ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/Coro/Changes
(Generate patch)

Comparing Coro/Changes (file contents):
Revision 1.349 by root, Sat May 24 18:01:50 2008 UTC vs.
Revision 1.610 by root, Wed Dec 19 23:36:16 2012 UTC

1Revision history for Perl extension Coro. 1Revision history for Perl extension Coro.
2 2
3TODO: should explore PerlIO::via::CoroCede and PerlIO::via::CoroAIO. 3TODO: should explore PerlIO::coroaio (perl leaks like hell).
4TODO: maybe implement a default message channel, very much like Erlang's 4TODO: channel->maxsize(newsize)?
5 actor model (which is cool in a lot of important aspects (failures!), 5TODO: __GCC_HAVE_DWARF2_CFI_ASM
6 but very lacking in others (higher level ipc)). 6
7 - improved Coro::State documentation a bit.
8 - Coro::Debug::command now flushes the output.
9 - add hack detection code for x32 abi, because the braindead slugs
10 who designed that made it look exactly like x86_64 without
11 providing proper compile time symbols to test for it. as a result,
12 this detection cannot work reliably.
13
146.23 Fri Dec 7 23:36:37 CET 2012
15 - use experimental fiber implementation on native windows
16 perls.
17 - use sizeof (void *) as multiplication factor for stack sizes,
18 to accomodate the totally braindamaged microsoft 64 bit "os".
19 - changed verifier host from win2k-ap510-32 to win7-sp516-32/64.
20 activeperl 5.16 crashes when PerlIO_define_layer is called due
21 to some bug in the perl dll, strawberry perl at least passes
22 the testsuite.
23 - implement Coro::Handle->peeraddr/host/port, for slightly
24 improved compatibility with LWP.
25 - implement 5.17 compatibility by almost blindly applying a
26 good-looking patch by Father Chrysostomos.
27 - move stack management functions into libcoro 3.
28 - libcoro version 3 "released".
29 - support magic values as timed_io_once args.
30 - recommend AnyEvent 7+ or EV 4+, also require EV
31 version 4 or newer for Coro::EV.
32
336.10 Tue Oct 9 01:14:27 CEST 2012
34 - updated ecb.h, it had a typo that caused it to not compile on many
35 big endian systems (reported by many people).
36 - disable memory fences in ecb.h to improve portability.
37
386.09 Sat Oct 6 23:25:02 CEST 2012
39 - Coro::EV I/O watchers were not interruptible by exceptions
40 (Coro::State::throw) (testcase by sten).
41 - ->throw now puts threads into the ready queue, as this seems to
42 be expected by existing code, and code that doesn't cope with spurious
43 wakeups needs fixing anyway.
44 - use fd -1 in mmap.
45 - cast I32 to int in error message printf.
46 - warn about broken so-called "hardened" kernels.
47
486.08 Fri Apr 13 12:05:47 CEST 2012
49 - be more aggressive about exiting like perl does - formerly,
50 exiting from the non-main thread would not execute END blocks.
51
526.07 Fri Nov 11 21:21:48 CET 2011
53 - work around a bug in PerlIO (setting $SIG{__WARN__} to a PVCV).
54 - update ecb.h.
55
566.06 Mon Aug 8 23:59:48 CEST 2011
57 - cygwin unfortunately patches the stack at runtime, so we use the pthreads
58 backend, which is an order of magnitude slower. unfortunately, cygwins
59 pthread implementation isn't very complete either, so allocate the stack
60 twice just to be sure.
61 (note: cygwin also enables mymalloc, which is NOT THREADSAFE ON WINDOWS,
62 in its ithreaded perl - best recompile cygwin and use the 'w'indows
63 backend for much better performance. also disable ithreads for
64 even better performance...).
65
666.05 Thu Aug 4 21:36:36 CEST 2011
67 - blush, condvar values would not be propagated from send to recv anymore
68 (reported by Chip Salzenberg).
69 - use exponential increase for the readline buffer length in
70 Coro::Handle. also reduce initial allocation to 1020 from 4096 bytes.
71
726.04 Wed Aug 3 17:15:45 CEST 2011
73 - use even more efficient and more compatible condvars for
74 compatibility to AnyEvent 6.x :)
75 - more inconsequential ecb.h updates.
76
776.03 Wed Aug 3 11:41:30 CEST 2011
78 - change how Coro patches AnyEvent condvars for compatibility to
79 AnyEvent 6.x.
80 - update ecb.h, to no longer include <pthread.h> in case WinNT.h
81 hasn't been included.
