Btrfs: Update metadata reservation for delayed allocation

Introduce metadata reservation context for delayed allocation
and update various related functions.

This patch also introduces EXTENT_FIRST_DELALLOC control bit for
set/clear_extent_bit. It tells set/clear_bit_hook whether they
are processing the first extent_state with EXTENT_DELALLOC bit
set. This change is important if set/clear_extent_bit involves
multiple extent_state.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>

1 files changed, 30 insertions, 33 deletions

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index d2d03684fab2..1a57c17d4029 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -335,21 +335,18 @@ static int merge_state(struct extent_io_tree *tree,
 }
 
 static int set_state_cb(struct extent_io_tree *tree,
-                         struct extent_state *state,
-                         unsigned long bits)
+                         struct extent_state *state, int *bits)
 {
         if (tree->ops && tree->ops->set_bit_hook) {
                 return tree->ops->set_bit_hook(tree->mapping->host,
-                                               state->start, state->end,
-                                               state->state, bits);
+                                               state, bits);
         }
 
         return 0;
 }
 
 static void clear_state_cb(struct extent_io_tree *tree,
-                           struct extent_state *state,
-                           unsigned long bits)
+                           struct extent_state *state, int *bits)
 {
         if (tree->ops && tree->ops->clear_bit_hook)
                 tree->ops->clear_bit_hook(tree->mapping->host, state, bits);
@@ -367,9 +364,10 @@ static void clear_state_cb(struct extent_io_tree *tree,
  */
 static int insert_state(struct extent_io_tree *tree,
                         struct extent_state *state, u64 start, u64 end,
-                        int bits)
+                        int *bits)
 {
         struct rb_node *node;
+        int bits_to_set = *bits & ~EXTENT_CTLBITS;
         int ret;
 
         if (end < start) {
@@ -384,9 +382,9 @@ static int insert_state(struct extent_io_tree *tree,
         if (ret)
                 return ret;
 
-        if (bits & EXTENT_DIRTY)
+        if (bits_to_set & EXTENT_DIRTY)
                 tree->dirty_bytes += end - start + 1;
-        state->state |= bits;
+        state->state |= bits_to_set;
         node = tree_insert(&tree->state, end, &state->rb_node);
         if (node) {
                 struct extent_state *found;
@@ -456,13 +454,13 @@ static int split_state(struct extent_io_tree *tree, struct extent_state *orig,
  * struct is freed and removed from the tree
  */
 static int clear_state_bit(struct extent_io_tree *tree,
-                            struct extent_state *state, int bits, int wake,
-                            int delete)
+                            struct extent_state *state,
+                            int *bits, int wake)
 {
-        int bits_to_clear = bits & ~EXTENT_DO_ACCOUNTING;
+        int bits_to_clear = *bits & ~EXTENT_CTLBITS;
         int ret = state->state & bits_to_clear;
 
-        if ((bits & EXTENT_DIRTY) && (state->state & EXTENT_DIRTY)) {
+        if ((bits_to_clear & EXTENT_DIRTY) && (state->state & EXTENT_DIRTY)) {
                 u64 range = state->end - state->start + 1;
                 WARN_ON(range > tree->dirty_bytes);
                 tree->dirty_bytes -= range;
@@ -471,9 +469,8 @@ static int clear_state_bit(struct extent_io_tree *tree,
         state->state &= ~bits_to_clear;
         if (wake)
                 wake_up(&state->wq);
-        if (delete || state->state == 0) {
+        if (state->state == 0) {
                 if (state->tree) {
-                        clear_state_cb(tree, state, state->state);
                         rb_erase(&state->rb_node, &tree->state);
                         state->tree = NULL;
                         free_extent_state(state);
@@ -514,6 +511,10 @@ int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
         int set = 0;
         int clear = 0;
 
+        if (delete)
+                bits |= ~EXTENT_CTLBITS;
+        bits |= EXTENT_FIRST_DELALLOC;
+
         if (bits & (EXTENT_IOBITS | EXTENT_BOUNDARY))
                 clear = 1;
 again:
@@ -580,8 +581,7 @@ hit_next:
                 if (err)
                         goto out;
                 if (state->end <= end) {
-                        set |= clear_state_bit(tree, state, bits, wake,
-                                               delete);
+                        set |= clear_state_bit(tree, state, &bits, wake);
                         if (last_end == (u64)-1)
                                 goto out;
                         start = last_end + 1;
@@ -602,7 +602,7 @@ hit_next:
                 if (wake)
                         wake_up(&state->wq);
 
