We allow two log transactions at a time, but use same flag
to mark dirty tree-log btree blocks. So we may flush dirty
blocks belonging to newer log transaction when committing a
log transaction. This patch fixes the issue by using two
flags to mark dirty tree-log btree blocks.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: fix panic when trying to destroy a newly allocated
Btrfs: allow more metadata chunk preallocation
Btrfs: fallback on uncompressed io if compressed io fails
Btrfs: find ideal block group for caching
Btrfs: avoid null deref in unpin_extent_cache()
Btrfs: skip btrfs_release_path in btrfs_update_root and btrfs_del_root
Btrfs: fix some metadata enospc issues
Btrfs: fix how we set max_size for free space clusters
Btrfs: cleanup transaction starting and fix journal_info usage
Btrfs: fix data allocation hint start
On an FS where all of the space has not been allocated into chunks yet,
the enospc can return enospc just because the existing metadata chunks
are full.
We get around this by allowing more metadata chunks to be allocated up
to a certain limit, and finding the right limit is a little fuzzy. The
problem is the reservations for delalloc would preallocate way too much
of the FS as metadata. We need to start saying no and just force some
IO to happen.
But we also need to let a reasonable amount of the FS become metadata.
This bumps the hard limit up, later releases will have a better system.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch changes a few things. Hopefully the comments are helpfull, but
I'll try and be as verbose here.
Problem:
My fedora box was taking 1 minute and 21 seconds to boot with btrfs as root.
Part of this problem was we pick the first block group we can find and start
caching it, even if it may not have enough free space. The other problem is
we only search for cached block groups the first time around, which we won't
find any cached block groups because this is a newly mounted fs, so we end up
caching several block groups during bootup, which with alot of fragmentation
takes around 30-45 seconds to complete, which bogs down the system. So
Solution:
1) Don't cache block groups willy-nilly at first. Instead try and figure out
which block group has the most free, and therefore will take the least amount
of time to cache.
2) Don't be so picky about cached block groups. The other problem is once
we've filled up a cluster, if the block group isn't finished caching the next
time we try and do the allocation we'll completely ignore the cluster and
start searching from the beginning of the space, which makes us cache more
block groups, which slows us down even more. So instead of skipping block
groups that are not finished caching when we have a hint, only skip the block
group if it hasn't started caching yet.
There is one other tweak in here. Before if we allocated a chunk and still
couldn't find new space, we'd end up switching the space info to force another
chunk allocation. This could make us end up with way too many chunks, so keep
track of this particular case.
With this patch and my previous cluster fixes my fedora box now boots in 43
seconds, and according to the bootchart is not held up by our block group
caching at all.
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: always pin metadata in discard mode
Btrfs: enable discard support
Btrfs: add -o discard option
Btrfs: properly wait log writers during log sync
Btrfs: fix possible ENOSPC problems with truncate
Btrfs: fix btrfs acl #ifdef checks
Btrfs: streamline tree-log btree block writeout
Btrfs: avoid tree log commit when there are no changes
Btrfs: only write one super copy during fsync
We have an optimization in btrfs to allow blocks to be
immediately freed if they were allocated in this transaction and never
written. Otherwise they are pinned and freed when the transaction
commits.
This isn't optimal for discard mode because immediately freeing
them means immediately discarding them. It is better to give the
block to the pinning code and letting the (slow) discard happen later.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The discard support code in btrfs currently is guarded by ifdefs for
BIO_RW_DISCARD, which is never defines as it's the name of an enum
memeber. Just remove the useless ifdefs to actually enable the code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Enable discard by default is not a good idea given the the trim speed
of SSD prototypes we've seen, and the carecteristics for many high-end
arrays. Turn of discards by default and require the -o discard option
to enable them on.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: fix file clone ioctl for bookend extents
Btrfs: fix uninit compiler warning in cow_file_range_nocow
Btrfs: constify dentry_operations
Btrfs: optimize back reference update during btrfs_drop_snapshot
Btrfs: remove negative dentry when deleting subvolumne
Btrfs: optimize fsync for the single writer case
Btrfs: async delalloc flushing under space pressure
Btrfs: release delalloc reservations on extent item insertion
Btrfs: delay clearing EXTENT_DELALLOC for compressed extents
Btrfs: cleanup extent_clear_unlock_delalloc flags
Btrfs: fix possible softlockup in the allocator
Btrfs: fix deadlock on async thread startup
This patch reading level 0 tree blocks that already use full backrefs.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch moves the delalloc flushing that occurs when we are under space
pressure off to a async thread pool. This helps since we only free up
metadata space when we actually insert the extent item, which means it takes
quite a while for space to be free'ed up if we wait on all ordered extents.
However, if space is freed up due to inline extents being inserted, we can
wake people who are waiting up early, and they can finish their work.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch fixes an issue with the delalloc metadata space reservation
code. The problem is we used to free the reservation as soon as we
allocated the delalloc region. The problem with this is if we are not
inserting an inline extent, we don't actually insert the extent item until
after the ordered extent is written out. This patch does 3 things,
1) It moves the reservation clearing stuff into the ordered code, so when
we remove the ordered extent we remove the reservation.
2) It adds a EXTENT_DO_ACCOUNTING flag that gets passed when we clear
delalloc bits in the cases where we want to clear the metadata reservation
when we clear the delalloc extent, in the case that we do an inline extent
or we invalidate the page.
3) It adds another waitqueue to the space info so that when we start a fs
wide delalloc flush, anybody else who also hits that area will simply wait
for the flush to finish and then try to make their allocation.
This has been tested thoroughly to make sure we did not regress on
performance.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Like the cluster allocating stuff, we can lockup the box with the normal
allocation path. This happens when we
1) Start to cache a block group that is severely fragmented, but has a decent
amount of free space.
2) Start to commit a transaction
3) Have the commit try and empty out some of the delalloc inodes with extents
that are relatively large.
The inodes will not be able to make the allocations because they will ask for
allocations larger than a contiguous area in the free space cache. So we will
wait for more progress to be made on the block group, but since we're in a
commit the caching kthread won't make any more progress and it already has
enough free space that wait_block_group_cache_progress will just return. So,
if we wait and fail to make the allocation the next time around, just loop and
go to the next block group. This keeps us from getting stuck in a softlockup.
Thanks,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
If an ioctl-initiated transaction is open, we can't force a commit during
the free space checks in order to free up pinned extents or else we
deadlock. Just ENOSPC instead.
A more satisfying solution that reserves space for the entire user
transaction up front is forthcoming...
Signed-off-by: Sage Weil <sage@newdream.net>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
At the start of a transaction we do a btrfs_reserve_metadata_space() and
specify how many items we plan on modifying. Then once we've done our
modifications and such, just call btrfs_unreserve_metadata_space() for
the same number of items we reserved.
For keeping track of metadata needed for data I've had to add an extent_io op
for when we merge extents. This lets us track space properly when we are doing
sequential writes, so we don't end up reserving way more metadata space than
what we need.
The only place where the metadata space accounting is not done is in the
relocation code. This is because Yan is going to be reworking that code in the
near future, so running btrfs-vol -b could still possibly result in a ENOSPC
related panic. This patch also turns off the metadata_ratio stuff in order to
allow users to more efficiently use their disk space.
This patch makes it so we track how much metadata we need for an inode's
delayed allocation extents by tracking how many extents are currently
waiting for allocation. It introduces two new callbacks for the
extent_io tree's, merge_extent_hook and split_extent_hook. These help
us keep track of when we merge delalloc extents together and split them
up. Reservations are handled prior to any actually dirty'ing occurs,
and then we unreserve after we dirty.
btrfs_unreserve_metadata_for_delalloc() will make the appropriate
unreservations as needed based on the number of reservations we
currently have and the number of extents we currently have. Doing the
reservation outside of doing any of the actual dirty'ing lets us do
things like filemap_flush() the inode to try and force delalloc to
happen, or as a last resort actually start allocation on all delalloc
inodes in the fs. This has survived dbench, fs_mark and an fsx torture
test.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
We now do extra checks before a balance to make sure
there is room for the balance to take place. One of
the checks was testing to see if we were trying to
balance away the last block group of a given type.
If there is no space available for new chunks, we
should not try and balance away the last block group
of a give type. But, the code wasn't checking for
available chunk space, and so it was exiting too soon.
