OSDN Git Service

android-x86/frameworks-base.git
9 years agoam 5ef53783: Merge "WindowManagerService not propagating X and Y steps"
Olawale Ogunwale [Mon, 1 Jun 2015 14:59:08 +0000 (14:59 +0000)]
am 5ef53783: Merge "WindowManagerService not propagating X and Y steps"

* commit '5ef53783743c7a7fb2339274ddb0886b0eb789ec':
  WindowManagerService not propagating X and Y steps

9 years agoMerge "WindowManagerService not propagating X and Y steps"
Olawale Ogunwale [Mon, 1 Jun 2015 14:42:36 +0000 (14:42 +0000)]
Merge "WindowManagerService not propagating X and Y steps"

9 years agoam e761938c: Merge "Fix javadocs based on recent case consistency change"
Neil Fuller [Mon, 1 Jun 2015 12:14:47 +0000 (12:14 +0000)]
am e761938c: Merge "Fix javadocs based on recent case consistency change"

* commit 'e761938c39c1cafa80d0c474e107041316a49cbd':
  Fix javadocs based on recent case consistency change

9 years agoMerge "Fix javadocs based on recent case consistency change"
Neil Fuller [Mon, 1 Jun 2015 12:01:44 +0000 (12:01 +0000)]
Merge "Fix javadocs based on recent case consistency change"

9 years agoam dd783301: Merge "Add optional additional entropy to finish"
Chad Brubaker [Fri, 29 May 2015 21:31:52 +0000 (21:31 +0000)]
am dd783301: Merge "Add optional additional entropy to finish"

* commit 'dd78330140ad0e02fc693fc1f5bf041adcf64a1f':
  Add optional additional entropy to finish

9 years agoMerge "Add optional additional entropy to finish"
Chad Brubaker [Fri, 29 May 2015 21:17:54 +0000 (21:17 +0000)]
Merge "Add optional additional entropy to finish"

9 years agoAdd optional additional entropy to finish
Chad Brubaker [Fri, 29 May 2015 19:32:51 +0000 (12:32 -0700)]
Add optional additional entropy to finish

If provided the extra entropy will be added to the device before calling
finish. If entropy is provided and the device does not support supplying
additional entropy then finish will fail with KM_ERROR_UNIMPLEMENTED.

Change-Id: If26be118bf382604f6f8e96e833b76e6f9e94d58

9 years agoam 4993da92: Merge "Patches for IKeyguardStateCallback"
Narayan Kamath [Fri, 29 May 2015 09:39:43 +0000 (09:39 +0000)]
am 4993da92: Merge "Patches for IKeyguardStateCallback"

* commit '4993da924c6f25194aded71e70eefd07208eee81':
  Patches for IKeyguardStateCallback

9 years agoMerge "Patches for IKeyguardStateCallback"
Narayan Kamath [Fri, 29 May 2015 09:26:16 +0000 (09:26 +0000)]
Merge "Patches for IKeyguardStateCallback"

9 years agoPatches for IKeyguardStateCallback
dooyoung.hwang [Thu, 21 May 2015 07:09:43 +0000 (16:09 +0900)]
Patches for IKeyguardStateCallback

1. Also notify input is restricted or not when IKeyguardStateCallback callback is added.
When first Keyguard is showing after boot, the mInputRestricted is not updated. So,
inKeyguardRestrictedInputMode still returns false, although device is locked.

2. If mShowing remains false during setup time of KeyguardViewMediator, mInputRestricted
is not updated. This is because updateInputRestrictedLocked is called only if showing
value is changed during setup time.
For this reason, when device is not provisioned, inKeyguardRestrictedInputMode returns
false, and device does not restrict key.

3. If one of callbacks causes RemoteException, the other callbacks are not called.
This is because catching RemoteException breaks for-loop.

Change-Id: I0281405127bcc75285b908d01485da3ac9b40b52
Signed-off-by: Dooyoung Hwang <dooyoung.hwang@lge.com>
9 years agoam 90cfc192: Merge "Close race condition in binderDied()"
Narayan Kamath [Fri, 29 May 2015 09:25:16 +0000 (09:25 +0000)]
am 90cfc192: Merge "Close race condition in binderDied()"

* commit '90cfc1923ae0a9fe98b97d123b2250f01e8bc3f2':
  Close race condition in binderDied()

9 years agoMerge "Close race condition in binderDied()"
Narayan Kamath [Fri, 29 May 2015 08:34:02 +0000 (08:34 +0000)]
Merge "Close race condition in binderDied()"

9 years agoClose race condition in binderDied()
Christopher Tate [Wed, 20 May 2015 01:16:58 +0000 (18:16 -0700)]
Close race condition in binderDied()

It was possible for a binderDied() call to occur while the death
recipient list containing the object was being iterated, in which
case we could invalidate an object reference out from under the
iteration, causing a VM abort.  We now interlock the binderDied()
deref operation with the list's locking semantics to prevent this.

