#include <linux/slab.h>
#include <linux/ratelimit.h>
#include <linux/mount.h>
+#include <linux/btrfs.h>
#include "compat.h"
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
-#include "ioctl.h"
#include "print-tree.h"
#include "ordered-data.h"
#include "xattr.h"
static struct extent_io_ops btrfs_extent_io_ops;
static struct kmem_cache *btrfs_inode_cachep;
+static struct kmem_cache *btrfs_delalloc_work_cachep;
struct kmem_cache *btrfs_trans_handle_cachep;
struct kmem_cache *btrfs_transaction_cachep;
struct kmem_cache *btrfs_path_cachep;
[S_IFLNK >> S_SHIFT] = BTRFS_FT_SYMLINK,
};
-static int btrfs_setsize(struct inode *inode, loff_t newsize);
+static int btrfs_setsize(struct inode *inode, struct iattr *attr);
static int btrfs_truncate(struct inode *inode);
static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent);
static noinline int cow_file_range(struct inode *inode,
struct page *locked_page,
u64 start, u64 end, int *page_started,
unsigned long *nr_written, int unlock);
+static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
+ u64 len, u64 orig_start,
+ u64 block_start, u64 block_len,
+ u64 orig_block_len, int type);
static int btrfs_init_inode_security(struct btrfs_trans_handle *trans,
struct inode *inode, struct inode *dir,
em->start = async_extent->start;
em->len = async_extent->ram_size;
em->orig_start = em->start;
+ em->mod_start = em->start;
+ em->mod_len = em->len;
em->block_start = ins.objectid;
em->block_len = ins.offset;
+ em->orig_block_len = ins.offset;
em->bdev = root->fs_info->fs_devices->latest_bdev;
em->compress_type = async_extent->compress_type;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
set_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
+ em->generation = -1;
while (1) {
write_lock(&em_tree->lock);
ret = add_extent_mapping(em_tree, em);
+ if (!ret)
+ list_move(&em->list,
+ &em_tree->modified_extents);
write_unlock(&em_tree->lock);
if (ret != -EEXIST) {
free_extent_map(em);
* required to start IO on it. It may be clean and already done with
* IO when we return.
*/
-static noinline int cow_file_range(struct inode *inode,
- struct page *locked_page,
- u64 start, u64 end, int *page_started,
- unsigned long *nr_written,
- int unlock)
+static noinline int __cow_file_range(struct btrfs_trans_handle *trans,
+ struct inode *inode,
+ struct btrfs_root *root,
+ struct page *locked_page,
+ u64 start, u64 end, int *page_started,
+ unsigned long *nr_written,
+ int unlock)
{
- struct btrfs_root *root = BTRFS_I(inode)->root;
- struct btrfs_trans_handle *trans;
u64 alloc_hint = 0;
u64 num_bytes;
unsigned long ram_size;
int ret = 0;
BUG_ON(btrfs_is_free_space_inode(inode));
- trans = btrfs_join_transaction(root);
- if (IS_ERR(trans)) {
- extent_clear_unlock_delalloc(inode,
- &BTRFS_I(inode)->io_tree,
- start, end, locked_page,
- EXTENT_CLEAR_UNLOCK_PAGE |
- EXTENT_CLEAR_UNLOCK |
- EXTENT_CLEAR_DELALLOC |
- EXTENT_CLEAR_DIRTY |
- EXTENT_SET_WRITEBACK |
- EXTENT_END_WRITEBACK);
- return PTR_ERR(trans);
- }
- trans->block_rsv = &root->fs_info->delalloc_block_rsv;
num_bytes = (end - start + blocksize) & ~(blocksize - 1);
num_bytes = max(blocksize, num_bytes);
disk_num_bytes = num_bytes;
- ret = 0;
/* if this is a small write inside eof, kick off defrag */
if (num_bytes < 64 * 1024 &&
em->orig_start = em->start;
ram_size = ins.offset;
em->len = ins.offset;
+ em->mod_start = em->start;
+ em->mod_len = em->len;
em->block_start = ins.objectid;
em->block_len = ins.offset;
+ em->orig_block_len = ins.offset;
em->bdev = root->fs_info->fs_devices->latest_bdev;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
+ em->generation = -1;
while (1) {
write_lock(&em_tree->lock);
ret = add_extent_mapping(em_tree, em);
+ if (!ret)
+ list_move(&em->list,
+ &em_tree->modified_extents);
write_unlock(&em_tree->lock);
if (ret != -EEXIST) {
free_extent_map(em);
alloc_hint = ins.objectid + ins.offset;
start += cur_alloc_size;
}
- ret = 0;
out:
- btrfs_end_transaction(trans, root);
-
return ret;
+
out_unlock:
extent_clear_unlock_delalloc(inode,
&BTRFS_I(inode)->io_tree,
goto out;
}
+static noinline int cow_file_range(struct inode *inode,
+ struct page *locked_page,
+ u64 start, u64 end, int *page_started,
+ unsigned long *nr_written,
+ int unlock)
+{
+ struct btrfs_trans_handle *trans;
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ int ret;
+
+ trans = btrfs_join_transaction(root);
+ if (IS_ERR(trans)) {
+ extent_clear_unlock_delalloc(inode,
