summaryrefslogtreecommitdiffstats
path: root/fs/btrfs/file.c
diff options
context:
space:
mode:
authorGreg Kroah-Hartman <gregkh@linuxfoundation.org>2012-10-14 22:41:27 -0700
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>2012-10-14 22:41:27 -0700
commitc362495586e8a3a6487a318fcd82eaf15ffe2142 (patch)
tree86f7b195d36ba198f24f86be327f21a8d24ec248 /fs/btrfs/file.c
parentb70936d9ffbf0f45f4fa13a03122f015f13ecdb0 (diff)
parentddffeb8c4d0331609ef2581d84de4d763607bd37 (diff)
Merge 3.7-rc1 into tty-linus
This syncs up the tty-linus branch to the latest in Linus's tree to get all of the UAPI stuff needed for the next set of patches to merge. Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'fs/btrfs/file.c')
-rw-r--r--fs/btrfs/file.c449
1 files changed, 409 insertions, 40 deletions
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 5caf285c6e4..9ab1bed8811 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -39,6 +39,7 @@
#include "tree-log.h"
#include "locking.h"
#include "compat.h"
+#include "volumes.h"
/*
* when auto defrag is enabled we
@@ -458,14 +459,15 @@ int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode,
* this drops all the extents in the cache that intersect the range
* [start, end]. Existing extents are split as required.
*/
-int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
- int skip_pinned)
+void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
+ int skip_pinned)
{
struct extent_map *em;
struct extent_map *split = NULL;
struct extent_map *split2 = NULL;
struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
u64 len = end - start + 1;
+ u64 gen;
int ret;
int testend = 1;
unsigned long flags;
@@ -477,11 +479,14 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
testend = 0;
}
while (1) {
+ int no_splits = 0;
+
if (!split)
split = alloc_extent_map();
if (!split2)
split2 = alloc_extent_map();
- BUG_ON(!split || !split2); /* -ENOMEM */
+ if (!split || !split2)
+ no_splits = 1;
write_lock(&em_tree->lock);
em = lookup_extent_mapping(em_tree, start, len);
@@ -490,6 +495,7 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
break;
}
flags = em->flags;
+ gen = em->generation;
if (skip_pinned && test_bit(EXTENT_FLAG_PINNED, &em->flags)) {
if (testend && em->start + em->len >= start + len) {
free_extent_map(em);
@@ -506,6 +512,8 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
compressed = test_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
clear_bit(EXTENT_FLAG_PINNED, &em->flags);
remove_extent_mapping(em_tree, em);
+ if (no_splits)
+ goto next;
if (em->block_start < EXTENT_MAP_LAST_BYTE &&
em->start < start) {
@@ -518,12 +526,13 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
split->block_len = em->block_len;
else
split->block_len = split->len;
-
+ split->generation = gen;
split->bdev = em->bdev;
split->flags = flags;
split->compress_type = em->compress_type;
ret = add_extent_mapping(em_tree, split);
BUG_ON(ret); /* Logic error */
+ list_move(&split->list, &em_tree->modified_extents);
free_extent_map(split);
split = split2;
split2 = NULL;
@@ -537,6 +546,7 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
split->bdev = em->bdev;
split->flags = flags;
split->compress_type = em->compress_type;
+ split->generation = gen;
if (compressed) {
split->block_len = em->block_len;
@@ -550,9 +560,11 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
ret = add_extent_mapping(em_tree, split);
BUG_ON(ret); /* Logic error */
+ list_move(&split->list, &em_tree->modified_extents);
free_extent_map(split);
split = NULL;
}
+next:
write_unlock(&em_tree->lock);
/* once for us */
@@ -564,7 +576,6 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
free_extent_map(split);
if (split2)
free_extent_map(split2);
- return 0;
}
/*
@@ -576,13 +587,13 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
* it is either truncated or split. Anything entirely inside the range
* is deleted from the tree.
