| blob | 429e5681fdc24750351ef066a485f4a47563f2b7 |
1 Sun Dec 31 01:02:01 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3 * ace/OS (mmap): Added a new parameter to the argument list for
4 mmap(). This parameter is an ACE_HANDLE * and is used in the
5 Win32 implementation, which requires CreateFileMapping to open a
6 new HANDLE. This handle is passed back to the caller, which is
7 then responsible for freeing it... I'll probably need to modify
8 the ACE_Mem_Map class to manage this new handle for the WIN32
9 version.
11 * ace: Updated all the SPIPE files in preparation to porting them
12 to Windows NT, where they will be implemented using Win32 Named
13 Pipes!
15 * ace/Typed_SV_Message: Fixed a mysterious braino where the method
16 definitions for Typed_SV_Message were actually within
17 Typed_SV_Message_Queue.cpp. I'm not sure why this happened, but
18 it's clearly been this way for eons... Also, changed the name
19 of the method "max" to "max_size" to avoid conflicts with macros
20 on Windows NT...
22 * ace: Added a new config symbol: ACE_HAS_SYSV_IPC. If this is
23 *not* set (e.g., Win32) then don't try to compile any of the
24 System V IPC mechanisms (e.g., shared memory, message queues,
25 semaphores).
27 * ace/Malloc: Reimplemented the Malloc-family to factor out common
28 code. Also added a new method called "trybind" which atomically
29 tries to bind a new void *pointer to a char *name. If there is
30 no name with the same value, the bind succeeds. Otherwise, the
31 bind "fails", but returns the existing pointer.
33 * ace/Map_Manager: Reimplemented the Map_Manager to factor out
34 common code. Also added a new method called "trybind" which
35 atomically tries to bind a new int_id to an ext_id. If there is
36 no ext_id with the same name, the bind succeeds. Otherwise, the
37 bind "fails", but returns the existing int_id.
39 * ace/Memory_Pool.cpp: Commented out all the diagnostic messages
40 in the various ACE_*_Memory_Pools. Now that we're actually
41 using this within ACE, they were distracting.
43 * ace/CORBA_Handler.cpp (ACE_ST_CORBA_Handler): Gave iterations_
44 an initial value of "5" for backwards compatibility. Thanks to
45 Irfan for doing this.
47 * ace/Map_Manager.cpp (ACE_Map_Manager): Fixed some more braino in
48 the constructors of Map_Manager. During the recent changes I'd
49 forgotten to initialize the Map_Manager's instance variables to
50 0. Thanks to Irfan for finding this out the hard way... ;-)
52 * ace/OS.i (gettimeofday): Added a WIN32 implementation of
53 gettimeofday. This is pretty nasty, but it should centralize
54 all of this processing in one place in ACE!
56 Sat Dec 30 00:50:20 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
58 * ace/Timer_Queue: Removed the current_time() method from the
59 Timer_Queue class. All uses will be replaced with calls to
60 ACE_OS::gettimeofday(). Not only does this factor out more
61 code, but it is also reentrant!
63 * ace/ACE.cpp (handle_timed_accept): Consolidated code so that
64 select() is always used, rather than poll().
66 * ace: Generalized the ACE file locking wrappers (both at the
67 Synch-level and the ACE_OS class level) to enable them to lock
68 arbitrary regions (aka "records") of a file. Before, they could
69 only lock the entire file (which remains the default behavior).
70 This new behavior works for both NT and UNIX.
72 * ace/Synch: Removed the definition and use of the
73 ACE_SYNCH_ERROR_RETURN macro from all the ACE Synch* methods.
74 This behavior has been integrated into the ACE_OS class instead
75 since it cleans up lots of code that would otherwise have been
76 hard to write.
78 * ace/INET_Addr.cpp (get_host_name): Fixed a bug that arose by
79 trying to use the reentrant version of gethostbyaddr all the
80 time. In order to fix this, I've added a new get_host_name()
81 method that allows the caller to supply a buffer for the
82 hostname.
84 * ace: Cleaned up all the POSIX_TIMER junk that permeated ACE.
85 Now all platforms will have the POSIX timer structures (e.g.,
86 timestruc_t and timespec), though only those that actually
87 support nano-second timing will be able to do anything useful
88 with this... As part of this cleanup, the poll, select, and
89 cond_timedwait method in ACE_OS now take ACE_Time_Value * rather
90 than int, timeval *, and timestruc_t, respectively. This
91 *greatly* cleans up the clutter throughout the rest of ACE.
93 * ace/ACE: Added two new methods called tv2msec and msec2tv that
94 converts ACE_Time_Value format into millisecond format and vice
95 versa. This will make it easier to port code between NT and
96 UNIX.
98 * ace/Reactor: Removed the code in the Reactor that previously
99 allowed it to be resized after it had been initialized. This
100 code no longer was used due to the recent enhancements that
101 ensure the Reactor is only initialized once per-process.
103 * ace/Reactor: Modified the Reactor so that it now compiles on NT.
104 It doesn't work yet, but at least it compiles!
106 * ace/Reactor: Moved the "max3" method from the Reactor to OS.h,
107 where it has become a template function usable throughout ACE.
109 * ace/OS.cpp: Added extern "C" before the NT implementations of
110 readv() and writev().
112 * ace/OS.i: Fixed the declaration of ::readv() and ::writev() for
113 cases where the OS doesn't provide them natively.
115 * ace/Event_Handler: Modified handle_signal() to take an int
116 rather than an ACE_HANDLE. The old way broke under NT...
118 * ace: Made lots of minor changes to resolve signed int
119 vs. unsigned int mismatches. Thanks to the MSVC++ 4.0 compiler
120 for pointing these out. At least it's good for
121 something... ;-).
123 Fri Dec 29 15:01:10 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
125 * ace/ACE.cpp (handle_timed_accept): Reimplemented the core
126 select() loop to make the code portable to Win32.
128 * ace/Map_Manager.cpp (ACE_Map_Manager): Fixed a minor bug caused
129 by recent changes (forgot to initialize max_size_ in the
130 constructor...).
132 * Updated all the manual pages to reflect the latest changes.
134 * ace/ACE.cpp: Fixed up the ACE class so that all of the methods
135 should compile on both Windows NT and UNIX!
137 * ace/OS: Moved the get_file_size() method from class ACE and put
138 it into class ACE_OS, where it's been renamed to filesize().
139 This is really more of an "OS" function, e.g., it's supported on
140 Win32 directly.
142 * ace/OS.cpp: added readv() and writev() implementations for those
143 platforms (e.g., Windows NT) that require them.
145 * ace/Makefile: Added a new regex to the Svc_Conf_l.cpp target
146 that will substitute <stdio> for <unistd>. This should solve a
147 portability problem with Windows NT...
149 * ace/Malloc: Modified the sync() and protect() methods to use
150 void * rather than char *. This is now consistent with the OS
151 classes.
153 * ace/OS: Changed things so that all the mmap() related OS
154 wrappers use void * in their external interface. Any
155 discrepancies between platforms are now dealt with internally,
156 in the implementation of the methods. Thanks to Chris Lahey
157 (clahey@ix.netcom.com) for suggesting this.
159 * ace/Local_Name_Space: Modified how ACE_NS_String is used to
160 instantiate ACE_Map_Manager in order to work around problems
161 with AIX C++. Thanks to Chris Lahey (clahey@ix.netcom.com) for
162 reporting this.
164 * ace/Synch.i: Added a definition for ACE_RW_Mutex::acquire().
166 Fri Dec 29 00:34:16 1995 Chris Lahey (clahey@ix.netcom.com)
168 * ace/OS.h: Changed the #ifdef for ACE_HAS_COMPLEX_LOCKS to
169 ACE_HAS_COMPLEX_LOCK.
171 Changed prototypes for ACE_OS:mprotect(), msync(), and munmap()
172 to accept ACE_MMAP_TYPE for their first arg. The AIX versions
173 of these calls all expect void * for their pointer variable.
174 Note that madvise() takes a caddr_t for its 1st arg and
175 therefore does not need to be altered. (See related changes in
176 OS.i and Mem_Map.i)
178 Changed prototype for ACE_OS::getsockname(), arg3 to int * (See
179 related change in OS.i)
181 Changed prototype for ACE_OS::setsockopt(), arg4 to const char *
182 and arg 5 to int (See related change in OS.i)
184 * ace/OS.i: Changed function definition for ACE_OS::mprotect(),
185 msync(), and munmap() to accept ACE_MMAP_TYPE as their 1st arg.
187 The following changes relate to the inconsistencies in the
188 sockets calls on AIX. I thought it would make more sense to
189 leave the public interfaces alone and then handle the
190 differences internally. That way it won't break any exisiting
191 code.
193 Changed function definition for ACE_OS::getsockname() to accept
194 arg3 as int *. Made similar changes in ACE_OS::setsockopt() -
195 change definition to accept const char * for arg4 and int for
196 arg5. Add the following code at the start of the function:
197 Also made changes to ACE_OS::recvfrom() - change arg 6 back so it is
198 int *.
200 * ace/Mem_Map.i: In ACE_Mem_Map::advise(), cast 1st arg to
201 ACE_OS:madvise() to (caddr_t).
203 * ace/Memory_Pool.cpp: In ACE_MMAP_Memory_Pool::acquire(), make
204 cast to char * in the return statement. In
205 ACE_MMAP_Memory_Pool::handle_signal(), also make same cast in
206 the ACE_DEBUG statement near the top of the function.
208 * ace/Reactor.cpp and ace/Reactor.h: ACE_Reactor::handler_i()
209 doesn't return properly if *eh != 0. Remove last "else" so
210 function always returns 0 if handler != 0.
212 In ACE_Reactor::owner(), we need to change signature since
213 thread_t is not an integer. This has been changed to int
214 ACE_Reactor::owner(thread_t *t_id); The last two lines have been
215 changed to: *t_id = this->owner_; return 0;
217 * ace/Synch.cpp: In ACE_Condition_Mutex::wait(), the call to
218 cond_timedwait() does not exist, but ACE_OS::cond_timedwait()
219 does exist. It takes the same arguments.
221 * ace/Synch.h and ace/Synch_T.h: At the end of both files, before
222 the #define INLINE check for this define's existence to avoid
223 causing compilers to choke.
225 * ace/Mem_Map.i: In ACE_Mem_Map::operator(), the return statement
226 looked very odd. In the assignment to addr, why is the ", 0" in
227 the statement?
229 Fri Dec 29 00:34:16 1995 Douglas C. Schmidt (schmidt@lambada.cs.wustl.edu)
231 * ace: Fixed up many minor problems with the various types of
232 Shared_Memory wrappers. This mostly entailed changing from char
233 * to void * to be more consistent with the other parts of ACE.
234 In addition, I now only include the *.i files if __INLINE__ is
235 enabled.
237 * ace/Memory_Pool.cpp (map_file): Fixed a minor bug by enabling
238 MAP_FIXED if we're trying to map a file using a fixed address.
240 Thu Dec 28 18:39:16 1995 Douglas C. Schmidt (schmidt@merengue.cs.wustl.edu)
242 * ace/Local_Name_Space.cpp: Finished implementing changes to the
243 ACE_Malloc stuff and the Local_Name_Space that Irfan had begun
244 last semester.
246 * ace/SString.cpp: Added new optimizations to the assignment
247 operators of the ACE_[WCS]String classes to avoid
248 freeing/allocating memory if the current size is large enough.
250 * ace/SString: Added copy constructors...
252 * ace/Local_Name_Space.cpp: Modified the implementation of bind()
253 to avoid a race condition.
255 * ace/Map_Manager.cpp (bind): Changed the semantics of bind() so
256 that callers can control whether or not INT_IDs are reassigned.
257 Also changed things so that callers can get back the existing
258 value if they choose not to reassign.
260 * ace/SString.cpp: Removed the "P" (persistent) versions of all
261 these classes and merged them into the default versions. The
262 persistence is now determined by the type of allocator passed in
263 as a parameter to the constructor (or by using the "default"
264 allocator that is accessible via
265 ACE_Service_Config::allocator()).
267 * ace/Map_Manager.cpp: Removed the "P" (persistent) versions of
268 all these classes and merged them into the default versions.
269 The persistence is now determined by the type of allocator
270 passed in as a parameter to the constructor (or by using the
271 "default" allocator that is accessible via
272 ACE_Service_Config::allocator()).
274 * ace/Malloc_T: Modified the bind() algorithm so that it will
275 allow the caller to determine whether duplicates are allowed or
276 not... Also changed things so that callers can get back the
277 existing value if they choose not to allow duplicates. Irfan
278 questions the sanity of this... ;-)
280 * ace/Synch: To be consistent, added acquire_write and
281 acquire_read methods to ACE_Mutex. These are useful for cases
282 where we are going to be parameterizing classes like ACE_Malloc
283 with some type of mutex wrapper (e.g., ACE_RW_Mutex, ACE_Mutex).
285 * ace/Malloc_T.cpp (calloc): Added a new method called "calloc"
286 that will not only allocate the memory dynamically, but will
287 also give it an initial value (e.g., '\0').
289 Thu Dec 28 01:10:43 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
291 * ace/OS.i: Modified the implementation of the TLI methods so that
292 they are conditionally compiled away if the platform doesn't
293 support TLI... Thanks to Neil Cohen for pointing out the need
294 for this...
296 * ace/OS.i: Modified the implementation of dlopen() and dlsym() to
297 account for limitations with SunOS 4.x...
299 * ace/Service_Repository.cpp: Modified this class to use the
300 ACE_OS::dlclose method.
302 * ace/Parse_Node.cpp: Modified this class to use the
303 ACE_OS::dlopen and ACE_OS::dlsym methods.
305 Wed Dec 27 23:25:03 1995 Jesper S. M|ller (stophph@diku.dk)
307 * Added appropriate flags in OS.h for successful inclusion of
308 several header files, such as Malloc.h and friends. Also added
309 mprotect to ACE_OS for Win32 to avoid linker errors.
311 * Added a return value (from map_file) to
312 ACE_MMAP_Memory_Pool::handle_signal. The compiler was
313 complaining. Likewise on
314 ACE_MMAP_Memory_Pool::commit_backing_store (there 0 seemed like
315 the correct off-the-end return value).
317 * Removed include of some SV semaphore file, if not needed in
318 Malloc.h. Worse for Memory_Pool.h, since the SV_Semaphore
319 wrappers consistently differ completely from the other
320 synchronization routines. I guess we'll need some variant of
321 ACE_Process_Semaphore that takes an integer as a key instead of
322 a string, and does not perform automatic init/destroy.
324 * Also upgraded fd's from int to ACE_HANDLEs and changes -1 to
325 ACE_INVALID_HANDLE where appropriate.
327 * By help of aggresive preprocessor conditionals, made Handle_Set
328 work with Win32. There is a fd_set abstraction with appropriate
329 FD_xxx macros for WinSock, and it requires no sync'ing. The
330 Handle_Set_Iterator should also work. The good looks of
331 Handle_Set was impaired, though. Right now, the trigger is the
332 ACE_WIN32 flag, although it should be something like
333 ACE_FD_SET_USES_BITS for the unix approach, or like
334 ACE_FD_SET_USES_VECTOR for the WinSock approach.
336 * Reactor almost compiles now, but not quite: Handlers are
337 addressed by fd, this is no longer possible - what a mess - we
338 need to step through the handle indices somehow. This will
339 likely get into some nasty quadratic complexities when iterating
340 the handles to find the event_handler for each file
341 handle. Mabye we can circumvent this with a hash table lookup
342 for those opaque handles.
344 * There was a problem with the Svc_Conf.l.cpp and the
345 Svc_Conf.y.cpp and the extremely lame MSVC++ 2.0. I had to
346 rename them to Svc_Conf_y.cpp, etc. Awful!
348 * Added code to inquire the 'host' name for Win32 in the ACE_Client_Id
349 class. Another candidate for ACE_OS.
351 * About 60 or so ACE .cpp files compile now, which reminds me that
352 I'll be working on some precompiled header scheme soon, since it
353 takes for ever and ever, even on my 32meg system with pretty
354 good SCSI drives. Mabye it's because I'm only running a 66MHz
355 i486 DX2 (yawn).
357 * Dozens of other changes have been made to various files, all
358 pertaining to int vs. ACE_HANDLE. In Service_Manager.h, a signal
359 number was an ACE_HANDLE, but just close by, some fd's were
360 ints. Fixed that, too.
362 Sat Dec 23 14:15:16 1995 Jesper S. M|ller (stophph@diku.dk)
364 * I've also grouped the .i and .cpp functions by class,
365 and #if'ed the sbrk and SysV pools out, since
366 Win32 cannot support those.
368 * Added 'inline' to day accessor functions in Date_Time.
370 * Two things that missed me in Trace.cpp:
371 1) The inclusion order was so that I couldn't
372 build Trace.o, since it didn't know the MALLOC_HOOK thing.
373 I moved the #include of Trace.h down below that of ACE.h.
374 2) The declaration and definition of the ACE_Trace constructor
375 differed by a const, which I added.
376 3) When linking, multiple symbols were found for some
377 OS calls like mutex_init, corresponding to the larger
378 OS Win32 functions. I has ACE_INLINED_OS_CALLS off,
379 so sometimes the OS functions were inlined, sometimes
380 not.
381 I disabled the inline hack in Trace.cpp for WIN32 for now.
382 This seemed to resolve matters.
384 * There was a major problem in OS.i, where Synch.h was needed for
385 mutex support (in conjunction with TSS info linked list). I
386 chose to make thr_keycreate and some others non-inlined.
388 * More 'is-it-a-handle-or-success' confusion in SOCK*.* Changed
389 ACE_SOCK_Dgram to use int status. Changed ACE_SOCK_CODgram to
390 use int status. Changed ACE_LSOCK to use int status. Also
391 changed a few derived classes
393 Fri Dec 22 14:03:15 1995 Douglas C. Schmidt (schmidt@lambada.cs.wustl.edu)
395 * ace/OS.h: Added a new type for rwlock_t for platforms that don't
396 support threading... Thanks to Neil Cohen for reporting this.
398 * ace: Merged in all of Jesper's changes for NT and Chris Lahey's
399 changes for AIX.
401 * ace: Merged tli.h into OS.h to avoid name conflicts on Win32.
403 * Fixed duplicated names where case is the only difference.
404 ace/thread.cpp
405 ace/Thread.cpp
406 ace/TLI.h
407 ace/tli.h
408 apps/Token_Server/mutex_example/simple_token_client.cpp
409 apps/Token_Server/mutex_example/Simple_Token_Client.cpp
410 apps/Token_Server/script_example/scripting_token_client.cpp
411 apps/Token_Server/script_example/Scripting_Token_Client.cpp
412 tests/IPC_SAP/SOCK_SAP/FD-client.cpp
413 tests/IPC_SAP/SOCK_SAP/fd-client.cpp
414 tests/IPC_SAP/SOCK_SAP/FD-server.cpp
415 tests/IPC_SAP/SOCK_SAP/fd-server.cpp
417 Wed Dec 20 22:26:24 1995 Jesper S. M|ller (stophph@diku.dk)
419 * FIFO.cpp: One last comparison of 'open' return status where and
420 ACE_INVALID_HANDLE should be changed to -1.
422 * One more last wrong comparison in FIFO_Recv_Msg.cpp.
423 (Never say never again!)
425 * Put an appropriate conditional around the inclusion of
426 Handle_Set.h for ACE_WIN32 (for now).
428 * Introduced a few backslashes in the macros of Dump.h that were
429 somehow left out... Had to add a really stupid constructor for
430 struct Tuple. I haven't a clue why MSVC++ wants this! Replaced
431 'return ACE_ODB;' with 'return ACE_ODB::instance_;' in the
432 Singleton access function.
434 * Added a constructor for ACE_ODB. It may be private, but it still
435 needs to be defined...
437 * Stack and Set compiles fine for Win32 (not surprisingly)
439 * If'ed out the guts of ACE_IPC_SAP enable/disable calls for
440 Win32, since async IO doesn't work that way on there.
442 * Event_Handler now uses ACE_INVALID_HANDLE for default args.
444 * SOCK compiles except for the ACE_SOCKET <> ACE_HANDLE
445 problem. Refer to issue #3 above.
447 * We're swamping the ace directory with .mak files. I used the
448 winntace.mak file, which is compilable and linkable.
450 Thu Dec 21 15:37:42 1995 Douglas C. Schmidt (schmidt@merengue.cs.wustl.edu)
452 * ace: Greatly simplified the include files so that (1) compilers
453 are faster and (2) the ACE_Trace mechanism works correctly.
455 * ace/Trace: Modified the ACE_Trace_TSS class so that it is no
456 longer a template. This was overkill since we know the type
457 head of time (i.e., ACE_Trace_State).
459 * ace: Added a number of new changes to ACE to make things work on
460 AIX. Thanks to Chris Lahey <clahey@ix.netcom.com> for these
461 changes.
463 * ace/Thread_Manager.cpp: Removed the get_max_thr_id() method
464 since it was not portable to NT or POSIX pthreads. Nothing in
465 ACE depended on it anyway...
467 * ace/OS: Added a new static data member called NULL_thread that
468 is a "zero'd" out thread id. This is necessary to deal with the
469 POSIX pthread implementation of a thread id, which is not always
470 implemented as a number... By using this new approach, all the
471 special-case code in Synch.cpp disappears...
473 * ace/OS: Added a new method to class ACE_OS called thr_equal()
474 that masks the differences between comparing thread ids using
475 the threading libraries. This simplifies the code in Synch.cpp
476 and Reactor.cpp.
478 * ace: Updated all uses of mmap() to use the new ACE_MMAP_TYPE
479 typedef. This deals with the fact that different versions of
480 UNIX define mmap() inconsistently (e.g., some use void * and
481 some use char *...).
483 * ace: Removed testconfig.h and merged all of those #defines into
484 OS.h. This centralizes changes in one place and also eliminates
485 the need for ./apps and ./tests to include extraneous files...
487 * ace/testconfig.h: Changed all uses of 0666 to ACE_DEFAULT_PERMS.
488 Also, removed all PERMS enums and substituted ACE_DEFAULT_PERMS
489 instead.
491 Wed Dec 20 17:29:55 1995 Tim H. Harrison (harrison@merengue.cs.wustl.edu)
493 * Rewrote the Token deadlock detection algorithm. It now performs
494 "best-effort" detection of deadlock for remote acquires and
495 complete deadlock detection for local acquires.
497 Wed Dec 20 02:37:39 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
499 * ace/OS.h: The select() system call has its own include file on
500 AIX: <sys/select.h>. We can wrap the #include with a variable
501 #if defined (ACE_HAS_SELECT_H).
503 * include/makeinclude/platform_aix.GNU: Added new support for AIX.
504 Thanks to Chris Lahey <clahey@ix.netcom.com> for these changes.
505 Added a number of other minor changes, as well.
507 * ace: Made a bunch of minor changes to get ACE to compile on
508 SunOS4.x, Linux, and SGI.
510 * ace/config-linux.h: added ACE_HAS_MSG (thanks to Neil Cohen
511 for this).
513 * ace/OS.h: Conditionally compile for ACE_HAS_UTIME for
514 <sys/utime.h>.
516 * Released an alpha version of ACE containing the new Windows NT
517 port.
519 * Integrated and tested all the new Windows NT changes in class
520 ACE_OS, as well as the Synch* and Threads* classes. We're now
521 able to build Jesper's win32_test.cpp file in ./ace! A complete
522 NT port should be just around the corner...
524 Tue Dec 19 17:59:04 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
526 * ace: Move the remaining contents of sysincludes.h into ACE.h.
