write_dev_supers is called in sequence. First is it called with wait == 0,
which starts IO on all of the super blocks for a given device. Then it is
called with wait == 1 to make sure they all reach the disk.
It doesn't currently pin the buffers between the two calls, and it also
assumes the buffers won't go away between the two calls, leading to
an oops if the VM manages to free the buffers in the middle of the sync.
This fixes that assumption and updates the code to return an error if things
are not up to date when the wait == 1 run is done.
Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
On multi-device filesystems, btrfs writes supers to all of the devices
before considering a sync complete. There wasn't any additional
locking between super writeout and the device list management code
because device management was done inside a transaction and
super writeout only happened with no transation writers running.
With the btrfs fsync log and other async transaction updates, this
has been racey for some time. This adds a mutex to protect
the device list. The existing volume mutex could not be reused due to
transaction lock ordering requirements.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
There's no need to preserve this abstraction; it used to let us use
hardware crc32c support directly, but libcrc32c is already doing that for us
through the crypto API -- so we're already using the Intel crc32c
acceleration where appropriate.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
During mount, btrfs will check the queue nonrot flag
for all the devices found in the FS. If they are all
non-rotating, SSD mode is enabled by default.
If the FS was mounted with -o nossd, the non-rotating
flag is ignored.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Once a metadata block has been written, it must be recowed, so the
btrfs dirty balancing call has a check to make sure a fair amount of metadata
was actually dirty before it started writing it back to disk.
A previous commit had changed the dirty tracking for metadata without
updating the btrfs dirty balancing checks. This commit switches it
to use the correct counter.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This commit introduces a new kind of back reference for btrfs metadata.
Once a filesystem has been mounted with this commit, IT WILL NO LONGER
BE MOUNTABLE BY OLDER KERNELS.
When a tree block in subvolume tree is cow'd, the reference counts of all
extents it points to are increased by one. At transaction commit time,
the old root of the subvolume is recorded in a "dead root" data structure,
and the btree it points to is later walked, dropping reference counts
and freeing any blocks where the reference count goes to 0.
The increments done during cow and decrements done after commit cancel out,
and the walk is a very expensive way to go about freeing the blocks that
are no longer referenced by the new btree root. This commit reduces the
transaction overhead by avoiding the need for dead root records.
When a non-shared tree block is cow'd, we free the old block at once, and the
new block inherits old block's references. When a tree block with reference
count > 1 is cow'd, we increase the reference counts of all extents
the new block points to by one, and decrease the old block's reference count by
one.
This dead tree avoidance code removes the need to modify the reference
counts of lower level extents when a non-shared tree block is cow'd.
But we still need to update back ref for all pointers in the block.
This is because the location of the block is recorded in the back ref
item.
We can solve this by introducing a new type of back ref. The new
back ref provides information about pointer's key, level and in which
tree the pointer lives. This information allow us to find the pointer
by searching the tree. The shortcoming of the new back ref is that it
only works for pointers in tree blocks referenced by their owner trees.
This is mostly a problem for snapshots, where resolving one of these
fuzzy back references would be O(number_of_snapshots) and quite slow.
The solution used here is to use the fuzzy back references in the common
case where a given tree block is only referenced by one root,
and use the full back references when multiple roots have a reference
on a given block.
This commit adds per subvolume red-black tree to keep trace of cached
inodes. The red-black tree helps the balancing code to find cached
inodes whose inode numbers within a given range.
This commit improves the balancing code by introducing several data
structures to keep the state of balancing. The most important one
is the back ref cache. It caches how the upper level tree blocks are
referenced. This greatly reduce the overhead of checking back ref.
The improved balancing code scales significantly better with a large
number of snapshots.
This is a very large commit and was written in a number of
pieces. But, they depend heavily on the disk format change and were
squashed together to make sure git bisect didn't end up in a
bad state wrt space balancing or the format change.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
These debugging WARN_ONs make too much console noise during regular
IO failures. An IO failure will still generate a number of messages
as we verify checksums etc, but these two are not needed.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Just happened to notice a bunch of %llu vs u64 warnings. Here's a patch
to cast them all.
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Btrfs has printks for various IO errors, including bad checksums and
mismatches between what we expect the block headers to contain and what
we actually find on the disk.
Longer term we need a real reporting mechanism for this, but for now
printk is going to have to do.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch makes the chunk allocator keep a good ratio of metadata vs data
block groups. By default for every 8 data block groups, we'll allocate 1
metadata chunk, or about 12% of the disk will be allocated for metadata. This
can be changed by specifying the metadata_ratio mount option.
