| 1 | Revision history for Perl extension Coro. |
1 | Revision history for Perl extension Coro. |
| 2 | |
2 | |
| 3 | TODO: better (earlier) error messages on broken transfer's |
3 | TODO: should explore PerlIO::coroaio (perl leaks like hell). |
| 4 | TODO: destroy gets only called on cancel: fix(?) |
4 | TODO: channel->maxsize(newsize)? |
|
|
5 | TODO: __GCC_HAVE_DWARF2_CFI_ASM |
|
|
6 | TODO: swap_sv, maybe add scope_swap_sv, andallow to unswap by repeat? |
|
|
7 | TODO: swap_sv not undone in pool |
|
|
8 | TODO: croak when async_pool tries to run canceled thread? |
| 5 | |
9 | |
| 6 | 3.3 |
10 | - port to stableperl-5.22.0-1.001. |
|
|
11 | |
|
|
12 | 6.43 Thu Jun 4 15:38:14 CEST 2015 |
|
|
13 | - use stability canary. |
|
|
14 | - update libecb. |
|
|
15 | |
|
|
16 | 6.42 Wed Feb 11 20:29:52 CET 2015 |
|
|
17 | - Coro::SemaphoreSet->try did not actually work (analyzed by |
|
|
18 | SATO Kentaro). |
|
|
19 | - upgrade libecb to be C11 compliant. |
|
|
20 | |
|
|
21 | 6.41 Sat Sep 6 22:08:46 CEST 2014 |
|
|
22 | - restore portability to perl 5.8.x. |
|
|
23 | - give new Coro's a valid GvHV(PL_hintgv) - this is slow and takes |
|
|
24 | up some memory, but fixes "use feature" and similar modules |
|
|
25 | when used inside a Coro. |
|
|
26 | - allow zero as argument to Coro::Channel to mean the same thing |
|
|
27 | as no argument. this works with older versions as well, |
|
|
28 | but wasn't legal until now. |
|
|
29 | - slightly better c header file detection. |
|
|
30 | |
|
|
31 | 6.39 Mon Jun 2 00:00:08 CEST 2014 |
|
|
32 | - work around more incompatible changes in 5.20. |
|
|
33 | |
|
|
34 | 6.38 Sun Jun 1 21:54:23 CEST 2014 |
|
|
35 | - check that perl slots actually have enough space to hold |
|
|
36 | interpreter variables. |
|
|
37 | - untested port to perl 5.19 (64 bit tmps indices) (reported |
|
|
38 | by Andreas König). |
|
|
39 | - croak when cancel is called without a thread context. |
|
|
40 | |
|
|
41 | 6.37 Tue Mar 4 13:27:33 CET 2014 |
|
|
42 | - *sigh*, removed leftover debugging code from debugging a |
|
|
43 | perl bug, of all things. |
|
|
44 | |
|
|
45 | 6.36 Tue Mar 4 07:11:59 CET 2014 |
|
|
46 | - semaphores would not clear the destroy hook when interrupted |
|
|
47 | by ->throw, causing segfaults or worse. |
|
|
48 | - ->throw on a thread waiting for a semaphore did not acquire |
|
|
49 | the semaphore, but also didn't wake up other waiters, |
|
|
50 | possibly causing a deadlock. |
|
|
51 | - "FATAL: $Coro::IDLE blocked itself" will now use Carp::confess |
|
|
52 | to report a full stacktrace, which should help find out |
|
|
53 | where the actual call is. |
|
|
54 | - "atomically" destroy data in slf_destroy, because it is |
|
|
55 | the right thing to do, just in case. |
|
|
56 | - disable ecb.h libm dependency, to compile on ancient systems |
|
|
57 | or under adverse conditions. |
|
|
58 | |
|
|
59 | 6.33 Mon Nov 18 11:26:27 CET 2013 |
|
|
60 | - do not crash when freeing padlists with holes (in 5.18). |
|
|
61 | - tentative SVt_BIND 5.19 port/fix. |
|
|
62 | |
|
|
63 | 6.32 Tue Nov 5 15:35:35 CET 2013 |
|
|
64 | - use a new algorithm to derive padlists for perl 5.18. The old |
|
|
65 | one could lead to 0-pointer accesses inside perl (reported |
|
|
66 | by Darin McBride). |
|
|
67 | |
|
|
68 | 6.31 Thu May 9 07:39:48 CEST 2013 |
|
|
69 | - Coro::AIO requests would crash if the thread was ready'd |
|
|
70 | while the request was ongoing. |
|
|
71 | |
|
|
72 | 6.29 Wed May 8 02:55:18 CEST 2013 |
|
|
73 | - when an on_destroy handler destructs the coro currently being |
|
|
74 | destructed a perl scalar could be accessed after being freed, |
|
|
75 | likely causing a crash. |
|
|
76 | |
|
|
77 | 6.28 Wed Mar 6 06:58:02 CET 2013 |
|
|
78 | - clean remnants of existing __DIE__ and __WARN__ handlers so |
|
|
79 | they lose their magic and will not cause segfaults later |
|
|
80 | (testcase by Andrey Sagulin). |
|
|
81 | - improved Coro::State documentation a bit. |
|
|
82 | - Coro::Debug::command now flushes the output. |
|
|
83 | - add hack detection code for x32 abi, because the braindead slugs |
|
|
84 | who designed that made it look exactly like x86_64 without |
|
|
85 | providing proper compile time symbols to test for it. as a result, |
|
|
86 | this detection cannot work reliably. |
|
|
87 | - valgrind stack registering was broken. |
|
|
88 | - do not rely on Time::HiRes anymore in Coro::Debug. |
|
|
89 | |
|
|
90 | 6.23 Fri Dec 7 23:36:37 CET 2012 |
|
|
91 | - use experimental fiber implementation on native windows |
|
|
92 | perls. |
|
|
93 | - use sizeof (void *) as multiplication factor for stack sizes, |
|
|
94 | to accomodate the totally braindamaged microsoft 64 bit "os". |
|
|
95 | - changed verifier host from win2k-ap510-32 to win7-sp516-32/64. |
|
|
96 | activeperl 5.16 crashes when PerlIO_define_layer is called due |
|
|
97 | to some bug in the perl dll, strawberry perl at least passes |
|
|
98 | the testsuite. |
|
|
99 | - implement Coro::Handle->peeraddr/host/port, for slightly |
|
|
100 | improved compatibility with LWP. |
|
|
101 | - implement 5.17 compatibility by almost blindly applying a |
|
|
102 | good-looking patch by Father Chrysostomos. |
|
|
103 | - move stack management functions into libcoro 3. |
|
|
104 | - libcoro version 3 "released". |
|
|
105 | - support magic values as timed_io_once args. |
|
|
106 | - recommend AnyEvent 7+ or EV 4+, also require EV |
|
|
107 | version 4 or newer for Coro::EV. |
|
|
108 | |
|
|
109 | 6.10 Tue Oct 9 01:14:27 CEST 2012 |
|
|
110 | - updated ecb.h, it had a typo that caused it to not compile on many |
|
|
111 | big endian systems (reported by many people). |
|
|
112 | - disable memory fences in ecb.h to improve portability. |
|
|
113 | |
|
|
114 | 6.09 Sat Oct 6 23:25:02 CEST 2012 |
|
|
115 | - Coro::EV I/O watchers were not interruptible by exceptions |
|
|
116 | (Coro::State::throw) (testcase by sten). |
|
|
117 | - ->throw now puts threads into the ready queue, as this seems to |
|
|
118 | be expected by existing code, and code that doesn't cope with spurious |
|
|
119 | wakeups needs fixing anyway. |
|
|
120 | - use fd -1 in mmap. |
|
|
121 | - cast I32 to int in error message printf. |
|
|
122 | - warn about broken so-called "hardened" kernels. |
|
|
123 | |
|
|
124 | 6.08 Fri Apr 13 12:05:47 CEST 2012 |
|
|
125 | - be more aggressive about exiting like perl does - formerly, |
|
|
126 | exiting from the non-main thread would not execute END blocks. |
|
|
127 | |
|
|
128 | 6.07 Fri Nov 11 21:21:48 CET 2011 |
|
|
129 | - work around a bug in PerlIO (setting $SIG{__WARN__} to a PVCV). |
|
|
