OSDN Git Service
Stephen Hines [Mon, 18 Aug 2014 19:36:43 +0000 (19:36 +0000)]
Merge "Collapse code paths for single- and multi-input kernels."
Chris Wailes [Tue, 15 Jul 2014 17:56:14 +0000 (10:56 -0700)]
Collapse code paths for single- and multi-input kernels.
This patch simplifies the RenderScript JNI layer by replacing six functions
with a single funciton. This new function now handles all previous cases.
Functions in android.renderscript.script have been updated to use this new JNI
function.
Change-Id: I6cd5448534c38123d51a589339bbeb7e98453e73
Elliott Hughes [Mon, 18 Aug 2014 19:28:31 +0000 (19:28 +0000)]
Merge "WifiP2pWfdInfo set device type bugfix"
Grigoriy Kraynov [Sat, 16 Aug 2014 10:42:44 +0000 (14:42 +0400)]
WifiP2pWfdInfo set device type bugfix
Change-Id: I37a53aed80574e78c80dc4a4192bd8cc4320de9d
Signed-off-by: Grigoriy Kraynov <kraynov@mgpro.ru>
Elliott Hughes [Mon, 18 Aug 2014 17:40:42 +0000 (17:40 +0000)]
Merge "Use gettid directly in frameworks JNI."
Elliott Hughes [Mon, 18 Aug 2014 17:26:52 +0000 (10:26 -0700)]
Use gettid directly in frameworks JNI.
Bug:
17048545
Change-Id: I1cd45d85368c799b75f76e778722979cbbb68ed2
Piotr Jastrzebski [Mon, 18 Aug 2014 08:27:56 +0000 (08:27 +0000)]
Merge "Update ziparchive usage to the new API."
Brian Carlstrom [Sun, 17 Aug 2014 23:48:02 +0000 (23:48 +0000)]
Merge "PackageInfo is a class, not a struct"
Brian Carlstrom [Sun, 17 Aug 2014 23:02:33 +0000 (16:02 -0700)]
PackageInfo is a class, not a struct
(cherry picked from commit
646d1b8ea801d92b04ead6cbae2899d8256ec3ff)
Change-Id: Id9ddc520fbfc4787c3b267df9a35a45e8cc52d65
Brian Carlstrom [Fri, 15 Aug 2014 17:22:07 +0000 (17:22 +0000)]
Merge "Disable verification based on vold.decrypt. [frameworks/base]"
Brian Carlstrom [Fri, 15 Aug 2014 16:31:27 +0000 (09:31 -0700)]
Disable verification based on vold.decrypt. [frameworks/base]
Bug:
15165413
(cherry picked from commit
6d05323e97e98022be7fc1a803115e2628f2d1f4)
Change-Id: I5454488a9db0fd70c64e8fc78b622c56ea96e6fd
Piotr Jastrzebski [Wed, 13 Aug 2014 06:50:20 +0000 (07:50 +0100)]
Update ziparchive usage to the new API.
Bug:
16162465
Change-Id: If3064e2be3e7e85366b07434294a7879b0e5f9bc
Calin Juravle [Fri, 15 Aug 2014 11:35:59 +0000 (11:35 +0000)]
Merge "Integrate NativeActivity with NativeBridge interfaces"
Yong WU [Mon, 28 Jul 2014 13:20:18 +0000 (21:20 +0800)]
Integrate NativeActivity with NativeBridge interfaces
Bug:
16884833
Change-Id: I73aab8e212860ba5aee9444d801806d3da326a41
Jeff Sharkey [Tue, 12 Aug 2014 22:12:17 +0000 (22:12 +0000)]
Merge "Prevent system_server shutdown when trying to format external storage"
Stephen Hines [Wed, 13 Aug 2014 17:32:49 +0000 (17:32 +0000)]
Merge "Revert "Collapse code paths for single- and multi-input kernels.""
Stephen Hines [Wed, 13 Aug 2014 17:32:19 +0000 (17:32 +0000)]
Revert "Collapse code paths for single- and multi-input kernels."
This reverts commit
eb3470219dea322efa93eb4b5457813ce71d0c5d.
Change-Id: Id943abf953e832ef831318e6699d4b46e9b46201
Stephen Hines [Wed, 13 Aug 2014 16:57:04 +0000 (16:57 +0000)]
Merge "Collapse code paths for single- and multi-input kernels."
