mozilla
/
pjs
зеркало из https://github.com/mozilla/pjs.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность
264 762 коммитов 40 Ветки 136 Теги 1,1 GiB
Граф коммитов

1531 Коммитов

Автор SHA1 Сообщение Дата
jag@tty.nl e0d2da14d0 Bug 416843: gqi.py generates dependency information that MSYS doesn't understand. r=bsmedberg, a=stuart 2008-02-13 00:13:33 -08:00
benjamin@smedbergs.us f54b2d462b Bug 412320 - generate QI for additional classes which, according to stats, show up in profiles as frequent callers of nsID::Equals, r=biesi,ted,sicking 2008-02-12 11:46:26 -08:00
jwalden@mit.edu eef0eb3f84 Bug 397929 - Convert GetMessage APIs to GetMessageMoz APIs, at the C++ symbol level only (vtables remain the same), to work around brain-dead, idiotic, insane Windows API macros. r=bsmedberg, a=schrep 2008-02-11 20:11:07 -08:00
dbaron@dbaron.org 30cc3e8b80 Warn when Unlink doesn't do what it should. b=414972 r+sr=peterv a=not part of default build (DEBUG_CC only) 2008-02-08 11:48:51 -08:00
dbaron@dbaron.org 0192ff9b4b And actually get the right bug number. b=415426 (which is what the previous checkin was for as well) 2008-02-03 11:59:07 -08:00
dbaron@dbaron.org 9c31c31e0d Switch back to evil casts, which are the only solution that works across SDK version. b=415252 2008-02-03 11:51:14 -08:00
dbaron@dbaron.org f7ee9cf9f6 Match new (more-const) signature for PENUMLOADED_MODULES_CALLBACK and PENUMLOADED_MODULES_CALLBACK64. b=415426 Fixing bustage on Windows tinderbox from SDK upgrade. 2008-02-03 11:34:31 -08:00
peterv@propagandism.org 0e982bae66 Fix for bug 407034 (JS_Assert "!rt->gcRunning" unbinding link elements in cycle collector with JS protocol handlers), r/sr=dbaron. 2008-01-29 18:05:43 -08:00
peterv@propagandism.org 84bde55fcb Try to fix orange 2008-01-29 13:42:05 -08:00
peterv@propagandism.org 0d4ee454ec Fix for bug 407034 (JS_Assert "!rt->gcRunning" unbinding link elements in cycle collector with JS protocol handlers), r/sr=dbaron. 2008-01-29 12:48:32 -08:00
benjamin@smedbergs.us b0487ed29a Bug 412320 - Try code-generation for QI impls, r=roc+luser 2008-01-25 09:14:48 -08:00
reed@reedloden.com 2f64c1bcab Bug 376530 - "Fix CycleCollector building with uClibc library" [p=romaxa@gmail.com (romaxa) r+sr=peterv a1.9=schrep] 2008-01-21 16:17:41 -08:00
dbaron@dbaron.org 58d61560c6 Fix DEBUG_CC bustage treating all XPConnect roots as expected garbage. b=410116 r+sr=peterv Not part of default build (DEBUG_CC-only). 2008-01-11 14:24:04 -08:00
dbaron@dbaron.org e1238a0757 Fix DEBUG_CC bustage from bug 402966. All changes #ifdef DEBUG_CC (not part of the default build). 2008-01-10 14:27:07 -08:00
peterv@propagandism.org 7f474b0c00 Fix for bug 402966 (JS_Assert "!rt->gcRunning"). r/sr=dbaron. 2008-01-10 06:10:03 -08:00
mozilla@weilbacher.org 7ee592b381 [OS/2] Bug 404407: improve memory consumption on OS/2. Part 1, call heapmin after cycle collection, r=dbaron 2008-01-09 13:14:08 -08:00
bzbarsky@mit.edu c593efcead Make non-threadsafe addreff/release in nsAgg assert the owning thread. Bug 410746, r+sr=bsmedberg 2008-01-09 13:14:02 -08:00
bzbarsky@mit.edu 81f1689c47 Report cycle collector faults to the error console and the observer service. Bug 410295, r+sr=peterv, a=schrep 2008-01-06 14:05:10 -08:00
jruderman@hmc.edu 2aadf2131c Fix DEBUG_CC compilation error (bug 410036). r+sr=peterv, a=blocking1.9+. DEBUG_CC-only change. 2008-01-04 19:57:09 -08:00
karlt+@karlt.net 5077dfb36a Bug 326594 - Consider using backtrace() for Unix stack traces: Add safety checks to i386/ppc unwind and use libgcc_s for amd64 and others r=dbaron,bsmedberg a=schrep 2007-11-11 20:45:40 -08:00
bent.mozilla@gmail.com e07ff086bf Bug 396369 - "Overload forget method on nsCOMPtr/nsRefPtr to work with out parameters". r=dbaron, sr=bsmedberg, a=damons. 2007-11-11 10:56:45 -08:00
jst@mozilla.org 25d6ed4518 Landing patch for bug 401687. Stop reference counting JS objects in the cycle collector. Patch by peterv@propagandism.org, r+sr=dbaron@mozilla.com,igor@mir2.org, a=dsicore@mozilla.com 2007-11-01 15:51:57 -07:00
peterv@propagandism.org 99066a28f9 Part 1 of fix for bug 379718 (using trace API for reference counts) and bug 386912 (cycle collector faults after tracing "JS object but unknown to the JS GC"). r=igor/jst, sr=jst, a=blocking1.9+/M9 (for bug 386912). 2007-10-29 06:45:07 -07:00
peterv@propagandism.org af78151f86 Backing out once more to fix orange. 2007-10-28 05:49:30 -07:00
peterv@propagandism.org d15713f435 Part 1 of fix for bug 379718 (using trace API for reference counts) and bug 386912 (cycle collector faults after tracing "JS object but unknown to the JS GC"). r=igor/jst, sr=jst, a=blocking1.9+/M9 (for bug 386912). 2007-10-28 04:14:40 -07:00
peterv@propagandism.org b8200cb3ce Backing out to fix orange. 2007-10-26 07:15:28 -07:00
peterv@propagandism.org f262d431c4 Part 1 of fix for bug 379718 (using trace API for reference counts) and bug 386912 (cycle collector faults after tracing "JS object but unknown to the JS GC"). r=igor/jst, sr=jst, a=blocking1.9+/M9 (for bug 386912). 2007-10-26 06:25:32 -07:00
Olli.Pettay@helsinki.fi 0a4c65920e Landing (again) Bug 373462, bug 385322, Better scheduling of cycle collection/gc, r+sr=sicking,jst 2007-10-22 14:42:25 -07:00
Olli.Pettay@helsinki.fi f756e7f2cd backout Bug 373462, bug 385322. Still regress tp/talos 2007-10-21 09:46:54 -07:00
Olli.Pettay@helsinki.fi 8b43021094 Test-landing (again) Bug 373462, bug 385322, Better scheduling of cycle collection/gc, r+sr=sicking,jst 2007-10-21 07:09:29 -07:00
Olli.Pettay@helsinki.fi 5faf21e36b back out Bug 373462 / bug 385322 (again :( ) 2007-09-29 10:33:22 -07:00
Olli.Pettay@helsinki.fi 3b8e5357c6 Bug 373462 / bug 385322, Better scheduling of cycle collection/gc, r+sr=jst 2007-09-29 08:02:57 -07:00
jag@tty.nl dfa403949b Fix bustage from previous checkin for bug 392055. Move #define of XPCOM_GLUE_AVOID_NSPR from nsISupportsImpl.h to nscore.h. r/a=bsmedberg 2007-09-28 13:33:32 -07:00
Olli.Pettay@helsinki.fi 88f4d45513 backout the test patch for Bug 373462 and 385322 2007-09-23 09:10:12 -07:00
Olli.Pettay@helsinki.fi ed1c2cd917 Testland a patch for Bug 373462 and 385322, 'Better scheduling of cycle collection/gc' 2007-09-23 07:05:27 -07:00
Olli.Pettay@helsinki.fi 4451d6e7c5 backout patch for Bug 373462, Bug 385322 2007-09-21 03:00:47 -07:00
Olli.Pettay@helsinki.fi f37cbb3f11 Bug 373462, Bug 385322, Better scheduling of cycle collection/gc, r=jst, sr=sicking, a=sicking 2007-09-21 01:54:36 -07:00
