 Roberto Ierusalimschy
|
1b0f943da7
Bug: new metatable in weak table can fool the GC
|
hai 2 meses |
|
Roberto Ierusalimschy
|
de794a6527
Towards release 5.4.7
|
hai 1 ano |
|
Roberto Ierusalimschy
|
9b4f39ab14
More disciplined use of 'getstr' and 'tsslen'
|
%!s(int64=2) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
6b51133a98
Thread stacks resized in the atomic phase
|
%!s(int64=2) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
9be74ccc21
Several functions turned 'static'
|
%!s(int64=2) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
fa2f294dd1
Reduce calls to 'luaC_step' when GC is stopped
|
%!s(int64=2) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
9ede317c70
Threads are created like other objects
|
%!s(int64=2) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
413a393e62
Stack indices changed to union's
|
%!s(int64=2) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
a1f77a234a
Bug: set correct pause when (re)entering gen. collection.
|
%!s(int64=3) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
597a53bbc6
Bug: finalizer calling exit can corrupt finalization order
|
%!s(int64=3) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
066e0f93c4
Fix debug information about finalizers
|
%!s(int64=3) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
0bfc572e51
Bug: GC is not reentrant
|
%!s(int64=3) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
1537d6680b
New control for reentrancy of emergency collections
|
%!s(int64=4) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
59c88f846d
Broadening the use of branch hints
|
%!s(int64=4) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
d282652561
Bug when growing a stack
|
%!s(int64=4) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
52c8679760
Fixed bug of keys removed from tables vs 'next'
|
%!s(int64=4) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
c23cc86c54
Details
|
%!s(int64=4) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
5aa36e894f
No more field 'lua_State.stacksize'
|
%!s(int64=4) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
8a89da07ba
Better control of gray objects
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
f849885a4b
Small changes in macros that change GC colors
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
f7ce7e5faa
TOUCHED2 objects are not always black
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
65141832d2
Open upvalues should be gray when entering gen. mode
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
68109afcdb
Detail (in asserts)
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
9cf3299faf
Threads don't need to always go to 'grayagain'
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
b9b554e0f6
Clearer handling of gray lists when entering generational mode
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
0dc5deca1c
Optimization in 'markold'
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
b4c353434f
Details
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
71f70df327
OLD1 ages advanced by 'markold'
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
663f83f647
Same changes around 'correctgraylist'
|
%!s(int64=5) %!d(string=hai) anos |
|
Roberto Ierusalimschy
|
d2c2e32e8a
All objects are kept 'new' in incremental GC
|
%!s(int64=5) %!d(string=hai) anos |
