User space interface for GPIO interrupts / Status Report Week 4

Christian Krämer uddka at student.kit.edu
Thu Jun 14 09:33:37 UTC 2018


Hi,

during last week I fixed the handler for the poll() and select() syscalls and implemented a ioctl request to retrieve the configuration. I also made some improvements to the test utilities and the driver. Furthermore I thought about automated long-term test scenarios and came up with two practical ideas:

- Implementation of a pseudo random number generator on a second device whose output is connected to a GPIO pin of the device under test. This scenario can be validated by implementing edge recovery and replaying the PRNG with the same seed on the DUT.
- Implementation of a discrete-time stochastic process as source for GPIO interrupts. For validation of this method either the PRGN can also be replayed on the DUT with the same seed to predict the timestamps when events should occur and compare it to the actual timestamp when the event arrives or another communication channel between the two devices to share information (like the count of occurred events for instance) can be used.

In the course of this week I am going to implement a handler for the kqueue()/kevent() syscalls and start to prepare everything for the hardware tests an start with their implementation.

The source code of the driver can be found in my fork of the FreeBSD src tree on GitHub [1]. The utilities for configuration and testing are located in a seperate repository [2].

Regards,
Chris

[1] <https://github.com/ckraemer/gsoc2018-utils>
[2] <https://github.com/ckraemer/freebsd/tree/gsoc2018>


More information about the soc-status mailing list