We have a few places where we skip doing csums if we mounted with one of
the rescue options that ignores bad csum roots.  In the future when
there are multiple csum roots it'll be costly to check and see if there
are any missing csum roots, so simply add a flag to indicate the fs
should skip loading csums in case of errors.
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

We used to need the root for btrfs_reserve_metadata_bytes to check the
orphan cleanup state, but we no longer need that, we simply need the
fs_info.  Change btrfs_reserve_metadata_bytes() to use the fs_info, and
change both btrfs_block_rsv_refill() and btrfs_block_rsv_add() to do the
same as they simply call btrfs_reserve_metadata_bytes() and then
manipulate the block_rsv that is being used.
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Now that we don't care about the stage of the orphan_cleanup_state,
simply replace it with a bit on ->state to make sure we don't call the
orphan cleanup every time we wander into this root.
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

I forgot to convert this over when I introduced the global reserve
stealing code to the space flushing code.  Evict was simply trying to
make its reservation and then if it failed it would steal from the
global rsv, which is racey because it's outside of the normal ticketing
code.

Fix this by setting ticket->steal if we are BTRFS_RESERVE_FLUSH_EVICT,
and then make the priority flushing path do the steal for us.
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Instead of getting the btrfs_item for this, simply pass in the slot of
the item and then use the btrfs_item_size_nr() helper inside of
btrfs_file_extent_inline_item_len().
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

When doing a direct IO write against a file range that either has
preallocated extents in that range or has regular extents and the file
has the NOCOW attribute set, the write fails with -ENOSPC when all of
the following conditions are met:

1) There are no data blocks groups with enough free space matching
   the size of the write;

2) There's not enough unallocated space for allocating a new data block
   group;

3) The extents in the target file range are not shared, neither through
   snapshots nor through reflinks.

This is wrong because a NOCOW write can be done in such case, and in fact
it's possible to do it using a buffered IO write, since when failing to
allocate data space, the buffered IO path checks if a NOCOW write is
possible.

The failure in direct IO write path comes from the fact that early on,
at btrfs_dio_iomap_begin(), we try to allocate data space for the write
and if it that fails we return the error and stop - we never check if we
can do NOCOW. But later, at btrfs_get_blocks_direct_write(), we check
if we can do a NOCOW write into the range, or a subset of the range, and
then release the previously reserved data space.

Fix this by doing the data reservation only if needed, when we must COW,
at btrfs_get_blocks_direct_write() instead of doing it at
btrfs_dio_iomap_begin(). This also simplifies a bit the logic and removes
the inneficiency of doing unnecessary data reservations.

The following example test script reproduces the problem:

  $ cat dio-nocow-enospc.sh
  #!/bin/bash

  DEV=/dev/sdj
  MNT=/mnt/sdj

  # Use a small fixed size (1G) filesystem so that it's quick to fill
  # it up.
  # Make sure the mixed block groups feature is not enabled because we
  # later want to not have more space available for allocating data
  # extents but still have enough metadata space free for the file writes.
  mkfs.btrfs -f -b $((1024 * 1024 * 1024)) -O ^mixed-bg $DEV
  mount $DEV $MNT

  # Create our test file with the NOCOW attribute set.
  touch $MNT/foobar
  chattr +C $MNT/foobar

  # Now fill in all unallocated space with data for our test file.
  # This will allocate a data block group that will be full and leave
  # no (or a very small amount of) unallocated space in the device, so
  # that it will not be possible to allocate a new block group later.
  echo
  echo "Creating test file with initial data..."
  xfs_io -c "pwrite -S 0xab -b 1M 0 900M" $MNT/foobar

  # Now try a direct IO write against file range [0, 10M[.
  # This should succeed since this is a NOCOW file and an extent for the
  # range was previously allocated.
  echo
  echo "Trying direct IO write over allocated space..."
  xfs_io -d -c "pwrite -S 0xcd -b 10M 0 10M" $MNT/foobar

  umount $MNT

When running the test:

  $ ./dio-nocow-enospc.sh
  (...)

  Creating test file with initial data...
  wrote 943718400/943718400 bytes at offset 0
  900 MiB, 900 ops; 0:00:01.43 (625.526 MiB/sec and 625.5265 ops/sec)

  Trying direct IO write over allocated space...
  pwrite: No space left on device

A test case for fstests will follow, testing both this direct IO write
scenario as well as the buffered IO write scenario to make it less likely
to get future regressions on the buffered IO case.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NFilipe Manana <fdmanana@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

The root argument passed to btrfs_unlink_inode() and its callee,
__btrfs_unlink_inode(), always matches the root of the given directory and
the given inode. So remove the argument and make __btrfs_unlink_inode()
use the root of the directory.
Signed-off-by: NFilipe Manana <fdmanana@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This reverts commit 4c2bf276.

