OSDN Git Service

android-x86/frameworks-base.git
8 years agoMerge "Don't use timestamps with all zero"
Eino-Ville Talvala [Tue, 17 Nov 2015 01:55:38 +0000 (01:55 +0000)]
Merge "Don't use timestamps with all zero"

8 years agoMerge "Insert Android Keystore JCA Provider at the correct position."
Alex Klyubin [Mon, 16 Nov 2015 20:14:43 +0000 (20:14 +0000)]
Merge "Insert Android Keystore JCA Provider at the correct position."

8 years agoInsert Android Keystore JCA Provider at the correct position.
Alex Klyubin [Sat, 14 Nov 2015 01:36:45 +0000 (17:36 -0800)]
Insert Android Keystore JCA Provider at the correct position.

Security.insertProviderAt uses 1-based positions whereas the
AndroidKeyStoreProvider.install code was incorrectly passing in
0-based positions, thus installing the AndroidKeyStoreBCWorkaround
provider one level higher than intended. This change fixes the issue
in AndroidKeyStoreProvider.

Bug: 25399691
Change-Id: I4a66bf37c0d151edb9a2349db9d91939064c0574

8 years agoMerge "Revert "jni: is loggable speedup, not within signal""
Mark Salyzyn [Mon, 16 Nov 2015 16:30:57 +0000 (16:30 +0000)]
Merge "Revert "jni: is loggable speedup, not within signal""

8 years agoRevert "jni: is loggable speedup, not within signal"
Mark Salyzyn [Mon, 16 Nov 2015 16:24:52 +0000 (16:24 +0000)]
Revert "jni: is loggable speedup, not within signal"

This reverts commit c954fde0b6e31ec99509faf0bbc33c4c9e4d012a.

Bug: 25693940
Change-Id: I568051aad39e215d677318cd5758962e562187ef

8 years agoMerge "Remove obsolete profile handling from AndroidRuntime."
Calin Juravle [Mon, 16 Nov 2015 11:08:06 +0000 (11:08 +0000)]
Merge "Remove obsolete profile handling from AndroidRuntime."

8 years agoMerge "Various fixes in setting globals in a script group"
Yang Ni [Fri, 13 Nov 2015 21:11:25 +0000 (21:11 +0000)]
Merge "Various fixes in setting globals in a script group"

8 years agoVarious fixes in setting globals in a script group
Yang Ni [Tue, 10 Nov 2015 21:27:04 +0000 (13:27 -0800)]
Various fixes in setting globals in a script group

Bug: 25602504

1) Passing floating point values into a script group was broken,
since they were casted to long values. Fixed that in the frameworks
implementation by taking the raw bits instead.

2) Passing 64-bit values into a script group was broken on 32-bit
platforms, since they were casted to pointer-sized integers
(uintptr_t) in the JNI code. Fixed that by casting to int64_t
instead.

3) Setting global variables of Allocation type in a script group was
broken. The special size value -1 was used to indicate the value is an
Allocation. However, size was casted to size_t in the JNI code.
Fixed that by using signed integers.

Change-Id: Ifff099a76be7707df7b67c388395f5a00f9cae66

8 years agoMerge "jni: is loggable speedup, not within signal"
Mark Salyzyn [Fri, 13 Nov 2015 15:16:48 +0000 (15:16 +0000)]
Merge "jni: is loggable speedup, not within signal"

8 years agoDon't use timestamps with all zero
Michal Piechowski [Wed, 15 Apr 2015 11:03:50 +0000 (13:03 +0200)]
Don't use timestamps with all zero

Although all zero (0000:00:00) is valid time, in most cases it means
that value is not present. According to http://www.exiv2.org/Exif2-2.PDF
in such case those values should be omitted, however
some cameras set them to 0 anyway. With this commit such timestamps
will be treated as they were empty.

Change-Id: I9c762b1fa04ea6bf9c0fba9e2459a20430c71c90

8 years agoMerge "Add hidden support for arrays of raw file descriptors"
Casey Dahlin [Thu, 12 Nov 2015 23:50:05 +0000 (23:50 +0000)]
Merge "Add hidden support for arrays of raw file descriptors"

8 years agojni: is loggable speedup, not within signal
Mark Salyzyn [Tue, 10 Nov 2015 01:56:42 +0000 (17:56 -0800)]
jni: is loggable speedup, not within signal

Speedup by telling __android_is_loggable interface it will not
be called within a signal handler. This saves two system calls.

