This change places BufferQueue into a predictable state where
allocation is allowed whenever a producer connects. This allows clients
to disconnect and reconnect without having to worry about being locked
out of allocation.
Bug: 20554276
Change-Id: Ic0f920a3d4204f2cafdfa69e46f3bb4204571d7e
This changes the way that SurfaceFlinger's shadow buffer management
works such that instead of tracking the size of the shadow queue in the
BufferQueue, SF tracks the last frame number it has seen, and passes
that into the acquireBuffer call. BufferQueueConsumer then ensures that
it never returns a buffer newer than that frame number, even if that
means that it must return PRESENT_LATER for an otherwise valid buffer.
Change-Id: I3fcb45f683ed660c3f18a8b85ae1f8a962ba6f0e
This completes the plumbing from ConsumerListener::onFrameReplaced into
SurfaceFlinger (and other consumers that may care).
Change-Id: I376e78ace95d6748e8662e6b4d47c0dfa697a300
Clears the frame number of a slot when it is freed, since it is used
to determine if a released buffer is stale.
Bug: 20445852
Change-Id: I02415e7b25a1eafe7414d6eb1cedf62ac5543cd9
SurfaceFlinger's (Layer's) shadow copy of the BufferQueue queue was
getting out of sync for a few reasons. This change fixes these by
doing the following:
- Adds a check to re-synchronize the shadow copy every time we
successfully acquire a buffer by first dropping stale buffers before
removing the current buffer.
- Avoids trying to perform updates for buffers which have been rejected
(for incorrect dimensions) by SurfaceFlinger.
- Adds IGraphicBufferConsumer::setShadowQueueSize, which allows the
consumer to notify the BufferQueue that it is maintaining a shadow
copy of the queue and prevents it from dropping so many buffers
during acquireBuffer that it ends up returning a buffer for which the
consumer has not yet received an onFrameAvailable call.
Bug: 20096136
Change-Id: I78d0738428005fc19b3be85cc8f1db498043612f
(cherry picked from commit 2e36f2283f)
Surface::{cancel,queue}Buffer and GraphicBufferMapper::lockAsyncYCbCr take the
ownership of |fenceFd|s. Though they don't close it on error cases.
Change-Id: I49a7ce8837d5c510c4ac4ad4649f310d18610e80
(cherry picked from commit c7263fb432)
GraphicBufferMapper::lockAsync{,YCbCr} close the fence FD even when the FD
is invalid.
Change-Id: Ia2b4dae3b2c06426e34f623f19ba92435f486ab7
(cherry picked from commit dcfe91e1f3)
Adds a NATIVE_WINDOW_BUFFER_AGE query, which returns the age of the
contents of the most recently dequeued buffer as the number of frames
that have elapsed since it was last queued.
Change-Id: Ib6fd62945cb62d1e60133a65beee510363218a23
(cherry picked from commit 49f810c72df8d1d64128e376757079825c8decd4)
Implements the GET_SIDEBAND_STREAM case of
IGraphicBufferConsumer::onTransact
Change-Id: I46138b87ce2d70dde48b3561dcd378ce5fd8f383
(cherry picked from commit 17166698d9)
Changes Surface::detachNextBuffer to return an sp<GraphicBuffer>
instead of an ANativeWindowBuffer* to ensure that reference counting
works correctly.
Bug: 20092217
Change-Id: I3979ea6121aaf14845f0554477b778770413581e
(cherry picked from commit 8b2daa3ca29492b181fb67840640d771c4a2b3ac)
Adds a new method IGBP::allowAllocation, which controls whether
dequeueBuffer is permitted to allocate a new buffer. If allocation is
disallowed, dequeueBuffer will block or return an error as it
normally would (as controlled by *ControlledByApp).
If there are free buffers, but they are not of the correct dimensions,
format, or usage, they may be freed if a more suitable buffer is not
found first.
Bug: 19801715
Change-Id: I0d604958b78b2fd775c2547690301423f9a52165
BufferQueue used to choose free buffers by scanning through its array
of slots and picking one based on timestamp. This changes that
mechanism to use a pair of free lists: one with buffers attached and
one without. This makes it easier to choose either type of free slot
depending on the needs of the current operation.
Fixes an issue with the first version of this change, found in bugs
20482952, 20443314, and 20464549.
Bug: 13175420
Change-Id: I9b6e83cfe8f9b4329a976025cb8e291d51fb6d4a
When the surface damage code went in, it incorrectly assumed that if
an application was doing CPU rendering, it would be using lock and
unlockAndPost instead of dequeue and queue, so it repurposed the dirty
region too aggressively. This change keeps it from clobbering the
dirty region if a CPU producer is attached.
Bug: 20431815
Change-Id: Id4dfd71378311ea822f0289f6de2d20a7bd84014
These tests hadn't been run in a while and had accumulated some rot.
