aboutsummaryrefslogtreecommitdiff
path: root/fs/btrfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/btrfs')
-rw-r--r--fs/btrfs/acl.c13
-rw-r--r--fs/btrfs/backref.c58
-rw-r--r--fs/btrfs/btrfs_inode.h73
-rw-r--r--fs/btrfs/check-integrity.c57
-rw-r--r--fs/btrfs/compression.c241
-rw-r--r--fs/btrfs/compression.h78
-rw-r--r--fs/btrfs/ctree.c208
-rw-r--r--fs/btrfs/ctree.h257
-rw-r--r--fs/btrfs/delayed-inode.c105
-rw-r--r--fs/btrfs/delayed-inode.h22
-rw-r--r--fs/btrfs/delayed-ref.c56
-rw-r--r--fs/btrfs/delayed-ref.h16
-rw-r--r--fs/btrfs/dev-replace.c18
-rw-r--r--fs/btrfs/dev-replace.h5
-rw-r--r--fs/btrfs/dir-item.c86
-rw-r--r--fs/btrfs/disk-io.c370
-rw-r--r--fs/btrfs/disk-io.h22
-rw-r--r--fs/btrfs/export.c18
-rw-r--r--fs/btrfs/extent-tree.c934
-rw-r--r--fs/btrfs/extent_io.c729
-rw-r--r--fs/btrfs/extent_io.h135
-rw-r--r--fs/btrfs/extent_map.c10
-rw-r--r--fs/btrfs/extent_map.h3
-rw-r--r--fs/btrfs/file-item.c91
-rw-r--r--fs/btrfs/file.c310
-rw-r--r--fs/btrfs/free-space-cache.c73
-rw-r--r--fs/btrfs/free-space-cache.h7
-rw-r--r--fs/btrfs/free-space-tree.c41
-rw-r--r--fs/btrfs/hash.c5
-rw-r--r--fs/btrfs/inode-map.c8
-rw-r--r--fs/btrfs/inode.c1801
-rw-r--r--fs/btrfs/ioctl.c213
-rw-r--r--fs/btrfs/lzo.c45
-rw-r--r--fs/btrfs/ordered-data.c93
-rw-r--r--fs/btrfs/ordered-data.h17
-rw-r--r--fs/btrfs/print-tree.c7
-rw-r--r--fs/btrfs/props.c11
-rw-r--r--fs/btrfs/qgroup.c465
-rw-r--r--fs/btrfs/qgroup.h96
-rw-r--r--fs/btrfs/raid56.c98
-rw-r--r--fs/btrfs/reada.c38
-rw-r--r--fs/btrfs/relocation.c59
-rw-r--r--fs/btrfs/root-tree.c16
-rw-r--r--fs/btrfs/scrub.c583
-rw-r--r--fs/btrfs/send.c328
-rw-r--r--fs/btrfs/super.c95
-rw-r--r--fs/btrfs/sysfs.c41
-rw-r--r--fs/btrfs/tests/btrfs-tests.c1
-rw-r--r--fs/btrfs/tests/extent-io-tests.c2
-rw-r--r--fs/btrfs/tests/inode-tests.c46
-rw-r--r--fs/btrfs/transaction.c123
-rw-r--r--fs/btrfs/transaction.h6
-rw-r--r--fs/btrfs/tree-log.c408
-rw-r--r--fs/btrfs/tree-log.h14
-rw-r--r--fs/btrfs/ulist.c10
-rw-r--r--fs/btrfs/ulist.h8
-rw-r--r--fs/btrfs/volumes.c982
-rw-r--r--fs/btrfs/volumes.h27
-rw-r--r--fs/btrfs/xattr.c18
-rw-r--r--fs/btrfs/zlib.c29
60 files changed, 5520 insertions, 4209 deletions
diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 247b8dfaf6e5..8d8370ddb6b2 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -78,12 +78,6 @@ static int __btrfs_set_acl(struct btrfs_trans_handle *trans,
switch (type) {
case ACL_TYPE_ACCESS:
name = XATTR_NAME_POSIX_ACL_ACCESS;
- if (acl) {
- ret = posix_acl_update_mode(inode, &inode->i_mode, &acl);
- if (ret)
- return ret;
- }
- ret = 0;
break;
case ACL_TYPE_DEFAULT:
if (!S_ISDIR(inode->i_mode))
@@ -119,6 +113,13 @@ out:
int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
{
+ int ret;
+
+ if (type == ACL_TYPE_ACCESS && acl) {
+ ret = posix_acl_update_mode(inode, &inode->i_mode, &acl);
+ if (ret)
+ return ret;
+ }
return __btrfs_set_acl(NULL, inode, acl, type);
}
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 8299601a3549..f723c11bb763 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -16,7 +16,7 @@
* Boston, MA 021110-1307, USA.