527 There is no more sysincludes.h (at long last!).
529 * ace: Rearranged things so that the OS.[hi] files contain all the
530 OS-specific #defines and #includes that were previously in
531 sysincludes.h.
533 * ace: Renamed the ACE_Misc class to class ACE, which is more
534 representative of what it does.
536 * INSTALL: updated the installation instructions to explain how
537 the new directory structure works. Thanks to Neil Cohen for
538 pointing out the discrepancy beteen the source and the
539 documentation ;-).
541 * Reorganized (by flattening) the ACE library directory structure
542 to make it work for both Windows NT and UNIX. Note that Windows
543 NT does not have symbolic links, so the old way of doing things
544 didn't port... The consequences of these changes is there is
545 now a single directory called $WRAPPER_ROOT/ace that contains
546 the *.[hi] and *.cpp source files. This is also where the
547 libACE.a and libACE.so libraries are built by default. Also,
548 note that the suffix used by ACE has been changed from *.C to
549 *.cpp to deal with the lame Visual C++ compiler...
551 Tue Dec 19 01:26:54 1995 Douglas C. Schmidt (schmidt@merengue.cs.wustl.edu)
553 * tests/ASX/Event_Server/Event_Server/Supplier_Router.cpp (put):
554 Fixed up a problem with the Event_Server. Somewhere along the
555 way it stopped being concurrent! The new version fixes this.
556 Thanks to Alex V Maclinvosky <alexm@teltrunk1.tait.co.nz> for
557 noticing this...
559 * Added a billion new changes to class OS to integrate the Windows
560 NT port with the rest of ACE. We are getting very close...
562 Mon Dec 18 12:45:13 1995 Jesper S. Møller (stophph@diku.dk)
564 * Major problems in the IPC_SAP. First, I detected that some open
565 calls returned handles, some returned -1 or 0. For instance
566 ACE_SOCK has a constructor, that calls this->open and expects a
567 handle (which it even sets as the current handle
568 value!). Unfortunately, the return from open is either 0 or -1,
569 depending on how well the open went. ACE_HANDLE being
570 typedef'ed to a void* really reveals some problems! I have
571 solved the problem for the FIFO*.* family and for SOCK.*
573 Mon Dec 18 12:43:46 1995 Douglas C. Schmidt (schmidt@merengue.cs.wustl.edu)
575 * tests/Reactor/CODgram/CODgram.C (main): Fixed a bug in main()
576 caused by initializing the Reactor twice. Thanks to Aurelio
577 Nocerino <aurelio@irsipcs2-27-le0.irsip.na.cnr.it> for reporting
578 this.
580 * libsrc/Service_Configurator/Service_Config.C Added a new
581 constructor to ACE_Service_Config that can be used to initialize
582 internal variables without performing a full configuration.
584 Mon Dec 18 01:22:22 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
586 * libsrc/Service_Config: renamed tokens.h to sc_tokens.h to avoid
587 file name collision on Win32. Thanks to Jesper S. M|ller
588 <stophph@diku.dk> for reporting this.
590 * tests/Mem_Map/IO_Test: renamed io_test.C to test_io.C to avoid
591 file name collision on Win32. Thanks to Jesper S. M|ller
592 <stophph@diku.dk> for reporting this.
594 * libsrc/Misc/OS: Implemented condition variables for Windows NT.
596 Sun Dec 17 21:23:32 1995 Douglas C. Schmidt (schmidt@merengue.cs.wustl.edu)
598 * libsrc/Misc/Trace: Fixed up the ACE_Trace class so that it
599 doesn't include any calls that will improperly trigger infinite
600 recursion! This allows the ACE library to be traced completely.
602 Sat Dec 16 19:12:37 1995 Jesper S. Møller (stophph@diku.dk)
604 * Changes to OS.*:
605 * Added conditionals to the OS.h for Win32, for the special ACE/Win32
606 thr_destroy_tss(), which is automatic for Solaris.
607 * Changed declarations of thr_create, thr_suspend, thr_continue,
608 thr_join and thr_[gs]etprio to use hthread_t instead of thread_t.
609 * Changed return type of OS::open from int to ACE_HANDLE.
610 * Provided non-obtrusive handling of thr_handle in OS::thr_create
611 for non-Win32.
612 * Copied ACE::writev to OS::writev for Win32. Something similar
613 should happen for other systems...
615 * Changes to Thread.*:
616 * Changed join, suspend and continue to take handles.
617 * Added thread handle return address to spawn.
619 * Changes to Thread_Manager.*:
620 * Added thread handles to the Thread_Descriptor structure.
621 * Modified insert_thr and append_thr functions to also take
622 handles.
623 * Made ACE_Thread_Descriptor::Thread_State (a enum) public
624 so Thread_Manager::append_thr could use it as a formal
625 parameter (I suspect this is a bug in MSVC++).
626 * Changes the various flavours of suspend, continue, etc. to
627 use handles instead of id's. Keep in mind that none of this
628 is destructive for Unix, but vital for Win32.
629 * There is still a problem with ACE_Thread_Control registering
630 itself with the Thread_Manager - the thread does not have means
631 of finding it's own handle. In this particular case, we could
632 have the thread duplicate it's pseudohandle into a real one,
633 and pass that one along. I'd really, really, really rather not
634 have to maintain a thread id->handle mapping myself; It would
635 slow down thread manipulation and give some troublesome memory
636 overhead.
638 * Changes to Log_Msg.*:
639 * OS::getpid instead of ::getpid.
641 * Changes to Misc.*:
642 * Sections inappropriate for Win32 #if'd out...
643 * Provided naive timestamp function (No date) for Win32.
645 * Changes to IPC_SAP.*:
646 * Uses OS::ioctl instead of ::ioctl
647 * Uses ACE_INVALID_HANDLE instead of ACE::INVALID_HANDLE
649 * Changes to FIFO.*, :
650 * Uses ACE_INVALID_HANDLE instead of ACE::INVALID_HANDLE
652 * Changes to FIFO_Send.*:
653 * open now returns an int insted of an ACE_HANDLE.
654 * Uses ACE_INVALID_HANDLE instead of ACE::INVALID_HANDLE
656 * Changes to FIFO_Send_Msg.*:
657 * open now returns an int insted of an ACE_HANDLE.
658 * Uses ACE_INVALID_HANDLE instead of ACE::INVALID_HANDLE
659 * Uses OS::writev instead of ::writev
661 * Changes to FIFO_Recv.*:
662 * open now returns an int insted of an ACE_HANDLE.
663 * Uses ACE_INVALID_HANDLE instead of ACE::INVALID_HANDLE
665 * Changes to FIFO_Recv_Msg.*:
666 * open now returns an int insted of an ACE_HANDLE.
667 * Uses ACE_INVALID_HANDLE instead of ACE::INVALID_HANDLE
669 * Changes to sysincludes.h:
670 * Uses ACE_INVALID_HANDLE instead of ACE::INVALID_HANDLE
672 Sat Dec 16 12:29:06 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
674 * libsrc/IPC_SAP/Addr/INET_Addr.C: Changed the implementation so
675 that the reentrant versions of the library calls are always used
676 (class OS sorts this all out). This reduces the amount of
677 clutter in the code.
679 * libsrc/Threads/Synch: Added a new implementation of
680 "readers/writer" locks for both Windows NT and POSIX pthreads.
681 By default, these threads packages don't have this feature.
683 * libsrc/Threads/Synch: Added a new macro called
684 ACE_SYNCH_ERROR_RETURN that greatly simplifies the definition of
685 the synchronization wrappers.
687 * include/ace/sysincludes.h: Removed the pthreads_map.h file and
688 integrated this into sysincludes.h.
690 Fri Dec 15 02:39:25 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
692 * libsrc/Misc/OS.i: Finished updating class OS so that it will
693 compile correctly for OS platforms that don't support threads,
694 as well as platforms that support POSIX Pthreads vs. Solaris
695 threads, etc.
697 * libsrc/Misc/OS: integrated Jesper S. M|ller <stophph@diku.dk>
698 support for POSIX regular expressions (e.g., compile() and
699 step()) and sysinfo() into ACE.
701 * Integrated and tested all of Irfan's changes to ACE to support
702 the new persistence Name_Server. This added some new classes in
703 various places that use the ACE_Allocator.
705 Thu Dec 14 00:23:43 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
707 * tests/ASX/Event_Server/Event_Server: Fixed an "off by 1" bug in
708 the declaration of char *argv[3] in the open() methods of
709 Supplier_Router.C and Consumer_Router.C. Thanks to the "ever
710 dissatisfied" Alex V Maclinvosky <alexm@teltrunk1.tait.co.nz>
711 for reporting this.
713 * libsrc/Service_Configurator/Service_Config: Added a new
714 "Singleton" access point for the global ACE_Allocator.
716 * libsrc/Naming/Name_Options.C (parse_args): Made it possible to
717 change the name of the process on the command line via the -P
718 option. The default is still to use argv[0] as the process
719 name.
721 * include/ace/testconfig.h (ACE_DEFAULT_GLOBALNAME): Changed the
722 value from "/localnames" to "/globalnames" to avoid a conflict.
724 * libsrc/Misc/SString.C: Added Irfan's new "persistent" string
725 mechanisms that use the ACE_Allocator_Manager. We may
726 eventually update the existing SS* stuff to use this.
728 * Merged in Jesper S. M|ller <stophph@diku.dk> updates for Windows
729 NT.
731 * include/ace/Malloc.h: Added a forward declaration for
732 ACE_Malloc_Iterator right before ACE_Malloc. Thanks to Antonio
733 Tortorici <antonio@rh0011.roma.tlsoft.it> for reporting it.
735 * libsrc/Threads/Synch_T.h: Fixed a small typo in
736 ACE_Null_Condition. Thanks to Antonio Tortorici
737 <antonio@rh0011.roma.tlsoft.it> for reporting it.
739 * include: Added new config-linux.h and platform_macros.GNU config
740 files for Linux. Thanks to Timothy Newell
741 <910430n@dragon.acadiau.ca> for this stuff.
743 Wed Dec 13 23:08:11 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
745 * libsrc/IPC_SAP/Addr/INET_Addr.C (set): Fixed a small typo in the
746 ACE_INET_Addr::set (const char port_name[], ACE_UINT32
747 inet_address) method where the test should be != rather than ==
748 0. Thanks to Bill Lear (rael@anarchy.cybercom.net) for
749 reporting this.
751 Sun Dec 10 12:23:54 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
753 * libsrc/Service_Configurator/Service_Record.C (init): Fixed a
754 small problem where a NULL pointer was being returned if a user
755 passes a non-existent config file (using the -f option) to
756 ACE_Service_Config::open(). Thanks to Dieter Quehl
757 (quehl@erlh.siemens.de) for noticing this.
759 * libsrc/Log_Msg/Log_Msg.C (log): added a new option (%T) that
760 causes the current timestamp to be inserted into the log msg.
761 Thanks to Dieter Quehl (quehl@erlh.siemens.de) for suggesting
762 this.
764 * libsrc/Misc/Misc: Added a new timestamp() method that returns
765 the current timestamp in the form
766 "hour:minute:second:microsecond." The month, day, and year are
767 also stored in the beginning of the date_and_time array.
769 * tests/Shared_Malloc: Added Irfan Pyarali <ip1@cec.wustl.edu> new
770 test program that exercises the new features of ACE_Malloc that
771 are described in the following bullet.
773 * libsrc/Shared_Malloc: Added Irfan Pyarali <ip1@cec.wustl.edu>
774 cool new versions of ACE_Malloc and ACE_MMAP_Memory_Pool. These
775 new versions provide the following enhancements:
777 1. Persistence -- via the sync() method
778 2. Protection -- via the prot() method
779 3. Named malloc chunks -- via bind(), find(), unbind().
780 4. Named iteration -- via ACE_Malloc_Iterator
782 Thu Dec 7 00:31:26 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
784 * libsrc/Reactor/Signal.C: Replaced ACE_Guard with ACE_TSS_Guard
785 to ensure that locks are correctly released even if
786 Event_Handler::handle_signal() callbacks invoke
787 ACE_Thread::exit(). Thanks to Detlef Becker
788 (beckerd@erlh.siemens.de) for suggesting this.
790 * libsrc/Threads/Synch_T: Used inheritance to factor out the
791 shared code between ACE_Guard, ACE_Read_Guard, and
792 ACE_Write_Guard in order to reduce unnecessary duplication.
794 * libsrc/Threads/Synch_T: Changed the implementation of ACE_Guard
795 to keep a pointer to the LOCK rather than a reference. This
796 makes the code more flexible and aids in factorization.
798 * libsrc/Threads/Synch_T: Changed the implementation of ACE_Guard
799 to incorporate ACE_Try_Guard semantics. This reduces the amount
800 of classes and duplicate code in ACE.
802 Wed Dec 6 21:36:02 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
804 * include/ace/Trace: Changed the type of the string passed into
805 ACE_Trace from char * to const char *.
807 * libsrc/Reactor/Reactor.C: Had forgotten to initialize the
808 initialized_ and requeue_position_ members appropriately in the
809 constructor of the ACE_Reactor if ACE_MT_SAFE is *disabled*.
810 Thanks to Mark Zusman <marklz@topaz.technion.ac.il> for pointing
811 this out.
813 * libsrc/Threads/Synch_T.C: changed the order of some methods so
814 they would be properly defined if ACE_HAS_THREADS and
815 ACE_HAS_THREAD_SPECIFIC are *not* enabled. This should fix
816 remaining problems with ACE on SunOS 4 and SGI. Thanks to
817 Aniruddha Gokhale <gokhale@cs.wustl.edu> for noticing this.
819 Tue Dec 5 01:06:14 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
821 * Changed things around so that the Name_Server and Token_Server
822 client interfaces are include in libACE.
824 * libsrc/Shared_Malloc: added a const char *pool_name to
825 ACE_Malloc::ACE_Malloc. This is passed into the constructor of
826 the MEMORY_POOL to make it easier to dynamically name memory
827 pools.
829 Mon Dec 4 21:20:28 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
831 * libsrc/Threads/Synch_T: Modified the place that the
832 ACE_Thread_Specific methods are defined so that they will be
833 compiled correctly for platforms without threads or
834 thread-specific storage.
836 Sun Dec 3 23:04:41 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
838 * libsrc/ASX/Message_Block: Modified clone() so that it takes an
839 extra parameter that specifies which flags SHOULD NOT BE
840 TRANSFERED to the clone. This defaults to
841 ACE_Message_Block::DONT_DELETE. Thanks to Alex V Maclinvosky
842 <alexm@teltrunk1.tait.co.nz> for suggesting this.
844 Sun Dec 3 17:43:11 1995 Tim H. Harrison (harrison@lambada.cs.wustl.edu)
846 * The number of files in the Token library has been reduced. As a
847 result, many of the files have been renamed. Also, the .i files
848 have been purged of relatively large methods (these have been
849 moved to the .C files.) Lastly, most uses of dynamic memory
850 allocation have been removed from the library. This was
851 important to reduce the amount of serialization in threaded
852 applications caused by the dynamic allocation.
854 Sun Dec 3 17:29:54 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
856 * libsrc/Threads/Synch: Modified all the synchronization classes
857 so that they disallow assignment and initialization!
859 * libsrc/Threads/Synch: Moved the ACE_Thread_Specific class from
860 its own files into the Threads/Synch.[Chi] files. This should
861 eliminate a set of horrible interdependencies among files.
863 Sat Dec 2 16:07:40 1995 Douglas C. Schmidt (schmidt@mambo.cs.wustl.edu)
865 * libsrc/Misc/SString.C (ACE_CString): Added a destructor to
866 CString. Thanks to Tim for noticing this...
868 Tue Nov 28 17:25:02 1995 Prashant Jain (pjain@tango.cs.wustl.edu)
870 * apps/Name_Server/client/lib/Name_Proxy: Changed Name_Proxy to
871 inherit from ACE_Service_Object rather than ACE_Event_Handler.
872 This will allow a Name_Proxy to be dynamically linked into an
873 application via the svc.conf file.
875 * apps/Name_Server/client/lib/Name_Proxy: Added new methods
876 init, fini, and info.
878 Tue Nov 28 00:56:52 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
880 * tests/Threads/test_thread_specific.C: Added code to test the new
881 ACE_TSS_Guard class...
883 * libsrc/Threads/Synch_T: Added the new ACE_TSS_Guard class. This
884 class is similar to the existing ACE_Guard, except that it uses
885 thread-specific storage to ensure locks are released even if a
886 thread exits via thr_exit()!
888 Mon Nov 27 20:39:58 1995 Prashant Jain (pjain@tango.cs.wustl.edu)
890 * apps/Name_Server/{client,server}/tests: Added new tests to
891 exercise the key features of the name server.
893 * apps/Name_Server/client/lib: Simplified access to the nameserver
894 database. For example, all the network-byte ordering code was
895 moved out of the Local and Remote Name_Spaces. This code is no
896 longer necessary since we'll be providing a different means to
897 get portable name service repositories via a tool that extracts
898 the "key/value/type" tuples and stores them in a
899 machine-independent format. The result of this change is that
900 the Name_Server code is much easier to understand and modify.
902 * apps/Name_Server/client/lib: Split local and remote name spaces
903 into two classes to simplify usage. This approach uses the
904 Bridge pattern to have a baseclass called Binding and then have
905 subclasses called Local_Name_Space and Remote_Name_Space. The
906 constructor of Naming_Context then allocates the right one
907 (i.e., it serves as a factory). This approach remove all the
908 switch statements in the code and makes it much easier to use
909 and extend.
911 * apps/Name_Server: updated all the Name_Server components to
912 use ACE naming conventions.
914 * apps/Name_Server/client/lib: Changed ACE_Name_Options to use the
915 Singleton pattern rather than the global variable...
917 Mon Nov 27 00:47:32 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
919 * libsrc/Shared_Malloc/Malloc.h: Named the previously anonymous
920 struct inside of ACE_Malloc_Header to be
921 ACE_Malloc_Control_Block in order to keep the OSE tools working.
922 Thanks to Karlheinz for reporting this.
924 * libsrc/Threads/Synch: Fixed a major bug in the ACE_Semaphore
925 wrappers. Apparently, these had slipped between the cracks and
926 were not correctly mapping onto the ACE return value scheme
927 where -1 corresponds to an error with errno set to indicate the
928 type of error.
930 * tests: Cleaned up some files that mistakenly were defining
931 LSOCK* objects even for platforms (i.e., Linux) that don't
932 support them. Thanks to Timothy Newell
933 <910430n@dragon.acadiau.ca> for noticing this.
935 Sun Nov 26 12:42:51 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
937 * libsrc/Shared_Malloc/Malloc: Added a new accessor method
938 (memory_pool) that returns a reference to the MEMORY_POOL object
939 used to configure ACE_Malloc.
941 * libsrc/Shared_Malloc/Malloc: Changed the constructor of
942 ACE_Malloc to take advantage of the new init_release method on
943 all the ACE_*_Memory_Pools.
945 * libsrc/Shared_Malloc/Memory_Pool: Added a new method to all the
946 ACE_*_Memory_Pool classes called "init_release". This is called
947 when ACE_Malloc finishes initializing the dynamic memory
948 manager. This method is necessary in order to initialize the
949 ACE_MMAP_Memory_Pool correctly if there are multiple processes
950 that try to initialize ACE_Malloc simultaneously. In addition,
951 also added an ACE_SV_Semaphore_Complex to the
952 ACE_MMAP_Memory_Pool in order to serialize initialization
953 correctly.
955 * libsrc/Shared_Malloc/Memory_Pool: Added sync() and protect()
956 methods to all of the ACE_*_Memory_Pool classes in order to
957 support a uniform interface for use with the "robust memory"
958 mechanism we're building.
960 * libsrc/Mem_Map/Mem_Map: Added two new "protect" methods to
961 ACE_Map_Manager in order to allow clients to change the
962 protection of memory-mapped regions.
964 * libsrc/ASX/Map_Manager: Changed Map_Manager methods to use
965 "const" for EXT_ID and INT_ID at request of Prashant Jain
966 (pjain@cs.wustl.edu).
968 * libsrc/Shared_Malloc/Memory_Pool: Revised the
969 ACE_Shared_Memory_Pool so that it would actually work
970 correctly... This code hadn't been reviewed carefully before
971 and there were some bugs...
973 * libsrc/Shared_Malloc/Malloc: added "INLINE" to the
974 ACE_Allocator_Adapter components.
976 * libsrc/Shared_Malloc/Memory_Pool: Greatly simplified and
977 consolidated the mmap(2) code for ACE_MMAP_Memory_Pool.
979 * libsrc/Shared_Malloc/Memory_Pool.i (release): Removed the
980 vestigal this->mmap_.remove() call in release(), but added the
981 appropriate code to close the fd and unmap the region. Thanks
982 to Irfan Pyarali <ip1@cec.wustl.edu> for noticing this.
984 Sat Nov 25 18:12:55 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
986 * libsrc/CORBA/CORBA_Handler.C (activate_service):
987 Modified activate_service so that
989 if (service_name != 0
990 && this->register_service (service_name, marker_name,
991 service_location) == -1)
992 return -1;
994 is changed to
996 if (service_name != 0 && service_location != 0
997 && this->register_service (service_name, marker_name,
998 service_location) == -1)
999 return -1;
1001 Irfan Pyarali <ip1@cec.wustl.edu> requested this.
1003 Sat Nov 25 16:41:15 1995 Douglas C. Schmidt (schmidt@lambada.cs.wustl.edu)
1005 * libsrc/Misc/Set: Changed all the find() and size() methods to
1006 be const member functions. Thanks to Irfan Pyarali
1007 (ip1@cec.wustl.edu) for requesting this.
1009 Fri Nov 24 02:51:59 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1011 * libsrc/Reactor/Reactor.C: Fixed the Reactor so that if we're
1012 compiling with DEADLOCK_DETECTION enabled it will create a
1013 uniquely named mutex by stringifying "this".
1015 Thu Nov 23 21:45:43 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1017 * libsrc/IPC_SAP: Changed
1019 * libsrc/Service_Configurator/Service_Config: Added three new
1020 methods to allow programmers to set the process-wide Singletons
1021 for Reactor, Service_Repository, and Thread_Manager.
1023 Tue Nov 21 01:59:06 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1025 * libsrc/Mem_Map/Mem_Map.C: Fixed a stupid typo in map_it() that
1026 was causing problems due to unsigned arithmetic semantics...
1027 This should make the file_reverse.C Mem_Map tests work correctly
1028 now... Thanks to Bill Lear <rael@anarchy.cybercom.net> for
1029 noticing the problem.
1031 * tests/Mem_Map/IO-tests/io_test.C (parse_args): Fixed a
1032 classic C/C++ braino in io_test.C:
1034 The for loop in parse_args()
1036 for (int c; (c = get_opt () != -1); )
1038 should be
1040 for (int c; ((c = get_opt ()) != -1); )
1042 How embarrassing. Maybe Bertrand Meyer is right -- C++ is too
1043 dangerous for its own good... ;-). Thanks to Bill Lear
1044 <rael@anarchy.cybercom.net> for finding this!
1046 * Removed all traces of ACE_Condition<ACE_Mutex> from the library.
1047 This should make life much easier for compilers (like G++) that
1048 can't grok templates very well...
1050 * libsrc/Threads/Thread_Specific.C (ts_object): Added a new method
1051 that allows you to simultaneously "test and set" thread-specific
1052 data! This is used to support a neat new trick that ensures
1053 locks are released even if a thr_exit() is called!
1055 * libsrc/Service_Configurator/Service_Repository: Removed the
1056 Recursive_Mutex from Service_Repository and replaced it with a
1057 regular (non-recursive) mutex. Had to rewrite the code a bit so
1058 that it wouldn't try to call internal methods that acquired the
1059 mutex (this leads to instant deadlock!)