This fixes most of the compile warnings and some tests which fail
incorrectly, but 2 failures remain on N6.
Change-Id: I7fee78cd058a32f4d7df40ec9f899ff767f68517
BufferItem and GraphicBuffer were not parceling correctly, which had
not been noticed because the libgui tests (specifically, one that
tests placing a BufferQueue in a separate process from the IGBP/C)
had not been run recently. This change fixes the errors found in
those classes.
Change-Id: Ie224361a534a79115a3481d83ff97f21d154d4f5
BufferQueue used to choose free buffers by scanning through its array
of slots and picking one based on timestamp. This changes that
mechanism to use a pair of free lists: one with buffers attached and
one without. This makes it easier to choose either type of free slot
depending on the needs of the current operation.
Bug: 13175420
Change-Id: Ic8398e7511bd11a60a1c82e3ad2ee271c9822be1
This change adds support for passing surface damage all of the way
down from the EGL interface through the consumer side of the
BufferQueue. Depends on system/core change
Ie645e6a52b37b5c1b3be19481e8348570d1aa62c
Bug: 11239309
Change-Id: I4457ea826e9ade4ec187f973851d855b7b93a31b
Fence was flattening incorrectly, causing issues with any fields
flattened after it, and Region was relying on being the last object
flattened. This change addresses both.
Change-Id: If49de061de6e2f11f4d846b4d8c18627dfb7e109
Will be used by the system_server watchdog to monitor the
availability of binder threads in the process to handle
incoming IPC requests.
Bug: 19297165
Change-Id: I39175f3869ad14da5620fddb47f454e6e4ee2b25
There is memory corruption in below code
const Rect* prev = &dst[prevIndex];
dst.add(Rect(prev->right, top, right, bottom));
prev points to a memory of vector dst, when dst resize in add()
call, the memory that prev points to will be copy to the new
allocated vector memory and the old memory will become undefined
Avoid pointer in this case, use a local copy instead
Change-Id: I4d95ceedd00c8fb615ac153082ade1b1ce0d0fa8
Exposes IGraphicBufferConsumer::detachBuffer as a ConsumerBase
method. attachBuffer is not currently exposed, because all current
clients will be recycling buffers through the allocator.
Bug: 19628705
Change-Id: I3e519767fa43d5d880c1d5695e31b60f6ad588af
Exposes the attachBuffer and detachNextBuffer calls from
IGraphicBufferProducer to the public Surface interface. Also moves
the version of connect that takes a producer callback from protected
to public.
Bug: 19628705
Change-Id: I9ebc3013c4d9c84c4e8ef150c00e03f8af80319e
(cherry picked from commit c14ecb9de2)
It's a security best practice for size_t values to be rejected if
they are greater than INT32_SIZE. This is intended to prevent the
common error of inadvertently passing a negative int value to a
function, which after conversion to an unsigned type, becomes a huge
number, defeating the purpose of bounds checking.
This patch also addresses a bug where the call to:
Parcel::write(buf, (size_t) -1);
would call writeInPlace() which uses PAD_SIZE on the supplied
argument. This would then cause an integer overflow, with PAD_SIZE
returning a small value, but the memcpy in Parcel::write using the
old large length value.
Bug: 19573085
Change-Id: Ib11bfb3dae4f3be91cd17b2c676926700972c7b8
This makes apps linked against the previous non-static versions
work with the new libbinder.
Bug: 19060033
Bug: 19773812
Change-Id: I3b5c78cbb4f4c0681ca1042e4d8503d98f969502
In restricted mode, only CTS tests can register for sensors or call flush() on them. The requests
from other applications will be ignored.
Change-Id: Ieb923df3e0cfe3390fe2d052af776da79589744b
Removes IGraphicBufferConsumer::BufferItem. Depends on the
following changes:
I187b3a7d05196b6289596afac8fb9a9d4aebff76
I0ddd38df37500cfd6b21d1e768ed14e39c5cd9fc
Cherry-pick of Id1fa56d092188f2cb712768d5d2fc6a9027fb73c
Change-Id: I3edf0db8fba656fd78e18a5a7f1137f0fb6b237d
Currently, there are two instances of BufferItem: one inside of
IGraphicBufferConsumer, and a standalone one inside of libgui. They
only differ in the name of one of the fields, and we want to remove
the IGBC version. This changes things so that client code may be
incrementally switched over to the libgui version.
