Commit Graph

112242 Commits

Author SHA1 Message Date
Tim Taubert
ff5fcb3a25 Bug 802073 - Receive input event twice from input tag type:time and type:date r=vingtetun a=blocking-basecamp 2012-11-07 12:53:24 +01:00
Ed Morley
7955e755b0 Backout e89f1fce980d (bug 509052) for an assortment of timeouts and crashes 2012-11-07 10:47:27 +00:00
Robert Longson
699d16a0e0 Bug 779971 - Make nsSVGTextPathProperty::DoUpdate trigger nsSVGTextFrame::NotifyGlyphMetricsChange() off an asynchronous change hint (to avoid calling nsLayoutUtils::FrameNeedsReflow synchronously under nsISVGChildFrame::ReflowSVG or during frame teardown, and avoid infinite loops caused by using an event queue event). r=jwatt. 2012-11-07 09:53:44 +00:00
Paul Adenot
9dbfcb81b2 Bug 779997 - Import SoundTouch Library in the tree. r=khuey,rstrong 2012-08-26 22:36:44 -07:00
Paul Adenot
0c5720f386 Backed out changeset 98f9d1044e54 because r= bit is missing in commit message. 2012-11-07 10:30:49 +01:00
Paul Adenot
9352d73358 Bug 779997 - Import SoundTouch Library in the tree. r= 2012-08-26 22:36:44 -07:00
Dave Hylands
8896b0866f Bug 794599 - Backout 847cedab81ae due to debug orange 2012-11-07 01:03:52 -08:00
Bas Schouten
18e314acfd Bug 509052: Add new, faster blurring code. r=derf 2012-11-07 09:29:54 +01:00
Shian-Yow Wu
b52583e740 Bug 809006 - B2G Network Manager: Unable to set default route and DNS when required system properties not available. r=vicamo 2012-11-07 15:12:52 +08:00
Matt Woodrow
9ef6a10cc9 Bug 805948 - Use the local perspective value instead of trying to retrieve it from the nsStyleDisplay again. r=cjones 2012-11-07 19:56:57 +13:00
Matt Woodrow
2146637de4 Bug 808469 - Use gfxQuartzSurface for OMTC drawing on mac. r=cjones 2012-11-07 19:56:56 +13:00
Matt Woodrow
f63b23664a Bug 808469 - Make gfxSharedImageSurface take a template parameter for the base surface type. r=cjones
--HG--
rename : gfx/thebes/gfxSharedImageSurface.cpp => gfx/thebes/gfxBaseSharedMemorySurface.cpp
rename : gfx/thebes/gfxSharedImageSurface.h => gfx/thebes/gfxBaseSharedMemorySurface.h
2012-11-07 19:56:54 +13:00
Dave Hylands
1b7759a62f Bug 794599 - Add a preference to disable logging. r=bzbarsky 2012-11-06 22:34:33 -08:00
Martijn Wargers
8ffe368942 Bug 802312 - test_power_basics.html is failing on b2g mochitest, r=kchen 2012-10-21 05:19:00 -07:00
Karl Tomlinson
3aa8a55355 b=808873 restore null check on gFocusWindow accidentally dropped in dfe49308dc99 r=trivial
--HG--
extra : transplant_source : n5S%82%B8n%18h%FD%86%A4%D9%19%96%D1%29%20%02%0D%CE
2012-11-07 18:05:01 +13:00
Wes Johnston
05e54fd0c4 Bug 808663 - Use nearest neighbor scaling for favicons in awesomescreen. r=bnicholson 2012-11-06 19:51:06 -08:00
Mike Habicher
5bf2c472b4 Bug 808099 - Write the correct rotational meta-data to the video stream. r=cjones 2012-11-06 22:32:01 -05:00
Nicholas D. Matsakis
e0497e260e Bug 807480 - Add Rooted<T> roots to PerThread state. r=luke
Currently there are a number of global fields in JSRuntime* which are basically
tracking per-thread state.  This makes sense on the current trunk since there
is only ever a single thread associated with a runtime, but as Parallel JS (nee
Rivertrail) starts to land this assumption no longer holds.