dbaron@dbaron.org 781a22bce6 Remove incorrect use of NS_METHOD. b=394566 r+sr=peterv a=bzbarsky 2007-09-17 17:30:06 -07:00
jwalden@mit.edu 8d8754b0e8 Bug 348748 - Remove the NS_.*_CAST macros completely! r=bsmedberg, a=no-functionality-change 2007-09-05 15:31:12 -07:00
jwalden@mit.edu 22298007e4 Unbreak DEBUG_CC on Windows (calling convention mismatch); apparently nobody really uses it there. NPOTDB, rs=sicking 2007-08-30 14:47:26 -07:00
bent.mozilla@gmail.com 956779be18 Bug 392493 - "Add forget() method to nsCOMPtr and nsRefPtr". r=dbaron, a=bzbarsky. 2007-08-28 16:37:22 -07:00
dbaron@dbaron.org 21fd23e1f5 Back out a small part of bug 391848 to fix bustage for some people (and add a comment warning about the problem). 2007-08-27 22:42:25 -07:00
dbaron@dbaron.org 4ae4444c8a Do stack walking on a single thread to avoid deadlocks during thread creation. b=391321 r=robarnold@mozilla.com sr=jmathies@mozilla.com (delegated by bsmedberg) a1.9=bzbarsky@mit.edu 2007-08-27 22:42:04 -07:00
dbaron@dbaron.org 0383ca7d80 Fix bustage. 2007-08-22 18:29:54 -07:00
dbaron@dbaron.org d410891057 Print path to expected garbage along with refcount imbalances. b=387224 r=graydon a=brendan 2007-08-22 17:47:47 -07:00
dbaron@dbaron.org e1aa5de18c When doing ExplainLiveExpectedGarbage, suspect all nsXPCWrappedJS so that we get more objects in the graph. b=387224 r=graydon a=brendan 2007-08-22 17:47:24 -07:00
dbaron@dbaron.org 7057a344e9 Make ExplainLiveExpectedGarbage print *all* references to JS objects with refcount imbalances. b=387224 r=graydon a=brendan 2007-08-22 17:46:43 -07:00
bzbarsky@mit.edu 1aade1f2bb Make HTML content objects use QI tables. Bug 389797, r+sr=jst, a=sicking. 2007-08-20 15:55:06 -07:00
dbaron@dbaron.org fee77550e0 Allow builds compiled with VC8 to function with older versions of dbghelp.dll. b=391848 r+a=bsmedberg 2007-08-19 14:38:31 -07:00
dbaron@dbaron.org 3c01b39a9d Debugging code and cleanup from fixing Windows stack traces to work on VC8-compiled builds in addition to VC7.1-compiled builds. b=391848 r=bsmedberg a=bzbarsky 2007-08-19 14:37:55 -07:00
dbaron@dbaron.org 79a14f243a Enable Linux stack walking code on Mac OS X. b=336517 r+a=bsmedberg 2007-08-15 17:03:29 -07:00
dbaron@dbaron.org 03daa47a88 Include file name and line number in Windows stack traces. b=391850 r+a=bsmedberg 2007-08-14 09:34:32 -07:00
dbaron@dbaron.org 2537e2adc1 Merge nsStackFrameWin and nsStackFrameUnix into nsStackWalk. b=391793 r+a=bsmedberg 2007-08-13 12:37:36 -07:00
dbaron@dbaron.org 5385fc565a Remove nsTraceRefcnt::LoadLibrarySymbols, which is no longer needed due to the error handling in SymGetModuleInfoEspecial. b=391793 r+a=bsmedberg 2007-08-13 12:36:57 -07:00
ginn.chen@sun.com 73535c643d Bug 374689 Fix build issues r+a=dbaron 2007-08-12 22:29:55 -07:00
dbaron@dbaron.org 422a0fc1a3 Convert trace-malloc to using XPCOM stack walking API. b=374829 r+a=brendan 2007-08-10 15:20:48 -07:00
dbaron@dbaron.org a8f872d6ae Export a stack walking API (usable from C or C++) from XPCOM. b=374689 r=bsmedberg a=bzbarsky 2007-08-10 14:32:49 -07:00
dbaron@dbaron.org 5db8176c4c Split the address->symbol mapping from the rest of the stack walking code. b=374829 r=bsmedberg a=bzbarsky 2007-08-10 14:29:37 -07:00
dbaron@dbaron.org c69f04c4b5 Export a stack walking API (usable from C or C++) from XPCOM. b=374689 r=bsmedberg a=bzbarsky 2007-08-10 14:28:22 -07:00
bzbarsky@mit.edu bf49f30281 Add a stack-and-abort value of XPCOM_DEBUG_BREAK 2007-08-10 12:42:58 -07:00
benjamin@smedbergs.us a8f4f63a68 Bug 386025 - Expose cycle-collection symbols, r=graydon 2007-08-06 07:34:02 -07:00
dcamp@mozilla.com d554b1cee2 Add an API to access <input type="file"> data. r=smontagu (for charset stuff), r+sr=jonas 2007-07-24 21:29:47 -07:00
dbaron@dbaron.org d31f5c81a7 Explain another possibility in this debugging printf. DEBUG_CC only. 2007-07-14 12:06:10 -07:00
neil@parkwaycc.co.uk 314a7d97d4 Followup to bug 348748 to fix the alignment of trailing backslashes in macros 
[xxx_cast<foo*>(bar) is three characters shorter than NS_XXX_CAST(foo*, bar)]
 2007-07-11 01:46:44 -07:00
dbaron@dbaron.org 78eff2581b Make ExplainLiveExpectedGarbage traverse the suspectCurrent roots, since now that it's more accurate, more graph is always better. b=387005 r=graydon 2007-07-08 12:26:44 -07:00
jwalden@mit.edu 12e960c504 Bug 348748 - Replace all instances of NS_STATIC_CAST and friends with C++ casts (and simultaneously bitrot nearly every patch in existence). r=bsmedberg on the script that did this. Tune in next time for Macro Wars: Episode II: Attack on the LL_* Macros. 2007-07-08 00:08:04 -07:00
dbaron@dbaron.org cffccfd62f Make Fault give more information, #ifdef DEBUG_CC. b=386947 r=graydon 2007-07-05 15:38:38 -07:00
dbaron@dbaron.org d74585d1d5 In ExplainLiveExpectedGarbage, print only the externally-referenced nodes from which the expected garbage is reachable. b=387005 r=graydon 2007-07-05 15:20:16 -07:00
dbaron@dbaron.org 63ccfec235 Make suspectCurrent assertion not fire constantly after we Fault. b=386664 r+sr=peterv 2007-07-03 14:21:57 -07:00
dbaron@dbaron.org b5932e37cc Change cycle collector aging constant from 10 to 0. b=381199 r+sr=peterv 2007-06-27 13:41:26 -07:00
dbaron@dbaron.org b6b7f7a826 Fix gcc warning about casting string literal to char*, DEBUG-only. 2007-06-27 13:40:52 -07:00
dbaron@dbaron.org 2664dc19bc Keep the purple bit in sync with the purple buffer when suspect and forget calls are dropped on the floor. b=385548 r=graydon 2007-06-27 10:51:48 -07:00
dbaron@dbaron.org a290d344b5 Back out 385548 to try to fix test orange. 2007-06-26 22:26:38 -07:00
dbaron@dbaron.org d17ffc8ccc Don't set the purple bit when suspect calls are dropped on the floor. b=385548 r=graydon 2007-06-26 21:47:40 -07:00
dbaron@dbaron.org 9bfd8d42d0 Print the warning message for failure to suspect or unlink only for the objects that are the roots of what would have been collected. b=385549 r=graydon 2007-06-25 13:36:24 -07:00
dbaron@dbaron.org 964ff03daf Back out again. 2007-06-21 17:30:41 -07:00
dbaron@dbaron.org d612a8c004 See if using mScanDelay of 1 instead of 0 fixes the leak regression. 2007-06-21 16:49:37 -07:00
dbaron@dbaron.org b2d6b62837 Change cycle collector aging constant from 10 to 0. b=381199 r+sr=peterv 2007-06-21 16:02:53 -07:00
graydon@mozilla.com 3574f6ee0a Bug 384942, dump graph on cycle collection fault. 
