Convert the default impl into a static library,
android.hardware.graphics.allocator@2.0-passthrough.
Test: boots and VTS
Change-Id: I8ec8b30766462ecb3fb789af7c6dbb3c088ccf57
Add a header-only library
android.hardware.graphics.allocator@2.0-hal that can be used by
implementations. An imlpementation can
class VendorHal : public AllocatorHal { ... };
auto allocator = std::make_unique<Allocator>();
allocator->init(std::make_unique<VendorHal>(...));
Or, if vendor extensions are to be added to the IAllocator,
class AlocatorHalExt : public AllocatorHal { ... };
class VendorHal : public AllocatorHalExt { ... };
class AllocatorExt : public AllocatorImpl<IAllocatorExt, AllocatorHalExt> { ... };
auto allocator = std::make_unique<AllocatorExt>();
allocator->init(std::make_unique<VendorHal>(...));
Test: builds
Change-Id: I7cb7a4888316b871e5c49d96524b1642fc708f2d
Move libgralloc1-adapter from 2.0/default/ to
2.0/utils/gralloc1-adapter/. Fix build issues after the move.
Test: builds
Change-Id: I674fe60c724a4ffc1540c796b92209a1dbf36438
ComposerClient destroys its internal model of the display while handling
the onHotPlug event from the Hwc. Subsequently SurfaceFlinger destroys
its model of the display, and destroys all Hwc layers associated with
the display.
This fixes the code for destroying layers to not dereference an invalid
iterator if the display does not exist, allowing destruction to
continue.
It also fixes a similar issue which could occur if a HWC layer is being
created for a display at around the same time as the disconnect event.
Test: hotplug disconnect no longer crashes
Bug: 38464421
Change-Id: I0f2d28fe89fccf997b4bbb9fa6b5c0e6a6e49b93
The interface lib has been in VNDK-SP because
android.hardware.graphics.mapper@1.0 was using it. However, since the
dependency has gone [1], there is no need keep it in VNDK-SP. The
VNDK-SP set should be kept as small as possible because libs in VNDK-SP
are subject to double-loading.
Unmark the 'support_system_process' property to exclude the lib from
VNDK-SP.
This commit re-lands I8722c1ac15ddf56a627a12a0c649b4d734e5e5cd because
it was reverted during O-MR1 push to AOSP-master.
Bug: 69480083
Test: walleye boots to the UI
Change-Id: I0af8115dceb9711c6c451ffaeedda6c823ec2905
Merged-In: I8722c1ac15ddf56a627a12a0c649b4d734e5e5cd
(cherry picked from commit ec44d18dbe)
There is no cap defined for PRESENT_OR_VALIDATE_DISPLAY in HIDL so
it must always work. Make sure it does not call HWC2 presentDisplay
when the underlying HWC2 does not support
HWC2_CAPABILITY_SKIP_VALIDATE.
Bug: 70407085
Test: boots
Change-Id: I54a4400e09e669c5064f05739f595ed978dcc713
Removing whenever I see these in code reviews.
Test: none
Merged-In: I4322f533a837d55618ec2ed2125e8966ace9d61d
Change-Id: I4322f533a837d55618ec2ed2125e8966ace9d61d
Removing whenever I see these in code reviews.
Test: none
Merged-In: I4322f533a837d55618ec2ed2125e8966ace9d61d
Change-Id: I4322f533a837d55618ec2ed2125e8966ace9d61d
Removing whenever I see these in code reviews.
Test: none
Merged-In: I4322f533a837d55618ec2ed2125e8966ace9d61d
Change-Id: I4322f533a837d55618ec2ed2125e8966ace9d61d
The interface lib has been in VNDK-SP because
android.hardware.graphics.mapper@1.0 was using it. However, since the
dependency has gone [1], there is no need keep it in VNDK-SP. The
VNDK-SP set should be kept as small as possible because libs in VNDK-SP
are subject to double-loading.
Unmark the 'support_system_process' property to exclude the lib from
VNDK-SP.