130 | - update ecb.h. |
|
|
131 | |
|
|
132 | 6.06 Mon Aug 8 23:59:48 CEST 2011 |
|
|
133 | - cygwin unfortunately patches the stack at runtime, so we use the pthreads |
|
|
134 | backend, which is an order of magnitude slower. unfortunately, cygwins |
|
|
135 | pthread implementation isn't very complete either, so allocate the stack |
|
|
136 | twice just to be sure. |
|
|
137 | (note: cygwin also enables mymalloc, which is NOT THREADSAFE ON WINDOWS, |
|
|
138 | in its ithreaded perl - best recompile cygwin and use the 'w'indows |
|
|
139 | backend for much better performance. also disable ithreads for |
|
|
140 | even better performance...). |
|
|
141 | |
|
|
142 | 6.05 Thu Aug 4 21:36:36 CEST 2011 |
|
|
143 | - blush, condvar values would not be propagated from send to recv anymore |
|
|
144 | (reported by Chip Salzenberg). |
|
|
145 | - use exponential increase for the readline buffer length in |
|
|
146 | Coro::Handle. also reduce initial allocation to 1020 from 4096 bytes. |
|
|
147 | |
|
|
148 | 6.04 Wed Aug 3 17:15:45 CEST 2011 |
|
|
149 | - use even more efficient and more compatible condvars for |
|
|
150 | compatibility to AnyEvent 6.x :) |
|
|
151 | - more inconsequential ecb.h updates. |
|
|
152 | |
|
|
153 | 6.03 Wed Aug 3 11:41:30 CEST 2011 |
|
|
154 | - change how Coro patches AnyEvent condvars for compatibility to |
|
|
155 | AnyEvent 6.x. |
|
|
156 | - update ecb.h, to no longer include <pthread.h> in case WinNT.h |
|
|
157 | hasn't been included. |
|
|
158 | |
|
|
159 | 6.02 Wed Jul 13 04:35:19 CEST 2011 |
|
|
160 | - "improve portability to Gentoo" - gentoo manages to put perl variables |
|
|
161 | in memory areas that are farther than 2gb apart, which the jit couldn't |
|
|
162 | handle and barfed. now it's just a bit slower on gentoo and similar |
|
|
163 | systems. |
|
|
164 | |
|
|
165 | 6.01 Sun Jul 3 12:31:14 CEST 2011 |
|
|
166 | - workarounds are good, but the test for whether pthreads are used |
|
|
167 | was not good. this one should be better. |
|
|
168 | - check differently whether gcc generates cfi instructions itself. |
|
|
169 | |
|
|
170 | 6.0 Wed Jun 29 19:43:35 CEST 2011 |
|
|
171 | - INCOMPATIBLE CHANGE: unreferenced coro objects will now be |
|
|
172 | destroyed and cleaned up automatically (e.g. async { schedule }). |
|
|
173 | - implement a JIT compiler for part of the thread switch code, |
|
|
174 | which gives a 50% speed improvement on threaded perls, and |
|
|
175 | about 4% on non-threaded perls (so threaded perls now finally |
|
|
176 | reach about half the speed of non-threaded perls). |
|
|
177 | - slightly modernise Coro::Intro, add section about rouse functions. |
|
|
178 | - avoid DEFSV and ERRSV, giving another 10% improvement |
|
|
179 | in thread switching. |
|
|
180 | - Coro::State->is_destroyed is now called is_zombie. |
|
|
181 | - implement a Coro->safe_cancel method that might fail, but |
|
|
182 | cancels in a "safer" way if it succeeds. |
|
|
183 | - add preliminary support for DEBUGGING perls. |
|
|
184 | - get rid of two hash-accesses when initialising a new Coro - this |
|
|
185 | speeds up coro creation by almost a factor of two. |
|
|
186 | - croak when a coro that is being cancelled tries to block |
|
|
187 | (e.g. while executing a guard block), instead of crashing or |
|
|
188 | deadlocking. |
|
|
189 | - use a more robust and also faster method to identify Coro::State |
|
|
190 | objects - speeds up everything a bit. |
|
|
191 | - implement Coro->cancel in XS for a 20% speed improvement, and to |
|
|
192 | be able to implement mutual cancellation. |
|
|
193 | - speed up context switches by a percent or two by more efficiently |
|
|
194 | allocating context stack entries. |
|
|
195 | - implement Coro->join and Coro->on_destroy in XS for a speedup and |
|
|
196 | a reduction in memory use. |
|
|
197 | - cancelling a coro while it itself is cancelling another coro is |
|
|
198 | now supported and working, instead of triggering an assertion. |
|
|
199 | - be a bit more crash-resistant when calling (buggy) on_destroy |
|
|
200 | callbacks (best effort). |
|
|
201 | - move on_destroy into the slf_frame, to allow extension slf |
|
|
202 | functions to have destructors. |
|
|
203 | - get rid if coro refcounting - simply crash in other interpreter |
|
|
204 | threads by nulling the pointers on clone. |
|
|
205 | - simplify warn/die hook handling when loading Coro - the convoluted |
|
|
206 | logic seems to be no longer neccessary. |
|
|
207 | - use libecb instead of our own home-grown gcc hacks. |
|
|
208 | - document alternatives to Coro::LWP. Please use them :) |
|
|
209 | - work around another mindless idiotic NEEDLESS bug in openbsd/mirbsds |
|
|
210 | sigaltstack. Really. wine suffers from it, erlang suffers from it, |
|
|
211 | and it's known since at least 2006. |
|
|
212 | |
|
|
213 | 5.372 Wed Feb 23 06:14:30 CET 2011 |
|
|
214 | - apparently mingw doesn't provide a working gettimeofday, try to |
|
|
215 | work around that by relying on Time::HiRes (indirectly brought to |
|
|
216 | my attention by Max Maischein). |
|
|
217 | - fix some portability issues when Time::HiRes was used. |
|
|
218 | |
|
|
219 | 5.371 Mon Feb 21 14:36:08 CET 2011 |
|
|
220 | - backport to windows process emulation code again. |
|
|
221 | |
|
|
222 | 5.37 Sat Feb 19 07:49:44 CET 2011 |
|
|
223 | - add a big "Coro thread life cycle" section to "man Coro". |
|
|
224 | - try a tentative workaround against the breakage that 5.13 has |
|
|
225 | introduced without depreciation period. sigh. |
|
|
226 | - no longer use Time::HiRes if gettimeofday is available, which |
|
|
227 | saves quite a lot of memory. |
|
|
228 | |
|
|
229 | 5.36 Sun Feb 13 05:33:41 CET 2011 |
|
|
230 | - automatically load Coro::Channel, Coro::RWLock, Coro::Semaphore, |
|
|
231 | Coro::SemaphoreSet, Coro::Signal and Coro::Specific on first "new" |
|
|
232 | method call. |
|
|
233 | - undocument Coro::Timer::sleep and obsolete whole module. |
|
|
234 | - optimise Coro::Timer::timeout memory and cpu usage. |
|
|
235 | - slightly updated Coro::Intro for recent changes. |
|
|
236 | - do not initialise PL_dirty anymore. |
|
|
237 | |
|
|
238 | 5.25 Thu Nov 11 01:08:39 CET 2010 |
|
|
239 | - try a different approach on netbsd - netbsd 5 finally has marginally |
|
|
240 | working pthreads, but still broken ucontext/sigaltstack. |
|
|
241 | - openbsd 4.8 finally got their act together, Coro works out of the box |
|
|
242 | with asm, setjmp and pthreads (no change, just informational). |
|
|
243 | |
|
|
244 | 5.24 Sat Oct 23 11:27:12 CEST 2010 |
|
|
245 | - port to the EV 4.0 API. |
|
|
