Wed Feb 5 01:45:03 1997 Douglas C. Schmidt * ace/OS.i: Changed the Win32 implementation of all methods that call WaitForMultipleObjects() so that errno is set to WAIT_ABANDONED if the mutex is abandoned. This makes it possible for users to understand why ACE_OS methods that use synchronization operations fail. Thanks to Ivan Murphy for reporting this problem. * ace/IOStream.cpp (ACE_Streambuf): Moved the const initialization from the header file to the constructor since this is the only portable way to do this. Thanks to Michael Maxie for reporting this. * ace/OS.i: Updated the call to thr_setprio() for pthreads so that it first determines the existing scheduling parameters and then updates the priority. Thanks to Thilo Kielmann for help with this. * ace/config-osf1-4.0*.h: Updated ACE_NEEDS_HUGE_THREAD_STACKSIZE to be 1 mega to minimize porting surprises. Thanks to Thilo Kielmann for help with this. * tests/Naming_Test.cpp: Cleaned up a few minor things related to int i = 0 vs. size_t i = 0. * ace/SOCK_Dgram.cpp: Added a new macro called ACE_HAS_SOCKADDR_MSG_NAME, which is used on platforms (e.g., DG/UX) that require a cast to (struct sockaddr *) for the msg_name field of struct msghdr. Wed Feb 05 10:45:41 1997 David L. Levine * ace/IOStream.h: replaced __*manip with expanded types, because Sun C++ doesn't have those typedefs; and added complete prototypes for send, recv, send_n, and recv_n. * ace/IOStream.cpp: fixed "#define ACE_IOSTREAM_C" to keep g++ sane. * ace/Hash_Map_Manager.{h,cpp},Map_Manager.{h,cpp}: replaced DEFAULT_SIZE with ACE_DEFAULT_MAP_SIZE so that Naming_Test runs on SunC++ 4.1. * ace/High_Res_Timer.*: added optional scale factor. * examples/IOStream/{client,server}/Makefile: addd -g++ if building with gcc. * tests/Naming_Test.cpp: moved declaration of "i" out of for loop because it is used after the loop. * examples/ASX/UPIPE_Event_server/{Consumer,Supplier}_Router.cpp (open): return 1 if init () succeeds. There's a commented-out call to activate the objects that would (should?) return a value. * examples/IPC_SAP/FIFO_SAP/FIFO-client.cpp (main): declare n as ssize_t instead of size_t to avoid compiler warning about signed/unsigned comparison. Tue Feb 4 19:06:52 1997 Douglas C. Schmidt * ace/Signal.cpp: Fixed all the ACE_Sig_Handler[s] methods so that they now use ACE_Guard, rather than ACE_TSS_Guard, *unless* they are actually performing callbacks to user-code. Not only is this faster, but it also avoids thread-specific storage key leaks on platforms like Solaris! * ace/OS.cpp (thr_keyfree): Added a new #ifdef for ACE_HAS_THR_KEYDELETE, which is defined for UNIXWARE. This fixes a nasty problem with leaking thread-specific keys. Thanks to Nigel Owen for reporting this. * ace/Message_Block.cpp: Added a #include for "ace/Service_Config.h". * ace/Malloc.i: Fixed a few warnings by removing the names of unused parameters. * ace: Added many minor changes to ACE so that it will work on DG/UX. Thanks to Michael Maxie for all these. * ace/LSOCK_Stream.cpp: Applied the same fix as to LSOCK.cpp. * ace/LSOCK.cpp: Removed the cast of 0 to (char *) for msg_name since some platforms use (struct sockaddr *). By omitting the cast the compiler will figure it out correctly. Thanks to Michael Maxie for this. * ace/IOStream: Added a new template adapter for creating an iostream-like object using an ACE IPC Stream for the actual I/O. Thanks to James CE Johnson for this. * examples/Misc/test_iostreams.cpp: Added a test that illustrates how the new ACE_iostreams mechanism works. Thanks to James CE Johnson for this. * ace/Message_Block: Updated the ACE_Message_Block implementation to use ACE_Service_Config::alloc() if the user doesn't supply a customize version. This also means that we never delete the allocator_strategy_, so we can remove this flag and reduce the amount of storage by 4 bytes. Thanks to Irfan for noticing that we can do this. * ace/platform_dgux4_epc.h: Added a new platform config file for DG/UX. Thanks to Michael Maxie for this. * ace/config-dgux-4.11-epc.h: Added a new config file for DG/UX. Thanks to Michael Maxie for this. Tue Feb 04 18:26:01 1997 * ace/Service_Config.cpp (alloc): The implementation now creates a ACE_New_Allocator rather than a ACE_Allocator_Adaptor. If you still want the old behavior, simply provide your own custom allocator singleton for ACE_Service_Config. * ace/OS.i (mktemp): Added the missing ACE_INLINE directive. * ace/OS.h: ACE_HAS_INLINED_OSCALLS is now automatically defined when __ACE_INLINE__ is defined. This avoids mysterious linking problems that occur otherwise. * ace/Malloc: Added a virtual destructor to ACE_Allocator. Also added a new class ACE_New_Allocator which is a sub-class of ACE_Allocator. This class uses the new/delete operators to allocate and free up memory. Please note that the only methods that are supported are malloc() and free(). All other methods are no-ops. If you require more functionality, please use: ACE_Allocator_Adapter > This will allow you to use the added functionality of bind/find/etc. while using the new/delete operators. Mon Feb 3 15:35:24 1997 Douglas C. Schmidt * ace/config-aix-*.h: Added a (char *) cast to ACE_DEFAULT_BASE_ADDR. Thanks to John Cosby for reporting this. * ace/OS.cpp: Added a new set() method that sets the Time_Value from a double. Thanks to Brad Flood for this suggestion. * ace/OS: Changed ACE_OS::flock_t to ACE_OS::ace_flock_t to avoid a name clash on DG/UX. Thanks to Michael Maxie for reporting this. Tue Feb 4 03:02:52 1997 Irfan Pyarali * tests/Naming_Test.cpp: Added timing and made debug print outs optional. Also added randomization of the test data. * ace/OS.h: Fixed prototype for gethrtime () Mon Feb 03 21:00:56 1997 David L. Levine * ace/OS.h: define ACE_SCOPE_PROCESS and ACE_SCOPE_THREAD even if ACE_HAS_THREADS is not defined. Thanks to Michael Maxie for pointing this out. * examples/Threads/{process_mutex.cpp,process_semaphore.cpp, task_three.cpp,thread_manager.cpp,thread_specific.cpp} (main): added ACE_UNUSED_ARG (sa);. * examples/Threads/reader_writer.cpp: fixed declaration of "shared_data" to be "static volatile" instead of "volatile static", and removed unused variable "n". * examples/Threads/thread_pool.cpp: removed unused variable "result". Sun Feb 02 21:24:02 1997 David L. Levine * ace/OS.h: changed hrtime_t to ACE_hrtime_t. * ace/{config-win*.h,README}: added #define ACE_HAS_PENTIUM to enable gethrtime () support. * ace/High_Res_Time.{h,i,cpp}: removed platform-specific #ifdef's, instead added "unsupported ()" static member function; changed changed signature of elapsed_time to take time reference as argument instead of return value; added elapsed_time (ACE_Time_Value &) and elapsed_time_incr (ACE_Time_Value &). * examples/Thread/context_switch.cpp: use ACE_High_Res_Timer instead of the old Timer adapter. * ace/config-aix-4.1.x.h: fixed typo: #defind ACE_HAS_LONGLONG_T. * ace/config-aix-4.2.x.h: added #define ACE_HAS_LONGLONG_T. * ace/Thread_Priority.