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

Comparing Coro/Changes (file contents):
Revision 1.547 by root, Sun Feb 13 04:39:14 2011 UTC vs.
Revision 1.627 by root, Tue Mar 4 12:32:02 2014 UTC

1Revision history for Perl extension Coro. 1Revision history for Perl extension Coro.
2 2
3TODO: should explore PerlIO::coroaio (perl leaks like hell). 3TODO: should explore PerlIO::coroaio (perl leaks like hell).
4TODO: unready_all
5TODO: myhttpd header parsing
6TODO: channel->maxsize(newsize)? 4TODO: channel->maxsize(newsize)?
7TODO: http://www.microsoft.com/msj/archive/s2ce.aspx 5TODO: __GCC_HAVE_DWARF2_CFI_ASM
6TODO: swap_sv, maybe add scope_swap_sv, andallow to unswap by repeat?
7TODO: swap_sv not undone in pool
8
96.37 Tue Mar 4 13:27:33 CET 2014
10 - *sigh*, removed leftover dbeugging code from debugging a
11 perl bug, of all things.
12
136.36 Tue Mar 4 07:11:59 CET 2014
14 - semaphores would not clear the destroy hook when interrupted
15 by ->throw, causing segfaults or worse.
16 - ->throw on a thread waiting for a semaphore did not acquire
17 the semaphore, but also didn't wake up other waiters,
18 possibly causing a deadlock.
19 - "FATAL: $Coro::IDLE blocked itself" will now use Carp::confess
20 to report a full stacktrace, which should help find out
21 where the actual call is.
22 - "atomically" destroy data in slf_destroy, because it is
23 the right thing to do, just in case.
24 - disable ecb.h libm dependency, to compile on ancient systems
25 or under adverse conditions.
26
276.33 Mon Nov 18 11:26:27 CET 2013
28 - do not crash when freeing padlists with holes (in 5.18).
29 - tentative SVt_BIND 5.19 port/fix.
30
316.32 Tue Nov 5 15:35:35 CET 2013
32 - use a new algorithm to derive padlists for perl 5.18. The old
33 one could lead to 0-pointer accesses inside perl (reported
34 by Darin McBride).
35
366.31 Thu May 9 07:39:48 CEST 2013
37 - Coro::AIO requests would crash if the thread was ready'd
38 while the request was ongoing.
39
406.29 Wed May 8 02:55:18 CEST 2013
41 - when an on_destroy handler destructs the coro currently being
42 destructed a perl scalar could be accessed after being freed,
43 likely causing a crash.
44
456.28 Wed Mar 6 06:58:02 CET 2013
46 - clean remnants of existing __DIE__ and __WARN__ handlers so
47 they lose their magic and will not cause segfaults later
48 (testcase by Andrey Sagulin).
49 - improved Coro::State documentation a bit.
50 - Coro::Debug::command now flushes the output.
51 - add hack detection code for x32 abi, because the braindead slugs
52 who designed that made it look exactly like x86_64 without
53 providing proper compile time symbols to test for it. as a result,
54 this detection cannot work reliably.
55 - valgrind stack registering was broken.
56 - do not rely on Time::HiRes anymore in Coro::Debug.
57
586.23 Fri Dec 7 23:36:37 CET 2012
59 - use experimental fiber implementation on native windows
60 perls.
61 - use sizeof (void *) as multiplication factor for stack sizes,
62 to accomodate the totally braindamaged microsoft 64 bit "os".
63 - changed verifier host from win2k-ap510-32 to win7-sp516-32/64.
64 activeperl 5.16 crashes when PerlIO_define_layer is called due
65 to some bug in the perl dll, strawberry perl at least passes
66 the testsuite.
67 - implement Coro::Handle->peeraddr/host/port, for slightly
68 improved compatibility with LWP.
69 - implement 5.17 compatibility by almost blindly applying a
70 good-looking patch by Father Chrysostomos.
71 - move stack management functions into libcoro 3.
72 - libcoro version 3 "released".
73 - support magic values as timed_io_once args.
74 - recommend AnyEvent 7+ or EV 4+, also require EV
75 version 4 or newer for Coro::EV.
76
776.10 Tue Oct 9 01:14:27 CEST 2012
78 - updated ecb.h, it had a typo that caused it to not compile on many
79 big endian systems (reported by many people).
80 - disable memory fences in ecb.h to improve portability.
81
826.09 Sat Oct 6 23:25:02 CEST 2012
83 - Coro::EV I/O watchers were not interruptible by exceptions
84 (Coro::State::throw) (testcase by sten).