The fix here is to combine some of the checks and make
sure we try to allocate new chunks when we're balancing
the last block group.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
After a balance it is briefly possible for the space info
field in the inode to be NULL. This adds some checks
to make sure things properly deal with the NULL value.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
As we get closer to proper -ENOSPC handling in btrfs, we need more accurate
space accounting for the space info's. Currently we exclude the free space for
the super mirrors, but the space they take up isn't accounted for in any of the
counters. This patch introduces bytes_super, which keeps track of the amount
of bytes used for a super mirror in the block group cache and space info. This
makes sure that our free space caclucations will be completely accurate.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch removes a bunch of dead code from the snapshot removal stuff. It
was confusing me when doing the metadata ENOSPC stuff so I killed it.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The box can get locked up in the allocator if we happen upon a block group
under these conditions:
1) During a commit, so caching threads cannot make progress
2) Our block group currently is in the middle of being cached
3) Our block group currently has plenty of free space in it
4) Our block group is so fragmented that it ends up having no free space chunks
larger than min_bytes calculated by btrfs_find_space_cluster.
What happens is we try and do btrfs_find_space_cluster, which fails because it
is unable to find enough free space chunks that are large than min_bytes and
are close enough together. Since the block group is not cached we do a
wait_block_group_cache_progress, which waits for the number of bytes we need,
except the block group already has _plenty_ of free space, its just severely
fragmented, so we loop and try again, ad infinitum. This patch keeps us from
waiting on the block group to finish caching if we failed to find a free space
cluster before. It also makes sure that we don't even try to find a free space
cluster if we are on our last loop in the allocator, since we will have tried
everything at this point at it is futile.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Currently, we can panic the box if the first block group we go to move is of a
type where there is no space left to move those extents. For example, if we
fill the disk up with data, and then we try to balance and we have no room to
move the data nor room to allocate new chunks, we will panic. Change this by
checking to see if we have room to move this chunk around, and if not, return
-ENOSPC and move on to the next chunk. This will make sure we remove block
groups that are moveable, like if we have alot of empty metadata block groups,
and then that way we make room to be able to balance our data chunks as well.
Tested this with an fs that would panic on btrfs-vol -b normally, but no longer
panics with this patch.
V1->V2:
-actually search for a free extent on the device to make sure we can allocate a
chunk if need be.
-fix btrfs_shrink_device to make sure we actually try to relocate all the
chunks, and then if we can't return -ENOSPC so if we are doing a btrfs-vol -r
we don't remove the device with data still on it.
-check to make sure the block group we are going to relocate isn't the last one
in that particular space
-fix a bug in btrfs_shrink_device where we would change the device's size and
not fix it if we fail to do our relocate
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch adds snapshot/subvolume destroy ioctl. A subvolume that isn't being
used and doesn't contains links to other subvolumes can be destroyed.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch contains two changes to avoid unnecessary tree block reads during
snapshot dropping.
First, check tree block's reference count and flags before reading the tree
block. if reference count > 1 and there is no need to update backrefs, we can
avoid reading the tree block.
Second, save when snapshot was created in root_key.offset. we can compare block
pointer's generation with snapshot's creation generation during updating
backrefs. If a given block was created before snapshot was created, the
snapshot can't be the tree block's owner. So we can avoid reading the block.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch gets rid of two limitations of async block group caching.
The old code delays handling pinned extents when block group is in
caching. To allocate logged file extents, the old code need wait
until block group is fully cached. To get rid of the limitations,
This patch introduces a data structure to track the progress of
caching. Base on the caching progress, we know which extents should
be added to the free space cache when handling the pinned extents.
The logged file extents are also handled in a similar way.
This patch also changes how pinned extents are tracked. The old
code uses one tree to track pinned extents, and copy the pinned
extents tree at transaction commit time. This patch makes it use
two trees to track pinned extents. One tree for extents that are
pinned in the running transaction, one tree for extents that can
be unpinned. At transaction commit time, we swap the two trees.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
blk_ioctl_discard duplicates large amounts of code from blkdev_issue_discard,
the only difference between the two is that blkdev_issue_discard needs to
send a barrier discard request and blk_ioctl_discard a non-barrier one,
and blk_ioctl_discard needs to wait on the request. To facilitates this
add a flags argument to blkdev_issue_discard to control both aspects of the
behaviour. This will be very useful later on for using the waiting
funcitonality for other callers.
Based on an earlier patch from Matthew Wilcox <matthew@wil.cx>.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The async caching thread can end up looping forever if a given
search puts it at the last key in a leaf. It will end up calling
btrfs_next_leaf and then checking if it needs to politely drop
the read semaphore.
Most of the time this looping isn't noticed because it is able to
make progress the next time around. But, during log replay,
we wait on the async caching thread to finish, and the async thread
is waiting on the commit, and no progress is really made.
The fix used here is to copy the key out of the next leaf,
that way our search lands there properly.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The semaphore used by the async caching threads can prevent a
transaction commit, which can make the FS appear to stall. This
releases the semaphore more often when a transaction commit is
in progress.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The async block group caching code uses the commit_root pointer
to get a stable version of the extent allocation tree for scanning.
This copy of the tree root isn't going to change and it significantly
reduces the complexity of the scanning code.
During a commit, we have a loop where we update the extent allocation
tree root. We need to loop because updating the root pointer in
the tree of tree roots may allocate blocks which may change the
extent allocation tree.
Right now the commit_root pointer is changed inside this loop. It
is more correct to change the commit_root pointer only after all the
looping is done.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
- don't stop the caching thread until btrfs_commit_super return.
- if caching is interrupted by umount, set last to (u64)-1.
otherwise the un-scanned range of block group will be considered
as free extent.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
We are racy with async block caching and unpinning extents. This patch makes
things much less complicated by only unpinning the extent if the block group is
cached. We check the block_group->cached var under the block_group->lock spin
lock. If it is set to BTRFS_CACHE_FINISHED then we update the pinned counters,
and unpin the extent and add the free space back. If it is not set to this, we
start the caching of the block group so the next time we unpin extents we can
unpin the extent. This keeps us from racing with the async caching threads,
lets us kill the fs wide async thread counter, and keeps us from having to set
DELALLOC bits for every extent we hit if there are caching kthreads going.
One thing that needed to be changed was btrfs_free_super_mirror_extents. Now
instead of just looking for LOCKED extents, we also look for DIRTY extents,
since we could have left some extents pinned in the previous transaction that
will never get freed now that we are unmounting, which would cause us to leak
memory. So btrfs_free_super_mirror_extents has been changed to
btrfs_free_pinned_extents, and it will clear the extents locked for the super
mirror, and any remaining pinned extents that may be present. Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Btrfs allocates individual extents from block groups, and each
block group has a specific type. It may hold metadata, data
mirrored or striped etc.
When we balance space (btrfs-vol -b) or remove a drive (btrfs-vol -r)
we free block groups. Once a block group is freed, the space it was
using on the device may be available for use by new block groups.
btrfs_remove_block_group was clearing the flag that said
'our devices are full, don't even try to allocate new block groups',
but it was only clearing that flag for a specific type of block group.
This commit clears the full flag for all of the types of block groups,
making it much more likely that we'll be able to balance space when
the drive is close to full.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch moves the caching of the block group off to a kthread in order to
allow people to allocate sooner. Instead of blocking up behind the caching
mutex, we instead kick of the caching kthread, and then attempt to make an
allocation. If we cannot, we wait on the block groups caching waitqueue, which
the caching kthread will wake the waiting threads up everytime it finds 2 meg
worth of space, and then again when its finished caching. This is how I tested
the speedup from this
mkfs the disk
mount the disk
fill the disk up with fs_mark
unmount the disk
mount the disk
time touch /mnt/foo
Without my changes this took 11 seconds on my box, with these changes it now
takes 1 second.
Another change thats been put in place is we lock the super mirror's in the
pinned extent map in order to keep us from adding that stuff as free space when
caching the block group. This doesn't really change anything else as far as the
pinned extent map is concerned, since for actual pinned extents we use
EXTENT_DIRTY, but it does mean that when we unmount we have to go in and unlock
those extents to keep from leaking memory.
I've also added a check where when we are reading block groups from disk, if the
amount of space used == the size of the block group, we go ahead and mark the
block group as cached. This drastically reduces the amount of time it takes to
cache the block groups. Using the same test as above, except doing a dd to a
file and then unmounting, it used to take 33 seconds to umount, now it takes 3
seconds.
