On Sat, Jun 03, 2023 at 12:07:40AM +0800, Kent Gibson wrote: > > One other thing - is it necessary for gpio-sim to log hogs? > > For the long running cases this results in a lot of logging, which could > be causing issues in itself, and further muddying the waters. > Ah, I've been looking at bisecting the problem the wrong way. I should've been bisecting the trigger condition. If I remove the hogs from the basic_sim then the test doesn't fail. It also explains why I haven't been seeing this more often - most of my tests don't include hogs. Only the info tests use hogs, the rest just use bare sims with no named lines or hogs (aka simpletons). Does that help isolate the root cause? Cheers, Kent.