*/
-int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode,
- u64 start, u64 end, u64 *hint_byte, int drop_cache)
+int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root, struct inode *inode,
+ struct btrfs_path *path, u64 start, u64 end,
+ u64 *drop_end, int drop_cache)
{
- struct btrfs_root *root = BTRFS_I(inode)->root;
struct extent_buffer *leaf;
struct btrfs_file_extent_item *fi;
- struct btrfs_path *path;
struct btrfs_key key;
struct btrfs_key new_key;
u64 ino = btrfs_ino(inode);
@@ -597,14 +608,12 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode,
int recow;
int ret;
int modify_tree = -1;
+ int update_refs = (root->ref_cows || root == root->fs_info->tree_root);
+ int found = 0;
if (drop_cache)
btrfs_drop_extent_cache(inode, start, end - 1, 0);
- path = btrfs_alloc_path();
- if (!path)
- return -ENOMEM;
-
if (start >= BTRFS_I(inode)->disk_i_size)
modify_tree = 0;
@@ -666,6 +675,7 @@ next_slot:
goto next_slot;
}
+ found = 1;
search_start = max(key.offset, start);
if (recow || !modify_tree) {
modify_tree = -1;
@@ -707,14 +717,13 @@ next_slot:
extent_end - start);
btrfs_mark_buffer_dirty(leaf);
- if (disk_bytenr > 0) {
+ if (update_refs && disk_bytenr > 0) {
ret = btrfs_inc_extent_ref(trans, root,
disk_bytenr, num_bytes, 0,
root->root_key.objectid,
new_key.objectid,
start - extent_offset, 0);
BUG_ON(ret); /* -ENOMEM */
- *hint_byte = disk_bytenr;
}
key.offset = start;
}
@@ -734,10 +743,8 @@ next_slot:
btrfs_set_file_extent_num_bytes(leaf, fi,
extent_end - end);
btrfs_mark_buffer_dirty(leaf);
- if (disk_bytenr > 0) {
+ if (update_refs && disk_bytenr > 0)
inode_sub_bytes(inode, end - key.offset);
- *hint_byte = disk_bytenr;
- }
break;
}
@@ -753,10 +760,8 @@ next_slot:
btrfs_set_file_extent_num_bytes(leaf, fi,
start - key.offset);
btrfs_mark_buffer_dirty(leaf);
- if (disk_bytenr > 0) {
+ if (update_refs && disk_bytenr > 0)
inode_sub_bytes(inode, extent_end - start);
- *hint_byte = disk_bytenr;
- }
if (end == extent_end)
break;
@@ -777,12 +782,13 @@ next_slot:
del_nr++;
}
- if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
+ if (update_refs &&
+ extent_type == BTRFS_FILE_EXTENT_INLINE) {
inode_sub_bytes(inode,
extent_end - key.offset);
extent_end = ALIGN(extent_end,
root->sectorsize);
- } else if (disk_bytenr > 0) {
+ } else if (update_refs && disk_bytenr > 0) {
ret = btrfs_free_extent(trans, root,
disk_bytenr, num_bytes, 0,
root->root_key.objectid,
@@ -791,7 +797,6 @@ next_slot:
BUG_ON(ret); /* -ENOMEM */
inode_sub_bytes(inode,
extent_end - key.offset);
- *hint_byte = disk_bytenr;
}
if (end == extent_end)
@@ -806,7 +811,7 @@ next_slot:
del_nr);
if (ret) {
btrfs_abort_transaction(trans, root, ret);
- goto out;
+ break;
}
del_nr = 0;
@@ -825,7 +830,24 @@ next_slot:
btrfs_abort_transaction(trans, root, ret);
}
-out:
+ if (drop_end)
+ *drop_end = found ? min(end, extent_end) : end;
+ btrfs_release_path(path);
+ return ret;
+}
+
+int btrfs_drop_extents(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root, struct inode *inode, u64 start,
+ u64 end, int drop_cache)
+{
+ struct btrfs_path *path;
+ int ret;
+
+ path = btrfs_alloc_path();
+ if (!path)
+ return -ENOMEM;
+ ret = __btrfs_drop_extents(trans, root, inode, path, start, end, NULL,
+ drop_cache);
btrfs_free_path(path);
return ret;
}
@@ -892,8 +914,6 @@ int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
int ret;
u64 ino = btrfs_ino(inode);
- btrfs_drop_extent_cache(inode, start, end - 1, 0);
-
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
@@ -935,12 +955,16 @@ again:
btrfs_set_item_key_safe(trans, root, path, &new_key);
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
