Peter Barker

Merged, thanks!

Easiest fix here is probably to move those checks to AP_Arming_Copter where the arming library has access to the failsafe object.

A PR exists to remove the line from AP_Arming now: https://github.com/ArduPilot/ardupilot/pull/18358 There was some resistance at the DevCall where it was brought up for merge.

Maybe split out the enum change into another PR? I'd also suggest `option_enabled(...)` or `option_is_set(...)`rather than `option_set` as the latter could be read as "set this option"

> a small nitpick is it would be nice if we were consistent with how we pass the GCS_MAVLINK argument into methods. sometimes we have "class" in front, sometimes we...

> why is this costing more flash?  I think that says that the AP_Vehicle var_info constructor got larger for some reason. I can't fathom that so far.

`MatekF405-Wing` ``` Binary Name Text [B] Data [B] BSS (B) Total Flash Change [B] (%) Flash Free After PR (B) --------------- ------------- ----------- ------------- ---------------------------- ------------------------- antennatracker 20 (+0.0030%) 0...

Test infinite-loop missions in autotest

@Williangalvani is simply reverting that commit going to cause you problems with your workflow? It's biting more people, so we should probably do *something* here :-)

On Tue, 28 Nov 2017, Fnoop wrote: > If LOG_BACKEND_TYPE is set to 3, and a mavlink connection is not present for > whatever reason, then PREARM fails. It would...