Skip to content

Add Egis Technology (LighTuning) Match-On-Chip driver (egismoc)

Joshua Grisham requested to merge joshuagrisham/libfprint:egismoc into master

This merge request adds a new libfprint driver that I have called egismoc with the aim and hope that it can support multiple Egis Technology (LighTuning) Match-On-Chip devices.

I am not sure how many different kind of Egis MoC devices exist "in the wild" but it seems that several of the newer Samsung Galaxy Book series notebooks carry some kind of variant of this (per #569 (closed) and #570 (closed) there are at least two different product IDs which seem to carry exactly the same endpoints, and even the same product number, and there could be more), so I have chosen the generic name egismoc for this and not something specific to just this one device that I am testing with.

I have tested this now for a few days with my Samsung Galaxy Book2 Pro as written in #569 (closed) and included also a mock test plus test capture data which seem to work quite well now that I have ironed out quite a few bugs that it helped to flush out (super great feature, by the way--I ended up needing to rewrite quite a lot of things just to fully pass the test capture process but I am positive that it made a significant difference on the quality and stability of the final result).

I thought it was also maybe worth to mention that it has been a bit of time since I have personally worked in any serious way with C, and have never worked with GLib before, but have tried to take care in reading the GLib documentation and adhering to their concepts around memory management as well as to take lead from examples that already exist in libfprint. With that said, I would ask to have a bit of help to make sure I have not left any leaks or done anything "weird" in this driver and any feedback is welcome!! 😄

There are a few "TODO" comments written in the code; I thought we could try to address these with conversations for each one within this merge request -- I will go ahead and try to kick them off along with what my contextual "question" was and then we can adjust and/or clean away the "TODO" comments as each one is addressed and resolved.

Fixes: #569 (closed)

Edited by Marco Trevisan

Merge request reports