On debug kernels, we call xfs_dir3_leaf_check_int() multiple times
on every directory modification. The robust hash ordering checks it
does on every entry in the leaf on every call results in a massive
CPU overhead which slows down debug kernels by a large amount.

We use xfs_dir3_leaf_check_int() for the verifiers as well, so we
can't just gut the function to reduce overhead. What we can do,
however, is reduce the work it does when it is called from the
debug interfaces, just leaving the high level checks in place and
leaving the robust validation to the verifiers. This means the debug
checks will catch gross errors, but subtle bugs might not be caught
until a verifier is run.

It is easy enough to restore the existing debug behaviour if the
developer needs it (just change a call parameter in the debug code),
but overwise the overhead makes testing large directory block sizes
on debug kernels very slow.

Profile at an unlink rate of ~80k file/s on a 64k block size
filesystem before the patch:

  40.30%  [kernel]  [k] xfs_dir3_leaf_check_int
  10.98%  [kernel]  [k] __xfs_dir3_data_check
   8.10%  [kernel]  [k] xfs_verify_dir_ino
   4.42%  [kernel]  [k] memcpy
   2.22%  [kernel]  [k] xfs_dir2_data_get_ftype
   1.52%  [kernel]  [k] do_raw_spin_lock

Profile after, at an unlink rate of ~125k files/s (+50% improvement)
has largely dropped the leaf verification debug overhead out of the
profile.

  16.53%  [kernel]  [k] __xfs_dir3_data_check
  12.53%  [kernel]  [k] xfs_verify_dir_ino
   7.97%  [kernel]  [k] memcpy
   3.36%  [kernel]  [k] xfs_dir2_data_get_ftype
   2.86%  [kernel]  [k] __pv_queued_spin_lock_slowpath

Create shows a similar change in profile and a +25% improvement in
performance.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

Every call to xfs_da_state_alloc() also requires setting up state->args
and state->mp

Change xfs_da_state_alloc() to receive an xfs_da_args_t as argument and
return a xfs_da_state_t with both args and mp already set.
Signed-off-by: NCarlos Maiolino <cmaiolino@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
[darrick: reduce struct typedef usage]
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Check the owner field of dir3 free block headers and reject the metadata
if there's something wrong with it.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NAllison Collins <allison.henderson@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

If we decide that a directory free block is corrupt, we must take care
not to leak a buffer pointer to the caller.  After xfs_trans_brelse
returns, the buffer can be freed or reused, which means that we have to
set *bpp back to NULL.

Callers are supposed to notice the nonzero return value and not use the
buffer pointer, but we should code more defensively, even if all current
callers handle this situation correctly.

Fixes: de14c5f5 ("xfs: verify free block header fields")
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

xfs_verifier_error is supposed to be called on a corrupt metadata buffer
from within a buffer verifier function, whereas xfs_buf_mark_corrupt
is the function to be called when a piece of code has read a buffer and
catches something that a read verifier cannot.  The first function sets
b_error anticipating that the low level buffer handling code will see
the nonzero b_error and clear XBF_DONE on the buffer, whereas the second
function does not.

Since xfs_dir3_free_header_check examines fields in the dir free block
header that require more context than can be provided to read verifiers,
we must call xfs_buf_mark_corrupt when it finds a problem.

Switching the calls has a secondary effect that we no longer corrupt the
buffer state by setting b_error and leaving XBF_DONE set.  When /that/
happens, we'll trip over various state assertions (most commonly the
b_error check in xfs_buf_reverify) on a subsequent attempt to read the
buffer.

Fixes: bc1a09b8 ("xfs: refactor verifier callers to print address of failing check")
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Add a helper function to get rid of buffers that we have decided are
corrupt after the verifiers have run.  This function is intended to
handle metadata checks that can't happen in the verifiers, such as
inter-block relationship checking.  Note that we now mark the buffer
stale so that it will not end up on any LRU and will be purged on
release.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Use the xfs_da_get_buf_daddr function directly for the two callers
that pass a mapped disk address, and then remove the mappedbno argument.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Move the code for reading an already mapped block into
xfs_da3_node_read_mapped, which is the only caller ever passing a block
number in the mappedbno argument and replace the mappedbno argument with
the simple xfs_dabuf_get flags.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

