Introduce a new in-core fork for storing copy-on-write delalloc
reservations and allocated extents that are in the process of being
written out.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Only non-rt files can be reflinked, so check that when we load an
inode.  Also, don't leak the attr fork if there's a failure.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Report the reflink feature in the XFS geometry so that xfs_info and
friends know the filesystem has this feature.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Return the range of file blocks that bunmapi didn't free.  This hint
is used by CoW and reflink to figure out what part of an extent
actually got freed so that it can set up the appropriate atomic
remapping of just the freed range.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Implement deferred versions of the inode block map/unmap functions.
These will be used in subsequent patches to make reflink operations
atomic.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Pass BMAPI_ flags from bunmapi into bmap_del_extent and extend
BMAPI_REMAP (which means "don't touch the allocator or the quota
accounting") to apply to bunmapi as well.  This will be used to
implement the unmap operation, which will be used by swapext.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Teach the bmap routine to know how to map a range of file blocks to a
specific range of physical blocks, instead of simply allocating fresh
blocks.  This enables reflink to map a file to blocks that are already
in use.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Provide a mechanism for higher levels to create BUI/BUD items, submit
them to the log, and a stub function to deal with recovered BUI items.
These parts will be connected to the rmapbt in a later patch.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Create bmbt update intent/done log items to record redo information in
the log.  Because we roll transactions multiple times for reflink
operations, we also have to track the status of the metadata updates
that will be recorded in the post-roll transactions in case we crash
before committing the final transaction.  This mechanism enables log
recovery to finish what was already started.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

These functions will be used by the other reflink functions to find
the maximum length of a range of shared blocks.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.coM>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Reduce the max AG usable space size so that we always have space for
the refcount btree root.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

When we're unmapping blocks from a reflinked file, decrease the
refcount of the affected blocks and free the extents that are no
longer in use.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Plumb in the upper level interface to schedule and finish deferred
refcount operations via the deferred ops mechanism.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Provide functions to adjust the reference counts for an extent of
physical blocks stored in the refcount btree.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Provide a mechanism for higher levels to create CUI/CUD items, submit
them to the log, and a stub function to deal with recovered CUI items.
These parts will be connected to the refcountbt in a later patch.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Create refcount update intent/done log items to record redo
information in the log.  Because we need to roll transactions between
updating the bmbt mapping and updating the reverse mapping, we also
have to track the status of the metadata updates that will be recorded
in the post-roll transactions, just in case we crash before committing
the final transaction.  This mechanism enables log recovery to finish
what was already started.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Implement the generic btree operations required to manipulate refcount
btree blocks.  The implementation is similar to the bmapbt, though it
will only allocate and free blocks from the AG.

Since the refcount root and level fields are separate from the
existing roots and levels array, they need a separate logging flag.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
[hch: fix logging of AGF refcount btree fields]
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Every time we allocate or free a data extent, we might need to split
the refcount btree.  Reserve some blocks in the transaction to handle
this possibility.  Even though the deferred refcount code can roll a
transaction to avoid overloading the transaction, we can still exceed
the reservation.

Certain pathological workloads (1k blocks, no cowextsize hint, random
directio writes), cause a perfect storm wherein a refcount adjustment
of a large range of blocks causes full tree splits in two separate
extents in two separate refcount tree blocks; allocating new refcount
tree blocks causes rmap btree splits; and all the allocation activity
causes the freespace btrees to split, blowing the reservation.

(Reproduced by generic/167 over NFS atop XFS)
Signed-off-by: NChristoph Hellwig <hch@lst.de>
[darrick.wong@oracle.com: add commit message]
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Start constructing the refcount btree implementation by establishing
the on-disk format and everything needed to read, write, and
manipulate the refcount btree blocks.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Since XFS reserves a small amount of space in each AG as the minimum
free space needed for an operation, save some more space in case we
touch the refcount btree.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Add new per-AG refcount btree definitions to the per-AG structures.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

When adding a new remote attribute, we write the attribute to the
new extent before the allocation transaction is committed. This
means we cannot reuse busy extents as that violates crash
consistency semantics. Hence we currently treat remote attribute
extent allocation like userdata because it has the same overwrite
ordering constraints as userdata.

