aboutsummaryrefslogtreecommitdiff
path: root/fs/btrfs
AgeCommit message (Collapse)AuthorFilesLines
2012-01-16Btrfs: do not reduce profile in do_chunk_alloc()Ilya Dryomov2-1/+19
Every caller of do_chunk_alloc() feeds it the reduced allocation profile, so stop trying to reduce it one more time. Instead check the validity of the passed profile. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: virtual address space subset filterIlya Dryomov2-0/+21
Select chunks which have at least one byte located inside a given [vstart, vend) virtual address space range. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: devid subset filterIlya Dryomov2-0/+47
Select chunks which have at least one byte of at least one stripe located on a device with devid X in a given [pstart,pend) physical address range. This filter only works when devid filter is turned on. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: devid filterIlya Dryomov2-0/+24
Relocate chunks which have at least one stripe located on a device with devid X. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: usage filterIlya Dryomov2-0/+37
Select chunks that are less than X percent full. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: profiles filterIlya Dryomov2-0/+28
Select chunks based on a given profile mask. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: add basic infrastructure for selective balancingIlya Dryomov3-2/+71
This allows to have a separate set of filters for each chunk type (data,meta,sys). The code however is generic and switch on chunk type is only done once. This commit also adds a type filter: it allows to balance for example meta and system chunks w/o touching data ones. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: add basic restriper infrastructureIlya Dryomov6-41/+281
Add basic restriper infrastructure: extended balancing ioctl and all related ioctl data structures, add data structure for tracking restriper's state to fs_info, etc. The semantics of the old balancing ioctl are fully preserved. Explicitly disallow any volume operations when balance is in progress. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: make avail_*_alloc_bits fields dynamicIlya Dryomov1-0/+20
Currently when new chunks are created respective avail_alloc_bits field is updated to reflect profiles of all chunks present in the system. However when chunks are removed profile bits are never cleared. This patch clears profile bit of respective avail_alloc_bits field when the last chunk with that profile is removed. Restriper needs this to properly operate when "downgrading". Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: add BTRFS_AVAIL_ALLOC_BIT_SINGLE bitIlya Dryomov2-9/+36
Right now on-disk BTRFS_BLOCK_GROUP_* profile bits are used for avail_{data,metadata,system}_alloc_bits fields, which gather info about available allocation profiles in the FS. When chunk is created or read from disk, its profile is OR'ed with the corresponding avail_alloc_bits field. Since SINGLE is denoted by 0 in the on-disk format, currently there is no way to tell when such chunks become avaialble. Restriper needs that information, so add a separate bit for SINGLE profile. This bit is going to be in-memory only, it should never be written out to disk, so it's not a disk format change. However to avoid remappings in future, reserve corresponding on-disk bit. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: introduce masks for chunk type and profileIlya Dryomov3-27/+22
Chunk's type and profile are encoded in u64 flags field. Introduce masks to easily access them. Also fix the type of BTRFS_BLOCK_GROUP_* constants, it should be ULL. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-16Btrfs: get rid of *_alloc_profile fieldsIlya Dryomov4-16/+6
{data,metadata,system}_alloc_profile fields have been unused for a long time now. Get rid of them. Signed-off-by: Ilya Dryomov <[email protected]>
2012-01-12mm: compaction: introduce sync-light migration for use by compactionMel Gorman1-2/+3
