Yet another cleanup patch broke code for which no xfstest exists.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Instead of doing another extent tree search if the first search failed
to find a metadata item, check if the previous item in the leaf is an
extent item and use it if it is, otherwise do the second tree search
for an extent item.
Signed-off-by: NFilipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

When debugging ENOSPC issues, it's nice to be able to see which
reservations failed as well as the ones which succeeded.
Signed-off-by: NJeff Mahoney <jeffm@suse.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

fs/btrfs/compat.h only contained trivial macro wrappers of drop_nlink()
and inc_nlink().  This doesn't belong in mainline.
Signed-off-by: NZach Brown <zab@redhat.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

move_pages() has an inefficient backwards byte copy of regions of two
different pages.  They're different pages so the regions won't overlap
and it could use memcpy().

At that point, though, move_pages() would be a slightly dimmer
re-implementation of copy_pages() that lacked the test for overlapping
page regions.

So remove move_pages() and just call copy_pages().
Signed-off-by: NZach Brown <zab@redhat.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Signed-off-by: NZach Brown <zab@redhat.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

When a directory has a default ACL and a subdirectory is created
under that directory, btrfs_init_acl() is called when the
subdirectory's inode is created to initialize the inode's ACL
(inherited from the parent directory) but it was clearing the ACL
from the inode after setting it if posix_acl_create() returned
success, instead of clearing it only if it returned an error.

To reproduce this issue:

$ mkfs.btrfs -f /dev/loop0
$ mount /dev/loop0 /mnt
$ mkdir /mnt/acl
$ setfacl -d --set u::rwx,g::rwx,o::- /mnt/acl
$ getfacl /mnt/acl
user::rwx
group::rwx
other::r-x
default:user::rwx
default:group::rwx
default:other::---

$ mkdir /mnt/acl/dir1
$ getfacl /mnt/acl/dir1
user::rwx
group::rwx
other::---

After unmounting and mounting again the filesystem, fgetacl returned the
expected ACL:

$ umount /mnt/acl
$ mount /dev/loop0 /mnt
$ getfacl /mnt/acl/dir1
user::rwx
group::rwx
other::---
default:user::rwx
default:group::rwx
default:other::---

Meaning that the underlying xattr was persisted.
Reported-by: NGiuseppe Fierro <giuseppe@fierro.org>
Signed-off-by: NFilipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Due to an off-by-one error, it is possible to reproduce a bug
when the inode cache is used.

The same inode number is assigned twice, the second time this
leads to an EEXIST in btrfs_insert_empty_items().

The issue can happen when a file is removed right after a subvolume
is created and then a new inode number is created before the
inodes in free_inode_pinned are processed.
unlink() calls btrfs_return_ino() which calls start_caching() in this
case which adds [highest_ino + 1, BTRFS_LAST_FREE_OBJECTID] by
searching for the highest inode (which already cannot find the
unlinked one anymore in btrfs_find_free_objectid()). So if this
unlinked inode's number is equal to the highest_ino + 1 (or >= this value
instead of > this value which was the off-by-one error), we mustn't add
the inode number to free_ino_pinned (caching_thread() does it right).
In this case we need to try directly to add the number to the inode_cache
which will fail in this case.

When this inode number is allocated while it is still in free_ino_pinned,
it is allocated and still added to the free inode cache when the
pinned inodes are processed, thus one of the following inode number
allocations will get an inode that is already in use and fail with EEXIST
in btrfs_insert_empty_items().

One example which was created with the reproducer below:
Create a snapshot, work in the newly created snapshot for the rest.
In unlink(inode 34284) call btrfs_return_ino() which calls start_caching().
start_caching() calls add_free_space [34284, 18446744073709517077].
In btrfs_return_ino(), call start_caching pinned [34284, 1] which is wrong.
mkdir() call btrfs_find_ino_for_alloc() which returns the number 34284.
btrfs_unpin_free_ino calls add_free_space [34284, 1].
mkdir() call btrfs_find_ino_for_alloc() which returns the number 34284.
EEXIST when the new inode is inserted.