1061 Mon Nov 20 01:05:47 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1063 * libsrc/Threads/Synch: Finally gave in and accepted the fact that
1064 GNU G++ is a horrible, horrible blight on mankind... Changed
1065 the internals of Synch.[Chi] to completely remove any
1066 dependencies on templates. I hope this fixes some portability
1067 problems deep in the libraries... Added several new classes to
1068 reflect this change: ACE_Mutex_Guard, ACE_Mutex_Condition, and
1069 ACE_Recursive_Mutex.
1071 * libsrc/IPC_SAP/UPIPE_SAP/UPIPE_Acceptor.C (ACE_UPIPE_Acceptor):
1072 Fixed a stupid typo that was causing the constructor of
1073 UPIPE_Acceptor to become inlined!!!
1075 * apps/Gateway/Gateway: Fixed up the Makefile so that the gatewayd
1076 executable would compile with G++.
1078 * include/ace/config-irix5.3-sgic++.h: Changed the SGI config.h
1079 file so that it uses select() rather than poll() by default.
1080 The select() version of the Reactor is implemented more
1081 efficiently in ACE...
1083 * include: Added config files for AIX courtesy of Byron Walton
1084 <bwalton@hughes.scg.hac.com>.
1086 * libsrc/Reactor/Signal.C: #ifdef'd out some code in Signal.[hC]
1087 that was causing problems for the HPUX C++ compiler. What a
1088 *horrible* compiler!
1090 * libsrc/Reactor/Event_Handler.h: Added two new methods to
1091 ACE_Event_Handler: get_priority() and set_priority(). These
1092 will be used to control the behavior of the ACE_Reactor's
1093 dispatching. This also required added a new data member to
1094 ACE_Event_Handler that stores the priority (which defaults to
1095 MIN_PRIORITY).
1097 * libsrc/Misc/Stack.C (enqueue): Added a new class called
1098 ACE_Unbounded_Queue. This is required for Tim's Token_Server
1099 connection repository stuff...
1101 * libsrc/CORBA/CORBA_Handler.C: Added new support to both the
1102 single-threaded and multi-threaded versions of CORBA_Handler so
1103 that it will now suspend/resume the services associated with a
1104 CORBA_Handler.
1106 * libsrc/CORBA/CORBA_Handler.C (ACE_MT_CORBA_Handler): Updated the
1107 constructor to use the associated Thread_Manager to spawn a new
1108 thread. This will allow the suspend() and resume() methods to
1109 atomically resume and suspend the daemon thread and its event
1110 handler.
1112 * libsrc/Threads/Synch.h: Fixed up the order of #includes to solve
1113 problems with circular header dependencies!
1115 * Had to add zillions of little changes to avoid problems with
1116 circular includes for G++...
1118 * tests/ASX/Event_Server/Event_Server/Peer_Router.C (bind_peer):
1119 Fixed a typo that had been dormant for ages due to C++'s very
1120 late binding of templates... G++ picked this right up!
1122 Sun Nov 19 11:55:02 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1124 * libsrc/Reactor/Signal: Replaced the static HANDLER_SET stuff
1125 withing ACE_Sig_Handlers with a different implementation using
1126 ACE_Fixed_Set. This should fix nasty problems with static
1127 variables on HP/UX...
1129 * Fixed the entire library so that all occurrences of ace_log_msg
1130 are changed to use the ACE_LOG_MSG macro. This macro has magic
1131 properties that should greatly simplify the effort required to
1132 compile ACE robustly on platforms with lame C++ compilers...
1134 * libsrc/Log_Msg/Log_Msg: Finally got my act together with respect
1135 to thread-specific data and the ACE_Log_Msg class. The new
1136 version can be conditionally compiled such that it won't use the
1137 ACE_Thread_Specific smart-pointer wrapper mechanism unless the
1138 platform supports thread-specific storage and threads. This
1139 will make life much easier for lame C++ compilers that can't
1140 handle static data member templates correctly...
1142 * apps/Name_Server/{client,server}/lib/Makefile (LIBS): Added a
1143 reference to -lACE in the Makefiles of the ./lib directories in
1144 order to get GCC to work correctly...
1146 * Added a bunch of fixes courtesy of Bill Lear
1147 <rael@anarchy.cybercom.net> to make ACE compile more cleanly
1148 with GNU GCC 2.7.x.
1150 Sat Nov 18 11:27:40 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1152 * libsrc/Misc/Trace.C: Fixed up the ACE_Trace class so that it
1153 will compile better on platforms that don't support
1154 thread-specific storage. The trick was to make a new Singleton
1155 called ACE_Nest_Depth and then conditionally compile it
1156 according to whether the platform supports thread-specific data
1157 or not!
1159 * include/makeinclude/platform_sunos5_g++.GNU: Added the new
1160 config file that should allow ACE to build correctly with GCC
1161 2.7.x. Todd L. Montgomery <tmont@cerc.wvu.edu> deserves a big
1162 round of applause for getting this stuff to work! If I had a
1163 budget, I'd give him a raise ;-)
1165 * libsrc/Misc/Set.C (remove): Tightened up the semantics of all
1166 the *Set::remove() methods so that they return 1 if they
1167 succeed, 0 if the item isn't in the set, and -1 if they fail.
1168 Thanks to Tim Harrison for noticing the inconsistencies.
1170 Fri Nov 17 01:34:51 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1172 * apps/Logger/Reactor_Logger: Fixed up this example app so that it
1173 would compile with GNU G++ correctly. There was a problem with
1174 the order of include files! Thanks to Todd L. Montgomery
1175 <tmont@cerc.wvu.edu> for reporting this.
1177 * libsrc/ASX/Module.C (writer): Fixed ACE_READER so that it is
1178 properly qualified with ACE_Task_Flags::ACE_READER. Thanks to
1179 Neil Cohen (nbc@metsci.com) for finding this!
1181 * apps/Gateway/Gateway/Routing_Entry: Replaced ACE_Fixed_Set with
1182 ACE_Unbounded_Set to get around the seemingly endless problems
1183 with HP/UX...
1185 * libsrc/Misc/Set: Fixed a braino in ACE_Unbounded_Set, where I'd
1186 forgotten to define cur_size_. Thanks to Todd L. Montgomery
1187 <tmont@cerc.wvu.edu> for reporting this.
1189 * libsrc/Connection/Acceptor.C: Change the #undefs of the
1190 shorthand names PA_AC_1, PA_AC_2, and PA_AD in
1191 libsrc/Connection/Acceptor.C to PR_AC_1, PR_AC_2, and PR_AD
1192 (same as in Connector.C and Strategies.C). Thanks to
1193 Dieter Quehl <quehl@erlh.siemens.de> for reporting this.
1195 Thu Nov 16 02:14:22 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1197 * libsrc/Misc/Trace: Fixed the ACE_Trace stuff by making the
1198 "nesting_indent_" a static data member. This will undoubtedly
1199 break some lame C++ compilers, but it's easy enough to remove
1200 this stuff via conditional compilation using the ACE_NTRACE
1201 #define in ./include/ace/config.h.
1203 * libsrc/Threads/Thread_Specific.C (cleanup): Under no
1204 circumstances should ACE_Thread_Specific be allowed to
1205 called ACE_Trace. Otherwise, chaos will result!
1207 * apps/Name_Server/{client,server}/tests/Makefile: Switched the
1208 order of the -lACE and -lName* so that -lACE came *afterwards*
1209 in order for the SGI linker to pick up the symbols correctly.
1211 * libsrc/Reactor/Signal: I'd accidentally put the "dump" methods
1212 in the *.i file rather than the *.C file. This was causing
1213 problems for G++.
1215 * libsrc/ASX/Task: Changed all "Q_" prefixes to "ACE_" prefixes in
1216 order to work around a bug with HP/UX...
1218 * Recompiled everything on SunOS 4.x with SunC++ 4.0.1. It seems
1219 to work!
1221 Thu Nov 16 18:05:03 1995 Tim H. Harrison (harrison@tango.cs.wustl.edu)
1223 * I moved around some class declarations so that g++ can properly
1224 instantiate templates for ACE_Token_Collection and
1225 ACE_Token_Manager.
1227 Wed Nov 15 00:26:40 1995 Tim H. Harrison (harrison@lambada.cs.wustl.edu)
1229 * The token library should now compile on platforms which do not
1230 support threads. All token components will be compiled into the
1231 library, but use null mutexes and condition variables. The
1232 remote mutex library shall run properly on single-threaded
1233 platforms, so it compiles as usual.
1235 Wed Nov 15 01:05:38 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1237 * Fixed a bunch of minor problems with the SGI port. Things
1238 should work now on IRIX 5.3!
1240 * man/man3: Completely regenerated all the manual pages to
1241 reflect all the recent updates.
1243 * libsrc: Added ACE_TRACE macros to every single method in ACE!
1244 This will help with debugging...
1246 Wed Nov 15 00:26:40 1995 Tim H. Harrison (harrison@lambada.cs.wustl.edu)
1248 * The token library should now compile on platforms which do not
1249 support threads. All token components will be compiled into the
1250 library, but use null mutexes and condition variables. The
1251 remote mutex library shall run properly on single-threaded
1252 platforms, so it compiles as usual.
1254 Tue Nov 14 01:58:47 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1256 * tests/CORBA: Added a new test to illustrate the
1257 ACE_MT_CORBA_Handler.
1259 * libsrc/ASX/Task.C (ACE_Task_Exit): Fixed a niggling problem with
1260 ACE_Task_Exit. Originally, if a thread exited via an implicit
1261 ACE_Thread::exit() (i.e., "falling off the end of the
1262 ACE_Task::svc_run function) then the ACE_Thread_Control on the
1263 thread stack would automatically unregister the thread from the
1264 ACE_Thread_Manager. However, this did *not* occur if a thread
1265 explicitly called ACE_Thread::exit() since in that case the
1266 destructor for ACE_Thread_Control never got called. The
1267 solution to this is a trivial change to ACE_Task_Exit, which now
1268 maintains an instance of ACE_Thread_Control internally, which
1269 will automatically be released when the thread-specific
1270 thread-exit-hook destructor gets called to cleanup upon thread
1271 exit. Since this happens regardless of whether the thread
1272 terminates implicitly or explicitly the ACE_Thread_Manager will
1273 always be able to clean up it's internal resources (unless the
1274 process exit()s, which is a whole different issue... ;-)).
1276 * Changed all occurrences of ACE_CORBA_Handler to
1277 ACE_ST_CORBA_Handler. Now we've got two types of
1278 CORBA_Handlers: single-threaded (ST) and multi-threaded (MT).
1279 Take a look at ./apps/Orbix-examples/Event_Comm/{Supplier,Consumer}
1280 for an example of how to leverage this in a portable manner!
1282 * libsrc/CORBA/CORBA_Handler: Totally redesigned the ACE
1283 CORBA_Handler stuff so that it now uses the Singleton pattern
1284 explicitly, rather than using the half-baked static method
1285 version... Changed the test code to reflect this update!
1287 * libsrc/CORBA/CORBA_Handler.C (activate_service): Changed the
1288 check so that if <service_name> == 0 then we don't try to do a
1289 putit (previously, this check used <marker_name>, which is less
1290 useful...).
1292 * libsrc/CORBA/CORBA_Handler.C (activate_service): Removed the
1293 <default_iterations> flag from the constructor so that we can
1294 make the CORBA_Handler and the MT_CORBA_Handler equivalent.
1296 * include/ace/sysincludes.h: Added a new macro called
1297 ACE_THREAD_EXIT_HOOK that enables application threads
1298 (including the main thread) to register a Task * who's
1299 close method will automatically be called when the thread
1300 exits, either implicitly (by falling off the end) or explicitly
1301 (by a thread calling ACE_Thread::exit ()).
1303 * libsrc/ASX/Task: Moved ACE_Task_Exit into the interface part of
1304 ACE_Task so that Irfan can maybe use this in his stuff ;-).
1306 * libsrc/Misc: Incorporated the ACE_Date_Time class into the
1307 release. This is an *interface* for a system independent
1308 representation of date and time. Implementation will follow...
1310 * libsrc/CORBA: Added a new MT_CORBA_Handler, which makes it
1311 very straightforward to integrate the ACE_Reactor with
1312 multi-threaded (MT) Orbix.
1314 * libsrc/Threads/Token.C: Yow, fixed a dumb problem in Tokens.C
1315 where I was failing to include the *.i file if we weren't
1316 compiling with inlining on!!!!
1318 Mon Nov 13 01:13:37 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1320 * libsrc/ASX/Message_Block: Added several new methods to
1321 Message_Block to allow users to access and set various flags.
1322 Thanks to Alex V Maclinvosky <alexm@teltrunk1.tait.co.nz> for
1323 suggesting this and providing a prototype implementation.
1325 * include/ace/sysincludes.h: Added a new macro called ACE_NDEBUG
1326 that can be used to toggle support for "live object dumping".
1327 If this macro is enabled then
1329 * libsrc/Misc/Dump.h: Added support for "live object dumping" into
1330 ACE. This technique is implemented with the "External
1331 Polymorphism" pattern described in a paper available at
1332 http://www.cs.wustl.edu/~schmidt/EuroPLoP-96.ps.Z.
1334 * libsrc/Threads: Added new classes called *_Process_* and
1335 *_Thread_* for ACE_RW_Mutex, ACE_Barrier, ACE_Condition, and
1336 ACE_Semphaore. This will allow programmers explicitly state the
1337 scope of their locks more explicitly and will also facilitate
1338 portability to Windows NT...
1340 * libsrc/Threads: Updated the constructors of all the Synch and
1341 Synch_T classes to take an option const char * called "name."
1342 At the moment, this doesn't do anything, but when we port to
1343 Window NT it will be used to ensure that we can name our
1344 process-global synchronization objects.
1346 * libsrc/Threads/Synch: Added a new tryacquire method to
1347 ACE_RW_Mutex to be consistent with acquire and release...
1349 * libsrc/Threads/Synch.C (ACE_Mutex): Added a new "name" parameter
1350 to an ACE_Mutex in order to support process-semantics on NT...
1352 * include/ace/sysincludes.h: Added a #define for EDEADLK in case
1353 some systems don't support it (this is needed by the new
1354 ACE_*_Token stuff).
1356 Sun Nov 12 14:17:01 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1358 * apps/Name_Server: Modified the client and server test programs
1359 so that they can both be linked statically and/or dynamically.
1360 If svc.conf is present then dynamic linking is used, otherwise
1361 static linking is used. This makes it much easier to test!!!
1363 * apps/Name_Server/client/lib/Naming_Context.C (local): Changed
1364 things a bit so that if our server host name is "localhost" then
1365 we assume we are local no matter what...
1367 * apps/Name_Server: Moved directories around a bit so that all the
1368 tests and libs for the client/server portions of the
1369 ACE_Name_Server build correctly...
1371 * apps/Name_Server/server: Replaced the ad hoc Name_Acceptor and
1372 Name_Handler to use the official ACE_Acceptor and
1373 ACE_Svc_Handler... This cleans up the code considerably...
1375 * libsrc/Threads/Synch: Added a new wrapper for the the UNIX file
1376 locking mechanism called ACE_File_Lock. This has the same
1377 interface as the other locking mechanisms (e.g., ACE_Mutex and
1378 ACE_Semaphore). Therefore, it can be used in the ACE_Guard
1379 class!
1381 Sat Nov 11 13:53:48 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1383 * libsrc/Shared_Malloc/Memory_Pool: Improved MMAP_Memory_Pool so
1384 that it will be smarter about remapping the file if we aren't
1385 trying to force a fixed addr to be used for the mapping. In
1386 particular, if we *aren't* forcing a particular address, the new
1387 solution will be more flexible and allow the OS to determine
1388 where to remap the memory should we need to grow the backing
1389 store and the mapping range...
1391 * libsrc/Misc/Set: Implemented a simple version of the
1392 Unbounded_Set class and its iterator.
1394 * libsrc/Misc/SString: Changed all the implementations of methods
1395 in SString classes to use new/delete rather than
1396 strdup()/free().
1398 * apps/Name_Server/client/Name_Proxy: Changed rcv_reply() to
1399 recv_reply().
1401 * apps/Name_Server/client/Name_Proxy: Modified ACE_Name_Proxy so
1402 that the constructor and open take an ACE_Synch_Options.
1404 * apps/Name_Server/client/Name_Options.C (ACE_Name_Options): Added
1405 a default for the Name_Space directory called
1406 (ACE_DEFAULT_NAMESPACE_DIR) to the testconfig.h file...
1408 Sat Nov 11 00:24:37 1995 Tim H. Harrison (harrison@lambada.cs.wustl.edu)
1410 * tests/Tokens/test_token_manager/test_token_manager.C: The token
1411 manager has a working deadlock detection algorithm.
1412 ACE_Local_Mutex uses this to report deadlock situations.
1413 Changes were made to ACE_Local_Mutex to fix a couple bugs which
1414 were causing deadlock. Also, a new test application has been
1415 added to test deadlock detection using the local mutex.
1417 Fri Nov 10 17:16:06 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1419 * libsrc/Mem_Map/Mem_Map.h: Changed the type of <length_> from
1420 long to size_t.
1422 * tests/Threads/test_thread_manager.C (main): Added a main() for
1423 the case when there's
1425 * tests/ASX/UPIPE_Event_Server/event_server.C (main): Fixed a
1426 typo that was causing problems on SunOS 4.x.
1428 * tests/Service_Configurator/IPC-tests/client/local_spipe_client_test.C:
1429 added #include "ace/Log_Msg.h" (why was this working?!).
1431 * include/makeinclude/rules.lib.GNU: Changed the order of operations
1432 in rules.lib.GNU from
1434 $(AR) $(ARFLAGS) $@ $?
1435 -$(RANLIB) $@
1436 -chmod a+r $@
1438 to
1440 $(AR) $(ARFLAGS) $@ $?
1441 -chmod a+r $@
1442 -$(RANLIB) $@
1444 to keep ranlib happy on SunOS 4.x.
1446 * apps/Time_Server: Added a Makefile
1448 * include/ace/sysincludes.h: Removed the <termios.h> include
1449 since it was causing major problems on SunOS 4.x...
1451 * apps/Name_Server/Naming_Context.C: Fixed up a couple of problems
1452 with casts. Thanks to Jack Erickson <jack@cibc.com> for
1453 pointing this out.
1455 Thu Nov 9 15:49:40 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1457 * libsrc/Threads/Synch_T: moved Atomic_Op from ./Misc to ./Threads
1458 and merged it into the Synch_T files. This makes more sense as
1459 a place to put it...
1461 * libsrc/Misc/SString.C (operator +=): Fixed a typo that would
1462 have caused mistakes when memcpy() was used. Thanks to Prashant
1463 Jain <pjain@wuerl.wustl.edu> for noticing this.
1465 Wed Nov 8 22:36:51 1995 Tim H. Harrison (harrison@lambada.cs.wustl.edu)
1467 * libsrc/Reactor/Reactor.C (TOKEN_GUARD): The reactor now uses the
1468 TOKEN_GUARD macro. The Reactor can now be conditionally
1469 compiled with ACE_REACTOR_HAS_DEADLOCK_DETECTION to use the new
1470 ACE_Local_Mutex in conjunction with the ACE_Token_Manager to
1471 detect deadlock. The deadlock detection algorithm has not yet
1472 been implemented, but the hooks are now in place.
1474 * libsrc/Tokens: This subdir includes ACE's new Token library.
1475 Check out libsrc/Tokens/README for more info.
1477 * tests/Tokens: There are a couple test applications for the new
1478 Token library. See tests/Tokens/README for more info.
1480 * apps/Token_Server: This directory contains some new client
1481 interfaces to the old ACE TokenServer, now Token_Server. There
1482 are also some example use cases included in this directory. As
1483 always, see apps/Token_Server/README for more info.
1485 Wed Nov 8 00:14:55 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1487 * libsrc/IPC_SAP/IO_SAP: Added the new classes for IO_SAP from
1488 Gerhard Lenzer (lenzer@csaserv.erlh.siemens.de). This
1489 eliminates UNIX I/O-specific features from ACE.
1491 * libsrc/Misc/Set: Added a new find() method to each of the
1492 ACE_*_Set classes...
1494 * include/ace/sysincludes.h (ACE_ALLOC_HOOK_DECLARE): Changed
1495 enum __Ace { __ACE } to struct __ACE {}. I think this
1496 will cause less problems with M.I...
1498 * libsrc/ASX/Message_Block: Added a new "allocator" parameter to a
1499 Message_Block. This allows the memory stored by a Message_Block
1500 to come from someplace besides the normal heap (e.g., a shared
1501 memory segment!).
1503 * libsrc/ASX/Message_Block.C (ACE_Message_Block): Yow, fixed a bug
1504 in the destructor where ACE_BIT_ENABLED was being used in place
1505 of ACE_BIT_DISABLED! I think this was causing a memory leak...
1507 * libsrc/Reactor/Reactor: added a pair of methods that allow a
1508 thread to set/get the notion of who "owns" the event loop. Only
1509 the owner of the loop can do a handle_events() call. Also
1510 changed things so that the requeue_position() are available for
1511 both threaded and non-threaded implementations (just to have a
1512 uniform interface...).
1514 * libsrc/ASX/Message_Block: added a new constructor and a new
1515 init() method that just assume ownership of a char * passed
1516 as the parameter. Note that this does *not* set any other
1517 fields in the Message_Block and is only used to ferry around
1518 totally opaque data in the Message Block!
1520 Tue Nov 7 00:52:15 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1522 * tests/Log_Msg/test_log_msg.C (main): Added some new tests to
1523 make sure that op_status() and errnum() work on ace_log_msg.
1524 Thanks to Tim Harrison for these tests.
1526 Mon Nov 6 12:55:46 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1528 * libsrc/Threads/Thread_Specific: Changed a couple of places in
1529 the Thread_Specific code so that key_ is initialized to 0 and
1530 the ts_obj * is initialized to 0. If these *aren't* 0 then
1531 weird bugs happen on Solaris... Thanks to Tim Harrison for
1532 noticing this and suggesting the fix!
1534 * libsrc/Threads/Synch_T: Added an accessor method to obtain
1535 the underlying mutex within ACE_Condition.
1537 * libsrc/Threads/Synch.C (ACE_Process_Mutex): Added a dummy
1538 argument of type const char * to ACE_Process_Mutex. This is in
1539 anticipation of the info required in Windows NT to support a
1540 named process-wide Mutex. On UNIX, this argument is ignored...
1542 Fri Nov 3 19:02:54 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1544 * libsrc/IPC_SAP/SOCK_SAP: Back from C++ World... Fixed a couple
1545 of typos in SOCK_Dgram_Bcast and SOCK_CODgram that included the
1546 *.i files multiple times... Thanks to Alex V Maclinvosky
1547 <alexm@teltrunk1.tait.co.nz> for reporting this.
1549 Tue Oct 31 02:12:13 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1551 * libsrc/Reactor/Reactor: Modifed the MT_SAFE Reactor so that the
1552 its open() method keeps track of which thread originally created
1553 it. This thread is then considered the "owner" of the Reactor.
1554 If a different thread tries to run the handle_events() method
1555 then an error is returned.
1557 * libsrc/ASX/Task.C: Added a "group id" field to the ACE_Task.
1558 This can be used to suspend and resume a group of tasks
1559 atomically. Also added default suspend() and resume() methods
1560 that do just that!
1562 * libsrc/Reactor/Reactor: Made the open() method
1563 thread-safe. Thanks to Detlef Becker (beckerd@erlh.siemens.de)
1564 for pointing out the need for this.
