thread I want
+> to cancel has PTHREAD_CANCEL_ASYNCHRONOUS, however, this piece of code
+> blocks on the join():
+>
+> if ((retv = Pthread_cancel( recvThread )) == 0)
+> {
+> retv = Pthread_join( recvThread, 0 );
+> }
+>
+> Pthread_* are just macro's; they call pthread_*.
+>
+> The thread recvThread seems to block on a select() call. It doesn't get
+> cancelled.
+>
+> Two questions:
+>
+> 1) is this normal behaviour?
+>
+> 2) if not, how does the cancel mechanism work? I'm not very familliar to
+> win32 programming, so I don't really understand how the *Event() family of
+> calls work.
+
+The answer to your first question is, normal POSIX behaviour would
+be to asynchronously cancel the thread. However, even that doesn't
+guarantee cancelation as the standard only says it should be
+cancelled as soon as possible.
+
+Snapshot 99-11-02 or earlier only partially supports asynchronous cancellation.
+Snapshots since then simulate async cancelation by poking the address of
+a cancelation routine into the PC of the threads context. This requires
+the thread to be resumed in some way for the cancelation to actually
+proceed. This is not true async cancelation, but it is as close as we've
+been able to get to it.
+
+If the thread you're trying to cancel is blocked (for instance, it could be
+waiting for data from the network), it will only get cancelled when it unblocks
+(when the data arrives). For true pre-emptive cancelation in these cases,
+pthreads-win32 from snapshot 2004-05-16 can automatically recognise and use the
+QueueUserAPCEx package by Panagiotis E. Hadjidoukas. This package is available
+from the pthreads-win32 ftp site and is included in the pthreads-win32
+self-unpacking zip from 2004-05-16 onwards.
+
+Using deferred cancelation would normally be the way to go, however,
+even though the POSIX threads standard lists a number of C library
+functions that are defined as deferred cancelation points, there is
+no hookup between those which are provided by Windows and the
+pthreads-win32 library.
+
+Incidently, it's worth noting for code portability that the older POSIX
+threads standards cancelation point lists didn't include "select" because
+(as I read in Butenhof) it wasn't part of POSIX. However, it does appear in
+the SUSV3.
+
+Effectively, the only mandatory cancelation points that pthreads-win32
+recognises are those the library implements itself, ie.
+
+ pthread_testcancel
+ pthread_cond_wait
+ pthread_cond_timedwait
+ pthread_join
+ sem_wait
+ sem_timedwait
+ pthread_delay_np
+
+The following routines from the non-mandatory list in SUSV3 are
+cancelation points in pthreads-win32:
+
+ pthread_rwlock_wrlock
+ pthread_rwlock_timedwrlock
+
+The following routines from the non-mandatory list in SUSV3 are not
+cancelation points in pthreads-win32:
+
+ pthread_rwlock_rdlock
+ pthread_rwlock_timedrdlock
+
+Pthreads-win32 also provides two functions that allow you to create
+cancelation points within your application, but only for cases where
+a thread is going to block on a Win32 handle. These are:
+
+ pthreadCancelableWait(HANDLE waitHandle) /* Infinite wait */
+
+ pthreadCancelableTimedWait(HANDLE waitHandle, DWORD timeout)
+
+------------------------------------------------------------------------------
+
+
+Q 10 How do I create thread-safe applications using
+---- pthreadGCE.dll, libpthreadw32.a and Mingw32?
+
+This should not be a problem with recent versions of MinGW32.
+
+For early versions, see Thomas Pfaff's email at:
+http://sources.redhat.com/ml/pthreads-win32/2002/msg00000.html
+------------------------------------------------------------------------------
+
diff --git a/GNUmakefile b/GNUmakefile
index f479778..af70a56 100644
--- a/GNUmakefile
+++ b/GNUmakefile
@@ -46,8 +46,8 @@ CP = cp -f
AR = ar
-#OPT = -g
-#OPT = -O3 -DTEST_ICE
+#OPT = -g -O0
+#OPT = -O3
OPT = -O3 -finline-functions
XOPT =
diff --git a/Makefile b/Makefile
index f39412e..8eccb96 100644
--- a/Makefile
+++ b/Makefile
@@ -14,9 +14,10 @@ DLLS = pthreadVCE.dll pthreadVSE.dll pthreadVC.dll
INLINED_STAMPS = pthreadVCE.stamp pthreadVSE.stamp pthreadVC.stamp
OPTIM = /O2
+#OPTIM =
-#CFLAGS = /W3 /MD /nologo /Yd /Zi /I. /D_WIN32_WINNT=0x400 /DHAVE_CONFIG_H /DTEST_ICE
-CFLAGS = /W3 /MD /nologo /Yd /Zi /I. /D_WIN32_WINNT=0x400 /DHAVE_CONFIG_H
+CFLAGS = /W3 /MD /nologo /Yd /I. /D_WIN32_WINNT=0x400 /DHAVE_CONFIG_H
+#CFLAGS = /W3 /MD /nologo /Yd /Zi /I. /D_WIN32_WINNT=0x400 /DHAVE_CONFIG_H
# C++ Exceptions
VCEFLAGS = /GX /TP /D__CLEANUP_CXX $(CFLAGS)
diff --git a/NEWS b/NEWS
index 695ff5d..d299b17 100644
--- a/NEWS
+++ b/NEWS
@@ -1,3 +1,69 @@
+SNAPSHOT 2004-11-03
+-------------------
+
+DLLs produced from this snapshot cannot be used with older applications without
+recompiling the application, due to a change to pthread_t to provide unique POSIX
+thread IDs.
+
+Although this snapshot passes the extended test suite, many of the changes are
+fairly major, and some applications may show different behaviour than previously,
+so adopt with care. Hopefully, any changed behaviour will be due to the library
+being better at it's job, not worse.
+
+Bug fixes
+---------
+
+* pthread_create() no longer accepts NULL as the thread reference arg.
+A segfault (memory access fault) will result, and no thread will be
+created.
+
+* pthread_barrier_wait() no longer acts as a cancelation point.
+
+* Fix potential race condition in pthread_once()
+- Tristan Savatier <tristan at mpegtv.com>
+
+* Changes to pthread_cond_destroy() exposed some coding weaknesses in several
+test suite mini-apps because pthread_cond_destroy() now returns EBUSY if the CV
+is still in use.
+
+New features
+------------
+
+* Added for compatibility:
+PTHREAD_RECURSIVE_MUTEX_INITIALIZER,
+PTHREAD_ERRORCHECK_MUTEX_INITIALIZER,
+PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP,
+PTHREAD_ERRORCHECK_MUTEX_INITIALIZER_NP
+
+* Initial support for Digital Mars compiler
+- Anuj Goyal <anuj.goyal at gmail.com>
+
+* Faster Mutexes. These have been been rewritten following a model provided by
+Alexander Terekhov that reduces kernel space checks, and eliminates some additional
+critical sections used to manage a race between timedlock expiration and unlock.
+Please be aware that the new mutexes do not enforce strict absolute FIFO scheduling
+of mutexes, however any out-of-order lock acquisition should be very rare.
+
+* Faster semaphores. Following a similar model to mutexes above, these have been
+rewritten to use preliminary users space checks.
+
+* sem_getvalue() now returns the number of waiters.
+
+* The POSIX thread ID now has much stronger uniqueness characteristics. The library
+garrantees not to reuse the same thread ID for at least 2^(wordsize) thread
+destruction/creation cycles.
+
+New tests
+---------
+
+* semaphore4.c: Tests cancelation of the new sem_wait().
+
+* semaphore4t.c: Likewise for sem_timedwait().
+
+* rwlock8.c: Tests and times the slow execution paths of r/w locks, and the CVs,
+mutexes, and semaphores that they're built on.
+
+
SNAPSHOT 2004-05-16
-------------------
diff --git a/PROGRESS b/PROGRESS
index 3e90bec..9abf0bc 100644
--- a/PROGRESS
+++ b/PROGRESS
@@ -1,4 +1,4 @@
-Please see the ANNOUNCE file "Level of Standards Conformance"
-or the web page:
-
-http://sources.redhat.com/pthreads-win32/conformance.html
+Please see the ANNOUNCE file "Level of Standards Conformance"
+or the web page:
+
+http://sources.redhat.com/pthreads-win32/conformance.html
diff --git a/README b/README
index 4f30a46..b7eafdf 100644
--- a/README
+++ b/README
@@ -1,479 +1,444 @@
-PTHREADS-WIN32
-==============
-
-Pthreads-win32 is free software, distributed under the GNU Lesser
-General Public License (LGPL). See the file 'COPYING.LIB' for terms
-and conditions. Also see the file 'COPYING' for information
-specific to pthreads-win32, copyrights and the LGPL.
-
-
-What is it?
------------
-
-Pthreads-win32 is an Open Source Software implementation of the
-Threads component of the POSIX 1003.1c 1995 Standard (or later)
-for Microsoft's Win32 environment. Some functions from POSIX
-1003.1b are also supported including semaphores. Other related
-functions include the set of read-write lock functions. The
-library also supports some of the functionality of the Open
-Group's Single Unix specification, version 2, namely mutex types,
-plus some common and pthreads-win32 specific non-portable
-routines (see README.NONPORTABLE).
-
-See the file "ANNOUNCE" for more information including standards
-conformance details and the list of supported and unsupported
-routines.
-
-
-Prerequisites
--------------
-MSVC or GNU C (MSys - MinGW32)
- To build from source.
-
-QueueUserAPCEx by Panagiotis E. Hadjidoukas
- For true async cancelation of threads (including blocked threads).
- This is a DLL and Windows driver that provides pre-emptive APC
- by forcing threads into an alertable state when the APC is queued.
- Both the DLL and driver are provided with the pthreads-win32.exe
- self-unpacking ZIP, and on the pthreads-win32 FTP site (in source
- and pre-built forms). Currently this is a separate LGPL package to
- pthreads-win32. See the README in the QueueUserAPCEx folder for
- installation instructions.
-
- Pthreads-win32 will automatically detect if the QueueUserAPCEx DLL
- QuserEx.DLL is available and whether the driver AlertDrv.sys is
- loaded. If it is not available, pthreads-win32 will simulate async
- cancelation, which means that it cannot pre-empt blocked threads.
-
-
-Library naming
---------------
-
-Because the library is being built using various exception
-handling schemes and compilers - and because the library
-may not work reliably if these are mixed in an application,
-each different version of the library has it's own name.
-
-Note 1: the incompatibility is really between EH implementations
-of the different compilers. It should be possible to use the
-standard C version from either compiler with C++ applications
-built with a different compiler. If you use an EH version of
-the library, then you must use the same compiler for the
-application. This is another complication and dependency that
-can be avoided by using only the standard C library version.
-
-Note 2: if you use a standard C pthread*.dll with a C++
-application, then any functions that you define that are
-intended to be called via pthread_cleanup_push() must be
-__cdecl.
-
-Note 3: the intention is to also name either the VC or GC
-version (it should be arbitrary) as pthread.dll, including
-pthread.lib and libpthread.a as appropriate.
-
-In general:
- pthread[VG]{SE,CE,C}.dll
- pthread[VG]{SE,CE,C}.lib
-
-where:
- [VG] indicates the compiler
- V - MS VC
- G - GNU C
-
- {SE,CE,C} indicates the exception handling scheme
- SE - Structured EH
- CE - C++ EH
- C - no exceptions - uses setjmp/longjmp
-
-For example:
- pthreadVSE.dll (MSVC/SEH)
- pthreadGCE.dll (GNUC/C++ EH)
- pthreadGC.dll (GNUC/not dependent on exceptions)
-
-The GNU library archive file names have changed to:
-
- libpthreadGCE.a
- libpthreadGC.a
-
-
-Which of the several dll versions to use?
------------------------------------------
-or,
----
-What are all these pthread*.dll and pthread*.lib files?
--------------------------------------------------------
-
-Simple, use either pthreadGC.* if you use GCC, or pthreadVC.* if you
-use MSVC.
-
-Otherwise, you need to choose carefully and know WHY.
-
-The most important choice you need to make is whether to use a
-version that uses exceptions internally, or not. There are versions
-of the library that use exceptions as part of the thread
-cancelation and exit implementation. The default version uses
-setjmp/longjmp.
-
-There is some contension amongst POSIX threads experts as
-to how POSIX threads cancelation and exit should work
-with languages that use exceptions, e.g. C++ and even C
-(Microsoft's Structured Exceptions).
-
-The issue is: should cancelation of a thread in, say,
-a C++ application cause object destructors and C++ exception
-handlers to be invoked as the stack unwinds during thread
-exit, or not?
-
-There seems to be more opinion in favour of using the
-standard C version of the library (no EH) with C++ applications
-for the reason that this appears to be the assumption commercial
-pthreads implementations make. Therefore, if you use an EH version
-of pthreads-win32 then you may be under the illusion that
-your application will be portable, when in fact it is likely to
-behave differently when linked with other pthreads libraries.
-
-Now you may be asking: then why have you kept the EH versions of
-the library?
-
-There are a couple of reasons:
-- there is division amongst the experts and so the code may
- be needed in the future. Yes, it's in the repository and we
- can get it out anytime in the future, but it would be difficult
- to find.
-- pthreads-win32 is one of the few implementations, and possibly
- the only freely available one, that has EH versions. It may be
- useful to people who want to play with or study application
- behaviour under these conditions.
-
-Notes:
-
-[If you use either pthreadVCE or pthreadGCE]
-
-1. [See also the discussion in the FAQ file - Q2, Q4, and Q5]
-
-If your application contains catch(...) blocks in your POSIX
-threads then you will need to replace the "catch(...)" with the macro
-"PtW32Catch", eg.
-
- #ifdef PtW32Catch
- PtW32Catch {
- ...
- }
- #else
- catch(...) {
- ...
- }
- #endif
-
-Otherwise neither pthreads cancelation nor pthread_exit() will work
-reliably when using versions of the library that use C++ exceptions
-for cancelation and thread exit.
-
-This is due to what is believed to be a C++ compliance error in VC++
-whereby you may not have multiple handlers for the same exception in
-the same try/catch block. GNU G++ doesn't have this restriction.
-
-
-Other name changes
-------------------
-
-All snapshots prior to and including snapshot 2000-08-13
-used "_pthread_" as the prefix to library internal
-functions, and "_PTHREAD_" to many library internal
-macros. These have now been changed to "ptw32_" and "PTW32_"
-respectively so as to not conflict with the ANSI standard's
-reservation of identifiers beginning with "_" and "__" for
-use by compiler implementations only.
-
-If you have written any applications and you are linking
-statically with the pthreads-win32 library then you may have
-included a call to _pthread_processInitialize. You will
-now have to change that to ptw32_processInitialize.
-
-
-Cleanup code default style
---------------------------
-
-Previously, if not defined, the cleanup style was determined automatically
-from the compiler used, and one of the following was defined accordingly:
-
- __CLEANUP_SEH MSVC only
- __CLEANUP_CXX C++, including MSVC++, GNU G++
- __CLEANUP_C C, including GNU GCC, not MSVC
-
-These defines determine the style of cleanup (see pthread.h) and,
-most importantly, the way that cancelation and thread exit (via
-pthread_exit) is performed (see the routine ptw32_throw() in private.c).
-
-In short, the exceptions versions of the library throw an exception
-when a thread is canceled or exits (via pthread_exit()), which is
-caught by a handler in the thread startup routine, so that the
-the correct stack unwinding occurs regardless of where the thread
-is when it's canceled or exits via pthread_exit().
-
-In this snapshot, unless the build explicitly defines (e.g. via a
-compiler option) __CLEANUP_SEH, __CLEANUP_CXX, or __CLEANUP_C, then
-the build NOW always defaults to __CLEANUP_C style cleanup. This style
-uses setjmp/longjmp in the cancelation and pthread_exit implementations,
-and therefore won't do stack unwinding even when linked to applications
-that have it (e.g. C++ apps). This is for consistency with most/all
-commercial Unix POSIX threads implementations.
-
-Although it was not clearly documented before, it is still necessary to
-build your application using the same __CLEANUP_* define as was
-used for the version of the library that you link with, so that the
-correct parts of pthread.h are included. That is, the possible
-defines require the following library versions:
-
- __CLEANUP_SEH pthreadVSE.dll
- __CLEANUP_CXX pthreadVCE.dll or pthreadGCE.dll
- __CLEANUP_C pthreadVC.dll or pthreadGC.dll
-
-THE POINT OF ALL THIS IS: if you have not been defining one of these
-explicitly, then the defaults as described at the top of this
-section were being used.
-
-THIS NOW CHANGES, as has been explained above, but to try to make this
-clearer here's an example:
-
-If you were building your application with MSVC++ i.e. using C++
-exceptions (rather than SEH) and not explicitly defining one of
-__CLEANUP_*, then __CLEANUP_C++ was defined for you in pthread.h.
-You should have been linking with pthreadVCE.dll, which does
-stack unwinding.
-
-If you now build your application as you had before, pthread.h will now
-set __CLEANUP_C as the default style, and you will need to link
-with pthreadVC.dll. Stack unwinding will now NOT occur when a thread
-is canceled, or the thread calls pthread_exit().
-
-Your application will now most likely behave differently to previous
-versions, and in non-obvious ways. Most likely is that locally
-instantiated objects may not be destroyed or cleaned up after a thread
-is canceled.
-
-If you want the same behaviour as before, then you must now define
-__CLEANUP_C++ explicitly using a compiler option and link with
-pthreadVCE.dll as you did before.
-
-
-WHY ARE WE MAKING THE DEFAULT STYLE LESS EXCEPTION-FRIENDLY?
-Because no commercial Unix POSIX threads implementation allows you to
-choose to have stack unwinding. Therefore, providing it in pthread-win32
-as a default is dangerous. We still provide the choice but unless
-you consciously choose to do otherwise, your pthreads applications will
-now run or crash in similar ways irrespective of the threads platform
-you use. Or at least this is the hope.
-
-
-
-Building under VC++ using C++ EH, Structured EH, or just C
-----------------------------------------------------------
-
-From the source directory run one of the following:
-
-nmake clean VC (builds the VC setjmp/longjmp version of pthreadVC.dll)
-
-or:
-
-nmake clean VCE (builds the VC++ C++ EH version pthreadVCE.dll)
-
-or:
-
-nmake clean VSE (builds the VC++ structured EH version pthreadVSE.dll)
-
-You can run the testsuite by changing to the "tests" directory and
-running the target corresponding to the DLL version you built:
-
-nmake clean VC
-
-or:
-
-nmake clean VCE
-
-or:
-
-nmake clean VSE
-
-or:
-
-nmake clean VCX (tests the VC version of the library with C++ (EH)
- applications)
-
-
-Building under Mingw32
-----------------------
-
-The dll can be built easily with recent versions of Mingw32.
-(The distributed versions are built using Mingw32 and MsysDTK
-from www.mingw32.org.)
-
-From the source directory, run
-
-make clean GC
-
-or:
-
-make clean GCE
-
-You can run the testsuite by changing to the "tests" directory and
-running
-
-make clean GC
-
-or:
-
-make clean GCE
-
-or:
-
-make clean GCX (tests the GC version of the library with C++ (EH)
- applications)
-
-
-Building the library under Cygwin
----------------------------------
-
-Not tested by me although I think some people have done this.
-Not sure how successfully though.
-
-Cygwin is implementing it's own POSIX threads routines and these
-will be the ones to use if you develop using Cygwin.
-
-
-Ready to run binaries
----------------------
-
-For convenience, the following ready-to-run files can be downloaded
-from the FTP site (see under "Availability" below):
-
- pthread.h
- semaphore.h
- sched.h
- pthread.def
- pthreadVC.dll - built with MSVC compiler using C setjmp/longjmp
- pthreadVC.lib
- pthreadVCE.dll - built with MSVC++ compiler using C++ EH
- pthreadVCE.lib
- pthreadVSE.dll - built with MSVC compiler using SEH
- pthreadVSE.lib
- pthreadGC.dll - built with Mingw32 GCC
- libpthreadGC.a - derived from pthreadGC.dll
- pthreadGCE.dll - built with Mingw32 G++
- libpthreadGCE.a - derived from pthreadGCE.dll
-
-As of August 2003 pthreads-win32 pthreadG* versions are built and tested
-using the MinGW + MsysDTK environment current as of that date or later.
-The following file MAY be needed for older MinGW environments.
-
- gcc.dll - needed to build and run applications that use
- pthreadGCE.dll.
-
-
-Building applications with GNU compilers
-----------------------------------------
-
-If you're using pthreadGC.dll:
-
-With the three header files, pthreadGC.dll and libpthreadGC.a in the
-same directory as your application myapp.c, you could compile, link
-and run myapp.c under Mingw32 as follows:
-
- gcc -o myapp.exe myapp.c -I. -L. -lpthreadGC
- myapp
-
-Or put pthreadGC.dll in an appropriate directory in your PATH,
-put libpthreadGC.a in your system lib directory, and
-put the three header files in your system include directory,
-then use:
-
- gcc -o myapp.exe myapp.c -lpthreadGC
- myapp
-
-
-If you're using pthreadGCE.dll:
-
-With the three header files, pthreadGCE.dll, gcc.dll and libpthreadGCE.a
-in the same directory as your application myapp.c, you could compile,
-link and run myapp.c under Mingw32 as follows:
-
- gcc -x c++ -o myapp.exe myapp.c -I. -L. -lpthreadGCE
- myapp
-
-Or put pthreadGCE.dll and gcc.dll in an appropriate directory in
-your PATH, put libpthreadGCE.a in your system lib directory, and
-put the three header files in your system include directory,
-then use:
-
- gcc -x c++ -o myapp.exe myapp.c -lpthreadGCE
- myapp
-
-
-Availability
-------------
-
-The complete source code in either unbundled, self-extracting
-Zip file, or tar/gzipped format can be found at:
-
- ftp://sources.redhat.com/pub/pthreads-win32
-
-The pre-built DLL, export libraries and matching pthread.h can
-be found at:
-
- ftp://sources.redhat.com/pub/pthreads-win32/dll-latest
-
-Home page:
-
- http://sources.redhat.com/pthreads-win32/
-
-
-Mailing list
-------------
-
-There is a mailing list for discussing pthreads on Win32.
-To join, send email to:
-
- pthreads-win32-subscribe@sources.redhat.com
-
-Unsubscribe by sending mail to:
-
- pthreads-win32-unsubscribe@sources.redhat.com
-
-
-Acknowledgements
-----------------
-
-Pthreads-win32 is based substantially on a Win32 Pthreads
-implementation contributed by John E. Bossom.
-
-Many others have contributed important new code,
-improvements and bug fixes. Thanks go to Alexander Terekhov
-and Louis Thomas for their implementation of condition variables
-(see README.CV).
-
-Thanks also to the authors of the following paper, which served as
-the first CV design, and which identifies the important issues:
-"Strategies for Implementing POSIX Condition Variables on Win32"
-- http://www.cs.wustl.edu/~schmidt/win32-cv-1.html
-
-See the 'CONTRIBUTORS' file for the list of contributors.
-
-As much as possible, the ChangeLog file attributes
-contributions and patches that have been incorporated
-in the library to the individuals responsible.
-
-Finally, thanks to all those who work on and contribute to the
-POSIX and Single Unix Specification standards. The maturity of an
-industry can be measured by it's open standards.
-
-----
-Ross Johnson
-<rpj@callisto.canberra.edu.au>
-
-
-
-
-
-
-
-
+PTHREADS-WIN32
+==============
+
+Pthreads-win32 is free software, distributed under the GNU Lesser
+General Public License (LGPL). See the file 'COPYING.LIB' for terms
+and conditions. Also see the file 'COPYING' for information
+specific to pthreads-win32, copyrights and the LGPL.
+
+
+What is it?
+-----------
+
+Pthreads-win32 is an Open Source Software implementation of the
+Threads component of the POSIX 1003.1c 1995 Standard (or later)
+for Microsoft's Win32 environment. Some functions from POSIX
+1003.1b are also supported including semaphores. Other related
+functions include the set of read-write lock functions. The
+library also supports some of the functionality of the Open
+Group's Single Unix specification, version 2, namely mutex types,
+plus some common and pthreads-win32 specific non-portable
+routines (see README.NONPORTABLE).
+
+See the file "ANNOUNCE" for more information including standards
+conformance details and the list of supported and unsupported
+routines.
+
+
+Prerequisites
+-------------
+MSVC or GNU C (MinGW32 MSys development kit)
+ To build from source.
+
+QueueUserAPCEx by Panagiotis E. Hadjidoukas
+ For true async cancelation of threads (including blocked threads).
+ This is a DLL and Windows driver that provides pre-emptive APC
+ by forcing threads into an alertable state when the APC is queued.
+ Both the DLL and driver are provided with the pthreads-win32.exe
+ self-unpacking ZIP, and on the pthreads-win32 FTP site (in source
+ and pre-built forms). Currently this is a separate LGPL package to
+ pthreads-win32. See the README in the QueueUserAPCEx folder for
+ installation instructions.
+
+ Pthreads-win32 will automatically detect if the QueueUserAPCEx DLL
+ QuserEx.DLL is available and whether the driver AlertDrv.sys is
+ loaded. If it is not available, pthreads-win32 will simulate async
+ cancelation, which means that it cannot pre-empt blocked threads.
+
+
+Library naming
+--------------
+
+Because the library is being built using various exception
+handling schemes and compilers - and because the library
+may not work reliably if these are mixed in an application,
+each different version of the library has it's own name.
+
+Note 1: the incompatibility is really between EH implementations
+of the different compilers. It should be possible to use the
+standard C version from either compiler with C++ applications
+built with a different compiler. If you use an EH version of
+the library, then you must use the same compiler for the
+application. This is another complication and dependency that
+can be avoided by using only the standard C library version.
+
+Note 2: if you use a standard C pthread*.dll with a C++
+application, then any functions that you define that are
+intended to be called via pthread_cleanup_push() must be
+__cdecl.
+
+Note 3: the intention is to also name either the VC or GC
+version (it should be arbitrary) as pthread.dll, including
+pthread.lib and libpthread.a as appropriate.
+
+In general:
+ pthread[VG]{SE,CE,C}.dll
+ pthread[VG]{SE,CE,C}.lib
+
+where:
+ [VG] indicates the compiler
+ V - MS VC
+ G - GNU C
+
+ {SE,CE,C} indicates the exception handling scheme
+ SE - Structured EH
+ CE - C++ EH
+ C - no exceptions - uses setjmp/longjmp
+
+For example:
+ pthreadVSE.dll (MSVC/SEH)
+ pthreadGCE.dll (GNUC/C++ EH)
+ pthreadGC.dll (GNUC/not dependent on exceptions)
+
+The GNU library archive file names have changed to:
+
+ libpthreadGCE.a
+ libpthreadGC.a
+
+
+Which of the several dll versions to use?
+-----------------------------------------
+or,
+---
+What are all these pthread*.dll and pthread*.lib files?
+-------------------------------------------------------
+
+Simple, use either pthreadGC.* if you use GCC, or pthreadVC.* if you
+use MSVC.
+
+Otherwise, you need to choose carefully and know WHY.
+
+The most important choice you need to make is whether to use a
+version that uses exceptions internally, or not. There are versions
+of the library that use exceptions as part of the thread
+cancelation and exit implementation. The default version uses
+setjmp/longjmp.
+
+There is some contension amongst POSIX threads experts as
+to how POSIX threads cancelation and exit should work
+with languages that use exceptions, e.g. C++ and even C
+(Microsoft's Structured Exceptions).
+
+The issue is: should cancelation of a thread in, say,
+a C++ application cause object destructors and C++ exception
+handlers to be invoked as the stack unwinds during thread
+exit, or not?
+
+There seems to be more opinion in favour of using the
+standard C version of the library (no EH) with C++ applications
+for the reason that this appears to be the assumption commercial
+pthreads implementations make. Therefore, if you use an EH version
+of pthreads-win32 then you may be under the illusion that
+your application will be portable, when in fact it is likely to
+behave differently when linked with other pthreads libraries.
+
+Now you may be asking: then why have you kept the EH versions of
+the library?
+
+There are a couple of reasons:
+- there is division amongst the experts and so the code may
+ be needed in the future. Yes, it's in the repository and we
+ can get it out anytime in the future, but it would be difficult
+ to find.
+- pthreads-win32 is one of the few implementations, and possibly
+ the only freely available one, that has EH versions. It may be
+ useful to people who want to play with or study application
+ behaviour under these conditions.
+
+Notes:
+
+[If you use either pthreadVCE or pthreadGCE]
+
+1. [See also the discussion in the FAQ file - Q2, Q4, and Q5]
+
+If your application contains catch(...) blocks in your POSIX
+threads then you will need to replace the "catch(...)" with the macro
+"PtW32Catch", eg.
+
+ #ifdef PtW32Catch
+ PtW32Catch {
+ ...
+ }
+ #else
+ catch(...) {
+ ...
+ }
+ #endif
+
+Otherwise neither pthreads cancelation nor pthread_exit() will work
+reliably when using versions of the library that use C++ exceptions
+for cancelation and thread exit.
+
+This is due to what is believed to be a C++ compliance error in VC++
+whereby you may not have multiple handlers for the same exception in
+the same try/catch block. GNU G++ doesn't have this restriction.
+
+
+Other name changes
+------------------
+
+All snapshots prior to and including snapshot 2000-08-13
+used "_pthread_" as the prefix to library internal
+functions, and "_PTHREAD_" to many library internal
+macros. These have now been changed to "ptw32_" and "PTW32_"
+respectively so as to not conflict with the ANSI standard's
+reservation of identifiers beginning with "_" and "__" for
+use by compiler implementations only.
+
+If you have written any applications and you are linking
+statically with the pthreads-win32 library then you may have
+included a call to _pthread_processInitialize. You will
+now have to change that to ptw32_processInitialize.
+
+
+Cleanup code default style
+--------------------------
+
+Previously, if not defined, the cleanup style was determined automatically
+from the compiler used, and one of the following was defined accordingly:
+
+ __CLEANUP_SEH MSVC only
+ __CLEANUP_CXX C++, including MSVC++, GNU G++
+ __CLEANUP_C C, including GNU GCC, not MSVC
+
+These defines determine the style of cleanup (see pthread.h) and,
+most importantly, the way that cancelation and thread exit (via
+pthread_exit) is performed (see the routine ptw32_throw()).
+
+In short, the exceptions versions of the library throw an exception
+when a thread is canceled, or exits via pthread_exit(). This exception is
+caught by a handler in the thread startup routine, so that the
+the correct stack unwinding occurs regardless of where the thread
+is when it's canceled or exits via pthread_exit().
+
+In this snapshot, unless the build explicitly defines (e.g. via a
+compiler option) __CLEANUP_SEH, __CLEANUP_CXX, or __CLEANUP_C, then
+the build NOW always defaults to __CLEANUP_C style cleanup. This style
+uses setjmp/longjmp in the cancelation and pthread_exit implementations,
+and therefore won't do stack unwinding even when linked to applications
+that have it (e.g. C++ apps). This is for consistency with most/all
+commercial Unix POSIX threads implementations.
+
+Although it was not clearly documented before, it is still necessary to
+build your application using the same __CLEANUP_* define as was
+used for the version of the library that you link with, so that the
+correct parts of pthread.h are included. That is, the possible
+defines require the following library versions:
+
+ __CLEANUP_SEH pthreadVSE.dll
+ __CLEANUP_CXX pthreadVCE.dll or pthreadGCE.dll
+ __CLEANUP_C pthreadVC.dll or pthreadGC.dll
+
+It is recommended that you let pthread.h use it's default __CLEANUP_C
+for both library and application builds. That is, don't define any of
+the above, and then link with pthreadVC.lib (MSVC or MSVC++) and
+libpthreadGC.a (MinGW GCC or G++). The reason is explained below, but
+another reason is that the prebuilt pthreadVCE.dll is currently broken.
+Versions built with MSVC++ later than version 6 may not be broken, but I
+can't verify this yet.
+
+WHY ARE WE MAKING THE DEFAULT STYLE LESS EXCEPTION-FRIENDLY?
+Because no commercial Unix POSIX threads implementation allows you to
+choose to have stack unwinding. Therefore, providing it in pthread-win32
+as a default is dangerous. We still provide the choice but unless
+you consciously choose to do otherwise, your pthreads applications will
+now run or crash in similar ways irrespective of the pthreads platform
+you use. Or at least this is the hope.
