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

Comparing libev/Changes (file contents):
Revision 1.265 by root, Thu May 3 16:00:47 2012 UTC vs.
Revision 1.326 by root, Thu Jun 20 22:44:59 2019 UTC

1Revision history for libev, a high-performance and full-featured event loop. 1Revision history for libev, a high-performance and full-featured event loop.
2 2
3TODO: ev_loop_wakeup 3 - new experimental linux aio backend (linux 4.18+).
4TODO: EV_STANDALONE == NO_HASSEL (do not use clock_gettime in ev_standalone) 4 - removed redundant 0-ptr check in ev_once.
5 - updated/extended ev_set_allocator documentation.
6 - replaced EMPTY2 macro by array_needsize_noinit.
7 - minor code cleanups.
5 8
6TODO: document WSA_EV_USE_SOCKET in win32 part 94.25 Fri Dec 21 07:49:20 CET 2018
7TODO: ^ OR use WSASend/WSARecv on the handle, which always works 10 - INCOMPATIBLE CHANGE: EV_THROW was renamed to EV_NOEXCEPT
8TODO: assert on fd watcher linked list pointing to itself 11 (EV_THROW still provided) and now uses noexcept on C++11 or newer.
9TODO: faq, process a thing in each iteration 12 - move the darwin select workaround highe rin ev.c, as newer versions of
10TODO: abuse idle watchers for their side effect 13 darwin managed to break their broken select even more.
11TODO: abuse check watchers for their side effect 14 - ANDROID => __ANDROID__ (reported by enh@google.com).
15 - disable epoll_create1 on android because it has broken header files
16 and google is unwilling to fix them (reported by enh@google.com).
17 - avoid a minor compilation warning on win32.
18 - c++: remove deprecated dynamic throw() specifications.
19 - c++: improve the (unsupported) bad_loop exception class.
20 - backport perl ev_periodic example to C, untested.
21 - update libecb, biggets change is to include a memory fence
22 in ECB_MEMORY_FENCE_RELEASE on x86/amd64.
23 - minor autoconf/automake modernisation.
24
254.24 Wed Dec 28 05:19:55 CET 2016
26 - bump version to 4.24, as the release tarball inexplicably
27 didn't have the right version in ev.h, even though the cvs-tagged
28 version did have the right one (reported by Ales Teska).
29
304.23 Wed Nov 16 18:23:41 CET 2016
31 - move some declarations at the beginning to help certain retarded
32 microsoft compilers, even though their documentation claims
33 otherwise (reported by Ruslan Osmanov).
34
354.22 Sun Dec 20 22:11:50 CET 2015
36 - when epoll detects unremovable fds in the fd set, rebuild
37 only the epoll descriptor, not the signal pipe, to avoid
38 SIGPIPE in ev_async_send. This doesn't solve it on fork,
39 so document what needs to be done in ev_loop_fork
40 (analyzed by Benjamin Mahler).
41 - remove superfluous sys/timeb.h include on win32
42 (analyzed by Jason Madden).
43 - updated libecb.
44
454.20 Sat Jun 20 13:01:43 CEST 2015
46 - prefer noexcept over throw () with C++ 11.
47 - update ecb.h due to incompatibilities with c11.
48 - fix a potential aliasing issue when reading and writing
49 watcher callbacks.
50
514.19 Thu Sep 25 08:18:25 CEST 2014
52 - ev.h wasn't valid C++ anymore, which tripped compilers other than
53 clang, msvc or gcc (analyzed by Raphael 'kena' Poss). Unfortunately,
54 C++ doesn't support typedefs for function pointers fully, so the affected
55 declarations have to spell out the types each time.
56 - when not using autoconf, tighten the check for clock_gettime and related
57 functionality.
58
594.18 Fri Sep 5 17:55:26 CEST 2014
60 - events on files were not always generated properly with the
61 epoll backend (testcase by Assaf Inbal).
62 - mark event pipe fd as cloexec after a fork (analyzed by Sami Farin).
63 - (ecb) support m68k, m88k and sh (patch by Miod Vallat).
64 - use a reasonable fallback for EV_NSIG instead of erroring out
65 when we can't detect the signal set size.
66 - in the absence of autoconf, do not use the clock syscall
67 on glibc >= 2.17 (avoids the syscall AND -lrt on systems
68 doing clock_gettime in userspace).
69 - ensure extern "C" function pointers are used for externally-visible
70 loop callbacks (not watcher callbacks yet).
71 - (ecb) work around memory barriers and volatile apparently both being
72 broken in visual studio 2008 and later (analysed and patch by Nicolas Noble).
73
744.15 Fri Mar 1 12:04:50 CET 2013
75 - destroying a non-default loop would stop the global waitpid
76 watcher (Denis Bilenko).
77 - queueing pending watchers of higher priority from a watcher now invokes
78 them in a timely fashion (reported by Denis Bilenko).
12 - add throw() to all libev functions that cannot throw exceptions, for 79 - add throw() to all libev functions that cannot throw exceptions, for
13 further code size decrease when compiling for C++. 80 further code size decrease when compiling for C++.