Bug 15831054

(cherry picked from commit 090c08ff2a7525c830b4edfd1a79aa8ee3893c60)

Change-Id: If0027d3ac4da1153284a425dd9b2819a203481ab

9 years agoam 3976a569: Merge "Replace --include-cfi with --generate-debug-info."
David Srbecky [Thu, 28 May 2015 18:45:36 +0000 (18:45 +0000)]
am 3976a569: Merge "Replace --include-cfi with --generate-debug-info."

* commit '3976a56904e359eb975395725e908122d7d6e466':
  Replace --include-cfi with --generate-debug-info.

9 years agoam 495ebb11: Merge "Update asan/app_process."
Evgenii Stepanov [Thu, 28 May 2015 18:45:30 +0000 (18:45 +0000)]
am 495ebb11: Merge "Update asan/app_process."

* commit '495ebb119bd1cea64f985367ba3bb87594e4fa81':
  Update asan/app_process.

9 years agoMerge "Replace --include-cfi with --generate-debug-info."
David Srbecky [Thu, 28 May 2015 18:37:35 +0000 (18:37 +0000)]
Merge "Replace --include-cfi with --generate-debug-info."

9 years agoMerge "Update asan/app_process."
Evgenii Stepanov [Thu, 28 May 2015 18:31:07 +0000 (18:31 +0000)]
Merge "Update asan/app_process."

9 years agoFix javadocs based on recent case consistency change
Neil Fuller [Thu, 28 May 2015 17:19:02 +0000 (18:19 +0100)]
Fix javadocs based on recent case consistency change

Commit a40d2447b6f516116135ee7f126579771ba98a2c made the
case consistently "as for the beginning of a sentence".
This updates the docs to reflect that. Previously the
result depended on the string returned (e.g. "Tomorrow"
was capitalized in English) and the casing varied
within and between locales.

Bug: 20247811
Change-Id: If4c6b3a13f61bd0cb64c82fc4c96d053130ae4c8

9 years agoReplace --include-cfi with --generate-debug-info.
David Srbecky [Thu, 28 May 2015 16:16:09 +0000 (17:16 +0100)]
Replace --include-cfi with --generate-debug-info.

This follows a change in dex2oat.

Change-Id: I46b61298ca00bee2e83080456b180d63efd95b8f

9 years agoam d2b63ff8: Merge "Fix bug regarding the comparator sorts Alarms into increasing...
Chris Tate [Thu, 28 May 2015 00:08:54 +0000 (00:08 +0000)]
am d2b63ff8: Merge "Fix bug regarding the comparator sorts Alarms into increasing time order."

* commit 'd2b63ff8b2706e54bae2678d42db2681db812113':
  Fix bug regarding the comparator sorts Alarms into increasing time order.

9 years agoMerge "Fix bug regarding the comparator sorts Alarms into increasing time order."
Chris Tate [Wed, 27 May 2015 23:53:32 +0000 (23:53 +0000)]
Merge "Fix bug regarding the comparator sorts Alarms into increasing time order."

9 years agoFix bug regarding the comparator sorts Alarms into increasing time order.
jinho.park [Wed, 27 May 2015 05:44:18 +0000 (14:44 +0900)]
Fix bug regarding the comparator sorts Alarms into increasing time order.

The comparator used variable "when" in IncreasingTimeOrder class.
variable "when" means "elapsed time" or "real time"(RTC) and these types have different unit of time.
so, I recommend that you use a "whenElapsed" which has the equal unit of time as its default value.

Change-Id: I248863ef8a5c49cc7114cb2965f3d6b78f4faa59
Signed-off-by: Jinho Park <jinho.park@lge.com>
9 years agoam 3f453164: Merge "Pass charset to XmlPullParser.setInput instead of null"
Narayan Kamath [Fri, 22 May 2015 09:39:25 +0000 (09:39 +0000)]
am 3f453164: Merge "Pass charset to XmlPullParser.setInput instead of null"

* commit '3f453164ec884d26a556477027b430cb22a9b7e3':
  Pass charset to XmlPullParser.setInput instead of null

9 years agoMerge "Pass charset to XmlPullParser.setInput instead of null"
Narayan Kamath [Fri, 22 May 2015 09:30:45 +0000 (09:30 +0000)]
Merge "Pass charset to XmlPullParser.setInput instead of null"

9 years agoam d3916c88: Merge "Fix application moves."
Narayan Kamath [Fri, 22 May 2015 09:27:31 +0000 (09:27 +0000)]
am d3916c88: Merge "Fix application moves."