This argument is always hard coded to -1, so remove it.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Instead of causing a relatively expensive indirect call for each
hashing and comparism of a file name in a directory just use an
inline function and a simple branch on the ASCII CI bit.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
[darrick: fix unused variable warning]
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Convert the last of the open coded corruption check and report idioms to
use the XFS_IS_CORRUPT macro.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

There is no real need for xfs_dir2_data_freescan wrapper, so rename
xfs_dir2_data_freescan_int to xfs_dir2_data_freescan and let the
callers dereference the mount pointer from the inode.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Replace the ->data_get_ftype and ->data_put_ftype dir ops methods with
directly called xfs_dir2_data_get_ftype and xfs_dir2_data_put_ftype
helpers that takes care of the differences between the directory format
with and without the file type field.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Replace the ->data_bestfree_p dir ops method with a directly called
xfs_dir2_data_bestfree_p helper that takes care of the differences
between the v4 and v5 on-disk format.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Move the data block fixed offsets towards our structure for dir/attr
geometry parameters.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Replace the ->data_entry_tag_p dir ops method with a directly called
xfs_dir2_data_entry_tag_p helper that takes care of the differences
between the directory format with and without the file type field.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Replace the ->data_entsize dir ops method with a directly called
xfs_dir2_data_entsize helper that takes care of the differences between
the directory format with and without the file type field.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Now that the max bests value is in struct xfs_da_geometry both instances
of ->db_to_fdb and ->db_to_fdindex are identical.  Replace them with
local xfs_dir2_db_to_fdb and xfs_dir2_db_to_fdindex functions in
xfs_dir2_node.c.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Move the max free bests count towards our structure for dir/attr
geometry parameters.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Move the free header size towards our structure for dir/attr geometry
parameters.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

All but two callers of the ->free_bests_p dir operation already have a
struct xfs_dir3_icfree_hdr from a previous call to
xfs_dir2_free_hdr_from_disk at hand.  Add a pointer to the bests to
struct xfs_dir3_icfree_hdr to clean up this pattern.  To optimize this
pattern, pass the struct xfs_dir3_icfree_hdr to xfs_dir2_free_log_bests
instead of recalculating the pointer there.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Return the xfs_dir3_icfree_hdr used by the helpers called from
xfs_dir2_node_addname_int to the main function to prepare for the
next round of changes where we'll use the ichdr in xfs_dir3_icfree_hdr
to avoid extra operations to find the bests pointers.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Replace the ->free_hdr_to_disk dir ops method with a directly called
xfs_dir2_free_hdr_to_disk helper that takes care of the differences
between the v4 and v5 on-disk format.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Replace the ->free_hdr_from_disk dir ops method with a directly called
xfs_dir_free_hdr_from_disk helper that takes care of the differences
between the v4 and v5 on-disk format.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Move the max leaf entries count towards our structure for dir/attr
geometry parameters.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Move the leaf header size towards our structure for dir/attr geometry
parameters.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

All callers of the ->node_tree_p dir operation already have a struct
xfs_dir3_icleaf_hdr from a previous call to xfs_da_leaf_hdr_from_disk at
hand, or just need slight changes to the calling conventions to do so.
Add a pointer to the entries to struct xfs_dir3_icleaf_hdr to clean up
this pattern.  To make this possible the xfs_dir3_leaf_log_ents function
grow a new argument to pass the xfs_dir3_icleaf_hdr that call callers
already have, and xfs_dir2_leaf_lookup_int returns the
xfs_dir3_icleaf_hdr to the callers so that they can later use it.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Replace the ->leaf_hdr_to_disk dir ops method with a directly called
xfs_dir_leaf_hdr_to_disk helper that takes care of the differences
between the v4 and v5 on-disk format.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Replace the ->leaf_hdr_from_disk dir ops method with a directly called
xfs_dir2_leaf_hdr_from_disk helper that takes care of the differences
between the v4 and v5 on-disk format.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Make sure we log something to dmesg whenever we return -EFSCORRUPTED up
the call stack.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NCarlos Maiolino <cmaiolino@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

When a directory is growing rapidly, new blocks tend to get added at
the end of the directory. These end up at the end of the freespace
index, and when the directory gets large finding these new
freespaces gets expensive. The code does a linear search across the
frespace index from the first block in the directory to the last,
hence meaning the newly added space is the last index searched.

