Commit Graph

143 Commits

Author SHA1 Message Date
Chris Mason 65b51a009e btrfs_search_slot: reduce lock contention by cowing in two stages
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>
2008-09-25 11:04:06 -04:00
Yan bcc63abbf3 Btrfs: implement memory reclaim for leaf reference cache
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>
2008-09-25 11:04:05 -04:00
Yan Zheng 31153d8128 Btrfs: Add a leaf reference cache
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>
2008-09-25 11:04:05 -04:00
Yan 9652480bf4 Fix path slots selection in btrfs_search_forward
We should decrease the found slot by one as btrfs_search_slot does
when bin_search return 1 and node level > 0.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:05 -04:00
Josef Bacik 7b12876623 Btrfs: Create orphan inode records to prevent lost files after a crash
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:05 -04:00
Chris Mason 0bd40a7184 btrfs_next_leaf: do readahead when skip_locking is turned on
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:04 -04:00
Chris Mason 7d9eb12c87 Btrfs: Add locking around volume management (device add/remove/balance)
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:04 -04:00
Chris Mason f9efa9c784 Btrfs: Reduce contention on the root node
This calls unlock_up sooner in btrfs_search_slot in order to decrease the
amount of work done with the higher level tree locks held.

Also, it changes btrfs_tree_lock to spin for a big against the page lock
before scheduling.  This makes a big difference in context switch rate under
highly contended workloads.

Longer term, a better locking structure is needed than the page lock.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:04 -04:00
Chris Mason 3f157a2fd2 Btrfs: Online btree defragmentation fixes
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>
2008-09-25 11:04:04 -04:00
Chris Mason e7a84565bc Btrfs: Add btree locking to the tree defragmentation code
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>
2008-09-25 11:04:03 -04:00
Chris Mason a74a4b97b6 Btrfs: Replace the transaction work queue with kthreads
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>
2008-09-25 11:04:03 -04:00
Chris Mason 333db94cdd Btrfs: Fix snapshot deletion to release the alloc_mutex much more often.
This lowers the impact of snapshot deletion on the rest of the FS.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:03 -04:00
Chris Mason 5cd57b2cbb Btrfs: Add a skip_locking parameter to struct path, and make various funcs honor it
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>
2008-09-25 11:04:03 -04:00
Chris Mason 168fd7d271 Fix btrfs_next_leaf to check for new items after dropping locks
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:03 -04:00
Chris Mason 594a24eb0e Fix btrfs_del_ordered_inode to allow forcing the drop during unlinks
This allows us to delete an unlinked inode with dirty pages from the list
instead of forcing commit to write these out before deleting the inode.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:03 -04:00
Chris Mason 051e1b9f74 Drop locks in btrfs_search_slot when reading a tree block.
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>
2008-09-25 11:04:03 -04:00
Chris Mason a213501153 Btrfs: Replace the big fs_mutex with a collection of other locks
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>
2008-09-25 11:04:03 -04:00
Chris Mason 925baeddc5 Btrfs: Start btree concurrency work.
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>
2008-09-25 11:04:03 -04:00
Chris Mason 0ef3e66b67 Btrfs: Allocator fix variety pack
* 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>
2008-09-25 11:04:03 -04:00
Chris Mason 1259ab75c6 Btrfs: Handle write errors on raid1 and raid10
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>
2008-09-25 11:04:03 -04:00
Chris Mason ca7a79ad8d Btrfs: Pass down the expected generation number when reading tree blocks
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:03 -04:00
Chris Mason bce4eae986 Btrfs: Fix balance_level to free the middle block if there is room in the left one
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>
2008-09-25 11:04:02 -04:00
Chris Mason 971a1f6648 Btrfs: Don't empty the middle buffer in push_nodes_for_insert
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:02 -04:00
Chris Mason c448acf0a0 Btrfs: Fix split_node to require more empty slots in the node as well
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:02 -04:00
Chris Mason 1514794e42 Btrfs: Make sure nodes have enough room for a double split
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:02 -04:00
Chris Mason 699122f559 Btrfs: Don't wait on tree block writeback before freeing them anymore
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>
2008-09-25 11:04:01 -04:00
Chris Mason e17cade25f Btrfs: Add chunk uuids and update multi-device back references
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>
2008-09-25 11:04:01 -04:00
Chris Mason 85d824c4a4 Btrfs: Disable extra debugging checks on tree blocks
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:01 -04:00
Chris Mason f188591e98 Btrfs: Retry metadata reads in the face of checksum failures
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:01 -04:00
Chris Mason ce9adaa5a7 Btrfs: Do metadata checksums for reads via a workqueue
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>
2008-09-25 11:04:01 -04:00
Chris Mason cea9e4452e Change btrfs_map_block to return a structure with mappings for all stripes
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:01 -04:00
Chris Mason 0ef8b2428a Btrfs: Properly dirty buffers in the split corner cases
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:01 -04:00
Chris Mason 0999df54f8 Btrfs: Verify checksums on tree blocks found without read_tree_block
Checksums were only verified by btrfs_read_tree_block, which meant the
functions to probe the page cache for blocks were not validating checksums.
Normally this is fine because the buffers will only be in cache if they
have already been validated.

