diff options
author | Peter Drahoš <drahosp@gmail.com> | 2010-10-01 03:22:32 +0200 |
---|---|---|
committer | Peter Drahoš <drahosp@gmail.com> | 2010-10-01 03:22:32 +0200 |
commit | 89d9c98af1ac352ba4d49d660e61b0853d6e1a86 (patch) | |
tree | 15c56d2ce66b4ab147171c0f674cdb4a435ff13f /CHANGES | |
download | lanes-89d9c98af1ac352ba4d49d660e61b0853d6e1a86.tar.gz lanes-89d9c98af1ac352ba4d49d660e61b0853d6e1a86.tar.bz2 lanes-89d9c98af1ac352ba4d49d660e61b0853d6e1a86.zip |
Import to git
Diffstat (limited to 'CHANGES')
-rw-r--r-- | CHANGES | 72 |
1 files changed, 72 insertions, 0 deletions
@@ -0,0 +1,72 @@ | |||
1 | |||
2 | CHANGES: | ||
3 | |||
4 | CHANGE X: | ||
5 | |||
6 | CHANGE 12 (bug fix on Windows, 2.0.3) AKa 25-Jan-2009: | ||
7 | Did CHANGE 9 the way it should be done. | ||
8 | |||
9 | CHANGE 11 (new feature, 2.0.3) AKa 23-Jan-2009: | ||
10 | Finalizers ('set_finalizer()') for being able to do cleanup of a lane's | ||
11 | resources, whether it returned succesfully or via an error. | ||
12 | |||
13 | CHANGE 10 (new feature, 2.0.3) AKa 23-Jan-2009: | ||
14 | Call stack showing where an error occurred is not merged with the error | ||
15 | message, but delivered as a separate stack table ({ "filename:line" [, ...] }). | ||
16 | Getting call stacks of errorred lanes is now possible. | ||
17 | |||
18 | CHANGE 9 (bug fix on Windows) AKa 10-Dec-2008 (> 2.0.2): | ||
19 | Applied patch from Kriss Daniels to avoid issues on 'now_time()' in Win32 | ||
20 | (http://luaforge.net/forum/forum.php?thread_id=22704&forum_id=1781). | ||
21 | |||
22 | CHANGE 8 (bug fix) AKa 26-Oct-2008: | ||
23 | Avoids occasional segfault at process exit (on multicore CPUs). Does this | ||
24 | by keeping track of "free running" threads (s.a. the time thread) and | ||
25 | cancelling them at process exit. | ||
26 | |||
27 | Tested (2.0.2) on Linux 64,x86, OS X, WinXP. | ||
28 | |||
29 | CHANGE 7 (bug fix) AKa 15-Oct-2008: | ||
30 | Recursive functions that use themselves as direct upvalue can now be | ||
31 | passed to other lanes, and used as a lane function. | ||
32 | |||
33 | CHANGE 6 (bug fix) AKa 15-Oct-2008: | ||
34 | Added local caches of the following to src/lanes.lua (was otherwise getting | ||
35 | errors at least in 'tests/irayo_recursive.lua'). | ||
36 | |||
37 | local assert= assert | ||
38 | local string_gmatch= assert( string.gmatch ) | ||
39 | local select= assert( select ) | ||
40 | local type= assert( type ) | ||
41 | local pairs= assert( pairs ) | ||
42 | local tostring= assert( tostring ) | ||
43 | local error= assert( error ) | ||
44 | local setmetatable= assert( setmetatable ) | ||
45 | local rawget= assert( rawget ) | ||
46 | |||
47 | Thanks to Irayo for detecting and reporting this. | ||
48 | |||
49 | CHANGE 5 (new feature): | ||
50 | Modifying Makefile so it's better suited to LuaRocks. | ||
51 | |||
52 | CHANGE 4 (new feature): | ||
53 | Metatable copying, allowing Lua objects to be copied across lanes. | ||
54 | |||
55 | CHANGE 3 (bug fix) AKa 5-Aug-2008: | ||
56 | The '__gc' method was not tied to thread userdata, at all. Caused memory | ||
57 | lifespan problems at least on OS X when threads were cancelled (EINVAL). | ||
58 | |||
59 | CHANGE 2 (bug fix) AKa 5-Aug-2008: | ||
60 | Better calculation of timeouts, always making them absolute (even in Win32) | ||
61 | to allow for events that wake the lane up but don't read/write the Linda | ||
62 | key that it was observing. | ||
63 | |||
64 | CHANGE 1 (bug fix) AKa 4-Aug-2008: | ||
65 | Signalling woke up only one waiting thread, not all. This caused i.e. | ||
66 | receive to not wake up if there was another thread waiting on the same | ||
67 | Linda object. | ||
68 | |||
69 | PThread fix: using 'pthread_cond_broadcast()' instead of 'pthread_cond_signal()' | ||
70 | Win32 fix: using manual events and 'PulseEvent()' | ||
71 | |||
72 | (end) | ||