The problem is:
	Task0(device scan task)		Task1(device replace task)
	scan_one_device()
	mutex_lock(&uuid_mutex)
	device = find_device()
					mutex_lock(&device_list_mutex)
					lock_chunk()
					rm_and_free_source_device
					unlock_chunk()
					mutex_unlock(&device_list_mutex)
	check device

Destroying the target device if device replace fails also has the same problem.

We fix this problem by locking uuid_mutex during destroying source device or
target device, just like the device remove operation.

It is a temporary solution, we can fix this problem and make the code more
clear by atomic counter in the future.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

->total_bytes,->disk_total_bytes,->bytes_used is protected by chunk
lock when we change them, but sometimes we read them without any lock,
and we might get unexpected value. We fix this problem like inode's
i_size.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

device->bytes_used will be changed when allocating a new chunk, and
disk_total_size will be changed if resizing is successful.
Meanwhile, the on-disk super blocks of the previous transaction
might not be updated. Considering the consistency of the metadata
in the previous transaction, We should use the size in the previous
transaction to check if the super block is beyond the boundary
of the device.

Though it is not big problem because we don't use it now, but anyway
it is better that we make it be consistent with the common metadata,
maybe we will use it in the future.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

total_size will be changed when resizing a device, and disk_total_size
will be changed if resizing is successful. Meanwhile, the on-disk super
blocks of the previous transaction might not be updated. Considering
the consistency of the metadata in the previous transaction, We should
use the size in the previous transaction to check if the super block is
beyond the boundary of the device. Fix it.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

We didn't protect the assignment of the target device, it might cause the
problem that the super block update was skipped because we might find wrong
size of the target device during the assignment. Fix it by moving the
assignment sentences into the initialization function of the target device.
And there is another merit that we can check if the target device is suitable
more early.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

The member variants - num_can_discard - of fs_devices structure
are set, but no one use them to do anything. so remove them.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

The member variants - latest_devid and latest_trans - of fs_devices structure
are set, but no one use them to do anything. so remove them.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

The io error might happen during writing out the device stats, and the
device stats information and dirty flag would be update at that time,
but the current code didn't consider this case, just clear the dirty
flag, it would cause that we forgot to write out the new device stats
information. Fix it.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

The lock in btrfs_device structure was far away from its protected data, it would
make CPU load the cache line twice when we accessed them, move them together.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <clm@fb.com>

The deadlock happened when we mount degraded filesystem, the reproduced
steps are following:
 # mkfs.btrfs -f -m raid1 -d raid1 <dev0> <dev1>
 # echo 1 > /sys/block/`basename <dev0>`/device/delete
 # mount -o degraded <dev1> <mnt>

The reason was that the counter -- bi_remaining was wrong. If the missing
or unwriteable device was the last device in the mapping array, we would
not submit the original bio, so we shouldn't increase bi_remaining of it
in btrfs_end_bio(), or we would skip the final endio handle.

Fix this problem by adding a flag into btrfs bio structure. If we submit
the original bio, we will set the flag, and we increase bi_remaining counter,
or we don't.

Though there is another way to fix it -- decrease bi_remaining counter of the
original bio when we make sure the original bio is not submitted, this method
need add more check and is easy to make mistake.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Reviewed-by: NLiu Bo <bo.li.liu@oracle.com>
Signed-off-by: NChris Mason <clm@fb.com>

This started as debugging helper, to watch the effects of converting
between raid levels on multiple devices, but could be useful standalone.

In my case the usage filter was not finegrained enough and led to
converting too many chunks at once. Another example use is in connection
with drange+devid or vrange filters that allow to work with a specific
chunk or even with a chunk on a given device.

The limit filter applies last, the value of 0 means no limiting.

CC: Ilya Dryomov <idryomov@gmail.com>
CC: Hugo Mills <hugo@carfax.org.uk>
Signed-off-by: NDavid Sterba <dsterba@suse.cz>
Signed-off-by: NChris Mason <clm@fb.com>

Since the "_struct" suffix is mainly used for distinguish the differnt
btrfs_work between the original and the newly created one,
there is no need using the suffix since all btrfs_workers are changed
into btrfs_workqueue.

