Commit Graph

14189 Commits

Author SHA1 Message Date
Michael Layzell
0097c41e01 Bug 1201190 - Part 3: Mark every consumer of GUARD_OBJECT as MOZ_RAII, r=ehsan 2015-09-12 16:53:33 -04:00
Michael Layzell
1eaae658aa Bug 1156802 - Part 2: Remove all explicit move constructors, r=ehsan 2015-09-12 12:34:45 -04:00
Nathan Froyd
85d7b073ac Bug 1202828 - use nsEventQueue::HasPendingEvent in nsThread.cpp; r=mccr8
nsEventQueue's HasPending event is defined to simply:

  return GetEvent(false, nullptr);

So we can substitute HasPendingEvent for this particular GetEvent call
to make the code clearer.
2015-09-04 20:39:10 -04:00
Honza Bambas
3506e6ec97 Bug 1199775 - mozilla::Tokenizer improvements vol 2. r=nfroyd 2015-09-02 06:20:00 +02:00
Ehsan Akhgari
97be810099 Bug 1196430 follow-up: Hide the usage of gCodeAddressService behind #ifdef MOZ_STACKWALKING
This is needed to fix the build bustage on B2G Desktop Windows.
Landed on a CLOSED TREE
2015-09-11 00:06:40 -04:00
Nathan Froyd
89639e1fda Bug 1196430 - part 8 - use less reinterpret_cast in nsTraceRefcnt.cpp; r=mccr8 2015-08-26 22:57:32 -04:00
Nathan Froyd
3953358c0b Bug 1196430 - part 5 - dump allocation stacks for leaked objects in XPCOM_MEM_LOG_CLASSES; r=mccr8 2015-08-26 19:10:22 -04:00
Nathan Froyd
d6c3776152 Bug 1196430 - part 4 - record allocation stacks for classes in XPCOM_MEM_LOG_CLASSES; r=mccr8 2015-08-26 18:47:03 -04:00
Nathan Froyd
04974cb31c Bug 1196430 - part 3 - remove unnecessary nsString.h include from nsTraceRefcnt.cpp; r=mccr8 2015-08-26 18:28:35 -04:00
Nathan Froyd
e9b6ac22f9 Bug 1196430 - part 2 - give SerialNumberRecord a proper constructor; r=mccr8
Using C++ style is so much nicer for these sorts of things.  This change
also paves the way for easily using non-POD things in SerialNumberRecord.
2015-08-26 18:25:35 -04:00
Nathan Froyd
1acef79801 Bug 1196430 - part 1 - rename serialNumberRecord to SerialNumberRecord; r=mccr8
This is a small cleanup to make the class name conform to coding standards.
2015-08-26 18:24:13 -04:00
Ms2ger
ce50a0bf86 Bug 1194603 - Remove INTERNAL_TOOLS; r=mshal
Its only purpose is to disable PGO. Where that was not already explicitly done,
or irrelevant (because the directory only contains python), I disabled it in
moz.build.
2015-09-10 13:49:19 +02:00
Nikhil Marathe
cd9d0eb7ab Bug 1198230 - Respect FetchEvent.preventDefault(). r=jdm
Update web-platform-tests expected data

--HG--
extra : commitid : 4dM6kivKeAc
extra : rebase_source : 64e40258a98b79064d5073422811fb8e1f688735
2015-09-04 12:00:24 -07:00
Nicholas Nethercote
ffb1fefc9f Bug 1202526 (part 8) - Use PLDHashTable::RemoveEntry() in the cycle collector. r=mccr8.
This avoids repeating the hash table search in order to remove a CC graph
entry, which is good because it's a common operation.
2015-09-07 21:15:32 -07:00
Nicholas Nethercote
de8fceb34d Bug 1202526 (part 1) - Add PLDHashTable::RemoveEntry(). r=froydnj.
This patch also consolidates the shrink handling so it's now entirely within
ShrinkIfAppropriate().
2015-09-07 19:20:12 -07:00
Wes Kocher
ba9de6a2f7 Backed out changeset f5b2a11c64c9 (bug 1201271) for bustage 2015-09-09 15:56:21 -07:00
Andrew McCreight
3e590f4e78 Bug 1201271 - Warn about unused results for more methods of nsTArray. r=froydnj
This leaves alone the AppendElement methods.
2015-09-09 15:39:10 -07:00
Carsten "Tomcat" Book
c040626933 Merge mozilla-central to fx-team 2015-09-09 14:10:46 +02:00
Milan Sreckovic
9a9792e17d Bug 1128472 - Part 3. Vendor string on windows. r=aklotz 2015-09-03 13:10:00 +02:00
Milan Sreckovic
79db27187c Bug 1128472 - Part 2. Linux support. r=gfritzsche 2015-09-08 14:35:00 +02:00
Milan Sreckovic
c2e36204f0 Bug 1128472 - Part 1. Mac and Win for model, stepping, cores, cache, cpu speed; VM max on Win only, vendor on Mac only. r=gfritzsche 2015-09-01 14:48:00 +02:00
Randell Jesup
5e0e7e846c Bug 1197152: Alternative to remove all sleep/wake functionality from Timers r=froydnj
CLOSED TREE