+
+
+Building under VC++ using C++ EH, Structured EH, or just C
+----------------------------------------------------------
+
+From the source directory run one of the following:
+
+nmake clean VC-inlined (builds the VC setjmp/longjmp version of
+pthreadVC.dll)
+
+or:
+
+nmake clean VCE-inlined (builds the VC++ C++ EH version pthreadVCE.dll)
+
+or:
+
+nmake clean VSE-inlined (builds the VC++ structured EH version
+pthreadVSE.dll)
+
+You can run the testsuite by changing to the "tests" directory and
+running the target corresponding to the DLL version you built:
+
+nmake clean VC
+
+or:
+
+nmake clean VCE
+
+or:
+
+nmake clean VSE
+
+or:
+
+nmake clean VCX (tests the VC version of the library with C++ (EH)
+ applications)
+
+
+Building under Mingw32
+----------------------
+
+The dll can be built easily with recent versions of Mingw32.
+(The distributed versions are built using Mingw32 and MsysDTK
+from www.mingw32.org.)
+
+From the source directory, run
+
+make clean GC-inlined
+
+or:
+
+make clean GCE-inlined
+
+You can run the testsuite by changing to the "tests" directory and
+running
+
+make clean GC
+
+or:
+
+make clean GCE
+
+or:
+
+make clean GCX (tests the GC version of the library with C++ (EH)
+ applications)
+
+
+Building the library under Cygwin
+---------------------------------
+
+Cygwin is implementing it's own POSIX threads routines and these
+will be the ones to use if you develop using Cygwin.
+
+
+Ready to run binaries
+---------------------
+
+For convenience, the following ready-to-run files can be downloaded
+from the FTP site (see under "Availability" below):
+
+ pthread.h
+ semaphore.h
+ sched.h
+ pthreadVC.dll - built with MSVC compiler using C setjmp/longjmp
+ pthreadVC.lib
+ pthreadVCE.dll - built with MSVC++ compiler using C++ EH
+ pthreadVCE.lib
+ pthreadVSE.dll - built with MSVC compiler using SEH
+ pthreadVSE.lib
+ pthreadGC.dll - built with Mingw32 GCC
+ libpthreadGC.a - derived from pthreadGC.dll
+ pthreadGCE.dll - built with Mingw32 G++
+ libpthreadGCE.a - derived from pthreadGCE.dll
+
+As of August 2003 pthreads-win32 pthreadG* versions are built and tested
+using the MinGW + MsysDTK environment current as of that date or later.
+The following file MAY be needed for older MinGW environments.
+
+ gcc.dll - needed to build and run applications that use
+ pthreadGCE.dll.
+
+
+Building applications with GNU compilers
+----------------------------------------
+
+If you're using pthreadGC.dll:
+
+With the three header files, pthreadGC.dll and libpthreadGC.a in the
+same directory as your application myapp.c, you could compile, link
+and run myapp.c under Mingw32 as follows:
+
+ gcc -o myapp.exe myapp.c -I. -L. -lpthreadGC
+ myapp
+
+Or put pthreadGC.dll in an appropriate directory in your PATH,
+put libpthreadGC.a in your system lib directory, and
+put the three header files in your system include directory,
+then use:
+
+ gcc -o myapp.exe myapp.c -lpthreadGC
+ myapp
+
+
+If you're using pthreadGCE.dll:
+
+With the three header files, pthreadGCE.dll, gcc.dll and libpthreadGCE.a
+in the same directory as your application myapp.c, you could compile,
+link and run myapp.c under Mingw32 as follows:
+
+ gcc -x c++ -o myapp.exe myapp.c -I. -L. -lpthreadGCE
+ myapp
+
+Or put pthreadGCE.dll and gcc.dll in an appropriate directory in
+your PATH, put libpthreadGCE.a in your system lib directory, and
+put the three header files in your system include directory,
+then use:
+
+ gcc -x c++ -o myapp.exe myapp.c -lpthreadGCE
+ myapp
+
+
+Availability
+------------
+
+The complete source code in either unbundled, self-extracting
+Zip file, or tar/gzipped format can be found at:
+
+ ftp://sources.redhat.com/pub/pthreads-win32
+
+The pre-built DLL, export libraries and matching pthread.h can
+be found at:
+
+ ftp://sources.redhat.com/pub/pthreads-win32/dll-latest
+
+Home page:
+
+ http://sources.redhat.com/pthreads-win32/
+
+
+Mailing list
+------------
+
+There is a mailing list for discussing pthreads on Win32.
+To join, send email to:
+
+ pthreads-win32-subscribe@sources.redhat.com
+
+Unsubscribe by sending mail to:
+
+ pthreads-win32-unsubscribe@sources.redhat.com
+
+
+Acknowledgements
+----------------
+
+See the ANNOUNCE file for acknowledgements.
+See the 'CONTRIBUTORS' file for the list of contributors.
+
+As much as possible, the ChangeLog file attributes
+contributions and patches that have been incorporated
+in the library to the individuals responsible.
+
+Finally, thanks to all those who work on and contribute to the
+POSIX and Single Unix Specification standards. The maturity of an
+industry can be measured by it's open standards.
+
+----
+Ross Johnson
+<rpj@callisto.canberra.edu.au>
+
+
+
+
+
+
+
+
diff --git a/README.NONPORTABLE b/README.NONPORTABLE
index ae76862..aa43297 100644
--- a/README.NONPORTABLE
+++ b/README.NONPORTABLE
@@ -1,285 +1,285 @@
-This file documents non-portable functions and other issues.
-
-Non-portable functions included in pthreads-win32
--------------------------------------------------
-
-BOOL
-pthread_win32_test_features_np(int mask)
-
- This routine allows an application to check which
- run-time auto-detected features are available within
- the library.
-
- The possible features are:
-
- PTW32_SYSTEM_INTERLOCKED_COMPARE_EXCHANGE
- Return TRUE if the native version of
- InterlockedCompareExchange() is being used.
- PTW32_ALERTABLE_ASYNC_CANCEL
- Return TRUE is the QueueUserAPCEx package
- QUSEREX.DLL is available and the AlertDrv.sys
- driver is loaded into Windows, providing
- alertable (pre-emptive) asyncronous threads
- cancelation. If this feature returns FALSE
- then the default async cancel scheme is in
- use, which cannot cancel blocked threads.
-
- Features may be Or'ed into the mask parameter, in which case
- the routine returns TRUE if any of the Or'ed features would
- return TRUE. At this stage it doesn't make sense to Or features
- but it may some day.
-
-
-void *
-pthread_timechange_handler_np(void *)
-
- To improve tolerance against operator or time service
- initiated system clock changes.
-
- This routine can be called by an application when it
- receives a WM_TIMECHANGE message from the system. At
- present it broadcasts all condition variables so that
- waiting threads can wake up and re-evaluate their
- conditions and restart their timed waits if required.
-
- It has the same return type and argument type as a
- thread routine so that it may be called directly
- through pthread_create(), i.e. as a separate thread.
-
- Parameters
-
- Although a parameter must be supplied, it is ignored.
- The value NULL can be used.
-
- Return values
-
- It can return an error EAGAIN to indicate that not
- all condition variables were broadcast for some reason.
- Otherwise, 0 is returned.
-
- If run as a thread, the return value is returned
- through pthread_join().
-
- The return value should be cast to an integer.
-
-
-HANDLE
-pthread_getw32threadhandle_np(pthread_t thread);
-
- Returns the win32 thread handle that the POSIX
- thread "thread" is running as.
-
- Applications can use the win32 handle to set
- win32 specific attributes of the thread.
-
-
-int
-pthread_mutexattr_setkind_np(pthread_mutexattr_t * attr, int kind)
-
-int
-pthread_mutexattr_getkind_np(pthread_mutexattr_t * attr, int *kind)
-
- These two routines are included for Linux compatibility
- and are direct equivalents to the standard routines
- pthread_mutexattr_settype
- pthread_mutexattr_gettype
-
- pthread_mutexattr_setkind_np accepts the following
- mutex kinds:
- PTHREAD_MUTEX_FAST_NP
- PTHREAD_MUTEX_ERRORCHECK_NP
- PTHREAD_MUTEX_RECURSIVE_NP
-
- These are really just equivalent to (respectively):
- PTHREAD_MUTEX_NORMAL
- PTHREAD_MUTEX_ERRORCHECK
- PTHREAD_MUTEX_RECURSIVE
-
-int
-pthread_delay_np (const struct timespec *interval);
-
- This routine causes a thread to delay execution for a specific period of time.
- This period ends at the current time plus the specified interval. The routine
- will not return before the end of the period is reached, but may return an
- arbitrary amount of time after the period has gone by. This can be due to
- system load, thread priorities, and system timer granularity.
-
- Specifying an interval of zero (0) seconds and zero (0) nanoseconds is
- allowed and can be used to force the thread to give up the processor or to
- deliver a pending cancelation request.
-
- This routine is a cancelation point.
-
- The timespec structure contains the following two fields:
-
- tv_sec is an integer number of seconds.
- tv_nsec is an integer number of nanoseconds.
-
- Return Values
-
- If an error condition occurs, this routine returns an integer value
- indicating the type of error. Possible return values are as follows:
-
- 0 Successful completion.
- [EINVAL] The value specified by interval is invalid.
-
-int
-pthread_num_processors_np
-
- This routine (found on HPUX systems) returns the number of processors
- in the system. This implementation actually returns the number of
- processors available to the process, which can be a lower number
- than the system's number, depending on the process's affinity mask.
-
-BOOL
-pthread_win32_process_attach_np (void);
-
-BOOL
-pthread_win32_process_detach_np (void);
-
-BOOL
-pthread_win32_thread_attach_np (void);
-
-BOOL
-pthread_win32_thread_detach_np (void);
-
- These functions contain the code normally run via dllMain
- when the library is used as a dll but which need to be
- called explicitly by an application when the library
- is statically linked.
-
- You will need to call pthread_win32_process_attach_np() before
- you can call any pthread routines when statically linking.
- You should call pthread_win32_process_detach_np() before
- exiting your application to clean up.
-
- pthread_win32_thread_attach_np() is currently a no-op, but
- pthread_win32_thread_detach_np() is needed to clean up
- the implicit pthread handle that is allocated to a Win32 thread if
- it calls certain pthreads routines. Call this routine when the
- Win32 thread exits.
-
- These functions invariably return TRUE except for
- pthread_win32_process_attach_np() which will return FALSE
- if pthreads-win32 initialisation fails.
-
-int
-pthreadCancelableWait (HANDLE waitHandle);
-
-int
-pthreadCancelableTimedWait (HANDLE waitHandle, DWORD timeout);
-
- These two functions provide hooks into the pthread_cancel
- mechanism that will allow you to wait on a Windows handle
- and make it a cancellation point. Both functions block
- until either the given w32 handle is signaled, or
- pthread_cancel has been called. It is implemented using
- WaitForMultipleObjects on 'waitHandle' and a manually
- reset w32 event used to implement pthread_cancel.
-
-
-Non-portable issues
--------------------
-
-Thread priority
-
- POSIX defines a single contiguous range of numbers that determine a
- thread's priority. Win32 defines priority classes and priority
- levels relative to these classes. Classes are simply priority base
- levels that the defined priority levels are relative to such that,
- changing a process's priority class will change the priority of all
- of it's threads, while the threads retain the same relativity to each
- other.
-
- A Win32 system defines a single contiguous monotonic range of values
- that define system priority levels, just like POSIX. However, Win32
- restricts individual threads to a subset of this range on a
- per-process basis.
-
- The following table shows the base priority levels for combinations
- of priority class and priority value in Win32.
-
- Process Priority Class Thread Priority Level
- -----------------------------------------------------------------
- 1 IDLE_PRIORITY_CLASS THREAD_PRIORITY_IDLE
- 1 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_IDLE
- 1 NORMAL_PRIORITY_CLASS THREAD_PRIORITY_IDLE
- 1 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_IDLE
- 1 HIGH_PRIORITY_CLASS THREAD_PRIORITY_IDLE
- 2 IDLE_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
- 3 IDLE_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
- 4 IDLE_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
- 4 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
- 5 IDLE_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
- 5 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
- 5 Background NORMAL_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
- 6 IDLE_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
- 6 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
- 6 Background NORMAL_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
- 7 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
- 7 Background NORMAL_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
- 7 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
- 8 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
- 8 NORMAL_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
- 8 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
- 8 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
- 9 NORMAL_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
- 9 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
- 9 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
- 10 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
- 10 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
- 11 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
- 11 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
- 11 HIGH_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
- 12 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
- 12 HIGH_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
- 13 HIGH_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
- 14 HIGH_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
- 15 HIGH_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
- 15 HIGH_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
- 15 IDLE_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
- 15 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
- 15 NORMAL_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
- 15 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
- 16 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_IDLE
- 17 REALTIME_PRIORITY_CLASS -7
- 18 REALTIME_PRIORITY_CLASS -6
- 19 REALTIME_PRIORITY_CLASS -5
- 20 REALTIME_PRIORITY_CLASS -4
- 21 REALTIME_PRIORITY_CLASS -3
- 22 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
- 23 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
- 24 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
- 25 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
- 26 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
- 27 REALTIME_PRIORITY_CLASS 3
- 28 REALTIME_PRIORITY_CLASS 4
- 29 REALTIME_PRIORITY_CLASS 5
- 30 REALTIME_PRIORITY_CLASS 6
- 31 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
-
- Windows NT: Values -7, -6, -5, -4, -3, 3, 4, 5, and 6 are not supported.
-
-
- As you can see, the real priority levels available to any individual
- Win32 thread are non-contiguous.
-
- An application using pthreads-win32 should not make assumptions about
- the numbers used to represent thread priority levels, except that they
- are monotonic between the values returned by sched_get_priority_min()
- and sched_get_priority_max(). E.g. Windows 95, 98, NT, 2000, XP make
- available a non-contiguous range of numbers between -15 and 15, while
- at least one version of WinCE (3.0) defines the minimum priority
- (THREAD_PRIORITY_LOWEST) as 5, and the maximum priority
- (THREAD_PRIORITY_HIGHEST) as 1.
-
- Internally, pthreads-win32 maps any priority levels between
- THREAD_PRIORITY_IDLE and THREAD_PRIORITY_LOWEST to THREAD_PRIORITY_LOWEST,
- or between THREAD_PRIORITY_TIME_CRITICAL and THREAD_PRIORITY_HIGHEST to
- THREAD_PRIORITY_HIGHEST. Currently, this also applies to
- REALTIME_PRIORITY_CLASSi even if levels -7, -6, -5, -4, -3, 3, 4, 5, and 6
- are supported.
-
- If it wishes, a Win32 application using pthreads-win32 can use the Win32
- defined priority macros THREAD_PRIORITY_IDLE through
- THREAD_PRIORITY_TIME_CRITICAL.
+This file documents non-portable functions and other issues.
+
+Non-portable functions included in pthreads-win32
+-------------------------------------------------
+
+BOOL
+pthread_win32_test_features_np(int mask)
+
+ This routine allows an application to check which
+ run-time auto-detected features are available within
+ the library.
+
+ The possible features are:
+
+ PTW32_SYSTEM_INTERLOCKED_COMPARE_EXCHANGE
+ Return TRUE if the native version of
+ InterlockedCompareExchange() is being used.
+ PTW32_ALERTABLE_ASYNC_CANCEL
+ Return TRUE is the QueueUserAPCEx package
+ QUSEREX.DLL is available and the AlertDrv.sys
+ driver is loaded into Windows, providing
+ alertable (pre-emptive) asyncronous threads
+ cancelation. If this feature returns FALSE
+ then the default async cancel scheme is in
+ use, which cannot cancel blocked threads.
+
+ Features may be Or'ed into the mask parameter, in which case
+ the routine returns TRUE if any of the Or'ed features would
+ return TRUE. At this stage it doesn't make sense to Or features
+ but it may some day.
+
+
+void *
+pthread_timechange_handler_np(void *)
+
+ To improve tolerance against operator or time service
+ initiated system clock changes.
+
+ This routine can be called by an application when it
+ receives a WM_TIMECHANGE message from the system. At
+ present it broadcasts all condition variables so that
+ waiting threads can wake up and re-evaluate their
+ conditions and restart their timed waits if required.
+
+ It has the same return type and argument type as a
+ thread routine so that it may be called directly
+ through pthread_create(), i.e. as a separate thread.
+
+ Parameters
+
+ Although a parameter must be supplied, it is ignored.
+ The value NULL can be used.
+
+ Return values
+
+ It can return an error EAGAIN to indicate that not
+ all condition variables were broadcast for some reason.
+ Otherwise, 0 is returned.
+
+ If run as a thread, the return value is returned
+ through pthread_join().
+
+ The return value should be cast to an integer.
+
+
+HANDLE
+pthread_getw32threadhandle_np(pthread_t thread);
+
+ Returns the win32 thread handle that the POSIX
+ thread "thread" is running as.
+
+ Applications can use the win32 handle to set
+ win32 specific attributes of the thread.
+
+
+int
+pthread_mutexattr_setkind_np(pthread_mutexattr_t * attr, int kind)
+
+int
+pthread_mutexattr_getkind_np(pthread_mutexattr_t * attr, int *kind)
+
+ These two routines are included for Linux compatibility
+ and are direct equivalents to the standard routines
+ pthread_mutexattr_settype
+ pthread_mutexattr_gettype
+
+ pthread_mutexattr_setkind_np accepts the following
+ mutex kinds:
+ PTHREAD_MUTEX_FAST_NP
+ PTHREAD_MUTEX_ERRORCHECK_NP
+ PTHREAD_MUTEX_RECURSIVE_NP
+
+ These are really just equivalent to (respectively):
+ PTHREAD_MUTEX_NORMAL
+ PTHREAD_MUTEX_ERRORCHECK
+ PTHREAD_MUTEX_RECURSIVE
+
+int
+pthread_delay_np (const struct timespec *interval);
+
+ This routine causes a thread to delay execution for a specific period of time.
+ This period ends at the current time plus the specified interval. The routine
+ will not return before the end of the period is reached, but may return an
+ arbitrary amount of time after the period has gone by. This can be due to
+ system load, thread priorities, and system timer granularity.
+
+ Specifying an interval of zero (0) seconds and zero (0) nanoseconds is
+ allowed and can be used to force the thread to give up the processor or to
+ deliver a pending cancelation request.
+
+ This routine is a cancelation point.
+
+ The timespec structure contains the following two fields:
+
+ tv_sec is an integer number of seconds.
+ tv_nsec is an integer number of nanoseconds.
+
+ Return Values
+
+ If an error condition occurs, this routine returns an integer value
+ indicating the type of error. Possible return values are as follows:
+
+ 0 Successful completion.
+ [EINVAL] The value specified by interval is invalid.
+
+int
+pthread_num_processors_np
+
+ This routine (found on HPUX systems) returns the number of processors
+ in the system. This implementation actually returns the number of
+ processors available to the process, which can be a lower number
+ than the system's number, depending on the process's affinity mask.
+
+BOOL
+pthread_win32_process_attach_np (void);
+
+BOOL
+pthread_win32_process_detach_np (void);
+
+BOOL
+pthread_win32_thread_attach_np (void);
+
+BOOL
+pthread_win32_thread_detach_np (void);
+
+ These functions contain the code normally run via dllMain
+ when the library is used as a dll but which need to be
+ called explicitly by an application when the library
+ is statically linked.
+
+ You will need to call pthread_win32_process_attach_np() before
+ you can call any pthread routines when statically linking.
+ You should call pthread_win32_process_detach_np() before
+ exiting your application to clean up.
+
+ pthread_win32_thread_attach_np() is currently a no-op, but
+ pthread_win32_thread_detach_np() is needed to clean up
+ the implicit pthread handle that is allocated to a Win32 thread if
+ it calls certain pthreads routines. Call this routine when the
+ Win32 thread exits.
+
+ These functions invariably return TRUE except for
+ pthread_win32_process_attach_np() which will return FALSE
+ if pthreads-win32 initialisation fails.
+
+int
+pthreadCancelableWait (HANDLE waitHandle);
+
+int
+pthreadCancelableTimedWait (HANDLE waitHandle, DWORD timeout);
+
+ These two functions provide hooks into the pthread_cancel
+ mechanism that will allow you to wait on a Windows handle
+ and make it a cancellation point. Both functions block
+ until either the given w32 handle is signaled, or
+ pthread_cancel has been called. It is implemented using
+ WaitForMultipleObjects on 'waitHandle' and a manually
+ reset w32 event used to implement pthread_cancel.
+
+
+Non-portable issues
+-------------------
+
+Thread priority
+
+ POSIX defines a single contiguous range of numbers that determine a
+ thread's priority. Win32 defines priority classes and priority
+ levels relative to these classes. Classes are simply priority base
+ levels that the defined priority levels are relative to such that,
+ changing a process's priority class will change the priority of all
+ of it's threads, while the threads retain the same relativity to each
+ other.
+
+ A Win32 system defines a single contiguous monotonic range of values
+ that define system priority levels, just like POSIX. However, Win32
+ restricts individual threads to a subset of this range on a
+ per-process basis.
+
+ The following table shows the base priority levels for combinations
+ of priority class and priority value in Win32.
+
+ Process Priority Class Thread Priority Level
+ -----------------------------------------------------------------
+ 1 IDLE_PRIORITY_CLASS THREAD_PRIORITY_IDLE
+ 1 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_IDLE
+ 1 NORMAL_PRIORITY_CLASS THREAD_PRIORITY_IDLE
+ 1 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_IDLE
+ 1 HIGH_PRIORITY_CLASS THREAD_PRIORITY_IDLE
+ 2 IDLE_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
+ 3 IDLE_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
+ 4 IDLE_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
+ 4 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
+ 5 IDLE_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
+ 5 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
+ 5 Background NORMAL_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
+ 6 IDLE_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
+ 6 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
+ 6 Background NORMAL_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
+ 7 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
+ 7 Background NORMAL_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
+ 7 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
+ 8 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
+ 8 NORMAL_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
+ 8 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
+ 8 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
+ 9 NORMAL_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
+ 9 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
+ 9 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
+ 10 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
+ 10 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
+ 11 Foreground NORMAL_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
+ 11 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
+ 11 HIGH_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
+ 12 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
+ 12 HIGH_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
+ 13 HIGH_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
+ 14 HIGH_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
+ 15 HIGH_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
+ 15 HIGH_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
+ 15 IDLE_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
+ 15 BELOW_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
+ 15 NORMAL_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
+ 15 ABOVE_NORMAL_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
+ 16 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_IDLE
+ 17 REALTIME_PRIORITY_CLASS -7
+ 18 REALTIME_PRIORITY_CLASS -6
+ 19 REALTIME_PRIORITY_CLASS -5
+ 20 REALTIME_PRIORITY_CLASS -4
+ 21 REALTIME_PRIORITY_CLASS -3
+ 22 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_LOWEST
+ 23 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_BELOW_NORMAL
+ 24 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_NORMAL
+ 25 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_ABOVE_NORMAL
+ 26 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_HIGHEST
+ 27 REALTIME_PRIORITY_CLASS 3
+ 28 REALTIME_PRIORITY_CLASS 4
+ 29 REALTIME_PRIORITY_CLASS 5
+ 30 REALTIME_PRIORITY_CLASS 6
+ 31 REALTIME_PRIORITY_CLASS THREAD_PRIORITY_TIME_CRITICAL
+
+ Windows NT: Values -7, -6, -5, -4, -3, 3, 4, 5, and 6 are not supported.
+
+
+ As you can see, the real priority levels available to any individual
+ Win32 thread are non-contiguous.
+
+ An application using pthreads-win32 should not make assumptions about
+ the numbers used to represent thread priority levels, except that they
+ are monotonic between the values returned by sched_get_priority_min()
+ and sched_get_priority_max(). E.g. Windows 95, 98, NT, 2000, XP make
+ available a non-contiguous range of numbers between -15 and 15, while
+ at least one version of WinCE (3.0) defines the minimum priority
+ (THREAD_PRIORITY_LOWEST) as 5, and the maximum priority
+ (THREAD_PRIORITY_HIGHEST) as 1.
+
+ Internally, pthreads-win32 maps any priority levels between
+ THREAD_PRIORITY_IDLE and THREAD_PRIORITY_LOWEST to THREAD_PRIORITY_LOWEST,
+ or between THREAD_PRIORITY_TIME_CRITICAL and THREAD_PRIORITY_HIGHEST to
+ THREAD_PRIORITY_HIGHEST. Currently, this also applies to
+ REALTIME_PRIORITY_CLASSi even if levels -7, -6, -5, -4, -3, 3, 4, 5, and 6
+ are supported.
+
+ If it wishes, a Win32 application using pthreads-win32 can use the Win32
+ defined priority macros THREAD_PRIORITY_IDLE through
+ THREAD_PRIORITY_TIME_CRITICAL.
diff --git a/TODO b/TODO
index f798e76..fa9efc4 100644
--- a/TODO
+++ b/TODO
@@ -1,7 +1,7 @@
- Things that aren't done yet
- ---------------------------
-
-1. Implement PTHREAD_PROCESS_SHARED for semaphores, mutexes,
- condition variables, read/write locks, barriers.
-
-
+ Things that aren't done yet
+ ---------------------------
+
+1. Implement PTHREAD_PROCESS_SHARED for semaphores, mutexes,
+ condition variables, read/write locks, barriers.
+
+
diff --git a/cleanup.c b/cleanup.c
index 39f8404..3b82876 100644
--- a/cleanup.c
+++ b/cleanup.c
@@ -73,7 +73,7 @@ ptw32_pop_cleanup (int execute)
* ------------------------------------------------------
*/
{
- ptw32_cleanup_t *cleanup = NULL;
+ ptw32_cleanup_t *cleanup;
cleanup = (ptw32_cleanup_t *) pthread_getspecific (ptw32_cleanupKey);
@@ -106,7 +106,7 @@ ptw32_push_cleanup (ptw32_cleanup_t * cleanup,
* popped and invoked with the argument 'arg' when
* a) the thread exits by calling 'pthread_exit',
* b) when the thread acts on a cancellation request,
- * c) or when the thrad calls pthread_cleanup_pop with a nonzero
+ * c) or when the thread calls pthread_cleanup_pop with a nonzero
* 'execute' argument
*
* PARAMETERS
diff --git a/create.c b/create.c
index 048088b..e71ba38 100644
--- a/create.c
+++ b/create.c
@@ -84,6 +84,7 @@ pthread_create (pthread_t * tid,
*/
{
pthread_t thread;
+ ptw32_thread_t * tp;
register pthread_attr_t a;
HANDLE threadH = 0;
int result = EAGAIN;
@@ -98,7 +99,7 @@ pthread_create (pthread_t * tid,
* Make sure that the assignment below can't be optimised out by the compiler.
* This is assured by conditionally assigning *tid again at the end.
*/
- *tid = NULL;
+ tid->x = 0;
if (attr != NULL)
{
@@ -109,12 +110,14 @@ pthread_create (pthread_t * tid,
a = NULL;
}
- if ((thread = ptw32_new ()) == NULL)
+ if ((thread = ptw32_new ()).p == NULL)
{
goto FAIL0;
}
- priority = thread->sched_priority;
+ tp = (ptw32_thread_t *) thread.p;
+
+ priority = tp->sched_priority;
if ((parms = (ThreadParms *) malloc (sizeof (*parms))) == NULL)
{
@@ -128,12 +131,12 @@ pthread_create (pthread_t * tid,
if (a != NULL)
{
stackSize = a->stacksize;
- thread->detachState = a->detachstate;
+ tp->detachState = a->detachstate;
priority = a->param.sched_priority;
#if HAVE_SIGSET_T
- thread->sigmask = a->sigmask;
+ tp->sigmask = a->sigmask;
#endif /* HAVE_SIGSET_T */
@@ -162,7 +165,7 @@ pthread_create (pthread_t * tid,
* system adjustment. This is not the case for POSIX threads.
*/
pthread_t self = pthread_self ();
- priority = self->sched_priority;
+ priority = ((ptw32_thread_t *) self.p)->sched_priority;
}
#endif
@@ -176,29 +179,30 @@ pthread_create (pthread_t * tid,
stackSize = PTHREAD_STACK_MIN;
}
- thread->state = run ? PThreadStateInitial : PThreadStateSuspended;
+ tp->state = run ? PThreadStateInitial : PThreadStateSuspended;
- thread->keys = NULL;
+ tp->keys = NULL;
/*
* Threads must be started in suspended mode and resumed if necessary
* after _beginthreadex returns us the handle. Otherwise we set up a
* race condition between the creating and the created threads.
* Note that we also retain a local copy of the handle for use
- * by us in case thread->threadH gets NULLed later but before we've
+ * by us in case thread.p->threadH gets NULLed later but before we've
* finished with it here.
*/
#if ! defined (__MINGW32__) || defined (__MSVCRT__) || defined (__DMC__)
- thread->threadH = threadH = (HANDLE) _beginthreadex ((void *) NULL, /* No security info */
- (unsigned) stackSize, /* default stack size */
- ptw32_threadStart,
- parms,
- (unsigned)
- CREATE_SUSPENDED,
- (unsigned *) &(thread->
- thread));
+ tp->threadH =
+ threadH =
+ (HANDLE) _beginthreadex ((void *) NULL, /* No security info */
+ (unsigned) stackSize, /* default stack size */
+ ptw32_threadStart,
+ parms,
+ (unsigned)
+ CREATE_SUSPENDED,
+ (unsigned *) &(tp->thread));
if (threadH != 0)
{
@@ -219,17 +223,19 @@ pthread_create (pthread_t * tid,
* This lock will force pthread_threadStart() to wait until we have
* the thread handle and have set the priority.
*/
- (void) pthread_mutex_lock (&thread->cancelLock);
+ (void) pthread_mutex_lock (&tp->cancelLock);
- thread->threadH = threadH = (HANDLE) _beginthread (ptw32_threadStart, (unsigned) stackSize, /* default stack size */
- parms);
+ tp->threadH =
+ threadH =
+ (HANDLE) _beginthread (ptw32_threadStart, (unsigned) stackSize, /* default stack size */
+ parms);
/*
* Make the return code match _beginthreadex's.
*/
if (threadH == (HANDLE) - 1L)
{
- thread->threadH = threadH = 0;
+ tp->threadH = threadH = 0;
}
else
{
@@ -249,7 +255,7 @@ pthread_create (pthread_t * tid,
}
}
- (void) pthread_mutex_unlock (&thread->cancelLock);
+ (void) pthread_mutex_unlock (&tp->cancelLock);
#endif /* __MINGW32__ && ! __MSVCRT__ */
@@ -270,7 +276,7 @@ FAIL0:
{
ptw32_threadDestroy (thread);
- thread = NULL;
+ tp = NULL;
if (parms != NULL)
{
diff --git a/dll.c b/dll.c
index d9c3176..ff75c68 100644
--- a/dll.c
+++ b/dll.c
@@ -37,10 +37,6 @@
#include "pthread.h"
#include "implement.h"
-#if 1
-#include <stdio.h>
-#endif
-
#ifdef _MSC_VER
/*
* lpvReserved yields an unreferenced formal parameter;
@@ -86,6 +82,7 @@ DllMain (HINSTANCE hinstDll, DWORD fdwReason, LPVOID lpvReserved)
result = pthread_win32_process_detach_np ();
break;
}
+
return (result);
} /* DllMain */
diff --git a/global.c b/global.c
index f3f18cb..d2f8252 100644
--- a/global.c
+++ b/global.c
@@ -40,7 +40,8 @@
int ptw32_processInitialized = PTW32_FALSE;
-pthread_t ptw32_threadReuseTop = PTW32_THREAD_REUSE_BOTTOM;
+ptw32_thread_t * ptw32_threadReuseTop = PTW32_THREAD_REUSE_EMPTY;
+ptw32_thread_t * ptw32_threadReuseBottom = PTW32_THREAD_REUSE_EMPTY;
pthread_key_t ptw32_selfThreadKey = NULL;
pthread_key_t ptw32_cleanupKey = NULL;
pthread_cond_t ptw32_cond_list_head = NULL;
diff --git a/implement.h b/implement.h
index 1f6179e..ed8f2a0 100644
--- a/implement.h
+++ b/implement.h
@@ -38,6 +38,9 @@
#ifndef _IMPLEMENT_H
#define _IMPLEMENT_H
+#ifdef _WIN32_WINNT
+#undef _WIN32_WINNT
+#endif
#define _WIN32_WINNT 0x400
#include <windows.h>
@@ -97,6 +100,8 @@ typedef enum
PThreadStateInitial = 0, /* Thread not running */
PThreadStateRunning, /* Thread alive & kicking */
PThreadStateSuspended, /* Thread alive but suspended */
+ PThreadStateCancelPending, /* Thread alive but is */
+ /* has cancelation pending. */
PThreadStateCanceling, /* Thread alive but is */
/* in the process of terminating */
/* due to a cancellation request */
@@ -107,15 +112,18 @@ typedef enum
PThreadState;
-struct pthread_t_
+typedef struct ptw32_thread_t_ ptw32_thread_t;
+
+struct ptw32_thread_t_
{
#ifdef _UWIN
DWORD dummy[5];
#endif
DWORD thread;
- HANDLE threadH; /* POSIX thread is invalid if threadH == 0 */
- pthread_t prevReuse; /* Links threads on reuse stack */
- PThreadState state;
+ HANDLE threadH; /* Win32 thread handle - POSIX thread is invalid if threadH == 0 */
+ pthread_t ptHandle; /* This thread's permanent pthread_t handle */
+ ptw32_thread_t * prevReuse; /* Links threads on reuse stack */
+ volatile PThreadState state;
void *exitStatus;
void *parms;
int ptErrno;
@@ -440,10 +448,11 @@ extern PTW32_INTERLOCKED_LONG (WINAPI *
extern DWORD (*ptw32_register_cancelation) (PAPCFUNC, HANDLE, DWORD);
/* Thread Reuse stack bottom marker. Must not be NULL or any valid pointer to memory. */
-#define PTW32_THREAD_REUSE_BOTTOM ((pthread_t) 1)
+#define PTW32_THREAD_REUSE_EMPTY ((ptw32_thread_t *) 1)
extern int ptw32_processInitialized;
-extern pthread_t ptw32_threadReuseTop;
+extern ptw32_thread_t * ptw32_threadReuseTop;
+extern ptw32_thread_t * ptw32_threadReuseBottom;
extern pthread_key_t ptw32_selfThreadKey;
extern pthread_key_t ptw32_cleanupKey;
extern pthread_cond_t ptw32_cond_list_head;
diff --git a/pthread.h b/pthread.h
index 9975eb1..f2f6323 100644
--- a/pthread.h
+++ b/pthread.h
@@ -519,20 +519,30 @@ extern "C"
#if defined(_UWIN) && PTW32_LEVEL >= PTW32_LEVEL_MAX
# include <sys/types.h>
#else
-typedef struct pthread_t_ * volatile pthread_t;
-typedef struct pthread_attr_t_ * volatile pthread_attr_t;
+/*
+ * Generic handle type - intended to extend uniqueness beyond
+ * that available with a simple pointer. It should scale for either
+ * IA-32 or IA-64.