This is simply the number of data block groups that have to be allocated to
force a metadata chunk allocation. By making sure we allocate metadata chunks
more often, we are less likely to get into situations where the whole disk
has been allocated as data block groups.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Btrfs is using WRITE_SYNC_PLUG to send down synchronous IOs with a
higher priority. But, the checksumming helper threads prevent it
from being fully effective.
There are two problems. First, a big queue of pending checksumming
will delay the synchronous IO behind other lower priority writes. Second,
the checksumming uses an ordered async work queue. The ordering makes sure
that IOs are sent to the block layer in the same order they are sent
to the checksumming threads. Usually this gives us less seeky IO.
But, when we start mixing IO priorities, the lower priority IO can delay
the higher priority IO.
This patch solves both problems by adding a high priority list to the async
helper threads, and a new btrfs_set_work_high_prio(), which is used
to make put a new async work item onto the higher priority list.
The ordering is still done on high priority IO, but all of the high
priority bios are ordered separately from the low priority bios. This
ordering is purely an IO optimization, it is not involved in data
or metadata integrity.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Part of reducing fsync/O_SYNC/O_DIRECT latencies is using WRITE_SYNC for
writes we plan on waiting on in the near future. This patch
mirrors recent changes in other filesystems and the generic code to
use WRITE_SYNC when WB_SYNC_ALL is passed and to use WRITE_SYNC for
other latency critical writes.
Btrfs uses async worker threads for checksumming before the write is done,
and then again to actually submit the bios. The bio submission code just
runs a per-device list of bios that need to be sent down the pipe.
This list is split into low priority and high priority lists so the
WRITE_SYNC IO happens first.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: BUG to BUG_ON changes
Btrfs: remove dead code
Btrfs: remove dead code
Btrfs: fix typos in comments
Btrfs: remove unused ftrace include
Btrfs: fix __ucmpdi2 compile bug on 32 bit builds
Btrfs: free inode struct when btrfs_new_inode fails
Btrfs: fix race in worker_loop
Btrfs: add flushoncommit mount option
Btrfs: notreelog mount option
Btrfs: introduce btrfs_show_options
Btrfs: rework allocation clustering
Btrfs: Optimize locking in btrfs_next_leaf()
Btrfs: break up btrfs_search_slot into smaller pieces
Btrfs: kill the pinned_mutex
Btrfs: kill the block group alloc mutex
Btrfs: clean up find_free_extent
Btrfs: free space cache cleanups
Btrfs: unplug in the async bio submission threads
Btrfs: keep processing bios for a given bdev if our proc is batching
Remove an unneeded return statement and conditional
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Because btrfs is copy-on-write, we end up picking new locations for
blocks very often. This makes it fairly difficult to maintain perfect
read patterns over time, but we can at least do some optimizations
for writes.
This is done today by remembering the last place we allocated and
trying to find a free space hole big enough to hold more than just one
allocation. The end result is that we tend to write sequentially to
the drive.
This happens all the time for metadata and it happens for data
when mounted -o ssd. But, the way we record it is fairly racey
and it tends to fragment the free space over time because we are trying
to allocate fairly large areas at once.
This commit gets rid of the races by adding a free space cluster object
with dedicated locking to make sure that only one process at a time
is out replacing the cluster.
The free space fragmentation is somewhat solved by allowing a cluster
to be comprised of smaller free space extents. This part definitely
adds some CPU time to the cluster allocations, but it allows the allocator
to consume the small holes left behind by cow.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch removes the pinned_mutex. The extent io map has an internal tree
lock that protects the tree itself, and since we only copy the extent io map
when we are committing the transaction we don't need it there. We also don't
need it when caching the block group since searching through the tree is also
protected by the internal map spin lock.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: try to free metadata pages when we free btree blocks
Btrfs: add extra flushing for renames and truncates
Btrfs: make sure btrfs_update_delayed_ref doesn't increase ref_mod
Btrfs: optimize fsyncs on old files
Btrfs: tree logging unlink/rename fixes
Btrfs: Make sure i_nlink doesn't hit zero too soon during log replay
Btrfs: limit balancing work while flushing delayed refs
Btrfs: readahead checksums during btrfs_finish_ordered_io
Btrfs: leave btree locks spinning more often
Btrfs: Only let very young transactions grow during commit
Btrfs: Check for a blocking lock before taking the spin
Btrfs: reduce stack in cow_file_range
Btrfs: reduce stalls during transaction commit
Btrfs: process the delayed reference queue in clusters
Btrfs: try to cleanup delayed refs while freeing extents
Btrfs: reduce stack usage in some crucial tree balancing functions
Btrfs: do extent allocation and reference count updates in the background
Btrfs: don't preallocate metadata blocks during btrfs_search_slot
Renames and truncates are both common ways to replace old data with new
data. The filesystem can make an effort to make sure the new data is
on disk before actually replacing the old data.
