linux.git/fs/btrfs/extent_map.c, branch v5.6

Btrfs: fix race between using extent maps and merging them

2020-02-12T16:16:46+00:00

We have a few cases where we allow an extent map that is in an extent map
tree to be merged with other extents in the tree. Such cases include the
unpinning of an extent after the respective ordered extent completed or
after logging an extent during a fast fsync. This can lead to subtle and
dangerous problems because when doing the merge some other task might be
using the same extent map and as consequence see an inconsistent state of
the extent map - for example sees the new length but has seen the old start
offset.

With luck this triggers a BUG_ON(), and not some silent bug, such as the
following one in __do_readpage():

  $ cat -n fs/btrfs/extent_io.c
  3061  static int __do_readpage(struct extent_io_tree *tree,
  3062                           struct page *page,
  (...)
  3127                  em = __get_extent_map(inode, page, pg_offset, cur,
  3128                                        end - cur + 1, get_extent, em_cached);
  3129                  if (IS_ERR_OR_NULL(em)) {
  3130                          SetPageError(page);
  3131                          unlock_extent(tree, cur, end);
  3132                          break;
  3133                  }
  3134                  extent_offset = cur - em->start;
  3135                  BUG_ON(extent_map_end(em) <= cur);
  (...)

Consider the following example scenario, where we end up hitting the
BUG_ON() in __do_readpage().

We have an inode with a size of 8KiB and 2 extent maps:

  extent A: file offset 0, length 4KiB, disk_bytenr = X, persisted on disk by
            a previous transaction

  extent B: file offset 4KiB, length 4KiB, disk_bytenr = X + 4KiB, not yet
            persisted but writeback started for it already. The extent map
	    is pinned since there's writeback and an ordered extent in
	    progress, so it can not be merged with extent map A yet

The following sequence of steps leads to the BUG_ON():

1) The ordered extent for extent B completes, the respective page gets its
   writeback bit cleared and the extent map is unpinned, at that point it
   is not yet merged with extent map A because it's in the list of modified
   extents;

2) Due to memory pressure, or some other reason, the MM subsystem releases
   the page corresponding to extent B - btrfs_releasepage() is called and
   returns 1, meaning the page can be released as it's not dirty, not under
   writeback anymore and the extent range is not locked in the inode's
   iotree. However the extent map is not released, either because we are
   not in a context that allows memory allocations to block or because the
   inode's size is smaller than 16MiB - in this case our inode has a size
   of 8KiB;

3) Task B needs to read extent B and ends up __do_readpage() through the
   btrfs_readpage() callback. At __do_readpage() it gets a reference to
   extent map B;

4) Task A, doing a fast fsync, calls clear_em_loggin() against extent map B
   while holding the write lock on the inode's extent map tree - this
   results in try_merge_map() being called and since it's possible to merge
   extent map B with extent map A now (the extent map B was removed from
   the list of modified extents), the merging begins - it sets extent map
   B's start offset to 0 (was 4KiB), but before it increments the map's
   length to 8KiB (4kb + 4KiB), task A is at:

   BUG_ON(extent_map_end(em) <= cur);

   The call to extent_map_end() sees the extent map has a start of 0
   and a length still at 4KiB, so it returns 4KiB and 'cur' is 4KiB, so
   the BUG_ON() is triggered.

So it's dangerous to modify an extent map that is in the tree, because some
other task might have got a reference to it before and still using it, and
needs to see a consistent map while using it. Generally this is very rare
since most paths that lookup and use extent maps also have the file range
locked in the inode's iotree. The fsync path is pretty much the only
exception where we don't do it to avoid serialization with concurrent
reads.

Fix this by not allowing an extent map do be merged if if it's being used
by tasks other then the one attempting to merge the extent map (when the
reference count of the extent map is greater than 2).

Reported-by: ryusuke1925 
Reported-by: Koki Mitani 
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=206211
CC: stable@vger.kernel.org # 4.4+
Reviewed-by: Josef Bacik 
Signed-off-by: Filipe Manana 
Signed-off-by: David Sterba

btrfs: remove extent_map::bdev

2019-11-18T22:43:44+00:00

We can now remove the bdev from extent_map. Previous patches made sure
that bio_set_dev is correctly in all places and that we don't need to
grab it from latest_bdev or pass it around inside the extent map.

Signed-off-by: David Sterba

btrfs: assert extent_map bdevs and lookup_map and split

2019-11-18T11:47:01+00:00

This is a preparatory patch for removing extent_map::bdev. There's some
history behind the code so this is only precaution to catch if things
break before the actual removal happens.

Logically, comparing a raw low-level block device (bdev) does not make
sense for extent maps (high-level objects). This had no effect in
practice but was quite confusing in the code.  The lookup_map is set iff
EXTENT_FLAG_FS_MAPPING is set.