+ &BTRFS_I(inode)->io_tree,
+ start, end, locked_page,
+ EXTENT_CLEAR_UNLOCK_PAGE |
+ EXTENT_CLEAR_UNLOCK |
+ EXTENT_CLEAR_DELALLOC |
+ EXTENT_CLEAR_DIRTY |
+ EXTENT_SET_WRITEBACK |
+ EXTENT_END_WRITEBACK);
+ return PTR_ERR(trans);
+ }
+ trans->block_rsv = &root->fs_info->delalloc_block_rsv;
+
+ ret = __cow_file_range(trans, inode, root, locked_page, start, end,
+ page_started, nr_written, unlock);
+
+ btrfs_end_transaction(trans, root);
+
+ return ret;
+}
+
/*
* work queue call back to started compression on a file and pages
*/
u64 extent_offset;
u64 disk_bytenr;
u64 num_bytes;
+ u64 disk_num_bytes;
int extent_type;
int ret, err;
int type;
extent_offset = btrfs_file_extent_offset(leaf, fi);
extent_end = found_key.offset +
btrfs_file_extent_num_bytes(leaf, fi);
+ disk_num_bytes =
+ btrfs_file_extent_disk_num_bytes(leaf, fi);
if (extent_end <= start) {
path->slots[0]++;
goto next_slot;
btrfs_release_path(path);
if (cow_start != (u64)-1) {
- ret = cow_file_range(inode, locked_page, cow_start,
- found_key.offset - 1, page_started,
- nr_written, 1);
+ ret = __cow_file_range(trans, inode, root, locked_page,
+ cow_start, found_key.offset - 1,
+ page_started, nr_written, 1);
if (ret) {
btrfs_abort_transaction(trans, root, ret);
goto error;
em = alloc_extent_map();
BUG_ON(!em); /* -ENOMEM */
em->start = cur_offset;
- em->orig_start = em->start;
+ em->orig_start = found_key.offset - extent_offset;
em->len = num_bytes;
em->block_len = num_bytes;
em->block_start = disk_bytenr;
+ em->orig_block_len = disk_num_bytes;
em->bdev = root->fs_info->fs_devices->latest_bdev;
+ em->mod_start = em->start;
+ em->mod_len = em->len;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
- set_bit(EXTENT_FLAG_PREALLOC, &em->flags);
+ set_bit(EXTENT_FLAG_FILLING, &em->flags);
+ em->generation = -1;
while (1) {
write_lock(&em_tree->lock);
ret = add_extent_mapping(em_tree, em);
+ if (!ret)
+ list_move(&em->list,
+ &em_tree->modified_extents);
write_unlock(&em_tree->lock);
if (ret != -EEXIST) {
free_extent_map(em);
}
if (cow_start != (u64)-1) {
- ret = cow_file_range(inode, locked_page, cow_start, end,
- page_started, nr_written, 1);
+ ret = __cow_file_range(trans, inode, root, locked_page,
+ cow_start, end,
+ page_started, nr_written, 1);
if (ret) {
btrfs_abort_transaction(trans, root, ret);
goto error;
spin_lock(&root->fs_info->delalloc_lock);
BTRFS_I(inode)->delalloc_bytes += len;
- root->fs_info->delalloc_bytes += len;
+ __percpu_counter_add(&root->fs_info->delalloc_bytes, len,
+ root->fs_info->delalloc_batch);
if (do_list && list_empty(&BTRFS_I(inode)->delalloc_inodes)) {
list_add_tail(&BTRFS_I(inode)->delalloc_inodes,
&root->fs_info->delalloc_inodes);
btrfs_free_reserved_data_space(inode, len);
spin_lock(&root->fs_info->delalloc_lock);
- root->fs_info->delalloc_bytes -= len;
+ __percpu_counter_add(&root->fs_info->delalloc_bytes, -len,
+ root->fs_info->delalloc_batch);
BTRFS_I(inode)->delalloc_bytes -= len;
if (do_list && BTRFS_I(inode)->delalloc_bytes == 0 &&
unsigned long bio_flags)
{
struct btrfs_root *root = BTRFS_I(page->mapping->host)->root;
- struct btrfs_mapping_tree *map_tree;
u64 logical = (u64)bio->bi_sector << 9;
u64 length = 0;
u64 map_length;
return 0;
length = bio->bi_size;
- map_tree = &root->fs_info->mapping_tree;
map_length = length;
- ret = btrfs_map_block(map_tree, READ, logical,
+ ret = btrfs_map_block(root->fs_info, READ, logical,
&map_length, NULL, 0);
- /* Will always return 0 or 1 with map_multi == NULL */
+ /* Will always return 0 with map_multi == NULL */
BUG_ON(ret < 0);
if (map_length < length + size)
return 1;
u64 bio_offset)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
- return btrfs_map_bio(root, rw, bio, mirror_num, 1);
+ int ret;
+
+ ret = btrfs_map_bio(root, rw, bio, mirror_num, 1);
+ if (ret)
+ bio_endio(bio, ret);
+ return ret;
}
/*
int ret = 0;
int skip_sum;
int metadata = 0;
+ int async = !atomic_read(&BTRFS_I(inode)->sync_writers);
skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
if (!(rw & REQ_WRITE)) {
ret = btrfs_bio_wq_end_io(root->fs_info, bio, metadata);
if (ret)
- return ret;
+ goto out;
if (bio_flags & EXTENT_BIO_COMPRESSED) {
- return btrfs_submit_compressed_read(inode, bio,
- mirror_num, bio_flags);
+ ret = btrfs_submit_compressed_read(inode, bio,
+ mirror_num,
+ bio_flags);
+ goto out;
} else if (!skip_sum) {
ret = btrfs_lookup_bio_sums(root, inode, bio, NULL);
if (ret)
- return ret;
+ goto out;
}
goto mapit;