246 | - work around bugs in mingw32, making strawberry perl work |
|
|
247 | out of the box. |
|
|
248 | - correctly modify Coro::AIO function prototypes |
|
|
249 | so that they reflect the "no optional parameters" rule. |
|
|
250 | - "ported" libcoro to C++. |
|
|
251 | |
|
|
252 | 5.23 Mon May 17 18:50:42 CEST 2010 |
|
|
253 | - be more resistant to ordering changes when initialising |
|
|
254 | Coro::AnyEvent, Coro::EV and Coro::Event (reported by Matthias |
|
|
255 | Waldorf). |
|
|
256 | - document that perl 5.12 deliberately removed support for cloning. |
|
|
257 | |
|
|
258 | 5.22 Wed Apr 14 03:55:35 CEST 2010 |
|
|
259 | - correctly return udnef on errors in Coro::Handle::read/write |
|
|
260 | (testcase by Marc Mims). |
|
|
261 | - convert Coro::Util into a "perl compatibility wrapper" - the functions |
|
|
262 | are less useful now, but are drop-in replacements for existing |
|
|
263 | functions, listing better alternatives in the documentation. This also |
|
|
264 | fixes a bug in Coro::LWP which naively substituted Socket::inet_aton |
|
|
265 | with Coro::Util::inet_aton. |
|
|
266 | - do not override $Coro::idle unconditionally in Coro.pm, as other |
|
|
267 | modules could have provided their own idle coro already |
|
|
268 | (for exmaple, Coro::AnyEvent). |
|
|
269 | - fix Coro::Util::gethost* functions. |
|
|
270 | - Coro::Timer corretcly exports it's symbols (reported by Hideki Yamamura). |
|
|
271 | |
|
|
272 | 5.21 Wed Dec 16 07:19:51 CET 2009 |
|
|
273 | - automatically load Coro::AnyEvent when AnyEvent and Coro are used |
|
|
274 | together. |
|
|
275 | - add some examples on how to combine other event loops with Coro in |
|
|
276 | Coro::AnyEvent, and how to run it (and not to block). Seems to be |
|
|
277 | the most common source of confusion. |
|
|
278 | - try to catch people naively blocking in an event callback. |
|
|
279 | - work around the perl filehandle bug issue in conjunction with |
|
|
280 | older common::sense (as indirectly pointed out by ZSystem). |
|
|
281 | - clarify the "not from signal handlers" section. |
|
|
282 | |
|
|
283 | 5.2 Sun Oct 4 14:54:24 CEST 2009 |
|
|
284 | - Coro::Storable destroyed the prototypes of the functions it wrapped. |
|
|
285 | - export rouse_cb and rouse_wait by default now. |
|
|
286 | - fix various prototype mismatches in Coro::AnyEvent and Coro::Handle. |
|
|
287 | - new method $state->swap_sv. |
|
|
288 | - added section on "windows process emulation" to the manpage, |
|
|
289 | after a not-so-fruitful (nor-friendly) "discussion" with chip |
|
|
290 | salzenberg (discussion implies arguments, but his only arguments |
|
|
291 | were ad-hominems, one wonders why he started it in the first |
|
|
292 | place). I hope this explains it well enough for him to understand, |
|
|
293 | and maybe well enough for others to understand. |
|
|
294 | - use common::sense everywhere now. |
|
|
295 | - idle callbacks are no longer supported, use idle coros instead. |
|
|
296 | - print a thread listing when a deadlock is detected. |
|
|
297 | |
|
|
298 | 5.17 Sat Aug 22 23:09:31 CEST 2009 |
|
|
299 | - work around a bug in the perl debugger causing crashes |
|
|
300 | when running under the debugger by marking _pool_handler as nodebug. |
|
|
301 | - speed up Coro::async considerably. |
|
|
302 | - try some hacks to get netbsd to work "more often" - their broken |
|
|
303 | setjmp/longjmp, ucontext *and* phtreads are really hard on Coro. |
|
|
304 | - convert Coro to AE 5.0 API. |
|
|
305 | |
|
|
306 | 5.162 Tue Jul 28 04:04:03 CEST 2009 |
|
|
307 | - perl 5.8.2 is now minimum requirement. |
|
|
308 | - skip t/19_handle.t on broken windows perls. |
|
|
309 | |
|
|
310 | 5.161 Wed Jul 22 04:47:38 CEST 2009 |
|
|
311 | - Coro::AnyEvent::poll could have a different prototype when EV was |
|
|
312 | used as backend (analyzed by Tatsuhiko Miyagawa). |
|
|
313 | - Coro::AnyEvent errornously initialised the event loop when loaded, |
|
|
314 | not on demand. |
|
|
315 | - try to workaround rare */t/01_unblock.t failures. |
|
|
316 | |
|
|
317 | 5.16 Tue Jul 21 01:44:37 CEST 2009 |
|
|
318 | - Coro::AnyEvent failed to hook into the event loop |
|
|
319 | when no threads had been readied between detecting |
|
|
320 | the event loop and actually running it. |
|
|
321 | - considerably speed up Coro::Select by taking avdantage |
|
|
322 | of AnyEvent > 4.8 and some other optimisations. |
|
|
323 | - implement paragraph readline mode in Coro::Handle |
|
|
324 | (based on patches by Zsbán Ambrus). |
|
|
325 | - replace WSAEINPROGRESS by WSAEWOULDBLOCK (reported |
|
|
326 | and analyzed by Yasuhiro MATSUMOTO). |
|
|
327 | - clarified libcoro license and copyright. |
|
|
328 | - someone stole my EXTRA_META!!! |
|
|
329 | - implement Coro::Select::patch_pp_sselect and it's brother, |
|
|
330 | for hardcode select overriding. |
|
|
331 | |
|
|
332 | 5.151 Mon Jul 6 05:41:57 CEST 2009 |
|
|
333 | - backport to windows process emulation code again (patch by |
|
|
334 | Yasuhiro MATSUMOTO). |
|
|
335 | - slightly update Coro::MakeMaker. |
|
|
336 | |
|
|
337 | 5.15 Tue Jun 30 10:28:06 CEST 2009 |
|
|
338 | - deprecate Coro::Socket, document how to get ipv6 support via |
|
|
339 | AnyEvent::Socket instead. |
|
|
340 | - implement signal->wait ($cb) interface, similar to semaphores. |
|
|
341 | - work around SvOK not supporting getmagic, so we have to getmagic |
|
|
342 | to test for undef :/ (reported by Matthias Waldorf). |
|
|
343 | - load Coro::AnyEvent in all modules using AnyEvent. |
|
|
344 | - work around perl corrupting our internal data structures, |
|
|
345 | reported by Tokuhiro Matsuno. |
|
|
346 | - enable per-coroutine real and cpu time gathering |
|
|
347 | (Coro::State::enable_times). |
|
|
348 | |
|
|
349 | 5.14 Wed Jun 24 01:37:48 CEST 2009 |
|
|
350 | - provide explicit functions to "cede" to the event loop in Coro::AnyEvent, |
|
|
351 | as this seems to have been a difficult concept (poll, sleep, idle, |
|
|
352 | idle_upto). |
|
|
353 | - add Coro::AnyEvent::readable/writable functions. |
|
|
354 | - clarify Coro::EV/Event/AnyEvent manpages. |
|
|
355 | - free per-thread global scalars in the thread calling ->cancel, to |
|
|
356 | avoid crashes when $_, $@ etc., are magical but some of those |
|
|
357 | had already been freed. |
|
|
358 | - "unexperimentalise" the callback interface for Coro::Semaphore. |
|
|
359 | - speed up ready queue management/context switching by using a linked |
|
|
360 | list instead of an array (~5-10%). |
|
|
361 | - implement "watch" command in Coro::Debug shells. |
|
|
362 | - for fun, implement time-slicing as an example in the manpage. |