+ */
+typedef struct {
+ void * p; /* Pointer to actual object */
+ unsigned int x; /* Extra information - reuse count etc */
+} ptw32_handle_t;
+
+typedef ptw32_handle_t pthread_t;
+typedef struct pthread_attr_t_ * pthread_attr_t;
typedef struct pthread_once_t_ pthread_once_t;
-typedef struct pthread_key_t_ * volatile pthread_key_t;
-typedef struct pthread_mutex_t_ * volatile pthread_mutex_t;
-typedef struct pthread_mutexattr_t_ * volatile pthread_mutexattr_t;
-typedef struct pthread_cond_t_ * volatile pthread_cond_t;
-typedef struct pthread_condattr_t_ * volatile pthread_condattr_t;
+typedef struct pthread_key_t_ * pthread_key_t;
+typedef struct pthread_mutex_t_ * pthread_mutex_t;
+typedef struct pthread_mutexattr_t_ * pthread_mutexattr_t;
+typedef struct pthread_cond_t_ * pthread_cond_t;
+typedef struct pthread_condattr_t_ * pthread_condattr_t;
#endif
-typedef struct pthread_rwlock_t_ * volatile pthread_rwlock_t;
-typedef struct pthread_rwlockattr_t_ * volatile pthread_rwlockattr_t;
-typedef struct pthread_spinlock_t_ * volatile pthread_spinlock_t;
-typedef struct pthread_barrier_t_ * volatile pthread_barrier_t;
-typedef struct pthread_barrierattr_t_ * volatile pthread_barrierattr_t;
+typedef struct pthread_rwlock_t_ * pthread_rwlock_t;
+typedef struct pthread_rwlockattr_t_ * pthread_rwlockattr_t;
+typedef struct pthread_spinlock_t_ * pthread_spinlock_t;
+typedef struct pthread_barrier_t_ * pthread_barrier_t;
+typedef struct pthread_barrierattr_t_ * pthread_barrierattr_t;
/*
* ====================
diff --git a/pthread_cancel.c b/pthread_cancel.c
index 69cec3b..c12a44d 100644
--- a/pthread_cancel.c
+++ b/pthread_cancel.c
@@ -121,6 +121,7 @@ pthread_cancel (pthread_t thread)
int result;
int cancel_self;
pthread_t self;
+ ptw32_thread_t * tp;
/* This is the proper way to test thread validity. */
result = pthread_kill (thread, 0);
@@ -129,7 +130,7 @@ pthread_cancel (pthread_t thread)
return result;
}
- if ((self = pthread_self ()) == NULL)
+ if ((self = pthread_self ()).p == NULL)
{
return ENOMEM;
};
@@ -150,35 +151,37 @@ pthread_cancel (pthread_t thread)
*/
cancel_self = pthread_equal (thread, self);
+ tp = (ptw32_thread_t *) thread.p;
+
/*
* Lock for async-cancel safety.
*/
- (void) pthread_mutex_lock (&thread->cancelLock);
+ (void) pthread_mutex_lock (&tp->cancelLock);
- if (thread->cancelType == PTHREAD_CANCEL_ASYNCHRONOUS
- && thread->cancelState == PTHREAD_CANCEL_ENABLE
- && thread->state < PThreadStateCanceling)
+ if (tp->cancelType == PTHREAD_CANCEL_ASYNCHRONOUS
+ && tp->cancelState == PTHREAD_CANCEL_ENABLE
+ && tp->state < PThreadStateCanceling)
{
if (cancel_self)
{
- thread->state = PThreadStateCanceling;
- thread->cancelState = PTHREAD_CANCEL_DISABLE;
+ tp->state = PThreadStateCanceling;
+ tp->cancelState = PTHREAD_CANCEL_DISABLE;
- (void) pthread_mutex_unlock (&thread->cancelLock);
+ (void) pthread_mutex_unlock (&tp->cancelLock);
ptw32_throw (PTW32_EPS_CANCEL);
/* Never reached */
}
else
{
- HANDLE threadH = thread->threadH;
+ HANDLE threadH = tp->threadH;
SuspendThread (threadH);
if (WaitForSingleObject (threadH, 0) == WAIT_TIMEOUT)
{
- thread->state = PThreadStateCanceling;
- thread->cancelState = PTHREAD_CANCEL_DISABLE;
+ tp->state = PThreadStateCanceling;
+ tp->cancelState = PTHREAD_CANCEL_DISABLE;
/*
* If alertdrv and QueueUserAPCEx is available then the following
* will result in a call to QueueUserAPCEx with the args given, otherwise
@@ -186,7 +189,7 @@ pthread_cancel (pthread_t thread)
* the threadH arg will be used.
*/
ptw32_register_cancelation (ptw32_cancel_callback, threadH, 0);
- (void) pthread_mutex_unlock (&thread->cancelLock);
+ (void) pthread_mutex_unlock (&tp->cancelLock);
ResumeThread (threadH);
}
}
@@ -196,13 +199,20 @@ pthread_cancel (pthread_t thread)
/*
* Set for deferred cancellation.
*/
- if (thread->state >= PThreadStateCanceling
- || !SetEvent (thread->cancelEvent))
+ if (tp->state < PThreadStateCancelPending)
+ {
+ tp->state = PThreadStateCancelPending;
+ if (!SetEvent (tp->cancelEvent))
+ {
+ result = ESRCH;
+ }
+ }
+ else if (tp->state >= PThreadStateCanceling)
{
result = ESRCH;
}
- (void) pthread_mutex_unlock (&thread->cancelLock);
+ (void) pthread_mutex_unlock (&tp->cancelLock);
}
return (result);
diff --git a/pthread_cond_destroy.c b/pthread_cond_destroy.c
index 7a34068..b788f9c 100644
--- a/pthread_cond_destroy.c
+++ b/pthread_cond_destroy.c
@@ -38,7 +38,6 @@
#include "pthread.h"
#include "implement.h"
-
int
pthread_cond_destroy (pthread_cond_t * cond)
/*
diff --git a/pthread_cond_wait.c b/pthread_cond_wait.c
index 6db2a48..241f464 100644
--- a/pthread_cond_wait.c
+++ b/pthread_cond_wait.c
@@ -284,6 +284,17 @@ ptw32_cond_wait_cleanup (void *args)
int result;
/*
+ * XSH: Upon successful return, the mutex has been locked and is owned
+ * by the calling thread. This must be done before any cancelation
+ * cleanup handlers are run.
+ */
+ if ((result = pthread_mutex_lock (cleanup_args->mutexPtr)) != 0)
+ {
+ *resultPtr = result;
+ return;
+ }
+
+ /*
* Whether we got here as a result of signal/broadcast or because of
* timeout on wait or thread cancellation we indicate that we are no
* longer waiting. The waiter is responsible for adjusting waiters
@@ -340,16 +351,6 @@ ptw32_cond_wait_cleanup (void *args)
return;
}
}
-
- /*
- * XSH: Upon successful return, the mutex has been locked and is owned
- * by the calling thread
- */
- if ((result = pthread_mutex_lock (cleanup_args->mutexPtr)) != 0)
- {
- *resultPtr = result;
- }
-
} /* ptw32_cond_wait_cleanup */
static INLINE int
diff --git a/pthread_delay_np.c b/pthread_delay_np.c
index ee9b2d9..c699a31 100644
--- a/pthread_delay_np.c
+++ b/pthread_delay_np.c
@@ -86,6 +86,7 @@ pthread_delay_np (struct timespec *interval)
DWORD millisecs;
DWORD status;
pthread_t self;
+ ptw32_thread_t * sp;
if (interval == NULL)
{
@@ -124,34 +125,36 @@ pthread_delay_np (struct timespec *interval)
#pragma enable_message (124)
#endif
- if (NULL == (self = pthread_self ()))
+ if (NULL == (self = pthread_self ()).p)
{
return ENOMEM;
}
- if (self->cancelState == PTHREAD_CANCEL_ENABLE)
+ sp = (ptw32_thread_t *) self.p;
+
+ if (sp->cancelState == PTHREAD_CANCEL_ENABLE)
{
/*
* Async cancelation won't catch us until wait_time is up.
* Deferred cancelation will cancel us immediately.
*/
if (WAIT_OBJECT_0 ==
- (status = WaitForSingleObject (self->cancelEvent, wait_time)))
+ (status = WaitForSingleObject (sp->cancelEvent, wait_time)))
{
/*
* Canceling!
*/
- (void) pthread_mutex_lock (&self->cancelLock);
- if (self->state < PThreadStateCanceling)
+ (void) pthread_mutex_lock (&sp->cancelLock);
+ if (sp->state < PThreadStateCanceling)
{
- self->state = PThreadStateCanceling;
- self->cancelState = PTHREAD_CANCEL_DISABLE;
- (void) pthread_mutex_unlock (&self->cancelLock);
+ sp->state = PThreadStateCanceling;
+ sp->cancelState = PTHREAD_CANCEL_DISABLE;
+ (void) pthread_mutex_unlock (&sp->cancelLock);
ptw32_throw (PTW32_EPS_CANCEL);
}
- (void) pthread_mutex_unlock (&self->cancelLock);
+ (void) pthread_mutex_unlock (&sp->cancelLock);
return ESRCH;
}
else if (status != WAIT_TIMEOUT)
diff --git a/pthread_detach.c b/pthread_detach.c
index a376cf2..df2c542 100644
--- a/pthread_detach.c
+++ b/pthread_detach.c
@@ -75,6 +75,7 @@ pthread_detach (pthread_t thread)
*/
{
int result;
+ ptw32_thread_t * tp = (ptw32_thread_t *) thread.p;
/* This is the proper way to test for a valid thread. */
result = pthread_kill (thread, 0);
@@ -83,14 +84,14 @@ pthread_detach (pthread_t thread)
return result;
}
- if (thread->detachState == PTHREAD_CREATE_DETACHED)
+ if (tp->detachState == PTHREAD_CREATE_DETACHED)
{
result = EINVAL;
}
else
{
result = 0;
- thread->detachState = PTHREAD_CREATE_DETACHED;
+ tp->detachState = PTHREAD_CREATE_DETACHED;
}
return (result);
diff --git a/pthread_equal.c b/pthread_equal.c
index f7965e4..377316b 100644
--- a/pthread_equal.c
+++ b/pthread_equal.c
@@ -69,7 +69,7 @@ pthread_equal (pthread_t t1, pthread_t t2)
* We also accept NULL == NULL - treating NULL as a thread
* for this special case, because there is no error that we can return.
*/
- result = ( t1 == t2 );
+ result = ( t1.p == t2.p && t1.x == t2.x );
return (result);
diff --git a/pthread_exit.c b/pthread_exit.c
index c8fb4fb..aadff5c 100644
--- a/pthread_exit.c
+++ b/pthread_exit.c
@@ -65,20 +65,20 @@ pthread_exit (void *value_ptr)
* ------------------------------------------------------
*/
{
- pthread_t self;
+ ptw32_thread_t * sp;
/*
* Don't use pthread_self() to avoid creating an implicit POSIX thread handle
* unnecessarily.
*/
- self = (pthread_t) pthread_getspecific (ptw32_selfThreadKey);
+ sp = (ptw32_thread_t *) pthread_getspecific (ptw32_selfThreadKey);
#ifdef _UWIN
if (--pthread_count <= 0)
exit ((int) value_ptr);
#endif
- if (NULL == self)
+ if (NULL == sp)
{
/*
* A POSIX thread handle was never created. I.e. this is a
@@ -97,7 +97,7 @@ pthread_exit (void *value_ptr)
/* Never reached */
}
- self->exitStatus = value_ptr;
+ sp->exitStatus = value_ptr;
ptw32_throw (PTW32_EPS_EXIT);
diff --git a/pthread_getschedparam.c b/pthread_getschedparam.c
index d968aa5..edeb5b4 100644
--- a/pthread_getschedparam.c
+++ b/pthread_getschedparam.c
@@ -69,7 +69,7 @@ pthread_getschedparam (pthread_t thread, int *policy,
* for the target thread. It must not return the actual thread
* priority as altered by any system priority adjustments etc.
*/
- param->sched_priority = thread->sched_priority;
+ param->sched_priority = ((ptw32_thread_t *)thread.p)->sched_priority;
return 0;
}
diff --git a/pthread_getw32threadhandle_np.c b/pthread_getw32threadhandle_np.c
index fcf1288..9ac81a7 100644
--- a/pthread_getw32threadhandle_np.c
+++ b/pthread_getw32threadhandle_np.c
@@ -49,5 +49,5 @@
HANDLE
pthread_getw32threadhandle_np (pthread_t thread)
{
- return (thread != NULL) ? (thread->threadH) : 0;
+ return ((ptw32_thread_t *)thread.p)->threadH;
}
diff --git a/pthread_join.c b/pthread_join.c
index 89ee1a5..ba2d3da 100644
--- a/pthread_join.c
+++ b/pthread_join.c
@@ -82,6 +82,7 @@ pthread_join (pthread_t thread, void **value_ptr)
{
int result;
pthread_t self;
+ ptw32_thread_t * tp = (ptw32_thread_t *) thread.p;
/*
* Possibilities for the target thread on entry to pthread_join():
@@ -134,13 +135,13 @@ pthread_join (pthread_t thread, void **value_ptr)
* so that we can use the reuse_lock to ensure the thread isn't destroyed
* and reused before we've finished with the POSIX thread struct.
*/
- if (NULL == thread
- || NULL == thread->threadH
- || THREAD_PRIORITY_ERROR_RETURN == GetThreadPriority (thread->threadH))
+ if (tp == NULL
+ || NULL == tp->threadH
+ || THREAD_PRIORITY_ERROR_RETURN == GetThreadPriority (tp->threadH))
{
result = ESRCH;
}
- else if (PTHREAD_CREATE_DETACHED == thread->detachState)
+ else if (PTHREAD_CREATE_DETACHED == tp->detachState)
{
result = EINVAL;
}
@@ -152,7 +153,9 @@ pthread_join (pthread_t thread, void **value_ptr)
LeaveCriticalSection (&ptw32_thread_reuse_lock);
- if (NULL == (self = pthread_self ()))
+ self = pthread_self();
+
+ if (NULL == self.p)
{
result = ENOENT;
}
@@ -169,7 +172,7 @@ pthread_join (pthread_t thread, void **value_ptr)
* pthreadCancelableWait will not return if we
* are canceled.
*/
- result = pthreadCancelableWait (thread->threadH);
+ result = pthreadCancelableWait (tp->threadH);
if (0 == result)
{
@@ -177,7 +180,7 @@ pthread_join (pthread_t thread, void **value_ptr)
#if ! defined (__MINGW32__) || defined (__MSVCRT__) || defined (__DMC__)
if (value_ptr != NULL
- && !GetExitCodeThread (thread->threadH, (LPDWORD) value_ptr))
+ && !GetExitCodeThread (tp->threadH, (LPDWORD) value_ptr))
{
result = ESRCH;
}
@@ -198,7 +201,7 @@ pthread_join (pthread_t thread, void **value_ptr)
*/
if (value_ptr != NULL)
{
- *value_ptr = thread->exitStatus;
+ *value_ptr = tp->exitStatus;
}
/*
diff --git a/pthread_kill.c b/pthread_kill.c
index c40235c..4cb62c1 100644
--- a/pthread_kill.c
+++ b/pthread_kill.c
@@ -74,12 +74,16 @@ pthread_kill (pthread_t thread, int sig)
*/
{
int result = 0;
+ ptw32_thread_t * tp;
EnterCriticalSection (&ptw32_thread_reuse_lock);
- if (NULL == thread
- || NULL == thread->threadH
- || THREAD_PRIORITY_ERROR_RETURN == GetThreadPriority (thread->threadH))
+ tp = (ptw32_thread_t *) thread.p;
+
+ if (NULL == tp
+ || thread.x != tp->ptHandle.x
+ || NULL == tp->threadH
+ || THREAD_PRIORITY_ERROR_RETURN == GetThreadPriority (tp->threadH))
{
result = ESRCH;
}
diff --git a/pthread_mutex_init.c b/pthread_mutex_init.c
index 0709690..4a6a47b 100644
--- a/pthread_mutex_init.c
+++ b/pthread_mutex_init.c
@@ -84,7 +84,7 @@ pthread_mutex_init (pthread_mutex_t * mutex, const pthread_mutexattr_t * attr)
mx->recursive_count = 0;
mx->kind = (attr == NULL || *attr == NULL
? PTHREAD_MUTEX_DEFAULT : (*attr)->kind);
- mx->ownerThread = NULL;
+ mx->ownerThread.p = NULL;
mx->event = CreateEvent (NULL, PTW32_FALSE, /* manual reset = No */
PTW32_FALSE, /* initial state = not signaled */
diff --git a/pthread_mutex_unlock.c b/pthread_mutex_unlock.c
index 9002df5..113ca68 100644
--- a/pthread_mutex_unlock.c
+++ b/pthread_mutex_unlock.c
@@ -73,7 +73,7 @@ pthread_mutex_unlock (pthread_mutex_t * mutex)
* release one waiter if possible, otherwise
* it will just reset the event.
*/
- if (PulseEvent (mx->event) == 0)
+ if (SetEvent (mx->event) == 0)
{
result = EINVAL;
}
@@ -94,13 +94,13 @@ pthread_mutex_unlock (pthread_mutex_t * mutex)
if (mx->kind != PTHREAD_MUTEX_RECURSIVE
|| 0 == --mx->recursive_count)
{
- mx->ownerThread = NULL;
+ mx->ownerThread.p = NULL;
if ((LONG) PTW32_INTERLOCKED_EXCHANGE ((LPLONG) &mx->lock_idx,
(LONG) 0) < 0)
{
/* Someone may be waiting on that mutex */
- if (PulseEvent (mx->event) == 0)
+ if (SetEvent (mx->event) == 0)
{
result = EINVAL;
}
diff --git a/pthread_self.c b/pthread_self.c
index 5cad22a..a505a8f 100644
--- a/pthread_self.c
+++ b/pthread_self.c
@@ -37,7 +37,6 @@
#include "pthread.h"
#include "implement.h"
-
pthread_t
pthread_self (void)
/*
@@ -61,23 +60,30 @@ pthread_self (void)
*/
{
pthread_t self;
+ pthread_t nil = {NULL, 0};
+ ptw32_thread_t * sp;
#ifdef _UWIN
if (!ptw32_selfThreadKey)
- return (NULL);
+ return nil;
#endif
- self = (pthread_t) pthread_getspecific (ptw32_selfThreadKey);
+ sp = (ptw32_thread_t *) pthread_getspecific (ptw32_selfThreadKey);
- if (self == NULL)
+ if (sp != NULL)
+ {
+ self = sp->ptHandle;
+ }
+ else
{
/*
* Need to create an implicit 'self' for the currently
* executing thread.
*/
self = ptw32_new ();
+ sp = (ptw32_thread_t *) self.p;
- if (self != NULL)
+ if (sp != NULL)
{
/*
* This is a non-POSIX thread which has chosen to call
@@ -85,9 +91,9 @@ pthread_self (void)
* it isn't joinable, but we do assume that it's
* (deferred) cancelable.
*/
- self->implicit = 1;
- self->detachState = PTHREAD_CREATE_DETACHED;
- self->thread = GetCurrentThreadId ();
+ sp->implicit = 1;
+ sp->detachState = PTHREAD_CREATE_DETACHED;
+ sp->thread = GetCurrentThreadId ();
#ifdef NEED_DUPLICATEHANDLE
/*
@@ -99,17 +105,21 @@ pthread_self (void)
* Therefore, you should not pass the handle to
* other threads for whatever purpose.
*/
- self->threadH = GetCurrentThread ();
+ sp->threadH = GetCurrentThread ();
#else
if (!DuplicateHandle (GetCurrentProcess (),
GetCurrentThread (),
GetCurrentProcess (),
- &self->threadH,
+ &sp->threadH,
0, FALSE, DUPLICATE_SAME_ACCESS))
{
- /* Thread structs are never freed. */
+ /*
+ * Should not do this, but we have no alternative if
+ * we can't get a Win32 thread handle.
+ * Thread structs are never freed.
+ */
ptw32_threadReusePush (self);
- return (NULL);
+ return nil;
}
#endif
@@ -117,10 +127,10 @@ pthread_self (void)
* No need to explicitly serialise access to sched_priority
* because the new handle is not yet public.
*/
- self->sched_priority = GetThreadPriority (self->threadH);
- }
+ sp->sched_priority = GetThreadPriority (sp->threadH);
- pthread_setspecific (ptw32_selfThreadKey, self);
+ pthread_setspecific (ptw32_selfThreadKey, (void *) sp);
+ }
}
return (self);
diff --git a/pthread_setcancelstate.c b/pthread_setcancelstate.c
index 4432a6f..337fb58 100644
--- a/pthread_setcancelstate.c
+++ b/pthread_setcancelstate.c
@@ -81,8 +81,9 @@ pthread_setcancelstate (int state, int *oldstate)
{
int result = 0;
pthread_t self = pthread_self ();
+ ptw32_thread_t * sp = (ptw32_thread_t *) self.p;
- if (self == NULL
+ if (sp == NULL
|| (state != PTHREAD_CANCEL_ENABLE && state != PTHREAD_CANCEL_DISABLE))
{
return EINVAL;
@@ -91,32 +92,32 @@ pthread_setcancelstate (int state, int *oldstate)
/*
* Lock for async-cancel safety.
*/
- (void) pthread_mutex_lock (&self->cancelLock);
+ (void) pthread_mutex_lock (&sp->cancelLock);
if (oldstate != NULL)
{
- *oldstate = self->cancelState;
+ *oldstate = sp->cancelState;
}
- self->cancelState = state;
+ sp->cancelState = state;
/*
* Check if there is a pending asynchronous cancel
*/
if (state == PTHREAD_CANCEL_ENABLE
- && self->cancelType == PTHREAD_CANCEL_ASYNCHRONOUS
- && WaitForSingleObject (self->cancelEvent, 0) == WAIT_OBJECT_0)
+ && sp->cancelType == PTHREAD_CANCEL_ASYNCHRONOUS
+ && WaitForSingleObject (sp->cancelEvent, 0) == WAIT_OBJECT_0)
{
- self->state = PThreadStateCanceling;
- self->cancelState = PTHREAD_CANCEL_DISABLE;
- ResetEvent (self->cancelEvent);
- (void) pthread_mutex_unlock (&self->cancelLock);
+ sp->state = PThreadStateCanceling;
+ sp->cancelState = PTHREAD_CANCEL_DISABLE;
+ ResetEvent (sp->cancelEvent);
+ (void) pthread_mutex_unlock (&sp->cancelLock);
ptw32_throw (PTW32_EPS_CANCEL);
/* Never reached */
}
- (void) pthread_mutex_unlock (&self->cancelLock);
+ (void) pthread_mutex_unlock (&sp->cancelLock);
return (result);
diff --git a/pthread_setcanceltype.c b/pthread_setcanceltype.c
index 9486353..5f88251 100644
--- a/pthread_setcanceltype.c
+++ b/pthread_setcanceltype.c
@@ -81,8 +81,9 @@ pthread_setcanceltype (int type, int *oldtype)
{
int result = 0;
pthread_t self = pthread_self ();
+ ptw32_thread_t * sp = (ptw32_thread_t *) self.p;
- if (self == NULL
+ if (sp == NULL
|| (type != PTHREAD_CANCEL_DEFERRED
&& type != PTHREAD_CANCEL_ASYNCHRONOUS))
{
@@ -92,32 +93,32 @@ pthread_setcanceltype (int type, int *oldtype)
/*
* Lock for async-cancel safety.
*/
- (void) pthread_mutex_lock (&self->cancelLock);
+ (void) pthread_mutex_lock (&sp->cancelLock);
if (oldtype != NULL)
{
- *oldtype = self->cancelType;
+ *oldtype = sp->cancelType;
}
- self->cancelType = type;
+ sp->cancelType = type;
/*
* Check if there is a pending asynchronous cancel
*/
- if (self->cancelState == PTHREAD_CANCEL_ENABLE
+ if (sp->cancelState == PTHREAD_CANCEL_ENABLE
&& type == PTHREAD_CANCEL_ASYNCHRONOUS
- && WaitForSingleObject (self->cancelEvent, 0) == WAIT_OBJECT_0)
+ && WaitForSingleObject (sp->cancelEvent, 0) == WAIT_OBJECT_0)
{
- self->state = PThreadStateCanceling;
- self->cancelState = PTHREAD_CANCEL_DISABLE;
- ResetEvent (self->cancelEvent);
- (void) pthread_mutex_unlock (&self->cancelLock);
+ sp->state = PThreadStateCanceling;
+ sp->cancelState = PTHREAD_CANCEL_DISABLE;
+ ResetEvent (sp->cancelEvent);
+ (void) pthread_mutex_unlock (&sp->cancelLock);
ptw32_throw (PTW32_EPS_CANCEL);
/* Never reached */
}
- (void) pthread_mutex_unlock (&self->cancelLock);
+ (void) pthread_mutex_unlock (&sp->cancelLock);
return (result);
diff --git a/pthread_setschedparam.c b/pthread_setschedparam.c
index d23e5d0..d3af098 100644
--- a/pthread_setschedparam.c
+++ b/pthread_setschedparam.c
@@ -72,6 +72,7 @@ ptw32_setthreadpriority (pthread_t thread, int policy, int priority)
{
int prio;
int result;
+ ptw32_thread_t * tp = (ptw32_thread_t *) thread.p;
prio = priority;
@@ -99,12 +100,12 @@ ptw32_setthreadpriority (pthread_t thread, int policy, int priority)
#endif
- result = pthread_mutex_lock (&thread->threadLock);
+ result = pthread_mutex_lock (&tp->threadLock);
if (0 == result)
{
/* If this fails, the current priority is unchanged. */
- if (0 == SetThreadPriority (thread->threadH, prio))
+ if (0 == SetThreadPriority (tp->threadH, prio))
{
result = EINVAL;
}
@@ -114,10 +115,10 @@ ptw32_setthreadpriority (pthread_t thread, int policy, int priority)
* Must record the thread's sched_priority as given,
* not as finally adjusted.
*/
- thread->sched_priority = priority;
+ tp->sched_priority = priority;
}
- (void) pthread_mutex_unlock (&thread->threadLock);
+ (void) pthread_mutex_unlock (&tp->threadLock);
}
return result;
diff --git a/pthread_setspecific.c b/pthread_setspecific.c
index b7c82d0..ed8253f 100644
--- a/pthread_setspecific.c
+++ b/pthread_setspecific.c
@@ -76,7 +76,7 @@ pthread_setspecific (pthread_key_t key, const void *value)
* thread if one wasn't explicitly created
*/
self = pthread_self ();
- if (self == NULL)
+ if (self.p == NULL)
{
return ENOENT;
}
@@ -87,11 +87,21 @@ pthread_setspecific (pthread_key_t key, const void *value)
* Resolve catch-22 of registering thread with selfThread
* key
*/
- self = (pthread_t) pthread_getspecific (ptw32_selfThreadKey);
- if (self == NULL)
- {
- self = (pthread_t) value;
- }
+ ptw32_thread_t * sp;
+
+ sp = (ptw32_thread_t *) pthread_getspecific (ptw32_selfThreadKey);
+ if (sp == NULL)
+ {
+ if (value == NULL)
+ {
+ return ENOENT;
+ }
+ self = *((pthread_t *) value);
+ }
+ else
+ {
+ self = sp->ptHandle;
+ }
}
result = 0;
@@ -100,8 +110,10 @@ pthread_setspecific (pthread_key_t key, const void *value)
{
ThreadKeyAssoc *assoc;
- if (self != NULL && key->destructor != NULL && value != NULL)
+ if (self.p != NULL && key->destructor != NULL && value != NULL)
{
+ ptw32_thread_t * sp = (ptw32_thread_t *) self.p;
+
/*
* Only require associations if we have to
* call user destroy routine.
@@ -111,7 +123,7 @@ pthread_setspecific (pthread_key_t key, const void *value)
* on the association; setting assoc to NULL short
* circuits the search.
*/
- assoc = (ThreadKeyAssoc *) self->keys;
+ assoc = (ThreadKeyAssoc *) sp->keys;
/*
* Locate existing association
*/
diff --git a/pthread_testcancel.c b/pthread_testcancel.c
index 04c214a..2a1c080 100644
--- a/pthread_testcancel.c
+++ b/pthread_testcancel.c
@@ -69,21 +69,34 @@ pthread_testcancel (void)
*/
{
pthread_t self = pthread_self ();
+ ptw32_thread_t * sp = (ptw32_thread_t *) self.p;
- (void) pthread_mutex_lock (&self->cancelLock);
+ if (sp == NULL)
+ {
+ return;
+ }
+
+ /*
+ * Pthread_cancel() will have set sp->state to PThreadStateCancelPending
+ * and set an event, so no need to enter kernel space if
+ * sp->state != PThreadStateCancelPending - that only slows us down.
+ */
+ if (sp->state != PThreadStateCancelPending)
+ {
+ return;
+ }
+
+ (void) pthread_mutex_lock (&sp->cancelLock);
- if (self != NULL
- && self->cancelState != PTHREAD_CANCEL_DISABLE
- && WaitForSingleObject (self->cancelEvent, 0) == WAIT_OBJECT_0)
+ if (sp->cancelState != PTHREAD_CANCEL_DISABLE)
{
- /*
- * Canceling!
- */
- self->state = PThreadStateCanceling;
- self->cancelState = PTHREAD_CANCEL_DISABLE;
- (void) pthread_mutex_unlock (&self->cancelLock);
+ ResetEvent(sp->cancelEvent);
+ sp->state = PThreadStateCanceling;
+ (void) pthread_mutex_unlock (&sp->cancelLock);
+ sp->cancelState = PTHREAD_CANCEL_DISABLE;
+ (void) pthread_mutex_unlock (&sp->cancelLock);
ptw32_throw (PTW32_EPS_CANCEL);
}
- (void) pthread_mutex_unlock (&self->cancelLock);
+ (void) pthread_mutex_unlock (&sp->cancelLock);
} /* pthread_testcancel */
diff --git a/pthread_win32_attach_detach_np.c b/pthread_win32_attach_detach_np.c
index 4aedccc..6fd7f27 100644
--- a/pthread_win32_attach_detach_np.c
+++ b/pthread_win32_attach_detach_np.c
@@ -92,8 +92,6 @@ pthread_win32_process_attach_np ()
#endif
-#ifndef TEST_ICE
-
/*
* Load KERNEL32 and try to get address of InterlockedCompareExchange
*/
@@ -136,12 +134,6 @@ pthread_win32_process_attach_np ()
ptw32_features |= PTW32_SYSTEM_INTERLOCKED_COMPARE_EXCHANGE;
}
-#else /* TEST_ICE */
-
- ptw32_interlocked_compare_exchange = ptw32_InterlockedCompareExchange;
-
-#endif /* TEST_ICE */
-
/*
* Load QUSEREX.DLL and try to get address of QueueUserAPCEx
*/
@@ -204,16 +196,19 @@ pthread_win32_process_detach_np ()
{
if (ptw32_processInitialized)
{
- pthread_t self = (pthread_t) pthread_getspecific (ptw32_selfThreadKey);
+ ptw32_thread_t * sp = (ptw32_thread_t *) pthread_getspecific (ptw32_selfThreadKey);
- /*
- * Detached threads have their resources automatically
- * cleaned up upon exit (others must be 'joined').
