diff options
99 files changed, 8095 insertions, 7377 deletions
@@ -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. @@ -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. @@ -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 = @@ -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)
@@ -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 ------------------- @@ -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 @@ -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. @@ -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. + + @@ -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 @@ -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) { @@ -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 */ @@ -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; @@ -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 } } @@ -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 */ } @@ -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 Binary files differnew file mode 100644 index 0000000..68ecb6d --- /dev/null +++ b/tests/Debug.ncb diff --git a/tests/Debug.opt b/tests/Debug.opt Binary files differnew file mode 100644 index 0000000..f263d4f --- /dev/null +++ b/tests/Debug.opt 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. */ |