Craig Mautner [Tue, 12 Aug 2014 21:09:21 +0000 (21:09 +0000)]
Merge "Revert "Fix a bug: unable to start activity when AM is killing the process.""
Craig Mautner [Tue, 12 Aug 2014 21:09:07 +0000 (21:09 +0000)]
Revert "Fix a bug: unable to start activity when AM is killing the process."
This reverts commit
5073145f85c8197ec4387d5e2599442a7291ca9c.
I spoke to Dianne and she said that if this is making a problem go away then it is only hiding it and there is a deeper problem we need to fix.
I think if killedByApp is true then we should set app and app.thread to null. I'll look at it further when I have more time.
Change-Id: I44c29e15ca654666440472eab2d2af1c5af80b9a
Leo Hsu [Thu, 3 Jul 2014 08:51:24 +0000 (16:51 +0800)]
Fix a bug: unable to start activity when AM is killing the process.
Symptom: No activity was started for startActivity, low repro rate.
Root Cause: Starting activity on a process which was killed by ActivityManagerService.killUnneededProcessLocked() and before receiving AppDeathRecipient of it.
Solution: Choose the flow of starting a new process if ProcessRecord.killedByAm is true.
Change-Id: Ida2639b0fb4631222fc92d65aadc9fd2da637b45
Craig Mautner [Tue, 12 Aug 2014 16:59:39 +0000 (16:59 +0000)]
Merge "Fix a bug: unable to start activity when AM is killing the process."
Adam Powell [Mon, 11 Aug 2014 17:46:53 +0000 (17:46 +0000)]
Merge "Don't unnecessarily set LayoutParams on children"
Calin Juravle [Thu, 7 Aug 2014 16:38:41 +0000 (17:38 +0100)]
Fix the instruction set for dex file during (re)moving ops.
This complements https://android-review.googlesource.com/#/c/103231/
where the instruction set was mapped just for comptilation. The same
should have been done for move,remove and getSize.
The list of dex code ISAs is alo de-duped.
Bug:
16185267
Change-Id: I8fe453a800812e382e8f41b5f7922997aa9c18a9
Calin Juravle [Thu, 7 Aug 2014 13:45:53 +0000 (14:45 +0100)]
Pass the name of the native bridge library to the runtime
Bug:
16843953
Change-Id: Ia0f57a9b3e30a74fe4e93eb4453f841c1ef512eb
Chris Wailes [Tue, 15 Jul 2014 17:56:14 +0000 (10:56 -0700)]
Collapse code paths for single- and multi-input kernels.
This patch simplifies the RenderScript JNI layer by replacing six functions
with a single funciton. This new function now handles all previous cases.
Functions in android.renderscript.script have been updated to use this new JNI
function.
Change-Id: I77e4b155cc7ca1581b05bf901c70ae53a9ff0b12
Piotr Jastrzebski [Mon, 11 Aug 2014 09:54:23 +0000 (09:54 +0000)]
Merge "Remove memory leak."
Piotr Jastrzebski [Fri, 8 Aug 2014 11:52:39 +0000 (12:52 +0100)]
Remove memory leak.
Call EndIteration in endIteration to free memory allocated for cookie in
StartIteration.
Change-Id: I3d58a461c2f3676565cfd7922dc6d8d5c35feae9
Calin Juravle [Fri, 8 Aug 2014 17:48:02 +0000 (17:48 +0000)]
Merge "Fix the instruction set for dex file during (re)moving ops."
Calin Juravle [Fri, 8 Aug 2014 17:47:43 +0000 (17:47 +0000)]
Merge "Pass the name of the native bridge library to the runtime"
Calin Juravle [Thu, 7 Aug 2014 12:47:07 +0000 (12:47 +0000)]
Merge "Use native ISA for dex code (in case there's a NativeBridge)"
Calin Juravle [Mon, 4 Aug 2014 18:17:06 +0000 (19:17 +0100)]
Use native ISA for dex code (in case there's a NativeBridge)
In the presence of a native bridge it is more efficient to compile the
dex directly to the native ISA than to use the shared library ISA as a
reference.
This can be achieve by configuring the readonly system properties to map
between the .so ISA and the desired dex code .ISA (e.g.
ro.dalvik.vm.isa.ISA1=ISA2).
