Create a new mount workqueue and delayed_work to enable background
scanning and freeing of eofblocks inodes. The scanner kicks in once
speculative preallocation occurs and stops requeueing itself when
no eofblocks inodes exist.

The scan interval is based on the new
'speculative_prealloc_lifetime' tunable (default to 5m). The
background scanner performs unfiltered, best effort scans (which
skips inodes under lock contention or with a dirty cache mapping).
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

The XFS_IOC_FREE_EOFBLOCKS ioctl allows users to invoke an EOFBLOCKS
scan. The xfs_eofblocks structure is defined to support the command
parameters (scan mode).
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

xfs_inodes_free_eofblocks() implements scanning functionality for
EOFBLOCKS inodes. It uses the AG iterator to walk the tagged inodes
and free post-EOF blocks via the xfs_inode_free_eofblocks() execute
function. The scan can be invoked in best-effort mode or wait
(force) mode.

A best-effort scan (default) handles all inodes that do not have a
dirty cache and we successfully acquire the io lock via trylock. In
wait mode, we continue to cycle through an AG until all inodes are
handled.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Genericize xfs_inode_ag_walk() to support an optional radix tree tag
and args argument for the execute function. Create a new wrapper
called xfs_inode_ag_iterator_tag() that performs a tag based walk
of perag's and inodes.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Add the XFS_ICI_EOFBLOCKS_TAG inode tag to identify inodes with
speculatively preallocated blocks beyond EOF. An inode is tagged
when speculative preallocation occurs and untagged either via
truncate down or when post-EOF blocks are freed via release or
reclaim.

The tag management is intentionally not aggressive to prefer
simplicity over the complexity of handling all the corner cases
under which post-EOF blocks could be freed (i.e., forward
truncation, fallocate, write error conditions, etc.). This means
that a tagged inode may or may not have post-EOF blocks after a
period of time. The tag is eventually cleared when the inode is
released or reclaimed.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

The inode cache functions remaining in xfs_iget.c can be moved to xfs_icache.c
along with the other inode cache functions. This removes all functionality from
xfs_iget.c, so the file can simply be removed.

This move results in various functions now only having the scope of a single
file (e.g. xfs_inode_free()), so clean up all the definitions and exported
prototypes in xfs_icache.[ch] and xfs_inode.h appropriately.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

xfs_sync.c now only contains inode reclaim functions and inode cache
iteration functions. It is not related to sync operations anymore.
Rename to xfs_icache.c to reflect it's contents and prepare for
consolidation with the other inode cache file that exists
(xfs_iget.c).
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Both callers of xfs_quiesce_attr() are in xfs_super.c, and there's
nothing really sync-specific about this functionality so it doesn't
really matter where it lives. Move it to benext to it's callers, so
all the remount/sync_fs code is in the one place.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

With the syncd functions moved to the log and/or removed, the syncd
workqueue is the only remaining bit left. It is used by the log
covering/ail pushing work, as well as by the inode reclaim work.

Given how cheap workqueues are these days, give the log and inode
reclaim work their own work queues and kill the syncd work queue.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NBen Myers <bpm@sgi.com>

We don't do any data writeback from XFS any more - the VFS is
completely responsible for that, including for freeze. We can
replace the remaining caller with a VFS level function that
achieves the same thing, but without conflicting with current
writeback work.

This means we can remove the flush_work and xfs_flush_inodes() - the
VFS functionality completely replaces the internal flush queue for
doing this writeback work in a separate context to avoid stack
overruns.

This does have one complication - it cannot be called with page
locks held.  Hence move the flushing of delalloc space when ENOSPC
occurs back up into xfs_file_aio_buffered_write when we don't hold
any locks that will stall writeback.

Unfortunately, writeback_inodes_sb_if_idle() is not sufficient to
trigger delalloc conversion fast enough to prevent spurious ENOSPC
whent here are hundreds of writers, thousands of small files and GBs
of free RAM.  Hence we need to use sync_sb_inodes() to block callers
while we wait for writeback like the previous xfs_flush_inodes
implementation did.

That means we have to hold the s_umount lock here, but because this
call can nest inside i_mutex (the parent directory in the create
case, held by the VFS), we have to use down_read_trylock() to avoid
potential deadlocks. In practice, this trylock will succeed on
almost every attempt as unmount/remount type operations are
exceedingly rare.