* commit 'd3916c885ba382d78b6c64bc0773dd38638a6c3f':
  Fix application moves.

9 years agoam 8f9339d0: Merge "Throw a checked exception on inconsistent installs."
Narayan Kamath [Fri, 22 May 2015 09:27:25 +0000 (09:27 +0000)]
am 8f9339d0: Merge "Throw a checked exception on inconsistent installs."

* commit '8f9339d0d76bf92f7a55b4eaf569e4cfb6e3212c':
  Throw a checked exception on inconsistent installs.

9 years agoam 3178e9f6: Merge "Scan package to derive ABIs before optimizing."
Narayan Kamath [Fri, 22 May 2015 09:27:19 +0000 (09:27 +0000)]
am 3178e9f6: Merge "Scan package to derive ABIs before optimizing."

* commit '3178e9f6e823b01d1af382149ecafec741ecd2d0':
  Scan package to derive ABIs before optimizing.

9 years agoam 2e06258c: Merge "Do not create oat directory for an app in ASEC container"
Narayan Kamath [Fri, 22 May 2015 09:27:13 +0000 (09:27 +0000)]
am 2e06258c: Merge "Do not create oat directory for an app in ASEC container"

* commit '2e06258c4588b5c381277ad0269c00f7d05d1016':
  Do not create oat directory for an app in ASEC container

9 years agoPass charset to XmlPullParser.setInput instead of null
Wojciech Staszkiewicz [Fri, 8 May 2015 13:58:46 +0000 (14:58 +0100)]
Pass charset to XmlPullParser.setInput instead of null

Passing null to XmlPullParser.setInput forces it to do additional
work, which can be easily avoided if we know the charset beforehand.

bug: b/20849543

(cherry picked from commit 9e9e2e73c6ec7bece20268196dc89ad0c8bafad4)

Change-Id: Iaff97be9df2d0f99d7af8f19f65934439c9658e2

9 years agoMerge "Fix application moves."
Narayan Kamath [Fri, 22 May 2015 08:30:07 +0000 (08:30 +0000)]
Merge "Fix application moves."

9 years agoFix application moves.
Narayan Kamath [Thu, 21 May 2015 09:50:35 +0000 (10:50 +0100)]
Fix application moves.

We don't dex2oat during application moves, so we must scan
the package again scanPackageDirtyLI to deduce its ABI.

This is unnecessary (since a move cannot change ABIs), but we
need some additional refactoring to avoid a second scan.

bug: 21337469

(cherry picked from commit cd251fa382e887f59c278d4d7cd0a858812c6114)

Change-Id: Id1ed3bdfabb41e05e6a2f7efbd05d2103a67c663

9 years agoMerge "Throw a checked exception on inconsistent installs."
Narayan Kamath [Fri, 22 May 2015 08:29:45 +0000 (08:29 +0000)]
Merge "Throw a checked exception on inconsistent installs."

9 years agoThrow a checked exception on inconsistent installs.
Narayan Kamath [Wed, 20 May 2015 10:11:55 +0000 (11:11 +0100)]
Throw a checked exception on inconsistent installs.

Allows us to proceed without crashing the system process. Also,
complete an incomplete error message. Follow up comments from
change b904863476991d8540d37d5.

bug: 21144503

(cherry picked from commit c52ae28006605fb2da38b305b60b9bb419aa2f80)

Change-Id: I47b8762449714162fcd5c16eba6fcc146b93120b

9 years agoMerge "Scan package to derive ABIs before optimizing."
Narayan Kamath [Fri, 22 May 2015 08:29:20 +0000 (08:29 +0000)]
Merge "Scan package to derive ABIs before optimizing."

9 years agoScan package to derive ABIs before optimizing.
Narayan Kamath [Mon, 18 May 2015 09:18:35 +0000 (10:18 +0100)]
Scan package to derive ABIs before optimizing.

This requires a minor refactor to extract the ABI detection logic
out of scanPackageDirtyLI.

Note that there's a minor regression here : we ignore the
cpuAbiOverride from the package settings when calculating the
CPU ABI. This is OK (and possibly better behaviour) because this
is only a debug only option (for adb install) AND because the instructions
require users to specify the abi override on every adb install
invocation. Furthermore, the behaviour when an ADB installed app
(with an override) is auto-updated is more consistent.

bug: 21144503

(cherry picked from commit b904863476991d8540d37d542c0a49b78deab680)

Change-Id: I1eb88b808fd2e90e14c32322131659220aafdb7a

9 years agoMerge "Do not create oat directory for an app in ASEC container"
Narayan Kamath [Fri, 22 May 2015 08:28:50 +0000 (08:28 +0000)]
Merge "Do not create oat directory for an app in ASEC container"

9 years agoDo not create oat directory for an app in ASEC container
Fyodor Kupolov [Wed, 22 Apr 2015 23:03:48 +0000 (16:03 -0700)]
Do not create oat directory for an app in ASEC container

Currently installation fails, if an app is located on SD card (inside ASEC).
If an app is located inside ASEC, dexopt output should go to
/data/dalvik-cache. dexopt also needs to be performed at a later stage,
because the name of the oat file contains the absolute path to the dex file.

