OSDN Git Service

android-x86/frameworks-native.git
8 years agoMerge "Allow setting alpha to 0 when creating native client buffers." into nyc-dev
Craig Donner [Mon, 6 Jun 2016 16:18:23 +0000 (16:18 +0000)]
Merge "Allow setting alpha to 0 when creating native client buffers." into nyc-dev

8 years agoAllow setting alpha to 0 when creating native client buffers.
Craig Donner [Sat, 4 Jun 2016 00:54:25 +0000 (17:54 -0700)]
Allow setting alpha to 0 when creating native client buffers.

Note that RGBA sizes are checked later in the function anyway, so there was no
reason to disallow 0, but it did preclude setting alpha to 0 for 565 buffers.

Fixes: 29126799

Change-Id: Iba11649976221ab385227fc022ae74b14c342203

8 years agolibvulkan: reduce log level for swapchain usage errors
Jesse Hall [Fri, 3 Jun 2016 19:39:49 +0000 (12:39 -0700)]
libvulkan: reduce log level for swapchain usage errors

Bug: 26927424
Change-Id: Id9346356436663db83a2d6fa8fefc4ea24d7854b

8 years agoMerge "libvulkan: Slightly better handling of swapchain re-creation" into nyc-dev
Jesse Hall [Thu, 2 Jun 2016 23:57:02 +0000 (23:57 +0000)]
Merge "libvulkan: Slightly better handling of swapchain re-creation" into nyc-dev

8 years agoMerge "vulkan: remove EnumerateDeviceLayerProperties internally" into nyc-dev
Chia-I Wu [Thu, 2 Jun 2016 22:36:26 +0000 (22:36 +0000)]
Merge "vulkan: remove EnumerateDeviceLayerProperties internally" into nyc-dev

8 years agoMerge "Camera: Add media/camera NDK documents" into nyc-dev
Yin-Chia Yeh [Thu, 2 Jun 2016 18:12:21 +0000 (18:12 +0000)]
Merge "Camera: Add media/camera NDK documents" into nyc-dev

8 years agolibvulkan: Slightly better handling of swapchain re-creation
Jesse Hall [Tue, 31 May 2016 05:40:14 +0000 (22:40 -0700)]
libvulkan: Slightly better handling of swapchain re-creation

Previously we would fail vkCreateSwapchainKHR if
VkSwapchainCreateInfo::oldSwapchain was used, because we were unable
to dequeue all buffers for the new swapchain from an already-used
native window.

Now we disconnect and re-connect to the native window in order to
fully reset state, and allow us to dequeue all buffers. Additionally,
we tag the old swapchain as out-of-date, so future image acquires and
presents will fail with VK_ERROR_OUT_OF_DATE_KHR on that swapchain.
This is less than ideal, but better than what we had, and the best we
have time to do for N.

Bug: 26927424
Change-Id: Ifaa5048376f72a63ecb1dca3d1ff85dbee2c24d0

8 years agoMerge "BQ: Don't return fences from dequeue in shared buffer mode" into nyc-dev
Pablo Ceballos [Thu, 2 Jun 2016 17:14:00 +0000 (17:14 +0000)]
Merge "BQ: Don't return fences from dequeue in shared buffer mode" into nyc-dev

8 years agoMerge "Push empty sync transaction before taking screenshot in tests" into nyc-dev
Pablo Ceballos [Thu, 2 Jun 2016 17:10:06 +0000 (17:10 +0000)]
Merge "Push empty sync transaction before taking screenshot in tests" into nyc-dev

8 years agoMerge "Create primary profile file together with app data folders" into nyc-dev
Calin Juravle [Thu, 2 Jun 2016 10:10:35 +0000 (10:10 +0000)]
Merge "Create primary profile file together with app data folders" into nyc-dev

8 years agovulkan: remove EnumerateDeviceLayerProperties internally
Chia-I Wu [Thu, 2 Jun 2016 03:25:47 +0000 (11:25 +0800)]
vulkan: remove EnumerateDeviceLayerProperties internally

We never use it.  Since it is also deprecated, we will never need it.

Bug: 27910383
Change-Id: If4e556bb69890153be2987b165482e43ab06a890

8 years agoPush empty sync transaction before taking screenshot in tests
Pablo Ceballos [Thu, 2 Jun 2016 01:53:40 +0000 (18:53 -0700)]
Push empty sync transaction before taking screenshot in tests

Because of ag/1043009 it's now necessary to push an empty sync
transaction before taking a screenshot to ensure that all transactions
have been applied.