Bug: 25563384
Change-Id: Ib8188641c8c76e5aa848476da98596be6b8e5040

8 years agoAdd hidden support for arrays of raw file descriptors
Casey Dahlin [Thu, 5 Nov 2015 20:19:13 +0000 (12:19 -0800)]
Add hidden support for arrays of raw file descriptors

Change-Id: I4013e0700369764a26485d8620ebf16d8bea1951
Test: Built and ran Android in an emulator
Bug: 25242023
Signed-off-by: Casey Dahlin <sadmac@google.com>
8 years agoMerge "Support X509TrustManagerExtensions methods"
Chad Brubaker [Thu, 12 Nov 2015 19:28:54 +0000 (19:28 +0000)]
Merge "Support X509TrustManagerExtensions methods"

8 years agoMerge "Use duck typing in X509TrustManagerExtensions"
Chad Brubaker [Thu, 12 Nov 2015 19:28:21 +0000 (19:28 +0000)]
Merge "Use duck typing in X509TrustManagerExtensions"

8 years agoMerge "System crash fixed when turn off MagnificationGesture"
Svetoslav Ganov [Thu, 12 Nov 2015 16:00:28 +0000 (16:00 +0000)]
Merge "System crash fixed when turn off MagnificationGesture"

8 years agoMerge "Avoid configuration of callback being reset after relayout."
Wale Ogunwale [Thu, 12 Nov 2015 15:55:18 +0000 (15:55 +0000)]
Merge "Avoid configuration of callback being reset after relayout."

8 years agoAvoid configuration of callback being reset after relayout.
riddle_hsu [Thu, 12 Nov 2015 06:07:12 +0000 (14:07 +0800)]
Avoid configuration of callback being reset after relayout.

mPendingConfiguration is a parameter of IWindowSession.relayout.
And IWindowSession.aidl declared "out Configuration outConfig",
it will always create a new configuration for remote side to write.
If remote side does not write (WMS does not have config change),
the new default configuration will be returned.

In original code passes mPendingConfiguration to updateConfiguration
directly, then callbacks (sConfigCallbacks) receive the same
instance of mPendingConfiguration. And because the implementation
of callback may use the configuration after relayout has reset
the configuration to default, then it may have timing that results
"showing hybrid of portrait and landscape modes" which try to fix
in commit e36d6e27.

To avoid this, always create a copy to updateConfiguration.
MSG_RESIZED_REPORT from dispatchResized also did the same thing.

Related commit:
e36d6e277e49475076b7872d36ea6a5c5b996e9d
694f79b5d1196640d1beb680b7d1fc68e6e77cbd

Change-Id: Ic1abd596e384918224b3a7020583d9a04641cccc

8 years agoSupport X509TrustManagerExtensions methods
Chad Brubaker [Tue, 10 Nov 2015 23:50:08 +0000 (15:50 -0800)]
Support X509TrustManagerExtensions methods

Change-Id: I14a405e90f139b8d73eb9f88597fac804a7c18f3

8 years agoUse duck typing in X509TrustManagerExtensions
Chad Brubaker [Tue, 10 Nov 2015 23:40:26 +0000 (15:40 -0800)]
Use duck typing in X509TrustManagerExtensions

X509TrustManagerExtensions assumes that the default X509TrustManager is
an instance of conscrypt's TrustManagerImpl. That's no longer going to
always be the case. Instead use duck typing to support any
X509TrustManagers that have the extra methods required for
X509TrustManagerExtensions.

Change-Id: If23471bda590d5e131bb1e802a60599957bc7f37

8 years agoMerge "Add NetworkSecurityConfigProvider"
Chad Brubaker [Wed, 11 Nov 2015 20:25:04 +0000 (20:25 +0000)]
Merge "Add NetworkSecurityConfigProvider"

8 years agoMerge "Fix system watchdog timeout when reading too many usage events."
Amith Yamasani [Wed, 11 Nov 2015 20:24:39 +0000 (20:24 +0000)]
Merge "Fix system watchdog timeout when reading too many usage events."

8 years agoMerge "Add support for debug-overrides configuration"
Chad Brubaker [Wed, 11 Nov 2015 19:03:30 +0000 (19:03 +0000)]
Merge "Add support for debug-overrides configuration"

8 years agoAdd NetworkSecurityConfigProvider
Chad Brubaker [Tue, 10 Nov 2015 20:26:18 +0000 (12:26 -0800)]
Add NetworkSecurityConfigProvider

Change-Id: I321e3ca94cc2a8d5e0e5d82a83b255ff5b8a71d2

8 years agoMerge "Fix memory corruption in SpotShadow"
John Reck [Wed, 11 Nov 2015 17:52:08 +0000 (17:52 +0000)]
Merge "Fix memory corruption in SpotShadow"

8 years agoMerge "Parse method-trace properties only in debug builds."
Calin Juravle [Wed, 11 Nov 2015 16:54:52 +0000 (16:54 +0000)]
Merge "Parse method-trace properties only in debug builds."