Bug: 20452651
Bug: 21360107

(cherry picked from commit 94056d1cb8183bde3e942336735b289b9654deb1)

Change-Id: Id1ced7159e5af9c05be0b637437b3b260e859c1d

9 years agoam 3385b616: Merge "Added in flag to ask context to wait for debugger attach."
Stephen Hines [Thu, 21 May 2015 16:35:53 +0000 (16:35 +0000)]
am 3385b616: Merge "Added in flag to ask context to wait for debugger attach."

* commit '3385b6169f47fd663d893fc787173c2d57a0a05f':
  Added in flag to ask context to wait for debugger attach.

9 years agoMerge "Added in flag to ask context to wait for debugger attach."
Stephen Hines [Thu, 21 May 2015 16:23:36 +0000 (16:23 +0000)]
Merge "Added in flag to ask context to wait for debugger attach."

9 years agoam 2db531ee: (-s ours) Merge "[DO NOT MERGE] Allow compilation with Jack."
Yohann Roussel [Thu, 21 May 2015 09:53:38 +0000 (09:53 +0000)]
am 2db531ee: (-s ours) Merge "[DO NOT MERGE] Allow compilation with Jack."

* commit '2db531ee1b3a980ea9ee6c7e72ad6d11983500e2':
  [DO NOT MERGE] Allow compilation with Jack.

9 years agoMerge "[DO NOT MERGE] Allow compilation with Jack."
Yohann Roussel [Thu, 21 May 2015 09:23:38 +0000 (09:23 +0000)]
Merge "[DO NOT MERGE] Allow compilation with Jack."

9 years agoMerge "Clean RenderScript warnings"
Jean-Luc Brouillet [Wed, 20 May 2015 18:24:58 +0000 (18:24 +0000)]
Merge "Clean RenderScript warnings"

9 years agoClean RenderScript warnings
Jean-Luc Brouillet [Tue, 19 May 2015 21:59:42 +0000 (14:59 -0700)]
Clean RenderScript warnings

Change-Id: Ia7a98c1c977c839c74ae8081e70b5deb4b60c25d

9 years agoMerge "Allow creating a managed profile if there is only one user."
Fyodor Kupolov [Tue, 19 May 2015 17:47:02 +0000 (17:47 +0000)]
Merge "Allow creating a managed profile if there is only one user."

9 years agoMerge "[RenderScript] fix Allocation.copyToFieldPacker"
Miao Wang [Tue, 19 May 2015 16:45:07 +0000 (16:45 +0000)]
Merge "[RenderScript] fix Allocation.copyToFieldPacker"

9 years ago[DO NOT MERGE] Allow compilation with Jack.
Yohann Roussel [Mon, 8 Sep 2014 12:53:12 +0000 (14:53 +0200)]
[DO NOT MERGE] Allow compilation with Jack.

Adapt to Jack compilation specifics.

Fix bad import in VideoDumpActivity.
The extra semicolumn is ignored by javac because of
https://bugs.openjdk.java.net/browse/JDK-8027682 but is not
permitted by ecj or Jack.

This includes cherry-picks of the following changes:
248ef6f9c414db26fd1bc7ec3d4ee4f3b87d54cf
705555dc468e46560d02db102dfcedc1e5c70d9e
f50a51fb4c8d5ca973ce910a21bddf9f8466cb51
bf5ad8d9943c0e1b0b3f1b4524c277f369065d08

Change-Id: Ifbe063e3822a7d63b1681921b9788cfca898ba64

9 years agoMerge "Switch from getSpi to getCurrentSpi."
Alex Klyubin [Mon, 18 May 2015 19:08:32 +0000 (19:08 +0000)]
Merge "Switch from getSpi to getCurrentSpi."

9 years agoAllow creating a managed profile if there is only one user.
Nicolas Prevot [Wed, 13 May 2015 19:15:03 +0000 (12:15 -0700)]
Allow creating a managed profile if there is only one user.

BUG:21119929

Change-ID: Ice1cf25f8ae8199228f828d22118c94b9e11b567

9 years agoSwitch from getSpi to getCurrentSpi.
Alex Klyubin [Mon, 18 May 2015 16:52:29 +0000 (09:52 -0700)]
Switch from getSpi to getCurrentSpi.

