Commit Graph

4184 Commits

Author SHA1 Message Date
Nicholas Nethercote
b38b84f869 Bug 925584 - Remove some unnecessary jsapi.h inclusions from .cpp files. r=Ms2ger.
--HG--
extra : rebase_source : 41fcb0e922a519ef679c1c1b6293c2b638e83a48
2013-10-10 15:22:35 -07:00
Phil Ringnalda
0f6fec1509 Back out f872d288480b:9b86b4e60b29 (bug 929513) for failing to build on Windows
CLOSED TREE
2013-10-27 15:38:40 -07:00
David Zbarsky
645cba6965 Bug 929513 Part 11: Convert ImageLayer's mScaleToSize to use LayerIntSize r=nical 2013-10-27 17:53:29 -04:00
Phil Ringnalda
6f8ea3541e Back out 338760b1955c (bug 902922) for discovering that we have a cycle collecter OOM 100% of the time in Win7 mochitest-2 shutdown 2013-10-27 09:38:35 -07:00
Benoit Jacob
c46d04715d Bug 902922 - OOM assertions in the cycle collector should be fatal in debug builds - r=mccr8 2013-10-27 07:51:31 -04:00
Ed Morley
4c8f0397a2 Backed out changeset 82d5d92a3eac (bug 915555) for build failures on a CLOSED TREE 2013-10-25 17:59:24 +01:00
Jeff Walden
a130aa5512 Bug 915555 - Don't redundantly cast an unnamed expression to an rvalue reference, because 1) it's dumb, and 2) gcc 4.4 warns about casting an rvalue reference to an rvalue reference (which is harmless due to C++11's rvalue semantics). f=hub, r=trivial 2013-09-12 10:55:14 -07:00
Ed Morley
d339853f4e Backed out changeset 2d6c5cbcc213 (bug 902587) for failures during linking on a CLOSED TREE
--HG--
rename : xpcom/build/PoisonIOInterposer.h => xpcom/build/mozPoisonWrite.h
rename : xpcom/build/PoisonIOInterposerBase.cpp => xpcom/build/mozPoisonWriteBase.cpp
rename : xpcom/build/PoisonIOInterposerMac.cpp => xpcom/build/mozPoisonWriteMac.cpp
rename : xpcom/build/PoisonIOInterposerWin.cpp => xpcom/build/mozPoisonWriteWin.cpp
2013-10-25 14:24:34 +01:00
Cykesiopka
e0bcd6aebc Bug 914270 - Part 1: Simple/Automated moves. r=joey 2013-10-24 18:51:00 +01:00
Jonas Finnemann Jensen
25dc345fdc Bug 902587 - Part 2A: Refactor late-write-checks as client of IO Interposer r=BenWa
--HG--
rename : xpcom/build/mozPoisonWrite.h => xpcom/build/LateWriteChecks.h
rename : xpcom/build/mozPoisonWrite.h => xpcom/build/PoisonIOInterposer.h
rename : xpcom/build/mozPoisonWriteBase.cpp => xpcom/build/PoisonIOInterposerBase.cpp
rename : xpcom/build/mozPoisonWriteMac.cpp => xpcom/build/PoisonIOInterposerMac.cpp
rename : xpcom/build/mozPoisonWriteWin.cpp => xpcom/build/PoisonIOInterposerWin.cpp
2013-10-24 17:02:00 +01:00
Carsten "Tomcat" Book
14a3507643 Merge Mozilla-Inbound to Mozilla-Central 2013-10-25 10:12:00 +02:00
Karl Tomlinson
870bbe7dd2 back out 242bb2279283 from bug 928222 on suspicion of causing bugs 930495 930479 930428 2013-10-25 17:40:39 +13:00
Mike Hommey
f1c0c07862 Bug 929905 - Consolidate sources in moz.build. r=gps 2013-10-25 08:23:05 +09:00
Mike Hommey
80c623bb6a Bug 921492 - Make StrictOrderingOnAppendList use actual alphabetical sorting. r=mshal,r=gps 2013-10-24 08:05:43 +09:00
Mike Hommey
5072f65f8c Bug 913268 - Make CPP_SOURCES a StrictOrderingOnAppendList. r=mshal 2013-10-24 08:00:23 +09:00
Andrew McCreight
ab63e479c1 Bug 928647 - Remove CycleCollectedRuntime::mObjectToUnlink. r=smaug 2013-10-19 08:59:10 -07:00
Karl Tomlinson
7356a80cad b=928222 remove nsThreadPool per-thread event queues r=bsmedberg
The way idle nsThreadPool threads wait with a timeout doesn't work well with
shutdown from nsThreadManager.

nsThreadPool doesn't need to use nsIThread for its threads.  The nsIThread
interface is not useful for threads running in an nsThreadPool.
The nsIEventTarget on the nsIThreadPool should be used for dispatching events,
not an interface on the individual threads, and the threads don't need an
nsEventQueue because they use the nsEventQueue on the nsThreadPool.