1566 Mon Oct 30 00:02:53 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1568 * libsrc/Shared_Malloc/Memory_Pool.C (ACE_MMAP_Memory_Pool):
1569 Made it possible to configure the ACE_MMAP_Memory_Pool with a
1570 (backing_store_) file name so that it is possible for multiple
1571 processes to share one wellknown file, as well as to have
1572 multiple pools per process. Check out the
1573 ./tests/Shared_Malloc/test_malloc.C file for an example of how
1574 this works (try the -p -e -L10000 -t3 options).
1576 * libsrc/Reactor/Reactor: Moved handler_i into the *.C file to
1577 avoid problems with inline functions on HP/UX.
1579 * libsrc/Misc/SString: Added a new class called ACE_CString. This
1580 class is similar to ACE_WString (which has "wide character"
1581 size), though CString has regular "char" size. Note that both
1582 of these classes are "true" string classes, unlike ACE_SString,
1583 which is a very simple string class that is only to be used for
1584 very specific purposes...
1586 * libsrc/Shared_Malloc/Memory_Pool.C (acquire): Fixed an
1587 "off-by-one" error in ACE_MMAP_Memory_Pool::acquire() that was
1588 causing an extra byte to be written to the backing store file...
1589 This was causing the ./tests/Shared_Malloc/test_malloc.C program
1590 to fail when -p was given. After this fix the test works
1591 again... (thank God!).
1593 Sun Oct 29 22:43:25 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1595 * libsrc/Shared_Malloc/Malloc: Added two new classes:
1596 ACE_Allocator and ACE_Allocator_Adapter. ACE_Allocator uses
1597 inheritance and dynamic binding to provide extensible mechanisms
1598 for allocating and deallocating memory. ACE_Allocator_Adapter
1599 implements the Adapter pattern to enable ACE_Allocator to be
1600 used with instantiations of the ACE_Malloc<> template class.
1602 Sat Oct 28 13:51:07 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1604 * tests/ASX/Event_Server: Added a new directory called
1605 UPIPE_Event_Server, which tests a version of the Event_Server
1606 that is instantiated with ACE_UPIPE_* IPC mechanisms,
1607 rather than by the ACE_SOCK_* mechanisms.
1609 * libsrc/IPC_SAP/Addr: Created a new file called UPIPE_Addr.h.
1610 This file contains a typedef of ACE_SPIPE_Addr to
1611 ACE_UPIPE_Addr. The purpose of doing this is to "logically"
1612 decouple the ACE_UPIPE* classes from the ACE_SPIPE* classes
1613 (even though they share the same *physical* representation at
1614 this point...).
1616 * libsrc/IPC_SAP: Fixed a small bug in TLI_Connector.connect() and
1617 SOCK_Connector.connect() that failed to set the
1618 new_stream::handle_ to ACE::INVALID_HANDLE when the connection
1619 failed. Thanks to the ever-astute Mark Patton
1620 (mark_patton@tx72.mot.com) for noticing this.
1622 Thu Oct 26 15:08:22 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1624 * tests/Threads: created a new test program called
1625 test_thread_manager.C that tests the new semantics for managing
1626 groups of threads.
1628 * libsrc/Reactor/Signal.C (ACE_Sig_Action): Added a new method
1629 that allows me to create a Sig_Action object that contains both
1630 the handler and the signal to register for. This handler is
1631 register to handle the signal in the constructor of the
1632 object...
1634 * libsrc/Threads/Thread_Manager: enhanced the Thread_Manager to
1635 add support for operations (i.e., suspend, resume, kill) on a
1636 group of threads.
1638 * libsrc/Threads/Thread_Manager: Added a new method that allows
1639 the Thread_Manager to resize itself automatically when it's
1640 internal table gets full.
1642 * libsrc/Threads/Thread_Manager: Updated the return value of
1643 Thread_Manager::spawn() and Thread_Manager::spawn_n() so that
1644 they return -1 on failure *and the group id* on success.
1645 Originally, they returned 0 on success, but this new return
1646 value is more useful since it can be used to control groups of
1647 threads atomically.
1649 Wed Oct 25 01:03:32 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1651 * apps/Name_Server: Began integrating the newly donated name
1652 server into ACE. There are a number of dependencies on
1653 RogueWave that must be removed, but things are looking good so
1654 far.
1656 * libsrc/Misc/Set: Changed the name of ACE_Unordered_Set to
1657 ACE_Fixed_Set to be consistent with the ACE Stack classes...
1659 * libsrc/Misc/Set: Added new ACE classes called ACE_Unbounded_Set,
1660 ACE_Bounded_Set, ACE_Unbounded_Set_Iterator, and
1661 ACE_Bounded_Set_Iterator to deal with ACE name service
1662 stuff.
1664 * libsrc/Misc/SString: Added a new class called ACE_WString that
1665 gives a very simple "wide-character" string representation for
1666 ACE. This is needed for the ACE name service stuff.
1668 * libsrc/Misc/SString.C (operator =): Fixed this so that ::strdup
1669 is matched up with ::free, rather than with delete...
1671 * libsrc/Log_Msg/Log_Msg.C (log): Updated the ACE_Log_Msg class to
1672 store an ostream * in thread-specific storage. This can be used
1673 in conjunction with the dump() method on each ACE class, as well
1674 as with the ACE_ERROR and ACE_DEBUG logging macros.
1676 Tue Oct 24 00:19:13 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1678 * Added a definition of the dump() method to every class in ACE in
1679 anticipation of the changes requested for Siemens.
1681 * Changed all occurrences of ACE_Signal_* to ACE_Sig_* to make the
1682 ACE naming of signal handling wrappers consistent.
1684 * libsrc/Threads/Thread: Fixed some obscure bugs with the way that
1685 ACE_Thread was compiling on platforms that don't support
1686 threads.
1688 * include/ace/config-irix*.h: Removed all TLI support from the SGI
1689 platform. It seems to be totally screwed up from the tests that
1690 I've run and there's no sense in bending over backwards to
1691 support an IPC API that is brain-damaged to begin with...
1693 * apps/Logger/Service_Configurator_Logger/Thr_Server_Logger: Fixed
1694 up the code so that the Thr_Logging_Server stuff will compile
1695 correctly even for platforms that lack threads!
1697 * libsrc/Connection/Strategies: Added new strategies that provide
1698 Singleton creation and Process concurrency policies.
1700 * libsrc/Connection/Connector.C (handle_close): Make sure that all
1701 pending timer objects are removed from the Reactor's timer queue
1702 when a Connector shuts down. In addition, make sure to remove
1703 and delete all dynamically allocated ASTs, as well. Thanks to
1704 Karl-Heinz Dorn (kdorn@erlh.siemens.de) for suggesting this.
1706 * libsrc/Threads/Thread_Manager.C (wait): Modified wait() so that
1707 it now takes an ACE_Time_Value *, which can be used to wait
1708 until all threads terminate or a timeout occurs...
1710 * libsrc/Threads/Thread_Manager: Moved Thr_Descriptor from within
1711 the ACE_Thread_Manager class to become ACE_Thr_Descriptor at
1712 file scope. This is in anticipation of the cool new thread
1713 manager stuff on the way...
1715 * libsrc: Updated every class in ACE to include a "dump" method
1716 with the signature "void dump (void);" This method soon will be
1717 tied into the ACE_Dump mechanism to allow all live objects in
1718 ACE to have their state dumped automagically on-demand. This is
1719 useful for debugging and tracing etc.
1721 * libsrc: Updated every class in ACE to include a "Alloc hook."
1722 This hook will enable all ACE classes to be allocated from a
1723 particular memory pool.
1725 * include/makeinclude/rules.local.GNU: fixed the clean target to
1726 delete the *.rpo files (else gcc -frepo can get some crazy
1727 errors) and combines all rm-commands for the target realclean
1728 into one command (in some cases the last two lines do not have
1729 any files to delete, in which case rm produces an usage
1730 message). Thanks to John Huchinson (hutchiso@epi.syr.ge.com)
1731 for the patch.
1733 * libsrc/IPC_SAP/Addr/INET_Addr: Fixed all the uses of gethost*
1734 and getserv* to use the reentrant get*_r functions if
1735 ACE_HAS_REENTRANT_FUNCTIONS is set and ACE_MT_SAFE is set.
1736 Thanks to Bill Tang <tang@tekats.com> for suggesting this.
1738 * libsrc/Misc/OS: Started adding support for the POSIX *_r
1739 functions to the OS class.
1741 * include/ace: Added a new #define called
1742 ACE_HAS_REENTRANT_FUNCTIONS. This indicates that the platform
1743 supports reentrant functions (i.e., all the POSIX *_r functions
1744 like gethostbyname_r).
1746 Mon Oct 23 21:15:50 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1748 * include/makeinclude/platform_irix5.3.GNU (LIBS): Removed the
1749 link of the nsl lib since it seems not to be present on this
1750 platform. Thanks to Karel Zuiderveld
1751 <Karel.Zuiderveld@cv.ruu.nl> for noticing this.
1753 * libsrc/IPC_SAP/Addr/INET_Addr.i (operator ==): Enhanced the
1754 semantics of comparison to check both the port number and IP
1755 address. Thanks to Mark Patton (mark_patton@tx72.mot.com) for
1756 suggesting this.
1758 Thu Oct 19 00:10:28 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1760 * libsrc/IPC_SAP/Addr/INET_Addr.C (set): Fixed a stupid bug that
1761 failed to set errno appropriately if things go wrong with this
1762 method. Thanks to Mark Patton (mark_patton@tx72.mot.com) for
1763 noticing this!
1765 Sat Oct 14 12:07:16 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1767 * libsrc/Threads/Thread_Manager: Modified the Thread_Manager so
1768 that even in the case where we are compiling for non-MT systems
1769 the Thread_Manager still has the same method interface (all of
1770 which are no-ops).
1772 Wed Oct 11 00:12:57 1995 Douglas C. Schmidt (schmidt@mambo.cs.wustl.edu)
1774 * Finished commenting all the classes in ACE. Updated all manual
1775 pages accordingly.
1777 Fri Oct 6 14:17:17 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1779 * tests/ASX/Event_Server/Event_Server: Changed
1781 ACE_SOCK_Acceptor &sa = (ACE_SOCK_Acceptor &) *this->acceptor_;
1783 to
1785 ACE_SOCK_Acceptor &sa = this->acceptor_->acceptor();
1787 Thanks to Dieter Quehl (quehl@erlh.siemens.de) for suggesting
1788 this.
1790 Thu Oct 5 00:22:56 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1792 * libsrc/ASX/Map_Manager: Added a new find() method to the
1793 Map_Manager. This method only checks for the existence of an
1794 EXTERNAL_ID, and doesn't return the INTERNAL_ID.
1796 * Introduced the new "trait"-based *_Connector, *_Acceptor, and
1797 *_Stream interfaces for all the IPC_SAP classes. Basically, for
1798 each class like
1799 {SOCK,TLI,SPIPE,etc,}_{Acceptor,Connector,Stream}, there's now a
1800 typedef for the appropriate type of ACE_*_Addr subclass and the
1801 appropriate type of ACE_*_Stream class. For example, here's
1802 what's in ACE_SOCK_Acceptor (same goes for ACE_SOCK_Connector):
1804 class ACE_SOCK_Acceptor
1805 {
1806 public:
1807 // ...
1809 // = Traits.
1810 typedef ACE_INET_Addr PEER_ADDR;
1811 typedef ACE_SOCK_Stream PEER_STREAM;
1812 };
1814 Once C++ compilers can grok template typedefs correct, this new
1815 approach will allow much greater simplification of code, so that
1816 template classes like
1818 template <class SVC_HANDLER, class PEER_ACCEPTOR, class PEER_ADDR>
1819 class ACE_Acceptor { /* ... */
1820 virtual int open (const PEER_ADDR &);
1821 };
1823 that are currently used like this:
1825 ACE_Acceptor <My_Svc_Handler, ACE_SOCK_Acceptor, ACE_INET_Addr> acc;
1827 can be replaced with
1829 template <class SVC_HANDLER, class PEER_ACCEPTOR>
1830 class ACE_Acceptor { /* ... */
1831 virtual int open (const PEER_ACCEPTOR::PEER_ADDR &);
1832 };
1834 that are used like this:
1836 ACE_Acceptor <My_Svc_Handler, ACE_SOCK_Acceptor> acc;
1838 i.e., the ACE_SOCK_Acceptor maintains a "trait" that the
1839 ACE_Acceptor uses to determine the appropriate addr. This is
1840 more concise and less error-prone than the current scheme that
1841 requires you to pass into the ACE_INET_Addr separately (ugh).
1843 Note that the trick to making all this work is to typedef the
1844 PEER_ADDR trait into the ACE_SOCK_Acceptor class, as follows:
1846 class ACE_SOCK_Acceptor {
1847 public:
1848 typedef ACE_INET_Addr PEER_ADDR; // Trait...
1849 };
1851 Unfortunately, none of the C++ compilers (e.g., SunC++ 4.0.1,
1852 G++ 2.7.0) support this stuff yet... Therefore, I've added
1853 a set of macros that that will toggle back and forth between
1854 whichever approach works, depending on the setting of
1855 ACE_HAS_TEMPLATE_TYPEDEFS. Hopefully, over time compilers will
1856 support this stuff correctly and life will become easier.
1858 * libsrc/ASX/Message_Queue: Changed all occurrences of the S_MUTEX
1859 and S_CONDITION macros to ACE_SYNCH_MUTEX and
1860 ACE_SYNCH_CONDITION in order to conform to the new ACE namespace
1861 control conventions.
1863 * libsrc/Connection/Acceptor: Modified the Acceptor class to split
1864 it into two classes: ACE_Acceptor (which implements a very
1865 simple, very concise version of the Acceptor pattern, without
1866 all the extra strategy mechanisms, etc.). These strategies have
1867 now been factored out into a new class called
1868 ACE_Strategy_Acceptor. The point of these changes is to "keep
1869 simple things simple, but enable powerful extensions when
1870 necessary." All the test programs that utilized the earlier
1871 strategy version of the Acceptor have been updated to the the
1872 Strategy_Acceptor instead.
1874 * libsrc/Connection/Connector: added two new protected methods,
1875 connect_svc_handler() and activate_svc_handler(). These methods
1876 allow subclasses to override the Connector's strategies for
1877 connection establishment and concurrency. In addition, it makes
1878 the pattern much easier to explain via the Connector pattern and
1879 also is more similar to the Acceptor.
1881 Wed Oct 4 18:45:58 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1883 * apps/Orbix-Examples/Event_Comm: Fixed the examples in this
1884 directory so that they compile with the new ACE.
1886 Tue Oct 3 17:33:46 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1888 * include/makeinclude/rules.nested.GNU: Added support so that
1889 ACE can now be built in parallel using the -jN option of GNU
1890 make. To accomplish this, just type
1892 % make MAKEFLAGS=-j2
1894 on the command line and the "-j2" flag will be passed through to
1895 the make hierarchy. Once the SunC++ compiler supports parallel
1896 makes correctly (Template.DB causes problems) this will allow me
1897 to tke advantage of my new dual-CPU SPARCstation 20!!!
1899 Mon Oct 2 13:34:14 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1901 * libsrc/Connection: Cleaned up some of the code related to
1902 Time_Values * by adding a new method call time_value() to the
1903 ACE_Synch_Options class.
1905 * libsrc/Service_Config/Service_Record: Changed things a bit so
1906 that we don't have to include "ace/Stream.h" in
1907 Service_Record.h, but instead use forward decls of the
1908 appropriate types. This breaks a circular dependency that was
1909 causing problems for GNU C++. Thanks to John Hutchinson
1910 (hutchiso@epi.syr.ge.com) for finding this problem.
1912 Sat Sep 30 13:45:08 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1914 * Changed all uses of typedef PARENT to inherited, which is less
1915 obtrusive.
1917 Fri Sep 29 01:33:54 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1919 * libsrc/Misc: Completed the new OS class, which will be used
1920 shortly to remove all direct UNIX system calls in ACE.
1922 * libsrc: Completed redocumented the header files for almost all
1923 of ACE. The goal was to document all the methods in all the
1924 classes. This has most been achieved (only a few minor changes
1925 remain). The result is *much* better manual page entires in
1926 ./man/man3, as well as much more consistent header files and
1927 overall documentation for ACE.
1929 * libsrc/Reactor/Reactor: Changed all uses of get() to
1930 handler_i(). Also changed the check_connections() method to
1931 check_handles(). Note that all these changes are invisible to
1932 apps...
1934 Thu Sep 28 01:22:36 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1936 * libsrc/IPC_SAP/Addr/SPIPE_Addr: Changed all occurrences of
1937 {get,set}_user_id() and {get,set}_group_id() to user_id() and
1938 group_id() in order to be more consistent with other parts of
1939 ACE.
1941 * libsrc/ASX/Task.C (module): Modified the implementation of
1942 several Task helper methods (e.g., sibling()) so that they don't
1943 crash if there is not Module associated with the Task.
1945 * libsrc/ASX/Stream: Removed the sync_ data member from the
1946 private part of Stream since it didn't seem to be used for
1947 anything.
1949 * libsrc/ASX/Map_Manager.C (bind): Changed the type of the INT_ID
1950 to bind() from INT_ID to const INT_ID &.
1952 * libsrc/IPC_SAP/Addr/Addr: Moved some methods around in the *.i
1953 file to the *.C file in order to be smarter about inlining.
1955 * libsrc/Misc/Auto_Ptr.h: Added a new pair of classes called
1956 "auto_ptr" and "auto_array_ptr". These implement the ANSI/ISO
1957 C++ standard auto_ptr mechanism, which helps to write
1958 exception-safe code. The code is based on material from Jack
1959 Reeves (jack@fx.com) and Dr. Harald M. Mueller
1960 (mueller@garwein.hai.siemens.co.at).
1962 * libsrc/Threads/Synch_T.h (ACE_Null_Condition): Changed the
1963 behavior of Null_Condition::{signal,broadcast} so that they
1964 return 0 rather than setting errno = ETIME and returning -1.
1966 Wed Sep 27 00:16:01 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
1968 * libsrc/Threads/Synch: Fixed a bug with class ACE_Barrier: count_
1969 wasn't being set properly.
1971 * tests/Threads/test_barrier.C (main): Fixed a stupid bug
1972 in the ACE_Barrier -- the main function was exiting, thereby
1973 destroying the barrier!
1975 * tests/IPC_SAP/{SOCK,TLI}_SAP: Added test cases to exercise the
1976 new reuse_addr feature of SOCK_Connector and TLI_Connector.
1978 * libsrc/IPC_SAP/{SOCK,TLI}_SAP/{SOCK,TLI}_Connector: Implemented
1979 the new behavior for reusing a local address. Also fixed a few
1980 error cases that would have lead to descriptor leaks.
1982 * libsrc/IPC_SAP/SOCK_SAP/SOCK.C (open): Modified a few return
1983 values and comparisons to ease the transition to WIN32.
1985 * libsrc/ASX/Task.C (ACE_Task): Changed the behavior of
1986 Task::activate() so that it uses the
1987 ACE_Service_Config::thr_mgr() Singleton if no thread manager has
1988 been associated with a Task when it becomes an active object.
1990 * libsrc/Service_Configurator/Service_Config: Added a new static
1991 method to class Service_Config called thr_mgr(). This static
1992 method behaves as a "Singleton" and provides a convenient
1993 default thread manager that is available to all threads
1994 throughout a process.
1996 * libsrc/IPC_SAP: Changed all the IPC_SAP/*_SAP/*_Connector.[hiC]
1997 files and the Connection/Connector.[Chi] class in order to add
1998 "reuse_addr" behavior consistent with the Reactor. This enables
1999 the client to specify that the "local_addr" should be reused
2000 (e.g., via SOCKREUSEADDR), even if its "2 minute wait" time
2001 hasn't elapsed yet. The libsrc/Connection/Connector.[Chi] files
2002 were also changed to support this new interface.
2004 * libsrc/Log_Msg/Log_Msg.C (log): Fixed a mistake in Log_Record
2005 caused by the new thread-safe storage enhancement. The data
2006 being logged was being rounded up incorrectly. Thanks to Daniel
2007 Proulx (daproulx@qc.bell.ca) for reporting this.
2009 * libsrc/Reactor/Signal.i (operator): Made a minor change to the
2010 definition of operator struct sigaction * to work around a bug
2011 with the HP/UX C++ compiler (lame, lame, lame)...
2013 * libsrc/Service_Config: Changed all occurrences of
2014 ACE_Service_Config::reactor (which was originally a globally
2015 accessible public method of class ACE_Service_Config) to
2016 ACE_Service_Config::reactor () (i.e., made the reactor a
2017 Singleton implemented by a static method). This was long
2018 overdue and will fix all sorts of niggling problems with order
2019 of initialization since the reactor() method can do "lazy
2020 creation" of Reactors now, i.e., create one the first time it is
2021 referenced. In addition, changed the
2022 ACE_Service_Config::svc_rep to ACE_Service_Config::svc_rep () in
2023 the same manner.
2025 * Changed all occurrences of the form:
2027 *_Stream stream;
2028 *_Connector con (stream, remote_addr);
2030 if (stream.get_handle () == ACE::INVALID_HANDLE)
2031 ...
2033 To
2035 *_Stream stream;
2036 *_Connector con;
2038 if (con.connect (stream, remote_addr) == -1)
2039 ...
2041 which is cleaner and less error prone with respect to
2042 UPIPE_Stream (which doesn't have a valid handle once it's
2043 connected).
2045 * Changed all occurrences of get_handle () == -1 to get_handle ()
2046 == ACE::INVALID_HANDLE to help smooth the move to WIN32...
2048 * libsrc: Created a new directory called Shared_Memory and moved
2049 the Shared_Malloc_MM and Shared_Malloc_SV classes from the
2050 ./Shared_Malloc directory to here, where they are now called
2051 Shared_Memory_MM and Shared_Memory_SV. This is a better
2052 name/place for them since they never really had anything to do
2053 with malloc in the first place! What remains the Shared_Malloc
2054 is the Malloc.* and Memory_Pool.* classes, which are typically
2055 used for truly shared malloc/free.
2057 * apps/Gateway/Gateway: added new logic to the Gateway to enable
2058 it to specify which local port to bind() to.
2060 Tue Sep 26 21:17:29 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2062 * libsrc/Connection/Strategies.C (open): Added a return 0; at the
2063 end of the ACE_Thread_Strategy::open method.
2065 * tests/Service_Configurator/IPC-tests/server/Handle_Thr_Stream.C:
2066 Made CLI_Stream a template so that we don't have to worry about
2067 multiple includes of classes...
2069 Mon Sep 25 01:41:27 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2071 * libsrc/IPC_SAP: Revised the close() method of TLI_SAP and
2072 SOCK_SAP so that they don't try to close down a handle that ==
2073 ACE::INVALID_HANDLE.
2075 * Changed all uses of ::free ((char *) ...) to ::free
2076 (ACE_MALLOC_T (...)) to work around inconsistent prototype
2077 problems with some compilers.
2079 * include/ace/sysincludes.h: Changed ACE_MALLOC_TYPE to
2080 ACE_MALLOC_T which is a bit less verbose...
2082 * libsrc/Service_Configurator/Makefile (BUILD): Fixed the argument
2083 to sed from -s (which is invalid) to -e (which is valid).
2084 Thanks to Doug Ritter (dougr@guilder.datalytics.com) for
2085 noticing this. Also used sed to do a more elegant fix on the
2086 age-old problem of inconsistent prototypes for free() and
2087 realloc()...
2089 * libsrc/IPC_SAP/UPIPE_SAP: Rewrote all the UPIPE_* classes to
2090 remove any dependencies on class UPIPE. This class appears to
2091 be unnecessary since it mimics the behavior of class SPIPE. To
2092 simplify the behavior of the UPIPE_* classes, they now inherit
2093 from the SPIPE_* classes where appropriate.
