Граф коммитов

346 Коммитов

Автор SHA1 Сообщение Дата
bsmedberg%covad.net ad6994b382 Bug 319999 - XPCOM Shutdown: xpcom-shutdown-threads notification and cleanup of event queues before component manager shutdown, r=darin 2006-01-17 15:41:40 +00:00
dbaron%dbaron.org 8e1263ab28 Make PL_InitEvent callers stop doing potentially dangerous casting of function pointers. b=305278 r=dougt sr=darin 2005-11-14 20:09:51 +00:00
dbaron%dbaron.org 51ae9d48e4 Fix bad advice in comment on how to use PL_InitEvent. b=305278 r=dougt 2005-11-12 21:11:33 +00:00
bsmedberg%covad.net dff1dea638 Bug 313309 part 2 - change the NS_DEFINE_STATIC_IID_ACCESSOR macro to NS_DECLARE, and make a NS_DEFINE_ macro that's outside the class declaration, r=shaver 2005-11-11 14:36:26 +00:00
bsmedberg%covad.net 1172c33dfc Bug 305949 Part 1 - Separate the frozen XPCOM exports with the XPCOM_API macro; add compatibility methods to the frozen string API; convert most of the tests/tools to use the xpcom glue when appropriate r=darin This checking does not actually stop exporting the nonfrozen methods yet, pending a bunch of test cleanup throughout the tree. 2005-11-08 18:17:49 +00:00
bzbarsky%mit.edu 8343828bd5 Make timers a little safer to use from JS. Bug 304669, patch by Giorgio Maone
<g.maone@informaction.com>, r=dougt, sr=dbaron
2005-11-02 00:46:37 +00:00
roc+%cs.cmu.edu f8cddec44f Backing out fix for bug 303484 since it exposed some other issues and doesn't fix any known crashers. 2005-10-18 01:41:11 +00:00
timeless%mozdev.org d60f1862ce Bug 310487 thread leak
patch by colin@theblakes.com r=timeless sr=darin
2005-09-30 20:46:14 +00:00
bsmedberg%covad.net 1d7b0362c3 Bug 307168 - Use the XPCOM (NS_COM) declarations for plevent symbols, instead of the NSPR declarations, since NSPR has different configure tests and compile flags r=darin 2005-09-15 18:29:21 +00:00
timeless%mozdev.org b642130d40 Bug 308404 Access violation [@ ntdll!RtlDeleteCriticalSection+0x0000000f]
r=darin sr=darin
2005-09-14 18:18:43 +00:00
dougt%meer.net 1ebff7a81a WinCE only. ShellExecute is now in the shunt lib, so I can remove a WINCE #define. Also, for process creation WinCE uses NSPR like a good child.... i wonder why the other platforms don't as well. 2005-09-02 18:00:23 +00:00
roc+%cs.cmu.edu 56b0974569 Bug 303484. Limit the nesting of modal event loops so we avoid overflowing the C++ stack. r+sr=brendan,a=asa 2005-08-16 20:59:39 +00:00
dougt%meer.net e0ae533dc4 Disabling USE_TIMER on windows ce. posting directly to the message queue seams to be far more effective. a=asa. WINCE only change 2005-07-26 04:22:11 +00:00
joshmoz%gmail.com 4be2400544 part 2 of landing for bug 282940. actually make the switch to CFRunLoop. a=asa 2005-06-21 21:04:37 +00:00
joshmoz%gmail.com fd84d70bfc part 1 of landing for bug 282940 (move FF and TB to CFRunLoop) - simply clean up old Mac OS classic code. a=asa 2005-06-21 20:43:11 +00:00
timeless%mozdev.org 8e30ab2aea Bug 296506 [@ nsThread::Main]
change Init to handle alloc failures and refuse multiple concurrent runs of a thread instance
r=dougt a=asa
2005-06-08 07:33:39 +00:00
bryner%brianryner.com 592dbdc176 Add support for caching content viewers in session history to speed up back/forward (bug 274784). This initial landing has the feature disabled by default; set browser.sessionhistory.max_viewers to the maximum number of pages to cache to enable the feature. r=bzbarsky, sr/a=brendan. 2005-05-04 20:22:32 +00:00
timeless%mozdev.org 69e8e54ec3 Bug 291988 build warnings in plevent.c
patch by bugzilla@gemal.dk r=alecf sr=darin a=asa
2005-04-29 06:09:42 +00:00
bzbarsky%mit.edu 12781e0d34 Fix revokeEvents to reliably revoke them. Bug 284389, r=darin, sr=dbaron,
a=asa
2005-04-15 03:17:13 +00:00
bsmedberg%covad.net 20b57929a4 Bug 288626 - Begin switch from MOZILLA_STRICT_API to MOZILLA_INTERNAL_API rs=darin for the Makefile var additions. Should not affect the build yet, until we flip a few switches at the end. 2005-04-01 18:22:57 +00:00
brendan%mozilla.org b227540370 Propagate failure when we can't make a queue (287846, r=dougt, sr=darin). 2005-03-29 22:31:14 +00:00
darin%meer.net 226a6d53db fixes bug 279852 "nsWeakReference not threadsafe assertions due to TimerThread::Init" r=sfraser sr=bryner 2005-03-14 06:37:01 +00:00
timeless%mozdev.org af00efe42b Bug 283429 javascript undefined object -> crash in nsIProcess::Init
r=dougt
2005-02-24 18:41:50 +00:00
darin%meer.net 14bc4f011a s/BumpPriority/AdjustPriority/ (bug 278531) r=shaver sr=brendan 2005-02-21 22:21:21 +00:00
darin%meer.net 6d8ab11f00 build the header file!! 2005-02-21 21:08:46 +00:00
darin%meer.net f2311e1369 fixes bug 278531 "generic request prioritization (loadgroup prioritization)" r=biesi sr=bzbarsky 2005-02-21 20:58:01 +00:00
dougt%meer.net 4546ecd25b bug 282031. [BEOS] Our nsToolkit::CallMethod is broken. r=sergei_d, sr=dougt 2005-02-19 18:41:33 +00:00
dougt%meer.net d258f3e99d bug 280245. WinCE build and support changes. r=darin 2005-01-30 21:33:47 +00:00
timeless%mozdev.org fc1db9810d Bug 270107 nsProcess::Run is broken
r=dougt
2005-01-14 11:52:09 +00:00
smfr%smfr.org 1c53cc203b Commit the CFRunLoop patch from bug 271050 for plevent handling. This uses a CFRunLoopSource to process plevents from the applications's main event loop. r=pinkerton. 2005-01-04 02:02:01 +00:00
smfr%smfr.org eeab113656 More experimentation with plevent handling for Camino (only affects cocoa toolkit), replacing the CFRunLoop patch with the CFMessagePort patch for performance comparison. 2004-12-29 16:40:41 +00:00
smfr%smfr.org 07794c7d06 Checking in CFRunLoop patch in bug 271050 to test pageload performance in Camino. r=pinkerton 2004-12-29 06:07:31 +00:00
cbiesinger%web.de 41f430dca0 Bug 274281 [BeOS] BeOS lets events stay in port through the 'restarts' in LaunchChild causing crashes on faulty events.