This version uses the commit_root in the caching kthread, and then keeps track
of how many async caching threads are running at any given time so if one of the
async threads is still running as we cross transactions we can wait until its
finished before handling the pinned extents. Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Currently btrfs has a problem where it can use a ridiculous amount of RAM simply
tracking free space. As free space gets fragmented, we end up with thousands of
entries on an rb-tree per block group, which usually spans 1 gig of area. Since
we currently don't ever flush free space cache back to disk this gets to be a
bit unweildly on large fs's with lots of fragmentation.
This patch solves this problem by using PAGE_SIZE bitmaps for parts of the free
space cache. Initially we calculate a threshold of extent entries we can
handle, which is however many extent entries we can cram into 16k of ram. The
maximum amount of RAM that should ever be used to track 1 gigabyte of diskspace
will be 32k of RAM, which scales much better than we did before.
Once we pass the extent threshold, we start adding bitmaps and using those
instead for tracking the free space. This patch also makes it so that any free
space thats less than 4 * sectorsize we go ahead and put into a bitmap. This is
nice since we try and allocate out of the front of a block group, so if the
front of a block group is heavily fragmented and then has a huge chunk of free
space at the end, we go ahead and add the fragmented areas to bitmaps and use a
normal extent entry to track the big chunk at the back of the block group.
I've also taken the opportunity to revamp how we search for free space.
Previously we indexed free space via an offset indexed rb tree and a bytes
indexed rb tree. I've dropped the bytes indexed rb tree and use only the offset
indexed rb tree. This cuts the number of tree operations we were doing
previously down by half, and gives us a little bit of a better allocation
pattern since we will always start from a specific offset and search forward
from there, instead of searching for the size we need and try and get it as
close as possible to the offset we want.
I've given this a healthy amount of testing pre-new format stuff, as well as
post-new format stuff. I've booted up my fedora box which is installed on btrfs
with this patch and ran with it for a few days without issues. I've not seen
any performance regressions in any of my tests.
Since the last patch Yan Zheng fixed a problem where we could have overlapping
entries, so updating their offset inline would cause problems. Thanks,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Write dirty block groups may allocate new block, and so may add new delayed
back ref. btrfs_run_delayed_refs may make some block groups dirty.
commit_cowonly_roots does not handle the recursion properly, and some dirty
blocks can be left unwritten at commit time. This patch moves
btrfs_run_delayed_refs into the loop that writes dirty block groups, and makes
the code not break out of the loop until there are no dirty block groups or
delayed back refs.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Make an error msg look nicer by inserting a space between number and word.
Signed-off-by: Hu Tao <hu.taoo@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The new backref format has restriction on type of backref item. If a tree
block isn't referenced by its owner tree, full backrefs must be used for the
pointers in it. When a tree block loses its owner tree's reference, backrefs
for the pointers in it should be updated to full backrefs. Current
btrfs_drop_snapshot misses the code that updates backrefs, so it's unsafe for
general use.
This patch adds backrefs update code to btrfs_drop_snapshot. It isn't a
problem in the restricted form btrfs_drop_snapshot is used today, but for
general snapshot deletion this update is required.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
lookup_inline_extent_backref only checks for duplicate backref for data
extents. It assumes backrefs for tree block never conflict.
This patch makes lookup_inline_extent_backref check for duplicate backrefs
for both data and tree block, so that we can detect potential bug earlier.
This is a safety check, strictly speaking it is not required.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
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>
Some SSDs perform best when reusing block numbers often, while
others perform much better when clustering strictly allocates
big chunks of unused space.
The default mount -o ssd will find rough groupings of blocks
where there are a bunch of free blocks that might have some
allocated blocks mixed in.
mount -o ssd_spread will make sure there are no allocated blocks
mixed in. It should perform better on lower end SSDs.
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>
The btrfs allocator uses list_for_each to walk the available block
groups when searching for free blocks. It starts off with a hint
to help find the best block group for a given allocation.
The hint is resolved into a block group, but we don't properly check
to make sure the block group we find isn't in the middle of being
freed due to filesystem shrinking or balancing. If it is being
freed, the list pointers in it are bogus and can't be trusted. But,
the code happily goes along and uses them in the list_for_each loop,
leading to all kinds of fun.
The fix used here is to check to make sure the block group we find really
is on the list before we use it. list_del_init is used when removing
it from the list, so we can do a proper check.
The allocation clustering code has a similar bug where it will trust
the block group in the current free space cluster. If our allocation
flags have changed (going from single spindle dup to raid1 for example)
because the drives in the FS have changed, we're not allowed to use
the old block group any more.
The fix used here is to check the current cluster against the
current allocation flags.
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>
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>
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>
This patch removes the block group alloc mutex used to protect the free space
tree for allocations and replaces it with a spin lock which is used only to
protect the free space rb tree. This means we only take the lock when we are
directly manipulating the tree, which makes us a touch faster with
multi-threaded workloads.
This patch also gets rid of btrfs_find_free_space and replaces it with
btrfs_find_space_for_alloc, which takes the number of bytes you want to
allocate, and empty_size, which is used to indicate how much free space should
be at the end of the allocation.
It will return an offset for the allocator to use. If we don't end up using it
we _must_ call btrfs_add_free_space to put it back. This is the tradeoff to
kill the alloc_mutex, since we need to make sure nobody else comes along and
takes our space.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
I've replaced the strange looping constructs with a list_for_each_entry on
space_info->block_groups. If we have a hint we just jump into the loop with
the block group and start looking for space. If we don't find anything we
start at the beginning and start looking. We never come out of the loop with a
ref on the block_group _unless_ we found space to use, then we drop it after we
set the trans block_group.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
This patch cleans up the free space cache code a bit. It better documents the
idiosyncrasies of tree_search_offset and makes the code make a bit more sense.
I took out the info allocation at the start of __btrfs_add_free_space and put it
where it makes more sense. This was left over cruft from when alloc_mutex
existed. Also all of the re-searches we do to make sure we inserted properly.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
COW means we cycle though blocks fairly quickly, and once we
free an extent on disk, it doesn't make much sense to keep the pages around.
This commit tries to immediately free the page when we free the extent,
which lowers our memory footprint significantly.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The tree logging code allows individual files or directories to be logged
without including operations on other files and directories in the FS.
It tries to commit the minimal set of changes to disk in order to
fsync the single file or directory that was sent to fsync or O_SYNC.
The tree logging code was allowing files and directories to be unlinked
if they were part of a rename operation where only one directory
in the rename was in the fsync log. This patch adds a few new rules
to the tree logging.
1) on rename or unlink, if the inode being unlinked isn't in the fsync
log, we must force a full commit before doing an fsync of the directory
where the unlink was done. The commit isn't done during the unlink,
but it is forced the next time we try to log the parent directory.
Solution: record transid of last unlink/rename per directory when the
directory wasn't already logged. For renames this is only done when
renaming to a different directory.
mkdir foo/some_dir
normal commit
rename foo/some_dir foo2/some_dir
mkdir foo/some_dir
fsync foo/some_dir/some_file
The fsync above will unlink the original some_dir without recording
it in its new location (foo2). After a crash, some_dir will be gone
unless the fsync of some_file forces a full commit
2) we must log any new names for any file or dir that is in the fsync
log. This way we make sure not to lose files that are unlinked during
the same transaction.
2a) we must log any new names for any file or dir during rename
when the directory they are being removed from was logged.
2a is actually the more important variant. Without the extra logging
a crash might unlink the old name without recreating the new one
3) after a crash, we must go through any directories with a link count
of zero and redo the rm -rf
mkdir f1/foo
normal commit
rm -rf f1/foo
fsync(f1)
The directory f1 was fully removed from the FS, but fsync was never
called on f1, only its parent dir. After a crash the rm -rf must
be replayed. This must be able to recurse down the entire
directory tree. The inode link count fixup code takes care of the
ugly details.
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>
To avoid deadlocks and reduce latencies during some critical operations, some
transaction writers are allowed to jump into the running transaction and make
it run a little longer, while others sit around and wait for the commit to
finish.
This is a bit unfair, especially when the callers that jump in do a bunch
of IO that makes all the others procs on the box wait. This commit
reduces the stalls this produces by pre-reading file extent pointers
during btrfs_finish_ordered_io before the transaction is joined.
It also tunes the drop_snapshot code to politely wait for transactions
that have started writing out their delayed refs to finish. This avoids
new delayed refs being flooded into the queue while we're trying to
close off the transaction.
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>
When extents are freed, it is likely that we've removed the last
delayed reference update for the extent. This checks the delayed
ref tree when things are freed, and if no ref updates area left it
immediately processes the delayed ref.