This is a squashed commit containing the following changes:
I64f495105f56cbf5803cea4aa6b072ea29b70cf5
I1394e693314429ada93427889f10b7b01c948053
I9c3bc8037fa9438d4d9080b8afb694219ef2f71f
I699ed0a6837076867ca756b28d1ffb2238f7a0d9
Iac8425e1241774304a131da2fb9dec6e82922f13
Change-Id: Ic4d51f5df6dbc70b376d13fceba2335b9bae4f3d
- Wire up new dataSpace parameter through buffer queue stack
- Update tests to include the parameter
- Switch eglApi to using dataSpace to indicate sRGB gamma/linear
difference
- Remove RAW_SENSOR in favor of RAW16
- Remove use of sRGB format enums
- Add default dataspace to buffer queue core
- Add query for default dataspace
Cherry pick of I070bd2e7c56506055c419004c29e2e3feac725df
Change-Id: I461952389c18051176c6b75e664f20ad369f5760
An #ifdef was causing multiple returns in this function, which was
triggering an unreachable code error on some build targets. This
adds an #else to ensure that only one return is exposed in any given
build.
Cherry pick of I9ed7c3e769bc9094ad31d810944227d1d8302886
Change-Id: Ibddc5706159d168031d1805fcce3eedc068296cf
This change allows clients of Surface to provide an IProducerListener
callback object to Surface::connect, which will be passed down to the
underlying IGraphicBufferProducer.
Cherry pick of I5ea5229bf3a329bf02c6bd20e7247039c75d136b
Change-Id: I6f8f52c72654e4cee649721383819bafe378f964
Since some variables had been switched from signed to unsigned, there
was a section of code that was guaranteed to be incorrect because it
effectively did 'if (a < b) { c = a - b; }'. This change fixes it.
Cherry pick of I9cdd6c9a0179801addebb5d6dc1fbaddf8f53c62
Bug: 19346631
Change-Id: Id13a46f74c9ae7278463ce22b586f4dc21b5e453
There shouldn't be more than 4096 fds (probably signficantly smaller) and
there shouldn't be more than 4096 ints.
Cherry pick of I3a3e50ee3078a4710e9737114e65afc923ed0573
Bug: 18076253
Change-Id: I82a883572b401f115d252dcd3d00aa7252b49b0e
Enables -Weverything and -Werror, with just a few exceptions for
warnings we can't (or shouldn't need to) work around.
Cherry pick of I034abec27bf4020d84af60d7acc1939c59986dd6 plus a
couple of minor changes to CpuConsumer.cpp to make it work with a
prior change:
Uncomment CC_LOGV on line 46
Change C-style cast to static_cast on line 71
Change-Id: Iaec610477ea0122317b0578fb74caf2383d4cf08
Adds CpuConsumer::{detachNextBuffer,attachAndReleaseBuffer}, which
can be used to more carefully manage the ownership of GraphicBuffers.
Bug: 19628705
Change-Id: Ia7a7e30da6d81eb2367241998f14988db0afc3bf
Exposes the attachBuffer and detachNextBuffer calls from
IGraphicBufferProducer to the public Surface interface. Also moves
the version of connect that takes a producer callback from protected
to public.
Bug: 19628705
Change-Id: I9ebc3013c4d9c84c4e8ef150c00e03f8af80319e
Adds CpuConsumer::{detachNextBuffer,attachAndReleaseBuffer}, which
can be used to more carefully manage the ownership of GraphicBuffers.
Bug: 19628705
Change-Id: Ia7aa1ac59c2f768f2d8a0f35ad23067936a7427c
Removes IGraphicBufferConsumer::BufferItem. Depends on the
following changes:
I187b3a7d05196b6289596afac8fb9a9d4aebff76
I0ddd38df37500cfd6b21d1e768ed14e39c5cd9fc
Change-Id: Id1fa56d092188f2cb712768d5d2fc6a9027fb73c
One of the overloads of BufferQueueConsumer::acquireBuffer was
calling itself infinitely instead of calling the other overload.
This fixes that issue.
Bug: 19733425
Change-Id: Iac8425e1241774304a131da2fb9dec6e82922f13
Switches some dependencies from IGraphicBufferConsumer::BufferItem to
android::BufferItem and adds some methods to facilitate incrementally
changing client code to do the same.
Change-Id: I699ed0a6837076867ca756b28d1ffb2238f7a0d9
Adds an overload of IGraphicBufferConsumer::acquireBuffer which takes
an android::BufferItem instead of an IGBC::BufferItem.
Change-Id: I9c3bc8037fa9438d4d9080b8afb694219ef2f71f
Since some variables had been switched from signed to unsigned, there
was a section of code that was guaranteed to be incorrect because it
effectively did 'if (a < b) { c = a - b; }'. This change fixes it.
Bug: 19346631
Change-Id: I9cdd6c9a0179801addebb5d6dc1fbaddf8f53c62
- Wire up new dataSpace parameter through buffer queue stack
- Update tests to include the parameter
- Switch eglApi to using dataSpace to indicate sRGB gamma/linear
difference
- Remove RAW_SENSOR in favor of RAW16
- Remove use of sRGB format enums
- Add default dataspace to buffer queue core
- Add query for default dataspace
Change-Id: I070bd2e7c56506055c419004c29e2e3feac725df
This change allows clients of Surface to provide an IProducerListener
callback object to Surface::connect, which will be passed down to the
underlying IGraphicBufferProducer.