The kmaps in compression code are still needed and cause crashes on
32bit machines (ARM, x86). Reproducible eg. by running fstest btrfs/004
with enabled LZO or ZSTD compression.

Link: https://lore.kernel.org/all/CAJCQCtT+OuemovPO7GZk8Y8=qtOObr0XTDp8jh4OHD6y84AFxw@mail.gmail.com/
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=214839Signed-off-by: NDavid Sterba <dsterba@suse.com>

The member btrfs_bio::logical is only initialized by two call sites:

- btrfs_repair_one_sector()
  No corresponding site to utilize it.

- btrfs_submit_direct()
  The corresponding site to utilize it is btrfs_check_read_dio_bio().

However for btrfs_check_read_dio_bio(), we can grab the file_offset from
btrfs_dio_private::file_offset directly.

Thus it turns out we don't really need that btrfs_bio::logical member at
all.

For btrfs_bio, the logical bytenr can be fetched from its
bio->bi_iter.bi_sector directly.

So let's just remove the member to save 8 bytes for structure btrfs_bio.
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

The naming of "logical_offset" can be confused with logical bytenr of
the dio range.

In fact it's file offset, and the naming "file_offset" is already widely
used in all other sites.

Just do the rename to avoid confusion.
Signed-off-by: NQu Wenruo <wqu@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Instead of checking whether qgroup processing for a dealyed ref has to
happen in the core of delayed ref, simply pull the check at init time of
respective delayed ref structures. This eliminates the final use of
real_root in delayed-ref core paving the way to making this member
optional.
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

In order to make 'real_root' used only in ref-verify it's required to
have the necessary context to perform the same checks that this member
is used for. So add 'mod_root' which will contain the root on behalf of
which a delayed ref was created and a 'skip_group' parameter which
will contain callsite-specific override of skip_qgroup.
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

We have a few flags that are inconsistently used to describe the fs in
different states of failure.  As of 5963ffca ("btrfs: always abort
the transaction if we abort a trans handle") we will always set
BTRFS_FS_STATE_ERROR if we abort, so we don't have to check both ABORTED
and ERROR to see if things have gone wrong.  Add a helper to check
BTRFS_FS_STATE_ERROR and then convert all checkers of FS_STATE_ERROR to
use the helper.

The TRANS_ABORTED bit check was added in af722733 ("Btrfs: clean up
resources during umount after trans is aborted") but is not actually
specific.
Reviewed-by: NAnand Jain <anand.jain@oracle.com>
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Currently we will abort the transaction if we get a random error (like
-EIO) while trying to remove the directory entries from the root log
during rename.

However since these are simply log tree related errors, we can mark the
trans as needing a full commit.  Then if the error was truly
catastrophic we'll hit it during the normal commit and abort as
appropriate.
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NFilipe Manana <fdmanana@suse.com>
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

For compressed write, we use a mechanism called async COW, which unlike
regular run_delalloc_cow() or cow_file_range() will also unlock the
first page.

This mechanism allows us to continue handling next ranges, without
waiting for the time consuming compression.

But this has a problem for subpage case, as we could have the following
delalloc range for a page:

0		32K		64K
|	|///////|	|///////|
		\- A		\- B

In the above case, if we pass both ranges to cow_file_range_async(),
both range A and range B will try to unlock the full page [0, 64K).

And which one finishes later than the other one will try to do other
page operations like end_page_writeback() on a unlocked page, triggering
VM layer BUG_ON().

To make subpage compression work at least partially, here we add another
restriction for it, only allow compression if the delalloc range is
fully page aligned.

By that, async extent is always ensured to unlock the first page
exclusively, just like it used to be for regular sectorsize.

In theory, we only need to make sure the delalloc range fully covers its
first page, but the tail page will be locked anyway, blocking later
writeback until the compression finishes.

Thus here we choose to make sure the range is fully page aligned before
doing the compression.

