| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| ... | |
| * | really stop collection during finalizers | Roberto Ierusalimschy | 2010-05-17 | 1 | -2/+2 |
| * | avoid "strong" cast | Roberto Ierusalimschy | 2010-05-11 | 1 | -2/+3 |
| * | corrected some places where an old object could end up in front | Roberto Ierusalimschy | 2010-05-10 | 1 | -5/+4 |
| * | new macro 'resetoldbit' | Roberto Ierusalimschy | 2010-05-10 | 1 | -6/+6 |
| * | avoid 'else assert' (which may result in an empty else) | Roberto Ierusalimschy | 2010-05-07 | 1 | -4/+4 |
| * | bug: incremental sweep was not cleaning old bits (as it stopped in the | Roberto Ierusalimschy | 2010-05-07 | 1 | -28/+28 |
| * | some cleaning in 'sweeplist' (threads do not need to be traversed | Roberto Ierusalimschy | 2010-05-06 | 1 | -28/+29 |
| * | new function 'luaC_changemode' + | Roberto Ierusalimschy | 2010-05-05 | 1 | -13/+42 |
| * | better documentation of GC costs | Roberto Ierusalimschy | 2010-05-05 | 1 | -12/+26 |
| * | full GC must run finalizers only after finishing everything else | Roberto Ierusalimschy | 2010-05-04 | 1 | -5/+5 |
| * | no need of a KGC_FORCED collection kind; it has the same behavior of | Roberto Ierusalimschy | 2010-05-03 | 1 | -5/+4 |
| * | no more 'finalize' phase in GC; finalizers are called along the | Roberto Ierusalimschy | 2010-05-03 | 1 | -25/+28 |
| * | items in 'tobefnz' are kept black (as before recent change) and changed | Roberto Ierusalimschy | 2010-05-03 | 1 | -8/+11 |
| * | "gray lists" only need to be valid when 'keepinvariant' is true | Roberto Ierusalimschy | 2010-04-30 | 1 | -13/+8 |
| * | nasty GC bug: upvalue must be turned white when not keeping invariant, | Roberto Ierusalimschy | 2010-04-29 | 1 | -1/+21 |
| * | 'luaC_linkupval' moved into 'lfunc.c' code + new way to control GC speed | Roberto Ierusalimschy | 2010-04-29 | 1 | -76/+60 |
| * | lots of new comments + small changes in loop controls + other small | Roberto Ierusalimschy | 2010-04-26 | 1 | -117/+161 |
| * | "to-be-finalized" objects marked and sweeped like all other objects | Roberto Ierusalimschy | 2010-04-20 | 1 | -13/+13 |
| * | new macro LUA_NUMTAGS | Roberto Ierusalimschy | 2010-04-12 | 1 | -2/+2 |
| * | cannot sweep main thread in sweepstring phase | Roberto Ierusalimschy | 2010-04-05 | 1 | -7/+16 |
| * | added proper code to trace garbage collection | Roberto Ierusalimschy | 2010-04-02 | 1 | -7/+1 |
| * | global table now is only kept in the registry | Roberto Ierusalimschy | 2010-03-29 | 1 | -3/+1 |
| * | no more fenvs! | Roberto Ierusalimschy | 2010-03-26 | 1 | -5/+2 |
| * | major collections in generational mode | Roberto Ierusalimschy | 2010-03-25 | 1 | -6/+14 |
| * | 'mainthread' is not inserted in the 'allgc' list anymore, but swept | Roberto Ierusalimschy | 2010-03-25 | 1 | -9/+10 |
| * | generational mode no longer sweep old objects | Roberto Ierusalimschy | 2010-03-24 | 1 | -18/+38 |
| * | userdata with finalizers are kept in a separated list ('udgc'), instead | Roberto Ierusalimschy | 2010-03-24 | 1 | -26/+30 |
| * | BUG: emergency collector might resize 'strt' (string table) when | Roberto Ierusalimschy | 2010-03-23 | 1 | -11/+27 |
| * | draft version of a generational mode for garbage collection. (Not well | Roberto Ierusalimschy | 2010-03-22 | 1 | -27/+54 |
| * | no more pseudoindex LUA_GLOBALSINDEX; global table now accessible | Roberto Ierusalimschy | 2009-12-22 | 1 | -4/+3 |
| * | allocator function receives the tag of object being allocated in 'osize' | Roberto Ierusalimschy | 2009-12-17 | 1 | -4/+12 |
| * | better to keep GC state numbers sequential, to optimize switch in | Roberto Ierusalimschy | 2009-12-11 | 1 | -8/+8 |
| * | new function 'luaC_runtilstate' to advance GC until a "valid" state | Roberto Ierusalimschy | 2009-12-11 | 1 | -8/+20 |
| * | comment typos | Roberto Ierusalimschy | 2009-11-26 | 1 | -2/+2 |
| * | (huge) simplification of GC management | Roberto Ierusalimschy | 2009-11-18 | 1 | -91/+42 |
| * | registry and global table may be changed through the API without a | Roberto Ierusalimschy | 2009-11-09 | 1 | -1/+4 |
| * | avoid using 'ttype' when there is an explicit test + macro 'checkdeadkey' | Roberto Ierusalimschy | 2009-11-06 | 1 | -5/+7 |
| * | better control over accesses to TValue fields | Roberto Ierusalimschy | 2009-11-05 | 1 | -4/+4 |
| * | no more one environment per thread: all threads share a single global | Roberto Ierusalimschy | 2009-10-23 | 1 | -5/+4 |
| * | information about upvalues (where they come from) kept in Proto structure, | Roberto Ierusalimschy | 2009-09-28 | 1 | -4/+4 |
| * | some operations may shrink g->totalbytes so g->estimate must be | Roberto Ierusalimschy | 2009-09-28 | 1 | -4/+9 |
| * | bug: stack must be cleared until its end (including extra size) + | Roberto Ierusalimschy | 2009-07-16 | 1 | -26/+9 |
| * | new macro 'condmovestack' instead of 'condhardstacktests' | Roberto Ierusalimschy | 2009-06-08 | 1 | -2/+2 |
| * | errors in finalizers are propagated with code LUA_ERRGCMM (ERRor in | Roberto Ierusalimschy | 2009-05-21 | 1 | -6/+15 |
| * | resize string hash table only when new size is smaller than current one | Roberto Ierusalimschy | 2009-04-29 | 1 | -3/+7 |
| * | new way to GC stacks: the entire stack must be correct all the times; | Roberto Ierusalimschy | 2009-04-28 | 1 | -30/+40 |
| * | 'CallInfo' stack implemented as double-linked list instead of an array | Roberto Ierusalimschy | 2009-04-17 | 1 | -20/+19 |
| * | yielding across lua_call (first version) | Roberto Ierusalimschy | 2009-03-10 | 1 | -2/+2 |
| * | finalizers always called protected | Roberto Ierusalimschy | 2009-02-17 | 1 | -2/+8 |
| * | simplification in the handling of finalizers: no more 'tmudata' list + | Roberto Ierusalimschy | 2008-06-26 | 1 | -72/+34 |
