aboutsummaryrefslogtreecommitdiffstats
path: root/fs/btrfs/extent_io.h
diff options
context:
space:
mode:
authorJosef Bacik <josef@redhat.com>2009-09-11 16:12:44 -0400
committerChris Mason <chris.mason@oracle.com>2009-09-28 16:29:42 -0400
commit9ed74f2dba6ebf9f30b80554290bfc73cc3ef083 (patch)
tree763d58a4a11ceca26dcdaedefb1fd662c4e2fa8b /fs/btrfs/extent_io.h
parentc65ddb52dc412c9b67681b1aa16cd1bac8434e24 (diff)
Btrfs: proper -ENOSPC handling
At the start of a transaction we do a btrfs_reserve_metadata_space() and specify how many items we plan on modifying. Then once we've done our modifications and such, just call btrfs_unreserve_metadata_space() for the same number of items we reserved. For keeping track of metadata needed for data I've had to add an extent_io op for when we merge extents. This lets us track space properly when we are doing sequential writes, so we don't end up reserving way more metadata space than what we need. The only place where the metadata space accounting is not done is in the relocation code. This is because Yan is going to be reworking that code in the near future, so running btrfs-vol -b could still possibly result in a ENOSPC related panic. This patch also turns off the metadata_ratio stuff in order to allow users to more efficiently use their disk space. This patch makes it so we track how much metadata we need for an inode's delayed allocation extents by tracking how many extents are currently waiting for allocation. It introduces two new callbacks for the extent_io tree's, merge_extent_hook and split_extent_hook. These help us keep track of when we merge delalloc extents together and split them up. Reservations are handled prior to any actually dirty'ing occurs, and then we unreserve after we dirty. btrfs_unreserve_metadata_for_delalloc() will make the appropriate unreservations as needed based on the number of reservations we currently have and the number of extents we currently have. Doing the reservation outside of doing any of the actual dirty'ing lets us do things like filemap_flush() the inode to try and force delalloc to happen, or as a last resort actually start allocation on all delalloc inodes in the fs. This has survived dbench, fs_mark and an fsx torture test. Signed-off-by: Josef Bacik <jbacik@redhat.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Diffstat (limited to 'fs/btrfs/extent_io.h')
-rw-r--r--fs/btrfs/extent_io.h13
1 files changed, 11 insertions, 2 deletions
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 14ed16fd862d..4794ec891fed 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -60,8 +60,13 @@ struct extent_io_ops {
60 struct extent_state *state, int uptodate); 60 struct extent_state *state, int uptodate);
61 int (*set_bit_hook)(struct inode *inode, u64 start, u64 end, 61 int (*set_bit_hook)(struct inode *inode, u64 start, u64 end,
62 unsigned long old, unsigned long bits); 62 unsigned long old, unsigned long bits);
63 int (*clear_bit_hook)(struct inode *inode, u64 start, u64 end, 63 int (*clear_bit_hook)(struct inode *inode, struct extent_state *state,
64 unsigned long old, unsigned long bits); 64 unsigned long bits);
65 int (*merge_extent_hook)(struct inode *inode,
66 struct extent_state *new,
67 struct extent_state *other);
68 int (*split_extent_hook)(struct inode *inode,
69 struct extent_state *orig, u64 split);
65 int (*write_cache_pages_lock_hook)(struct page *page); 70 int (*write_cache_pages_lock_hook)(struct page *page);
66}; 71};
67 72
@@ -79,10 +84,14 @@ struct extent_state {
79 u64 start; 84 u64 start;
80 u64 end; /* inclusive */ 85 u64 end; /* inclusive */
81 struct rb_node rb_node; 86 struct rb_node rb_node;
87
88 /* ADD NEW ELEMENTS AFTER THIS */
82 struct extent_io_tree *tree; 89 struct extent_io_tree *tree;
83 wait_queue_head_t wq; 90 wait_queue_head_t wq;
84 atomic_t refs; 91 atomic_t refs;
85 unsigned long state; 92 unsigned long state;
93 u64 split_start;
94 u64 split_end;
86 95
87 /* for use by the FS */ 96 /* for use by the FS */
88 u64 private; 97 u64 private;