aboutsummaryrefslogtreecommitdiff
path: root/lundump.h (unfollow)
Commit message (Collapse)AuthorFilesLines
2010-05-11wrong type being assigned (not detected because of obj2gco macro)Roberto Ierusalimschy1-2/+2
2010-05-10udata in 'tobefnz' list be have old bit on (it will be clearedRoberto Ierusalimschy1-2/+1
when udata is moved to 'allgc' list)
2010-05-10corrected some places where an old object could end up in frontRoberto Ierusalimschy4-10/+15
of a new one + minimal documentation about this problem
2010-05-10new macro 'resetoldbit'Roberto Ierusalimschy2-7/+9
2010-05-10macros 'lua_number2int' and 'lua_number2uint' for Visual StudioRoberto Ierusalimschy1-11/+12
corrected and tested
2010-05-10avoid reserving LUA_RIDX_LAST slots in any table used by referenceRoberto Ierusalimschy1-15/+9
system. Store free list in a field indexed by a unique name, instead of using a numerical index.
2010-05-10corrected definition of 'lua_register' (there is no LUA_ENVIRONINDEXRoberto Ierusalimschy1-3/+2
anymore)
2010-05-07slightly better definition for 'changenvalue'Roberto Ierusalimschy1-3/+2
2010-05-07'lua_assert' can be empty when assertions are offRoberto Ierusalimschy1-2/+2
2010-05-07slightly better definition for 'isgray'Roberto Ierusalimschy1-2/+3
2010-05-07avoid 'else assert' (which may result in an empty else)Roberto Ierusalimschy1-4/+4
2010-05-07details (to avoid too long strings in assertions)Roberto Ierusalimschy1-4/+4
2010-05-07bug: incremental sweep was not cleaning old bits (as it stopped in theRoberto Ierusalimschy1-28/+28
first old object) + bug: moving udata to 'udgc' list might put old object in front a new one + some new macros + generational mode may be in 'pause' state (it just skips 'markroot')
2010-05-07stupid bug when calling 'luaC_changemode' (in function lua_gc)Roberto Ierusalimschy1-3/+6
2010-05-07more tests in 'lua_checkmemory' + more information in functionRoberto Ierusalimschy1-17/+29
'gccolor'
2010-05-07new macro 'isgenerational' + new macro 'isold' + better deffinitionRoberto Ierusalimschy1-3/+7
for 'isdead', compatible with the code used by 'sweeplist'
2010-05-06some cleaning in 'sweeplist' (threads do not need to be traversedRoberto Ierusalimschy1-28/+29
if they are going to be collected + use of bit masks to unify differences in generational and incremental modes)
2010-05-06commentsRoberto Ierusalimschy1-18/+9
2010-05-06more tests in 'lua_checkmemory'Roberto Ierusalimschy1-13/+48
2010-05-05new function 'luaC_changemode' +Roberto Ierusalimschy1-13/+42
bug: objects entering the 'allgc' list must have their OLDBIT cleared (upvalues being cleared) + bug: in 'checkSizes', KGC_EMERGENCY is stored in 'gckind' field, not in 'gcstate' + current white changes when entering sweep phase (so there are dead objects only in that phase)
2010-05-05new function 'luaC_changemode'Roberto Ierusalimschy2-10/+6
2010-05-05typos in commentsRoberto Ierusalimschy3-7/+7
2010-05-05better documentation of GC costsRoberto Ierusalimschy1-12/+26
2010-05-04changes in patch to monitor garbage collectionRoberto Ierusalimschy1-10/+6
2010-05-04full GC must run finalizers only after finishing everything elseRoberto Ierusalimschy1-5/+5
(including a return to generational mode if needed)
2010-05-04details (comments)Roberto Ierusalimschy1-3/+3
2010-05-04missing parentheses in declarationRoberto Ierusalimschy1-2/+2
2010-05-04no need to handle '\0' differently from other control chars inRoberto Ierusalimschy1-2/+2
format '%q'
2010-05-03no need of a KGC_FORCED collection kind; it has the same behavior ofRoberto Ierusalimschy2-9/+7
a normal collection
2010-05-03no more 'finalize' phase in GC; finalizers are called along theRoberto Ierusalimschy3-31/+33
entire cycle
2010-05-03items in 'tobefnz' are kept black (as before recent change) and changedRoberto Ierusalimschy2-9/+15
to white only when needed (being moved to 'allgc' when not keeping invariant).
2010-05-03invariant must be kept in atomic 'phase' tooRoberto Ierusalimschy1-9/+9
2010-04-30"gray lists" only need to be valid when 'keepinvariant' is trueRoberto Ierusalimschy2-15/+11
2010-04-30added comment explaining a bit about the invariants of the collectorRoberto Ierusalimschy1-1/+14
2010-04-30detailRoberto Ierusalimschy1-3/+2
2010-04-30wrong commentRoberto Ierusalimschy1-2/+2
2010-04-30default PAUSE should be 200 (differences may be corrected inRoberto Ierusalimschy1-6/+6
internal multiplers) + check for unbilt stack equal the one used by lgc.c
2010-04-29nasty GC bug: upvalue must be turned white when not keeping invariant,Roberto Ierusalimschy3-9/+26
but barrier was not being called when uv->v were already white.
2010-04-29removed commented-out debugging codeRoberto Ierusalimschy1-3/+1
2010-04-29new way to control GC speedRoberto Ierusalimschy2-16/+26
2010-04-29'luaC_linkupval' embedded into call siteRoberto Ierusalimschy1-4/+10
2010-04-29improved 'lua_checkmemory', with better control over gray objectsRoberto Ierusalimschy1-24/+45
2010-04-29'luaC_linkupval' moved into 'lfunc.c' code + new way to control GC speedRoberto Ierusalimschy2-80/+80
2010-04-29new way to control GC speed (keeping a 'debt' counter)Roberto Ierusalimschy3-10/+11
2010-04-26lots of new comments + small changes in loop controls + other smallRoberto Ierusalimschy1-117/+161
"janitor work"
2010-04-20"to-be-finalized" objects marked and sweeped like all other objectsRoberto Ierusalimschy1-13/+13
2010-04-20upvalue barriers for Lua functions must act on the upvalue itself,Roberto Ierusalimschy1-6/+10
not on its closure
2010-04-19missing parentheses around 'luaL_pushresultsize' declarationRoberto Ierusalimschy1-2/+2
2010-04-19bug with io.read(op, "*n")Roberto Ierusalimschy1-4/+14
2010-04-19allows thread switches (when/if there are thread switches) in the sameRoberto Ierusalimschy1-5/+7
places that finalizers can run (so they should be safe...)
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-08-13 22:23:21 +0000