r=dbaron
 2007-06-18 18:29:10 -07:00
dbaron@dbaron.org 8164e1ba8d Back out bug 381199. 2007-06-13 17:12:16 -07:00
dbaron@dbaron.org a499511012 Change cycle collector aging constant from 10 to 0. b=381199 r+sr=peterv 2007-06-13 16:22:19 -07:00
peterv@propagandism.org 132a3c131e Fix for bug 383651 (Re-entrancy into nsCycleCollector::Collect can result in a crash). Patch by ak.miller@auckland.ac.nz, r=graydon, sr=dbaron. 2007-06-08 03:17:21 -07:00
dbaron@dbaron.org 780ee712cd Don't loop through all the possible shutdown collections when we suspect all native wrappers. b=383234 r+sr=peterv 2007-06-06 15:09:00 -07:00
dbaron@dbaron.org 810148e879 Shutdown in one cycle when we set mScanDelay to 0. b=383234 r+sr=peterv 2007-06-06 15:08:25 -07:00
dbaron@dbaron.org a337c41e62 strdup when allocating mName so we have useful pointers. b=383234 r+sr=peterv 2007-06-06 15:07:29 -07:00
dbaron@dbaron.org 28961934db Print info about things that are suspected on the extra collections at shutdown to catch deficiencies in traversal that cause us to need the extra collections. b=383234 r+sr=peterv 2007-06-06 15:06:56 -07:00
dbaron@dbaron.org af57aaaad5 When explaining why expected garbage was not collected, list the known references when some were unknown. b=383234 r+sr=peterv 2007-06-06 15:06:00 -07:00
dbaron@dbaron.org e5778015f1 Make nsTraceRefcnt print addresses along with serial numbers, and fix warning about casting to a function pointer. b=383234 r=peterv 2007-06-06 15:05:25 -07:00
dbaron@dbaron.org f3346ea718 Make DEBUG_CC compile again. b=383234 r+sr=peter 2007-06-06 15:04:26 -07:00
peterv@propagandism.org 9638d6c9f8 Fix warning. 2007-05-25 03:57:38 -07:00
peterv@propagandism.org 418a6ec9e1 Fix for bug 368774 (Make cycle collector work with refcounted non-XPCOM objects). r=dbaron, sr=sicking. 2007-05-24 07:10:02 -07:00
jonas@sicking.cc c8e8237fd4 Relanding bug 378987. Don't try to collect windows, documents or elements of actively viewed pages. r=bz/jst sr=jst 2007-05-10 13:21:12 -07:00
jst@mozilla.org 3484a36643 Backing out Jonas' last changes due to tinderbox orange, per Jonas' request. 2007-05-09 17:42:33 -07:00
jonas@sicking.cc eb28533fb6 Relanding bug 378987. Don't try to collect windows, documents or elements of actively viewed pages. r=bz/jst sr=jst 2007-05-09 16:33:28 -07:00
jonas@sicking.cc 77e2d700f6 back out last patch since leaks went through the roof 2007-05-07 18:03:08 -07:00
jonas@sicking.cc fe815718d5 Don't attempt to cycle collect documents, windows and elements that are currently being viewed. r=bz, sr=jst b=378987 2007-05-07 16:45:25 -07:00
dbaron@dbaron.org 8ca3a248f0 Remove extra class names to address Neil's comments. b=378514 2007-05-03 16:07:26 -07:00
dbaron@dbaron.org 63ebd5de73 Record results of cycle collection traversal so we only have to call the traversal APIs once per collection. b=378514 sr=peterv r=graydon 2007-04-25 14:12:11 -07:00
peterv@propagandism.org 2e4c901958 Fix for bug 376854 (CycleCollector depends on uninitialized values (ageSelectionCallback and nsPurpleBuffer::BumpGeneration)). r/sr=dbaron. 2007-04-25 11:33:55 -07:00
peterv@propagandism.org 37ea48ef85 Fix for bug 378595 (Improve cycle collection QI performance). r/sr=sicking. 2007-04-25 09:35:27 -07:00
dbaron@dbaron.org 1c62a2ab46 Followup fixes to making cycle collector use more efficient hashtables: shrink size of table and fix warnings on 64-bit machines. b=377606 sr=peterv r=graydon 2007-04-23 16:34:33 -07:00
peterv@propagandism.org 5dbe6da8e9 Fix for bug 372110 (Make cycle-collection debugging features optional at compile time). r=graydon, sr=jst. 2007-04-21 10:43:19 -07:00
peterv@propagandism.org 02d8395a6d Reorder members to make struct smaller and try to lower MH down again. 2007-04-20 01:51:31 -07:00
peterv@propagandism.org 4a00a8787c Fix for bug 377606 (Switch cycle collector to more efficient hashtables). Patch by graydon, r=peterv, sr=dbaron. 2007-04-20 01:01:01 -07:00
peterv@propagandism.org c96265ae6d Fix for bug 373693 (Cycle collector makes AddRef about 6x slower). r=graydon, sr=dbaron. 2007-04-18 03:10:35 -07:00
peterv@propagandism.org b21a47335d Fix for bug 372713 (Add cycle collection to RDF datasources). r=bsmedberg, sr=dbaron. 2007-04-10 15:05:41 -07:00
peterv@propagandism.org bc607785b9 Fix for bug 375075 (Stop leaking the cycle collector on shutdown). r=bsmedberg, sr=dbaron. 2007-03-27 02:49:06 -07:00
peterv@propagandism.org 85ec2e5b12 Backing out change for debugging tinderbox orange. 2007-03-23 09:10:21 -07:00
peterv@propagandism.org 31c71d20c8 Temporarily enable cycle collection timing for debug builds to figure out tinderbox orange. Will back out after qm-xserve01 tinderbox has cycled. 2007-03-23 08:11:53 -07:00
roc+@cs.cmu.edu 0054412272 Bug 374866. Reftests for text-transform. r=dbaron 2007-03-22 16:01:14 -07:00
bent.mozilla%gmail.com 647e6cb371 Bug 430718 - "Better objc exception logging with stack traces in debug builds". r=josh, sr=roc. a=DEBUG-only change. 2008-04-25 20:55:50 +00:00
timeless%mozdev.org 71f819b50c Bug 394114 Interfaces missing from various QI implementations. 