{h,cpp}: moved #include of OS.h from .cpp to .h file. Sun Feb 02 10:30:29 1997 David L. Levine * ace/Local_Name_Space.cpp: removed ACE_Map_Manager, ACE_Map_Iterator, and ACE_Map_Entry: they don't appear to be used and, without including their headers, this file won't compile. * ace/Hash_Map_Manager.cpp (Hash_Map_Manager (ACE_Allocator *)): reordered initializers to match declaration order. Sat Feb 1 15:11:28 1997 Douglas C. Schmidt * include/makeinclude/platform_aix4.2.GNU (LD): Removed the -qnoro flag since it is no longer needed. Thanks to Rob Jordan for reporting this. Sat Feb 1 03:25:21 1997 Irfan Pyarali * ace/SString.cpp: Added two methods to the wide string implementation for high-performance. Even though these methods are somewhat ugly and not the "OO" way of doing things, performance was at stake here. * ace/Registry_Name_Space.cpp (resolve): Fixed the implementation by correctly figuring out the size of the value in the Registry. This allows us to dynamically allocated the correct amount of memory. * ace/Registry: Fixed the constant MAX_CONTEXT_NAME_SIZE to be the maximum path length. * ace/Naming_Context.cpp: Fixed various memory leaks. * ace/Local_Name_Space_T.h: Replaced the simple-minded Map_Manager with the new (improved) Hash_Map_Manager. * ace/Local_Name_Space.cpp: Added new methods operator!= and hash required by the map. * ace/Hash_Map_Manager.cpp (ACE_Hash_Map_Manager): Added an extra constructor. * tests/Naming_Test.cpp: Removed memory leaks. Fri Jan 31 19:39:48 1997 David L. Levine * ace/Hash_Map_Manager.cpp (ctor): reordered initializers to match declaration order. * tests/Time_Value_Test.cpp: cast double argument to long because there no longer is a constructor that takes a double. * ace/Hash_Map_Manager.{h,cpp}: removed #include of ace/Hash_Map_Manager.i because it doesn't exist. Protected .cpp file against multiple includes. Moved #include of "ace/Malloc.h" from .h to .cpp file. * ace/Map_Manager.cpp (current_size, total_size): cast -1 to size_t to avoid warning about signed-to-unsigned conversion. * ace/OS.h: moved hrtime_t typedef for ACE_WIN32 so it doesn't get defined twice. * ace/OS.i (getservbyname): one last ACE_HAS_NONCONST_GETBY change: missed "lproto" to "proto". * bin/create_ace_build: added checks for existing symlinks, so now the script can be re-run safely at any time on a build tree to update it. Fri Jan 31 02:16:30 1997 Douglas C. Schmidt * ace/config-aix-*.h: Made the default value for ACE_DEFAULT_BASE_ADDR 0x80000000 to work around restrictions with AIX. Thanks to Rob Jordan for this. * examples/Map_Manager: Added a new example of the Hash_Map_Manager. * tests/Map_Manager: Integrated the Hash_Map_Manager into the Map_Manager test and greatly improved the test (e.g., by adding timing). * ace: Added a new implementation of the Map_Manager called Hash_Map_Manager. This, of course, uses hashing rather than linear search. The result should be a much faster implementation for many types of applications. Thu Jan 30 13:55:08 1997 Douglas C. Schmidt * ace/Map_Manager: Changed the return type of total_size() and current_size() to size_t. * ace/OS: Had to remove the ACE_Time_Value (double) constructor since it caused ambiguities with ACE_Time_Value (long). Thanks to Per Andersson for figuring this out. * include/makeinclude/platform_linux_lxpthread.GNU (CXX): Added -D_REENTRANT to make things work correctly in MT programs. Thanks to Marius for this fix. * ace/OS.h: Added several extra constructors for ACE_Time_Value so that it won't cause problems with the new ACE_Time_Value (double) constructor. * ace/Log_Record.cpp (ACE_Log_Record): Added an "L" to the 0 initial value used for time_stamp_. Thanks to Marius for reporting this. * ace: Added the new config-aix-4.2.x.h file, which supports the AIX4.2 platform. Thanks to Rob Jordan for this. * include/makeinclude/platform_aix4.2.GNU: Added support for the new AIX4.2 platform. Thanks to Rob Jordan for this. * ace/OS.h: It's now possible to override ACE_DEFAULT_BASE_ADDR in a config.h file. This is important for platforms that need to put shared memory segments at particular address ranges. Thanks to Rob Jordan for suggesting this. Thu Jan 30 16:01:37 1997 David L. Levine * ace/{config-aix-4.1.x.h,README}: added ACE_HAS_AIX_HIRES_TIMER, see change to gettimeofday in OS.i below; and ACE_HAS_LONGLONG_T. * ace/OS.i (gettimeofday, gethrtime): use read_real_time () if ACE_HAS_AIX_HIRES_TIMER. Thanks to Bob Olson for this suggestion. * ace/OS.h: define hrtime_t as long long if ACE_HAS_AIX_HIRES_TIMER. * ace/OS.{i,cpp}: uninlined ACE_OS::gethostbyname (), and ACE_OS::inet_ntoa () on VxWorks, because they have static char arrays. * ace/OS.i (gethostbyaddr, getservbyname, gethostbyaddr, gethostbyname, getservbyname_r): with ACE_HAS_NONCONST_GETBY, instead of copying the const argument, just cast it to (char *). Thanks to Michael Maxie for pointing out the problem with the nonstatic char arrays with sizes determined at runtime. * ace/High_Res_Timer.*: added ACE_HAS_AIX_HIRES_TIMER support. Also, renamed get_time () to elapsed_microseconds (). * examples/Threads/{Timer.h,context_switch_time.cpp}: ACE_High_Res_Timer::elapsed_microseconds () instead of ACE_High_Res_Timer::get_time (). * bin/create_ace_build: added this script: it creates an ACE build tree in directory "build/" below the current directory, which must be a ACE "top level" directory (such as $WRAPPER_ROOT). The build tree directory structure mirrors that of the ACE top level directory structure, except that instead of containing any plain files, it contains only symlinks to the files in the ACE top level structure. Thu Jan 30 06:57:50 1997 David L. Levine * ace/Thread_Priority.cpp (convert_to_os_priority): changed switch (priority_class) to switch (priority_class_). Thanks to Chris Lahey for pointing this out. Also, return -1 on other than Solaris, Win32, and VxWorks. Wed Jan 29 22:05:01 1997 Douglas C. Schmidt * tests/Naming_Test.cpp: Added the prefix "test_" to the bind(), find(), rebind(), and unbind() functions. This avoids name clashes with the bind() call from the socket API. Thanks to Marius for this info. * ace/OS.cpp (ACE_Time_Value): Added a new constructor that makes it possible to construct an ACE_Time_Value from a double. Thanks to Brad Flood for this. * ace/OS.i (thr_join): If ACE_HAS_THREADS is defined, but none of the threading APIs match (e.g., AIX) then we'll return ACE_NOTSUP_RETURN(-1). Thanks to Chris Lahey for help with this. * include/makeinclude/platform_aix.GNU: Made some changes to the type of dynamic linker and libraries used by AIX 4.1. Thanks to Chris Lahey for help with this. * ace/config-aix-4.1.x.h: Changed from BSD 4.3 to 4.4. Thanks to Chris Lahey for help with this. Wed Jan 29 09:03:48 1997 David L. Levine * ace/OS.h: moved ACE_id_t and ACE_pri_t typedefs out of the ACE_HAS_THREADS section. * ace/OS.cpp: fixed ACE_OS::set_sched_params on Win32 and VxWorks. * ace/Scheduling_Parameters.*, ace/Thread_Priority.