OSDN Git Service

android-x86/hardware-interfaces.git
6 years agoFix some stk api Radio Vts tests.
Sanket Padawe [Wed, 31 May 2017 23:06:04 +0000 (16:06 -0700)]
Fix some stk api Radio Vts tests.

Bug: 37502436
Test: Manual vts
Change-Id: Idd9a1fd38413aed89aaa152cb9a91c001f18b339
Merged-in: Idd9a1fd38413aed89aaa152cb9a91c001f18b339

7 years agoAudio VTS: Enforce policy configuration format
Kevin Rocard [Wed, 24 May 2017 18:01:34 +0000 (11:01 -0700)]
Audio VTS: Enforce policy configuration format

Validate the audio_policy_configuration.xml supplied in /vendor against
the format schema.

Test: Run the VTS test on target
Test: vts-tradefed run commandAndExit vts --skip-all-system-status-check --primary-abi-only --skip-preconditions --module VtsHalAudioV2_0Target -t CheckConfig.audioPolicyConfigurationValidation
Bug: 35700978
Change-Id: I800e692a6ff4f64655007c33af7e34d879ee1132
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio VTS: Test now use Android.mk
Kevin Rocard [Wed, 31 May 2017 00:15:28 +0000 (17:15 -0700)]
Audio VTS: Test now use Android.mk

To validate the audio_policy_configuration.xml against a schema,
libxml2 is used.
Unfortunately, libxml2 uses Android.mk in oc-dev thus can not be used
from the tests which uses Android.bp.

It has been deemed safer to transform the tests to use Android.mk rather
than backporting the Android.bp patch from master.

Test: Compile
Bug: 35700978
Merged-In: I800e692a6ff4f64655007c33af7e34d879ee1132
Change-Id: Ie3f2069d60be369b4c37c27e9c1cf9372323a19c
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoMerge "[RenderScript] Use OVERRIDE_RS_DRIVER string in the default implementation...
Miao Wang [Fri, 2 Jun 2017 23:20:57 +0000 (23:20 +0000)]
Merge "[RenderScript] Use OVERRIDE_RS_DRIVER string in the default implementation." into oc-dev

7 years ago[RenderScript] Use OVERRIDE_RS_DRIVER string in the default
Miao Wang [Tue, 16 May 2017 22:36:54 +0000 (15:36 -0700)]
[RenderScript] Use OVERRIDE_RS_DRIVER string in the default
implementation.

Bug: 37035983
Test: mm
Test: manually push to sailfish & angler, verify GPU driver successfully
load, and tests are passing.

Change-Id: Ic55c74228e7eae599702365c513009aa3454656b
Merged-In: Ic55c74228e7eae599702365c513009aa3454656b

7 years agoVTS Audio Effect: Do not check the result of IEffect::offload method
Mikhail Naganov [Thu, 1 Jun 2017 16:58:06 +0000 (09:58 -0700)]
VTS Audio Effect: Do not check the result of IEffect::offload method

HAL may not support offloading, and a call to IEffect::offload
is allowed to return an error code.

Update the test to only check that the call to the method doesn't
crash.

Bug: 38437110
Test: still passes
Change-Id: I6cfd25998edb9c68d2e585e572ba8ba3a3427e98

7 years agoMerge "Compile manager-lib shared library" into oc-dev
TreeHugger Robot [Thu, 1 Jun 2017 22:07:00 +0000 (22:07 +0000)]
Merge "Compile manager-lib shared library" into oc-dev

7 years agoMerge "Audio: Add XML schema for audio policy configuration" into oc-dev
TreeHugger Robot [Thu, 1 Jun 2017 19:56:20 +0000 (19:56 +0000)]
Merge "Audio: Add XML schema for audio policy configuration" into oc-dev

7 years agoMerge "Revert "audiohal: Add diagnostics to investigate HAL call crashes"" into oc-dev
TreeHugger Robot [Thu, 1 Jun 2017 16:32:28 +0000 (16:32 +0000)]
Merge "Revert "audiohal: Add diagnostics to investigate HAL call crashes"" into oc-dev

7 years agoAnnotate DeleteKey test with bug number.
Shawn Willden [Mon, 24 Apr 2017 16:48:16 +0000 (10:48 -0600)]
Annotate DeleteKey test with bug number.

Bug: 37623742
Change-Id: I2cb4bb39bacb3aabf5fff7937f60c7355c9636e3
Test: VTS test, manually run

7 years agoCheck return from DeleteKey correctly.
Shawn Willden [Thu, 1 Jun 2017 13:44:27 +0000 (07:44 -0600)]
Check return from DeleteKey correctly.

DeleteKey may legitimately return ErrorCode::UNIMPLEMENTED rather than
ErrorCode::OK, but the VTS test didn't allow that in all cases.  In many
case the return code was also left unchecked.

Test: adb shell/data/nativetest64/VtsHalKeymasterV3_0TargetTest/VtsHalKeymasterV3_0TargetTest
Bug: 62193967
Change-Id: I19a90a87850675b0700baf7409e57098e0584d54

7 years agoaudiohal: Prevent logspam when calling get_presentation_position
Mikhail Naganov [Fri, 19 May 2017 23:46:28 +0000 (16:46 -0700)]
audiohal: Prevent logspam when calling get_presentation_position

get_presentation_position can return ENODATA if the stream
has stopped but the write is still querying the position.

Bug: 38376424
Change-Id: I9d516a19fb347843a8ec8e8e9a0f5eab0b0798e6
Test: no log messages from StreamHAL about get_presentation_position
(cherry picked from commit 96a3a19beda45bef7c34fc06a0048174f3a7e362)

7 years agoAudio: Add XML schema for audio policy configuration
Eric Laurent [Thu, 18 May 2017 01:32:36 +0000 (18:32 -0700)]
Audio: Add XML schema for audio policy configuration

The schema describes the XML format for audio policy
configuration.
It will be enforced in a follow up vts test.