- } else if (!skip_sum) {
+ } else if (async && !skip_sum) {
/* csum items have already been cloned */
if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID)
goto mapit;
/* we're doing a write, do the async checksumming */
- return btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
+ ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
inode, rw, bio, mirror_num,
bio_flags, bio_offset,
__btrfs_submit_bio_start,
__btrfs_submit_bio_done);
+ goto out;
+ } else if (!skip_sum) {
+ ret = btrfs_csum_one_bio(root, inode, bio, 0, 0);
+ if (ret)
+ goto out;
}
mapit:
- return btrfs_map_bio(root, rw, bio, mirror_num, 0);
+ ret = btrfs_map_bio(root, rw, bio, mirror_num, 0);
+
+out:
+ if (ret < 0)
+ bio_endio(bio, ret);
+ return ret;
}
/*
int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
struct extent_state **cached_state)
{
- if ((end & (PAGE_CACHE_SIZE - 1)) == 0)
- WARN_ON(1);
+ WARN_ON((end & (PAGE_CACHE_SIZE - 1)) == 0);
return set_extent_delalloc(&BTRFS_I(inode)->io_tree, start, end,
cached_state, GFP_NOFS);
}
if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) {
BUG_ON(!list_empty(&ordered_extent->list)); /* Logic error */
- ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent);
- if (!ret) {
- if (nolock)
- trans = btrfs_join_transaction_nolock(root);
- else
- trans = btrfs_join_transaction(root);
- if (IS_ERR(trans)) {
- ret = PTR_ERR(trans);
- trans = NULL;
- goto out;
- }
- trans->block_rsv = &root->fs_info->delalloc_block_rsv;
- ret = btrfs_update_inode_fallback(trans, root, inode);
- if (ret) /* -ENOMEM or corruption */
- btrfs_abort_transaction(trans, root, ret);
+ btrfs_ordered_update_i_size(inode, 0, ordered_extent);
+ if (nolock)
+ trans = btrfs_join_transaction_nolock(root);
+ else
+ trans = btrfs_join_transaction(root);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ trans = NULL;
+ goto out;
}
+ trans->block_rsv = &root->fs_info->delalloc_block_rsv;
+ ret = btrfs_update_inode_fallback(trans, root, inode);
+ if (ret) /* -ENOMEM or corruption */
+ btrfs_abort_transaction(trans, root, ret);
goto out;
}
add_pending_csums(trans, inode, ordered_extent->file_offset,
&ordered_extent->list);
- ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent);
- if (!ret || !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
- ret = btrfs_update_inode_fallback(trans, root, inode);
- if (ret) { /* -ENOMEM or corruption */
- btrfs_abort_transaction(trans, root, ret);
- goto out_unlock;
- }
- } else {
- btrfs_set_inode_last_trans(trans, inode);
+ btrfs_ordered_update_i_size(inode, 0, ordered_extent);
+ ret = btrfs_update_inode_fallback(trans, root, inode);
+ if (ret) { /* -ENOMEM or corruption */
+ btrfs_abort_transaction(trans, root, ret);
+ goto out_unlock;
}
ret = 0;
out_unlock:
static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
struct extent_state *state, int mirror)
{
- size_t offset = start - ((u64)page->index << PAGE_CACHE_SHIFT);
+ size_t offset = start - page_offset(page);
struct inode *inode = page->mapping->host;
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
char *kaddr;
continue;
}
nr_truncate++;
+
+ /* 1 for the orphan item deletion. */
+ trans = btrfs_start_transaction(root, 1);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ goto out;
+ }
+ ret = btrfs_orphan_add(trans, inode);
+ btrfs_end_transaction(trans, root);
+ if (ret)
+ goto out;
+
ret = btrfs_truncate(inode);
} else {
nr_unlink++;
struct btrfs_inode_item *item,
struct inode *inode)
{
- btrfs_set_inode_uid(leaf, item, i_uid_read(inode));
- btrfs_set_inode_gid(leaf, item, i_gid_read(inode));
- btrfs_set_inode_size(leaf, item, BTRFS_I(inode)->disk_i_size);
- btrfs_set_inode_mode(leaf, item, inode->i_mode);
- btrfs_set_inode_nlink(leaf, item, inode->i_nlink);
+ struct btrfs_map_token token;
+
+ btrfs_init_map_token(&token);
+
+ btrfs_set_token_inode_uid(leaf, item, i_uid_read(inode), &token);
+ btrfs_set_token_inode_gid(leaf, item, i_gid_read(inode), &token);
+ btrfs_set_token_inode_size(leaf, item, BTRFS_I(inode)->disk_i_size,
+ &token);
+ btrfs_set_token_inode_mode(leaf, item, inode->i_mode, &token);
+ btrfs_set_token_inode_nlink(leaf, item, inode->i_nlink, &token);
- btrfs_set_timespec_sec(leaf, btrfs_inode_atime(item),
- inode->i_atime.tv_sec);
- btrfs_set_timespec_nsec(leaf, btrfs_inode_atime(item),
- inode->i_atime.tv_nsec);
+ btrfs_set_token_timespec_sec(leaf, btrfs_inode_atime(item),
+ inode->i_atime.tv_sec, &token);
+ btrfs_set_token_timespec_nsec(leaf, btrfs_inode_atime(item),