8 years agoFix memory corruption in SpotShadow
Keith Mok [Wed, 11 Nov 2015 00:52:05 +0000 (16:52 -0800)]
Fix memory corruption in SpotShadow

The array list is too small and causing stack corruption

Change-Id: I0e34dad39357fb63977d2ce6f183ced7b6a632be

8 years agoParse method-trace properties only in debug builds.
Calin Juravle [Tue, 10 Nov 2015 19:30:45 +0000 (19:30 +0000)]
Parse method-trace properties only in debug builds.

Bug: 25612377
Change-Id: Ia1d8e6c5afbdd70486d5aa8c059e70e27b03367d

8 years agoMerge "Fix HWUI Path Cache dangling pointer"
John Reck [Tue, 10 Nov 2015 23:20:03 +0000 (23:20 +0000)]
Merge "Fix HWUI Path Cache dangling pointer"

8 years agoAdd support for debug-overrides configuration
Chad Brubaker [Mon, 9 Nov 2015 21:38:51 +0000 (13:38 -0800)]
Add support for debug-overrides configuration

Debug overrides are only used if the application is debuggable in
order to help local debugging and development by trusting additional
CAs. In a non-debuggable version of the application the debug-overrides
are ignored.

Trust anchors in the debug override configuration have two key
differences from those in base-config and domain-config:
1) trust anchors in the debug-overrides are trusted for all connections
in addition to any trust anchors included in the relevant base/domain
configs.
2) By default trust anchors in the debug config override pins, as their
purpose is for connecting to non-standard servers for debugging and
testing and those servers should not be pinned in the production
configuration.

Change-Id: I15ee98eae182be0ffaa49b06bc5e1c6c3d22baee

8 years agoMerge "Avoid NPE in printspooler when changing language"
Svetoslav Ganov [Tue, 10 Nov 2015 00:58:28 +0000 (00:58 +0000)]
Merge "Avoid NPE in printspooler when changing language"

8 years agoMerge "system_server: allow /proc read access to other UIDs"
Nick Kralevich [Mon, 9 Nov 2015 20:05:22 +0000 (20:05 +0000)]
Merge "system_server: allow /proc read access to other UIDs"

8 years agoFix HWUI Path Cache dangling pointer
Digish Pandya [Wed, 4 Nov 2015 05:30:28 +0000 (11:00 +0530)]
Fix HWUI Path Cache dangling pointer

When precache, PathTexture is added to PathCache, and it is released after drawn if we want to clean it.
But the PathCache LRU still holds the entry of the PathTexture object. When trim the cache in
the end of each frame, LRU finds that its mListener is not NULL and invoke the functor, however,
mListerer points to the released PathTexture object and is a dangling pointer, thus leads to crash.
Smart pointer don't help here since they only manage scopes, while PathTexture is also controled by
its cleanup field.
The fix is to also remove the LRU entry of PathTexture*, it will also release the texture object
and there won't be texture leaks.

Change-Id: Iaa0621df5dc71532e9e75b38ad94384353930b95

8 years agoFix system watchdog timeout when reading too many usage events.
Mark Lu [Fri, 6 Nov 2015 07:26:15 +0000 (15:26 +0800)]
Fix system watchdog timeout when reading too many usage events.

Symptom:
As issue link:
https://code.google.com/p/android/issues/detail?id=193100

RootCause:
UsageStatsService.onDisplayChanged executed in system main thread,
If calling UsageStatsManager.queryEvents with too many events before it
(i.e. in daily usage event file,
too many activity resume / pause events or configuration change
during monkey test),
System will be blocked then watchdog timeout.

Solution:
Let display listener executed in background thread handler

Change-Id: Ic894d112612400ed8fb7ba843b3309fdc4f66fe1

8 years agosystem_server: allow /proc read access to other UIDs
Nick Kralevich [Sun, 8 Nov 2015 01:04:04 +0000 (17:04 -0800)]
system_server: allow /proc read access to other UIDs

Add system_server to AID_READPROC, to allow system_server to read
/proc entries associated with other UIDs.

Bug: 23310674
Change-Id: I1602b8ee30670cde4ed56e2f901524c105895352

8 years agoMerge "Support nested domain-config elements"
Chad Brubaker [Sat, 7 Nov 2015 22:11:04 +0000 (22:11 +0000)]
Merge "Support nested domain-config elements"

8 years agoMerge "Add xml source for network security configuration"
Chad Brubaker [Sat, 7 Nov 2015 22:10:50 +0000 (22:10 +0000)]
Merge "Add xml source for network security configuration"

8 years agoSupport nested domain-config elements
Chad Brubaker [Sat, 7 Nov 2015 07:02:37 +0000 (23:02 -0800)]
Support nested domain-config elements

Nested domain-config inherit unset parameters from the domain-config
they are nested in. This helps avoid copy and pasted configs that are
almost the same except a few minor differences for a domain with
slightly different requirements.