Bug: 35700978
Test: schema is parsed with xmllint with no errors
Test: xmllint  --xinclude --noout --schema audio_policy_configuration.xsd  out/target/product/sailfish/vendor/etc/audio_policy_configuration.xml

Change-Id: Ic6830446b54b3e02e4bca9b0a392523a8fb359e1
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoGenerate serial number randomly in VTS unit test
sqian [Wed, 31 May 2017 21:04:36 +0000 (14:04 -0700)]
Generate serial number randomly in VTS unit test

Any fixed integer can be returned by vendor's APIs response.
Ideally it should be a random value for testing a vendor library.

Applying this to all the unit tests serial number.

Cherry-pick from:
  https://googleplex-android-review.git.corp.google.com/#/c/2330821/

Bug: 38499889
Test: adb shell data/local/tmp/VtsHalRadioV1_0TargetTest
Change-Id: I491db829cca607fad9f6a21987de92269d115253
Merged-In: I491db829cca607fad9f6a21987de92269d115253
(cherry picked from commit 0e8a3be41c7c0b0a8de9740524b8cac49c685c99)

7 years agoRevert "audiohal: Add diagnostics to investigate HAL call crashes"
Mikhail Naganov [Wed, 31 May 2017 19:59:21 +0000 (19:59 +0000)]
Revert "audiohal: Add diagnostics to investigate HAL call crashes"

This reverts commit 6c0f76a684bc58eb10181f71d9f45c5799ca7c6a.

Since the root cause of the crash has been established, there is no need to keep this code around.

Bug: 36225019
Change-Id: I74e570e863a0cdec5d9029f1672e2e8066c246b5

7 years agoOmx vts tests.
Ram Mohan M [Mon, 29 May 2017 13:20:01 +0000 (18:50 +0530)]
Omx vts tests.

Bug:32023356
Test: make vts.

bug fix : fix issues in port reconfiguration

soft and vendor omx components communicate differently. The existing
design doesnt cover this. This is now fixed

Change-Id: I898e60c93961f468e749ab5f13db996c5fd53967

add support for additional color formats

tested encoder for bgra & 420sp color format inputs.
added code to get/set encoder latency.
some more necessary code cleanup.

Change-Id: I05b79d0a7706e20afaf91c88caf3e523c98b4540

IOmxBufferSource callback tests

Change-Id: I2826597c6ab110a0bf52290e3c52ff3fb3c6ea23

7 years ago[WIFI] Remove dependency on vendor-specific library
Etan Cohen [Wed, 3 May 2017 16:12:50 +0000 (09:12 -0700)]
[WIFI] Remove dependency on vendor-specific library

The libcld80211 is vendor-specific. Remove from dependecy list.

(cherry-pick of commit aa281aa266d6c8701c2ecbf3f49f505a2ecddae3)

Bug: 37901207
Test: builds and runs
Merged-In: If251002c2dd7a0e30ca18620edd905a63217f28a
Change-Id: If251002c2dd7a0e30ca18620edd905a63217f28a

7 years agoMerge "configstore-utils: vendor_available" into oc-dev
Jayant Chowdhary [Wed, 31 May 2017 03:59:44 +0000 (03:59 +0000)]
Merge "configstore-utils: vendor_available" into oc-dev

7 years agoMerge "GNSS HAL radio group for E911 SUPL" into oc-dev
TreeHugger Robot [Wed, 31 May 2017 03:56:10 +0000 (03:56 +0000)]
Merge "GNSS HAL radio group for E911 SUPL" into oc-dev

7 years agoconfigstore-utils: vendor_available
Steven Moreland [Fri, 28 Apr 2017 01:32:39 +0000 (18:32 -0700)]
configstore-utils: vendor_available

Test: with BOARD_VNDK_VERSION := current

(cherry picked from commit 65af0612ce5bd5f8d86e489b83971c2bd17badb2)

Merged-In: Ied49d6affe69099bc08a58e1d5e38052d018a2e8
Change-Id: I6278d659957829bce4ebdf44615ad25b371658d8

7 years agoMerge "Camera: fix release fence FD leaks" into oc-dev
TreeHugger Robot [Tue, 30 May 2017 22:27:58 +0000 (22:27 +0000)]
Merge "Camera: fix release fence FD leaks" into oc-dev

7 years agoCompile manager-lib shared library
Alex Yakavenka [Fri, 12 May 2017 00:18:13 +0000 (17:18 -0700)]
Compile manager-lib shared library

bug: 38424350
Test: Local build
Change-Id: Idd03b8bac365f17b7fc418016d2c250a199384fe

7 years agoGNSS HAL radio group for E911 SUPL
Wyatt Riley [Tue, 2 May 2017 00:07:26 +0000 (17:07 -0700)]
GNSS HAL radio group for E911 SUPL

Bug: 37730994
Test: Builds, boots, basic GPS works
Change-Id: I2ceda822d6414348f61c183765f68068764d36cc
Merged-In: I2ceda822d6414348f61c183765f68068764d36cc

7 years agoCamera: fix release fence FD leaks
Yin-Chia Yeh [Fri, 26 May 2017 21:01:32 +0000 (14:01 -0700)]
Camera: fix release fence FD leaks

Test: use sw_sync to fake release fence FDs, GCA, CTS
Bug: 62070085
Change-Id: Iae77a2e112df5363e55e4177656a5dd41b830cbd

7 years agoBluetooth: Remove scheduler-service dependency
Myles Watson [Fri, 26 May 2017 17:09:07 +0000 (10:09 -0700)]
Bluetooth: Remove scheduler-service dependency

Bug: 38178265
Test: Bluetooth enable/disable
      adb shell stop && adb shell VtsHalBluetoothV1_TargetTest

Change-Id: I7f892bad73f72b46b7b1720bda267fdddad21948

7 years agoMerge "Omx vts test." into oc-dev
Zhuoyao Zhang [Fri, 26 May 2017 00:15:41 +0000 (00:15 +0000)]
Merge "Omx vts test." into oc-dev

7 years agoMerge "Fix typos in IEvsEnumerator descriptive comment" into oc-dev
TreeHugger Robot [Fri, 26 May 2017 00:14:46 +0000 (00:14 +0000)]
Merge "Fix typos in IEvsEnumerator descriptive comment" into oc-dev

7 years agoOmx vts test.
Ram Mohan M [Thu, 11 May 2017 13:22:47 +0000 (18:52 +0530)]
Omx vts test.