+ inode->i_atime.tv_nsec, &token);
- btrfs_set_timespec_sec(leaf, btrfs_inode_mtime(item),
- inode->i_mtime.tv_sec);
- btrfs_set_timespec_nsec(leaf, btrfs_inode_mtime(item),
- inode->i_mtime.tv_nsec);
+ btrfs_set_token_timespec_sec(leaf, btrfs_inode_mtime(item),
+ inode->i_mtime.tv_sec, &token);
+ btrfs_set_token_timespec_nsec(leaf, btrfs_inode_mtime(item),
+ inode->i_mtime.tv_nsec, &token);
- btrfs_set_timespec_sec(leaf, btrfs_inode_ctime(item),
- inode->i_ctime.tv_sec);
- btrfs_set_timespec_nsec(leaf, btrfs_inode_ctime(item),
- inode->i_ctime.tv_nsec);
+ btrfs_set_token_timespec_sec(leaf, btrfs_inode_ctime(item),
+ inode->i_ctime.tv_sec, &token);
+ btrfs_set_token_timespec_nsec(leaf, btrfs_inode_ctime(item),
+ inode->i_ctime.tv_nsec, &token);
- btrfs_set_inode_nbytes(leaf, item, inode_get_bytes(inode));
- btrfs_set_inode_generation(leaf, item, BTRFS_I(inode)->generation);
- btrfs_set_inode_sequence(leaf, item, inode->i_version);
- btrfs_set_inode_transid(leaf, item, trans->transid);
- btrfs_set_inode_rdev(leaf, item, inode->i_rdev);
- btrfs_set_inode_flags(leaf, item, BTRFS_I(inode)->flags);
- btrfs_set_inode_block_group(leaf, item, 0);
+ btrfs_set_token_inode_nbytes(leaf, item, inode_get_bytes(inode),
+ &token);
+ btrfs_set_token_inode_generation(leaf, item, BTRFS_I(inode)->generation,
+ &token);
+ btrfs_set_token_inode_sequence(leaf, item, inode->i_version, &token);
+ btrfs_set_token_inode_transid(leaf, item, trans->transid, &token);
+ btrfs_set_token_inode_rdev(leaf, item, inode->i_rdev, &token);
+ btrfs_set_token_inode_flags(leaf, item, BTRFS_I(inode)->flags, &token);
+ btrfs_set_token_inode_block_group(leaf, item, 0, &token);
}
/*
struct btrfs_trans_handle *trans;
struct inode *inode = dentry->d_inode;
int ret;
- unsigned long nr = 0;
trans = __unlink_start_trans(dir, dentry);
if (IS_ERR(trans))
}
out:
- nr = trans->blocks_used;
__unlink_end_trans(trans, root);
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
return ret;
}
int err = 0;
struct btrfs_root *root = BTRFS_I(dir)->root;
struct btrfs_trans_handle *trans;
- unsigned long nr = 0;
if (inode->i_size > BTRFS_EMPTY_DIR_SIZE)
return -ENOTEMPTY;
if (!err)
btrfs_i_size_write(inode, 0);
out:
- nr = trans->blocks_used;
__unlink_end_trans(trans, root);
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
return err;
}
if (ret)
goto out;
- ret = -ENOMEM;
again:
page = find_or_create_page(mapping, index, mask);
if (!page) {
btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE);
+ ret = -ENOMEM;
goto out;
}
goto out_unlock;
}
- ret = 0;
if (offset != PAGE_CACHE_SIZE) {
if (!len)
len = PAGE_CACHE_SIZE - offset;
block_end - cur_offset, 0);
if (IS_ERR(em)) {
err = PTR_ERR(em);
+ em = NULL;
break;
}
last_byte = min(extent_map_end(em), block_end);
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;
return err;
}
-static int btrfs_setsize(struct inode *inode, loff_t newsize)
+static int btrfs_setsize(struct inode *inode, struct iattr *attr)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
loff_t oldsize = i_size_read(inode);
+ loff_t newsize = attr->ia_size;
+ int mask = attr->ia_valid;
int ret;
if (newsize == oldsize)
return 0;
+ /*
+ * The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a
+ * special case where we need to update the times despite not having
+ * these flags set. For all other operations the VFS set these flags
+ * explicitly if it wants a timestamp update.
+ */
+ if (newsize != oldsize && (!(mask & (ATTR_CTIME | ATTR_MTIME))))
+ inode->i_ctime = inode->i_mtime = current_fs_time(inode->i_sb);
+
if (newsize > oldsize) {
truncate_pagecache(inode, oldsize, newsize);
ret = btrfs_cont_expand(inode, oldsize, newsize);
set_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
&BTRFS_I(inode)->runtime_flags);
+ /*
+ * 1 for the orphan item we're going to add
+ * 1 for the orphan item deletion.
+ */
+ trans = btrfs_start_transaction(root, 2);
+ if (IS_ERR(trans))
+ return PTR_ERR(trans);
+
+ /*
+ * We need to do this in case we fail at _any_ point during the
+ * actual truncate. Once we do the truncate_setsize we could
+ * invalidate pages which forces any outstanding ordered io to
+ * be instantly completed which will give us extents that need
+ * to be truncated. If we fail to get an orphan inode down we
+ * could have left over extents that were never meant to live,
+ * so we need to garuntee from this point on that everything
+ * will be consistent.