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>
Storage allocated to different raid levels in btrfs is tracked by
a btrfs_space_info structure, and all of the current space_infos are
collected into a list_head.
Most filesystems have 3 or 4 of these structs total, and the list is
only changed when new raid levels are added or at unmount time.
This commit adds rcu locking on the list head, and properly frees
things at unmount time. It also clears the space_info->full flag
whenever new space is added to the FS.
The locking for the space info list goes like this:
reads: protected by rcu_read_lock()
writes: protected by the chunk_mutex
At unmount time we don't need special locking because all the readers
are gone.
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>
This fixes a problem where we could return -ENOSPC when we may actually have
plenty of space, the space is just pinned. Instead of returning -ENOSPC
immediately, commit the transaction first and then try and do the allocation
again.
This patch also does chunk allocation for metadata if we pass the 80%
threshold for metadata space. This will help with stack usage since the chunk
allocation will happen early on, instead of when the allocation is happening.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
This is a step in the direction of better -ENOSPC handling. Instead of
checking the global bytes counter we check the space_info bytes counters to
make sure we have enough space.
If we don't we go ahead and try to allocate a new chunk, and then if that fails
we return -ENOSPC. This patch adds two counters to btrfs_space_info,
bytes_delalloc and bytes_may_use.
bytes_delalloc account for extents we've actually setup for delalloc and will
be allocated at some point down the line.
bytes_may_use is to keep track of how many bytes we may use for delalloc at
some point. When we actually set the extent_bit for the delalloc bytes we
subtract the reserved bytes from the bytes_may_use counter. This keeps us from
not actually being able to allocate space for any delalloc bytes.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
btrfs_record_root_in_trans needs the trans_mutex held to make sure two
callers don't race to setup the root in a given transaction. This adds
it to all the places that were missing it.
Signed-off-by: Yan Zheng <zheng.yan@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>
Larger metadata clusters can significantly improve writeback performance
on ssd drives with large erasure blocks. The larger clusters make it
more likely a given IO will completely overwrite the ssd block, so it
doesn't have to do an internal rwm cycle.
On spinning media, lager metadata clusters end up spreading out the
metadata more over time, which makes fsck slower, so we don't want this
to be the default.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Theres a slight problem with finish_current_insert, if we set all to 1 and then
go through and don't actually skip any of the extents on the pending list, we
could exit right after we've added new extents.
This is a problem because by inserting the new extents we could have gotten new
COW's to happen and such, so we may have some pending updates to do or even
more inserts to do after that.
So this patch will only exit if we have never skipped any of the extents in the
pending list, and we have no extents to insert, this will make sure that all of
the pending work is truly done before we return. I've been running with this
patch for a few days with all of my other testing and have not seen issues.
Thanks,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Every transaction in btrfs creates a new snapshot, and then schedules the
snapshot from the last transaction for deletion. Snapshot deletion
works by walking down the btree and dropping the reference counts
on each btree block during the walk.
If if a given leaf or node has a reference count greater than one,
the reference count is decremented and the subtree pointed to by that
node is ignored.
If the reference count is one, walking continues down into that node
or leaf, and the references of everything it points to are decremented.
The old code would try to work in small pieces, walking down the tree
until it found the lowest leaf or node to free and then returning. This
was very friendly to the rest of the FS because it didn't have a huge
impact on other operations.
But it wouldn't always keep up with the rate that new commits added new
snapshots for deletion, and it wasn't very optimal for the extent
allocation tree because it wasn't finding leaves that were close together
on disk and processing them at the same time.
This changes things to walk down to a level 1 node and then process it
in bulk. All the leaf pointers are sorted and the leaves are dropped
in order based on their extent number.
The extent allocation tree and commit code are now fast enough for
this kind of bulk processing to work without slowing the rest of the FS
down. Overall it does less IO and is better able to keep up with
snapshot deletions under high load.
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>
When a block goes through cow, we update the reference counts of
everything that block points to. The internal pointers of the block
can be in just about any order, and it is likely to have clusters of
things that are close together and clusters of things that are not.
To help reduce the seeks that come with updating all of these reference
counts, sort them by byte number before actual updates are done.
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>
replace_one_extent searches tree leaves for references to a given extent. It
stops searching if it goes beyond the last possible position.
The last possible position is computed by adding the starting offset of a found
file extent to the full size of the extent. The code uses physical size of the
extent as the full size. This is incorrect when compression is used.
The fix is get the full size from ram_bytes field of file extent item.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
btrfs_extent_post_op calls finish_current_insert and del_pending_extents. They
both may enter infinite loops.
finish_current_insert enters infinite loop if it only finds some backrefs to
update. The fix is to check for pending backref updates before restarting the
loop.
The infinite loop in del_pending_extents is due to a the skipped variable
not being properly reset before looping around.
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>
This patch contains following things.
1) Limit the max size of btrfs_ordered_sum structure to PAGE_SIZE. This
struct is kmalloced so we want to keep it reasonable.
2) Replace copy_extent_csums by btrfs_lookup_csums_range. This was
duplicated code in tree-log.c
3) Remove replay_one_csum. csum items are replayed at the same time as
replaying file extents. This guarantees we only replay useful csums.
4) nbytes accounting fix.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This is a patch to fix discard semantic to make Btrfs work with FTL and SSD.
We can improve FTL's performance by telling it which sectors are freed by file
system. But if we don't tell FTL the information of free sectors in proper
time, the transaction mechanism of Btrfs will be destroyed and Btrfs could not
roll back the previous transaction under the power loss condition.
There are some problems in the old implementation:
1, In __free_extent(), the pinned down extents should not be discarded.
2, In free_extents(), the free extents are all pinned, so they need to
be discarded in transaction committing time instead of free_extents().
3, The reserved extent used by log tree should be discard too.
This patch change discard behavior as follows:
1, For the extents which need to be free at once,
we discard them in update_block_group().
2, Delay discarding the pinned extent in btrfs_finish_extent_commit()
when committing transaction.
3, Remove discarding from free_extents() and __free_extent()
4, Add discard interface into btrfs_free_reserved_extent()
5, Discard sectors before updating the free space cache, otherwise,
FTL will destroy file system data.
There is a race in relocate_inode_pages, it happens when
find_delalloc_range finds the delalloc extent before the
boundary bit is set. Thank you,
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This adds the missing block accounting code to finish_current_insert and makes
block accounting for root item properly protected by the delalloc spin lock.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Btrfs maintains a cache of blocks available for allocation in ram. The
code that frees extents was marking the extents free and then deleting
the checksum items.
This meant it was possible the extent would be reallocated before the
checksum item was actually deleted, leading to races and other
problems as the checksums were updated for the newly allocated extent.
The fix is to delete the checksum before marking the extent free.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The delalloc lock doesn't need to have irqs disabled, nobody that
changes the number of delalloc bytes in the FS is running with irqs off.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Checksums on data can be disabled by mount option, so it's
possible some data extents don't have checksums or have
invalid checksums. This causes trouble for data relocation.
This patch contains following things to make data relocation
work.
1) make nodatasum/nodatacow mount option only affects new
files. Checksums and COW on data are only controlled by the
inode flags.
2) check the existence of checksum in the nodatacow checker.
If checksums exist, force COW the data extent. This ensure that
checksum for a given block is either valid or does not exist.
3) update data relocation code to properly handle the case
of checksum missing.
Signed-off-by: Yan Zheng <zheng.yan@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>
The block group structs are referenced in many different
places, and it's not safe to free while balancing. So, those block
group structs were simply leaked instead.
This patch replaces the block group pointer in the inode with the starting byte
offset of the block group and adds reference counting to the block group
struct.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This finishes off the new checksumming code by removing csum items
for extents that are no longer in use.
The trick is doing it without racing because a single csum item may
hold csums for more than one extent. Extra checks are added to
btrfs_csum_file_blocks to make sure that we are using the correct
csum item after dropping locks.
A new btrfs_split_item is added to split a single csum item so it
can be split without dropping the leaf lock. This is used to
remove csum bytes from the middle of an item.
Signed-off-by: Chris Mason <chris.mason@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>
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>
This the lockdep complaint by having a different mutex to gaurd caching the
block group, so you don't end up with this backwards dependancy. Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.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>
* open/close_bdev_excl -> open/close_bdev_exclusive
* blkdev_issue_discard takes a GFP mask now
* Fix blkdev_issue_discard usage now that it is enabled
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch fixes what I hope is the last early ENOSPC bug left. I did not know
that pinned extents would merge into one big extent when inserted on to the
pinned extent tree, so I was adding free space to a block group that could
possibly span multiple block groups.