r=jag sr=jag a=dsicore
 2008-04-06 12:28:42 +00:00
dbaron%dbaron.org caf78b3e20 Print shortest path to expected garbage for unlink deficiencies as well. b=422848 r+sr=peterv a=DEBUG_CC only (not part of default build) 2008-03-27 17:09:51 +00:00
timeless%mozdev.org 4dd59e4169 reverting, wrong patch, missing AMBIGUOUS 2008-03-19 21:55:28 +00:00
timeless%mozdev.org 8f189a966c Bug 394114 Interfaces missing from various QI implementations. 
r=jag sr=jag a=dsicore
 2008-03-19 21:19:42 +00:00
dbaron%dbaron.org f94f6d2b1d Add knowledge of edge names to cycle collector, ifdef DEBUG_CC. b=420514 r+sr=peterv a=damons 2008-03-17 23:11:10 +00:00
reed%reedloden.com c5ac5a30fa Bug 418131 - "Need a way to detect GTK+ version so we don't break users on major update" [p=ventnor.bugzilla@yahoo.com.au (Michael Ventnor) r=bsmedberg sr=dveditz a=blocking-firefox3+] 2008-03-14 15:42:35 +00:00
reed%reedloden.com 915cc2f649 Bug 414063 - "annotate xpidl-generated c++ headers with attributes for scriptable and deprecated methods" [p=Pidgeot18@gmail.com (Joshua Cranmer) r=dbradley a1.9=damons] 2008-03-12 11:00:59 +00:00
blassey%mozilla.com b8929025f0 backing out: switching narrow char system calls to wide char in windows b=418703 2008-03-11 22:46:32 +00:00
blassey%mozilla.com fc37ea022b switching narrow char system calls to wide char in windows b=418703 r=bsmedberg r=dougt also looked at by stuart and roc a=beltzner and a=damon 2008-03-11 21:49:58 +00:00
dbaron%dbaron.org c287957e87 Fix DEBUG_CC bustage on Windows resulting from bug 402966 (and bug 407034). r+sr=peterv a=Not part of the default build (DEBUG_CC only) 2008-03-07 17:55:51 +00:00
vladimir%pobox.com 9f7e5e50a1 leftover bit from earlier 392351; get rid of compiler warnings 2008-03-06 06:39:36 +00:00
vladimir%pobox.com 3bbe768e44 b=392351, implement about:memory framework core, relading with forward-declaration victory; r=shaver,sr=bsdmedberg 2008-03-06 01:28:36 +00:00
vladimir%pobox.com 8ebc317d8e backed out 2008-03-06 00:54:15 +00:00
vladimir%pobox.com 15a69af68a b=392351, implement about:memory framework core r=shaver, sr=bsmedberg 2008-03-06 00:40:59 +00:00
blassey%mozilla.com d66d0b8772 backing out b=418703, caused test 40118 to fail on WINNT 5.2 qm-win2k3-01 dep unit test tinderbox 2008-03-05 22:56:48 +00:00
blassey%mozilla.com fbc9687a30 switching windows narrow char system calls to wide char b=418703 r=bsmedberg a=beltzner 2008-03-05 21:39:35 +00:00
reed%reedloden.com 4443155b7a Back out Ventron's patch from bug 418131 due to build failures on debug machines. 2008-03-05 11:36:37 +00:00
reed%reedloden.com 1060545a7b Bustage fix for bug 418131 (try #3). 2008-03-05 11:26:06 +00:00
reed%reedloden.com fe213ba4ab Bustage fix for bug 418131 (try #2). 2008-03-05 11:10:35 +00:00
reed%reedloden.com 3888554ba4 Bustage fix for bug 418131. 2008-03-05 10:56:33 +00:00
reed%reedloden.com ff4634aba0 Bug 418131 - "Need a way to detect GTK+ version so we don't break users on major update" [p=ventnor.bugzilla@yahoo.com.au (Michael Ventnor) r=bsmedberg sr=dveditz a=blocking-firefox3+] 2008-03-05 10:30:30 +00:00
Olli.Pettay%helsinki.fi 97ccc330e6 Bug 415025, try to improve CC scheduling, r=peterv, sr=jst 2008-02-25 17:47:26 +00:00
roc+%cs.cmu.edu 99a887fa96 Bug 417903. Don't crash if _SymGetLineFromAddr64 fails. r=benjamin,a=beltzner 2008-02-25 01:16:41 +00:00
dougt%meer.net 70510911c5 Fixes for Windows CE. b=418662, r=bsmedberg, a=damons 2008-02-23 00:22:48 +00:00
joshmoz%gmail.com 061f03bf81 touch nsObjCExceptions.h to force a rebuild. 2008-02-22 21:26:33 +00:00
bent.mozilla%gmail.com fed5cfb6e1 Bug 414977 - "insufficient unlink methods in some DOM classes?". Move the warning until after XPConnect deferred releases, r=dbaron, sr=peterv, a=blocking1.9+. 2008-02-22 20:25:12 +00:00
benjamin%smedbergs.us 624a4d7ef0 Back out bug 412320 and various followups because the perf impact was minimal and the code is more complicated 2008-02-21 20:39:23 +00:00
joshmoz%gmail.com cfde4b9f48 nsnull version of wrapping macros. b=417563 sr=roc 2008-02-20 02:23:52 +00:00
joshmoz%gmail.com 436df0cd15 obj-c exceptions work. b=417558 r=shebs sr=roc 2008-02-18 17:30:59 +00:00
peterv%propagandism.org 15af60e9b6 Fix DEBUG_CC builds. 2008-02-15 22:23:16 +00:00
peterv%propagandism.org 84d1520ba9 Fix for bug 405128 (Remove some nsDeque use from cycle collector). r/sr=dbaron, a=beltzner. 2008-02-15 13:12:55 +00:00
joshmoz%gmail.com c1783c02ea Infrastructure for catching Obj-C exceptions. Cocoa appshell usage. b=163260 r=shebs sr=roc 2008-02-13 23:11:11 +00:00
jag%tty.nl a731880086 Bug 416843: gqi.py generates dependency information that MSYS doesn't understand. r=bsmedberg, a=stuart 2008-02-13 08:13:33 +00:00
benjamin%smedbergs.us b27c6528eb Bug 412320 - generate QI for additional classes which, according to stats, show up in profiles as frequent callers of nsID::Equals, r=biesi,ted,sicking 2008-02-12 19:46:50 +00:00
jwalden%mit.edu c5afd9b749 Bug 397929 - Convert GetMessage APIs to GetMessageMoz APIs, at the C++ symbol level only (vtables remain the same), to work around brain-dead, idiotic, insane Windows API macros. r=bsmedberg, a=schrep 2008-02-12 04:11:08 +00:00
dbaron%dbaron.org 9d5185de45 Warn when Unlink doesn't do what it should. b=414972 r+sr=peterv a=not part of default build (DEBUG_CC only) 2008-02-08 19:48:51 +00:00
dbaron%dbaron.org b9801a94d3 And actually get the right bug number. b=415426 (which is what the previous checkin was for as well) 2008-02-03 19:59:07 +00:00
dbaron%dbaron.org c6cda8ec8f Switch back to evil casts, which are the only solution that works across SDK version. b=415252 2008-02-03 19:51:14 +00:00