82
836.02 Wed Jul 13 04:35:19 CEST 2011
84 - "improve portability to Gentoo" - gentoo manages to put perl variables
85 in memory areas that are farther than 2gb apart, which the jit couldn't
86 handle and barfed. now it's just a bit slower on gentoo and similar
87 systems.
88
896.01 Sun Jul 3 12:31:14 CEST 2011
90 - workarounds are good, but the test for whether pthreads are used
91 was not good. thisone should be better.
92 - check differently whether gcc generates cfi instructions itself.
93
946.0 Wed Jun 29 19:43:35 CEST 2011
95 - INCOMPATIBLE CHANGE: unreferenced coro objects will now be
96 destroyed and cleaned up automatically (e.g. async { schedule }).
97 - implement a JIT compiler for part of the thread switch code,
98 which gives a 50% speed improvement on threaded perls, and
99 about 4% on non-thraeded perls (so threaded perls now finally
100 reach about half the speed of non-threaded perls).
101 - slightly modernise Coro::Intro, add section about rouse functions.
102 - avoid DEFSV and ERRSV, giving another 10% improvement
103 in thread switching.
104 - Coro::State->is_destroyed is now called is_zombie.
105 - implement a Coro->safe_cancel method that might fail, but
106 cancels in a "safer" way if it succeeds.
107 - add preliminary support for DEBUGGING perls.
108 - get rid of two hash-accesses when initialising a new Coro - this
109 speeds up coro creation by almost a factor of two.
110 - croak when a coro that is being cancelled tries to block
111 (e.g. while executing a guard block), instead of crashing or
112 deadlocking.
113 - use a more robust and also faster method to identify Coro::State
114 objects - speeds up everything a bit.
115 - implement Coro->cancel in XS for a 20% speed improvement, and to
116 be able to implement mutual cancellation.
117 - speed up context switches by a percent or two by more efficiently
118 allocating context stack entries.
119 - implement Coro->join and Coro->on_destroy in XS for a speedup and
120 a reduction in memory use.
121 - cancelling a coro while it itself is cancelling another coro is
122 now supported and working, instead of triggering an assertion.
123 - be a bit more crash-resistant when calling (buggy) on_destroy
124 callbacks (best effort).
125 - move on_destroy into the slf_frame, to allow extension slf
126 functions to have destructors.
127 - get rid if coro refcounting - simply crash in other interpreter
128 threads by nulling the pointers on clone.
129 - simplify warn/die hook handling when loading Coro - the convoluted
130 logic seems to be no longer neccessary.
131 - use libecb instead of our own home-grown gcc hacks.
132 - document alternatives to Coro::LWP. Please use them :)
133 - work around another mindless idiotic NEEDLESS bug in openbsd/mirbsds
134 sigaltstack. Really. wine suffers from it, erlang suffers from it,
135 and it's known since at least 2006.
136
1375.372 Wed Feb 23 06:14:30 CET 2011
138 - apparently mingw doesn't provide a working gettimeofday, try to
139 work around that by relying on Time::HiRes (indirectly brought to
140 my attention by Max Maischein).
141 - fix some portability issues when Time::HiRes was used.
142
1435.371 Mon Feb 21 14:36:08 CET 2011
144 - backport to windows process emulation code again.
145
1465.37 Sat Feb 19 07:49:44 CET 2011
147 - add a big "Coro thread life cycle" section to "man Coro".
148 - try a tentative workaround against the breakage that 5.13 has
149 introduced without depreciation period. sigh.
150 - no longer use Time::HiRes if gettimeofday is available, which
151 saves quite a lot of memory.
152
1535.36 Sun Feb 13 05:33:41 CET 2011
154 - automatically load Coro::Channel, Coro::RWLock, Coro::Semaphore,
155 Coro::SemaphoreSet, Coro::Signal and Coro::Specific on first "new"
156 method call.
157 - undocument Coro::Timer::sleep and obsolete whole module.
158 - optimise Coro::Timer::timeout memory and cpu usage.
159 - slightly updated Coro::Intro for recent changes.
160 - do not initialise PL_dirty anymore.
161
1625.25 Thu Nov 11 01:08:39 CET 2010
163 - try a different approach on netbsd - netbsd 5 finally has marginally
164 working pthreads, but still broken ucontext/sigaltstack.