For example: Consider a domain-config for example.com that, among other
settings, does not enforce hsts. Now if you want the rules for
example.com to apply to secure.example.com except that hsts _is_
enforced you can make a nested domain-config for secure.example.com
under example.com that sets hstsEnforced="true" and nothing else.

Change-Id: I9e33f7e62127fd7f4f15c3560fff2f2626477bd4

8 years agoAdd xml source for network security configuration
Chad Brubaker [Thu, 5 Nov 2015 07:55:29 +0000 (23:55 -0800)]
Add xml source for network security configuration

XmlConfigSource parses an ApplicationConfig from an xml resource.
Currently this supports app-wide default configuration via the
base-config element, per domain via the domain-config element and
inheritance of unset properties at parse time.

Inheritance of unset properties is currently only:
domain-config -> base-config -> platform default configuration
Where the most specific value is used.
For example: If the base-config specifies trust anchors, all connections
will use those anchors except for connections to a domain which has a
domain-config that specifies trust anchors, in which case the
domain-config's trust anchors will be used. If the domain-config or
base-config don't set trust anchors, or don't exist, then the platform
default trust anchors will be used.

Nested domain-config entries, debug-overrides, and thorough
documentation of the xml format will follow in later commits.

Change-Id: I1232ff1e8079a81b340bc12e142f0889f6947aa0

8 years agoMerge "Fix incorrect hstsEnforced lookup"
Chad Brubaker [Sat, 7 Nov 2015 02:37:45 +0000 (02:37 +0000)]
Merge "Fix incorrect hstsEnforced lookup"

8 years agoFix incorrect hstsEnforced lookup
Chad Brubaker [Sat, 7 Nov 2015 02:08:12 +0000 (18:08 -0800)]
Fix incorrect hstsEnforced lookup

Change-Id: I00364d8c1d3311bebd3ad88e557860e694ba5018

8 years agoMerge "Fix directory check in UserCertificateSource"
Chad Brubaker [Fri, 6 Nov 2015 21:02:23 +0000 (21:02 +0000)]
Merge "Fix directory check in UserCertificateSource"

8 years agoMerge "Fix incorrect hasPerDomainConfigs check"
Chad Brubaker [Fri, 6 Nov 2015 21:02:01 +0000 (21:02 +0000)]
Merge "Fix incorrect hasPerDomainConfigs check"

8 years agoFix directory check in UserCertificateSource
Chad Brubaker [Fri, 6 Nov 2015 20:52:12 +0000 (12:52 -0800)]
Fix directory check in UserCertificateSource

If the user has not added any CAs to the user trust store the user-added
directory will not have been created.

Change-Id: I8b5f73af3c0761c56969874231004fedbf7badda

8 years agoFix incorrect hasPerDomainConfigs check
Chad Brubaker [Fri, 6 Nov 2015 20:50:31 +0000 (12:50 -0800)]
Fix incorrect hasPerDomainConfigs check

Change-Id: Iaeedbbcc5f12475d346a77dba84b38a1d3d8d346

8 years agoMerge "Use a builder for NetworkSecurityConfig"
Chad Brubaker [Fri, 6 Nov 2015 20:21:29 +0000 (20:21 +0000)]
Merge "Use a builder for NetworkSecurityConfig"

8 years agoSystem crash fixed when turn off MagnificationGesture
Masaaki Iwaguchi [Fri, 6 Nov 2015 11:38:27 +0000 (12:38 +0100)]
System crash fixed when turn off MagnificationGesture

Symptom:
WindowManagerService throws exception and system crashes.

Root cause:
When the MagnificationGesture is turned off during animation of scale
up/down, WindowManagerService.mAccessibilityController can be null so
exception is thrown.

Solution:
Cancel ongoing animation before turning off MagnificationGesture.

Change-Id: I469d84e024bdd453e3729a69229398233ed2eb61

8 years agoMerge "Take into account default app for default grants"
Svetoslav Ganov [Fri, 6 Nov 2015 00:34:21 +0000 (00:34 +0000)]
Merge "Take into account default app for default grants"

8 years agoTake into account default app for default grants
Svetoslav [Fri, 6 Nov 2015 00:33:02 +0000 (16:33 -0800)]
Take into account default app for default grants

Change-Id: Ib78c7dbda65d405bbf74544cad18e5e10bfcdeec

8 years agoMerge "Revert "Take into account default app for default grants""
Svetoslav Ganov [Fri, 6 Nov 2015 00:24:59 +0000 (00:24 +0000)]
Merge "Revert "Take into account default app for default grants""

8 years agoRevert "Take into account default app for default grants"
Svetoslav Ganov [Fri, 6 Nov 2015 00:24:47 +0000 (00:24 +0000)]
Revert "Take into account default app for default grants"

Apparently OEMs can set the chooser so this patch is not enough since it assumes the chooser is a platform component.