But, there is a window while the buffer is being read from disk where
it could be up to date in the cache but not yet verified.  This patch
makes sure all buffers go through checksum verification before they
are used.

This is safer, and it prevents modification of buffers before they go
through the csum code.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:01 -04:00
Chris Mason 63b10fc487 Reorder the flags field in struct btrfs_header and record a flag on writeout
This allows detection of blocks that have already been written in the
running transaction so they can be recowed instead of modified again.
It is step one in trusting the transid field of the block pointers.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:01 -04:00
Chris Mason 0b86a832a1 Btrfs: Add support for multiple devices per filesystem
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:00 -04:00
Yan 2f375ab9c5 Call btrfs_cow_block while lowering tree level.
When freeing root block of a tree,  btrfs_free_extent' parameter
'ref_generation' is from root block itseft.  When freeing non-root
block,  'ref_generation' is from its parent. so when converting a
non-root block to root block, we must guarantee its generation is
equal to its parent's generation.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:00 -04:00
Chris Mason 5a01a2e3a9 Btrfs: Copy correct tree when inserting into slot 0
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:00 -04:00
Chris Mason 9c58309d6c Btrfs: Add inode item and backref in one insert, reducing cpu usage
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:00 -04:00
Chris Mason 85e21bac16 Btrfs: During deletes and truncate, remove many items at once from the tree
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:04:00 -04:00
Chris Mason dc17ff8f11 Btrfs: Add data=ordered support
This forces file data extents down the disk along with the metadata that
references them.  The current implementation is fairly simple, and just
writes out all of the dirty pages in an inode before the commit.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:59 -04:00
Chris Mason 98ed51747b Btrfs: Force inlining off in a few places to save stack usage
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:58 -04:00
Chris Mason 8f662a76c6 Btrfs: Add readahead to the online shrinker, and a mount -o alloc_start= for testing
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:58 -04:00
Chris Mason 01f4665805 Btrfs: Less aggressive readahead on deletes
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:58 -04:00
Chris Mason 4aec2b5232 kmalloc a few large stack objects in the btrfs_ioctl path
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:58 -04:00
Chris Mason be20aa9dba Btrfs: Add mount option to turn off data cow
A number of workloads do not require copy on write data or checksumming.
mount -o nodatasum to disable checksums and -o nodatacow to disable
both copy on write and checksumming.

In nodatacow mode, copy on write is still performed when a given extent
is under snapshot.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:58 -04:00
Chris Mason 7bb86316c3 Btrfs: Add back pointers from extents to the btree or file referencing them
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:58 -04:00
Chris Mason 74493f7a59 Btrfs: Implement generation numbers in block pointers
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:58 -04:00
Yan eef1c494a2 Btrfs: Properly update right_nritems in push_leaf_left
The codes that fixup the right leaf and the codes that dirty the
extnet buffer use the variable 'right_nritems' ,  both of them expect
'right_nritems' is the number of items in right leaf after the push.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:58 -04:00
Chris Mason 34a3821873 Btrfs: Change push_leaf_{leaf,right} to empty the src leave during item deletion
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:57 -04:00
Chris Mason 081e95736d Btrfs: Make defrag check nodes against the progress key to prevent repeating work
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25 11:03:57 -04:00