One possible reproducer is this one:
 #!/bin/sh
 # preparation
TEST_DEV=/dev/sdc1
TEST_MNT=/mnt
umount ${TEST_MNT} 2>/dev/null || true
mkfs.btrfs -f ${TEST_DEV}
mount ${TEST_DEV} ${TEST_MNT} -o \
 rw,relatime,compress=lzo,space_cache,inode_cache
btrfs subv create ${TEST_MNT}/s1
for i in `seq 34027`; do touch ${TEST_MNT}/s1/${i}; done
btrfs subv snap ${TEST_MNT}/s1 ${TEST_MNT}/s2
FILENAME=`find ${TEST_MNT}/s1/ -inum 4085 | sed 's|^.*/\([^/]*\)$|\1|'`
rm ${TEST_MNT}/s2/$FILENAME
touch ${TEST_MNT}/s2/$FILENAME
 # the following steps can be repeated to reproduce the issue again and again
[ -e ${TEST_MNT}/s3 ] && btrfs subv del ${TEST_MNT}/s3
btrfs subv snap ${TEST_MNT}/s2 ${TEST_MNT}/s3
rm ${TEST_MNT}/s3/$FILENAME
touch ${TEST_MNT}/s3/$FILENAME
ls -alFi ${TEST_MNT}/s?/$FILENAME
touch ${TEST_MNT}/s3/_1 || logger FAILED
ls -alFi ${TEST_MNT}/s?/_1
touch ${TEST_MNT}/s3/_2 || logger FAILED
ls -alFi ${TEST_MNT}/s?/_2
touch ${TEST_MNT}/s3/__1 || logger FAILED
ls -alFi ${TEST_MNT}/s?/__1
touch ${TEST_MNT}/s3/__2 || logger FAILED
ls -alFi ${TEST_MNT}/s?/__2
 # if the above is not enough, add the following loop:
for i in `seq 3 9`; do touch ${TEST_MNT}/s3/__${i} || logger FAILED; done
 #for i in `seq 3 34027`; do touch ${TEST_MNT}/s3/__${i} || logger FAILED; done
 # one of the touch(1) calls in s3 fail due to EEXIST because the inode is
 # already in use that btrfs_find_ino_for_alloc() returns.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Reviewed-by: NJan Schmidt <list.btrfs@jan-o-sch.net>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

If we decrement the key type, we must reset its offset to the largest
possible offset (u64)-1. If we decrement the key's objectid, then we
must reset the key's type and offset to their largest possible values,
(u8)-1 and (u64)-1 respectively. Not doing so can make us miss an
items in the tree.
Signed-off-by: NFilipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Avoid repeated tree searches by processing all inode ref items in
a leaf at once instead of processing one at a time, followed by a
path release and a tree search for a key with a decremented offset.
Signed-off-by: NFilipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Use memdup_user rather than duplicating its implementation
This is a little bit restricted to reduce false positives

The semantic patch that makes this report is available
in scripts/coccinelle/api/memdup_user.cocci.

More information about semantic patching is available at
http://coccinelle.lip6.fr/Signed-off-by: NGeyslan G. Bem <geyslan@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Comparison of an inode's last modified transaction with the last committed
transaction is incorrect. Fix it.
Signed-off-by: Nchandan <chandan@linux.vnet.ibm.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

If the path allocation failed, we would return without decrementing
the reference count in the delayed node we got before, resulting
in a leak.
Signed-off-by: NFilipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

If the user remounts the filesystem read-only while the uuid-tree
scan and rebuild task is still running (this happens once after the
filesystem was mounted with an old kernel, or when forced with the
mount options), the remount should wait on the tasks completion
before setting the filesystem read-only. Otherwise the background
task continues to write to the filesystem which is apparently not
what users expect.

The reproducer:

TEST_DEV=/dev/sdzzzzz1
TEST_MNT=/mnt
mkfs.btrfs -f $TEST_DEV
mount $TEST_DEV $TEST_MNT
for i in `seq 50000`; do btrfs subvolume create ${TEST_MNT}/$i; done
umount $TEST_MNT
mount $TEST_DEV $TEST_MNT -o rescan_uuid_tree
sleep 1
ps -elf | fgrep '[btrfs-uuid]' | grep -v grep
mount $TEST_DEV $TEST_MNT -o ro,remount
ps -elf | fgrep '[btrfs-uuid]' | grep -v grep
sleep 1
umount $TEST_MNT
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Device stats are only initialized (read from tree items) on mount.
Trying to read device stats after adding or replacing new devices will
return errors.

btrfs_init_new_device() and btrfs_init_dev_replace_tgtdev() are the two
functions that allocate and initialize new btrfs_device structures after
a filesystem is mounted. They set the device stats to zero by using
kzalloc() which is correct for new devices. The only missing thing was
to declare these stats as being valid (device->dev_stats_valid = 1) and
this patch adds this missing code.