*: updated comments. * examples/Threads/{context_switch_time.cpp,Timer.h,Makefile}: added context switch timing test. Currently only works on Suns because that's the only platform for which ACE has a high-resolution timer. * Makefile: added chmod of VERSION file in TIMESTAMP macro. Wed Jan 29 00:59:12 1997 Douglas C. Schmidt * include/makeinclude/platform_linux_lxpthread.GNU (CXX): Removed the -D_MIT_POSIX_PTHREADS symbol since we're using Xavier's threads instead. Thanks to Marius Kjeldahl for reporting this. * ace/config-linux-lxpthreads.h: Removed the _MIT_POSIX_PTHREADS symbol since we're using Xavier's threads instead. Thanks to Marius Kjeldahl for reporting this. Tue Jan 28 23:32:58 1997 Douglas C. Schmidt * ace/SString: Added hash() methods to each ACE String class so that we can use this with the new ACE_Hash_Set implementation of ACE_Map_Manager. * ace/ACE.h (ACE): Added a UNICODE version of hash_pjw() so that we can use it to hash ACE_WString. * ace/OS.i (thr_setprio): Fixed a typo that was left over from a previous change. Thanks to Fred LaBar for reporting this. * apps/Gateway/Gateway/{proxy_config,consumer_config}: Added comments to these config files so they are easier to understand. Thanks to Chuck Gehr for motivating this. * apps/Gateway/Peer/Peer.cpp: Changed the name from connection_id_ to proxy_id_, which is more accurate. * apps/Gateway/Peer/Peer.cpp: Added a destructor to Peer_Handler so that it can clean itself up properly when shutdown with a signal. This fixes a nasty bug. * ace/config-win32-msvc4.x.h (ACE_HAS_MFC): By default, ACE now enables ACE_HAS_MFC. This makes it possible to use AfxBeginThread() with the ACE_THR_AFX flag. * ace/OS.cpp: Fixed a typo in the call to ::AfxBeginThread() function. Thanks to Karlheinz for reporting this. * apps/Gateway/Peer/Peer.cpp: Removed all uses of the Map_Manager from the Peer. There's only ever one connection from a Gateway per-Peer, so what was this doing here anyway?! It was causing problems due to multiple deletions, so removing it improves robustness of the tests. * ace/OS.i: Commented out the "extern char **_sys_siglist" definition since this was conflicting with the definition in Solaris. If some platforms needs this, we'll need to add a specific #ifdef. Tue Jan 28 21:57:45 1997 David L. Levine * ace/OS.{h,cpp}: added ACE_OS::set_sched_params (). It permits platform-independent setting of thread priorities and priority classes. It is currently supported on Solaris, Win32, and VxWorks. * ace/Scheduling_Parameters.{h,i,cpp}, ace/Thread_Priority.{h,i,cpp}, ace/Makefile: Added these two classes to support ACE_OS::set_sched_params (). Tue Jan 28 12:42:39 1997 Irfan Pyarali * ace/ace.mdp: Added the "Using MFC in a Shared DLL" flags. * tests/test_config.h (ACE_INIT_LOG): Made this UNICODE safe. * tests/Pipe_Test.cpp (main): Added an iteration count to control the number of children created by the main process. * ace/Memory_Pool.cpp (init_acquire): cast this->minimum_bytes_ to size_t to avoid compiler warning if size_t is unsigned. * ace/OS.h: Added strncat to the OS class. Tue Jan 28 09:13:00 1997 David L. Levine * ace/ACE.cpp (ACE::ldfind), ace/Local_Name_Space_T.cpp, ace/Local_Token.cpp (ACE_TPQ_Entry::ACE_TPQ_Entry), ace/Log_Msg.cpp (ACE_Log_Msg::msg, ACE_Log_Msg::find), Memory_Pool.cpp (ACE_MMAP_Memory_Pool::ACE_MMAP_Memory_Pool), ace/Naming_Context.cpp (ACE_Naming_Context::open), ace/Parse_Node.cpp (ACE_Location_Node::open_handle), ace/Registry.cpp (ACE_Predefined_Naming_Contexts::is_local_host, ACE_Registry::Binding_Iterator::Object_Iteration::next_n, ACE_Registry::Binding_Iterator::Context_Iteration::next_n), tests/UPIPE_SAP_Test.cpp: changed "sizeof char" to "sizeof (char)" because g++ and Sun C++ (properly) choked on it. * ace/Local_Name_Space_T.cpp (create_manager_i): -- ::strncat () instead of ACE_OS::strcat () -- "lock_name_for_backing_store" instead of "lock_name_for_local_name_space" * ace/Naming_Context.cpp (both constructors): reordered initializers to match declaration order. * include/makeinclude/platform_sunos5_g++.GNU: removed -fno-strict-prototypes because it's not needed with SunOS 5.5.1. Mon Jan 27 21:33:00 1997 Irfan Pyarali * tests/Naming_Test.cpp (main): Made modifications to the test to make it more dynamic. Also made it work with UNICODE and ACE_Naming_Context. * ace/Local_Name_Space_T.cpp (create_manager_i): Added a better naming scheme for the locks. * ace: Changed the calculations of string lengths such that the size of the character type is carefully considered. The following were affected: ACE.cpp (ACE::ldfind) Local_Token.cpp (ACE_TPQ_Entry::ACE_TPQ_Entry) Log_Msg.cpp (ACE_Log_Msg::msg, ACE_Log_Msg::find) Memory_Pool.cpp (ACE_MMAP_Memory_Pool::ACE_MMAP_Memory_Pool) Naming_Context.cpp (ACE_Naming_Context::open) Parse_Node.cpp (ACE_Location_Node::open_handle) Registry.cpp (ACE_Predefined_Naming_Contexts::is_local_host, ACE_Registry::Binding_Iterator::Object_Iteration::next_n, ACE_Registry::Binding_Iterator::Context_Iteration::next_n) * ace/OS.h: Added UNICODE support for ACE_DEFAULT_LOCALNAME. * ace/Proactor.cpp: Redid "lost" changes to add UNICODE support to: /* Proactor */ ACE_Overlapped_File::ACE_Overlapped_File ACE_Overlapped_File::open * ace/Registry_Name_Space.h: Changed the order of #include and #define as that the correct set of MACROS are defined for the include files. * tests/MM_Shared_Memory_Test.cpp: Added UNICODE support. Mon Jan 27 09:16:03 1997 David L. Levine * include/makeinclude/platform_sunos5_sunc++.GNU, include/makeinclude/platform_sunos5_sunc++_orbix.GNU, ace/config-sunos5.5-sunc++-4.x-orbix.h ace/config-sunos5.5-sunc++-4.x.h: disabled exception handling by commenting out ACE_HAS_EXCEPTIONS in the config files and adding -noex to the platform files, because it causes trouble, at least on MP machines. Mon Jan 27 14:54:31 1997 Douglas C. Schmidt * ace/Memory_Pool.h: Added code to ACE_MMAP_Memory_Pool_Options to take a minimum_bytes parameter. This is needed to prevent mmap remapping when a large segment is requested. Thanks to Fred LaBar for this. Sun Jan 26 13:39:13 1997 Douglas C. Schmidt * ace/OS.i: Changed the typo sis_siglist to sys_siglist. Also, changed the logic from #if !defined (ACE_HAS_SYS_SIGLIST) to #if defined (ACE_HAS_SYS_SIGLIST) since this makes more sense... Thanks to Nanbor Wang for noticing this. Sat Jan 25 20:30:23 1997 David L. Levine * ace/config-sunos5.5-sunc++-4.1.h: commented out ACE_HAS_EXCEPTIONS #define because -noex is required with it, and that causes core dump during thread exit on multiprocessor UltraSparcs. * include/makeinclude/platform_sunos5_sunc++_4.1.GNU: added -noex back because of core dump during thread exit on multiprocessor UltraSparcs. * examples/ASX/Event_Server/Event_Server/Event_Analyzer.cpp, examples/ASX/UPIPE_Event_Server/Event_Analyzer.cpp (control): added "default" case to switch statement to eliminate compiler warnings about enumeration values not being handled in switch. * examples/ASX/UPIPE_Event_Server/event_server.cpp (consumer): removed unused variable "iter". Sat Jan 25 12:50:24 1997 Douglas C. Schmidt * ace/OS.i: Updated all the cond_timedwait() and cond_wait() methods so that they decrement the count of waiters by one even when errors occur. Thanks to Irfan for finding this. * ace/Map_Manager.cpp: Updated the ACE_Map_Iterator code so that it no longer crashes if the Map has been closed. Thanks to Slawomir Kuzniar for reporting this. Sat Jan 25 11:34:56 1997 Irfan Pyarali * ace/SOCK_Connector.cpp (complete): Added a sleep of zero seconds which allows non-blocking connects to work properly on NT. * ace/config-winnt-4.0-msvc4.x.h and config-win32-msvc4.x.h: Added ACE_HAS_BROKEN_NON_BLOCKING_CONNECTS. * ace/OS.i: Moved out all code for Time_Value from #if region. It is common to all. * ace/Reactor: Added helper routines to stop the VC++ compiler from bitching about destructors in exception handling code. * ace/ace.