Also this patch fixed some codes whose code style is changed due to the
too long "_struct" suffix.
Signed-off-by: NQu Wenruo <quwenruo@cn.fujitsu.com>
Tested-by: NDavid Sterba <dsterba@suse.cz>
Signed-off-by: NJosef Bacik <jbacik@fb.com>

Much like the fs_info->workers, replace the fs_info->submit_workers
use the same btrfs_workqueue.
Signed-off-by: NQu Wenruo <quwenruo@cn.fujitsu.com>
Tested-by: NDavid Sterba <dsterba@suse.cz>
Signed-off-by: NJosef Bacik <jbacik@fb.com>

During device replace test, we hit a null pointer deference (It was very easy
to reproduce it by running xfstests' btrfs/011 on the devices with the virtio
scsi driver). There were two bugs that caused this problem:
- We might allocate new chunks on the replaced device after we updated
  the mapping tree. And we forgot to replace the source device in those
  mapping of the new chunks.
- We might get the mapping information which including the source device
  before the mapping information update. And then submit the bio which was
  based on that mapping information after we freed the source device.

For the first bug, we can fix it by doing mapping tree update and source
device remove in the same context of the chunk mutex. The chunk mutex is
used to protect the allocable device list, the above method can avoid
the new chunk allocation, and after we remove the source device, all
the new chunks will be allocated on the new device. So it can fix
the first bug.

For the second bug, we need make sure all flighting bios are finished and
no new bios are produced during we are removing the source device. To fix
this problem, we introduced a global @bio_counter, we not only inc/dec
@bio_counter outsize of map_blocks, but also inc it before submitting bio
and dec @bio_counter when ending bios.

Since Raid56 is a little different and device replace dosen't support raid56
yet, it is not addressed in the patch and I add comments to make sure we will
fix it in the future.
Reported-by: NQu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: NWang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NJosef Bacik <jbacik@fb.com>

Pack the structure btrfs_device in volumes.h to eliminate holes detected
by pahole, thus reducing binary memory footprint.
Signed-off-by: NDulshani Gunawardhana <dulshani.gunawardhana89@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>

Originally, we introduced scrub_super_lock to synchronize
tree log code with scrubbing super.

However we can replace scrub_super_lock with device_list_mutex,
because writing super will hold this mutex, this will reduce an extra
lock holding when writing supers in sync log code.
Signed-off-by: NWang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Currently btrfs_device is allocated ad-hoc in a few different places,
and as a result not all fields are initialized properly.  In particular,
readahead state is only initialized in device_list_add (at scan time),
and not in btrfs_init_new_device (when the new device is added with
'btrfs dev add').  Fix this by adding an allocation helper and switch
everybody but __btrfs_close_devices to it.  (__btrfs_close_devices is
dealt with in a later commit.)
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>

If the filesystem was mounted with an old kernel that was not
aware of the UUID tree, this is detected by looking at the
uuid_tree_generation field of the superblock (similar to how
the free space cache is doing it). If a mismatch is detected
at mount time, a thread is started that does two things:
1. Iterate through the UUID tree, check each entry, delete those
   entries that are not valid anymore (i.e., the subvol does not
   exist anymore or the value changed).
2. Iterate through the root tree, for each found subvolume, add
   the UUID tree entries for the subvolume (if they are not
   already there).

This mechanism is also used to handle and repair errors that
happened during the initial creation and filling of the tree.
The update of the uuid_tree_generation field (which indicates
that the state of the UUID tree is up to date) is blocked until
all create and repair operations are successfully completed.
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>

This tree is not created by mkfs.btrfs. Therefore when a filesystem
is mounted writable and the UUID tree does not exist, this tree is
created if required. The tree is also added to the fs_info structure
and initialized, but this commit does not yet read or write UUID tree
elements.
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>

Before applying this patch, we cached the csum value into the extent state
tree when reading some data from the disk, this operation increased the lock
contention of the state tree.