Bug:32023356
Test: make vts.

Merged-In: I83b454b999203fb94bc4cc50c91bd99f788131b3

secure buffer allocation for secure components

As of now, we see a crash in encoder test application on secure components
This is due to nullptr access. setting property media.mediadrmservice.enable
to 1 causes a different crash sigabrt(). In decoder securebufferallocation
call passes but we havent done anything with it. This commit needs more work.

Change-Id: I19127e39ad7daf66ac5277406e3857ec45c99e0a

add debug code

This can come in handy to view the decoded/encoded content or
for computation of psnr

Change-Id: I80e60349c76c02e5098df667223a0227f59b8324

video encoder test with anw input buffer

In this commit, anw buffers are provided as input for encoding
as opposed to byte buffers. In process IOmxNode, IGraphicBufferSource api
calls are tested

Change-Id: I7ec4af0746fe59221de42e56b344852d8fadc4f2

add support for additional color formats

Change-Id: Ia88ef9c95882958a68fee5cc68e146c3502a1b48

add eos and timestamp deviation test to video encoder

Change-Id: I9ebeaa53a986ec3f8a2ef55306877aec808d2add

add metamode support for video decoders

Change-Id: I7d6d6991cdc0ed36241d0e99c7a23675664acb10

add a timeout while processing input buffers

do not wait on input buffers processing to complete for an
indefinite amount of time. wait for a predefined duration and
leave

Change-Id: Icceaf0737a52e62f47bc052367e7ddbdc1868a46

move duplicate code to a library

move routines common across audio, video, component folders to a static
library

Change-Id: I55bf21e47571490e989b52b82c48c6e9a4b23745

add end of stream test for audio encoder

add eos test for audio encoder and some more code cleanup

Change-Id: I9d670ed53f6bba5802f919ec915e67bb0fa83518

Ensure all the buffers allocated are cycled

use all buffers provided by the component (even if it is greater than the
number of buffers needed by the bitstream)

Change-Id: I83b454b999203fb94bc4cc50c91bd99f788131b3

move duplicate code to a library - II

Change-Id: I375603feb9da303adc6a2d75ef59e43a11c3bd6d

Change-Id: I83b454b999203fb94bc4cc50c91bd99f788131b3
(cherry picked from commit 1da607c513c74612127cd18e83dcca53afb2f38f)

7 years agoCamera: Silence log spam
Shuzhen Wang [Thu, 25 May 2017 17:34:13 +0000 (10:34 -0700)]
Camera: Silence log spam

HAL is allowed to return captureResult from different thread contexts.
Do not do ALOGW if this is the case.

Test: Run camera and check logspam is gone
Bug: 62068207
Change-Id: I86e77fc578953cefecf10ebc4d395edf33627e8c

7 years agoMerge "Fix more radio VTS test." into oc-dev
TreeHugger Robot [Thu, 25 May 2017 17:06:28 +0000 (17:06 +0000)]
Merge "Fix more radio VTS test." into oc-dev

7 years agoMerge "camera.device@3.2 wrapper: Close release fence FDs" into oc-dev
TreeHugger Robot [Thu, 25 May 2017 01:50:45 +0000 (01:50 +0000)]
Merge "camera.device@3.2 wrapper: Close release fence FDs" into oc-dev

7 years agocamera.device@3.2 wrapper: Close release fence FDs
Eino-Ville Talvala [Wed, 24 May 2017 22:05:56 +0000 (15:05 -0700)]
camera.device@3.2 wrapper: Close release fence FDs

Freeing the native handle isn't enough; we need to close the fence FDs
after sending them to the camera service, since that's the behavior
that the legacy HAL interface requires.

Test: Camera CTS continues to pass.
Bug: 38482274
Change-Id: I879de566dcafecc1b899d79e68b38ea39b91a37e

7 years agoMerge changes from topic 'vintf_compat_mat_update' into oc-dev
Yifan Hong [Tue, 23 May 2017 22:35:10 +0000 (22:35 +0000)]
Merge changes from topic 'vintf_compat_mat_update' into oc-dev

* changes:
  Add <kernel> entry to framework compat mat
  Update framework compatibility matrix
  Fix automotive.* hals for oc-dev
  Add evs and vehicle hal to compatibility matrix

7 years agoMerge "Temporary disable tests failing tests for O" into oc-dev
TreeHugger Robot [Tue, 23 May 2017 21:39:10 +0000 (21:39 +0000)]
Merge "Temporary disable tests failing tests for O" into oc-dev

7 years agoFix more radio VTS test.
Sanket Padawe [Wed, 10 May 2017 18:29:33 +0000 (11:29 -0700)]
Fix more radio VTS test.

+ Increase timeout to 65s to fix some tests failing because of timeouts.

Bug: 35800570
Test: vts
Change-Id: Idf0204908592eac5f2a954d4425123df75999b1c

7 years agoMerge "Increase poll max size, avoid performance issue when batching" into oc-dev
Peng Xu [Tue, 23 May 2017 20:33:28 +0000 (20:33 +0000)]
Merge "Increase poll max size, avoid performance issue when batching" into oc-dev

7 years agoAdd <kernel> entry to framework compat mat
Yifan Hong [Tue, 16 May 2017 01:04:06 +0000 (18:04 -0700)]
Add <kernel> entry to framework compat mat

3.18, 4.4 and 4.9 are the supported kernels in O.