165 - openbsd 4.8 finally got their act together, Coro works out of the box
166 with asm, setjmp and pthreads (no change, just informational).
167
1685.24 Sat Oct 23 11:27:12 CEST 2010
169 - port to the EV 4.0 API.
170 - work around bugs in mingw32, making strawberry perl work
171 out of the box.
172 - correctly modify Coro::AIO function prototypes
173 so that they reflect the "no optional parameters" rule.
174 - "ported" libcoro to C++.
175
1765.23 Mon May 17 18:50:42 CEST 2010
177 - be more resistant to ordering changes when initialising
178 Coro::AnyEvent, Coro::EV and Coro::Event (reported by Matthias
179 Waldorf).
180 - document that perl 5.12 deliberately removed support for cloning.
181
1825.22 Wed Apr 14 03:55:35 CEST 2010
183 - correctly return udnef on errors in Coro::Handle::read/write
184 (testcase by Marc Mims).
185 - convert Coro::Util into a "perl compatibility wrapper" - the functions
186 are less useful now, but are drop-in replacements for existing
187 functions, listing better alternatives in the documentation. This also
188 fixes a bug in Coro::LWP which naively substituted Socket::inet_aton
189 with Coro::Util::inet_aton.
190 - do not override $Coro::idle unconditionally in Coro.pm, as other
191 modules could have provided their own idle coro already
192 (for exmaple, Coro::AnyEvent).
193 - fix Coro::Util::gethost* functions.
194 - Coro::Timer corretcly exports it's symbols (reported by Hideki Yamamura).
195
1965.21 Wed Dec 16 07:19:51 CET 2009
197 - automatically load Coro::AnyEvent when AnyEvent and Coro are used
198 together.
199 - add some examples on how to combine other event loops with Coro in
200 Coro::AnyEvent, and how to run it (and not to block). Seems to be
201 the most common source of confusion.
202 - try to catch people naively blocking in an event callback.
203 - work around the perl filehandle bug issue in conjunction with
204 older common::sense (as indirectly pointed out by ZSystem).
205 - clarify the "not from signal handlers" section.
206
2075.2 Sun Oct 4 14:54:24 CEST 2009
208 - Coro::Storable destroyed the prototypes of the functions it wrapped.
209 - export rouse_cb and rouse_wait by default now.
210 - fix various prototype mismatches in Coro::AnyEvent and Coro::Handle.
211 - new method $state->swap_sv.
212 - added section on "windows process emulation" to the manpage,
213 after a not-so-fruitful (nor-friendly) "discussion" with chip
214 salzenberg (discussion implies arguments, but his only arguments
215 were ad-hominems, one wonders why he started it in the first
216 place). I hope this explains it well enough for him to understand,
217 and maybe well enough for others to understand.
218 - use common::sense everywhere now.
219 - idle callbacks are no longer supported, use idle coros instead.
220 - print a thread listing when a deadlock is detected.
221
2225.17 Sat Aug 22 23:09:31 CEST 2009
223 - work around a bug in the perl debugger causing crashes
224 when running under the debugger by marking _pool_handler as nodebug.
225 - speed up Coro::async considerably.
226 - try some hacks to get netbsd to work "more often" - their broken
227 setjmp/longjmp, ucontext *and* phtreads are really hard on Coro.
228 - convert Coro to AE 5.0 API.
229
2305.162 Tue Jul 28 04:04:03 CEST 2009
231 - perl 5.8.2 is now minimum requirement.
232 - skip t/19_handle.t on broken windows perls.
233
2345.161 Wed Jul 22 04:47:38 CEST 2009
235 - Coro::AnyEvent::poll could have a different prototype when EV was
236 used as backend (analyzed by Tatsuhiko Miyagawa).
237 - Coro::AnyEvent errornously initialised the event loop when loaded,
238 not on demand.
239 - try to workaround rare */t/01_unblock.t failures.
240
2415.16 Tue Jul 21 01:44:37 CEST 2009
242 - Coro::AnyEvent failed to hook into the event loop
243 when no threads had been readied between detecting
244 the event loop and actually running it.
245 - considerably speed up Coro::Select by taking avdantage
246 of AnyEvent > 4.8 and some other optimisations.
247 - implement paragraph readline mode in Coro::Handle
248 (based on patches by Zsbán Ambrus).
249 - replace WSAEINPROGRESS by WSAEWOULDBLOCK (reported
250 and analyzed by Yasuhiro MATSUMOTO).