+ btrfs_set_file_extent_generation(leaf, fi,
+ trans->transid);
btrfs_set_file_extent_num_bytes(leaf, fi,
extent_end - end);
btrfs_set_file_extent_offset(leaf, fi,
end - orig_offset);
fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
struct btrfs_file_extent_item);
+ btrfs_set_file_extent_generation(leaf, fi,
+ trans->transid);
btrfs_set_file_extent_num_bytes(leaf, fi,
end - other_start);
btrfs_mark_buffer_dirty(leaf);
@@ -958,12 +982,16 @@ again:
struct btrfs_file_extent_item);
btrfs_set_file_extent_num_bytes(leaf, fi,
start - key.offset);
+ btrfs_set_file_extent_generation(leaf, fi,
+ trans->transid);
path->slots[0]++;
new_key.offset = start;
btrfs_set_item_key_safe(trans, root, path, &new_key);
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
+ btrfs_set_file_extent_generation(leaf, fi,
+ trans->transid);
btrfs_set_file_extent_num_bytes(leaf, fi,
other_end - start);
btrfs_set_file_extent_offset(leaf, fi,
@@ -991,12 +1019,14 @@ again:
leaf = path->nodes[0];
fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
struct btrfs_file_extent_item);
+ btrfs_set_file_extent_generation(leaf, fi, trans->transid);
btrfs_set_file_extent_num_bytes(leaf, fi,
split - key.offset);
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
+ btrfs_set_file_extent_generation(leaf, fi, trans->transid);
btrfs_set_file_extent_offset(leaf, fi, split - orig_offset);
btrfs_set_file_extent_num_bytes(leaf, fi,
extent_end - split);
@@ -1056,12 +1086,14 @@ again:
struct btrfs_file_extent_item);
btrfs_set_file_extent_type(leaf, fi,
BTRFS_FILE_EXTENT_REG);
+ btrfs_set_file_extent_generation(leaf, fi, trans->transid);
btrfs_mark_buffer_dirty(leaf);
} else {
fi = btrfs_item_ptr(leaf, del_slot - 1,
struct btrfs_file_extent_item);
btrfs_set_file_extent_type(leaf, fi,
BTRFS_FILE_EXTENT_REG);
+ btrfs_set_file_extent_generation(leaf, fi, trans->transid);
btrfs_set_file_extent_num_bytes(leaf, fi,
extent_end - key.offset);
btrfs_mark_buffer_dirty(leaf);
@@ -1173,8 +1205,8 @@ again:
clear_extent_bit(&BTRFS_I(inode)->io_tree, start_pos,
last_pos - 1, EXTENT_DIRTY | EXTENT_DELALLOC |
- EXTENT_DO_ACCOUNTING, 0, 0, &cached_state,
- GFP_NOFS);
+ EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
+ 0, 0, &cached_state, GFP_NOFS);
unlock_extent_cached(&BTRFS_I(inode)->io_tree,
start_pos, last_pos - 1, &cached_state,
GFP_NOFS);
@@ -1514,16 +1546,24 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
trace_btrfs_sync_file(file, datasync);
+ /*
+ * We write the dirty pages in the range and wait until they complete
+ * out of the ->i_mutex. If so, we can flush the dirty pages by
+ * multi-task, and make the performance up.
+ */
+ ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+ if (ret)
+ return ret;
+
mutex_lock(&inode->i_mutex);
/*
- * we wait first, since the writeback may change the inode, also wait
- * ordered range does a filemape_write_and_wait_range which is why we
- * don't do it above like other file systems.
+ * We flush the dirty pages again to avoid some dirty pages in the
+ * range being left.
*/
- root->log_batch++;
+ atomic_inc(&root->log_batch);
btrfs_wait_ordered_range(inode, start, end);
- root->log_batch++;
+ atomic_inc(&root->log_batch);
/*
* check the transaction that last modified this inode
@@ -1544,6 +1584,14 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
BTRFS_I(inode)->last_trans <=
root->fs_info->last_trans_committed) {
BTRFS_I(inode)->last_trans = 0;
+
+ /*
+ * We'v had everything committed since the last time we were
+ * modified so clear this flag in case it was set for whatever
+ * reason, it's no longer relevant.
