diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2018-04-04 13:03:38 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2018-04-04 13:03:38 -0700 |
commit | 94514bbe9e5c402c4232af158a295a8fdfd72a2c (patch) | |
tree | c990c722cbac5abe8a3b28e0564effa722b7c80e /fs/btrfs/dev-replace.c | |
parent | 547c43d777968228b1060b6f1b152b96215eb7b2 (diff) | |
parent | 57599c7e7722daf5f8c2dba4b0e4628f5c500771 (diff) |
Merge tag 'for-4.17-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs updates from David Sterba:
"There are a several user visible changes, the rest is mostly invisible
and continues to clean up the whole code base.
User visible changes:
- new mount option nossd_spread (pair for ssd_spread)
- mount option subvolid will detect junk after the number and fail
the mount
- add message after cancelled device replace
- direct module dependency on libcrc32, removed own crc wrappers
- removed user space transaction ioctls
- use lighter locking when reading /proc/self/mounts, RCU instead of
mutex to avoid unnecessary contention
Enhancements:
- skip writeback of last page when truncating file to same size
- send: do not issue unnecessary truncate operations
- mount option token specifiers: use %u for unsigned values, more
validation
- selftests: more tree block validations
qgroups:
- preparatory work for splitting reservation types for data and
metadata, this should allow for more accurate tracking and fix some
issues with underflows or do further enhancements
- split metadata reservations for started and joined transaction so
they do not get mixed up and are accounted correctly at commit time
- with the above, it's possible to revert patch that potentially
deadlocks when trying to make more space by explicitly committing
when the quota limit is hit
- fix root item corruption when multiple same source snapshots are
created with quota enabled
RAID56:
- make sure target is identical to source when raid56 rebuild fails
after dev-replace
- faster rebuild during scrub, batch by stripes and not
block-by-block
- make more use of cached data when rebuilding from a missing device
Fixes:
- null pointer deref when device replace target is missing
- fix fsync after hole punching when using no-holes feature
- fix lockdep splat when allocating percpu data with wrong GFP flags
Cleanups, refactoring, core changes:
- drop redunant parameters from various functions
- kill and opencode trivial helpers
- __cold/__exit function annotations
- dead code removal
- continued audit and documentation of memory barriers
- error handling: handle removal from uuid tree
- error handling: remove handling of impossible condtitons
- more debugging or error messages
- updated tracepoints
- one VLA use removal (and one still left)"
* tag 'for-4.17-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (164 commits)
btrfs: lift errors from add_extent_changeset to the callers
Btrfs: print error messages when failing to read trees
btrfs: user proper type for btrfs_mask_flags flags
btrfs: split dev-replace locking helpers for read and write
btrfs: remove stale comments about fs_mutex
btrfs: use RCU in btrfs_show_devname for device list traversal
btrfs: update barrier in should_cow_block
btrfs: use lockdep_assert_held for mutexes
btrfs: use lockdep_assert_held for spinlocks
btrfs: Validate child tree block's level and first key
btrfs: tests/qgroup: Fix wrong tree backref level
Btrfs: fix copy_items() return value when logging an inode
Btrfs: fix fsync after hole punching when using no-holes feature
btrfs: use helper to set ulist aux from a qgroup
Revert "btrfs: qgroups: Retry after commit on getting EDQUOT"
btrfs: qgroup: Update trace events for metadata reservation
btrfs: qgroup: Use root::qgroup_meta_rsv_* to record qgroup meta reserved space
btrfs: delayed-inode: Use new qgroup meta rsv for delayed inode and item
btrfs: qgroup: Use separate meta reservation type for delalloc
btrfs: qgroup: Introduce function to convert META_PREALLOC into META_PERTRANS
...
