Linux 4.9 rc1

I usually do the releases on a Sunday afternoon, but occasionally cutthe merge window short by a day just to keep people on their toes, andmake sure people learn not to send in last-minute pull requests. Nogaming the merge window to the last day. This is one such release.To be fair, the reason I did it a day early this time around is lessto stop people from trying to time their pull requests, and mostlybecause this has been a pretty big merge window, and not hugelyenjoyable. I ended up stopping doing pulls twice during the mergewindow just because I was chasing down some random problem. That tendsto turn my busy merge window time from "busy" to "somewhat stressful".But hey, it's all good now, and while 4.9 looks to be a big releaseand we had a couple of hiccups, on the whole things look normal. Thebig new thing is the greybus addition, which Greg swears is actuallygetting used. But the bulk of the changes by far is actually a lot ofsmall details under the hood, as usual.My own favorite "small detail under the hood" happens to be AndyLutomirski's new virtually mapped kernel stack allocations. They makeit easier to find and recover from stack overflows, but the effortalso cleaned up some code, and added a kernel stack mapping cache toavoid any performance downsides. Al has also been working on some vfsand uaccess cleanups (particularly a goo splice model cleanup) that Ifollow. But realistically, what _I_ consider cool small details isjust my own personal thing, there's things all over.The virtual stack mapping also happens to mean that people who try todo DMA from temporary buffers on the stack ("Don't do it!") now reallyneed to change their evil ways. So there is some fallout from this,and I expect a couple of drivers to need minor fixes. But it's all fora good cause, really (and it isn't all that common, because doing DMAfrom the stack really has never been a good idea, and is generally noteven workable in most situations).But there really is a lot of other things going on, and the shortlogthat I do for other releases is much too big during rc1. So as usual,I'm appending my "mergelog" instead, which gives a very high-levelview of what I merged and from whom. And as usual, I want to point outthat the person I merge from is not necessarily the person who did thework: we had 1500 people involved in this release, only the top-levelmaintainers get credited in my mergelog.Go forth and test,                 Linus