Relocation code will drop btrfs_root::reloc_root as soon as
merge_reloc_root() finishes.

However later qgroup code will need to access btrfs_root::reloc_root
after merge_reloc_root() for delayed subtree rescan.

So alter the timming of resetting btrfs_root:::reloc_root, make it
happens after transaction commit.

With this patch, we will introduce a new btrfs_root::state,
BTRFS_ROOT_DEAD_RELOC_TREE, to info part of btrfs_root::reloc_tree user
that although btrfs_root::reloc_tree is still non-NULL, but still it's
not used any more.

The lifespan of btrfs_root::reloc tree will become:
          Old behavior            |              New
------------------------------------------------------------------------
btrfs_init_reloc_root()      ---  | btrfs_init_reloc_root()      ---
  set reloc_root              |   |   set reloc_root              |
                              |   |                               |
                              |   |                               |
merge_reloc_root()            |   | merge_reloc_root()            |
|- btrfs_update_reloc_root() ---  | |- btrfs_update_reloc_root() -+-
     clear btrfs_root::reloc_root |      set ROOT_DEAD_RELOC_TREE |
                                  |      record root into dirty   |
                                  |      roots rbtree             |
                                  |                               |
                                  | reloc_block_group() Or        |
                                  | btrfs_recover_relocation()    |
                                  | | After transaction commit    |
                                  | |- clean_dirty_subvols()     ---
                                  |     clear btrfs_root::reloc_root

During ROOT_DEAD_RELOC_TREE set lifespan, the only user of
btrfs_root::reloc_tree should be qgroup.

Since reloc root needs a longer life-span, this patch will also delay
btrfs_drop_snapshot() call.
Now btrfs_drop_snapshot() is called in clean_dirty_subvols().

This patch will increase the size of btrfs_root by 16 bytes.
Signed-off-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This function is a simple wrapper over btrfs_get_extent that returns
either:

a) A real extent in the passed range or
b) Adjusted extent based on whether delalloc bytes are found backing up
   a hole.

To support these semantics it doesn't need the page/pg_offset/create
arguments which are passed to btrfs_get_extent in case an extent is to
be created. So simplify the function by removing the unused arguments.
No functional changes.
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Since this function is no longer a callback there is no need to have
its first argument obfuscated with a void *. Change it directly to a
pointer to an inode. No functional changes.
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

The cleaner thread usually takes care of delayed iputs, with the
exception of the btrfs_end_transaction_throttle path.  Delaying iputs
means we are potentially delaying the eviction of an inode and it's
respective space.  The cleaner thread only gets woken up every 30
seconds, or when we require space.  If there are a lot of inodes that
need to be deleted we could induce a serious amount of latency while we
wait for these inodes to be evicted.  So instead wakeup the cleaner if
it's not already awake to process any new delayed iputs we add to the
list.  If we suddenly need space we will less likely be backed up
behind a bunch of inodes that are waiting to be deleted, and we could
possibly free space before we need to get into the flushing logic which
will save us some latency.
Reviewed-by: NFilipe Manana <fdmanana@suse.com>
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

We weren't doing any of the accounting cleanup when we aborted
transactions.  Fix this by making cleanup_ref_head_accounting global and
calling it from the abort code, this fixes the issue where our
accounting was all wrong after the fs aborts.

The test generic/475 on a 2G VM can trigger the problems eg.:

  [ 8502.136957] WARNING: CPU: 0 PID: 11064 at fs/btrfs/extent-tree.c:5986 btrfs_free_block_grou +ps+0x3dc/0x410 [btrfs]
  [ 8502.148372] CPU: 0 PID: 11064 Comm: umount Not tainted 5.0.0-rc1-default+ #394
  [ 8502.150807] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.2-0-gf9626 +cc-prebuilt.qemu-project.org 04/01/2014
  [ 8502.154317] RIP: 0010:btrfs_free_block_groups+0x3dc/0x410 [btrfs]
  [ 8502.160623] RSP: 0018:ffffb1ab84b93de8 EFLAGS: 00010206
  [ 8502.161906] RAX: 0000000001000000 RBX: ffff9f34b1756400 RCX: 0000000000000000
  [ 8502.163448] RDX: 0000000000000002 RSI: 0000000000000001 RDI: ffff9f34b1755400
  [ 8502.164906] RBP: ffff9f34b7e8c000 R08: 0000000000000001 R09: 0000000000000000
  [ 8502.166716] R10: 0000000000000000 R11: 0000000000000001 R12: ffff9f34b7e8c108
  [ 8502.168498] R13: ffff9f34b7e8c158 R14: 0000000000000000 R15: dead000000000100
  [ 8502.170296] FS:  00007fb1cf15ffc0(0000) GS:ffff9f34bd400000(0000) knlGS:0000000000000000
  [ 8502.172439] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  [ 8502.173669] CR2: 00007fb1ced507b0 CR3: 000000002f7a6000 CR4: 00000000000006f0
  [ 8502.175094] Call Trace:
  [ 8502.175759]  close_ctree+0x17f/0x350 [btrfs]
  [ 8502.176721]  generic_shutdown_super+0x64/0x100
  [ 8502.177702]  kill_anon_super+0x14/0x30
  [ 8502.178607]  btrfs_kill_super+0x12/0xa0 [btrfs]
  [ 8502.179602]  deactivate_locked_super+0x29/0x60
  [ 8502.180595]  cleanup_mnt+0x3b/0x70
  [ 8502.181406]  task_work_run+0x98/0xc0
  [ 8502.182255]  exit_to_usermode_loop+0x83/0x90
  [ 8502.183113]  do_syscall_64+0x15b/0x180
  [ 8502.183919]  entry_SYSCALL_64_after_hwframe+0x49/0xbe

Corresponding to

  release_global_block_rsv() {
  ...
  WARN_ON(fs_info->delayed_refs_rsv.reserved > 0);

CC: stable@vger.kernel.org
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
[ add log dump ]
Signed-off-by: NDavid Sterba <dsterba@suse.com>

1) keeping a copy in btrfs_fs_info is completely pointless - we never
use it for anything.  Getting rid of that allows for simpler calling
conventions for setup_security_options() (caller is responsible for
freeing mnt_opts in all cases).

2) on remount we want to use ->sb_remount(), not ->sb_set_mnt_opts(),
same as we would if not for FS_BINARY_MOUNTDATA.  Behaviours *are*
close (in fact, selinux sb_set_mnt_opts() ought to punt to
sb_remount() in "already initialized" case), but let's handle
that uniformly.  And the only reason why the original btrfs changes
didn't go for security_sb_remount() in btrfs_remount() case is that
it hadn't been exported.  Let's export it for a while - it'll be
going away soon anyway.
Reviewed-by: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

When debugging some weird extent reference bug I suspected that we were
changing a snapshot while we were deleting it, which could explain my
bug.  This was indeed what was happening, and this patch helped me
verify my theory.  It is never correct to modify the snapshot once it's
being deleted, so mark the root when we are deleting it and make sure we
complain about it when it happens.
Reviewed-by: NFilipe Manana <fdmanana@suse.com>
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Now with the delayed_refs_rsv we can now know exactly how much pending
delayed refs space we need.  This means we can drastically simplify
btrfs_check_space_for_delayed_refs by simply checking how much space we
have reserved for the global rsv (which acts as a spill over buffer) and
the delayed refs rsv.  If our total size is beyond that amount then we
know it's time to commit the transaction and stop any more delayed refs
from being generated.

With the introduction of dealyed_refs_rsv infrastructure, namely
btrfs_update_delayed_refs_rsv we now know exactly how much pending
delayed refs space is required.
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

A nice thing we gain with the delayed refs rsv is the ability to flush
the delayed refs on demand to deal with enospc pressure.  Add states to
flush delayed refs on demand, and this will allow us to remove a lot of
ad-hoc work around checking to see if we should commit the transaction
to run our delayed refs.
Signed-off-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Traditionally we've had voodoo in btrfs to account for the space that
delayed refs may take up by having a global_block_rsv.  This works most
of the time, except when it doesn't.  We've had issues reported and seen
in production where sometimes the global reserve is exhausted during
transaction commit before we can run all of our delayed refs, resulting
in an aborted transaction.  Because of this voodoo we have equally
dubious flushing semantics around throttling delayed refs which we often
get wrong.