Bug:
16185267
Change-Id: I50baa7b37e1465b9adf72d6f6b96f526a08d59c7
Brian Carlstrom [Tue, 5 Aug 2014 17:47:12 +0000 (17:47 +0000)]
Merge "Make system use patchoat to relocate during runtime."
Alex Light [Fri, 18 Jul 2014 23:34:38 +0000 (16:34 -0700)]
Make system use patchoat to relocate during runtime.
Make PackageManagerService aware of patchoat and make it use it when
appropriate.
Bug:
15358152
Change-Id: Ibe92d8b55a24bbf718b0416a21b76e5df7a2de26
Mathieu Chartier [Wed, 6 Aug 2014 20:57:39 +0000 (20:57 +0000)]
Merge "Fix JNI error in exception reporting."
Mathieu Chartier [Wed, 6 Aug 2014 20:39:17 +0000 (13:39 -0700)]
Fix JNI error in exception reporting.
There was a JNI error where when you got an OOM and called
report_exception, it would call two NewStringUTF in a row without
checking the return values. This could mean that the first one
threw a new OOME and the second one would cause a JNI error when
it also attempted to throw an OOME with a pending OOME.
Bug:
16843627
Change-Id: Ie4f9f9a5f8b7993cd3655d42a6718c0a5e1199f8
Brian Carlstrom [Mon, 4 Aug 2014 18:27:38 +0000 (18:27 +0000)]
Merge changes I69c064b0,I1972c5af
* changes:
Move image classes options from art to AndroidRuntime [frameworks/base]
Disable compilation based on vold.decrypt. [frameworks/base]
Brian Carlstrom [Mon, 4 Aug 2014 18:27:35 +0000 (18:27 +0000)]
Merge "Move from dalvik.vm.image-dex2oat-flags to dalvik.vm.dex2oat-filter [frameworks/base]"
Brian Carlstrom [Wed, 30 Jul 2014 22:13:17 +0000 (15:13 -0700)]
Move image classes options from art to AndroidRuntime [frameworks/base]
Bug:
15165413
Change-Id: I69c064b05f62a2e8a3801dc924f5b032dedc8e51
Brian Carlstrom [Tue, 29 Jul 2014 02:13:28 +0000 (19:13 -0700)]
Move from dalvik.vm.image-dex2oat-flags to dalvik.vm.dex2oat-filter [frameworks/base]
This will allow us to conditionally change the compiler-filter based on other properties.
Bug:
15165413
Change-Id: I1676987c69a3cad4217dc4b56465ef3ae886122a
Brian Carlstrom [Wed, 30 Jul 2014 19:11:41 +0000 (12:11 -0700)]
Disable compilation based on vold.decrypt. [frameworks/base]
Bug:
15165413
Change-Id: I1972c5af5f392c2ed247888ef33c03e9ffe4d588
John Reck [Thu, 31 Jul 2014 18:28:07 +0000 (18:28 +0000)]
Merge "[View] Clear Hardware-Layers before system destroys HardwareRenderer."
Craig Mautner [Thu, 31 Jul 2014 16:04:18 +0000 (16:04 +0000)]
Merge "Fix a bug that the RecentTask list of other’s may show up to non-primary users."
Eunae Kim [Fri, 25 Jul 2014 00:16:13 +0000 (09:16 +0900)]
Fix a bug that the RecentTask list of other’s may show up to non-primary users.
When switching to a newly created user, the user may face this bug when he opens the RecentTask screen.
A possible bug scenario is described as follows:
A user id of a removed user may be recycled when created a new user.
However, mRecentTasks is not correctly controlled so that old information may still remain and be possibly mapped to wrong user.
This patch prevents this bug by explicitly removing old information in mRecentTasks when removing existing user.
Change-Id: I1874dbd604598a5d740ae1e034981e21214c15c6
Signed-off-by: Eunae Kim <eunae.kim@lge.com>
Craig Mautner [Thu, 31 Jul 2014 03:53:15 +0000 (03:53 +0000)]
Merge "[ActivityManager] Prevent coredump file may not complete for testing."
riddle_hsu [Wed, 30 Jul 2014 16:15:05 +0000 (00:15 +0800)]
[ActivityManager] Prevent coredump file may not complete for testing.
Sympton:
During testing, skip kill native crash process manually because it will continue to die by default.
Root Cause:
Large process may take some time to do coredump.In auto test, crash process will be killed immediately that results incomplete coredump file.
