It's Sunday afternoon, so it must be time for a release candidate!
Things are looking fairly normal, and there are fixes all over, with
drivers and architecture code leading the charge as usual, but there's
stuff spread out all over the place, including filesystems,
networking, mm, library helpers, etc etc.
There's still a known nfs regression pending, but nobody outside of
some explicit stress-testing seems to have noticed, so I made a rc2
release despite knowing of the problem. I'll get a pull request from
Al later and we'll have that fixed for rc3. In the meantime I expect
that the only people who would run into it are doing filesystem
stress-testing and they already are testing the fix.
There's a late non-fix I took even though the merge window is over,
because I've been wanting it for a while. I doubt anybody notices the
actual effects of a pty change/cleanup that means that our old
disgusting DEVPTS_MULTIPLE_INSTANCES kernel config option is gone,
because the cleanup means that it is no longer needed. We used to have
the situation that certain distributions wanted that config option
disabled for legacy reasons, and others needed it enabled for modern
behavior. Our /dev/ptmx behavior has been cleaned up and fixed so
that it JustWorks(tm) in both cases, and that nasty wart is just gone.
Plus the code is nicer anyway.
I mention that non-fix probably mainly because I feel a bit guilty
about taking it, since I would probably have shouted at some
submaintainer that tried to call that cleanup a late fix. Quod licet
Iovi, and all that jazz.. But really, no excuses except for "finally".
Anyway, so there's a little bit of everything in there, and it's small
enough that you can peruse the appended Shortlog for details.
Linus