Bug: 69480083
Test: walleye boots to the UI
Change-Id: I8722c1ac15ddf56a627a12a0c649b4d734e5e5cd
FB (framebuffer) HAL has been replaced by HWC HAL for 5+ years, but
we still support the legacy path in SurfaceFlinger. Devices using
the legacy path cannot be Treblized.
This change allows such devices to use HIDL IComposer, by adding
support for FB HAL in the default implementation.
Test: boots hikey960
Change-Id: Ie9050bbcaac0fd5b134786f4f9f0f5075f4ebd0c
After initialization or onRefresh, we want to make sure
validateDisplay is called before presentDisplay.
Bug: 67505273
Test: manual
Change-Id: Id876d9251586aaaf552ca82c52f8f902af364251
The hardware composer service has a rule that only one client can be
connected at a time. The surface flinger process, when transitioning
composer ownership from surface flinger to vr flinger, will destroy the
current client on one thread and create a new client on another
thread. Although surface flinger ensures that these events happen in the
expected sequence (delete then create), the requests sometimes land in
the hardware composer service in inverted order, causing the creation
request to fail with an error.
Instead of failing with an error, block for a brief period (1 second)
until the existing client is removed, then proceed to initialize the new
client. This gives us enough time to ensure an inverted
creation/destruction order doesn't cause client creation to fail, while
avoiding a deadlock if the existing client is never destroyed.
Bug: 62925812
Test: - Transitioned to/from vr flinger hundreds of times, and confirmed
I no longer see sporadic composer client creation failure due to an
already existing client.
- Ran the vts graphics composer tests and confirmed they all pass.
Change-Id: I40be1fb0cb3d42ddb5a9fc159188886e9f5b6267
After initialization or onRefresh, we want to make sure
validateDisplay is called before presentDisplay.
Bug: 67505273
Test: manual
Change-Id: Id876d9251586aaaf552ca82c52f8f902af364251
To prevent property name collisions between properties of system and
vendor, 'vendor.' prefix must be added to a vendor HAL service name.
You can see the details in http://go/treble-sysprop-compatibility.
Test: succeeded building and tested on a walleye device
Bug: 36796459
Change-Id: I4e8fbee791ec917a8f627a1366f4d44ec7e6febc
Our use of message queues is synchronous. If there are already data
in the queue when writeQueue is called, we know they are stale and
can be discarded.
Bug: 65449888
Test: manual
Change-Id: Ie29b8a7386c9733c183a6c3569e3572efa62cbc2
* For registerTestService/getServiceName, no need to pass
any hard coded string of HAL service FQName.
* Affect test: VtsHalCameraProviderV2_4TargetTest,
VtsHalGraphicsComposerV2_1TargetTest,
VtsHalNeuralnetworksV1_0TargetTest
Bug: 62946472
Bug: 64203181
Test: make vts
vts-tradefed run vts -m VtsHalCameraProviderV2_4Target
Merged-In: If365ab2ed9a91eb4013d71769804b9d4bf089d66
Change-Id: Id0bddbc2949337147557f45cc60dbfaa114ce25e
(cherry picked from commit d71b654d6d)
* For registerTestService/getServiceName, no need to pass
any hard coded string of HAL service FQName.
* Affect test: VtsHalCameraProviderV2_4TargetTest,
VtsHalGraphicsComposerV2_1TargetTest,
VtsHalNeuralnetworksV1_0TargetTest
Bug: 62946472
Bug: 64203181
Test: make vts
vts-tradefed run vts -m VtsHalCameraProviderV2_4Target
Change-Id: Id0bddbc2949337147557f45cc60dbfaa114ce25e
* Update to run against different service names.
Bug:38348289
Bug:64203181
Test: make vts
vts-tradefed run vts -m VtsHalGraphicsComposerV2_1TargetTest
Change-Id: I7c4643935cf5f0ea63b0a97d3079559d9c4f5ec2
This reverts commit 8ac1971678.
Reason for revert: Didn't remove automotive changes from this CL.
