Driver madness 7: Deep dark

[okawo80085]

All socket logic was is now handled by lazy-sockets.

Posers are actually notified about bad packets and posers will get notified on driver exit. Also improved manager handling.

Split the entire driver from a giant cpp file, into a bunch of small dedicated files for simpler management.

Plus comments and generally nicer code.

[okawo80085]

Turns out lazy-sockets had issues building on Linux, thanks to @guysoft for catching it.

The issue is already fixed with lazy-sockets itself, we just need to update it's submodules.

[okawo80085]

So there is a Windows bug with the tracking stream, still need to track it down, hope its a lazy-sockets issue.

Thanks to @SimLeek for catching it.

[okawo80085]

Ok turns out the issue is a bit deeper, the manager device is not getting activated at all on Windows.

No idea why yet, its getting activated normally on Linux... Feels like im up for another hell of a headache with Windows :/

[okawo80085]

Well i can't handle the MR right now because of war.

@SimLeek can you try to get the windows version up and running? affected files are

driver_hobovr.cpp
tracking_references.h
tracking_references.cpp

The manager device class is called HobovrTrackingRef_SettManager, in driver_hobovr.cpp it should be getting activated on line 152, but on windows for some reason it never gets to the Activate method of the manager class, no idea why.

Everything else should work fine.

[okawo80085]

Sike, i have internet again, can't run tests tho, don't have access to VR capable hardware at the moment :/

[okawo80085]

Windows build is broken by gaze master plugins, i'll fix it tomorrow.

[okawo80085]

CI builds are currently broken cuz virtual-environments/issues#2667

Will push a dependency fix later

[Minothor]

if needs be, I can write a hotfix to the ci to pull in the necessary lib manually.

[okawo80085]

Thats pretty much the only way to get it right now, but we can just take the steps for getting it from that github issue