*/
-#include <linux/vmalloc.h>
+#include <linux/mm.h>
#include <linux/rbtree.h>
#include "ctree.h"
#include "disk-io.h"
@@ -26,6 +26,11 @@
#include "delayed-ref.h"
#include "locking.h"
+enum merge_mode {
+ MERGE_IDENTICAL_KEYS = 1,
+ MERGE_IDENTICAL_PARENTS,
+};
+
/* Just an arbitrary number so we can be sure this happened */
#define BACKREF_FOUND_SHARED 6
@@ -533,7 +538,7 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path,
* slot==nritems. In that case, go to the next leaf before we continue.
*/
if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
- if (time_seq == (u64)-1)
+ if (time_seq == SEQ_LAST)
ret = btrfs_next_leaf(root, path);
else
ret = btrfs_next_old_leaf(root, path, time_seq);
@@ -577,7 +582,7 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path,
eie = NULL;
}
next:
- if (time_seq == (u64)-1)
+ if (time_seq == SEQ_LAST)
ret = btrfs_next_item(root, path);
else
ret = btrfs_next_old_item(root, path, time_seq);
@@ -629,7 +634,7 @@ static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info,
if (path->search_commit_root)
root_level = btrfs_header_level(root->commit_root);
- else if (time_seq == (u64)-1)
+ else if (time_seq == SEQ_LAST)
root_level = btrfs_header_level(root->node);
else
root_level = btrfs_old_root_level(root, time_seq);
@@ -640,7 +645,7 @@ static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info,
}
path->lowest_level = level;
- if (time_seq == (u64)-1)
+ if (time_seq == SEQ_LAST)
ret = btrfs_search_slot(NULL, root, &ref->key_for_search, path,
0, 0);
else
@@ -809,14 +814,12 @@ static int __add_missing_keys(struct btrfs_fs_info *fs_info,
/*
* merge backrefs and adjust counts accordingly
*
- * mode = 1: merge identical keys, if key is set
- * FIXME: if we add more keys in __add_prelim_ref, we can merge more here.
- * additionally, we could even add a key range for the blocks we
- * looked into to merge even more (-> replace unresolved refs by those
- * having a parent).
- * mode = 2: merge identical parents
+ * FIXME: For MERGE_IDENTICAL_KEYS, if we add more keys in __add_prelim_ref
+ * then we can merge more here. Additionally, we could even add a key
+ * range for the blocks we looked into to merge even more (-> replace
+ * unresolved refs by those having a parent).
*/
-static void __merge_refs(struct list_head *head, int mode)
+static void __merge_refs(struct list_head *head, enum merge_mode mode)
{
struct __prelim_ref *pos1;
@@ -829,7 +832,7 @@ static void __merge_refs(struct list_head *head, int mode)
if (!ref_for_same_block(ref1, ref2))
continue;
- if (mode == 1) {
+ if (mode == MERGE_IDENTICAL_KEYS) {
if (!ref1->parent && ref2->parent)
swap(ref1, ref2);
} else {
@@ -956,8 +959,7 @@ static int __add_delayed_refs(struct btrfs_delayed_ref_head *head, u64 seq,
/*
* add all inline backrefs for bytenr to the list
*/
-static int __add_inline_refs(struct btrfs_fs_info *fs_info,
- struct btrfs_path *path, u64 bytenr,
+static int __add_inline_refs(struct btrfs_path *path, u64 bytenr,
int *info_level, struct list_head *prefs,
struct ref_root *ref_tree,
u64 *total_refs, u64 inum)
@@ -1197,7 +1199,7 @@ static int __add_keyed_refs(struct btrfs_fs_info *fs_info,
*
* NOTE: This can return values > 0
*
- * If time_seq is set to (u64)-1, it will not search delayed_refs, and behave
+ * If time_seq is set to SEQ_LAST, it will not search delayed_refs, and behave
* much like trans == NULL case, the difference only lies in it will not
* commit root.