This is especially important for rename, which many application use as
though it were atomic for both the data and the metadata involved. The
current btrfs code will happily replace a file that is fully on disk
with one that was just created and still has pending IO.
If we crash after transaction commit but before the IO is done, we'll end
up replacing a good file with a zero length file. The solution used
here is to create a list of inodes that need special ordering and force
them to disk before the commit is done. This is similar to the
ext3 style data=ordering, except it is only done on selected files.
Btrfs is able to get away with this because it does not wait on commits
very often, even for fsync (which use a sub-commit).
For renames, we order the file when it wasn't already
on disk and when it is replacing an existing file. Larger files
are sent to filemap_flush right away (before the transaction handle is
opened).
For truncates, we order if the file goes from non-zero size down to
zero size. This is a little different, because at the time of the
truncate the file has no dirty bytes to order. But, we flag the inode
so that it is added to the ordered list on close (via release method). We
also immediately add it to the ordered list of the current transaction
so that we can try to flush down any writes the application sneaks in
before commit.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
btrfs_mark_buffer dirty would set dirty bits in the extent_io tree
for the buffers it was dirtying. This may require a kmalloc and it
was not atomic. So, anyone who called btrfs_mark_buffer_dirty had to
set any btree locks they were holding to blocking first.
This commit changes dirty tracking for extent buffers to just use a flag
in the extent buffer. Now that we have one and only one extent buffer
per page, this can be safely done without losing dirty bits along the way.
This also introduces a path->leave_spinning flag that callers of
btrfs_search_slot can use to indicate they will properly deal with a
path returned where all the locks are spinning instead of blocking.
Many of the btree search callers now expect spinning paths,
resulting in better btree concurrency overall.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The delayed reference queue maintains pending operations that need to
be done to the extent allocation tree. These are processed by
finding records in the tree that are not currently being processed one at
a time.
This is slow because it uses lots of time searching through the rbtree
and because it creates lock contention on the extent allocation tree
when lots of different procs are running delayed refs at the same time.
This commit changes things to grab a cluster of refs for processing,
using a cursor into the rbtree as the starting point of the next search.
This way we walk smoothly through the rbtree.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The extent allocation tree maintains a reference count and full
back reference information for every extent allocated in the
filesystem. For subvolume and snapshot trees, every time
a block goes through COW, the new copy of the block adds a reference
on every block it points to.
If a btree node points to 150 leaves, then the COW code needs to go
and add backrefs on 150 different extents, which might be spread all
over the extent allocation tree.
These updates currently happen during btrfs_cow_block, and most COWs
happen during btrfs_search_slot. btrfs_search_slot has locks held
on both the parent and the node we are COWing, and so we really want
to avoid IO during the COW if we can.
This commit adds an rbtree of pending reference count updates and extent
allocations. The tree is ordered by byte number of the extent and byte number
of the parent for the back reference. The tree allows us to:
1) Modify back references in something close to disk order, reducing seeks
2) Significantly reduce the number of modifications made as block pointers
are balanced around
3) Do all of the extent insertion and back reference modifications outside
of the performance critical btrfs_search_slot code.
#3 has the added benefit of greatly reducing the btrfs stack footprint.
The extent allocation tree modifications are done without the deep
(and somewhat recursive) call chains used in the past.
These delayed back reference updates must be done before the transaction
commits, and so the rbtree is tied to the transaction. Throttling is
implemented to help keep the queue of backrefs at a reasonable size.
Since there was a similar mechanism in place for the extent tree
extents, that is removed and replaced by the delayed reference tree.
Yan Zheng <yan.zheng@oracle.com> helped review and fixup this code.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
btrfs_tree_locked was being used to make sure a given extent_buffer was
properly locked in a few places. But, it wasn't correct for UP compiled
kernels.