Crypto primitives' getSpi has a side-effect which modifies the state
of the primitive: it selects an SPI implementation if it hasn't been
selected yet (e.g., Cipher.getInstance("AES") doesn't select an SPI
implementation until Cipher.init). The new method getCurrentSpi has
no side-effects: it simply returns null if no SPI implementation is
selected. The switch to getCurrentSpi lets us avoid side-effects and
throw a more pertinent exception when no SPI is yet selected.

Bug: 18088752
Change-Id: I6d62eb67f904a15782673d16bcdae240cdc8f979

9 years agoWindowManagerService not propagating X and Y steps
Richard MacGregor [Fri, 17 Apr 2015 16:26:36 +0000 (09:26 -0700)]
WindowManagerService not propagating X and Y steps

WindowManagerService received X and Y offset steps but failed to pass
them off to the relavent wallpaperservice.

Live wallpapers expect a valid value for xOffsetStep and yOffsetStep
when the WallpaperService.Engine::onOffsetsChanged function is called
As specified here:
https://developer.android.com/reference/android/service/wallpaper/WallpaperService.Engine.html
onOffsetsChanged(float,float, float, float, int, int)

See bug report:
https://code.google.com/p/android/issues/detail?id=173607

Change-Id: I532dedf2db055e27d6eca813e30346e37f52dc65
(cherry picked from commit e03ea68d0a7f7d7c60663feae587225cca3a3a5b)

9 years agoMerge "Cleanup Keystore API"
Chad Brubaker [Fri, 15 May 2015 19:54:32 +0000 (19:54 +0000)]
Merge "Cleanup Keystore API"

9 years agoMerge "Use the correct parent size to initialize animations"
Olawale Ogunwale [Fri, 15 May 2015 17:23:07 +0000 (17:23 +0000)]
Merge "Use the correct parent size to initialize animations"

9 years agoMerge "Fix lost singleton provider after force-stopping user or package."
Olawale Ogunwale [Fri, 15 May 2015 17:11:27 +0000 (17:11 +0000)]
Merge "Fix lost singleton provider after force-stopping user or package."

9 years agoMerge "Also restart provider if there is external handle."
Olawale Ogunwale [Fri, 15 May 2015 16:19:46 +0000 (16:19 +0000)]
Merge "Also restart provider if there is external handle."

9 years agoFix lost singleton provider after force-stopping user or package.
riddle_hsu [Fri, 15 May 2015 03:33:03 +0000 (11:33 +0800)]
Fix lost singleton provider after force-stopping user or package.

Case 1 (name == null):
Switch user from guest to owner. All processes of guest
will be killed, it will not include processes which singleton
components live in, but singleton provider records are still
collected and removed.

When the user switch is complete and the process of removed
singleton provider is still alive, there is someone access
the provider, it will create a new ContentProviderRecord and
wait but no one will notify it because the provider process
is alive with different ContentProviderRecord.

Then the access cannot get response unless the process of target
provider is died and restarted.

Case 2 (name != null):
Switch to another non-guest user, launch an application which
contains singleton provider. Go to Settings to force-stop the
package then switch back to owner user. Launch an application
which will access the singleton provider. It will also cannot
get response that similar as case 1.

Solution:
Only collect singleton provider if target user is all or owner.

Change-Id: Ic6828da66645172d1378cfb1f66d092df5966516

9 years agoCleanup Keystore API
Chad Brubaker [Tue, 12 May 2015 22:19:52 +0000 (15:19 -0700)]
Cleanup Keystore API

Rename confusingly named methods, add userID arguments to all methods
that operate on user state and delete methods that have been replaced by
the onUser* methods.

Some of the old methods have been kept in KeyStore.java in order to ease
the transition of various system packages to the new methods.

Change-Id: Ic271689d62c36d255c5adee26c7abc2e7ed24df5

9 years agoAdded in flag to ask context to wait for debugger attach.
Stephen McGroarty [Fri, 8 May 2015 14:56:58 +0000 (15:56 +0100)]
Added in flag to ask context to wait for debugger attach.

Adds in flag CREATE_FLAG_WAIT_FOR_ATTACH with value 0x0008 to the RenderScript API
to be passed down to the C++ implementation to tell it to wait for a debugger to be
attached before executing the kernel.

Change-Id: Ibc4e903efbed1b1fb14cf378d1a8517d5c8d1f26
Signed-off-by: Stephen McGroarty <stephen@codeplay.com>
9 years agoMerge "AlarmClockInfo: fix redundant word in doc comment"
Chris Tate [Thu, 14 May 2015 01:20:52 +0000 (01:20 +0000)]
Merge "AlarmClockInfo: fix redundant word in doc comment"

9 years agoAlarmClockInfo: fix redundant word in doc comment
Shuhrat Dehkanov [Wed, 13 May 2015 08:16:29 +0000 (17:16 +0900)]
AlarmClockInfo: fix redundant word in doc comment

Removed one of unintentional double "intent"s.