Solution:
If the tester (IActivityController) will handle app crash event,
Do not kill native crashed process if the rom is debuggable.
Change-Id: Ia360af147d694125d440e5ba2f958c4759a50494
Craig Mautner [Thu, 31 Jul 2014 03:52:14 +0000 (03:52 +0000)]
Merge "[ActivityManager] Avoid mistaking visibility by finishing task."
riddle_hsu [Wed, 30 Jul 2014 16:26:51 +0000 (00:26 +0800)]
[ActivityManager] Avoid mistaking visibility by finishing task.
Sympton:
Next activity only adds to history but does not launch/resume then results ANR.
Root Cause:
In a rare timing, some windows are switched at the same time,
it will cause some finishing records on the top temporarily,
then set startIt to false that skip to resume the real top activty.
Solution:
If all activities in a task are finishing, do not use it to check.
The behavior/checking is the same concept as in JellyBean:
// If starting in an existing task, find where that is...
boolean startIt = true;
for (int i = NH-1; i >= 0; i--) {
ActivityRecord p = mHistory.get(i);
if (p.finishing) { // <--
continue;
}
Change-Id: I9d81a7b5182400c52e173da23eee61c74692beee
Narayan Kamath [Thu, 31 Jul 2014 12:13:23 +0000 (12:13 +0000)]
Merge "Don't assume languages are 2 letter codes."
Narayan Kamath [Wed, 30 Jul 2014 14:42:25 +0000 (15:42 +0100)]
Don't assume languages are 2 letter codes.
Also, note that this method never worked. Locale
settings were stored with underscores (like Locale.toString)
but matched against AssetManager.getLocales() which
returned language-tag like output.
bug:
10090157
Change-Id: I36ffea6e39ff7d1907b2787f8fd4545c24d326a8
Narayan Kamath [Thu, 31 Jul 2014 11:29:56 +0000 (11:29 +0000)]
Merge "Switch TextUtils over to new ICU API."
Narayan Kamath [Wed, 30 Jul 2014 15:51:27 +0000 (16:51 +0100)]
Switch TextUtils over to new ICU API.
.. and use Locale.getScript() instead of ICU.getScript.
bug:
15876704
Change-Id: Idf9462d8ef568dbc88c95e65a971184952e97872
henry.uh_chen [Fri, 18 Jul 2014 07:11:18 +0000 (15:11 +0800)]
[View] Clear Hardware-Layers before system destroys HardwareRenderer.
Symptom: Sometimes HardwareLayer display is corrupted.
Root Cause: In some cases, ViewRootImpl destroys HardwareRenderer without
clearing HardwareLayers in the view hirarchy. If the system trims memory fully
and kill the glContext at the same time, HardwareLayer may corrupt.
Solution: Clear Hardware-Layers before system destroys HardwareRenderer.
Reproduce steps: Modify the cache's trim memory mechanism to make it easier to
trim fully, HardwareLayers implemented in app will easily be corrupted.
Modification of the cache is not required but make it easier to reproduce.
Change-Id: If57c802cdacf8dc37cbbcc199e74f9482834f981
Tim Murray [Thu, 24 Jul 2014 20:59:33 +0000 (20:59 +0000)]
Merge "Update makefile for clang host build."
Chris Craik [Wed, 23 Jul 2014 00:14:48 +0000 (00:14 +0000)]
Merge "Restore untrusted glStencilMask value on resume after functor"
Chris Craik [Tue, 22 Jul 2014 19:30:39 +0000 (12:30 -0700)]
Restore untrusted glStencilMask value on resume after functor
Fixes an issue where the stencil buffer is corrupted
(used in overdraw debugging, path clipping) when used in
conjunction with a HW accelerated WebView.
bug:
15023700
bug:
16465551
Change-Id: If201b503931380f7107a34e9a9a2f877a50f72b1
Chris Craik [Tue, 22 Jul 2014 19:08:23 +0000 (19:08 +0000)]
Merge "[Bitmap] Add null pointer protection in Bitmap_sameAs()"
Chris Craik [Tue, 22 Jul 2014 19:07:19 +0000 (19:07 +0000)]
Merge "[HWUI] Fix invisible views are still shown in popup window"
Tim Murray [Thu, 17 Jul 2014 23:37:42 +0000 (16:37 -0700)]
Update makefile for clang host build.