So instead give them their own block_rsv.  This way we can always know
exactly how much outstanding space we need for delayed refs.  This
allows us to make sure we are constantly filling that reservation up
with space, and allows us to put more precise pressure on the enospc
system.  Instead of doing math to see if its a good time to throttle,
the normal enospc code will be invoked if we have a lot of delayed refs
pending, and they will be run via the normal flushing mechanism.

For now the delayed_refs_rsv will hold the reservations for the delayed
refs, the block group updates, and deleting csums.  We could have a
separate rsv for the block group updates, but the csum deletion stuff is
still handled via the delayed_refs so that will stay there.

Historical background:

The global reserve has grown to cover everything we don't reserve space
explicitly for, and we've grown a lot of weird ad-hoc heuristics to know
if we're running short on space and when it's time to force a commit.  A
failure rate of 20-40 file systems when we run hundreds of thousands of
them isn't super high, but cleaning up this code will make things less
ugly and more predictible.

Thus the delayed refs rsv.  We always know how many delayed refs we have
outstanding, and although running them generates more we can use the
global reserve for that spill over, which fits better into it's desired
use than a full blown reservation.  This first approach is to simply
take how many times we're reserving space for and multiply that by 2 in
order to save enough space for the delayed refs that could be generated.
This is a niave approach and will probably evolve, but for now it works.
Signed-off-by: NJosef Bacik <jbacik@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com> # high-level review
[ added background notes from the cover letter ]
Signed-off-by: NDavid Sterba <dsterba@suse.com>

After the rw semaphore has been added, the custom blocking using
::blocking_readers and ::read_lock_wq is redundant.

The blocking logic in __btrfs_map_block is replaced by extending the
time the semaphore is held, that has the same blocking effect on writes
as the previous custom scheme that waited until ::blocking_readers was
zero.
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This is the first part of removing the custom locking and waiting scheme
used for device replace. It was probably copied from extent buffer
locking, but there's nothing that would require more than is provided by
the common locking primitives.

The rw spinlock protects waiting tasks counter in case of incompatible
locks and the waitqueue. Same as rw semaphore.

This patch only switches the locking primitive, for better
bisectability.  There should be no functional change other than the
overhead of the locking and potential sleeping instead of spinning when
the lock is contended.
Signed-off-by: NDavid Sterba <dsterba@suse.com>

The first auto-assigned value to enum is 0, we can use that and not
initialize all members where the auto-increment does the same. This is
used for values that are not part of on-disk format.
Reviewed-by: NOmar Sandoval <osandov@fb.com>
Reviewed-by: NQu Wenruo <wqu@suse.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

We can use simple enum for values that are not part of on-disk format:
root tree flags.
Reviewed-by: NOmar Sandoval <osandov@fb.com>
Reviewed-by: NQu Wenruo <wqu@suse.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

We can use simple enum for values that are not part of on-disk format:
internal filesystem states.
Reviewed-by: NOmar Sandoval <osandov@fb.com>
Reviewed-by: NQu Wenruo <wqu@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

We can use simple enum for values that are not part of on-disk format:
block reserve types.
Reviewed-by: NOmar Sandoval <osandov@fb.com>
Reviewed-by: NQu Wenruo <wqu@suse.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

We can use simple enum for values that are not part of on-disk format:
global filesystem states.
Reviewed-by: NOmar Sandoval <osandov@fb.com>
Reviewed-by: NQu Wenruo <wqu@suse.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This function really checks whether adding more data to the bio will
straddle a stripe/chunk. So first let's give it a more appropraite name
- btrfs_bio_fits_in_stripe. Secondly, the offset parameter was never
used to just remove it. Thirdly, pages are submitted to either btree or
data inodes so it's guaranteed that tree->ops is set so replace the
check with an ASSERT. Finally, document the parameters of the function.
No functional changes.
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Currently btrfs_fs_info structure contains a copy of the
fsid/metadata_uuid fields. Same values are also contained in the
btrfs_fs_devices structure which fs_info has a reference to. Let's
reduce duplication by removing the fields from fs_info and always refer
to the ones in fs_devices. No functional changes.
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This field is going to be used when the user wants to change the UUID
of the filesystem without having to rewrite all metadata blocks. This
field adds another level of indirection such that when the FSID is
changed what really happens is the current UUID (the one with which the
fs was created) is copied to the 'metadata_uuid' field in the superblock
as well as a new incompat flag is set METADATA_UUID. When the kernel
detects this flag is set it knows that the superblock in fact has 2
UUIDs:

1. Is the UUID which is user-visible, currently known as FSID.
2. Metadata UUID - this is the UUID which is stamped into all on-disk
   datastructures belonging to this file system.