--HG--
extra : amend_source : 8c363c570bea046904f2f9d7bf72b107012653d7
2015-09-08 11:34:09 -04:00
Nathan Froyd
444fb07360 Bug 1202667 - make TaskQueue task running slightly more efficient; r=mccr8
We can transfer the reference out of the queue of runnables instead of
taking a new reference right before we drop the old one.
2015-09-07 20:02:16 -04:00
Nathan Froyd
fbb4e33fce Bug 1202667 - make TaskQueue dispatching slightly more efficient; r=mccr8
We're already holding a reference to the Runner prior to dispatching it
to the thread pool; we can pass that reference in rather than requiring
the thread pool to take a new reference to it.
2015-09-07 19:51:54 -04:00
Nathan Froyd
17c37f82b5 Bug 1195767 - part 5 - use signaling instead of broadcast when work items are placed in nsEventQueue; r=gerald
There's no reason to wake up all the threads in a thread pool when one
item gets placed in the queue.  Waking up one will serve the same
purpose and is significantly more efficient for thread pools with large
numbers of threads.
2015-09-03 16:38:18 -04:00
Nathan Froyd
8502eaeea7 Bug 1195767 - part 4 - remove nsEventQueue::GetReentrantMonitor; r=gerald
The last commit eliminated the only client of this method, so we can
remove it now.
2015-09-03 16:37:51 -04:00
Nathan Froyd
aafe5d489a Bug 1195767 - part 3 - modify nsThreadPool to use a non-reentrant monitor; r=gerald
There's no reason nsThreadPool needs to use a reentrant monitor for
locking its event queue.  Having it use a non-reentrant one should be
slightly more efficient, both in the general operation of the monitor,
and that we're not performing redundant locking in methods like
nsThreadPool::Run.  This change also eliminates the only usage of
nsEventQueue::GetReentrantMonitor.
2015-09-03 15:38:28 -04:00
Nathan Froyd
92e6eccda2 Bug 1195767 - part 2 - create an nsEventQueueBase templated over the monitor type; r=gerald
Clients of nsEventQueue don't always need fully reentrant monitors.
Let's account for that by having a base class templated on the monitor
type.  This change also opens up the possibility of having the monitor
for the event queue not owned by the event queue itself, but by the
client class, which makes a lot more sense than the current design.
2015-08-28 13:26:17 -04:00
Nathan Froyd
1e05c1710d Bug 1195767 - part 1 - remove nsCOMPtr temporary from nsEventQueue::PutEvent; r=gerald
The comment here suggests that we might AddRef/Release, but we really do
no such thing.  Let's deal with the transfer of ownership directly,
rather than going through nsCOMPtr.  This change makes the code slightly
smaller, and it also makes later refactorings to pull the lock out of
this function easier to do, since we don't have to consider how to hold
the lock within the lifetime of the nsCOMPtr temporary.
2015-08-28 14:19:49 -04:00
Miko Mynttinen
2337336eca Bug 1201287 - Cleanup nsSupportsPrimitives.cpp. r=smaug
--HG--
extra : rebase_source : b0828cd31ba5fd7778852185fdb39af180750287
2015-09-06 19:53:33 +03:00
Phil Ringnalda
90583f7067 Backed out 10 changesets (bug 1123237) for hazard build failures and Windows test failures
CLOSED TREE