2095 * libsrc/IPC_SAP/UPIPE_SAP: Changed the semantics of the
2096 UPIPE_Connector and UPIPE_Acceptor connection methods so that
2097 they close down the SPIPE_Stream after a connection is
2098 established successfully. This is important to conserve
2099 descriptors.
2101 * libsrc/Misc: Created a new class called OS, which encapsulates
2102 *all* of the UNIX system calls and library routines within a
2103 single class. The rest of ACE will program only to the methods
2104 in this interface, which will make it much easier to port to
2105 other versions of UNIX (and WIN32!).
2107 Sun Sep 24 11:49:47 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2109 * libsrc/IPC_SAP/SOCK_SAP: Conditionally compile the LSOCK_* stuff
2110 if the OS platform (e.g., Linux) doesn't support it.
2112 * Changed all occurrences of ACE_HAS_NO_... to ACE_LACKS_... in
2113 all the config*.h files and the headers/source. This reads
2114 better...
2116 * Changed all uses of (1) Acceptor::peer_acceptor_ to
2117 Acceptor::acceptor(), (2) Svc_Handler::peer_ to
2118 Svc_Handler::peer(), and (3) Connector::peer_connector_ to
2119 Connector::connector() to make the code more abstract and
2120 resilient to future changes.
2122 * include/ace/sysincludes.h: Added a new macro called
2123 ACE_NEW_RETURN that provides a useful abstraction for
2124 expressions involving operator new since we can change memory
2125 allocation error handling policies (e.g., depending on whether
2126 ANSI/ISO exception handling semantics are being used).
2128 * libsrc/Threads/Thread.C: Changed things a bit so that if
2129 ACE_MT_SAFE == 0 then Thread::self() returns 1, regardless of
2130 whether the platform supports threads.
2132 * Went through the entire library and (hopefully) made sure that
2133 all calls to global system calls and library routines are
2134 prefixed with "::".
2136 * libsrc/Misc/Get_Opt.C: Changed this class to use the Log_Msg
2137 logging mechanism rather than stderr...
2139 * tests/Misc: Added a new test program to test the Profile_Timer.
2141 * tests/Reactor/misc: Added a new test program to test the
2142 Handle_Set.
2144 * libsrc/Service_Configurator: Changed the error messages in
2145 Svc_Conf.y and Svc_Conf.l to go to the ACE Log_Msg logging
2146 mechanism rather than stderr...
2148 * libsrc/Connection/Acceptor: Modified the Acceptor and
2149 Oneshot_Acceptor classes so that they take advantage of the new
2150 ACE_Creation_Strategy, ACE_Accept_Strategy, and
2151 ACE_Concurrency_Strategy components. This will make is easy to
2152 define Acceptors that can be flexibly configured to use various
2153 creation strategies (e.g., dynamic linking, singletons, dynamic
2154 memory creation, etc.) for making Svc_Handlers.
2156 * libsrc/Connection/Svc_Handler: Added several new classes --
2157 ACE_Creation_Strategy, ACE_Accept_Strategy, and
2158 ACE_Concurrency_Strategy -- that form the heart of the new
2159 ACE_Acceptor implementation. These classes define the creation,
2160 passive connection acceptance, and concurrency strategies
2161 employed by the Acceptor factory when it receives a connection
2162 from a client.
2164 * libsrc/Connection/Acceptor: Modified the implementation of the
2165 Acceptor and the Oneshot_Acceptor. These classes are no longer
2166 related by inheritance since they behave in fundamentally
2167 different ways and sharing code was making it hard to write and
2168 use these classes correctly.
2170 * libsrc/Misc/Misc: Moved the enum INVALID_HANDLE from IPC_SAP.h
2171 to Misc.h since this is actually more general than just the
2172 sockets stuff and we need to be portable to WIN32... Therefore,
2173 any code that originally said ACE_IPC_SAP::INVALID_HANDLE should
2174 now be ACE::INVALID_HANDLE.
2176 * Removed the ACE_OMIT_SERVICE_CONFIGURATOR flags. These are
2177 annoying -- anyone who wants to subset ACE should be responsible
2178 for doing this stuff.
2180 Fri Sep 22 22:36:33 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2182 * libsrc/Threads/Thread_Spawn.C: Greatly simplified the
2183 implementation of Thread_Spawn by leveraging the new class Task
2184 semantics for activate(). Now, all the concurrency activation
2185 mechanisms necessary to create active objects are associated
2186 with a Task and all the Thread_Spawn needs to do is to define a
2187 new make_svc_handler() Factory Method to create a SVC_HANDLER
2188 and activate it!
2190 * libsrc/ASX/Task.C: Tightened up the semantics of Task::activate
2191 (which turns a passive object into an active object, i.e., one
2192 with its own thread). Now, if an object is activated more than
2193 once it ignores the other requests, unless the force_active
2194 parameter is enabled. Moreover, activate() now takes a
2195 parameter that indicates the number of threads to allocate for
2196 the Task. This makes it simple to have a "thread pool"
2197 associated with a Task.
2199 Thu Sep 21 00:49:55 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2201 * tests/Connection: Updated the CPP-acceptor and CPP-connector to
2202 use the underlying Acceptor and Connector patterns more
2203 robustly. Also added new hooks to include a Service_Config
2204 component so that integrating this with the ACE dynamic linking
2205 scheme will be a snap!
2207 * libsrc/Connection/Acceptor.C: Revised the Oneshot_Acceptor to be
2208 robust in situations where it isn't given a Reactor *...
2210 * apps/Logger: Fixed a couple of minor problems with *.i files
2211 being included when __INLINE__ is set. This should help G++
2212 compiler ACE better...
2214 * libsrc/Reactor/Reactor: Enhanced the Reactor::notify() method so
2215 that it takes both an Event_Hander * and a Reactor_Mask, which
2216 it passes to the Reactor's main event loop thread via the pipe.
2217 The event loop thread uses this mask to determine which method
2218 to invoke. This new feature is due to the insight of Karl-Heinz
2219 Dorn (kdorn@erlh.siemens.de).
2221 * libsrc/Connection/Connector: Fixed a braino in
2222 Connector::handle_output, which was using ::getpeername() to
2223 check whether a connection has been established with a peer.
2224 Naturally, this only works for sockets, and doesn't work at all
2225 for TLI or SPIPEs...
2227 * libsrc/IPC_SAP/Addr/SPIPE_Addr.C (set): Changed the return type
2228 of this method to "int" in order to conform to the types
2229 expected by the Acceptor/Connector patterns.
2231 * Removed all uses of the global scope "::" for all variables and
2232 methods that are not UNIX system calls or library calls in
2233 preparation to change over to the new OS class (requested by
2234 Siemens).
2236 Wed Sep 20 14:39:08 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2238 * libsrc/IPC_SAP/SPIPE_SAP: Renamed the SPIPE_IO classes to
2239 SPIPE_Stream to be consistent with the other parts of the ACE
2240 library. Updated all parts of the library to reflect this
2241 change.
2243 * tests/Service_Configurator: Moved the current contents of this
2244 directory into a new directory called IPC-tests. Created a new
2245 directory called Connection-tests where the dynamic linking
2246 examples for Siemens will go.
2248 * libsrc/Reactor/Signal: Made a bunch of minor changes to support
2249 signal handling on SunOS 4.x platforms. Thanks to Aniruddha
2250 Gokhale <gokhale@cs.wustl.edu> for help with this.
2252 * libsrc/Threads/Synch.h: Modified the #ifdefs a bit so that
2253 Synch_T.h gets included regardless of whether ACE_HAS_THREADS.
2255 * libsrc/Threads/Synch_Options: Added a new set() method to make
2256 it possible to initialize the Synch_Options from outside the
2257 constructor.
2259 * libsrc/Connection/Svc_Handler.C: Added checks within the
2260 ACE_Svc_Handler so that if we are given a NULL Reactor we don't
2261 crash!
2263 * libsrc/Mem_Map: Tidied up this class and added comments to the
2264 header so that the class2man has something to generate manual
2265 pages from!
2267 * libsrc/Misc/Misc: Added a new method called "round_to_pagesize",
2268 which was previously in Mem_Map. This is a better place for it
2269 since other parts of ACE (e.g., Shared_Memory) use it.
2271 * libsrc/Misc/Misc: Added a new method called "get_file_size",
2272 which was previously in Mem_Map. This is a better place for it
2273 since other parts of ACE might want to use it.
2275 Tue Sep 19 00:24:41 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2277 * Released a beta version of ACE 3.3.1 for G++ testing.
2279 * libsrc/IPC_SAP/SOCK_SAP/SOCK_Dgram.C: Removed a stray INLINE
2280 that was causing problems for G++. Thanks to E. Jason Scheck
2281 <jasons@ims.com> for reporting this.
2283 * libsrc/IPC_SAP/Addr/INET_Addr: Improved the documentation for
2284 this class and also added new semantics to the constructor and
2285 set() method so that a "ip_addr:port_number" tuple can be given
2286 as a single string (e.g., "1234:tango.cs.wustl.edu" or
2287 "1234:128.252.166.57"). This is useful since it gives a uniform
2288 interface for addressing for Internet domain, UNIX domain, and
2289 SPIPE domain addresses... Updated the ./tests/Connection tests
2290 to use this form (which is nice since now they are all very
2291 orthogonal!)
2293 * tests/Connection: Created a whole new suite of tests that
2294 exercise the connection patterns for all of the relevant IPC
2295 mechanisms (e.g., SOCK_SAP, TLI_SAP, SPIPE_SAP, and UPIPE_SAP).
2297 * bin/clone.C: fixed the first #include so that it uses #include
2298 "ace/sysincludes.h". Thanks to Alex V Maclinvosky
2299 <alexm@teltrunk1.tait.co.nz> for noticing this.
2301 Mon Sep 18 01:52:07 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2303 * libsrc/ASX/Message_Queue.C (close): Fixed yet another problem
2304 with variables being defined in for loops. G++ is good for
2305 detecting this ;-).
2307 * libsrc/Threads: Moved all the template classes from Synch.* into
2308 Synch_T.* in order to make it possible to compile templates with
2309 G++. Thanks to E. Jason Scheck <jasons@ims.com> for suggesting
2310 this.
2312 * libsrc/IPC_SAP/UPIPE_SAP: Make a number of changes to ensure
2313 that all the UPIPE_SAP classes conform to the same interface as
2314 all the other ACE IPC classes.
2316 * Changed all occurrences of THR_FUNC to ACE_THR_FUNC to protect
2317 the global namespace better.
2319 Sun Sep 17 13:36:23 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2321 * tests/ASX/Message_Queue: Fixed a typo in these test programs
2322 that caused a segfault since a vararg parameter was omitted on a
2323 call to the Log_Msg::log() method. Maybe iostreams isn't so bad
2324 after all... ;-)
2326 * apps/Logger/Service_Configurator_Logger/Thr_Server_Logger: fixed
2327 a stupid error that was caused by some mods I made after ECOOP
2328 to update the threaded logging server. When I changed the
2329 Thr_Logging_Acceptor so that it no longer inherited from the
2330 Logging_Acceptor I forgot to redefine the init() method...
2332 * man: Totally regenerated the ACE manual pages using the new
2333 versions of the OSE tools provided by Karl-Heinz Dorn
2334 (kdorn@erlh.siemens.de).
2336 * Changed all occurrences of MT_SYNCH and NULL_SYNCH to
2337 ACE_MT_SYNCH and ACE_NULL_SYNCH, respectively. This is
2338 consistent with the ACE naming conventions and had been an
2339 oversight when I renamed everything earlier.
2341 * Fully built and tested ACE with the __INLINE__ flag enabled.
2342 This will generate code with many small methods in the library
2343 inlined. Had to fix a bunch of minor things to allow this to
2344 work without compilation-order dependency problems.
2346 * libsrc: Made a bunch of changes to the way that header files are
2347 included internally to ACE in order to break compilation-order
2348 dependencies. This is necessary to support GNU G++'s lame
2349 handling of templates. None of this stuff should affect
2350 application code.
2352 * tests: Added a whole new slew of tests for the remaining parts
2353 of ACE that weren't currently included in ./tests. This stuff
2354 exercises ACE components like the Log_Msg logger and factors all
2355 the #if defined (DEBUGGING) code out of the ./libsrc directory
2356 tree and puts it in the ./tests directory tree, where it
2357 belongs.
2359 * Changed all uses of the error macros LM_* to ACE_* in order to
2360 avoid name collisions with other libraries, frameworks, and
2361 toolkits. In addition, changed all uses of the error enumerals
2362 LOG_ to LM_ in order to avoid a conflict with system #defines in
2363 <sys/syslog.h>.
2365 In order to change these automatically, I used the following
2366 UNIX command sequence:
2368 % find . -type f -print | xargs perl -p -i -e 's/LM_/ACE_/g'
2369 % find . -type f -print | xargs perl -p -i -e 's/LOG_/LM_/g'
2371 Sat Sep 16 11:55:18 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2373 * tests/Threads: Added a new test for Thread_Specific storage.
2375 * Added a new platform/compiler configuration flag called
2376 ACE_TEMPLATES_REQUIRE_SOURCE that must be set for compilers
2377 (e.g., GNU G++) whose template mechanism must be able to
2378 see the source code (i.e., the *.C files). Changed *lots* of
2379 header files to enable this... Also had to change the
2380 corresponding *.C files so that they wouldn't get included
2381 twice...
2383 * libsrc/Connection: redid the implementation of Acceptor,
2384 Connector, and Svc_Handler to get more control over the scope of
2385 #defines like #define SH SVC_HANDLER, etc.
2387 * libsrc/Threads/Thread_Spawn: Totally redid the implementation of
2388 Thread_Spawn to use the Acceptor class template. This greatly
2389 reduces the amount of code to implement the Thread_Spawn!
2391 * libsrc/Threads/Thread_Spawn: Moved the Thread_Spawn
2392 implementation into the ./libsrc/Threads directory rather than
2393 in the ./libsrc/Service_Configurator directory since it deals
2394 with threading and thus belongs in the other place.
2396 Fri Sep 15 00:25:51 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2398 * libsrc/Threads/Token.i (tryacquire): Added a cast to fix passing
2399 a const pointer to a function that was expecting a non-const.
2400 Thanks to E. Jason Scheck <jasons@ims.com> for reporting this.
2402 * libsrc: Added a bunch of changes to enable G++ to compile ACE.
2403 A lot of this involves moving around info in header files so
2404 that templates can be dealt with using the relatively lame GNU
2405 C++ repository scheme. Thanks to E. Jason Scheck
2406 <jasons@ims.com> for all his help in this.
2408 * libsrc/Synch: created Synch_Options.C out of Svc_Handler.C, so
2409 that Svc_Handler.C could be "template pure"; all the other files
2410 were already separated. Thanks to E. Jason Scheck
2411 <jasons@ims.com> for recommending this.
2413 * Makefile (clone): Added a tiny fix that solves a weird problem
2414 that arises with symbolic links on HP/UX. Thanks to Jam Hamidi
2415 (jh1@osi.com) for tips on how to fix this.
2417 Thu Sep 14 10:55:30 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2419 * apps/Gateway/Gateway/Channel.C (route_message): Updated the main
2420 routing code to check for whether a channel is_active()
2421 (i.e., is actually connected) before trying to send messages
2422 to it. This was originally done in the Set class iterator, but
2423 I revised that to make it more reuseable and to workaround bugs
2424 in the HP/UX compiler so I had to update the main code...
2426 * libsrc/Threads/Thread_Specific: Added a new method to called
2427 ts_object () to class ACE_Thread_Specific that get the
2428 thread-specific object for the key associated with this object.
2429 Returns 0 if the data has never been initialized, otherwise
2430 returns a pointer to the data. This is useful since now you can
2431 query a thread-specific storage mechanism to see if there's ever
2432 been a thread-specific object created *without* having to
2433 actually create one if one hasn't been created yet! Thanks to
2434 Detlef Becker (beckerd@erlh.siemens.de) for pointing this out.
2436 * include/makeinclude/platform_sunos5_sunc++_orbix.GNU (LIBS):
2437 Changed the default compilation strategy for SunOS 5.x machines
2438 to use -mt. This seems to be necessary to build robust
2439 libraries that are thread-safe. Thanks to Phil Mesnier
2440 <phil@yakko.envision.com> and Chris Cleeland
2441 <chris@envision.com> for hounding me until I changed the
2442 existing solution. If you *don't* want this behavior you need
2443 to remove -mt from the platform_macros.GNU file and and remove
2444 the ACE_HAS_THREADS and ACE_MT_SAFE flags, etc. from the
2445 config-sunos5.*.h files.
2447 * libsrc/Shared_Malloc/Memory_Pool.C (ACE_MMAP_Memory_Pool): Added
2448 a new parameter to the constructor called write_each_page that
2449 if enabled forces a write to each page to ensure that space is
2450 allocated from the file system (otherwise, we can end up failing
2451 due to optimisitic resource allocation...). Thanks to Phil
2452 Brooks <phil_brooks@mentorg.com> for detecting this issue and
2453 implementing a solution.
2455 * include/ace/sysincludes.h: Added #ifdef support for HP/UX, which
2456 fails to properly wrap <sys/mman.h> with an extern "C" block.
2458 Mon Sep 11 01:39:35 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2460 * libsrc/Connection/Connector.C (handle_input): Changed the call
2461 to Svc_Handler::close() to accept the default value of 0 rather
2462 than -1... Thanks to Mark Seaborn
2463 <mseaborn@itthp1.comm.mot.com> for noticing this.
2465 * libsrc/Connection: Changed all uses of ADDR to PEER_ADDR to
2466 avoid a name clash in Linux... Thanks to James Morris
2467 <jmorris@aurora.apana.org.au> for finding this.
2469 Sun Sep 10 03:56:18 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2471 * libsrc/Threads/Synch: Added support for "barrier
2472 synchronization" to ACE in the form of ACE_Barrier. Thanks to
2473 Bruce Worden (bruce@betsy.gps.caltech.edu) for suggesting this
2474 and pointing me in the right direction.
2476 * tests/Threads: Added a test for the new barrier synchronization
2477 mechanism.
2479 Sat Sep 9 11:58:16 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2481 * libsrc/Reactor/Reactor: Changed Reactor::max to Reactor::max3 to
2482 avoid problems with conflicting names of macros in Sun header
2483 files!
2485 Mon Sep 4 14:34:52 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2487 * libsrc/Reactor/Signal: The new ACE_Signal_Handlers mechanism
2488 appears to be working! There's a test in ./tests/Reactor/misc
2489 called test-signal.C that illustrates how all this works.
2491 * libsrc/Reactor/Signal: Added several new methods to
2492 ACE_Sig_Action to make life easier to implement the new
2493 ACE_Signal_Handler code...
2495 * libsrc/Reactor/Signal.C (ACE_Sig_Action): Changed the order of
2496 the arguments to the ACE_Sig_Action constructor. It's almost
2497 always the case that you want to vary the SignalHandler, but
2498 only rarely do you want to vary the mask or flags. By
2499 reordering this, it's easier to get the correct default values
2500 without adding extra junk...
2502 Mon Sep 4 01:11:29 1995 Tim Harrison (harrison@tango.cs.wustl.edu)
2504 * libsrc/IPC_SAP/SOCK_SAP/SOCK_Dgram_Multicast: added support to
2505 allow a port to be reused for multicast sockets. This is useful
2506 if you are multicasting to multiple processes, some of which are
2507 on the same host!
2509 * libsrc/ASX/Map_Manager: Fixed an odd bug that must have gone
2510 undetected for a long time somehow. Basically, the "is_free_"
2511 field of the ACE_Search_Structure struct was never being set to
2512 it's correct initial value of 1. I don't know how this didn't
2513 surface before... At any rate, it is fixed now...
2515 * include/ace/sysincludes.h: Changed the inline methods for
2516 SET'ing and CLR'ing bits to be macros in order to get cheap
2517 polymorphic behavior... Also changed them to use the prefix
2518 "ACE_" to avoid namespace pollution. Changed all dependencies
2519 in the source code (only a few...).
2521 * libsrc/Reactor/Signal: Added a new class called
2522 ACE_Signal_Handlers, which subclasses from ACE_Signal_Handler.
2523 This new class implements the semantics required for Siemens.
2524 For example, this class allows multiple signal handlers to be
2525 registered for the same signal. It also makes SA_RESTART the
2526 default mode. Note that by default, the Reactor uses the
2527 original ACE_Signal_Handler semantics. If you want the new
2528 semantics, simply pass the Reactor a pointer to
2529 ACE_Signal_Handlers.
2531 * include/ace/sysincludes.h: Removed the automatic inclusion of
2532 ./libsrc/Misc/Misc.h from sysincludes.h since this was causing
2533 problems with include file ordering.
2535 Sun Sep 3 00:22:11 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2537 * libsrc/Reactor/Signal.C (remove_handler): Moved a definition of
2538 ACE_Sig_Action out of an inner block to avoid portability
2539 problems.
2541 * libsrc/ASX/Map_Manager: Improved the documentation of this class
2542 and fixed up a couple of problems with the scope of loop
2543 indexes.
2545 * libsrc/Misc/{Stack,Set}: Added this new file that contains a set
2546 of Stack and Unordered Set implementations. The Unordered Set
2547 is used in various places in ACE (e.g., libsrc/Reactor/Signal
2548 and apps/Gateway/Gateway/Routing_Entry).
2550 * libsrc/Log_Msg/Log_Msg: Moved the definition and declaration of
2551 Thread_Specific<Log_Msg> ace_log_msg from this directory to
2552 ./libsrc/Misc/Misc.h in order to cleanup the namespace and also
2553 to help make this work on DEC platforms...
2555 * libsrc/Reactor/Signal: Changed ACE_Signal_Handler from a static
2556 class to a non-static class in order to allow subclassing. This
2557 is necessary to support the Siemens requirements.
2559 * libsrc/Shared_Malloc/Memory_Pool: Modified ACE_MMAP_Memory_Pool
2560 so that it no longer stores the name of the backing store in a
2561 static character array (which made it impossible to have more
2562 than one of these at a time...). The new version stores the
2563 name in each ACE_MMAP_Memory_Pool object. It also selects a
2564 name that won't conflict with other names by using ::mktemp.
2566 * tests/Shared_Malloc/test_malloc.C (parse_args): Fixed two stupid
2567 omissions of "break" when parsing command-line arguments. Isn't
2568 C++ great?! (NOT)...
2570 * apps/Synch-Benchmarks: Came up with a killer solution to the
2571 nagging problem of POSIX Pthread's lack of an integral thread id
2572 (a la Solaris threads thr_self()). The solution leverages off
2573 of our new ACE_Thread_Specific wrapper to provide the necessary
2574 functionality. Thanks to Reginald S. Perry (perry@zso.dec.com)
2575 for triggering the thoughts that lead to this solution...
2577 Sat Sep 2 17:00:46 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2579 * libsrc/IPC_SAP/TLI_SAP/TLI_Acceptor: Changed the implementation
2580 of ACE_TLI_Acceptor so that ACE_TLI_Request is defined inside
2581 the TLI_Acceptor.C file (this is the so-called "Cheshire Cat"
2582 technique). This clean up the code and the global namespace a
2583 bit.
2585 Fri Sep 1 00:53:47 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2587 * Released new version of ACE 3.3
2589 * libsrc/IPC_SAP/UPIPE_SAP/UPIPE.C: Fixed a small problem with
2590 failing to #ifdef this file correctly if we don't have
2591 threads... Thanks to John Morey (jmorey@hitel.com) for
2592 reporting this.
