OSDN Git Service
Nick Kralevich [Wed, 26 Mar 2014 15:41:14 +0000 (15:41 +0000)]
Merge "Note libselinux dependency on packages.list format changes."
Stephen Smalley [Wed, 26 Mar 2014 13:19:12 +0000 (09:19 -0400)]
Note libselinux dependency on packages.list format changes.
Change-Id: I3c34a86f5706c4fca826a8634936131e4e4fc297
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Chris Craik [Tue, 25 Mar 2014 20:11:33 +0000 (20:11 +0000)]
Merge "Solve three memory leaks related to PatchCache"
Kenny Root [Tue, 25 Mar 2014 17:06:54 +0000 (17:06 +0000)]
Merge "Use X509ExtendedTrustManager and not Conscrypt"
Craig Mautner [Tue, 25 Mar 2014 17:01:27 +0000 (17:01 +0000)]
Merge "Don't wait for finishing animation when second display is disconnected"
Narayan Kamath [Tue, 25 Mar 2014 13:56:51 +0000 (13:56 +0000)]
Merge "Fix scaleNinePatch."
Koji Fukui [Fri, 14 Mar 2014 12:20:21 +0000 (21:20 +0900)]
Don't wait for finishing animation when second display is disconnected
Basically WindowManagerService wait for finishing animation when
a window is removed. But when second display is disconnected, windows
on second display can't be shown even if animation is waited for.
On the contrary, it keeps on waiting for finishing the animation
in special case.
With this fix windows are immediately removed without waiting for
animation when second display is disconnected.
Change-Id: I1354c193c04db394a21a11c174e10c8e7da17a0e
Jens Gulin [Tue, 4 Feb 2014 16:38:02 +0000 (17:38 +0100)]
Solve three memory leaks related to PatchCache
A Patch can be fairly large, holding bitmap data, but
is also frequently leaked which adds to the severity.
The feature is used in many important processes such
as Home, SystemUI and Chrome.
The following leaks are solved:
1. The Patch itself was not always freed.
PatchCache::removeDeferred() can mark patches to be
cared for by PatchCache::clearGarbage(). But
mCache.remove() would only destroy the container
and the pointer, not the Patch object itself.
2. The vertices stored in the Patch at Patch::createMesh()
would always leak. The empty/default destructor in Patch
would not properly destroy "vertices" since it's just a
pointer.
3. A BufferBlock that's added to the mFreeBlocks
in PatchCache could leak. The leak happened when a
patch later needed the entire free block, because the
object was removed from the list but never deleted
in PatchCache::setupMesh().
Change-Id: I41e60824479230b67426fc546d3dbff294c8891f
Kenny Root [Mon, 24 Mar 2014 17:18:18 +0000 (10:18 -0700)]
Use X509ExtendedTrustManager and not Conscrypt
This allows the new X509ExtendedTrustManager to be used instead of
Conscrypt directly.
Bug:
13103812
Change-Id: I736f475342395f26657f2371828a6d3a0065a50a
Craig Mautner [Mon, 24 Mar 2014 14:31:38 +0000 (14:31 +0000)]
Merge "Set ScreenshotSurface secure if any secure content is shown."
Narayan Kamath [Tue, 11 Mar 2014 16:50:30 +0000 (16:50 +0000)]
Fix scaleNinePatch.
Use yDivs for yDivs, and not xDivs.
bug:
13394494
(cherry picked from commit
42a51ae8812bccde7ff370cc2688f7955e489ad4)
Change-Id: Ia0a7d701a170945216ab247a483e7f972b6fe17e
Alex Klyubin [Fri, 21 Mar 2014 18:36:07 +0000 (18:36 +0000)]
Merge "Adjust to the change in the private SSLParameterImpl API."
Alex Klyubin [Fri, 21 Mar 2014 18:13:50 +0000 (11:13 -0700)]
Adjust to the change in the private SSLParameterImpl API.
Conscrypt's SSLParameterImpl getDefaultTrustManager was renamed to
getDefaultX509TrustManager.