* The special case is for qgroup to search roots in commit_transaction().
@@ -1244,7 +1246,7 @@ static int find_parent_nodes(struct btrfs_trans_handle *trans,
path->skip_locking = 1;
}
- if (time_seq == (u64)-1)
+ if (time_seq == SEQ_LAST)
path->skip_locking = 1;
/*
@@ -1274,9 +1276,9 @@ again:
#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
if (trans && likely(trans->type != __TRANS_DUMMY) &&
- time_seq != (u64)-1) {
+ time_seq != SEQ_LAST) {
#else
- if (trans && time_seq != (u64)-1) {
+ if (trans && time_seq != SEQ_LAST) {
#endif
/*
* look if there are updates for this ref queued and lock the
@@ -1284,10 +1286,10 @@ again:
*/
delayed_refs = &trans->transaction->delayed_refs;
spin_lock(&delayed_refs->lock);
- head = btrfs_find_delayed_ref_head(trans, bytenr);
+ head = btrfs_find_delayed_ref_head(delayed_refs, bytenr);
if (head) {
if (!mutex_trylock(&head->mutex)) {
- atomic_inc(&head->node.refs);
+ refcount_inc(&head->node.refs);
spin_unlock(&delayed_refs->lock);
btrfs_release_path(path);
@@ -1354,7 +1356,7 @@ again:
if (key.objectid == bytenr &&
(key.type == BTRFS_EXTENT_ITEM_KEY ||
key.type == BTRFS_METADATA_ITEM_KEY)) {
- ret = __add_inline_refs(fs_info, path, bytenr,
+ ret = __add_inline_refs(path, bytenr,
&info_level, &prefs,
ref_tree, &total_refs,
inum);
@@ -1375,7 +1377,7 @@ again:
if (ret)
goto out;
- __merge_refs(&prefs, 1);
+ __merge_refs(&prefs, MERGE_IDENTICAL_KEYS);
ret = __resolve_indirect_refs(fs_info, path, time_seq, &prefs,
extent_item_pos, total_refs,
@@ -1383,7 +1385,7 @@ again:
if (ret)
goto out;
- __merge_refs(&prefs, 2);
+ __merge_refs(&prefs, MERGE_IDENTICAL_PARENTS);
while (!list_empty(&prefs)) {
ref = list_first_entry(&prefs, struct __prelim_ref, list);
@@ -2303,7 +2305,7 @@ struct btrfs_data_container *init_data_container(u32 total_bytes)
size_t alloc_bytes;
alloc_bytes = max_t(size_t, total_bytes, sizeof(*data));
- data = vmalloc(alloc_bytes);
+ data = kvmalloc(alloc_bytes, GFP_KERNEL);
if (!data)
return ERR_PTR(-ENOMEM);
@@ -2337,9 +2339,9 @@ struct inode_fs_paths *init_ipath(s32 total_bytes, struct btrfs_root *fs_root,
if (IS_ERR(fspath))
return (void *)fspath;
- ifp = kmalloc(sizeof(*ifp), GFP_NOFS);
+ ifp = kmalloc(sizeof(*ifp), GFP_KERNEL);
if (!ifp) {
- vfree(fspath);
+ kvfree(fspath);
return ERR_PTR(-ENOMEM);
}
@@ -2354,6 +2356,6 @@ void free_ipath(struct inode_fs_paths *ipath)
{
if (!ipath)
return;
- vfree(ipath->fspath);
+ kvfree(ipath->fspath);
kfree(ipath);
}
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 1a8fa46ff87e..d87ac27a5f2b 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -125,6 +125,13 @@ struct btrfs_inode {
u64 delalloc_bytes;
/*
+ * Total number of bytes pending delalloc that fall within a file
+ * range that is either a hole or beyond EOF (and no prealloc extent
+ * exists in the range). This is always <= delalloc_bytes.