The two pointers were stored in the same bytes and used potentially in
two meanings. Now they're split, so the asserts are in place to check
that the condition will not change.

The lookup map pointer misused bdev, this has been changed in commit
95617d69326c ("btrfs: cleanup, stop casting for extent_map->lookup
everywhere") to the explicit type. But the semantics hasn't changed and
bdev was not actually used to decide if maps are mergeable.

Signed-off-by: David Sterba

btrfs: assert extent map tree lock in add_extent_mapping

2019-09-09T12:59:00+00:00

As add_extent_mapping is called from several functions, let's add the
lock annotation. The tree is going to be modified so it must be the
exclusive lock.

Reviewed-by: Nikolay Borisov 
Signed-off-by: David Sterba

btrfs: Optimize unallocated chunks discard

2019-04-29T17:02:38+00:00

Currently unallocated chunks are always trimmed. For example
2 consecutive trims on large storage would trim freespace twice
irrespective of whether the space was actually allocated or not between
those trims.

Optimise this behavior by exploiting the newly introduced alloc_state
tree of btrfs_device. A new CHUNK_TRIMMED bit is used to mark
those unallocated chunks which have been trimmed and have not been
allocated afterwards. On chunk allocation the respective underlying devices'
physical space will have its CHUNK_TRIMMED flag cleared. This avoids
submitting discards for space which hasn't been changed since the last
time discard was issued.

This applies to the single mount period of the filesystem as the
information is not stored permanently.

Signed-off-by: Nikolay Borisov 
Signed-off-by: David Sterba

btrfs: replace pending/pinned chunks lists with io tree

2019-04-29T17:02:37+00:00

The pending chunks list contains chunks that are allocated in the
current transaction but haven't been created yet. The pinned chunks
list contains chunks that are being released in the current transaction.
Both describe chunks that are not reflected on disk as in use but are
unavailable just the same.

The pending chunks list is anchored by the transaction handle, which
means that we need to hold a reference to a transaction when working
with the list.

The way we use them is by iterating over both lists to perform
comparisons on the stripes they describe for each device. This is
backwards and requires that we keep a transaction handle open while
we're trimming.

This patchset adds an extent_io_tree to btrfs_device that maintains
the allocation state of the device.  Extents are set dirty when
chunks are first allocated -- when the extent maps are added to the
mapping tree. They're cleared when last removed -- when the extent
maps are removed from the mapping tree. This matches the lifespan
of the pending and pinned chunks list and allows us to do trims
on unallocated space safely without pinning the transaction for what
may be a lengthy operation. We can also use this io tree to mark
which chunks have already been trimmed so we don't repeat the operation.

Signed-off-by: Jeff Mahoney 
Signed-off-by: Nikolay Borisov 
Signed-off-by: David Sterba

btrfs: Remove impossible condition from mergable_maps

2019-02-25T13:13:21+00:00

We can never have extents marked as EXTENT_MAP_DELALLOC since this
value is only ever used by btrfs_get_extent_fiemap. In this case the
extent map is created by btrfs_get_extent_fiemap and is never really
published, this flag is used to return the corresponding userspace one.
Considering this, it's pointless having a check for EXTENT_MAP_DELALLOC
in mergable_maps. Just remove it.

Signed-off-by: Nikolay Borisov 
Reviewed-by: David Sterba 
Signed-off-by: David Sterba

btrfs: Fix typos in comments and strings

2018-12-17T13:51:50+00:00

The typos accumulate over time so once in a while time they get fixed in
a large patch.

Signed-off-by: Andrea Gelmini 
Reviewed-by: David Sterba 
Signed-off-by: David Sterba

Btrfs: extent_map: use rb_first_cached

2018-10-15T15:23:33+00:00

rb_first_cached() trades an extra pointer "leftmost" for doing the
same job as rb_first() but in O(1).

As evict_inode_truncate_pages() removes all extent mapping by always
looking for the first rb entry, it's helpful to use rb_first_cached
instead.

For more details about the optimization see patch "Btrfs: delayed-refs:
use rb_first_cached for href_root".

Tested-by: Holger Hoffstätte 
Signed-off-by: Liu Bo 
Reviewed-by: David Sterba 
Signed-off-by: David Sterba

btrfs: change remove_extent_mapping to return void

2018-10-15T15:23:31+00:00

remove_extent_mapping uses the variable "ret" for return value, but it
is not modified after initialzation. Further, I find that any of the
callers do not handle the return value and the callees are only simple
functions so the return values does not need to be passed.

Signed-off-by: zhong jiang 
Reviewed-by: Nikolay Borisov 
Reviewed-by: David Sterba 
Signed-off-by: David Sterba