{mak,mdp}: Removed Time_Value.cpp from project. * tests/tests.{mak,mdp}: Corrected the creation of SOCK_Test and Message_Blob_Test executables and temporary files. Fri Jan 24 09:01:50 1997 David L. Levine * examples/Connection/misc/Connection_Handler.cpp: added specialization of ACE_Svc_Handler so that subproject will build with g++ without optimization. * apps/Gateway/Gateway/Proxy_Handler.cpp: added specializations of ACE_Svc_Handler and ACE_Svc_Handler so that subproject will build with g++ without optimization. * tests/Reactor_Exceptions_Test.cpp (main): added ACE_UNUSED_ARG for argc and argv when ACE_HAS_EXCEPTIONS is not defined. * performance-tests/Misc/test_mutex.cpp: made destructor of Thread_Mutex_Derived virtual because the class has other virtual functions. Thu Jan 23 11:51:45 1997 David L. Levine * ace/config-sunos5.5-g++.h: commented out ACE_HAS_EXCEPTIONS because -fhandle-exceptions is broken in g++ 2.7.2 (and is required in order to use exception handling); also, added ACE_HAS_INLINED_OSCALLS, if __ACE_INLINE__ is #defined. * ace/config-sunos5.5-sunc++-4.1.h: added ACE_HAS_INLINED_OSCALLS, if __ACE_INLINE__ is #defined. Wed Jan 22 15:04:47 1997 Douglas C. Schmidt * ace/OS.h: Moved the ACE_TIMER_SKEW macro to below the #inclusion of config.h and updated it so that it can be overridden in the config.h file if necessary. Thanks to Bob Olson for motivating this. * ace/OS.cpp: Fixed the implementation of ACE_Thread_ID::operator== so that it works correctly when thread ids and thread handles aren't ints. Thanks to Chuck Gehr for reporting this. * ace/OS.i (thr_getprio): Moved a stray #else into the right place. Thanks to Chuck Gehr for reporting this. * ace/OS.i (sema_post): this->count_++ should have been s->count_++. Thanks to Chuck Gehr for reporting this. * ace/Synch_T.cpp: Added an extern "C" definition for ACE_TSS_C_cleanup(). Thanks to Chuck Gehr for reporting this. * ace: Added a new #define called ACE_LACKS_SYS_NERR to the config files for MVS and VxWorks. This is used in OS.cpp as follows: #if defined (ACE_LACKS_SYS_NERR) int sys_nerr = ERRMAX + 1; #endif /* ACE_LACKS_SYS_NERR */ Thanks to Chuck Gehr for reporting this. * Removed all #inclusions of "ace/Time_Value.h." This file no longer contains anything besides a #include "ace/OS.h" and is only kept around due to backwards compatibility. * ace: Moved the contents of Time_Value.h into OS.h to reduce the confusion that arises from the #inclusion of "config.h" in Time_Value.h. Also merged the contents of Time_Value.{i,cpp} into OS.{i,cpp}. Thanks to the MDA guys for motivating this. * tests/Reactor_Exceptions_Test.cpp: Added a new test that exercises the Reactor with C++ exceptions. Thanks to Luca for the test. * ace/Reactor.cpp (handle_events): Added Win32 structured exception handling support to the ACE_Reactor::handle_events() method so that it correctly releases the token lock when Win32 SEH occur in callback routines. Thanks to Luca for reporting this problem. * ace: Added a new macro called ACE_HAS_EXCEPTIONS, which indicates that the compiler supports C++ exception handling. * include/makeinclude/platform_sunos5_sunc++*.GNU: Commented out the -noex flag for now. Hopefully, things will work correctly now. * ace/Thread_Manager.cpp (exit): Removed the final acquire() on ace_mon, which is no longer necessary since the Guard classes now keep track of whether they are "owned". Thanks to Karlheinz and Detlef for this fix. Wed Jan 22 09:05:31 1997 David L. Levine * Makefile,VERSION: modified top-level Makefile (and VERSION) to add a timestamp to the VERSION file if creating a release from the /project/adaptive/ACE_wrappers directory. If it causes problems, this will disable it: % make TIMESTAMP= release. * ace/Local_Tokens.cpp (dump): replaced "ACE_TSS" with "ACE_TPQ_ENTRY" so that ACE_NO_TSS_TOKENS is supported. * ace/Synch.h: fixed typo in comment header for ACE_Thread_Semaphore * ace/config-vxworks-ghs-1.8.h,config-vxworks5.2-g++.h, config-unixware-2.01-g++.h: protect ACE_HAS_INLINED_OSCALLS #defined with #if defined (__ACE_INLINE__). * apps/Gateway/Gateway/Event_Channel.cpp (open): added ACE_UNUSED_ARG (sig);. * examples/ASX/Event_Server/Event_Server/Peer_Router.h: declared a friend of class Peer_Router_Context because it has a private destructor, to eliminate compiler warning about not having any friends. * examples/ASX/Event_Server/Event_Server/Supplier_Router.cpp (open): added return 0 in "true" branch of if statement so that the function always returns a value. Tue Jan 21 09:19:44 1997 David L. Levine * tests/Reader_Writer_Test.cpp (reader): commented out unused variable "n"; also, commented out unused function print_usage_and_die. * tests/TSS_Test.cpp, Thread_Manager_Test.cpp (main): added ACE_UNUSED_ARG (sa);. * tests/Thread_Pool_Test.cpp (open): added return 0 at end. Tue Jan 21 11:00:12 1997 Douglas C. Schmidt * apps/Gateway/Gateway/Gateway.cpp (parse_args): There was a small bug in parse_args (latest beta), it mixed proxy and consumer conf files. Thanks to Per Andersson for reporting this. Mon Jan 20 13:29:28 1997 David L. Levine * ace/OS.i, ace/Task.h, and ace/Thread_Manager.h: changed a few occurrences of NULL to 0 (for other than Win32 platforms), primarily to avoid compiler warnings on VxWorks if ACE.h isn't #included before stddef.h. * netsvcs/clients/Naming/Dump_Restore/Dump_Restore.cpp (handle_input): replaced %d format with %hu for u_short arg. * netsvcs/clients/Tokens/deadlock/deadlock_detection_test.cpp (run_writer): initialized variable "t" to avoid compiler warning. * examples/Connection/misc/Connection_Handler.cpp (handle_timeout): added ACE_UNUSED_ARG (tv); also, added explicit int return type to definition of Connection_Handler::close (u_long). * examples/Threads/task_four.cpp (Invoker_Task::svc): moved declaration of "task" out of for loop because it is also used after the for loop. * netsvcs/lib/Name_Handler.cpp, Server_Logging_Handler.cpp, TS_Clerk_Handler.cpp, TS_Server_Handler.cpp, Token_Handler.cpp: added ACE_UNUSED_ARG (sig);. Sun Jan 19 11:10:49 1997 David L. Levine * include/makeinclude/platform_sunos5_sunc++.GNU, platform_sunos5_sunc++_4.1.GNU: added comment about compiler warnings with +w option. Also, added +w to platform_sunos5_sunc++_4.1.GNU. With the other changes in this ChangeLog entry, there no longer should be significant warnings from Sun C++ when building ACE. * examples/ASX/Event_Server/Event_Server/Peer_Router.cpp (Peer_Handler::put): added ACE_UNUSED_ARG (tv);. * examples/IPC_SAP/SPIPE_SAP/NPServer.cpp (main), netsvcs/clients/Tokens/invariant/invariant.cpp (main): commented out argc and argv. * examples/Mem_Map/file-reverse/file-reverse.cpp (main): removed unused variable "cp". * netsvcs/clients/Tokens/manual/manual.cpp (handle_input, handle_exception): added ACE_UNUSED_ARG (fd); and (create_proxy): added return statement at end: even though it should never be reached, it eliminates a compiler