This reverts commit fce33d58d87fd494686aae4e21332b58ec260b73.

Change-Id: I3b8e499d2f0a731a227d193f6c5d6ed528be2272

8 years agoMerge "When the incoming light source is invalid, don't generate any shadow"
Tenghui Zhu [Thu, 5 Nov 2015 23:56:56 +0000 (23:56 +0000)]
Merge "When the incoming light source is invalid, don't generate any shadow"

8 years agoWhen the incoming light source is invalid, don't generate any shadow
Teng-Hui Zhu [Thu, 5 Nov 2015 00:57:53 +0000 (16:57 -0800)]
When the incoming light source is invalid, don't generate any shadow

b/25417885

Change-Id: I4b87e35ca68091fd0409cb9fe9b9400af860a507

8 years agoUse a builder for NetworkSecurityConfig
Chad Brubaker [Thu, 5 Nov 2015 07:36:07 +0000 (23:36 -0800)]
Use a builder for NetworkSecurityConfig

The builder supports all the standard builder set* methods as well as
setting a parent builder to use when values are not set (recursively).
This allows us to have a level of inheretence in configurations without
complicating the lookup and trust checking logic by doing inheretence
when building the configs.

Change-Id: I054af83451e52761227479eadf9cb9803437505f

8 years agoMerge "Use std::unique_ptr instead of UniquePtr."
Dan Albert [Thu, 5 Nov 2015 19:32:43 +0000 (19:32 +0000)]
Merge "Use std::unique_ptr instead of UniquePtr."

8 years agoMerge "Add initial network security config implementation"
Chad Brubaker [Thu, 5 Nov 2015 18:46:24 +0000 (18:46 +0000)]
Merge "Add initial network security config implementation"

8 years agoMerge "Fixes for sm and pm usage"
Brian Carlstrom [Thu, 5 Nov 2015 18:38:56 +0000 (18:38 +0000)]
Merge "Fixes for sm and pm usage"

8 years agoFixes for sm and pm usage
Brian Carlstrom [Thu, 5 Nov 2015 18:33:40 +0000 (10:33 -0800)]
Fixes for sm and pm usage

sm showed IllegalArgumentException after usage.
pm inconsistently used "return showUsage()"

Bug: 20948199
Change-Id: I5828489a9d4ef2eabfb97dd408d66e1560a5a983

8 years agoMerge "Don't apply clang flag to gcc builds."
Dan Albert [Thu, 5 Nov 2015 18:18:45 +0000 (18:18 +0000)]
Merge "Don't apply clang flag to gcc builds."

8 years agoAvoid NPE in printspooler when changing language
Masaaki Iwaguchi [Fri, 3 Apr 2015 07:11:39 +0000 (16:11 +0900)]
Avoid NPE in printspooler when changing language

PrintActivity is recreated when the language setting is changed.
As a result of the recreation mCurrentPrinter is null, which causes
a NullPointerException when onActivityResult() is called.

The issue is solved with a simple null check.

Change-Id: Ic58f0ca01577d2ec02d494739f3bc2f06240dd44

8 years agoDon't apply clang flag to gcc builds.
Dan Albert [Thu, 5 Nov 2015 08:59:09 +0000 (00:59 -0800)]
Don't apply clang flag to gcc builds.

For whatever reason GCC doesn't fail the build on unknown flags unless
there is also a warning emitted. Since we're now deprecating
UniquePtr, the warning is thrown here and the build fails (despite not
being -Werror).

Bug: http://b/22403888
Change-Id: I4af7911a51f7af9aa47a596a76f3e9d79535585b

8 years agoUse std::unique_ptr instead of UniquePtr.
Dan Albert [Thu, 5 Nov 2015 08:57:12 +0000 (00:57 -0800)]
Use std::unique_ptr instead of UniquePtr.

We're deprecating UniquePtr, so we need to move to the real thing.

Bug: http://b/22403888
Change-Id: I5b7fdf4924dd8c12a8c7dba89278714ca6fdc60a

8 years agoMerge "Take into account default app for default grants"
Svetoslav Ganov [Thu, 5 Nov 2015 02:19:12 +0000 (02:19 +0000)]
Merge "Take into account default app for default grants"

8 years agoMerge "Enforce weaker grant if app is default handler for miltiple action"
Svetoslav Ganov [Thu, 5 Nov 2015 02:19:00 +0000 (02:19 +0000)]
Merge "Enforce weaker grant if app is default handler for miltiple action"

8 years agoEnforce weaker grant if app is default handler for miltiple action
Svetoslav [Tue, 3 Nov 2015 02:40:18 +0000 (18:40 -0800)]
Enforce weaker grant if app is default handler for miltiple action

Change-Id: Ibc46e458e277412ebb276eb5ba6f2765b6b6c5e9

8 years agoAdd initial network security config implementation
Chad Brubaker [Fri, 23 Oct 2015 22:33:56 +0000 (15:33 -0700)]
Add initial network security config implementation

Initial implementation of a unified application wide static
network security configuration.