When the new incompat flag is present device scanning checks whether
both fsid/metadata_uuid of the scanned device match any of the
registered filesystems. When the flag is not set then both UUIDs are
equal and only the FSID is retained on disk, metadata_uuid is set only
in-memory during mount.

Additionally a new metadata_uuid field is also added to the fs_info
struct. It's initialised either with the FSID in case METADATA_UUID
incompat flag is not set or with the metdata_uuid of the superblock
otherwise.

This commit introduces the new fields as well as the new incompat flag
and switches all users of the fsid to the new logic.
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
[ minor updates in comments ]
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Depending on whether CONFIG_BTRFS_FS_RUN_SANITY_TESTS is set, some BTRFS
functions are either local to the file they are implemented in and thus
should be declared static or are called from within the test
implementation defined in a different file.

Introduce an EXPORT_FOR_TESTS macro which depending on
CONFIG_BTRFS_FS_RUN_SANITY_TESTS either adds the 'static' keyword to a
function or not.
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Snapshot is expected to be fast. But if there are writers steadily
creating dirty pages in our subvolume, the snapshot may take a very long
time to complete. To fix the problem, we use tagged writepage for
snapshot flusher as we do in the generic write_cache_pages(), so we can
omit pages dirtied after the snapshot command.

This does not change the semantics regarding which data get to the
snapshot, if there are pages being dirtied during the snapshotting
operation.  There's a sync called before snapshot is taken in old/new
case, any IO in flight just after that may be in the snapshot but this
depends on other system effects that might still sync the IO.

We do a simple snapshot speed test on a Intel D-1531 box:

fio --ioengine=libaio --iodepth=32 --bs=4k --rw=write --size=64G
--direct=0 --thread=1 --numjobs=1 --time_based --runtime=120
--filename=/mnt/sub/testfile --name=job1 --group_reporting & sleep 5;
time btrfs sub snap -r /mnt/sub /mnt/snap; killall fio

original: 1m58sec
patched:  6.54sec

This is the best case for this patch since for a sequential write case,
we omit nearly all pages dirtied after the snapshot command.

For a multi writers, random write test:

fio --ioengine=libaio --iodepth=32 --bs=4k --rw=randwrite --size=64G
--direct=0 --thread=1 --numjobs=4 --time_based --runtime=120
--filename=/mnt/sub/testfile --name=job1 --group_reporting & sleep 5;
time btrfs sub snap -r /mnt/sub /mnt/snap; killall fio

original: 15.83sec
patched:  10.35sec

The improvement is smaller compared to the sequential write case,
since we omit only half of the pages dirtied after snapshot command.
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NEthan Lien <ethanlien@synology.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This parameter was never used, yet was part of the interface of the
function ever since its introduction as extent_io_ops::writepage_end_io_hook
in e6dcd2dc ("Btrfs: New data=ordered implementation"). Now that
NULL is passed everywhere as a value for this parameter let's remove it
for good. No functional changes.
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

A later patch will implement swap file support for Btrfs, but before we
do that, we need to make sure that the various Btrfs ioctls cannot
change a swap file.

When a swap file is active, we must make sure that the extents of the
file are not moved and that they don't become shared. That means that
the following are not safe:

- chattr +c (enable compression)
- reflink
- dedupe
- snapshot
- defrag

Don't allow those to happen on an active swap file.

Additionally, balance, resize, device remove, and device replace are
also unsafe if they affect an active swapfile. Add a red-black tree of
block groups and devices which contain an active swapfile. Relocation
checks each block group against this tree and skips it or errors out for
balance or resize, respectively. Device remove and device replace check
the tree for the device they will operate on.