- */
- if (self != NULL && self->detachState == PTHREAD_CREATE_DETACHED)
+ if (sp != NULL)
{
- ptw32_threadDestroy (self);
- TlsSetValue (ptw32_selfThreadKey->key, NULL);
+ /*
+ * Detached threads have their resources automatically
+ * cleaned up upon exit (others must be 'joined').
+ */
+ if (sp->detachState == PTHREAD_CREATE_DETACHED)
+ {
+ ptw32_threadDestroy (sp->ptHandle);
+ TlsSetValue (ptw32_selfThreadKey->key, NULL);
+ }
}
/*
@@ -265,16 +260,19 @@ pthread_win32_thread_detach_np ()
* Don't use pthread_self() - to avoid creating an implicit POSIX thread handle
* unnecessarily.
*/
- pthread_t self = (pthread_t) pthread_getspecific (ptw32_selfThreadKey);
+ ptw32_thread_t * sp = (ptw32_thread_t *) pthread_getspecific (ptw32_selfThreadKey);
- /*
- * Detached threads have their resources automatically
- * cleaned up upon exit (others must be 'joined').
- */
- if (self != NULL && self->detachState == PTHREAD_CREATE_DETACHED)
+ if (sp != NULL)
{
- ptw32_threadDestroy (self);
- TlsSetValue (ptw32_selfThreadKey->key, NULL);
+ /*
+ * Detached threads have their resources automatically
+ * cleaned up upon exit (others must be 'joined').
+ */
+ if (sp->detachState == PTHREAD_CREATE_DETACHED)
+ {
+ ptw32_threadDestroy (sp->ptHandle);
+ TlsSetValue (ptw32_selfThreadKey->key, NULL);
+ }
}
}
diff --git a/ptw32_InterlockedCompareExchange.c b/ptw32_InterlockedCompareExchange.c
index 6da2bec..45f8d59 100644
--- a/ptw32_InterlockedCompareExchange.c
+++ b/ptw32_InterlockedCompareExchange.c
@@ -46,7 +46,7 @@
* We now use this version wherever possible so we can inline it.
*/
-INLINE PTW32_INTERLOCKED_LONG WINAPI
+PTW32_INTERLOCKED_LONG WINAPI
ptw32_InterlockedCompareExchange (PTW32_INTERLOCKED_LPLONG location,
PTW32_INTERLOCKED_LONG value,
PTW32_INTERLOCKED_LONG comparand)
@@ -159,7 +159,7 @@ ptw32_InterlockedCompareExchange (PTW32_INTERLOCKED_LPLONG location,
* We now use this version wherever possible so we can inline it.
*/
-INLINE LONG WINAPI
+LONG WINAPI
ptw32_InterlockedExchange (LPLONG location,
LONG value)
{
diff --git a/ptw32_callUserDestroyRoutines.c b/ptw32_callUserDestroyRoutines.c
index 90b0b73..081a60d 100644
--- a/ptw32_callUserDestroyRoutines.c
+++ b/ptw32_callUserDestroyRoutines.c
@@ -69,7 +69,7 @@ ptw32_callUserDestroyRoutines (pthread_t thread)
ThreadKeyAssoc *
assoc;
- if (thread != NULL)
+ if (thread.p != NULL)
{
/*
* Run through all Thread<-->Key associations
@@ -82,7 +82,7 @@ ptw32_callUserDestroyRoutines (pthread_t thread)
* by the current thread and must be released; otherwise
* the assoc will be destroyed when the key is destroyed.
*/
- nextP = (ThreadKeyAssoc **) & (thread->keys);
+ nextP = (ThreadKeyAssoc **) &((ptw32_thread_t *)thread.p)->keys;
assoc = *nextP;
while (assoc != NULL)
@@ -90,8 +90,9 @@ ptw32_callUserDestroyRoutines (pthread_t thread)
if (pthread_mutex_lock (&(assoc->lock)) == 0)
{
- pthread_key_t
- k;
+ pthread_key_t k;
+ pthread_t nil = {NULL, 0};
+
if ((k = assoc->key) != NULL)
{
/*
@@ -101,8 +102,7 @@ ptw32_callUserDestroyRoutines (pthread_t thread)
* key is valid and we can call the destroy
* routine;
*/
- void *
- value = NULL;
+ void * value = NULL;
value = pthread_getspecific (k);
if (value != NULL && k->destructor != NULL)
@@ -146,7 +146,7 @@ ptw32_callUserDestroyRoutines (pthread_t thread)
* mark assoc->thread as NULL to indicate the
* thread no longer references this association
*/
- assoc->thread = NULL;
+ assoc->thread = nil;
/*
* Remove association from the pthread_t chain
diff --git a/ptw32_new.c b/ptw32_new.c
index c2da5d5..21d08ce 100644
--- a/ptw32_new.c
+++ b/ptw32_new.c
@@ -42,37 +42,48 @@ pthread_t
ptw32_new (void)
{
pthread_t t;
+ pthread_t nil = {NULL, 0};
+ ptw32_thread_t * tp;
/*
* If there's a reusable pthread_t then use it.
*/
t = ptw32_threadReusePop ();
- if (NULL == t)
+ if (NULL != t.p)
{
- t = (pthread_t) calloc (1, sizeof (*t));
+ tp = (ptw32_thread_t *) t.p;
}
-
- if (t != NULL)
+ else
{
- t->sched_priority = THREAD_PRIORITY_NORMAL;
- t->detachState = PTHREAD_CREATE_JOINABLE;
- t->cancelState = PTHREAD_CANCEL_ENABLE;
- t->cancelType = PTHREAD_CANCEL_DEFERRED;
- t->cancelLock = PTHREAD_MUTEX_INITIALIZER;
- t->threadLock = PTHREAD_MUTEX_INITIALIZER;
- t->cancelEvent = CreateEvent (0, (int) PTW32_TRUE, /* manualReset */
- (int) PTW32_FALSE, /* setSignaled */
- NULL);
+ /* No reuse threads available */
+ tp = (ptw32_thread_t *) calloc (1, sizeof(ptw32_thread_t));
- if (t->cancelEvent == NULL)
+ if (tp == NULL)
{
- /*
- * Thread ID structs are never freed.
- */
- ptw32_threadReusePush (t);
- t = NULL;
+ return nil;
}
+
+ /* ptHandle.p needs to point to it's parent ptw32_thread_t. */
+ t.p = tp->ptHandle.p = tp;
+ t.x = tp->ptHandle.x = 0;
+ }
+
+ /* Set default state. */
+ tp->sched_priority = THREAD_PRIORITY_NORMAL;
+ tp->detachState = PTHREAD_CREATE_JOINABLE;
+ tp->cancelState = PTHREAD_CANCEL_ENABLE;
+ tp->cancelType = PTHREAD_CANCEL_DEFERRED;
+ tp->cancelLock = PTHREAD_MUTEX_INITIALIZER;
+ tp->threadLock = PTHREAD_MUTEX_INITIALIZER;
+ tp->cancelEvent = CreateEvent (0, (int) PTW32_TRUE, /* manualReset */
+ (int) PTW32_FALSE, /* setSignaled */
+ NULL);
+
+ if (tp->cancelEvent == NULL)
+ {
+ ptw32_threadReusePush (tp->ptHandle);
+ return nil;
}
return t;
diff --git a/ptw32_processTerminate.c b/ptw32_processTerminate.c
index 13d73aa..b36c5a7 100644
--- a/ptw32_processTerminate.c
+++ b/ptw32_processTerminate.c
@@ -62,10 +62,9 @@ ptw32_processTerminate (void)
* ------------------------------------------------------
*/
{
- pthread_t thread, nextThread;
-
if (ptw32_processInitialized)
{
+ ptw32_thread_t * tp, * tpNext;
if (ptw32_selfThreadKey != NULL)
{
@@ -89,12 +88,12 @@ ptw32_processTerminate (void)
EnterCriticalSection (&ptw32_thread_reuse_lock);
- thread = ptw32_threadReuseTop;
- while (thread != PTW32_THREAD_REUSE_BOTTOM)
+ tp = ptw32_threadReuseTop;
+ while (tp != PTW32_THREAD_REUSE_EMPTY)
{
- nextThread = thread->prevReuse;
- free (thread);
- thread = nextThread;
+ tpNext = tp->prevReuse;
+ free (tp);
+ tp = tpNext;
}
LeaveCriticalSection (&ptw32_thread_reuse_lock);
diff --git a/ptw32_reuse.c b/ptw32_reuse.c
index a21737a..7056395 100644
--- a/ptw32_reuse.c
+++ b/ptw32_reuse.c
@@ -39,10 +39,34 @@
/*
+ * How it works:
+ * A pthread_t is a struct (2x32 bit scalar types on IA-32, 2x64 bit on IA-64)
+ * which is normally passed/returned by value to/from pthreads routines.
+ * Applications are therefore storing a copy of the struct as it is at that
+ * time.
+ *
+ * The original pthread_t struct plus all copies of it contain the address of
+ * the thread state struct ptw32_thread_t_ (p), plus a reuse counter (x). Each
+ * ptw32_thread_t contains the original copy of it's pthread_t.
+ * Once malloced, a ptw2_thread_t_ struct is never freed.
+ *
* The thread reuse stack is a simple LIFO stack managed through a singly
- * linked list element in the pthread_t_ struct.
+ * linked list element in the ptw32_thread_t.
+ *
+ * Each time a thread is destroyed, the ptw32_thread_t address is pushed onto the
+ * reuse stack after it's ptHandle's reuse counter has been incremented.
+ *
+ * The following can now be said from this:
+ * - two pthread_t's are identical if their ptw32_thread_t reference pointers
+ * are equal and their reuse couters are equal. That is,
+ *
+ * equal = (a.p == b.p && a.x == b.x)
+ *
+ * - a pthread_t copy refers to a destroyed thread if the reuse counter in
+ * the copy is not equal to the reuse counter in the original.
+ *
+ * threadDestroyed = (copy.x != ((ptw32_thread_t *)copy.p)->ptHandle.x)
*
- * All thread structs on the stack are clean and ready for reuse.
*/
/*
@@ -51,20 +75,26 @@
pthread_t
ptw32_threadReusePop (void)
{
- pthread_t t;
+ pthread_t t = {NULL, 0};
EnterCriticalSection (&ptw32_thread_reuse_lock);
- t = ptw32_threadReuseTop;
-
- if (PTW32_THREAD_REUSE_BOTTOM != t)
- {
- ptw32_threadReuseTop = t->prevReuse;
- t->prevReuse = NULL;
- }
- else
+ if (PTW32_THREAD_REUSE_EMPTY != ptw32_threadReuseTop)
{
- t = NULL;
+ ptw32_thread_t * tp;
+
+ tp = ptw32_threadReuseTop;
+
+ ptw32_threadReuseTop = tp->prevReuse;
+
+ if (PTW32_THREAD_REUSE_EMPTY == ptw32_threadReuseTop)
+ {
+ ptw32_threadReuseBottom = PTW32_THREAD_REUSE_EMPTY;
+ }
+
+ tp->prevReuse = NULL;
+
+ t = tp->ptHandle;
}
LeaveCriticalSection (&ptw32_thread_reuse_lock);
@@ -75,15 +105,39 @@ ptw32_threadReusePop (void)
/*
* Push a clean pthread_t struct onto the reuse stack.
+ * Must be re-initialised when reused.
+ * All object elements (mutexes, events etc) must have been either
+ * detroyed before this, or never initialised.
*/
void
ptw32_threadReusePush (pthread_t thread)
{
+ ptw32_thread_t * tp = (ptw32_thread_t *) thread.p;
+ pthread_t t;
+
EnterCriticalSection (&ptw32_thread_reuse_lock);
- memset (thread, 0, sizeof (*thread));
- thread->prevReuse = ptw32_threadReuseTop;
- ptw32_threadReuseTop = thread;
+ t = tp->ptHandle;
+ memset(tp, 0, sizeof(ptw32_thread_t));
+
+ /* Must restore the original POSIX handle that we just wiped. */
+ tp->ptHandle = t;
+
+ /* Bump the reuse counter now */
+ tp->ptHandle.x++;
+
+ tp->prevReuse = PTW32_THREAD_REUSE_EMPTY;
+
+ if (PTW32_THREAD_REUSE_EMPTY != ptw32_threadReuseBottom)
+ {
+ ptw32_threadReuseBottom->prevReuse = tp;
+ }
+ else
+ {
+ ptw32_threadReuseTop = tp;
+ }
+
+ ptw32_threadReuseBottom = tp;
LeaveCriticalSection (&ptw32_thread_reuse_lock);
}
diff --git a/ptw32_semwait.c b/ptw32_semwait.c
index e8b7ace..3956a17 100644
--- a/ptw32_semwait.c
+++ b/ptw32_semwait.c
@@ -41,7 +41,7 @@
#include "implement.h"
-INLINE int
+int
ptw32_semwait (sem_t * sem)
/*
* ------------------------------------------------------
diff --git a/ptw32_threadDestroy.c b/ptw32_threadDestroy.c
index 58b23ff..9bca1b4 100644
--- a/ptw32_threadDestroy.c
+++ b/ptw32_threadDestroy.c
@@ -42,20 +42,21 @@
void
ptw32_threadDestroy (pthread_t thread)
{
- struct pthread_t_ threadCopy;
+ ptw32_thread_t * tp = (ptw32_thread_t *) thread.p;
+ ptw32_thread_t threadCopy;
- if (thread != NULL)
+ if (tp != NULL)
{
- (void) pthread_mutex_lock (&thread->cancelLock);
- thread->state = PThreadStateLast;
- (void) pthread_mutex_unlock (&thread->cancelLock);
+ (void) pthread_mutex_lock (&tp->cancelLock);
+ tp->state = PThreadStateLast;
+ (void) pthread_mutex_unlock (&tp->cancelLock);
ptw32_callUserDestroyRoutines (thread);
/*
* Copy thread state so that the thread can be atomically NULLed.
*/
- memcpy (&threadCopy, thread, sizeof (threadCopy));
+ memcpy (&threadCopy, tp, sizeof (threadCopy));
/*
* Thread ID structs are never freed. They're NULLed and reused.
diff --git a/ptw32_threadStart.c b/ptw32_threadStart.c
index 8d48621..a067aa2 100644
--- a/ptw32_threadStart.c
+++ b/ptw32_threadStart.c
@@ -69,7 +69,7 @@ ExceptionFilter (EXCEPTION_POINTERS * ep, DWORD * ei)
*/
pthread_t self = pthread_self ();
- (void) pthread_mutex_destroy (&self->cancelLock);
+ (void) pthread_mutex_destroy (&((ptw32_thread_t *)self.p)->cancelLock);
ptw32_callUserDestroyRoutines (self);
return EXCEPTION_CONTINUE_SEARCH;
@@ -109,10 +109,11 @@ static terminate_function
void
ptw32_terminate ()
{
- pthread_t
- self = pthread_self ();
+ pthread_t self = pthread_self ();
+ ptw32_thread_t * sp = (ptw32_thread_t *) self.p;
+
set_terminate (ptw32_oldTerminate);
- (void) pthread_mutex_destroy (&self->cancelLock);
+ (void) pthread_mutex_destroy (&sp->cancelLock);
ptw32_callUserDestroyRoutines (self);
terminate ();
}
@@ -127,13 +128,11 @@ void
#endif
ptw32_threadStart (void *vthreadParms)
{
- ThreadParms *
- threadParms = (ThreadParms *) vthreadParms;
- pthread_t
- self;
+ ThreadParms * threadParms = (ThreadParms *) vthreadParms;
+ pthread_t self;
+ ptw32_thread_t * sp;
void *(*start) (void *);
- void *
- arg;
+ void * arg;
#ifdef __CLEANUP_SEH
DWORD
@@ -141,14 +140,13 @@ ptw32_threadStart (void *vthreadParms)
#endif
#ifdef __CLEANUP_C
- int
- setjmp_rc;
+ int setjmp_rc;
#endif
- void *
- status = (void *) 0;
+ void * status = (void *) 0;
self = threadParms->tid;
+ sp = (ptw32_thread_t *) self.p;
start = threadParms->start;
arg = threadParms->arg;
@@ -159,21 +157,21 @@ ptw32_threadStart (void *vthreadParms)
* beginthread does not return the thread id and is running
* before it returns us the thread handle, and so we do it here.
*/
- self->thread = GetCurrentThreadId ();
+ sp->thread = GetCurrentThreadId ();
/*
* Here we're using cancelLock as a general-purpose lock
* to make the new thread wait until the creating thread
* has the new handle.
*/
- if (pthread_mutex_lock (&self->cancelLock) == 0)
+ if (pthread_mutex_lock (&sp->cancelLock) == 0)
{
- (void) pthread_mutex_unlock (&self->cancelLock);
+ (void) pthread_mutex_unlock (&sp->cancelLock);
}
#endif
- pthread_setspecific (ptw32_selfThreadKey, self);
+ pthread_setspecific (ptw32_selfThreadKey, sp);
- self->state = PThreadStateRunning;
+ sp->state = PThreadStateRunning;
#ifdef __CLEANUP_SEH
@@ -182,7 +180,7 @@ ptw32_threadStart (void *vthreadParms)
/*
* Run the caller's routine;
*/
- status = self->exitStatus = (*start) (arg);
+ status = sp->exitStatus = (*start) (arg);
#ifdef _UWIN
if (--pthread_count <= 0)
@@ -202,7 +200,7 @@ ptw32_threadStart (void *vthreadParms)
#endif
break;
case PTW32_EPS_EXIT:
- status = self->exitStatus;
+ status = sp->exitStatus;
break;
default:
status = PTHREAD_CANCELED;
@@ -214,7 +212,7 @@ ptw32_threadStart (void *vthreadParms)
#ifdef __CLEANUP_C
- setjmp_rc = setjmp (self->start_mark);
+ setjmp_rc = setjmp (sp->start_mark);
if (0 == setjmp_rc)
{
@@ -222,9 +220,8 @@ ptw32_threadStart (void *vthreadParms)
/*
* Run the caller's routine;
*/
- status = self->exitStatus = (*start) (arg);
+ status = sp->exitStatus = (*start) (arg);
}
-
else
{
switch (setjmp_rc)
@@ -233,7 +230,7 @@ ptw32_threadStart (void *vthreadParms)
status = PTHREAD_CANCELED;
break;
case PTW32_EPS_EXIT:
- status = self->exitStatus;
+ status = sp->exitStatus;
break;
default:
status = PTHREAD_CANCELED;
@@ -256,7 +253,7 @@ ptw32_threadStart (void *vthreadParms)
*/
try
{
- status = self->exitStatus = (*start) (arg);
+ status = sp->exitStatus = (*start) (arg);
}
catch (ptw32_exception &)
{
@@ -293,14 +290,14 @@ ptw32_threadStart (void *vthreadParms)
/*
* Thread was canceled.
*/
- status = self->exitStatus = PTHREAD_CANCELED;
+ status = sp->exitStatus = PTHREAD_CANCELED;
}
catch (ptw32_exception_exit &)
{
/*
* Thread was exited via pthread_exit().
*/
- status = self->exitStatus;
+ status = sp->exitStatus;
}
catch (...)
{
@@ -309,11 +306,11 @@ ptw32_threadStart (void *vthreadParms)
* terminate routine. We get control back within this block - cleanup
* and release the exception out of thread scope.
*/
- status = self->exitStatus = PTHREAD_CANCELED;
- (void) pthread_mutex_lock (&self->cancelLock);
- self->state = PThreadStateException;
- (void) pthread_mutex_unlock (&self->cancelLock);
- (void) pthread_mutex_destroy (&self->cancelLock);
+ status = sp->exitStatus = PTHREAD_CANCELED;
+ (void) pthread_mutex_lock (&sp->cancelLock);
+ sp->state = PThreadStateException;
+ (void) pthread_mutex_unlock (&sp->cancelLock);
+ (void) pthread_mutex_destroy (&sp->cancelLock);
(void) set_terminate (ptw32_oldTerminate);
ptw32_callUserDestroyRoutines (self);
throw;
@@ -333,7 +330,7 @@ ptw32_threadStart (void *vthreadParms)
#endif /* __CLEANUP_C */
#endif /* __CLEANUP_SEH */
- if (self->detachState == PTHREAD_CREATE_DETACHED)
+ if (sp->detachState == PTHREAD_CREATE_DETACHED)
{
/*
* We need to cleanup the pthread now in case we have
diff --git a/ptw32_throw.c b/ptw32_throw.c
index 8f67069..1a71b97 100644
--- a/ptw32_throw.c
+++ b/ptw32_throw.c
@@ -53,7 +53,7 @@ ptw32_throw (DWORD exception)
* Don't use pthread_self() to avoid creating an implicit POSIX thread handle
* unnecessarily.
*/
- pthread_t self = (pthread_t) pthread_getspecific (ptw32_selfThreadKey);
+ ptw32_thread_t * sp = (ptw32_thread_t *) pthread_getspecific (ptw32_selfThreadKey);
#ifdef __CLEANUP_SEH
DWORD exceptionInformation[3];
@@ -65,7 +65,7 @@ ptw32_throw (DWORD exception)
exit (1);
}
- if (NULL == self || self->implicit)
+ if (NULL == sp || sp->implicit)
{
/*
* We're inside a non-POSIX initialised Win32 thread
@@ -81,7 +81,7 @@ ptw32_throw (DWORD exception)
exitCode = (unsigned) PTHREAD_CANCELED;
break;
case PTW32_EPS_EXIT:
- exitCode = (unsigned) self->exitStatus;;
+ exitCode = (unsigned) sp->exitStatus;;
break;
}
@@ -109,7 +109,7 @@ ptw32_throw (DWORD exception)
#ifdef __CLEANUP_C
ptw32_pop_cleanup_all (1);
- longjmp (self->start_mark, exception);
+ longjmp (sp->start_mark, exception);
#else /* __CLEANUP_C */
diff --git a/ptw32_tkAssocCreate.c b/ptw32_tkAssocCreate.c
index c2a5c81..1e3e3e6 100644
--- a/ptw32_tkAssocCreate.c
+++ b/ptw32_tkAssocCreate.c
@@ -112,13 +112,13 @@ ptw32_tkAssocCreate (ThreadKeyAssoc ** assocP,
pthread_mutex_unlock (&(key->threadsLock));
- if (thread != NULL)
+ if (thread.p != NULL)
{
/*
* Register assoc with thread
*/
- assoc->nextKey = (ThreadKeyAssoc *) thread->keys;
- thread->keys = (void *) assoc;
+ assoc->nextKey = (ThreadKeyAssoc *) ((ptw32_thread_t *)thread.p)->keys;
+ ((ptw32_thread_t *)thread.p)->keys = (void *) assoc;
}
*assocP = assoc;
diff --git a/ptw32_tkAssocDestroy.c b/ptw32_tkAssocDestroy.c
index 682b896..1022dea 100644
--- a/ptw32_tkAssocDestroy.c
+++ b/ptw32_tkAssocDestroy.c
@@ -56,7 +56,7 @@ ptw32_tkAssocDestroy (ThreadKeyAssoc * assoc)
*/
{
- if ((assoc != NULL) && (assoc->key == NULL && assoc->thread == NULL))
+ if ((assoc != NULL) && (assoc->key == NULL && assoc->thread.p == NULL))
{
pthread_mutex_destroy (&(assoc->lock));
diff --git a/sem_timedwait.c b/sem_timedwait.c
index d1d9d37..59fcc18 100644
--- a/sem_timedwait.c
+++ b/sem_timedwait.c
@@ -52,15 +52,18 @@
#include "semaphore.h"
#include "implement.h"
-static inline void
+static void PTW32_CDECL
ptw32_sem_timedwait_cleanup (void * sem)
{
sem_t s = (sem_t) sem;
if (pthread_mutex_lock (&s->lock) == 0)
{
- s->value++;
- ReleaseSemaphore(s->sem, 1, 0);
+ ++s->value;
+ /*
+ * Don't release the W32 sema, it doesn't need adjustment
+ * because it doesn't record the number of waiters.
+ */
(void) pthread_mutex_unlock (&s->lock);
}
}
@@ -202,6 +205,8 @@ sem_timedwait (sem_t * sem, const struct timespec *abstime)
#else /* NEED_SEM */
+ pthread_testcancel();
+
if ((result = pthread_mutex_lock (&s->lock)) == 0)
{
int v = --s->value;
@@ -210,15 +215,21 @@ sem_timedwait (sem_t * sem, const struct timespec *abstime)
if (v < 0)
{
/* Must wait */
- pthread_cleanup_push(ptw32_sem_timedwait_cleanup, s);
+#ifdef _MSC_VER
+#pragma inline_depth(0)
+#endif
+ pthread_cleanup_push(ptw32_sem_timedwait_cleanup, (void *) s);
result = pthreadCancelableTimedWait (s->sem, milliseconds);
/*
- * Restore the semaphore counters if no longer waiting
+ * Restore the semaphore counter if no longer waiting
* and not taking the semaphore. This will occur if the
* thread is cancelled while waiting, or the wake was
* not the result of a post event given to us, e.g. a timeout.
*/
pthread_cleanup_pop(result);
+#ifdef _MSC_VER
+#pragma inline_depth()
+#endif
}
}
diff --git a/sem_wait.c b/sem_wait.c
index cead2cd..68131bf 100644
--- a/sem_wait.c
+++ b/sem_wait.c
@@ -45,15 +45,16 @@
#include "semaphore.h"
#include "implement.h"
-static inline void
+
+static void PTW32_CDECL
ptw32_sem_wait_cleanup(void * sem)
{
sem_t s = (sem_t) sem;
if (pthread_mutex_lock (&s->lock) == 0)
{
- s->value++;
- ReleaseSemaphore(s->sem, 1, 0);
+ ++s->value;
+ /* Don't release the W32 sema, it should always == 0. */
(void) pthread_mutex_unlock (&s->lock);
}
}
@@ -105,6 +106,12 @@ sem_wait (sem_t * sem)
#else /* NEED_SEM */
+ /*
+ * sem_wait is a cancelation point and it's easy to test before
+ * modifying the sem value
+ */
+ pthread_testcancel();
+
if ((result = pthread_mutex_lock (&s->lock)) == 0)
{
int v = --s->value;
@@ -114,18 +121,18 @@ sem_wait (sem_t * sem)
if (v < 0)
{
/* Must wait */
- pthread_cleanup_push(ptw32_sem_wait_cleanup, s);
+#ifdef _MSC_VER
+#pragma inline_depth(0)
+#endif
+ pthread_cleanup_push(ptw32_sem_wait_cleanup, (void *) s);
result = pthreadCancelableWait (s->sem);
- /*
- * Restore the semaphore counters if no longer waiting
- * and not taking the semaphore. This will occur if the
- * thread is cancelled while waiting, or the wake was
- * not the result of a post event given to us.
- */
- pthread_cleanup_pop(result);
+ pthread_cleanup_pop(result != 0);
+#ifdef _MSC_VER
+#pragma inline_depth()
+#endif
}
}
-
+
#endif /* NEED_SEM */
}
diff --git a/signal.c b/signal.c
index a7c3ef0..21747f6 100644
--- a/signal.c
+++ b/signal.c
@@ -101,7 +101,7 @@ pthread_sigmask (int how, sigset_t const *set, sigset_t * oset)
{
pthread_t thread = pthread_self ();
- if (thread == NULL)
+ if (thread.p == NULL)
{
return ENOENT;
}
@@ -126,7 +126,7 @@ pthread_sigmask (int how, sigset_t const *set, sigset_t * oset)
/* Copy the old mask before modifying it. */
if (oset != NULL)
{
- memcpy (oset, &(thread->sigmask), sizeof (sigset_t));
+ memcpy (oset, &(thread.p->sigmask), sizeof (sigset_t));
}
if (set != NULL)
@@ -137,7 +137,7 @@ pthread_sigmask (int how, sigset_t const *set, sigset_t * oset)
the size of a long integer. */
unsigned long *src = (unsigned long const *) set;
- unsigned long *dest = (unsigned long *) &(thread->sigmask);
+ unsigned long *dest = (unsigned long *) &(thread.p->sigmask);
switch (how)
{
@@ -156,7 +156,7 @@ pthread_sigmask (int how, sigset_t const *set, sigset_t * oset)
}
case SIG_SETMASK:
/* Replace the whole sigmask. */
- memcpy (&(thread->sigmask), set, sizeof (sigset_t));
+ memcpy (&(thread.p->sigmask), set, sizeof (sigset_t));
break;
}
}
diff --git a/tests/ChangeLog b/tests/ChangeLog
index 27f4eea..abc8260 100644
--- a/tests/ChangeLog
+++ b/tests/ChangeLog
@@ -1,4 +1,17 @@
-2004-10-23 Ross Johnson <rpj@callisto.canberra.edu.au>
+2004-10-29 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * semaphore4.c: New test.
+ * semaphore4t.c: New test.
+ * Debug.dsp (et al): Created MSVC Workspace project to aid debugging.
+ * All: Many tests have been modified to work with the new pthread
+ ID type; some other corrections were made after some library
+ functions were semantically strengthened. For example,
+ pthread_cond_destroy() no longer destroys a busy CV, which
+ required minor redesigns of some tests, including some where
+ the mutex associated with the CV was not locked during
+ signaling and broadcasting.
+
+2004-10-23 Ross Johnson <rpj@callisto.canberra.edu.au>
* condvar3.c: Fixed mutex operations that were incorrectly
placed in relation to their condition variable operations.
@@ -15,791 +28,791 @@
* condvar8.c: Likewise.
* condvar9.c: Likewise.
-2004-10-19 Ross Johnson <rpj@callisto.canberra.edu.au>
+2004-10-19 Ross Johnson <rpj@callisto.canberra.edu.au>
* semaphore3.c: New test.
-2004-10-14 Ross Johnson <rpj@callisto.canberra.edu.au>
+2004-10-14 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * rwlock7.c (main): Tidy up statistics reporting; randomise
+ update accesses.
+ * rwlock8.c: New test.
+
+2004-09-08 Alexandre Girao <alexgirao@gmail.com>
+
+ * cancel7.c (main): Win98 wants a valid (non-NULL) location
+ for the last arg of _beginthreadex().
+ * cancel8.c (main): Likewise.
+ * exit4.c (main): Likewise.
+ * exit5.c (main): Likewise.
+
+2004-08-26 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * create3.c: New test.
+
+2004-06-21 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * mutex2r.c: New test.
+ * mutex2e.c: New test.
+ * mutex3r.c: New test.
+ * mutex3e.c: New test.
+ * mutex6s.c: New test.
+ * mutex6rs.c: New test.
+ * mutex6es.c: New test.
+
+2004-05-21 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * join3.c: New test.
+
+2004-05-16 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * condvar2.c (WIN32_WINNT): Define to avoid redefinition warning
+ from inclusion of implement.h.
+ * convar2_1.c: Likewise.
+ * condvar3_1.c: Likewise.
+ * condvar3_2.c: Likewise.
+ * context1.c: Likewise.
+ * sizes.c: Likewise.
+ * Makefile: Don't define _WIN32_WINNT on compiler command line.
+ * GNUmakefile: Likewise.
+ * priority1.c (main): Add column to output for actual win32
+ priority.
+
+2004-05-16 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * cancel9.c: New test.
+ * cancel3.c: Remove inappropriate conditional compilation;
+ GNU C version of test suite no longer quietly skips this test.
+ * cancel5.c: Likewise.
+ * GNUmakefile: Can now build individual test app using default
+ C version of library using 'make clean testname.c'.
+ * Makefile: Likewise for VC using 'nmake clean test testname.c'.
+
+2003-10-14 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * Wmakefile: New makefile for Watcom testing.