This currently encompases:
* Trust decisions such as what trust anchors to use as well as static
  certificate pinning.
* Policy on what to do with cleartext traffic.

In order to prevent issues due to interplay of various components in an
application and their potentially different security requirements
configuration can be specified at a per-domain granularity in addition
to application wide defaults.

This change contains the internal data structures and trust management
code, hooking these up in application startup will come in a future
commit.

Change-Id: I53ce5ba510a4221d58839e61713262a8f4c6699c

8 years agoMerge "Built-in mic is not used when USB mic is disconnected"
Glenn Kasten [Wed, 4 Nov 2015 16:03:41 +0000 (16:03 +0000)]
Merge "Built-in mic is not used when USB mic is disconnected"

8 years agoBuilt-in mic is not used when USB mic is disconnected
Alejandro Ochoa [Sat, 31 Oct 2015 02:05:00 +0000 (20:05 -0600)]
Built-in mic is not used when USB mic is disconnected

In some cases when the USB microphone is disconnected,
audio stack does not switch to the built-in microphone.
It gets stuck in a state where it still recognizes the
USB mic is still connected. Current device removal
implementation only considers USB output devices such
as headset. The same process should be used for input
USB devices (microphone).

Bug: 24932354
Change-Id: Ic2089ef5a9a318cb47336ade405f79eccd7129f8
Signed-off-by: Alejandro Ochoa <alejandro.ochoa@intel.com>
8 years agoRemove obsolete profile handling from AndroidRuntime.
Calin Juravle [Wed, 4 Nov 2015 14:54:37 +0000 (14:54 +0000)]
Remove obsolete profile handling from AndroidRuntime.

Bug: 24698874
Change-Id: I61193800e8b1f285b0a8edd05a0430a3c779e5b1

8 years agoMerge "[ActivityManager] Fix NullPointerException in startNextMatchingActivity."
Wale Ogunwale [Tue, 3 Nov 2015 15:13:24 +0000 (15:13 +0000)]
Merge "[ActivityManager] Fix NullPointerException in startNextMatchingActivity."
am: 7810aed854

* commit '7810aed85408bddc4cc0c65e5be8d082c638d995':
  [ActivityManager] Fix NullPointerException in startNextMatchingActivity.

8 years agoMerge "[ActivityManager] Fix NullPointerException in startNextMatchingActivity."
Wale Ogunwale [Tue, 3 Nov 2015 15:09:32 +0000 (15:09 +0000)]
Merge "[ActivityManager] Fix NullPointerException in startNextMatchingActivity."

8 years ago[ActivityManager] Fix NullPointerException in startNextMatchingActivity.
joanne_chung [Mon, 2 Nov 2015 04:41:34 +0000 (12:41 +0800)]
[ActivityManager] Fix NullPointerException in startNextMatchingActivity.

Symptom:
NullPointerException crash while call startNextMatchingActivity.

Root Cause:
Activity starts with FLAG_DEBUG_LOG_RESOLUTION for debug dump.
If the current Activity is the last object of resolves list,
  we can not get the next Activity and aInfo is still null.

Solution:
Check null in debug log dump.

Change-Id: I4c5d14d9acd8d47768b36c95354763841579e050

8 years agoTake into account default app for default grants
Svetoslav [Tue, 3 Nov 2015 00:41:53 +0000 (16:41 -0800)]
Take into account default app for default grants

Use the correct API to query activities when granting
default permissions to ensure the preferred apps are
taken into account. This allows OEMs to customize which
app should get default permission grants if there is
more than one resolving to the query intent.

Change-Id: I047f22dfff773a2ba9de1a0144442e372607777c

8 years agoMerge "BluetoothMidiDevice: change write type to support JamStik"
Glenn Kasten [Mon, 2 Nov 2015 23:06:04 +0000 (23:06 +0000)]
Merge "BluetoothMidiDevice: change write type to support JamStik"
am: 6079b990f1

* commit '6079b990f1461011a9dedb47cc65db1ce062e199':
  BluetoothMidiDevice: change write type to support JamStik

8 years agoMerge "BLE-MIDI: change binding for BluetoothMidiService"
Glenn Kasten [Mon, 2 Nov 2015 23:04:42 +0000 (23:04 +0000)]
Merge "BLE-MIDI: change binding for BluetoothMidiService"
am: 5831944a6e