Unfortunately, this also allows the allocator to incorrectly apply
extent size hints to the remote attribute extent allocation. This
results in interesting failures, such as transaction block
reservation overruns and in-memory inode attribute fork corruption.

To fix this, we need to separate the busy extent reuse configuration
from the userdata configuration. This changes the definition of
XFS_BMAPI_METADATA slightly - it now means that allocation is
metadata and reuse of busy extents is acceptible due to the metadata
ordering semantics of the journal. If this flag is not set, it
means the allocation is that has unordered data writeback, and hence
busy extent reuse is not allowed. It no longer implies the
allocation is for user data, just that the data write will not be
strictly ordered. This matches the semantics for both user data
and remote attribute block allocation.

As such, This patch changes the "userdata" field to a "datatype"
field, and adds a "no busy reuse" flag to the field.
When we detect an unordered data extent allocation, we immediately set
the no reuse flag. We then set the "user data" flags based on the
inode fork we are allocating the extent to. Hence we only set
userdata flags on data fork allocations now and consider attribute
fork remote extents to be an unordered metadata extent.

The result is that remote attribute extents now have the expected
allocation semantics, and the data fork allocation behaviour is
completely unchanged.

It should be noted that there may be other ways to fix this (e.g.
use ordered metadata buffers for the remote attribute extent data
write) but they are more invasive and difficult to validate both
from a design and implementation POV. Hence this patch takes the
simple, obvious route to fixing the problem...
Reported-and-tested-by: NRoss Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Currently xfs_iomap_write_delay does up to lookups in the inode
extent tree, which is rather costly especially with the new iomap
based write path and small write sizes.

But it turns out that the low-level xfs_bmap_search_extents gives us
all the information we need in the regular delalloc buffered write
path:

 - it will return us an extent covering the block we are looking up
   if it exists.  In that case we can simply return that extent to
   the caller and are done
 - it will tell us if we are beyoned the last current allocated
   block with an eof return parameter.  In that case we can create a
   delalloc reservation and use the also returned information about
   the last extent in the file as the hint to size our delalloc
   reservation.
 - it can tell us that we are writing into a hole, but that there is
   an extent beyoned this hole.  In this case we can create a
   delalloc reservation that covers the requested size (possible
   capped to the next existing allocation).

All that can be done in one single routine instead of bouncing up
and down a few layers.  This reduced the CPU overhead of the block
mapping routines and also simplified the code a lot.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

One unfortunate quirk of the reference count and reverse mapping
btrees -- they can expand in size when blocks are written to *other*
allocation groups if, say, one large extent becomes a lot of tiny
extents.  Since we don't want to start throwing errors in the middle
of CoWing, we need to reserve some blocks to handle future expansion.
The transaction block reservation counters aren't sufficient here
because we have to have a reserve of blocks in every AG, not just
somewhere in the filesystem.

Therefore, create two per-AG block reservation pools.  One feeds the
AGFL so that rmapbt expansion always succeeds, and the other feeds all
other metadata so that refcountbt expansion never fails.

Use the count of how many reserved blocks we need to have on hand to
create a virtual reservation in the AG.  Through selective clamping of
the maximum length of allocation requests and of the length of the
longest free extent, we can make it look like there's less free space
in the AG unless the reservation owner is asking for blocks.

In other words, play some accounting tricks in-core to make sure that
we always have blocks available.  On the plus side, there's nothing to
clean up if we crash, which is contrast to the strategy that the rough
draft used (actually removing extents from the freespace btrees).
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

When xfs_defer_finish calls ->finish_item, it's possible that
(refcount) won't be able to finish all the work in a single
transaction.  When this happens, the ->finish_item handler should
shorten the log done item's list count, update the work item to
reflect where work should continue, and return -EAGAIN so that
defer_finish knows to retain the pending item on the pending list,
roll the transaction, and restart processing where we left off.

