Btrfs: fix enospc problems with delalloc
So I had this brilliant idea to use atomic counters for outstanding and reserved
extents, but this turned out to be a bad idea. Consider this where we have 1
outstanding extent and 1 reserved extent
Reserver Releaser
atomic_dec(outstanding) now 0
atomic_read(outstanding)+1 get 1
atomic_read(reserved) get 1
don't actually reserve anything because
they are the same
atomic_cmpxchg(reserved, 1, 0)
atomic_inc(outstanding)
atomic_add(0, reserved)
free reserved space for 1 extent
Then the reserver now has no actual space reserved for it, and when it goes to
finish the ordered IO it won't have enough space to do it's allocation and you
get those lovely warnings.
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 2bf4d2b..55d68ea 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1298,7 +1298,9 @@
if (!(orig->state & EXTENT_DELALLOC))
return 0;
- atomic_inc(&BTRFS_I(inode)->outstanding_extents);
+ spin_lock(&BTRFS_I(inode)->lock);
+ BTRFS_I(inode)->outstanding_extents++;
+ spin_unlock(&BTRFS_I(inode)->lock);
return 0;
}
@@ -1316,7 +1318,9 @@
if (!(other->state & EXTENT_DELALLOC))
return 0;
- atomic_dec(&BTRFS_I(inode)->outstanding_extents);
+ spin_lock(&BTRFS_I(inode)->lock);
+ BTRFS_I(inode)->outstanding_extents--;
+ spin_unlock(&BTRFS_I(inode)->lock);
return 0;
}
@@ -1339,10 +1343,13 @@
u64 len = state->end + 1 - state->start;
bool do_list = !is_free_space_inode(root, inode);
- if (*bits & EXTENT_FIRST_DELALLOC)
+ if (*bits & EXTENT_FIRST_DELALLOC) {
*bits &= ~EXTENT_FIRST_DELALLOC;
- else
- atomic_inc(&BTRFS_I(inode)->outstanding_extents);
+ } else {
+ spin_lock(&BTRFS_I(inode)->lock);
+ BTRFS_I(inode)->outstanding_extents++;
+ spin_unlock(&BTRFS_I(inode)->lock);
+ }
spin_lock(&root->fs_info->delalloc_lock);
BTRFS_I(inode)->delalloc_bytes += len;
@@ -1372,10 +1379,13 @@
u64 len = state->end + 1 - state->start;
bool do_list = !is_free_space_inode(root, inode);
- if (*bits & EXTENT_FIRST_DELALLOC)
+ if (*bits & EXTENT_FIRST_DELALLOC) {
*bits &= ~EXTENT_FIRST_DELALLOC;
- else if (!(*bits & EXTENT_DO_ACCOUNTING))
- atomic_dec(&BTRFS_I(inode)->outstanding_extents);
+ } else if (!(*bits & EXTENT_DO_ACCOUNTING)) {
+ spin_lock(&BTRFS_I(inode)->lock);
+ BTRFS_I(inode)->outstanding_extents--;
+ spin_unlock(&BTRFS_I(inode)->lock);
+ }
if (*bits & EXTENT_DO_ACCOUNTING)
btrfs_delalloc_release_metadata(inode, len);
@@ -6735,8 +6745,9 @@
ei->index_cnt = (u64)-1;
ei->last_unlink_trans = 0;
- atomic_set(&ei->outstanding_extents, 0);
- atomic_set(&ei->reserved_extents, 0);
+ spin_lock_init(&ei->lock);
+ ei->outstanding_extents = 0;
+ ei->reserved_extents = 0;
ei->ordered_data_close = 0;
ei->orphan_meta_reserved = 0;
@@ -6774,8 +6785,8 @@
WARN_ON(!list_empty(&inode->i_dentry));
WARN_ON(inode->i_data.nrpages);
- WARN_ON(atomic_read(&BTRFS_I(inode)->outstanding_extents));
- WARN_ON(atomic_read(&BTRFS_I(inode)->reserved_extents));
+ WARN_ON(BTRFS_I(inode)->outstanding_extents);
+ WARN_ON(BTRFS_I(inode)->reserved_extents);
/*
* This can happen where we create an inode, but somebody else also