Note: we always need to pass a count of zero to
generic_file_buffered_write() as the previously written byte count.
We only do this by accident before this patch by the virtue of ret
always being zero when there are no errors. Make this explicit
rather than needing to specifically zero ret in the ENOSPC retry
case.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Tested-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NBen Myers <bpm@sgi.com>

The only thing the periodic sync work does now is flush the AIL and
idle the log. These are really functions of the log code, so move
the work to xfs_log.c and rename it appropriately.

The only wart that this leaves behind is the xfssyncd_centisecs
sysctl, otherwise the xfssyncd is dead. Clean up any comments that
related to xfssyncd to reflect it's passing.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NBen Myers <bpm@sgi.com>

xfs_syncd_start and xfs_syncd_stop tie a bunch of unrelated
functionailty together that actually have different start and stop
requirements. Kill these functions and open code the start/stop
methods for each of the background functions.

Subsequent patches will move the start/stop functions around to the
correct places to avoid races and shutdown issues.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Timestamps on regular files are the last metadata that XFS does not update
transactionally.  Now that we use the delaylog mode exclusively and made
the log scode scale extremly well there is no need to bypass that code for
timestamp updates.  Logging all updates allows to drop a lot of code, and
will allow for further performance improvements later on.

Note that this patch drops optimized handling of fdatasync - it will be
added back in a separate commit.
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Since Linux 2.6.36 the writeback code has introduces various measures for
live lock prevention during sync().  Unfortunately some of these are
actively harmful for the XFS model, where the inode gets marked dirty for
metadata from the data I/O handler.

The older_than_this checks that are now more strictly enforced since

    writeback: avoid livelocking WB_SYNC_ALL writeback

by only calling into __writeback_inodes_sb and thus only sampling the
current cut off time once.  But on a slow enough devices the previous
asynchronous sync pass might not have fully completed yet, and thus XFS
might mark metadata dirty only after that sampling of the cut off time for
the blocking pass already happened.  I have not myself reproduced this
myself on a real system, but by introducing artificial delay into the
XFS I/O completion workqueues it can be reproduced easily.

Fix this by iterating over all XFS inodes in ->sync_fs and log all that
are dirty.  This might log inode that only got redirtied after the
previous pass, but given how cheap delayed logging of inodes is it
isn't a major concern for performance.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Tested-by: NMark Tinguely <tinguely@sgi.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Use the move from Linux 2.6 to Linux 3.x as an excuse to kill the
annoying subdirectories in the XFS source code.  Besides the large
amount of file rename the only changes are to the Makefile, a few
files including headers with the subdirectory prefix, and the binary
sysctl compat code that includes a header under fs/xfs/ from
kernel/.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Convert the inode reclaim shrinker to use the new per-sb shrinker
operations. This allows much bigger reclaim batches to be used, and
allows the XFS inode cache to be shrunk in proportion with the VFS
dentry and inode caches. This avoids the problem of the VFS caches
being shrunk significantly before the XFS inode cache is shrunk
resulting in imbalances in the caches during reclaim.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAlex Elder <aelder@sgi.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

All of the work xfssyncd does is background functionality. There is
no need for a thread per filesystem to do this work - it can al be
managed by a global workqueue now they manage concurrency
effectively.

Introduce a new gglobal xfssyncd workqueue, and convert the periodic
work to use this new functionality. To do this, use a delayed work
construct to schedule the next running of the periodic sync work
for the filesystem. When the sync work is complete, queue a new
delayed work for the next running of the sync work.

For laptop mode, we wait on completion for the sync works, so ensure
that the sync work queuing interface can flush and wait for work to
complete to enable the work queue infrastructure to replace the
current sequence number and wakeup that is used.

Because the sync work does non-trivial amounts of work, mark the
new work queue as CPU intensive.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAlex Elder <aelder@sgi.com>

With the reclaim code separated from the generic walking code, it is
simple to implement batched lookups for the generic walk code.
Separate out the inode validation from the execute operations and
modify the tree lookups to get a batch of inodes at a time.

Reclaim operations will be optimised separately.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAlex Elder <aelder@sgi.com>

The reclaim walk requires different locking and has a slightly
different walk algorithm, so separate it out so that it can be
optimised separately.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAlex Elder <aelder@sgi.com>

On the final put of a superblock the VFS already calls sync_filesystem
for us to write out all data and wait for it.  No need to start another
asynchronous writeback inside ->put_super.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Now the shrinker passes us a context, wire up a shrinker context per
filesystem. This allows us to remove the global mount list and the
locking problems that introduced. It also means that a shrinker call
does not need to traverse clean filesystems before finding a
filesystem with reclaimable inodes.  This significantly reduces
scanning overhead when lots of filesystems are present.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

On low memory boxes or those with highmem, kernel can OOM before the
background reclaims inodes via xfssyncd. Add a shrinker to run inode
reclaim so that it inode reclaim is expedited when memory is low.

