linux-toradex.git/fs/btrfs, branch v4.4.110

btrfs: add missing memset while reading compressed inline extents

2017-12-20T09:04:57+00:00

[ Upstream commit e1699d2d7bf6e6cce3e1baff19f9dd4595a58664 ]

This is a story about 4 distinct (and very old) btrfs bugs.

Commit c8b978188c ("Btrfs: Add zlib compression support") added
three data corruption bugs for inline extents (bugs #1-3).

Commit 93c82d5750 ("Btrfs: zero page past end of inline file items")
fixed bug #1:  uncompressed inline extents followed by a hole and more
extents could get non-zero data in the hole as they were read.  The fix
was to add a memset in btrfs_get_extent to zero out the hole.

Commit 166ae5a418 ("btrfs: fix inline compressed read err corruption")
fixed bug #2:  compressed inline extents which contained non-zero bytes
might be replaced with zero bytes in some cases.  This patch removed an
unhelpful memset from uncompress_inline, but the case where memset is
required was missed.

There is also a memset in the decompression code, but this only covers
decompressed data that is shorter than the ram_bytes from the extent
ref record.  This memset doesn't cover the region between the end of the
decompressed data and the end of the page.  It has also moved around a
few times over the years, so there's no single patch to refer to.

This patch fixes bug #3:  compressed inline extents followed by a hole
and more extents could get non-zero data in the hole as they were read
(i.e. bug #3 is the same as bug #1, but s/uncompressed/compressed/).
The fix is the same:  zero out the hole in the compressed case too,
by putting a memset back in uncompress_inline, but this time with
correct parameters.

The last and oldest bug, bug #0, is the cause of the offending inline
extent/hole/extent pattern.  Bug #0 is a subtle and mostly-harmless quirk
of behavior somewhere in the btrfs write code.  In a few special cases,
an inline extent and hole are allowed to persist where they normally
would be combined with later extents in the file.

A fast reproducer for bug #0 is presented below.  A few offending extents
are also created in the wild during large rsync transfers with the -S
flag.  A Linux kernel build (git checkout; make allyesconfig; make -j8)
will produce a handful of offending files as well.  Once an offending
file is created, it can present different content to userspace each
time it is read.

Bug #0 is at least 4 and possibly 8 years old.  I verified every vX.Y
kernel back to v3.5 has this behavior.  There are fossil records of this
bug's effects in commits all the way back to v2.6.32.  I have no reason
to believe bug #0 wasn't present at the beginning of btrfs compression
support in v2.6.29, but I can't easily test kernels that old to be sure.

It is not clear whether bug #0 is worth fixing.  A fix would likely
require injecting extra reads into currently write-only paths, and most
of the exceptional cases caused by bug #0 are already handled now.

Whether we like them or not, bug #0's inline extents followed by holes
are part of the btrfs de-facto disk format now, and we need to be able
to read them without data corruption or an infoleak.  So enough about
bug #0, let's get back to bug #3 (this patch).

An example of on-disk structure leading to data corruption found in
the wild:

        item 61 key (606890 INODE_ITEM 0) itemoff 9662 itemsize 160
                inode generation 50 transid 50 size 47424 nbytes 49141
                block group 0 mode 100644 links 1 uid 0 gid 0
                rdev 0 flags 0x0(none)
        item 62 key (606890 INODE_REF 603050) itemoff 9642 itemsize 20
                inode ref index 3 namelen 10 name: DB_File.so
        item 63 key (606890 EXTENT_DATA 0) itemoff 8280 itemsize 1362
                inline extent data size 1341 ram 4085 compress(zlib)
        item 64 key (606890 EXTENT_DATA 4096) itemoff 8227 itemsize 53
                extent data disk byte 5367308288 nr 20480
                extent data offset 0 nr 45056 ram 45056
                extent compression(zlib)

Different data appears in userspace during each read of the 11 bytes
between 4085 and 4096.  The extent in item 63 is not long enough to
fill the first page of the file, so a memset is required to fill the
space between item 63 (ending at 4085) and item 64 (beginning at 4096)
with zero.

Here is a reproducer from Liu Bo, which demonstrates another method
of creating the same inline extent and hole pattern:

Using 'page_poison=on' kernel command line (or enable
CONFIG_PAGE_POISONING) run the following:

	# touch foo
	# chattr +c foo
	# xfs_io -f -c "pwrite -W 0 1000" foo
	# xfs_io -f -c "falloc 4 8188" foo
	# od -x foo
	# echo 3 >/proc/sys/vm/drop_caches
	# od -x foo

This produce the following on my box:

Correct output:  file contains 1000 data bytes followed
by zeros:

	0000000 cdcd cdcd cdcd cdcd cdcd cdcd cdcd cdcd
	*
	0001740 cdcd cdcd cdcd cdcd 0000 0000 0000 0000
	0001760 0000 0000 0000 0000 0000 0000 0000 0000
	*
	0020000

Actual output:  the data after the first 1000 bytes
will be different each run:

	0000000 cdcd cdcd cdcd cdcd cdcd cdcd cdcd cdcd
	*
	0001740 cdcd cdcd cdcd cdcd 6c63 7400 635f 006d
	0001760 5f74 6f43 7400 435f 0053 5f74 7363 7400
	0002000 435f 0056 5f74 6164 7400 645f 0062 5f74
	(...)

