diff options
author | Josef Bacik <josef@redhat.com> | 2011-04-11 17:25:13 -0400 |
---|---|---|
committer | Josef Bacik <josef@redhat.com> | 2011-05-23 13:00:57 -0400 |
commit | a4abeea41adfa3c143c289045f4625dfaeba2212 (patch) | |
tree | 792e2a398d8ba77447ba3f9f2c4266a1ce2f611c /fs/btrfs/disk-io.c | |
parent | 2a1eb4614d984d5cd4c928784e9afcf5c07f93be (diff) | |
download | op-kernel-dev-a4abeea41adfa3c143c289045f4625dfaeba2212.zip op-kernel-dev-a4abeea41adfa3c143c289045f4625dfaeba2212.tar.gz |
Btrfs: kill trans_mutex
We use trans_mutex for lots of things, here's a basic list
1) To serialize trans_handles joining the currently running transaction
2) To make sure that no new trans handles are started while we are committing
3) To protect the dead_roots list and the transaction lists
Really the serializing trans_handles joining is not too hard, and can really get
bogged down in acquiring a reference to the transaction. So replace the
trans_mutex with a trans_lock spinlock and use it to do the following
1) Protect fs_info->running_transaction. All trans handles have to do is check
this, and then take a reference of the transaction and keep on going.
2) Protect the fs_info->trans_list. This doesn't get used too much, basically
it just holds the current transactions, which will usually just be the currently
committing transaction and the currently running transaction at most.
3) Protect the dead roots list. This is only ever processed by splicing the
list so this is relatively simple.
4) Protect the fs_info->reloc_ctl stuff. This is very lightweight and was using
the trans_mutex before, so this is a pretty straightforward change.
5) Protect fs_info->no_trans_join. Because we don't hold the trans_lock over
the entirety of the commit we need to have a way to block new people from
creating a new transaction while we're doing our work. So we set no_trans_join
and in join_transaction we test to see if that is set, and if it is we do a
wait_on_commit.
6) Make the transaction use count atomic so we don't need to take locks to
modify it when we're dropping references.
7) Add a commit_lock to the transaction to make sure multiple people trying to
commit the same transaction don't race and commit at the same time.
8) Make open_ioctl_trans an atomic so we don't have to take any locks for ioctl
trans.
I have tested this with xfstests, but obviously it is a pretty hairy change so
lots of testing is greatly appreciated. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Diffstat (limited to 'fs/btrfs/disk-io.c')
-rw-r--r-- | fs/btrfs/disk-io.c | 30 |
1 files changed, 15 insertions, 15 deletions
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 9d6c9e3..93ef254 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1551,22 +1551,22 @@ static int transaction_kthread(void *arg) vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE); mutex_lock(&root->fs_info->transaction_kthread_mutex); - spin_lock(&root->fs_info->new_trans_lock); + spin_lock(&root->fs_info->trans_lock); cur = root->fs_info->running_transaction; if (!cur) { - spin_unlock(&root->fs_info->new_trans_lock); + spin_unlock(&root->fs_info->trans_lock); goto sleep; } now = get_seconds(); if (!cur->blocked && (now < cur->start_time || now - cur->start_time < 30)) { - spin_unlock(&root->fs_info->new_trans_lock); + spin_unlock(&root->fs_info->trans_lock); delay = HZ * 5; goto sleep; } transid = cur->transid; - spin_unlock(&root->fs_info->new_trans_lock); + spin_unlock(&root->fs_info->trans_lock); trans = btrfs_join_transaction(root); BUG_ON(IS_ERR(trans)); @@ -1658,7 +1658,7 @@ struct btrfs_root *open_ctree(struct super_block *sb, INIT_LIST_HEAD(&fs_info->ordered_operations); INIT_LIST_HEAD(&fs_info->caching_block_groups); spin_lock_init(&fs_info->delalloc_lock); - spin_lock_init(&fs_info->new_trans_lock); + spin_lock_init(&fs_info->trans_lock); spin_lock_init(&fs_info->ref_cache_lock); spin_lock_init(&fs_info->fs_roots_radix_lock); spin_lock_init(&fs_info->delayed_iput_lock); @@ -1687,6 +1687,7 @@ struct btrfs_root *open_ctree(struct super_block *sb, fs_info->sb = sb; fs_info->max_inline = 8192 * 1024; fs_info->metadata_ratio = 0; + fs_info->trans_no_join = 0; fs_info->thread_pool_size = min_t(unsigned long, num_online_cpus() + 2, 8); @@ -1735,7 +1736,6 @@ struct btrfs_root *open_ctree(struct super_block *sb, fs_info->do_barriers = 1; - mutex_init(&fs_info->trans_mutex); mutex_init(&fs_info->ordered_operations_mutex); mutex_init(&fs_info->tree_log_mutex); mutex_init(&fs_info->chunk_mutex); @@ -3006,10 +3006,13 @@ static int btrfs_cleanup_transaction(struct btrfs_root *root) WARN_ON(1); - mutex_lock(&root->fs_info->trans_mutex); mutex_lock(&root->fs_info->transaction_kthread_mutex); + spin_lock(&root->fs_info->trans_lock); list_splice_init(&root->fs_info->trans_list, &list); + root->fs_info->trans_no_join = 1; + spin_unlock(&root->fs_info->trans_lock); + while (!list_empty(&list)) { t = list_entry(list.next, struct btrfs_transaction, list); if (!t) @@ -3034,23 +3037,18 @@ static int btrfs_cleanup_transaction(struct btrfs_root *root) t->blocked = 0; if (waitqueue_active(&root->fs_info->transaction_wait)) wake_up(&root->fs_info->transaction_wait); - mutex_unlock(&root->fs_info->trans_mutex); - mutex_lock(&root->fs_info->trans_mutex); t->commit_done = 1; if (waitqueue_active(&t->commit_wait)) wake_up(&t->commit_wait); - mutex_unlock(&root->fs_info->trans_mutex); - - mutex_lock(&root->fs_info->trans_mutex); btrfs_destroy_pending_snapshots(t); btrfs_destroy_delalloc_inodes(root); - spin_lock(&root->fs_info->new_trans_lock); + spin_lock(&root->fs_info->trans_lock); root->fs_info->running_transaction = NULL; - spin_unlock(&root->fs_info->new_trans_lock); + spin_unlock(&root->fs_info->trans_lock); btrfs_destroy_marked_extents(root, &t->dirty_pages, EXTENT_DIRTY); @@ -3064,8 +3062,10 @@ static int btrfs_cleanup_transaction(struct btrfs_root *root) kmem_cache_free(btrfs_transaction_cachep, t); } + spin_lock(&root->fs_info->trans_lock); + root->fs_info->trans_no_join = 0; + spin_unlock(&root->fs_info->trans_lock); mutex_unlock(&root->fs_info->transaction_kthread_mutex); - mutex_unlock(&root->fs_info->trans_mutex); return 0; } |