Re: [LSF/MM TOPIC] FS, MM, and stable trees

From: Greg KH
Date: Wed Feb 13 2019 - 04:18:09 EST

On Wed, Feb 13, 2019 at 11:01:25AM +0200, Amir Goldstein wrote:
> I think the main difference between these review announcements
> and true CI is what kind of guaranty you get for a release candidate
> from NOT getting a test failure response, which is one of the main
> reasons that where holding back xfs stable fixes for so long.

That's not true, I know to wait for some responses before doing a
release of these kernels.

> Best effort testing in timely manner is good, but a good way to
> improve confidence in stable kernel releases is a publicly
> available list of tests that the release went through.

We have that, you aren't noticing them...

> Do you have any such list of tests that you *know* are being run,
> that you (or Sasha) run yourself or that you actively wait on an
> ACK from a group before a release?

Yes, look at the responses to those messages from Guenter, Shuah, Jon,, Red Hat testing, the Linaro testing teams, and a few other
testers that come and go over time. Those list out all of the tests
that are being run, and the results of those tests.

I also get a number of private responses from different build systems
from companies that don't want to post in public, which is fine, I
understand the issues involved with that.

I would argue that the stable releases are better tested than Linus's
releases for that reason alone :)


greg k-h