+ */
+ clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+ &BTRFS_I(inode)->runtime_flags);
mutex_unlock(&inode->i_mutex);
goto out;
}
@@ -1599,6 +1647,7 @@ out:
static const struct vm_operations_struct btrfs_file_vm_ops = {
.fault = filemap_fault,
.page_mkwrite = btrfs_page_mkwrite,
+ .remap_pages = generic_file_remap_pages,
};
static int btrfs_file_mmap(struct file *filp, struct vm_area_struct *vma)
@@ -1610,11 +1659,328 @@ static int btrfs_file_mmap(struct file *filp, struct vm_area_struct *vma)
file_accessed(filp);
vma->vm_ops = &btrfs_file_vm_ops;
- vma->vm_flags |= VM_CAN_NONLINEAR;
return 0;
}
+static int hole_mergeable(struct inode *inode, struct extent_buffer *leaf,
+ int slot, u64 start, u64 end)
+{
+ struct btrfs_file_extent_item *fi;
+ struct btrfs_key key;
+
+ if (slot < 0 || slot >= btrfs_header_nritems(leaf))
+ return 0;
+
+ btrfs_item_key_to_cpu(leaf, &key, slot);
+ if (key.objectid != btrfs_ino(inode) ||
+ key.type != BTRFS_EXTENT_DATA_KEY)
+ return 0;
+
+ fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
+
+ if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG)
+ return 0;
+
+ if (btrfs_file_extent_disk_bytenr(leaf, fi))
+ return 0;
+
+ if (key.offset == end)
+ return 1;
+ if (key.offset + btrfs_file_extent_num_bytes(leaf, fi) == start)
+ return 1;
+ return 0;
+}
+
+static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode,
+ struct btrfs_path *path, u64 offset, u64 end)
+{
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct extent_buffer *leaf;
+ struct btrfs_file_extent_item *fi;
+ struct extent_map *hole_em;
+ struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
+ struct btrfs_key key;
+ int ret;
+
+ key.objectid = btrfs_ino(inode);
+ key.type = BTRFS_EXTENT_DATA_KEY;
+ key.offset = offset;
+
+
+ ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
+ if (ret < 0)
+ return ret;
+ BUG_ON(!ret);
+
+ leaf = path->nodes[0];
+ if (hole_mergeable(inode, leaf, path->slots[0]-1, offset, end)) {
+ u64 num_bytes;
+
+ path->slots[0]--;
+ fi = btrfs_item_ptr(leaf, path->slots[0],
+ struct btrfs_file_extent_item);
+ num_bytes = btrfs_file_extent_num_bytes(leaf, fi) +
+ end - offset;
+ btrfs_set_file_extent_num_bytes(leaf, fi, num_bytes);
+ btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes);
+ btrfs_set_file_extent_offset(leaf, fi, 0);
+ btrfs_mark_buffer_dirty(leaf);
+ goto out;
+ }
+
+ if (hole_mergeable(inode, leaf, path->slots[0]+1, offset, end)) {
+ u64 num_bytes;
+
+ path->slots[0]++;
+ key.offset = offset;
+ btrfs_set_item_key_safe(trans, root, path, &key);
+ fi = btrfs_item_ptr(leaf, path->slots[0],
+ struct btrfs_file_extent_item);
+ num_bytes = btrfs_file_extent_num_bytes(leaf, fi) + end -
+ offset;
+ btrfs_set_file_extent_num_bytes(leaf, fi, num_bytes);
+ btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes);
+ btrfs_set_file_extent_offset(leaf, fi, 0);
+ btrfs_mark_buffer_dirty(leaf);
+ goto out;
+ }
+ btrfs_release_path(path);
+