In the future, we could optimize the situation by properly increasing
subpage::writers number for the locked page, but that also means we need
to change how we run delalloc range of page.
(Instead of running each delalloc range we hit, we need to find and lock
all delalloc ranges covering the page, then run each of them).
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

[BUG]
With experimental subpage compression enabled, a simple fsstress can
lead to self deadlock on page 720896:

        mkfs.btrfs -f -s 4k $dev > /dev/null
        mount $dev -o compress $mnt
        $fsstress -p 1 -n 100 -w -d $mnt -v -s 1625511156

[CAUSE]
If we have a file layout looks like below:

	0	32K	64K	96K	128K
	|//|		|///////////////|
	   4K

Then we run delalloc range for the inode, it will:

- Call find_lock_delalloc_range() with @delalloc_start = 0
  Then we got a delalloc range [0, 4K).

  This range will be COWed.

- Call find_lock_delalloc_range() again with @delalloc_start = 4K
  Since find_lock_delalloc_range() never cares whether the range
  is still inside page range [0, 64K), it will return range [64K, 128K).

  This range meets the condition for subpage compression, will go
  through async COW path.

  And async COW path will return @page_started.

  But that @page_started is now for range [64K, 128K), not for range
  [0, 64K).

- writepage_dellloc() returned 1 for page [0, 64K)
  Thus page [0, 64K) will not be unlocked, nor its page dirty status
  will be cleared.

Next time when we try to lock page [0, 64K) we will deadlock, as there
is no one to release page [0, 64K).

This problem will never happen for regular page size as one page only
contains one sector.  After the first find_lock_delalloc_range() call,
the @delalloc_end will go beyond @page_end no matter if we found a
delalloc range or not

Thus this bug only happens for subpage, as now we need multiple runs to
exhaust the delalloc range of a page.

[FIX]
Fix the problem by ensuring the delalloc range we ran at least started
inside @locked_page.

So that we will never get incorrect @page_started.

And to prevent such problem from happening again:

- Make find_lock_delalloc_range() return false if the found range is
  beyond @end value passed in.

  Since @end will be utilized now, add an ASSERT() to ensure we pass
  correct @end into find_lock_delalloc_range().

  This also means, for selftests we needs to populate @end before calling
  find_lock_delalloc_range().

- New ASSERT() in find_lock_delalloc_range()
  Now we will make sure the @start/@end passed in at least covers part
  of the page.

- New ASSERT() in run_delalloc_range()
  To make sure the range at least starts inside @locked page.

- Use @delalloc_start as proper cursor, while @delalloc_end is always
  reset to @page_end.
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Introduce a new helper, submit_uncompressed_range(), for async cow cases
where we fallback to COW.

There are some new updates introduced to the helper:

- Proper locked_page detection
  It's possible that the async_extent range doesn't cover the locked
  page.  In that case we shouldn't unlock the locked page.

  In the new helper, we will ensure that we only unlock the locked page
  when:

  * The locked page covers part of the async_extent range
  * The locked page is not unlocked by cow_file_range() nor
    extent_write_locked_range()

  This also means extra comments are added focusing on the page locking.

- Add extra comment on some rare parameter used.
  We use @unlock_page = 0 for cow_file_range(), where only two call
  sites doing the same thing, including the new helper.

  It's definitely worth some comments.
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

In function compress_file_range(), when the compression is finished, the
function just rounds up @total_in to PAGE_SIZE.  This is fine for
regular sectorsize == PAGE_SIZE case, but not for subpage.

Just change the ALIGN(, PAGE_SIZE) to round_up(, sectorsize) so that
both regular sectorsize and subpage sectorsize will be happy.
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

There are several cleanups for extent_write_locked_range(), most of them
are pure cleanups, but with some preparation for future subpage support.

- Add a proper comment for which call sites are suitable
  Unlike regular synchronized extent write back, if async COW or zoned
  COW happens, we have all pages in the range still locked.

  Thus for those (only) two call sites, we need this function to submit
  page content into bios and submit them.

- Remove @mode parameter
  All the existing two call sites pass WB_SYNC_ALL. No need for @mode
  parameter.

- Better error handling
  Currently if we hit an error during the page iteration loop, we
  overwrite @ret, causing only the last error can be recorded.

  Here we add @found_error and @first_error variable to record if we hit
  any error, and the first error we hit.
  So the first error won't get lost.

- Don't reuse @start as the cursor
  We reuse the parameter @start as the cursor to iterate the range, not
  a big problem, but since we're here, introduce a proper @cur as the
  cursor.

- Remove impossible branch
  Since all pages are still locked after the ordered extent is inserted,
  there is no way that pages can get its dirty bit cleared.
  Remove the branch where page is not dirty and replace it with an
  ASSERT().
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

We have a big chunk of code inside a while() loop, with tons of strange
jumps for error handling.  It's definitely not to the code standard of
today.  Move the code into a new function, submit_one_async_extent().