+ */
+ u64 new_delalloc_bytes;
+
+ /*
* total number of bytes pending defrag, used by stat to check whether
* it needs COW.
*/
@@ -224,47 +231,45 @@ static inline void btrfs_insert_inode_hash(struct inode *inode)
__insert_inode_hash(inode, h);
}
-static inline u64 btrfs_ino(struct inode *inode)
+static inline u64 btrfs_ino(struct btrfs_inode *inode)
{
- u64 ino = BTRFS_I(inode)->location.objectid;
+ u64 ino = inode->location.objectid;
/*
* !ino: btree_inode
* type == BTRFS_ROOT_ITEM_KEY: subvol dir
*/
- if (!ino || BTRFS_I(inode)->location.type == BTRFS_ROOT_ITEM_KEY)
- ino = inode->i_ino;
+ if (!ino || inode->location.type == BTRFS_ROOT_ITEM_KEY)
+ ino = inode->vfs_inode.i_ino;
return ino;
}
-static inline void btrfs_i_size_write(struct inode *inode, u64 size)
+static inline void btrfs_i_size_write(struct btrfs_inode *inode, u64 size)
{
- i_size_write(inode, size);
- BTRFS_I(inode)->disk_i_size = size;
+ i_size_write(&inode->vfs_inode, size);
+ inode->disk_i_size = size;
}
-static inline bool btrfs_is_free_space_inode(struct inode *inode)
+static inline bool btrfs_is_free_space_inode(struct btrfs_inode *inode)
{
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_root *root = inode->root;
if (root == root->fs_info->tree_root &&
btrfs_ino(inode) != BTRFS_BTREE_INODE_OBJECTID)
return true;
- if (BTRFS_I(inode)->location.objectid == BTRFS_FREE_INO_OBJECTID)
+ if (inode->location.objectid == BTRFS_FREE_INO_OBJECTID)
return true;
return false;
}
-static inline int btrfs_inode_in_log(struct inode *inode, u64 generation)
+static inline int btrfs_inode_in_log(struct btrfs_inode *inode, u64 generation)
{
int ret = 0;
- spin_lock(&BTRFS_I(inode)->lock);
- if (BTRFS_I(inode)->logged_trans == generation &&
- BTRFS_I(inode)->last_sub_trans <=
- BTRFS_I(inode)->last_log_commit &&
- BTRFS_I(inode)->last_sub_trans <=
- BTRFS_I(inode)->root->last_log_commit) {
+ spin_lock(&inode->lock);
+ if (inode->logged_trans == generation &&
+ inode->last_sub_trans <= inode->last_log_commit &&
+ inode->last_sub_trans <= inode->root->last_log_commit) {
/*
* After a ranged fsync we might have left some extent maps
* (that fall outside the fsync's range). So return false
@@ -272,10 +277,10 @@ static inline int btrfs_inode_in_log(struct inode *inode, u64 generation)
* will be called and process those extent maps.
*/
smp_mb();
- if (list_empty(&BTRFS_I(inode)->extent_tree.modified_extents))
+ if (list_empty(&inode->extent_tree.modified_extents))
ret = 1;
}
- spin_unlock(&BTRFS_I(inode)->lock);
+ spin_unlock(&inode->lock);
return ret;
}
@@ -305,7 +310,8 @@ struct btrfs_dio_private {
* The original bio may be split to several sub-bios, this is
* done during endio of sub-bios
*/
- int (*subio_endio)(struct inode *, struct btrfs_io_bio *, int);
+ blk_status_t (*subio_endio)(struct inode *, struct btrfs_io_bio *,
+ blk_status_t);
};
/*
@@ -313,17 +319,34 @@ struct btrfs_dio_private {
* to grab i_mutex. It is used to avoid the endless truncate due to
* nonlocked dio read.