Note that we don't have to worry about chattr -C (disable nocow), which
we ignore for non-empty files, because an active swapfile must be
non-empty and can't be truncated. We also don't have to worry about
autodefrag because it's only done on COW files. Truncate and fallocate
are already taken care of by the generic code. Device add doesn't do
relocation so it's not an issue, either.
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This is the counterpart to merge_extent_hook, similarly, it's used only
for data/freespace inodes so let's remove it, rename it and call it
directly where necessary. No functional changes.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This callback is used only for data and free space inodes. Such inodes
are guaranteed to have their extent_io_tree::private_data set to the
inode struct. Exploit this fact to directly call the function. Also give
it a more descriptive name. No functional changes.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This is the counterpart to ex-set_bit_hook (now btrfs_set_delalloc_extent),
similar to what was done before remove clear_bit_hook and rename the
function. No functional changes.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This callback is used to properly account delalloc extents for data
inodes (ordinary file inodes and freespace v1 inodes). Those can be
easily identified since they have their extent_io trees ->private_data
member point to the inode. Let's exploit this fact to remove the
needless indirection through extent_io_hooks and directly call the
function. Also give the function a name which reflects its purpose -
btrfs_set_delalloc_extent.

This patch also modified test_find_delalloc so that the extent_io_tree
used for testing doesn't have its ->private_data set which would have
caused a crash in btrfs_set_delalloc_extent due to the btrfs_inode->root
member not being initialised. The old version of the code also didn't
call set_bit_hook since the extent_io ops weren't set for the inode.  No
functional changes.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This callback is ony ever called for data page writeout so there is no
need to actually abstract it via extent_io_ops. Lets just export it,
remove the definition of the callback and call it directly in the
functions that invoke the callback. Also rename the function to
btrfs_writepage_endio_finish_ordered since what it really does is
account finished io in the ordered extent data structures.  No
functional changes.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This hook is called only from __extent_writepage_io which is already
called only from the data page writeout path. So there is no need to
make an indirect call via extent_io_ops. This patch just removes the
callback definition, exports the callback function and calls it directly
at the only call site. Also give the function a more descriptive name.
No functional changes.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This callback is called only from writepage_delalloc which in turn is
guaranteed to be called from the data page writeout path. In the end
there is no reason to have the call to this function to be indrected via
the extent_io_ops structure. This patch removes the callback definition,
exports the function and calls it directly. No functional changes.
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
[ rename to btrfs_run_delalloc_range ]
Signed-off-by: NDavid Sterba <dsterba@suse.com>

When we are writing out a free space cache, during the transaction commit
phase, we can end up in a deadlock which results in a stack trace like the
following:

 schedule+0x28/0x80
 btrfs_tree_read_lock+0x8e/0x120 [btrfs]
 ? finish_wait+0x80/0x80
 btrfs_read_lock_root_node+0x2f/0x40 [btrfs]
 btrfs_search_slot+0xf6/0x9f0 [btrfs]
 ? evict_refill_and_join+0xd0/0xd0 [btrfs]
 ? inode_insert5+0x119/0x190
 btrfs_lookup_inode+0x3a/0xc0 [btrfs]
 ? kmem_cache_alloc+0x166/0x1d0
 btrfs_iget+0x113/0x690 [btrfs]
 __lookup_free_space_inode+0xd8/0x150 [btrfs]
 lookup_free_space_inode+0x5b/0xb0 [btrfs]
 load_free_space_cache+0x7c/0x170 [btrfs]
 ? cache_block_group+0x72/0x3b0 [btrfs]
 cache_block_group+0x1b3/0x3b0 [btrfs]
 ? finish_wait+0x80/0x80
 find_free_extent+0x799/0x1010 [btrfs]
 btrfs_reserve_extent+0x9b/0x180 [btrfs]
 btrfs_alloc_tree_block+0x1b3/0x4f0 [btrfs]
 __btrfs_cow_block+0x11d/0x500 [btrfs]
 btrfs_cow_block+0xdc/0x180 [btrfs]
 btrfs_search_slot+0x3bd/0x9f0 [btrfs]
 btrfs_lookup_inode+0x3a/0xc0 [btrfs]
 ? kmem_cache_alloc+0x166/0x1d0
 btrfs_update_inode_item+0x46/0x100 [btrfs]
 cache_save_setup+0xe4/0x3a0 [btrfs]
 btrfs_start_dirty_block_groups+0x1be/0x480 [btrfs]
 btrfs_commit_transaction+0xcb/0x8b0 [btrfs]