+
+2003-09-18 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * benchtest.h: Move old mutex code into benchlib.c.
+ * benchlib.c: New statically linked module to ensure that
+ bench apps don't inline the code and therefore have an unfair
+ advantage over the pthreads lib routines. Made little or no
+ difference.
+ * benchtest1.c: Minor change to avoid compiler warnings.
+ * benchtest5.c: Likewise.
+ * benchtest2.c: Fix misinformation in output report.
+ * README.BENCH: Add comments on results.
+
+2003-09-14 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * priority1.c: Reworked to comply with modified priority
+ management and provide additional output.
+ * priority2.c: Likewise.
+ * inherit1.c: Likewise.
+
+2003-09-03 Ross Johnson <rpj@callisto.canberra.edu.au>
+
+ * exit4.c: New test.
+ * exit5.c: New test.
+ * cancel7.c: New test.
+ * cancel8.c: New test.
+
+2003-08-13 Ross Johnson <rpj@ise.canberra.edu.au>
+
+ * reuse1.c: New test.
+ * reuse1.c: New test.
+ * valid1.c: New test.
+ * valid2.c: New test.
+ * kill1.c: New test.
+ * create2.c: Now included in test regime.
+
+2003-07-19 Ross Johnson <rpj@ise.canberra.edu.au>
+
+ * eyal1.c (waste_time): Make threads do more work to ensure that
+ all threads get to do some work.
+ * semaphore1.c: Make it clear that certain errors are expected.
+ * exception2.c (non_MSVC code sections): Change to include
+ C++ standard include file, i.e. change <new.h> to <exception>.
+ * exception3.c (non_MSVC code sections): Likewise; qualify std::
+ namespace entities where necessary.
+ * GNUmakefile: modified to work in the MsysDTK (newer MinGW)
+ environment; define CC as gcc or g++ as appropriate because
+ using gcc -x c++ doesn't link with required c++ libs by default,
+ but g++ does.
+
+2002-12-11 Ross Johnson <ross@special.ise.canberra.edu.au>
+
+ * mutex7e.c: Assert EBUSY return instead of EDEADLK.
+
+2002-06-03 Ross Johnson <rpj@digit.ise.canberra.edu.au>
+
+ * semaphore2.c: New test.
+
+2002-03-02 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * Makefile (CFLAGS): Changed /MT to /MD to link with
+ the correct library MSVCRT.LIB. Otherwise errno doesn't
+ work.
+
+2002-02-28 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * exception3.c: Correct recent change.
+
+ * semaphore1.c: New test.
+
+ * Makefile: Add rule to generate pre-processor output.
+
+2002-02-28 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * exception3.c (terminateFunction): For MSVC++, call
+ exit() rather than pthread_exit(). Add comments to explain
+ why.
+ * Notes from the MSVC++ manual:
+ * 1) A term_func() should call exit(), otherwise
+ * abort() will be called on return to the caller.
+ * abort() raises SIGABRT. The default signal handler
+ * for all signals terminates the calling program with
+ * exit code 3.
+ * 2) A term_func() must not throw an exception. Therefore
+ * term_func() should not call pthread_exit() if an
+ * an exception-using version of pthreads-win32 library
+ * is being used (i.e. either pthreadVCE or pthreadVSE).
+
+
+2002-02-23 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * rwlock2_t.c: New test.
+ * rwlock3_t.c: New test.
+ * rwlock4_t.c: New test.
+ * rwlock5_t.c: New test.
+ * rwlock6_t.c: New test.
+ * rwlock6_t2.c: New test.
+ * rwlock6.c (main): Swap thread and result variables
+ to correspond to actual thread functions.
+ * rwlock1.c: Change test description comment to correspond
+ to the actual test.
+
+ * condvar1_2.c: Loop over the test many times in the hope
+ of detecting any intermittent deadlocks. This is to
+ test a fixed problem in pthread_cond_destroy.c.
+
+ * spin4.c: Remove unused variable.
+
+2002-02-17 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * condvar1_1.c: New test.
+ * condvar1_2.c: New test.
+
+2002-02-07 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * delay1.c: New test.
+ * delay2.c: New test.
+ * exit4.c: New test.
+
+2002-02-02 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * mutex8: New test.
+ * mutex8n: New test.
+ * mutex8e: New test.
+ * mutex8r: New test.
+ * cancel6a: New test.
+ * cancel6d: New test.
+ * cleanup0.c: Add pragmas for inline optimisation control.
+ * cleanup1.c: Add pragmas for inline optimisation control.
+ * cleanup2.c: Add pragmas for inline optimisation control.
+ * cleanup3.c: Add pragmas for inline optimisation control.
+ * condvar7.c: Add pragmas for inline optimisation control.
+ * condvar8.c: Add pragmas for inline optimisation control.
+ * condvar9.c: Add pragmas for inline optimisation control.
+
+2002-01-30 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * cleanup1.c (): Must be declared __cdecl when compiled
+ as C++ AND testing the standard C library version.
+
+2002-01-16 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * spin4.c (main): Fix renamed function call.
+
+2002-01-14 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * exception3.c (main): Shorten wait time.
+
+2002-01-09 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * mutex7.c: New test.
+ * mutex7n.c: New test.
+ * mutex7e.c: New test.
+ * mutex7r.c: New test.
+ * mutex6.c: Modified to avoid leaving the locked mutex
+ around on exit.
+
+2001-10-25 Ross Johnson <rpj@setup1.ise.canberra.edu.au>
+
+ * condvar2.c: Remove reference to cv->nWaitersUnblocked.
+ * condvar2_1.c: Likewise; lower NUMTHREADS from 60 to 30.
+ * condvar3_1.c: Likewise.
+ * condvar3_2.c: Likewise.
+ * count1.c: lower NUMTHREADS from 60 to 30.
+ * inherit1.c: Determine valid priority values and then
+ assert values returned by POSIX routines are the same.
+ * priority1.c: Likewise.
+ * priority2.c: Likewise.
+
+2001-07-12 Ross Johnson <rpj@setup1.ise.canberra.edu.au>
+
+ * barrier5.c: Assert that precisely one thread receives
+ PTHREAD_BARRIER_SERIAL_THREAD at each barrier.
+
+2001-07-09 Ross Johnson <rpj@setup1.ise.canberra.edu.au>
+
+ * barrier3.c: Fixed.
+ * barrier4.c: Fixed.
+ * barrier5.c: New; proves that all threads in the group
+ reaching the barrier wait and then resume together. Repeats the test
+ using groups of 1 to 16 threads. Each group of threads must negotiate
+ a large number of barriers (10000).
+ * spin4.c: Fixed.
+ * test.h (error_string): Modified the success (0) value.
+
+2001-07-07 Ross Johnson <rpj@setup1.ise.canberra.edu.au>
+
+ * spin3.c: Changed test and fixed.
+ * spin4.c: Fixed.
+ * barrier3.c: Fixed.
+ * barrier4.c: Fixed.
+
+2001-07-05 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * spin1.c: New; testing spinlocks.
+ * spin2.c: New; testing spinlocks.
+ * spin3.c: New; testing spinlocks.
+ * spin4.c: New; testing spinlocks.
+ * barrier1.c: New; testing barriers.
+ * barrier2.c: New; testing barriers.
+ * barrier3.c: New; testing barriers.
+ * barrier4.c: New; testing barriers.
+ * GNUmakefile: Add new tests.
+ * Makefile: Add new tests.
+
+2001-07-01 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * benchtest3.c: New; timing mutexes.
+ * benchtest4.c: New; time mutexes.
+ * condvar3_1.c: Fixed bug - Alexander Terekhov
+ * condvar3_3.c: New test.
+
+2001-06-25 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * priority1.c: New test.
+ * priority2.c: New test.
+ * inherit1.c: New test.
+ * benchtest1.c: New; timing mutexes.
+ * benchtest2.c: New; timing mutexes.
+ * mutex4.c: Modified to test all mutex types.
+
+2001-06-8 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * mutex5.c: Insert inert change to quell compiler warnings.
+ * condvar3_2.c: Remove unused variable.
+
+2001-06-3 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * condvar2_1.c: New test.
+ * condvar3_1.c: New test.
+ * condvar3_2.c: New test.
+
+2001-05-30 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * mutex1n.c: New test.
+ * mutex1e.c: New test.
+ * mutex1r.c: New test.
+ * mutex4.c: Now locks and unlocks a mutex.
+ * mutex5.c: New test.
+ * mutex6.c: New test.
+ * mutex6n.c: New test.
+ * mutex6e.c: New test.
+ * mutex6r.c: New test.
+ * Makefile: Added new tests; reorganised.
+ * GNUmakefile: Likewise.
+ * rwlock6.c: Fix to properly prove read-while-write locking
+ and single writer locking.
+
+2001-05-29 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * Makefile: Reorganisation.
+ * GNUmakefile: Likewise.
+ - Thomas Pfaff <tpfaff@gmx.net>
+
+ * exception1.c: Add stdio.h include to define fprintf and stderr
+ in non-exception C version of main().
+ * exception2.c: Likewise.
+ * exception3.c: Likewise.
+
+ * Makefile (rwlock7): Add new test.
+ * GNUmakefile (rwlock7): Add new test.
+ * rwlock7.c: New test.
+ * rwlock6.c: Changed to test that writer has priority.
+
+ * eyal1.c (main): Unlock each mutex_start lock before destroying
+ it.
+
+2000-12-29 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * GNUmakefile: Add mutex4 test; ensure libpthreadw32.a is
+ removed for "clean" target.
+ * Makefile: Add mutex4 test.
+
+ * exception3.c: Remove SEH code; automatically pass the test
+ under SEH (which is an N/A environment).
+
+ * mutex4.c: New test.
+
+ * eyal1.c (do_work_unit): Add a dummy "if" to force the
+ optimiser to retain code; reduce thread work loads.
+
+ * condvar8.c (main): Add an additional "assert" for debugging;
+ increase pthread_cond_signal timeout.
+
+2000-12-28 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * eyal1.c: Increase thread work loads.
+ * exception2.c: New test.
+ * exception3.c: New test.
+ * Makefile: Add new tests exception2.c and exception3.c.
+ * GNUmakefile: Likewise.
+
+2000-12-11 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * cleanup3.c: Remove unused variable.
+ * cleanup2.c: Likewise.
+ * exception1.c: Throw an exception rather than use
+ a deliberate zero divide so that catch(...) will
+ handle it under Mingw32. Mingw32 now builds the
+ library correctly to pass all tests - see Thomas
+ Pfaff's detailed instructions re needed changes
+ to Mingw32 in the Pthreads-Win32 FAQ.
+
+2000-09-08 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * cancel5.c: New; tests calling pthread_cancel()
+ from the main thread without first creating a
+ POSIX thread struct for the non-POSIX main thread
+ - this forces pthread_cancel() to create one via
+ pthread_self().
+ * Makefile (cancel5): Add new test.
+ * GNUmakefile (cancel5): Likewise.
+
+2000-08-17 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * create2.c: New; Test that pthread_t contains
+ the W32 HANDLE before it calls the thread routine
+ proper.
+
+2000-08-13 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * condvar3.c: Minor change to eliminate compiler
+ warning.
+
+ * condvar4.c: ditto.
+
+ * condvar5.c: ditto.
+
+ * condvar6.c: ditto.
+
+ * condvar7.c: ditto.
+
+ * condvar8.c: ditto.
+
+ * condvar9.c: ditto.
+
+ * exit1.c: Function needed return statement.
+
+ * cleanup1.c: Remove unnecessary printf arg.
+
+ * cleanup2.c: Fix cast.
+
+ * rwlock6.c: Fix casts.
+
+ * exception1.c (PtW32CatchAll): Had the wrong name;
+ fix casts.
+
+ * cancel3.c: Remove unused waitLock variable.
+
+ * GNUmakefile: Change library/dll naming; add new tests;
+ general minor changes.
+
+ * Makefile: Change library/dll naming; add targets for
+ testing each of the two VC++ EH scheme versions;
+ default target now issues help message; compile warnings
+ now interpreted as errors to stop the make; add new
+ tests; restructure to remove prerequisites needed
+ otherwise.
+
+ * README: Updated.
+
+
+2000-08-10 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * eyal1.c (main): Change implicit cast to explicit
+ cast when passing "print_server" function pointer;
+ G++ no longer allows implicit func parameter casts.
+
+ * cleanup1.c: Remove unused "waitLock".
+ (main): Fix implicit parameter cast.
+
+ * cancel2.c (main): Fix implicit parameter cast.
+
+ * cancel4.c (main): Fix implicit parameter cast.
+
+ * cancel3.c (main): Fix implicit parameter cast.
+
+ * GNUmakefile: Renamed from Makefile; Add missing
+ cancel1 and cancel2 test targets.
+
+ * Makefile: Converted for use with MS nmake.
+
+2000-08-06 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * ccl.bat: Add /nologo to remove extraneous output.
+
+ * exception1.c (exceptionedThread): Init 'dummy';
+ put expression into if condition to prevent optimising away;
+ remove unused variable.
+
+ * cancel4.c (mythread): Cast return value to avoid warnings.
+
+ * cancel2.c (mythread): Missing #endif.
+
+ * condvar9.c (mythread): Cast return value to avoid warnings.
+
+ * condvar8.c (mythread): Cast return value to avoid warnings.
+
+ * condvar7.c (mythread): Cast return value to avoid warnings.
+
+ * cleanup3.c (mythread): Cast return value to avoid warnings.
+
+ * cleanup2.c (mythread): Cast return value to avoid warnings.
+
+ * cleanup1.c (mythread): Cast return value to avoid warnings.
+
+ * condvar5.c (mythread): Cast return value to avoid warnings.
+
+ * condvar3.c (mythread): Cast return value to avoid warnings.
+
+ * condvar6.c (mythread): Cast return value to avoid warnings.
+
+ * condvar4.c (mythread): Cast return value to avoid warnings.
+
+2000-08-05 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * cancel2.c: Use PtW32CatchAll macro if defined.
+
+ * exception1.c: Use PtW32CatchAll macro if defined.
+
+2000-08-02 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * tsd1.c: Fix typecasts of &result [g++ is now very fussy].
+
+ * test.h (assert): Return 0's explicitly to allay
+ g++ errors.
+
+ * join2.c: Add explicit typecasts.
+
+ * join1.c: Add explicit typecasts.
+
+ * join0.c: Add explicit typecasts.
+
+ * eyal1.c: Add explicit typecasts.
+
+ * count1.c (main): Add type cast to remove g++ parse warning
+ [gcc-2.95.2 seems to have tightened up on this].
+
+ * Makefile (GLANG): Use c++ explicitly.
+ Remove MSVC sections (was commented out).
+ Add target to generate cpp output.
+
+2000-07-25 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * runtest.bat: modified to work under W98.
+
+ * runall.bat: Add new tests; modified to work under W98.
+ It was ok under NT.
+
+ * Makefile: Add new tests.
+
+ * exception1.c: New; Test passing exceptions back to the
+ application and retaining library internal exceptions.
+
+ * join0.c: New; Test a single join.
+
+2000-01-06 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * cleanup1.c: New; Test cleanup handler executes (when thread is
+ canceled).
+
+ * cleanup2.c: New; Test cleanup handler executes (when thread is
+ not canceled).
+
+ * cleanup3.c: New; Test cleanup handler does not execute
+ (when thread is not canceled).
+
+2000-01-04 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * cancel4.c: New; Test cancelation does not occur in deferred
+ cancelation threads with no cancelation points.
+
+ * cancel3.c: New; Test asynchronous cancelation.
+
+ * context1.c: New; Test context switching method for async
+ cancelation.
+
+1999-11-23 Ross Johnson <rpj@special.ise.canberra.edu.au>
+
+ * test.h: Add header includes; include local header versions rather
+ than system versions; rearrange the assert macro defines.
+
+1999-11-07 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * loadfree.c: New. Test loading and freeing the library (DLL).
+
+1999-10-30 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * cancel1.c: New. Test pthread_setcancelstate and
+ pthread_setcanceltype functions.
+ * eyal1.c (waste_time): Change calculation to avoid FP exception
+ on Aplhas
+ - Rich Peters <rpeters@micro-magic.com>
+
+Oct 14 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * condvar7.c: New. Test broadcast after waiting thread is canceled.
+ * condvar8.c: New. Test multiple broadcasts.
+ * condvar9.c: New. Test multiple broadcasts with thread
+ cancelation.
+
+Sep 16 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * rwlock6.c: New test.
+
+Sep 15 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * rwlock1.c: New test.
+ * rwlock2.c: New test.
+ * rwlock3.c: New test.
+ * rwlock4.c: New test.
+ * rwlock5.c: New test.
+
+Aug 22 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * runall.bat (join2): Add test.
+
+Aug 19 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * join2.c: New test.
+
+Wed Aug 12 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * Makefile (LIBS): Add -L.
+
+Mon May 31 10:25:01 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * Makefile (GLANG): Add GCC language option.
+
+Sat May 29 23:29:04 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * runall.bat (condvar5): Add new test.
+
+ * runall.bat (condvar6): Add new test.
+
+ * Makefile (condvar5) : Add new test.
+
+ * Makefile (condvar6) : Add new test.
+
+ * condvar5.c: New test for pthread_cond_broadcast().
+
+ * condvar6.c: New test for pthread_cond_broadcast().
+
+Sun Apr 4 12:04:28 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * tsd1.c (mythread): Change Sleep(0) to sched_yield().
+ (sched.h): Include.
+
+ * condvar3.c (mythread): Remove redundant Sleep().
+
+ * runtest.bat: Re-organised to make more informative.
+
+Fri Mar 19 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * *.bat: redirect unwanted output to nul:
+
+ * runall.bat: new.
+
+ * cancel1.c: new. Not part of suite yet.
+
+Mon Mar 15 00:17:55 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * mutex1.c: only test mutex init and destroy; add assertions.
+
+ * count1.c: raise number of spawned threads to 60 (appears to
+ be the limit under Win98).
+
+Sun Mar 14 21:31:02 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * test.h (assert): add assertion trace option.
+ Use:
+ "#define ASSERT_TRACE 1" to turn it on,
+ "#define ASSERT_TRACE 0" to turn it off (default).
+
+ * condvar3.c (main): add more assertions.
+
+ * condvar4.c (main): add more assertions.
+
+ * condvar1.c (main): add more assertions.
+
+Fri Mar 12 08:34:15 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * condvar4.c (cvthing): switch the order of the INITIALIZERs.
+
+ * eyal1.c (main): Fix trylock loop; was not waiting for thread to lock
+ the "started" mutex.
+
+Wed Mar 10 10:41:52 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * tryentercs.c: Apply typo patch from bje.
+
+ * tryentercs2.c: Ditto.
+
+Sun Mar 7 10:41:52 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * Makefile (condvar3, condvar4): Add tests.
+
+ * condvar4.c (General): Reduce to simple test case; prerequisite
+ is condvar3.c; add description.
+
+ * condvar3.c (General): Reduce to simple test case; prerequisite
+ is condvar2.c; add description.
+
+ * condvar2.c (General): Reduce to simple test case; prerequisite
+ is condvar1.c; add description.
+
+ * condvar1.c (General): Reduce to simple test case; add
+ description.
+
+ * Template.c (Comments): Add generic test detail.
+
+1999-02-23 Ross Johnson <rpj@ise.canberra.edu.au>
+
+ * Template.c: Revamp.
+
+ * condvar1.c: Add.
+
+ * condvar2.c: Add.
+
+ * Makefile: Add condvar1 condvar2 tests.
+
+ * exit1.c, exit2.c, exit3.c: Cosmetic changes.
+
+1999-02-23 Ross Johnson <rpj@ise.canberra.edu.au>
+
+ * Makefile: Some refinement.
+
+ * *.c: More exhaustive checking through assertions; clean up;
+ add some more tests.
+
+ * Makefile: Now actually runs the tests.
+
+ * tests.h: Define our own assert macro. The Mingw32
+ version pops up a dialog but we want to run non-interactively.
+
+ * equal1.c: use assert a little more directly so that it
+ prints the actual call statement.
+
+ * exit1.c: Modify to return 0 on success, 1 on failure.
+
+1999-02-22 Ross Johnson <rpj@ise.canberra.edu.au>
+
+ * self2.c: Bring up to date.
+
+ * self3.c: Ditto.
+
+1999-02-21 Ben Elliston <bje@cygnus.com>
+
+ * README: Update.
+
+ * Makefile: New file. Run all tests automatically. Primitive tests
+ are run first; more complex tests are run last.
+
+ * count1.c: New test. Validate the thread count.
+
+ * exit2.c: Perform a simpler test.
+
+ * exit3.c: New test. Replaces exit2.c, since exit2.c needs to
+ perform simpler checking first.
+
+ * create1.c: Update to use the new testsuite exiting convention.
+
+ * equal1.c: Likewise.
+
+ * mutex1.c: Likewise.
+
+ * mutex2.c: Likewise.
+
+ * once1.c: Likewise.
+
+ * self2.c: Likewise.
+
+ * self3.c: Likewise.
+
+ * tsd1.c: Likewise.
+
+1999-02-20 Ross Johnson <rpj@ise.canberra.edu.au>
+
+ * mutex2.c: Test static mutex initialisation.
+
+ * test.h: New. Declares a table mapping error numbers to
+ error names.
+
+1999-01-17 Ross Johnson <rpj@ise.canberra.edu.au>
+
+ * runtest: New script to build and run a test in the tests directory.
+
+Wed Dec 30 11:22:44 1998 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * tsd1.c: Re-written. See comments at start of file.
+ * Template.c: New. Contains skeleton code and comment template
+ intended to fully document the test.
+
+Fri Oct 16 17:59:49 1998 Ross Johnson <rpj@swan.canberra.edu.au>
+
+ * tsd1.c (destroy_key): Add function. Change diagnostics.
+
+Thu Oct 15 17:42:37 1998 Ross Johnson <rpj@swan.canberra.edu.au>
+
+ * tsd1.c (mythread): Fix some casts and add some message
+ output. Fix inverted conditional.
+
+Mon Oct 12 02:12:29 1998 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
+
+ * tsd1.c: New. Test TSD using 1 key and 2 threads.
+
+1998-09-13 Ben Elliston <bje@cygnus.com>
+
+ * eyal1.c: New file; contributed by Eyal Lebedinsky
+ <eyal@eyal.emu.id.au>.
+
+1998-09-12 Ben Elliston <bje@cygnus.com>
+
+ * exit2.c (func): Return a value.
+ (main): Call the right thread entry function.
+
+1998-07-22 Ben Elliston <bje@cygnus.com>
+
+ * exit2.c (main): Fix size of pthread_t array.
+
+1998-07-10 Ben Elliston <bje@cygnus.com>
- * rwlock7.c (main): Tidy up statistics reporting; randomise
- update accesses.
- * rwlock8.c: New test.
+ * exit2.c: New file; test pthread_exit() harder.
-2004-09-08 Alexandre Girao <alexgirao@gmail.com>
-
- * cancel7.c (main): Win98 wants a valid (non-NULL) location
- for the last arg of _beginthreadex().
- * cancel8.c (main): Likewise.
- * exit4.c (main): Likewise.
- * exit5.c (main): Likewise.
-
-2004-08-26 Ross Johnson <rpj@callisto.canberra.edu.au>
-
- * create3.c: New test.
-
-2004-06-21 Ross Johnson <rpj@callisto.canberra.edu.au>
-
- * mutex2r.c: New test.
- * mutex2e.c: New test.
- * mutex3r.c: New test.
- * mutex3e.c: New test.
- * mutex6s.c: New test.
- * mutex6rs.c: New test.
- * mutex6es.c: New test.
-
-2004-05-21 Ross Johnson <rpj@callisto.canberra.edu.au>
-
- * join3.c: New test.
-
-2004-05-16 Ross Johnson <rpj@callisto.canberra.edu.au>
-
- * condvar2.c (WIN32_WINNT): Define to avoid redefinition warning
- from inclusion of implement.h.
- * convar2_1.c: Likewise.
- * condvar3_1.c: Likewise.
- * condvar3_2.c: Likewise.
- * context1.c: Likewise.
- * sizes.c: Likewise.
- * Makefile: Don't define _WIN32_WINNT on compiler command line.
- * GNUmakefile: Likewise.
- * priority1.c (main): Add column to output for actual win32
- priority.
-
-2004-05-16 Ross Johnson <rpj@callisto.canberra.edu.au>
-
- * cancel9.c: New test.
- * cancel3.c: Remove inappropriate conditional compilation;
- GNU C version of test suite no longer quietly skips this test.
- * cancel5.c: Likewise.
- * GNUmakefile: Can now build individual test app using default
- C version of library using 'make clean testname.c'.
- * Makefile: Likewise for VC using 'nmake clean test testname.c'.
-
-2003-10-14 Ross Johnson <rpj@callisto.canberra.edu.au>
-
- * Wmakefile: New makefile for Watcom testing.
-
-2003-09-18 Ross Johnson <rpj@callisto.canberra.edu.au>
-
- * benchtest.h: Move old mutex code into benchlib.c.
- * benchlib.c: New statically linked module to ensure that
- bench apps don't inline the code and therefore have an unfair
- advantage over the pthreads lib routines. Made little or no
- difference.
- * benchtest1.c: Minor change to avoid compiler warnings.
- * benchtest5.c: Likewise.
- * benchtest2.c: Fix misinformation in output report.
- * README.BENCH: Add comments on results.
-
-2003-09-14 Ross Johnson <rpj@callisto.canberra.edu.au>
-
- * priority1.c: Reworked to comply with modified priority
- management and provide additional output.
- * priority2.c: Likewise.
- * inherit1.c: Likewise.
-
-2003-09-03 Ross Johnson <rpj@callisto.canberra.edu.au>
-
- * exit4.c: New test.
- * exit5.c: New test.
- * cancel7.c: New test.
- * cancel8.c: New test.
-
-2003-08-13 Ross Johnson <rpj@ise.canberra.edu.au>
-
- * reuse1.c: New test.
- * reuse1.c: New test.
- * valid1.c: New test.
- * valid2.c: New test.
- * kill1.c: New test.
- * create2.c: Now included in test regime.
-
-2003-07-19 Ross Johnson <rpj@ise.canberra.edu.au>
-
- * eyal1.c (waste_time): Make threads do more work to ensure that
- all threads get to do some work.
- * semaphore1.c: Make it clear that certain errors are expected.
- * exception2.c (non_MSVC code sections): Change to include
- C++ standard include file, i.e. change <new.h> to <exception>.
- * exception3.c (non_MSVC code sections): Likewise; qualify std::
- namespace entities where necessary.
- * GNUmakefile: modified to work in the MsysDTK (newer MinGW)
- environment; define CC as gcc or g++ as appropriate because
- using gcc -x c++ doesn't link with required c++ libs by default,
- but g++ does.
-
-2002-12-11 Ross Johnson <ross@special.ise.canberra.edu.au>
-
- * mutex7e.c: Assert EBUSY return instead of EDEADLK.
-
-2002-06-03 Ross Johnson <rpj@digit.ise.canberra.edu.au>
-
- * semaphore2.c: New test.
-
-2002-03-02 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * Makefile (CFLAGS): Changed /MT to /MD to link with
- the correct library MSVCRT.LIB. Otherwise errno doesn't
- work.
-
-2002-02-28 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * exception3.c: Correct recent change.
-
- * semaphore1.c: New test.
-
- * Makefile: Add rule to generate pre-processor output.
-
-2002-02-28 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * exception3.c (terminateFunction): For MSVC++, call
- exit() rather than pthread_exit(). Add comments to explain
- why.
- * Notes from the MSVC++ manual:
- * 1) A term_func() should call exit(), otherwise
- * abort() will be called on return to the caller.
- * abort() raises SIGABRT. The default signal handler
- * for all signals terminates the calling program with
- * exit code 3.
- * 2) A term_func() must not throw an exception. Therefore
- * term_func() should not call pthread_exit() if an
- * an exception-using version of pthreads-win32 library
- * is being used (i.e. either pthreadVCE or pthreadVSE).
-
-
-2002-02-23 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * rwlock2_t.c: New test.
- * rwlock3_t.c: New test.
- * rwlock4_t.c: New test.
- * rwlock5_t.c: New test.
- * rwlock6_t.c: New test.
- * rwlock6_t2.c: New test.
- * rwlock6.c (main): Swap thread and result variables
- to correspond to actual thread functions.
- * rwlock1.c: Change test description comment to correspond
- to the actual test.
-
- * condvar1_2.c: Loop over the test many times in the hope
- of detecting any intermittent deadlocks. This is to
- test a fixed problem in pthread_cond_destroy.c.
-
- * spin4.c: Remove unused variable.
-
-2002-02-17 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * condvar1_1.c: New test.
- * condvar1_2.c: New test.
-
-2002-02-07 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * delay1.c: New test.
- * delay2.c: New test.
- * exit4.c: New test.
-
-2002-02-02 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * mutex8: New test.
- * mutex8n: New test.
- * mutex8e: New test.
- * mutex8r: New test.
- * cancel6a: New test.
- * cancel6d: New test.
- * cleanup0.c: Add pragmas for inline optimisation control.
- * cleanup1.c: Add pragmas for inline optimisation control.
- * cleanup2.c: Add pragmas for inline optimisation control.
- * cleanup3.c: Add pragmas for inline optimisation control.
- * condvar7.c: Add pragmas for inline optimisation control.
- * condvar8.c: Add pragmas for inline optimisation control.
- * condvar9.c: Add pragmas for inline optimisation control.
-
-2002-01-30 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * cleanup1.c (): Must be declared __cdecl when compiled
- as C++ AND testing the standard C library version.
-
-2002-01-16 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * spin4.c (main): Fix renamed function call.
-
-2002-01-14 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * exception3.c (main): Shorten wait time.
-
-2002-01-09 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * mutex7.c: New test.
- * mutex7n.c: New test.
- * mutex7e.c: New test.
- * mutex7r.c: New test.
- * mutex6.c: Modified to avoid leaving the locked mutex
- around on exit.
-
-2001-10-25 Ross Johnson <rpj@setup1.ise.canberra.edu.au>
-
- * condvar2.c: Remove reference to cv->nWaitersUnblocked.
- * condvar2_1.c: Likewise; lower NUMTHREADS from 60 to 30.
- * condvar3_1.c: Likewise.
- * condvar3_2.c: Likewise.
- * count1.c: lower NUMTHREADS from 60 to 30.
- * inherit1.c: Determine valid priority values and then
- assert values returned by POSIX routines are the same.
- * priority1.c: Likewise.
- * priority2.c: Likewise.
-
-2001-07-12 Ross Johnson <rpj@setup1.ise.canberra.edu.au>
-
- * barrier5.c: Assert that precisely one thread receives
- PTHREAD_BARRIER_SERIAL_THREAD at each barrier.
-
-2001-07-09 Ross Johnson <rpj@setup1.ise.canberra.edu.au>
-
- * barrier3.c: Fixed.
- * barrier4.c: Fixed.
- * barrier5.c: New; proves that all threads in the group
- reaching the barrier wait and then resume together. Repeats the test
- using groups of 1 to 16 threads. Each group of threads must negotiate
- a large number of barriers (10000).
- * spin4.c: Fixed.
- * test.h (error_string): Modified the success (0) value.
-
-2001-07-07 Ross Johnson <rpj@setup1.ise.canberra.edu.au>
-
- * spin3.c: Changed test and fixed.
- * spin4.c: Fixed.
- * barrier3.c: Fixed.
- * barrier4.c: Fixed.
-
-2001-07-05 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * spin1.c: New; testing spinlocks.
- * spin2.c: New; testing spinlocks.
- * spin3.c: New; testing spinlocks.
- * spin4.c: New; testing spinlocks.
- * barrier1.c: New; testing barriers.
- * barrier2.c: New; testing barriers.
- * barrier3.c: New; testing barriers.
- * barrier4.c: New; testing barriers.
- * GNUmakefile: Add new tests.
- * Makefile: Add new tests.
-
-2001-07-01 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * benchtest3.c: New; timing mutexes.
- * benchtest4.c: New; time mutexes.
- * condvar3_1.c: Fixed bug - Alexander Terekhov
- * condvar3_3.c: New test.
-
-2001-06-25 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * priority1.c: New test.
- * priority2.c: New test.
- * inherit1.c: New test.
- * benchtest1.c: New; timing mutexes.
- * benchtest2.c: New; timing mutexes.
- * mutex4.c: Modified to test all mutex types.
-
-2001-06-8 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * mutex5.c: Insert inert change to quell compiler warnings.
- * condvar3_2.c: Remove unused variable.
-
-2001-06-3 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * condvar2_1.c: New test.
- * condvar3_1.c: New test.