Merged-In: I8608c8f636c35f21e4246a805a9eff6d14124e0a
Change-Id: I1c660cffc8817ad0b33da9f6eceb3d88e7c48416
These tests now statically links to HAL def libs which are
not guaranteed to be on the device.
Bug: 64040096
Test: vts-tradefed run commandAndExit vts --skip-all-system-status-check
--skip-preconditions --module VtsHalGraphicsMapperV2_0Target
Test: vts-tradefed run commandAndExit vts --skip-all-system-status-check
--skip-preconditions --module VtsHalGraphicsComposerV2_1Target
Change-Id: I83b68dafbd161f88394f67dbda14f9957ddf7232
vndk-sp is not automatically tagged by hidl-gen.
For vndk-sp libs, "support_system_process: true" is manually added
in "vndk" property.
Bug: 63866913
Test: build and boot with BOARD_VNDK_VERSION=current
Change-Id: I2b18d691411e58dc55bcdfa39ecb3659242c8437
Update the Android.bp generated with hidl-gen.
Test: build with and without BOARD_VNDK_VERSION=current
Bug: 63866913
Change-Id: I1a9db1df49e0f13c5790da2b118ae9ec63ba34a7
Allow HAL definition libs to be static.
Bug: 32920003
Bug: 64040096
Test: update-all-google-makefiles.sh
Change-Id: I1483d572bea6799717d1614fb7d52fe225e31104
The hardware composer service has a rule that only one client can be
connected at a time. The surface flinger process, when transitioning
composer ownership from surface flinger to vr flinger, will destroy the
current client on one thread and create a new client on another
thread. Although surface flinger ensures that these events happen in the
expected sequence (delete then create), the requests sometimes land in
the hardware composer service in inverted order, causing the creation
request to fail with an error.
Instead of failing with an error, block for a brief period (1 second)
until the existing client is removed, then proceed to initialize the new
client. This gives us enough time to ensure an inverted
creation/destruction order doesn't cause client creation to fail, while
avoiding a deadlock if the existing client is never destroyed.
Bug: 62925812
Test: - Transitioned to/from vr flinger hundreds of times, and confirmed
I no longer see sporadic composer client creation failure due to an
already existing client.
- Ran the vts graphics composer tests and confirmed they all pass.
Change-Id: I40be1fb0cb3d42ddb5a9fc159188886e9f5b6267
We should call native_handle_clone instead of IMapper::importBuffer
on sideband stream handles.
Bug: 63707399
Test: manual
Change-Id: Ie07e50f6bb771f3c7a69d29e36f60b149dc6af95
This reverts commit 33c08a53ac. Some
gralloc implementations apparently have library initializers (executed
on dlopen) which create anonymous sockets, open device nodes, etc.
which Zygote and some other non-Zygote-child processes that use libui
don't have permissions to do. When the library initializer fails, it
crashes the process.
Reverting this until this initialization code can be made lazy (on
first use of gralloc).
Bug: 62732270
Test: boot fugu to launcher
In the a.h.graphics.mapper default implementation, load the gralloc
module in a static constructor that runs when the library is loaded.
This causes Zygote to load the gralloc module and its dependencies
when it loads a.h.graphics.mapper, so that it doesn't have to be
loaded during each activity launch.
Bug: 62353585
Test: boot to launcher, run calculator, confirm gralloc library is in
/proc/`pid zygote64`/maps
Change-Id: I56defd76b0ba0fba71dea48e2b0d338511563d1f
BufferUsage::SENSOR_DIRECT_DATA should be filtered out for consumer
usage. BufferUsage::GPU_DATA_BUFFER should be filtered out for
producer usage and be translated to
GRALLOC1_CONSUMER_USAGE_GPU_DATA_BUFFER for consumer usage.
Bug: 62491483
Test: boots
Change-Id: I171712adf45ae80dadeed7997fb9872ff176fa1a
Split up utility classes that can be re-used in other IComposer related
tests (such as the IVrComposerClient tests).
