xfs: deferred inode inactivation

This patch series implements deferred inode inactivation.  Inactivation
is what happens when an open file loses its last incore reference: if
the file has speculative preallocations, they must be freed, and if the
file is unlinked, all forks must be truncated, and the inode marked
freed in the inode chunk and the inode btrees.

Currently, all of this activity is performed in frontend threads when
the last in-memory reference is lost and/or the vfs decides to drop the
inode.  Three complaints stem from this behavior: first, that the time
to unlink (in the worst case) depends on both the complexity of the
directory as well as the the number of extents in that file; second,
that deleting a directory tree is inefficient and seeky because we free
the inodes in readdir order, not disk order; and third, the upcoming
online repair feature needs to be able to xfs_irele while scanning a
filesystem in transaction context.  It cannot perform inode inactivation
in this context because xfs does not support nested transactions.

The implementation will be familiar to those who have studied how XFS
scans for reclaimable in-core inodes -- we create a couple more inode
state flags to mark an inode as needing inactivation and being in the
middle of inactivation.  When inodes need inactivation, we set
NEED_INACTIVE in iflags and add it to a percpu work list.  Eventually, a
bounded percpu workqueue item will be scheduled to perform all the
on-disk metadata updates.  Once the inode has been inactivated, it is
left in the reclaim state and the background reclaim worker (or direct
reclaim) will get to it eventually.

Doing the inactivations from kernel threads solves the first problem by
constraining the amount of work done by the unlink() call to removing
the directory entry.  It solves the third problem by moving inactivation
to a separate process.  Performing the inactivations in batches
decreases the amount of time it takes to let go of an inode cluster if
we're deleting entire directory trees.

There are three big warts I can think of in this series: first, because
the actual freeing of nlink==0 inodes is now done in the background,
this means that the system will be busy making metadata updates for some
time after the unlink() call returns.  This temporarily reduces
available iops.  Second, in order to retain the behavior that deleting
100TB of unshared data should result in a free space gain of 100TB, the
statvfs and quota reporting ioctls wait for inactivation to finish,
which increases the long tail latency of those calls.  This behavior is,
unfortunately, key to not introducing regressions in fstests.  The third
problem is that the deferrals keep memory usage higher for longer.  The
final patch in the series (clumsily) addresses this by forcing the
inodegc workers to run when memory shrinkers get called and by
throttling the frontend xfs_inodegc_queue callers to wait for the
worker.

v1-v2: NYE patchbombs
v3: rebase against 5.12-rc2 for submission.
v4: combine the can/has eofblocks predicates, clean up incore inode tree
    walks, fix inobt deadlock
v5: actually freeze the inode gc threads when we freeze the filesystem,
    consolidate the code that deals with inode tagging, and use
    foreground inactivation during quotaoff to avoid cycling dquots
v6: rebase to 5.13-rc4, fix quotaoff not to require foreground inactivation,
    refactor to use inode walk goals, use atomic bitflags to control the
    scheduling of gc workers
v7: simplify the inodegc worker, which simplifies how flushes work, break
    up the patch into smaller pieces, flush inactive inodes on syncfs to
    simplify freeze/ro-remount handling, separate inode selection filtering
    in iget, refactor inode recycling further, change gc delay to 100ms,
    decrease the gc delay when space or quota are low, move most of the
    destroy_inode logic to mark_reclaimable, get rid of the fallocate flush
    scan thing, get rid of polled flush mode
v8: rebase against 5.14-rc2, hook the memory shrinkers so that we requeue
    inactivation immediately when memory starts to get tight and force
    callers queueing inodes for inactivation to wait for the inactivation
    workers to run (i.e. throttling the frontend) to reduce memory storms,
    add hch's quotaoff removal series as a dependency to shut down arguments
    about quota walks
v9: replace the entire mechanism with percpu lists and workers, clean out
    a ton of ratty code that nobody liked anyway :P
v10: remove an unnecessary assert, and fix some naming problems in tracepoints
-----BEGIN PGP SIGNATURE-----
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=18lt
-----END PGP SIGNATURE-----