Bug 29074285

Change-Id: I9f1b4a5b6b1f39e1ff26d536817aef6f40865177

8 years agoBQ: Don't return fences from dequeue in shared buffer mode
Pablo Ceballos [Wed, 1 Jun 2016 22:03:21 +0000 (15:03 -0700)]
BQ: Don't return fences from dequeue in shared buffer mode

Bug 29007505

Change-Id: I112ae6bd64ddc9a923ad9080e85a71adb96146af

8 years agoinstalld support for profman --dump-info-for
David Sehr [Tue, 17 May 2016 23:06:22 +0000 (16:06 -0700)]
installd support for profman --dump-info-for

Bug: 28748264
Change-Id: Idcd3ed86294c1d0c75c26ca938b88744a25e387a

8 years agoMerge "svcmgr: don't send reply to one-way txn (part deux)" into nyc-dev
Mark Salyzyn [Tue, 31 May 2016 19:44:18 +0000 (19:44 +0000)]
Merge "svcmgr: don't send reply to one-way txn (part deux)" into nyc-dev

8 years agoUse explicit path for /vendor/xbin/wlutil
Erik Kline [Mon, 30 May 2016 02:55:44 +0000 (11:55 +0900)]
Use explicit path for /vendor/xbin/wlutil

At this time /vendor/xbin is not in root's path.

Bug: 28850734
Change-Id: Ib6c2b6f8b0285854670c843dcc437cfcc84484a8

8 years agoMerge "vulkan: fix crash when only non-global layers are enabled" into nyc-dev
Chia-I Wu [Fri, 27 May 2016 08:22:49 +0000 (08:22 +0000)]
Merge "vulkan: fix crash when only non-global layers are enabled" into nyc-dev

8 years agoCamera: Add media/camera NDK documents
Yin-Chia Yeh [Fri, 27 May 2016 06:07:15 +0000 (23:07 -0700)]
Camera: Add media/camera NDK documents

Include media/camera ndk headers to codegen source.
Add docs/images directory for image links in generated documents.

Bug: 27102995
Change-Id: I9c75e2ceba3458c9962c26cf786e1402fd963696

8 years agovulkan: fix crash when only non-global layers are enabled
Chia-I Wu [Fri, 27 May 2016 02:18:25 +0000 (10:18 +0800)]
vulkan: fix crash when only non-global layers are enabled

We fail to set up get_instance_proc_addr_ and get_device_proc_addr_ when
there are layers but only non-global ones.

Bug: 28990561
Change-Id: I82704f946deb5a76640f9dfb792e6fd118c78aa7

8 years agoAdd "ip maddr" to dumpstate output for debugging.
Erik Kline [Fri, 27 May 2016 02:29:19 +0000 (11:29 +0900)]
Add "ip maddr" to dumpstate output for debugging.

Bug: 28966611
Change-Id: I8fdab94fbd41448e52f2fed406e8f43edf7f18ad

8 years agosvcmgr: don't send reply to one-way txn (part deux)
Mark Salyzyn [Wed, 25 May 2016 20:14:10 +0000 (13:14 -0700)]
svcmgr: don't send reply to one-way txn (part deux)

Do not split transaction into two when sending synchronous responses.

Partially revert changes associated with binder_send_reply from
previous incantations in commit 51f592c7bad13d3d8e8f7cfec25edfef42901802
that resulted in a performance regression.

Signed-off-by: Mark Salyzyn <salyzyn@google.com>
Bug: 27111612
Bug: 28696050
Bug: 28946245
Change-Id: I12edc10a2d1408a71507a7983afa7f17b5a6a81b

8 years agoMark the multiwindow divider as a trusted overlay.
Michael Wright [Thu, 26 May 2016 12:49:53 +0000 (13:49 +0100)]
Mark the multiwindow divider as a trusted overlay.

Having it be untrusted breaks the permissions granting dialog in
multiwindow mode.

Bug: 28828041
Change-Id: I9193dbce7315ebaa205673edd4f406d0b72665c8

8 years agoCreate primary profile file together with app data folders
Calin Juravle [Tue, 24 May 2016 14:33:12 +0000 (15:33 +0100)]
Create primary profile file together with app data folders

Bug: 28785119
Bug: 28625993
Change-Id: I12c0de85bc3dc35199acf32c10fe69f93e76d7fc

8 years agoMerge "SF: Force empty sync transactions to apply anyway" into nyc-dev
TreeHugger Robot [Tue, 24 May 2016 23:15:00 +0000 (23:15 +0000)]
Merge "SF: Force empty sync transactions to apply anyway" into nyc-dev

8 years agoSF: Force empty sync transactions to apply anyway
Robert Carr [Tue, 24 May 2016 18:41:28 +0000 (11:41 -0700)]
SF: Force empty sync transactions to apply anyway

Following the change to not flush transactions in the screen shot
path, this enables the use of empty synchronous transactions as a
flush mechanism for previous asynchronous transactions.

Bug: 27098060
Change-Id: I282fdfffb72db148edcbf1ded65b15816fc714a2

8 years agoMerge "SF: Remove forced transaction on screenshot" into nyc-dev
Dan Stoza [Tue, 24 May 2016 18:01:24 +0000 (18:01 +0000)]
Merge "SF: Remove forced transaction on screenshot" into nyc-dev

8 years agoMerge "HWC2: Fix breakage from header change" into nyc-dev
Dan Stoza [Tue, 24 May 2016 16:55:43 +0000 (16:55 +0000)]
Merge "HWC2: Fix breakage from header change" into nyc-dev

8 years agostagefright: define mpeg2 HighP level constant
Lajos Molnar [Tue, 24 May 2016 14:02:01 +0000 (07:02 -0700)]
stagefright: define mpeg2 HighP level constant

Bug: 28671284
Change-Id: Ic97fd9c3cb4ce04f9cfcd18a4daf68db3349694b

8 years agovulkan: advertise VK_EXT_debug_report
Chia-I Wu [Mon, 23 May 2016 07:31:02 +0000 (15:31 +0800)]
vulkan: advertise VK_EXT_debug_report

We use VK_EXT_debug_report to report most messages and we should always
advertise the extension.  This is complicated however because the HAL may
also advertise the extension.