Bug:
13563574
Change-Id: I601c651d631f5a2e4a04d21941186553988e5286
Colin Cross [Thu, 20 Mar 2014 17:28:11 +0000 (17:28 +0000)]
Merge "bootanimation: build 32-bit if 64-bit gl libraries are not available"
Narayan Kamath [Thu, 20 Mar 2014 10:08:37 +0000 (10:08 +0000)]
Merge "Don't register methods when there's nothing to register."
Minoru Aoi [Thu, 13 Mar 2014 03:00:34 +0000 (12:00 +0900)]
Set ScreenshotSurface secure if any secure content is shown.
The screenrecord command can record secure content during rotation.
This is because the ScreenshotSurface surface for rotation animation
is not set secure even though its screenshot contains secure content.
Change-Id: I5a70be71ddfc26de4230e7d4fe547210c2154210
Chris Craik [Thu, 20 Mar 2014 02:19:53 +0000 (02:19 +0000)]
Merge "Prevent Lockscreen album art from activating FLAG_KEEP_SCREEN_ON"
Nick Kralevich [Thu, 20 Mar 2014 01:37:38 +0000 (01:37 +0000)]
Merge "Get rid of noise during boot."
Dianne Hackborn [Thu, 20 Mar 2014 00:49:09 +0000 (17:49 -0700)]
Get rid of noise during boot.
This log is not an error, it is a warning, don't spam a stack
crawl when it happens.
Change-Id: I6038e3625cc0c16af9e54887b5e7ec451d9f864d
Colin Cross [Thu, 20 Mar 2014 01:14:43 +0000 (01:14 +0000)]
Merge "build app_process64 on 64-bit targets"
Colin Cross [Thu, 20 Mar 2014 01:09:01 +0000 (18:09 -0700)]
bootanimation: build 32-bit if 64-bit gl libraries are not available
Set LOCAL_32_BIT_ONLY if TARGET_32_BIT_SURFACEFLINGER is set.
Change-Id: I76396cc9fe3fe45d1e98a44bf911d6e70922555a
Colin Cross [Thu, 20 Mar 2014 00:47:39 +0000 (17:47 -0700)]
build app_process64 on 64-bit targets
On 64-bit platforms builds two copies of app_process, the 32-bit
app_process and 64-bit app_process64.
Change-Id: I59d739b5df398ad0bd040c954c57640ff7ab3e72
Glenn Kasten [Wed, 19 Mar 2014 21:02:38 +0000 (21:02 +0000)]
Merge "mediaeditor shouldn't depend on audioflinger."
Glenn Kasten [Wed, 19 Mar 2014 21:01:57 +0000 (21:01 +0000)]
Merge "Fix 64 bit compile failures in renderscript"
Nick Kralevich [Wed, 19 Mar 2014 20:17:40 +0000 (20:17 +0000)]
Merge "Allow PMS to restorecon directories under /data."
Mark Salyzyn [Wed, 19 Mar 2014 19:36:51 +0000 (19:36 +0000)]
Merge "androidfw: resolve 64-bit build issues"
Glenn Kasten [Wed, 19 Mar 2014 19:34:07 +0000 (19:34 +0000)]
Merge "PacProcessor should be 32 bit only."
Mark Salyzyn [Wed, 19 Mar 2014 18:00:06 +0000 (11:00 -0700)]
androidfw: resolve 64-bit build issues
- uid_t/gid_t cast to unsigned long
- unused argument warnings
- tab and space requirements
Change-Id: Ib446d8165b9082be02edb55e6b71fd1a03ea3431
Narayan Kamath [Wed, 19 Mar 2014 11:41:32 +0000 (11:41 +0000)]
mediaeditor shouldn't depend on audioflinger.
Change-Id: I7e1b999b03a3ee5eff2e10a460f64abeba16ba90
Robert Craig [Tue, 4 Mar 2014 16:57:23 +0000 (11:57 -0500)]
Allow PMS to restorecon directories under /data.
This change applies a relabel to both /data/data and
/data/user directories on boot. Not every boot will
apply this relabeling however. The appropriate
seapp_contexts is hashed and compared to
/data/system/seapp_hash to decide if the relabel
should occur.
Change-Id: I05e8b438950ddb908e46c9168ea6ee601e6d674f
Signed-off-by: rpcraig <rpcraig@tycho.ncsc.mil>
Narayan Kamath [Wed, 19 Mar 2014 10:15:51 +0000 (10:15 +0000)]
Fix 64 bit compile failures in renderscript
Use uintptr_t for pointers (and not uint32_t) and
also don't assume that size_t is 32 bits wide.