Signed-off-by: Zygo Blaxell 
Reviewed-by: Liu Bo 
Reviewed-by: Chris Mason 
Signed-off-by: Chris Mason 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

btrfs: clear space cache inode generation always

2017-12-05T10:22:50+00:00

commit 8e138e0d92c6c9d3d481674fb14e3439b495be37 upstream.

We discovered a box that had double allocations, and suspected the space
cache may be to blame.  While auditing the write out path I noticed that
if we've already setup the space cache we will just carry on.  This
means that any error we hit after cache_save_setup before we go to
actually write the cache out we won't reset the inode generation, so
whatever was already written will be considered correct, except it'll be
stale.  Fix this by _always_ resetting the generation on the block group
inode, this way we only ever have valid or invalid cache.

With this patch I was no longer able to reproduce cache corruption with
dm-log-writes and my bpf error injection tool.

Signed-off-by: Josef Bacik 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

btrfs: return the actual error value from from btrfs_uuid_tree_iterate

2017-11-30T08:37:28+00:00

[ Upstream commit 73ba39ab9307340dc98ec3622891314bbc09cc2e ]

In function btrfs_uuid_tree_iterate(), errno is assigned to variable ret
on errors. However, it directly returns 0. It may be better to return
ret. This patch also removes the warning, because the caller already
prints a warning.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=188731
Signed-off-by: Pan Bian 
Reviewed-by: Omar Sandoval 
[ edited subject ]
Signed-off-by: David Sterba 

Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

Btrfs: send, fix failure to rename top level inode due to name collision

2017-10-21T15:09:04+00:00

[ Upstream commit 4dd9920d991745c4a16f53a8f615f706fbe4b3f7 ]

Under certain situations, an incremental send operation can fail due to a
premature attempt to create a new top level inode (a direct child of the
subvolume/snapshot root) whose name collides with another inode that was
removed from the send snapshot.

Consider the following example scenario.

Parent snapshot:

  .                 (ino 256, gen 8)
  |---- a1/         (ino 257, gen 9)
  |---- a2/         (ino 258, gen 9)

Send snapshot:

  .                 (ino 256, gen 3)
  |---- a2/         (ino 257, gen 7)

In this scenario, when receiving the incremental send stream, the btrfs
receive command fails like this (ran in verbose mode, -vv argument):

  rmdir a1
  mkfile o257-7-0
  rename o257-7-0 -> a2
  ERROR: rename o257-7-0 -> a2 failed: Is a directory

What happens when computing the incremental send stream is:

1) An operation to remove the directory with inode number 257 and
   generation 9 is issued.

2) An operation to create the inode with number 257 and generation 7 is
   issued. This creates the inode with an orphanized name of "o257-7-0".

3) An operation rename the new inode 257 to its final name, "a2", is
   issued. This is incorrect because inode 258, which has the same name
   and it's a child of the same parent (root inode 256), was not yet
   processed and therefore no rmdir operation for it was yet issued.
   The rename operation is issued because we fail to detect that the
   name of the new inode 257 collides with inode 258, because their
   parent, a subvolume/snapshot root (inode 256) has a different
   generation in both snapshots.

So fix this by ignoring the generation value of a parent directory that
matches a root inode (number 256) when we are checking if the name of the
inode currently being processed collides with the name of some other
inode that was not yet processed.

We can achieve this scenario of different inodes with the same number but
different generation values either by mounting a filesystem with the inode
cache option (-o inode_cache) or by creating and sending snapshots across
different filesystems, like in the following example:

  $ mkfs.btrfs -f /dev/sdb
  $ mount /dev/sdb /mnt
  $ mkdir /mnt/a1
  $ mkdir /mnt/a2
  $ btrfs subvolume snapshot -r /mnt /mnt/snap1
  $ btrfs send /mnt/snap1 -f /tmp/1.snap
  $ umount /mnt

  $ mkfs.btrfs -f /dev/sdc
  $ mount /dev/sdc /mnt
  $ touch /mnt/a2
  $ btrfs subvolume snapshot -r /mnt /mnt/snap2
  $ btrfs receive /mnt -f /tmp/1.snap
  # Take note that once the filesystem is created, its current
  # generation has value 7 so the inode from the second snapshot has
  # a generation value of 7. And after receiving the first snapshot
  # the filesystem is at a generation value of 10, because the call to
  # create the second snapshot bumps the generation to 8 (the snapshot
  # creation ioctl does a transaction commit), the receive command calls
  # the snapshot creation ioctl to create the first snapshot, which bumps
  # the filesystem's generation to 9, and finally when the receive
  # operation finishes it calls an ioctl to transition the first snapshot
  # (snap1) from RW mode to RO mode, which does another transaction commit
  # and bumps the filesystem's generation to 10.
  $ rm -f /tmp/1.snap
  $ btrfs send /mnt/snap1 -f /tmp/1.snap
  $ btrfs send -p /mnt/snap1 /mnt/snap2 -f /tmp/2.snap
  $ umount /mnt