At cache_save_setup() we need to update the inode item of a block group's
cache which is located in the tree root (fs_info->tree_root), which means
that it may result in COWing a leaf from that tree. If that happens we
need to find a free metadata extent and while looking for one, if we find
a block group which was not cached yet we attempt to load its cache by
calling cache_block_group(). However this function will try to load the
inode of the free space cache, which requires finding the matching inode
item in the tree root - if that inode item is located in the same leaf as
the inode item of the space cache we are updating at cache_save_setup(),
we end up in a deadlock, since we try to obtain a read lock on the same
extent buffer that we previously write locked.

So fix this by using the tree root's commit root when searching for a
block group's free space cache inode item when we are attempting to load
a free space cache. This is safe since block groups once loaded stay in
memory forever, as well as their caches, so after they are first loaded
we will never need to read their inode items again. For new block groups,
once they are created they get their ->cached field set to
BTRFS_CACHE_FINISHED meaning we will not need to read their inode item.
Reported-by: NAndrew Nelson <andrew.s.nelson@gmail.com>
Link: https://lore.kernel.org/linux-btrfs/CAPTELenq9x5KOWuQ+fa7h1r3nsJG8vyiTH8+ifjURc_duHh2Wg@mail.gmail.com/
Fixes: 9d66e233 ("Btrfs: load free space cache if it exists")
Tested-by: NAndrew Nelson <andrew.s.nelson@gmail.com>
Reviewed-by: NJosef Bacik <josef@toxicpanda.com>
Signed-off-by: NFilipe Manana <fdmanana@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Change the remap_file_range functions to take a number of bytes to
operate upon and return the number of bytes they operated on.  This is a
requirement for allowing fs implementations to return short clone/dedupe
results to the user, which will enable us to obey resource limits in a
graceful manner.

A subsequent patch will enable copy_file_range to signal to the
->clone_file_range implementation that it can handle a short length,
which will be returned in the function's return value.  For now the
short return is not implemented anywhere so the behavior won't change --
either copy_file_range manages to clone the entire range or it tries an
alternative.

Neither clone ioctl can take advantage of this, alas.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NAmir Goldstein <amir73il@gmail.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Combine the clone_file_range and dedupe_file_range operations into a
single remap_file_range file operation dispatch since they're
fundamentally the same operation.  The differences between the two can
be made in the prep functions.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NAmir Goldstein <amir73il@gmail.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

The avg_delayed_ref_runtime can be referenced from the transaction
handle.
Signed-off-by: NLu Fengqi <lufq.fnst@cn.fujitsu.com>
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

It can be referenced from the transaction handle.
Signed-off-by: NLu Fengqi <lufq.fnst@cn.fujitsu.com>
Reviewed-by: NNikolay Borisov <nborisov@suse.com>
Reviewed-by: NDavid Sterba <dsterba@suse.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Btrfs's btree locking has two modes, spinning mode and blocking mode,
while searching btree, locking is always acquired in spinning mode and
then converted to blocking mode if necessary, and in some hot paths we may
switch the locking back to spinning mode by btrfs_clear_path_blocking().

When acquiring locks, both of reader and writer need to wait for blocking
readers and writers to complete before doing read_lock()/write_lock().

The problem is that btrfs_clear_path_blocking() needs to switch nodes
in the path to blocking mode at first (by btrfs_set_path_blocking) to
make lockdep happy before doing its actual clearing blocking job.

When switching to blocking mode from spinning mode, it consists of

step 1) bumping up blocking readers counter and
step 2) read_unlock()/write_unlock(),

this has caused serious ping-pong effect if there're a great amount of
concurrent readers/writers, as waiters will be woken up and go to
sleep immediately.