warning. * examples/OS/Process/process.cpp, examples/Reactor/Misc/test_reactors.cpp, netsvcs/clients/Tokens/deadlock/deadlock_detection_test.cpp, netsvcs/clients/Tokens/mutex/test_mutex.cpp, netsvcs/clients/Tokens/rw_lock/rw_locks.cpp, performance-tests/TTCP/ORBeline/ttcp_i.cpp, performance-tests/TTCP/Orbix/ttcp_i.cpp (main): return 0 instead of 42. Fri Jan 17 11:47:10 1997 Douglas C. Schmidt * ace/OS.h: Added an equivalent set of ACE_LSOCK_{STREAM,ACCEPTOR,CONNECTOR} macros to complement the ones for ACE_SOCK_*. Thanks to Gonzalo Diethelm for suggesting this. * ace/OS.cpp (ACE_Thread_Adapter): Added a new feature that allows logging features to be inherited by threads. It works with those OSs (e.g., Win32) that use ACE_Thread_Adapter class. At some point, we might do this for all the OSs. Thanks to Luca for this. * examples/Connection/misc/Connection_Handler.cpp: Added a new example that illustrates how to use the Acceptor pattern to create multiple threads, each running its own Reactor. * ace/Log_Msg.cpp (dump): Some of the variables in the dump() method have the wrong format specification type. These are: restart_ (%d), flags_ (%x), priority_mask_ (%d). They are now fixed, thanks to Luca. * tests/Priority_Task_Test.cpp (open): Add an ACE_ASSERT to make sure that the tasks activate() correctly. Thanks to David for finding this. * ace/config-linux-lxpthreads.h: Added #defines for ACE_HAS_AUTOMATIC_INIT_FINI and ACE_HAS_SVR4_DYNAMIC_LINKING. Thanks to Marius for this. * ace/Malloc.h: Changed the typedef of ACE_Malloc_Align to a macro called ACE_MALLOC_ALIGN to allow users to override this value in their config.h file. This is necessary for platforms that don't have "long" word alignments. Thanks to Fred LaBar for this suggestion. * ace/OS.h: Added an #ifdef check for ACE_HAS_IRIX62_THREADS and if so, we'll use the if (&a) /* null */; } form of the ACE_UNUSED_ARG macro to avoid unnecessary warnings. Thanks to Fred LaBar for testing this out. Fri Jan 17 16:33:25 1997 David L. Levine * ace/High_Res_Timer.*: added elapsed_time () and get_time (). * netsvcs/clients/Tokens/collection/collection.cpp, rw_locks.cpp: commented out unused variables to avoid compiler warnings; and return 0 from main (). * include/makeinclude/rules.local.GNU, netsvcs/lib/Makefile, examples/Service_Configurator/IPC-tests/server/Makefile: fixed ESOBUILD mechanism to work on all platforms, not just Linux. Thanks to Marius Kjeldahl for this fix. * ace/OS.i and OS.cpp: uninlined functions that use varargs, because they usually won't get inlined anyways. * ace/OS.cpp (thr_create): added #else clause to be sure that always sparam.sched_priority always get set to the priority argument. * netsvcs/clients/Naming/Client/Client_Test.cpp: commented out unused parameters and wrapped combined assignments/conditional values to avoid compiler (g++) warnings. * tests/Priority_Task_Test.cpp: added check for successful thread spawn, and ACE_NEW_THREAD to svc () so that messages get logged. Thu Jan 16 17:03:47 1997 Douglas C. Schmidt * include/makeinclude: Added support for shared object files on Linux. Thanks to Marius Kjeldahl for his help with this. * ace: Merged in the IRIX support, mainly the missing netdb reentrant functions. Thanks to Gonzalo Diethelm and Carlos O'Ryan for their help. * ace/Timer_{List,Heap,Queue}: Changed iterator() to iter() to avoid name clashes with STL (ugh). * ace/Timer_{Heap,List}.cpp: Added ACE_BUILD_DLL so that things will link. Thanks to John Morey for reporting this. * ace/Synch_T.h: Fixed: class ACE_Write_Guard: method: tryacquire_write so that it will call ...->lock_->tryacquire_write() and not ...->lock_->acquire_write(). Thanks to Alexandre Karev for this fix. Thu Jan 16 15:51:16 1997 David L. Levine * Dump.cpp (register_object): initialized "slot" so that g++ doesn't complain about uninitialized use. * Reactor.cpp (initialized): use ACE_GUARD_RETURN instead of ACE_GUARD because the function returns a value; (find): initialize "i" if handle is out of range so that g++ doesn't complain about uninitialized use. * Service_Repository.cpp (remove): removed unused variable "handle". * Thread_Manager.h: made destructor virtual now that spawn_i () is virtual. * Timer_Heap.cpp: changed some int loop indicies to size_t to make sure that comparisons are consistently signed or unsigned. * OS.h, config-vxworks*.h: removed ACE_HAS_GREENHILLS_SOCKETS ACE macro because the code it wrapped applies to all VxWorks compilers, not just GreenHills. Wed Jan 15 02:06:21 1997 Douglas C. Schmidt * ace/ACE.cpp (round_to_pagesize): If _SC_PAGESIZE is enabled then we set the ACE::pagesize_ variable using sysconf(2). Thanks to James Mansion for this suggestion. Wed Jan 15 02:06:21 1997 Douglas C. Schmidt * ace/Timer_Queue.cpp (expire): Changed a line of code from ACE_Timer_Queue_Iterator &iterator (this->iterator ()); to ACE_Timer_Queue_Iterator &iterator = this->iterator (); so that the Visual C++ compiler can grok it. Thanks to John Morey for reporting this. * ace/Reactor.cpp: Modified the close() method so that it can be called multiple times with no bad side-effects. * ace/Pipe.cpp: Modified the close() method so that it can be called multiple times with no bad side-effects. * ace/Reactor: Added an initialized() method that indicates whether the Reactor was successfully initialized. Also rearranged the code in ACE_Reactor::open() so that initialized_ is only set at the end of the method when everything succeeds. If things *don't* succeed, we call close() on the way out. Thanks to Luca for these suggestions. * ace/Reactor.cpp (close): When closing down the ACE_Reactor, we now set initialized_ back to 0 so that it can be reopened again later. * tests/Reader_Writer_Test.cpp: There were some problems with this test due to the fact that on many pthreads platforms pthread_t isn't an int (it's a struct). Thanks to Chuck Gehr for the fix. * ace/OS.i (exit): In WIN32 when the process exits through an ExitProcess() call, some process resources (e.g., file handles) are freed before the static variable destructors are called. This results in an application error when tracing is enabled. Removing the ACE_TRACE call in ~ACE_Event_Handler seems to fix this. Thanks to Luca for this fix. * ace/OS: Fixed the implementation of semaphores for the POSIX port (POSIX lacks semaphores). The trick was to add a "waiters" field, along with the semaphore count. Thanks to John Bossom for reporting this. * examples/Threads/wfmo.cpp: Added a new test program that seems to indicate that the Win32 WaitForMultipleObjects() function can be called in multiple threads, all of which wait on the same set of HANDLEs. Note that the dispatching of the threads is relatively "fair" (i.e., everyone gets a chance to process the various HANDLEs as they become active). Thanks to Ari Erev for suggesting this and providing the initial code. * ace/Thread_Manager.h: Made the spawn_i() method virtual and moved it into the protected section of the class so that it can be over-ridden by subclasses. Thanks to Tim Harrison for suggesting this. Tue Jan 14 15:57:55 1997 Douglas C. Schmidt * ace/OS.h: Changed the default size of ACE_DEFAULT_TIMERS to _SC_TIMER_MAX (which is typically around 40 or so). Thanks to Stuart Powell for suggesting this. * tests/Timer_Queue_Test.cpp (main): Added some additional logic so that we can test the performance