This is a big issue because first that space doesn't exist in that block group,
and second we won't actually use that space because there are a bunch of other
checks to make sure we're allocating within the constraints of the block group.
This patch fixes the problem by adding the btrfs_add_free_space to
btrfs_update_pinned_extents which makes sure we are adding the appropriate
amount of free space to the appropriate block group. Thanks much to Lee Trager
for running my myriad of debug patches to help me track this problem down.
Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
In insert_extents(), when ret==1 and last is not zero, it should
check if the current inserted item is the last item in this batching
inserts. If so, it should just break from loop. If not, 'cur =
insert_list->next' will make no sense because the list is empty now,
and 'op' will point to an unexpectable place.
There are also some trivial fixs in this patch including one comment
typo error and deleting two redundant lines.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
In my batch delete/update/insert patch I introduced a free space leak. The
extent that we do the original search on in free_extents is never pinned, so we
always update the block saying that it has free space, but the free space never
actually gets added to the free space tree, since op->del will always be 0 and
it's never actually added to the pinned extents tree.
This patch fixes this problem by making sure we call pin_down_bytes on the
pending extent op and set op->del to the return value of pin_down_bytes so
update_block_group is called with the right value. This seems to fix the case
where we were getting ENOSPC when there was plenty of space available.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Seed device is a special btrfs with SEEDING super flag
set and can only be mounted in read-only mode. Seed
devices allow people to create new btrfs on top of it.
The new FS contains the same contents as the seed device,
but it can be mounted in read-write mode.
This patch does the following:
1) split code in btrfs_alloc_chunk into two parts. The first part does makes
the newly allocated chunk usable, but does not do any operation that modifies
the chunk tree. The second part does the the chunk tree modifications. This
division is for the bootstrap step of adding storage to the seed device.
2) Update device management code to handle seed device.
The basic idea is: For an FS grown from seed devices, its
seed devices are put into a list. Seed devices are
opened on demand at mounting time. If any seed device is
missing or has been changed, btrfs kernel module will
refuse to mount the FS.
3) make btrfs_find_block_group not return NULL when all
block groups are read-only.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This patch adds mount ro and remount support. The main
changes in patch are: adding btrfs_remount and related
helper function; splitting the transaction related code
out of close_ctree into btrfs_commit_super; updating
allocator to properly handle read only block group.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
While profiling the allocator I noticed a good amount of time was being spent in
finish_current_insert and del_pending_extents, and as the filesystem filled up
more and more time was being spent in those functions. This patch aims to try
and reduce that problem. This happens two ways
1) track if we tried to delete an extent that we are going to update or insert.
Once we get into finish_current_insert we discard any of the extents that were
marked for deletion. This saves us from doing unnecessary work almost every
time finish_current_insert runs.
2) Batch insertion/updates/deletions. Instead of doing a btrfs_search_slot for
each individual extent and doing the needed operation, we instead keep the leaf
around and see if there is anything else we can do on that leaf. On the insert
case I introduced a btrfs_insert_some_items, which will take an array of keys
with an array of data_sizes and try and squeeze in as many of those keys as
possible, and then return how many keys it was able to insert. In the update
case we search for an extent ref, update the ref and then loop through the leaf
to see if any of the other refs we are looking to update are on that leaf, and
then once we are done we release the path and search for the next ref we need to
update. And finally for the deletion we try and delete the extent+ref in pairs,
so we will try to find extent+ref pairs next to the extent we are trying to free
and free them in bulk if possible.
This along with the other cluster fix that Chris pushed out a bit ago helps make
the allocator preform more uniformly as it fills up the disk. There is still a
slight drop as we fill up the disk since we start having to stick new blocks in
odd places which results in more COW's than on a empty fs, but the drop is not
nearly as severe as it was before.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
When we fail to allocate a new block group, we should still do the
checks to make sure allocations try again with the minimum requested
allocation size.
This also fixes a deadlock that come from a missed down_read in
the chunk allocation failure handling.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The allocator wasn't catching all of the cases where it needed to do
extra loops because the check to enforce them wasn't happening early
enough.
When the allocator decided to increase the size of the allocation
for metadata clustering, it wasn't always setting the empty_size to
include the extra (optional) bytes. This also fixes the empty_size field
to be correct.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The loop searching for free space would exit out too soon when
metadata clustering was trying to allocate a large extent. This makes
sure a full scan of the free space is done searching for only the
minimum extent size requested by the higher layers.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
When metadata allocation clustering has to fall back to unclustered
allocs because large free areas could not be found, it was sometimes
substracting too much from the total bytes to allocate. This would
make it wrap below zero.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
In comes cases the empty cluster was added twice to the total number of
bytes the allocator was trying to find.
With empty clustering on, the hint byte was sometimes outside of the
block group. Add an extra goto to find the correct block group.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This lowers the empty cluster target for metadata allocations. The lower
target makes it easier to do allocations and still seems to perform well.
It also fixes the allocator loop to drop the empty cluster when things
start getting difficult, avoiding false enospc warnings.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The allocator uses the last allocation as a starting point for metadata
allocations, and tries to allocate in clusters of at least 256k.
If the search for a free block fails to find the expected block, this patch
forces a new cluster to be found in the free list.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
When reading compressed extents, try to put pages into the page cache
for any pages covered by the compressed extent that readpages didn't already
preload.
Add an async work queue to handle transformations at delayed allocation processing
time. Right now this is just compression. The workflow is:
1) Find offsets in the file marked for delayed allocation
2) Lock the pages
3) Lock the state bits
4) Call the async delalloc code
The async delalloc code clears the state lock bits and delalloc bits. It is
important this happens before the range goes into the work queue because
otherwise it might deadlock with other work queue items that try to lock
those extent bits.
The file pages are compressed, and if the compression doesn't work the
pages are written back directly.
An ordered work queue is used to make sure the inodes are written in the same
order that pdflush or writepages sent them down.
This changes extent_write_cache_pages to let the writepage function
update the wbc nr_written count.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch updates btrfs-progs for fallocate support.
fallocate is a little different in Btrfs because we need to tell the
COW system that a given preallocated extent doesn't need to be
cow'd as long as there are no snapshots of it. This leverages the
-o nodatacow checks.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This patch simplifies the nodatacow checker. If all references
were created after the latest snapshot, then we can avoid COW
safely. This patch also updates run_delalloc_nocow to do more
fine-grained checking.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
When dropping middle part of an extent, btrfs_drop_extents truncates
the extent at first, then inserts a bookend extent.
Since truncation and insertion can't be done atomically, there is a small
period that the bookend extent isn't in the tree. This causes problem for
functions that search the tree for file extent item. The way to fix this is
lock the range of the bookend extent before truncation.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
finish_current_insert and del_pending_extents process extent tree modifications
that build up while we are changing the extent tree. It is a confusing
bit of code that prevents recursion.
Both functions run through a list of pending operations and both funcs
add to the list of pending operations. If you have two procs in either
one of them, they can end up looping forever making more work for each other.
This patch makes them walk forward through the list of pending changes instead
of always trying to process the entire list. At transaction commit
time, we catch any changes that were left over.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch adds transaction IDs to root tree pointers.
Transaction IDs in tree pointers are compared with the
generation numbers in block headers when reading root
blocks of trees. This can detect some types of IO errors.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This patch removes the giant fs_info->alloc_mutex and replaces it with a bunch
of little locks.
There is now a pinned_mutex, which is used when messing with the pinned_extents
extent io tree, and the extent_ins_mutex which is used with the pending_del and
extent_ins extent io trees.
The locking for the extent tree stuff was inspired by a patch that Yan Zheng
wrote to fix a race condition, I cleaned it up some and changed the locking
around a little bit, but the idea remains the same. Basically instead of
holding the extent_ins_mutex throughout the processing of an extent on the
extent_ins or pending_del trees, we just hold it while we're searching and when
we clear the bits on those trees, and lock the extent for the duration of the
operations on the extent.
Also to keep from getting hung up waiting to lock an extent, I've added a
try_lock_extent so if we cannot lock the extent, move on to the next one in the
tree and we'll come back to that one. I have tested this heavily and it does
not appear to break anything. This has to be applied on top of my
find_free_extent redo patch.
I tested this patch on top of Yan's space reblancing code and it worked fine.
