| 1bff005b3dfa358722d4f0e74448c6154008ec2c |
|
25-Apr-2017 |
bohu <bohu@google.com> |
Emulator: update goldfish to 0ff252b9
Bug: 37302997
Test: Build and run sdk_google_aw_x86-userdebug emulator
To commit
0ff252b95a21998caa20403da724d6d7a0f50174 oc-emu-dev
Author: bohu <bohu@google.com>
Date: Mon Jun 27 18:23:28 2016 -0700
Enable bugreport service
Change-Id: Ie04a6f7c3a19415b63db5850dc3b9aa0488af555
(cherry picked from commit bceb8ad4f4290b0076a4315644d79cbd1b713a67)
(cherry picked from commit 0319174f8eb0f97cfe27c70d95b91d006ee6c1ce)
/device/generic/goldfish/sensors/sensors_qemu.c
|
| ccfe555a7c6eceaa4110b6ec485cb0abacc3d673 |
|
11-May-2017 |
Jim Kaye <jameskaye@google.com> |
Fix the 'flush' function for emulated sensors
Flush operations could get lost by being overwritten
if a sensor value arrives before the 'flush' is seen.
This CL prevents overwriting the 'flush' if it hasn't
been accepted.
With this change, all 51 'SensorBatchingTests' pass.
Bug: 37465457
Test: CTS: CtsSensorTestCases android.hardware.cts.SensorBatchingTests
Change-Id: Iabf58ce0868f1470a8aa4b2c2b845dc30ca466cf
(cherry picked from commit 9ca9f56b11340307425ac09a497cde29af025cc6)
/device/generic/goldfish/sensors/sensors_qemu.c
|
| def4b9beed60faea4fd40988daba1c8bb259c21d |
|
20-Jan-2017 |
Lingfeng Yang <lfy@google.com> |
[goldfish-sensors] Sanitize v1.3 sensor settings
- Stop specifying maxDelay twice
- Set on-change for sensors that should take that setting.
Change-Id: I3d6ac101579de632a7ed11f24ab9013f1cf06e1e
/device/generic/goldfish/sensors/sensors_qemu.c
|
| d0dc61679414c81f211da4ad1eb54c3ee7e95c0b |
|
20-Jan-2017 |
Lingfeng Yang <lfy@google.com> |
[goldfish] update goldfish for O
1. use latest qemu_pipe / qemd header combos
2. self contain them in goldfish/ so we don't break
anything that doesn't use goldfish but uses qemu pipe
3. stahp building the "vibrator" device for now since it uses
a lotof legacy stuff that is delted in O
4. upgrade sensor HAL impl to version 1.3
bug: 34465648
after this, rotation and camera work at least
Change-Id: I6157ede893d720acf4f8fcbe42db68a911cad531
/device/generic/goldfish/sensors/sensors_qemu.c
|
| 289fb2f10b5df0fc6daafc09153003d0b841f1d0 |
|
29-Jun-2016 |
Yurii Zubrytskyi <zyy@google.com> |
[master][sensors] Fix the emulator sensor timestamp and set max/min delays
- Emulated sensors may never have their timestamps to go greater
than the current timestamp - otherwise they appear to be
coming from the future. CTS don't like time travelers :(
- CTS issues a warning to logcat that emulated sensors don't
have maxDelay set - let's fix it as well
Change-Id: I4d45953fa99ab659c00948f7ff5a57c95b7b9b69
(cherry picked from commit d8459a08cfd0e41fe5d6e3e9866f6b396f941517)
/device/generic/goldfish/sensors/sensors_qemu.c
|
| 0f0e98eab7e2e430be1f629fcbd7c1b15fcfc0c3 |
|
18-Jun-2016 |
Yurii Zubrytskyi <zyy@google.com> |
[master][Sensors] Set a minimal delay for all sensors
If the min delay is not set, Android assumes it's infinite; then
strange things start happening
2000 mcs is the minimum emulator is able to reliably keep up to:
the current implementation may only go down to 1 ms between
ticks, and that's not enough as it needs some non-0 time for the
event generation and processing.
Bug=b/29462288
Change-Id: I791b7a32dbaf9b360d61c55beed72f2861d820c3
(cherry picked from commit 1850cf9f4a1f1053c2c2babd2c3f5f2e302c584f)
/device/generic/goldfish/sensors/sensors_qemu.c
|
| 2f4b512afd0411dc7de3531358238db43bb5648e |
|
16-Jul-2016 |
Yahan Zhou <yahan@google.com> |
Fix the naming of sensor pipe
Making it boot in the emulator.
Change-Id: I12f190f037b2a130ba0819e5ffc8e613673cb41a
/device/generic/goldfish/sensors/sensors_qemu.c
|
| 09c319886ea246a7e78ce2479db4de1d1a36865f |
|
21-Jun-2016 |
David 'Digit' Turner <digit@google.com> |
resolve merge conflicts of d6b1ea8 to stage-aosp-master
Change-Id: Ib0b4f368dae0d8c76a1c031f24785318566216f0
|
| 51a0c12d0841d45824385a567ab0699d85d86122 |
|
16-Jun-2016 |
David 'Digit' Turner <digit@google.com> |
Use new <system/qemu_pipe.h> header.
Also also remove misc. compiler warnings about unused
parameters or mismatched function pointer types.
BUG=25875346
Change-Id: I763267e2b670730e194dc361ce356e10b6aa8180
/device/generic/goldfish/sensors/sensors_qemu.c
|
| 737bb41ef2946730b0b0ec7b702ec3445047863b |
|
30-Oct-2015 |
Grigory Dzhavadyan <grigoryj@google.com> |
Add support for new sensors in the QEMU-specific sensors HAL module
Change-Id: I596aa8e90ecdb81d5984d6c2eb33410ced0fd861
(cherry picked from commit d9c1ea0d820055ab0e262958938ff9e5e4fd91de)
/device/generic/goldfish/sensors/sensors_qemu.c
|
| 8b63ce54923bf66fff337b2495e73584bdfa91f8 |
|
07-Jan-2015 |
David 'Digit' Turner <digit@google.com> |
sensors: Fix thread-safety + cleanup source code.
This cleans up the source code of the goldfish sensors module a bit:
- Implement proper thread-safety with mutexes. The HAL is pretty
specific that module methods can be called from different threads
at any time, and that the code should be thread-safe. That was
not the case.
- Get rid of un-necessary native_handle processing and dup()-ing.
- Remove the artificial distinction between 'control' and 'data'
channels, which was probably introduced to handle calls to
poll() and set_delay() from two different threads (without
proper synchronization, yuck!).
Change-Id: I3300218ca8b40c9f6df637afb4b4b0da2a446199
/device/generic/goldfish/sensors/sensors_qemu.c
|
| 58ab34d952685a7b8a52151d2e39bd58ee4f6041 |
|
10-Sep-2014 |
bohu <bohu@google.com> |
Upgrade goldfish sensor HAL module to 1_0
The previouls goldfish sensor HAL module is too outdated and
is upgraded to 1_0 to make it work again.
Change-Id: Ia5ce531836d4eecda20b74f5fd91c9804dae18a0
/device/generic/goldfish/sensors/sensors_qemu.c
|
| 8a94683196406b83b14218d1beef66067f126a16 |
|
08-Mar-2013 |
keunyoung <keunyoung@google.com> |
migrate opengl and system from development/tools
- components under system are moved one directory up like all other HALs
Change-Id: I03b870b870d83b247ac398cadfb155f03c9adfa0
/device/generic/goldfish/sensors/sensors_qemu.c
|