dbaron%dbaron.org ccf2c076e8 Match new (more-const) signature for PENUMLOADED_MODULES_CALLBACK and PENUMLOADED_MODULES_CALLBACK64. b=415426 Fixing bustage on Windows tinderbox from SDK upgrade. 2008-02-03 19:34:31 +00:00
peterv%propagandism.org afcc825f59 Fix for bug 407034 (JS_Assert "!rt->gcRunning" unbinding link elements in cycle collector with JS protocol handlers), r/sr=dbaron. 2008-01-30 02:05:45 +00:00
peterv%propagandism.org 594825b911 Try to fix orange 2008-01-29 21:42:06 +00:00
peterv%propagandism.org adf10add78 Fix for bug 407034 (JS_Assert "!rt->gcRunning" unbinding link elements in cycle collector with JS protocol handlers), r/sr=dbaron. 2008-01-29 20:48:35 +00:00
benjamin%smedbergs.us 432ff728c7 Bug 412320 - Try code-generation for QI impls, r=roc+luser 2008-01-25 17:14:53 +00:00
reed%reedloden.com 72bbab5ee4 Bug 376530 - "Fix CycleCollector building with uClibc library" [p=romaxa@gmail.com (romaxa) r+sr=peterv a1.9=schrep] 2008-01-22 00:17:41 +00:00
dbaron%dbaron.org 993990d18c Fix DEBUG_CC bustage treating all XPConnect roots as expected garbage. b=410116 r+sr=peterv Not part of default build (DEBUG_CC-only). 2008-01-11 22:24:04 +00:00
dbaron%dbaron.org fb633e3fd5 Fix DEBUG_CC bustage from bug 402966. All changes #ifdef DEBUG_CC (not part of the default build). 2008-01-10 22:27:07 +00:00
peterv%propagandism.org 61924b8732 Fix for bug 402966 (JS_Assert "!rt->gcRunning"). r/sr=dbaron. 2008-01-10 14:10:08 +00:00
mozilla%weilbacher.org 6d2fb79650 [OS/2] Bug 404407: improve memory consumption on OS/2. Part 1, call heapmin after cycle collection, r=dbaron 2008-01-09 21:14:08 +00:00
bzbarsky%mit.edu edfce58396 Make non-threadsafe addreff/release in nsAgg assert the owning thread. Bug 
410746, r+sr=bsmedberg
 2008-01-09 21:14:02 +00:00
bzbarsky%mit.edu bc52549480 Report cycle collector faults to the error console and the observer 
service. Bug 410295, r+sr=peterv, a=schrep
 2008-01-06 22:05:10 +00:00
jruderman%hmc.edu 98edab664a Fix DEBUG_CC compilation error (bug 410036). r+sr=peterv, a=blocking1.9+. DEBUG_CC-only change. 2008-01-05 03:57:09 +00:00
karlt+%karlt.net 366cdd4a7a Bug 326594 ��� Consider using backtrace() for Unix stack traces: 
Add safety checks to i386/ppc unwind and use libgcc_s for amd64 and others
r=dbaron,bsmedberg a=schrep
 2007-11-12 04:45:40 +00:00
bent.mozilla%gmail.com 310f388fc2 Bug 396369 - "Overload forget method on nsCOMPtr/nsRefPtr to work with out parameters". r=dbaron, sr=bsmedberg, a=damons. 2007-11-11 18:56:45 +00:00
jst%mozilla.org 52ecd87ffb Landing patch for bug 401687. Stop reference counting JS objects in the cycle collector. Patch by peterv@propagandism.org, r+sr=dbaron@mozilla.com,igor@mir2.org, a=dsicore@mozilla.com 2007-11-01 22:51:59 +00:00
peterv%propagandism.org a41832fc13 Part 1 of fix for bug 379718 (using trace API for reference counts) and bug 386912 (cycle collector faults after tracing "JS object but unknown to the JS GC"). r=igor/jst, sr=jst, a=blocking1.9+/M9 (for bug 386912). 2007-10-29 13:45:10 +00:00
peterv%propagandism.org 6cfb25d584 Backing out once more to fix orange. 2007-10-28 12:49:33 +00:00
peterv%propagandism.org a76fba526b Part 1 of fix for bug 379718 (using trace API for reference counts) and bug 386912 (cycle collector faults after tracing "JS object but unknown to the JS GC"). r=igor/jst, sr=jst, a=blocking1.9+/M9 (for bug 386912). 2007-10-28 11:14:43 +00:00
peterv%propagandism.org 48c7f383e4 Backing out to fix orange. 2007-10-26 14:15:29 +00:00
peterv%propagandism.org c47e794776 Part 1 of fix for bug 379718 (using trace API for reference counts) and bug 386912 (cycle collector faults after tracing "JS object but unknown to the JS GC"). r=igor/jst, sr=jst, a=blocking1.9+/M9 (for bug 386912). 2007-10-26 13:25:35 +00:00
Olli.Pettay%helsinki.fi 1f2a78728d Landing (again) Bug 373462, bug 385322, Better scheduling of cycle collection/gc, r+sr=sicking,jst 2007-10-22 21:42:26 +00:00
Olli.Pettay%helsinki.fi 0d8ed891c3 backout Bug 373462, bug 385322. Still regress tp/talos 2007-10-21 16:46:56 +00:00
Olli.Pettay%helsinki.fi 00c264a0fe Test-landing (again) Bug 373462, bug 385322, Better scheduling of cycle collection/gc, r+sr=sicking,jst 2007-10-21 14:09:31 +00:00
Olli.Pettay%helsinki.fi b898a5ddce back out Bug 373462 / bug 385322 (again :( ) 2007-09-29 17:33:24 +00:00
Olli.Pettay%helsinki.fi 3462127628 Bug 373462 / bug 385322, Better scheduling of cycle collection/gc, r+sr=jst 2007-09-29 15:02:58 +00:00
jag%tty.nl ba8a8d1242 Fix bustage from previous checkin for bug 392055. Move #define of XPCOM_GLUE_AVOID_NSPR from nsISupportsImpl.h to nscore.h. r/a=bsmedberg 2007-09-28 20:33:33 +00:00
Olli.Pettay%helsinki.fi 697380898b backout the test patch for Bug 373462 and 385322 2007-09-23 16:10:13 +00:00
Olli.Pettay%helsinki.fi bb903a0cc2 Testland a patch for Bug 373462 and 385322, 'Better scheduling of cycle collection/gc' 2007-09-23 14:05:29 +00:00
Olli.Pettay%helsinki.fi cbd66d619b backout patch for Bug 373462, Bug 385322 2007-09-21 10:00:48 +00:00
Olli.Pettay%helsinki.fi 4d46f9de45 Bug 373462, Bug 385322, Better scheduling of cycle collection/gc, r=jst, sr=sicking, a=sicking 2007-09-21 08:54:36 +00:00
dbaron%dbaron.org 6d2fb0100d Remove incorrect use of NS_METHOD. b=394566 r+sr=peterv a=bzbarsky 2007-09-18 00:30:06 +00:00
jwalden%mit.edu 97daf72ebf Bug 348748 - Remove the NS_.*_CAST macros completely! r=bsmedberg, a=no-functionality-change 2007-09-05 22:31:12 +00:00
jwalden%mit.edu 29f22d595f Unbreak DEBUG_CC on Windows (calling convention mismatch); apparently nobody really uses it there. NPOTDB, rs=sicking 2007-08-30 21:47:26 +00:00
bent.mozilla%gmail.com b4fa7a162c Bug 392493 - "Add forget() method to nsCOMPtr and nsRefPtr". r=dbaron, a=bzbarsky. 2007-08-28 23:37:22 +00:00