Test: builds
Test: adb shell vintf (RuntimeInfo is compatible with
 framework compat mat)

Bug: 38324908
Change-Id: I8b2c98e8168af651215a309c8b0b93d3c342f97d

7 years agoUpdate framework compatibility matrix
Yifan Hong [Tue, 23 May 2017 00:35:52 +0000 (17:35 -0700)]
Update framework compatibility matrix

Test: builds (compat check passes with assemble_vintf)

Change-Id: I23a8b3ab1db2344c8f246349aabd99ed42a4ac92

7 years agoIncrease poll max size, avoid performance issue when batching
Peng Xu [Thu, 4 May 2017 23:36:37 +0000 (16:36 -0700)]
Increase poll max size, avoid performance issue when batching

Vts polls sensor events with max size of 1, which can cause
performance issue when batching. Increase it to 64 to help eliminate
potential performance issue.

Note this CL has no change in behavior of test.

Test: VTS target test pass on several devices
Bug: 62027156
Change-Id: I04a6791f407ec0a54eb6cd8df4d4ef41e8a89c12
Merged-In: I04a6791f407ec0a54eb6cd8df4d4ef41e8a89c12
(cherry picked from commit e337dfe3d250dabcbdc9d4c8e9620f76b96b4562)

7 years agoFix automotive.* hals for oc-dev
Yifan Hong [Tue, 23 May 2017 18:16:20 +0000 (11:16 -0700)]
Fix automotive.* hals for oc-dev

Test: builds
Change-Id: I12795a37d3bbd1508e059a79bb61e1b4bc1a9a3f

7 years agoAdd evs and vehicle hal to compatibility matrix
Scott Randolph [Thu, 11 May 2017 17:16:12 +0000 (10:16 -0700)]
Add evs and vehicle hal to compatibility matrix

Test: builds
Change-Id: I61dcc985d1d4dde4b4b53e36b0de1d4dcff8d3f4
Merged-In: I61dcc985d1d4dde4b4b53e36b0de1d4dcff8d3f4

7 years agoMerge "wifi(vts): Add retry for HAL start" into oc-dev
TreeHugger Robot [Tue, 23 May 2017 00:25:30 +0000 (00:25 +0000)]
Merge "wifi(vts): Add retry for HAL start" into oc-dev

7 years agoFix typos in IEvsEnumerator descriptive comment
Scott Randolph [Mon, 22 May 2017 23:35:39 +0000 (16:35 -0700)]
Fix typos in IEvsEnumerator descriptive comment

IEvsEnumerator::openCamera had a typo and the incorrect name of a related
interface in one of the descriptive comments.  This fixes those issues
and updates the checksum for the interface.

Bug: 38508696
Test: build evs_app and evs default driver.
Change-Id: I24f3ca2bc3652711a4ee2ea9a37428d4bdafc509

7 years agoMerge "Move media_profiles_V1_0.dtd to hardware/interfaces" into oc-dev
TreeHugger Robot [Mon, 22 May 2017 23:53:33 +0000 (23:53 +0000)]
Merge "Move media_profiles_V1_0.dtd to hardware/interfaces" into oc-dev

7 years agowifi(vts): Add retry for HAL start
Roshan Pius [Mon, 22 May 2017 21:16:06 +0000 (14:16 -0700)]
wifi(vts): Add retry for HAL start

Port of the retry logic added in HalDeviceManager for temporary IWifi.start()
failures when invoked immediately after IWifi.stop().

Bug: 38342536
Test: Run VTS tests
Change-Id: Ie58a08c45979bafa0c61fa1c8536a2c0a0cd8ed9

7 years agoUpdate makefiles. (2/2)
Steven Moreland [Fri, 19 May 2017 22:52:22 +0000 (15:52 -0700)]
Update makefiles. (2/2)

Adds default configuration for all hals.

Bug: 38415912
Test: pass
Change-Id: Idd1f3a2b7b16ad956d31784a513e93a066cdd02e

7 years agoUpdate makefiles.
Steven Moreland [Thu, 18 May 2017 23:14:16 +0000 (16:14 -0700)]
Update makefiles.

Adds default configuration to all hals.

Test: pass
Merged-In: Icdf6b56df82c4b7b7de815e7febea9a07ab96856
Change-Id: Icdf6b56df82c4b7b7de815e7febea9a07ab96856

7 years agoMerge "VTS Audio Effect: Allow 'NOT_SUPPORTED' from IEffect::offload method" into...
TreeHugger Robot [Fri, 19 May 2017 20:28:08 +0000 (20:28 +0000)]
Merge "VTS Audio Effect: Allow 'NOT_SUPPORTED' from IEffect::offload method" into oc-dev

7 years agoVTS Audio Effect: Allow 'NOT_SUPPORTED' from IEffect::offload method
Mikhail Naganov [Fri, 19 May 2017 17:55:17 +0000 (10:55 -0700)]
VTS Audio Effect: Allow 'NOT_SUPPORTED' from IEffect::offload method

HAL may not support offloading, and a call to IEffect::offload
is allowed to return an error code.

Update the test to accept either OK or NOT_SUPPORTED.

