Re: [Ksummit-discuss] new topic: regression test infrastructure? (was Draft Maintainer's Summit Agenda and Attendees - )

["Theodore Y. Ts'o" <tytso@mit.edu>]

On Tue, Oct 16, 2018 at 09:53:06PM +0200, Christian Borntraeger wrote:
> 
> 
> On 10/16/2018 04:12 AM, Theodore Y. Ts'o wrote:
> > This is the draft agenda for the Maintainer's Summit which I and the
> > program committee have come up with:
> > [..] 
> > 2:30 TBD
> > [..]
> > 4:30 TBD
> 
> Does it make sense to talk about existing test infrastructure 
> (like syzbot,Geert Uytterhoeven and Guenter Roecks reports as well as all the others)
> What architectures are missing? Is it ok to run tests in-house and only report errors?
> What test scenarios are missing?

This would probably be better suited at the Kernel Summit track in
Vancouver next month, as there will be a much larger number of kernel
developers there.  The Maintainer's Summit is intended to focus much
more process issues than specific technical questions.

Also, some of these questions are best answered by those who can bring
resources to the table.  For example, it's easy for someone to point
out that some architecture, like Itanium or pa-risc are "missing".
But the people to whom we would need to be having that conversation
would be Intel and HP, respectively, and both would likely say that
those architectures are obscelent, and so they would probably decline
to provide those resources.

Similarly, the question of whether, say, Power, or Z series, or Risc-V
are "missing" are best targetted at the companies which back those
architectures.

					- Ted