*/
-static inline void btrfs_inode_block_unlocked_dio(struct inode *inode)
+static inline void btrfs_inode_block_unlocked_dio(struct btrfs_inode *inode)
{
- set_bit(BTRFS_INODE_READDIO_NEED_LOCK, &BTRFS_I(inode)->runtime_flags);
+ set_bit(BTRFS_INODE_READDIO_NEED_LOCK, &inode->runtime_flags);
smp_mb();
}
-static inline void btrfs_inode_resume_unlocked_dio(struct inode *inode)
+static inline void btrfs_inode_resume_unlocked_dio(struct btrfs_inode *inode)
{
smp_mb__before_atomic();
- clear_bit(BTRFS_INODE_READDIO_NEED_LOCK,
- &BTRFS_I(inode)->runtime_flags);
+ clear_bit(BTRFS_INODE_READDIO_NEED_LOCK, &inode->runtime_flags);
+}
+
+static inline void btrfs_print_data_csum_error(struct btrfs_inode *inode,
+ u64 logical_start, u32 csum, u32 csum_expected, int mirror_num)
+{
+ struct btrfs_root *root = inode->root;
+
+ /* Output minus objectid, which is more meaningful */
+ if (root->objectid >= BTRFS_LAST_FREE_OBJECTID)
+ btrfs_warn_rl(root->fs_info,
+ "csum failed root %lld ino %lld off %llu csum 0x%08x expected csum 0x%08x mirror %d",
+ root->objectid, btrfs_ino(inode),
+ logical_start, csum, csum_expected, mirror_num);
+ else
+ btrfs_warn_rl(root->fs_info,
+ "csum failed root %llu ino %llu off %llu csum 0x%08x expected csum 0x%08x mirror %d",
+ root->objectid, btrfs_ino(inode),
+ logical_start, csum, csum_expected, mirror_num);
}
bool btrfs_page_exists_in_range(struct inode *inode, loff_t start, loff_t end);
diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c
index ab14c2e635ca..11d37c94ce05 100644
--- a/fs/btrfs/check-integrity.c
+++ b/fs/btrfs/check-integrity.c
@@ -94,7 +94,7 @@
#include <linux/mutex.h>
#include <linux/genhd.h>
#include <linux/blkdev.h>
-#include <linux/vmalloc.h>
+#include <linux/mm.h>
#include <linux/string.h>
#include "ctree.h"
#include "disk-io.h"
@@ -1638,12 +1638,7 @@ static int btrfsic_read_block(struct btrfsic_state *state,
struct bio *bio;
unsigned int j;
- bio = btrfs_io_bio_alloc(GFP_NOFS, num_pages - i);
- if (!bio) {
- pr_info("btrfsic: bio_alloc() for %u pages failed!\n",
- num_pages - i);
- return -1;
- }
+ bio = btrfs_io_bio_alloc(num_pages - i);
bio->bi_bdev = block_ctx->dev->bdev;
bio->bi_iter.bi_sector = dev_bytenr >> 9;
bio_set_op_attrs(bio, REQ_OP_READ, 0);
@@ -1668,14 +1663,8 @@ static int btrfsic_read_block(struct btrfsic_state *state,
dev_bytenr += (j - i) * PAGE_SIZE;
i = j;
}
- for (i = 0; i < num_pages; i++) {
+ for (i = 0; i < num_pages; i++)
block_ctx->datav[i] = kmap(block_ctx->pagev[i]);
- if (!block_ctx->datav[i]) {
- pr_info("btrfsic: kmap() failed (dev %s)!\n",
- block_ctx->dev->name);
- return -1;
- }
- }
return block_ctx->len;
}
@@ -2129,7 +2118,7 @@ static void btrfsic_bio_end_io(struct bio *bp)
/* mutex is not held! This is not save if IO is not yet completed
* on umount */
iodone_w_error = 0;
