Re: next-200110804 ARM build break (cpuidle_call_idle)

[Ingo Molnar <mingo@elte.hu>]

* Linus Torvalds <torvalds@linux-foundation.org> wrote:

> [...] In fact, I'm seriously considering a rather draconian measure 
> for next merge window: I'll fetch the -next tree when I open the 
> merge window, and if I get anything but trivial fixes that don't 
> show up in that "next tree at the point of merge window open", I'll 
> just ignore that pull request. Because clearly people are just not 
> being careful enough.

That would be really nice - in hindsight, had you done that in this 
cycle it would have solved about 80% of the merge window bugs i ran 
into!

I have trouble getting my own fixes out on time because -git is so 
untestable at the moment and i need to test on -git now that our new 
bits are upstream.

 [ In fact i'd even argue that the snapshot should be a couple of 
   days _before_ the merge window opening. The trivial annoyances and
   serious showstoppers need a couple of days of latency to get 
   discovered in linux-next and there's always a late rush of getting 
   stuff into linux-next just before the final release. linux-next
   also has weekend latency which can add up to 3 days to the 
   normally 2-3 days latency of discovering showstoppers. ]

People can still try to tempt you with late stuff, with an [RFC GIT 
PULL] tag, clearly and visibly declaring that it's a workflow 
exception and that it's fully discretional on your side with no 
maintenance problems if it's skipped to the next merge window.

Thanks,

	Ingo