This patch adds a lightweight sync migrate operation MIGRATE_SYNC_LIGHT mode that avoids writing back pages to backing storage. Async compaction maps to MIGRATE_ASYNC while sync compaction maps to MIGRATE_SYNC_LIGHT. For other migrate_pages users such as memory hotplug, MIGRATE_SYNC is used. This avoids sync compaction stalling for an excessive length of time, particularly when copying files to a USB stick where there might be a large number of dirty pages backed by a filesystem that does not support ->writepages. [[email protected]: This patch is heavily based on Andrea's work] [[email protected]: fix fs/nfs/write.c build] [[email protected]: fix fs/btrfs/disk-io.c build] Signed-off-by: Mel Gorman <[email protected]> Reviewed-by: Rik van Riel <[email protected]> Cc: Andrea Arcangeli <[email protected]> Cc: Minchan Kim <[email protected]> Cc: Dave Jones <[email protected]> Cc: Jan Kara <[email protected]> Cc: Andy Isaacson <[email protected]> Cc: Nai Xia <[email protected]> Cc: Johannes Weiner <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2012-01-12mm: compaction: determine if dirty pages can be migrated without blocking ↵Mel Gorman1-2/+2
within ->migratepage Asynchronous compaction is used when allocating transparent hugepages to avoid blocking for long periods of time. Due to reports of stalling, there was a debate on disabling synchronous compaction but this severely impacted allocation success rates. Part of the reason was that many dirty pages are skipped in asynchronous compaction by the following check; if (PageDirty(page) && !sync && mapping->a_ops->migratepage != migrate_page) rc = -EBUSY; This skips over all mapping aops using buffer_migrate_page() even though it is possible to migrate some of these pages without blocking. This patch updates the ->migratepage callback with a "sync" parameter. It is the responsibility of the callback to fail gracefully if migration would block. Signed-off-by: Mel Gorman <[email protected]> Reviewed-by: Rik van Riel <[email protected]> Cc: Andrea Arcangeli <[email protected]> Cc: Minchan Kim <[email protected]> Cc: Dave Jones <[email protected]> Cc: Jan Kara <[email protected]> Cc: Andy Isaacson <[email protected]> Cc: Nai Xia <[email protected]> Cc: Johannes Weiner <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2012-01-11Btrfs: fix possible deadlock when opening a seed deviceLi Zefan2-4/+7
The correct lock order is uuid_mutex -> volume_mutex -> chunk_mutex, but when we mount a filesystem which has backing seed devices, we have this lock chain: open_ctree() lock(chunk_mutex); read_chunk_tree(); read_one_dev(); open_seed_devices(); lock(uuid_mutex); and then we hit a lockdep splat. Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: update global block_rsv when creating a new block groupLi Zefan1-0/+1
A bug was triggered while using seed device: # mkfs.btrfs /dev/loop1 # btrfstune -S 1 /dev/loop1 # mount -o /dev/loop1 /mnt # btrfs dev add /dev/loop2 /mnt btrfs: block rsv returned -28 ------------[ cut here ]------------ WARNING: at fs/btrfs/extent-tree.c:5969 btrfs_alloc_free_block+0x166/0x396 [btrfs]() ... Call Trace: ... [<f7b7c31c>] btrfs_cow_block+0x101/0x147 [btrfs] [<f7b7eaa6>] btrfs_search_slot+0x1b8/0x55f [btrfs] [<f7b7f844>] btrfs_insert_empty_items+0x42/0x7f [btrfs] [<f7b7f8c1>] btrfs_insert_item+0x40/0x7e [btrfs] [<f7b8ac02>] btrfs_make_block_group+0x243/0x2aa [btrfs] [<f7bb3f53>] __btrfs_alloc_chunk+0x672/0x70e [btrfs] [<f7bb41ff>] init_first_rw_device+0x77/0x13c [btrfs] [<f7bb5a62>] btrfs_init_new_device+0x664/0x9fd [btrfs] [<f7bbb65a>] btrfs_ioctl+0x694/0xdbe [btrfs] [<c04f55f7>] do_vfs_ioctl+0x496/0x4cc [<c04f5660>] sys_ioctl+0x33/0x4f [<c07b9edf>] sysenter_do_call+0x12/0x38 ---[ end trace 906adac595facc7d ]--- Since seed device is readonly, there's no usable space in the filesystem. Afterwards we add a sprout device to it, and the kernel creates a METADATA block group and a SYSTEM block group where comes free space we can reserve, but we still get revervation failure because the global block_rsv hasn't been updated accordingly. Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: rewrite btrfs_trim_block_group()Li Zefan1-71/+164
There are various bugs in block group trimming: - It may trim from offset smaller than user-specified offset. - It may trim beyond user-specified range. - It may leak free space for extents smaller than specified minlen. - It may truncate the last trimmed extent thus leak free space. - With mixed extents+bitmaps, some extents may not be trimmed. - With mixed extents+bitmaps, some bitmaps may not be trimmed (even none will be trimmed). Even for those trimmed, not all the free space in the bitmaps will be trimmed. I rewrite btrfs_trim_block_group() and break it into two functions. One is to trim extents only, and the other is to trim bitmaps only. Before patching: # fstrim -v /mnt/ /mnt/: 1496465408 bytes were trimmed After patching: # fstrim -v /mnt/ /mnt/: 2193768448 bytes were trimmed And this matches the total free space: # btrfs fi df /mnt Data: total=3.58GB, used=1.79GB System, DUP: total=8.00MB, used=4.00KB System: total=4.00MB, used=0.00 Metadata, DUP: total=205.12MB, used=97.14MB Metadata: total=8.00MB, used=0.00 Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: simplfy calculation of stripe length for discard operationLi Zefan1-64/+31