Since we're here, also do the following changes:

- Comment style change
  To follow the current scheme

- Don't fallback to non-compressed write then hitting ENOSPC
  If we hit ENOSPC for compressed write, how could we reserve more space
  for non-compressed write?
  Thus we go error path directly.
  This removes the retry: label.

- Add more comment for super long parameter list
  Explain which parameter is for, so we don't need to check the
  prototype.

- Move the error handling to submit_one_async_extent()
  Thus no strange code like:

  out_free:
	...
	goto again;
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

As the last caller in compression.c has been removed, we don't need that
function anymore.
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Although in btrfs we have very limited usage of PageChecked flag, it's
still some page flag not yet subpage compatible.

Fix it by introducing btrfs_subpage::checked_offset to do the convert.

For most call sites, especially for free-space cache, COW fixup and
btrfs_invalidatepage(), they all work in full page mode anyway.

For other call sites, they work as subpage compatible mode.

Some call sites need extra modification:

- btrfs_drop_pages()
  Needs extra parameter to get the real range we need to clear checked
  flag.

  Also since btrfs_drop_pages() will accept pages beyond the dirtied
  range, update btrfs_subpage_clamp_range() to handle such case
  by setting @len to 0 if the page is beyond target range.

- btrfs_invalidatepage()
  We need to call subpage helper before calling __btrfs_releasepage(),
  or it will trigger ASSERT() as page->private will be cleared.

- btrfs_verify_data_csum()
  In theory we don't need the io_bio->csum check anymore, but it's
  won't hurt.  Just change the comment.
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Since async_extent holds the compressed page, it would trigger the new
ASSERT() in btrfs_mark_ordered_io_finished() which checks that the range
is inside the page.

Now btrfs_writepage_endio_finish_ordered() can accept @page == NULL,
just pass NULL to btrfs_writepage_endio_finish_ordered().
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

For structure async_chunk, we use a very strange member layout to grab
structure async_cow who owns this async_chunk.

At initialization, it goes like this:

		async_chunk[i].pending = &ctx->num_chunks;

Then at async_cow_free() we do a super weird freeing:

	/*
	 * Since the pointer to 'pending' is at the beginning of the array of
	 * async_chunk's, freeing it ensures the whole array has been freed.
	 */
	if (atomic_dec_and_test(async_chunk->pending))
		kvfree(async_chunk->pending);

This is absolutely an abuse of kvfree().

Replace async_chunk::pending with async_chunk::async_cow, so that we can
grab the async_cow structure directly, without this strange dancing.

And with this change, there is no requirement for any specific member
location.
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

When inserting a batch of items into a btree, we end up looping over the
data sizes array 3 times:

1) Once in the caller of btrfs_insert_empty_items(), when it populates the
   array with the data sizes for each item;

2) Once at btrfs_insert_empty_items() to sum the elements of the data
   sizes array and compute the total data size;

3) And then once again at setup_items_for_insert(), where we do exactly
   the same as what we do at btrfs_insert_empty_items(), to compute the
   total data size.

That is not bad for small arrays, but when the arrays have hundreds of
elements, the time spent on looping is not negligible. For example when
doing batch inserts of delayed items for dir index items or when logging
a directory, it's common to have 200 to 260 dir index items in a single
batch when using a leaf size of 16K and using file names between 8 and 12
characters. For a 64K leaf size, multiply that by 4. Taking into account
that during directory logging or when flushing delayed dir index items we
can have many of those large batches, the time spent on the looping adds
up quickly.

It's also more important to avoid it at setup_items_for_insert(), since
we are holding a write lock on a leaf and, in some cases, on upper nodes
of the btree, which causes us to block other tasks that want to access
the leaf and nodes for longer than necessary.

So change the code so that setup_items_for_insert() and
btrfs_insert_empty_items() no longer compute the total data size, and
instead rely on the caller to supply it. This makes us loop over the
array only once, where we can both populate the data size array and
compute the total data size, taking advantage of spatial and temporal
locality. To make this more manageable, use a structure to contain
all the relevant details for a batch of items (keys array, data sizes
array, total data size, number of items), and use it as an argument
for btrfs_insert_empty_items() and setup_items_for_insert().