Do the minimal and cache whether the HAL supports VK_EXT_debug_report in
class Hal.  More refactorings will be required if we want to be more
flexible.

Bug: 28753291
Change-Id: I0298b2c3024e2ed059411128dbbe264c0b289a4b

8 years agovulkan: add a wrapper for hwvulkan_device_t
Chia-I Wu [Mon, 23 May 2016 02:47:57 +0000 (10:47 +0800)]
vulkan: add a wrapper for hwvulkan_device_t

Add class Hal as a wrapper for hwvulkan_device_t.

Bug: 28753291
Change-Id: I9612ee5515c7f2266cfc1ef0c8583fa98c82dd86

8 years agoMerge "vulkan: do not use exported functions internally" into nyc-dev
Chia-I Wu [Mon, 23 May 2016 23:36:52 +0000 (23:36 +0000)]
Merge "vulkan: do not use exported functions internally" into nyc-dev

8 years agoMerge "Expose functions used to add entries to the zip file." into nyc-dev
TreeHugger Robot [Mon, 23 May 2016 20:01:25 +0000 (20:01 +0000)]
Merge "Expose functions used to add entries to the zip file." into nyc-dev

8 years agoExpose functions used to add entries to the zip file.
Felipe Leme [Thu, 19 May 2016 23:18:17 +0000 (16:18 -0700)]
Expose functions used to add entries to the zip file.

Currently only dumpstate.cpp is using the new functions and variables
used to add an entry to the zip file so they're defined locally, but
they should be exposed through the .h so they can be used by the
device-specific dumpstate_board() implementations.

This CL moves the following definitions to the .h:

- bugreport_dir
- ZIP_ROOT_DIR
- add_zip_entry()
- add_zip_entry_from_fd()

It also refactored functions that were taking a specific file location
to use bugreport_dir instead, to avoid pollution on main().

BUG: 28867848
Change-Id: Ic5faa203b0863bc950d640f22fb91977e52b6bfc

8 years agoomx: add asString for new enumerations and fix some issues
Lajos Molnar [Sat, 21 May 2016 02:38:40 +0000 (19:38 -0700)]
omx: add asString for new enumerations and fix some issues

Bug: 28671284
Change-Id: I54e98d727f974544d51b2a14b565ad24912ab242

8 years agoMerge "omx: add missing MPEG4 levels" into nyc-dev
Lajos Molnar [Mon, 23 May 2016 13:17:25 +0000 (13:17 +0000)]
Merge "omx: add missing MPEG4 levels" into nyc-dev

8 years agovulkan: do not use exported functions internally
Chia-I Wu [Fri, 20 May 2016 08:15:06 +0000 (16:15 +0800)]
vulkan: do not use exported functions internally

Our vkGet*ProcAddr uses the exported functions.  They will break when any
of the exported functions are overridden (e.g., through LD_PRELOAD).

Unexport and move all exported functions to vulkan::api namespace.
Re-export them by having vkFoo as a wrapper to vulkan::api::Foo.

Another option is to re-export vulkan::api::Foo by having vkFoo as an
alias using __attribute__((alias)).  That results in smaller binaries.
But we will not be able to catch mismatches between vulkan.h and
vulkan.api.

To avoid future breakage, define VK_NO_PROTOTYPES for all files except
api_gen.cpp.

Bug: 28886971
Change-Id: I08fde7ebb247f8c7e040ccf812b40b02094d3c7f

8 years agoMerge "Ensure memory ordering around libagl and EGL refcount operations" into nyc-dev
Jesse Hall [Sat, 21 May 2016 21:21:42 +0000 (21:21 +0000)]
Merge "Ensure memory ordering around libagl and EGL refcount operations" into nyc-dev

8 years agoEnsure memory ordering around libagl and EGL refcount operations
Jesse Hall [Fri, 20 May 2016 17:47:07 +0000 (10:47 -0700)]
Ensure memory ordering around libagl and EGL refcount operations

The android_atomic_inc/android_atomic_dec functions don't impose
sufficient memory ordering. Using them for object refcounting could
allow an object to be destroyed prior to writes by a different thread
being visible.

Bug: 28820690
Change-Id: Ie018091035174255a22ebc52852528cdaec2d648

8 years agoHWC2: Fix breakage from header change
Dan Stoza [Fri, 20 May 2016 21:02:39 +0000 (14:02 -0700)]
HWC2: Fix breakage from header change

Updates the shim and adapter to handle the header changes introduced
by "HWC2: Add format to VD create, surface damage" in libhardware.