Bug: 38437110
Test: still passes
Change-Id: I1822010eed73f61bb0e0ae1852739b47fa72eade

7 years agoMerge "Add <interface> / <instance> to fwk compat mat" into oc-dev
Yifan Hong [Fri, 19 May 2017 16:36:14 +0000 (16:36 +0000)]
Merge "Add <interface> / <instance> to fwk compat mat" into oc-dev

7 years agoMove media_profiles_V1_0.dtd to hardware/interfaces
Pawin Vongmasa [Thu, 18 May 2017 11:07:59 +0000 (04:07 -0700)]
Move media_profiles_V1_0.dtd to hardware/interfaces

Test: media_profiles_V1_0.dtd exists in system/etc
Bug: 36430270
Merged-In: I43d6eb68cc866a280cbe24fcf0f0af75a3c3ae67
Change-Id: I43d6eb68cc866a280cbe24fcf0f0af75a3c3ae67

7 years agoMerge changes from topic 'fix-vts-test' into oc-dev
Shuo Qian [Thu, 18 May 2017 22:56:14 +0000 (22:56 +0000)]
Merge changes from topic 'fix-vts-test' into oc-dev

* changes:
  update .hal file to fix vts test
  Fix vts tests

7 years agoupdate .hal file to fix vts test
sqian [Wed, 17 May 2017 03:29:19 +0000 (20:29 -0700)]
update .hal file to fix vts test

Test: Manual
Bug: 35671595
Change-Id: Ic97cc8e9a713bfa22f9986c18ddb5af624142b17

7 years agoFix vts tests
sqian [Tue, 16 May 2017 20:01:20 +0000 (13:01 -0700)]
Fix vts tests

- apduReq

(with QC UIM document)
- handleStkCallSetupRequestFromSim
- supplyNetworkDepersonalization

Test: make vts; vts-tradefed; run vts -m VtsHalRadioV1_0Target
Bug: 35671595
Change-Id: Ie7427e027eb93e89a9790260cfdbbb2d01c50cc1

7 years agoAdd <interface> / <instance> to fwk compat mat
Yifan Hong [Thu, 18 May 2017 20:46:57 +0000 (13:46 -0700)]
Add <interface> / <instance> to fwk compat mat

Test: bullhead, m manifest.xml -j
Test: marlin, m manifest.xml -j
Test: new devices, m manifest.xml -j
(for marlin and new devices, assemble_vintf
is invoked to check compatibility)

Bug: 38395193
Change-Id: I4201bec4753e7efe41678a736dd0a124b995fbf9

7 years agoMerge "audio HAL: fix getParameters status check." into oc-dev
TreeHugger Robot [Thu, 18 May 2017 17:24:29 +0000 (17:24 +0000)]
Merge "audio HAL: fix getParameters status check." into oc-dev

7 years agoaudio HAL: fix getParameters status check.
Eric Laurent [Thu, 18 May 2017 01:32:36 +0000 (18:32 -0700)]
audio HAL: fix getParameters status check.

Bug: 38326193
Test: VTS tests for module VtsHalAudioV2_0TargetTest.

Change-Id: If4109cea68120d722a3b1084ad583170e950cfa0

7 years agoTemporary disable tests failing tests for O
Kevin Rocard [Fri, 12 May 2017 22:25:09 +0000 (15:25 -0700)]
Temporary disable tests failing tests for O

Those tests are failing because their fix
(I0ea1a4bf952527ac0711ed3cdd31c6aa53db4d2e) is dangerous to merge in O
so close to the final release.

The fix implements a rejection feature for unsupported audio format.
This rejection is required by the hal interface
but was only partially implemented in the hal.
The problem is that the framework was never tested with a hal supporting
this behavior nor some calls implemented with this requirement.
As a result I do not feel confident in merging it in oc-dev,
it might complex, hard to detect bugs.

The fix will be merged in master thus will be part of mr1.
The current patch will be reverted in master after the fix is merged.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: Ia9e1dcec5442e2755f0bcb13d8d502663f507dbe
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agosound trigger VTS: used default HAL service name
Eric Laurent [Wed, 17 May 2017 19:50:15 +0000 (12:50 -0700)]
sound trigger VTS: used default HAL service name

Bug: 37864311
Test: run VTS for module VtsHalSoundtriggerV2_0Target on marlin
Change-Id: If727dcc5921742ece6252dff839ce8100f24ccd1

7 years agoMerge changes from topic 'audio-primary-hal-vts-fix' into oc-dev
TreeHugger Robot [Tue, 16 May 2017 18:52:39 +0000 (18:52 +0000)]
Merge changes from topic 'audio-primary-hal-vts-fix' into oc-dev

* changes:
  Audio HAL: A speech volume outside of [0,1] is an error
  Audio HAL VTS: Some methods are not optional

7 years agoMerge "Revert "Fix resource leaks in drm hal"" into oc-dev
Jeff Tinker [Tue, 16 May 2017 18:25:50 +0000 (18:25 +0000)]
Merge "Revert "Fix resource leaks in drm hal"" into oc-dev

7 years agoRevert "Fix resource leaks in drm hal"
Jeff Tinker [Tue, 16 May 2017 18:21:39 +0000 (18:21 +0000)]
Revert "Fix resource leaks in drm hal"

This caused b/38343526 which is a chase list issue, reverting for now.

This reverts commit 4e1fe736af1b88485bba762b11bc0c11a447819c.

Change-Id: Ib9ffcd11e575e3f05f54cbea17fda2f2eb3780b5

7 years agoMerge "Omx vts tests." into oc-dev
Zhuoyao Zhang [Mon, 15 May 2017 23:10:36 +0000 (23:10 +0000)]
Merge "Omx vts tests." into oc-dev

7 years agoAudio HAL: A speech volume outside of [0,1] is an error
Kevin Rocard [Tue, 9 May 2017 00:08:11 +0000 (17:08 -0700)]
Audio HAL: A speech volume outside of [0,1] is an error

Hals are supposed to received normalized volumes, between 0 and 1.
Previously volumes outside [0,1] were clamp to this range.
This clamping has the capability to hide bugs thus return an error if
such volume is received.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: Iab70f9c651540ea2434d10939d28c1c842db19e0
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoNFC: vts: Add commands to reset the NFC controller before other commands
Ruchi Kandoi [Fri, 12 May 2017 21:19:25 +0000 (14:19 -0700)]
NFC: vts: Add commands to reset the NFC controller before other commands

Bandwidth and WriteInvalidAndThenValidCommand test assumes that the NFC
controller will be initialized before the test. This is not necessarily
true for all devices. Send CORE_RESET_CMD and CORE_INIT_CMD before
creating a loop-back connection.