Change-Id: I0ce4224964afae0587ce55b6b4775370eff89d26
Signed-off-by: Shuhrat Dehkanov <k@efir.uz>
9 years agoMerge "Fix issues found during CTS testing of FP16"
Jason Sams [Thu, 14 May 2015 00:53:16 +0000 (00:53 +0000)]
Merge "Fix issues found during CTS testing of FP16"

9 years agoFix issues found during CTS testing of FP16
Jason Sams [Wed, 13 May 2015 20:21:30 +0000 (13:21 -0700)]
Fix issues found during CTS testing of FP16

bug 20822382

This corrects a copy&paste error placing F16 into F32.
Fixes an error disallowing vectors of fp16 types.

Change-Id: I34be1aa9a323d61a7121581ab9599d2c2d32dd42

9 years agoMerge "Replace JNI primitive array critical calls with non-critical ones."
Hiroshi Yamauchi [Wed, 13 May 2015 22:47:59 +0000 (22:47 +0000)]
Merge "Replace JNI primitive array critical calls with non-critical ones."

9 years agoReplace JNI primitive array critical calls with non-critical ones.
Hiroshi Yamauchi [Wed, 13 May 2015 20:10:33 +0000 (13:10 -0700)]
Replace JNI primitive array critical calls with non-critical ones.

The files generated by glgen + manually edited util.cpp.

(cherry pick commit 5b406cb7b548443b589f0acb235c8a7465ffb82b)

Bug: 19235243
Change-Id: I9ab79c36e4ac6e123296b06ee8e4c04c4f583104

9 years agoAlso restart provider if there is external handle.
riddle_hsu [Wed, 13 May 2015 10:59:03 +0000 (18:59 +0800)]
Also restart provider if there is external handle.

e.g. ContentResolver.getType will call
ActivityManagerService.getProviderMimeType
that will not have connection but increase
externalProcessNoHandleCount.

Change-Id: I649c0b2390a749c77c6be5e7dfadc1acb689ec4c

9 years agoMerge "Add keystore onUserAdded/Removed methods"
Chad Brubaker [Tue, 12 May 2015 23:26:34 +0000 (23:26 +0000)]
Merge "Add keystore onUserAdded/Removed methods"

9 years agoAdd keystore onUserAdded/Removed methods
Chad Brubaker [Tue, 12 May 2015 20:00:02 +0000 (13:00 -0700)]
Add keystore onUserAdded/Removed methods

Change-Id: I73fe9344ec5660e58425d5c85d14381820533d57

9 years agoMerge "Don't apply animation clip to dialog activities"
Olawale Ogunwale [Tue, 12 May 2015 16:48:27 +0000 (16:48 +0000)]
Merge "Don't apply animation clip to dialog activities"

9 years ago[RenderScript] fix Allocation.copyToFieldPacker
Miao Wang [Mon, 4 May 2015 22:29:25 +0000 (15:29 -0700)]
[RenderScript] fix Allocation.copyToFieldPacker

Change-Id: Iaba3333f9741238745fdb5c94f1b27b1015d50b2

9 years agoMerge "Cleanup keystore password changing and unlocking"
Chad Brubaker [Mon, 11 May 2015 19:30:40 +0000 (19:30 +0000)]
Merge "Cleanup keystore password changing and unlocking"

9 years agoMerge "[RenderScript] (minor) fix validation of L3 BLAS"
Miao Wang [Mon, 11 May 2015 19:20:47 +0000 (19:20 +0000)]
Merge "[RenderScript] (minor) fix validation of L3 BLAS"

9 years ago[RenderScript] (minor) fix validation of L3 BLAS
Miao Wang [Fri, 1 May 2015 22:32:24 +0000 (15:32 -0700)]
[RenderScript] (minor) fix validation of L3 BLAS

bug: 21028875

Change-Id: I9fd48c57b6518e06ee1d9d7d4ca44faa96d887b0

9 years agoCleanup keystore password changing and unlocking
Chad Brubaker [Thu, 7 May 2015 17:02:22 +0000 (10:02 -0700)]
Cleanup keystore password changing and unlocking

Add KeyStore.onUserPasswordChanged for the lockscreen to call when
the user changes their password. Keystore will then handle the logic of
deleting keys. Instead of calling Keystore.password_uid for both
unlocking and password changes the behavior has been split into
Keystore.unlock and onUserPasswordChanged.

