Re: [PATCHv3 0/4] hfsplus: stop using write_supers and s_dirt

[Andrew Morton <akpm@linux-foundation.org>]

On Thu, 12 Jul 2012 17:26:27 +0300
Artem Bityutskiy <dedekind1@gmail.com> wrote:

> Hi Andrew, here is v3 of the patches, could you please pick them instead of the
> ones you have in your tree?

I really do dislike dropping patches and replacing them.  For a whole
bunch of reasons.  One of which is that a wholesale replacement
requires a full re-review.  Another is that wholesale replacement makes
it hard to see what was changed.

> The differences to v2 are:
> 
> 1. checkpatch.pl warnings fixed
> 2. removed 'cancel_delayed_work()' from 'hfsplus_sync_fs()' because it is not
>    enough to just cancel the job, we need to set the work_queued flag to zero
>    as well. I could do this, but it is simpler to just remove this tiny
>    optimization - we do not gain much with it. I caught this bug while doing
>    some more testing.

And lo, when I look at "what was changed", I see that this patchset
actually does *not* remove the cancel_delayed_work() call.  What's up
with that?

This latest patchset is identical to the code which is presently in -mm.