Also uses java longs for passing pointers across
JNI boundaries.
Change-Id: Ie52bd8ae967fbddc911eda3a43cc799d53bbce66
Narayan Kamath [Wed, 19 Mar 2014 14:13:27 +0000 (14:13 +0000)]
Don't register methods when there's nothing to register.
Change-Id: I6d0233f79bb90089426d702b4076ec5750e82475
Narayan Kamath [Wed, 19 Mar 2014 11:39:43 +0000 (11:39 +0000)]
PacProcessor should be 32 bit only.
It depends on external/v8, which isn't ready for 64 bit yet.
Change-Id: Ibc9705665b81d3667826dedd6217f2eb6d451c81
Dianne Hackborn [Tue, 18 Mar 2014 18:28:16 +0000 (18:28 +0000)]
Merge "Insufficient ProcessRecord cleanup when persistent process is killed"
Narayan Kamath [Tue, 18 Mar 2014 09:34:43 +0000 (09:34 +0000)]
Merge "Pass int32_t for JNI calls to java Input/Output streams."
Akira Numata [Thu, 13 Mar 2014 04:36:31 +0000 (13:36 +0900)]
Insufficient ProcessRecord cleanup when persistent process is killed
When persistent process with Service restarts, ActivityManagerService
does not reset ProcessRecord#hasClientActivites to false
(because ProcessRecord of persistent process is continued using
after killing).
It disturbs updating LRU list in ActivityManagerService, and then,
when new process calls ActivityManagerProxy#publishContentProviders,
SecurityException happens because of no entry in the list.
Bug:
13517358
Change-Id: I46b064f71a4f7025ade1bf117801352a7ab22e6a
Jeff Sharkey [Mon, 17 Mar 2014 16:32:05 +0000 (16:32 +0000)]
Merge "Null pointer exception in FileRotator.java"
Ashok Bhat [Wed, 5 Mar 2014 12:40:53 +0000 (12:40 +0000)]
Pass int32_t for JNI calls to java Input/Output streams.
Passing size_t is problematic on 64 bit platforms where
it's 8 bytes in size. Conversion to int32_t is safe because
the size argument is always clamped to fCapacity, which is
4 bytes wide.
Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
Change-Id: I58558561a4f56451485f1a5fc6cdeda677247071
Steve Moyer [Sun, 16 Mar 2014 17:26:43 +0000 (17:26 +0000)]
Merge "Don't attempt to launch disabled activities in smoke test."
Dave Allison [Fri, 14 Mar 2014 21:15:25 +0000 (21:15 +0000)]
Merge "Move options buffers to top scope"
Dave Allison [Fri, 14 Mar 2014 15:54:33 +0000 (08:54 -0700)]
Move options buffers to top scope
It is important that the char buffers for options do
not go out of scope as the mOptions.add() does not copy
the buffer. This moves all the buffers to the top
level scope of the function to prevent accidental
overwriting when they go out of scope.
Bug:
13448497
Change-Id: I5a97ddd32ff34f237915927906e1e1f833ff036e
Victoria Lease [Fri, 14 Mar 2014 17:48:00 +0000 (17:48 +0000)]
Merge "remove unnecessary switchUser call"
Narayan Kamath [Fri, 14 Mar 2014 14:12:35 +0000 (14:12 +0000)]
Merge "Fix layoutlib build."
Narayan Kamath [Fri, 14 Mar 2014 14:10:30 +0000 (14:10 +0000)]
Fix layoutlib build.
Was using private data structures from libcore.
Change-Id: I5bd43190081891659e18b5c871686e6078efcc81
Dave Platt [Thu, 13 Mar 2014 23:35:15 +0000 (23:35 +0000)]
Merge "NsdService does not clean up after exiting clients"
Dave Platt [Fri, 7 Mar 2014 22:48:22 +0000 (14:48 -0800)]
NsdService does not clean up after exiting clients
When a client of the NsdService exits, NsdService should
clean up the requests it has sent to the mDNS daemon:
cancel any pending resource-discovery and resource-resolution
queries, and remove any services registered by this client.