- * condvar3_2.c: New test.
-
-2001-05-30 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * mutex1n.c: New test.
- * mutex1e.c: New test.
- * mutex1r.c: New test.
- * mutex4.c: Now locks and unlocks a mutex.
- * mutex5.c: New test.
- * mutex6.c: New test.
- * mutex6n.c: New test.
- * mutex6e.c: New test.
- * mutex6r.c: New test.
- * Makefile: Added new tests; reorganised.
- * GNUmakefile: Likewise.
- * rwlock6.c: Fix to properly prove read-while-write locking
- and single writer locking.
-
-2001-05-29 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * Makefile: Reorganisation.
- * GNUmakefile: Likewise.
- - Thomas Pfaff <tpfaff@gmx.net>
-
- * exception1.c: Add stdio.h include to define fprintf and stderr
- in non-exception C version of main().
- * exception2.c: Likewise.
- * exception3.c: Likewise.
-
- * Makefile (rwlock7): Add new test.
- * GNUmakefile (rwlock7): Add new test.
- * rwlock7.c: New test.
- * rwlock6.c: Changed to test that writer has priority.
-
- * eyal1.c (main): Unlock each mutex_start lock before destroying
- it.
-
-2000-12-29 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * GNUmakefile: Add mutex4 test; ensure libpthreadw32.a is
- removed for "clean" target.
- * Makefile: Add mutex4 test.
-
- * exception3.c: Remove SEH code; automatically pass the test
- under SEH (which is an N/A environment).
-
- * mutex4.c: New test.
-
- * eyal1.c (do_work_unit): Add a dummy "if" to force the
- optimiser to retain code; reduce thread work loads.
-
- * condvar8.c (main): Add an additional "assert" for debugging;
- increase pthread_cond_signal timeout.
-
-2000-12-28 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * eyal1.c: Increase thread work loads.
- * exception2.c: New test.
- * exception3.c: New test.
- * Makefile: Add new tests exception2.c and exception3.c.
- * GNUmakefile: Likewise.
-
-2000-12-11 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * cleanup3.c: Remove unused variable.
- * cleanup2.c: Likewise.
- * exception1.c: Throw an exception rather than use
- a deliberate zero divide so that catch(...) will
- handle it under Mingw32. Mingw32 now builds the
- library correctly to pass all tests - see Thomas
- Pfaff's detailed instructions re needed changes
- to Mingw32 in the Pthreads-Win32 FAQ.
-
-2000-09-08 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * cancel5.c: New; tests calling pthread_cancel()
- from the main thread without first creating a
- POSIX thread struct for the non-POSIX main thread
- - this forces pthread_cancel() to create one via
- pthread_self().
- * Makefile (cancel5): Add new test.
- * GNUmakefile (cancel5): Likewise.
-
-2000-08-17 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * create2.c: New; Test that pthread_t contains
- the W32 HANDLE before it calls the thread routine
- proper.
-
-2000-08-13 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * condvar3.c: Minor change to eliminate compiler
- warning.
-
- * condvar4.c: ditto.
-
- * condvar5.c: ditto.
-
- * condvar6.c: ditto.
-
- * condvar7.c: ditto.
-
- * condvar8.c: ditto.
-
- * condvar9.c: ditto.
-
- * exit1.c: Function needed return statement.
-
- * cleanup1.c: Remove unnecessary printf arg.
-
- * cleanup2.c: Fix cast.
-
- * rwlock6.c: Fix casts.
-
- * exception1.c (PtW32CatchAll): Had the wrong name;
- fix casts.
-
- * cancel3.c: Remove unused waitLock variable.
-
- * GNUmakefile: Change library/dll naming; add new tests;
- general minor changes.
-
- * Makefile: Change library/dll naming; add targets for
- testing each of the two VC++ EH scheme versions;
- default target now issues help message; compile warnings
- now interpreted as errors to stop the make; add new
- tests; restructure to remove prerequisites needed
- otherwise.
-
- * README: Updated.
-
-
-2000-08-10 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * eyal1.c (main): Change implicit cast to explicit
- cast when passing "print_server" function pointer;
- G++ no longer allows implicit func parameter casts.
-
- * cleanup1.c: Remove unused "waitLock".
- (main): Fix implicit parameter cast.
-
- * cancel2.c (main): Fix implicit parameter cast.
-
- * cancel4.c (main): Fix implicit parameter cast.
-
- * cancel3.c (main): Fix implicit parameter cast.
-
- * GNUmakefile: Renamed from Makefile; Add missing
- cancel1 and cancel2 test targets.
-
- * Makefile: Converted for use with MS nmake.
-
-2000-08-06 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * ccl.bat: Add /nologo to remove extraneous output.
-
- * exception1.c (exceptionedThread): Init 'dummy';
- put expression into if condition to prevent optimising away;
- remove unused variable.
-
- * cancel4.c (mythread): Cast return value to avoid warnings.
-
- * cancel2.c (mythread): Missing #endif.
-
- * condvar9.c (mythread): Cast return value to avoid warnings.
-
- * condvar8.c (mythread): Cast return value to avoid warnings.
-
- * condvar7.c (mythread): Cast return value to avoid warnings.
-
- * cleanup3.c (mythread): Cast return value to avoid warnings.
-
- * cleanup2.c (mythread): Cast return value to avoid warnings.
-
- * cleanup1.c (mythread): Cast return value to avoid warnings.
-
- * condvar5.c (mythread): Cast return value to avoid warnings.
-
- * condvar3.c (mythread): Cast return value to avoid warnings.
-
- * condvar6.c (mythread): Cast return value to avoid warnings.
-
- * condvar4.c (mythread): Cast return value to avoid warnings.
-
-2000-08-05 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * cancel2.c: Use PtW32CatchAll macro if defined.
-
- * exception1.c: Use PtW32CatchAll macro if defined.
-
-2000-08-02 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * tsd1.c: Fix typecasts of &result [g++ is now very fussy].
-
- * test.h (assert): Return 0's explicitly to allay
- g++ errors.
-
- * join2.c: Add explicit typecasts.
-
- * join1.c: Add explicit typecasts.
-
- * join0.c: Add explicit typecasts.
-
- * eyal1.c: Add explicit typecasts.
-
- * count1.c (main): Add type cast to remove g++ parse warning
- [gcc-2.95.2 seems to have tightened up on this].
-
- * Makefile (GLANG): Use c++ explicitly.
- Remove MSVC sections (was commented out).
- Add target to generate cpp output.
-
-2000-07-25 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * runtest.bat: modified to work under W98.
-
- * runall.bat: Add new tests; modified to work under W98.
- It was ok under NT.
-
- * Makefile: Add new tests.
-
- * exception1.c: New; Test passing exceptions back to the
- application and retaining library internal exceptions.
-
- * join0.c: New; Test a single join.
-
-2000-01-06 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * cleanup1.c: New; Test cleanup handler executes (when thread is
- canceled).
-
- * cleanup2.c: New; Test cleanup handler executes (when thread is
- not canceled).
-
- * cleanup3.c: New; Test cleanup handler does not execute
- (when thread is not canceled).
-
-2000-01-04 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * cancel4.c: New; Test cancelation does not occur in deferred
- cancelation threads with no cancelation points.
-
- * cancel3.c: New; Test asynchronous cancelation.
-
- * context1.c: New; Test context switching method for async
- cancelation.
-
-1999-11-23 Ross Johnson <rpj@special.ise.canberra.edu.au>
-
- * test.h: Add header includes; include local header versions rather
- than system versions; rearrange the assert macro defines.
-
-1999-11-07 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * loadfree.c: New. Test loading and freeing the library (DLL).
-
-1999-10-30 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * cancel1.c: New. Test pthread_setcancelstate and
- pthread_setcanceltype functions.
- * eyal1.c (waste_time): Change calculation to avoid FP exception
- on Aplhas
- - Rich Peters <rpeters@micro-magic.com>
-
-Oct 14 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * condvar7.c: New. Test broadcast after waiting thread is canceled.
- * condvar8.c: New. Test multiple broadcasts.
- * condvar9.c: New. Test multiple broadcasts with thread
- cancelation.
-
-Sep 16 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * rwlock6.c: New test.
-
-Sep 15 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * rwlock1.c: New test.
- * rwlock2.c: New test.
- * rwlock3.c: New test.
- * rwlock4.c: New test.
- * rwlock5.c: New test.
-
-Aug 22 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * runall.bat (join2): Add test.
-
-Aug 19 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * join2.c: New test.
-
-Wed Aug 12 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * Makefile (LIBS): Add -L.
-
-Mon May 31 10:25:01 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * Makefile (GLANG): Add GCC language option.
-
-Sat May 29 23:29:04 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * runall.bat (condvar5): Add new test.
-
- * runall.bat (condvar6): Add new test.
-
- * Makefile (condvar5) : Add new test.
-
- * Makefile (condvar6) : Add new test.
-
- * condvar5.c: New test for pthread_cond_broadcast().
-
- * condvar6.c: New test for pthread_cond_broadcast().
-
-Sun Apr 4 12:04:28 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * tsd1.c (mythread): Change Sleep(0) to sched_yield().
- (sched.h): Include.
-
- * condvar3.c (mythread): Remove redundant Sleep().
-
- * runtest.bat: Re-organised to make more informative.
-
-Fri Mar 19 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * *.bat: redirect unwanted output to nul:
-
- * runall.bat: new.
-
- * cancel1.c: new. Not part of suite yet.
-
-Mon Mar 15 00:17:55 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * mutex1.c: only test mutex init and destroy; add assertions.
-
- * count1.c: raise number of spawned threads to 60 (appears to
- be the limit under Win98).
-
-Sun Mar 14 21:31:02 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * test.h (assert): add assertion trace option.
- Use:
- "#define ASSERT_TRACE 1" to turn it on,
- "#define ASSERT_TRACE 0" to turn it off (default).
-
- * condvar3.c (main): add more assertions.
-
- * condvar4.c (main): add more assertions.
-
- * condvar1.c (main): add more assertions.
-
-Fri Mar 12 08:34:15 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * condvar4.c (cvthing): switch the order of the INITIALIZERs.
-
- * eyal1.c (main): Fix trylock loop; was not waiting for thread to lock
- the "started" mutex.
-
-Wed Mar 10 10:41:52 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * tryentercs.c: Apply typo patch from bje.
-
- * tryentercs2.c: Ditto.
-
-Sun Mar 7 10:41:52 1999 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * Makefile (condvar3, condvar4): Add tests.
-
- * condvar4.c (General): Reduce to simple test case; prerequisite
- is condvar3.c; add description.
-
- * condvar3.c (General): Reduce to simple test case; prerequisite
- is condvar2.c; add description.
-
- * condvar2.c (General): Reduce to simple test case; prerequisite
- is condvar1.c; add description.
-
- * condvar1.c (General): Reduce to simple test case; add
- description.
-
- * Template.c (Comments): Add generic test detail.
-
-1999-02-23 Ross Johnson <rpj@ise.canberra.edu.au>
-
- * Template.c: Revamp.
-
- * condvar1.c: Add.
-
- * condvar2.c: Add.
-
- * Makefile: Add condvar1 condvar2 tests.
-
- * exit1.c, exit2.c, exit3.c: Cosmetic changes.
-
-1999-02-23 Ross Johnson <rpj@ise.canberra.edu.au>
-
- * Makefile: Some refinement.
-
- * *.c: More exhaustive checking through assertions; clean up;
- add some more tests.
-
- * Makefile: Now actually runs the tests.
-
- * tests.h: Define our own assert macro. The Mingw32
- version pops up a dialog but we want to run non-interactively.
-
- * equal1.c: use assert a little more directly so that it
- prints the actual call statement.
-
- * exit1.c: Modify to return 0 on success, 1 on failure.
-
-1999-02-22 Ross Johnson <rpj@ise.canberra.edu.au>
-
- * self2.c: Bring up to date.
-
- * self3.c: Ditto.
-
-1999-02-21 Ben Elliston <bje@cygnus.com>
-
- * README: Update.
-
- * Makefile: New file. Run all tests automatically. Primitive tests
- are run first; more complex tests are run last.
-
- * count1.c: New test. Validate the thread count.
-
- * exit2.c: Perform a simpler test.
-
- * exit3.c: New test. Replaces exit2.c, since exit2.c needs to
- perform simpler checking first.
-
- * create1.c: Update to use the new testsuite exiting convention.
-
- * equal1.c: Likewise.
-
- * mutex1.c: Likewise.
-
- * mutex2.c: Likewise.
-
- * once1.c: Likewise.
-
- * self2.c: Likewise.
-
- * self3.c: Likewise.
-
- * tsd1.c: Likewise.
-
-1999-02-20 Ross Johnson <rpj@ise.canberra.edu.au>
-
- * mutex2.c: Test static mutex initialisation.
-
- * test.h: New. Declares a table mapping error numbers to
- error names.
-
-1999-01-17 Ross Johnson <rpj@ise.canberra.edu.au>
-
- * runtest: New script to build and run a test in the tests directory.
-
-Wed Dec 30 11:22:44 1998 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * tsd1.c: Re-written. See comments at start of file.
- * Template.c: New. Contains skeleton code and comment template
- intended to fully document the test.
-
-Fri Oct 16 17:59:49 1998 Ross Johnson <rpj@swan.canberra.edu.au>
-
- * tsd1.c (destroy_key): Add function. Change diagnostics.
-
-Thu Oct 15 17:42:37 1998 Ross Johnson <rpj@swan.canberra.edu.au>
-
- * tsd1.c (mythread): Fix some casts and add some message
- output. Fix inverted conditional.
-
-Mon Oct 12 02:12:29 1998 Ross Johnson <rpj@ixobrychus.canberra.edu.au>
-
- * tsd1.c: New. Test TSD using 1 key and 2 threads.
-
-1998-09-13 Ben Elliston <bje@cygnus.com>
-
- * eyal1.c: New file; contributed by Eyal Lebedinsky
- <eyal@eyal.emu.id.au>.
-
-1998-09-12 Ben Elliston <bje@cygnus.com>
-
- * exit2.c (func): Return a value.
- (main): Call the right thread entry function.
-
-1998-07-22 Ben Elliston <bje@cygnus.com>
-
- * exit2.c (main): Fix size of pthread_t array.
-
-1998-07-10 Ben Elliston <bje@cygnus.com>
-
- * exit2.c: New file; test pthread_exit() harder.
-
- * exit1.c: New file; test pthread_exit().
+ * exit1.c: New file; test pthread_exit().
diff --git a/tests/Debug.dsp b/tests/Debug.dsp
new file mode 100644
index 0000000..0ba51bb
--- /dev/null
+++ b/tests/Debug.dsp
@@ -0,0 +1,93 @@
+# Microsoft Developer Studio Project File - Name="Debug" - Package Owner=<4>
+# Microsoft Developer Studio Generated Build File, Format Version 6.00
+# ** DO NOT EDIT **
+
+# TARGTYPE "Win32 (x86) Console Application" 0x0103
+
+CFG=Debug - Win32 Debug
+!MESSAGE This is not a valid makefile. To build this project using NMAKE,
+!MESSAGE use the Export Makefile command and run
+!MESSAGE
+!MESSAGE NMAKE /f "Debug.mak".
+!MESSAGE
+!MESSAGE You can specify a configuration when running NMAKE
+!MESSAGE by defining the macro CFG on the command line. For example:
+!MESSAGE
+!MESSAGE NMAKE /f "Debug.mak" CFG="Debug - Win32 Debug"
+!MESSAGE
+!MESSAGE Possible choices for configuration are:
+!MESSAGE
+!MESSAGE "Debug - Win32 Release" (based on "Win32 (x86) Console Application")
+!MESSAGE "Debug - Win32 Debug" (based on "Win32 (x86) Console Application")
+!MESSAGE
+
+# Begin Project
+# PROP AllowPerConfigDependencies 0
+# PROP Scc_ProjName ""
+# PROP Scc_LocalPath ""
+CPP=cl.exe
+RSC=rc.exe
+
+!IF "$(CFG)" == "Debug - Win32 Release"
+
+# PROP BASE Use_MFC 0
+# PROP BASE Use_Debug_Libraries 0
+# PROP BASE Output_Dir "Release"
+# PROP BASE Intermediate_Dir "Release"
+# PROP BASE Target_Dir ""
+# PROP Use_MFC 0
+# PROP Use_Debug_Libraries 0
+# PROP Output_Dir "Release"
+# PROP Intermediate_Dir "Release"
+# PROP Target_Dir ""
+# ADD BASE CPP /nologo /W3 /GX /O2 /D "WIN32" /D "NDEBUG" /D "_CONSOLE" /D "_MBCS" /YX /FD /c
+# ADD CPP /nologo /W3 /GX /O2 /D "WIN32" /D "NDEBUG" /D "_CONSOLE" /D "_MBCS" /YX /FD /c
+# ADD BASE RSC /l 0xc09 /d "NDEBUG"
+# ADD RSC /l 0xc09 /d "NDEBUG"
+BSC32=bscmake.exe
+# ADD BASE BSC32 /nologo
+# ADD BSC32 /nologo
+LINK32=link.exe
+# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /machine:I386
+# ADD LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /machine:I386
+
+!ELSEIF "$(CFG)" == "Debug - Win32 Debug"
+
+# PROP BASE Use_MFC 0
+# PROP BASE Use_Debug_Libraries 1
+# PROP BASE Output_Dir "Debug"
+# PROP BASE Intermediate_Dir "Debug"
+# PROP BASE Target_Dir ""
+# PROP Use_MFC 0
+# PROP Use_Debug_Libraries 1
+# PROP Output_Dir "Debug"
+# PROP Intermediate_Dir "Debug"
+# PROP Ignore_Export_Lib 0
+# PROP Target_Dir ""
+# ADD BASE CPP /nologo /W3 /Gm /GX /ZI /Od /D "WIN32" /D "_DEBUG" /D "_CONSOLE" /D "_MBCS" /YX /FD /GZ /c
+# ADD CPP /nologo /MDd /Ze /W3 /WX /Gm /vd1 /GR- /GX- /ZI /Od /I ".." /D "WIN32" /D "_DEBUG" /D "_CONSOLE" /D "_MBCS" /D "CLEANUP_C" /FR /YX /FD /GZ /c
+# ADD BASE RSC /l 0xc09 /d "_DEBUG"
+# ADD RSC /l 0xc09 /d "_DEBUG"
+BSC32=bscmake.exe
+# ADD BASE BSC32 /nologo
+# ADD BSC32 /nologo
+LINK32=link.exe
+# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /debug /machine:I386 /pdbtype:sept
+# ADD LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib pthreadVC.lib /nologo /subsystem:console /debug /machine:I386 /pdbtype:sept /libpath:".."
+
+!ENDIF
+
+# Begin Target
+
+# Name "Debug - Win32 Release"
+# Name "Debug - Win32 Debug"
+# Begin Source File
+
+SOURCE=.\Debug.txt
+# End Source File
+# Begin Source File
+
+SOURCE=.\semaphore4.c
+# End Source File
+# End Target
+# End Project
diff --git a/tests/Debug.dsw b/tests/Debug.dsw
new file mode 100644
index 0000000..5fd6af3
--- /dev/null
+++ b/tests/Debug.dsw
@@ -0,0 +1,29 @@
+Microsoft Developer Studio Workspace File, Format Version 6.00
+# WARNING: DO NOT EDIT OR DELETE THIS WORKSPACE FILE!
+
+###############################################################################
+
+Project: "Debug"=.\Debug.dsp - Package Owner=<4>
+
+Package=<5>
+{{{
+}}}
+
+Package=<4>
+{{{
+}}}
+
+###############################################################################
+
+Global:
+
+Package=<5>
+{{{
+}}}
+
+Package=<3>
+{{{
+}}}
+
+###############################################################################
+
diff --git a/tests/Debug.ncb b/tests/Debug.ncb
new file mode 100644
index 0000000..68ecb6d
--- /dev/null
+++ b/tests/Debug.ncb
Binary files differ
diff --git a/tests/Debug.opt b/tests/Debug.opt
new file mode 100644
index 0000000..f263d4f
--- /dev/null
+++ b/tests/Debug.opt
Binary files differ
diff --git a/tests/Debug.plg b/tests/Debug.plg
new file mode 100644
index 0000000..9b26377
--- /dev/null
+++ b/tests/Debug.plg
@@ -0,0 +1,35 @@
+<html>
+<body>
+<pre>
+<h1>Build Log</h1>
+<h3>
+--------------------Configuration: Debug - Win32 Debug--------------------
+</h3>
+<h3>Command Lines</h3>
+Creating temporary file "C:\DOCUME~1\ADMINI~1\LOCALS~1\Temp\RSP8EA.tmp" with contents
+[
+/nologo /MDd /Ze /W3 /WX /Gm /vd1 /GR- /GX- /ZI /Od /I ".." /D "WIN32" /D "_DEBUG" /D "_CONSOLE" /D "_MBCS" /D "CLEANUP_C" /FR"Debug/" /Fp"Debug/Debug.pch" /YX /Fo"Debug/" /Fd"Debug/" /FD /GZ /c
+"E:\PTHREADS\pthreads\tests\semaphore4.c"
+]
+Creating command line "cl.exe @C:\DOCUME~1\ADMINI~1\LOCALS~1\Temp\RSP8EA.tmp"
+Creating temporary file "C:\DOCUME~1\ADMINI~1\LOCALS~1\Temp\RSP8EB.tmp" with contents
+[
+kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib pthreadVC.lib /nologo /subsystem:console /incremental:yes /pdb:"Debug/Debug.pdb" /debug /machine:I386 /out:"Debug/Debug.exe" /pdbtype:sept /libpath:".."
+.\Debug\semaphore4.obj
+]
+Creating command line "link.exe @C:\DOCUME~1\ADMINI~1\LOCALS~1\Temp\RSP8EB.tmp"
+<h3>Output Window</h3>
+Compiling...
+semaphore4.c
+Linking...
+Creating command line "bscmake.exe /nologo /o"Debug/Debug.bsc" .\Debug\semaphore4.sbr"
+Creating browse info file...
+<h3>Output Window</h3>
+
+
+
+<h3>Results</h3>
+Debug.exe - 0 error(s), 0 warning(s)
+</pre>
+</body>
+</html>
diff --git a/tests/Debug.txt b/tests/Debug.txt
new file mode 100644
index 0000000..5323874
--- /dev/null
+++ b/tests/Debug.txt
@@ -0,0 +1,6 @@
+This project is used to debug individual test case programs.
+
+To build and debug a test case:
+- add the .c file to this project;
+- remove any .c files from other test cases from this project.
+- build and debug as usual. \ No newline at end of file
diff --git a/tests/GNUmakefile b/tests/GNUmakefile
index aecd718..0e07abf 100644
--- a/tests/GNUmakefile
+++ b/tests/GNUmakefile
@@ -50,8 +50,8 @@ MAKE = make
#
XXCFLAGS =
XXLIBS = -lws2_32
-CFLAGS = -O3 -UNDEBUG -Wall $(XXCFLAGS)
-#CFLAGS = -g -O0 -UNDEBUG -Wall $(XXCFLAGS)
+#CFLAGS = -O3 -UNDEBUG -Wall $(XXCFLAGS)
+CFLAGS = -g -UNDEBUG -Wall $(XXCFLAGS)
BUILD_DIR = ..
INCLUDES = -I.
@@ -81,7 +81,9 @@ TESTS = sizes loadfree \
mutex4 mutex6 mutex6n mutex6e mutex6r \
mutex6s mutex6es mutex6rs \
mutex7 mutex7n mutex7e mutex7r mutex8 mutex8n mutex8e mutex8r \
- count1 once1 tsd1 self2 cancel1 cancel2 \
+ count1 once1 tsd1 self2 \
+ cancel1 cancel2 \
+ semaphore4 semaphore4t \
delay1 delay2 eyal1 \
condvar3 condvar3_1 condvar3_2 condvar3_3 \
condvar4 condvar5 condvar6 condvar7 condvar8 condvar9 \
@@ -253,6 +255,8 @@ self2.pass: create1.pass
semaphore1.pass:
semaphore2.pass:
semaphore3.pass: semaphore2.pass
+semaphore4.pass: semaphore3.pass cancel1.pass
+semaphore4t.pass: semaphore4.pass
sizes.pass:
spin1.pass:
spin2.pass: spin1.pass
diff --git a/tests/Makefile b/tests/Makefile
index 71154b3..10dd7fe 100644
--- a/tests/Makefile
+++ b/tests/Makefile
@@ -1,349 +1,353 @@
-# Makefile for the pthreads test suite.
-# If all of the .pass files can be created, the test suite has passed.
-#
-# --------------------------------------------------------------------------
-#
-# Pthreads-win32 - POSIX Threads Library for Win32
-# Copyright(C) 1998 John E. Bossom
-# Copyright(C) 1999,2003 Pthreads-win32 contributors
-#
-# Contact Email: rpj@callisto.canberra.edu.au
-#
-# The current list of contributors is contained
-# in the file CONTRIBUTORS included with the source
-# code distribution. The list can also be seen at the
-# following World Wide Web location:
-# http://sources.redhat.com/pthreads-win32/contributors.html
-#
-# This library is free software; you can redistribute it and/or
-# modify it under the terms of the GNU Lesser General Public
-# License as published by the Free Software Foundation; either
-# version 2 of the License, or (at your option) any later version.
-#
-# This library is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
-# Lesser General Public License for more details.
-#
-# You should have received a copy of the GNU Lesser General Public
-# License along with this library in the file COPYING.LIB;
-# if not, write to the Free Software Foundation, Inc.,
-# 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
-#
-
-
-CP = copy
-RM = erase
-CAT = type
-MKDIR = mkdir
-TOUCH = echo Passed >
-ECHO = @echo
-
-QAPC = ..\QueueUserAPCEx\User\quserex.dll
-
-CPHDR = pthread.h semaphore.h sched.h
-
-OPTIM = /O2 /Ob0
-
-XXLIBS = ws2_32.lib
-
-# C++ Exceptions
-VCEFLAGS = /GX /TP /DPtW32NoCatchWarn /D__CLEANUP_CXX
-VCELIB = pthreadVCE.lib
-VCEDLL = pthreadVCE.dll
-# Structured Exceptions
-VSEFLAGS = /D__CLEANUP_SEH
-VSELIB = pthreadVSE.lib
-VSEDLL = pthreadVSE.dll
-# C cleanup code
-VCFLAGS = /D__CLEANUP_C
-VCLIB = pthreadVC.lib
-VCDLL = pthreadVC.dll
-# C++ Exceptions in application - using VC version of pthreads dll
-VCXFLAGS = /GX /TP /D__CLEANUP_C
-
-# Defaults
-CPLIB = pthreadVC.lib
-CPDLL = pthreadVC.dll
-
-CFLAGS= $(OPTIM) /W3 /WX /MD /nologo /Yd /Zi
-LFLAGS= /INCREMENTAL:NO
-INCLUDES=-I.
-BUILD_DIR=..
-
-COPYFILES = $(CPHDR) $(CPLIB) $(CPDLL) $(QAPC)
-
-TEST =
-EHFLAGS =
-
-# If a test case returns a non-zero exit code to the shell, make will
-# stop.
-
-PASSES= sizes.pass loadfree.pass \
- self1.pass mutex5.pass \
+# Makefile for the pthreads test suite.
+# If all of the .pass files can be created, the test suite has passed.
+#
+# --------------------------------------------------------------------------
+#
+# Pthreads-win32 - POSIX Threads Library for Win32
+# Copyright(C) 1998 John E. Bossom
+# Copyright(C) 1999,2003 Pthreads-win32 contributors
+#
+# Contact Email: rpj@callisto.canberra.edu.au
+#
+# The current list of contributors is contained
+# in the file CONTRIBUTORS included with the source
+# code distribution. The list can also be seen at the
+# following World Wide Web location:
+# http://sources.redhat.com/pthreads-win32/contributors.html
+#
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2 of the License, or (at your option) any later version.
+#
+# This library is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with this library in the file COPYING.LIB;
+# if not, write to the Free Software Foundation, Inc.,
+# 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
+#
+
+
+CP = copy
+RM = erase
+CAT = type
+MKDIR = mkdir
+TOUCH = echo Passed >
+ECHO = @echo
+
+QAPC = ..\QueueUserAPCEx\User\quserex.dll
+
+CPHDR = pthread.h semaphore.h sched.h
+
+OPTIM = /O2 /Ob0
+
+XXLIBS = ws2_32.lib
+
+# C++ Exceptions
+VCEFLAGS = /GX /TP /DPtW32NoCatchWarn /D__CLEANUP_CXX
+VCELIB = pthreadVCE.lib
+VCEDLL = pthreadVCE.dll
+# Structured Exceptions
+VSEFLAGS = /D__CLEANUP_SEH
+VSELIB = pthreadVSE.lib
+VSEDLL = pthreadVSE.dll
+# C cleanup code
+VCFLAGS = /D__CLEANUP_C
+VCLIB = pthreadVC.lib
+VCDLL = pthreadVC.dll
+# C++ Exceptions in application - using VC version of pthreads dll
+VCXFLAGS = /GX /TP /D__CLEANUP_C
+
+# Defaults
+CPLIB = pthreadVC.lib
+CPDLL = pthreadVC.dll
+
+CFLAGS= $(OPTIM) /W3 /WX /MD /nologo /Yd /Zi
+LFLAGS= /INCREMENTAL:NO
+INCLUDES=-I.
+BUILD_DIR=..
+
+COPYFILES = $(CPHDR) $(CPLIB) $(CPDLL) $(QAPC)
+
+TEST =
+EHFLAGS =
+
+# If a test case returns a non-zero exit code to the shell, make will
+# stop.
+
+PASSES= sizes.pass loadfree.pass \
+ self1.pass mutex5.pass \
mutex1.pass mutex1n.pass mutex1e.pass mutex1r.pass \
semaphore1.pass semaphore2.pass semaphore3.pass \
- mutex2.pass mutex3.pass \
- mutex2r.pass mutex2e.pass mutex3r.pass mutex3e.pass \
- condvar1.pass condvar1_1.pass condvar1_2.pass condvar2.pass condvar2_1.pass \
- exit1.pass create1.pass create2.pass reuse1.pass reuse2.pass equal1.pass \
- kill1.pass valid1.pass valid2.pass \
- exit2.pass exit3.pass exit4 exit5 \
- join0.pass join1.pass join2.pass join3.pass \
- mutex4.pass mutex6.pass mutex6n.pass mutex6e.pass mutex6r.pass \
- mutex6s.pass mutex6es.pass mutex6rs.pass \
- mutex7.pass mutex7n.pass mutex7e.pass mutex7r.pass \
- mutex8.pass mutex8n.pass mutex8e.pass mutex8r.pass \
- count1.pass once1.pass tsd1.pass \
- self2.pass \
- cancel1.pass cancel2.pass \
- delay1.pass delay2.pass eyal1.pass \
- condvar3.pass condvar3_1.pass condvar3_2.pass condvar3_3.pass \
- condvar4.pass condvar5.pass condvar6.pass \
- condvar7.pass condvar8.pass condvar9.pass \
- errno1.pass \
- rwlock1.pass rwlock2.pass rwlock3.pass rwlock4.pass \
- rwlock5.pass rwlock6.pass rwlock7.pass rwlock8.pass \
- rwlock2_t.pass rwlock3_t.pass rwlock4_t.pass rwlock5_t.pass rwlock6_t.pass rwlock6_t2.pass \
- context1.pass \
- cancel3.pass cancel4.pass cancel5.pass cancel6a.pass cancel6d.pass \
- cancel7 cancel8 \
- cleanup0.pass cleanup1.pass cleanup2.pass cleanup3.pass \
- priority1.pass priority2.pass inherit1.pass \
- spin1.pass spin2.pass spin3.pass spin4.pass \
- barrier1.pass barrier2.pass barrier3.pass barrier4.pass barrier5.pass \
- exception1.pass exception2.pass exception3.pass \
- cancel9 create3
-
-BENCHRESULTS = \
- benchtest1.bench benchtest2.bench benchtest3.bench benchtest4.bench benchtest5.bench
-
-help:
- @ $(ECHO) Run one of the following command lines:
- @ $(ECHO) nmake clean VC (to test using VC dll with VC (no EH) applications)
- @ $(ECHO) nmake clean VCX (to test using VC dll with VC++ (EH) applications)
- @ $(ECHO) nmake clean VCE (to test using the VCE dll with VC++ EH applications)
- @ $(ECHO) nmake clean VSE (to test using VSE dll with VC (SEH) applications)
- @ $(ECHO) nmake clean VC-bench (to benchtest using VC dll with C bench app)
- @ $(ECHO) nmake clean VCX-bench (to benchtest using VC dll with C++ bench app)
- @ $(ECHO) nmake clean VCE-bench (to benchtest using VCE dll with C++ bench app)
- @ $(ECHO) nmake clean VSE-bench (to benchtest using VSE dll with SEH bench app)
-
-all:
- @ nmake clean VC
- @ nmake clean VCX
- @ nmake clean VCE
- @ nmake clean VSE
- @ nmake clean VC-bench
-
-# This allows an individual test application to be made using the default lib.