This patch makes a struct, currently called |JS::PerThreadData|, that stores
per-thread data from the runtime.  There is one instance of this struct
embedded in "Runtime" itself (the field |mainThread|).  For now I have only
migrated the debug GC fields into |PerThread|, those are the ones causing me
immediate pain.  Eventually more fields will want to move into there.

The eventual goal is to distinguish thread-safe code, which will take as
argument a |JS::PerThread*|, from non-thread-safe code, which will take a
|JSRuntime*| or |JSContext*|.
2012-11-06 22:32:01 -05:00
Steven Lee
c1e57fbdec Bug 800263 - [FM Radio] Launch the FM Radio with headphone already plugged-in will cause screen freeze. r=cjones 2012-11-06 22:32:00 -05:00
Ryan VanderMeulen
1e6a5f0d3a Backout c3933841e06e (bug 793013) for possible Win7 reftest failures. 2012-11-06 21:27:09 -05:00
Bobby Holley
2d5f3aaee1 Bug 808730 - Implement the 'IfFound' parent of ReparentWrapperIfFound. r=mrbkap
This is a pretty dumb bug, but we never actually hit this codepath at the moment.
We currently end up in wrapper by either explicitly calling
ReparentWrappedNativeIfFound (in nsNodeUtils and in nsHTMLDocument), or by moving
existing wrappers during orphan fixup. The former always passes a wrapper-cached
object, and the latter guarantees that we always have a wrapper. So we'd never
hit the "no wrapper cache, no wrapper" case in the current code.
2012-11-06 17:56:49 -08:00
Nicholas Nethercote
8bd3070a4a Bug 808467 - Rewrite the cycle collector's memory reporter. r=mccr8.
--HG--
extra : rebase_source : 9944d59772654a1df8db3227f1a8b18ae3dcfeda
2012-11-06 17:38:29 -08:00
Ehsan Akhgari
bb35cf3c91 Merge mozilla-central into mozilla-inbound 2012-11-06 20:16:41 -05:00
Gregory Szorc
96087ea75a Bug 807974 - Handle make errors more gracefully; r=jhammel
We now return the status code from executed processes. The API to
require a successful status code has been changed from ignore_errors to
ensure_exit_code. The build mach command no longer spews a stack trace
if make fails.

DONTBUILD (NPOTB)
2012-11-06 17:01:08 -08:00
Gregory Szorc
a015660155 Bug 808346 - Don't scan sys.path to discover mach commands; r=jhammel
All mach modules are now explicitly listed in the mach driver.

--HG--
rename : python/mozbuild/mach/commands/build.py => python/mozbuild/mozbuild/mach_commands.py
2012-11-06 17:00:19 -08:00
Gregory Szorc
47aa1a9366 Bug 808336 - Part 2: Discover mach settings providers via decorators; r=jhammel
With this patch, mach is now decoupled from the build system and is
truly a generic command dispatching framework.
2012-11-06 16:58:13 -08:00
Mark Hammond
a9911ca0b2 Bug 807571 - popup now resizes when sizeTo is called with the same height while open. r=enndeakin 2012-11-07 11:58:01 +11:00
Gregory Szorc
203a9e64f9 Bug 808336 - Part 1: Refactor mach command handler management; r=jhammel
Previously we were tighly coupled with MozbuildObject. This was not in
the spirit of mach being a generic tool. Now, instead of passing multiple
arguments to __init__ of the class providing the mach command we either
pass 0 or 1. The number of arguments is detected when processing the
@CommandProvider decorator. The optional argument is a named tuple
containing mach run-time state.

Capturing of mach command provider information is now captured in a
class (as opposed to an anoymous tuple). We also capture these in a rich
data structure which is passed as part of the run-time data to the
command provider class. This allows mach commands to inspect the mach
environment.

Mach decorators have been moved to mach.decorators. mach.base is
reserved for generic mach data/container classes.

Existing mach command classes derived from MozbuildObject have been
replaced with either object or mozbuild.base.MachCommandBase. This
latter class translates the mach context instance passed to __init__
into the constructor arguments for MozbuildObject.__init__.