If this isn't done, several bad things happen. The daemon will
continue to run unnecessarily, will report service discoveries
that can't be forwarded on to the client, and will continue to
advertise service ports for an application which is no longer
running until the device is rebooted (mDNS pollution).
Bug:
9801184
Change-Id: I0aa7311480322aefcff16f902fbbf34f50985d38
Mikael Gullstrand [Fri, 13 Dec 2013 09:44:50 +0000 (10:44 +0100)]
Null pointer exception in FileRotator.java
Sometimes a null pointer exception is thrown when examining files
managed by the file rotator.
The logs from the test show that the Wifi state is changed a large
number of times. On every state change, a write operation is
initiated on the file system. This will eventually result in out
of memory and the call to mBasePath.list() in the maybeRotate(...)
method in FileRotator.java will return null so the iteration will
throw a NullPointerException.
Change-Id: I5d5980d9593bc9ec75281169ca27ee591809903f
Jianzheng Zhou [Thu, 10 Oct 2013 06:02:09 +0000 (14:02 +0800)]
remove unnecessary switchUser call
If userId doesn't change, no need to do switchuser.or it will
introduce some tweak for gps modules in quick off/on because
they need cost time in off/on.
Change-Id: Ie5e408dc8be69c8a63a18bcb2cfe478cef4d70e6
Signed-off-by: Jianzheng Zhou <jianzheng.zhou@freescale.com>
Deepanshu Gupta [Tue, 11 Mar 2014 21:22:39 +0000 (21:22 +0000)]
Merge "Fix the delegate method signatures in LayoutLib"
Narayan Kamath [Tue, 11 Mar 2014 15:09:44 +0000 (15:09 +0000)]
Merge "Remove java.io.tmpdir assignment in AndroidRuntime (again)."
Narayan Kamath [Tue, 11 Mar 2014 15:02:10 +0000 (15:02 +0000)]
Remove java.io.tmpdir assignment in AndroidRuntime (again).
This was brought back by a bad merge conflict resolution
in change
0efbd9a463c848118c7685f4bfc8765a82caa761.
Change-Id: I0c7cbe8ee396293619eabf4d0a3c2f06c76bdd6e
Narayan Kamath [Tue, 11 Mar 2014 12:00:54 +0000 (12:00 +0000)]
Merge "Remove stray logging statement."
Narayan Kamath [Tue, 11 Mar 2014 11:08:41 +0000 (11:08 +0000)]
Remove stray logging statement.
bug: https://code.google.com/p/android/issues/detail?id=66974
Change-Id: Ie34c102de143cf6aeb16c78778dda3ef5e928380
Jeff Brown [Tue, 11 Mar 2014 00:06:40 +0000 (00:06 +0000)]
Merge "Round off when calculate next step in Seekbar"
Jeff Brown [Tue, 11 Mar 2014 00:06:24 +0000 (00:06 +0000)]
Merge "Fixed typo errors in documentation"
Jeff Brown [Tue, 11 Mar 2014 00:06:01 +0000 (00:06 +0000)]
Merge "Hang off call by power button while screen is on"
Jeff Brown [Tue, 11 Mar 2014 00:04:49 +0000 (00:04 +0000)]
Merge "take account of RotationVector's new extra parameter"
Brian Carlstrom [Mon, 10 Mar 2014 17:44:30 +0000 (17:44 +0000)]
Merge "Only pass -Xprofile-* options to ART"
Brian Carlstrom [Mon, 10 Mar 2014 17:20:01 +0000 (10:20 -0700)]
Only pass -Xprofile-* options to ART
Bug:
13391896
Change-Id: I5d6a3b900c9b20f02e1d4ccb73f712e9260c7dfd
Narayan Kamath [Mon, 10 Mar 2014 10:43:57 +0000 (10:43 +0000)]
Merge "LP64: Make 9 patches architecture agnostic."
Narayan Kamath [Mon, 3 Mar 2014 17:12:03 +0000 (17:12 +0000)]
LP64: Make 9 patches architecture agnostic.
The Res_png_9patch struct had several pointer members
whose size differed between 32 and 64 bit platforms.