* commit '5831944a6e99eff746e70d6ee9532ec3e3f32918':
  BLE-MIDI: change binding for BluetoothMidiService

8 years agoMerge "MidiBluetoothService: add new device to HashMap"
Glenn Kasten [Mon, 2 Nov 2015 23:03:18 +0000 (23:03 +0000)]
Merge "MidiBluetoothService: add new device to HashMap"
am: 9b07ba4bc3

* commit '9b07ba4bc33319d6db8de173a24eb13b2d7f0b50':
  MidiBluetoothService: add new device to HashMap

8 years agoMerge "MidiManager: fix double BTLE callback"
Glenn Kasten [Mon, 2 Nov 2015 23:01:49 +0000 (23:01 +0000)]
Merge "MidiManager: fix double BTLE callback"
am: bbf34d0bd1

* commit 'bbf34d0bd1bc4f85c9af614291925f5da1fa0459':
  MidiManager: fix double BTLE callback

8 years agoMerge "MidiService: update listener with current status when registered"
Glenn Kasten [Mon, 2 Nov 2015 23:00:15 +0000 (23:00 +0000)]
Merge "MidiService: update listener with current status when registered"
am: d4e9b9e4ac

* commit 'd4e9b9e4ac4196d37b0a5e59867f5a51018f4087':
  MidiService: update listener with current status when registered

8 years agoMerge "BluetoothMidiDevice: change write type to support JamStik"
Glenn Kasten [Mon, 2 Nov 2015 22:51:29 +0000 (22:51 +0000)]
Merge "BluetoothMidiDevice: change write type to support JamStik"

8 years agoMerge "BLE-MIDI: change binding for BluetoothMidiService"
Glenn Kasten [Mon, 2 Nov 2015 22:51:18 +0000 (22:51 +0000)]
Merge "BLE-MIDI: change binding for BluetoothMidiService"

8 years agoMerge "MidiBluetoothService: add new device to HashMap"
Glenn Kasten [Mon, 2 Nov 2015 22:51:12 +0000 (22:51 +0000)]
Merge "MidiBluetoothService: add new device to HashMap"

8 years agoMerge "MidiManager: fix double BTLE callback"
Glenn Kasten [Mon, 2 Nov 2015 22:51:02 +0000 (22:51 +0000)]
Merge "MidiManager: fix double BTLE callback"

8 years agoMerge "MidiService: update listener with current status when registered"
Glenn Kasten [Mon, 2 Nov 2015 22:50:50 +0000 (22:50 +0000)]
Merge "MidiService: update listener with current status when registered"

8 years agoBluetoothMidiDevice: change write type to support JamStik
Phil Burk [Tue, 22 Sep 2015 19:07:31 +0000 (12:07 -0700)]
BluetoothMidiDevice: change write type to support JamStik

Temporarily use a Request instead of a Command so that the JamStik
will be properly enabled.

Bug: 24207964
Change-Id: I3d11732958c46e6039405b5969d5af0db8032fb9
Signed-off-by: Phil Burk <philburk@google.com>
(cherry picked from commit 95129f50c68c734c5ebdf32ff6b8b9c63cc1ada7)

8 years agoBLE-MIDI: change binding for BluetoothMidiService
Phil Burk [Thu, 3 Sep 2015 21:37:03 +0000 (14:37 -0700)]
BLE-MIDI: change binding for BluetoothMidiService

Add a remote call addBluetoothDevice() using AIDL.
This was needed because onBind() is only called once.

Bug: 23219556
Bug: 23760886
Change-Id: Id7554ca55d596352d11dbd6ae3e403138a29c864
Signed-off-by: Phil Burk <philburk@google.com>
(cherry picked from commit 7cd06c0b9e087a555d2c5dd4cab5b7eac8497526)

8 years agoMidiBluetoothService: add new device to HashMap
Phil Burk [Fri, 21 Aug 2015 22:39:10 +0000 (15:39 -0700)]
MidiBluetoothService: add new device to HashMap

The HashMap is being used but devices never added to the map.

Bug: 23429459
Change-Id: I1da5305a56f5bd48e1c5d9345e721ea8dd2eed4e
Signed-off-by: Phil Burk <philburk@google.com>
(cherry picked from commit e5f722e55c5708c2b48b8c9688c135e451225032)

8 years agoMidiManager: fix double BTLE callback
Phil Burk [Fri, 21 Aug 2015 22:22:35 +0000 (15:22 -0700)]
MidiManager: fix double BTLE callback

Stop openBluetoothDevice() from calling its callback twice.