-                set |= clear_state_bit(tree, prealloc, bits, wake, delete);
+                set |= clear_state_bit(tree, prealloc, &bits, wake);
 
                 prealloc = NULL;
                 goto out;
@@ -613,7 +613,7 @@ hit_next:
         else
                 next_node = NULL;
 
-        set |= clear_state_bit(tree, state, bits, wake, delete);
+        set |= clear_state_bit(tree, state, &bits, wake);
         if (last_end == (u64)-1)
                 goto out;
         start = last_end + 1;
@@ -706,19 +706,19 @@ out:
 
 static int set_state_bits(struct extent_io_tree *tree,
                            struct extent_state *state,
-                           int bits)
+                           int *bits)
 {
         int ret;
+        int bits_to_set = *bits & ~EXTENT_CTLBITS;
 
         ret = set_state_cb(tree, state, bits);
         if (ret)
                 return ret;
-
-        if ((bits & EXTENT_DIRTY) && !(state->state & EXTENT_DIRTY)) {
+        if ((bits_to_set & EXTENT_DIRTY) && !(state->state & EXTENT_DIRTY)) {
                 u64 range = state->end - state->start + 1;
                 tree->dirty_bytes += range;
         }
-        state->state |= bits;
+        state->state |= bits_to_set;
 
         return 0;
 }
@@ -757,6 +757,7 @@ static int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
         u64 last_start;
         u64 last_end;
 
+        bits |= EXTENT_FIRST_DELALLOC;
 again:
         if (!prealloc && (mask & __GFP_WAIT)) {
                 prealloc = alloc_extent_state(mask);
@@ -778,7 +779,7 @@ again:
          */
         node = tree_search(tree, start);
         if (!node) {
-                err = insert_state(tree, prealloc, start, end, bits);
+                err = insert_state(tree, prealloc, start, end, &bits);
                 prealloc = NULL;
                 BUG_ON(err == -EEXIST);
                 goto out;
@@ -802,7 +803,7 @@ hit_next:
                         goto out;
                 }
 
-                err = set_state_bits(tree, state, bits);
+                err = set_state_bits(tree, state, &bits);
                 if (err)
                         goto out;
 
@@ -852,7 +853,7 @@ hit_next:
                 if (err)
                         goto out;
                 if (state->end <= end) {
-                        err = set_state_bits(tree, state, bits);
+                        err = set_state_bits(tree, state, &bits);
                         if (err)
                                 goto out;
                         cache_state(state, cached_state);
@@ -877,7 +878,7 @@ hit_next:
                 else
                         this_end = last_start - 1;
                 err = insert_state(tree, prealloc, start, this_end,
-                                   bits);
+                                   &bits);
                 BUG_ON(err == -EEXIST);
                 if (err) {
                         prealloc = NULL;
@@ -903,7 +904,7 @@ hit_next:
                 err = split_state(tree, state, prealloc, end + 1);
                 BUG_ON(err == -EEXIST);
 
-                err = set_state_bits(tree, prealloc, bits);
+                err = set_state_bits(tree, prealloc, &bits);
                 if (err) {
                         prealloc = NULL;
                         goto out;
@@ -966,8 +967,7 @@ int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
 {
         return clear_extent_bit(tree, start, end,
                                 EXTENT_DIRTY | EXTENT_DELALLOC |
-                                EXTENT_DO_ACCOUNTING, 0, 0,
-                                NULL, mask);
+                                EXTENT_DO_ACCOUNTING, 0, 0, NULL, mask);
 }
 
 int set_extent_new(struct extent_io_tree *tree, u64 start, u64 end,
@@ -1435,9 +1435,6 @@ int extent_clear_unlock_delalloc(struct inode *inode,
         if (op & EXTENT_CLEAR_DELALLOC)
                 clear_bits |= EXTENT_DELALLOC;
 
-        if (op & EXTENT_CLEAR_ACCOUNTING)
-                clear_bits |= EXTENT_DO_ACCOUNTING;
-
         clear_extent_bit(tree, start, end, clear_bits, 1, 0, NULL, GFP_NOFS);
         if (!(op & (EXTENT_CLEAR_UNLOCK_PAGE | EXTENT_CLEAR_DIRTY |
                     EXTENT_SET_WRITEBACK | EXTENT_END_WRITEBACK |