+ ret = btrfs_insert_file_extent(trans, root, btrfs_ino(inode), offset,
+ 0, 0, end - offset, 0, end - offset,
+ 0, 0, 0);
+ if (ret)
+ return ret;
+
+out:
+ btrfs_release_path(path);
+
+ hole_em = alloc_extent_map();
+ if (!hole_em) {
+ btrfs_drop_extent_cache(inode, offset, end - 1, 0);
+ set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+ &BTRFS_I(inode)->runtime_flags);
+ } else {
+ hole_em->start = offset;
+ hole_em->len = end - offset;
+ hole_em->orig_start = offset;
+
+ hole_em->block_start = EXTENT_MAP_HOLE;
+ hole_em->block_len = 0;
+ hole_em->bdev = root->fs_info->fs_devices->latest_bdev;
+ hole_em->compress_type = BTRFS_COMPRESS_NONE;
+ hole_em->generation = trans->transid;
+
+ do {
+ btrfs_drop_extent_cache(inode, offset, end - 1, 0);
+ write_lock(&em_tree->lock);
+ ret = add_extent_mapping(em_tree, hole_em);
+ if (!ret)
+ list_move(&hole_em->list,
+ &em_tree->modified_extents);
+ write_unlock(&em_tree->lock);
+ } while (ret == -EEXIST);
+ free_extent_map(hole_em);
+ if (ret)
+ set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+ &BTRFS_I(inode)->runtime_flags);
+ }
+
+ return 0;
+}
+
+static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
+{
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct extent_state *cached_state = NULL;
+ struct btrfs_path *path;
+ struct btrfs_block_rsv *rsv;
+ struct btrfs_trans_handle *trans;
+ u64 mask = BTRFS_I(inode)->root->sectorsize - 1;
+ u64 lockstart = (offset + mask) & ~mask;
+ u64 lockend = ((offset + len) & ~mask) - 1;
+ u64 cur_offset = lockstart;
+ u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
+ u64 drop_end;
+ unsigned long nr;
+ int ret = 0;
+ int err = 0;
+ bool same_page = (offset >> PAGE_CACHE_SHIFT) ==
+ ((offset + len) >> PAGE_CACHE_SHIFT);
+
+ btrfs_wait_ordered_range(inode, offset, len);
+
+ mutex_lock(&inode->i_mutex);
+ if (offset >= inode->i_size) {
+ mutex_unlock(&inode->i_mutex);
+ return 0;
+ }
+
+ /*
+ * Only do this if we are in the same page and we aren't doing the
+ * entire page.
+ */
+ if (same_page && len < PAGE_CACHE_SIZE) {
+ ret = btrfs_truncate_page(inode, offset, len, 0);
+ mutex_unlock(&inode->i_mutex);
+ return ret;
+ }
+
+ /* zero back part of the first page */
+ ret = btrfs_truncate_page(inode, offset, 0, 0);
+ if (ret) {
+ mutex_unlock(&inode->i_mutex);
+ return ret;
+ }
+
+ /* zero the front end of the last page */
+ ret = btrfs_truncate_page(inode, offset + len, 0, 1);
+ if (ret) {
+ mutex_unlock(&inode->i_mutex);
+ return ret;
+ }
+
+ if (lockend < lockstart) {
+ mutex_unlock(&inode->i_mutex);
+ return 0;
+ }
+
+ while (1) {
+ struct btrfs_ordered_extent *ordered;
+
+ truncate_pagecache_range(inode, lockstart, lockend);
+
+ lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend,
+ 0, &cached_state);
+ ordered = btrfs_lookup_first_ordered_extent(inode, lockend);
+
+ /*
+ * We need to make sure we have no ordered extents in this range
+ * and nobody raced in and read a page in this range, if we did
+ * we need to try again.