+ */
+ ret = btrfs_orphan_add(trans, inode);
+ btrfs_end_transaction(trans, root);
+ if (ret)
+ return ret;
+
/* we don't support swapfiles, so vmtruncate shouldn't fail */
truncate_setsize(inode, newsize);
ret = btrfs_truncate(inode);
+ if (ret && inode->i_nlink)
+ btrfs_orphan_del(NULL, inode);
}
return ret;
return err;
if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE)) {
- err = btrfs_setsize(inode, attr->ia_size);
+ err = btrfs_setsize(inode, attr);
if (err)
return err;
}
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_block_rsv *rsv, *global_rsv;
u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
- unsigned long nr;
int ret;
trace_btrfs_inode_evict(inode);
goto no_delete;
}
+ ret = btrfs_commit_inode_delayed_inode(inode);
+ if (ret) {
+ btrfs_orphan_del(NULL, inode);
+ goto no_delete;
+ }
+
rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP);
if (!rsv) {
btrfs_orphan_del(NULL, inode);
* inode item when doing the truncate.
*/
while (1) {
- ret = btrfs_block_rsv_refill_noflush(root, rsv, min_size);
+ ret = btrfs_block_rsv_refill(root, rsv, min_size,
+ BTRFS_RESERVE_FLUSH_LIMIT);
/*
* Try and steal from the global reserve since we will
goto no_delete;
}
- trans = btrfs_start_transaction_noflush(root, 1);
+ trans = btrfs_join_transaction(root);
if (IS_ERR(trans)) {
btrfs_orphan_del(NULL, inode);
btrfs_free_block_rsv(root, rsv);
break;
trans->block_rsv = &root->fs_info->trans_block_rsv;
- ret = btrfs_update_inode(trans, root, inode);
- BUG_ON(ret);
-
- nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
trans = NULL;
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
}
btrfs_free_block_rsv(root, rsv);
root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID))
btrfs_return_ino(root, btrfs_ino(inode));
- nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
no_delete:
clear_inode(inode);
return;
if (S_ISREG(mode)) {
if (btrfs_test_opt(root, NODATASUM))
BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM;
- if (btrfs_test_opt(root, NODATACOW) ||
- (BTRFS_I(dir)->flags & BTRFS_INODE_NODATACOW))
+ if (btrfs_test_opt(root, NODATACOW))
BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW;
}
ret = btrfs_insert_dir_item(trans, root, name, name_len,
parent_inode, &key,
btrfs_inode_type(inode), index);
- if (ret == -EEXIST)
+ if (ret == -EEXIST || ret == -EOVERFLOW)
goto fail_dir_item;
else if (ret) {
btrfs_abort_transaction(trans, root, ret);
int err;
int drop_inode = 0;
u64 objectid;
- unsigned long nr = 0;
u64 index = 0;
if (!new_valid_dev(rdev))
d_instantiate(dentry, inode);
}
out_unlock:
- nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
if (drop_inode) {
inode_dec_link_count(inode);
iput(inode);
struct btrfs_trans_handle *trans;
struct btrfs_root *root = BTRFS_I(dir)->root;
struct inode *inode = NULL;
- int drop_inode = 0;
+ int drop_inode_on_err = 0;
int err;
- unsigned long nr = 0;
u64 objectid;
u64 index = 0;
err = PTR_ERR(inode);
goto out_unlock;
}
+ drop_inode_on_err = 1;
err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
- if (err) {
- drop_inode = 1;
+ if (err)
+ goto out_unlock;
+
+ err = btrfs_update_inode(trans, root, inode);
+ if (err)
goto out_unlock;
- }
/*
* If the active LSM wants to access the inode during
err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
if (err)
- drop_inode = 1;
- else {
- inode->i_mapping->a_ops = &btrfs_aops;
- inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
- BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
- d_instantiate(dentry, inode);
- }
+ goto out_unlock;
+
+ inode->i_mapping->a_ops = &btrfs_aops;
+ inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
+ BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
+ d_instantiate(dentry, inode);
+
out_unlock:
- nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
- if (drop_inode) {
+ if (err && drop_inode_on_err) {
inode_dec_link_count(inode);
iput(inode);
}
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
return err;
}
struct btrfs_root *root = BTRFS_I(dir)->root;
struct inode *inode = old_dentry->d_inode;
u64 index;
- unsigned long nr = 0;
int err;
int drop_inode = 0;
inode_inc_iversion(inode);
inode->i_ctime = CURRENT_TIME;
ihold(inode);
+ set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags);
err = btrfs_add_nondir(trans, dir, dentry, inode, 1, index);
btrfs_log_new_name(trans, inode, NULL, parent);
}
- nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
fail:
if (drop_inode) {
inode_dec_link_count(inode);
iput(inode);
}
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
return err;
}
int drop_on_err = 0;
u64 objectid = 0;
u64 index = 0;
- unsigned long nr = 1;
/*
* 2 items for inode and ref
drop_on_err = 0;
out_fail:
- nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
if (drop_on_err)
iput(inode);
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
return err;
}
if (start + len <= found_key.offset)
goto not_found;
em->start = start;
+ em->orig_start = start;
em->len = found_key.offset - start;
goto not_found_em;
}
em->len = extent_end - extent_start;
em->orig_start = extent_start -
btrfs_file_extent_offset(leaf, item);
+ em->orig_block_len = btrfs_file_extent_disk_num_bytes(leaf,
+ item);
bytenr = btrfs_file_extent_disk_bytenr(leaf, item);
if (bytenr == 0) {
em->block_start = EXTENT_MAP_HOLE;
set_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
em->compress_type = compress_type;
em->block_start = bytenr;
- em->block_len = btrfs_file_extent_disk_num_bytes(leaf,
- item);
+ em->block_len = em->orig_block_len;
} else {
bytenr += btrfs_file_extent_offset(leaf, item);
em->block_start = bytenr;
em->start = extent_start + extent_offset;
em->len = (copy_size + root->sectorsize - 1) &
~((u64)root->sectorsize - 1);
- em->orig_start = EXTENT_MAP_INLINE;
+ em->orig_block_len = em->len;
+ em->orig_start = em->start;
if (compress_type) {
set_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
em->compress_type = compress_type;
extent_map_end(em) - 1, NULL, GFP_NOFS);
goto insert;
} else {
- printk(KERN_ERR "btrfs unknown found_type %d\n", found_type);
- WARN_ON(1);
+ WARN(1, KERN_ERR "btrfs unknown found_type %d\n", found_type);
}
not_found:
em->start = start;
+ em->orig_start = start;
em->len = len;
not_found_em:
em->block_start = EXTENT_MAP_HOLE;
return em;
if (em) {
/*
- * if our em maps to a hole, there might
- * actually be delalloc bytes behind it
+ * if our em maps to
+ * - a hole or
+ * - a pre-alloc extent,
+ * there might actually be delalloc bytes behind it.