For btrfs raid, while discarding a range of space, we'll need to know the start offset and length to discard for each device, and it's done in btrfs_map_block(). However the calculation is a bit complex for raid0 and raid10, so I reimplement it based on a fact that: dev1 dev2 dev3 (raid0) ----------------------------------- s0 s3 s6 s1 s4 s7 s2 s5 Each device has (total_stripes / nr_dev) stripes, or plus one. Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: don't pre-allocate btrfs bioLi Zefan1-46/+21
We pre-allocate a btrfs bio with fixed size, and then may re-allocate memory if we find stripes are bigger than the fixed size. But this pre-allocation is not necessary. Also we don't have to calcuate the stripe number twice. Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: don't pass a trans handle unnecessarily in volumes.cLi Zefan3-14/+9
Some functions never use the transaction handle passed to them. Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: reserve metadata space in btrfs_ioctl_setflags()Li Zefan1-1/+1
Check and reserve space for btrfs_update_inode(). Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: remove BUG_ON()s in btrfs_ioctl_setflags()Li Zefan1-4/+14
We can recover from errors and return -errno to user space. Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: check the return value of io_ctl_init()Li Zefan1-2/+7
It can return -ENOMEM. Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: avoid possible NULL deref in io_ctl_drop_pages()Li Zefan1-3/+5
If we run into some failure path in io_ctl_prepare_pages(), io_ctl->pages[] array may have some NULL pointers. Signed-off-by: Li Zefan <[email protected]>
2012-01-11Btrfs: add pinned extents to on-disk free space cache correctlyLi Zefan1-21/+20
I got this while running xfstests: [24256.836098] block group 317849600 has an wrong amount of free space [24256.836100] btrfs: failed to load free space cache for block group 317849600 We should clamp the extent returned by find_first_extent_bit(), so the start of the extent won't smaller than the start of the block group. Signed-off-by: Li Zefan <[email protected]>
2012-01-11Merge branch 'for-linus' of ↵Li Zefan33-2009/+6337
git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs into for-linus
2012-01-10Merge branch 'writeback-for-linus' of ↵Linus Torvalds1-1/+2
git://git.kernel.org/pub/scm/linux/kernel/git/wfg/linux * 'writeback-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/wfg/linux: writeback: move MIN_WRITEBACK_PAGES to fs-writeback.c writeback: balanced_rate cannot exceed write bandwidth writeback: do strict bdi dirty_exceeded writeback: avoid tiny dirty poll intervals writeback: max, min and target dirty pause time writeback: dirty ratelimit - think time compensation btrfs: fix dirtied pages accounting on sub-page writes writeback: fix dirtied pages accounting on redirty writeback: fix dirtied pages accounting on sub-page writes writeback: charge leaked page dirties to active tasks writeback: Include all dirty inodes in background writeback
2012-01-10btrfs: pass __GFP_WRITE for buffered write page allocationsJohannes Weiner1-1/+1
Tell the page allocator that pages allocated for a buffered write are expected to become dirty soon. Signed-off-by: Johannes Weiner <[email protected]> Reviewed-by: Rik van Riel <[email protected]> Acked-by: Mel Gorman <[email protected]> Cc: Minchan Kim <[email protected]> Cc: Michal Hocko <[email protected]> Cc: KAMEZAWA Hiroyuki <[email protected]> Cc: Christoph Hellwig <[email protected]> Cc: Wu Fengguang <[email protected]> Cc: Dave Chinner <[email protected]> Cc: Jan Kara <[email protected]> Cc: Shaohua Li <[email protected]> Cc: Chris Mason <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2012-01-08btrfs: take allocation of ->tree_root into open_ctree()Al Viro3-10/+5
now that we don't need it for sget() anymore... Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: let ->s_fs_info point to fs_info, not root...Al Viro5-45/+45
the latter can be obtained from the former (by looking as ->tree_root) just as cheaply as we currently are doing the other way round. Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: consolidate failure exits in btrfs_mount() a bitAl Viro1-16/+5
Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: make free_fs_info() call ->kill_sb() unconditionalAl Viro1-6/+2
... and don't bother with it after btrfs_fill_super() failure - ->kill_sb() (unlike ->put_super()) will be called even if we have not got non-NULL ->s_root. Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: merge free_fs_info() calls on fill_super failuresAl Viro2-3/+1
... all the way up into btrfs_mount(). Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: kill pointless reassignment of ->s_fs_info in btrfs_fill_super()Al Viro1-5/+2
We do not (fortunately) modify ->s_fs_info of superblock on the fly in btrfs_fill_super(); apparent assignment is a no-op. Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: make open_ctree() return intAl Viro3-13/+13
It returns either ERR_PTR(-ve) or sb->s_fs_info. The latter can be found by caller just as well, TYVM, no need to return it. Just return -ve or 0... Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: sanitizing ->fs_info, part 5Al Viro1-11/+11
close_ctree() uses a weird mix of accesses to root->fs_info and its value at the beginning of function stored in local variable. Since ->fs_info *never* changes, let's just use the local variable to avoid confusion. Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: sanitizing ->fs_info, part 4Al Viro3-20/+18
A new helper: btrfs_alloc_root(fs_info); allocates btrfs_root and sets ->fs_info. All places allocating the suckers converted to it. At that point we *never* reassign ->fs_info of btrfs_root; it's set before anyone sees the address of newly allocated struct btrfs_root and never assigned anywhere else. Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: sanitizing ->fs_info, part 3Al Viro1-7/+5
move assignments to ->fs_info in open_ctree() up, to the place just after the original allocations. Assignment for tree_root becomes a no-op - we'd obtained fs_info from tree_root->fs_info in the first place. Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: sanitizing ->fs_info, part 2Al Viro1-1/+4
lift assignment to callers of find_and_setup_root() Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: sanitizing ->fs_info, part 1Al Viro1-1/+6
take assignment of ->fs_info to callers of __setup_root() Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: fix a deadlock in btrfs_scan_one_device()Al Viro1-3/+2
pathname resolution under a global mutex, taken on some paths in ->mount() is a Bad Idea(tm) - think what happens if said pathname resolution triggers automount of some btrfs instance and walks into attempt to grab the same mutex. Deadlock - we are waiting for daemon to finish walking the path, daemon is waiting for us to release the mutex... Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: fix mount/umount raceAl Viro1-9/+4
Do *NOT* skip doomed superblocks in btrfs_test_super(); we want sget() to wait for their shutdown to complete. Since we don't mutilate ->s_fs_info in ->put_super() anymore (or free what it used to point to until the superblock is past being findable by sget()), we can just DTRT there and report a match. Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: get ->kill_sb() of its ownAl Viro1-9/+18
... and move free_fs_info() to that, out of ->put_super(). Do NOT set ->s_fs_info to NULL in the latter; we need it for sget() to be able to see and wait for fs in the middle of umount if we get a mount/umount race. Signed-off-by: Al Viro <[email protected]>
2012-01-08btrfs: preparation to fixing mount/umount raceAl Viro2-3/+6
We need fs_info and root to live until the moment when the victim superblock leaves the list, so we need to postpone free_fs_info() until after ->put_super(). The call is buried in close_ctree(), though, so we need to lift it into the callers (including btrfs_put_super()) first. Signed-off-by: Al Viro <[email protected]>
2012-01-08Merge branch 'for-linus' of ↵Linus Torvalds2-2/+2