|
|
363 | - if AnyEvent detects EV or Event, but we don't have Coro::EV or |
|
|
364 | Coro::Event, use the normal AnyEvent handling instead of dieing |
|
|
365 | (the same is true for Coro::Handle). |
|
|
366 | - properly document Coro::EV::timed_io_once. |
|
|
367 | - avoid unneccessary ->cancel calls in Coro::Handle. |
|
|
368 | - maybe make it work on mingw32 with win32 backend |
|
|
369 | (based on patch by Yasuhiro Matsumoto). |
|
|
370 | |
|
|
371 | 5.132 Fri May 29 09:00:39 CEST 2009 |
|
|
372 | - do not keep a reference to the argument itself in |
|
|
373 | Coro::Semaphore::guard, as it could change later. |
|
|
374 | - support SO_RCVBUF/SO_SNDBUF nonstandard Coro::Socket options, |
|
|
375 | should support a prepare callback. |
|
|
376 | |
|
|
377 | 5.131 Mon Mar 16 23:20:37 CET 2009 |
|
|
378 | - implement and document Coro->suspend, Coro->resume. |
|
|
379 | - fix Coro::Select implementation to not (often) close |
|
|
380 | the passed file descriptors (testcase provided by pippijn). |
|
|
381 | |
|
|
382 | 5.13 Mon Dec 15 21:51:42 CET 2008 |
|
|
383 | - EXPERIMENTAL: implement dynamic winds (on_enter/on_leave). |
|
|
384 | - don't set diehook to C<undef> but instead to NULL, to avoid |
|
|
385 | spurious warnings. |
|
|
386 | - fix a lot of bugs in Coro::SemaphoreSet. |
|
|
387 | - Coro::SemaphoreSet will less often create a semaphore needlessly. |
|
|
388 | - add Coro::SemaphoreSet::count and wait methods. |
|
|
389 | - take advantage of the new Guard module. |
|
|
390 | - deprecate Coro::guard. |
|
|
391 | - try to fix the dreaded 01_unblock tests once more. I hate it when |
|
|
392 | testsuites need more fixing than the code they are supposed to test. |
|
|
393 | - croak in more cases when a required callback isn't resolvable. |
|
|
394 | - fix some minor issues in Coro::State->call/eval. |
|
|
395 | - use current coroutine context instead of a temporary one |
|
|
396 | when temporarily switching to another coroutine. |
|
|
397 | - do not call C-level on_destroy handlers during global destruction, |
|
|
398 | to avoid needless segfaults. |
|
|
399 | |
|
|
400 | 5.12 Sun Dec 7 13:30:38 CET 2008 |
|
|
401 | - add default config for MirOS, which seems to be a bug-to-bug |
|
|
402 | compatible fork of openbsd ("world domination by releasing |
|
|
403 | openbsd cvs before the openbsd folks do it" or so :). |
|
|
404 | - free_padlist did destroy the names pad, not good, but didn't |
|
|
405 | seem to bother perl - this could fix issues such as eval "" |
|
|
406 | inside a function called from multiple coroutines. |
|
|
407 | - use a different method to detect destruction time. |
|
|
408 | - be more careful when freeing padlists just before global |
|
|
409 | destruction. |
|
|
410 | - fixed and expanded the call/cc example. |
|
|
411 | - renamed _terminate to _coro_run. |
|
|
412 | - new method Coro::Channel->shutdown. |
|
|
413 | - try pthreads on openbsd <4.4 (broken sigaltstack, will |
|
|
414 | pthreads fare better?). |
|
|
415 | - be less picky about destroying "a" running coroutine. |
|
|
416 | |
|
|
417 | 5.11 Tue Nov 25 21:49:05 CET 2008 |
|
|
418 | - DEBUGGING in 5.10.0 is a mess: it constantly flags perfectly |
|
|
419 | working code with failed assertions, introducing more bugs than |
|
|
420 | it fixes, requiring elaborate workarounds :( |
|
|
421 | |
|
|
422 | 5.1 Mon Nov 24 08:54:59 CET 2008 |
|
|
423 | - wrote a small introductory tutorial - Coro::Intro. |
|
|
424 | - convert Coro::Timer, Coro::Select and Coro::Util to rouse API. |
|
|
425 | - Coro::Select did errornously dup the file descriptors |
|
|
426 | and didn't work with all AnyEvent backends. |
|
|
427 | - Coro::Select wasn't imported correctly form Coro::LWP, causing blocking |
|
|
428 | LWP data transfers. |
|
|
429 | - disassociate c contexts from coro objects - this is agruably more |
|
|
430 | correct, but mostly allows sharing of cctxs between coro and state |
|
|
431 | objects, for added memory savings and speed increases. |
|
|
432 | - bumped $Coro::POOL_RSS up to 32kb by default. |
|
|
433 | - no longer set the optype to OP_CUSTOM, as B::* understandably |
|
|
434 | doesn't like this very much (and we *are* a type of entersub). |
|
|
435 | - implement state cloning, just to prove that call/cc can be done. |
|
|
436 | - automatically load Coro::AnyEvent in Coro::Handle. |
|
|
437 | - wrap ->cancel calls in eval inside Coro::Handle as EV watchers |
|
|
438 | do not have this method (and don't need it either). |
|
|
439 | - speed up generic anyevent methods in Coro::Handle by using rouse |
|
|
440 | callbacks. |
|
|
441 | - allow coroutines in $Coro::IDLE, speeding up Coro::AnyEvent and |
|
|
442 | others. It also makes the debugger happier, as you can trace |
|
|
443 | through the idle threads now. |
|
|
444 | - add comppad_name* and hints ($^H, %^H) to per-thread variables. |
|
|
445 | - eg/event was pretty broken. |
|
|
446 | - better 5.8.6 compatibility. |
|
|
447 | |
|
|
448 | 5.0 Thu Nov 20 10:35:05 CET 2008 |
|
|
449 | - NEW ARCHITECTURE: use the latest 4.x version if you experience |
|
|
450 | stability issues. |
|
|
451 | - bump API version to 7 - all dependents must be recompiled. |
|
|
452 | - removed timed_* functions - they were not being used anyways |
|
|
453 | and should be replaced by a more generic mechanism - |
|
|
454 | and were annoying to support anyways :) |
|
|
455 | - removed SemaphoreSet's waiter method - use sem method instead. |
|
|
456 | - Coro::Semaphore->adjust didn't correctly wake up enough waiters. |
|
|
457 | - async_pool did free a scalar value twice |
|
|
458 | ("Attempt to unreference..."). |
|
|
459 | - fix a longstanding bug where calling terminate on a coro that |
|
|
460 | was waiting for a semaphore that was just becoming available |
|
|
461 | would cause a deadlock (semaphore would get into a state where |
|
|
462 | it was available but waiters were still blocked). |
|
|
463 | - calling throw on a coroutine that is waiting for a semaphore will |
|
|
464 | no longer make it acquire the semaphore (and thus leak a count). |
|
|
465 | - perl's process emulation is now not even theoretically supported |
|
|
466 | anymore. |
|
|
467 | - new functions Coro::rouse_cb and Coro::rouse_wait for easier |
|
|
468 | conversion of callback-style to blocking-style. |
|
|
469 | - new methods $coro->schedule_to and ->cede_to, to specifically |
|
|
470 | schedule or cede to a specific coroutine. |
|
|
471 | - new function Coro::Semaphore::wait. |
|
|
472 | - use named constants in Coro::Channel (Richard Hundt). |
|
|
473 | - directly patch the entersub opcode calling SLF functions (cede, |