CORE_RESET_RSP doesn't necessary return the configuration that was sent
in CORE_RESET_CMD. For instance, if CORE_RESET_CMD requested keeping
the configuration, and if the controller doesn't support that, the
configuration will be resetted and CORE_RESET_RSP will reflect that.

Bug: 38267593
Test: vts-tradefed run commandAndExit vts --skip-all-system-status-check
      --primary-abi-only --skip-preconditions
      --module VtsHalNfcV1_0Target -l INFO
Change-Id: I65ae9746054ca767b928bdd461faa02f03719aea
Signed-off-by: Ruchi Kandoi <kandoiruchi@google.com>
(cherry picked from commit 209a274c31f8654000561b080f64ee84dd221ed9)

7 years agoMerge "Fix resource leaks in drm hal" into oc-dev
TreeHugger Robot [Mon, 15 May 2017 19:54:09 +0000 (19:54 +0000)]
Merge "Fix resource leaks in drm hal" into oc-dev

7 years agoOmx vts tests.
Ram Mohan M [Fri, 28 Apr 2017 06:48:35 +0000 (12:18 +0530)]
Omx vts tests.

Bug: 32023356
Test: make vts.

cover basic encoder & decoder test to all components

Till the previous commit, few components went untested.
This commit covers all softomx components.

Test: tested on softomx components

Change-Id: I4c996a3743a6190d9c6d4c835aa89daefdc14658

enumerate profile and level for encoder components

Change-Id: I4b9d3b3106d33a0241a7e36bd7189c1b3ef40ef0

bug fix: omx android compile as 32bit on 64bit platforms

thanks lajos

Change-Id: If11b20eeb6a75319f9e0cd6a62524dadab2ba8af

Add more VTS tests for IOmxStore and IOmxNode

Change-Id: I3401025b4f0b821f469222707c75b49fe4f0b0c5

avoid false flagging of errors

If testing of the given component is not supported application do not flag
an error, instead exit gracefully.

Change-Id: Ic91a92614d18b204cc50b7d8f6dc1a0ca81ff782

Add support for g711, gsm and raw audio decoder components

Change-Id: Ied70abdabb397a7a63e13293a8b35365a6b23e4f

split eos test

the eos test in decoders handles three different cases. These are now
separated in to different test fixtures.

Change-Id: I30c102109ce60a345f878d03d481337462fc817f

Change-Id: I3401025b4f0b821f469222707c75b49fe4f0b0c5

7 years agoAudio HAL VTS: Some methods are not optional
Kevin Rocard [Wed, 3 May 2017 18:16:05 +0000 (11:16 -0700)]
Audio HAL VTS: Some methods are not optional

Some mandatory methods were allowed to return NOT_SUPPORTED
although their implementations is mandatory.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: Ibe4b3cf73257309975ed11269a38315051fa9064
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoMerge "graphics: set maxThreads to 4 for the composer" into oc-dev
Chia-I Wu [Fri, 12 May 2017 19:50:55 +0000 (19:50 +0000)]
Merge "graphics: set maxThreads to 4 for the composer" into oc-dev

7 years agographics: set maxThreads to 4 for the composer
Chia-I Wu [Thu, 11 May 2017 22:20:10 +0000 (15:20 -0700)]
graphics: set maxThreads to 4 for the composer

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

7 years agoBluetooth: Add a second thread for the scheduler
Myles Watson [Thu, 11 May 2017 23:39:59 +0000 (16:39 -0700)]
Bluetooth: Add a second thread for the scheduler

Test: service call bluetooth_manager 6 # (Stop Bluetooth)
      /data/nativetest64/VtsHalBluetoothV1_0TargetTest/\
          VtsHalBluetoothV1_0TargetTest
Bug: 3817826534461621
Change-Id: Ia8f6683ca47c9deac43d1047846caee5688f1eca
(cherry picked from commit 74c601361c5f2d68561b545b7b7a52a3916ba8e9)

7 years agoMerge changes from topic 'audio-primary-hal-vts-fix' into oc-dev
Kevin Rocard [Thu, 11 May 2017 22:34:39 +0000 (22:34 +0000)]
Merge changes from topic 'audio-primary-hal-vts-fix' into oc-dev

* changes:
  Audio HAL: A volume/gain outside of [0,1] is an error
  Audio HAL VTS: Log test unexpected behaviour
  Audio HAL VTS: Improve ASSERT of Result and Return
  Audio HAL VTS: stopped stream state getters may return INVALID_STATE
  Audio HAL VTS: setGain allow -0.0
  Audio HAL VTS: Some methods are optional
  Audio HAL VTS: GetPresentationPosition may return 0 on stop stream
  Audio HAL VTS: getDevice() == NONE => not supported
  Audio HAL VTS: Getter test assert logic was incorrect
  Audio HAL VTS: debugDump can only test for crash
  Audio HAL VTS: Sanitize prepareFor{Writing,Reading} input size
  Audio HAL VTS: refactor prepareFor{Reading,Writing}
  Audio HAL VTS: Sanitize setMode input
  Audio HAL VTS: Allow OK when setting a non existing parameter
  Audio HAL VTS: differentiate getParam success/failure/not_implemented
  Audio HAL VTS: Fix style on modified files

7 years agoMerge changes from topic 'fix-hidl-generate' into oc-dev
Steven Moreland [Wed, 10 May 2017 20:02:38 +0000 (20:02 +0000)]
Merge changes from topic 'fix-hidl-generate' into oc-dev

* changes:
  Update makefiles for c++-sources and c++-headers.
  Update makefiles for c++-sources and c++-headers. (2/2)

7 years agoMerge "Start processes that need real-time with CAP_SYS_NICE." into oc-dev
Martijn Coenen [Wed, 10 May 2017 17:55:46 +0000 (17:55 +0000)]
Merge "Start processes that need real-time with CAP_SYS_NICE." into oc-dev

7 years agoUpdate makefiles for c++-sources and c++-headers.
Steven Moreland [Tue, 9 May 2017 21:34:19 +0000 (14:34 -0700)]
Update makefiles for c++-sources and c++-headers.