bug
16172793
Change-Id: If7484c5dbcccce7d925bec97bff0a3e4c30e9434
Neil Fuller [Tue, 22 Jul 2014 15:26:08 +0000 (15:26 +0000)]
Merge "Rewriting android.text.format.Time without the native _tz functions"
Neil Fuller [Wed, 25 Jun 2014 10:13:25 +0000 (11:13 +0100)]
Rewriting android.text.format.Time without the native _tz functions
Bug:
15765976
Change-Id: I666b72ecf9da8a9dcfb97cc503006b415909a558
Craig Mautner [Mon, 21 Jul 2014 15:26:58 +0000 (15:26 +0000)]
Merge "Skip broadcasting to a receiver if the receiver seems to be dead"
Koji Fukui [Fri, 21 Feb 2014 12:31:14 +0000 (21:31 +0900)]
Skip broadcasting to a receiver if the receiver seems to be dead
If an application that has a broadcast receiver is killed
during broadcasting, thread variable of ProcessRecord becomes null
so that IIntentReceiver#performReceive() is called
in BroadcastQueue#performReceiveLocked(). But if binder driver has not
noticed the death of the application yet, it can't throw
DeadObjectException. After that, binder driver notices. But it can't
notify DeadObjectException to the caller because performReceive() is
async call. So broadcasting keeps on waiting for finishing
performReceive() until timeout.
This change checks the death of the application before calling
performReceive() and skips broadcasting to the receiver
if the application has already died.
Change-Id: Ifa02b8b1a7e7b6fd314de90fedff5b7a5326825d
Yohann Roussel [Fri, 18 Jul 2014 10:54:42 +0000 (10:54 +0000)]
Merge "Add support.multidex tests app for an update bug."
Narayan Kamath [Tue, 22 Jul 2014 11:15:42 +0000 (11:15 +0000)]
Merge "Remove copy of Locale.adjustLanguageCode."
Narayan Kamath [Fri, 18 Jul 2014 15:09:36 +0000 (16:09 +0100)]
Remove copy of Locale.adjustLanguageCode.
Change-Id: Iba8c67890f95d50fadcd1245dc083d646fe6bcc1
Chris Craik [Thu, 17 Jul 2014 19:28:32 +0000 (19:28 +0000)]
Merge "[HWUI]: fix residual line on FrameBuffer"
Craig Mautner [Wed, 16 Jul 2014 17:17:46 +0000 (17:17 +0000)]
Merge "[ActivityManager]: Update home process when home activity resumed"
louis_chang [Mon, 28 Apr 2014 08:59:22 +0000 (16:59 +0800)]
[ActivityManager]: Update home process when home activity resumed
Symptom:
When switch between two home activities, the home process might not be updated to the correct one. In that case, the home activity might be killed easily.
Root Cause:
The home process is updated only when a home activity is newly created or being restarted. ActivityManager did not update the home process when simply resume a home activity.
Solution:
Update home process when home activity resumed
Reproduce Steps:
1. Install a launcher application, such as Apex Launcher
2. Press home key to change to Apex Launcher by "Just once" option
3. Press home key to switch back to original home activity by "Just once" option.
(The home process is still the Apex Launcher's process because ActivityManager does not update the home process when resuming the original home activity.)
Change-Id: I046279ca7ba851a283ee67ea19202890f7b3f343
Yohann Roussel [Tue, 15 Oct 2013 08:10:47 +0000 (10:10 +0200)]
Add support.multidex tests app for an update bug.
The failing scenario is
- install v1
- run
- update to v2
- update to v3
- run
last run was failing with initial multidex library versions because
it's still running on v1 extracted secondary dex files.
(cherry picked from commit
05e2a94c8b510131f43a686f5188d4c0f2a5eebd)
Change-Id: Ibb5d16642b127ee4c0baddc4f6ba461c11d25f90
Craig Mautner [Wed, 16 Jul 2014 15:46:30 +0000 (15:46 +0000)]
Merge "[ActivityManager] Do not finish root activity when reset task"
louis_chang [Wed, 16 Jul 2014 10:15:35 +0000 (18:15 +0800)]
[ActivityManager] Do not finish root activity when reset task
Symptom:
The root activity not always located at index 0 of the task.
For example, the index 1 activity will become the new root
when the original root activity (at index 0) finished.