|
|
474 | transfer and so on). this does speed up context switching, but |
|
|
475 | more importanly, it frees us from the hardcoded behaviour of |
|
|
476 | entersub, so we might actually be able to return something from |
|
|
477 | those functions and atcually create new ones. |
|
|
478 | - take advantage of __builtin_frame_address on gcc. |
|
|
479 | - expose THX in coroapi (not sure whether this was a wise decision, |
|
|
480 | as "threaded" perls are running at half speed anyways). |
|
|
481 | - implement execute_slf (schedule-like-function) interface that makes |
|
|
482 | it possible to implement schedule-like-functions in XS. |
|
|
483 | - use new SLF interface to massively speed up Coro::EV by roughly a |
|
|
484 | factor of two. |
|
|
485 | - used new SLF interface to massively speed up Coro::Semaphore by a |
|
|
486 | factor of three. |
|
|
487 | - used new SLF interface to speed up Coro::AIO by roughly a factor of |
|
|
488 | four and reduce its memory usage considerably. |
|
|
489 | - implement Coro::SemaphoreSet purely in terms of Coro::Semaphore, |
|
|
490 | for a nice speedup and vastly more correct behaviour. Also implement |
|
|
491 | a new method "sem" to get at the underlying semaphore object. |
|
|
492 | - implement Coro::Channel in terms of Coro::Semaphore, for a moderate |
|
|
493 | (in comparison) 20-40% speedup. |
|
|
494 | - used new SLF interface to reimplement Coro::Signal gaining |
|
|
495 | some unknown (because I was too lazy), but certain, speedup, and also |
|
|
496 | making signals reliable for the first time. |
|
|
497 | - used new SLF interface and other optimisations to speed up async_pool |
|
|
498 | by a factor of two. It also doesn't rely on perl's exception mechanism |
|
|
499 | to exit anymore. The overhead for terminating an async_pool, coro over |
|
|
500 | a normal async is now very small. |
|
|
501 | - sped up coroutine creation/destruction by 40%. |
|
|
502 | - forgot to include Coro/libcoro/README in the dist for all these years. |
|
|
503 | - work around a freebsd pthreads bug (manual testcancel is required as |
|
|
504 | pthread_cond_wait isn't a cancellation point on freebsd). |
|
|
505 | - use new rouse functions to speed up and simplify Coro::BDB. |
|
|
506 | - make "prefer perl native functions" work with threaded perls. |
|
|
507 | - condense Coro::Debug ps output, hint at v and w flags. |
|
|
508 | - (libcoro) lots of minor cleanups and portability improvements. |
|
|
509 | |
|
|
510 | 4.914 Wed Nov 19 12:54:18 CET 2008 |
|
|
511 | - fix a disastrous bug in the readline optimisation |
|
|
512 | introduced in 4.801. |
|
|
513 | |
|
|
514 | 4.913 Sat Nov 15 07:58:28 CET 2008 |
|
|
515 | - async_pool did free a scalar value twice |
|
|
516 | ("Attempt to unreference..."). |
|
|
517 | |
|
|
518 | 4.912 Thu Nov 13 18:31:23 CET 2008 |
|
|
519 | - minor cleanups. |
|
|
520 | - use much larger stacks on linux and perl < 5.8.8. |
|
|
521 | - Coro::Debug::new_unix_server did not unlink the socket |
|
|
522 | when destroyed. |
|
|
523 | |
|
|
524 | 4.911 Tue Nov 11 04:26:17 CET 2008 |
|
|
525 | - "port" to threaded perls. |
|
|
526 | |
|
|
527 | 4.91 Mon Nov 10 05:36:38 CET 2008 |
|
|
528 | - the ->throw exception object no longer leaks. |
|
|
529 | - creating a new cctx leaked a scopestack entry (memleak). |
|
|
530 | - new coroutines didn't get created with a zero flags field |
|
|
531 | (unknown impact). |
|
|
532 | - calling ->throw on a not-yet-started coroutine should now work |
|
|
533 | instead of being ignored. |
|
|
534 | - ->throw is now supported on Coro::State objects. |
|
|
535 | - clean up cctx creation code a bit. |
|
|
536 | - entersub is actually a UNOP, not a LOGOP (not a bugfix). |
|
|
537 | |
|
|
538 | 4.9 Sat Nov 8 17:45:27 CET 2008 |
|
|
539 | - (libcoro) did not preserve rbp with CORO_ASM (we are getting there). |
|
|
540 | - (libcoro) no longer leak threads in the experimental pthread backend, |
|
|
541 | also speed it up considerably. |
|
|
542 | - (libcoro) do not rely on makecontext passing void *'s unscathed. |
|
|
543 | - fix compiletime dependencies on libcoro in the Makefile. |
|
|
544 | - cctx_count wasn't always updated properly. |
|
|
545 | - Coro::State::cctx_stacksize wasn't applied correctly. |
|
|
546 | - new function Coro::State::cctx_max_idle. |
|
|
547 | - the default max number of idle C contexts is now 4. |
|
|
548 | - (libcoro) try harder to get _setjmp/_longjmp. |
|
|
549 | - (libcoro) cleanup and extend the libcoro API to officially |
|
|
550 | allow the creation of empty source contexts. |
|
|
551 | - very experimental workaround for the totally broken netbsd platform. |
|
|
552 | - tried to hack around openbsd bugs. |
|
|
553 | |
|
|
554 | 4.804 Wed Nov 5 16:36:00 CET 2008 |
|
|
555 | - Coro::Debug::new_unix_server would not create a non-blocking listening |
|
|
556 | socket, sometimes causing freezes. |
|
|
557 | - (libcoro) fix misaligned stack points for setjmp and assembly |
|
|
558 | methods, which can cause crashes on x86/x86_64 with a sufficiently |
|
|
559 | aggressive compiler. |
|
|
560 | - new function: Coro::Debug::new_tcp_server. |
|
|
561 | - move ->throw into the Coro class because it only works on coro objects. |
|
|
562 | |
|
|
563 | 4.803 Mon Nov 3 17:16:12 CET 2008 |
|
|
564 | - (libcoro) use a global asm statement to become independent of gcc |
|
|
565 | otpimisations for CORO_ASM (thanks to pippijn for the idea). |
|
|
566 | - try to workaround yet another broken bsd, this time dragonfly. |
|
|
567 | |
|
|
568 | 4.802 Thu Oct 30 10:56:12 CET 2008 |
|
|
569 | - support -fno-omit-frame-pointer on x86 with the assembly method. |
|
|
570 | - tune 01_unblock.t tests a bit. |
|
|
571 | |
|
|
572 | 4.801 Wed Oct 22 18:33:37 CEST 2008 |
|
|
573 | - improve readline speed for very long "lines". |
|
|
574 | - backport to 5.8.8. |
|
|
575 | |
|
|
576 | 4.8 Thu Oct 2 13:34:40 CEST 2008 |
|
|
577 | - new function Coro::AIO::aio_wait. |
|
|
578 | - Coro.:AIO and Coro::BDB now "use Coro::AnyEvent". |
|
|
579 | - greatly speed up and reduce memory usage of Coro::AIO requests. |
|
|
580 | - implement some other µ-optimisations. |
|
|
581 | |
|
|
582 | 4.749 Mon Sep 29 14:40:12 CEST 2008 |
|
|
583 | - port to slow and broken pseudo-threaded perls. (courtesy pippijn). |
|
|
584 | |
|
|
585 | 4.748 Sat Sep 27 14:03:19 CEST 2008 |
|
|
586 | - implement, but do not document, PerlIO::cede(granularity). |
|
|
587 | - Coro::Storable forgot to wrap Storable::pstore. |
|
|
588 | - work around the multitude of leaks and memory corruption |
|
|
589 | bugs in PerlIO::via by using our own C-level perliol. As a side |