Now, we just store the csum value into the bio structure or other unshared
structure, so we can reduce the lock contention.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

When adjusting the enospc rules for relocation I ran into a deadlock because we
were relocating the only system chunk and that forced us to try and allocate a
new system chunk while holding locks in the chunk tree, which caused us to
deadlock.  To fix this I've moved all of the dev extent addition and chunk
addition out to the delayed chunk completion stuff.  We still keep the in-memory
stuff which makes sure everything is consistent.

One change I had to make was to search the commit root of the device tree to
find a free dev extent, and hold onto any chunk em's that we allocated in that
transaction so we do not allocate the same dev extent twice.  This has the side
effect of fixing a bug with balance that has been there ever since balance
existed.  Basically you can free a block group and it's dev extent and then
immediately allocate that dev extent for a new block group and write stuff to
that dev extent, all within the same transaction.  So if you happen to crash
during a balance you could come back to a completely broken file system.  This
patch should keep these sort of things from happening in the future since we
won't be able to allocate free'd dev extents until after the transaction
commits.  This has passed all of the xfstests and my super annoying stress test
followed by a balance.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

There are several functions whose code is similar, such as
  btrfs_find_last_root()
  btrfs_read_fs_root_no_radix()

Besides that, some functions are invoked twice, it is unnecessary,
for example, we are sure that all roots which is found in
  btrfs_find_orphan_roots()
have their orphan items, so it is unnecessary to check the orphan
item again.

So cleanup it.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

Btrfs has been pointer tagging bi_private and using bi_bdev
to store the stripe index and mirror number of failed IOs.

As bios bubble back up through the call chain, we use these
to decide if and how to retry our IOs.  They are also used
to count IO failures on a per device basis.

Recently a bio tracepoint was added lead to crashes because
we were abusing bi_bdev.

This commit adds a btrfs bioset, and creates explicit fields
for the mirror number and stripe index.  The plan is to
extend this structure for all of the fields currently in
struct btrfs_bio, which will mean one less kmalloc in
our IO path.
Signed-off-by: NChris Mason <chris.mason@fusionio.com>
Reported-by: NTejun Heo <tj@kernel.org>

Big patch, but all it does is add statics to functions which
are in fact static, then remove the associated dead-code fallout.

removed functions:

btrfs_iref_to_path()
__btrfs_lookup_delayed_deletion_item()
__btrfs_search_delayed_insertion_item()
__btrfs_search_delayed_deletion_item()
find_eb_for_page()
btrfs_find_block_group()
range_straddles_pages()
extent_range_uptodate()
btrfs_file_extent_length()
btrfs_scrub_cancel_devid()
btrfs_start_transaction_lflush()

btrfs_print_tree() is left because it is used for debugging.
btrfs_start_transaction_lflush() and btrfs_reada_detach() are
left for symmetry.

ulist.c functions are left, another patch will take care of those.
Signed-off-by: NEric Sandeen <sandeen@redhat.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

The header file will then be installed under /usr/include/linux so that
userspace applications can refer to Btrfs ioctls by name and use the same
structs used internally in the kernel.
Signed-off-by: NFilipe Brandenburger <filbranden@google.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

This builds on David Woodhouse's original Btrfs raid5/6 implementation.
The code has changed quite a bit, blame Chris Mason for any bugs.

Read/modify/write is done after the higher levels of the filesystem have
prepared a given bio.  This means the higher layers are not responsible
for building full stripes, and they don't need to query for the topology
of the extents that may get allocated during delayed allocation runs.
It also means different files can easily share the same stripe.

But, it does expose us to incorrect parity if we crash or lose power
while doing a read/modify/write cycle.  This will be addressed in a
later commit.

Scrub is unable to repair crc errors on raid5/6 chunks.