Test: pass
Bug: 38174080
Merged-In: I683455736eed35e6ced24e2763c2bdb5fb6c9601
Change-Id: I683455736eed35e6ced24e2763c2bdb5fb6c9601

7 years agoUpdate makefiles for c++-sources and c++-headers. (2/2)
Steven Moreland [Wed, 10 May 2017 02:51:19 +0000 (19:51 -0700)]
Update makefiles for c++-sources and c++-headers. (2/2)

Test: pass
Bug: 38174080
Change-Id: Icad451f9f9ad5f46412356a4171a8a1222f8e545

7 years agoMerge "Bugfix: Add HIDL_FETCH_IScheduleTest declaration" into oc-dev
PO HUNG CHEN [Wed, 10 May 2017 02:28:10 +0000 (02:28 +0000)]
Merge "Bugfix: Add HIDL_FETCH_IScheduleTest declaration" into oc-dev

7 years agoMerge "Update IRadioResponse.hal documentation to fix failing VTS" into oc-dev
Sanket Padawe [Wed, 10 May 2017 02:27:04 +0000 (02:27 +0000)]
Merge "Update IRadioResponse.hal documentation to fix failing VTS" into oc-dev

7 years agoUpdate IRadioResponse.hal documentation to fix failing VTS
Naina Nalluri [Fri, 5 May 2017 18:00:05 +0000 (11:00 -0700)]
Update IRadioResponse.hal documentation to fix failing VTS

update IRadioResponse.hal documentation with
nas/data error codes

Bug: 35671595
Test: Only doc update.
Change-Id: I5a106e05b487f7deecb208af6465e5ea5238b26e

7 years agoCamera: Remove TODOs in interfaces
Eino-Ville Talvala [Tue, 9 May 2017 18:13:20 +0000 (11:13 -0700)]
Camera: Remove TODOs in interfaces

These are tracked elsewhere and won't be addresed for the initial
interface versions.

Update current.txt hashes; comment-only update

Test: Compiles
Bug: 34108979
Bug: 33269977
Bug: 33262893
Change-Id: I0c0521cf69649d871f64a67773ce822db2487044

7 years agoFix resource leaks in drm hal
Jeff Tinker [Tue, 9 May 2017 02:27:10 +0000 (19:27 -0700)]
Fix resource leaks in drm hal

bug:36408047
test: manual tests to verify leaks are fixed
Change-Id: Idd79e99e211d81f8b3e58cf52c9b01ce9d370dc4

7 years agoMerge "Camera: Clean up TODOs that aren't going to be addressed." into oc-dev
TreeHugger Robot [Tue, 9 May 2017 17:30:35 +0000 (17:30 +0000)]
Merge "Camera: Clean up TODOs that aren't going to be addressed." into oc-dev

7 years agoMerge "Add DOOR_LOCK to default config." into oc-dev
TreeHugger Robot [Tue, 9 May 2017 17:21:13 +0000 (17:21 +0000)]
Merge "Add DOOR_LOCK to default config." into oc-dev

7 years agoStart processes that need real-time with CAP_SYS_NICE.
Martijn Coenen [Tue, 9 May 2017 16:57:00 +0000 (09:57 -0700)]
Start processes that need real-time with CAP_SYS_NICE.

Bug: 37293077
Test: verified CAP_SYS_NICE granted in /proc/pid/status
Change-Id: If404450cbeb077d331b231bda1af2d6eae774b75

7 years agoAudio HAL: A volume/gain outside of [0,1] is an error
Kevin Rocard [Tue, 9 May 2017 00:08:11 +0000 (17:08 -0700)]
Audio HAL: A volume/gain outside of [0,1] is an error

Hals are supposed to received normalized volumes, between 0 and 1.
Previously volumes outside [0,1] were clamp to this range.
This clamping has the capability to hide bugs thus return an error if
such volume is received.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: Ia4880bdff6111cbcdae6a4ebee921eddae141ee4
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoCamera: Clean up TODOs that aren't going to be addressed.
Eino-Ville Talvala [Mon, 8 May 2017 22:29:28 +0000 (15:29 -0700)]
Camera: Clean up TODOs that aren't going to be addressed.

These TODOs aren't in scope for the wrapper libraries.

Leave the few TODOs that have bugs tracking them, and the ones that
are from autogenerated metadata.

Test: Builds
Bug: 34108979
Change-Id: Ib28467a4fbcfaab4a0809334bc62eabb9290797a

7 years agoAudio HAL VTS: Log test unexpected behaviour
Kevin Rocard [Mon, 8 May 2017 18:53:07 +0000 (11:53 -0700)]
Audio HAL VTS: Log test unexpected behaviour

Some test output infos that are useful to understand how they run.
Unfortunately the xml report does not seem to be saved by VTS tradefed.
Thus output them in logcat.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: I9a2cc10160c3b1c8f81db0464efbc6b26600cadc
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: Improve ASSERT of Result and Return
Kevin Rocard [Wed, 3 May 2017 02:21:58 +0000 (19:21 -0700)]
Audio HAL VTS: Improve ASSERT of Result and Return

Previously tests on Result and Return were using ASSERT_NO_FATAL_FAILURE
and helper methods.
This leaded to complex error messages were the error did not pointed to
the helper methods instead of the ASSERT_RESULT call.
Additionally SCOPE_TRACE messages are repeated
for each ASSERT_NO_FATAL_FAILURE level.