Change-Id: I5ea5229bf3a329bf02c6bd20e7247039c75d136b
There shouldn't be more than 4096 fds (probably signficantly smaller) and
there shouldn't be more than 4096 ints.
Bug: 18076253
Change-Id: I3a3e50ee3078a4710e9737114e65afc923ed0573
When compiling 32 bit binaries against a 64 bit binder interface,
implicit promotions of intptr_t types to uint64_t for fields in the
binder ioctl structures can result in invalid pointers because of sign
extension.
The Region::set(uint32_t,uint32_t) ABI is used by the Nexus 10 graphics
driver blobs.
After commit dd883c0b08, Nexus 10 blobs
need a rebuild unless the old ABI is added back
Change-Id: I138dcc1cc281b8d3cde0021282bf874054ec4eaf
Signed-off-by: Bernhard Rosenkraenzer <Bernhard.Rosenkranzer@linaro.org>
Enables -Weverything and -Werror, with just a few exceptions for
warnings we can't (or shouldn't need to) work around.
This is a squashed commit based on an initial change with a couple of
fixes to avoid breaking certain targets. The source commits are:
d723bd766900d504c06e429ba89cd2
Change-Id: I034abec27bf4020d84af60d7acc1939c59986dd6
An #ifdef was causing multiple returns in this function, which was
triggering an unreachable code error on some build targets. This
adds an #else to ensure that only one return is exposed in any given
build.
Change-Id: I9ed7c3e769bc9094ad31d810944227d1d8302886
Enables -Weverything and -Werror, with just a few exceptions for
warnings we can't (or shouldn't need to) work around.
Change-Id: I034abec27bf4020d84af60d7acc1939c59986dd6
Enables -Weverything and -Werror, with just a few exceptions for
warnings we can't (or shouldn't need to) work around.
Change-Id: I470a267e9b1c373f2d5c960f005c3b0e02b2db63
(cherry picked from commit 303b9a5123)
Enables clang and C++11 for libui/libgui/surfaceflinger, and
eliminates all compile-time warnings.
Change-Id: Ie237fdb5ae44f2bfcddaa884f9c65ec3f08ae50f
(cherry picked from commit f10c46ef85)
Enables -Weverything and -Werror, with just a few exceptions for
warnings we can't (or shouldn't need to) work around.
Change-Id: I470a267e9b1c373f2d5c960f005c3b0e02b2db63
Adds readUint32 and writeUint32 methods to the Parcel class. This
saves a lot of static_casting in anything implementing a Binder
interface on the native side.
Change-Id: Iafc73b0633654a3a4c49767f41806b56906c924f
Now that we're in C++11 mode by default, clang complains about switch
statements with case values not matching the type -- since some
binder_driver_return_protocol values are > 0x7fffffff, we need to make
the switch statements operate on uint32_t rather than int32_t.
BUG: 18466763
Change-Id: Iedbfd5c7a3d3d9f087d2eab4ff21343ad7a2a448
Signed-off-by: Bernhard Rosenkränzer <Bernhard.Rosenkranzer@linaro.org>
* Explicit conversion for atomic_uintptr_t initialization.
* Fix string literal concatenation to not be a UD literal.
* Use __typeof__ instead of typeof (should become decltype once this
actually moves to C++11).
Bug: 18466763
Change-Id: I4eedddfb945a2a703ed27317cb6e2b3041b1ebfc
so we can do NULL checks again, and update calls to IInterface::asBinder()
to use the new static version.
Change-Id: Ia7b10eb38ca55b72278bfd33d3bf647f338b4e6a
so we can do NULL checks again, and update calls to IInterface::asBinder()
to use the new static version.
Change-Id: Ia7b10eb38ca55b72278bfd33d3bf647f338b4e6a
...in Parcel destructor
Don't use a Mutux object, just use simple posix mutex primitives,
to avoid static init/destroy order problems.
Change-Id: Ic012d94297564c0a55d58869f8276d7d10545fbc
Enables clang and C++11 for libui/libgui/surfaceflinger, and
eliminates all compile-time warnings.
Change-Id: Ie237fdb5ae44f2bfcddaa884f9c65ec3f08ae50f
Passes the BufferItem for the queued buffer to the onFrameAvailable
callback so the consumer can track the BufferQueue's contents. Also
adds an onFrameReplaced callback, which is necessary if the consumer
wants to do anything more than simple queue length tracking.
Bug: 18111837
Change-Id: If9d07229c9b586c668e5f99074e9b63b0468feb0