Shutdown of single event threads is easier than running nested event loops for
nsIThreads, avoiding the multilevel nested event loop situations when several
threads finish and are shutdown.  While the ThreadFunc is running, a nested
event loop is still used in Shutdown() in case some consumers might need it
and because that is the documented API.

This also simplifies thread creation, avoiding races that could mean there was
temporarily an extra thread or more.

--HG--
extra : transplant_source : %F7%14%16%12%EF%E9%84%19D%26%3C%FE%1F%EC%FF%A3%BAG%C4%F3
2013-10-24 09:31:15 +13:00
Karl Tomlinson
479f466a6d b=928222 move ReentrantMonitorAutoExit from VideoUtils.h to ReentrantMonitor.h r=bsmedberg
--HG--
extra : transplant_source : %13O%CF%90x%02%B8%B6%A3%EA%AD%03%99%9A7%1B%A0%21%C4%A3
2013-10-24 07:22:53 +13:00
Karl Tomlinson
e914e6bb1a b=926522 remove unecessary shutdown code from ~nsThreadPool r=bsmedberg
While threads are processing their initial runnable, which is the nsThreadPool, it will not be
destroyed.  The threads remove themselves from mThreads before returning from Run().

This includes removal of changes from bug 900711.

--HG--
extra : transplant_source : %EF%E9%94L%E9d%F5%99b%09r%CB%024%86%88%BA%EFi%29
2013-10-24 07:21:37 +13:00
Nathan Froyd
66aef536b3 Bug 927551 - make darwin xptcall sources compilable in a singlepass universal build; r=bsmedberg,glandium 2013-10-16 15:25:36 -04:00
Nathan Froyd
7efcdc0a70 Bug 927427 - use __LP64__ in FileUtils.cpp for Linux; r=bsmedberg 2013-10-16 11:14:05 -04:00
Nicholas Nethercote
35059be3df Bug 929791 - Fix invalid JSON generation when dumping memory reports to file after viewing about:memory in the presence of child processes. r=mccr8. 2013-10-22 16:58:33 -07:00
Nicholas Nethercote
d58b8ff6f4 Bug 918207 - Support per-tab memory profiling. r=billm,mccr8,till,smaug,nfroyd.
--HG--
extra : rebase_source : 7603eaedbd761c6b231b1b911f719388864755e6
2013-10-22 14:53:26 -07:00
Nathan Froyd
fa82c2129e Bug 929494 - use template typedefs, not inheritance, to define nsTArray element copiers; r=ehsan
There's no reason to use inheritance here, and using plain typedefs avoids massive
amounts of code duplication for the common case of copying with mem*.  Code savings
on Android come in at about 570K (!), or ~2% of libxul .text size, which is a massive
win.
2013-10-22 10:36:34 -04:00
Nicholas Nethercote
5b8e9b8bf7 Bug 927705 (part 3) - Add a C++ unit test for pldhash. r=jorendorff.
--HG--
extra : rebase_source : 58cddfff7da6116af7512097c3a6172ca1c06b87
2013-10-21 15:44:03 -07:00
Nicholas Nethercote
bff6066f1d Bug 927705 (part 2b) - Never overload pldhash past 75% full, because performance plummets near the end. r=jorendorff.
--HG--
extra : rebase_source : 7d3a0594e49ceb89550861debca03a40d3e479c1
2013-10-21 15:44:01 -07:00
Nicholas Nethercote
091ff4359a Bug 927705 (part 2a) - Increase pldhash's max capacity from 1<<23 to 1<<26, and protect against uint32_t overflow when computing storage size. r=jorendorff.
--HG--
extra : rebase_source : 9293ec75e5ba8ad416d66ad6bf0c0788c0d35b02
2013-10-21 15:36:45 -07:00
Nicholas Nethercote
ff98ed6578 Bug 927705 (part 1) - Remove PL_DHashTableSetAlphaBounds() and the supporting machinery. r=jorendorff.
--HG--
extra : rebase_source : 0b97b3276ec91d417fcfc12fe871d93b0cbd2263
2013-10-20 20:17:48 -07:00
Mike Hommey