Bug: 22767098
Change-Id: I8142301c121d9b5602a4455f3fbadefefd685cf0

8 years agovulkan: Update vulkan.h to 1.0.13
Jesse Hall [Fri, 20 May 2016 14:13:52 +0000 (07:13 -0700)]
vulkan: Update vulkan.h to 1.0.13

vulkan.h and vk_platform.h imported from
github.com:KhronosGroup/Vulkan-Docs @ 2656f459333b3a1dc63619a9ebd83490eea22e93

Updated vulkan.api to reflect the changes in vulkan.h. There were no
changes to any of the code we generate. I didn't update the
vulkan_h.tmpl, since we're not currently using it and it was already
out of date.

Bug: 28878216
Change-Id: I71676683148fae7ad8799bd39386a931319aa395

8 years agoMerge "libvulkan: Don't try to load compressed/unaligned libraries from APK" into...
Jesse Hall [Fri, 20 May 2016 13:58:01 +0000 (13:58 +0000)]
Merge "libvulkan: Don't try to load compressed/unaligned libraries from APK" into nyc-dev

8 years agoFix the DUMPSYS skip.
Wei Liu [Thu, 19 May 2016 20:59:01 +0000 (13:59 -0700)]
Fix the DUMPSYS skip.

b/28869921

Change-Id: Idc108b2d3cfd237f459467a3c399a0ba5e248d24

8 years agolibvulkan: Don't try to load compressed/unaligned libraries from APK
Jesse Hall [Thu, 19 May 2016 17:58:35 +0000 (10:58 -0700)]
libvulkan: Don't try to load compressed/unaligned libraries from APK

Bug: 28825642
Change-Id: I46ea3a54010cccf9e23696a4aff52a42a31d92b1

8 years agoomx: add missing MPEG4 levels
Lajos Molnar [Thu, 19 May 2016 22:27:31 +0000 (15:27 -0700)]
omx: add missing MPEG4 levels

Bug: 28671284
Change-Id: I96b50580cc29c11732fd2b496ba6d832b7067256

8 years agoMerge "vulkan: fix double dlopen for layer libraries" into nyc-dev
Chia-I Wu [Wed, 18 May 2016 22:47:45 +0000 (22:47 +0000)]
Merge "vulkan: fix double dlopen for layer libraries" into nyc-dev

8 years agoMerge "Preserve parameters to getProcAddress functions" into nyc-dev
Mike Stroyan [Wed, 18 May 2016 18:07:00 +0000 (18:07 +0000)]
Merge "Preserve parameters to getProcAddress functions" into nyc-dev

8 years agoMerge "SF: Fix arbitrary rotation with inverse display transform" into nyc-dev
Pablo Ceballos [Wed, 18 May 2016 17:59:02 +0000 (17:59 +0000)]
Merge "SF: Fix arbitrary rotation with inverse display transform" into nyc-dev

8 years agovulkan: fix double dlopen for layer libraries
Chia-I Wu [Wed, 18 May 2016 00:50:15 +0000 (08:50 +0800)]
vulkan: fix double dlopen for layer libraries

Bug: 28826014
Change-Id: I10693426f987e63bb490ac8d1f4445ce2d195d8d

8 years agoSF: Fix arbitrary rotation with inverse display transform
Pablo Ceballos [Tue, 17 May 2016 20:34:45 +0000 (13:34 -0700)]
SF: Fix arbitrary rotation with inverse display transform

Bug 28043535

Change-Id: I8b939e17f91cf7e5b617bbcceb9df4efcc9d0161

8 years agoPreserve parameters to getProcAddress functions
Mike Stroyan [Tue, 17 May 2016 20:03:49 +0000 (14:03 -0600)]
Preserve parameters to getProcAddress functions

The contents of register r0 parameter were overwritten after commit
4690754. That used an "r" register to hold larger offsets, but doing
that overwrote r0 before it was passed on to the next call.
Applying the offset as two parts allows the total largest offset of
3588+1020 to be reached without using a register to hold that value.
Use an "addne" instruction to add the _api offset of up to 1020 within
the extensions array.
Use a "ldrne" offset to add the offset to the start of the extensions
array.

Bug: 26541117
Change-Id: Ie19e4aa584c9fa6fdec4de18e858777ac59d8dbd

8 years agoHWC2: Set composition type before setting buffer
Dan Stoza [Tue, 17 May 2016 17:48:38 +0000 (10:48 -0700)]
HWC2: Set composition type before setting buffer

SurfaceFlinger was incorrectly setting the buffer handle before
setting the composition type. This led to cases where the layer was
initially marked as Invalid or Client, allowing HWC2 to ignore the
buffer handle, even though the layer was later changed to Device
before presentation.