Plumb in the code and document how this mechanism is supposed to work.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Provide a helper method to count the number of blocks in a short form
btree.  The refcount and rmap btrees need to know the number of blocks
already in use to set up their per-AG block reservations during mount.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Create a helper to generate AG btree height calculator functions.
This will be used (much) later when we get to the refcount btree.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Remove the xfs_btree_bigkey mess and simply make xfs_btree_key big enough
to hold both keys in-core.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Use variable length array declarations for RUI log items,
and replace the open coded sizeof formulae with a single function.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Christoph reports slab corruption when a deferred refcount update
aborts during _defer_finish().  The cause of this was broken log item
state tracking in xfs_defer_pending -- upon an abort,
_defer_trans_abort() will call abort_intent on all intent items,
including the ones that have already had a done item attached.

This is incorrect because each intent item has 2 refcount: the first
is released when the intent item is committed to the log; and the
second is released when the _done_ item is committed to the log, or
by the intent creator if there is no done item.  In other words, once
we log the done item, responsibility for releasing the intent item's
second refcount is transferred to the done item and /must not/ be
performed by anything else.

The dfp_committed flag should have been tracking whether or not we had
a done item so that _defer_trans_abort could decide if it needs to
abort the intent item, but due to a thinko this was not the case.  Rip
it out and track the done item directly so that we do the right thing
w.r.t. intent item freeing.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reported-by: NChristoph Hellwig <hch@infradead.org>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

From inspection, the superblock sb_inprogress check is done in the
verifier and triggered only for the primary superblock via a
"bp->b_bn == XFS_SB_DADDR" check.

Unfortunately, the primary superblock is an uncached buffer, and
hence it is configured by xfs_buf_read_uncached() with:

	bp->b_bn = XFS_BUF_DADDR_NULL;  /* always null for uncached buffers */

And so this check never triggers. Fix it.

cc: <stable@vger.kernel.org>
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

If the initial LOOKUP_LE in the simple query range fails to find
anything, we should attempt to increment the btree cursor to see
if there actually /are/ records for what we're trying to find.
Without this patch, a bnobt range query of (0, $agsize) returns
no results because the leftmost record never has a startblock
of zero.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

We only need the record's high key for the first record that we look
at; for all records, we /definitely/ need the regular record key.
Therefore, fix how the simple range query function gets its keys.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

When we're logging the last non-spare field in the AGF, we don't
need to log the spare fields, so plumb in a new AGF logging flag
to help us avoid that.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

If the caller passes in a cursor to a zero-height btree (which is
impossible), we never set block to anything but NULL, which causes the
later dereference of it to crash.  Instead, just return -EFSCORRUPTED.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

When we're really tight on space, xfs_alloc_ag_vextent_small() can
allocate a block from the AGFL and give it to the caller.  Since the
caller is never the AGFL-fixing method, we must remove the OWN_AG
reverse mapping because it will clash with whatever rmap the caller
wants to set up.  This bug was discovered by running generic/299
repeatedly.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Track the number of blocks used for the rmapbt in the AGF.  When we
get to the AG reservation code we need this counter to quickly
make our reservation during mount.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Rename the deferred bmap-free to extent_free and make them only
trigger when we're really running deferred ops.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Nothing ever uses the extent array in the rmap update done redo
item, so remove it before it is fixed in the on-disk log format.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

We only need the temporary cursor in _btree_lshift if we're shifting
in an overlapped btree.  Therefore, factor that into a single block
of code so we avoid unnecessary cursor duplication.

Also fix use of the wrong cursor when checking for corruption in
xfs_btree_rshift().
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>