Discard does not work on raid5/6 (yet)

The stripe size is fixed at 64KiB per disk.  This will be tunable
in a later commit.
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Raid properties can be shared among raid calculation code, we can put
them into a global table to keep it simple.
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

This commit contains all the essential changes to the core code
of Btrfs for support of the device replace procedure.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

This adds a new file to the sources together with the header file
and the changes to ioctl.h and ctree.h that are required by the
new C source file. Additionally, 4 new functions are added to
volume.c that deal with device creation and destruction.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

This patch adds some code to disallow operations on the device that
is used as the target for the device replace operation.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

A small number of functions that are used in a device replace
procedure when the operation is resumed at mount time are unable
to pass the same root pointer that would be used in the regular
(ioctl) context. And since the root pointer is not required, only
the fs_info is, the root pointer argument is replaced with the
fs_info pointer argument.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

This new function is used by the device replace procedure in
a later patch.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

This is required for the device replace procedure in a later step.
Two calling functions also had to be changed to have the fs_info
pointer: repair_io_failure() and scrub_setup_recheck_block().
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

This is required for the device replace procedure in a later step.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

The new function btrfs_find_device_missing_or_by_path() will be
used for the device replace procedure. This function itself calls
the second new function btrfs_find_device_by_path().
Unfortunately, it is not possible to currently make the rest of the
code use these functions as well, since all functions that look
similar at first view are all a little bit different in what they
are doing. But in the future, new code could benefit from these
two new functions, and currently, device replace uses them.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

The device replace procedure makes use of the scrub code. The scrub
code is the most efficient code to read the allocated data of a disk,
i.e. it reads sequentially in order to avoid disk head movements, it
skips unallocated blocks, it uses read ahead mechanisms, and it
contains all the code to detect and repair defects.
This commit is a first preparation step to adapt the scrub code to
be shareable for the device replace procedure.
The block device will be removed from the scrub context state
structure in a later step. It used to be the source block device.
The scrub code as it is used for the device replace procedure reads
the source data from whereever it is optimal. The source device might
even be gone (disconnected, for instance due to a hardware failure).
Or the drive can be so faulty so that the device replace procedure
tries to avoid access to the faulty source drive as much as possible,
and only if all other mirrors are damaged, as a last resort, the
source disk is accessed.
The modified scrub code operates as if it would handle the source
drive and thereby generates an exact copy of the source disk on the
target disk, even if the source disk is not present at all. Therefore
the block device pointer to the source disk is removed in a later
patch, and therefore the context structure is renamed (this is the
goal of the current patch) to reflect that no source block device
scope is there anymore.

Summary:
This first preparation step consists of a textual substitution of the
term "dev" to the term "ctx" whereever the scrub context is used.
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

Commit 442a4f63 added btrfs device
statistic counters for detected IO and checksum errors to Linux 3.5.
The statistic part that counts checksum errors in
end_bio_extent_readpage() can cause a BUG() in a subfunction:
"kernel BUG at fs/btrfs/volumes.c:3762!"
That part is reverted with the current patch.
However, the counting of checksum errors in the scrub context remains
active, and the counting of detected IO errors (read, write or flush
errors) in all contexts remains active.

Cc: stable <stable@vger.kernel.org> # 3.5
Signed-off-by: NStefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

This will be used in conjunction with btrfs device ready <dev>.  This is
needed for initrd's to have a nice and lightweight way to tell if all of the
devices needed for a file system are in the cache currently.  This keeps
them from having to do mount+sleep loops waiting for devices to show up.
Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

Commit c11d2c23 (Btrfs: add ioctl to get and reset the device
stats) introduced two ioctls doing almost the same thing distinguished
by just the ioctl number which encodes "do reset after read". I have
suggested

http://www.mail-archive.com/linux-btrfs@vger.kernel.org/msg16604.html

to implement it via the ioctl args. This hasn't happen, and I think we
should use a more clean way to pass flags and should not waste ioctl
numbers.

CC: Stefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: NDavid Sterba <dsterba@suse.cz>