These members have been replaced by uint32_t offsets
to serialized data. The serialized form for 9patches
places a Res_png_9patch object at the beginning of
serialized data, followed by int32_t arrays of xDivs,
yDivs and colors.
Note that these offsets are not strictly required,
since they can be computed from the values of numXDivs,
numYDivs & numColors, however they are called in tight
loops so having them computed once is a beneficial.
This change also removed the unused patch_equals function
from aapt's Image.cpp.
Change-Id: I3b9ac8ae5c05510d41377cae4dff1c69b40c2531
Dave Allison [Fri, 7 Mar 2014 22:25:42 +0000 (22:25 +0000)]
Merge "ART profiler usage."
Dave Allison [Thu, 30 Jan 2014 22:19:51 +0000 (14:19 -0800)]
ART profiler usage.
This is a change to add args to some of the profiler related
functions, including installd commands.
Also read properties and set command line options for the runtime
profiling parameters.
Changed calls to isDexOptNeeded() to isDexOptNeededInternal(). This
needs additional arguments passed for profiles.
Bug:
12877748
Change-Id: I1a426c9309d760bac0cf92daa298defee62287c1
Conflicts:
core/jni/AndroidRuntime.cpp
Narayan Kamath [Fri, 7 Mar 2014 16:04:50 +0000 (16:04 +0000)]
Merge "Don't assume that size_t is 32-bit"
Narayan Kamath [Fri, 7 Mar 2014 16:04:27 +0000 (16:04 +0000)]
Merge "Don't use size_t for variables that store uint32_t value"
Narayan Kamath [Fri, 7 Mar 2014 13:50:03 +0000 (13:50 +0000)]
Merge "Inform libcore of time format pref. changes."
Narayan Kamath [Thu, 19 Dec 2013 14:49:36 +0000 (14:49 +0000)]
Inform libcore of time format pref. changes.
- Introduce a boolean extra for intent TIME_CHANGED that
specifies if the user wants a 24 hour format or not.
- Have the ActivityManagerService inform running processes
of changes to this preference.
- Add plumbing in ActivityThread to inform j.t.DateFormat
Change-Id: I05fafb903ae54e39c03a048b7a219dc5a93fd472
Martijn Coenen [Thu, 6 Mar 2014 14:33:10 +0000 (14:33 +0000)]
Merge "NFC: getNdefMessage() should throw exception if no tag in field."
Ashok Bhat [Wed, 5 Mar 2014 11:51:15 +0000 (11:51 +0000)]
Don't assume that size_t is 32-bit
Crypto data was being copied to java jint array in chunks of
size_t. This will not work on LP64 as size_t will be 64-bit.
This patch changes copy to use int32_t chunks instead of size_t.
Change-Id: I75d910a1182ad2f58f432cd172127f048b4c393b
Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
Ashok Bhat [Tue, 4 Mar 2014 21:13:52 +0000 (21:13 +0000)]
Don't use size_t for variables that store uint32_t value
The local variables width, height, displayWidth and
displayHeight are set to the corresponding members
of VideoFrame class. As the members of VideoFrame are
uint32_t, the local variables have been changed to
uint32_t.
As these local variables are passed to a java method,
the change of size_t to uint32_t ensures that the size
of the actual parameter (i.e uint32_t) is now same as
the size of the formal parameter (i.e. jint) for all
platforms.
Change-Id: Icd14de0142bfd4e6ba52a3e6aff3d80b323a0de4
Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
Deepanshu Gupta [Thu, 6 Mar 2014 03:23:48 +0000 (19:23 -0800)]
Fix the delegate method signatures in LayoutLib
Change-Id: I262c285c1047958afdb024724959d5afb3552802
Jason Sams [Thu, 6 Mar 2014 00:50:24 +0000 (00:50 +0000)]
Merge "Validate objects are from the correct context."
Jason Sams [Thu, 6 Mar 2014 00:09:02 +0000 (16:09 -0800)]
Validate objects are from the correct context.
Change-Id: I7d87b0e253b8d2e36d1aed790cfe3a7dd23e158f
David 'Digit' Turner [Wed, 5 Mar 2014 23:10:44 +0000 (23:10 +0000)]
Merge "aidl: Support building with updated host sysroot."
