Should be safe to revert now, I assume?
Edit: Initially this timed out, but it seems this does no longer timeout as of today...
Good to merge?
Should be safe to revert now, I assume?
Edit: Initially this timed out, but it seems this does no longer timeout as of today...
Good to merge?
Implements #6211.
This reverts commit a33b7c9cb545985771d074748c0e368ca2d06702.
Not yet. It's building quickly now because it's been incrementally cached. The issue will be fresh builds with no cache (#7165 as an example, which will essentially be a complete rebuild).
I think we should work on speeding up tests first. I played around with parallelizing the rpc tests yesterday with reasonable success.
Additionally, https://github.com/travis-ci/travis-build/pull/706 will buy us a few extra minutes, as we'll be able to switched to the stripped down image that should startup quicker. @MarcoFalke to see where we stand, you could try merging #7165 in here and seeing how long it takes.
Maybe we can build against the platform qt in the meantime? The qt tests aren't very worthwhile, but building against qt was a good way to make sure a full compilation still works.
Concept NACK per above discussion but I am leaving this open as an issue to indicate something needs to be done here.