of the preallocated and non-preallocated ACE_Timer_Heap. * ace/Timer_Heap: Added a new "preallocate nodes" mechanism to the ACE_Timer_Heap. If this is enabled by the constructor, then we preallocate as many ACE_Timer_Nodes as there are slots in the heap. This allows us to completely remove the need for dynamic memory allocation, which is important for real-time systems. * ace/Thread_Manager.cpp (remove_thr): Optimized the code just a tad by decrementing the current_count_ *first*. Thanks to Per Andersson for suggesting this. * ace/ReactorEx: Updated the implementation so that the handle_events() method now in fact *does* return the number of event handler's dispatched (i.e., the implementation matches the documentation...). Thanks to Hamutal Yanay for reporting this. * ace/Service_Repository.cpp (remove): Removed an extra call to ACE_OS::dlclose(). This is already being called in the destructor of ACE_Service_Record. Thanks to Karlheinz for reporting this. * man: Updated all the nroff and html documentation. * ace/Timer_Heap: Implemented the spiffy new timer cancellation algorithm that requires only O(log N) time to remove a timer from the ACE_Timer_Heap. This is about 50 to 100 times faster than using the ACE_Timer_List implementation. Tue Jan 14 21:01:36 1997 David L. Levine * ace/OS.h: Replace "Linux" preprocessor symbol with "linux". Mon Jan 13 17:57:50 1997 Douglas C. Schmidt * ace/Synch: Added lock() accessor methods to both ACE_Process_Semaphore and ACE_Process_Mutex. Thanks to Hamutal Yanay for suggesting this. * ace/config-mvs.h: Added a #define for ACE_HAS_GETPAGESIZE. Thanks to Chuck Gehr for this information. * ace/LSOCK_Stream.h: Added a new typedef for PEER_ADDR that is associated with ACE_UNIX_Addr. Thanks to Mark Rabotnikov for suggesting this. Mon Jan 13 22:24:09 1997 David L. Levine * ace/OS.h: Added #undefs of sigemptyset and sigfillset on Linux when __OPTIMIZE__ is enabled. Yes, the system header on Linux really #defines them when optimizing! Thanks to an anonymous ACE user for pointing this out. * tests/tests_config.h: Dropped ACE_MAX_ITERATIONS back to 10. Sun Jan 12 16:59:52 1997 Douglas C. Schmidt * ace/Timer_List.cpp (schedule): Cleanup the code so that (1) timer_id never == -1 (in order to distinguish it from error return values and (2) so that we avoid an unnecessary check for operator new failure. * ace/OS.h: Added a new #define called ACE_DEFAULT_MAX_TIMERS, which specifies the maximum number of ACE_Event_Handlers supported by ACE_Timer_Heap. * ace/Proactor.cpp (ACE_Proactor): Updated the allocation of ACE_Timer_Queue so that it really points to an ACE_Timer_List by default. Naturally, this can be overridden by users. * ace/ReactorEx.cpp (open): Updated the allocation of ACE_Timer_Queue so that it really points to an ACE_Timer_List by default. Naturally, this can be overridden by users. * ace/Reactor.cpp (open): Updated the allocation of ACE_Timer_Queue so that it really points to an ACE_Timer_List by default. Naturally, this can be overridden by users. * ace: Added two new classes called ACE_Timer_List and ACE_Timer_Heap, which derive from ACE_Timer_Queue. ACE_Timer_List is essentially the old behavior of ACE_Timer_Queue, i.e., it uses an implementation of a linked list of ACE_Timer_Nodes. ACE_Timer_Heap, on the other hand, uses a heap-based callout queue, which is more appropriate for certain real-time applications. * ace/Timer_Queue: Revised the ACE_Timer_Queue implementation so that it's now an abstract base class with most of the implementation details pushed into the ACE_Timer_List and ACE_Timer_Heap subclasses. * ace/Timer_Queue.h: Made all the ACE_Timer_Queue methods public so that we can override them in a subclass (e.g., the new ACE_Timer_Heap). * examples/Logger/Acceptor-server/server_loggerd.cpp: Fixed this example application so that it only uses a single Reactor (the REACTOR::instance()) rather than two Reactors. Thanks to Sandro Doro for reporting this. * netsvcs/lib/Logging_Strategy.cpp: Fixed the #include of fstream.h and iostream.h to use the '<' '>' notation rather than the '"' and '"' notation. This avoids problems with running "make depend". Thanks to Rino Simioni for reporting this fix. Sun Jan 12 18:20:49 1997 David L. Levine * ace/ACE.cpp (round_to_pagesize): fixed typo in reference to ACE::pagesize_. * ace/ACE.cpp: initialize ACE::pagesize_ to 0. * ace/config-irix*.h and config-win*.h: removed ACE_PAGE_SIZE definition because these platforms no longer use it. * tests/UNIX*.conf and Win32server.conf: replaced net_svcs with netsvcs. Sat Jan 11 15:11:28 1997 Douglas C. Schmidt * ace/OS.i (thr_setprio): The POSIX implementation of ACE_OS:thr_setprio did not set the priority. It calls: pthread_setschedparam( thr_id, &policy, ¶m ) and then does prio = param.sched_priority. instead of param.sched_priority = prio; This is now fixed. Thanks to John Bossom for reporting this. * ace/config-irix*.h: Modified all the IRIX config files to use the new ACE_HAS_GETPAGESIZE feature since this is necessary to differentiate between IRIX and IRIX64. Thanks to Fred LaBar for reporting this. * ace/ACE.cpp: Modified the round_to_pagesize() so that if a config.h file enables ACE_HAS_GETPAGESIZE then we use that function in preference to the value in config.h. Thanks to Fred LaBar for suggesting this. * ace/ACE.cpp: Modified the round_to_pagesize() so that on platforms that allow the page size to be obtained dynamically we'll get that rather than replying on what it set in the config.h file. We'll cache this value in a static variable to speedup searches. Thanks to James Mansion for suggesting this. * apps/Gateway/Gateway: Finished adding support for thread-safe reference counting to the uses of ACE_Message_Block. * examples/ASX/Event_Server/Event_Server/Peer_Router.cpp: Replaced the use of ACE_MT_SYNCH with ACE_SYNCH so that this will compile correctly on non-MT platforms. Fri Jan 10 12:38:42 1997 Douglas C. Schmidt * Added a COPYING file that explicitly states the copyright status of ACE. * ace/config-win32-msvc4.x.h and ace/config-winnt-4.0-msvc4.x.h: By default, we'll revert to *not* enabling ACE_HAS_TEMPLATE_TYPEDEFS for MSVC++ 4.2, even though this is well-support by the compiler since it breaks some application code that expects to use 3 template parameters for ACE_Svc_Handler<>. The "fix" is to use the ACE_SOCK_STREAM macros (rather than using the "ACE_SOCK_Stream, ACE_INET_Addr" tuple) in order to achieve portability in either case. Thanks to Karlheinz for pointing out the need for backwards compatibility. Wed Jan 8 01:00:05 1997 Douglas C. Schmidt * tests/Message_Block_Test.cpp: Enhanced the torture test of the ACE_Message_Block so that it stresses the reference counting implementation in a multi-threaded environment. Wed Jan 08 10:26:45 1997 David L. Levine * ace/Message_Block.cpp (~ACE_Data_Block): assigned 0 to allocator_strategry_ after deleting it; fixed template specializations (for g++) by using macros instead of class names; commented out template specializations because Service_Record.cpp already has them, and they should only appear once in the library in order to avoid link clashes * ace/Service_Object.cpp (ACE_Service_Object): reordered initializations in