The new root activity might be finished unexpectedly before
the original root activity actually destroyed.
Solution:
Check frontOfTask to avoid finishing the root activity unexpectedly
Change-Id: I623ab97e9c95c83b3cfe7c9dfc151a291a391ea4
Craig Mautner [Wed, 16 Jul 2014 15:42:50 +0000 (15:42 +0000)]
Merge "[ActivityManager] Move top task to top in window manager"
louis_chang [Wed, 16 Jul 2014 09:42:04 +0000 (17:42 +0800)]
[ActivityManager] Move top task to top in window manager
Symptom:
In certain situations, application starts activity while
it is in background. When an existing background activity
starts a new activity which results to be kept in the same
background task, the new activity won't be resumed.
In that case, the background task (sourceTask) should not be
moved to top in window manager.
Solution:
Move top task of the target stack to top in window manager
Change-Id: Id7a37ea67ce1f80e0c2b5399865c51fd7113deb8
Narayan Kamath [Wed, 16 Jul 2014 13:31:58 +0000 (13:31 +0000)]
Merge "Removed useless test"
Narayan Kamath [Wed, 16 Jul 2014 12:51:30 +0000 (12:51 +0000)]
Merge "Add an internal API to get all asset locales."
Deepanshu Gupta [Tue, 15 Jul 2014 20:07:22 +0000 (20:07 +0000)]
Merge "Optimize Blend composites."
Deepanshu Gupta [Tue, 15 Jul 2014 20:07:08 +0000 (20:07 +0000)]
Merge "Add BlendComposite.java"
Narayan Kamath [Mon, 14 Jul 2014 15:28:24 +0000 (16:28 +0100)]
Add an internal API to get all asset locales.
This will allow us to not copy paste this code verbatim into
bundled apps.
bug:
10090157
Change-Id: I008dc683ecbef2ad8b7a26968cb3cbda7e5a8388
Narayan Kamath [Mon, 14 Jul 2014 08:49:13 +0000 (08:49 +0000)]
Merge "Allow 3 letter language codes in InputMethodUtils."
Narayan Kamath [Fri, 11 Jul 2014 10:50:24 +0000 (11:50 +0100)]
Allow 3 letter language codes in InputMethodUtils.
Replace locale.substring(0, 2) with a function that always
returns the first component of the locale (assumed to be the
language).
bug:
10090157
(cherry picked from commit
cefc79c6b18bb2e824c299e7b9e212071e5ebc43)
Change-Id: I56b02dd0c6dc222daa3ceade841a30879449bb4a
Christopher Tate [Thu, 10 Jul 2014 18:51:58 +0000 (18:51 +0000)]
Merge "Logging in LocalTransport on as default"
Henrik Baard [Thu, 10 Jul 2014 06:46:39 +0000 (08:46 +0200)]
Logging in LocalTransport on as default
The logging is LocalTransport.java is enabled by
default.
Change default to false to remove logs, as this class
states that it is only for debug.
Change-Id: Iae63c7b62edbd503c606f0aee671a9579e78843d
Colin Cross [Thu, 10 Jul 2014 17:49:35 +0000 (17:49 +0000)]
Merge "GpsLocationProvider: look for config file in /etc/gps.<ro.hardware.gps>.conf"
Stephen Hines [Thu, 10 Jul 2014 00:10:05 +0000 (00:10 +0000)]
Merge "Update copyFrom(BaseObj[]) for large objects."
Tim Murray [Tue, 1 Jul 2014 23:56:18 +0000 (16:56 -0700)]
Update copyFrom(BaseObj[]) for large objects.
Change-Id: I99cadbd1ad925cad0dd357c9abbd1a49c07785c9
Elliott Hughes [Wed, 9 Jul 2014 23:59:58 +0000 (23:59 +0000)]
Merge "Switch frameworks/base/libs/androidfw to the new icu."
Elliott Hughes [Wed, 9 Jul 2014 23:58:28 +0000 (16:58 -0700)]
Switch frameworks/base/libs/androidfw to the new icu.
Change-Id: Id82830750dd7e83bbc66811a1133b403cc697c97
Elliott Hughes [Wed, 9 Jul 2014 23:56:11 +0000 (16:56 -0700)]
Switch frameworks/base/core/jni to the new icu.
