@kiwi
You have so far not answered any concerns what the actual plan is moving forward / how to address them more than "I want to push this".

Not sure why this is opened in Phabricator and not bugzilla and I'm kinda leaning towards not until we have a majority of local patches submitted (and preferably reviewed) upstream rather than keep ploughing on. Additionally there's no mentioning on any testing as far as consumers go.

In D60836#1395163, @fuz wrote:

In D60836#1395161, @diizzy wrote:

Well,

In D60836#1395155, @fuz wrote:

You don't really need to patch the test suite to exit with failure if it does not pass, as the make test target is mainly for human operators to look at.
That said, it's your choice and it's ok to do it if you prefer that.
No PORTREVISION bump needed.

Uhm,.. No, https://cgit.freebsd.org/ports/tree/Mk/Features/testing.mk

Yeah nobody uses that as only about half the test suites actually pass. Good luck with that.

In D60836#1395155, @fuz wrote:

You don't really need to patch the test suite to exit with failure if it does not pass, as the make test target is mainly for human operators to look at.
That said, it's your choice and it's ok to do it if you prefer that.
No PORTREVISION bump needed.

Given that you actually do a "make test" for each update accordingly you hardly need to read the whole log as your claim and it's pointless if its stale in which you need to read the log anyway to figure out when its from anyway.
There's no need at all to clutter the Makefile, you have commit messages for documentation and git log Makefile is hardly a much of an effort.

Just mentioned it in the commit message, there's no need to clutter the Makefile with that information (which also tends to go stable because people forget)...

Can we please try to add unit test when possible to at least attempt to ensure that basic functionality works as intended?