This is the reproducer:

TEST_DEV1=/dev/sdzzzzz1
TEST_DEV2=/dev/sdzzzzz2
TEST_DEV3=/dev/sdzzzzz3
TEST_MNT=/mnt
mkfs.btrfs $TEST_DEV1
mount $TEST_DEV1 $TEST_MNT
btrfs device add $TEST_DEV2 $TEST_MNT
btrfs device stat $TEST_MNT
btrfs replace start -B $TEST_DEV2 $TEST_DEV3 $TEST_MNT
btrfs device stat $TEST_MNT
umount $TEST_MNT
Reported-by: NOndrej Kunc <kunc88@gmail.com>
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

When we fail to add a reference after a non-inline insertion by some reasons,
eg. ENOSPC, we'll abort the transaction, but we don't return this error to
the caller who has to walk around again to find something wrong, that's
unnecessary.

Also fixup other error paths to keep it simple.
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

For both balance and replace, cancelling involves changing the on-disk
state and committing a transaction, which is not a good thing to do on
read-only filesystems.

Cc: Stefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NIlya Dryomov <idryomov@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

struct btrfs_ioctl_dev_replace_args memory is leaked if replace is
requested on a read-only filesystem.  Fix it.
Signed-off-by: NIlya Dryomov <idryomov@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

On mount failures, __btrfs_close_devices can be called well before
dev-replace state is read and ->is_tgtdev_for_dev_replace is set.  This
leads to a bogus decrement of ->rw_devices and sets off a WARN_ON in
__btrfs_close_devices if replace target device happens to be on the
lists and we fail early in the mount sequence.  Fix this by checking
the devid instead of ->is_tgtdev_for_dev_replace before the decrement:
for replace targets devid is always equal to BTRFS_DEV_REPLACE_DEVID.

Cc: Stefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NIlya Dryomov <idryomov@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

In add_inode_ref() function:

Initializes local pointers.

Reduces the logical condition with the __add_inode_ref() return
value by using only one 'goto out'.

Centralizes the exiting, ensuring the freeing of all used memory.
Signed-off-by: NGeyslan G. Bem <geyslan@gmail.com>
Reviewed-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

After commit de78b51a
(btrfs: remove cache only arguments from defrag path), @blockptr is no more
used.
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

@is_extent is no more needed since we don't defrag extent root.
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

The btrfs_insert_empty_item() function doesn't modify its
key argument.
Signed-off-by: NFilipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: NZach Brown <zab@redhat.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

alloc_extent_buffer() uses radix_tree_lookup() when radix_tree_insert()
fails with EEXIST. That part of the code is very similar to the code in
find_extent_buffer(). This patch replaces radix_tree_lookup() and
surrounding code in alloc_extent_buffer() with find_extent_buffer().

Note that radix_tree_lookup() does not need to be protected by
tree->buffer_lock. It is protected by eb->refs.

While at it, this patch
  - changes the other usage of radix_tree_lookup() in alloc_extent_buffer()
    with find_extent_buffer() to reduce redundancy.
  - removes the unused argument 'len' to find_extent_buffer().
Signed-Off-by: NChandra Seetharaman <sekharan@us.ibm.com>
Reviewed-by: NZach Brown <zab@redhat.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Whoever wrote this was braindead.  Also it doesn't work right if you have
VACANCY's since we assumed you would only have that at the end of the file,
which won't be the case in the near future.  I tested this with generic/285 and
generic/286 as well as the btrfs tests that use fssum since it uses
seek_hole/seek_data to verify things are ok.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

I was hitting weird issues when trying to remove hole extents and it turned out
it was because I was sending non-aligned offsets down to
btrfs_lookup_csums_range.  So add an assert for this in case somebody trips over
this in the future.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

I added an assert to make sure we were looking up aligned offsets for csums and
I tripped it when running xfstests.  This is because log_one_extent was checking
if block_start == 0 for a hole instead of EXTENT_MAP_HOLE.  This worked out fine
in practice it seems, but it adds a lot of extra work that is uneeded.  With
this fix I'm no longer tripping my assert.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Btrfs_get_extent was not handling this case properly, add a test to make sure we
don't regress.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

While trying to kill our hole extents I noticed I was seeing problems where we
seek into a file and then start writing and then try to fiemap that file later.
This is because we search for offset 0, don't find anything and so back up one
slot, which puts us at the inode ref or something like that, which means we goto
not_found and create an extent map for our entire search area.  This isn't quite
what we want, we want to move forward one slot and see if there is an extent
there so we can limit our hole extent.  This patch fixes this problem, I will
add a testcase for this as well.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Stefan was hitting a panic in the async worker stuff because we had outstanding
read bios while we were stopping the worker threads.  You could reproduce this
easily if you mount -o nospace_cache and ran generic/273.  This is because the
caching thread stuff is still going and we were stopping all the worker threads.
We need to stop the workers after this work is done, and the free block groups
code will wait for all the caching threads to stop first so we don't run into
this problem.  With this patch we no longer panic.  Thanks,