(cherry-picked from commit a91a8504191d91d288c55821caa5bf00c9be26a2)

Change-Id: I324914c00195d762cbaa8c63084e41fa796b7df8

9 years agoMerge "Fix KeyStoreTest now that begin requires parameters."
Alex Klyubin [Mon, 11 May 2015 17:37:23 +0000 (17:37 +0000)]
Merge "Fix KeyStoreTest now that begin requires parameters."

9 years agoFix KeyStoreTest now that begin requires parameters.
Alex Klyubin [Mon, 11 May 2015 17:31:12 +0000 (10:31 -0700)]
Fix KeyStoreTest now that begin requires parameters.

Keystore's begin operation now requires parameters which describe the
operation (e.g., algorithm, block mode, padding). This adjusts
KeyStoreTest to provide the necessary parameters.

Bug: 19509156
Change-Id: I8205288d8cbdd37196acde32548a0994763be38c

9 years agoMerge "BootReceiver: scrub APANIC"
Mark Salyzyn [Fri, 8 May 2015 22:47:03 +0000 (22:47 +0000)]
Merge "BootReceiver: scrub APANIC"

9 years agoMerge "Fix print format error."
Chih-Hung Hsieh [Fri, 8 May 2015 22:22:22 +0000 (22:22 +0000)]
Merge "Fix print format error."

9 years agoMerge "Switch libjnigraphics.so to hash-style=both"
Dmitriy Ivanov [Fri, 8 May 2015 21:17:08 +0000 (21:17 +0000)]
Merge "Switch libjnigraphics.so to hash-style=both"

9 years agoSwitch libjnigraphics.so to hash-style=both
Dmitriy Ivanov [Fri, 8 May 2015 21:03:19 +0000 (14:03 -0700)]
Switch libjnigraphics.so to hash-style=both

Bug: http://b/19059885
Change-Id: I15f70b434b1d450e5b62ca7c204c628e2dc9edcc

9 years agoFix print format error.
Chih-Hung Hsieh [Fri, 8 May 2015 18:05:12 +0000 (11:05 -0700)]
Fix print format error.

Change-Id: I284c5e9c44f6dd56a7bb6ff93a1a86cc8f2deefa

9 years agoMerge "Moved limit constants for kernel and script group"
Yang Ni [Fri, 8 May 2015 17:41:25 +0000 (17:41 +0000)]
Merge "Moved limit constants for kernel and script group"

9 years agoUse the correct parent size to initialize animations
tiger_huang [Mon, 27 Apr 2015 12:36:46 +0000 (20:36 +0800)]
Use the correct parent size to initialize animations

The original logic would use out-of-date parent sizes to initialize
animations. If the screen size is changed after assigning mAnimDw and
mAnimDh in the constructor of WindowStateAnimator, the fromDeltaY
(in the most cases) of TranslateAnimation would be initialized
incorrectly.

In this change, we always use up-to-date parent sizes to initialize
animations to prevent the issue.

https://code.google.com/p/android/issues/detail?id=170348

Change-Id: Ib9c609121228934bdb463263feb1924eb389c1d2

9 years agoMerge "[RenderScript] Make L2 BLAS methods pulic field."
Miao Wang [Fri, 8 May 2015 00:58:03 +0000 (00:58 +0000)]
Merge "[RenderScript] Make L2 BLAS methods pulic field."

9 years ago[RenderScript] Make L2 BLAS methods pulic field.
Miao Wang [Fri, 8 May 2015 00:56:05 +0000 (17:56 -0700)]
[RenderScript] Make L2 BLAS methods pulic field.

Change-Id: Ia5f6a747c4998203f8155c67111aea70301af24a

9 years agoMerge "[RenderScript] typo fix in BLAS, TRMM & TSMM"
Miao Wang [Thu, 7 May 2015 23:29:46 +0000 (23:29 +0000)]
Merge "[RenderScript] typo fix in BLAS, TRMM & TSMM"

9 years agoMerge "[RenderScript] typo fix in rs BLAS jni, to correctly handle complex params."
Miao Wang [Thu, 7 May 2015 23:29:18 +0000 (23:29 +0000)]
Merge "[RenderScript] typo fix in rs BLAS jni, to correctly handle complex params."

9 years agoMerge "[RenderScript] fix edges cases about Transpose for SYMM & SYR2K"
Miao Wang [Thu, 7 May 2015 23:28:36 +0000 (23:28 +0000)]
Merge "[RenderScript] fix edges cases about Transpose for SYMM & SYR2K"

9 years agoMerge "[RenderScript] L2 BLAS, fix element type in ZHER"
Miao Wang [Thu, 7 May 2015 23:28:12 +0000 (23:28 +0000)]
Merge "[RenderScript] L2 BLAS, fix element type in ZHER"

9 years agoMerge "[RenderScript] fixes for L3 BLAS APIs"
Miao Wang [Thu, 7 May 2015 23:27:48 +0000 (23:27 +0000)]
Merge "[RenderScript] fixes for L3 BLAS APIs"

9 years agoMerge "[RenderScript] improve & minor fixes of L2 BLAS validation."
Miao Wang [Thu, 7 May 2015 23:27:25 +0000 (23:27 +0000)]
Merge "[RenderScript] improve & minor fixes of L2 BLAS validation."