David 'Digit' Turner [Tue, 4 Mar 2014 15:43:41 +0000 (16:43 +0100)]
aidl: Support building with updated host sysroot.
This patch fixes tools/aidl/Type.cpp to compile with
a more recent SDK host toolchain sysroot (that has not
been submitted yet).
The main issue is that the code uses ssize_t which is
defined in <sys/types.h>, but didn't include the header
directly.
Apparently, this no longer compiles when using the Ubuntu
10.04 libc6-dev headers (the current SDK toolchain is
based on Ubuntu 8.04), so perform an explicit include
to fix this.
NOTE: This doesn't break the Windows SDK build.
Change-Id: Idcacf8f8d1c606b39bf8aa75ab409aa0c2a3cd5a
Brian Carlstrom [Wed, 5 Mar 2014 19:23:48 +0000 (19:23 +0000)]
Merge "Add dex2oat-flags plumbing to AndroidRuntime"
Brian Carlstrom [Sat, 1 Mar 2014 07:27:22 +0000 (23:27 -0800)]
Add dex2oat-flags plumbing to AndroidRuntime
Change-Id: Idaa7e5351e146d76e1972cbe4d93af69f0b999a6
Dave Platt [Tue, 4 Mar 2014 21:55:56 +0000 (21:55 +0000)]
Merge "Harden NsdManager against null-dereference crashes"
Dave Platt [Fri, 28 Feb 2014 00:16:20 +0000 (16:16 -0800)]
Harden NsdManager against null-dereference crashes
Due to race conditions or programming errors, the NsdManager
can attempt to process an asynchronous status message (and issue
a callback to the listener) after the listener has already been
removed from the NsdManager state. This causes dereferencing of
null objects, and a crash.
Split out the three async-queue message cases: these are ones
in which message.arg2 does not hold an NsdManager array index
and the code should not interpret this field as if it were.
Add an explicit check for "null listener" (the array index in the
message has already been released), log a warning, and exit early.
Safeguard accesses to the "NSD service type" string from a (possibly
null) NsdServiceInfo object... return a constant "?" string rather
than crashing.
Bug:
9016259
Change-Id: I40aabdfc65d86fdd0eaac7a1e7e56e6ff69796cf
Craig Mautner [Tue, 4 Mar 2014 15:19:17 +0000 (15:19 +0000)]
Merge "Avoid duplication of AppWindowToken in created Task"
Yevgen Pronenko [Wed, 19 Feb 2014 09:33:07 +0000 (10:33 +0100)]
Avoid duplication of AppWindowToken in created Task
When WindowManagerService creates a new Task, it passes atoken to the
Task constructor. In this case atoken is added to mAppTokens list by
the Task constructor and then it is added manually again by calling
newTask.mAppTokens.add(atoken). As a result, the same atoken is present
in mAppTokens list twice.
When another window token is added to the list, it may be placed in
beetwen duplicated tokens and corresponding window will be hidden from
the screen by corresponding duplicated windows.
From user perspective it means that some windows will not be displayed
on the screen.
The issue is fixed by avoid adding atoken to mAppTokens list manually
when it was added already in Task constructor.
Change-Id: I10628e68186160fffdde07beb7d84ab3cecb7051
Victoria Lease [Mon, 3 Mar 2014 23:38:53 +0000 (23:38 +0000)]
Merge "Add casual/cursive/sans-serif-smallcaps font families."
Tim Murray [Mon, 3 Mar 2014 19:09:07 +0000 (19:09 +0000)]
Merge "Fix null check for BaseObj.equals."
Victoria Lease [Fri, 28 Feb 2014 23:34:26 +0000 (15:34 -0800)]
Add casual/cursive/sans-serif-smallcaps font families.
Bug:
12384866
Change-Id: I7482d228279a63ea92cb605de314b2f7d0ad483f
Ying Wang [Sat, 1 Mar 2014 17:25:10 +0000 (17:25 +0000)]
Merge "Fix indirect misuse of $(my-dir)."
Ying Wang [Sat, 1 Mar 2014 02:04:37 +0000 (18:04 -0800)]
Fix indirect misuse of $(my-dir).
Change-Id: I20a98a8e1378da3a53e6a6d599c8fc0100f7f87a
Tim Murray [Sat, 1 Mar 2014 00:57:47 +0000 (16:57 -0800)]
Fix null check for BaseObj.equals.
