xfs: drain deferred work items when scrubbing

The design doc for XFS online fsck contains a long discussion of the
eventual consistency models in use for XFS metadata.  In that chapter,
we note that it is possible for scrub to collide with a chain of
deferred space metadata updates, and proposes a lightweight solution:
The use of a pending-intents counter so that scrub can wait for the
system to drain all chains.

This patchset implements that scrub drain.  The first patch implements
the basic mechanism, and the subsequent patches reduce the runtime
overhead by converting the implementation to use sloppy counters and
introducing jump labels to avoid walking into scrub hooks when it isn't
running.  This last paradigm repeats elsewhere in this megaseries.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
-----BEGIN PGP SIGNATURE-----

iQIzBAABCgAdFiEEUzaAxoMeQq6m2jMV+H93GTRKtOsFAmNJ0hgACgkQ+H93GTRK
tOv2FA/9FUDa+ArPkGgbeEmSMAeyaOmTErmaXtjTPIR8C/4hCBkqzZSMCUglWdL9
BFae+jsXEbFhXIWs2atfyjce6PoXbtmYeAv9LvcjJoD62RF06glbPzf6cDf59Y2/
2ZRSVcksXTFvzjxbUbAonmiVqdHG4Nirh710EzrKljVvUQuTlIOtKubNoCsZd2lX
Fp8JIANDzx38ExWXDpnuBxIjq+xACNxghDS5qs2QvzYaY0tmKAkJIu4eW3AF/U/1
reJO4URRXSxDOCKX3TLjpIsLzuUjvdMEjrJ7/HDrpruAIwcw+Ay0brICD3YrpdXJ
vAkafPCxK1Anh/mshQOJU4eFmwDR3GMAkazLZmti3tO14a37XK2gClzzM6J80sFQ
Q8dj8DXoku42SA1eIyJnEitmdcIP0kPMGh+c7coGaF2MEXBc/NA+jzZrwL05f/x0
CiZDNC2TCmRGE2U1ckmJAV7HEFwn4tch4VcVZ+CXPXOMS/k57+3uqXaWFM/BIya/
oUsBBwRpFSBHMmPK2MV0YG8LQWXxY+9sabF/dTvY5BwESKd3L0f4VTHOD4D/guEM
kPsObH0qoDD/XGlzrnUHz4uNCOw4dDlE10/Ghk46HJpdA/4FcTvkHfRaAI+XvTk7
jAdWGrpkYQq7ExSwo0bVgOB8ZWIyzx0RjpUXoutD2/gnoAxAelA=
=2QuQ
-----END PGP SIGNATURE-----