After installing the driver it must be executed.
Due to the wide range of Android targets, some of which actively block the installation of native executables there is no single method to achieve this. If there is sufficient interest an Android service based implementation of this driver will be developed to provide a more consistent approach to invoking the driver.
Currently a method appropriate to the target system must be selected. The following are some example methods.
Run directly from terminal
Enter the following command in the terminal
/system/vendor/updd/upddenv /system/vendor/updd/updd -e
This is useful for initially evaluating the driver.
The command must be reentered if the system is restarted.
Using Tasker and other 3rd party tools
The Tasker tool provides a simple method to launch processes
you should configure to run the command
at system reboot time.
This method is useful for deploying to small numbers of devices
Other tools that may be of interest: init.d scripts support; Universal Init.d APK
All these tools are used at your own risk!
Android init system
In principle adding the command
to the file /init.rc will launch the driver at startup.
Unfortunately the root location ( / ) is not persistent. So any changes are lost on reboot.
There are 3 possible approaches
- If the system init.rc (or any file referenced by it) is on a persistant storage location (e.g. /system) then the referenced file can be edited.
- If you have access to the boot image file boot.img you can unpack it, edit /init.rc and repack as described in this article . The boot.img file contains content that is copied to / at startup.
- If you are building the Android kernel from source you can prepare init.rc with the required edit prior to building.
If viable for your system, this approach is useful for deploying to a large number of target devices.