Re: [git pull] vfs.git

[Linus Torvalds <torvalds@linux-foundation.org>]

On Mon, May 16, 2016 at 11:27 PM, Al Viro <viro@zeniv.linux.org.uk> wrote:
>
> FWIW, I think the least PITA would be if I send #work.lookups + backmerge from
> #ovl-fixes to resolve the fs/overlay/super.c conflict.  If you think that
> it needs to done in even smaller steps (after all, it's about 3/4 of the
> entire pile), please say so.

That looked ok and at least separates out the other fluff from the real meat.

I would probably have preferred to get the "lookup" part separately as
a first merge, and then do the readdir work as an independent pull a
day or two later, just because they are reasonably independent.

But I took it this way as one bigger pull instead. With the
lookup_open() work being in the middle there, it wasn't a
black-and-white totally independent series anyway.

I tried to turn your description of the work into a reasonable merge
message, because I think that's a useful explanation of what went on
if there are issues. And while I hope that everything "just works",
this is definitely conceptually one of the biggest vfs changes we've
had in a long time. That directory semaphore serialization has been
there since pretty much day one. I'm happy with how relatively small
the actual changes end up being, but still..

I'll push out after my test-build, can you please then double-check
that everything looks good?

Thanks,
                 Linus