|
|
590 | effect, Coro::Storable is now much, much, much faster. |
|
|
591 | |
|
|
592 | 4.747 Tue Sep 23 01:59:41 CEST 2008 |
|
|
593 | - fix a per-cv memleak (one empty array was leaked per |
|
|
594 | code reference). |
|
|
595 | - avoid a crash in coro->call|rss when the coroutine was already |
|
|
596 | destroyed (most noticably when using Coro::Debug::ps :) |
|
|
597 | - also protect *Storable::FILE. |
|
|
598 | - push up default storable granularity to 20ms. |
|
|
599 | |
|
|
600 | 4.746 Sun Sep 21 03:22:20 CEST 2008 |
|
|
601 | - be more insistent on locking Storable against reentrancy |
|
|
602 | in Coro::Storable. |
|
|
603 | - move swap_def?v and throw to Coro::State, as documented. |
|
|
604 | |
|
|
605 | 4.745 Thu Jul 24 00:14:38 CEST 2008 |
|
|
606 | - remove debugging code related to MgPV_nolen_const, also try to |
|
|
607 | make it compile with perl 5.8.6 (yes, apple apparently loves |
|
|
608 | outdated software). Reported by John S. |
|
|
609 | |
|
|
610 | 4.744 Tue Jul 8 22:06:35 CEST 2008 |
|
|
611 | - correctly provide default DIE/WARN handlers as documented. |
|
|
612 | - also overwrite PL_vtbl_sigelem.svt_clear, even though current |
|
|
613 | implementations inside perl work fine for us. |
|
|
614 | |
|
|
615 | 4.743 Mon Jun 16 00:21:57 CEST 2008 |
|
|
616 | - when using Coro::EV without running EV::loop it could |
|
|
617 | result in busy-waiting for events, this has been fixed. |
|
|
618 | - reduce codesize and improve performance by using EV_DEFAULT_UC. |
|
|
619 | |
|
|
620 | 4.742 Sat May 31 14:10:21 CEST 2008 |
|
|
621 | - implement a workaround for (some) perl <5.8.8 versions. |
|
|
622 | - require EV 3.3+. |
|
|
623 | |
|
|
624 | 4.741 Fri May 30 23:33:09 CEST 2008 |
|
|
625 | - tell netbsd how utterly broken their imitation of an OS is |
|
|
626 | and refuse to build by default if pthreads are in use. |
|
|
627 | - switch to "s" method on all bsd's by default, as their ucontext |
|
|
628 | stuff seems just too broken. |
|
|
629 | - fix a bug in Coro::Select. |
|
|
630 | |
|
|
631 | 4.74 Thu May 29 20:05:31 CEST 2008 |
|
|
632 | - do not test Coro::LWP for lack of dependencies. |
|
|
633 | - do not test Coro::Select for lack of working perls. |
|
|
634 | |
|
|
635 | 4.73 Thu May 29 2008 |
|
|
636 | - fix a bug in Coro::EV which would cause it to block despite |
|
|
637 | there being runnable coroutines. |
|
|
638 | - sprinkle "no warnings" freely over everything, also suppress |
|
|
639 | warnings for some other modules. |
|
|
640 | - fix typo in WSAEWOULDBLOCK. |
|
|
641 | |
|
|
642 | 4.72 Sun May 25 05:14:36 CEST 2008 |
|
|
643 | - tweak META.yaml a bit, unfortunately, there is no documented way |
|
|
644 | to have optional dependencies with CPAN. doh :( |
|
|
645 | - avoid running some tests on windows because they would fail due to |
|
|
646 | perl bug (broken fork, broken pipes...). |
|
|
647 | - work around perl on windows bugs where perl returns undocumented |
|
|
648 | error codes for sysread, syswrite etc. by taking advantage |
|
|
649 | of AnyEvent's workaround for that problem. |
|
|
650 | - use AnyEvent::Util::fh_nonblocking in Coro::Handle to work around |
|
|
651 | a common perl implementation bug on windows. |
|
|
652 | - use unix domain sockets in testsuite to work around |
|
|
653 | common perl implementation bugs on widows (they are emulated by |
|
|
654 | tcp sockets on windows. ugh.) |
|
|
655 | |
|
|
656 | 4.71 Sat May 24 20:01:27 CEST 2008 |
|
|
657 | - fix a bug in Coro::AnyEvent ("Usage: Coro::AnyEvent::_schedule()"). |
|
|
658 | - take advantage of async name resolution of AnyEvent::Util. |
|
|
659 | - work around brutal inet_aton override in Coro::LWP. |
|
|
660 | - take advantage of the readyhook in Coro::EV, for smoother |
|
|
661 | scheduling. |
|
|
662 | |
|
|
663 | 4.7 Sun May 11 00:32:19 CEST 2008 |
|
|
664 | - completely reworked the Coro manpage. |
|
|
665 | - added Coro::AnyEvent, generic event loop integration. |
|
|
666 | - implement cancel, ready and kill commands in Coro::Debug. |
|
|
667 | - document find_coro in Coro::Debug. |
|
|
668 | - incompatible change: rename has_stack to has_cctx. |
|
|
669 | - Coro::AIO and Coro::BDB no longer force event model detection, |
|
|
670 | use AnyEvent::AIO and AnyEvent::BDB. |
|
|
671 | |
|
|
672 | 4.6 Sat Apr 26 10:05:14 CEST 2008 |
|
|
673 | - INCOMPATIBLE CHANGE: sub/code attributes are no longer supported |
|
|
674 | by the Coro module. It was a mistake to have it in the first place. |
|
|
675 | - (experimental) support for activestate perl 5.10 (method "w"). |
|
|
676 | - (experimental) support for strawberry perl (method "a"). |
|
|
677 | - coro_sigelem_set did not return a value although it had to, |
|
|
678 | actual impact unknown. |
|
|
679 | |
|
|
680 | 4.51 Mon Apr 14 13:28:27 CEST 2008 |
|
|
681 | - make it compile again on 5.8. |
|
|
682 | |
|
|
683 | 4.50 Thu Apr 10 09:43:17 CEST 2008 |
|
|
684 | - I did it twice! (see 4.49). |
|
|
685 | |
|
|
686 | 4.49 Sun Apr 6 21:23:31 CEST 2008 |
|
|
687 | - grr, instead of compiling the recent changes on 5.10 only they |
|
|
688 | were compiled on 5.8 only. |
|
|
689 | |
|
|
690 | 4.48 Sun Apr 6 20:36:46 CEST 2008 |
|
|
691 | - allow coroutine switches during eval's under 5.10.x, as it seems |
|
|
692 | the parser is a per-interpreter option now, so this is safe |
|
|
693 | (this might fix the odd crashing bug). |
|
|
694 | - drop support for 5.9.x versions: they are dead, jim. |
|
|
695 | |
|
|
696 | 4.47 Sun Apr 6 00:37:52 CEST 2008 |
|
|
697 | - force cctx allocation on API calls: we know we need to force one |
|
|
698 | and gcc actually manages to confuse our heuristic nowadays, |
|
|
699 | leading to crashes and worse. |
|
|
700 | - document force_cctx. |
|
|
701 | |
|
|
702 | 4.46 Fri Apr 4 22:05:43 CEST 2008 |
|
|
703 | - upgrade libcoro, resulting in pthread-backend (which was only created |
|
|
704 | to fulfill the rules of the programming languages shootout). |
|
|
705 | |
|
|
706 | 4.45 Thu Mar 13 11:55:36 CET 2008 |
|
|
707 | - fix a file leaking bug in eg/mhyttpd that would allow |
|
|
708 | downloading of any file (reported by oesi). |
|
|
709 | - fix deadlock bug in Coro::Channel (reported by Richard Hundt) |
|
|
710 | (also add testcase). |
|
|
711 | - support Broadcast option in Coro::Socket (patch by Richard Hundt, |
|
|
712 | apparently having loads of fun with that). |
|
|
713 | - implement $state->swap_defsv, ->swap_defav and document ->throw. |
|
|
714 | |
|
|
715 | 4.4 Wed Feb 13 16:44:29 CET 2008 |
|
|
716 | - only assume .cfi pseudo isns exist on GNU/Linux. |
|
|
717 | - add get_timed method to Coro::Channels. |