Bug: 23218338
Change-Id: I2e738b22a124c0336c46bef05680a82ee1f59e79
Signed-off-by: Phil Burk <philburk@google.com>
(cherry picked from commit 0ca998d79fe95ded26ec4ff0a1e0a8dc4802684b)

8 years agoMidiService: update listener with current status when registered
Phil Burk [Thu, 6 Aug 2015 06:32:54 +0000 (23:32 -0700)]
MidiService: update listener with current status when registered

This will give an app an opportunity to learn whether an input port is busy
before the user tries to connect and then fails.

Bug: 22825043
Change-Id: Ifede60f166dfe66ea15453044fce06f4a8452b18
Signed-off-by: Phil Burk <philburk@google.com>
(cherry picked from commit b2355940e3e697deeecd1b6511cd5094f9dc553a)

8 years agoMerge "[ActivityManager] Fix IndexOutOfBoundsException"
Wale Ogunwale [Tue, 27 Oct 2015 15:19:56 +0000 (15:19 +0000)]
Merge "[ActivityManager] Fix IndexOutOfBoundsException"
am: a61515720a

* commit 'a61515720ab95f6aa307c1fa4337c5c2078bad26':
  [ActivityManager] Fix IndexOutOfBoundsException

8 years agoMerge "[ActivityManager] Fix IndexOutOfBoundsException"
Wale Ogunwale [Tue, 27 Oct 2015 15:12:56 +0000 (15:12 +0000)]
Merge "[ActivityManager] Fix IndexOutOfBoundsException"

8 years ago[ActivityManager] Fix IndexOutOfBoundsException
louis_chang [Tue, 27 Oct 2015 02:45:53 +0000 (10:45 +0800)]
[ActivityManager] Fix IndexOutOfBoundsException

Symptom:
The stack was deleted during cleanup, so the stack
list size has reduced by one and cause this issue.

IndexOutOfBoundsException: Invalid index 1, size is 1

Change-Id: Ic6722aed98a0ad789b331d00e89cb74bf3b36d4c

8 years agoMerge "SystemUI: Reload the EndNowButton label"
Julia Reynolds [Mon, 26 Oct 2015 15:05:34 +0000 (15:05 +0000)]
Merge "SystemUI: Reload the EndNowButton label"
am: 40b52bbe48

* commit '40b52bbe480436fbdc3bfb8a1a70c6144e71ff8f':
  SystemUI: Reload the EndNowButton label

8 years agoMerge "SystemUI: Reload the EndNowButton label"
Julia Reynolds [Mon, 26 Oct 2015 14:57:34 +0000 (14:57 +0000)]
Merge "SystemUI: Reload the EndNowButton label"

8 years agoSystemUI: Reload the EndNowButton label
Johan Redestig [Fri, 23 Oct 2015 16:23:40 +0000 (18:23 +0200)]
SystemUI: Reload the EndNowButton label

Reload the EndNowButton label when configuration
changes.

Change-Id: I042d31f0bb011996aa8219c3c412e7b71987244c

8 years agoMerge "Adjust for changes to libcore/Docs.mk"
Neil Fuller [Mon, 26 Oct 2015 10:11:23 +0000 (10:11 +0000)]
Merge "Adjust for changes to libcore/Docs.mk"
am: f5e3398596

* commit 'f5e33985964c3e7acf6b9cd16d94a110e4f9a42d':
  Adjust for changes to libcore/Docs.mk

8 years agoMerge "Adjust for changes to libcore/Docs.mk"
Neil Fuller [Mon, 26 Oct 2015 10:04:30 +0000 (10:04 +0000)]
Merge "Adjust for changes to libcore/Docs.mk"

8 years agoMerge "Add Workman keyboard layout"
Kenny Root [Fri, 23 Oct 2015 23:39:24 +0000 (23:39 +0000)]
Merge "Add Workman keyboard layout"
am: 9a65c36022

* commit '9a65c36022482d7fd4217e537fa434cc79b69f39':
  Add Workman keyboard layout

8 years agoMerge "Add Workman keyboard layout"
Kenny Root [Fri, 23 Oct 2015 23:32:01 +0000 (23:32 +0000)]
Merge "Add Workman keyboard layout"

8 years agoMerge "[RenderScript] Add null check for return values of JNI Get<TYPE>ArrayElements."
Miao Wang [Fri, 23 Oct 2015 17:05:12 +0000 (17:05 +0000)]
Merge "[RenderScript] Add null check for return values of JNI Get<TYPE>ArrayElements."
am: 773756eeed

* commit '773756eeed21f67fa7f7eabc663df46c789129aa':
  [RenderScript] Add null check for return values of JNI Get<TYPE>ArrayElements.

8 years agoMerge "[RenderScript] Add null check for return values of JNI Get<TYPE>ArrayElements."
Miao Wang [Fri, 23 Oct 2015 17:00:31 +0000 (17:00 +0000)]
Merge "[RenderScript] Add null check for return values of JNI Get<TYPE>ArrayElements."