This patch is part of a small patchset that is comprised of the following
patches:

  btrfs: loop only once over data sizes array when inserting an item batch
  btrfs: unexport setup_items_for_insert()
  btrfs: use single bulk copy operations when logging directories

This is patch 1/3 and performance results, and the specific tests, are
included in the changelog of patch 3/3.
Signed-off-by: NFilipe Manana <fdmanana@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Previously we had "struct btrfs_bio", which records IO context for
mirrored IO and RAID56, and "strcut btrfs_io_bio", which records extra
btrfs specific info for logical bytenr bio.

With "btrfs_bio" renamed to "btrfs_io_context", we are safe to rename
"btrfs_io_bio" to "btrfs_bio" which is a more suitable name now.

The struct btrfs_bio changes meaning by this commit. There was a
suggested name like btrfs_logical_bio but it's a bit long and we'd
prefer to use a shorter name.

This could be a concern for backports to older kernels where the
different meaning could possibly cause confusion or bugs. Comparing the
new and old structures, there's no overlap among the struct members so a
build would break in case of incorrect backport.

We haven't had many backports to bio code anyway so this is more of a
theoretical cause of bugs and a matter of precaution but we'll need to
keep the semantic change in mind.
Signed-off-by: NQu Wenruo <wqu@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

After the first time we log a directory in the current transaction, for
each directory item in a changed leaf of the subvolume tree, we have to
check if we previously logged the item, in order to overwrite it in case
its data changed or skip it in case its data hasn't changed.

Checking if we have logged each item before not only wastes times, but it
also adds lock contention on the log tree. So in order to minimize the
number of times we do such checks, keep track of the offset of the last
key we logged for a directory and, on the next time we log the directory,
skip the checks for any new keys that have an offset greater than the
offset we have previously saved. This is specially effective for index
keys, because the offset for these keys comes from a monotonically
increasing counter.

This patch is part of a patchset comprised of the following 5 patches:

  btrfs: remove root argument from btrfs_log_inode() and its callees
  btrfs: remove redundant log root assignment from log_dir_items()
  btrfs: factor out the copying loop of dir items from log_dir_items()
  btrfs: insert items in batches when logging a directory when possible
  btrfs: keep track of the last logged keys when logging a directory

This is patch 5/5.

The following test was used on a non-debug kernel to measure the impact
it has on a directory fsync:

  $ cat test-dir-fsync.sh
  #!/bin/bash

  DEV=/dev/nvme0n1
  MNT=/mnt/nvme0n1

  NUM_NEW_FILES=100000
  NUM_FILE_DELETES=1000

  mkfs.btrfs -f $DEV
  mount -o ssd $DEV $MNT

  mkdir $MNT/testdir

  for ((i = 1; i <= $NUM_NEW_FILES; i++)); do
      echo -n > $MNT/testdir/file_$i
  done

  # fsync the directory, this will log the new dir items and the inodes
  # they point to, because these are new inodes.
  start=$(date +%s%N)
  xfs_io -c "fsync" $MNT/testdir
  end=$(date +%s%N)

  dur=$(( (end - start) / 1000000 ))
  echo "dir fsync took $dur ms after adding $NUM_NEW_FILES files"

  # sync to force transaction commit and wipeout the log.
  sync

  del_inc=$(( $NUM_NEW_FILES / $NUM_FILE_DELETES ))
  for ((i = 1; i <= $NUM_NEW_FILES; i += $del_inc)); do
      rm -f $MNT/testdir/file_$i
  done

  # fsync the directory, this will only log dir items, there are no
  # dentries pointing to new inodes.
  start=$(date +%s%N)
  xfs_io -c "fsync" $MNT/testdir
  end=$(date +%s%N)

  dur=$(( (end - start) / 1000000 ))
  echo "dir fsync took $dur ms after deleting $NUM_FILE_DELETES files"

  umount $MNT

Test results with NUM_NEW_FILES set to 100 000 and 1 000 000:

**** before patchset, 100 000 files, 1000 deletes ****

dir fsync took 848 ms after adding 100000 files
dir fsync took 175 ms after deleting 1000 files

**** after patchset, 100 000 files, 1000 deletes ****

dir fsync took 758 ms after adding 100000 files  (-11.2%)
dir fsync took 63 ms after deleting 1000 files   (-94.1%)

**** before patchset, 1 000 000 files, 1000 deletes ****

dir fsync took 9945 ms after adding 1000000 files
dir fsync took 473 ms after deleting 1000 files

**** after patchset, 1 000 000 files, 1000 deletes ****

dir fsync took 8677 ms after adding 1000000 files (-13.6%)
dir fsync took 146 ms after deleting 1000 files   (-105.6%)
Signed-off-by: NFilipe Manana <fdmanana@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Prepare for allowing preallocation for relocation inodes.
Reviewed-by: NNaohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: NJohannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

There are several places in our codebase where we check if a root is the
root of the data reloc tree and subsequent patches will introduce more.

