Is: automated testing before David M picks them up? Was: Re: [Xen-devel] [PATCH net v2] xen-netback: bookkeep number of active queues in our own module

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

> > Your previous patch didn't do this, and I really am suspect as to
> > whether you functionally tested and verified this aspect of your
> > change at a ll.
> > 
> 
> I've done some testing.
> 

Hey David M,

First of sorry for hijacking this thread - but it seemed like the
perfect thread.

David Vrabel raised an interesting point is that we do have an automated
build/test system that finds bugs - and in fact the 3.16-rc1 netfront
and netback bugs were discovered by that (Boris watches them like
hawks and pounces on them immediately).

That being said - all patches that go through the Xen tree (xen/tip.git)
go through that - and also the ones that maintainers do a git pull
on (say for drivers/block/*). But for xen-net* we don't do that since
well.. it hasn't been a problem in the past and we never formulated
a policy for that. (It was easy with Jens' because he likes the
GIT PULL mechanism and we just created a branch in the tree - and I also
managed to cause some embbarasing bugs so to save my face I am now
testing it religiously).

Having said that - how do you handle such situation with sub-maintainers
wanting to do some testing before they are cleared to go your way?