94e89c1def Bug 927775 - Cleanup xpcom/typelib/xpt/tools/Makefile.in. r=mshal 2013-10-22 10:02:49 +09:00
Benjamin Smedberg
918d9a5ff9 Bug 927944 part B - move AutoCriticalSection to nsWindowsHelpers.h and use it instead of manual lock handling in nsWindowsDllBlocklist.cpp, r=ehsan
--HG--
extra : rebase_source : 54e55b729263bffbb5a92c3336eacf248e99a659
2013-10-18 14:24:51 -04:00
Ms2ger
39b292cc30 Merge inbound to m-c. 2013-10-20 10:40:40 +02:00
Ms2ger
34421d0936 Bug 908142 - Part c: Move FAIL_ON_WARNINGS to moz.build in xpcom/tests/; r=gps 2013-10-20 09:25:19 +02:00
Ms2ger
88f73ee0bb Bug 923395 - Part b: Remove makefiles that only set LOCAL_INCLUDES and DEFINES; r=gps 2013-10-20 09:25:19 +02:00
Andrew McCreight
8cfa1d97d8 Bug 923486, part 2 - Define NS_IsMainThread out-of-line in ASAN builds. r=ehsan 2013-10-03 11:58:41 -07:00
Andrew McCreight
e9c045806a Bug 923486, part 1 - Harmonize ordering of definitions of NS_IsMainThread(). r=ehsan 2013-10-17 06:24:30 -07:00
Ehsan Akhgari
4a6090fdcb Bug 928434 - Fix the invalid format string in nsMemoryImpl::IsLowMemoryPlatform; r=bsmedgerg
--HG--
extra : rebase_source : 115fb93cb90a2004168032a4ad1d998a5dfb192d
2013-10-18 20:34:57 -04:00
Ehsan Akhgari
b02212d91b Bug 928040 - Remove some prtypes.h inclusions from xpcom/; r=bsmedberg
--HG--
extra : rebase_source : 2f238320ba8330a27aa986af1146446dcb5d7128
2013-10-18 20:34:57 -04:00
Masatoshi Kimura
11b1ec2573 Bug 925599 - Introduce version test functions using VerifyVersionInfo(). r=bbondy 2013-10-19 03:14:31 +09:00
Olli Pettay
129aa70c07 Bug 927813 - Some micro-optimizations for Suspect, r=mccr8 2013-10-17 17:05:13 +03:00
Ehsan Akhgari
9bf1bff8b4 Bug 895047 - Make char16_t available everywhere and use that to define PRUnichar and jschar; r=jcranmer,jorendorff,glandium 2013-10-17 00:36:13 -04:00
Phil Ringnalda
214fbe3acd Back out ee100983f921 (bug 895047) for build bustage
CLOSED TREE
2013-10-16 21:51:55 -07:00
Ehsan Akhgari
6f865c67f5 Bug 895047 - Make char16_t available everywhere and use that to define PRUnichar and jschar; r=jcranmer,jorendorff,glandium 2013-10-17 00:36:13 -04:00
Mike Hommey
5badab1842 Bug 926906 - Kill core_abspath, core_realpath and core_winabspath. r=gps 2013-10-17 07:55:16 +09:00
Narihiro Nakamura
e0ce23d5df Bug 899477 - Remove unused COMPONENT_DIRECTORY macro. r=bsmedberg 2013-10-15 13:54:10 -04:00
Gregory Szorc
c194b3b09b Bug 917753 - Add XPIDL files to clean target; r=glandium 2013-10-11 07:32:49 -07:00
Nathan Froyd
2fdb664d3b Bug 925173 - don't depend on HAVE_64BIT_OS in FileUtils.cpp; use the standard __LP64__ macro instead; r=bsmedberg
This point in the code is Mac-specific, so we are free to use Mac-specific macros
here.  Just use __LP64__ instead of HAVE_64BIT_OS.
2013-10-09 17:15:53 -04:00
Birunthan Mohanathas
3bd29c41c6 Bug 784739 - Switch from NULL to nullptr in xpcom/ (3/3); r=ehsan 2013-10-10 16:42:16 -04:00
Birunthan Mohanathas
3926b031ae Bug 784739 - Switch from NULL to nullptr in xpcom/ (2/3); r=ehsan 2013-10-10 16:41:39 -04:00
Birunthan Mohanathas
c09e07a17c Bug 784739 - Switch from NULL to nullptr in xpcom/ (1/3); r=ehsan 2013-10-10 16:41:00 -04:00
Benjamin Smedberg
d8185b20ac Bug 753687 - nsCategoryCache implementation doesn't free old category entries if their contract mapping is removed using .unregisterFactory. Store the factory objects directly in the map, instead of keeping both a map and a separate list. r=froydnj 2013-10-10 08:48:03 -04:00