|
|
718 | - fixed Coro::Debug manpage. |
|
|
719 | - perl 5.11 compatbility improvement based on patch by |
|
|
720 | Andreas König. |
|
|
721 | |
|
|
722 | 4.37 Sun Jan 20 11:25:23 CET 2008 |
|
|
723 | - soften the check for an active parser for perl 5.10 (avoiding |
|
|
724 | "Coro::State::transfer called while parsing" in many cases). |
|
|
725 | |
|
|
726 | 4.36 Sun Jan 13 10:53:56 CET 2008 |
|
|
727 | - reset diehook when terminating from an async_pool as to not |
|
|
728 | trigger any __DIE__ handlers. |
|
|
729 | |
|
|
730 | 4.35 Sun Jan 13 04:14:13 CET 2008 |
|
|
731 | - "bt" debug command now displays any exceptions |
|
|
732 | from longmess and also skips the innermost |
|
|
733 | stackframes, giving more useufl output. |
|
|
734 | - allow backtraces in coroutines blocked in Coro::EV, |
|
|
735 | at a <1% speed hit. |
|
|
736 | - handle localising of $SIG{__WARN__} and __DIE__ |
|
|
737 | properly (with a proper amount of dirty hacking). |
|
|
738 | |
|
|
739 | 4.34 Sat Dec 22 17:49:53 CET 2007 |
|
|
740 | - upgrade to EV version 2.0 API. |
|
|
741 | |
|
|
742 | 4.33 Mon Dec 17 08:36:12 CET 2007 |
|
|
743 | - make Coro::AIO etc. loadable in the absence of EV. |
|
|
744 | |
|
|
745 | 4.32 Mon Dec 17 07:46:02 CET 2007 |
|
|
746 | - majorly improved Coro::Handle's performance with Coro::EV. |
|
|
747 | - implemented the until now mythical Coro::BDB module. |
|
|
748 | - specialcase EV in Coro::AIO and Coro::BDB for extra speed. |
|
|
749 | |
|
|
750 | 4.31 Wed Dec 5 12:32:39 CET 2007 |
|
|
751 | - remove warn statement form Coro::Util. |
|
|
752 | |
|
|
753 | 4.3 Tue Dec 4 20:33:14 CET 2007 |
|
|
754 | - calls to the idle function could cause stack corruption |
|
|
755 | when the stack changed. |
|
|
756 | - do no longer rely on the presence of EV::DNS if EV is used |
|
|
757 | (because it is gone), but instead take avdantage of EV::ADNS |
|
|
758 | if available. |
|
|
759 | - add ($) prototypes to all functions in Coro::Storable. |
|
|
760 | - use a conventional (safer) idle callback in Coro::EV. |
|
|
761 | - do accept quickly in Coro::Debug to avoid endless loops. |
|
|
762 | |
|
|
763 | 4.22 Fri Nov 30 16:04:04 CET 2007 |
|
|
764 | - really use optimised versions for Event and EV in Coro::Util |
|
|
765 | and Coro::Handle. |
|
|
766 | |
|
|
767 | 4.21 Sun Nov 25 10:48:59 CET 2007 |
|
|
768 | - fix a spurious memory read. |
|
|
769 | - Coro::EV no longer keeps the eventloop "alive". |
|
|
770 | |
|
|
771 | 4.2 Fri Nov 9 20:47:05 CET 2007 |
|
|
772 | - enable/disable tracing from a new coroutine, not a pooled one. |
|
|
773 | - fix a memleak in Coro::Event. |
|
|
774 | - removed killall call from fork_eval. |
|
|
775 | - made sure store_fd is already loaded so that fork_eval |
|
|
776 | does not have to parse autoload in each subprocess. |
|
|
777 | - only use assembly method if -O switch is in $Config{optimize}. |
|
|
778 | - add (optional) Coro::EV module, so far the best event loop module |
|
|
779 | directly supported by Coro. |
|
|
780 | - if the event model is EV, use EV::DNS to resolve |
|
|
781 | stuff in Coro::Util. |
|
|
782 | - don't get confused by multiple event notifications in Coro::Handle. |
|
|
783 | - initial support for EV (libevent interface). |
|
|
784 | - require Event and EV using configure_requires, to force their existance. |
|
|
785 | |
|
|
786 | 4.13 Wed Oct 24 07:26:45 CEST 2007 |
|
|
787 | - add Coro::Storable::blocking_thaw. |
|
|
788 | - use a vastly more complicated technique to localise |
|
|
789 | $SIG{__WARN/DIE__} that also works on perls <= 5.8.8. |
|
|
790 | - use a coroutine for the idle callback Coro::Event, |
|
|
791 | instead of running Event in the current coroutine context. |
|
|
792 | This also catches recursive invocations. |
|
|
793 | - actually report fork errors in gethostbyname and inet_aton. |
|
|
794 | |
|
|
795 | 4.11 Thu Oct 11 02:40:24 CEST 2007 |
|
|
796 | - port to threaded perls. |
|
|
797 | |
|
|
798 | 4.1 Thu Oct 11 02:38:16 CEST 2007 |
|
|
799 | - incompatible change: $SIG{__DIE__} and $SIG{__WARN__} will now |
|
|
800 | be local to each coro (see Coro::State). |
|
|
801 | - incompatible change: for very deep reasons, cede and cede_notself |
|
|
802 | cannot return anything, so nothing will be returned. |
|
|
803 | - possibly bring back 5.10 compatibility (untested). |
|
|
804 | - work around stupid (and wrong) warning on 5.10 :(. |
|
|
805 | - overlay the saved state over the context stack. This saves |
|
|
806 | a few hundred bytes per coroutine on average and also |
|
|
807 | speeds up context switching a bit. |
|
|
808 | - further tune default stack sizes. |
|
|
809 | - (more) correctly calculate stack usage in coro_rss. |
|
|
810 | - Coro::Storable::blocking_* did not properly lock |
|
|
811 | resulting in races between coroutines. |
|
|
812 | - added Coro::Storable::guard. |
|
|
813 | - stopping to trace a coroutine could destroy the cctx of |
|
|
814 | an unrelated coroutine. |
|
|
815 | - explain the relationship between Perl and C coroutines in |
|
|
816 | more detail in Coro::State. |
|
|
817 | - Coro::Util::inet_aton did not short-circuit dotted quad forms, |
|
|
818 | causing a fork per resolve. This also affected Coro::Socket. |
|
|
819 | - switch to a separate stack in $coro->call/eval to avoid |
|
|
820 | invalidating pointers. |
|
|
821 | |
|
|
822 | 4.03 Sat Oct 6 21:24:00 CEST 2007 |
|
|
823 | - added Coro::throw method. |
|
|
824 | - minor code cleanups. |
|
|
825 | |
|
|
826 | 4.02 Sat Oct 6 02:36:47 CEST 2007 |
|
|
827 | - fix a very minor per-coroutine memleak (a single codereference). |
|
|
828 | - fixed a bug where the currently in-use c context would be freed |
|
|
829 | prematurely (can happen only when programs change the stacksize |
|
|
830 | or use tracing). |
|
|
831 | - tracing can no longer keep a coro alive after it terminated. |
|
|
832 | - do static branch prediction in the common path for gcc. gives |
|
|
833 | about 2-5% speed improvement here. |
|
|
834 | |
|
|
835 | 4.01 Fri Oct 5 22:10:49 CEST 2007 |
|
|
836 | - instead of recreating *a* standard output handle we simply |
|
|
837 | use STDOUT, which is faster and hopefully more robust. |
|
|
838 | |
|
|
839 | 4.0 Fri Oct 5 12:56:00 CEST 2007 |
|
|
840 | - incompatibly changed Coro::Storable::freeze. |
|
|
841 | - major new feature: added Coro::Debug, for interactive coroutine |
|
|
842 | debugging, tracing and much more. |
|
|
843 | - major bug fix: unbelievable, but true: $_, $/ and many other |
|
|
844 | "saved" variables actually weren't being saved. This has been fixed, |
|
|
845 | of course, while increasing performance while losing all the save |