constructor * ace/Service_Record.cpp (ACE_Service_Record): reordered initializations in constructor * ace/config-linux-lxpthreads.h: added ACE_HAS_RECURSIVE_THR_EXIT_SEMANTICS, thanks to James CE Johnson for tracking down this problem and testing the fix. Tue Jan 7 13:03:25 1997 Douglas C. Schmidt * ace/Message_Block: Continued to improve the ACE_Message_Block/ACE_Data_Block implementation. * ace/Message_Block: Added a new constructor to ACE_Message_Block that takes an ACE_Data_Block * and "owns" it. Thanks to Tim for this idea. * ace/Log_Msg.cpp: Now that we've prevented signal handlers from occurring within critical sections of ACE_Log_Msg::log() we don't need to use the ACE_Recursive_Thread_Mutex anymore. Instead, we just need ACE_Thread_Mutex. * ace/Log_Msg.cpp (log): Added an ACE_Sig_Guard to the block in ACE_Log_Msg::log() that acquires the mutex that serializes output. This prevents nasty problems with recursive to ACE_Log_Msg::log() from within signal handlers. * ace/Service_Config.cpp (end_reactor_event_loop): Added a timeout of ACE_Time_Value::zero to the ACE_Reactor::notify() method when called in the ACE_Service_Config::end_reactor_event_loop(). This prevents the Reactor from blocking indefinitely if there's no longer a thread to receive from the notification pipe. * netsvcs/clients/Naming/Dump_Restore: Removed the vestigal nametest.cpp and nametest.h files. I'm not sure why they were still there, but they shouldn't have been! * ace/Synch.i (remove): this->owner_ should be set to -1 before calling this->release(). Thanks to Per Andersson for suggesting this. * ace/Thread_Manager.cpp (exit): Added a flag called "do_thr_exit" to the ACE_Thread_Control::exit() method. This controls whether we call ACE_OS::thr_exit() after removing the thread from the Thread Manager's internal table. This is necessary to work around pthread bugs on certain platforms (e.g., AIX and Linux). * ace/Synch.h: Removed the ACE_Recursive_Thread_Mutex::set_nesting() method since this isn't being used and is potentially dangerous. Thanks to Per Andersson for suggesting this. * ace/XtReactor: Updated the XtReactor so that it uses the next internal Reactor implementation. Thanks to Mark Rabotnikov for reporting this. * ace/Task.i: Updated the grp_id(int) method to cache the group id in the task and then set it in the Thread_Manager, if there is one. Thanks to Hamutal Yanay for suggesting this. * ace/Log_Record.h: Updated the ACE_Log_Record class to use the new ACE_MAXLOGMSGLEN macro. * ace/OS.h: Added a new #define called ACE_MAXLOGMSGLEN. By default this is 4k. However, you can reset it in the config.h file if you'd like to increase or decrease the size. Thanks to Anthony McConnell for suggesting this. * examples/Threads/tss[12].cpp: Tidied up the tss1.cpp and tss2.cpp examples and improved the comments. Thanks to Luca for suggesting this. * ace/Service_Object.cpp (fini): Added the "virtual" keyword to the new destructor of ACE_Service_Type. Thanks to David for reporting this. * ace/Message_Block.i: Rearranged the order of the ACE_Data_Block and ACE_Message_Block locking_strategy() methods so that they'd be defined as inline before being used. Thanks to David for reporting this. Mon Jan 6 16:46:06 1997 Douglas C. Schmidt * ace/Service_Manager.cpp (fini): Only try to remove the ACE_Service_Manager from the Reactor if it's actually been registered in the first place (i.e., if it's get_handle() method does not return ACE_INVALID_HANDLE). * netsvcs/servers/svc.conf: Changed the name of the net services .so file from libnet_svcs to netsvcs so that it will work correctly with our new "cross-platform" svc.conf ACE::ldfind() strategy. Thanks to Marius Kjeldahl for reporting this. * ace/Reactor.cpp (find): Added a strategic break and a check for handle_in_range(). Thanks to Tim for this help! * netsvcs/clients/Naming/Client/svc.conf: Removed the "lib" part from the executable shared object file names in the svc.conf file so that it will work with the new "cross-platform" svc.conf ACE::ldfind() strategy. Thanks to Marius Kjeldahl for reporting this. * ace/Service_Record.cpp (fini): Added an extra assignment to a temporary const void *obj to avoid problems with MSVC++ and it's cast mechanism. * ace/ReactorEx.h: Added a default parameter to ACE_ReactorEx::notify(). Thanks to Tim for noticing this. Sun Jan 5 10:48:23 1997 Douglas C. Schmidt * tests: Added a new test for ACE_Message_Block, which is called Message_Block_Test.cpp (no surprise there ;-)). This test exercises the reference counting mechanisms. * ace/Message_Block.h: Made the destructors for ACE_Message_Block and ACE_Data_Block virtual, so that these classes can be subclassed and ACE_Message_Blocks can delete these subclasses (either via release() or delete). Thanks to Tilo Christ for suggesting this. * ace/Message_Block.cpp: Modified the implementation of ACE_Message_Block/ACE_Data_Block so that we always create an allocator_strategy_. By default, this strategy will be ACE_Allocator_Adapter >; This greatly simplifies the code. * ace/Message_Block.h: It's now possible to assign a Message_Block's priority when you create it. * ace/Reactor: Sped things up some more by merging the Reactor's any_ready() and fill_in_ready() methods together. This reduces redundant computations. * ace/Reactor: Changed the name of ACE_Reactor::detach() to ACE_Reactor::remove_handler_i() and changed the name of ACE_Reactor::attach() to ACE_Reactor::register_handler_i(). This is more consistent with the other naming schemes. * ace/Reactor.cpp: Reworked the implementation of the Handler Repository's unbind() method in order to simplfy the logic. Also merged in all the Reactor's detach() functionality into the unbind() method in order to localize all the handler removal processing. * ace/Reactor[Ex].cpp: Added more descriptive errno entries for failed operations in the handler repository. * ace/Reactor.cpp: Modified the implementation of the Handler Repository's bind() method so that it reports an error if you try to register an Event_Handler for an invalid HANDLE. * ace/Reactor.cpp: Reimplemented the internals of the Reactor to prepare for the upcoming merge of the Reactor and ReactorEx to use a single code base. * ace/Timer_Queue.cpp: Modified the behavior of the expire() methods so that they return the number of ACE_Event_Handlers whose handle_timeout() method was called. Sat Jan 4 11:47:05 1997 Douglas C. Schmidt * ace/Reactor.h: Removed the #ifdefs for ACE_HAS_POLL. This will be recast as a strategy in subsequent implementations. * ace/Reactor.cpp: Added a new call, ACE_Handler_Repository::handle_in_range(), which checks to make sure that the HANDLE is within the range of 0 .. max_handlep1_. This factors out some redundant code. * ace/Reactor.cpp: Sped up the performance of the Reactor by removing redundant calls to ACE_Handler_Repository::invalid_handle() and ACE_Handler_Repository::find(). * tests/Reactors_Test.cpp: Added a test for the new Reactor::notify() method with timeouts. * examples/Reactor/Misc/notification.cpp (svc): Added a test for the new Reactor::notify() method with timeouts. * ace/Reactor[Ex]: Modified the Reactor and ReactorEx so that they use timed send() operations for their notify() methods. This