Diffstat (limited to 'fs/btrfs/dev-replace.c')
-rw-r--r-- | fs/btrfs/dev-replace.c | 130 |
1 files changed, 68 insertions, 62 deletions
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c index 7efbc4d1128b..0d203633bb96 100644 --- a/fs/btrfs/dev-replace.c +++ b/fs/btrfs/dev-replace.c @@ -44,7 +44,6 @@ static void btrfs_dev_replace_update_device_in_mapping_tree( struct btrfs_fs_info *fs_info, struct btrfs_device *srcdev, struct btrfs_device *tgtdev); -static u64 __btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info); static int btrfs_dev_replace_kthread(void *data); static int btrfs_dev_replace_continue_on_mount(struct btrfs_fs_info *fs_info); @@ -174,8 +173,14 @@ no_valid_dev_replace_entry_found: } set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &dev_replace->tgtdev->dev_state); - btrfs_init_dev_replace_tgtdev_for_resume(fs_info, - dev_replace->tgtdev); + + WARN_ON(fs_info->fs_devices->rw_devices == 0); + dev_replace->tgtdev->io_width = fs_info->sectorsize; + dev_replace->tgtdev->io_align = fs_info->sectorsize; + dev_replace->tgtdev->sector_size = fs_info->sectorsize; + dev_replace->tgtdev->fs_info = fs_info; + set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, + &dev_replace->tgtdev->dev_state); } break; } @@ -200,13 +205,13 @@ int btrfs_run_dev_replace(struct btrfs_trans_handle *trans, struct btrfs_dev_replace_item *ptr; struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; - btrfs_dev_replace_lock(dev_replace, 0); + btrfs_dev_replace_read_lock(dev_replace); if (!dev_replace->is_valid || !dev_replace->item_needs_writeback) { - btrfs_dev_replace_unlock(dev_replace, 0); + btrfs_dev_replace_read_unlock(dev_replace); return 0; } - btrfs_dev_replace_unlock(dev_replace, 0); + btrfs_dev_replace_read_unlock(dev_replace); key.objectid = 0; key.type = BTRFS_DEV_REPLACE_KEY; @@ -264,7 +269,7 @@ int btrfs_run_dev_replace(struct btrfs_trans_handle *trans, ptr = btrfs_item_ptr(eb, path->slots[0], struct btrfs_dev_replace_item); - btrfs_dev_replace_lock(dev_replace, 1); + btrfs_dev_replace_write_lock(dev_replace); if (dev_replace->srcdev) btrfs_set_dev_replace_src_devid(eb, ptr, dev_replace->srcdev->devid); @@ -287,7 +292,7 @@ int btrfs_run_dev_replace(struct btrfs_trans_handle *trans, btrfs_set_dev_replace_cursor_right(eb, ptr, dev_replace->cursor_right); dev_replace->item_needs_writeback = 0; - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); btrfs_mark_buffer_dirty(eb); @@ -307,7 +312,7 @@ void btrfs_after_dev_replace_commit(struct btrfs_fs_info *fs_info) static char* btrfs_dev_name(struct btrfs_device *device) { - if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state)) + if (!device || test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state)) return "<missing disk>"; else return rcu_str_deref(device->name); @@ -352,7 +357,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, return PTR_ERR(trans); } - btrfs_dev_replace_lock(dev_replace, 1); + btrfs_dev_replace_write_lock(dev_replace); switch (dev_replace->replace_state) { case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED: case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED: @@ -390,7 +395,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, dev_replace->item_needs_writeback = 1; atomic64_set(&dev_replace->num_write_errors, 0); atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0); - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device); if (ret) @@ -402,7 +407,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { ret = PTR_ERR(trans); - btrfs_dev_replace_lock(dev_replace, 1); + btrfs_dev_replace_write_lock(dev_replace); goto leave; } @@ -426,7 +431,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, leave: dev_replace->srcdev = NULL; dev_replace->tgtdev = NULL; - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device); return ret; } @@ -493,18 +498,18 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, /* don't allow cancel or unmount to disturb the finishing procedure */ mutex_lock(&dev_replace->lock_finishing_cancel_unmount); - btrfs_dev_replace_lock(dev_replace, 0); + btrfs_dev_replace_read_lock(dev_replace); /* was the operation canceled, or is it finished? */ if (dev_replace->replace_state != BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED) { - btrfs_dev_replace_unlock(dev_replace, 0); + btrfs_dev_replace_read_unlock(dev_replace); mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); return 0; } tgt_device = dev_replace->tgtdev; src_device = dev_replace->srcdev; - btrfs_dev_replace_unlock(dev_replace, 0); + btrfs_dev_replace_read_unlock(dev_replace); /* * flush all outstanding I/O and inode extent mappings before the @@ -529,7 +534,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, /* keep away write_all_supers() during the finishing procedure */ mutex_lock(&fs_info->fs_devices->device_list_mutex); mutex_lock(&fs_info->chunk_mutex); - btrfs_dev_replace_lock(dev_replace, 1); + btrfs_dev_replace_write_lock(dev_replace); dev_replace->replace_state = scrub_ret ? BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED : BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED; @@ -549,7 +554,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, btrfs_dev_name(src_device), src_device->devid, rcu_str_deref(tgt_device->name), scrub_ret); - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); mutex_unlock(&fs_info->chunk_mutex); mutex_unlock(&fs_info->fs_devices->device_list_mutex); mutex_unlock(&uuid_mutex); @@ -586,7 +591,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, list_add(&tgt_device->dev_alloc_list, &fs_info->fs_devices->alloc_list); fs_info->fs_devices->rw_devices++; - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); btrfs_rm_dev_replace_blocked(fs_info); @@ -679,7 +684,7 @@ void btrfs_dev_replace_status(struct btrfs_fs_info *fs_info, { struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; - btrfs_dev_replace_lock(dev_replace, 0); + btrfs_dev_replace_read_lock(dev_replace); /* even if !dev_replace_is_valid, the values are good enough for * the replace_status ioctl */ args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR; @@ -691,41 +696,36 @@ void btrfs_dev_replace_status(struct btrfs_fs_info *fs_info, args->status.num_uncorrectable_read_errors = atomic64_read(&dev_replace->num_uncorrectable_read_errors); args->status.progress_1000 = btrfs_dev_replace_progress(fs_info); - btrfs_dev_replace_unlock(dev_replace, 0); + btrfs_dev_replace_read_unlock(dev_replace); } -int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info, - struct btrfs_ioctl_dev_replace_args *args) -{ - args->result = __btrfs_dev_replace_cancel(fs_info); - return 0; -} - -static u64 __btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info) +int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info) { struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; struct btrfs_device *tgt_device = NULL; + struct btrfs_device *src_device = NULL; struct btrfs_trans_handle *trans; struct btrfs_root *root = fs_info->tree_root; - u64 result; + int result; int ret; if (sb_rdonly(fs_info->sb)) return -EROFS; mutex_lock(&dev_replace->lock_finishing_cancel_unmount); - btrfs_dev_replace_lock(dev_replace, 1); + btrfs_dev_replace_write_lock(dev_replace); switch (dev_replace->replace_state) { case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED: case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED: case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED: result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED; - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); goto leave; case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED: case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED: result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR; tgt_device = dev_replace->tgtdev; + src_device = dev_replace->srcdev; dev_replace->tgtdev = NULL; dev_replace->srcdev = NULL; break; @@ -733,7 +733,7 @@ static u64 __btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info) dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED; dev_replace->time_stopped = get_seconds(); dev_replace->item_needs_writeback = 1; - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); btrfs_scrub_cancel(fs_info); trans = btrfs_start_transaction(root, 0); @@ -743,6 +743,12 @@ static u64 __btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info) } ret = btrfs_commit_transaction(trans); WARN_ON(ret); + + btrfs_info_in_rcu(fs_info, + "dev_replace from %s (devid %llu) to %s canceled", + btrfs_dev_name(src_device), src_device->devid, + btrfs_dev_name(tgt_device)); + if (tgt_device) btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device); @@ -756,7 +762,7 @@ void btrfs_dev_replace_suspend_for_unmount(struct btrfs_fs_info *fs_info) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; mutex_lock(&dev_replace->lock_finishing_cancel_unmount); - btrfs_dev_replace_lock(dev_replace, 1); + btrfs_dev_replace_write_lock(dev_replace); switch (dev_replace->replace_state) { case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED: case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED: @@ -772,7 +778,7 @@ void btrfs_dev_replace_suspend_for_unmount(struct btrfs_fs_info *fs_info) break; } - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); } @@ -782,12 +788,12 @@ int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info) struct task_struct *task; struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; - btrfs_dev_replace_lock(dev_replace, 1); + btrfs_dev_replace_write_lock(dev_replace); switch (dev_replace->replace_state) { case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED: case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED: case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED: - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); return 0; case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED: break; @@ -801,10 +807,10 @@ int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info) "cannot continue dev_replace, tgtdev is missing"); btrfs_info(fs_info, "you may cancel the operation after 'mount -o degraded'"); - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); return 0; } - btrfs_dev_replace_unlock(dev_replace, 1); + btrfs_dev_replace_write_unlock(dev_replace); WARN_ON(test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)); task = kthread_run(btrfs_dev_replace_kthread, fs_info, "btrfs-devrepl"); @@ -873,37 +879,37 @@ int btrfs_dev_replace_is_ongoing(struct btrfs_dev_replace *dev_replace) return 1; } -void btrfs_dev_replace_lock(struct btrfs_dev_replace *dev_replace, int rw) +void btrfs_dev_replace_read_lock(struct btrfs_dev_replace *dev_replace) { - if (rw == 1) { - /* write */ -again: - wait_event(dev_replace->read_lock_wq, - atomic_read(&dev_replace->blocking_readers) == 0); - write_lock(&dev_replace->lock); - if (atomic_read(&dev_replace->blocking_readers)) { - write_unlock(&dev_replace->lock); - goto again; - } - } else { - read_lock(&dev_replace->lock); - atomic_inc(&dev_replace->read_locks); - } + read_lock(&dev_replace->lock); + atomic_inc(&dev_replace->read_locks); +} + +void btrfs_dev_replace_read_unlock(struct btrfs_dev_replace *dev_replace) +{ + ASSERT(atomic_read(&dev_replace->read_locks) > 0); + atomic_dec(&dev_replace->read_locks); + read_unlock(&dev_replace->lock); } -void btrfs_dev_replace_unlock(struct btrfs_dev_replace *dev_replace, int rw) +void btrfs_dev_replace_write_lock(struct btrfs_dev_replace *dev_replace) { - if (rw == 1) { - /* write */ - ASSERT(atomic_read(&dev_replace->blocking_readers) == 0); +again: + wait_event(dev_replace->read_lock_wq, + atomic_read(&dev_replace->blocking_readers) == 0); + write_lock(&dev_replace->lock); + if (atomic_read(&dev_replace->blocking_readers)) { write_unlock(&dev_replace->lock); - } else { - ASSERT(atomic_read(&dev_replace->read_locks) > 0); - atomic_dec(&dev_replace->read_locks); - read_unlock(&dev_replace->lock); + goto again; } } +void btrfs_dev_replace_write_unlock(struct btrfs_dev_replace *dev_replace) +{ + ASSERT(atomic_read(&dev_replace->blocking_readers) == 0); + write_unlock(&dev_replace->lock); +} + /* inc blocking cnt and release read lock */ void btrfs_dev_replace_set_lock_blocking( struct btrfs_dev_replace *dev_replace) |