Change-Id: I4f060697db175cbf0b5617c1220db853bbe53f1f
Nick Kralevich [Fri, 28 Feb 2014 21:02:20 +0000 (21:02 +0000)]
Merge "Fix descriptor leak after accepting connections"
Dave Platt [Fri, 6 Dec 2013 22:03:47 +0000 (14:03 -0800)]
Fix descriptor leak after accepting connections
After accepting a connection on a listening socket and
storing the resulting FileDescriptor into a newly created
LocalSocketImpl, the new impl's "descriptor was created
locally and should be closed normally" flag should be set.
(cherrypicked from
3e7305c6bf6062b5cb1e2ddcec6c6d30b4a8bc0d)
Bug:
11805817
Change-Id: I0a1a7c62ec3fbcf647a44a22110ddc778d14e5f4
Narayan Kamath [Fri, 28 Feb 2014 13:41:58 +0000 (13:41 +0000)]
Merge "Various format string fixes for 64bit"
Narayan Kamath [Fri, 28 Feb 2014 13:22:24 +0000 (13:22 +0000)]
Merge "AArch64: Use long[] for RS id array"
Xavier Ducrohet [Thu, 27 Feb 2014 23:28:13 +0000 (23:28 +0000)]
Merge "Update layoutlib method for SystemClock"
Christopher Tate [Thu, 27 Feb 2014 22:01:28 +0000 (22:01 +0000)]
Merge "Fixed upgrading from forward-lock application to system application"
Narayan Kamath [Thu, 27 Feb 2014 13:18:06 +0000 (13:18 +0000)]
Merge "AArch64: Make eglGetDisplay(int) work for EGL_DEFAULT_DISPLAY"
Narayan Kamath [Thu, 27 Feb 2014 13:13:08 +0000 (13:13 +0000)]
Merge "Make getAssetInt throw unconditionally."
Ashok Bhat [Wed, 26 Feb 2014 10:00:48 +0000 (10:00 +0000)]
AArch64: Make eglGetDisplay(int) work for EGL_DEFAULT_DISPLAY
Note that files changed in this patch have been automatically
generated by running frameworks/native/opengl/tools/glgen/gen
script
This will allow eglGetDisplay(int) to work on both 32-bit and
64-bit systems when EGL_DEFAULT_DISPLAY is passed as a parameter.
Change-Id: Idf27d6e00d623d331bb2d4d7f85fa450e0db26c9
Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
Chris Craik [Wed, 26 Feb 2014 18:48:27 +0000 (18:48 +0000)]
Merge "Fix AOSP build - DO NOT MERGE"
Chris Craik [Wed, 26 Feb 2014 18:47:59 +0000 (18:47 +0000)]
Merge "Fix graphics corruption caused by HWUI caches"
Chris Craik [Wed, 26 Feb 2014 18:44:33 +0000 (10:44 -0800)]
Fix AOSP build - DO NOT MERGE
Change-Id: I42b420c2ce89ce364a2809b28b827964e6923fa1
Sangkyu Lee [Thu, 9 Jan 2014 05:11:57 +0000 (14:11 +0900)]
Fix graphics corruption caused by HWUI caches
Some caches(PatchCache, TextureCache, PathCache) for HWUI
uses deferred removal for their cache entries even though
actual resource objects are immediately freed by
ResourceCache.
For this reason, the uniqueness of a resource address in
the caches is not guaranteed in specific cases.
(Because malloc() can return the same address when malloc()
and free() called very frequently.)
So it can be possible the cache have two cache entries for
two different resources but the same memory address.
(Of course one of the resources is already freed.)
It also can be possible mGarbage vector in PatchCache has
duplicated addresses and this can lead to duplicated free
blocks in the free block list and graphics corruption.
(Deferred removal was implmeneted based on an assumption of
unique resource addresses.)
So this patch makes sure resource objects are freed after
the resources are removed from the caches to guarantee
the uniqueness of a resource address and prevent graphics
corruption.
Change-Id: I040f033a4fc783d2c4bc04b113589657c36fb15b
Signed-off-by: Sangkyu Lee <sk82.lee@lge.com>