dbaron%dbaron.org 255ff3969c Back out a small part of bug 391848 to fix bustage for some people (and add a comment warning about the problem). 2007-08-28 05:42:25 +00:00
dbaron%dbaron.org 17351774fa Do stack walking on a single thread to avoid deadlocks during thread creation. b=391321 r=robarnold@mozilla.com sr=jmathies@mozilla.com (delegated by bsmedberg) a1.9=bzbarsky@mit.edu 2007-08-28 05:42:04 +00:00
dbaron%dbaron.org 9b71a1692b Fix bustage. 2007-08-23 01:29:54 +00:00
dbaron%dbaron.org 8f482fde98 Print path to expected garbage along with refcount imbalances. b=387224 r=graydon a=brendan 2007-08-23 00:48:28 +00:00
dbaron%dbaron.org c13d4e5ee3 When doing ExplainLiveExpectedGarbage, suspect all nsXPCWrappedJS so that we get more objects in the graph. b=387224 r=graydon a=brendan 2007-08-23 00:48:27 +00:00
dbaron%dbaron.org f757dfe784 Make ExplainLiveExpectedGarbage print *all* references to JS objects with refcount imbalances. b=387224 r=graydon a=brendan 2007-08-23 00:48:26 +00:00
bzbarsky%mit.edu 7a163b9617 Make HTML content objects use QI tables. Bug 389797, r+sr=jst, a=sicking. 2007-08-20 22:55:08 +00:00
dbaron%dbaron.org 69efaea510 Allow builds compiled with VC8 to function with older versions of dbghelp.dll. b=391848 r+a=bsmedberg 2007-08-19 21:38:31 +00:00
dbaron%dbaron.org 84d808923d Debugging code and cleanup from fixing Windows stack traces to work on VC8-compiled builds in addition to VC7.1-compiled builds. b=391848 r=bsmedberg a=bzbarsky 2007-08-19 21:37:55 +00:00
dbaron%dbaron.org dd944c0956 Enable Linux stack walking code on Mac OS X. b=336517 r+a=bsmedberg 2007-08-16 00:03:29 +00:00
dbaron%dbaron.org c01336480d Include file name and line number in Windows stack traces. b=391850 r+a=bsmedberg 2007-08-14 16:34:32 +00:00
dbaron%dbaron.org 0c334e1759 Merge nsStackFrameWin and nsStackFrameUnix into nsStackWalk. b=391793 r+a=bsmedberg 2007-08-13 19:37:36 +00:00
dbaron%dbaron.org fcb9984a4e Remove nsTraceRefcnt::LoadLibrarySymbols, which is no longer needed due to the error handling in SymGetModuleInfoEspecial. b=391793 r+a=bsmedberg 2007-08-13 19:36:57 +00:00
ginn.chen%sun.com 367563acf6 Bug 374689 Fix build issues 
r+a=dbaron
 2007-08-13 05:29:55 +00:00
dbaron%dbaron.org d834a42cdd Convert trace-malloc to using XPCOM stack walking API. b=374829 r+a=brendan 2007-08-10 22:20:49 +00:00
dbaron%dbaron.org d9d01ad030 Split the address->symbol mapping from the rest of the stack walking code. b=374829 r=bsmedberg a=bzbarsky 2007-08-10 21:32:50 +00:00
dbaron%dbaron.org 322b9b6ea1 Export a stack walking API (usable from C or C++) from XPCOM. b=374689 r=bsmedberg a=bzbarsky 2007-08-10 21:32:49 +00:00
bzbarsky%mit.edu 72e242e9e6 Add a stack-and-abort value of XPCOM_DEBUG_BREAK 2007-08-10 19:42:58 +00:00
benjamin%smedbergs.us bbb6d901f3 Bug 386025 - Expose cycle-collection symbols, r=graydon 2007-08-06 14:34:05 +00:00
dcamp%mozilla.com 70e3c46599 Add an API to access <input type="file"> data. b=371432, r=smontagu (for charset stuff), r+sr=jonas 2007-07-25 04:29:49 +00:00
dbaron%dbaron.org 7c256de140 Explain another possibility in this debugging printf. DEBUG_CC only. 2007-07-14 19:06:10 +00:00
neil%parkwaycc.co.uk 11d0936e48 Followup to bug 348748 to fix the alignment of trailing backslashes in macros 
[xxx_cast<foo*>(bar) is three characters shorter than NS_XXX_CAST(foo*, bar)]
 2007-07-11 08:46:49 +00:00
dbaron%dbaron.org 35f82f7d85 Make ExplainLiveExpectedGarbage traverse the suspectCurrent roots, since now that it's more accurate, more graph is always better. b=387005 r=graydon 2007-07-08 19:26:44 +00:00
jwalden%mit.edu ef68fcf595 Bug 348748 - Replace all instances of NS_STATIC_CAST and friends with C++ casts (and simultaneously bitrot nearly every patch in existence). r=bsmedberg on the script that did this. Tune in next time for Macro Wars: Episode II: Attack on the LL_* Macros. 2007-07-08 07:08:56 +00:00
dbaron%dbaron.org e76eaa9f3e Make Fault give more information, #ifdef DEBUG_CC. b=386947 r=graydon 2007-07-05 22:38:38 +00:00
dbaron%dbaron.org e3c7499467 In ExplainLiveExpectedGarbage, print only the externally-referenced nodes from which the expected garbage is reachable. b=387005 r=graydon 2007-07-05 22:20:16 +00:00
dbaron%dbaron.org 744bfc7a0c Make suspectCurrent assertion not fire constantly after we Fault. b=386664 r+sr=peterv 2007-07-03 21:21:57 +00:00
dbaron%dbaron.org 17afb32c85 Change cycle collector aging constant from 10 to 0. b=381199 r+sr=peterv 2007-06-27 20:41:26 +00:00
dbaron%dbaron.org ec61f386d7 Fix gcc warning about casting string literal to char*, DEBUG-only. 2007-06-27 20:40:52 +00:00
dbaron%dbaron.org 7c54a43209 Keep the purple bit in sync with the purple buffer when suspect and forget calls are dropped on the floor. b=385548 r=graydon 2007-06-27 17:51:48 +00:00
dbaron%dbaron.org bd26de47a7 Back out 385548 to try to fix test orange. 2007-06-27 05:26:39 +00:00
dbaron%dbaron.org bcef420e0b Don't set the purple bit when suspect calls are dropped on the floor. b=385548 r=graydon 2007-06-27 04:47:40 +00:00
dbaron%dbaron.org ff5099f1c4 Print the warning message for failure to suspect or unlink only for the objects that are the roots of what would have been collected. b=385549 r=graydon 2007-06-25 20:36:24 +00:00
dbaron%dbaron.org 900577f8aa Back out again. 2007-06-22 00:30:41 +00:00
dbaron%dbaron.org 0ce6e8791b See if using mScanDelay of 1 instead of 0 fixes the leak regression. 2007-06-21 23:49:37 +00:00
dbaron%dbaron.org a8ccb62785 Change cycle collector aging constant from 10 to 0. b=381199 r+sr=peterv 2007-06-21 23:02:53 +00:00
graydon%mozilla.com 1314e6df62 Bug 384942, dump graph on cycle collection fault. 