The only thing that has changed since the last version is I pulled out all my
debugging stuff, apparently I forgot to run guilt refresh before I sent the
last patch out. Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
So there is an odd case where we can possibly return -ENOSPC when there is in
fact space to be had. It only happens with Metadata writes, and happens _very_
infrequently. What has to happen is we have to allocate have allocated out of
the first logical byte on the disk, which would set last_alloc to
first_logical_byte(root, 0), so search_start == orig_search_start. We then
need to allocate for normal metadata, so BTRFS_BLOCK_GROUP_METADATA |
BTRFS_BLOCK_GROUP_DUP. We will do a block lookup for the given search_start,
block_group_bits() won't match and we'll go to choose another block group.
However because search_start matches orig_search_start we go to see if we can
allocate a chunk.
If we are in the situation that we cannot allocate a chunk, we fail and ENOSPC.
This is kind of a big flaw of the way find_free_extent works, as it along with
find_free_space loop through _all_ of the block groups, not just the ones that
we want to allocate out of. This patch completely kills find_free_space and
rolls it into find_free_extent. I've introduced a sort of state machine into
this, which will make it easier to get cache miss information out of the
allocator, and will work well with my locking changes.
The basic flow is this: We have the variable loop which is 0, meaning we are
in the hint phase. We lookup the block group for the hint, and lookup the
space_info for what we want to allocate out of. If the block group we were
pointed at by the hint either isn't of the correct type, or just doesn't have
the space we need, we set head to space_info->block_groups, so we start at the
beginning of the block groups for this particular space info, and loop through.
This is also where we add the empty_cluster to total_needed. At this point
loop is set to 1 and we just loop through all of the block groups for this
particular space_info looking for the space we need, just as find_free_space
would have done, except we only hit the block groups we want and not _all_ of
the block groups. If we come full circle we see if we can allocate a chunk.
If we cannot of course we exit with -ENOSPC and we are good. If not we start
over at space_info->block_groups and loop through again, with loop == 2. If we
come full circle and haven't found what we need then we exit with -ENOSPC.
I've been running this for a couple of days now and it seems stable, and I
haven't yet hit a -ENOSPC when there was plenty of space left.
Also I've made a groups_sem to handle the group list for the space_info. This
is part of my locking changes, but is relatively safe and seems better than
holding the space_info spinlock over that entire search time. Thanks,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
This patch improves the space balancing code to keep more sharing
of tree blocks. The only case that breaks sharing of tree blocks is
data extents get fragmented during balancing. The main changes in
this patch are:
Add a 'drop sub-tree' function. This solves the problem in old code
that BTRFS_HEADER_FLAG_WRITTEN check breaks sharing of tree block.
Remove relocation mapping tree. Relocation mappings are stored in
struct btrfs_ref_path and updated dynamically during walking up/down
the reference path. This reduces CPU usage and simplifies code.
This patch also fixes a bug. Root items for reloc trees should be
updated in btrfs_free_reloc_root.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This is a large change for adding compression on reading and writing,
both for inline and regular extents. It does some fairly large
surgery to the writeback paths.
Compression is off by default and enabled by mount -o compress. Even
when the -o compress mount option is not used, it is possible to read
compressed extents off the disk.
If compression for a given set of pages fails to make them smaller, the
file is flagged to avoid future compression attempts later.
* While finding delalloc extents, the pages are locked before being sent down
to the delalloc handler. This allows the delalloc handler to do complex things
such as cleaning the pages, marking them writeback and starting IO on their
behalf.
* Inline extents are inserted at delalloc time now. This allows us to compress
the data before inserting the inline extent, and it allows us to insert
an inline extent that spans multiple pages.
* All of the in-memory extent representations (extent_map.c, ordered-data.c etc)
are changed to record both an in-memory size and an on disk size, as well
as a flag for compression.
From a disk format point of view, the extent pointers in the file are changed
to record the on disk size of a given extent and some encoding flags.
Space in the disk format is allocated for compression encoding, as well
as encryption and a generic 'other' field. Neither the encryption or the
'other' field are currently used.
In order to limit the amount of data read for a single random read in the
file, the size of a compressed extent is limited to 128k. This is a
software only limit, the disk format supports u64 sized compressed extents.
In order to limit the ram consumed while processing extents, the uncompressed
size of a compressed extent is limited to 256k. This is a software only limit
and will be subject to tuning later.
Checksumming is still done on compressed extents, and it is done on the
uncompressed version of the data. This way additional encodings can be
layered on without having to figure out which encoding to checksum.
Compression happens at delalloc time, which is basically singled threaded because
it is usually done by a single pdflush thread. This makes it tricky to
spread the compression load across all the cpus on the box. We'll have to
look at parallel pdflush walks of dirty inodes at a later time.
Decompression is hooked into readpages and it does spread across CPUs nicely.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Due to the optimization for truncate, tree leaves only containing
checksum items can be deleted without being COW'ed first. This causes
reference cache misses. The way to fix the miss is create cache
entries for tree leaves only contain checksum.
This patch also fixes a -EEXIST issue in shared reference cache.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
The offset field in struct btrfs_extent_ref records the position
inside file that file extent is referenced by. In the new back
reference system, tree leaves holding references to file extent
are recorded explicitly. We can scan these tree leaves very quickly, so the
offset field is not required.
This patch also makes the back reference system check the objectid
when extents are in deleting.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This patch makes btrfs count space allocated to file in bytes instead
of 512 byte sectors.
Everything else in btrfs uses a byte count instead of sector sizes or
blocks sizes, so this fits better.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
The tree logging code was trying to separate tree log allocations
from normal metadata allocations to improve writeback patterns during
an fsync.
But, the code was not effective and ended up just mixing tree log
blocks with regular metadata. That seems to be working fairly well,
so the last_log_alloc code can be removed.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This fixes a deadlock that happens between the alloc_mutex and chunk_mutex.
Process A comes in, decides to do a do_chunk_alloc, which takes the
chunk_mutex, and is holding the alloc_mutex because the only way you get to
do_chunk_alloc is by holding the alloc_mutex. btrfs_alloc_chunk does its thing
and goes to insert a new item, which results in a cow of the block.
We get into del_pending_extents from there, where if we need to be rescheduled
we drop the alloc_mutex and schedule. At this point process B comes in to do
an allocation and gets the alloc_mutex, and because process A did not do the
chunk allocation completely it thinks its a good time to do a chunk allocation
as well, and hangs on the chunk_mutex.
Process A wakes up and tries to take the alloc_mutex and cannot. The way to
fix this is do a mutex_trylock() on chunk_mutex. If we return 0 we didn't get
the lock, and if this is just a "hey it may be a good time to allocate a chunk"
then we just exit. If we are trying to force an allocation then we reschedule
and keep trying to acquire the chunk_mutex. If once we acquire it the space is
already full then we can just exit, otherwise we can continue with the chunk
allocation. Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
When reading in block groups, a global mask of the available raid policies
should be adjusted based on the types of block groups found on disk. This
global mask is then used to decide which raid policy to use for new
block groups.
The recent allocator changes dropped the call that updated the global
mask, making all the block groups allocated at run time single striped
onto a single drive.
This also fixes the async worker threads to set any thread that uses
the requeue mechanism as busy. This allows us to avoid blocking
on get_request_wait for the async bio submission threads.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch fixes a problem where we end up seeking too much when *last_ptr is
valid. This happens because btrfs_lookup_first_block_group only returns a
block group that starts on or after the given search start, so if the
search_start is in the middle of a block group it will return the block group
after the given search_start, which is suboptimal.
This patch fixes that by doing a btrfs_lookup_block_group, which will return
the block group that contains the given search start. If we fail to find a
block group, we fall back on btrfs_lookup_first_block_group so we can find the
next block group, not sure if this is absolutely needed, but better safe than
sorry.
Also if we can't find the block group that we need, or it happens to not be of
the right type, we need to add empty_cluster since *last_ptr could point to a
mismatched block group, which means we need to start over with empty_cluster
added to total needed. Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch updates the space balancing code to utilize the new
backref format. Before, btrfs-vol -b would break any COW links
on data blocks or metadata. This was slow and caused the amount
of space used to explode if a large number of snapshots were present.
The new code can keeps the sharing of all data extents and
most of the tree blocks.
To maintain the sharing of data extents, the space balance code uses
a seperate inode hold data extent pointers, then updates the references
to point to the new location.