14 - add throw () to callbacks that must not throw exceptions (allocator, 81 - add throw () to callbacks that must not throw exceptions (allocator,
15 syserr, loop acquire/release, periodic reschedule cbs). 82 syserr, loop acquire/release, periodic reschedule cbs).
16 - fix event_base_loop return code, add event_get_callback, event_base_new, 83 - fix event_base_loop return code, add event_get_callback, event_base_new,
17 event_base_get_method calls to improve libevent 1.x emulation and add 84 event_base_get_method calls to improve libevent 1.x emulation and add
18 some libevent 2.x funcitonality (based on a patch by Jeff Davey). 85 some libevent 2.x functionality (based on a patch by Jeff Davey).
19 - add more memory fences to fix a bug reported by Jeff Davey. Better 86 - add more memory fences to fix a bug reported by Jeff Davey. Better
20 be overfenced than underprotected. 87 be overfenced than underprotected.
21 - ev_run now returns a boolean status (true meaning watchers are 88 - ev_run now returns a boolean status (true meaning watchers are
22 still active). 89 still active).
23 - ev_once: undef EV_ERROR in ev_kqueue.c, to avoid clashing with 90 - ev_once: undef EV_ERROR in ev_kqueue.c, to avoid clashing with
… …
25 - (ecb) add memory fence support for xlC (Darin McBride). 92 - (ecb) add memory fence support for xlC (Darin McBride).
26 - (ecb) add memory fence support for gcc-mips (Anton Kirilov). 93 - (ecb) add memory fence support for gcc-mips (Anton Kirilov).
27 - (ecb) add memory fence support for gcc-alpha (Christian Weisgerber). 94 - (ecb) add memory fence support for gcc-alpha (Christian Weisgerber).
28 - work around some kernels losing file descriptors by leaking 95 - work around some kernels losing file descriptors by leaking
29 the kqueue descriptor in the child. 96 the kqueue descriptor in the child.
97 - work around linux inotify not reporting IN_ATTRIB changes for directories
98 in many cases.
30 - include sys/syscall.h instead of plain syscall.h. 99 - include sys/syscall.h instead of plain syscall.h.
100 - check for io watcher loops in ev_verify, check for the most
101 common reported usage bug in ev_io_start.
102 - choose socket vs. WSASocket at compiletime using EV_USE_WSASOCKET.
103 - always use WSASend/WSARecv directly on windows, hoping that this
104 works in all cases (unlike read/write/send/recv...).
105 - try to detect signals around a fork faster (test program by
106 Denis Bilenko).
107 - work around recent glibc versions that leak memory in realloc.
108 - rename ev::embed::set to ev::embed::set_embed to avoid clashing
109 the watcher base set (loop) method.
110 - rewrite the async/signal pipe logic to always keep a valid fd, which
111 simplifies (and hopefully correctifies :) the race checking
112 on fork, at the cost of one extra fd.
113 - add fat, msdos, jffs2, ramfs, ntfs and btrfs to the list of
114 inotify-supporting filesystems.
115 - move orig_CFLAGS assignment to after AC_INIT, as newer autoconf
116 versions ignore it before
117 (https://bugzilla.redhat.com/show_bug.cgi?id=908096).
118 - add some untested android support.
119 - enum expressions must be of type int (reported by Juan Pablo L).
31 120
324.11 Sat Feb 4 19:52:39 CET 2012 1214.11 Sat Feb 4 19:52:39 CET 2012
33 - INCOMPATIBLE CHANGE: ev_timer_again now clears the pending status, as 122 - INCOMPATIBLE CHANGE: ev_timer_again now clears the pending status, as
34 was documented already, but not implemented in the repeating case. 123 was documented already, but not implemented in the repeating case.
35 - new compiletime symbols: EV_NO_SMP and EV_NO_THREADS. 124 - new compiletime symbols: EV_NO_SMP and EV_NO_THREADS.
… …
219 reinitialising the default loop actually works (patch by 308 reinitialising the default loop actually works (patch by
220 Malek Hadj-Ali). 309 Malek Hadj-Ali).
221 - implement ev_suspend and ev_resume. 310 - implement ev_suspend and ev_resume.
222 - new EV_CUSTOM revents flag for use by applications. 311 - new EV_CUSTOM revents flag for use by applications.
223 - add documentation section about priorities. 312 - add documentation section about priorities.
224 - add a glossary to the dcoumentation. 313 - add a glossary to the documentation.
225 - extend the ev_fork description slightly. 314 - extend the ev_fork description slightly.
226 - optimize a jump out of call_pending. 315 - optimize a jump out of call_pending.
227 316
2283.53 Sun Feb 15 02:38:20 CET 2009 3173.53 Sun Feb 15 02:38:20 CET 2009
229 - fix a bug in event pipe creation on win32 that would cause a 318 - fix a bug in event pipe creation on win32 that would cause a

Diff Legend

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