diff options
author | Liu Bo <bo.li.liu@oracle.com> | 2015-07-17 16:49:19 +0800 |
---|---|---|
committer | David Sterba <dsterba@suse.com> | 2016-02-23 13:10:10 +0100 |
commit | 73beece9ca07c003e0e4f4825b12be167334d4ad (patch) | |
tree | b0575466fc5e0c7a493826963e7a4ffa49e31210 /fs/btrfs/volumes.c | |
parent | 18558cae0272f8fd9647e69d3fec1565a7949865 (diff) | |
download | op-kernel-dev-73beece9ca07c003e0e4f4825b12be167334d4ad.zip op-kernel-dev-73beece9ca07c003e0e4f4825b12be167334d4ad.tar.gz |
Btrfs: fix lockdep deadlock warning due to dev_replace
Xfstests btrfs/011 complains about a deadlock warning,
[ 1226.649039] =========================================================
[ 1226.649039] [ INFO: possible irq lock inversion dependency detected ]
[ 1226.649039] 4.1.0+ #270 Not tainted
[ 1226.649039] ---------------------------------------------------------
[ 1226.652955] kswapd0/46 just changed the state of lock:
[ 1226.652955] (&delayed_node->mutex){+.+.-.}, at: [<ffffffff81458735>] __btrfs_release_delayed_node+0x45/0x1d0
[ 1226.652955] but this lock took another, RECLAIM_FS-unsafe lock in the past:
[ 1226.652955] (&fs_info->dev_replace.lock){+.+.+.}
and interrupts could create inverse lock ordering between them.
[ 1226.652955]
other info that might help us debug this:
[ 1226.652955] Chain exists of:
&delayed_node->mutex --> &found->groups_sem --> &fs_info->dev_replace.lock
[ 1226.652955] Possible interrupt unsafe locking scenario:
[ 1226.652955] CPU0 CPU1
[ 1226.652955] ---- ----
[ 1226.652955] lock(&fs_info->dev_replace.lock);
[ 1226.652955] local_irq_disable();
[ 1226.652955] lock(&delayed_node->mutex);
[ 1226.652955] lock(&found->groups_sem);
[ 1226.652955] <Interrupt>
[ 1226.652955] lock(&delayed_node->mutex);
[ 1226.652955]
*** DEADLOCK ***
Commit 084b6e7c7607 ("btrfs: Fix a lockdep warning when running xfstest.") tried
to fix a similar one that has the exactly same warning, but with that, we still
run to this.
The above lock chain comes from
btrfs_commit_transaction
->btrfs_run_delayed_items
...
->__btrfs_update_delayed_inode
...
->__btrfs_cow_block
...
->find_free_extent
->cache_block_group
->load_free_space_cache
->btrfs_readpages
->submit_one_bio
...
->__btrfs_map_block
->btrfs_dev_replace_lock
However, with high memory pressure, tasks which hold dev_replace.lock can
be interrupted by kswapd and then kswapd is intended to release memory occupied
by superblock, inodes and dentries, where we may call evict_inode, and it comes
to
[ 1226.652955] [<ffffffff81458735>] __btrfs_release_delayed_node+0x45/0x1d0
[ 1226.652955] [<ffffffff81459e74>] btrfs_remove_delayed_node+0x24/0x30
[ 1226.652955] [<ffffffff8140c5fe>] btrfs_evict_inode+0x34e/0x700
delayed_node->mutex may be acquired in __btrfs_release_delayed_node(), and it leads
to a ABBA deadlock.
To fix this, we can use "blocking rwlock" used in the case of extent_buffer, but
things are simpler here since we only needs read's spinlock to blocking lock.
With this, btrfs/011 no more produces warnings in dmesg.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Diffstat (limited to 'fs/btrfs/volumes.c')
-rw-r--r-- | fs/btrfs/volumes.c | 24 |
1 files changed, 14 insertions, 10 deletions
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 366b335..858a2e4 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1714,12 +1714,12 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path) } while (read_seqretry(&root->fs_info->profiles_lock, seq)); num_devices = root->fs_info->fs_devices->num_devices; - btrfs_dev_replace_lock(&root->fs_info->dev_replace); + btrfs_dev_replace_lock(&root->fs_info->dev_replace, 0); if (btrfs_dev_replace_is_ongoing(&root->fs_info->dev_replace)) { WARN_ON(num_devices < 1); num_devices--; } - btrfs_dev_replace_unlock(&root->fs_info->dev_replace); + btrfs_dev_replace_unlock(&root->fs_info->dev_replace, 0); if ((all_avail & BTRFS_BLOCK_GROUP_RAID10) && num_devices <= 4) { ret = BTRFS_ERROR_DEV_RAID10_MIN_NOT_MET; @@ -3686,12 +3686,12 @@ int btrfs_balance(struct btrfs_balance_control *bctl, } num_devices = fs_info->fs_devices->num_devices; - btrfs_dev_replace_lock(&fs_info->dev_replace); + btrfs_dev_replace_lock(&fs_info->dev_replace, 0); if (btrfs_dev_replace_is_ongoing(&fs_info->dev_replace)) { BUG_ON(num_devices < 1); num_devices--; } - btrfs_dev_replace_unlock(&fs_info->dev_replace); + btrfs_dev_replace_unlock(&fs_info->dev_replace, 0); allowed = BTRFS_AVAIL_ALLOC_BIT_SINGLE; if (num_devices == 1) allowed |= BTRFS_BLOCK_GROUP_DUP; @@ -5062,10 +5062,10 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len) ret = 1; free_extent_map(em); - btrfs_dev_replace_lock(&fs_info->dev_replace); + btrfs_dev_replace_lock(&fs_info->dev_replace, 0); if (btrfs_dev_replace_is_ongoing(&fs_info->dev_replace)) ret++; - btrfs_dev_replace_unlock(&fs_info->dev_replace); + btrfs_dev_replace_unlock(&fs_info->dev_replace, 0); return ret; } @@ -5325,10 +5325,12 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw, if (!bbio_ret) goto out; - btrfs_dev_replace_lock(dev_replace); + btrfs_dev_replace_lock(dev_replace, 0); dev_replace_is_ongoing = btrfs_dev_replace_is_ongoing(dev_replace); if (!dev_replace_is_ongoing) - btrfs_dev_replace_unlock(dev_replace); + btrfs_dev_replace_unlock(dev_replace, 0); + else + btrfs_dev_replace_set_lock_blocking(dev_replace); if (dev_replace_is_ongoing && mirror_num == map->num_stripes + 1 && !(rw & (REQ_WRITE | REQ_DISCARD | REQ_GET_READ_MIRRORS)) && @@ -5751,8 +5753,10 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw, bbio->mirror_num = map->num_stripes + 1; } out: - if (dev_replace_is_ongoing) - btrfs_dev_replace_unlock(dev_replace); + if (dev_replace_is_ongoing) { + btrfs_dev_replace_clear_lock_blocking(dev_replace); + btrfs_dev_replace_unlock(dev_replace, 0); + } free_extent_map(em); return ret; } |