makes it possible for users to bound the amount of time they're willing to wait to try and notify the Reactor. This is mostly useful to avoid infinite blocking if there is congestion in the Reactor's notification mechanism. * ace/ACE.cpp: Added a new timed send() method that uses the write() system call on UNIX and the socket send() call on Win32. This new method is used in the new ACE_Reactor::notify() method, which now supported timed notifications. * ace/Reactor.cpp (ACE_Handler_Repository): To simplify the code and reduce unnecessary #ifdefs, I've merged the max_handlep1_ and cur_size_ fields (which are now called max_handlep1_). Fri Jan 3 10:47:15 1997 Douglas C. Schmidt * ace: Replaced all uses of "delete mb" with mb->release (); * ace/Stream_Modules.cpp: Replaced the use of explicit bit twiddling with the ACE_BIT* macros. * ace/Message_Block.cpp: Make sure that we use the allocator_strategy_ to create the memory for the reference count since this may need to go into shared memory if that's the memory pool where the Message_Block allocations are coming from. * ace/OS.h: Added two new macros, ACE_ALLOCATOR_RETURN and ACE_ALLOCATOR, which are similar to ACE_NEW_RETURN and ACE_NEW, except that these * ace/Message_Block.cpp (release): Make sure to "delete this" outside the scope of the locking_strategy_. * ace/Service_Object.cpp: Added a destructor to ACE_Service_Type. Thanks to Per.Andersson@hfera.ericsson.se (Per Andersson) for suggesting this. * ace/Service_Object.i: Be smarter about how we reassign the name_ pointer, i.e., delete the old one and make a copy. Thanks to Per.Andersson@hfera.ericsson.se (Per Andersson) for reporting this. * ace/Module.cpp (open): Rearranged the assignments to reader_q->mod_ and writer_q->mod_ so that we don't try to initialize through NULL pointers. Thanks to Per.Andersson@hfera.ericsson.se (Per Andersson) for reporting this. * ace/Service_Record.cpp (ACE_Service_Record): Initialized name_ to NULL so that the following change works correctly now. Thanks to Per.Andersson@hfera.ericsson.se (Per Andersson) for reporting this. * ace/Service_Record.i (name): Make sure to delete [] (char *) this->name_ before allocating a new one. Thanks to Per.Andersson@hfera.ericsson.se (Per Andersson) for reporting this. * ace/Message_Block: Reworked the reference counting implemention so that reference counts are shared correctly amongst their various owners. This requires making a deep copy the "header" portion, but a shallow copy of the "data." * ace/Message_Block.cpp (ACE_Message_Block): Updated all three ACE_Message_Block constructors so that they all call the init() method. This centralizes all the initialization logic in one place. Thu Jan 2 00:42:21 1997 Douglas C. Schmidt * ace/Message_Block.cpp (ACE_Message_Block): Make sure to set the cont_ field to 0 after "releasing" it so that we don't mistakenly think it's still around later on. This problem arose in the ACE_Message_Queue::close() method, which was trying to count the number of bytes being freed. * ace/Message_Queue.cpp (close): Fixed a subtle bug where we weren't actually deleting messages from the ACE_Message_Queue::close() routine. This should work now... * ace/Message_Queue.cpp (close): Replaced the use of "delete mb" with "mb->release()" since the Message_Blocks are now reference counted. * ace/Message_Block: Enhanced the reference counting scheme so that you can increment and decrement the count by an arbitrary amount. This is particular useful when you know you'll be sending the same Message_Block to N consumers. * ace/Singleton: The dump() must be used same as instance() (without supplying an object) so it must be declarated *static*, i.e., static void dump (void); Thanks to Sandro Doro for reporting this. * examples/ASX/Event_Server: Completely rewrote and retested the ACE Event Server example. The new code is *much* easier to understand, has many more comments, is more robust, and compiles much faster since I removed many of the templates. * examples/ASX/Event_Server/Tranceiver/tranceiver.cpp: Fixed the tranceiver so that it shuts down correctly when the Event_Server exits. * examples/Connection/non_blocking/CPP-connector.cpp: Fixed a problem where we were trying to select() on ACE_STDIN on Win32. Naturally, this doesn't work, so we used the ACE::register_stdin_handler() to fix this. Thanks to Samuel_Bercovici for reporting this. * examples/ASX/Event_Server/Event_Server/Options: Changed the Options class to be a Singleton... * ace/Task.h: Added "const" method qualifier to accessor methods like is_reader() and is_writer(). Wed Jan 1 00:10:47 1997 Douglas C. Schmidt * ace/Message_Block: Added a new static release() method that behaves like the non-static method release(), except that it checks if is 0. This is similar to CORBA::release(), which is useful if you want to eliminate lots of checks for NULL pointers before calling release() on them. * Updated all places in ACE where we use ACE_Event_Handler::READ_MASK when we should be using ACE_Event_Handler::ACCEPT_MASK. * examples/Service_Configurator/IPC-tests/server/Handle_Thr_Stream.cpp: Changed the name of the acceptor class from Handle_Thr_Stream to Handle_Thr_Acceptor, which is more accurate. * ace/Reactor: Since we've now got an ACE_Event_Handler::ACCEPT_MASK (intended primarily to accept connections asynchronously using Win32 overlapped I/O) we can now use this with the ACE_Reactor, as well. In particular, any time that we are registering a "passive-mode" socket acceptor we can now specify ACE_Event_Handler::ACCEPT_MASK. This is much more intuitive than using the READ_MASK (which never made any sense...). The ACE_Reactor now treats the ACCEPT_MASK as a READ_MASK internally, so you can still use READ_MASK if you'd like (i.e., this change doesn't break any existing code). * ace/Event_Handler: Changed "get_priority" and "set_priority" to simply "priority(void)" and "priority(int)" to be more consistent with other parts of ACE. * apps/Gateway/Gateway: Updated the Gateway so that the concurrency strategies can now be specified on the command-line (or in the svc.conf file), rather than being determined at compile-time. This is much more flexible. See the ./apps/Gateway/Gateway/README file for details. * apps/Gateway/Gateway: Verified that all the multi-threading strategies still work. * ace/Message_Block.cpp (ACE_Message_Block): Make sure that we "release" the continuation field, rather than delete it, since its reference count may be > 1. * ace/Log_Msg.cpp (log): Change the abort_prog logic a bit so that we always print a message to stderr if we're exiting, regardless... Thanks to David Levine for pointing this out. * apps/Gateway/Gateway: Moved all of the configuration file parsing logic *outside* of the Event_Channel into the Gateway class so that we wouldn't have unnecessary dependencies. * apps/Gateway/Gateway: Redesigned the Gateway so that the Proxy_Handlers (i.e., the Consumer_Proxy and Supplier_Proxy) most of their work to the Event_Channel. This "lightweight proxy" design is an improvement since it is now possible to emulate the COS Event Channel semantics within the Event_Channel "kernel." * Happy new year! Let's start a new ChangeLog to celebrate the new year.