Use ::testing::AssertionResult to improve drastically the error messages.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: If705502546606d678df3f1966b0cd9f3ef8c2529
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: stopped stream state getters may return INVALID_STATE
Kevin Rocard [Wed, 3 May 2017 17:57:06 +0000 (10:57 -0700)]
Audio HAL VTS: stopped stream state getters may return INVALID_STATE

A never started stream should have its render position and next write
timestamp at 0 or indicate that the state is invalid.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: I62e16066bb22101ee8f75154fc6c85a66be2f402
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: setGain allow -0.0
Kevin Rocard [Wed, 3 May 2017 18:19:25 +0000 (11:19 -0700)]
Audio HAL VTS: setGain allow -0.0

Previously -0.0 was consider an invalid value as not in the range [0,1].
But it is quite difficult in C++ to differentiate -0.0 and 0.0 as -0.0 == 0.0
and such difference has no impact in practice.

Thus leave the implementation support or not -0.0.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: Ia0ebcb325f77adcf8471620f418da1cbe8995e36
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: Some methods are optional
Kevin Rocard [Wed, 3 May 2017 18:16:05 +0000 (11:16 -0700)]
Audio HAL VTS: Some methods are optional

Although the method documentation does not say it,
some HIDL interface methods are optional.

Update the tests to allow NOT_SUPPORTED to be returned.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: If31acc2dbdb6d1d563910e85c99401c48f4f3f86
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: GetPresentationPosition may return 0 on stop stream
Kevin Rocard [Wed, 3 May 2017 17:52:43 +0000 (10:52 -0700)]
Audio HAL VTS: GetPresentationPosition may return 0 on stop stream

GetPresentationPosition returns the last time a sample was written to
the hardware.
Calling it on a stop stream should return 0 sample written, but the
timestamps of that measure is of little importance.
Thus allow this timestamp to be 0 indicating that the measure was never
actually made.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: I3c33b60f98e8cbea269a7739cc1889af932dcff5
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: getDevice() == NONE => not supported
Kevin Rocard [Wed, 3 May 2017 02:34:29 +0000 (19:34 -0700)]
Audio HAL VTS: getDevice() == NONE => not supported

getDevice does not return a Result, thus it can not return NOT_SUPPORTED.
Consider NONE as not supported.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: I3b6f7a1fbc1d1535faf549f5b031461cb39d1722
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: Getter test assert logic was incorrect
Kevin Rocard [Wed, 3 May 2017 01:58:58 +0000 (18:58 -0700)]
Audio HAL VTS: Getter test assert logic was incorrect

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: Iaf2d71829a15b12dcf56e825773c8a697896a264
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: debugDump can only test for crash
Kevin Rocard [Wed, 3 May 2017 01:41:46 +0000 (18:41 -0700)]
Audio HAL VTS: debugDump can only test for crash

DebugDump does not return an error code, thus the implementation can not
return not implemented.

As a result, the test can not expect any output from the function.
Only test that the call does not crash and add a log if the function is
probably not implemented in the test.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: I2c18958bceb1eb638491f9afce9d8e8025ccd3ec
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: Sanitize prepareFor{Writing,Reading} input size
Kevin Rocard [Wed, 3 May 2017 01:34:59 +0000 (18:34 -0700)]
Audio HAL VTS: Sanitize prepareFor{Writing,Reading} input size

Return an error if framesCount or frameSize are null to avoid a division
by zero when calculating the buffer size.

The message queues are allocated with a buffer size but if two big they will
assert not return an error.
Thus take some margin on the buffer size check.

Note that both function should be refactored as 99% identical.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: I0576e9016ef2e567c8d4e171c6237883d9865db9
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: refactor prepareFor{Reading,Writing}
Kevin Rocard [Wed, 3 May 2017 01:31:24 +0000 (18:31 -0700)]
Audio HAL VTS: refactor prepareFor{Reading,Writing}

Those functions had lots of copy paste on errors and
the following patch will even add more error detections.
Refactor the hidl_cb call to avoid all duplication.
Note that both function should be refactored as 99% identical.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: I40d6926b4f9f5e3aba51e878f55fb013f4ca09c1
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: Sanitize setMode input
Kevin Rocard [Wed, 3 May 2017 01:16:00 +0000 (18:16 -0700)]
Audio HAL VTS: Sanitize setMode input

Some values of AudioMode are exposed although implementation detail.
Make sure the client can not use them.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: If513c2a06efa8a92459f0af80c63232fc63302b0
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoAudio HAL VTS: Allow OK when setting a non existing parameter
Kevin Rocard [Wed, 3 May 2017 17:43:21 +0000 (10:43 -0700)]
Audio HAL VTS: Allow OK when setting a non existing parameter

setHwAvSync and setParameters were implemented in the pre-hidl interface
as set_parameters.
Unfortunately set_parameters did not return an error if a key was not
implemented.

As most HIDL implementation will be a wrapper around the pre-hidl
interface, allow those functions to return OK on not implemented key.

Test: vts-tradefed run vts --module VtsHalAudioV2_0Target
Test: call/play music/record/video...
Bug: 36311550
Change-Id: Icfcaa02b7d63e03375fddc90dc5a803754c1874f
Signed-off-by: Kevin Rocard <krocard@google.com>
7 years agoMerge "Revert "Set omx and configstore as optional."" into oc-dev
Yifan Hong [Mon, 8 May 2017 21:10:56 +0000 (21:10 +0000)]
Merge "Revert "Set omx and configstore as optional."" into oc-dev