Changes:
* Move GraphicsComposerCallback into separate module
* Extract TestCommandReader in separate module
* Extract Composer::execute() from test class
Bug: 38203529
Test: Compiled
Change-Id: I0d1193469e9dd94c477a63058f98111c162dfa99
There are implmentations that can dead lock when calling certain
functions in certain states. It works fine before HIDL because
SurfaceFlinger calls those functions in another thread. We allow
for 4 hwbinder threads in this commit to simulate how SurfaceFlinger
calls hwcomposer.
Bug: 38183197
Test: camera, videos, multi windows, screencap, screenrecord
Change-Id: Ie05b2ca349b8c1ed9a3ac962981434f2efee92d9
This should help buffer allocation performance. We choose 4 because
that is the number of binder threads in SurfaceFlinger.
Bug: 38243574
Test: manual
Change-Id: I704aacb271fb7fd6a91c9231ab8a5256531eebc9
android.hidl.base@1.0 and android.hidl.manager@1.0 are built into libhidltransport.
Test: links
Bug: 33276472
Merged-In: I08aaad80f7e2fc262aa3a8b66fe932e8133a928d
Change-Id: I08aaad80f7e2fc262aa3a8b66fe932e8133a928d
When setLayerBuffer() was called on a layer previously destroyed by
destroyLayer() the code would crash. Instead, return an error.
Bug: 37159844
Test: Ran vr flinger in a way that would trigger the crash, and
confirmed that I now get error logs instead of crashing. Unfortunately
the error code is consumed by the Composer wrapper class and not
propagated back to the caller, but that's a separate issue (b/37901601).
Change-Id: I75a5b954d47a1deac44d03851f60f347085eca89
android.hidl.base@1.0 and android.hidl.manager@1.0 are built into libhidltransport.
Test: links
Bug: 33276472
Change-Id: I08aaad80f7e2fc262aa3a8b66fe932e8133a928d
Note, current.txt change (as opposed to addition) is only allowed
because this is an unreleased interface.
Bug: 32593292
Test: builds
Change-Id: I95381d1a3a456aeadbc472e2dbbffde19cb43e6c
HALs are not expected to see YCBCR_422_888, YCBCR_444_888,
FLEX_RGB_888, and FLEX_RGBA_8888. JPEG is purely wrong.
Note, current.txt change (as opposed to addition) is only allowed
because this is an unreleased interface.
Bug: 32593292
Test: builds
Change-Id: I432e6cb75cfc347a41e8e1632ae1898eed3350c8
Make the docs more self-contained, not referring to stuff not
defined by the HIDL interfaces.
Mark JPEG, YCBCR_422_888, YCBCR_444_888, FLEX_RGB_888,
FLEX_RGBA_8888 as "MUST NOT USE". These format are never supposed
to be seen by gralloc and do not belong to the HIDL interfaces.
Note, current.txt change (as opposed to addition) is only allowed
because this is an unreleased interface.
Bug: 32593292
Test: builds
Change-Id: Ia7607cc73dd2f58dfa27e71b0004b21c7ca6904d
Sort the enum values in their numerical order. There is no other
change.
Note, current.txt change (as opposed to addition) is only allowed
because this is an unreleased interface.
Bug: 32593292
Test: builds
Change-Id: I47914a7971f922f5602f982b41dd7602842e853b
We must use the mapper HAL instead of gralloc0/gralloc1 from the
composer.
Bug: 37540361
Test: boots on marlin, angler, and ryu
Change-Id: I5a3ff6a025bf51a3507a4f33fa77e9506a6f1ec9
A buffer handle recieved from a HAL is by definition raw and needs
to be imported. But because of passthrough HALs, such a raw handle
may have been imported already. Explicitly specify that an
implementation must accept such a raw handle.
Bug: 37540361
Test: boots on angler, ryu and marlin
Change-Id: I5ecf526e59b27cc4a8f7f5d5ec27477da0946ece
GraphicBufferMapper is valid during process termination. IMapper
must stay valid as well. It should not rely on global/static
objects that may have been destructed.