Instead, do a reverse order index search, starting from the last
block and index in the freespace index. This makes most lookups for
free space on rapidly growing directories O(1) instead of O(N), but
should not have any impact on random insert workloads because the
average search length is the same regardless of which end of the
array we start at.

The result is a major improvement in large directory grow rates:

		create time(sec) / rate (files/s)
 File count     vanilla             Prev commit		Patched
  10k	      0.41 / 24.3k	   0.42 / 23.8k       0.41 / 24.3k
  20k	      0.74 / 27.0k	   0.76 / 26.3k       0.75 / 26.7k
 100k	      3.81 / 26.4k	   3.47 / 28.8k       3.27 / 30.6k
 200k	      8.58 / 23.3k	   7.19 / 27.8k       6.71 / 29.8k
   1M	     85.69 / 11.7k	  48.53 / 20.6k      37.67 / 26.5k
   2M	    280.31 /  7.1k	 130.14 / 15.3k      79.55 / 25.2k
  10M	   3913.26 /  2.5k                          552.89 / 18.1k
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

When running a "create millions inodes in a directory" test
recently, I noticed we were spending a huge amount of time
converting freespace block headers from disk format to in-memory
format:

 31.47%  [kernel]  [k] xfs_dir2_node_addname
 17.86%  [kernel]  [k] xfs_dir3_free_hdr_from_disk
  3.55%  [kernel]  [k] xfs_dir3_free_bests_p

We shouldn't be hitting the best free block scanning code so hard
when doing sequential directory creates, and it turns out there's
a highly suboptimal loop searching the the best free array in
the freespace block - it decodes the block header before checking
each entry inside a loop, instead of decoding the header once before
running the entry search loop.

This makes a massive difference to create rates. Profile now looks
like this:

  13.15%  [kernel]  [k] xfs_dir2_node_addname
   3.52%  [kernel]  [k] xfs_dir3_leaf_check_int
   3.11%  [kernel]  [k] xfs_log_commit_cil

And the wall time/average file create rate differences are
just as stark:

		create time(sec) / rate (files/s)
File count	     vanilla		    patched
  10k		   0.41 / 24.3k		   0.42 / 23.8k
  20k		   0.74	/ 27.0k		   0.76 / 26.3k
 100k		   3.81	/ 26.4k		   3.47 / 28.8k
 200k		   8.58	/ 23.3k		   7.19 / 27.8k
   1M		  85.69	/ 11.7k		  48.53 / 20.6k
   2M		 280.31	/  7.1k		 130.14 / 15.3k

The larger the directory, the bigger the performance improvement.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Simplify the logic in xfs_dir2_node_addname_int() by factoring out
the free block index lookup code that finds a block with enough free
space for the entry to be added. The code that is moved gets a major
cleanup at the same time, but there is no algorithm change here.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Factor out the code that adds a data block to a directory from
xfs_dir2_node_addname_int(). This makes the code flow cleaner and
more obvious and provides clear isolation of upcoming optimsations.
Signed-off-By: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

This gets rid of the need for a forward  declaration of the static
function xfs_dir2_addname_int() and readies the code for factoring
of xfs_dir2_addname_int().
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Continue our game of replacing ASSERTs for corrupt ondisk metadata with
EFSCORRUPTED returns.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NBill O'Donnell <billodo@redhat.com>

There are many, many xfs header files which are included but
unneeded (or included twice) in the xfs code, so remove them.

nb: xfs_linux.h includes about 9 headers for everyone, so those
explicit includes get removed by this.  I'm not sure what the
preference is, but if we wanted explicit includes everywhere,
a followup patch could remove those xfs_*.h includes from
xfs_linux.h and move them into the files that need them.
Or it could be left as-is.
Signed-off-by: NEric Sandeen <sandeen@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

We need to derive the mount pointer from a buffer in a lot of place.
Add a direct pointer to short cut the pointer chasing.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

The inode geometry structure isn't related to ondisk format; it's
support for the mount structure.  Move it to xfs_shared.h.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Remove typedefs and consolidate local variable initialization.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NNick Desaulniers <ndesaulniers@google.com>