Need to apply the sub-HAL index byte to ensure handles provided in
DYNAMIC_SENSOR_META events align properly with the [dis]connected
callbacks.
Bug: 228879057
Test: try on device with dynamic sensors sub-HAL at index > 0
Merged-In: Ic7290b51733bb829a5494ef41e9430b51dc6074e
Change-Id: Ic7290b51733bb829a5494ef41e9430b51dc6074e
* Required by some OEM sensor implementations, e.g.
ASUS ZenFone 8, ASUS ROG Phone 3, OnePlus 9. This
is done so /dev/input/event* nodes can be read by
the QCOM sensor HAL.
Test: Verified light and proximity sensor work on ASUS
ZenFone 8 & ROG Phone 3 (android-12.1.0_r7).
Change-Id: I1a2d7384da0b159a4b59fcf38c37117d43b3b127
VTS ensures that 2.0 HALs don't expose the hinge angle sensor,
but the 2.1 HALs must run 2.0 VTS tests. This CL ensures that
we don't expose the hinge angle sensor on the 2.0 HAL, to avoid
failing those VTS tests.
Bug: 173135479
Test: run vts -a x86 -m VtsHalSensorsV2_0TargetTest
Change-Id: I944bfb0b01be0831e45cfd3d6ed7dab6770d1b80
Moving to asynchronous capture listener allows us to fix previous race
on GraphicBuffer and make sure it won't be blocked by any incident e.g.
fence timeout.
Also, guard updateScreenBuffer() with mutex to make sure lastScreenUpdate
can be updated safely.
Extends ThreadPoolMaxThreadCount to 1 because 0 is a confusing value.
Although we set max threads to 0 previously, extra binder thread was
still started according to the warning message in startThreadPool(),
which led to the race noted above as sigaction can also be run on binder
thread.
Also add verbose logging so we can do quick debugging with LOG_NDEBUG=0.
Change-Id: I4733919bc20b3c1b382e60b71c5ceaa10dedaf49
Signed-off-by: LibXZR <i@xzr.moe>
* That way video recording apps will choose
16-bit PCM by default instead of 24-bit
one which happens to have quite low volume
on this device.
Change-Id: I2130c6a759afcd50a985b2f034b5c0b5374287af
- while lz4hc,9 provides better compression ratio, it has the cost of
tremendous compression time and potential hit while decompressing
- internal tests on aarch64 seem to suggest the decompression hit
is larger than x86_64
Change-Id: I5c102c03f95ace982b85a6ffa5c33aa9ce4637a4
Signed-off-by: Omkar Chandorkar <gotenksIN@aospa.co>
Since commit: eaaf5331d6
multiple intensities configuration flag has been changed to an interger
reflect that change in here to allow multiple intensity vibration pattern in settings.
Change-Id: I13f9dbf6280623ad1c954ce96ceecbe363140062
Signed-off-by: UtsavBalar1231 <utsavbalar1231@gmail.com>
* I may tweak this at a later point, or split it up across devices,
depending on user feedback
* Below around 15% force 90 hz
* To get the brightness in integer from the UI percentage, use
"settings get system screen_brightness"
The missing diag property actually broke the ability to use diag mode.
Change-Id: I135dc2f739e969a30a269652f599f4a79bbafceb
Signed-off-by: Omkar Chandorkar <gotenksIN@aosip.dev>
Yixuan Wang
Brian Duddie
Alexander Koskovich
karthik bharadwaj
Stan Rokita
LuK1337
LibXZR
Fabian Leutenegger
Omkar Chandorkar
Rituj Beniwal
TimoDo
Jake Weinstein
John Galt
Vishalcj17
UtsavBalar1231
Adithya R
Tyler Trephan
Lisa Liu
Danny Baumann
Omkar Chandorkar