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

Comparing Coro/Changes (file contents):
Revision 1.406 by root, Thu Nov 13 06:15:37 2008 UTC vs.
Revision 1.459 by root, Sat Nov 22 15:48:43 2008 UTC

… …
3TODO: should explore PerlIO::coroaio (perl leaks like hell). 3TODO: should explore PerlIO::coroaio (perl leaks like hell).
4TODO: maybe implement a default message channel, very much like Erlang's 4TODO: maybe implement a default message channel, very much like Erlang's
5 actor model (which is cool in a lot of important aspects (failures!), 5 actor model (which is cool in a lot of important aspects (failures!),
6 but very lacking in others (higher level ipc)). 6 but very lacking in others (higher level ipc)).
7 7
8 TODO: Coro/Intro
9 - convert Coro::Timer, Coro::Select and Coro::Util to rouse API.
10 - Coro::Select did errornously dup the file descriptors
11 and didn't work with all AnyEvent backends.
12 - disassociate c contexts from coro objects - this is agruably more
13 correct, but mostly allows sharing of cctxs between coro and state
14 objects, for added memory savings and speed increases.
15 - bumped $Coro::POOL_RSS up to 32kb by default.
16 - no longer set the optype to OP_CUSTOM, as B::* understandably
17 doesn't like this very much (and we *are* a type of entersub).
18 - implement state cloning, just to prove that call/cc can be done.
19 - better 5.8.6 compatibility.
20
215.0 Thu Nov 20 10:35:05 CET 2008
22 - NEW ARCHITECTURE: use the latest 4.x version if you experience
23 stability issues.
24 - bump API version to 7 - all dependents must be recompiled.
25 - removed timed_* functions - they were not being used anyways
26 and should be replaced by a more generic mechanism -
27 and were annoying to support anyways :)
28 - removed SemaphoreSet's waiter method - use sem method instead.
29 - Coro::Semaphore->adjust didn't correctly wake up enough waiters.
30 - async_pool did free a scalar value twice
31 ("Attempt to unreference...").
32 - fix a longstanding bug where calling terminate on a coro that
33 was waiting for a semaphore that was just becoming available
34 would cause a deadlock (semaphore would get into a state where
35 it was available but waiters were still blocked).
36 - calling throw on a coroutine that is waiting for a semaphore will
37 no longer make it acquire the semaphore (and thus leak a count).
38 - perl's process emulation is now not even theoretically supported
39 anymore.
40 - new functions Coro::rouse_cb and Coro::rouse_wait for easier
41 conversion of callback-style to blocking-style.
42 - new methods $coro->schedule_to and ->cede_to, to specifically
43 schedule or cede to a specific coroutine.
44 - new function Coro::Semaphore::wait.
45 - use named constants in Coro::Channel (Richard Hundt).
46 - directly patch the entersub opcode calling SLF functions (cede,
47 transfer and so on). this does speed up context switching, but
48 more importanly, it frees us from the hardcoded behaviour of
49 entersub, so we might actually be able to return something from
50 those functions and atcually create new ones.
51 - take advantage of __builtin_frame_address on gcc.
52 - expose THX in coroapi (not sure whether this was a wise decision,
53 as "threaded" perls are running at half speed anyways).
54 - implement execute_slf (schedule-like-function) interface that makes
55 it possible to implement schedule-like-functions in XS.
56 - use new SLF interface to massively speed up Coro::EV by roughly a
57 factor of two.
58 - used new SLF interface to massively speed up Coro::Semaphore by a
59 factor of three.
60 - used new SLF interface to speed up Coro::AIO by roughly a factor of
61 four and reduce its memory usage considerably.
62 - implement Coro::SemaphoreSet purely in terms of Coro::Semaphore,
63 for a nice speedup and vastly more correct behaviour. Also implement
64 a new method "sem" to get at the underlying semaphore object.
65 - implement Coro::Channel in terms of Coro::Semaphore, for a moderate
66 (in comparison) 20-40% speedup.
67 - used new SLF interface to reimplement Coro::Signal gaining
68 some unknown (because I was too lazy), but certain, speedup, and also
69 making signals reliable for the first time.
70 - used new SLF interface and other optimisations to speed up async_pool
71 by a factor of two. It also doesn't rely on perl's exception mechanism
72 to exit anymore. The overhead for terminating an async_pool, coro over
73 a normal async is now very small.
74 - sped up coroutine creation/destruction by 40%.
75 - forgot to include Coro/libcoro/README in the dist for all these years.
76 - work around a freebsd pthreads bug (manual testcancel is required as
77 pthread_cond_wait isn't a cancellation point on freebsd).
78 - use new rouse functions to speed up and simplify Coro::BDB.
79 - make "prefer perl native functions" work with threaded perls.
80 - condense Coro::Debug ps output, hint at v and w flags.
81 - (libcoro) lots of minor cleanups and portability improvements.
82
834.914 Wed Nov 19 12:54:18 CET 2008
84 - fix a disastrous bug in the readline optimisation
85 introduced in 4.801.
86
874.913 Sat Nov 15 07:58:28 CET 2008
88 - async_pool did free a scalar value twice
89 ("Attempt to unreference...").
90
914.912 Thu Nov 13 18:31:23 CET 2008
8 - minor cleanups. 92 - minor cleanups.
9 - use much larger stacks on linux and perl < 5.8.8. 93 - use much larger stacks on linux and perl < 5.8.8.
10 - Coro::Debug::new_unix_server did not unlink the socket 94 - Coro::Debug::new_unix_server did not unlink the socket
11 when destroyed. 95 when destroyed.
12 96

Diff Legend

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