This is more complex than it needs to be because the VM folk don't
want a context added to the shrinker infrastructure. Hence we need
to add a global list of XFS mount structures so the shrinker can
traverse them.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Just minor housekeeping, a lot more functions can be trivially made
static; others could if we reordered things a bit...
Signed-off-by: NEric Sandeen <sandeen@sandeen.net>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Make the inode tree reclaim walk exclusive to avoid races with
concurrent sync walkers and lookups. This is a version of a patch
posted by Christoph Hellwig that avoids all the code duplication.
Signed-off-by: NDave Chinner <david@fromorbit.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Currently the reclaim code for the case where we don't reclaim the
final reclaim is overly complicated.  We know that the inode is clean
but instead of just directly reclaiming the clean inode we go through
the whole process of marking the inode reclaimable just to directly
reclaim it from the calling context.  Besides being overly complicated
this introduces a race where iget could recycle an inode between
marked reclaimable and actually being reclaimed leading to panics.

This patch gets rid of the existing reclaim path, and replaces it with
a simple call to xfs_ireclaim if the inode was clean.  While we're at
it we also use the slightly more lax xfs_inode_clean check we'd use
later to determine if we need to flush the inode here.

Finally get rid of xfs_reclaim function and place the remaining small
bits of reclaim code directly into xfs_fs_destroy_inode.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reported-by: NPatrick Schreurs <patrick@news-service.com>
Reported-by: NTommy van Leeuwen <tommy@news-service.com>
Tested-by: NPatrick Schreurs <patrick@news-service.com>
Reviewed-by: NAlex Elder <aelder@sgi.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

A lot more functions could be made static, but they need
forward declarations; this does some easy ones, and also
found a few unused functions in the process.
Signed-off-by: NEric Sandeen <sandeen@sandeen.net>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NFelix Blyakher <felixb@sgi.com>

The locking in xfs_iget_cache_hit currently has numerous problems:

 - we clear the reclaim tag without i_flags_lock which protects
   modifications to it
 - we call inode_init_always which can sleep with pag_ici_lock
   held (this is oss.sgi.com BZ #819)
 - we acquire and drop i_flags_lock a lot and thus provide no
   consistency between the various flags we set/clear under it

This patch fixes all that with a major revamp of the locking in
the function.  The new version acquires i_flags_lock early and
only drops it once we need to call into inode_init_always or before
calling xfs_ilock.

This patch fixes a bug seen in the wild where we race modifying the
reclaim tag.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NFelix Blyakher <felixb@sgi.com>
Reviewed-by: NEric Sandeen <sandeen@sandeen.net>
Signed-off-by: NFelix Blyakher <felixb@sgi.com>

The locking in xfs_iget_cache_hit currently has numerous problems:

 - we clear the reclaim tag without i_flags_lock which protects
   modifications to it
 - we call inode_init_always which can sleep with pag_ici_lock
   held (this is oss.sgi.com BZ #819)
 - we acquire and drop i_flags_lock a lot and thus provide no
   consistency between the various flags we set/clear under it

This patch fixes all that with a major revamp of the locking in
the function.  The new version acquires i_flags_lock early and
only drops it once we need to call into inode_init_always or before
calling xfs_ilock.

This patch fixes a bug seen in the wild where we race modifying the
reclaim tag.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NFelix Blyakher <felixb@sgi.com>
Reviewed-by: NEric Sandeen <sandeen@sandeen.net>
Signed-off-by: NFelix Blyakher <felixb@sgi.com>

A lot more functions could be made static, but they need
forward declarations; this does some easy ones, and also
found a few unused functions in the process.
Signed-off-by: NEric Sandeen <sandeen@sandeen.net>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NFelix Blyakher <felixb@sgi.com>

SYNC_BDFLUSH is a leftover from IRIX and rather misnamed for todays
code.  Make xfs_sync_fsdata and xfs_dq_sync use the SYNC_TRYLOCK flag
for not blocking on logs just as the inode sync code already does.