Cc: stable@vger.kernel.org
Reported-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

I'm going to be removing hole extents in the near future so I wanted to make a
sanity test for btrfs_get_extent to make sure I don't break anything in the
meantime.  This patch just puts btrfs_get_extent through its paces by giving it
a completely unreasonable mapping to look at and make sure it is giving us back
maps that make sense.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

So both Liu and I made huge messes of find_lock_delalloc_range trying to fix
stuff, me first by fixing extent size, then him by fixing something I broke and
then me again telling him to fix it a different way.  So this is obviously a
candidate for some testing.  This patch adds a pseudo fs so we can allocate fake
inodes for tests that need an inode or pages.  Then it addes a bunch of tests to
make sure find_lock_delalloc_range is acting the way it is supposed to.  With
this patch and all of our previous patches to find_lock_delalloc_range I am sure
it is working as expected now.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

While trying to track down a reserved space leak I noticed a few places where we
won't properly clean up reserved space if we have an error, this patch fixes
those up.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

In trying to track down where we were leaking reserved space I noticed our
reserve extent tracepoints are a little off.  First we were saying that the
reserved space had been alloced in btrfs_reserve_extent, which isn't the case,
this needs to be triggered when we actually allocate the space when we run the
delayed ref.  We were also missing a few places where we should have been
tracing the btrfs_reserve_extent_free tracepoint.  With these in place I was
able to put together where we were leaking reserved space.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

If we abort a transaction we will do the tree log cleanup at unmount, but this
happens after we free up the block groups.  This makes all the leak detection
warnings go off because we think we've leaked space but in reality we just
haven't cleaned it up yet.  So instead do the block group cleanup stuff after
free'ing the fs roots so we don't get these warnings.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

On error we will wait and free the tree log at unmount without a transaction.
This means that the actual freeing of the blocks doesn't happen which means we
complain about space leaks on unmount.  So to fix this just skip the transaction
specific cleanup part of the tree log free'ing if we don't have a transaction
and that way we can free up our reserved space and our counters stay happy.
Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

The transactions should be cleaning up their reservations on failure, this just
causes us to have warnings on unmount because we go negative by free'ing
reservations that have already been free'ed.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Structures of the types tree_mod_elem and qgroup_update are allocated
during transaction commit but were not being released if the call to
btrfs_run_delayed_items() returned an error.

Stack trace reported by kmemleak:

unreferenced object 0xffff880679f0b398 (size 128):
  comm "umount", pid 21508, jiffies 4295967793 (age 36718.112s)
  hex dump (first 32 bytes):
    60 b5 f0 79 06 88 ff ff 00 00 00 00 00 00 00 00  `..y............
    00 00 00 00 00 00 00 00 50 1c 00 00 00 00 00 00  ........P.......
  backtrace:
    [<ffffffff81742d26>] kmemleak_alloc+0x26/0x50
    [<ffffffff811889c2>] kmem_cache_alloc_trace+0x112/0x200
    [<ffffffffa046f2d3>] tree_mod_log_insert_key.constprop.45+0x93/0x150 [btrfs]
    [<ffffffffa04720f9>] __btrfs_cow_block+0x299/0x4f0 [btrfs]
    [<ffffffffa0472510>] btrfs_cow_block+0x120/0x1f0 [btrfs]
    [<ffffffffa0476679>] btrfs_search_slot+0x449/0x930 [btrfs]
    [<ffffffffa048eecf>] btrfs_lookup_inode+0x2f/0xa0 [btrfs]
    [<ffffffffa04eb49c>] __btrfs_update_delayed_inode+0x1c/0x1d0 [btrfs]
    [<ffffffffa04eb9e2>] __btrfs_run_delayed_items+0x162/0x1e0 [btrfs]
    [<ffffffffa04eba63>] btrfs_delayed_inode_exit+0x3/0x20 [btrfs]
    [<ffffffffa0499c63>] btrfs_commit_transaction+0x203/0xa50 [btrfs]
    [<ffffffffa046b519>] btrfs_sync_fs+0x69/0x110 [btrfs]
    [<ffffffff811cb210>] __sync_filesystem+0x30/0x60
    [<ffffffff811cb2bb>] sync_filesystem+0x4b/0x70
    [<ffffffff8119ce7b>] generic_shutdown_super+0x3b/0xf0
    [<ffffffff8119cfc6>] kill_anon_super+0x16/0x30
unreferenced object 0xffff880677e0dd88 (size 32):
  comm "umount", pid 21508, jiffies 4295967793 (age 36718.112s)
  hex dump (first 32 bytes):
    78 75 11 a9 06 88 ff ff 00 c0 e0 77 06 88 ff ff  xu.........w....
    40 c3 a2 70 06 88 ff ff 00 00 00 00 00 00 00 00  @..p............
  backtrace:
    [<ffffffff81742d26>] kmemleak_alloc+0x26/0x50
    [<ffffffff811889c2>] kmem_cache_alloc_trace+0x112/0x200
    [<ffffffffa04fa54f>] btrfs_qgroup_record_ref+0xf/0x90 [btrfs]
    [<ffffffffa04e1914>] btrfs_add_delayed_tree_ref+0xf4/0x170 [btrfs]
    [<ffffffffa048518a>] btrfs_free_tree_block+0x9a/0x220 [btrfs]
    [<ffffffffa0472163>] __btrfs_cow_block+0x303/0x4f0 [btrfs]
    [<ffffffffa0472510>] btrfs_cow_block+0x120/0x1f0 [btrfs]
    [<ffffffffa0476679>] btrfs_search_slot+0x449/0x930 [btrfs]
    [<ffffffffa048eecf>] btrfs_lookup_inode+0x2f/0xa0 [btrfs]
    [<ffffffffa04eb49c>] __btrfs_update_delayed_inode+0x1c/0x1d0 [btrfs]
    [<ffffffffa04eb9e2>] __btrfs_run_delayed_items+0x162/0x1e0 [btrfs]
    [<ffffffffa04eba63>] btrfs_delayed_inode_exit+0x3/0x20 [btrfs]
    [<ffffffffa0499c63>] btrfs_commit_transaction+0x203/0xa50 [btrfs]
    [<ffffffffa046b519>] btrfs_sync_fs+0x69/0x110 [btrfs]
    [<ffffffff811cb210>] __sync_filesystem+0x30/0x60
    [<ffffffff811cb2bb>] sync_filesystem+0x4b/0x70
Signed-off-by: NFilipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Replace progresses strictly from lower to higher offsets, and the
progress is tracked in chunks, by storing the physical offset of the
dev_extent which is being copied in the cursor_left field of
btrfs_dev_replace_item.  When we are done copying the chunk,
left_cursor is updated to point one byte past the dev_extent, so that
on resume we can skip the dev_extents that have already been copied.

There is a major bug (which goes all the way back to the inception of
dev-replace in 3.8) in the way left_cursor is bumped: the bump is done
unconditionally, without any regard to the scrub_chunk return value.
On suspend (and also on any kind of error) scrub_chunk returns early,
i.e. without completing the copy.  This leads to us skipping the chunk
that hasn't been fully copied yet when resuming.

Fix this by doing the cursor_left update only if scrub_chunk ret is 0.
(On suspend scrub_chunk returns with -ECANCELED, so this fix covers
both suspend and error cases.)

Cc: Stefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NIlya Dryomov <idryomov@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Currently the hash value used for adding an inode to the VFS's inode
hash table consists of the plain inode number, which is a 64 bits
integer. This results in hash table buckets (hlist_head lists) with
too many elements for at least 2 important scenarios:

1) When we have many subvolumes. Each subvolume has its own btree
   where its files and directories are added to, and each has its
   own objectid (inode number) namespace. This means that if we have
   N subvolumes, and all have inode number X associated to a file or
   directory, the corresponding inodes all map to the same hash table
   entry, resulting in a bucket (hlist_head list) with N elements;

2) On 32 bits machines. Th VFS hash values are unsigned longs, which
   are 32 bits wide on 32 bits machines, and the inode (objectid)
   numbers are 64 bits unsigned integers. We simply cast the inode
   numbers to hash values, which means that for all inodes with the
   same 32 bits lower half, the same hash bucket is used for all of
   them. For example, all inodes with a number (objectid) between
   0x0000_0000_ffff_ffff and 0xffff_ffff_ffff_ffff will end up in
   the same hash table bucket.

This change ensures the inode's hash value depends both on the
objectid (inode number) and its subvolume's (btree root) objectid.
For 32 bits machines, this change gives better entropy by making
the hash value depend on both the upper and lower 32 bits of the
64 bits hash previously computed.
Signed-off-by: NFilipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>