r=dbaron
 2007-06-19 01:29:10 +00:00
dbaron%dbaron.org 3bacfd472a Back out bug 381199. 2007-06-14 00:12:16 +00:00
dbaron%dbaron.org 409af12b1c Change cycle collector aging constant from 10 to 0. b=381199 r+sr=peterv 2007-06-13 23:22:19 +00:00
peterv%propagandism.org 5aa4486015 Fix for bug 383651 (Re-entrancy into nsCycleCollector::Collect can result in a crash). Patch by ak.miller@auckland.ac.nz, r=graydon, sr=dbaron. 2007-06-08 10:17:21 +00:00
dbaron%dbaron.org c78656f5ea Don't loop through all the possible shutdown collections when we suspect all native wrappers. b=383234 r+sr=peterv 2007-06-06 22:09:00 +00:00
dbaron%dbaron.org bfe9d47793 Shutdown in one cycle when we set mScanDelay to 0. b=383234 r+sr=peterv 2007-06-06 22:08:25 +00:00
dbaron%dbaron.org 236d50d667 strdup when allocating mName so we have useful pointers. b=383234 r+sr=peterv 2007-06-06 22:07:29 +00:00
dbaron%dbaron.org 95cf743124 Print info about things that are suspected on the extra collections at shutdown to catch deficiencies in traversal that cause us to need the extra collections. b=383234 r+sr=peterv 2007-06-06 22:06:56 +00:00
dbaron%dbaron.org 7c337eba46 When explaining why expected garbage was not collected, list the known references when some were unknown. b=383234 r+sr=peterv 2007-06-06 22:06:00 +00:00
dbaron%dbaron.org e736230c4a Make nsTraceRefcnt print addresses along with serial numbers, and fix warning about casting to a function pointer. b=383234 r=peterv 2007-06-06 22:05:25 +00:00
dbaron%dbaron.org b4abaa24fe Make DEBUG_CC compile again. b=383234 r+sr=peter 2007-06-06 22:04:26 +00:00
peterv%propagandism.org eea1671eb5 Fix warning. 2007-05-25 10:57:38 +00:00
peterv%propagandism.org 9eca95ccdc Fix for bug 368774 (Make cycle collector work with refcounted non-XPCOM objects). r=dbaron, sr=sicking. 2007-05-24 14:10:04 +00:00
jonas%sicking.cc 360162166e Relanding bug 378987. Don't try to collect windows, documents or elements of actively viewed pages. r=bz/jst sr=jst 2007-05-10 20:21:13 +00:00
jst%mozilla.org 6cd05e346d Backing out Jonas' last changes due to tinderbox orange, per Jonas' request. 2007-05-10 00:42:33 +00:00
jonas%sicking.cc 917c6cc079 Relanding bug 378987. Don't try to collect windows, documents or elements of actively viewed pages. r=bz/jst sr=jst 2007-05-09 23:33:29 +00:00
jonas%sicking.cc c21c489595 back out last patch since leaks went through the roof 2007-05-08 01:03:08 +00:00
jonas%sicking.cc 9652a963bd Don't attempt to cycle collect documents, windows and elements that are currently being viewed. r=bz, sr=jst b=378987 2007-05-07 23:45:26 +00:00
dbaron%dbaron.org 4c091a4965 Remove extra class names to address Neil's comments. b=378514 2007-05-03 23:07:26 +00:00
dbaron%dbaron.org c663e7b961 Record results of cycle collection traversal so we only have to call the traversal APIs once per collection. b=378514 sr=peterv r=graydon 2007-04-25 21:12:11 +00:00
peterv%propagandism.org 1db166f2c6 Fix for bug 376854 (CycleCollector depends on uninitialized values (ageSelectionCallback and nsPurpleBuffer::BumpGeneration)). r/sr=dbaron. 2007-04-25 18:33:55 +00:00
peterv%propagandism.org d0d256596c Fix for bug 378595 (Improve cycle collection QI performance). r/sr=sicking. 2007-04-25 16:35:31 +00:00
dbaron%dbaron.org 4da3802f28 Followup fixes to making cycle collector use more efficient hashtables: shrink size of table and fix warnings on 64-bit machines. b=377606 sr=peterv r=graydon 2007-04-23 23:34:33 +00:00
peterv%propagandism.org e0337217d2 Fix for bug 372110 (Make cycle-collection debugging features optional at compile time). r=graydon, sr=jst. 2007-04-21 17:43:19 +00:00
peterv%propagandism.org b9af28af5a Reorder members to make struct smaller and try to lower MH down again. 2007-04-20 08:51:31 +00:00
peterv%propagandism.org 93712e056d Fix for bug 377606 (Switch cycle collector to more efficient hashtables). Patch by graydon, r=peterv, sr=dbaron. 2007-04-20 08:01:01 +00:00
peterv%propagandism.org 1395012304 Fix for bug 373693 (Cycle collector makes AddRef about 6x slower). r=graydon, sr=dbaron. 2007-04-18 10:10:37 +00:00
peterv%propagandism.org 727a2d25e8 Fix for bug 372713 (Add cycle collection to RDF datasources). r=bsmedberg, sr=dbaron. 2007-04-10 22:05:44 +00:00
peterv%propagandism.org 84e7da2624 Fix for bug 375075 (Stop leaking the cycle collector on shutdown). r=bsmedberg, sr=dbaron. 2007-03-27 09:49:06 +00:00
peterv%propagandism.org d020bc5120 Backing out change for debugging tinderbox orange. 2007-03-23 16:10:21 +00:00
peterv%propagandism.org af97f9c441 Temporarily enable cycle collection timing for debug builds to figure out tinderbox orange. Will back out after qm-xserve01 tinderbox has cycled. 2007-03-23 15:11:53 +00:00
jst%mozilla.org b04f820608 Fixing bug 374239. Add debugging code that will print time spent in various parts of nsCycleCollector::Collect(). r+sr=peterv@propagandism.org 2007-03-19 23:21:31 +00:00
peterv%propagandism.org 09b00a6b2d Fix for bug 374096 (Cycle collector doesn't collect all the cycles it could). r=graydon, sr=dbaron. 2007-03-16 12:52:47 +00:00
graydon%mozilla.com ff918fa55c Bug 368869, previous commit had a typo: s/bool/PRBool/, s/false/PR_FALSE/. 2007-03-15 23:47:01 +00:00
graydon%mozilla.com d474d7e1b4 Bug 368869 - Make cycle collection suspect all native wrapper roots 
r=brendan, sr=jst
 2007-03-15 23:05:03 +00:00
joshmoz%gmail.com 2f5d37badf XPCOM cycle collector is initialized statically, don't do that. b=369336 r=graydon sr=vlad 2007-03-14 06:15:10 +00:00
dbaron%dbaron.org adcc767817 Add debugging code to cycle collector to give information about objects it has been told ought to be freed. b=372766 r=graydon sr=peterv 2007-03-09 23:00:14 +00:00
peterv%propagandism.org 21bb9cfe9c Take care of reviewer comment (bug 373219). 2007-03-09 13:49:45 +00:00
peterv%propagandism.org 9ed7625526 Fix for bug 373219 (ASSERTION: Fault in cycle collector: null XPCOM pointer returned (ptr: 0)). r/sr=dbaron. 2007-03-09 13:14:06 +00:00
peterv%propagandism.org 76fe81f310 Try to fix bustage. 2007-03-08 11:31:14 +00:00
peterv%propagandism.org b0d71960ba Fix for bug 368773 (Add a bunch of DOM objects to cycle collection). r=sicking, sr=jst. 2007-03-08 11:17:16 +00:00
bzbarsky%mit.edu 22392fa747 Relanding bug 368523 now that we've established it's the reason for the Txul change. 2007-03-06 19:25:10 +00:00
bzbarsky%mit.edu 2b907fbe12 Back out bug 368523 to test performance impact 2007-03-06 18:23:15 +00:00
dbaron%dbaron.org 01dced5738 Make nsCycleCollector::Fault do something in debug builds. b=372320 r=graydon sr=peterv 2007-03-05 21:07:51 +00:00
bzbarsky%mit.edu d2efe4be57 Don't hook malloc by default (still hook it if asked to via the env vars), and 
when it's hooked ignore free(0) calls.  Bug 368523, r=graydon, sr=dbaron
 2007-03-02 01:08:03 +00:00
peterv%propagandism.org 13bf3e89cd Fix for bug 367779 (Make cycle collection through JS objects more reliable). r=jst, sr=brendan. 2007-02-18 20:05:32 +00:00
peterv%propagandism.org 7c0a0af3b0 Fix for bug 368549 (Cycle collector doesn't handle tearoffs correctly). r=graydon, sr=jst. 2007-02-18 14:38:04 +00:00
jonas%sicking.cc 9eea4cd5fe Bug 370332: Kill nsIBindingManager and nsIStyleRuleSupplier. r=dbaron for the layout/style parts, r/sr=jst for the rest. 2007-02-16 23:02:08 +00:00
gerv%gerv.net e0e8fe16df Bug 236613: change to MPL/LGPL/GPL tri-license. 2007-02-14 17:33:38 +00:00
philringnalda%gmail.com f780e390c6 Bug 349526: BeOS crashes at startup due to mState buffer overrun, p=tqh <thesuckiestemail@yahoo.se>, r+sr=vlad 2007-01-27 05:11:35 +00:00
jst%mozilla.org 5ff3c88f95 Landing graydon's fix for bug 366578. Make the cycle collector run at shutdown. 2007-01-25 00:24:08 +00:00
martijn.martijn%gmail.com 6e5d945b23 Bug 366119 � Mingw build error in nsCycleCollector.cpp, r=graydon, sr=jst 2007-01-23 00:39:25 +00:00
graydon%mozilla.com 5b07354fbe Bug 366063: better fix to critical crasher, replacing band-aid submitted in 333078. 
