… | |
… | |
7 | |
7 | |
8 | 5.0 |
8 | 5.0 |
9 | - NEW ARCHITECTURE: use 4.912 for the stable version. |
9 | - NEW ARCHITECTURE: use 4.912 for the stable version. |
10 | - bump API version to 7 - all dependents must be recompiled. |
10 | - bump API version to 7 - all dependents must be recompiled. |
11 | - Coro::Semaphore->adjust didn't correctly wake up enough waiters. |
11 | - Coro::Semaphore->adjust didn't correctly wake up enough waiters. |
12 | - async_poll did free a scalar value twice. |
12 | - async_pool did free a scalar value twice |
|
|
13 | ("Attempt to unreference..."). |
13 | - perl's process emulation is now not even theoretically supported |
14 | - perl's process emulation is now not even theoretically supported |
14 | anymore. |
15 | anymore. |
15 | - directly patch the entersub opcode calling SLF functions (cede, |
16 | - directly patch the entersub opcode calling SLF functions (cede, |
16 | transfer and so on). this does speed up context switching, but |
17 | transfer and so on). this does speed up context switching, but |
17 | more importanly, it frees us from the hardcoded behaviour of |
18 | more importanly, it frees us from the hardcoded behaviour of |
… | |
… | |
20 | - take advantage of __builtin_frame_address on gcc. |
21 | - take advantage of __builtin_frame_address on gcc. |
21 | - expose THX in coroapi (not sure whether this was a wise decision, |
22 | - expose THX in coroapi (not sure whether this was a wise decision, |
22 | as "threaded" perls are running at half speed anyways). |
23 | as "threaded" perls are running at half speed anyways). |
23 | - implement execute_slf (schedule-like-function) interface that makes |
24 | - implement execute_slf (schedule-like-function) interface that makes |
24 | it possible to implement schedule-like-functions in XS. |
25 | it possible to implement schedule-like-functions in XS. |
25 | - use new SLF interface to massively speed up Coro::EV (it's roughly |
26 | - use new SLF interface to massively speed up Coro::EV by roughly a factor |
26 | twice as fast now). |
27 | of two. |
27 | - removed timed_down and timed_guard functions - they were not being used |
28 | - removed timed_down and timed_guard functions - they were not being used |
28 | anyways and should be replaced by a more generic mechanism. |
29 | anyways and should be replaced by a more generic mechanism - and were |
|
|
30 | annoying to support anyways : |
|
|
31 | - used new SLF interface to massively speed up Coro::Semaphore by a factor |
|
|
32 | of three. |
|
|
33 | |
|
|
34 | 4.913 Sat Nov 15 07:58:28 CET 2008 |
|
|
35 | - async_pool did free a scalar value twice |
|
|
36 | ("Attempt to unreference..."). |
29 | |
37 | |
30 | 4.912 Thu Nov 13 18:31:23 CET 2008 |
38 | 4.912 Thu Nov 13 18:31:23 CET 2008 |
31 | - minor cleanups. |
39 | - minor cleanups. |
32 | - use much larger stacks on linux and perl < 5.8.8. |
40 | - use much larger stacks on linux and perl < 5.8.8. |
33 | - Coro::Debug::new_unix_server did not unlink the socket |
41 | - Coro::Debug::new_unix_server did not unlink the socket |