This switches it to using assert_spin_locked instead, and renames it to
btrfs_assert_tree_locked to better reflect how it was really being used.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Btrfs is currently using spin_lock_nested with a nested value based
on the tree depth of the block. But, this doesn't quite work because
the max tree depth is bigger than what spin_lock_nested can deal with,
and because locks are sometimes taken before the level field is filled in.
The solution here is to use lockdep_set_class_and_name instead, and to
set the class before unlocking the pages when the block is read from the
disk and just after init of a freshly allocated tree block.
btrfs_clear_path_blocking is also changed to take the locks in the proper
order, and it also makes sure all the locks currently held are properly
set to blocking before it tries to retake the spinlocks. Otherwise, lockdep
gets upset about bad lock orderin.
The lockdep magic cam from Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Most of the btrfs metadata operations can be protected by a spinlock,
but some operations still need to schedule.
So far, btrfs has been using a mutex along with a trylock loop,
most of the time it is able to avoid going for the full mutex, so
the trylock loop is a big performance gain.
This commit is step one for getting rid of the blocking locks entirely.
btrfs_tree_lock takes a spinlock, and the code explicitly switches
to a blocking lock when it starts an operation that can schedule.
We'll be able get rid of the blocking locks in smaller pieces over time.
Tracing allows us to find the most common cause of blocking, so we
can start with the hot spots first.
The basic idea is:
btrfs_tree_lock() returns with the spin lock held
btrfs_set_lock_blocking() sets the EXTENT_BUFFER_BLOCKING bit in
the extent buffer flags, and then drops the spin lock. The buffer is
still considered locked by all of the btrfs code.
If btrfs_tree_lock gets the spinlock but finds the blocking bit set, it drops
the spin lock and waits on a wait queue for the blocking bit to go away.
Much of the code that needs to set the blocking bit finishes without actually
blocking a good percentage of the time. So, an adaptive spin is still
used against the blocking bit to avoid very high context switch rates.
btrfs_clear_lock_blocking() clears the blocking bit and returns
with the spinlock held again.
btrfs_tree_unlock() can be called on either blocking or spinning locks,
it does the right thing based on the blocking bit.
ctree.c has a helper function to set/clear all the locked buffers in a
path as blocking.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Before metadata is written to disk, it is updated to reflect that writeout
has begun. Once this update is done, the block must be cow'd before it
can be modified again.
This update was originally synchronized by using a per-fs spinlock. Today
the buffers for the metadata blocks are locked before writeout begins,
and everyone that tests the flag has the buffer locked as well.
So, the per-fs spinlock (called hash_lock for no good reason) is no
longer required.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Tracing shows the delay between when an async thread goes to sleep
and when more work is added is often very short. This commit adds
a little bit of delay and extra checking to the code right before
we schedule out.
It allows more work to be added to the worker
without requiring notifications from other procs.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
To improve performance, btrfs_sync_log merges tree log sync
requests. But it wrongly merges sync requests for different
tree logs. If multiple tree logs are synced at the same time,
only one of them actually gets synced.
This patch has following changes to fix the bug:
Move most tree log related fields in btrfs_fs_info to
btrfs_root. This allows merging sync requests separately
for each tree log.
Don't insert root item into the log root tree immediately
after log tree is allocated. Root item for log tree is
inserted when log tree get synced for the first time. This
allows syncing the log root tree without first syncing all
log trees.
At tree-log sync, btrfs_sync_log first sync the log tree;
then updates corresponding root item in the log root tree;
sync the log root tree; then update the super block.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Merge list_for_each* and list_entry to list_for_each_entry*
Signed-off-by: Qinghuang Feng <qhfeng.kernel@gmail.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
kthread_run() returns the kthread or ERR_PTR(-ENOMEM), not NULL.
Signed-off-by: Qinghuang Feng <qhfeng.kernel@gmail.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The data in fs_info->super_for_commit are zeros before the
first transaction commit. If tree log sync and system crash
both occur before the first transaction commit, super block
will get corrupted.
This fixes it by properly filling in the super_for_commit field at
open time.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
bio_end_io for reads without checksumming on and btree writes were
happening without using async thread pools. This means the extent_io.c
code had to use spin_lock_irq and friends on the rb tree locks for
extent state.
There were some irq safe vs unsafe lock inversions between the delallock
lock and the extent state locks. This patch gets rid of them by moving
all end_io code into the thread pools.
To avoid contention and deadlocks between the data end_io processing and the
metadata end_io processing yet another thread pool is added to finish
off metadata writes.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch makes seed device possible to be shared by
multiple mounted file systems. The sharing is achieved
by cloning seed device's btrfs_fs_devices structure.