Bug: 37480313
Test: libgui_test
Change-Id: Icb8079153306f2465c26c0f1ce812895ad83f21b
Bug: 33241851
Test: links
Test: try build with BOARD_VNDK_VERSION current, no errors related to
this module
Test: (sanity) YouTube
Change-Id: Ia2c3922bf8bf1ab51396a8c8c5fa8d909c442994
By setting vendor_available, the following may become true:
* a prebuilt library from this release may be used at runtime by
in a later releasse (by vendor code compiled against this release).
so this library shouldn't depend on runtime state that may change
in the future.
* this library may be loaded twice into a single process (potentially
an old version and a newer version). The symbols will be isolated
using linker namespaces, but this may break assumptions about 1
library in 1 process (your singletons will run twice).
Background:
This means that these modules may be built and installed twice --
once for the system partition and once for the vendor partition. The
system version will build just like today, and will be used by the
framework components on /system. The vendor version will build
against a reduced set of exports and libraries -- similar to, but
separate from, the NDK. This means that all your dependencies must
also mark vendor_available.
At runtime, /system binaries will load libraries from /system/lib*,
while /vendor binaries will load libraries from /vendor/lib*. There
are some exceptions in both directions -- bionic(libc,etc) and liblog
are always loaded from /system. And SP-HALs (OpenGL, etc) may load
/vendor code into /system processes, but the dependencies of those
libraries will load from /vendor until it reaches a library that's
always on /system. In the SP-HAL case, if both framework and vendor
libraries depend on a library of the same name, both versions will be
loaded, but they will be isolated from each other.
It's possible to compile differently -- reducing your source files,
exporting different include directories, etc. For details see:
https://android-review.googlesource.com/368372
None of this is enabled unless the device opts into the system/vendor
split with BOARD_VNDK_VERSION := current.
Bug: 36426473
Bug: 36079834
Test: m -j libhwcomposer-client
Test: attempt to compile with BOARD_VNDK_VERSION := current
Change-Id: I735b861fcc25bc1048ce0ce3ad48432980248b06
Revise IAllocator and IMapper to reduce IPC and to support gralloc0
devices.
Specifically, IAllocator is trimmed down to have essentially only
allocate(BufferDescriptor descriptor, uint32_t count)
generates (Error error,
uint32_t stride,
vec<handle> buffers);
The ability to allocate buffers with shared backing store is
removed. ProducerUsage and ConsumerUsage are moved to the
graphics.common package and are merged and renamed to BufferUsage.
BufferUsage's bits follow gralloc0.
IMapper gains
typedef vec<uint32_t> BufferDescriptor;
createDescriptor(BufferDescriptorInfo descriptorInfo)
generates (Error error,
BufferDescriptor descriptor);
where BufferDescriptor is an implementation-defined blob. lockFlex
is replaced by lockYCbCr. All getters are removed.
Reference counting with retain/release is replaced by
importBuffer/freeBuffer.
Most if not all gralloc1 features are not used by the runtime yet.
There is also not too much test written for them. As such, they
tend to behave differently between implementations and cannot be
used reliably.
Bug: 36481301
Test: builds and boots on Pixel
Change-Id: I1d31105120517ea2c128c7a19297acf3bfd312bb
RAW16 format could be used together with depth dataspace.
The buffer contents in this case will be phase detection
statistics data.
Bug: 36015382
Test: testPDStats
Change-Id: Ied1c179193abef62a87fcec6f0906528ed58ea2b
(cherry picked from commit 23699d6314)
Files relying on transitive include of utils/Log.h (and things that it
includes) from MQDescriptor.h
Test: pass
Merged-In: Iff316b21bef556bb026378b7f89e97ded3febef4
Change-Id: Iff316b21bef556bb026378b7f89e97ded3febef4
Converting from ..graphics.allocator@2.0 usage to gralloc0 usage isn't
quite as simple as ORing and truncating, which is what the default
implementation was doing. Switch to using library functions that do it
correctly.
Test: boot bullhead
Change-Id: I40ae00e9aad92b374f281569207972b7461a3e55
