RFC V3 [1/2] test-in-container

[Florian Weimer]

On 06/28/2018 11:07 PM, DJ Delorie wrote:
> Are there good guidelines for dealing with failures in the test harness
> itself, rather than errors resulting from the actual thing being tested?
> Or do we lump them all as "fail" and let the developers sort it out?

We treat it as FAIL if we cannot attribute it to a valid, expected 
environmental limitation (such as a kernel which is too old, but still 
supported).  I don't think there is any other way to do this.

Thanks,
Florian