patch by tqh <thesuckiestemail@yahoo.se> r=sergei_d sr=shaver (for the xpcom/ part)
2004-12-28 13:56:17 +00:00
darin%meer.net 8d1164cb8c fixes bug 273819 "ASSERTION: Native event queues should only be used on the main thread" r=danm, sr=bienvenu 2004-12-17 19:47:06 +00:00
jst%mozilla.jstenback.com 8d53dd0302 Fixing bug 273961. Don't execute unix process creation code on windows. r+sr=darin@meer.net 2004-12-10 01:06:37 +00:00
bsmedberg%covad.net 7ccf6e4965 Bug 273876 - libxul step 2 (everything through widget, except spidermonkey) r=darin; again, this should not affect non-xulrunner trees. 2004-12-09 19:28:35 +00:00
darin%meer.net d877d3daef alternate fix for bug 234620 "Unknown random SEGV/seg fault/core dumps/crashes, only thing on is Mail/IMAP"
in an attempt to avoid crash bug 269585, r=dbaron sr=bienvenu a=asa
2004-11-19 19:06:58 +00:00
jst%mozilla.jstenback.com 83953b0991 Fixing bugstage. 2004-11-11 20:48:45 +00:00
jst%mozilla.jstenback.com 471617b8f8 Fixing bug 267269. Make nsIProcess::Run() use NSPR code for running the process if arguments are passed. r=pinkerton@aol.net, sr=bzbarsky@mit.edu 2004-11-11 19:42:26 +00:00
darin%meer.net 1fd6b4e56b fixes bug 234620 "Unknown random SEGV/seg fault/core dumps/crashes, only thing on is Mail/IMAP" r=dbaron sr=bienvenu 2004-11-04 02:34:11 +00:00
mkaply%us.ibm.com cdeb6187d6 OS/2 build bustage - don't ask 2004-10-25 20:45:17 +00:00
jst%mozilla.jstenback.com 50ced49104 Fixing problem where mExitValue is unconditionally -1 on the mac after running a process. No bug. r+sr=dbaron@dbaron.org, a=asa@mozilla.org 2004-10-22 23:55:56 +00:00
jst%mozilla.jstenback.com f242538f08 Fixing bug 263429. Making nsProcessCommon::Run() work on Mac OS X. r=peterv@propagandism.org, sr=dveditz@cruzio.com 2004-10-12 05:31:37 +00:00
brendan%mozilla.org 0839a484e6 Retract last change, it's wrong -- thanks again to lcook@sybase.com. 2004-09-22 02:37:49 +00:00
brendan%mozilla.org 6f067fcd44 Followup to last checkin (252324, r+sr=bz). 2004-09-22 01:04:33 +00:00
brendan%mozilla.org 353221b100 Patch from Larry Cook <lcook@sybase.com> to fix long-delay timer queue order bug (252324, r+sr=me). 2004-09-21 01:11:47 +00:00
timeless%mozdev.org a5c885b9b3 Fixing nondebug mAddr is a debug paranoia thing. 2004-09-10 06:02:07 +00:00
timeless%mozdev.org 85b8339503 Bug 252733 ###!!! ASSERTION: Potential deadlock between Monitor@12f0a8 and Monitor@0: 'Error', file r:/mozilla/xpcom/threads/nsAutoLock.cpp, line 299
Ignore null monitors - they aren't real and xpconnect promises not to enter them so you can't possibly reach a deadlock involving them.
r=brendan sr=brendan
2004-09-10 05:13:38 +00:00
bsmedberg%covad.net 21c11cf95c bug 257188 - check initialization of nsThread patch by CTho <cst@andrew.cmu.edu> r=dougt 2004-08-28 20:11:15 +00:00
timeless%mozdev.org 98bf634481 Bug 256615 crash when calling nsIThread.state before calling nsIThread.init [@ PR_GetThreadState]
patch by cst@andrew.cmu.edu r=dougt
2004-08-26 12:31:56 +00:00