Support for registering function handlers with mach has been removed.
All handlers must be inside classes.

--HG--
rename : python/mach/mach/base.py => python/mach/mach/decorators.py
2012-11-06 16:57:41 -08:00
Ehsan Akhgari
7efa74d459 Backed out changeset 5042f0a60460 (bug 805766) on the suspicion of causing bug 808287 2012-11-06 19:54:43 -05:00
Ryan VanderMeulen
a330ccbd1e Backout 3f0e69962962, 84a1afc5b15e, c62f225d0dbb, 24870ebc9665, and dd7663e68657 (bug 799344) for mochitest-1 orange. 2012-11-06 19:36:49 -05:00
Felipe Gomes
72c3adf341 Bug 808171. Follow-up. Check for and avoid setting the same value twice inside the _setEnabled function instead of the observer. r=markh 2012-11-06 16:30:04 -08:00
Richard Newman
3dcbea7f2e Merge m-c to s-c. 2012-11-06 15:36:31 -08:00
Doug Turner
bfcb7f0001 Backed out changeset 30b340a71112. r=me 2012-11-06 14:31:04 -08:00
Shane Caraveo
0b61fdb7c5 Bug 807217 - Disable Social in private browsing mode. r=gavin,felipe 2012-11-06 12:50:08 -08:00
Mark Hammond
1d1b8439e5 Bug 808171 - social.enabled preference observer checks if enabled state really needs to change. r=felipe 2012-11-07 11:11:28 +11:00
Doug Turner
6cf66eef3a Bug 805965 - Geolocation fix jerks around. r=jdm/bz. a=blocking-basecamp 2012-11-06 12:34:03 -08:00
Doug Turner
bebc200792 Backed out changeset 9035907b68ec 2012-11-06 12:45:25 -08:00
Doug Turner
40d8fe576a Bug 805965 - Geolocation fix jerks around. r=jdm/bz. a=blocking-basecamp 2012-11-06 12:34:03 -08:00
Olli Pettay
799158b668 Bug 809003 - Don't dispatch load and error events on generated content images, r=bz 2012-11-06 20:54:02 +02:00
Olli Pettay
fd975256a9 Bug 808993, ASSERTION: must have binding parent when in native anonymous subtree with a parent node, r=bz 2012-11-06 20:52:46 +02:00
Doug Turner
396ad276f6 Bug 808866 - do not create location providers in child process. r=jdm/amarchesini a=blocking-basecamp 2012-11-06 10:42:27 -08:00
Richard Newman
3f259525f3 Bug 808750 - More logging and safety in resource callbacks. r=gps 2012-11-05 17:31:12 -08:00
Richard Newman
5098856c67 Bug 693427 - More logging and fixes for filelog cleanup. r=gps 2012-11-05 16:05:06 -08:00
Eric Chou
843f440877 Bug 808193 - return 'boolean' if PROP_DEVICE_CONNECTED_TYPE is not found, r=qdot 2012-11-05 15:48:34 -08:00
Jonathan Griffin
c587d86099 Bug 804366 - Include test names in Marionette failure messages, r=ahal 2012-11-05 10:49:26 -08:00
Ed Morley
a5572ca7ff Merge last PGO-green changeset of mozilla-inbound to mozilla-central 2012-11-05 15:23:36 +00:00
Ed Morley
44c0c46e66 Bug 808419 - remoteautomation.py's waitForFinish() should print to the log when maxTime is reached; r=jmaher 2012-11-05 13:03:55 +00:00
Ed Morley
1c5718e096 Bug 808417 - Make runTests() exception reporting consistent across test harnesses; r=jmaher 2012-11-05 13:03:55 +00:00
Ed Morley
43bd031bcd Bug 808417 - remotereftest.py should print the actual exception if one occurred runTests(); r=jmaher 2012-11-05 13:03:54 +00:00
Lucas Rocha
c1fe0c501b Bug 808027 - Change favicons handling to be Bitmap-based (r=mfinkle) 2012-11-05 11:42:25 +00:00