Backed out changeset 9c26b3b787f8 (bug 1123237)
Backed out changeset 1fcec0dc93d5 (bug 1123237)
Backed out changeset 390033ceebb6 (bug 1123237)
Backed out changeset e8a1845876d6 (bug 1123237)
Backed out changeset 714ec40715fa (bug 1123237)
Backed out changeset 5ed952e011c3 (bug 1123237)
Backed out changeset c785df6c0cdf (bug 1123237)
Backed out changeset d69e2d195a24 (bug 1123237)
Backed out changeset 1f328807da1d (bug 1123237)
Backed out changeset a1546857dce9 (bug 1123237)
2015-09-05 13:49:39 -07:00
Kan-Ru Chen
ab9d4cc8c0 Bug 1123237 - Part 10. Expose SwapElements from nsBaseHashtable. r=nfroyd 2015-09-06 00:13:26 +08:00
Wes Kocher
85ea8cdaff Merge m-c to inbound, a=merge 2015-09-04 15:41:19 -07:00
Wes Kocher
a5709b4620 Merge inbound to central, a=merge 2015-09-04 15:34:42 -07:00
Eric Rahm
42afa97523 Bug 1199400 - Part 2: Add tests for possible nsDeque corner cases. r=froydnj 2015-09-04 15:05:01 -07:00
Andrew McCreight
13445609b8 Backed out changeset 351d5f864f9e for not compiling. 2015-09-04 09:52:24 -07:00
Andrew McCreight
cc00f8d5c9 Bug 1201271 - Warn about unused results for more methods of nsTArray. r=froydnj
This leaves alone the AppendElement methods.
2015-09-04 09:45:44 -07:00
Andrew McCreight
e46abbfe4f Bug 1200795, part 2 - Add memory reporting for CCGraph::mPtrToNodeMap. r=njn 2015-09-04 09:45:44 -07:00
Andrew McCreight
bd863a70b2 Bug 1200795, part 1 - Eliminate excessive detail from cycle collector graph memory reporting. r=njn 2015-09-04 09:45:44 -07:00
Carsten "Tomcat" Book
ba11394eac Merge mozilla-central to fx-team 2015-09-04 16:25:40 +02:00
Mike Hommey
fba2225be7 Bug 1201453 - Make TestTArray's test_fallible use array sizes slightly less than 128MB. r=nfroyd 2015-09-04 14:35:53 +09:00
Eric Rahm
f8b8e77b55 Bug 1199400 - Part 1: Use CheckedInt when growing nsDeque capacity. r=froydnj 2015-09-03 13:33:46 -07:00
Eric Rahm
42bed38c5e Bug 1199400 - Part 0: Remove unused nsDequeIterator. r=froydnj 2015-09-04 15:04:58 -07:00
Florian Quèze
e966fcb5ac Bug 1169459 - remove the loadFromJars/jarURIs prefs, r=Mossop. 2015-09-04 11:50:49 +02:00
David Major
1415ae55cb Bug 1201205 part 2: Restore protection on the nop space separately from the function. r=m_kato
--HG--
extra : rebase_source : c3b5d44c04b41dc4133e9f3f50a0394c964ac673
2015-09-04 14:24:05 -04:00
David Major
2f23a1b1e6 Bug 1201205 part 1: Add an AutoVirtualProtect helper class to make the next patch easier. r=m_kato
--HG--
extra : rebase_source : c32b1cd483bdbfe09760fcb7c0a36cecf3b9940c
2015-09-04 14:23:33 -04:00
Terrence Cole
0716d4888f Bug 1199843 - Part 3: Strongly type JS::TraceChildren; r=jonco r=mccr8
--HG--
extra : rebase_source : c779897338dc4b73dca98517acb8d0eef7e0d7a5
2015-09-02 11:34:36 -07:00
Ben Kelly
10350eb242 Bug 1184607 P4 Handle the RequestRedirect mode during service worker interception. r=nsm
* * *
Bug 1184607 P4 interdiff 001 fix manual redirect assertion for navigations r=nsm
* * *
Bug 1184607 P4 interdiff 002 dom/worker nits
2015-08-31 14:26:29 -07:00
Alphan Chen
daa714c8ea Bug 1154435 - [nsGZFileWriter] Add one more mode for writing a GZFile (Create/Append). r=froydnj 2015-06-05 11:20:16 +08:00
Nicholas Nethercote
f44287005f Bug 1198334 (part 1) - Replace the opt-in FAIL_ON_WARNINGS with the opt-out ALLOW_COMPILER_WARNINGS. r=glandium.
The patch removes 455 occurrences of FAIL_ON_WARNINGS from moz.build files, and
adds 78 instances of ALLOW_COMPILER_WARNINGS. About half of those 78 are in
code we control and which should be removable with a little effort.

--HG--
extra : rebase_source : 82e3387abfbd5f1471e953961d301d3d97ed2973
2015-08-27 20:44:53 -07:00