-# e.g. nmake clean test cancel3.exe
-test: $(CPLIB) $(CPDLL) $(CPHDR) $(QAPC)
-
-tests: $(CPLIB) $(CPDLL) $(CPHDR) $(QAPC) $(PASSES)
- @ $(ECHO) ALL TESTS PASSED! Congratulations!
-
-benchtests: $(CPLIB) $(CPDLL) $(CPHDR) $(XXLIBS) $(BENCHRESULTS)
- @ $(ECHO) ALL BENCH TESTS DONE.
-
-sizes.pass: sizes.exe
- @ $(ECHO) ... Running $(TEST) test: $*.exe
- @ .\$*.exe > SIZES.$(TEST)
- @ $(CAT) SIZES.$(TEST)
- @ $(ECHO) ...... Passed
- @ $(TOUCH) $*.pass
-
-$(PASSES): $*.exe
- @ $(ECHO) ... Running $(TEST) test: $*.exe
- @ .\$*.exe
- @ $(ECHO) ...... Passed
- @ $(TOUCH) $*.pass
-
-$(BENCHRESULTS): $*.exe
- @ $(ECHO) ... Running $(TEST) benchtest: $*.exe
- @ .\$*.exe
- @ $(ECHO) ...... Done
- @ $(TOUCH) $*.bench
-
-VCE:
- @ nmake TEST="$@" CPLIB="$(VCELIB)" CPDLL="$(VCEDLL)" EHFLAGS="$(VCEFLAGS)" tests
-
-VSE:
- @ nmake TEST="$@" CPLIB="$(VSELIB)" CPDLL="$(VSEDLL)" EHFLAGS="$(VSEFLAGS)" tests
-
-VC:
- @ nmake TEST="$@" CPLIB="$(VCLIB)" CPDLL="$(VCDLL)" EHFLAGS="$(VCFLAGS)" tests
-
-VCX:
- @ nmake TEST="$@" CPLIB="$(VCLIB)" CPDLL="$(VCDLL)" EHFLAGS="$(VCXFLAGS)" tests
-
-VCE-bench:
- @ nmake TEST="$@" CPLIB="$(VCELIB)" CPDLL="$(VCEDLL)" EHFLAGS="$(VCEFLAGS)" XXLIBS="benchlib.o" benchtests
-
-VSE-bench:
- @ nmake TEST="$@" CPLIB="$(VSELIB)" CPDLL="$(VSEDLL)" EHFLAGS="$(VSEFLAGS)" XXLIBS="benchlib.o" benchtests
-
-VC-bench:
- @ nmake TEST="$@" CPLIB="$(VCLIB)" CPDLL="$(VCDLL)" EHFLAGS="$(VCFLAGS)" XXLIBS="benchlib.o" benchtests
-
-VCX-bench:
- @ nmake TEST="$@" CPLIB="$(VCLIB)" CPDLL="$(VCDLL)" EHFLAGS="$(VCXFLAGS)" XXLIBS="benchlib.o" benchtests
-
-.c.exe:
- @ $(ECHO) $(CC) $(EHFLAGS) $(CFLAGS) $(INCLUDES) $< /Fe$@ /link $(LFLAGS) $(CPLIB) $(XXLIBS)
- @ $(CC) $(EHFLAGS) $(CFLAGS) $(INCLUDES) $< /Fe$@ /link $(LFLAGS) $(CPLIB) $(XXLIBS)
-
-.c.o:
- @ $(ECHO) $(CC) $(EHFLAGS) /c $(CFLAGS) $(INCLUDES) $< /Fo$@
- @ $(CC) $(EHFLAGS) $(CFLAGS) /c $(INCLUDES) $< /Fo$@
-
-.c.i:
- @ $(CC) /P $(EHFLAGS) $(CFLAGS) $(INCLUDES) $<
-
-$(COPYFILES):
- @ $(ECHO) Copying $@
- @ $(CP) $(BUILD_DIR)\$@ .
-
-pthread.dll: $(CPDLL)
- @ $(CP) $(CPDLL) pthread.dll
- @ $(CP) $(CPLIB) pthread.lib
-
-clean:
- - $(RM) *.dll
- - $(RM) *.lib
- - $(RM) pthread.h
- - $(RM) semaphore.h
- - $(RM) sched.h
- - $(RM) *.e
- - $(RM) *.i
- - $(RM) *.obj
- - $(RM) *.pdb
- - $(RM) *.o
- - $(RM) *.asm
- - $(RM) *.exe
- - $(RM) *.pass
- - $(RM) *.bench
- - $(RM) *.log
-
-benchtest1.bench:
-benchtest2.bench:
-benchtest3.bench:
-benchtest4.bench:
-benchtest5.bench:
-barrier1.pass:
-barrier2.pass: barrier1.pass
-barrier3.pass: barrier2.pass
-barrier4.pass: barrier3.pass
-barrier5.pass: barrier4.pass
-cancel1.pass: create1.pass
-cancel2.pass: cancel1.pass
-cancel3.pass: context1.pass
-cancel4.pass: cancel3.pass
-cancel5.pass: cancel3.pass
-cancel6a.pass: cancel3.pass
-cancel6d.pass: cancel3.pass
-cancel7.pass: kill1.pass
-cancel8.pass: cancel7.pass
-cleanup0.pass: cancel5.pass
-cleanup1.pass: cleanup0.pass
-cleanup2.pass: cleanup1.pass
-cleanup3.pass: cleanup2.pass
-condvar1.pass:
-condvar1_1.pass: condvar1.pass
-condvar1_2.pass: join2.pass
-condvar2.pass: condvar1.pass
-condvar2_1.pass: condvar2.pass join2.pass
-condvar3.pass: create1.pass condvar2.pass
-condvar3_1.pass: condvar3.pass join2.pass
-condvar3_2.pass: condvar3_1.pass
-condvar3_3.pass: condvar3_2.pass
-condvar4.pass: create1.pass
-condvar5.pass: condvar4.pass
-condvar6.pass: condvar5.pass
-condvar7.pass: condvar6.pass cleanup1.pass
-condvar8.pass: condvar7.pass
-condvar9.pass: condvar8.pass
-context1.pass: cancel2.pass
-count1.pass: join1.pass
-create1.pass: mutex2.pass
-create2.pass: create1.pass
-create3.pass:
-delay1.pass:
-delay2.pass: delay1.pass
-equal1.pass: create1.pass
-errno1.pass: mutex3.pass
-exception1.pass: cancel4.pass
-exception2.pass: exception1.pass
-exception3.pass: exception2.pass
-exit1.pass:
-exit2.pass: create1.pass
-exit3.pass: create1.pass
-exit4.pass:
-exit5.pass: kill1.pass
-eyal1.pass: tsd1.pass
-inherit1.pass: join1.pass priority1.pass
-join0.pass: create1.pass
-join1.pass: create1.pass
-join2.pass: create1.pass
-join3.pass: join2.pass
-kill1.pass:
-loadfree.pass: pthread.dll
-mutex1.pass: self1.pass
-mutex1n.pass: mutex1.pass
-mutex1e.pass: mutex1.pass
-mutex1r.pass: mutex1.pass
-mutex2.pass: mutex1.pass
-mutex2r.pass: mutex2.pass
-mutex2e.pass: mutex2.pass
-mutex3.pass: create1.pass
-mutex3r.pass: mutex3.pass
-mutex3e.pass: mutex3.pass
-mutex4.pass: mutex3.pass
-mutex5.pass:
-mutex6.pass: mutex4.pass
-mutex6n.pass: mutex4.pass
-mutex6e.pass: mutex4.pass
-mutex6r.pass: mutex4.pass
-mutex6s.pass: mutex6.pass
-mutex6rs.pass: mutex6r.pass
-mutex6es.pass: mutex6e.pass
-mutex7.pass: mutex6.pass
-mutex7n.pass: mutex6n.pass
-mutex7e.pass: mutex6e.pass
-mutex7r.pass: mutex6r.pass
-mutex8.pass: mutex7.pass
-mutex8n.pass: mutex7n.pass
-mutex8e.pass: mutex7e.pass
-mutex8r.pass: mutex7r.pass
-once1.pass: create1.pass
-priority1.pass: join1.pass
-priority2.pass: priority1.pass barrier3.pass
-reuse1.pass: create2.pass
-reuse2.pass: reuse1.pass
-rwlock1.pass: condvar6.pass
-rwlock2.pass: rwlock1.pass
-rwlock3.pass: rwlock2.pass
-rwlock4.pass: rwlock3.pass
-rwlock5.pass: rwlock4.pass
-rwlock6.pass: rwlock5.pass
-rwlock7.pass: rwlock6.pass
-rwlock8.pass: rwlock7.pass
-rwlock2_t.pass: rwlock2.pass
-rwlock3_t.pass: rwlock2_t.pass
-rwlock4_t.pass: rwlock3_t.pass
-rwlock5_t.pass: rwlock4_t.pass
-rwlock6_t.pass: rwlock5_t.pass
-rwlock6_t2.pass: rwlock6_t.pass
-self1.pass:
-self2.pass: create1.pass
-semaphore1.pass:
-semaphore2.pass:
-sizes.pass:
-spin1.pass:
-spin2.pass: spin1.pass
-spin3.pass: spin2.pass
-spin4.pass: spin3.pass
-tsd1.pass: join1.pass
-valid1.pass: join1.pass
-valid2.pass: valid1.pass
-cancel9.pass: cancel8.pass
+ mutex2.pass mutex3.pass \
+ mutex2r.pass mutex2e.pass mutex3r.pass mutex3e.pass \
+ condvar1.pass condvar1_1.pass condvar1_2.pass condvar2.pass condvar2_1.pass \
+ exit1.pass create1.pass create2.pass reuse1.pass reuse2.pass equal1.pass \
+ kill1.pass valid1.pass valid2.pass \
+ exit2.pass exit3.pass exit4 exit5 \
+ join0.pass join1.pass join2.pass join3.pass \
+ mutex4.pass mutex6.pass mutex6n.pass mutex6e.pass mutex6r.pass \
+ mutex6s.pass mutex6es.pass mutex6rs.pass \
+ mutex7.pass mutex7n.pass mutex7e.pass mutex7r.pass \
+ mutex8.pass mutex8n.pass mutex8e.pass mutex8r.pass \
+ count1.pass once1.pass tsd1.pass \
+ self2.pass \
+ cancel1.pass cancel2.pass \
+ semaphore4.pass semaphore4t.pass \
+ delay1.pass delay2.pass eyal1.pass \
+ condvar3.pass condvar3_1.pass condvar3_2.pass condvar3_3.pass \
+ condvar4.pass condvar5.pass condvar6.pass \
+ condvar7.pass condvar8.pass condvar9.pass \
+ errno1.pass \
+ rwlock1.pass rwlock2.pass rwlock3.pass rwlock4.pass \
+ rwlock5.pass rwlock6.pass rwlock7.pass rwlock8.pass \
+ rwlock2_t.pass rwlock3_t.pass rwlock4_t.pass rwlock5_t.pass rwlock6_t.pass rwlock6_t2.pass \
+ context1.pass \
+ cancel3.pass cancel4.pass cancel5.pass cancel6a.pass cancel6d.pass \
+ cancel7 cancel8 \
+ cleanup0.pass cleanup1.pass cleanup2.pass cleanup3.pass \
+ priority1.pass priority2.pass inherit1.pass \
+ spin1.pass spin2.pass spin3.pass spin4.pass \
+ barrier1.pass barrier2.pass barrier3.pass barrier4.pass barrier5.pass \
+ exception1.pass exception2.pass exception3.pass \
+ cancel9 create3
+
+BENCHRESULTS = \
+ benchtest1.bench benchtest2.bench benchtest3.bench benchtest4.bench benchtest5.bench
+
+help:
+ @ $(ECHO) Run one of the following command lines:
+ @ $(ECHO) nmake clean VC (to test using VC dll with VC (no EH) applications)
+ @ $(ECHO) nmake clean VCX (to test using VC dll with VC++ (EH) applications)
+ @ $(ECHO) nmake clean VCE (to test using the VCE dll with VC++ EH applications)
+ @ $(ECHO) nmake clean VSE (to test using VSE dll with VC (SEH) applications)
+ @ $(ECHO) nmake clean VC-bench (to benchtest using VC dll with C bench app)
+ @ $(ECHO) nmake clean VCX-bench (to benchtest using VC dll with C++ bench app)
+ @ $(ECHO) nmake clean VCE-bench (to benchtest using VCE dll with C++ bench app)
+ @ $(ECHO) nmake clean VSE-bench (to benchtest using VSE dll with SEH bench app)
+
+all:
+ @ nmake clean VC
+ @ nmake clean VCX
+ @ nmake clean VCE
+ @ nmake clean VSE
+ @ nmake clean VC-bench
+
+# This allows an individual test application to be made using the default lib.
+# e.g. nmake clean test cancel3.exe
+test: $(CPLIB) $(CPDLL) $(CPHDR) $(QAPC)
+
+tests: $(CPLIB) $(CPDLL) $(CPHDR) $(QAPC) $(PASSES)
+ @ $(ECHO) ALL TESTS PASSED! Congratulations!
+
+benchtests: $(CPLIB) $(CPDLL) $(CPHDR) $(XXLIBS) $(BENCHRESULTS)
+ @ $(ECHO) ALL BENCH TESTS DONE.
+
+sizes.pass: sizes.exe
+ @ $(ECHO) ... Running $(TEST) test: $*.exe
+ @ .\$*.exe > SIZES.$(TEST)
+ @ $(CAT) SIZES.$(TEST)
+ @ $(ECHO) ...... Passed
+ @ $(TOUCH) $*.pass
+
+$(PASSES): $*.exe
+ @ $(ECHO) ... Running $(TEST) test: $*.exe
+ @ .\$*.exe
+ @ $(ECHO) ...... Passed
+ @ $(TOUCH) $*.pass
+
+$(BENCHRESULTS): $*.exe
+ @ $(ECHO) ... Running $(TEST) benchtest: $*.exe
+ @ .\$*.exe
+ @ $(ECHO) ...... Done
+ @ $(TOUCH) $*.bench
+
+VCE:
+ @ nmake TEST="$@" CPLIB="$(VCELIB)" CPDLL="$(VCEDLL)" EHFLAGS="$(VCEFLAGS)" tests
+
+VSE:
+ @ nmake TEST="$@" CPLIB="$(VSELIB)" CPDLL="$(VSEDLL)" EHFLAGS="$(VSEFLAGS)" tests
+
+VC:
+ @ nmake TEST="$@" CPLIB="$(VCLIB)" CPDLL="$(VCDLL)" EHFLAGS="$(VCFLAGS)" tests
+
+VCX:
+ @ nmake TEST="$@" CPLIB="$(VCLIB)" CPDLL="$(VCDLL)" EHFLAGS="$(VCXFLAGS)" tests
+
+VCE-bench:
+ @ nmake TEST="$@" CPLIB="$(VCELIB)" CPDLL="$(VCEDLL)" EHFLAGS="$(VCEFLAGS)" XXLIBS="benchlib.o" benchtests
+
+VSE-bench:
+ @ nmake TEST="$@" CPLIB="$(VSELIB)" CPDLL="$(VSEDLL)" EHFLAGS="$(VSEFLAGS)" XXLIBS="benchlib.o" benchtests
+
+VC-bench:
+ @ nmake TEST="$@" CPLIB="$(VCLIB)" CPDLL="$(VCDLL)" EHFLAGS="$(VCFLAGS)" XXLIBS="benchlib.o" benchtests
+
+VCX-bench:
+ @ nmake TEST="$@" CPLIB="$(VCLIB)" CPDLL="$(VCDLL)" EHFLAGS="$(VCXFLAGS)" XXLIBS="benchlib.o" benchtests
+
+.c.exe:
+ @ $(ECHO) $(CC) $(EHFLAGS) $(CFLAGS) $(INCLUDES) $< /Fe$@ /link $(LFLAGS) $(CPLIB) $(XXLIBS)
+ @ $(CC) $(EHFLAGS) $(CFLAGS) $(INCLUDES) $< /Fe$@ /link $(LFLAGS) $(CPLIB) $(XXLIBS)
+
+.c.o:
+ @ $(ECHO) $(CC) $(EHFLAGS) /c $(CFLAGS) $(INCLUDES) $< /Fo$@
+ @ $(CC) $(EHFLAGS) $(CFLAGS) /c $(INCLUDES) $< /Fo$@
+
+.c.i:
+ @ $(CC) /P $(EHFLAGS) $(CFLAGS) $(INCLUDES) $<
+
+$(COPYFILES):
+ @ $(ECHO) Copying $@
+ @ $(CP) $(BUILD_DIR)\$@ .
+
+pthread.dll: $(CPDLL)
+ @ $(CP) $(CPDLL) pthread.dll
+ @ $(CP) $(CPLIB) pthread.lib
+
+clean:
+ - $(RM) *.dll
+ - $(RM) *.lib
+ - $(RM) pthread.h
+ - $(RM) semaphore.h
+ - $(RM) sched.h
+ - $(RM) *.e
+ - $(RM) *.i
+ - $(RM) *.obj
+ - $(RM) *.pdb
+ - $(RM) *.o
+ - $(RM) *.asm
+ - $(RM) *.exe
+ - $(RM) *.pass
+ - $(RM) *.bench
+ - $(RM) *.log
+
+benchtest1.bench:
+benchtest2.bench:
+benchtest3.bench:
+benchtest4.bench:
+benchtest5.bench:
+barrier1.pass:
+barrier2.pass: barrier1.pass
+barrier3.pass: barrier2.pass
+barrier4.pass: barrier3.pass
+barrier5.pass: barrier4.pass
+cancel1.pass: create1.pass
+cancel2.pass: cancel1.pass
+cancel3.pass: context1.pass
+cancel4.pass: cancel3.pass
+cancel5.pass: cancel3.pass
+cancel6a.pass: cancel3.pass
+cancel6d.pass: cancel3.pass
+cancel7.pass: kill1.pass
+cancel8.pass: cancel7.pass
+cleanup0.pass: cancel5.pass
+cleanup1.pass: cleanup0.pass
+cleanup2.pass: cleanup1.pass
+cleanup3.pass: cleanup2.pass
+condvar1.pass:
+condvar1_1.pass: condvar1.pass
+condvar1_2.pass: join2.pass
+condvar2.pass: condvar1.pass
+condvar2_1.pass: condvar2.pass join2.pass
+condvar3.pass: create1.pass condvar2.pass
+condvar3_1.pass: condvar3.pass join2.pass
+condvar3_2.pass: condvar3_1.pass
+condvar3_3.pass: condvar3_2.pass
+condvar4.pass: create1.pass
+condvar5.pass: condvar4.pass
+condvar6.pass: condvar5.pass
+condvar7.pass: condvar6.pass cleanup1.pass
+condvar8.pass: condvar7.pass
+condvar9.pass: condvar8.pass
+context1.pass: cancel2.pass
+count1.pass: join1.pass
+create1.pass: mutex2.pass
+create2.pass: create1.pass
+create3.pass:
+delay1.pass:
+delay2.pass: delay1.pass
+equal1.pass: create1.pass
+errno1.pass: mutex3.pass
+exception1.pass: cancel4.pass
+exception2.pass: exception1.pass
+exception3.pass: exception2.pass
+exit1.pass:
+exit2.pass: create1.pass
+exit3.pass: create1.pass
+exit4.pass:
+exit5.pass: kill1.pass
+eyal1.pass: tsd1.pass
+inherit1.pass: join1.pass priority1.pass
+join0.pass: create1.pass
+join1.pass: create1.pass
+join2.pass: create1.pass
+join3.pass: join2.pass
+kill1.pass:
+loadfree.pass: pthread.dll
+mutex1.pass: self1.pass
+mutex1n.pass: mutex1.pass
+mutex1e.pass: mutex1.pass
+mutex1r.pass: mutex1.pass
+mutex2.pass: mutex1.pass
+mutex2r.pass: mutex2.pass
+mutex2e.pass: mutex2.pass
+mutex3.pass: create1.pass
+mutex3r.pass: mutex3.pass
+mutex3e.pass: mutex3.pass
+mutex4.pass: mutex3.pass
+mutex5.pass:
+mutex6.pass: mutex4.pass
+mutex6n.pass: mutex4.pass
+mutex6e.pass: mutex4.pass
+mutex6r.pass: mutex4.pass
+mutex6s.pass: mutex6.pass
+mutex6rs.pass: mutex6r.pass
+mutex6es.pass: mutex6e.pass
+mutex7.pass: mutex6.pass
+mutex7n.pass: mutex6n.pass
+mutex7e.pass: mutex6e.pass
+mutex7r.pass: mutex6r.pass
+mutex8.pass: mutex7.pass
+mutex8n.pass: mutex7n.pass
+mutex8e.pass: mutex7e.pass
+mutex8r.pass: mutex7r.pass
+once1.pass: create1.pass
+priority1.pass: join1.pass
+priority2.pass: priority1.pass barrier3.pass
+reuse1.pass: create2.pass
+reuse2.pass: reuse1.pass
+rwlock1.pass: condvar6.pass
+rwlock2.pass: rwlock1.pass
+rwlock3.pass: rwlock2.pass
+rwlock4.pass: rwlock3.pass
+rwlock5.pass: rwlock4.pass
+rwlock6.pass: rwlock5.pass
+rwlock7.pass: rwlock6.pass
+rwlock8.pass: rwlock7.pass
+rwlock2_t.pass: rwlock2.pass
+rwlock3_t.pass: rwlock2_t.pass
+rwlock4_t.pass: rwlock3_t.pass
+rwlock5_t.pass: rwlock4_t.pass
+rwlock6_t.pass: rwlock5_t.pass
+rwlock6_t2.pass: rwlock6_t.pass
+self1.pass:
+self2.pass: create1.pass
+semaphore1.pass:
+semaphore2.pass:
+semaphore3.pass: semaphore2.pass
+semaphore4.pass: semaphore3.pass cancel1.pass
+semaphore4t.pass: semaphore4.pass
+sizes.pass:
+spin1.pass:
+spin2.pass: spin1.pass
+spin3.pass: spin2.pass
+spin4.pass: spin3.pass
+tsd1.pass: join1.pass
+valid1.pass: join1.pass
+valid2.pass: valid1.pass
+cancel9.pass: cancel8.pass
diff --git a/tests/README b/tests/README
index b767f89..784c4f7 100644
--- a/tests/README
+++ b/tests/README
@@ -5,19 +5,19 @@ These make scripts expect to be able to copy the dll, libarary
and header files from this directory's parent directory,
which should be the pthreads-win32 source directory.
-MS VC++ nmake
+MS VC nmake
-------------
Run the target corresponding to the DLL version being tested:
-nmake clean VCE
+nmake clean VC
or:
-nmake clean VSE
+nmake clean VS
-GNU G++ make
+GNU GCC make
------------
Run "make clean" and then "make". See the "Known bugs" section
@@ -36,3 +36,9 @@ Tests written in this test suite should behave in the following manner:
* No diagnostic output should appear when the test is succeeding.
Diagnostic output may be emitted if something in the test
fails, to help determine the cause of the test failure.
+
+Notes:
+------
+
+Many test cases use knowledge of implementation internals which are supposed
+to be opaque to portable applications.
diff --git a/tests/README.benchtests b/tests/README.benchtests
index 01051a2..e02cb3e 100644
--- a/tests/README.benchtests
+++ b/tests/README.benchtests
@@ -1,97 +1,97 @@
-
-------------
-Benchmarking
-------------
-There is a new but growing set a benchmarking programs in the
-"tests" directory. These should be runnable using the
-following command-lines corresponding to each of the possible
-library builds:
-
-MSVC:
-nmake clean VC-bench
-nmake clean VCE-bench
-nmake clean VSE-bench
-
-Mingw32:
-make clean GC-bench
-make clean GCE-bench
-
-UWIN:
-The benchtests are run as part of the testsuite.
-
-
-Mutex benchtests
-----------------
-
-benchtest1 - Lock plus unlock on an unlocked mutex.
-benchtest2 - Lock plus unlock on a locked mutex.
-benchtest3 - Trylock on a locked mutex.
-benchtest4 - Trylock plus unlock on an unlocked mutex.
-
-
-Each test times up to three alternate synchronisation
-implementations as a reference, and then times each of
-the four mutex types provided by the library. Each is
-described below:
-
-Simple Critical Section
-- uses a simple Win32 critical section. There is no
-additional overhead for this case as there is in the
-remaining cases.
-
-POSIX mutex implemented using a Critical Section
-- The old implementation which uses runtime adaptation
-depending on the Windows variant being run on. When
-the pthreads DLL was run on WinNT or higher then
-POSIX mutexes would use Win32 Critical Sections.
-
-POSIX mutex implemented using a Win32 Mutex
-- The old implementation which uses runtime adaptation
-depending on the Windows variant being run on. When
-the pthreads DLL was run on Win9x then POSIX mutexes
-would use Win32 Mutexes (because TryEnterCriticalSection
-is not implemented on Win9x).
-
-PTHREAD_MUTEX_DEFAULT
-PTHREAD_MUTEX_NORMAL
-PTHREAD_MUTEX_ERRORCHECK
-PTHREAD_MUTEX_RECURSIVE
-- The current implementation supports these mutex types.
-The underlying basis of POSIX mutexes is now the same
-irrespective of the Windows variant, and should therefore
-have consistent performance.
-
-
-In all benchtests, the operation is repeated a large
-number of times and an average is calculated. Loop
-overhead is measured and subtracted from all test times.
-
-Comment on the results
-----------------------
-The gain in performance for Win9x systems is enormous - up to
-40 times faster for unlocked mutexes (2 times faster for locked
-mutexes).
-
-Pthread_mutex_trylock also appears to be faster for locked mutexes.
-
-The price for the new consistency between WinNT and Win9x is
-slower performance (up to twice as long) across a lock/unlock
-sequence. It is difficult to get a good split timing for lock
-and unlock operations, but by code inspection, it is the unlock
-operation that is slowing the pair down in comparison with the
-old-style CS mutexes, even for the fast PTHREAD_MUTEX_NORMAL mutex
-type with no other waiting threads. However, comparitive
-performance for operations on already locked mutexes is very close.
-
-When this is translated to real-world applications, the overall
-camparitive performance should be almost identical on NT class
-systems. That is, applications with heavy mutex contention should
-have almost equal performance, while applications with only light
-mutex contention should also have almost equal performance because
-the most critical operation in this case is the lock operation.
-
-Overall, the newer pthreads-win32 mutex routines are only slower
-(on NT class systems) where and when it is least critical.
-
-Thanks go to Thomas Pfaff for the current implementation of mutex
-routines.
+
+------------
+Benchmarking
+------------
+There is a new but growing set a benchmarking programs in the
+"tests" directory. These should be runnable using the
+following command-lines corresponding to each of the possible
+library builds:
+
+MSVC:
+nmake clean VC-bench
+nmake clean VCE-bench
+nmake clean VSE-bench
+
+Mingw32:
+make clean GC-bench
+make clean GCE-bench
+
+UWIN:
+The benchtests are run as part of the testsuite.
+
+
+Mutex benchtests
+----------------
+
+benchtest1 - Lock plus unlock on an unlocked mutex.
+benchtest2 - Lock plus unlock on a locked mutex.
+benchtest3 - Trylock on a locked mutex.
+benchtest4 - Trylock plus unlock on an unlocked mutex.
+
+
+Each test times up to three alternate synchronisation
+implementations as a reference, and then times each of
+the four mutex types provided by the library. Each is
+described below:
+
+Simple Critical Section
+- uses a simple Win32 critical section. There is no
+additional overhead for this case as there is in the
+remaining cases.
+
+POSIX mutex implemented using a Critical Section
+- The old implementation which uses runtime adaptation
+depending on the Windows variant being run on. When
+the pthreads DLL was run on WinNT or higher then
+POSIX mutexes would use Win32 Critical Sections.
+
+POSIX mutex implemented using a Win32 Mutex
+- The old implementation which uses runtime adaptation
+depending on the Windows variant being run on. When
+the pthreads DLL was run on Win9x then POSIX mutexes
+would use Win32 Mutexes (because TryEnterCriticalSection
+is not implemented on Win9x).
+
+PTHREAD_MUTEX_DEFAULT
+PTHREAD_MUTEX_NORMAL
+PTHREAD_MUTEX_ERRORCHECK
+PTHREAD_MUTEX_RECURSIVE
+- The current implementation supports these mutex types.
+The underlying basis of POSIX mutexes is now the same
+irrespective of the Windows variant, and should therefore
+have consistent performance.
+
+
+In all benchtests, the operation is repeated a large
+number of times and an average is calculated. Loop
+overhead is measured and subtracted from all test times.
+
+Comment on the results
+----------------------
+The gain in performance for Win9x systems is enormous - up to
+40 times faster for unlocked mutexes (2 times faster for locked
+mutexes).
+
+Pthread_mutex_trylock also appears to be faster for locked mutexes.
+
+The price for the new consistency between WinNT and Win9x is
+slower performance (up to twice as long) across a lock/unlock
+sequence. It is difficult to get a good split timing for lock
+and unlock operations, but by code inspection, it is the unlock
+operation that is slowing the pair down in comparison with the
+old-style CS mutexes, even for the fast PTHREAD_MUTEX_NORMAL mutex
+type with no other waiting threads. However, comparitive
+performance for operations on already locked mutexes is very close.
+
+When this is translated to real-world applications, the overall
+camparitive performance should be almost identical on NT class
+systems. That is, applications with heavy mutex contention should
+have almost equal performance, while applications with only light
+mutex contention should also have almost equal performance because
+the most critical operation in this case is the lock operation.
+
+Overall, the newer pthreads-win32 mutex routines are only slower
+(on NT class systems) where and when it is least critical.
+
+Thanks go to Thomas Pfaff for the current implementation of mutex
+routines.
diff --git a/tests/cancel1.c b/tests/cancel1.c
index 390db9f..55409d6 100644
--- a/tests/cancel1.c
+++ b/tests/cancel1.c
@@ -129,7 +129,7 @@ main()
int i;
pthread_t t[NUMTHREADS + 1];
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
diff --git a/tests/cancel2.c b/tests/cancel2.c
index d317ef9..dbc8c85 100644
--- a/tests/cancel2.c
+++ b/tests/cancel2.c
@@ -167,7 +167,7 @@ main()
int i;
pthread_t t[NUMTHREADS + 1];
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
assert(pthread_mutex_lock(&waitLock) == 0);
for (i = 1; i <= NUMTHREADS; i++)
diff --git a/tests/cancel3.c b/tests/cancel3.c
index c3657d6..3258fd0 100644
--- a/tests/cancel3.c
+++ b/tests/cancel3.c
@@ -126,7 +126,7 @@ main ()
int i;
pthread_t t[NUMTHREADS + 1];
- assert ((t[0] = pthread_self ()) != NULL);
+ assert ((t[0] = pthread_self ()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
diff --git a/tests/cancel4.c b/tests/cancel4.c
index b77f921..5b1b7e7 100644
--- a/tests/cancel4.c
+++ b/tests/cancel4.c
@@ -128,7 +128,7 @@ main()
int i;
pthread_t t[NUMTHREADS + 1];
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
diff --git a/tests/cancel6a.c b/tests/cancel6a.c
index 4aea81a..644cd4a 100644
--- a/tests/cancel6a.c
+++ b/tests/cancel6a.c
@@ -115,7 +115,7 @@ main()
int i;
pthread_t t[NUMTHREADS + 1];
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
diff --git a/tests/cancel6d.c b/tests/cancel6d.c
index 9a69adb..d0ad7ac 100644
--- a/tests/cancel6d.c
+++ b/tests/cancel6d.c
@@ -119,7 +119,7 @@ main()
int i;
pthread_t t[NUMTHREADS + 1];
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
@@ -167,11 +167,6 @@ main()
int fail = 0;
int result = 0;
- /*
- * The thread does not contain any cancelation points, so
- * a return value of PTHREAD_CANCELED confirms that async
- * cancelation succeeded.