Thanks you,
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This patch implements superblock duplication. Superblocks
are stored at offset 16K, 64M and 256G on every devices.
Spaces used by superblocks are preserved by the allocator,
which uses a reverse mapping function to find the logical
addresses that correspond to superblocks. Thank you,
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Btrfs stores checksums for each data block. Until now, they have
been stored in the subvolume trees, indexed by the inode that is
referencing the data block. This means that when we read the inode,
we've probably read in at least some checksums as well.
But, this has a few problems:
* The checksums are indexed by logical offset in the file. When
compression is on, this means we have to do the expensive checksumming
on the uncompressed data. It would be faster if we could checksum
the compressed data instead.
* If we implement encryption, we'll be checksumming the plain text and
storing that on disk. This is significantly less secure.
* For either compression or encryption, we have to get the plain text
back before we can verify the checksum as correct. This makes the raid
layer balancing and extent moving much more expensive.
* It makes the front end caching code more complex, as we have touch
the subvolume and inodes as we cache extents.
* There is potentitally one copy of the checksum in each subvolume
referencing an extent.
The solution used here is to store the extent checksums in a dedicated
tree. This allows us to index the checksums by phyiscal extent
start and length. It means:
* The checksum is against the data stored on disk, after any compression
or encryption is done.
* The checksum is stored in a central location, and can be verified without
following back references, or reading inodes.
This makes compression significantly faster by reducing the amount of
data that needs to be checksummed. It will also allow much faster
raid management code in general.
The checksums are indexed by a key with a fixed objectid (a magic value
in ctree.h) and offset set to the starting byte of the extent. This
allows us to copy the checksum items into the fsync log tree directly (or
any other tree), without having to invent a second format for them.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch gives us the space we will need in order to have different csum
algorithims at some point in the future. We save the csum algorithim type
in the superblock, and use those instead of define's.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
This needs to be applied on top of my previous patches, but is needed for more
than just my new stuff. We're going to the wrong label when we have an error,
we try to stop the workers, but they are started below all of this code. This
fixes it so we go to the right error label and not panic when we fail one of
these cases.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
This adds the necessary disk format for handling compatibility flags
in the future to handle disk format changes. We have a compat_flags,
compat_ro_flags and incompat_flags set for the super block. Compat
flags will be to hold the features that are compatible with older
versions of btrfs, compat_ro flags have features that are compatible
with older versions of btrfs if the fs is mounted read only, and
incompat_flags has features that are incompatible with older versions
of btrfs. This also axes the compat_flags field for the inode and
just makes the flags field a 64bit field, and changes the root item
flags field to 64bit.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Shut up various sparse warnings about symbols that should be either
static or have their declarations in scope.
Signed-off-by: Christoph Hellwig <hch@lst.de>
The log replay produces dirty roots. These dirty roots
should be dropped immediately if the fs is mounted as
ro. Otherwise they can be added to the dirty root list
again when remounting the fs as rw. Thank you,
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
The btrfs git kernel trees is used to build a standalone tree for
compiling against older kernels. This commit makes the standalone tree
work with 2.6.27
Signed-off-by: Chris Mason <chris.mason@oracle.com>
fsync log replay can change the filesystem, so it cannot be delayed until
mount -o rw,remount, and it can't be forgotten entirely. So, this patch
changes btrfs to do with reiserfs, ext3 and xfs do, which is to do the
log replay even when mounted readonly.
On a readonly device if log replay is required, the mount is aborted.
Getting all of this right had required fixing up some of the error
handling in open_ctree.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
While building large bios in writepages, btrfs may end up waiting
for other page writeback to finish if WB_SYNC_ALL is used.
While it is waiting, the bio it is building has a number of pages with the
writeback bit set and they aren't getting to the disk any time soon. This
lowers the latencies of writeback in general by sending down the bio being
built before waiting for other pages.
The bio submission code tries to limit the total number of async bios in
flight by waiting when we're over a certain number of async bios. But,
the waits are happening while writepages is building bios, and this can easily
lead to stalls and other problems for people calling wait_on_page_writeback.
The current fix is to let the congestion tests take care of waiting.
sync() and others make sure to drain the current async requests to make
sure that everything that was pending when the sync was started really get
to disk. The code would drain pending requests both before and after
submitting a new request.
But, if one of the requests is waiting for page writeback to finish,
the draining waits might block that page writeback. This changes the
draining code to only wait after submitting the bio being processed.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