Bug: 28814745
Change-Id: I41189495dcb254f04e6b2b329c044301c5cff97c

8 years agoSF: Remove forced transaction on screenshot
Dan Stoza [Mon, 16 May 2016 20:36:29 +0000 (13:36 -0700)]
SF: Remove forced transaction on screenshot

Removes the forced transaction that screenshots used to perform
before capturing the screen. This was initially added to prevent
screenshot capture from sneaking in between when a transaction is
closed and when it is actually applied (at the next vsync), but in
the presence of deferred transactions today, forcing a transaction
to occur doesn't guarantee that the last transaction that
WindowManager pushed has actually been committed yet, since it may be
waiting for a buffer with which to synchronize.

If WindowManager wants to guarantee some arrangement of windows is
present for a screenshot, it needs to set the synchronous flag to
true when calling closeGlobalTransaction and wait for the call to
return.

Bug: 27098060
Change-Id: Iaa36d76df521076cc8ed16576f651543bae419a8

8 years agoMerge "dumpstate: convert sprintfs to snprintfs" into nyc-dev
TreeHugger Robot [Mon, 16 May 2016 17:12:56 +0000 (17:12 +0000)]
Merge "dumpstate: convert sprintfs to snprintfs" into nyc-dev

8 years agodumpstate: convert sprintfs to snprintfs
Nick Kralevich [Sat, 14 May 2016 23:47:44 +0000 (16:47 -0700)]
dumpstate: convert sprintfs to snprintfs

Bug: 28731007
Change-Id: Icfa4d6dfaf69e989ec785146a6bde1afb6a6f345

8 years agovulkan: check flags for early debug report callbacks
Chia-I Wu [Fri, 13 May 2016 02:45:59 +0000 (10:45 +0800)]
vulkan: check flags for early debug report callbacks

We have a different path to invoke early (that is, vkCreateInstance) debug
report callbacks.  Check for "flags" in this path as well.

Bug: 28753666
Change-Id: I49b5f2920b38ffdf69036f352166ae3b0aca62c5

8 years agolibvulkan: Load layer libraries into the app namespace
Jesse Hall [Thu, 12 May 2016 05:56:29 +0000 (22:56 -0700)]
libvulkan: Load layer libraries into the app namespace

Bug: 28213888
Change-Id: I79901693c30f0e69730643b424e504a8d3628624

8 years agolibvulkan: Discover layer libraries in entire search path
Jesse Hall [Sun, 1 May 2016 19:04:40 +0000 (21:04 +0200)]
libvulkan: Discover layer libraries in entire search path

Previously we only searched the unpacked native library path for layer
libraries. Now we search all directories in the dynamic linker's
search path, including unpacked libraries in the APK itself.

Bug: 28213888
Change-Id: I37772fd90f705592ae1725c7c8603b0174e6b133

8 years agoMerge changes from topic 'vk-debug-report' into nyc-dev
Chia-I Wu [Thu, 12 May 2016 04:25:21 +0000 (04:25 +0000)]
Merge changes from topic 'vk-debug-report' into nyc-dev

* changes:
  vulkan: improve vulkan::driver logcat messages
  vulkan: use Logger in vulkan::api
  vulkan: use Logger in the generated code
  vulkan: add DebugReportLogger
  vulkan: make debug report callbacks available in DeviceData
  vulkan: avoid duplicated app messages
  vulkan: constify DebugReportCallbackList::Message
  vulkan: refactor DebugReportCallbackList

8 years agoMerge "SF: Apply translations to transparent region" into nyc-dev
Dan Stoza [Wed, 11 May 2016 17:01:24 +0000 (17:01 +0000)]
Merge "SF: Apply translations to transparent region" into nyc-dev

8 years agoMerge "SF: Set state modified flag when changing sizes" into nyc-dev
Pablo Ceballos [Wed, 11 May 2016 17:00:22 +0000 (17:00 +0000)]
Merge "SF: Set state modified flag when changing sizes" into nyc-dev

8 years agoMerge "Move both CE/DE storage areas to adopted storage." into nyc-dev
Jeff Sharkey [Wed, 11 May 2016 03:41:25 +0000 (03:41 +0000)]
Merge "Move both CE/DE storage areas to adopted storage." into nyc-dev

8 years agoMerge "Search both CE/DE storage when clearing caches." into nyc-dev
Jeff Sharkey [Wed, 11 May 2016 03:39:58 +0000 (03:39 +0000)]
Merge "Search both CE/DE storage when clearing caches." into nyc-dev

8 years agoSF: Set state modified flag when changing sizes
Pablo Ceballos [Wed, 11 May 2016 00:15:24 +0000 (17:15 -0700)]
SF: Set state modified flag when changing sizes

The BufferRejecter updates the Layer state when it gets the first
buffer of the new size after a size change has occurred. When this
happens make sure to set the state modified flag.

Bug 27780983

Change-Id: Id8c628c40164110e75f8defd68ca895d72ed2e83

8 years agoMove both CE/DE storage areas to adopted storage.
Jeff Sharkey [Tue, 10 May 2016 23:46:39 +0000 (17:46 -0600)]
Move both CE/DE storage areas to adopted storage.

Public APIs now support separate CE/DE storage areas on all devices,
so they both need to be moved together.