85 - ->throw now puts threads into the ready queue, as this seems to
86 be expected by existing code, and code that doesn't cope with spurious
87 wakeups needs fixing anyway.
88 - use fd -1 in mmap.
89 - cast I32 to int in error message printf.
90 - warn about broken so-called "hardened" kernels.
91
926.08 Fri Apr 13 12:05:47 CEST 2012
93 - be more aggressive about exiting like perl does - formerly,
94 exiting from the non-main thread would not execute END blocks.
95
966.07 Fri Nov 11 21:21:48 CET 2011
97 - work around a bug in PerlIO (setting $SIG{__WARN__} to a PVCV).
98 - update ecb.h.
99
1006.06 Mon Aug 8 23:59:48 CEST 2011
101 - cygwin unfortunately patches the stack at runtime, so we use the pthreads
102 backend, which is an order of magnitude slower. unfortunately, cygwins
103 pthread implementation isn't very complete either, so allocate the stack
104 twice just to be sure.
105 (note: cygwin also enables mymalloc, which is NOT THREADSAFE ON WINDOWS,
106 in its ithreaded perl - best recompile cygwin and use the 'w'indows
107 backend for much better performance. also disable ithreads for
108 even better performance...).
109
1106.05 Thu Aug 4 21:36:36 CEST 2011
111 - blush, condvar values would not be propagated from send to recv anymore
112 (reported by Chip Salzenberg).
113 - use exponential increase for the readline buffer length in
114 Coro::Handle. also reduce initial allocation to 1020 from 4096 bytes.
115
1166.04 Wed Aug 3 17:15:45 CEST 2011
117 - use even more efficient and more compatible condvars for
118 compatibility to AnyEvent 6.x :)
119 - more inconsequential ecb.h updates.
120
1216.03 Wed Aug 3 11:41:30 CEST 2011
122 - change how Coro patches AnyEvent condvars for compatibility to
123 AnyEvent 6.x.
124 - update ecb.h, to no longer include <pthread.h> in case WinNT.h
125 hasn't been included.
126
1276.02 Wed Jul 13 04:35:19 CEST 2011
128 - "improve portability to Gentoo" - gentoo manages to put perl variables
129 in memory areas that are farther than 2gb apart, which the jit couldn't
130 handle and barfed. now it's just a bit slower on gentoo and similar
131 systems.
132
1336.01 Sun Jul 3 12:31:14 CEST 2011
134 - workarounds are good, but the test for whether pthreads are used
135 was not good. thisone should be better.
136 - check differently whether gcc generates cfi instructions itself.
137
1386.0 Wed Jun 29 19:43:35 CEST 2011
139 - INCOMPATIBLE CHANGE: unreferenced coro objects will now be
140 destroyed and cleaned up automatically (e.g. async { schedule }).
141 - implement a JIT compiler for part of the thread switch code,
142 which gives a 50% speed improvement on threaded perls, and
143 about 4% on non-thraeded perls (so threaded perls now finally
144 reach about half the speed of non-threaded perls).
145 - slightly modernise Coro::Intro, add section about rouse functions.
146 - avoid DEFSV and ERRSV, giving another 10% improvement
147 in thread switching.
148 - Coro::State->is_destroyed is now called is_zombie.
149 - implement a Coro->safe_cancel method that might fail, but
150 cancels in a "safer" way if it succeeds.
151 - add preliminary support for DEBUGGING perls.
152 - get rid of two hash-accesses when initialising a new Coro - this
153 speeds up coro creation by almost a factor of two.
154 - croak when a coro that is being cancelled tries to block
155 (e.g. while executing a guard block), instead of crashing or
156 deadlocking.
157 - use a more robust and also faster method to identify Coro::State
158 objects - speeds up everything a bit.
159 - implement Coro->cancel in XS for a 20% speed improvement, and to
160 be able to implement mutual cancellation.
161 - speed up context switches by a percent or two by more efficiently
162 allocating context stack entries.
163 - implement Coro->join and Coro->on_destroy in XS for a speedup and
164 a reduction in memory use.
165 - cancelling a coro while it itself is cancelling another coro is
166 now supported and working, instead of triggering an assertion.
167 - be a bit more crash-resistant when calling (buggy) on_destroy
168 callbacks (best effort).
169 - move on_destroy into the slf_frame, to allow extension slf
170 functions to have destructors.
171 - get rid if coro refcounting - simply crash in other interpreter
172 threads by nulling the pointers on clone.
173 - simplify warn/die hook handling when loading Coro - the convoluted
174 logic seems to be no longer neccessary.
175 - use libecb instead of our own home-grown gcc hacks.
