Seems that the fast blinking notification LEDs experienced on some
Samsung devices was due to the removal of these two lines that clean
the 'blink' array. Propose re-inserting to correct the fast blinking
LED issue.
Change-Id: I88ea828a50468aae494b1dd54290607308de543a
* Because we're using #ifdef in set_light_buttons(),
defining VAR_BUTTON_BRIGHTNESS to anything, including 0,
results in the button brightness code being enabled
* That results in broken touchkey backlight for devices
that don't override the samsung_lights header, or defined
this macro to 0 in their device's samsung_lights header
* Comment out the macro to correct this logic
Change-Id: Icf554b0deac975f729f1a70e517341dae74f3657
* Try to read the hardware revision from the
bootloader before parsing /proc/cpuinfo
* Set more properties to catch all recent device
needs as well
* Introduce a property to let other services (cbd)
know they can start
Change-Id: I50d8ec37921ec2559bdd7fe852d830f10fe1c12e
AOSP power hints can have a NULL data parameter,
only our custom hints are guaranteed to provide data.
Change-Id: Ibea874e75a7092b3f9c1c56dc5182cdf73f35532
The hispeed_freq value is read only in the inizialization process.
If we decide to change it later writing a new value to the node
this new value wont be taken into consideration by the power hal.
Lets update the hispeed_freq value in the power hal before using
it to limit the max cpu freq when we go into power save mode.
Change-Id: I4b8c7f67259750b2931b18c3871874022aa6054d
* As of now the code account only max_brightness > 255
-> Check if max_brightness != 255 and apply the scaling
even if the max_brightness is less then 255
Change-Id: I809ece5844c8ccf20ae6507f23d308fe3ddfe94b
* Simplify interactive changes
* Better support for non-touchkey devices
* Only perform read I/O in samsung_power_set_interactive
if needed (improves wake->panel_on delay)
Change-Id: If7bd6a1833ac7605db437ef3664230f9a452d58b
* Use O_WRONLY for write
* Return errno for write
* Log all failures
* Fix read_int():
We would always treat strtol(...) = 0 as an error,
although the real value can very well be 0.
Check the end pointer to fix this.
Change-Id: Id2bc9acc103a7a6400dd06f3904af37cbb0b5174
Some devices (often T-Mobile variants) have extra hardware for
incall audio processing.
Audio must be routed to the ES IC properly so it can be passed to
ALSA, otherwise the RX/TX streams are lost.
Change-Id: Ib29c747d5728a09726e14bab00f26ad273400aba
When we change the device during a phone call we need to completely
restart it. This is required by the modem.
Change-Id: I902c78bf50dfb5dc3da51fca0fb862f872df86ca
The old exynos4x12 lights HAL had logic for scaling the brightness of
lights according to the light's type and also scaling each R/G/B colour.
Bring this logic across, and make the battery LED not eye-searing at
night.
Based on work by Daniel Hillenbrand.
REGRESSION-66
Change-Id: Ia6108468f57c3b6c6a1a68da76a1c3362fd982a5
Exynos4 devices send two ints as a response to this request,
causing responseFailCause to think the result is a LastCallFailCause struct,
which it then tries to treat as a Java UTF-8 encoded string, causing a
crash.
Work around this by forcing responseInts (the original behaviour)
on Exynos4 devices.
Change-Id: Ied14948db3049066ffc760b6d3df09b4cefa2df0
Jason Vidal