Bug: 25861755
Change-Id: Ibe64a0c9ce0ef286adff2b66bfd432d14f1041d0

8 years agoSearch both CE/DE storage when clearing caches.
Jeff Sharkey [Tue, 10 May 2016 23:21:13 +0000 (17:21 -0600)]
Search both CE/DE storage when clearing caches.

Now that we have two data storage locations, we need to scan them
both when freeing up disk space.  Clean up logic to use new
get_known_users() API.

Bug: 26056125
Change-Id: Ic05c0022d6f5eb438967b69402b1900678f3858b

8 years agoSF: Apply translations to transparent region
Dan Stoza [Tue, 10 May 2016 23:19:53 +0000 (16:19 -0700)]
SF: Apply translations to transparent region

Correctly applies translation-only transforms to the transparent
region so that it aligns with the layer bounds in screen space.

Bug: 28220791
Change-Id: If9137b873f4b89890127671d8c45745e079f1cbc

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f6...
Marco Nelissen [Mon, 9 May 2016 21:38:19 +0000 (21:38 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9 am: 853702ce3d am: 775f2e6dbf am: 82afbe2258 am: c2c470c700 am: 56dea28189 am: bca20d8376 am: a8244873a8 am: 93bf31f54d
am: f91c66a9a8

* commit 'f91c66a9a889f0ed851953dd0c1b2ff70e622dda':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: Ib4a4fe7d4ca8291fb82fcbbd91c3a0803a1e1b30

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f6...
Marco Nelissen [Mon, 9 May 2016 21:32:52 +0000 (21:32 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9 am: 853702ce3d am: 775f2e6dbf am: 82afbe2258 am: c2c470c700 am: 56dea28189 am: bca20d8376 am: a8244873a8
am: 93bf31f54d

* commit '93bf31f54d56617baf0192a451f2269ad78e6c49':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: I436ae48cf652ac717ea74e2eff8a92359562e413

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f6...
Marco Nelissen [Mon, 9 May 2016 21:27:37 +0000 (21:27 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9 am: 853702ce3d am: 775f2e6dbf am: 82afbe2258 am: c2c470c700 am: 56dea28189 am: bca20d8376
am: a8244873a8

* commit 'a8244873a881f18cf5cea22dc5f566dbe6394f48':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: I351dad18e1f3b1903553a8cdfacde18c64f3b6d4

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f6...
Marco Nelissen [Mon, 9 May 2016 21:22:14 +0000 (21:22 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9 am: 853702ce3d am: 775f2e6dbf am: 82afbe2258 am: c2c470c700 am: 56dea28189
am: bca20d8376

* commit 'bca20d8376708da7ed835e71442ab52197d1e9c0':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: Ia0845cc9f58fcb2cca204118cbaff6e251ed0b8e

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f6...
Marco Nelissen [Mon, 9 May 2016 21:16:30 +0000 (21:16 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9 am: 853702ce3d am: 775f2e6dbf am: 82afbe2258 am: c2c470c700
am: 56dea28189

* commit '56dea28189ff0894d24ebb63650ad5159031549c':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: I283fd9ad78ddb434d831adaef58db22a57b335f6

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f6...
Marco Nelissen [Mon, 9 May 2016 21:11:03 +0000 (21:11 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9 am: 853702ce3d am: 775f2e6dbf am: 82afbe2258
am: c2c470c700

* commit 'c2c470c700a4e0fbbcd5b99da144f02cee7b9b3e':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: I68084d5f6bfd4b209c738b94a0f31fe9db113d9f

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f6...
Marco Nelissen [Mon, 9 May 2016 21:05:18 +0000 (21:05 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9 am: 853702ce3d am: 775f2e6dbf
am: 82afbe2258

* commit '82afbe2258c5c5e18ea6e9496d590cff177e67d4':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: I2c52991c2e6b1c5ef2eebbc4649b36c30dd9120d

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f6...
Marco Nelissen [Mon, 9 May 2016 20:59:50 +0000 (20:59 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9 am: 853702ce3d
am: 775f2e6dbf

* commit '775f2e6dbf757d1b1853337c19bf15f15a127e31':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: Ibfc37203a33c603df365ed9cbb85632d34acafbd

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9
Marco Nelissen [Mon, 9 May 2016 20:54:25 +0000 (20:54 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72 am: 275c9f60f9
am: 853702ce3d

* commit '853702ce3d1ba5e45ce58f332ed1d40008a44375':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: I2b8f6070ecc873d67be5a4c72ca870606af93a3d

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72
Marco Nelissen [Mon, 9 May 2016 20:48:32 +0000 (20:48 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle am: 1de7966c72
am: 275c9f60f9

* commit '275c9f60f94780bd686eca9750ec41cc1fafa333':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: I6516dea7eac82d06e1ffd1d269dbb6415fece948

8 years agoCorrectly handle dup() failure in Parcel::readNativeHandle
Marco Nelissen [Mon, 9 May 2016 20:43:16 +0000 (20:43 +0000)]
Correctly handle dup() failure in Parcel::readNativeHandle
am: 1de7966c72