Change-Id: Id12ce02da377ce78f318e10633c47f500237d9d9
Elliott Hughes [Wed, 9 Jul 2014 23:59:14 +0000 (23:59 +0000)]
Merge "Switch frameworks/base/core/jni to the new icu."
Deepanshu Gupta [Mon, 7 Jul 2014 21:22:19 +0000 (14:22 -0700)]
Optimize Blend composites.
Removed redundant array allocations to improve performance for various
blending modes.
Change-Id: Iaba1d6ff3ad03eebdc859c599b610cc593370438
Deepanshu Gupta [Mon, 7 Jul 2014 17:35:15 +0000 (10:35 -0700)]
Add BlendComposite.java
The class is adapted from a demo tool for Blending Modes written by
Romain Guy (romainguy@android.com). The tool is available at
http://www.curious-creature.org/2006/09/20/new-blendings-modes-for-java2d/
Change-Id: I8f7c7ca08d3078106056764a4e2f1ce95d990137
Craig Mautner [Mon, 7 Jul 2014 21:14:33 +0000 (21:14 +0000)]
Merge "[ActivityManager] Prevent potential deadlock from system error dialog and InputMethodManager."
riddle_hsu [Mon, 7 Jul 2014 20:30:19 +0000 (04:30 +0800)]
[ActivityManager] Prevent potential deadlock from system error dialog and InputMethodManager.
Sympton:
When application crash in a special timing and system server's InputMethodManager is binding, deadlock may occur.
Root Cause:
Thread(1): When using InputMethodManager in system server, it will lock mH(handler) of InputMethodManager,
and sometimes it will call to InputMethodManagerService::showCurrentInputLocked and will call bindService at some condition,
then it will also lock ActivityManagerService.
Thread(2): When an application crashed, it will lock ActivityManager when showing crash dialog, inside the dialog,
it will call setEnabled of Button and lock mH of InputMethodManager.setEnable of TextView will lock IMM's handler.
So the deadlock happened as the flow: (2) lock AMS -> (1)lock mH -> (2)wait mH -> (1) wait AMS
Solution:
Reduce nested lock of error dialog: post message to let (2) lock mH after release AMS lock.
Change-Id: Id85c29406236db3b5fca9655fde1fcaf0afd1337
Brian Carlstrom [Mon, 7 Jul 2014 20:48:48 +0000 (20:48 +0000)]
Merge "Remove"
Brian Carlstrom [Tue, 17 Jun 2014 20:57:58 +0000 (13:57 -0700)]
Remove
(cherry picked from commit
9881203920c8445e9a4bdd9e4a90ba0014f34c5e)
Change-Id: Ib40823b7c03af4c71da902b0e5c504e074cf0dd9
Narayan Kamath [Mon, 7 Jul 2014 08:52:10 +0000 (08:52 +0000)]
Merge "Fix issue when converting fil->tl."
Narayan Kamath [Sat, 5 Jul 2014 14:33:28 +0000 (15:33 +0100)]
Fix issue when converting fil->tl.
We should call ResourceTable::getLocales directly,
and not AssetManager::getLocales. The latter will convert
"tl" to "fil" so we'll end up thinking we have resources
for "fil" when we don't really have any.
bug:
15873165
Change-Id: I9753e4608aaecede328a40ee1f3ee6b016d0dedc
henry.uh_chen [Thu, 3 Jul 2014 12:40:22 +0000 (20:40 +0800)]
[Bitmap] Add null pointer protection in Bitmap_sameAs()
Symptom: SkBitmap::getAddr(int, int) may return NULL due to unrecognized config
(ex: kRLE_Index8_Config). This will cause memcmp method to crash. Since bm0 and
bm1 both have pixel data() (have passed NULL == getPixels() check), those 2
bitmaps should be valid (only unrecognized), we return JNI_FALSE to warn user
those 2 unrecognized config bitmaps may be different.
Change-Id: I6970c27de412110a3035d0a783112c4cd3ebc35b
henry.uh_chen [Thu, 3 Jul 2014 10:01:37 +0000 (18:01 +0800)]
[HWUI] Fix invisible views are still shown in popup window
Symptom: Invisible views are still shown in popup window
Root Cause: glClear isn't called if there is no draw command
Solution: Even if there is no draw command, it still calls glClear
by startFrame().
Reproduce steps: App set one view in a popup window to invisible,
it will be still visible.
Change-Id: I36bc7b752434ad388fbb80ea63341ac778f48d35