*/
- if (em->block_start != EXTENT_MAP_HOLE)
+ if (em->block_start != EXTENT_MAP_HOLE &&
+ !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
return em;
else
hole_em = em;
*/
em->block_start = hole_em->block_start;
em->block_len = hole_len;
+ if (test_bit(EXTENT_FLAG_PREALLOC, &hole_em->flags))
+ set_bit(EXTENT_FLAG_PREALLOC, &em->flags);
} else {
em->start = range_start;
em->len = found;
}
static struct extent_map *btrfs_new_extent_direct(struct inode *inode,
- struct extent_map *em,
u64 start, u64 len)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
- struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
+ struct extent_map *em;
struct btrfs_key ins;
u64 alloc_hint;
int ret;
- bool insert = false;
-
- /*
- * Ok if the extent map we looked up is a hole and is for the exact
- * range we want, there is no reason to allocate a new one, however if
- * it is not right then we need to free this one and drop the cache for
- * our range.
- */
- if (em->block_start != EXTENT_MAP_HOLE || em->start != start ||
- em->len != len) {
- free_extent_map(em);
- em = NULL;
- insert = true;
- btrfs_drop_extent_cache(inode, start, start + len - 1, 0);
- }
trans = btrfs_join_transaction(root);
if (IS_ERR(trans))
return ERR_CAST(trans);
- if (start <= BTRFS_I(inode)->disk_i_size && len < 64 * 1024)
- btrfs_add_inode_defrag(trans, inode);
-
trans->block_rsv = &root->fs_info->delalloc_block_rsv;
alloc_hint = get_extent_allocation_hint(inode, start, len);
goto out;
}
- if (!em) {
- em = alloc_extent_map();
- if (!em) {
- em = ERR_PTR(-ENOMEM);
- goto out;
- }
- }
-
- em->start = start;
- em->orig_start = em->start;
- em->len = ins.offset;
-
- em->block_start = ins.objectid;
- em->block_len = ins.offset;
- em->bdev = root->fs_info->fs_devices->latest_bdev;
-
- /*
- * We need to do this because if we're using the original em we searched
- * for, we could have EXTENT_FLAG_VACANCY set, and we don't want that.
- */
- em->flags = 0;
- set_bit(EXTENT_FLAG_PINNED, &em->flags);
-
- while (insert) {
- write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em);
- write_unlock(&em_tree->lock);
- if (ret != -EEXIST)
- break;
- btrfs_drop_extent_cache(inode, start, start + em->len - 1, 0);
- }
+ em = create_pinned_em(inode, start, ins.offset, start, ins.objectid,
+ ins.offset, ins.offset, 0);
+ if (IS_ERR(em))
+ goto out;
ret = btrfs_add_ordered_extent_dio(inode, start, ins.objectid,
ins.offset, ins.offset, 0);
static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
u64 len, u64 orig_start,
u64 block_start, u64 block_len,
- int type)
+ u64 orig_block_len, int type)
{
struct extent_map_tree *em_tree;
struct extent_map *em;
em->start = start;
em->orig_start = orig_start;
+ em->mod_start = start;
+ em->mod_len = len;
em->len = len;
em->block_len = block_len;
em->block_start = block_start;
em->bdev = root->fs_info->fs_devices->latest_bdev;
+ em->orig_block_len = orig_block_len;
+ em->generation = -1;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
if (type == BTRFS_ORDERED_PREALLOC)
- set_bit(EXTENT_FLAG_PREALLOC, &em->flags);
+ set_bit(EXTENT_FLAG_FILLING, &em->flags);
do {
btrfs_drop_extent_cache(inode, em->start,
em->start + em->len - 1, 0);
write_lock(&em_tree->lock);
ret = add_extent_mapping(em_tree, em);
+ if (!ret)
+ list_move(&em->list,
+ &em_tree->modified_extents);
write_unlock(&em_tree->lock);
} while (ret == -EEXIST);
goto must_cow;
if (can_nocow_odirect(trans, inode, start, len) == 1) {
- u64 orig_start = em->start;
+ u64 orig_start = em->orig_start;
+ u64 orig_block_len = em->orig_block_len;
if (type == BTRFS_ORDERED_PREALLOC) {
free_extent_map(em);
em = create_pinned_em(inode, start, len,
orig_start,
- block_start, len, type);
+ block_start, len,
+ orig_block_len, type);
if (IS_ERR(em)) {
btrfs_end_transaction(trans, root);
goto unlock_err;
* it above
*/
len = bh_result->b_size;
- em = btrfs_new_extent_direct(inode, em, start, len);
+ free_extent_map(em);
+ em = btrfs_new_extent_direct(inode, start, len);
if (IS_ERR(em)) {
ret = PTR_ERR(em);
goto unlock_err;
struct btrfs_root *root = BTRFS_I(inode)->root;
int ret;
