The method parsePackage failes to delete some resources
in case of failures.
Added delete of "package" in the cases that was previously
missed.
Change-Id: I183e9ec5864c5ed18bb48410ab41317cb3d96bda
Symptom: memory leakage
Root Cause: new SortedVector but not free it in native cpp
Solution: free the SortedVector
Project: AOSP
Note:
Change-Id: Iab5a7f2e8d8509631301e7231427927d4797c856
Issue: When the layer of previous frame is destroyed, it doesn't clear the
texture id in mBoundTextures[mTextureUnit], so in the next frame, if
glGenTexture returns same texture ID of the previous frame,
the new texture is not bound.
CRs-fixed: 671736
Change-Id: Ifc5fd2115fb1863b3c006ab14b0faabeaeb4eab4
Google optimized the uploadToTexture with GLES3.0+,
This allows us to upload a sub-rectangle of texture,
but this has compliance issue with GLES 2.0.
With OpenGL ES 2.0 we need to copy the bitmap in a
temporary buffer if the stride doesn't match the width,
and then upload entire temporary stripes.
Change-Id: I8987cd8347913c61b7fa45bd47be1189aa6bdb05
Signed-off-by: Shenghua Lu <shenghua.lu@intel.com>
Signed-off-by: Shuo Gao <shuo.gao@intel.com>
Signed-off-by: Guobin Zhang<guobin.zhang@intel.com>
When enabled defer rendering, it will do precache for DrawPathOp.
The paint used for creating PathTask in precache just get the address
of mFilteredPaint of OpenGLRenderer. So for the following defer
operation like DrawTextOp has possibility change the mFilteredPaint
by getPaint while another WorkerThread in PathCache is using the paint
which pointed to the same address of mFilteredPaint to generate bitmap.
As a result, it will generate a wrong bitmap for generateTexture in
PathCache. To fix it, do a copy of paint when creating PathTask.
CRs-Fixed: 664244
Change-Id: I5516f5b143458b88d3573d15b7ebb34f688800c7
the correct value to copy in bitmap is alpha component from
the cachebuffer. currently it is offset to red channel if format is
RGBA which is wrong.
this is followup fix for https://android-review.googlesource.com/#/c/93943/
Change-Id: I96b5ba35df5ccaef44caf2542d35d6585ba1df80
Signed-off-by: Digish Pandya <digishp@codeaurora.org>
fixes the glyph cachebuffer index to account for pixelbuffer
format size.
issue: in launcher home screen create icon group and rename the
group to some text string with emo-icons (smilies). without this
change the drop shadows for emo-icons will look bad.
Change-Id: I525a9219d35f4541363fd64e7fc5fd1053269725
Signed-off-by: Digish Pandya <digishp@codeaurora.org>
Once the RS::init failed, go through the original single thread path,
this will avoid blocking the main thread(ANR), but we also should have
some implement in RenderScript component.
Change-Id: I3a21395ffd5bc144c15254893f425b1dd2643ba4
Signed-off-by: Shenghua Lu <shenghua.lu@intel.com>
Signed-off-by: Shuo Gao <shuo.gao@intel.com>
Declare a new method, Display.getState() to retrieve the actual
power state of a display.
Improved documentation for Intent.ACTION_SCREEN_ON and
Intent.ACTION_SCREEN_OFF to clarify what they really mean in
terms of the interactive state of the device.
Deprecated PowerManager.isScreenOn() and replaced it with
PowerManager.isInteractive() with a more suggestive name and
better documentation.
Redirect display power state changes to go through the display
manager first and only then head over to the power manager for
legacy compatibility.
Eliminated the bright here and woke here policy flags since they
were unused. Simplified the input dispatch policy somewhat.
Ensure that screen wake locks are respected up until the point
when dozing really begins.
Fixed a regression in DreamService where onDreamingStarted
might be called before onWindowAttached.
Bug: 13133142
Bug: 13472578
Bug: 13929355
Bug: 13760290
Change-Id: Iabef96921dd554ce3768fb18619cefc3230b5fb0
The method is in libhwui.so but used by libandroid_runtime.so
(via android_view_GLES20Canvas.cpp). If the compiler decides not to
inline this method, the linker won't be able to resolve the dependency
when it has hidden visibility.
The problem is found when testing GCC 4.9.
Change-Id: I3b4d096c7767f0bef0796886dfb2c5ba2ecff48c
BUTTON_[1-9] are intended to be misc. buttons so the kernel will
assign them to non-gamepad devices. As such, we shouldn't treat them
as belonging to gamepads at all.
Bug: 13432364
Change-Id: I9495760f2379e44785b9ce1cf018b6db7f6cb0f4
* commit 'd020fd33d147c88a9cda4976d2fa559e55b4ab03':
Generate and respect ACTION_CANCEL for joystick fallbacks. DO NOT MERGE
Adds API for determining confirm and cancel keys.
Changes in this patch include
[x] Use %zu for size_t, %zd for ssize_t
[x] Some minor changes have been done to conform with
standard JNI practice (e.g. use of jint instead of int
in JNI function prototypes)
Change-Id: Id1aaa7894a7d0b85ac7ecd7b2bfd8cc40374261f
Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
Signed-off-by: Craig Barber <craig.barber@arm.com>
Signed-off-by: Kévin PETIT <kevin.petit@arm.com>
Signed-off-by: Marcus Oakland <marcus.oakland@arm.com>
Also change MotionEvent.PointerCoords bit packing and unpacking
methods to be consistent with BitSets which are now used on the
native PointerCoords object.
Bug: 11480300
Change-Id: Ib18c99b94ac555104c69eac526860aa501e89e03