2594 * tests/ASX/CCM_App/svc.conf: Fixed the svc.conf file so that it
2595 looked in the .shobj directory rather than the .obj directory.
2597 * Released new version of ACE 3.2.9...
2599 * libsrc/Log_Msg/Log_Msg: Added a new field called "restart_" to
2600 the thread-specific storage. This will be used to control
2601 whether system calls are restarted when interrupted.
2603 * Changed inheritance syntax from
2605 class xxx
2606 : public yyyy
2607 {
2608 };
2610 to
2612 class xxx : public yyyy
2613 {
2614 };
2616 so that the OSE tools would work correctly.
2618 Thu Aug 31 00:12:40 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2620 * libsrc/IPC_SAP/UPIPE_SAP: Changed all uses of UPIPE_Addr to
2621 SPIPE_Addr since they were the same thing. This also allows
2622 much reuse of code...
2624 * libsrc/IPC_SAP/{TLI_SAP,SOCK_SAP,SPIPE_SAP}: fixed the
2625 "complete" method for these classes so that it uses the new
2626 ACE::handle_timed_complete() method in libsrc/Misc. This cleans
2627 up the code by merging common logic.
2629 * libsrc/IPC_SAP/{DEV_SAP,FILE_SAP,SPIPE_SAP}: fixed the "connect"
2630 method for these classes to conform to the API used by the
2631 SOCK_SAP and TLI_SAP wrappers. In addition, cleaned up the code
2632 so that common logic was pushed into a new method in libsrc/Misc
2633 called ACE::handle_timed_connect().
2635 Wed Aug 30 00:20:18 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2637 * libsrc/Connection/Acceptor.C (open): Fixed this method so that
2638 it returns a value on all paths through the code...
2640 * libsrc/IPC_SAP/TLI_SAP/TLI.C (ACE_TLI): Moved the option
2641 allocation code from the TLI::open() method into the TLI
2642 constructor in order to make sure it is always called!
2644 * libsrc/IPC_SAP/DEV_SAP/DEV_Connector: Changed the signature of
2645 the DEV_Connector::connect method to allow users to specify
2646 flags, permissions, and timeouts (this is now consistent with
2647 other parts of ACE).
2649 * libsrc/IPC_SAP/FILE_SAP/FILE_Connector: Changed the signature of
2650 the FILE_Connector::connect method to allow users to specify
2651 flags, permissions, and timeouts (this is now consistent with
2652 other parts of ACE).
2654 * tests/IPC_SAP/TLI_SAP: Updated the test code to check the new
2655 timer support for connection establishment.
2657 * libsrc/IPC_SAP/TLI_SAP: Updated the code to make it conform to
2658 the interfaces provided by SOCK_SAP. This primarily affected
2659 the TLI_Acceptor and TLI_Connector classes in order to add
2660 support for timed connects and accepts.
2662 * libsrc/Reactor/Signal.C: Modified Signal_Handler::dispatch so
2663 that it saves/restores errno to prevent it from being corrupted
2664 by the handle_signal callback. Thanks to Detlef for suggesting
2665 this.
2667 * libsrc/Shared_Malloc/Memory_Pool: Changed the name of
2668 ACE_Local_Memory_Pool to ACE_Sbrk_Memory_Pool. Then added a new
2669 version of ACE_Local_Memory_Pool that uses the C++ operator new
2670 to acquire chunks of memory. This enables the ACE Malloc class
2671 to integrate with existing programs that use new/delete. Thanks
2672 to Karlheinz for suggesting this.
2674 * libsrc/IPC_SAP: Added the UPIPE mechanism donated by SIEMENS.
2675 This provides an intra-process IPC mechanism that has the same
2676 API as the interprocess and network mechanisms.
2678 * Reran catman on ./man/windex. Thanks to Dieter Quehl
2679 (quehl@csaserv.erlh.siemens.de) for reporting the need for this.
2681 * Released new version of ACE 3.2.9...
2683 * tests: Fixed a bunch of minor problems that occurred when
2684 building on HP/UX. Thanks to John Morey
2685 (jmorey@hitel.com) for reporting these.
2687 * apps/Gateway/Gateway/Peer_Message.h: Changed the default values
2688 of the parameters to Peer_Addr so they aren't trying to assign
2689 negative values to unsigned chars! Thanks to John Morey
2690 (jmorey@hitel.com) for noticing this...
2692 Tue Aug 29 18:52:17 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2694 * libsrc/Thread/Thread_Specific: Changed the constructor so
2695 that it will take an initial TYPE *, which will be used to
2696 initialize the thread-specific object. This is necessary to
2697 support the changes to ACE_Task described in the following
2698 bullet.
2700 * libsrc/ASX/Task: Added a new class called ACE_Task_Exit to
2701 Task.C. This class is used in conjunction with
2702 ACE_Thread_Specific to keep exit information for a Task in
2703 thread-specific storage. This ensures that the Task::close()
2704 method will get called no matter how the thread exits (e.g., via
2705 Thread::exit() or by "falling off the end of Task::svc_run").
2707 Mon Aug 28 09:54:35 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2709 * Released new version of ACE 3.2.9...
2711 * libsrc/Reactor/Handle_Set.C: changed the type of MSB_MASK from
2712 u_long to ACE_UINT32 to handle problems with the Alpha's 64 bit
2713 longs...
2715 * libsrc/Threads/Thread_Specific: Continued to try and get this
2716 class to build correctly on other platforms... I think I've
2717 just about got it working right now...
2719 * libsrc/IPC_SAP/IO_SAP/IO_SAP: Added installation flags that
2720 indicate whether the platform has terminal ioctl flags like
2721 TCGETS and TCSETS. I know that SunOS 5.x has these, but I'm not
2722 sure about other platforms...
2724 Sat Aug 26 13:55:45 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2726 * Released new version of ACE 3.2.9...
2728 Fri Aug 25 09:05:09 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2730 * libsrc/Threads/Thread_Specific.h: Fixed a typo that was causing
2731 this file to fail on HP/UX. Thanks to Neil Cohen
2732 (nbc@metsci.com) for finding this!
2734 * libsrc/IPC_SAP/DEV_SAP/DEV_IO: Added conditional support for
2735 getmsg/putmsg calls in the DEV_IO class. This will make things
2736 work correctly for platforms that don't support SVR4 STREAM
2737 pipes.
2739 * libsrc/IPC_SAP/FILE_SAP/FILE_IO: Added the same fixes to FILE_IO
2740 that I added to DEV_IO...
2742 * include/ace/sysincludes.h (MAXNAMELEN): Fixed a typo in
2743 sysincludes.h that caused problems for the G++ compiler.
2745 * libsrc/Connection/Acceptor.C (handle_close): Removed a
2746 diagnostic message that was getting printed if a
2747 Oneshot_Acceptor had already been removed from the reactor (it's
2748 ok for this call to fail). Thanks to Irfan Pyarali
2749 (ip1@cec.wustl.edu) for noticing this!
2751 * libsrc/ASX/Task.C (ACE_Task): Fixed a bug in the constructor
2752 that prevented a Message_Queue from being allocated
2753 automagically.
2755 Thu Aug 24 16:47:14 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2757 * include/ace/sysincludes.h: added support for 64 bit machines so
2758 that Internet addresses are 32 bits, as they must be!
2760 * libsrc/Shared_Malloc/Malloc: Changed things so that MALLOC_STATS
2761 is now *off* by default... (also changed the name to
2762 ACE_MALLOC_STATS).
2764 Wed Aug 23 15:21:25 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2766 * include/ace/config-osf1.h: Added support for thread-specific
2767 storage. Please let me know if this breaks on OSF/1!
2769 * libsrc/Threads/Thread: Added support for the thread-specific
2770 storage wrappers for both Pthreads and Solaris threads. Also,
2771 rearranged some of the code to emphasize similarities between
2772 Pthreads and Solaris threads.
2774 * libsrc/Threads/Thread_Specific: Updated this to remove "static"
2775 from all the data members in this class since those should be
2776 specific to an *instance* rather than to an instantiation of the
2777 particular template class. Tim claims this works...
2779 * libsrc/Threads/Token.C: Fixed a couple of typos that misspelled
2780 "assert" (jaysus...). Thanks to David Trumble
2781 (trumble@cvg.enet.dec.com) for noticing this.
2783 * libsrc/Threads/Token.C (release): Fixed a very stupid bug that
2784 was causing the Token never to become "unused"... No excuses
2785 for this, except that Pthreads is partly responsible ;-)
2787 Tue Aug 22 11:36:58 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2789 * libsrc/ASX/Task: Changed the behavior of task so that it doesn't
2790 try to delete the Message_Queue or Thread_Manager if it didn't
2791 allocate them! Thanks to Karl-Heinz Dorn
2792 (kdorn@erlh.siemens.de) for suggesting this fix.
2794 * libsrc/ASX/Message_Block: Changed the semantics for
2795 Message_Block::end() so that it returns a pointer to 1 past the
2796 end of the data. This is more consistent with toolkits like
2797 STL. Also changed the behavior of Message_Block::copy() so that
2798 it checks to make sure the data will fit in its buffer.
2800 * tests and apps: Revised a bunch of files to add #ifdef so that
2801 TLI tests and apps are not compiled if the platform doesn't
2802 support it...
2804 * Fixed a bunch of minor problems for HP/UX (which lacks TLI and
2805 other common OS features). Thanks to Mark Seaborn
2806 (mseaborn@itthp1.comm.mot.com) for noticing these.
2808 Mon Aug 21 00:19:29 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2810 * libsrc/Misc: Changed the name of the class Argument_Vector to
2811 ACE_ARGV. This is a bit more concise... UNIX programmer will
2812 know precisely what argv is...
2814 * Updated all of ACE to use the new naming scheme, which prepends
2815 all ACE classes with "ACE_". This is an important change since
2816 it prevents ACE from polluting the namespace of applications
2817 (unless they also prefix "ACE_" in front of their classes, which
2818 is not very likely!).
2820 Since this change basically affects every single file in the
2821 entire release, as well as all existing user code I've written a
2822 perl script called "rename-ace.pl" to automate all of this.
2823 This script is in $WRAPPER_ROOT/bin. To use it, simply type:
2825 % find . -type f -print | egrep '[Chi]$' | xargs rename-ace.pl
2827 Note that you will need to change the first line of
2828 rename-ace.pl to point it to whereever perl is located on your
2829 system.
2831 I've tested this on the entire ACE source code base and it seems
2832 to work fine. Please be careful using it on your code, however,
2833 since it may conflict with names that you use. When in doubt,
2834 remove the '-pi' from the first line of the rename-ace.pl perl
2835 script and replace it with '-p' (which is non-destructive).
2836 Then run the commands above and check the output carefully.
2837 When you're convinced that everything is ok, add the '-pi' back
2838 again. Let me know immediately if you find any problems with
2839 this scheme!
2841 * libsrc/Connection/Acceptor.C: Added a virtual destructor to the
2842 Oneshot_Acceptor to make sure that descriptors are closed down
2843 correctly. Thanks to Irfan (irfan@wuerl.wustl.edu) for
2844 suggesting this.
2846 * libsrc: Change all occurrences of Shared_Memory to
2847 SV_Shared_Memory to firmly indicate the origins of this
2848 wrapper...
2850 Sun Aug 20 23:12:03 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2852 * libsrc/IPC_SAP: added the IO_SAP, DEV_SAP, and FILE_SAP
2853 components donated by SIEMENS to the ACE release.
2855 * libsrc/ASX: Split the Message_Queue.* files into Message_Block.*
2856 and Message_Queue.* in anticipation of the Windows NT port...
2858 Fri Aug 18 13:54:09 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2860 * libsrc/Threads/Thread_Specific: Move the operator->()
2861 method back into the *.C file in order to make HP/UX
2862 happy...
2864 * apps/Gateway/Peer: Removed the Options.* files since they
2865 weren't being used and they were causing problems on OSF/1...
2867 * libsrc/Misc/Profile_Timer: Factored common code together by
2868 making a new typedef called Rusage that defaults to either
2869 struct rusage or prusage_t, depending on installation flags.
2870 Fixed a couple places in the code that were depending on the
2871 prusage_t type (which is now the Profile_Timer::Rusage type...).
2873 Thu Aug 17 14:31:11 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2875 * libsrc/Connection/Connector.h: Changed some typedefs in order to
2876 keep the Centerline compiler from crapping out. I can't
2877 *believe* how screwed up that compiler is when it comes to
2878 templates...
2880 * Released new version of ACE 3.2.9...
2882 * libsrc/Threads/Synch: Fixed a bunch of typos that showed up
2883 on OSF/1. Also tried to fix some other problems by removing
2884 "const" from all the methods...
2886 Wed Aug 16 22:26:24 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2888 * libsrc/Threads/Thread_Specific.h: Fixed another stupid bug
2889 caused by a braino that failed to unconditionally include
2890 Thread_Specific.i. Stuart Powell found this also... (thanks!).
2892 * libsrc/Threads/Thread_Specific.h: In Thread_Specific.h the
2893 inlining macros were only included if threads were used. This
2894 obviously didn't work for people without threads... I
2895 moved the #endif for ACE_HAS_THREADS... back before the #ifdef
2896 __INLINE__ stuff (e.g. to line 71). Thanks to Stuart Powell
2897 (stuartp@ot.com.au) for suggesting this.
2899 * libsrc/Threads/Synch.h: Fixed a typo that manifested itself for
2900 pthreads: Condition count_nonzero_ should obviously be
2901 Condition<Mutex> count_nonzero_. Thanks to Rob Clairmont
2902 (rclairmo@bnr.ca) for reporting this.
2904 Tue Aug 15 00:31:44 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2906 * libsrc/Log_Msg/Log_Msg.C (log): Fixed a stupid typo (*format++
2907 should have been format++...).
2909 * libsrc/Misc/Trace.h: Move the class TSS_Int from within class
2910 Trace to file scope and changed the name to ACE_TSS_Int to
2911 handle problems with the HP/UX compiler.
2913 * include/ace/sysincludes.h: Added the word "struct" in front
2914 of rusage to make things work on HP/UX. Thanks to Neil Cohen
2915 (nbc@metsci.com) for reporting this fix.
2917 * apps/TokenServer/server/TokenMap.C: Fixed yet another problem
2918 with scope of variables defined in for loops...
2920 * Released new version of ACE 3.2.9...
2922 * libsrc/Connector: Fixed a braino whereby I didn't use consistent
2923 naming for my #defines (PA should have been PRC). Thanks to
2924 Alex (alexey@ace.elektra.ru) for noticing this.
2926 Mon Aug 14 18:13:46 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2928 * libsrc/IPC_SAP/SPIPE_Acceptor: Modified the interface of open()
2929 and accept() so they would conform to the SOCK_Acceptor and
2930 TLI_Acceptor. This makes it possible to use SPIPE_Acceptor in
2931 similar situations (e.g., Acceptor and Connector patterns).
2933 * libsrc/ASX/Stream: Fix a bug that occurred when trying to dump()
2934 a linked Stream.
2936 * libsrc/Connection/Acceptor: Moved the body of the init() method
2937 out of this class since it was unnecessarily dependent on the
2938 INET_Addr domain addressing types. This required changes to the
2939 ./apps and ./tests directory in order to add the init() method
2940 to classes that used the default behavior.
2942 * libsrc/IPC_SAP/SPIPE_SAP/SPIPE_Acceptor.C: Fixed close() so that
2943 it will call fdetach(2) *before* closing down the descriptor. I
2944 hope this will fix a problem noticed by people at SIEMENS.
2946 * tests/ASX/Event_Server/Event_Server/Options: Fixed a problem
2947 with the default port numbers (they weren't using the values
2948 from ./include/ace/testconfig.h).
2950 * include/ace/sysincludes.h (MAXNAMELEN): If MAXNAMELEN is not
2951 defined by a platform then ACE sets it to be FILENAME_MAX,
2952 which should be defined in stdio.h. Thanks to Todd Blanchard
2953 (tblancha@evolving.com) for this suggestion.
2955 Sun Aug 13 17:02:57 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2957 * libsrc/Service_Configurator/Parse_Node: Changed the name of
2958 Function_Node::symbol (const void *) so that it won't give
2959 those annoying warnings anymore...
2961 * libsrc/Reactor/Handle_Set: Moved the definition of MSB_MASK from
2962 the header file into a static const within the .C file in order
2963 to avoid overflow problems on certain compilers.
2965 * libsrc/Threads/Synch: Implemented bare-bones versions of
2966 Semaphore and RW_Mutex for the POSIX Pthreads wrappers so that
2967 they'll be more compatible with the Solaris threads wrappers.
2968 The semaphore implementation uses a Condition object and a
2969 Mutex, which should be a reasonable solution. The RW_Mutex is a
2970 cop-out for now and just uses a Mutex (i.e., no extra
2971 parallelism for readers...). If anyone has a good
2972 implementation of RW_Mutex that they'd like to share please let
2973 me know.
2975 * libsrc/Threads/Thread_Specific: Fixed the prototypes for copy
2976 constructor and operator=, which were broken... Thanks to Alex
2977 (alexey@ace.elektra.ru) for noticing this.
2979 * libsrc/Shared_Malloc/Memory_Pool.C: Added some casts to
2980 MAP_FAILED to handle OSF/1. Thanks to Alex
2981 (alexey@ace.elektra.ru) for noticing this.
2983 * libsrc/Threads/Token: Fixed things so that threads waiting for a
2984 token wouldn't get screwed up by signals that occur...
2986 * include/ace/sysincludes.h: Changed the #ifdef
2987 ACE_SELECT_USES_LONG to ACE_SELECT_USES_INT for HP/UX since
2988 believe it or not, it really does use int, not long!
2990 * libsrc/SV_Semaphores: Fixed some weird problems that the HP/UX
2991 compiler was having when trying to inline methods in this class.
2992 As a consequence, I've rearranged the class to avoid inlining
2993 non-trivial methods. Thanks to John Morey (jmorey@hitel.com)
2994 for reporting these problems.
2996 Wed Aug 9 01:29:16 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
2998 * libsrc/Threads/Token: Added a tryacquire() method to become
2999 interface compliant with other LOCK mechanisms.
3001 Sat Aug 5 09:18:29 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3003 * libsrc/Connection: Fixed a couple of bugs when using
3004 the Acceptor and Svc_Handler when ACE_OMIT_SERVICE_CONFIGURATOR
3005 is enabled. Thanks to Stuart Powell (stuartp@ot.com.au) for
3006 bringing this to my attention.
3008 * include/ace/sysincludes.h: Added a #ifdef for MAXNAMELEN to
3009 handle systems (like HP/UX) that don't support it.
3011 Thu Aug 3 22:59:13 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3013 * Released new version of ACE 3.2.9...
3015 * libsrc/Reactor/Time_Value.i (normalize): Added new code to
3016 perform normalization of Time_Values. Thanks to Hans Rohnert
3017 (Hans.Rohnert@zfe.siemens.de) of SIEMENS for the suggestion.
3019 Tue Aug 1 00:19:00 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3021 * Changed all code that used Log_Msg::log() directly to use
3022 LM_ERROR or LM_DEBUG. This will ensure that logging will work
3023 with the new thread-specific storage implementation.
3025 * libsrc/Threads/Thread_Manager.i (open): Fixed this method so
3026 that it is thread-safe when a Thread_Manager is resized.
3028 * libsrc/ASX/Map_Manager.i (open): Fixed this method so that it is
3029 thread-safe when a Map_Manager is resized.
3031 Mon Jul 31 12:56:17 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3033 * libsrc/Misc/Trace: Reimplemented the ACE Trace class to use the
3034 new thread-specific storage wrapper.
3036 * libsrc/Log_Msg: Reimplemented the Log_Msg class to use the new
3037 thread-specific storage wrapper. This shouldn't affect any
3038 existing code that was careful to only use the LM* macros...
3040 * libsrc/Threads: Added a new class called Thread_Specific which
3041 implements a C++ wrapper for SunOS 5.x thread-specific data
3042 (this will also work for POSIX pthreads I believe). Thanks to
3043 Tim Harrison (harrison@cs.wustl.edu) for coming up with the idea
3044 for using C++ "smart pointers" to implement this.
3046 * libsrc/Reactor/Timer_Queue.h: Moved Timer_Node from within
3047 Timer_Queue to outside Timer_Queue and renamed it to
3048 ACE_Timer_Node. Unfortunately, some compilers still don't like
3049 nested classes (ugh)...
3051 * Changed ACE_Synch_Options to be simply Synch_Options since it
3052 is *not* an ACE private class...
3054 Sun Jul 30 00:07:28 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3056 * include/makeinclude/README: Added a new #define called
3057 ACE_HAS_THREAD_SPECIFIC_STORAGE, which does exactly what it
3058 sounds like! So far, I know that Solaris defines this. I'm not
3059 sure which other platforms do (perhaps OSF/1 does?).
3061 Fri Jul 28 14:53:45 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3063 * Fixed a bunch more problems with loop variables that were
3064 uncovered with GCC 2.7. Thanks to Matt Stevens
3065 (mstevens@ent.mrj.com) for noticing this.
3067 * apps/Logger/Service_Configurator_Logger: Added a new flag called
3068 ACE_HAS_NO_STATIC_DATA_MEMBER_TEMPLATES which is necessary to
3069 workaround bugs with GNU G++... Thanks to Matt Stevens
3070 (mstevens@ent.mrj.com) for noticing this.
3072 * libsrc/Connection/Connector.C (cleanup_AST): Changed the
3073 parameter list just a tad to try and fix a problem with
3074 G++ (which doesn't seem to like unscoped typedefs in
3075 argument lists or return values.
3077 * libsrc/Misc/Profile_Timer.i: Changed ::getrusage to be getrusage
3078 to avoid problems with macros and scope operators...
3080 * include/ace/testconfig.h (ACE_DEFAULT_RENDEZVOUS): Changed the
3081 value from /tmp/foo to /tmp/fifo.ace. Thanks to Neil B. Cohen
3082 (nbc@metsci.com) for suggesting this.
3084 Thu Jul 27 12:30:06 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3086 * apps/TokenServer/server/TokenHandler.C (abandon): Fixed a minor
3087 bug that caused assert() to fail when a client abandoned a
3088 Token.
3090 * libsrc/Reactor: Fixed a bug with the design of the Reactor's
3091 Timer_Queue cancellation mechanism. The new mechanism is much
3092 more robust since it ensures that timer_ids (used to cancel
3093 pending timers) are unique up to values of greater than 2
3094 billion timers! As long as timers don't stay around longer than
3095 this there should be no problems with accidentally deleting the
3096 wrong timer. Thanks to Bill Sears (wsears@world.std.com) for
3097 noticing this problem.
3099 One nice consequence to this change is that legacy code that
3100 would have broken with the previous change is now 100%
3101 compatible!
3103 * Provided a definition of ACE_Synch_Options::arg(const void *),
3104 which I'd forgotten to define before (darn templates...).
3105 Thanks to Tim Harrison (harrison@cs.wustl.edu) for finding this!
3107 * include/ace/sysincludes.h: Added a total hack to get HP/UX to
3108 understand getrusage(). The trick is to use the following
3109 undocumented syscall:
3111 #define getrusage(a, b) syscall(SYS_GETRUSAGE, a, b);
3113 What a hack (but it works...).
3115 Tue Jul 25 13:20:58 1995 Douglas C. Schmidt (schmidt@kavita.cs.wustl.edu)
3117 * libsrc/Misc/Profile_Timer: Fixed a typo in Profile_Timer.h
3118 (ACE_HAS_RUSAGE_T should have been ACE_HAS_GETRUSAGE). Thanks
3119 to George Reynolds (george@dvcorp.com) for noticing this.
