summaryrefslogtreecommitdiff
path: root/fs/btrfs/transaction.c
AgeCommit message (Collapse)AuthorFilesLines
2008-09-25Btrfs: New data=ordered implementationChris Mason1-58/+9
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>
2008-09-25Btrfs: Drop some verbose printksChris Mason1-2/+0
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Online btree defragmentation fixesChris Mason1-34/+1
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>
2008-09-25Btrfs: Add a per-inode csum mutex to avoid races creating csum itemsChris Mason1-1/+1
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Replace the transaction work queue with kthreadsChris Mason1-70/+2
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>
2008-09-25Add btrfs_end_transaction_throttle to force writers to wait for pending commitsChris Mason1-21/+45
The existing throttle mechanism was often not sufficient to prevent new writers from coming in and making a given transaction run forever. This adds an explicit wait at the end of most operations so they will allow the current transaction to close. There is no wait inside file_write, inode updates, or cow filling, all which have different deadlock possibilities. This is a temporary measure until better asynchronous commit support is added. This code leads to stalls as it waits for data=ordered writeback, and it really needs to be fixed. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Replace the big fs_mutex with a collection of other locksChris Mason1-26/+16
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>
2008-09-25Btrfs: Start btree concurrency work.Chris Mason1-6/+9
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>
2008-09-25Btrfs: Invalidate dcache entry after creating snapshot andSven Wegener1-1/+7
We need to invalidate an existing dcache entry after creating a new snapshot or subvolume, because a negative dache entry will stop us from accessing the new snapshot or subvolume. --- ctree.h | 23 +++++++++++++++++++++++ inode.c | 4 ++++ transaction.c | 4 ++++ 3 files changed, 31 insertions(+) Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Fix race in running_transaction checksChris Mason1-1/+3
When a new transaction was started, the code would incorrectly set the pointer in fs_info before all the data structures were setup. fsync heavy workloads hit races on the setup of the ordered inode spinlock Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Add support for online device removalChris Mason1-3/+2
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>
2008-09-25Btrfs: Fixes for 2.6.18 enterprise kernelsChris Mason1-2/+6
2.6.18 seems to get caught in an infinite loop when cancel_rearming_delayed_workqueue is called more than once, so this switches to cancel_delayed_work, which is arguably more correct. Also, balance_dirty_pages can run into problems with 2.6.18 based kernels because it doesn't have the per-bdi dirty limits. This avoids calling balance_dirty_pages on the btree inode unless there is actually something to balance, which is a good optimization in general. Finally there's a compile fix for ordered-data.h Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Throttle file_write when data=ordered is flushing the inodeChris Mason1-2/+8
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Do metadata checksums for reads via a workqueueChris Mason1-1/+1
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>
2008-09-25Btrfs: Add support for multiple devices per filesystemChris Mason1-16/+34
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Lower stack usage in transaction.cChris Mason1-18/+24
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Add data block hints to SSD mode tooChris Mason1-0/+1
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Split the extent_map code into two partsChris Mason1-4/+4
There is now extent_map for mapping offsets in the file to disk and extent_io for state tracking, IO submission and extent_bufers. The new extent_map code shifts from [start,end] pairs to [start,len], and pushes the locking out into the caller. This allows a few performance optimizations and is easier to use. A number of extent_map usage bugs were fixed, mostly with failing to remove extent_map entries when changing the file. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Add mount -o ssd, which includes optimizations for seek free storageChris Mason1-0/+1
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Fix data=ordered vs wait_on_inode deadlock on older kernelsChris Mason1-17/+13
Using ilookup5 during data=ordered writeback could deadlock on I_LOCK. This saves a pointer to the inode instead. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Run igrab on data=ordered inodes to prevent deadlocks during writeoutChris Mason1-0/+1
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Rework btrfs_drop_inode to avoid schedulingChris Mason1-0/+2
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Add some simple throttling to wait for data=ordered and snapshot deletionChris Mason1-0/+4
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Move snapshot creation to commit timeChris Mason1-3/+78
It is very difficult to create a consistent snapshot of the btree when other writers may update the btree before the commit is done. This changes the snapshot creation to happen during the commit, while no other updates are possible. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Add data=ordered supportChris Mason1-0/+58
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>
2008-09-25Btrfs: Reduce stack usage in the resizer, fix 32 bit compilesChris Mason1-5/+15
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Back port to 2.6.18-el kernelsChris Mason1-0/+8
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: section mismatch warningsChristian Hesse1-1/+1
--Boundary-00=_CcOWHFYK4T+JwSj Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Hello everybody, compiling btrfs into the kernel results in section mismatch warnings. __exit functions are called where they are not allowed to. The attached patch fixes this for me. Not sure if it is correct though. Signed-off-by: Christian Hesse <mail@earthworm.de> -- Regards, Chris --Boundary-00=_CcOWHFYK4T+JwSj Content-Type: text/x-diff; charset="iso-8859-1"; name="btrfs-section_mismatches.patch" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="btrfs-section_mismatches.patch" Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Fix PAGE_CACHE_SHIFT shifts on 32 bit machinesChris Mason1-1/+1
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Defrag only leaves, and only when the parent node has a single objectidChris Mason1-2/+0
This allows us to defrag huge directories, but skip the expensive defrag case in more common usage, where it does not help as much. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Add an extent buffer LRU to reduce radix tree hitsChris Mason1-3/+2
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Add back the online defragging codeChris Mason1-2/+2
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Allow tree blocks larger than the page sizeChris Mason1-17/+22
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Change the remaining radix trees used by extent-tree.c to extent_map ↵Chris Mason1-2/+3
trees Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Fix extent_buffer and extent_state leaksChris Mason1-2/+3
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Create extent_buffer interface for large blocksizesChris Mason1-27/+34
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-09-25Btrfs: Use balance_dirty_pages_nr on btree blocksChris Mason1-4/+10
btrfs_btree_balance_dirty is changed to pass the number of pages dirtied for more accurate dirty throttling. This lets the VM make better decisions about when to force some writeback. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-09-11Btrfs: Find and remove dead roots the first time a root is loaded.Chris Mason1-2/+4
Dead roots are trees left over after a crash, and they were either in the process of being removed or were waiting to be removed when the box crashed. Before, a search of the entire tree of root pointers was done on mount looking for dead roots. Now, the search is done the first time we load a root. This makes mount faster when there are a large number of snapshots, and it enables the block accounting code to properly update the block counts on the latest root as old versions of the root are reaped after a crash. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-08-29Btrfs: Add per-root block accounting and sysfs entriesJosef Bacik1-1/+33
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-08-11Btrfs: delay commits during fsync to allow more writersJosef Bacik1-12/+32
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-08-10Btrfs: Btree defrag on the extent-mapping tree as wellChris Mason1-42/+49
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-08-09Btrfs: Further reduce the concurrency penalty of defrag and drop_snapshotChris Mason1-0/+3
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-08-09Btrfs: Replace extent tree preallocation code with some bit radix magic.Chris Mason1-0/+2
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-08-08Btrfs: Let some locks go during defrag and snapshot droppingChris Mason1-0/+8
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-08-08Btrfs: Add run time btree defrag, and an ioctl to force btree defragChris Mason1-18/+87
This adds two types of btree defrag, a run time form that tries to defrag recently allocated blocks in the btree when they are still in ram, and an ioctl that forces defrag of all btree blocks. File data blocks are not defragged yet, but this can make a huge difference in sequential btree reads. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-08-07Btrfs: Do snapshot deletion in smaller chunks.Chris Mason1-25/+38
Before, snapshot deletion was a single atomic unit. This caused considerable lock contention and required an unbounded amount of space. Now, the drop_progress field in the root item is used to indicate how far along snapshot deletion is, and to resume where it left off. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-07-11Btrfs: trivial include fixupsZach Brown1-1/+0
Almost none of the files including module.h need to do so, remove them. Include sched.h in extent-tree.c to silence a warning about cond_resched() being undeclared. Signed-off-by: Zach Brown <zach.brown@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-06-28Btrfs: crash recovery fixesChris Mason1-20/+34
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-06-26Btrfs: Fix super block updates during transaction commitChris Mason1-2/+6
The super block written during commit was not consistent with the state of the trees. This change adds an in-memory copy of the super so that we can make sure to write out consistent data during a commit. Signed-off-by: Chris Mason <chris.mason@oracle.com>
2007-06-22Btrfs: Documentation updateChris Mason1-1/+0
Signed-off-by: Chris Mason <chris.mason@oracle.com>
generated by cgit v1.2.3 (git 2.39.0) at 2025-09-24 09:34:38 +0300