* commit '1de7966c72981aebc3c7f9978ab129678ac89258':
  Correctly handle dup() failure in Parcel::readNativeHandle

Change-Id: Ie043622a17b241c489429273d369e9a478b7ebcc

8 years agoMerge "DO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity" into...
Nick Vaccaro [Mon, 9 May 2016 12:57:41 +0000 (12:57 +0000)]
Merge "DO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity" into cw-e-dev am: dd9012bf09  -s ours
am: d2ebf37473  -s ours

* commit 'd2ebf374732d6824fffc69fcffe9395b5a4a842c':
  DO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity

Change-Id: I43a30806ab719a017e9966570ba74725137c1cdd

8 years agoMerge "DO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity" into...
Peng Xu [Mon, 9 May 2016 12:54:22 +0000 (12:54 +0000)]
Merge "DO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity" into cw-e-dev
am: dd9012bf09  -s ours

* commit 'dd9012bf09405e038f2232a16a7748f64777a4aa':
  DO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity

Change-Id: I2a795b16c1083b03cad117b4a01dbd32df842636

8 years agoMerge "DO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity" into...
Nick Vaccaro [Mon, 9 May 2016 12:48:34 +0000 (12:48 +0000)]
Merge "DO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity" into cw-e-dev

8 years agoDO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity
Peng Xu [Thu, 23 Jul 2015 18:41:53 +0000 (11:41 -0700)]
DO NOT MERGE ANYWHERE Add AOSP Geomag and Game Rotation, and Gravity

Providing AOSP software implementation of Geomag Rotation Vector, Game
Rotation Vector and Gravity sensors for platforms that does not have
hardware implementation of these sensors but do have primitive sensors
(accelerometers, gyrometers and magnetometers).

Previously, AOSP Gravity sensor is enabled only when all primitive sensors are
available. This is changed so that AOSP Gravity will be available even
no magnetometer is in the device.

Related bug/feature request:
    * b/17508800
    * b/22610016

Change-Id: I4e2d3e544884047d66e7fdbce2282f1f8234eae9
(cherry picked from commit f66684a6fb2a2991e84a085673629db2a0494fc6)

8 years agoMerge "libgui: Prevent segfaulting in abandoned ConsumerBase" into nyc-dev
Pablo Ceballos [Fri, 6 May 2016 20:16:57 +0000 (20:16 +0000)]
Merge "libgui: Prevent segfaulting in abandoned ConsumerBase" into nyc-dev

8 years agoAdd feature strings for newly added sensors am: f50c85d263
Peng Xu [Fri, 6 May 2016 19:08:51 +0000 (19:08 +0000)]
Add feature strings for newly added sensors am: f50c85d263
am: 8778a3864c

* commit '8778a3864cedc7cb67aa893752f0b487c9de3f24':
  Add feature strings for newly added sensors

Change-Id: I4a02f3978c4c916b20b8c8b4e96231ffe4005a7f

8 years agoAdd feature strings for newly added sensors
Peng Xu [Fri, 6 May 2016 19:03:20 +0000 (19:03 +0000)]
Add feature strings for newly added sensors
am: f50c85d263

* commit 'f50c85d2631b41d9303b9726bc1312d7ce6fe1d2':
  Add feature strings for newly added sensors

Change-Id: I0f2be7c5a3dae1a690b32121d962eaa5c7384002

8 years agoMerge "atrace: introduce category for core services" into nyc-dev
Sergei Vasilinetc [Fri, 6 May 2016 17:56:44 +0000 (17:56 +0000)]
Merge "atrace: introduce category for core services" into nyc-dev

8 years agoMerge "BQ: Fix segfault in dump()" into nyc-dev
Pablo Ceballos [Fri, 6 May 2016 16:34:34 +0000 (16:34 +0000)]
Merge "BQ: Fix segfault in dump()" into nyc-dev

8 years agoAdd feature strings for newly added sensors
Peng Xu [Wed, 18 Nov 2015 03:02:41 +0000 (19:02 -0800)]
Add feature strings for newly added sensors

Added the feature strings for:
  * Ambient Temperature Sensor
  * Relative Humidity Sensor

Bug:25476695

Change-Id: Ib307bebe33c67ddb8513edfb37c697c1488f1a7f

8 years agovulkan: improve vulkan::driver logcat messages
Chia-I Wu [Tue, 3 May 2016 05:26:08 +0000 (13:26 +0800)]
vulkan: improve vulkan::driver logcat messages

Make it clear the errors are generated from internal vkGet*ProcAddr calls.
Log a message whenever stubhal is used.

Bug: 28120066
Change-Id: Iabf88757c3cb20b7ddcbe3c1d201f3d23dd3ed0b

8 years agovulkan: use Logger in vulkan::api
Chia-I Wu [Thu, 5 May 2016 03:57:23 +0000 (11:57 +0800)]
vulkan: use Logger in vulkan::api

Use debug report logger for actionable application errors (but not other
errors).