+ if (async_submit)
+ async_submit = !atomic_read(&BTRFS_I(inode)->sync_writers);
+
bio_get(bio);
if (!write) {
{
struct inode *inode = dip->inode;
struct btrfs_root *root = BTRFS_I(inode)->root;
- struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree;
struct bio *bio;
struct bio *orig_bio = dip->orig_bio;
struct bio_vec *bvec = orig_bio->bi_io_vec;
int async_submit = 0;
map_length = orig_bio->bi_size;
- ret = btrfs_map_block(map_tree, READ, start_sector << 9,
+ ret = btrfs_map_block(root->fs_info, READ, start_sector << 9,
&map_length, NULL, 0);
if (ret) {
bio_put(orig_bio);
bio->bi_end_io = btrfs_end_dio_bio;
map_length = orig_bio->bi_size;
- ret = btrfs_map_block(map_tree, READ, start_sector << 9,
+ ret = btrfs_map_block(root->fs_info, READ,
+ start_sector << 9,
&map_length, NULL, 0);
if (ret) {
bio_put(bio);
btrfs_submit_direct, 0);
}
+#define BTRFS_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC)
+
static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
__u64 start, __u64 len)
{
+ int ret;
+
+ ret = fiemap_check_flags(fieinfo, BTRFS_FIEMAP_FLAGS);
+ if (ret)
+ return ret;
+
return extent_fiemap(inode, fieinfo, start, len, btrfs_get_extent_fiemap);
}
return;
}
lock_extent_bits(tree, page_start, page_end, 0, &cached_state);
- ordered = btrfs_lookup_ordered_extent(inode,
- page_offset(page));
+ ordered = btrfs_lookup_ordered_extent(inode, page_offset(page));
if (ordered) {
/*
* IO on this page will never be started, so we need
int ret;
int err = 0;
struct btrfs_trans_handle *trans;
- unsigned long nr;
u64 mask = root->sectorsize - 1;
u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
/*
* 1 for the truncate slack space
- * 1 for the orphan item we're going to add
- * 1 for the orphan item deletion
* 1 for updating the inode.
*/
- trans = btrfs_start_transaction(root, 4);
+ trans = btrfs_start_transaction(root, 2);
if (IS_ERR(trans)) {
err = PTR_ERR(trans);
goto out;
min_size);
BUG_ON(ret);
- ret = btrfs_orphan_add(trans, inode);
- if (ret) {
- btrfs_end_transaction(trans, root);
- goto out;
- }
-
/*
* setattr is responsible for setting the ordered_data_close flag,
* but that is only tested during the last file release. That
break;
}
- nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
trans = btrfs_start_transaction(root, 2);
if (IS_ERR(trans)) {
ret = btrfs_orphan_del(trans, inode);
if (ret)
err = ret;
- } else if (ret && inode->i_nlink > 0) {
- /*
- * Failed to do the truncate, remove us from the in memory
- * orphan list.
- */
- ret = btrfs_orphan_del(NULL, inode);
}
if (trans) {
if (ret && !err)
err = ret;
- nr = trans->blocks_used;
ret = btrfs_end_transaction(trans, root);
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
}
out:
extent_io_tree_init(&ei->io_failure_tree, &inode->i_data);
ei->io_tree.track_uptodate = 1;
ei->io_failure_tree.track_uptodate = 1;
+ atomic_set(&ei->sync_writers, 0);
mutex_init(&ei->log_mutex);
mutex_init(&ei->delalloc_mutex);
btrfs_ordered_inode_tree_init(&ei->ordered_tree);
kmem_cache_destroy(btrfs_path_cachep);
if (btrfs_free_space_cachep)
kmem_cache_destroy(btrfs_free_space_cachep);
+ if (btrfs_delalloc_work_cachep)
+ kmem_cache_destroy(btrfs_delalloc_work_cachep);
}
int btrfs_init_cachep(void)
if (!btrfs_free_space_cachep)
goto fail;
+ btrfs_delalloc_work_cachep = kmem_cache_create("btrfs_delalloc_work",
+ sizeof(struct btrfs_delalloc_work), 0,
+ SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD,
+ NULL);
+ if (!btrfs_delalloc_work_cachep)
+ goto fail;
+
return 0;
fail:
btrfs_destroy_cachep();
if (S_ISDIR(old_inode->i_mode) && new_inode &&
new_inode->i_size > BTRFS_EMPTY_DIR_SIZE)
return -ENOTEMPTY;
+
+
+ /* check for collisions, even if the name isn't there */
+ ret = btrfs_check_dir_item_collision(root, new_dir->i_ino,
+ new_dentry->d_name.name,
+ new_dentry->d_name.len);
+
+ if (ret) {
+ if (ret == -EEXIST) {
+ /* we shouldn't get
+ * eexist without a new_inode */
+ if (!new_inode) {
+ WARN_ON(1);
+ return ret;
+ }
+ } else {
+ /* maybe -EOVERFLOW */
+ return ret;
+ }
+ }
+ ret = 0;
+
/*
* we're using rename to replace one file with another.