For xfs_sync_fsdata it's a trivial 1:1 replacement, but for xfs_qm_sync
I use the opportunity to decouple the non-blocking lock case from the
different flushing modes, similar to the inode sync code.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NEric Sandeen <sandeen@sandeen.net>

We want to wait for all I/O to finish when we do data integrity syncs.  So
there is no reason to keep SYNC_WAIT separate from SYNC_IOWAIT.  This
causes a little change in behaviour for the ENOSPC flushing code which now
does a second submission and wait of buffered I/O, but that should finish
ASAP as we already did an asynchronous writeout earlier.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJosef 'Jeff' Sipek <jeffpc@josefsipek.net>
Reviewed-by: NEric Sandeen <sandeen@sandeen.net>

xfs_sync_inodes is used to write back either file data or inode metadata.
In general we always do these separately, except for one fishy case in
xfs_fs_put_super that does both.  So separate xfs_sync_inodes into
separate xfs_sync_data and xfs_sync_attr functions.  In xfs_fs_put_super
we first call the data sync and then the attr sync as that was the previous
order.  The moved log force in that path doesn't make a difference because
we will force the log again as part of the real unmount process.

The filesystem readonly checks are not performed by the new function but
instead moved into the callers, given that most callers alredy have it
further up in the stack.  Also add debug checks that we do not pass in
incorrect flags in the new xfs_sync_data and xfs_sync_attr function and
fix the one place that did pass in a wrong flag.

Also remove a comment mentioning xfs_sync_inodes that has been incorrect
for a while because we always take either the iolock or ilock in the
sync path these days.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NEric Sandeen <sandeen@sandeen.net>

Use xfs_inode_ag_iterator instead of opencoding the inode walk in the
quota code.  Mark xfs_inode_ag_iterator and xfs_sync_inode_valid non-static
to allow using them from the quota code.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJosef 'Jeff' Sipek <jeffpc@josefsipek.net>
Reviewed-by: NEric Sandeen <sandeen@sandeen.net>

The noblock parameter of xfs_reclaim_inodes is only ever set to zero. Remove
it and all the conditional code that is never executed.
Signed-off-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NEric Sandeen <sandeen@sandeen.net>

xfs_flush_inodes() currently uses a magic timeout to wait for
some inodes to be flushed before returning. This isn't
really reliable but used to be the best that could be done
due to deadlock potential of waiting for the entire flush.

Now the inode flush is safe to execute while we hold page
and inode locks, we can wait for all the inodes to flush
synchronously. Convert the wait mechanism to a completion
to do this efficiently. This should remove all remaining
spurious ENOSPC errors from the delayed allocation reservation
path.

This is extracted almost line for line from a larger patch
from Mikulas Patocka.
Signed-off-by: NMikulas Patocka <mpatocka@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

When we are writing to a single file and hit ENOSPC, we trigger a background
flush of the inode and try again.  Because we hold page locks and the iolock,
the flush won't proceed until after we release these locks. This occurs once
we've given up and ENOSPC has been reported. Hence if this one is the only
dirty inode in the system, we'll get an ENOSPC prematurely.

To fix this, remove the async flush from the allocation routines and move
it to the top of the write path where we can do a synchronous flush
and retry the write again. Only retry once as a second ENOSPC indicates
that we really are ENOSPC.

This avoids a page cache deadlock when trying to do this flush synchronously
in the allocation layer that was identified by Mikulas Patocka.
Signed-off-by: NDave Chinner <david@fromorbit.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Currently xfs_device_flush calls sync_blockdev() which is
a no-op for XFS as all it's metadata is held in a different
address to the one sync_blockdev() works on.

Call xfs_sync_inodes() instead to flush all the delayed
allocation blocks out. To do this as efficiently as possible,
do it via two passes - one to do an async flush of all the
dirty blocks and a second to wait for all the IO to complete.
This requires some modification to the xfs-sync_inodes_ag()
flush code to do efficiently.
Signed-off-by: NDave Chinner <david@fromorbit.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Currently we call from the nicely abstracted linux quotaops into a ugly
multiplexer just to split the calls out at the same boundary again.
Rewrite the quota ops handling to remove that obfucation.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>

Prepare for removing the deleted inode list by marking inodes for reclaim
in the inode radix trees so that we can use the radix trees to find
reclaimable inodes.

SGI-PV: 988142

SGI-Modid: xfs-linux-melb:xfs-kern:32331a
Signed-off-by: NDavid Chinner <david@fromorbit.com>
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>
Signed-off-by: NChristoph Hellwig <hch@infradead.org>