To maintain the sharing of tree blocks, the space balance code uses
reloc trees to relocate tree blocks in reference counted roots.
There is one reloc tree for each subvol, and all reloc trees share
same root key objectid. Reloc trees are snapshots of the latest
committed roots of subvols (root->commit_root).
To relocate a tree block referenced by a subvol, there are two steps.
COW the block through subvol's reloc tree, then update block pointer in
the subvol to point to the new block. Since all reloc trees share
same root key objectid, doing special handing for tree blocks
owned by them is easy. Once a tree block has been COWed in one
reloc tree, we can use the resulting new block directly when the
same block is required to COW again through other reloc trees.
In this way, relocated tree blocks are shared between reloc trees,
so they are also shared between subvols.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Btrfs has a cache of reference counts in leaves, allowing it to
avoid reading tree leaves while deleting snapshots. To reduce
contention with multiple subvolumes, this cache is private to each
subvolume.
This patch adds shared reference cache support. The new space
balancing code plays with multiple subvols at the same time, So
the old per-subvol reference cache is not well suited.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* Reserved extent accounting: reserved extents have been
allocated in the rbtrees that track free space but have not
been allocated on disk. They were never properly accounted for
in the past, making it hard to know how much space was really free.
* btrfs_find_block_group used to return NULL for block groups that
had been removed by the space balancing code. This made it hard
to account for space during the final stages of a balance run.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The code to free block groups needs to drop the space info spin lock
before calling btrfs_remove_free_space_cache (which can schedule).
This is safe because at unmount time, nobody else is going to play
with the block groups.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch makes the back reference system to explicit record the
location of parent node for all types of extents. The location of
parent node is placed into the offset field of backref key. Every
time a tree block is balanced, the back references for the affected
lower level extents are updated.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Tree log blocks are only reserved, and should not ever get fully
allocated on disk. This check makes sure they stay out of the
extent tree.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
1) replace the per fs_info extent_io_tree that tracked free space with two
rb-trees per block group to track free space areas via offset and size. The
reason to do this is because most allocations come with a hint byte where to
start, so we can usually find a chunk of free space at that hint byte to satisfy
the allocation and get good space packing. If we cannot find free space at or
after the given offset we fall back on looking for a chunk of the given size as
close to that given offset as possible. When we fall back on the size search we
also try to find a slot as close to the size we want as possible, to avoid
breaking small chunks off of huge areas if possible.
2) remove the extent_io_tree that tracked the block group cache from fs_info and
replaced it with an rb-tree thats tracks block group cache via offset. also
added a per space_info list that tracks the block group cache for the particular
space so we can lookup related block groups easily.
3) cleaned up the allocation code to make it a little easier to read and a
little less complicated. Basically there are 3 steps, first look from our
provided hint. If we couldn't find from that given hint, start back at our
original search start and look for space from there. If that fails try to
allocate space if we can and start looking again. If not we're screwed and need
to start over again.
4) small fixes. there were some issues in volumes.c where we wouldn't allocate
the rest of the disk. fixed cow_file_range to actually pass the alloc_hint,
which has helped a good bit in making the fs_mark test I run have semi-normal
results as we run out of space. Generally with data allocations we don't track
where we last allocated from, so everytime we did a data allocation we'd search
through every block group that we have looking for free space. Now searching a
block group with no free space isn't terribly time consuming, it was causing a
slight degradation as we got more data block groups. The alloc_hint has fixed
this slight degredation and made things semi-normal.
There is still one nagging problem I'm working on where we will get ENOSPC when
there is definitely plenty of space. This only happens with metadata
allocations, and only when we are almost full. So you generally hit the 85%
mark first, but sometimes you'll hit the BUG before you hit the 85% wall. I'm
still tracking it down, but until then this seems to be pretty stable and make a
significant performance gain.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
cache block group had a few bugs in the error handling code,
this makes sure paths get properly released and the correct return value
goes out.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This is the same way the transaction code makes sure that all the
other tree blocks are safely on disk. There's an extent_io tree
for each root, and any blocks allocated to the tree logs are
recorded in that tree.
At tree-log sync, the extent_io tree is walked to flush down the
dirty pages and wait for them.
The main benefit is less time spent walking the tree log and skipping
clean pages, and getting sequential IO down to the drive.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Since tree log blocks get freed every transaction, they never really
need to be written to disk. This skips the step where we update
metadata to record they were allocated.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* Pin down data blocks to prevent them from being reallocated like so:
trans 1: allocate file extent
trans 2: free file extent
trans 3: free file extent during old snapshot deletion
trans 3: allocate file extent to new file
trans 3: fsync new file
Before the tree logging code, this was legal because the fsync
would commit the transation that did the final data extent free
and the transaction that allocated the extent to the new file
at the same time.
With the tree logging code, the tree log subtransaction can commit
before the transaction that freed the extent. If we crash,
we're left with two different files using the extent.
* Don't wait in start_transaction if log replay is going on. This
avoids deadlocks from iput while we're cleaning up link counts in the
replay code.
* Don't deadlock in replay_one_name by trying to read an inode off
the disk while holding paths for the directory
* Hold the buffer lock while we mark a buffer as written. This
closes a race where someone is changing a buffer while we write it.
They are supposed to mark it dirty again after they change it, but
this violates the cow rules.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
File syncs and directory syncs are optimized by copying their
items into a special (copy-on-write) log tree. There is one log tree per
subvolume and the btrfs super block points to a tree of log tree roots.
After a crash, items are copied out of the log tree and back into the
subvolume. See tree-log.c for all the details.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Far from the perfect fix, but these structs are small. TODO for the
next release. The block group cache structs are referenced in many
different places, and it isn't safe to just free them while resizing.
A real fix will be a larger change to the allocator so that it doesn't
have to carry about the block group cache structs to find good places
to search for free blocks.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* Make walk_down_tree wake up throttled tasks more often
* Make walk_down_tree call cond_resched during long loops
* As the size of the ref cache grows, wait longer in throttle
* Get rid of the reada code in walk_down_tree, the leaves don't get
read anymore, thanks to the ref cache.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
A btree block cow has two parts, the first is to allocate a destination
block and the second is to copy the old bock over.
The first part needs locks in the extent allocation tree, and may need to
do IO. This changeset splits that into a separate function that can be
called without any tree locks held.
btrfs_search_slot is changed to drop its path and start over if it has
to COW a contended block. This often means that many writers will
pre-alloc a new destination for a the same contended block, but they
cache their prealloc for later use on lower levels in the tree.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
While dropping snapshots, walk_down_tree does most of the work of checking
reference counts and limiting tree traversal to just the blocks that
we are freeing.
It dropped and held the allocation mutex in strange and confusing ways,
this commit changes it to only hold the mutex while actually freeing a block.
The rest of the checks around reference counts should be safe without the lock
because we only allow one process in btrfs_drop_snapshot at a time. Other
processes dropping reference counts should not drop it to 1 because
their tree roots already have an extra ref on the block.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This avoids waiting for transactions with pages locked by breaking out
the code to wait for the current transaction to close into a function
called by btrfs_throttle.
It also lowers the limits for where we start throttling.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The memory reclaiming issue happens when snapshot exists. In that
case, some cache entries may not be used during old snapshot dropping,
so they will remain in the cache until umount.
The patch adds a field to struct btrfs_leaf_ref to record create time. Besides,
the patch makes all dead roots of a given snapshot linked together in order of
create time. After a old snapshot was completely dropped, we check the dead
root list and remove all cache entries created before the oldest dead root in
the list.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
To check whether a given file extent is referenced by multiple snapshots, the
checker walks down the fs tree through dead root and checks all tree blocks in
the path.
We can easily detect whether a given tree block is directly referenced by other
snapshot. We can also detect any indirect reference from other snapshot by
checking reference's generation. The checker can always detect multiple
references, but can't reliably detect cases of single reference. So btrfs may
do file data cow even there is only one reference.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
A large reference cache is directly related to a lot of work pending
for the cleaner thread. This throttles back new operations based on
the size of the reference cache so the cleaner thread will be able to keep
up.
Overall, this actually makes the FS faster because the cleaner thread will
be more likely to find things in cache.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This changes the reference cache to make a single cache per root
instead of one cache per transaction, and to key by the byte number
of the disk block instead of the keys inside.
This makes it much less likely to have cache misses if a snapshot
or something has an extra reference on a higher node or a leaf while
the first transaction that added the leaf into the cache is dropping.