+ */
+ if ((!ordered ||
+ (ordered->file_offset + ordered->len < lockstart ||
+ ordered->file_offset > lockend)) &&
+ !test_range_bit(&BTRFS_I(inode)->io_tree, lockstart,
+ lockend, EXTENT_UPTODATE, 0,
+ cached_state)) {
+ if (ordered)
+ btrfs_put_ordered_extent(ordered);
+ break;
+ }
+ if (ordered)
+ btrfs_put_ordered_extent(ordered);
+ unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart,
+ lockend, &cached_state, GFP_NOFS);
+ btrfs_wait_ordered_range(inode, lockstart,
+ lockend - lockstart + 1);
+ }
+
+ path = btrfs_alloc_path();
+ if (!path) {
+ ret = -ENOMEM;
+ goto out;
+ }
+
+ rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP);
+ if (!rsv) {
+ ret = -ENOMEM;
+ goto out_free;
+ }
+ rsv->size = btrfs_calc_trunc_metadata_size(root, 1);
+ rsv->failfast = 1;
+
+ /*
+ * 1 - update the inode
+ * 1 - removing the extents in the range
+ * 1 - adding the hole extent
+ */
+ trans = btrfs_start_transaction(root, 3);
+ if (IS_ERR(trans)) {
+ err = PTR_ERR(trans);
+ goto out_free;
+ }
+
+ ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, rsv,
+ min_size);
+ BUG_ON(ret);
+ trans->block_rsv = rsv;
+
+ while (cur_offset < lockend) {
+ ret = __btrfs_drop_extents(trans, root, inode, path,
+ cur_offset, lockend + 1,
+ &drop_end, 1);
+ if (ret != -ENOSPC)
+ break;
+
+ trans->block_rsv = &root->fs_info->trans_block_rsv;
+
+ ret = fill_holes(trans, inode, path, cur_offset, drop_end);
+ if (ret) {
+ err = ret;
+ break;
+ }
+
+ cur_offset = drop_end;
+
+ ret = btrfs_update_inode(trans, root, inode);
+ if (ret) {
+ err = ret;
+ break;
+ }
+
+ nr = trans->blocks_used;
+ btrfs_end_transaction(trans, root);
+ btrfs_btree_balance_dirty(root, nr);
+
+ trans = btrfs_start_transaction(root, 3);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ trans = NULL;
+ break;
+ }
+
+ ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv,
+ rsv, min_size);
+ BUG_ON(ret); /* shouldn't happen */
+ trans->block_rsv = rsv;
+ }
+
+ if (ret) {
+ err = ret;
+ goto out_trans;
+ }
+
+ trans->block_rsv = &root->fs_info->trans_block_rsv;
+ ret = fill_holes(trans, inode, path, cur_offset, drop_end);
+ if (ret) {
+ err = ret;
+ goto out_trans;
+ }
+
+out_trans:
+ if (!trans)
+ goto out_free;
+
+ trans->block_rsv = &root->fs_info->trans_block_rsv;
+ ret = btrfs_update_inode(trans, root, inode);
+ nr = trans->blocks_used;
+ btrfs_end_transaction(trans, root);
+ btrfs_btree_balance_dirty(root, nr);
+out_free:
+ btrfs_free_path(path);
+ btrfs_free_block_rsv(root, rsv);
+out:
+ unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
+ &cached_state, GFP_NOFS);
+ mutex_unlock(&inode->i_mutex);
+ if (ret && !err)
+ err = ret;
+ return err;
+}
+
static long btrfs_fallocate(struct file *file, int mode,
loff_t offset, loff_t len)
{
@@ -1633,15 +1999,18 @@ static long btrfs_fallocate(struct file *file, int mode,
alloc_start = offset & ~mask;
alloc_end = (offset + len + mask) & ~mask;
- /* We only support the FALLOC_FL_KEEP_SIZE mode */
- if (mode & ~FALLOC_FL_KEEP_SIZE)
+ /* Make sure we aren't being give some crap mode */
+ if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
return -EOPNOTSUPP;
+ if (mode & FALLOC_FL_PUNCH_HOLE)
+ return btrfs_punch_hole(inode, offset, len);
+
/*
* Make sure we have enough space before we do the
* allocation.
*/
- ret = btrfs_check_data_free_space(inode, len);
+ ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start + 1);
if (ret)
return ret;
@@ -1748,7 +2117,7 @@ static long btrfs_fallocate(struct file *file, int mode,
out:
mutex_unlock(&inode->i_mutex);
/* Let go of our reservation. */
- btrfs_free_reserved_data_space(inode, len);
+ btrfs_free_reserved_data_space(inode, alloc_end - alloc_start + 1);
return ret;
}