176 - document alternatives to Coro::LWP. Please use them :)
177 - work around another mindless idiotic NEEDLESS bug in openbsd/mirbsds
178 sigaltstack. Really. wine suffers from it, erlang suffers from it,
179 and it's known since at least 2006.
180
1815.372 Wed Feb 23 06:14:30 CET 2011
182 - apparently mingw doesn't provide a working gettimeofday, try to
183 work around that by relying on Time::HiRes (indirectly brought to
184 my attention by Max Maischein).
185 - fix some portability issues when Time::HiRes was used.
186
1875.371 Mon Feb 21 14:36:08 CET 2011
188 - backport to windows process emulation code again.
189
1905.37 Sat Feb 19 07:49:44 CET 2011
191 - add a big "Coro thread life cycle" section to "man Coro".
192 - try a tentative workaround against the breakage that 5.13 has
193 introduced without depreciation period. sigh.
194 - no longer use Time::HiRes if gettimeofday is available, which
195 saves quite a lot of memory.
8 196
95.36 Sun Feb 13 05:33:41 CET 2011 1975.36 Sun Feb 13 05:33:41 CET 2011
10 - automatically load Coro::Channel, Coro::RWLock, Coro::Semaphore, 198 - automatically load Coro::Channel, Coro::RWLock, Coro::Semaphore,
11 Coro::SemaphoreSet, Coro::Signal and Coro::Specific on first "new" 199 Coro::SemaphoreSet, Coro::Signal and Coro::Specific on first "new"
12 method call. 200 method call.
… …
314 - ->throw is now supported on Coro::State objects. 502 - ->throw is now supported on Coro::State objects.
315 - clean up cctx creation code a bit. 503 - clean up cctx creation code a bit.
316 - entersub is actually a UNOP, not a LOGOP (not a bugfix). 504 - entersub is actually a UNOP, not a LOGOP (not a bugfix).
317 505
3184.9 Sat Nov 8 17:45:27 CET 2008 5064.9 Sat Nov 8 17:45:27 CET 2008
319 - (libcoro) did not preserve rbp with CORO_ASM (we are getting there). 507 - (libcoro) did not preserve rbp with CORO_ASM (we are getting there).
320 - (libcoro) no longer leak threads in the experimental pthread backend, 508 - (libcoro) no longer leak threads in the experimental pthread backend,
321 also speed it up considerably. 509 also speed it up considerably.
322 - (libcoro) do not rely on makecontext passing void *'s unscathed. 510 - (libcoro) do not rely on makecontext passing void *'s unscathed.
323 - fix compiletime dependencies on libcoro in the Makefile. 511 - fix compiletime dependencies on libcoro in the Makefile.
324 - cctx_count wasn't always updated properly. 512 - cctx_count wasn't always updated properly.
… …
386 - remove debugging code related to MgPV_nolen_const, also try to 574 - remove debugging code related to MgPV_nolen_const, also try to
387 make it compile with perl 5.8.6 (yes, apple apparently loves 575 make it compile with perl 5.8.6 (yes, apple apparently loves
388 outdated software). Reported by John S. 576 outdated software). Reported by John S.
389 577
3904.744 Tue Jul 8 22:06:35 CEST 2008 5784.744 Tue Jul 8 22:06:35 CEST 2008
391 - correctly provide default DIE/WARN handlers as documented. 579 - correctly provide default DIE/WARN handlers as documented.
392 - also overwrite PL_vtbl_sigelem.svt_clear, even though current 580 - also overwrite PL_vtbl_sigelem.svt_clear, even though current
393 implementations inside perl work fine for us. 581 implementations inside perl work fine for us.
394 582
3954.743 Mon Jun 16 00:21:57 CEST 2008 5834.743 Mon Jun 16 00:21:57 CEST 2008
396 - when using Coro::EV without running EV::loop it could 584 - when using Coro::EV without running EV::loop it could
… …
418 - sprinkle "no warnings" freely over everything, also suppress 606 - sprinkle "no warnings" freely over everything, also suppress
419 warnings for some other modules. 607 warnings for some other modules.
420 - fix typo in WSAEWOULDBLOCK. 608 - fix typo in WSAEWOULDBLOCK.
421 609
4224.72 Sun May 25 05:14:36 CEST 2008 6104.72 Sun May 25 05:14:36 CEST 2008
423 - tweak META.yaml a bit, unfortunately, there is no documented way 611 - tweak META.yaml a bit, unfortunately, there is no documented way
424 to have optional dependencies with CPAN. doh :( 612 to have optional dependencies with CPAN. doh :(
425 - avoid running some tests on windows because they would fail due to 613 - avoid running some tests on windows because they would fail due to
426 perl bug (broken fork, broken pipes...). 614 perl bug (broken fork, broken pipes...).