|
|
846 | flags. |
|
|
847 | - save flags are gone, and all the api functions dealing with them. |
|
|
848 | - added Coro::Semaphore::adjust. |
|
|
849 | - added Coro::Util::fork_eval. |
|
|
850 | - added Coro::Storable::{nfreeze,blocking_{freeze,nfreeze}}. |
|
|
851 | - added Coro::killall. |
|
|
852 | - reduce initial stack sizes to allow for "micro-coroutines". |
|
|
853 | - better async_pool resource management, moved parts of async_pool |
|
|
854 | handling to XS (major speed improvement). |
|
|
855 | - actually croak before modifying important data structures. |
|
|
856 | - refuse to transfer while compiling. |
|
|
857 | - possibly support eval EXPR better now. |
|
|
858 | - enable assembly per default on linux+bsd x86+amd64. |
|
|
859 | - all internal members were renamed _something for easier subclassing. |
|
|
860 | - many minor tweaks. |
|
|
861 | |
|
|
862 | 3.63 Wed May 16 14:10:06 CEST 2007 |
|
|
863 | - implement handcoded assembly for x86/amd64 SVR ABI. |
|
|
864 | |
|
|
865 | 3.62 Fri Apr 27 21:36:06 CEST 2007 |
|
|
866 | - upgrade libcoro (which might set unwind info correctly). |
|
|
867 | - change default on linux to setjmp/longjmp. |
|
|
868 | |
|
|
869 | 3.61 Thu Apr 19 12:36:18 CEST 2007 |
|
|
870 | - Coro::Storable caused an endless loop when thawing invalid |
|
|
871 | pst's sometimes. |
|
|
872 | - use a Semaphore in Coro::Storable, as Storable doesn't |
|
|
873 | seem to be reentrant (although it is documented to |
|
|
874 | be threadsafe...). |
|
|
875 | - fix Coro::Signal to bring back the original unreliable |
|
|
876 | but stateful semantics. |
|
|
877 | - fixed a lot of typos in Coro.pm (patch submitted by David |
|
|
878 | Steinbrunner, which applied flawlessly). |
|
|
879 | |
|
|
880 | 3.6 Sat Apr 14 17:13:31 CEST 2007 |
|
|
881 | - added some bugfixes to get eg/myhttpd working again. |
|
|
882 | - added Coro::Storable for often-cede'ing freeze/thaw. |
|
|
883 | - try to do a clean exit when a coroutine calls exit |
|
|
884 | (EXPERIMENTAL). |
|
|
885 | - got rid of indirect call through _coro_init. |
|
|
886 | - updated the partly antique examples in eg/ to |
|
|
887 | work again and be a bit less magic, too. |
|
|
888 | - fixed Coro::Signal semantics to work as documented again. |
|
|
889 | |
|
|
890 | 3.55 Sun Mar 25 01:20:47 CET 2007 |
|
|
891 | - add SAVE_DEFFH to save the default filehandle and enable |
|
|
892 | it by default. |
|
|
893 | - finally move socket-operations from Coro::Socket to Coro::Handle |
|
|
894 | to be able to unblock foreign sockets. |
|
|
895 | - add Coro::State::save_also and guarded_save. |
|
|
896 | - add count accessor to Coro::Semaphore. |
|
|
897 | - add Coro::State::cctx_stacksize. |
|
|
898 | - just for the fun of it, do not rely on implicit context, |
|
|
899 | which can dramatically improve performance, but people |
|
|
900 | using windows-process-emulation perls do not care much about |
|
|
901 | performance. |
|
|
902 | |
|
|
903 | 3.51 Sun Mar 4 14:18:04 CET 2007 |
|
|
904 | - fixed a problem when you weakened references to Coro::State's |
|
|
905 | (patch by Yuval Kogman). |
|
|
906 | |
|
|
907 | 3.501 Wed Feb 28 12:44:07 CET 2007 |
|
|
908 | - rename some global symbols as macosx from hell redefines |
|
|
909 | them without asking. |
|
|
910 | |
|
|
911 | 3.5 Tue Feb 13 20:22:53 CET 2007 |
|
|
912 | - do AnyEvent model detection earlier, avoiding problems |
|
|
913 | caused by first using AnyEvent and later Coro::Event. |
|
|
914 | - implement and document Coro::Event event objects. |
|
|
915 | - fix a potential problem in Coro::Event causing crashes. |
|
|
916 | - initialise PL_comppad when creating a new coroutine, |
|
|
917 | avoids crashes on early coro destruction. |
|
|
918 | |
|
|
919 | 3.41 Mon Jan 22 19:19:49 CET 2007 |
|
|
920 | - readline on Coro::Handle did not support undefined $/, |
|
|
921 | nor did it deliver partial lines on EOF or error. |
|
|
922 | - implement malloc fallback for stack allocation because |
|
|
923 | stupid broken idiotic OSX has a stupid broken |
|
|
924 | idiotic fits-the-whole-os mmap "implementation" and |
|
|
925 | my dick feels longer if Coro is portable even to |
|
|
926 | obsolete platforms. |
|
|
927 | |
|
|
928 | 3.4 Fri Jan 19 21:52:54 CET 2007 |
|
|
929 | - remove t/09_timer.t, as it isn't really testing much |
|
|
930 | but was rather flaky in practise. |
|
|
931 | - async_pool coro would keep arguments and callback alive until |
|
|
932 | it was reused. |
|
|
933 | - cancellation of a coroutine could cause spurious idle calls |
|
|
934 | in cede_notself. |
|
|
935 | |
|
|
936 | 3.3 Sat Jan 6 03:45:00 CET 2007 |
| 7 | - implement $coro->on_destroy. |
937 | - implement $coro->on_destroy. |
| 8 | - Coro::Event blocking semantics have been changed, |
938 | - Coro::Event blocking semantics have been changed, |
| 9 | documented and - hopefully - improved. |
939 | documented and - hopefully - improved. |
| 10 | - fix nice adding, not subtracting, from priority. |
940 | - fix nice adding, not subtracting, from priority. |
| 11 | - fix ->prio and api_is_ready (patch by Mark Hinds). |
941 | - fix ->prio and api_is_ready (patch by Mark Hinds). |
| 12 | - fixed an assert ("... == prev__cctx->idle_te") |
942 | - fixed an assert ("... == prev__cctx->idle_te") |
| 13 | that could errronously trigger. |
943 | that could errronously trigger. |
| 14 | - fix various large and small memleaks. |
944 | - fix various large and small memleaks. |
| 15 | - use a (hopefully) more stable cancel implementation |
945 | - use a (hopefully) more stable cancel implementation |
| 16 | that immediately frees the coroutine data. |
946 | that immediately frees the coroutine data. |
|
|
947 | - cede/cede_notself return a status now. |
| 17 | - added Coro::guard function. |
948 | - added Coro::guard function. |
| 18 | - added a global coroutine pool for jobs (on my machine, |
949 | - added a global coroutine pool for jobs (on my machine, |
| 19 | I can create and execute 48k simple coros/s with async, |
950 | I can create and execute 48k simple coros/s with async, |
| 20 | and 128k coros with async_pool). |
951 | and 128k coros with async_pool). |
|
|
952 | - Coro::AIO now uses the coroutine priority as io priority. |
| 21 | |
953 | |
| 22 | 3.2 Fri Dec 22 05:07:09 CET 2006 |
954 | 3.2 Fri Dec 22 05:07:09 CET 2006 |
| 23 | - improve portability to slightly older perls. |
955 | - improve portability to slightly older perls. |
| 24 | - use cleaner coroutine destruction. |
956 | - use cleaner coroutine destruction. |
| 25 | - simplify configuration for users. |
957 | - simplify configuration for users. |