Factor out the check into a small helper function instead of open coding
it multiple times.
Reviewed-by: NNaohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: NJohannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

In preparation to fix a bug in btrfs_show_devname().

Convert fs_devices::latest_bdev type from struct block_device to struct
btrfs_device and, rename the member to fs_devices::latest_dev.
So that btrfs_show_devname() can use fs_devices::latest_dev::name.
Tested-by: NSu Yue <l@damenly.su>
Signed-off-by: NAnand Jain <anand.jain@oracle.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

If we have written to the zone capacity, the device automatically
deactivates the zone. Sync up block group side (the active BG list and
zone_is_active flag) with it.

We need to do it both on data BGs and metadata BGs. On data side, we add a
hook to btrfs_finish_ordered_io(). On metadata side, we use
end_extent_buffer_writeback().

To reduce excess lookup of a block group, we mark the last extent buffer in
a block group with EXTENT_BUFFER_ZONE_FINISH flag. This cannot be done for
data (ordered_extent), because the address may change due to
REQ_OP_ZONE_APPEND.
Signed-off-by: NNaohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Replace the blk_poll interface that requires the caller to keep a queue
and cookie from the submissions with polling based on the bio.

Polling for the bio itself leads to a few advantages:

 - the cookie construction can made entirely private in blk-mq.c
 - the caller does not need to remember the request_queue and cookie
   separately and thus sidesteps their lifetime issues
 - keeping the device and the cookie inside the bio allows to trivially
   support polling BIOs remapping by stacking drivers
 - a lot of code to propagate the cookie back up the submission path can
   be removed entirely.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Tested-by: NMark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-15-hch@lst.deSigned-off-by: NJens Axboe <axboe@kernel.dk>

blk-cgroup.h pulls in blkdev.h and thus pretty much all the block
headers.  Break this dependency chain by turning wbc_blkcg_css into a
macro and dropping the blk-cgroup.h include.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20210920123328.1399408-2-hch@lst.deSigned-off-by: NJens Axboe <axboe@kernel.dk>

This reverts commit f2165627.

[BUG]
It's no longer possible to create compressed inline extent after commit
f2165627 ("btrfs: compression: don't try to compress if we don't
have enough pages").

[CAUSE]
For compression code, there are several possible reasons we have a range
that needs to be compressed while it's no more than one page.

- Compressed inline write
  The data is always smaller than one sector and the test lacks the
  condition to properly recognize a non-inline extent.

- Compressed subpage write
  For the incoming subpage compressed write support, we require page
  alignment of the delalloc range.
  And for 64K page size, we can compress just one page into smaller
  sectors.

For those reasons, the requirement for the data to be more than one page
is not correct, and is already causing regression for compressed inline
data writeback.  The idea of skipping one page to avoid wasting CPU time
could be revisited in the future.

[FIX]
Fix it by reverting the offending commit.
Reported-by: NZygo Blaxell <ce3g8jdj@umail.furryterror.org>
Link: https://lore.kernel.org/linux-btrfs/afa2742.c084f5d6.17b6b08dffc@tnonline.net
Fixes: f2165627 ("btrfs: compression: don't try to compress if we don't have enough pages")
CC: stable@vger.kernel.org # 4.4+
Signed-off-by: NQu Wenruo <wqu@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Creating subvolumes and snapshots is one of the core features of btrfs
and is even available to unprivileged users. Make it possible to use
subvolume and snapshot creation on idmapped mounts. This is a fairly
straightforward operation since all the permission checking helpers are
already capable of handling idmapped mounts. So we just need to pass
down the mount's userns.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Enable btrfs_permission() to handle idmapped mounts. This is just a
matter of passing down the mount's userns.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Enable btrfs_setattr() to handle idmapped mounts. This is just a matter
of passing down the mount's userns.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Enable btrfs_tmpfile() to handle idmapped mounts. This is just a matter
of passing down the mount's userns.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Enable btrfs_symlink() to handle idmapped mounts. This is just a matter
of passing down the mount's userns.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Enable btrfs_mkdir() to handle idmapped mounts. This is just a matter of
passing down the mount's userns.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>