251 - clarified libcoro license and copyright.
252 - someone stole my EXTRA_META!!!
253 - implement Coro::Select::patch_pp_sselect and it's brother,
254 for hardcode select overriding.
255
2565.151 Mon Jul 6 05:41:57 CEST 2009
257 - backport to windows process emulation code again (patch by
258 Yasuhiro MATSUMOTO).
259 - slightly update Coro::MakeMaker.
260
2615.15 Tue Jun 30 10:28:06 CEST 2009
262 - deprecate Coro::Socket, document how to get ipv6 support via
263 AnyEvent::Socket instead.
264 - implement signal->wait ($cb) interface, similar to semaphores.
265 - work around SvOK not supporting getmagic, so we have to getmagic
266 to test for undef :/ (reported by Matthias Waldorf).
267 - load Coro::AnyEvent in all modules using AnyEvent.
268 - work around perl corrupting our internal data structures,
269 reported by Tokuhiro Matsuno.
270 - enable per-coroutine real and cpu time gathering
271 (Coro::State::enable_times).
272
2735.14 Wed Jun 24 01:37:48 CEST 2009
274 - provide explicit functions to "cede" to the event loop in Coro::AnyEvent,
275 as this seems to have been a difficult concept (poll, sleep, idle,
276 idle_upto).
277 - add Coro::AnyEvent::readable/writable functions.
278 - clarify Coro::EV/Event/AnyEvent manpages.
279 - free per-thread global scalars in the thread calling ->cancel, to
280 avoid crashes when $_, $@ etc., are magical but some of those
281 had already been freed.
282 - "unexperimentalise" the callback interface for Coro::Semaphore.
283 - speed up ready queue management/context switching by using a linked
284 list instead of an array (~5-10%).
285 - implement "watch" command in Coro::Debug shells.
286 - for fun, implement time-slicing as an example in the manpage.
287 - if AnyEvent detects EV or Event, but we don't have Coro::EV or
288 Coro::Event, use the normal AnyEvent handling instead of dieing
289 (the same is true for Coro::Handle).
290 - properly document Coro::EV::timed_io_once.
291 - avoid unneccessary ->cancel calls in Coro::Handle.
292 - maybe make it work on mingw32 with win32 backend
293 (based on patch by Yasuhiro Matsumoto).
294
2955.132 Fri May 29 09:00:39 CEST 2009
296 - do not keep a reference to the argument itself in
297 Coro::Semaphore::guard, as it could change later.
298 - support SO_RCVBUF/SO_SNDBUF nonstandard Coro::Socket options,
299 should support a prepare callback.
300
3015.131 Mon Mar 16 23:20:37 CET 2009
302 - implement and document Coro->suspend, Coro->resume.
303 - fix Coro::Select implementation to not (often) close
304 the passed file descriptors (testcase provided by pippijn).
305
3065.13 Mon Dec 15 21:51:42 CET 2008
307 - EXPERIMENTAL: implement dynamic winds (on_enter/on_leave).
308 - don't set diehook to C<undef> but instead to NULL, to avoid
309 spurious warnings.
310 - fix a lot of bugs in Coro::SemaphoreSet.
311 - Coro::SemaphoreSet will less often create a semaphore needlessly.
312 - add Coro::SemaphoreSet::count and wait methods.
313 - take advantage of the new Guard module.
314 - deprecate Coro::guard.
315 - try to fix the dreaded 01_unblock tests once more. I hate it when
316 testsuites need more fixing than the code they are supposed to test.
317 - croak in more cases when a required callback isn't resolvable.
318 - fix some minor issues in Coro::State->call/eval.
319 - use current coroutine context instead of a temporary one
320 when temporarily switching to another coroutine.
321 - do not call C-level on_destroy handlers during global destruction,
322 to avoid needless segfaults.
323
3245.12 Sun Dec 7 13:30:38 CET 2008
325 - add default config for MirOS, which seems to be a bug-to-bug
326 compatible fork of openbsd ("world domination by releasing
327 openbsd cvs before the openbsd folks do it" or so :).
328 - free_padlist did destroy the names pad, not good, but didn't
329 seem to bother perl - this could fix issues such as eval ""
330 inside a function called from multiple coroutines.
331 - use a different method to detect destruction time.
332 - be more careful when freeing padlists just before global
333 destruction.
334 - fixed and expanded the call/cc example.