git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (53 commits) Kconfig: acpi: Fix typo in comment. misc latin1 to utf8 conversions devres: Fix a typo in devm_kfree comment btrfs: free-space-cache.c: remove extra semicolon. fat: Spelling s/obsolate/obsolete/g SCSI, pmcraid: Fix spelling error in a pmcraid_err() call tools/power turbostat: update fields in manpage mac80211: drop spelling fix types.h: fix comment spelling for 'architectures' typo fixes: aera -> area, exntension -> extension devices.txt: Fix typo of 'VMware'. sis900: Fix enum typo 'sis900_rx_bufer_status' decompress_bunzip2: remove invalid vi modeline treewide: Fix comment and string typo 'bufer' hyper-v: Update MAINTAINERS treewide: Fix typos in various parts of the kernel, and fix some comments. clockevents: drop unknown Kconfig symbol GENERIC_CLOCKEVENTS_MIGR gpio: Kconfig: drop unknown symbol 'CS5535_GPIO' leds: Kconfig: Fix typo 'D2NET_V2' sound: Kconfig: drop unknown symbol ARCH_CLPS7500 ... Fix up trivial conflicts in arch/powerpc/platforms/40x/Kconfig (some new kconfig additions, close to removed commented-out old ones)
2012-01-08Merge branch 'pm-for-linus' of ↵Linus Torvalds2-7/+3
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm * 'pm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (76 commits) PM / Hibernate: Implement compat_ioctl for /dev/snapshot PM / Freezer: fix return value of freezable_schedule_timeout_killable() PM / shmobile: Allow the A4R domain to be turned off at run time PM / input / touchscreen: Make st1232 use device PM QoS constraints PM / QoS: Introduce dev_pm_qos_add_ancestor_request() PM / shmobile: Remove the stay_on flag from SH7372's PM domains PM / shmobile: Don't include SH7372's INTCS in syscore suspend/resume PM / shmobile: Add support for the sh7372 A4S power domain / sleep mode PM: Drop generic_subsys_pm_ops PM / Sleep: Remove forward-only callbacks from AMBA bus type PM / Sleep: Remove forward-only callbacks from platform bus type PM: Run the driver callback directly if the subsystem one is not there PM / Sleep: Make pm_op() and pm_noirq_op() return callback pointers PM/Devfreq: Add Exynos4-bus device DVFS driver for Exynos4210/4212/4412. PM / Sleep: Merge internal functions in generic_ops.c PM / Sleep: Simplify generic system suspend callbacks PM / Hibernate: Remove deprecated hibernation snapshot ioctls PM / Sleep: Fix freezer failures due to racy usermodehelper_is_disabled() ARM: S3C64XX: Implement basic power domain support PM / shmobile: Use common always on power domain governor ... Fix up trivial conflict in fs/xfs/xfs_buf.c due to removal of unused XBT_FORCE_SLEEP bit
2012-01-07Btrfs: revamp clustered allocation logicAlexandre Oliva1-63/+49
Parameterize clusters on minimum total size, minimum chunk size and minimum contiguous size for at least one chunk, without limits on cluster, window or gap sizes. Don't tolerate any fragmentation for SSD_SPREAD; accept it for metadata, but try to keep data dense. Signed-off-by: Alexandre Oliva <[email protected]> Signed-off-by: Chris Mason <[email protected]>
2012-01-07Btrfs: don't set up allocation result twiceAlexandre Oliva1-3/+0
We store the allocation start and length twice in ins, once right after the other, but with intervening calls that may prevent the duplicate from being optimized out by the compiler. Remove one of the assignments. Signed-off-by: Alexandre Oliva <[email protected]> Signed-off-by: Chris Mason <[email protected]>
2012-01-06vfs: switch ->show_options() to struct dentry *Al Viro1-2/+2
Signed-off-by: Al Viro <[email protected]>
2012-01-06Btrfs: test free space only for unclustered allocationAlexandre Oliva1-11/+23
Since the clustered allocation may be taking extents from a different block group, there's no point in spin-locking and testing the current block group free space before attempting to allocate space from a cluster, even more so when we might refrain from even trying the cluster in the current block group because, after the cluster was set up, not enough free space remained. Furthermore, cluster creation attempts fail fast when the block group doesn't have enough free space, so the test was completely superfluous. I've move the free space test past the cluster allocation attempt, where it is more useful, and arranged for a cluster in the current block group to be released before trying an unclustered allocation, when we reach the LOOP_NO_EMPTY_SIZE stage, so that the free space in the cluster stands a chance of being combined with additional free space in the block group so as to succeed in the allocation attempt. Signed-off-by: Alexandre Oliva <[email protected]> Signed-off-by: Chris Mason <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.