Addressed all feedback, you might want to squash the commits since the second doesn't have a good commit message.

@Thomasdezeeuw Please go ahead and squash the commits on your end and force-push back to this branch.

@Susurrus can't you use the squash and merge button on GitHub? Force pushing to Github never seems to work for me.

We don't use GitHub to merge PRs, we rely on Bors, an external tool. It doesn't squash commits. So we rely on the PR authors to have their commits in order before we merge it.

@Susurrus I didn't know that, squashed the commits so this is ready to go.

It's somewhat mentioned in the CONYRIBUTING.MD document, but not very well. I've been meaning to update that actually as this comes up a lot.

It seems the build bot is failing, it times out after 5 minutes and then fails to kill it.

command timed out: 300 seconds without output running ['rustup', 'run', '1.20.0', 'cargo', 'test'], attempting to kill
SIGKILL failed to kill process
using fake rc=-1
program finished with exit code -1
remoteFailed: [Failure instance: Traceback from remote host -- exceptions.RuntimeError: SIGKILL failed to kill process
]


buildbot.process.remotecommand.RemoteException: exceptions.RuntimeError: SIGKILL failed to kill process
Traceback (most recent call last):
Failure: exceptions.RuntimeError: SIGKILL failed to kill process

And trying to digg in further fails to load (part of) the page.

It looks like my VM suddenly got about 10 times slower. Problems with the VM host? Meltdown/Spectra countermeasures? I don't know. I've bumped the timeouts and I'm restarting the builds.

bors r+

Build succeeded

• buildbot/nix-rust/nix amd64_fbsd11
• buildbot/nix-rust/nix i386_fbsd11
• continuous-integration/travis-ci/push