335 - renamed _terminate to _coro_run.
336 - new method Coro::Channel->shutdown.
337 - try pthreads on openbsd <4.4 (broken sigaltstack, will
338 pthreads fare better?).
339 - be less picky about destroying "a" running coroutine.
340
3415.11 Tue Nov 25 21:49:05 CET 2008
342 - DEBUGGING in 5.10.0 is a mess: it constantly flags perfectly
343 working code with failed assertions, introducing more bugs than
344 it fixes, requiring elaborate workarounds :(
345
3465.1 Mon Nov 24 08:54:59 CET 2008
347 - wrote a small introductory tutorial - Coro::Intro.
348 - convert Coro::Timer, Coro::Select and Coro::Util to rouse API.
349 - Coro::Select did errornously dup the file descriptors
350 and didn't work with all AnyEvent backends.
351 - Coro::Select wasn't imported correctly form Coro::LWP, causing blocking
352 LWP data transfers.
353 - disassociate c contexts from coro objects - this is agruably more
354 correct, but mostly allows sharing of cctxs between coro and state
355 objects, for added memory savings and speed increases.
356 - bumped $Coro::POOL_RSS up to 32kb by default.
357 - no longer set the optype to OP_CUSTOM, as B::* understandably
358 doesn't like this very much (and we *are* a type of entersub).
359 - implement state cloning, just to prove that call/cc can be done.
360 - automatically load Coro::AnyEvent in Coro::Handle.
361 - wrap ->cancel calls in eval inside Coro::Handle as EV watchers
362 do not have this method (and don't need it either).
363 - speed up generic anyevent methods in Coro::Handle by using rouse
364 callbacks.
365 - allow coroutines in $Coro::IDLE, speeding up Coro::AnyEvent and
366 others. It also makes the debugger happier, as you can trace
367 through the idle threads now.
368 - add comppad_name* and hints ($^H, %^H) to per-thread variables.
369 - eg/event was pretty broken.
370 - better 5.8.6 compatibility.
371
3725.0 Thu Nov 20 10:35:05 CET 2008
373 - NEW ARCHITECTURE: use the latest 4.x version if you experience
374 stability issues.
375 - bump API version to 7 - all dependents must be recompiled.
376 - removed timed_* functions - they were not being used anyways
377 and should be replaced by a more generic mechanism -
378 and were annoying to support anyways :)
379 - removed SemaphoreSet's waiter method - use sem method instead.
380 - Coro::Semaphore->adjust didn't correctly wake up enough waiters.
381 - async_pool did free a scalar value twice
382 ("Attempt to unreference...").
383 - fix a longstanding bug where calling terminate on a coro that
384 was waiting for a semaphore that was just becoming available
385 would cause a deadlock (semaphore would get into a state where
386 it was available but waiters were still blocked).
387 - calling throw on a coroutine that is waiting for a semaphore will
388 no longer make it acquire the semaphore (and thus leak a count).
389 - perl's process emulation is now not even theoretically supported
390 anymore.
391 - new functions Coro::rouse_cb and Coro::rouse_wait for easier
392 conversion of callback-style to blocking-style.
393 - new methods $coro->schedule_to and ->cede_to, to specifically
394 schedule or cede to a specific coroutine.
395 - new function Coro::Semaphore::wait.
396 - use named constants in Coro::Channel (Richard Hundt).
397 - directly patch the entersub opcode calling SLF functions (cede,
398 transfer and so on). this does speed up context switching, but
399 more importanly, it frees us from the hardcoded behaviour of
400 entersub, so we might actually be able to return something from
401 those functions and atcually create new ones.
402 - take advantage of __builtin_frame_address on gcc.
403 - expose THX in coroapi (not sure whether this was a wise decision,
404 as "threaded" perls are running at half speed anyways).
405 - implement execute_slf (schedule-like-function) interface that makes
406 it possible to implement schedule-like-functions in XS.
407 - use new SLF interface to massively speed up Coro::EV by roughly a
408 factor of two.
409 - used new SLF interface to massively speed up Coro::Semaphore by a
410 factor of three.
411 - used new SLF interface to speed up Coro::AIO by roughly a factor of
412 four and reduce its memory usage considerably.
413 - implement Coro::SemaphoreSet purely in terms of Coro::Semaphore,
414 for a nice speedup and vastly more correct behaviour. Also implement
415 a new method "sem" to get at the underlying semaphore object.
