btrfs: qgroup: do a reservation in a higher level.
There are two problems in qgroup:
a). The PAGE_CACHE is 4K, even when we are writing a data of 1K,
qgroup will reserve a 4K size. It will cause the last 3K in a qgroup
is not available to user.
b). When user is writing a inline data, qgroup will not reserve it,
it means this is a window we can exceed the limit of a qgroup.
The main idea of this patch is reserving the data size of write_bytes
rather than the reserve_bytes. It means qgroup will not care about
the data size btrfs will reserve for user, but only care about the
data size user is going to write. Then reserve it when user want to
write and release it in transaction committed.
In this way, qgroup can be released from the complex procedure in
btrfs and only do the reserve when user want to write and account
when the data is written in commit_transaction().
Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 4d37746..1eef4ee 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3331,7 +3331,7 @@
num_pages *= 16;
num_pages *= PAGE_CACHE_SIZE;
- ret = btrfs_check_data_free_space(inode, num_pages);
+ ret = btrfs_check_data_free_space(inode, num_pages, num_pages);
if (ret)
goto out_put;
@@ -3851,7 +3851,7 @@
* This will check the space that the inode allocates from to make sure we have
* enough space for bytes.
*/
-int btrfs_check_data_free_space(struct inode *inode, u64 bytes)
+int btrfs_check_data_free_space(struct inode *inode, u64 bytes, u64 write_bytes)
{
struct btrfs_space_info *data_sinfo;
struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -3969,7 +3969,7 @@
data_sinfo->flags, bytes, 1);
return -ENOSPC;
}
- ret = btrfs_qgroup_reserve(root, bytes);
+ ret = btrfs_qgroup_reserve(root, write_bytes);
if (ret)
goto out;
data_sinfo->bytes_may_use += bytes;
@@ -3995,7 +3995,6 @@
data_sinfo = root->fs_info->data_sinfo;
spin_lock(&data_sinfo->lock);
WARN_ON(data_sinfo->bytes_may_use < bytes);
- btrfs_qgroup_free(root, bytes);
data_sinfo->bytes_may_use -= bytes;
trace_btrfs_space_reservation(root->fs_info, "space_info",
data_sinfo->flags, bytes, 0);
@@ -5243,8 +5242,6 @@
u64 qgroup_reserved)
{
btrfs_block_rsv_release(root, rsv, (u64)-1);
- if (qgroup_reserved)
- btrfs_qgroup_free(root, qgroup_reserved);
}
/**
@@ -5478,11 +5475,8 @@
to_free = 0;
}
spin_unlock(&BTRFS_I(inode)->lock);
- if (dropped) {
- if (root->fs_info->quota_enabled)
- btrfs_qgroup_free(root, dropped * root->nodesize);
+ if (dropped)
to_free += btrfs_calc_trans_metadata_size(root, dropped);
- }
if (to_free) {
btrfs_block_rsv_release(root, block_rsv, to_free);
@@ -5524,9 +5518,6 @@
trace_btrfs_space_reservation(root->fs_info, "delalloc",
btrfs_ino(inode), to_free, 0);
- if (root->fs_info->quota_enabled) {
- btrfs_qgroup_free(root, dropped * root->nodesize);
- }
btrfs_block_rsv_release(root, &root->fs_info->delalloc_block_rsv,
to_free);
@@ -5551,7 +5542,7 @@
{
int ret;
- ret = btrfs_check_data_free_space(inode, num_bytes);
+ ret = btrfs_check_data_free_space(inode, num_bytes, num_bytes);
if (ret)
return ret;
@@ -5727,12 +5718,8 @@
set_extent_dirty(root->fs_info->pinned_extents, bytenr,
bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL);
- if (reserved) {
- btrfs_qgroup_update_reserved_bytes(root->fs_info,
- root->root_key.objectid,
- num_bytes, -1);
+ if (reserved)
trace_btrfs_reserved_extent_free(root, bytenr, num_bytes);
- }
return 0;
}
@@ -6470,9 +6457,6 @@
btrfs_put_block_group(cache);
trace_btrfs_reserved_extent_free(root, buf->start, buf->len);
pin = 0;
- btrfs_qgroup_update_reserved_bytes(root->fs_info,
- root->root_key.objectid,
- buf->len, -1);
}
out:
if (pin)
@@ -7205,9 +7189,6 @@
ret = btrfs_discard_extent(root, start, len, NULL);
btrfs_add_free_space(cache, start, len);
btrfs_update_reserved_bytes(cache, len, RESERVE_FREE, delalloc);
- btrfs_qgroup_update_reserved_bytes(root->fs_info,
- root->root_key.objectid,
- len, -1);
}
btrfs_put_block_group(cache);
@@ -7446,9 +7427,6 @@
BUG_ON(ret); /* logic error */
ret = alloc_reserved_file_extent(trans, root, 0, root_objectid,
0, owner, offset, ins, 1);
- btrfs_qgroup_update_reserved_bytes(root->fs_info,
- root->root_key.objectid,
- ins->offset, 1);
btrfs_put_block_group(block_group);
return ret;
}
@@ -7595,10 +7573,6 @@
return ERR_PTR(ret);
}
- btrfs_qgroup_update_reserved_bytes(root->fs_info,
- root_objectid,
- ins.offset, 1);
-
buf = btrfs_init_new_buffer(trans, root, ins.objectid, level);
BUG_ON(IS_ERR(buf)); /* -ENOMEM */