r=dbaron
 2007-01-09 01:33:02 +00:00
jst%mozilla.org 0817fed458 Landing 64-bit fixes for bug 333078. r=graydon@mozilla.com, sr=bugmail@sicking.cc 2007-01-05 23:59:18 +00:00
graydon%mozilla.com 89cc937929 Bug 333078: XPCOM cycle collector, first performance-related followup patch. 
Reorganizes collection throttling to coincide with existing JS_GC scheme.
 2007-01-05 01:44:42 +00:00
graydon%mozilla.com e6fbf7f335 Bug 333078: XPCOM cycle collector 
r=brendan, sicking, bz, dbaron, bsmedberg
 2007-01-04 22:31:26 +00:00
vladimir%pobox.com df954f6171 backing out bug 333078, xpcom cycle collector due to perf issues 2006-11-22 23:18:16 +00:00
vladimir%pobox.com c53729f4c5 b=333078, xpcom cycle collector; followup patch -- patch from graydon@mozilla.com, r=many 2006-11-22 21:31:33 +00:00
vladimir%pobox.com 6044ffd026 b=333078, xpcom cycle collector; followup patch -- patch from graydon@mozilla.com, r=many 2006-11-22 20:48:47 +00:00
vladimir%pobox.com fdf72a2c5b b=333078, xpcom cycle collector -- patch from graydon@mozilla.com, r=many 2006-11-22 19:23:02 +00:00
gavin%gavinsharp.com 47e1c35444 Bug 342670 2006-11-22 17:50:35 +00:00
cvshook%sicking.cc 5c7ed7c2dc Don't return nsDerivedSafe when .get() is explicitly used on nsCOMPtr and nsRefPtr. r/sr=dbaron 2006-11-21 21:28:14 +00:00
mats.palmgren%bredband.net 57cc88bce4 Make NS_LIKELY/NS_UNLIKELY accept pointers etc and always return 0 or 1. b=340244 r+sr=darin 2006-11-18 00:48:56 +00:00
benjamin%smedbergs.us 4543c02b83 Bug 313309 - Table-driven QI part 3 - use weak statics for interface IDs, r=shaver. This will break VC6 and gcc2.9x, which is ok 2006-11-14 21:21:37 +00:00
mats.palmgren%bredband.net cfb7879dc8 Backing out the NS_LIKELY/NS_UNLIKELY change to see if it fixes Tinderbox orange. b=340244 2006-11-09 22:52:19 +00:00
mats.palmgren%bredband.net 038bf019d7 Make NS_LIKELY/NS_UNLIKELY accept pointers etc. b=340244 r+sr=darin 2006-11-09 17:41:03 +00:00
vladimir%pobox.com ba3c1f2953 b=354984, patches to build with vista RC1 SDK, r=darin 2006-10-24 20:38:29 +00:00
cbiesinger%web.de 33e35afd71 bug 348748 make NS_*_CAST always use C++-style casts (in preparation for 
removing the macros altogether)
r=bsmedberg sr=dbaron
 2006-09-04 23:04:57 +00:00
vladimir%pobox.com 8827e3f637 b=327161, real and final fix for btek's orange in uuidgenerator glibc brokenness, r+sr=bz 2006-08-15 00:52:21 +00:00
vladimir%pobox.com 7126f1a5d8 b=327161, uuidgenerator stuff, attempt to fix btek from previous checkin, r=bz 2006-08-14 22:04:01 +00:00
vladimir%pobox.com eb040ffefa b=327161, nsUUIDGenerator subject to reseeding woes, r+sr=bz 2006-08-14 18:53:34 +00:00
benjamin%smedbergs.us 66d7b11b39 Bug 305949 - stop exporting nonfrozen XPCOM symbols when libxul is enabled, r=darin 2006-08-02 18:27:25 +00:00
gavin%gavinsharp.com f5c3506ad3 Bug 322169: Clear Private Data does not clear JS Console, patch by Simon B�nzli <zeniko@gmail.com>, r=mconnor, r=darin 2006-07-19 12:49:33 +00:00
sspitzer%mozilla.org e1ed597e97 fix for bug #341190 
Software update should forward information about specific operating system version.

r=darin
 2006-07-06 06:32:03 +00:00
mark%moxienet.com 34b5a77a24 Copy'n'paste (340071) and drag'n'drop (340890) between Mozilla app running natively and Mozilla app running under Rosetta translation does not work [properly]. Byte-swap UTF-16 text in private flavors when running under Rosetta. r=josh sr=pink a/1.8.1=me 2006-06-15 17:13:59 +00:00
bzbarsky%mit.edu 9968fe3d8b Make nsUUIDGenerator use the low bits, not the high ones, so we actually get 
random numbers on 64-bit systems.  Bug 335549, r=vlad, sr=roc
 2006-06-15 04:40:21 +00:00
mhammond%skippinet.com.au d5ad1dc2b9 Land DOM_AGNOSTIC3_BRANCH, bug 255942. r=a few people, sr=brendan. 2006-06-13 03:07:47 +00:00
bent.mozilla%gmail.com d77e79e3cb Bug 289689 - "Memory leak: nsDebugImpl::Assertion, Create Process". r=dougt, sr=darin. 2006-06-02 23:37:10 +00:00
joshmoz%gmail.com f081589556 add a comment for my last checkin, b=339987 2006-06-02 20:20:52 +00:00
joshmoz%gmail.com cd658ed6ba Make Mac OS X debug trap impl use raise(SIGTRAP) instead of Mac OS X debugging function so we're only required to set the XPCOM debug env variable and not also the Mac OS X debug break variable. Also make GNUC/x86 test more robust, include '__i386__' in the test because not all platforms define '__i386' (Mac OS X is one of those, though it doesn't really matter here since we do a separate test for Mac OS X). b=339987 r=mento sr=dougt 2006-06-02 20:06:43 +00:00