416 - implement Coro::Channel in terms of Coro::Semaphore, for a moderate
417 (in comparison) 20-40% speedup.
418 - used new SLF interface to reimplement Coro::Signal gaining
419 some unknown (because I was too lazy), but certain, speedup, and also
420 making signals reliable for the first time.
421 - used new SLF interface and other optimisations to speed up async_pool
422 by a factor of two. It also doesn't rely on perl's exception mechanism
423 to exit anymore. The overhead for terminating an async_pool, coro over
424 a normal async is now very small.
425 - sped up coroutine creation/destruction by 40%.
426 - forgot to include Coro/libcoro/README in the dist for all these years.
427 - work around a freebsd pthreads bug (manual testcancel is required as
428 pthread_cond_wait isn't a cancellation point on freebsd).
429 - use new rouse functions to speed up and simplify Coro::BDB.
430 - make "prefer perl native functions" work with threaded perls.
431 - condense Coro::Debug ps output, hint at v and w flags.
432 - (libcoro) lots of minor cleanups and portability improvements.
433
4344.914 Wed Nov 19 12:54:18 CET 2008
435 - fix a disastrous bug in the readline optimisation
436 introduced in 4.801.
437
4384.913 Sat Nov 15 07:58:28 CET 2008
439 - async_pool did free a scalar value twice
440 ("Attempt to unreference...").
441
4424.912 Thu Nov 13 18:31:23 CET 2008
443 - minor cleanups.
444 - use much larger stacks on linux and perl < 5.8.8.
445 - Coro::Debug::new_unix_server did not unlink the socket
446 when destroyed.
447
4484.911 Tue Nov 11 04:26:17 CET 2008
449 - "port" to threaded perls.
450
4514.91 Mon Nov 10 05:36:38 CET 2008
452 - the ->throw exception object no longer leaks.
453 - creating a new cctx leaked a scopestack entry (memleak).
454 - new coroutines didn't get created with a zero flags field
455 (unknown impact).
456 - calling ->throw on a not-yet-started coroutine should now work
457 instead of being ignored.
458 - ->throw is now supported on Coro::State objects.
459 - clean up cctx creation code a bit.
460 - entersub is actually a UNOP, not a LOGOP (not a bugfix).
461
4624.9 Sat Nov 8 17:45:27 CET 2008
463 - (libcoro) did not preserve rbp with CORO_ASM (we are getting there).
464 - (libcoro) no longer leak threads in the experimental pthread backend,
465 also speed it up considerably.
466 - (libcoro) do not rely on makecontext passing void *'s unscathed.
467 - fix compiletime dependencies on libcoro in the Makefile.
468 - cctx_count wasn't always updated properly.
469 - Coro::State::cctx_stacksize wasn't applied correctly.
470 - new function Coro::State::cctx_max_idle.
471 - the default max number of idle C contexts is now 4.
472 - (libcoro) try harder to get _setjmp/_longjmp.
473 - (libcoro) cleanup and extend the libcoro API to officially
474 allow the creation of empty source contexts.
475 - very experimental workaround for the totally broken netbsd platform.
476 - tried to hack around openbsd bugs.
477
4784.804 Wed Nov 5 16:36:00 CET 2008
479 - Coro::Debug::new_unix_server would not create a non-blocking listening
480 socket, sometimes causing freezes.
481 - (libcoro) fix misaligned stack points for setjmp and assembly
482 methods, which can cause crashes on x86/x86_64 with a sufficiently
483 aggressive compiler.
484 - new function: Coro::Debug::new_tcp_server.
485 - move ->throw into the Coro class because it only works on coro objects.
486
4874.803 Mon Nov 3 17:16:12 CET 2008
488 - (libcoro) use a global asm statement to become independent of gcc
489 otpimisations for CORO_ASM (thanks to pippijn for the idea).
490 - try to workaround yet another broken bsd, this time dragonfly.
491
4924.802 Thu Oct 30 10:56:12 CET 2008
493 - support -fno-omit-frame-pointer on x86 with the assembly method.
494 - tune 01_unblock.t tests a bit.
495
4964.801 Wed Oct 22 18:33:37 CEST 2008
497 - improve readline speed for very long "lines".
498 - backport to 5.8.8.
499
5004.8 Thu Oct 2 13:34:40 CEST 2008
501 - new function Coro::AIO::aio_wait.
502 - Coro.:AIO and Coro::BDB now "use Coro::AnyEvent".