Bug: 28120066
Change-Id: Ia9a3b568e415555bf42c8a76d1df6279835a710c

8 years agovulkan: use Logger in the generated code
Chia-I Wu [Wed, 4 May 2016 08:37:23 +0000 (16:37 +0800)]
vulkan: use Logger in the generated code

Except for INIT_PROC, where ALOGE is still used for layer/HAL bugs.

Bug: 28120066
Change-Id: Icd9ce99b6bfb77816b11c05f6c8b197a6782b7cd

8 years agovulkan: add DebugReportLogger
Chia-I Wu [Tue, 3 May 2016 04:19:55 +0000 (12:19 +0800)]
vulkan: add DebugReportLogger

Add DebugReportLogger that can be used during vkCreateInstance and other
commands.  Add vulkan::driver::Logger to create a temporary
DebugReportLogger from a dispatchable object.  With them, we can log to
debug report by, for example,

  Logger(instance).Warn(obj, "failed with %s", err);

Bug: 28120066
Change-Id: If059d38d3cfcfe591031ca135695d08cad42e0a4

8 years agovulkan: make debug report callbacks available in DeviceData
Chia-I Wu [Tue, 3 May 2016 01:12:35 +0000 (09:12 +0800)]
vulkan: make debug report callbacks available in DeviceData

This allows device commands to send messages to debug report.

Bug: 28120066
Change-Id: I43d217fc23799b06fd3fda73e1876d8773c723fc

8 years agovulkan: avoid duplicated app messages
Chia-I Wu [Tue, 3 May 2016 04:06:21 +0000 (12:06 +0800)]
vulkan: avoid duplicated app messages

When vkDebugReportMessageEXT is called, every component (layer,
vulkan::driver, or hal) supporting the extension knows how to handle it.
But only one of them should handle it.

Bug: 28120066
Change-Id: I4e746b80e0920c2c63f0aceceba1601536c6236d

8 years agovulkan: constify DebugReportCallbackList::Message
Chia-I Wu [Tue, 3 May 2016 04:04:47 +0000 (12:04 +0800)]
vulkan: constify DebugReportCallbackList::Message

Sending a message should not modify the state of DebugReportCallbackList,
conceptually.

Bug: 28120066
Change-Id: I7b30b4a41492b4e670408f34a6bffe8b56295afd

8 years agovulkan: refactor DebugReportCallbackList
Chia-I Wu [Thu, 5 May 2016 04:43:03 +0000 (12:43 +0800)]
vulkan: refactor DebugReportCallbackList

Simplify DebugReportCallbackList to be a thread-safe list with three
methods:

 - AddCallback adds a node to the list
 - RemoveCallback removes a node from the list
 - Message invokes each of the nodes on the list

Add some static methods for Node* and VkDebugReportCallbackEXT
conversions.

Bug: 28120066
Change-Id: I109c6eff368cacb37508e2549dbd0b5dfa23bcb3

8 years agoMerge "Revert "Add debug dump if about to segfault"" into nyc-dev
Pablo Ceballos [Thu, 5 May 2016 21:20:40 +0000 (21:20 +0000)]
Merge "Revert "Add debug dump if about to segfault"" into nyc-dev

8 years agoBQ: Fix segfault in dump()
Pablo Ceballos [Thu, 5 May 2016 20:49:03 +0000 (13:49 -0700)]
BQ: Fix segfault in dump()

When dequeueBuffer() is called we put the slot in mActiveBuffers, then
don't hold the BufferQueue lock while allocation occurs. So a slot
might be in mActiveBuffers but not have a buffer attached yet. Prevent
the dump function from segfaulting in this case.

Bug 27128710

Change-Id: Ie1480c0f9b2544554fc3281045e55ad30605e6ec

8 years agoHWC2: Add support for color modes
Dan Stoza [Mon, 14 Mar 2016 17:47:53 +0000 (10:47 -0700)]
HWC2: Add support for color modes

Adds support for setting color modes through HWC2On1Adapter and the
HWC2 C++ shim

Bug: 22767098
Change-Id: I0b7300093ab427be7105a2535fc9db4953b8af1c

8 years agolibgui: Prevent segfaulting in abandoned ConsumerBase
Pablo Ceballos [Wed, 4 May 2016 20:59:35 +0000 (13:59 -0700)]
libgui: Prevent segfaulting in abandoned ConsumerBase

mConsumer will be null if the ConsumerBase has been abandoned. Prevent
it from being dereferenced in those cases.

Bug 27718219

Change-Id: I9a3ecadb0655ec61cd2fd15ee98b3e1bef078cff

8 years agoHWC2: Hook up setColorTransform/setLayerDataspace
Dan Stoza [Thu, 24 Mar 2016 23:19:37 +0000 (16:19 -0700)]
HWC2: Hook up setColorTransform/setLayerDataspace

Plumbs the setColorTransform and setLayerDataspace calls through the
HWC2 C++ shim and implements a trivial versions in the adapter,
which drops non-HAL_DATASPACE_UNKNOWN layers to client composition,
and which drops all layers to client composition if a color transform
is applied.

Bug: 22767098
Change-Id: Ifffd19b77cf3b33ec86fde3f72257f6b97b4dd79