427 - work around perl on windows bugs where perl returns undocumented 615 - work around perl on windows bugs where perl returns undocumented
428 error codes for sysread, syswrite etc. by taking advantage 616 error codes for sysread, syswrite etc. by taking advantage
… …
432 - use unix domain sockets in testsuite to work around 620 - use unix domain sockets in testsuite to work around
433 common perl implementation bugs on widows (they are emulated by 621 common perl implementation bugs on widows (they are emulated by
434 tcp sockets on windows. ugh.) 622 tcp sockets on windows. ugh.)
435 623
4364.71 Sat May 24 20:01:27 CEST 2008 6244.71 Sat May 24 20:01:27 CEST 2008
437 - fix a bug in Coro::AnyEvent ("Usage: Coro::AnyEvent::_schedule()"). 625 - fix a bug in Coro::AnyEvent ("Usage: Coro::AnyEvent::_schedule()").
438 - take advantage of async name resolution of AnyEvent::Util. 626 - take advantage of async name resolution of AnyEvent::Util.
439 - work around brutal inet_aton override in Coro::LWP. 627 - work around brutal inet_aton override in Coro::LWP.
440 - take advantage of the readyhook in Coro::EV, for smoother 628 - take advantage of the readyhook in Coro::EV, for smoother
441 scheduling. 629 scheduling.
442 630
4434.7 Sun May 11 00:32:19 CEST 2008 6314.7 Sun May 11 00:32:19 CEST 2008
444 - completely reworked the Coro manpage. 632 - completely reworked the Coro manpage.
445 - added Coro::AnyEvent, generic event loop integration. 633 - added Coro::AnyEvent, generic event loop integration.
446 - implement cancel, ready and kill commands in Coro::Debug. 634 - implement cancel, ready and kill commands in Coro::Debug.
447 - document find_coro in Coro::Debug. 635 - document find_coro in Coro::Debug.
448 - incompatible change: rename has_stack to has_cctx. 636 - incompatible change: rename has_stack to has_cctx.
449 - Coro::AIO and Coro::BDB no longer force event model detection, 637 - Coro::AIO and Coro::BDB no longer force event model detection,
450 use AnyEvent::AIO and AnyEvent::BDB. 638 use AnyEvent::AIO and AnyEvent::BDB.
… …
574 762
5754.11 Thu Oct 11 02:40:24 CEST 2007 7634.11 Thu Oct 11 02:40:24 CEST 2007
576 - port to threaded perls. 764 - port to threaded perls.
577 765
5784.1 Thu Oct 11 02:38:16 CEST 2007 7664.1 Thu Oct 11 02:38:16 CEST 2007
579 - incompatible change: $SIG{__DIE__} and $SIG{__WARN__} will now 767 - incompatible change: $SIG{__DIE__} and $SIG{__WARN__} will now
580 be local to each coro (see Coro::State). 768 be local to each coro (see Coro::State).
581 - incompatible change: for very deep reasons, cede and cede_notself 769 - incompatible change: for very deep reasons, cede and cede_notself
582 cannot return anything, so nothing will be returned. 770 cannot return anything, so nothing will be returned.
583 - possibly bring back 5.10 compatibility (untested). 771 - possibly bring back 5.10 compatibility (untested).
584 - work around stupid (and wrong) warning on 5.10 :(. 772 - work around stupid (and wrong) warning on 5.10 :(.
… …
656 but stateful semantics. 844 but stateful semantics.
657 - fixed a lot of typos in Coro.pm (patch submitted by David 845 - fixed a lot of typos in Coro.pm (patch submitted by David
658 Steinbrunner, which applied flawlessly). 846 Steinbrunner, which applied flawlessly).
659 847
6603.6 Sat Apr 14 17:13:31 CEST 2007 8483.6 Sat Apr 14 17:13:31 CEST 2007
661 - added some bugfixes to get eg/myhttpd working again. 849 - added some bugfixes to get eg/myhttpd working again.
662 - added Coro::Storable for often-cede'ing freeze/thaw. 850 - added Coro::Storable for often-cede'ing freeze/thaw.
663 - try to do a clean exit when a coroutine calls exit 851 - try to do a clean exit when a coroutine calls exit
664 (EXPERIMENTAL). 852 (EXPERIMENTAL).
665 - got rid of indirect call through _coro_init. 853 - got rid of indirect call through _coro_init.
666 - updated the partly antique examples in eg/ to 854 - updated the partly antique examples in eg/ to

Diff Legend

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