  $ mkfs.btrfs -f /dev/sdd
  $ mount /dev/sdd /mnt
  $ btrfs receive /mnt /tmp/1.snap
  # Receive of snapshot snap2 used to fail.
  $ btrfs receive /mnt /tmp/2.snap

Signed-off-by: Robbie Ko 
Reviewed-by: Filipe Manana 
[Rewrote changelog to be more precise and clear]
Signed-off-by: Filipe Manana 

Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

btrfs: prevent to set invalid default subvolid

2017-10-05T07:41:47+00:00

commit 6d6d282932d1a609e60dc4467677e0e863682f57 upstream.

`btrfs sub set-default` succeeds to set an ID which isn't corresponding to any
fs/file tree. If such the bad ID is set to a filesystem, we can't mount this
filesystem without specifying `subvol` or `subvolid` mount options.

Fixes: 6ef5ed0d386b ("Btrfs: add ioctl and incompat flag to set the default mount subvol")
Signed-off-by: Satoru Takeuchi 
Reviewed-by: Qu Wenruo 
Reviewed-by: David Sterba 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

btrfs: propagate error to btrfs_cmp_data_prepare caller

2017-10-05T07:41:47+00:00

commit 78ad4ce014d025f41b8dde3a81876832ead643cf upstream.

btrfs_cmp_data_prepare() (almost) always returns 0 i.e. ignoring errors
from gather_extent_pages(). While the pages are freed by
btrfs_cmp_data_free(), cmp->num_pages still has > 0. Then,
btrfs_extent_same() try to access the already freed pages causing faults
(or violates PageLocked assertion).

This patch just return the error as is so that the caller stop the process.

Signed-off-by: Naohiro Aota 
Fixes: f441460202cb ("btrfs: fix deadlock with extent-same and readpage")
Reviewed-by: David Sterba 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

btrfs: fix NULL pointer dereference from free_reloc_roots()

2017-10-05T07:41:46+00:00

commit bb166d7207432d3c7d10c45dc052f12ba3a2121d upstream.

__del_reloc_root should be called before freeing up reloc_root->node.
If not, calling __del_reloc_root() dereference reloc_root->node, causing
the system BUG.

Fixes: 6bdf131fac23 ("Btrfs: don't leak reloc root nodes on error")
Signed-off-by: Naohiro Aota 
Reviewed-by: Nikolay Borisov 
Reviewed-by: David Sterba 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

btrfs: resume qgroup rescan on rw remount

2017-09-13T21:09:46+00:00

commit 6c6b5a39c4bf3dbd8cf629c9f5450e983c19dbb9 upstream.

Several distributions mount the "proper root" as ro during initrd and
then remount it as rw before pivot_root(2). Thus, if a rescan had been
aborted by a previous shutdown, the rescan would never be resumed.

This issue would manifest itself as several btrfs ioctl(2)s causing the
entire machine to hang when btrfs_qgroup_wait_for_completion was hit
(due to the fs_info->qgroup_rescan_running flag being set but the rescan
itself not being resumed). Notably, Docker's btrfs storage driver makes
regular use of BTRFS_QUOTA_CTL_DISABLE and BTRFS_IOC_QUOTA_RESCAN_WAIT
(causing this problem to be manifested on boot for some machines).

Cc: Jeff Mahoney 
Fixes: b382a324b60f ("Btrfs: fix qgroup rescan resume on mount")
Signed-off-by: Aleksa Sarai 
Reviewed-by: Nikolay Borisov 
Tested-by: Nikolay Borisov 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

btrfs: remove duplicate const specifier

2017-09-02T05:06:50+00:00

commit fb75d857a31d600cc0c37b8c7d914014f7fa3f9a upstream.

duplicate const is redundant so remove it

Signed-off-by: Colin Ian King 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

Btrfs: adjust outstanding_extents counter properly when dio write is split

2017-08-07T02:19:46+00:00

[ Upstream commit c2931667c83ded6504b3857e99cc45b21fa496fb ]

Currently how btrfs dio deals with split dio write is not good
enough if dio write is split into several segments due to the
lack of contiguous space, a large dio write like 'dd bs=1G count=1'
can end up with incorrect outstanding_extents counter and endio
would complain loudly with an assertion.

This fixes the problem by compensating the outstanding_extents
counter in inode if a large dio write gets split.

Reported-by: Anand Jain 
Tested-by: Anand Jain 
Signed-off-by: Liu Bo 
Signed-off-by: David Sterba 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman