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
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>
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
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These members have been replaced by uint32_t offsets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
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>
Using reinterpret_cast allows conversions between pointer types
and integers which don't have the same size. The corresponding
flags is -Wint-to-pointer-cast.
Change-Id: I8ff0c79d235fa6d07b8d9305edc185d946d5133d
Signed-off-by: Kévin PETIT <kevin.petit@arm.com>
Changes in this patch include
[x] Long(64-bit) is used to store native pointers in
AssetAtlasService and related classes as they can be 64-bit.
[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: Ib4c77c134e3ad5b21732e20cde9a54a0b16bdab1
Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
TinyHashMap used hash_t(key) to generate hashcode. This
would not work for 64-bit pointers as hash_t is declared as
an uint32_t.
Replaced the hash_t(key) call to more generic android::hash_type(key).
This function is a template function declared in TypeHelpers.h and
has a version available for all data types including pointers.
Change-Id: I612cf18b49ca7c30b63f9d6938df68fed7d80d08
Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
For positioning of glyphs within a bitmap, roundf(int + float) is used,
where the float is the glyph position and the int is the text position.
When the text position is varied, this may lead to the sum being rounded
in different directions, due to floating point rounding, caused by that
floating point numbers have different precision in different ranges.
This may therefore lead to slightly different positioning for glyphs and
therefore slightly different widths and heights for text strings,
depending on the position they are rendered at.
The solution in this patch is to use int + (int) roundf(float), which
has consistent rounding, and also enables us to use the full range of
ints.
Change-Id: Id1143cdfcbdfa9915ced878ae04df589a3e03cee
mBitmap and mTexture is not initialized to NULL which causes
illegal address access when it fails to be created from
oversized bitmap.
Change-Id: Iea54bec8788bc5f3a10040fdb43f416c0d41a14c
Fixes: This patch makes sure that the layer's alpha value is
up-to-date and does not reflect the previous view's alpha value.
it fixes the square block on marquee fading edge when text view
is applied transparency.
Change-Id: I110649db72a75fecc02ba56c201a05773f80328c
bug:10863163
This fixes two issues
The check for pure translation was incorrect. It was fixed and renamed
for clarity.
Certain matrix paths weren't setting kTypePositiveScale. For
simplicity (and because positive scale is simple to check) removed
flag in favor of dynamic checking.
Change-Id: Ic5ce235653ef49a68b8b242bd89fc2e95874ecc9
I guess we don't want to overwrite the last line in every RGBA glyph
with our one-line texture atlas spacer?
Bug: 10841207
Change-Id: Ief85ca58650c731e9d21dbf90942b7b44670abcc
bug:10761696
Avoids a case where a rect with top coordinate of (e.g.) 0.51f is
assumed to not draw in the first row of pixels, which leads to it not
being clipped. Since rounding can cause it to render in this first
pixel anyway, we very slightly expand geometry bounds.
Now, in ambiguous cases, the geometry bounds are expanded so clipping
is more likely to happen.
Change-Id: I119b7c7720de07bac1634549724ffb63935567fc
bug:9969358
Instead of storing DeferredDisplayState within an op (thus forcing ops
to be tied to a single state instance), associate each op with a new
state at DeferredDisplayList insertion time.
Now, DisplayLists (and the ops within) can be reused in a single
DeferredDisplayList draw call, as ops will use different state
instances at different points in the frame.
Change-Id: I525ab2abe0c3883679f2fa00b219b293e9ec53d9
Bug: 10677765
enableScissor() must precede setScissorFromClip() as
otherwise setScissorFromClip() doesn't do anything.
Also make sure to call setScissorFromClip() if
enableScissor() returns true as enableScissor() calls
resetScissor() if the scissor state has changed.
Change-Id: I9226b20bb256c92066aae344e4e6407540b6eae9
Bug #10347089
If an app clears its path cache before stopping background tasks, it could
get into an infinite loop in PathCache::trim().
Change-Id: Ieb865b762e7b00aebaba0c023769c2db286a94f5
Oops! kBW_Format was omitted from a couple of switch statements,
resulting in glyphs in that format being invisible.
Bug: 10206452
Change-Id: Ib2aa52250aeeecc0de1b1b78e3d0f568f368c73e
Also remove dead code from OpenGLRenderer.cpp
Change-Id: I7eb54ca19e77ee3c32f1fe9513a031e6b2e115cf
(cherry picked from commit 5c7d5ab878b26f855175a3305a14ac12fcacf25e)
Bug #10185769
The assets atlas contains assets that need to be blended and assets
that do not need to be blended. With a single merge id, currently
set to be the pointer to the atlas itself, draw ops merging could
generate batches of commands containing both opaque and translucent
assets. The blend state was chosen from only one of the assets in
the batch, leading either to inefficiencies (blending large opaque
assets) or incorrect behaviors (not blending translucent assets.)
This change introduces two new merge ids in the atlas: an opaque
key and a blend key. These keys are simple booleans set to false
and true respectively (the values do not matter really.) Their
memory addresses are used as the merge ids when createing draw ops
batches, allowing all opaque ops to be batched together and all
translucent ops to be batched together.
Change-Id: I114dba0533c44987e53864b471ccb28c811f2025