- */
assert(pthread_join(t[i], (void **) &result) == 0);
fail = (result != (int) PTHREAD_CANCELED);
diff --git a/tests/cancel7.c b/tests/cancel7.c
index 9ad7db0..0609710 100644
--- a/tests/cancel7.c
+++ b/tests/cancel7.c
@@ -109,7 +109,7 @@ Win32thread(void * arg)
assert(bag->started == 0);
bag->started = 1;
- assert((bag->self = pthread_self()) != NULL);
+ assert((bag->self = pthread_self()).p != NULL);
assert(pthread_kill(bag->self, 0) == 0);
for (i = 0; i < 100; i++)
@@ -191,7 +191,7 @@ main()
result = (int) PTHREAD_CANCELED;
#endif
- assert(threadbag[i].self != NULL);
+ assert(threadbag[i].self.p != NULL);
assert(pthread_kill(threadbag[i].self, 0) == ESRCH);
fail = (result != (int) PTHREAD_CANCELED);
diff --git a/tests/cancel8.c b/tests/cancel8.c
index 484472e..1a99640 100644
--- a/tests/cancel8.c
+++ b/tests/cancel8.c
@@ -111,7 +111,7 @@ Win32thread(void * arg)
assert(bag->started == 0);
bag->started = 1;
- assert((bag->self = pthread_self()) != NULL);
+ assert((bag->self = pthread_self()).p != NULL);
assert(pthread_kill(bag->self, 0) == 0);
assert(pthread_mutex_lock(&CVLock) == 0);
@@ -192,7 +192,7 @@ main()
result = (int) PTHREAD_CANCELED;
#endif
- assert(threadbag[i].self != NULL);
+ assert(threadbag[i].self.p != NULL);
assert(pthread_kill(threadbag[i].self, 0) == ESRCH);
fail = (result != (int) PTHREAD_CANCELED);
diff --git a/tests/cleanup0.c b/tests/cleanup0.c
index 03098d3..8171e8c 100644
--- a/tests/cleanup0.c
+++ b/tests/cleanup0.c
@@ -140,7 +140,7 @@ main()
int i;
pthread_t t[NUMTHREADS + 1];
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
diff --git a/tests/cleanup1.c b/tests/cleanup1.c
index 5397297..18518da 100644
--- a/tests/cleanup1.c
+++ b/tests/cleanup1.c
@@ -98,7 +98,6 @@ static void
#ifdef __CLEANUP_C
__cdecl
#endif
-
increment_pop_count(void * arg)
{
int * c = (int *) arg;
@@ -150,7 +149,7 @@ main()
int i;
pthread_t t[NUMTHREADS + 1];
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
diff --git a/tests/cleanup2.c b/tests/cleanup2.c
index 2549b0c..3f0f445 100644
--- a/tests/cleanup2.c
+++ b/tests/cleanup2.c
@@ -134,7 +134,7 @@ main()
int i;
pthread_t t[NUMTHREADS + 1];
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
diff --git a/tests/cleanup3.c b/tests/cleanup3.c
index dd43ccc..9c29be7 100644
--- a/tests/cleanup3.c
+++ b/tests/cleanup3.c
@@ -137,7 +137,7 @@ main()
int i;
pthread_t t[NUMTHREADS + 1];
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
diff --git a/tests/condvar1_2.c b/tests/condvar1_2.c
index 07b3816..884832a 100644
--- a/tests/condvar1_2.c
+++ b/tests/condvar1_2.c
@@ -79,8 +79,8 @@
#include "test.h"
enum {
- NUM_CV = 100,
- NUM_LOOPS = 100
+ NUM_CV = 5,
+ NUM_LOOPS = 5
};
static pthread_cond_t cv[NUM_CV];
diff --git a/tests/condvar3.c b/tests/condvar3.c
index 8fa9ef7..7a30531 100644
--- a/tests/condvar3.c
+++ b/tests/condvar3.c
@@ -120,7 +120,7 @@ main()
assert(pthread_attr_init(&attr) == 0);
assert(pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_DETACHED) == 0);
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
assert(pthread_cond_init(&cv, NULL) == 0);
diff --git a/tests/condvar4.c b/tests/condvar4.c
index 15a32ba..2da6708 100644
--- a/tests/condvar4.c
+++ b/tests/condvar4.c
@@ -119,7 +119,7 @@ main()
cvthing.shared = 0;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
assert(cvthing.notbusy == PTHREAD_COND_INITIALIZER);
diff --git a/tests/condvar5.c b/tests/condvar5.c
index ff5ad70..3b016b9 100644
--- a/tests/condvar5.c
+++ b/tests/condvar5.c
@@ -118,7 +118,7 @@ main()
cvthing.shared = 0;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
assert(cvthing.notbusy == PTHREAD_COND_INITIALIZER);
diff --git a/tests/condvar6.c b/tests/condvar6.c
index d642f9c..0cf92a4 100644
--- a/tests/condvar6.c
+++ b/tests/condvar6.c
@@ -149,7 +149,7 @@ main()
cvthing.shared = 0;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
assert(cvthing.notbusy == PTHREAD_COND_INITIALIZER);
@@ -164,7 +164,7 @@ main()
abstime.tv_sec += 5;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
awoken = 0;
diff --git a/tests/condvar7.c b/tests/condvar7.c
index a5bf4bf..9e0b29c 100644
--- a/tests/condvar7.c
+++ b/tests/condvar7.c
@@ -159,7 +159,7 @@ main()
cvthing.shared = 0;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
assert(cvthing.notbusy == PTHREAD_COND_INITIALIZER);
@@ -174,7 +174,7 @@ main()
abstime.tv_sec += 10;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
awoken = 0;
@@ -196,15 +196,15 @@ main()
*/
Sleep(1000);
- assert(pthread_mutex_lock(&cvthing.lock) == 0);
-
- cvthing.shared++;
-
/*
* Cancel one of the threads.
*/
- assert(pthread_cancel(t[3]) == 0);
- Sleep(500);
+ assert(pthread_cancel(t[1]) == 0);
+ assert(pthread_join(t[1], NULL) == 0);
+
+ assert(pthread_mutex_lock(&cvthing.lock) == 0);
+
+ cvthing.shared++;
/*
* Signal all remaining waiting threads.
@@ -214,9 +214,10 @@ main()
assert(pthread_mutex_unlock(&cvthing.lock) == 0);
/*
- * Give threads time to complete.
+ * Wait for all threads to complete.
*/
- Sleep(2000);
+ for (i = 2; i <= NUMTHREADS; i++)
+ assert(pthread_join(t[i], NULL) == 0);
/*
* Cleanup the CV.
diff --git a/tests/condvar8.c b/tests/condvar8.c
index 2ab67b3..572a5ac 100644
--- a/tests/condvar8.c
+++ b/tests/condvar8.c
@@ -158,7 +158,7 @@ main()
struct _timeb currSysTime;
const DWORD NANOSEC_PER_MILLISEC = 1000000;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
assert(cvthing.notbusy == PTHREAD_COND_INITIALIZER);
@@ -171,7 +171,7 @@ main()
abstime.tv_sec += 10;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
awoken = 0;
diff --git a/tests/condvar9.c b/tests/condvar9.c
index b24d71e..7eb9035 100644
--- a/tests/condvar9.c
+++ b/tests/condvar9.c
@@ -86,6 +86,7 @@ typedef struct bag_t_ bag_t;
struct bag_t_ {
int threadnum;
int started;
+ int finished;
/* Add more per-thread state variables here */
};
@@ -127,8 +128,8 @@ mythread(void * arg)
assert(pthread_mutex_lock(&cvthing.lock) == 0);
/*
- * pthread_cond_timedwait is a cancelation point and we
- * going to cancel one deliberately.
+ * pthread_cond_timedwait is a cancelation point and we're
+ * going to cancel some threads deliberately.
*/
#ifdef _MSC_VER
#pragma inline_depth(0)
@@ -146,6 +147,7 @@ mythread(void * arg)
assert(cvthing.shared > 0);
awoken++;
+ bag->finished = 1;
assert(pthread_mutex_unlock(&cvthing.lock) == 0);
@@ -164,7 +166,7 @@ main()
struct _timeb currSysTime;
const DWORD NANOSEC_PER_MILLISEC = 1000000;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
assert(cvthing.notbusy == PTHREAD_COND_INITIALIZER);
@@ -177,7 +179,7 @@ main()
abstime.tv_sec += 5;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
awoken = 0;
@@ -185,14 +187,15 @@ main()
first < NUMTHREADS;
first = last + 1, last = NUMTHREADS)
{
+ int ct;
+
assert(pthread_mutex_lock(&start_flag) == 0);
for (i = first; i <= last; i++)
{
- threadbag[i].started = 0;
+ threadbag[i].started = threadbag[i].finished = 0;
threadbag[i].threadnum = i;
assert(pthread_create(&t[i], NULL, mythread, (void *) &threadbag[i]) == 0);
- assert(pthread_detach(t[i]) == 0);
}
/*
@@ -207,46 +210,47 @@ main()
*/
Sleep(1000);
+ ct = (first + last) / 2;
+ assert(pthread_cancel(t[ct]) == 0);
+ canceledThreads++;
+ assert(pthread_join(t[ct], NULL) == 0);
+
assert(pthread_mutex_lock(&cvthing.lock) == 0);
cvthing.shared++;
- assert(pthread_cancel(t[(first + last) / 2]) == 0);
- canceledThreads++;
-
assert(pthread_cond_broadcast(&cvthing.notbusy) == 0);
assert(pthread_mutex_unlock(&cvthing.lock) == 0);
/*
- * Give threads time to complete.
+ * Standard check that all threads started - and wait for them to finish.
*/
- Sleep(1000);
- }
-
-
- /*
- * Standard check that all threads started.
- */
- for (i = 1; i <= NUMTHREADS; i++)
- {
- failed = !threadbag[i].started;
-
- if (failed)
- {
- fprintf(stderr, "Thread %d: started %d\n", i, threadbag[i].started);
+ for (i = first; i <= last; i++)
+ {
+ failed = !threadbag[i].started;
+
+ if (failed)
+ {
+ fprintf(stderr, "Thread %d: started %d\n", i, threadbag[i].started);
+ }
+ else
+ {
+ assert(pthread_join(t[i], NULL) == 0 || threadbag[i].finished == 0);
+// fprintf(stderr, "Thread %d: finished %d\n", i, threadbag[i].finished);
+ }
}
}
/*
* Cleanup the CV.
*/
-
+
assert(pthread_mutex_destroy(&cvthing.lock) == 0);
assert(cvthing.lock == NULL);
- assert(pthread_cond_destroy(&cvthing.notbusy) == 0);
+ assert_e(pthread_cond_destroy(&cvthing.notbusy), ==, 0);
assert(cvthing.notbusy == NULL);
diff --git a/tests/context1.c b/tests/context1.c
index 405f9da..fe36695 100644
--- a/tests/context1.c
+++ b/tests/context1.c
@@ -106,7 +106,7 @@ main()
assert(pthread_create(&t, NULL, func, NULL) == 0);
- hThread = t->threadH;
+ hThread = ((ptw32_thread_t *)t.p)->threadH;
Sleep(500);
diff --git a/tests/errno1.c b/tests/errno1.c
index f3231f5..9e6b46d 100644
--- a/tests/errno1.c
+++ b/tests/errno1.c
@@ -125,7 +125,7 @@ main()
pthread_mutex_lock(&stop_here);
errno = 0;
- assert((t[0] = pthread_self()) != NULL);
+ assert((t[0] = pthread_self()).p != NULL);
for (i = 1; i <= NUMTHREADS; i++)
{
diff --git a/tests/exception1.c b/tests/exception1.c
index 9b52636..0e63806 100644
--- a/tests/exception1.c
+++ b/tests/exception1.c
@@ -196,7 +196,7 @@ main()
pthread_t et[NUMTHREADS];
pthread_t ct[NUMTHREADS];
- assert((mt = pthread_self()) != NULL);
+ assert((mt = pthread_self()).p != NULL);
for (i = 0; i < NUMTHREADS; i++)
{
diff --git a/tests/exception2.c b/tests/exception2.c
index ed54a57..da835b5 100644
--- a/tests/exception2.c
+++ b/tests/exception2.c
@@ -129,7 +129,7 @@ main(int argc, char argv[])
exit(0);
}
- assert((mt = pthread_self()) != NULL);
+ assert((mt = pthread_self()).p != NULL);
for (i = 0; i < NUMTHREADS; i++)
{
diff --git a/tests/exception3.c b/tests/exception3.c
index af47705..428cf5b 100644
--- a/tests/exception3.c
+++ b/tests/exception3.c
@@ -158,7 +158,7 @@ main()
pthread_t et[NUMTHREADS];
pthread_mutexattr_t ma;
- assert((mt = pthread_self()) != NULL);
+ assert((mt = pthread_self()).p != NULL);
printf("See the notes inside of exception3.c re term_funcs.\n");
diff --git a/tests/exit5.c b/tests/exit5.c
index e896bf4..4812cac 100644
--- a/tests/exit5.c
+++ b/tests/exit5.c
@@ -109,7 +109,7 @@ Win32thread(void * arg)
assert(bag->started == 0);
bag->started = 1;
- assert((bag->self = pthread_self()) != NULL);
+ assert((bag->self = pthread_self()).p != NULL);
assert(pthread_kill(bag->self, 0) == 0);
/*
@@ -181,7 +181,7 @@ main()
result = 1;
#endif
- assert(threadbag[i].self != NULL && pthread_kill(threadbag[i].self, 0) == ESRCH);
+ assert(threadbag[i].self.p != NULL && pthread_kill(threadbag[i].self, 0) == ESRCH);
fail = (result != 1);
diff --git a/tests/reuse1.c b/tests/reuse1.c
index 2b74955..8e8ca54 100644
--- a/tests/reuse1.c
+++ b/tests/reuse1.c
@@ -82,7 +82,7 @@ static int washere = 0;
void * func(void * arg)
{
washere = 1;
- return (void *) 0;
+ return arg;
}
int
@@ -107,11 +107,16 @@ main()
for (i = 1; i < NUMTHREADS; i++)
{
washere = 0;
- assert(pthread_create(&t, &attr, func, NULL) == 0);
+ assert(pthread_create(&t, &attr, func, (void *) i) == 0);
pthread_join(t, &result);
- assert(result == 0);
+ assert((int) result == i);
assert(washere == 1);
- assert(t == last_t);
+ /* thread IDs should be unique */
+ assert(!pthread_equal(t, last_t));
+ /* thread struct pointers should be the same */
+ assert(t.p == last_t.p);
+ /* thread handle reuse counter should be different by one */
+ assert(t.x == last_t.x+1);
last_t = t;
}
diff --git a/tests/reuse2.c b/tests/reuse2.c
index 625611e..12aaa5f 100644
--- a/tests/reuse2.c
+++ b/tests/reuse2.c
@@ -53,7 +53,9 @@
* -
*
* Environment:
- * -
+ * - This test is implementation specific
+ * because it uses knowledge of internals that should be
+ * opaque to an application.
*
* Input:
* - None.
@@ -74,6 +76,9 @@
#include "test.h"
+/*
+ */
+
enum {
NUMTHREADS = 10000
};
@@ -95,11 +100,11 @@ main()
{
pthread_t t[NUMTHREADS];
pthread_attr_t attr;
- int i, j;
- int reuse,
- reuseMax = 0,
- reuseMin = NUMTHREADS,
- idTotal = 0;
+ int i;
+ unsigned int notUnique = 0,
+ totalHandles = 0,
+ reuseMax = 0,
+ reuseMin = NUMTHREADS;
assert(pthread_attr_init(&attr) == 0);
assert(pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_DETACHED) == 0);
@@ -120,43 +125,39 @@ main()
*/
for (i = 0; i < NUMTHREADS; i++)
{
- reuse = 0;
+ if (t[i].p != NULL)
+ {
+ unsigned int j, thisMax;
- if (t[i] == NULL)
- continue;
+ thisMax = t[i].x;
- for (j = 0; j < i; j++)
- {
- if (t[j] == t[i])
- {
- reuse++;
- t[j] = NULL;
- }
- }
- for (j = i + 1; j < NUMTHREADS; j++)
- {
- if (t[j] == t[i])
- {
- reuse++;
- t[j] = NULL;
- }
+ for (j = i+1; j < NUMTHREADS; j++)
+ if (t[i].p == t[j].p)
+ {
+ if (t[i].x == t[j].x)
+ notUnique++;
+ if (thisMax < t[j].x)
+ thisMax = t[j].x;
+ t[j].p = NULL;
+ }
+
+ if (reuseMin > thisMax)
+ reuseMin = thisMax;
+
+ if (reuseMax < thisMax)
+ reuseMax = thisMax;
}
- if (reuseMax < reuse)
- reuseMax = reuse;
- if (reuseMin > reuse)
- reuseMin = reuse;
}
for (i = 0; i < NUMTHREADS; i++)
- {
- if (t[i] != NULL)
- idTotal++;
- }
+ if (t[i].p != NULL)
+ totalHandles++;
printf("For %d total threads:\n", NUMTHREADS);
- printf("Reuse maximum = %d\n", reuseMax);
- printf("Reuse minimum = %d\n", reuseMin);
- printf("Total thread IDs allocated = %d\n", idTotal);
+ printf("Non-unique IDs = %d\n", notUnique);
+ printf("Reuse maximum = %d\n", reuseMax);
+ printf("Reuse minimum = %d\n", reuseMin);
+ printf("Total handles = %d\n", totalHandles);
return 0;
}
diff --git a/tests/self1.c b/tests/self1.c
index 5cffac8..347fa79 100644
--- a/tests/self1.c
+++ b/tests/self1.c
@@ -52,7 +52,11 @@ main(int argc, char * argv[])
* This should always succeed unless the system has no
* resources (memory) left.
*/
- assert(pthread_self() != NULL);
+ pthread_t self;
+
+ self = pthread_self();
+
+ assert(self.p != NULL);
return 0;
}
diff --git a/tests/self2.c b/tests/self2.c
index f8a68c9..93be0c7 100644
--- a/tests/self2.c
+++ b/tests/self2.c
@@ -64,14 +64,9 @@ main()
assert(pthread_create(&t, NULL, entry, NULL) == 0);
- Sleep(2000);
+ Sleep(100);
- /*
- * Not much more we can do here but bytewise compare t with
- * what pthread_self returned.
- */
- assert(t == me);
- assert(memcmp((const void *) t, (const void *) me, sizeof t) == 0);
+ assert(pthread_equal(t, me) != 0);
/* Success. */
return 0;
diff --git a/tests/semaphore3.c b/tests/semaphore3.c
index c6570f8..1b19494 100644
--- a/tests/semaphore3.c
+++ b/tests/semaphore3.c
@@ -80,9 +80,8 @@ sem_t s;
void *
thr (void * arg)
{
- assert(pthread_detach(pthread_self()) == 0);
assert(sem_wait(&s) == 0);
-
+ assert(pthread_detach(pthread_self()) == 0);
return NULL;
}
@@ -91,19 +90,21 @@ main()
{
int value = 0;
int i;
- pthread_t t[MAX_COUNT];
+ pthread_t t[MAX_COUNT+1];
- assert(sem_init(&s, PTHREAD_PROCESS_PRIVATE, 0) == 0);
+ assert(sem_init(&s, PTHREAD_PROCESS_PRIVATE, 0) == 0);
assert(sem_getvalue(&s, &value) == 0);
+// printf("Value = %d\n", value); fflush(stdout);
assert(value == 0);
-// printf("Value = %ld\n", value);
for (i = 1; i <= MAX_COUNT; i++)
{
- assert(pthread_create(&t[i-1], NULL, thr, NULL) == 0);
- sched_yield();
- assert(sem_getvalue(&s, &value) == 0);
-// printf("Value = %ld\n", value);
+ assert(pthread_create(&t[i], NULL, thr, NULL) == 0);
+ do {
+ sched_yield();
+ assert(sem_getvalue(&s, &value) == 0);
+ } while (value != -i);
+// printf("Value = %d\n", value); fflush(stdout);
assert(-value == i);
}
@@ -111,7 +112,7 @@ main()
{
assert(sem_post(&s) == 0);
assert(sem_getvalue(&s, &value) == 0);
-// printf("Value = %ld\n", value);
+// printf("Value = %d\n", value); fflush(stdout);
assert(-value == i);
}
diff --git a/tests/semaphore4.c b/tests/semaphore4.c
new file mode 100644
index 0000000..3ea0568
--- /dev/null
+++ b/tests/semaphore4.c
@@ -0,0 +1,138 @@
+/*
+ * File: semaphore4.c
+ *
+ *
+ * --------------------------------------------------------------------------
+ *
+ * Pthreads-win32 - POSIX Threads Library for Win32
+ * Copyright(C) 1998 John E. Bossom
+ * Copyright(C) 1999,2003 Pthreads-win32 contributors
+ *
+ * Contact Email: rpj@callisto.canberra.edu.au
+ *
+ * The current list of contributors is contained
+ * in the file CONTRIBUTORS included with the source
+ * code distribution. The list can also be seen at the
+ * following World Wide Web location:
+ * http://sources.redhat.com/pthreads-win32/contributors.html
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library in the file COPYING.LIB;
+ * if not, write to the Free Software Foundation, Inc.,
+ * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
+ *
+ * --------------------------------------------------------------------------
+ *
+ * Test Synopsis: Verify sem_getvalue returns the correct number of waiters
+ * after threads are cancelled.
+ * -
+ *
+ * Test Method (Validation or Falsification):
+ * - Validation
+ *
+ * Requirements Tested:
+ * -
+ *
+ * Features Tested:
+ * -
+ *
+ * Cases Tested:
+ * -
+ *
+ * Description:
+ * -
+ *
+ * Environment:
+ * -
+ *
+ * Input:
+ * - None.
+ *
+ * Output:
+ * - File name, Line number, and failed expression on failure.
+ * - No output on success.
+ *
+ * Assumptions:
+ * -
+ *
+ * Pass Criteria:
+ * - Process returns zero exit status.
+ *
+ * Fail Criteria:
+ * - Process returns non-zero exit status.
+ */
+
+#include "test.h"
+
+#define MAX_COUNT 100
+
+sem_t s;
+
+void *
+thr (void * arg)
+{
+ assert(sem_wait(&s) == 0);
+ return NULL;
+}
+
+int
+main()
+{
+ int value = 0;
+ int i;
+ pthread_t t[MAX_COUNT+1];
+
+ assert(sem_init(&s, PTHREAD_PROCESS_PRIVATE, 0) == 0);
+ assert(sem_getvalue(&s, &value) == 0);
+// printf("Value = %d\n", value); fflush(stdout);
+ assert(value == 0);
+
+ for (i = 1; i <= MAX_COUNT; i++)
+ {
+ assert(pthread_create(&t[i], NULL, thr, NULL) == 0);
+ do {
+ sched_yield();
+ assert(sem_getvalue(&s, &value) == 0);
+ } while (value != -i);
+// printf("Value = %d\n", value); fflush(stdout);
+ assert(-value == i);
+ }
+
+ assert(sem_getvalue(&s, &value) == 0);
+ assert(-value == MAX_COUNT);
+//printf("value = %d\n", -value); fflush(stdout);
+ assert(pthread_cancel(t[50]) == 0);
+ {
+ int result;
+ assert(pthread_join(t[50], (void **) &result) == 0);
+// printf("result = %d\n", result); fflush(stdout);
+ }
+ assert(sem_getvalue(&s, &value) == 0);
+//printf("value = %d\n", -value); fflush(stdout);
+ assert(-value == (MAX_COUNT - 1));
+
+ for (i = MAX_COUNT - 2; i >= 0; i--)
+ {
+ assert(sem_post(&s) == 0);
+ assert(sem_getvalue(&s, &value) == 0);
+// printf("Value = %d\n", value); fflush(stdout);
+ assert(-value == i);
+ }
+
+ for (i = 1; i <= MAX_COUNT; i++)
+ if (i != 50)
+ assert(pthread_join(t[i], NULL) == 0);
+
+ return 0;
+}
+
diff --git a/tests/semaphore4t.c b/tests/semaphore4t.c
new file mode 100644
index 0000000..bb0535c
--- /dev/null
+++ b/tests/semaphore4t.c
@@ -0,0 +1,132 @@
+/*
+ * File: semaphore4t.c
+ *
+ *
+ * --------------------------------------------------------------------------
+ *
+ * Pthreads-win32 - POSIX Threads Library for Win32
+ * Copyright(C) 1998 John E. Bossom
+ * Copyright(C) 1999,2003 Pthreads-win32 contributors
+ *
+ * Contact Email: rpj@callisto.canberra.edu.au
+ *
+ * The current list of contributors is contained
+ * in the file CONTRIBUTORS included with the source
+ * code distribution. The list can also be seen at the
+ * following World Wide Web location:
+ * http://sources.redhat.com/pthreads-win32/contributors.html
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library in the file COPYING.LIB;
+ * if not, write to the Free Software Foundation, Inc.,
+ * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
+ *
+ * --------------------------------------------------------------------------
+ *
+ * Test Synopsis: Verify sem_getvalue returns the correct number of waiters
+ * after threads are cancelled.
+ * -
+ *
+ * Test Method (Validation or Falsification):
+ * - Validation
+ *
+ * Requirements Tested:
+ * - sem_timedwait cancellation.
+ *
+ * Features Tested:
+ * -
+ *
+ * Cases Tested:
+ * -
+ *
+ * Description:
+ * -
+ *
+ * Environment:
+ * -
+ *
+ * Input:
+ * - None.
+ *
+ * Output:
+ * - File name, Line number, and failed expression on failure.
+ * - No output on success.
+ *
+ * Assumptions:
+ * -
+ *
+ * Pass Criteria:
+ * - Process returns zero exit status.
+ *
+ * Fail Criteria:
+ * - Process returns non-zero exit status.
+ */
+
+#include "test.h"
+
+#define MAX_COUNT 100
+
+sem_t s;
+
+void *
+thr (void * arg)
+{
+ assert(sem_timedwait(&s, NULL) == 0);
+ return NULL;
+}
+
+int
+main()
+{
+ int value = 0;
+ int i;
+ pthread_t t[MAX_COUNT+1];
+
+ assert(sem_init(&s, PTHREAD_PROCESS_PRIVATE, 0) == 0);
+ assert(sem_getvalue(&s, &value) == 0);
+// printf("Value = %d\n", value); fflush(stdout);
+ assert(value == 0);
+
+ for (i = 1; i <= MAX_COUNT; i++)
+ {
+ assert(pthread_create(&t[i], NULL, thr, NULL) == 0);
+ do {
+ sched_yield();
+ assert(sem_getvalue(&s, &value) == 0);
+ } while (value != -i);
+// printf("Value = %d\n", value); fflush(stdout);
+ assert(-value == i);
+ }
+
+ assert(sem_getvalue(&s, &value) == 0);
+ assert(-value == MAX_COUNT);
+ assert(pthread_cancel(t[50]) == 0);
+ assert(pthread_join(t[50], NULL) == 0);
+ assert(sem_getvalue(&s, &value) == 0);
+ assert(-value == MAX_COUNT - 1);
+
+ for (i = MAX_COUNT - 2; i >= 0; i--)
+ {
+ assert(sem_post(&s) == 0);
+ assert(sem_getvalue(&s, &value) == 0);
+// printf("Value = %d\n", value); fflush(stdout);
+ assert(-value == i);
+ }
+
+ for (i = 1; i <= MAX_COUNT; i++)
+ if (i != 50)
+ assert(pthread_join(t[i], NULL) == 0);
+
+ return 0;
+}
+
diff --git a/tests/sizes.c b/tests/sizes.c
index da7e115..6c077ad 100644
--- a/tests/sizes.c
+++ b/tests/sizes.c
@@ -8,7 +8,8 @@ main()
{
printf("Sizes of pthreads-win32 structs\n");
printf("-------------------------------\n");
- printf("%30s %4d\n", "pthread_t_", sizeof(struct pthread_t_));
+ printf("%30s %4d\n", "pthread_t", sizeof(pthread_t));
+ printf("%30s %4d\n", "ptw32_thread_t", sizeof(ptw32_thread_t));
printf("%30s %4d\n", "pthread_attr_t_", sizeof(struct pthread_attr_t_));
printf("%30s %4d\n", "sem_t_", sizeof(struct sem_t_));
printf("%30s %4d\n", "pthread_mutex_t_", sizeof(struct pthread_mutex_t_));
diff --git a/tests/test.h b/tests/test.h
index 072b797..b337674 100644
--- a/tests/test.h
+++ b/tests/test.h
@@ -45,6 +45,8 @@
#include <windows.h>
#include <stdio.h>
+#define PTW32_THREAD_NULL_ID {NULL,0}
+
char * error_string[] = {
"ZERO_or_EOK",
"EPERM",
@@ -122,6 +124,16 @@ char * error_string[] = {
(fprintf(stderr, "Assertion failed: (%s), file %s, line %d\n", \
#e, __FILE__, (int) __LINE__), exit(1), 0))
+int assertE;
+# define assert_e(e, o, r) \
+ (((assertE = e) o (r)) ? ((ASSERT_TRACE) ? fprintf(stderr, \
+ "Assertion succeeded: (%s), file %s, line %d\n", \
+ #e, __FILE__, (int) __LINE__), \
+ fflush(stderr) : \
+ 0) : \
+ (fprintf(stderr, "Assertion failed: (%s %s %s), file %s, line %d, error %s\n", \
+ #e,#o,#r, __FILE__, (int) __LINE__, error_string[assertE]), exit(1), 0))
+
#endif /* NDEBUG */
diff --git a/tests/valid2.c b/tests/valid2.c
index ad01b55..cef1020 100644
--- a/tests/valid2.c
+++ b/tests/valid2.c
@@ -76,15 +76,9 @@
int
main()
{
- pthread_t ptrToNull = NULL;
- /* This should be bigger than a pthread handle. */
- char corruptDummy[1000];
+ pthread_t NullThread = PTW32_THREAD_NULL_ID;
- memset(corruptDummy, 0x5A, sizeof(corruptDummy));
-
- assert(pthread_kill(NULL, 0) == ESRCH);
- assert(pthread_kill(ptrToNull, 0) == ESRCH);
- assert(pthread_kill((pthread_t) corruptDummy, 0) == ESRCH);
+ assert(pthread_kill(NullThread, 0) == ESRCH);
return 0;
}
diff --git a/w32_CancelableWait.c b/w32_CancelableWait.c
index a40fa4d..406bdbf 100644
--- a/w32_CancelableWait.c
+++ b/w32_CancelableWait.c
@@ -56,21 +56,25 @@ ptw32_cancelable_wait (HANDLE waitHandle, DWORD timeout)
{
int result;
pthread_t self;
+ ptw32_thread_t * sp;
HANDLE handles[2];
DWORD nHandles = 1;
DWORD status;
handles[0] = waitHandle;
- if ((self = pthread_self ()) != NULL)
+ self = pthread_self();
+ sp = (ptw32_thread_t *) self.p;
+
+ if (sp != NULL)
{
/*
* Get cancelEvent handle
*/
- if (self->cancelState == PTHREAD_CANCEL_ENABLE)
+ if (sp->cancelState == PTHREAD_CANCEL_ENABLE)
{
- if ((handles[1] = self->cancelEvent) != NULL)
+ if ((handles[1] = sp->cancelEvent) != NULL)
{
nHandles++;
}
@@ -83,7 +87,6 @@ ptw32_cancelable_wait (HANDLE waitHandle, DWORD timeout)
status = WaitForMultipleObjects (nHandles, handles, PTW32_FALSE, timeout);
-
switch (status - WAIT_OBJECT_0)
{
case 0:
@@ -105,23 +108,23 @@ ptw32_cancelable_wait (HANDLE waitHandle, DWORD timeout)
*/
ResetEvent (handles[1]);
- if (self != NULL)
+ if (sp != NULL)
{
/*
* Should handle POSIX and implicit POSIX threads..
* Make sure we haven't been async-canceled in the meantime.
*/
- (void) pthread_mutex_lock (&self->cancelLock);
- if (self->state < PThreadStateCanceling)
+ (void) pthread_mutex_lock (&sp->cancelLock);
+ if (sp->state < PThreadStateCanceling)
{
- self->state = PThreadStateCanceling;
- self->cancelState = PTHREAD_CANCEL_DISABLE;
- (void) pthread_mutex_unlock (&self->cancelLock);
+ sp->state = PThreadStateCanceling;
+ sp->cancelState = PTHREAD_CANCEL_DISABLE;
+ (void) pthread_mutex_unlock (&sp->cancelLock);
ptw32_throw (PTW32_EPS_CANCEL);
/* Never reached */
}
- (void) pthread_mutex_unlock (&self->cancelLock);
+ (void) pthread_mutex_unlock (&sp->cancelLock);
}
/* Should never get to here. */