3121 * Changed all uses of ::getopt() to use the ACE class Get_Opt
3122 get_opt. Also changed all uses of optarg to get_opt.optarg.
3123 Thanks to Bob Vistica (robertv@ims.com) for detecting and
3124 reporting some inconsistencies in my previous changes...
3126 Mon Jul 24 19:03:03 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3128 * include: I've added config and platform support for SGI IRIX5.3
3129 for use with the SGI C++ compiler. Thanks to Stuart Powell
3130 (stuartp@ot.com.au) for these config files.
3132 * libsrc/Connection/Acceptor.C (handle_input): Added the
3133 Event_Handler::DONT_CALL flag when removing the Oneshot_Acceptor
3134 from the Reactor so that we don't set it's peer_acceptor_ to -1.
3136 Mon Jul 24 12:46:10 1995 Douglas C. Schmidt (schmidt@kavita.cs.wustl.edu)
3138 * libsrc/Misc/Profile_Timer: Fixed problems stemming from the fact
3139 that HP/UX doesn't seem to support getrusage() (how odd).
3141 * tests/Connection: a new test directory to test out the Connector
3142 and Acceptor pattern implementations.
3144 Sun Jul 23 12:26:37 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3146 * libsrc/Connection/Acceptor.C: Fixed a minor bug with the
3147 Oneshot_Acceptor constructor. Thanks to John Morey
3148 (jmorey@hitel.com) for noticing this and the one below.
3150 * libsrc/Reactor/Timer_Queue: Moved the static method current_time
3151 from the Timer_Queue.i file to the Timer_Queue.C file. This
3152 avoids a bug with HP/UX C++.
3154 Sat Jul 22 15:54:27 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3156 * Released beta version 3.2.9. Once this compiles on all major
3157 platforms then it's time to move to version 3.3.
3159 * Added a number of miscellaneous changes for OSF/1 courtesy of
3160 David Trumble (trumble@cvg.enet.dec.com). One of these changes
3161 involved changing all uses of Reactor::{ADD,SET,CLR,GET} to
3162 Reactor::{ADD,SET,CLR,GET}_MASK in order to avoid a class with
3163 some symbols in OSF/1.
3165 Fri Jul 21 00:21:02 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3167 * Finally was able to test ACE on the SunOS 4.x platform (using
3168 the SunC++ 4.0.1 compiler). The entire release compiled
3169 correctly! This is a great relief after fighting with this
3170 stuff for months... If you are building ACE on SunOS 4.x I'd
3171 *strongly* recommend you get SunC++ 4.0.1 since it is the only
3172 compiler that seems to be able to grok the weirdness of SunOS
3173 4.x...
3175 * Finished updating all the ./tests and ./apps code so that it
3176 works correctly with the new Acceptor/Connector modifications.
3177 To see examples of these changes check out
3178 ./tests/IPC_SAP/SOCK_SAP/CPP-{nbclient,inserver}.C.
3180 * libsrc/Misc/Profile_Timer: Merged in the new Profile_Timer
3181 implementation from David Trumble (trumble@cvg.enet.dec.com).
3182 This will work with basically the same interface as the current
3183 scheme even if the OS platform doesn't support the prusage_t
3184 type...
3186 Thu Jul 20 01:07:23 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3188 * libsrc/Connection: Added a new class called ACE_Synch_Options
3189 which is used in both the Acceptor and Connector classes to
3190 consolidate options related to synchronous and asynchronous
3191 behavior.
3193 * libsrc/Connection: Added a new class called Oneshot_Acceptor.
3194 This class inherits from the Acceptor but it only accepts one
3195 connection at a time (i.e., it doesn't keep itself registered
3196 with the Reactor). Several examples in the tests/SOCK_SAP
3197 directory have been added to illustrate how all this works.
3199 * libsrc/Connection/Connector: completely redid the Connector and
3200 Acceptor class interfaces to incorporate the new changes for
3201 asynchronous and synchronous behavior. The new scheme should be
3202 much more general (and correct...). Thanks to Tim and Irfan for
3203 helping out with this.
3205 * libsrc/Threads/Synch: Fixed the implementation of Recursive_Lock
3206 so that it won't have race conditions when testing the thread id
3207 and nesting level in parallel threads.
3209 Wed Jul 19 13:15:05 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3211 * libsrc/Reactor/Timer_Queue: Added new support for cancellation
3212 of individual Event_Handlers in the Reactor. In addition, it is
3213 now possible to cancel all Event_Handlers with a single call to
3214 Timer_Queue::cancel(). Note that this new scheme will break
3215 existing code since Timer_Queue::schedule() (and
3216 Reactor::schedule_timer()) now return ACE_Timer_Node *'s rather
3217 than ints...
3219 * libsrc/Service_Configurator/Service_Config: Fixed a problem
3220 where the "-s" option didn't work since it was parsed when
3221 "Service_Config::open" was called
3222 (it sets Service_Config::signum_). However, previously the
3223 signal handler was already setup by the constructor. By moving
3224 this registration to the open() method the problem was solved.
3225 Thanks to Bob Vistica (robertv@ims.com) for noticing this.
3227 * libsrc/Service_Configurator/Service_Manager: Changed the
3228 Service_Manager::reconfigure_services to use this->signum_
3229 rather than to hard code SIGHUP. Thanks to Bob Vistica
3230 (robertv@ims.com) for this insight.
3232 Mon Jul 17 12:08:08 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3234 * libsrc/Threads/Token.C (renew): Fixed a typo so that we compare
3235 this->head_ == 0 rather than this->head_->next_ == 0.
3237 * libsrc/Reactor/Reactor.h: Changed the Reactor to use the
3238 Reactor_Token rather than the pure Token to make sure that the
3239 sleep_hook() is called to unblock the Reactor.
3241 Fri Jul 14 14:12:07 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3243 * libsrc/Connection: Changed the default behavior of the Connector
3244 and Acceptor classes such that they pass a -1 to
3245 svc_handler_->close() when things go wrong. This is useful as a
3246 flag to close() (e.g., if it needs to figure out what to do if
3247 is shuts down prematurely).
3249 * libsrc/Connection/Svc_Handler.C: Added a default definition of
3250 the put() method, which is defined as a pure virtual method in
3251 class Task.
3253 Thu Jul 13 23:10:35 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3255 * include/ace/sysincludes.h: Added support for the SGI IRIX 5.3
3256 OS. More screwy gettimeofday() hacks (what a pain...). Thanks
3257 to Matt Stevens (mstevens@kirk.softeng.infonautics.com) for
3258 these fixes.
3260 * include/makeinclude/platform_sunos5_centerline.GNU: Fixed a
3261 vexing problem that was causing the Centerline C++ compiler to
3262 puke when compiling ACE. Thanks to Chandra Venkatapathy
3263 (cvenkat@develop.bsis.com) for finding a solution!
3265 Tue Jul 11 00:01:15 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3267 * Released version 3.2.6 for use by Karl and Detlef.
3269 Mon Jul 10 00:28:51 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3271 * apps/TTCP: Added support for benchmarking the performance of
3272 Orbix, ORBeline, ACE SOCK_SAP, and C sockets. These results are
3273 available at URL http://www.cs.wustl.edu/~schmidt/COOTS-95.ps.Z
3275 * libsrc/Service_Configurator/Makefile: Added new commands to the
3276 Service Configurator Makefile so that all automatically
3277 generated flex and yacc symbols (i.e., the "yy" stuff) is
3278 renamed "ace_yy". The purpose is to avoid conflicts with other
3279 uses of flex/yacc lexers/parsers with ACE. Thanks to Steve
3280 Ritter (ritter@titan.com) for this suggestion and code.
3282 * libsrc/IPC_SAP/TLI: Changed the TLI files so they are
3283 conditionally compiled only if the platform supports TLI...
3285 * libsrc/Connection: Changed the inlining strategy of Connector,
3286 Acceptor, and Svc_Handler so that very short methods are always
3287 inlined, but anything larger is never inlined...
3289 Sun Jul 9 14:07:02 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3291 * include/ace/testconfig.h: Changed all occurrences of DEFAULT_*
3292 to ACE_DEFAULT_ to prevent namespace collision.
3294 * libsrc/Connection/Connector: Massively improved the Connector
3295 class so that its connect() method supports the following
3296 behavior (tv == "timeout value" and ur == "use Reactor"):
3298 Parameters | Description
3299 |
3300 tv | ur |
3301 -----|----------|-------------------------------
3302 | |
3303 NULL | yes | infinite timeout (using Reactor)
3304 | |
3305 time | yes | try asynch transaction for
3306 | | the specified time (using Reactor)
3307 | |
3308 0,0 | yes | poll; try, if EWOULDBLOCK,
3309 | | then return immediately
3310 | | (using Reactor)
3311 | |
3312 NULL | no | block forever (don't use Reactor)
3313 | |
3314 time | no | do a blocking transaction
3315 | | for the specified time
3316 | | (don't use Reactor)
3317 | |
3318 0,0 | no | poll; but do not initiate a
3319 | | nonblocking transaction
3320 | | (don't use Reactor)
3322 * libsrc/IPC_SAP/SOCK_SAP/SOCK_Connector: Changed the behavior of
3323 the SOCK_Connector::connect() method so that it uses Time_Values
3324 rather than a simple flag that indicates whether or not to use
3325 non-blocking connectors. The new scheme is an improvement since
3326 it allows greater control over synchronous and asynchronous
3327 timeouts for connection establishment. This behavior parallels
3328 that of SOCK_Acceptor, as well! If this works well then I'll
3329 update the TLI_Connector and SPIPE_Connector to match this
3330 interface.
3332 * libsrc: Continued to improve ACE's use of the global name space
3333 by prefixing all "helper" classes with "ACE_"
3335 * Updated SPIPE_Acceptor so that it would have the same basic
3336 interface as the SOCK_Acceptor and TLI_Acceptor.
3338 * libsrc/IPC_SAP: Consolidated the handle_timed_wait() methods
3339 used by the TLI, socket, and STREAM pipe wrappers so that they
3340 share the new ACE::handle_timed_accept() method, which is in
3341 libsrc/Misc.
3343 * libsrc/Misc: Modified the structure of Misc.[Ch]. Originally,
3344 this file contained a bunch of stand-alone C functions with the
3345 prefix "ace_" to keep them from conflicting with user's code.
3346 I've modified things now so that all the miscellaneous functions
3347 are now static methods in class ACE. This provides better scope
3348 control...
3350 * libsrc/Log_Msg/Log_Msg: Replaced the use of a Mutex in class
3351 Log_Msg with a Recursive_Lock<Mutex>. This is necessary to
3352 handle signals correctly...
3354 * libsrc/Reactor/Reactor: Modified the behavior of
3355 Reactor::notify() so that writers will block if the pipe is
3356 full. This solves some nasty flow-control problems.
3358 * libsrc/Reactor/Handle_Set: changed all uses of fd_set to
3359 ACE_FD_SET_TYPE * so that HP_UX would work correctly...
3361 * include/ace/config-hpux.h: Removed the ACE_HAS_XLI flag until I
3362 get a better idea which HP systems this is installed on.
3364 * libsrc/Threads/Thread: Added a new static method called
3365 spawn_n() that spawns "n" threads all running the same function.
3367 Sat Jul 8 14:14:34 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3369 * libsrc/Service_Configurator: Made some miscellaneous changes to
3370 "const" methods and parameters in order to handle the new
3371 Reactor changes (C++ can be such a pain about this
3372 sometimes...).
3374 * libsrc/Threads: Added "yield" and "sigsetmask" methods to class
3375 Thread. How did I manage to omit these before?!
3377 * libsrc/Reactor: Updated the Reactor to use the new Token class
3378 described below. This greatly simplies the structure of the
3379 multi-thread support in the Reactor code. In addition, it
3380 should improve performance because it cuts the number of context
3381 switches compared with the old scheme. Many thanks to
3382 Karl-Heinz and Detlef for encouraging me to redo the Reactor
3383 implementation.
3385 * libsrc/Threads: Added a new class called Token that provides a
3386 flexible and efficient recursive mutex scheme. Thanks to
3387 Karl-Heinz Dorn (kdorn@erlh.siemens.de) and Detlef Becker
3388 (beckerd@erlh.siemens.de) for sharing their original code for this.
3390 Thu Jul 6 10:37:45 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3392 * libsrc/Reactor/Event_Handler_T: Conditionally compiled this so
3393 that it will only compile if the compiler supports template
3394 typedefs (e.g., G++ doesn't seem to like this...).
3396 * Started to make changes in ACE to deal with the new ANSI C++
3397 semantics in the scope of variables defined within for loops.
3398 Thanks to Aniruddha Gokhale (gokhale@cs.wustl.edu) for noticing
3399 this in G++ 2.7...
3401 Wed Jul 5 21:50:39 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3403 * libsrc/Connection/Svc_Handler: Fixed the behavior of the
3404 Svc_Handler class so that it can be configured with a Reactor
3405 other than Service_Config::reactor, just like the Acceptor and
3406 Connector. Thanks to Karl-Heinz Dorn (kdorn@erlh.siemens.de)
3407 for suggesting this!
3409 Tue Jul 4 00:21:31 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3411 * Added some new changes for Linux courtesy of Charles Rennolet
3412 (clr@thurse.mn.org).
3414 * libsrc/Reactor/Reactor.C (bit_ops): Fixed a dumb error in
3415 bit_ops that was caused by checking "else if (Reactor::SET)"
3416 rather than "else if (ops == Reactor::SET)". Thanks to Mark
3417 Patton (mark_patton@tx72.mot.com) for finding this bug and
3418 reporting it along with the fix.
3420 * include/ace/sysincludes.h: Added new fixes for M_SYNC and
3421 ENOTSUP on SunOS 4.
3423 * libsrc/Reactor: Added new support for integrating X and the
3424 Reactor. These files are called XtReactor.* and XReactor.*.
3425 Thanks to Eric Vaughan (evaughan@arinc.com) for providing this
3426 stuff...
3428 Mon Jul 3 19:44:07 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3430 * Added support for SunOS 5.5 on i86 PC. Thanks to Bin Mu
3431 (mubin@wfg.com) for the config files and fixes.
3433 * libsrc/Misc/Profile_Timer: Added new support for versions of
3434 UNIX that don't support prusage_t. The new version of
3435 Profile_Timer will use gettimeofday() and getrusage() of
3436 prusage_t and /procfs isn't available... This code hasn't been
3437 tested yet (since I don't have SunOS 4.x) so I don't know if it
3438 works.
3440 * Hopefully fixed the SunOS4 prototype for gettimeofday(). Thanks
3441 to Andrew McGowan (ajm@se09.wg2.waii.com) for the suggested fix.
3443 * include/makeinclude/platform_hpux.GNU (CC): Added new support
3444 for building ACE and shared libraries on HP/UX. Thanks to Jam
3445 Hamidi (jh1@osi.com) for this code.
3447 * libsrc/Reactor/Handle_Set: Fixed Handle_Set_Iterator::operator++
3448 to check for index to be greater or equal than NUM_WORDS instead
3449 of just equal. This is better for sanity, although it may not be
3450 needed. Thanks to Carlos Garcia Braschi (cgarcia@caramba.tid.es)
3451 for suggesting this fix.
3453 Wed Jun 14 11:16:40 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3455 * libsrc/IPC_SAP/TLI_Acceptor: modified TLI_Acceptor so that it's
3456 public constructor/open() and accept() methods are more similar
3457 to the SOCK_Acceptor. In particular, added support for
3458 "SO_REUSEADDR" in TLI_Acceptor and time values for accept().
3459 Thanks to John P. Hearn (jph@ccrl.nj.nec.com), TLI_SAP now does
3460 the right thing for socket options.
3462 * libsrc/IPC_SAP: Fixed up the SOCK_Acceptor and TLI_Acceptor in
3463 order to add better support for "timed accepts." The new scheme
3464 uses Time_Values, which is more consistent with other parts of
3465 ACE. In addition, the Time_Value was moved from the
3466 constructor/open() to the accept() method, which allows more
3467 fine grained control over this behavior. Thanks to Irfan
3468 (ip1@cs.wustl.edu) for suggesting this.
3470 * libsrc/IPC_SAP: Cleaned up the SOCK_Acceptor::open and
3471 TLI_Acceptor::open routine. In addition to being more compact
3472 and robust, this routine now also let's open() determine which
3473 local port to bind to (if you pass in Addr::sap_any as the
3474 local_addr). Thanks to Irfan (ip1@cs.wustl.edu) for suggesting
3475 this.
3477 Tue Jun 13 16:09:13 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3479 * libsrc/Reactor/Timer_Queue: Fixed an odd problem with the
3480 Timer_Queue on Solaris. Apparently the select() call will
3481 return slightly earlier than the timeout dictates due to lack of
3482 granularity with the system clock. This was causing problems
3483 where the Reactor wasn't correctly dispatching the
3484 handle_timeout() method of Event_Handlers. The fix was to add a
3485 10 Millisec "fudge factor" when calling the
3486 Timer_Queue::expire() method. Please let me know if this causes
3487 any problems. Thanks to Giang Hoang Nguyen
3488 (yang@titan.com) to noticing this problem.
3490 * libsrc/Connection/Acceptor: Enhanced the Acceptor class so that
3491 it is possible to accept a new SVC_HANDLER synchronously, rather
3492 than always driving this acceptance out of the Reactor's event
3493 loop. Thanks to Irfan (ip1@cs.wustl.edu) for suggesting this.
3495 * libsrc/IPC_SAP/Addr: Changed all occurrences of ace_sap_any to
3496 Addr::sap_any (i.e., sap_any is now a static data member in
3497 class Addr). This is an improvement since it helps to limit the
3498 scope of what would otherwise be global variables...
3500 Wed Jun 7 17:26:31 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3502 * apps/Logger/Reactor_Logger/Client_Acceptor.i (handle_input):
3503 Fixed a braino that omitted an important block of code! Thanks
3504 to Ken Konecki (kenk@wfg.com) for finding this.
3506 Fri Jun 2 13:59:07 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3508 * libsrc/IPC_SAP/Addr/INET_Addr.i (set): Removed the special case
3509 for INADDR_ANY. This should just fall right out...
3511 Thu Jun 1 19:45:21 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3513 * libsrc/IPC_SAP/SOCK_SAP/SOCK_Connector: Added support to
3514 SOCK_Connector so that it is possible to bind the local TCP port
3515 number prior to establishing the connection. Also fixed the
3516 TLI_Connector to be consistent with this approach. Thanks to
3517 Mark Patton (mark_patton@tx72.mot.com) for this idea.
3519 * Fixed a problem with Mem_Map that was caused by the fact that
3520 some versions of Unix (e.g., SunOS 4.x) don't support MS_SYNC...
3521 Thanks to Andy McGowan (mcgowan@wg2.waii.com) for noticing this.
3523 * Fixed a very stupid bug in ./libsrc/Synch.h that accidentally
3524 omitted this->lock_.acquire() from the Guard class constructor.
3525 Thanks to Bin Mu (mubin@wfg.com) for noticing this!
3527 Fri May 26 13:20:38 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3529 * libsrc/Connection/Acceptor: Added a new method called
3530 make_svc_handler() that generalizes the dynamic creation of a
3531 SVC_HANDLER by the Acceptor's Template Method handle_input().
3532 This scheme is completely backwardly compatible with the
3533 original approach, but now allows transparent extension too!
3535 * libsrc/Connection/Acceptor: cleaned up the Template Method logic
3536 in Acceptor::handle_input. No longer do we set the listener
3537 socket into non-blocking mode and then rely on a -1 with
3538 EWOULDBLOCK to indicate there are no more connections to
3539 establish. This was causing problems for singleton Svc_Handlers
3540 since the Acceptor was setting their peer_stream_ to -1... The
3541 new approach should complete solve this problem.
3543 Mon May 22 15:10:27 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3545 * Released version 3.2.3 so that Bill Sears can test the new
3546 #include "ace/" features.
3548 * Modified all the ACE source, tests, and apps so that all
3549 #includes of its headers are refixed by ace/, e.g.:
3551 #include <ace/Reactor.h>
3553 There are two reasons for have the include files be of
3554 this form:
3556 First, you can tell at a glance by looking at the caller where
3557 the library is coming from.
3559 Secondly, you can build an include tree of the form:
3561 /include/lib1
3562 /include/lib2
3563 /include/lib3
3564 /include/ace
3566 which then links to WRAPPER_ROOT. Now, all you have to do
3567 when you build a make file is point to the include root, and
3568 put links in the include root. This is especially helpful
3569 if there are multiple versions.
3571 This means just one less thing that has to be modified in the
3572 Makefile. Thanks to Bill Sears (wsears@world.std.com) for
3573 recommending this change.
3575 Sat May 20 17:12:35 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3577 * config-irix5.3-sgic++.h: Added new configuration support for SGI
3578 IRIX5.3 courtesy of Stuart Powell (stuartp@ot.com.au).
3580 Sat May 13 20:44:06 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3582 * Generalized the SOCK_Connector::complete method so that it takes
3583 a Time_Value timeout. This allows applications to wait upto a
3584 certain limit before giving up on a non-blocking connection.
3585 Updated the tests in ./tests/IPC_SAP/SOCK_SAP/CPP-inclient.C to
3586 illustrate how this is used.
3588 * Added some interesting new tests to ./tests/Mem_Map that can be
3589 used to benchmark the performance of various strategies (e.g.,
3590 stdio, read/write, mmap, etc.) for copying files.
3592 Fri May 12 19:09:10 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3594 * libsrc/Mem_Map/Mem_Map: Added two new overloaded methods called
3595 sync() that are wrappers around the msync(3c) system call. Also
3596 added a new unmap() method that gives access to the full
3597 behavior of munmap(3c).
3599 Wed May 10 14:16:16 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3601 * Tried yet another fix for the gettimeofday() botch that both
3602 Centerline and SunOS 5.4 screw up in different ways. Thanks to
3603 Medhi Tabatabai (Mehdi.Tabatabai@ed.nce.sita.int) for the latest
3604 fixes.
3606 Tue May 9 19:05:58 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3608 * libsrc/Reactor/Handle_Set.C: Fixed some "off-by-one" errors that
3609 were occurring in the Handle_Set::sync() and set_max() methods.
3610 Thanks to Nigel Hooke (n.hooke@trl.oz.au) for finding and fixing
3611 these.
3613 Mon May 8 02:01:54 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3615 * libsrc/Reactor/Reactor.C: Moved the Null_Callback class from
3616 being nested within the Reactor class to outside the Reactor
3617 class. This fixes a problem with the SIG compiler.
3619 * Released version 3.2.2.
3621 * libsrc/Reactor/Signal.C: Made the Signal_Handler class
3622 thread-safe as well as signal-safe.
3624 * libsrc/Reactor/Reactor.C: Modified the new Reactor
3625 implementation slightly to handle signals correctly.
3627 * libsrc/Threads/Synch.h: Added a new conversion operator to
3628 Recursive_Lock that returns the underlying LOCK in case we need
3629 it for something (e.g., to initialize a Condition object).
3631 Sun May 7 04:17:46 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3633 * libsrc/Reactor/Reactor.C: Finished what I hope will be the last
3634 set of changes to make the Reactor robust in multi-threaded
3635 programs in situations where separate threads are used to
3636 register and remove handlers. The final solution is elegant in
3637 that it does not significantly penalize the performance of
3638 programs that do not use separate threads to register and remove
3639 handlers. Moreover, if you compile without the ACE_MT_SAFE flag
3640 the Reactor will not include additional state information
3641 related to the multi-threaded implementation.
3643 * libsrc/Threads/Synch.h: Subclassed the Guard class to make a new
3644 class Try_Guard that uses tryacquire() to obtain a LOCK.