9 years ago[RenderScript] typo fix in BLAS, TRMM & TSMM
Miao Wang [Fri, 1 May 2015 00:14:28 +0000 (17:14 -0700)]
[RenderScript] typo fix in BLAS, TRMM & TSMM

Change-Id: I78f014a864700369c7ccdfb46a257851a45abd7c

9 years ago[RenderScript] typo fix in rs BLAS jni, to correctly handle complex
Miao Wang [Thu, 30 Apr 2015 20:44:49 +0000 (13:44 -0700)]
[RenderScript] typo fix in rs BLAS jni, to correctly handle complex
params.

Change-Id: Ie44f430dc7ed4155f40aa9f9c2864fb0429c97fe

9 years ago[RenderScript] fix edges cases about Transpose for SYMM & SYR2K
Miao Wang [Thu, 30 Apr 2015 17:47:42 +0000 (10:47 -0700)]
[RenderScript] fix edges cases about Transpose for SYMM & SYR2K

Change-Id: Id86d5fe8475060a22f49b06b6e1f8f3eae7c7e9f

9 years ago[RenderScript] L2 BLAS, fix element type in ZHER
Miao Wang [Thu, 30 Apr 2015 01:14:55 +0000 (18:14 -0700)]
[RenderScript] L2 BLAS, fix element type in ZHER

Change-Id: I99f9f9ff8942522d6581ff60b9a2d75e33e66ff6

9 years ago[RenderScript] fixes for L3 BLAS APIs
Miao Wang [Fri, 24 Apr 2015 18:19:53 +0000 (11:19 -0700)]
[RenderScript] fixes for L3 BLAS APIs

  - Typo for validateUplo
  - Typo in ZHEMM, element should be FLOAT64_2.
  - For GEMM and SYMM, SYRK, 'CONJ_TRANSPOSE' should also be handled in the
    validation process.
  - For SYMM, check matrix A is symmetric.
  - For HERK, the dimension validation was switched for Transpose case.
    Also, only Conj Trans is allowed in this case.
  - FOR SYR2K, fix the dimension check for Matrix C.
  - For TRMM & TRSM, fix the validation part for dimension check.

Change-Id: I559b5c5695aa82604de2955ae2327b694236d3ed

9 years ago[RenderScript] improve & minor fixes of L2 BLAS validation.
Miao Wang [Thu, 23 Apr 2015 22:06:09 +0000 (15:06 -0700)]
[RenderScript] improve & minor fixes of L2 BLAS validation.

Change-Id: If8dd0f9d4c7db03df22763a80fa6d600539be7c1

9 years agoMerge "[RenderScript] Fix HEMM and SYRK in IntrinscBLAS to correctly handle complex...
Miao Wang [Thu, 7 May 2015 22:37:08 +0000 (22:37 +0000)]
Merge "[RenderScript] Fix HEMM and SYRK in IntrinscBLAS to correctly handle complex 'Alpha' and 'Beta'"

9 years ago[RenderScript] Fix HEMM and SYRK in IntrinscBLAS to correctly handle
Miao Wang [Wed, 22 Apr 2015 22:57:57 +0000 (15:57 -0700)]
[RenderScript] Fix HEMM and SYRK in IntrinscBLAS to correctly handle
complex 'Alpha' and 'Beta'

Change-Id: I0aed57d41b0199d5c6dd3c31ca33644a5634dc35

9 years agoMerge "Fix testAuthNeeded test"
Chad Brubaker [Thu, 7 May 2015 19:32:15 +0000 (19:32 +0000)]
Merge "Fix testAuthNeeded test"

9 years agoFix testAuthNeeded test
Chad Brubaker [Thu, 7 May 2015 18:31:32 +0000 (11:31 -0700)]
Fix testAuthNeeded test

begin now returns OP_AUTH_REQUIRED for per operations with per op
authorization instead of NO_ERROR.

Change-Id: I1f472125f46155833e03ab30bf18363ff51b2c58

9 years agoUpdate asan/app_process.
Evgenii Stepanov [Thu, 7 May 2015 17:55:59 +0000 (10:55 -0700)]
Update asan/app_process.

Build both 32 and 64-bit versions and don't limit it to eng.

Change-Id: I5d6cc8005291100d5fe0f1385e8a0b4ba5f4d630