* and the replacement file is large. Start IO on it now so
return ret;
}
+static void btrfs_run_delalloc_work(struct btrfs_work *work)
+{
+ struct btrfs_delalloc_work *delalloc_work;
+
+ delalloc_work = container_of(work, struct btrfs_delalloc_work,
+ work);
+ if (delalloc_work->wait)
+ btrfs_wait_ordered_range(delalloc_work->inode, 0, (u64)-1);
+ else
+ filemap_flush(delalloc_work->inode->i_mapping);
+
+ if (delalloc_work->delay_iput)
+ btrfs_add_delayed_iput(delalloc_work->inode);
+ else
+ iput(delalloc_work->inode);
+ complete(&delalloc_work->completion);
+}
+
+struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode,
+ int wait, int delay_iput)
+{
+ struct btrfs_delalloc_work *work;
+
+ work = kmem_cache_zalloc(btrfs_delalloc_work_cachep, GFP_NOFS);
+ if (!work)
+ return NULL;
+
+ init_completion(&work->completion);
+ INIT_LIST_HEAD(&work->list);
+ work->inode = inode;
+ work->wait = wait;
+ work->delay_iput = delay_iput;
+ work->work.func = btrfs_run_delalloc_work;
+
+ return work;
+}
+
+void btrfs_wait_and_free_delalloc_work(struct btrfs_delalloc_work *work)
+{
+ wait_for_completion(&work->completion);
+ kmem_cache_free(btrfs_delalloc_work_cachep, work);
+}
+
/*
* some fairly slow code that needs optimization. This walks the list
* of all the inodes with pending delalloc and forces them to disk.
*/
int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
{
- struct list_head *head = &root->fs_info->delalloc_inodes;
struct btrfs_inode *binode;
struct inode *inode;
+ struct btrfs_delalloc_work *work, *next;
+ struct list_head works;
+ struct list_head splice;
+ int ret = 0;
if (root->fs_info->sb->s_flags & MS_RDONLY)
return -EROFS;
+ INIT_LIST_HEAD(&works);
+ INIT_LIST_HEAD(&splice);
+
spin_lock(&root->fs_info->delalloc_lock);
- while (!list_empty(head)) {
- binode = list_entry(head->next, struct btrfs_inode,
+ list_splice_init(&root->fs_info->delalloc_inodes, &splice);
+ while (!list_empty(&splice)) {
+ binode = list_entry(splice.next, struct btrfs_inode,
delalloc_inodes);
+
+ list_del_init(&binode->delalloc_inodes);
+
inode = igrab(&binode->vfs_inode);
if (!inode)
- list_del_init(&binode->delalloc_inodes);
+ continue;
+
+ list_add_tail(&binode->delalloc_inodes,
+ &root->fs_info->delalloc_inodes);
spin_unlock(&root->fs_info->delalloc_lock);
- if (inode) {
- filemap_flush(inode->i_mapping);
- if (delay_iput)
- btrfs_add_delayed_iput(inode);
- else
- iput(inode);
+
+ work = btrfs_alloc_delalloc_work(inode, 0, delay_iput);
+ if (unlikely(!work)) {
+ ret = -ENOMEM;
+ goto out;
}
+ list_add_tail(&work->list, &works);
+ btrfs_queue_worker(&root->fs_info->flush_workers,
+ &work->work);
+
cond_resched();
spin_lock(&root->fs_info->delalloc_lock);
}
spin_unlock(&root->fs_info->delalloc_lock);
+ list_for_each_entry_safe(work, next, &works, list) {
+ list_del_init(&work->list);
+ btrfs_wait_and_free_delalloc_work(work);
+ }
+
/* the filemap_flush will queue IO into the worker threads, but
* we have to make sure the IO is actually started and that
* ordered extents get created before we return
}
atomic_dec(&root->fs_info->async_submit_draining);
return 0;
+out:
+ list_for_each_entry_safe(work, next, &works, list) {
+ list_del_init(&work->list);
+ btrfs_wait_and_free_delalloc_work(work);
+ }
+
+ if (!list_empty_careful(&splice)) {
+ spin_lock(&root->fs_info->delalloc_lock);
+ list_splice_tail(&splice, &root->fs_info->delalloc_inodes);
+ spin_unlock(&root->fs_info->delalloc_lock);
+ }
+ return ret;
}
static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
unsigned long ptr;
struct btrfs_file_extent_item *ei;
struct extent_buffer *leaf;
- unsigned long nr = 0;
name_len = strlen(symname) + 1;
if (name_len > BTRFS_MAX_INLINE_DATA_SIZE(root))
out_unlock:
if (!err)
d_instantiate(dentry, inode);
- nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
if (drop_inode) {
inode_dec_link_count(inode);
iput(inode);
}
- btrfs_btree_balance_dirty(root, nr);
+ btrfs_btree_balance_dirty(root);
return err;
}
em->len = ins.offset;
em->block_start = ins.objectid;
em->block_len = ins.offset;
+ em->orig_block_len = ins.offset;
em->bdev = root->fs_info->fs_devices->latest_bdev;
set_bit(EXTENT_FLAG_PREALLOC, &em->flags);
em->generation = trans->transid;