Some throttling is added to functions that free blocks heavily so they
wait for old transactions to drop.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Much of the IO done while dropping snapshots is done looking up
leaves in the filesystem trees to see if they point to any extents and
to drop the references on any extents found.
This creates a cache so that IO isn't required.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Before setting an extent to delalloc, the code needs to wait for
pending ordered extents.
Also, the relocation code needs to wait for ordered IO before scanning
the block group again. This is because the extents are not removed
until the IO for the new extents is finished
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This releases the alloc_mutex in a few places that hold it for over long
operations. btrfs_lookup_block_group is changed so that it doesn't need
the mutex at all.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This replaces the use of the page cache lock bit for locking, which wasn't
suitable for block size < page size and couldn't be used recursively.
The mutexes alone don't fix either problem, but they are the first step.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* In btrfs_delete_inode, wait for ordered extents after calling
truncate_inode_pages. This is much faster, and more correct
* Properly clear our the PageChecked bit everywhere we redirty the page.
* Change the writepage fixup handler to lock the page range and check to
see if an ordered extent had been inserted since the improperly dirtied
page was discovered
* Wait for ordered extents outside the transaction. This isn't required
for locking rules but does improve transaction latencies
* Reduce contention on the alloc_mutex by dropping it while incrementing
refs on a node/leaf and while dropping refs on a leaf.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
btrfs_drop_extents is always called with a range lock held on the inode.
But, it may operate on extents outside that range as it drops and splits
them.
This patch adds a per-inode mutex that is held while calling
btrfs_drop_extents and while inserting new extents into the tree. It
prevents races from two procs working against adjacent ranges in the tree.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The old data=ordered code would force commit to wait until
all the data extents from the transaction were fully on disk. This
introduced large latencies into the commit and stalled new writers
in the transaction for a long time.
The new code changes the way data allocations and extents work:
* When delayed allocation is filled, data extents are reserved, and
the extent bit EXTENT_ORDERED is set on the entire range of the extent.
A struct btrfs_ordered_extent is allocated an inserted into a per-inode
rbtree to track the pending extents.
* As each page is written EXTENT_ORDERED is cleared on the bytes corresponding
to that page.
* When all of the bytes corresponding to a single struct btrfs_ordered_extent
are written, The previously reserved extent is inserted into the FS
btree and into the extent allocation trees. The checksums for the file
data are also updated.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The btree defragger wasn't making forward progress because the new key wasn't
being saved by the btrfs_search_forward function.
This also disables the automatic btree defrag, it wasn't scaling well to
huge filesystems. The auto-defrag needs to be done differently.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This makes it possible for callers to check for extent_buffers in cache
without deadlocking against any btree locks held.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The online btree defragger is simplified and rewritten to use
standard btree searches instead of a walk up / down mechanism.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This creates one kthread for commits and one kthread for
deleting old snapshots. All the work queues are removed.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Allocations may need to read in block groups from the extent allocation tree,
which will require a tree search and take locks on the extent allocation
tree. But, those locks might already be held in other places, leading
to deadlocks.
Since the alloc_mutex serializes everything right now, it is safe to
skip the btree locking while caching block groups. A better fix will be
to either create a recursive lock or find a way to back off existing
locks while caching block groups.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
One lock per btree block can make for significant congestion if everyone
has to wait for IO at the high levels of the btree. This drops
locks held by a path when doing reads during a tree search.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Extent alloctions are still protected by a large alloc_mutex.
Objectid allocations are covered by a objectid mutex
Other btree operations are protected by a lock on individual btree nodes
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The allocation trees and the chunk trees are serialized via their own
dedicated mutexes. This means allocation location is still not very
fine grained.
The main FS btree is protected by locks on each block in the btree. Locks
are taken top / down, and as processing finishes on a given level of the
tree, the lock is released after locking the lower level.
The end result of a search is now a path where only the lowest level
is locked. Releasing or freeing the path drops any locks held.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* Force chunk allocation when find_free_extent has to do a full scan
* Record the max key at the start of defrag so it doesn't run forever
* Block groups might not be contiguous, make a forward search for the
next block group in extent-tree.c
* Get rid of extra checks for total fs size
* Fix relocate_one_reference to avoid relocating the same file data block
twice when referenced by an older transaction
* Use the open device count when allocating chunks so that we don't
try to allocate from devices that don't exist
Signed-off-by: Chris Mason <chris.mason@oracle.com>
When duplicate copies exist, writes are allowed to fail to one of those
copies. This changeset includes a few changes that allow the FS to
continue even when some IOs fail.
It also adds verification of the parent generation number for btree blocks.
This generation is stored in the pointer to a block, and it ensures
that missed writes to are detected.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Once part of a delalloc request fails the cow checks, just cow the
entire range
It is possible for the back references to all be from the same root,
but still have snapshots against an extent. The checks are now more strict,
forcing cow any time there are multiple refs against the data extent.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Before, nodatacow only checked to make sure multiple roots didn't have
references on a single extent. This check makes sure that multiple
inodes don't have references.
nodatacow needed an extra check to see if the block group was currently
readonly. This way cows forced by the chunk relocation code are honored.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This required a few structural changes to the code that manages bdev pointers:
The VFS super block now gets an anon-bdev instead of a pointer to the
lowest bdev. This allows us to avoid swapping the super block bdev pointer
around at run time.
The code to read in the super block no longer goes through the extent
buffer interface. Things got ugly keeping the mapping constant.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
balance level starts by trying to empty the middle block, and then
pushes from the right to the middle. This might empty the right block
and leave a small number of pointers in the middle.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The data read retry code needs to find the logical disk block before it
can resubmit new bios. But, finding this block isn't allowed to take
the fs_mutex because that will deadlock with a number of different callers.
This changes the retry code to use the extent map cache instead, but
that requires the extent map cache to have the extent we're looking for.
This is a problem because btrfs_drop_extent_cache just drops the entire
extent instead of the little tiny part it is invalidating.
The bulk of the code in this patch changes btrfs_drop_extent_cache to
invalidate only a portion of the extent cache, and changes btrfs_get_extent
to deal with the results.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This isn't required anymore because we don't reallocate blocks that
have already been written in this transaction.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Block headers now store the chunk tree uuid
Chunk items records the device uuid for each stripes
Device extent items record better back refs to the chunk tree
Block groups record better back refs to the chunk tree
The chunk tree format has also changed. The objectid of BTRFS_CHUNK_ITEM_KEY
used to be the logical offset of the chunk. Now it is a chunk tree id,
with the logical offset being stored in the offset field of the key.
This allows a single chunk tree to record multiple logical address spaces,
upping the number of bytes indexed by a chunk tree from 2^64 to
2^128.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
On huge machines, delayed allocation may try to allocate massive extents.
This change allows btrfs_alloc_extent to return something smaller than
the caller asked for, and the data allocation routines will loop over
the allocations until it fills the whole delayed alloc.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Fix for a endianess BUG when using btrfs v0.13 with kernels older than 2.6.23
Problem:
Has of v0.13, btrfs-progs is using crc32c.c equivalent to the one found on
linux-2.6.23/lib/libcrc32c.c Since crc32c_le() changed in linux-2.6.23, when
running btrfs v0.13 with older kernels we have a missmatch between the versions
of crc32c_le() from btrfs-progs and libcrc32c in the kernel. This missmatch
causes a bug when using btrfs on big endian machines.
Solution:
btrfs_crc32c() macro that when compiling for kernels older than 2.6.23, does
endianess conversion to parameters and return value of crc32c().
This endianess conversion nullifies the differences in implementation
of crc32c_le().
If kernel 2.6.23 or better, it calls crc32c().
Signed-off-by: Miguel Sousa Filipe <miguel.filipe@gmail.com>
---
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Before, metadata checksumming was done by the callers of read_tree_block,
which would set EXTENT_CSUM bits in the extent tree to show that a given
range of pages was already checksummed and didn't need to be verified
again.
But, those bits could go away via try_to_releasepage, and the end
result was bogus checksum failures on pages that never left the cache.
The new code validates checksums when the page is read. It is a little
tricky because metadata blocks can span pages and a single read may
end up going via multiple bios.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
When a block is freed, it can be immediately reused if it is from
the current transaction. But, an extra check is required to make sure
the block had not been written yet. If it were reused after being written,
the transid in the block header might match the transid of the
next time the block was allocated.
The parent node records the transaction ID of the block it is pointing to,
and this is used as part of validating the block on reads. So, there
can only be one version of a block per transaction.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Yan, Zheng