3646 Fri May 5 18:43:50 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3648 * Released version 3.2.1
3650 Thu May 4 19:43:01 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3652 * libsrc/IPC_SAP/TLI_SAP/TLI_Connector.C (connect): Fixed a
3653 problem whereby a t_bind() was being done twice. Since I didn't
3654 write this code I don't know if my fix will work generically on
3655 all platforms supporting TLI. Please let me know if there are
3656 any problems.
3658 Tue May 2 17:21:53 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3660 * libsrc/Reactor: Added a new notification feature that allows
3661 multiple threads of control to enqueue Event_Handler * to the
3662 main Reactor thread via the pipe used to wake up the main
3663 Reactor thread. The Reactor will call the handle_exception()
3664 method on Event_Handler with a HANDLE == -1 to notify the
3665 handler. This feature is very useful if you need to have
3666 certain operations (such as handler termination) performed in
3667 the main thread.
3669 Tue Apr 25 00:34:18 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3671 * libsrc/Reactor: Fixed a stupid pair of typos in the Reactor that
3672 failed to cast arguments to the pipe to char *.
3674 * libsrc/Connection: Fixed a portability problem with the
3675 Connector: we can't use RW_Mutex for synchronization since not
3676 all platforms support threads. The quick fix is to use
3677 Null_Mutex, but a better long term approach is on the way!
3679 * Released version 3.2.
3681 * Added richer support for the TLI_Connector to make it work
3682 better with protocols other than TCP/IP. Thanks to Mats
3683 Sundvall (sundvall@perrier.embnet.se) for these enhancements.
3685 Mon Apr 24 02:26:04 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3687 * Changed the SOCK_Dgram_Brdcast class name to SOCK_Dgram_Bcast
3688 and the SOCK_Mcast to SOCK_Dgram_Mcast to be more consistent.
3690 * Massively improved the Reactor's support for multi-threading.
3691 There was actually a major problem in previous versions of ACE
3692 that used poll() as the underlying event demultiplexor. Poll()
3693 produced incorrect results when the Reactor::handle_events()
3694 method was run in one thread, but another thread came along and
3695 registered or removed an Event_Handler. The new version fixes
3696 this problem so that the Reactor will now work correctly in
3697 multi-threaded programs.
3699 One consequence of this change is that sysincludes.h and the
3700 SunOS 5.x configuration files have been changed so that poll()
3701 is no longer used as the default demultiplexor. The reason is
3702 that poll() doesn't work as efficiently when used in
3703 multi-threaded programs (ugh). See the Reactor code for
3704 examples of how this all works now. Basically, the trick is to
3705 use the select()-style implementation for most of the code, and
3706 transform to poll()-style implementation only when necessary.
3708 Sat Apr 22 03:35:51 1995 Douglas C. Schmidt (schmidt@lambada.cs.wustl.edu)
3710 * libsrc/ASX/Message_Queue: Added a new method to the
3711 Message_Queue class called unblock(). This method allows one
3712 thread to release all other threads that are waiting for
3713 messages to be enqueued or dequeued on a Message_Queue.
3714 Modified the return value of methods like enqueue_head() and
3715 dequeue_head() so that if a thread unblocks() then these methods
3716 return -1 with errno == ESHUTDOWN. Also changed the name of
3717 some of the internal methods so that they would be more
3718 consistent.
3720 * libsrc/Connection/Connector: Fixed a stupid bug caused by
3721 failing to have a destructor for Connector. Thus, there are
3722 cases where unconnected Svc_Handlers are left around in the
3723 handler_map_. This causes problems for dynamic linking in
3724 conjunction with the Reactor, so now the destructor iterates
3725 through all the unconnected Svc_Handlers and removes them from
3726 the Reactor.
3728 Fri Apr 21 15:28:46 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3730 * libsrc/Reactor/Signal: Changed Sig_Set::addset() and
3731 Sig_Set::delset() to the more appropriately named sig_add() and
3732 sig_del(). Also added '_' in the Sig_Set method names to be
3733 more consistent...
3735 * Removed the ./libsrc/IPC_SAP/SOCK_SAP/misc.[hC] file and merged
3736 the one function in that file (ace_bind_port) into
3737 ./libsrc/Misc/Misc.[hC].
3739 * Added a bunch of fixes from various people to make ACE compile
3740 better on SGI, SunOS 4, and HP/UX. Thanks for all the fixes!
3742 * libsrc/IPC_SAP/Addr: Changed sap_any to ace_sap_any to avoid
3743 problems with the namespace.
3745 * Added a new portability flag that only compiles the SOCK_Mcast.*
3746 stuff if the platform supports multicast!
3748 * ./libsrc/Shared_Malloc: Fixed some portability bugs with
3749 Memory_Pool.C and added new support for SunOS 4.x running SunC++
3750 4.x. Thanks to Steve Warwick (swarwick@arinc.com) for this
3751 strategic help.
3753 Wed Apr 19 09:05:43 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3755 * Fixed up the ./libsrc/CORBA_Handler so that it will correctly
3756 compile on platforms that don't have Orbix...
3758 * libsrc/Misc: added a new "Simple String" class called SString.
3759 This is not really meant to be used by end-user applications.
3760 It is used by certain portions of ACE that need to have
3761 operator== and operator!= defined on a string (e.g., the
3762 Token_Server and the Orbix Event_Comm components that use the
3763 Map_Manager).
3765 Tue Apr 18 00:09:31 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3767 * libsrc/ASX: Fixed a stupid bug in Task::activate() that didn't
3768 spawn a thread if the Thread_Manager was NULL...
3770 Sat Apr 15 19:39:30 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3772 * libsrc/Reactor: Added a new class to the Reactor called
3773 Event_Handler_T.[hi]. This class makes it easy to integrate
3774 existing classes into the Reactor framework without requiring
3775 them to inherit from Event_Handler directly. This technique is
3776 a superset of the DEF_TIE approach used by IONA in Orbix as one
3777 of the ways to combine an existing class (that doesn't know
3778 anything about CORBA) with a CORBA interface. Thanks to Greg
3779 Lavender
3780 (g.lavender@isode.com) for the suggestion.
3782 Fri Apr 14 14:41:49 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3784 * Run catman on the ./man directory so that it now has a windex
3785 file. This facilitates "man -k".
3787 Thu Apr 13 23:39:22 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3789 * libsrc/Reactor: Made the Reactor be a stand-alone component that
3790 does not require any other ACE libraries.
3792 * Added two new methods to the Reactor so that you can now
3793 register or remove a set of HANDLEs in a single call. Thanks to
3794 Bill Sears
3795 (wsears@world.std.com) for this suggestion. Basically, now you
3796 can register or deregister a set of n HANDLEs with one operation
3797 (rather than doing n register_handler () or n remove_handler
3798 ()).
3800 * include/sysincludes.h: Changed the name of the macro MT to
3801 ACE_MT and DB to ACE_DB to avoid polluting the namespace.
3803 Wed Apr 12 11:14:46 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3805 * libsrc/Reactor/Reactor.C (close): Added code to close down the
3806 pipe handles when the Reactor is shut down. Thanks to Mark
3807 Patton (mark_patton@tx72.mot.com) for this info.
3809 Fri Apr 7 18:33:30 1995 Douglas C. Schmidt (schmidt@siesta.cs.wustl.edu)
3811 * Added a new method to CORBA_Handler so that handlers don't need
3812 to subclass from CORBA_Handler, but rather can call the
3813 CORBA_Handler::register_service() method (which is static).
3814 Thanks to Chris Tarr (ctarr@objectspace.co) for this suggestion.
3816 Mon Apr 3 13:09:45 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3818 * Added a new static method to Service_Config.[hC] called
3819 end_event_loop(). This can be used to stop the run_event_loop()
3820 method of the Service_Config. In addition, all the methods in
3821 Service_Config were changed to be static to emphasize the fact
3822 that the Service Configurator is a singleton...
3824 * include/Trace.h (ACE_TRACE): Changed the macro T to ACE_TRACE.
3825 This will prevent namespace pollution.
3827 * Added support for the Orbix CORBA implementation. If you don't
3828 have Orbix, you'll need to remove this flag from the
3829 config-sunos5-sunc++.4.x.h file if you are compiling on SunOS
3830 5.x.
3832 Sun Apr 2 01:12:19 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3834 * Added new support for C++ wrappers around the IP multicasting.
3835 There are three new files in ./libsrc/IPC_SAP/SOCK_SAP called
3836 SOCK_Mcast.[hiC] and a new test directory in
3837 ./tests/Reactor/multicast and ./tests/Reactor/ntalker to
3838 illustrate how to use this stuff. Thanks to Tim Harrison
3839 (harrison@cs.wustl.edu) for this code.
3841 Sat Apr 1 18:48:40 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3843 * Added support to the libsrc/Shared_Malloc/Memory_Pool.[hC] class
3844 for System V shared memory, as well as for mmap(2) and sbrk(2)
3845 memory. Please note that this has not been extensively tested
3846 yet, so use with caution...
3848 Thu Mar 30 21:50:00 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3850 * libsrc/Misc/Atomic_Op: Added the postfix versions of the ++ and
3851 -- operators. This stops the Sun CC compiler from griping about
3852 the "anachronistic" usage when using a++ rather than ++a.
3853 Thanks to Bruce Worden <bruce@betsy.gps.caltech.edu> for the
3854 suggestion.
3856 Wed Mar 29 22:26:37 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3858 * libsrc/Reactor/Handle_Set: Fixed a weird problem where I was
3859 using NOFILE as the max number of descriptors rather than
3860 FD_SETSIZE. I don't know why I was doing this.
3862 * Also fixed up a problem with Handle_Set that was causing the
3863 iterator to go nuts when it got a strange initial value of
3864 fd_set in Handle_Set. Orbix triggers this kind of nonsense in
3865 some cases...
3867 Tue Mar 28 21:01:36 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3869 * libsrc/Reactor/Reactor.C: swapped the order of dispatching so
3870 that handle_exception methods are dispatched before handle_input
3871 methods to improve response time for exceptions (e.g., OOB
3872 data). Thanks to Detlef Becker
3873 (beckerd@erlh.siemens.de) for suggesting this.
3875 * Fixed a bug in ./ASX/Stream.i that caused problems when a Module
3876 was popped off a Stream. Thanks to Paul Stephenson for noticing
3877 this and sending me the fix.
3879 * Changed the Reactor::dispatch methods so that they dispatch the
3880 timers *before* dispatching the I/O-based event handlers. This
3881 is helpful for systems that are time-delay sensitive.
3883 * libsrc/Reactor/Time_Value: Added new += and -= operators to
3884 Time_Value. Thanks to Alex V. Maclinovsky
3885 (garyh@teleng1.tait.co.nz) for this suggestion.
3887 Thu Mar 23 15:38:23 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3889 * libsrc/IPC_SAP/Addr/SPIPE_Addr.i (set): Fixed a problem that the
3890 HP/UX compiler had with complex arithmetic expressions.
3892 * libsrc/Service_Configurator: fixed up the Service_Configurator
3893 source code so that it will compile correctly on HP/UX
3894 platforms.
3896 Tue Mar 21 00:28:25 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3898 * libsrc/IPC_SAP/SOCK_SAP/LSOCK_Connector.h: Fixed an amazingly
3899 stupid bug with LSOCK_Connector, where I was passing in a
3900 default argument of PF_INET instead of PF_UNIX... Arrgh!
3902 Mon Mar 20 20:24:29 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3904 * Fixed up the $WRAPPER_ROOT/include/makeinclude/ directory to
3905 define a new set of platform-specific configuration files. This
3906 should greatly improve the portability of ACE to different
3907 OS/compiler platforms.
3909 * Added a bunch of changes to allow ACE to build with Lucid C++ on
3910 SunOs 4.x. This should also fix some other problems we've been
3911 having with SunOS 4.x. Thanks to Lee Baker (baker@ctis.af.mil)
3912 for these changes.
3914 Sun Mar 19 00:34:30 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3916 * Added manual pages for all of the ./libsrc libraries. These
3917 manual pages are all generated automagically from the libsrc
3918 header files.
3920 Sat Mar 18 10:48:46 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3922 * Finally broke down and fixed the Makefile scheme so that there
3923 is just one shared library (libACE.so) and one archive library
3924 (libACE.a). This should massively simplify the application
3925 development process, though it may require a bit of fixing of
3926 Makefiles to remove all the -lReactor -lIPC_SAP stuff that was
3927 in there before.
3929 * libsrc/IPC_SAP/TLI_SAP/: Fixed a stupid bug that was causing
3930 core dumps since the TLI option pointers weren't initialized to
3931 0 in the TLI::TLI constructor. Thanks to Ed Brown for noticing
3932 this (eebrown@netcom.com).
3934 Wed Mar 15 00:08:19 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3936 * Added some new fixes to get ACE to run on OSF1 and Linux.
3938 Tue Mar 14 13:36:31 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3940 * Added several changes to the source code and config-hpux.h
3941 configuration file based on comments from Alex Ranous
3942 (ranous@nsa.hp.com). In particular, changed ACE_HAS_NO_FDSET to
3943 ACE_SELECT_USES_LONG.
3945 Mon Mar 13 09:23:58 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3947 * Improved the Makefile scheme so that the use of PTDIRS is no
3948 longer required. The trick was to link the appropriate template
3949 *.C files into the WRAPPER_ROOT/include directory. This enables
3950 the compiler to find them in one single place.
3952 Sun Mar 12 22:35:50 1995 Douglas C. Schmidt (schmidt@tango.cs.wustl.edu)
3954 * Added a new mechanism to clean up the insane number of -D's in
3955 the Makefile. This new mechanism requires a bit more work at
3956 installation time, but should greatly reduce the effort during
3957 normal software development
3958 (once ACE is installed...). See the INSTALL file
3959 for more details.
3961 Fri Mar 10 17:29:02 1995 Douglas Schmidt <douglas@gtcterminator>
3963 * Fixed a stupid bug in Map_Manager.i which wasn't correctly
3964 setting default values for fields in one of the Map_Manager
3965 constructors.
3967 Fri Mar 10 08:54:42 1995 Doug Schmidt <douglas@gtccipher>
3969 * /libsrc/Connection/Acceptor.i: Fixed a potential memory leak in
3970 handle_input().
3972 Thu Mar 9 11:59:54 1995 Doug Schmidt <douglas@gtccipher>
3974 * Changed the Map_Manager::find() method to use class Read_Guard
3975 so that multiple finds may proceed in parallel on a multiple
3976 threaded application. Likewise, changed the implementation of
3977 bind() and unbind() to use class Write_Guard so that destructive
3978 operations will be serialized correctly. One consequence of
3979 this is that RW_Mutex must be used instead of Mutex...
3981 * Added a new overloaded Map_Manager::unbind() method with an
3982 INT_ID parameter. This parameter that passes back a reference
3983 to the internal id being unbound. This allows the caller to
3984 clean up any dynamic memory associated with the INT_ID.
3986 Tue Mar 7 13:32:17 1995 Doug Schmidt <douglas@gtccipher>
3988 * Changed a bunch more things to get ACE to compile with
3989 Centerline C++. This should also make ACE more portable to
3990 other cfront-based compilers.
3992 * Had to change the ASX source in ACE to work around a stupid
3993 problem with templates in cfront-based C++ compilers.
3994 Fortunately, it was possible to mask most of the problems using
3995 the C++ preprocessor.
3997 * There seem to be some problems with cfront-based compilers (such
3998 as centerline). They don't like the new SYNCH::MUTEX usage in
3999 the Message_Queue.
4001 * Removed several unused local variables in the INET_Addr::set()
4002 method.
4004 Wed Mar 1 00:35:11 1995 Douglas C. Schmidt (schmidt@tango)
4006 * Added a major performance boost on high-speed networks. It
4007 turns out that send(3) and revc(3) are slower than write(2) and
4008 read(2) (since the latter are system calls, and the former are
4009 library calls. Therefore, I added new overloaded methods to
4010 SOCK_IO so that write(2) and read(2) are available to the user!
4012 Tue Feb 28 10:13:09 1995 Douglas C. Schmidt (schmidt@tango)
4014 * libsrc/Reactor/Reactor.C: Fixed the implementation of the
4015 Reactor::open and Reactor::Reactor constructor so that it is now
4016 possible to resize the size of the Reactor *after* the
4017 constructor has run (by calling open()) *without* having to shut
4018 down the Reactor first.
4020 * Modified the Reactor::attach() method so that it will ADD the
4021 new bits to the bitmask rather than SET them. Thanks to Mark
4022 Patton for recommending this
4023 (mark_patton@tx72.mot.com).
4025 Sat Feb 25 15:08:04 1995 Douglas C. Schmidt (schmidt@tango)
4027 * Released version 3.0.5.
4029 Fri Feb 24 17:57:01 1995 Douglas C. Schmidt (schmidt@tango)
4031 * Used the OSE class2man scripts to generate UNIX manual pages for
4032 IPC_SAP, Thread, Connection, and Reactor. Other manual pages
4033 will be forthcoming...
4035 * libsrc/IPC_SAP/Addr/INET_Addr.i: Fixed a bug in one of the
4036 INET_Addr::set() methods that caused a segmentation fault if
4037 host_name was NULL. In addition, greatly cleaned up the code so
4038 that all the "set" methods share a common basis of code.
4040 Tue Feb 21 19:32:28 1995 Douglas C. Schmidt (schmidt@tango)
4042 * Updated the entire release to use the new SYNCH template
4043 interface for Task, Module, Stream, and Message_Queue. This is
4044 *much* cleaner, though it reveals some bugs with the Sun C++
4045 templates facility.
4047 Mon Feb 20 22:46:14 1995 Douglas C. Schmidt (schmidt@tango)
4049 * Changed the SPIPE_SAP, SOCK_SAP, and TLI_SAP *Connector class
4050 methods from open() to connect(). This is a more accurate name
4051 and it also fits in nicely with the Acceptor::accept() method.
4053 * Started integrating new versions of various header files that
4054 include hooks to automatically generate documentation.
4056 Tue Feb 14 20:52:13 1995 Douglas C. Schmidt (schmidt@tango)
4058 * Changed the order in which the Reactor dispatches descriptors.
4059 Originally, it dispatched the "read" descriptors *before* the
4060 "write" descriptors. Now, it dispatches the "write" descriptors
4061 first. This was necessary to handle weird behavior of sockets
4062 over TCP/IP when data is piggy-backed with the final ACK on a
4063 non-blocking connection.
4065 Mon Feb 13 15:49:21 1995 Douglas C. Schmidt (schmidt@tango)
4067 * Changed the name of Event_Handler::get_fd() to
4068 Event_Handler::get_handle() to be more consistent with other
4069 usage in ACE.
4071 Sat Feb 4 22:47:34 1995 Douglas C. Schmidt (schmidt@tango)
4073 * libsrc/Service_Configurator/Service_Object: changed the
4074 destructor to be virtual (thanks to Steffen Winther Sorensen
4075 <sts@dad.stibo.dk> for noticing this).
4077 Sat Jan 28 16:29:49 1995 Douglas C. Schmidt (schmidt@tango)
4079 * Changed the name of class Map_Manager's Search_Structure struct
4080 to Map_Entry. This is more specific to what that data structure
4081 really does.
4083 * Began adding support for Linux, courtesy of sts@dad.stibo.dk.
4085 Sun Jan 22 23:15:38 1995 Douglas C. Schmidt (schmidt@tango)
4087 * Changed the name of the ./libsrc/Connector-Acceptor directory to
4088 ./libsrc/Connection. Also changed the name of libConn_Acc.so to
4089 libConnection.so to be more consistent.
4091 Sat Jan 21 13:59:18 1995 Douglas C. Schmidt (schmidt@tango)
4093 * libsrc/IPC_SAP: Fixed the *Acceptor classes for SOCK_SAP and
4094 SPIPE_SAP so they behave just like FIFO_SAP and contain a method
4095 called remove(). This method closes the underlying descriptor
4096 and also unlinks the local address from the file system.
4098 * libsrc/ASX/Message_Queue.i (copy): Fixed this code so that it
4099 correctly stores starting at the wr_ptr rather than the rd_ptr.
4100 Thanks to Chris Cleeland
4101 (chris@milo.st-louis.mo.us) for pointing this out
4102 to me.
4104 Wed Jan 11 13:07:19 1995 Douglas C. Schmidt (schmidt@tango)
4106 * Changed all uses of class Thr_Manager to class Thread_Manager
4107 and class Thr_Cntl to Thread_Control. This is more readable and
4108 is now consistent with the documentation...
4110 Tue Jan 10 13:49:31 1995 Douglas C. Schmidt (schmidt@tango)
4112 * Modified the makefile scheme so that only the *.so files are
4113 built by default. If you want to build both *.a and *.so files,
4114 uncomment out the lines described in
4115 $WRAPPER_ROOT/include/makeinclude/rules.lib.GNU. Note that this
4116 will require you to run make on the ./libsrc directories twice
4117 in order to properly build and install both the *.a and *.so
4118 libraries.
4120 Mon Jan 9 22:57:29 1995 Douglas C. Schmidt (schmidt@tango)
4122 * libsrc/Reactor/Signal.h: Changed the name of the Signal_Block
4123 class to the Signal_Guard class to be more consistent with the
4124 Guard class in Synch.h.
4126 Sat Jan 7 19:49:46 1995 Douglas C. Schmidt (schmidt@tango)
4128 * tests/ASX/Event_Server/Event_Server: Revised the Event Server
4129 test example to use the Acceptor pattern components.
4131 Fri Jan 6 23:38:21 1995 Douglas C. Schmidt (schmidt@tango)
4133 * Added a new method to class Task that turns a task into an
4134 active object (i.e., associates a thread of control with the
4135 task). This is useful since it replaces all the places in
4136 application code that original said "this->thr_mgr_.spawn
4137 (THR_FUNC
4138 (&this->svc_run))", etc...
4140 Thu Jan 5 21:05:15 1995 Douglas C. Schmidt (schmidt@tango)
4142 * libsrc/ASX: Changed the capitalization of class STREAM to class
4143 Stream. This seems more reasonable since I don't see any
4144 particular reason to shout about Streams!
4146 * libsrc/ASX/Task: Changed name of method qreply() to reply() to
4147 reflect the fact that the name of the class is no longer Queue!
4149 * libsrc/ASX: Made both Task and Message_Queue into parameterized
4150 types. This greatly improves the ability to parameterized
4151 synchronization into an application.
4153 * Changed all occurrences of timestruc_t to use Time_Value
4154 instead. This helps to improve portability and reduce the
4155 "impedence mismatch" caused by mixing both C and C++ types in
4156 the ACE interfaces. Note that one drawback of this is that we
4157 lose nano-second timing accuracy. However, I don't know of any
4158 real OS platforms that support that degree of precision anyway!
4160 * libsrc/Reactor: Moved the static "zero" data member from the
4161 Timer_Queue class to the Time_Value class. Also added a new
4162 static data member called "zerop," which is a pointer to "zero".
4164 * libsrc/Threads/Synch: Changed the interface of class Condition
4165 so that it no longer has both wait() and a timedwait() methods.
4166 Since C++ has default values, these two methods were redundant.
4167 Now, there is only a single method called wait(). By default,
4168 it's argument is 0, which defaults to the original wait()
4169 semantics. If the argument is non-zero then the timewait
4170 semantics apply.
4172 * libsrc/Threads/Synch: Added a new class called Null_Condition.
4173 This is similar to the Null_Mutex class in the sense that it has
4174 the same interface as class Condition, but it's methods are all
4175 no-ops
4176 (however, wait() and signal() both set errno = ETIME
4177 before returning...). This class is useful for
4178 parameterizing synchronization into an application.