503 - greatly speed up and reduce memory usage of Coro::AIO requests.
504 - implement some other µ-optimisations.
505
5064.749 Mon Sep 29 14:40:12 CEST 2008
507 - port to slow and broken pseudo-threaded perls. (courtesy pippijn).
508
5094.748 Sat Sep 27 14:03:19 CEST 2008
510 - implement, but do not document, PerlIO::cede(granularity).
511 - Coro::Storable forgot to wrap Storable::pstore.
512 - work around the multitude of leaks and memory corruption
513 bugs in PerlIO::via by using our own C-level perliol. As a side
514 effect, Coro::Storable is now much, much, much faster.
515
5164.747 Tue Sep 23 01:59:41 CEST 2008
517 - fix a per-cv memleak (one empty array was leaked per
518 code reference).
519 - avoid a crash in coro->call|rss when the coroutine was already
520 destroyed (most noticably when using Coro::Debug::ps :)
521 - also protect *Storable::FILE.
522 - push up default storable granularity to 20ms.
523
5244.746 Sun Sep 21 03:22:20 CEST 2008
525 - be more insistent on locking Storable against reentrancy
526 in Coro::Storable.
527 - move swap_def?v and throw to Coro::State, as documented.
528
5294.745 Thu Jul 24 00:14:38 CEST 2008
530 - remove debugging code related to MgPV_nolen_const, also try to
531 make it compile with perl 5.8.6 (yes, apple apparently loves
532 outdated software). Reported by John S.
533
5344.744 Tue Jul 8 22:06:35 CEST 2008
535 - correctly provide default DIE/WARN handlers as documented.
536 - also overwrite PL_vtbl_sigelem.svt_clear, even though current
537 implementations inside perl work fine for us.
538
5394.743 Mon Jun 16 00:21:57 CEST 2008
540 - when using Coro::EV without running EV::loop it could
541 result in busy-waiting for events, this has been fixed.
542 - reduce codesize and improve performance by using EV_DEFAULT_UC.
543
5444.742 Sat May 31 14:10:21 CEST 2008
545 - implement a workaround for (some) perl <5.8.8 versions.
546 - require EV 3.3+.
547
5484.741 Fri May 30 23:33:09 CEST 2008
549 - tell netbsd how utterly broken their imitation of an OS is
550 and refuse to build by default if pthreads are in use.
551 - switch to "s" method on all bsd's by default, as their ucontext
552 stuff seems just too broken.
553 - fix a bug in Coro::Select.
554
5554.74 Thu May 29 20:05:31 CEST 2008
556 - do not test Coro::LWP for lack of dependencies.
557 - do not test Coro::Select for lack of working perls.
558
5594.73 Thu May 29 2008
560 - fix a bug in Coro::EV which would cause it to block despite
561 there being runnable coroutines.
562 - sprinkle "no warnings" freely over everything, also suppress
563 warnings for some other modules.
564 - fix typo in WSAEWOULDBLOCK.
565
5664.72 Sun May 25 05:14:36 CEST 2008
567 - tweak META.yaml a bit, unfortunately, there is no documented way
568 to have optional dependencies with CPAN. doh :(
569 - avoid running some tests on windows because they would fail due to
570 perl bug (broken fork, broken pipes...).
571 - work around perl on windows bugs where perl returns undocumented
572 error codes for sysread, syswrite etc. by taking advantage
573 of AnyEvent's workaround for that problem.
574 - use AnyEvent::Util::fh_nonblocking in Coro::Handle to work around
575 a common perl implementation bug on windows.
576 - use unix domain sockets in testsuite to work around
577 common perl implementation bugs on widows (they are emulated by
578 tcp sockets on windows. ugh.)
7 579
84.71 Sat May 24 20:01:27 CEST 2008 5804.71 Sat May 24 20:01:27 CEST 2008
9 - fix a bug in Coro::AnyEvent ("Usage: Coro::AnyEvent::_schedule()"). 581 - fix a bug in Coro::AnyEvent ("Usage: Coro::AnyEvent::_schedule()").
10 - take advantage of async name resolution of AnyEvent::Util. 582 - take advantage of async name resolution of AnyEvent::Util.
11 - work around brutal inet_aton override in Coro::LWP. 583 - work around brutal inet_aton override in Coro::LWP.

Diff Legend

– Removed lines
+ Added lines
< Changed lines
> Changed lines