Linux 3.15 rc1 out, merge window closed
Apr 14, 2014, 19:00 (0 Talkback[s])
(Other stories by Linus Torvalds)
It's been two weeks since 3.14 was released, and -rc1 of 3.15 is now
tagged and pushed out, and the patches and tar-balls are going through
the compressors on kernel.org as I write this. Which means that the
merge window is closed, and people should send me fixes only.
And quite frankly, it's about time. This release doesn't really have a
lot of odd things going on, but it's *big*. Sure, we've had releases
with more files and lines changed (3.7-rc1 and 3.11-rc1 in
particular), but those tended to have something particular going on
(3.7-rc1 saw the largely automated UAPI header file disintegration,
and 3.11 saw the bug staging lustre merge).
In comparison to those large releases, 3.15-rc1 is just big in
general. No single big thing, but just lots and lots of commits. Sure,
it has a few big new staging drivers (rtl8723au in particular), but
even when big, those aren't nearly the bulk of things. There's just a
lot going on
In fact, we have the biggest number of commits in recent history
(maybe ever), at just over 12000 non-merge commits (and about 800
And it really is all over the map. The bulk is driver changes, at
about three quarters of the actual patch. Staging shows up pretty
prominently, but it's really all over the driver map, with networking,
sound, media, gpu, block drivers..)
But there's tons of non-driver stuff too. Outside of the driver
subdirectories, architecture updates account for about half the
changes (with ARM leading the way, largely due to device-tree
descriptors, but there's mips, x86, powerpc, s390, blackfin..). And
the rest is pretty varied too, with core networking, documentation,
kernel, mm, tools etc.
So while drivers and architecture updates are the bulk of it, we
really do have a lot of core changes too.
Anyway, even more than usual, the -rc1 is much too big to include a
shortlog of all commits. But the shortlog of merges I've done might
give at least some overview of all the changes. As usual, the people
credited in the mergelog are the maintainers I pull from, not the
developers that wrote the code. You can see that in the full git logs.
Anyway, because -rc1 is already pretty darn big, I do *not* want to
hear about "sorry this missed the window, can I still sneak in". Fixes
The only exception to that is a couple of pending things that came in
during the merge window, but were explicitly delayed. So we do have a
fbdev file movement pending (I'll do the file movement after -rc1 just
to make things easier to see in the history and not mix up movement
with development). And there was one namespaces/mounting pull request
that I didn't pull, but that might still make it in after some more
commentary/work. That will *probably* be delayed 3.16, but we'll see
how much TLC that thing needs..