split->block_len = em->block_len;
else
split->block_len = split->len;
+ split->orig_block_len = max(split->block_len,
+ em->orig_block_len);
split->generation = gen;
split->bdev = em->bdev;
split->flags = flags;
split->flags = flags;
split->compress_type = em->compress_type;
split->generation = gen;
+ split->orig_block_len = max(em->block_len,
+ em->orig_block_len);
if (compressed) {
split->block_len = em->block_len;
} else {
split->block_len = split->len;
split->block_start = em->block_start + diff;
- split->orig_start = split->start;
+ split->orig_start = em->orig_start;
}
ret = add_extent_mapping(em_tree, split);
cond_resched();
- balance_dirty_pages_ratelimited_nr(inode->i_mapping,
- dirty_pages);
+ balance_dirty_pages_ratelimited(inode->i_mapping);
if (dirty_pages < (root->leafsize >> PAGE_CACHE_SHIFT) + 1)
btrfs_btree_balance_dirty(root);
return written ? written : err;
}
+static void update_time_for_write(struct inode *inode)
+{
+ struct timespec now;
+
+ if (IS_NOCMTIME(inode))
+ return;
+
+ now = current_fs_time(inode->i_sb);
+ if (!timespec_equal(&inode->i_mtime, &now))
+ inode->i_mtime = now;
+
+ if (!timespec_equal(&inode->i_ctime, &now))
+ inode->i_ctime = now;
+
+ if (IS_I_VERSION(inode))
+ inode_inc_iversion(inode);
+}
+
static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
const struct iovec *iov,
unsigned long nr_segs, loff_t pos)
ssize_t num_written = 0;
ssize_t err = 0;
size_t count, ocount;
+ bool sync = (file->f_flags & O_DSYNC) || IS_SYNC(file->f_mapping->host);
sb_start_write(inode->i_sb);
goto out;
}
- err = file_update_time(file);
- if (err) {
- mutex_unlock(&inode->i_mutex);
- goto out;
- }
+ /*
+ * We reserve space for updating the inode when we reserve space for the
+ * extent we are going to write, so we will enospc out there. We don't
+ * need to start yet another transaction to update the inode as we will
+ * update the inode when we finish writing whatever data we write.
+ */
+ update_time_for_write(inode);
start_pos = round_down(pos, root->sectorsize);
if (start_pos > i_size_read(inode)) {
}
}
+ if (sync)
+ atomic_inc(&BTRFS_I(inode)->sync_writers);
+
if (unlikely(file->f_flags & O_DIRECT)) {
num_written = __btrfs_direct_write(iocb, iov, nr_segs,
pos, ppos, count, ocount);
* this will either be one more than the running transaction
* or the generation used for the next transaction if there isn't
* one running right now.
+ *
+ * We also have to set last_sub_trans to the current log transid,
+ * otherwise subsequent syncs to a file that's been synced in this
+ * transaction will appear to have already occured.
*/
BTRFS_I(inode)->last_trans = root->fs_info->generation + 1;
+ BTRFS_I(inode)->last_sub_trans = root->log_transid;
if (num_written > 0 || num_written == -EIOCBQUEUED) {
err = generic_write_sync(file, pos, num_written);
if (err < 0 && num_written > 0)
num_written = err;
}
out:
+ if (sync)
+ atomic_dec(&BTRFS_I(inode)->sync_writers);
sb_end_write(inode->i_sb);
current->backing_dev_info = NULL;
return num_written ? num_written : err;
* out of the ->i_mutex. If so, we can flush the dirty pages by
* multi-task, and make the performance up.
*/
+ atomic_inc(&BTRFS_I(inode)->sync_writers);
ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+ atomic_dec(&BTRFS_I(inode)->sync_writers);
if (ret)
return ret;
hole_em->block_start = EXTENT_MAP_HOLE;
hole_em->block_len = 0;
+ hole_em->orig_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;
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 lockstart = round_up(offset, BTRFS_I(inode)->root->sectorsize);
+ u64 lockend = round_down(offset + len,
+ BTRFS_I(inode)->root->sectorsize) - 1;
u64 cur_offset = lockstart;
u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
u64 drop_end;
int ret = 0;
int err = 0;
- bool same_page = (offset >> PAGE_CACHE_SHIFT) ==
- ((offset + len) >> PAGE_CACHE_SHIFT);
+ bool same_page = ((offset >> PAGE_CACHE_SHIFT) ==
+ ((offset + len - 1) >> 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;
- }
-
+ /*
+ * We needn't truncate any page which is beyond the end of the file
+ * because we are sure there is no data there.
+ */
/*
* 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);
+ if (offset < round_up(inode->i_size, 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;
+ if (offset < round_up(inode->i_size, PAGE_CACHE_SIZE)) {
+ 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 (offset + len < round_up(inode->i_size, PAGE_CACHE_SIZE)) {
+ ret = btrfs_truncate_page(inode, offset + len, 0, 1);
+ if (ret) {
+ mutex_unlock(&inode->i_mutex);
+ return ret;
+ }
}
if (lockend < lockstart) {
* Make sure we have enough space before we do the
* allocation.
*/
- ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start + 1);
+ ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start);
if (ret)
return ret;
out:
mutex_unlock(&inode->i_mutex);
/* Let go of our reservation. */
- btrfs_free_reserved_data_space(inode, alloc_end - alloc_start + 1);
+ btrfs_free_reserved_data_space(inode, alloc_end - alloc_start);
return ret;
}
-static int find_desired_extent(struct inode *inode, loff_t *offset, int origin)
+static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct extent_map *em;
if (lockend <= lockstart)
lockend = lockstart + root->sectorsize;
+ lockend--;
len = lockend - lockstart + 1;
len = max_t(u64, len, root->sectorsize);
* before the position we want in case there is outstanding delalloc
* going on here.
*/
- if (origin == SEEK_HOLE && start != 0) {
+ if (whence == SEEK_HOLE && start != 0) {
if (start <= root->sectorsize)
em = btrfs_get_extent_fiemap(inode, NULL, 0, 0,
root->sectorsize, 0);
}
}
- if (origin == SEEK_HOLE) {
+ if (whence == SEEK_HOLE) {
*offset = start;
free_extent_map(em);
break;
}
} else {
- if (origin == SEEK_DATA) {
+ if (whence == SEEK_DATA) {
if (em->block_start == EXTENT_MAP_DELALLOC) {
if (start >= inode->i_size) {
free_extent_map(em);
}
}
- *offset = start;
- free_extent_map(em);
- break;
+ if (!test_bit(EXTENT_FLAG_PREALLOC,
+ &em->flags)) {
+ *offset = start;
+ free_extent_map(em);
+ break;
+ }
}
}
return ret;
}
-static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int origin)
+static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence)
{
struct inode *inode = file->f_mapping->host;
int ret;
mutex_lock(&inode->i_mutex);
- switch (origin) {
+ switch (whence) {
case SEEK_END:
case SEEK_CUR:
- offset = generic_file_llseek(file, offset, origin);
+ offset = generic_file_llseek(file, offset, whence);
goto out;
case SEEK_DATA:
case SEEK_HOLE:
return -ENXIO;
}
- ret = find_desired_extent(inode, &offset, origin);
+ ret = find_desired_extent(inode, &offset, whence);
if (ret) {
mutex_unlock(&inode->i_mutex);
return ret;