1) Killing this kind of ping-pong results in a big improvement in my 1600k
files creation script,

MNT=/mnt/btrfs
mkfs.btrfs -f /dev/sdf
mount /dev/def $MNT
time fsmark  -D  10000  -S0  -n  100000  -s  0  -L  1 -l /tmp/fs_log.txt \
        -d  $MNT/0  -d  $MNT/1 \
        -d  $MNT/2  -d  $MNT/3 \
        -d  $MNT/4  -d  $MNT/5 \
        -d  $MNT/6  -d  $MNT/7 \
        -d  $MNT/8  -d  $MNT/9 \
        -d  $MNT/10  -d  $MNT/11 \
        -d  $MNT/12  -d  $MNT/13 \
        -d  $MNT/14  -d  $MNT/15

w/o patch:
real    2m27.307s
user    0m12.839s
sys     13m42.831s

w/ patch:
real    1m2.273s
user    0m15.802s
sys     8m16.495s

1.1) latency histogram from funclatency[1]

Overall with the patch, there're ~50% less write lock acquisition and
the 95% max latency that write lock takes also reduces to ~100ms from
>500ms.

--------------------------------------------
w/o patch:
--------------------------------------------
Function = btrfs_tree_lock
     msecs               : count     distribution
         0 -> 1          : 2385222  |****************************************|
         2 -> 3          : 37147    |                                        |
         4 -> 7          : 20452    |                                        |
         8 -> 15         : 13131    |                                        |
        16 -> 31         : 3877     |                                        |
        32 -> 63         : 3900     |                                        |
        64 -> 127        : 2612     |                                        |
       128 -> 255        : 974      |                                        |
       256 -> 511        : 165      |                                        |
       512 -> 1023       : 13       |                                        |

Function = btrfs_tree_read_lock
     msecs               : count     distribution
         0 -> 1          : 6743860  |****************************************|
         2 -> 3          : 2146     |                                        |
         4 -> 7          : 190      |                                        |
         8 -> 15         : 38       |                                        |
        16 -> 31         : 4        |                                        |

--------------------------------------------
w/ patch:
--------------------------------------------
Function = btrfs_tree_lock
     msecs               : count     distribution
         0 -> 1          : 1318454  |****************************************|
         2 -> 3          : 6800     |                                        |
         4 -> 7          : 3664     |                                        |
         8 -> 15         : 2145     |                                        |
        16 -> 31         : 809      |                                        |
        32 -> 63         : 219      |                                        |
        64 -> 127        : 10       |                                        |

Function = btrfs_tree_read_lock
     msecs               : count     distribution
         0 -> 1          : 6854317  |****************************************|
         2 -> 3          : 2383     |                                        |
         4 -> 7          : 601      |                                        |
         8 -> 15         : 92       |                                        |

2) dbench also proves the improvement,
dbench -t 120 -D /mnt/btrfs 16

w/o patch:
Throughput 158.363 MB/sec

w/ patch:
Throughput 449.52 MB/sec

3) xfstests didn't show any additional failures.

One thing to note is that callers may set path->leave_spinning to have
all nodes in the path stay in spinning mode, which means callers are
ready to not sleep before releasing the path, but it won't cause
problems if they don't want to sleep in blocking mode.

[1]: https://github.com/iovisor/bcc/blob/master/tools/funclatency.pySigned-off-by: NLiu Bo <bo.liu@linux.alibaba.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

The replace_wait and bio_counter were mistakenly added to fs_info in
commit c404e0dc ("Btrfs: fix use-after-free in the finishing
procedure of the device replace"), but they logically belong to
fs_info::dev_replace. Besides, bio_counter is a very generic name and is
confusing in bare fs_info context.
Reviewed-by: NAnand Jain <anand.jain@oracle.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

This member seems to be copied from the extent_buffer locking scheme and
is at least used to assert that the read lock/unlock is properly nested.
In some way. While the _inc/_dec are called inside the read lock
section, the asserts are both inside and outside, so the ordering is not
guaranteed and we can see read/inc/dec ordered in any way
(theoretically).

A missing call of btrfs_dev_replace_clear_lock_blocking could cause
unexpected read_locks count, so this at least looks like a valid
assertion, but this will become unnecessary with later updates.
Reviewed-by: NAnand Jain <anand.jain@oracle.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>

Avoid the inline ifdefs and use two sections for self-tests enabled and
disabled.

Though there could be no ifdef and unconditional test_bit of
BTRFS_FS_STATE_DUMMY_FS_INFO, the static inline can help to optimize out
any code that would depend on conditions using btrfs_is_testing.

As this is only for the testing code, drop unlikely().
Reviewed-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NDavid Sterba <dsterba@suse.com>