- if (bp->bi_error)
+ if (bp->bi_status)
iodone_w_error = 1;
BUG_ON(NULL == block);
@@ -2143,7 +2132,7 @@ static void btrfsic_bio_end_io(struct bio *bp)
if ((dev_state->state->print_mask &
BTRFSIC_PRINT_MASK_END_IO_BIO_BH))
pr_info("bio_end_io(err=%d) for %c @%llu (%s/%llu/%d)\n",
- bp->bi_error,
+ bp->bi_status,
btrfsic_get_block_type(dev_state->state, block),
block->logical_bytenr, dev_state->name,
block->dev_bytenr, block->mirror_num);
@@ -2822,44 +2811,47 @@ static void __btrfsic_submit_bio(struct bio *bio)
dev_state = btrfsic_dev_state_lookup(bio->bi_bdev);
if (NULL != dev_state &&
(bio_op(bio) == REQ_OP_WRITE) && bio_has_data(bio)) {
- unsigned int i;
+ unsigned int i = 0;
u64 dev_bytenr;
u64 cur_bytenr;
- struct bio_vec *bvec;
+ struct bio_vec bvec;
+ struct bvec_iter iter;
int bio_is_patched;
char **mapped_datav;
+ unsigned int segs = bio_segments(bio);
dev_bytenr = 512 * bio->bi_iter.bi_sector;
bio_is_patched = 0;
if (dev_state->state->print_mask &
BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
pr_info("submit_bio(rw=%d,0x%x, bi_vcnt=%u, bi_sector=%llu (bytenr %llu), bi_bdev=%p)\n",
- bio_op(bio), bio->bi_opf, bio->bi_vcnt,
+ bio_op(bio), bio->bi_opf, segs,
(unsigned long long)bio->bi_iter.bi_sector,
dev_bytenr, bio->bi_bdev);
- mapped_datav = kmalloc_array(bio->bi_vcnt,
+ mapped_datav = kmalloc_array(segs,
sizeof(*mapped_datav), GFP_NOFS);
if (!mapped_datav)
goto leave;
cur_bytenr = dev_bytenr;
- bio_for_each_segment_all(bvec, bio, i) {
- BUG_ON(bvec->bv_len != PAGE_SIZE);
- mapped_datav[i] = kmap(bvec->bv_page);
+ bio_for_each_segment(bvec, bio, iter) {
+ BUG_ON(bvec.bv_len != PAGE_SIZE);
+ mapped_datav[i] = kmap(bvec.bv_page);
+ i++;
if (dev_state->state->print_mask &
BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH_VERBOSE)
pr_info("#%u: bytenr=%llu, len=%u, offset=%u\n",
- i, cur_bytenr, bvec->bv_len, bvec->bv_offset);
- cur_bytenr += bvec->bv_len;
+ i, cur_bytenr, bvec.bv_len, bvec.bv_offset);
+ cur_bytenr += bvec.bv_len;
}
btrfsic_process_written_block(dev_state, dev_bytenr,
- mapped_datav, bio->bi_vcnt,
+ mapped_datav, segs,
bio, &bio_is_patched,
NULL, bio->bi_opf);
- bio_for_each_segment_all(bvec, bio, i)
- kunmap(bvec->bv_page);
+ bio_for_each_segment(bvec, bio, iter)
+ kunmap(bvec.bv_page);
kfree(mapped_datav);
} else if (NULL != dev_state && (bio->bi_opf & REQ_PREFLUSH)) {
if (dev_state->state->print_mask &
@@ -2923,13 +2915,10 @@ int btrfsic_mount(struct btrfs_fs_info *fs_info,
fs_info->sectorsize, PAGE_SIZE);
return -1;
}
- state = kzalloc(sizeof(*state), GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT);
+ state = kvzalloc(sizeof(*state), GFP_KERNEL);
if (!state) {
- state = vzalloc(sizeof(*state));
- if (!state) {
- pr_info("btrfs check-integrity: vzalloc() failed!\n");
- return -1;
- }
+ pr_info("btrfs check-integrity: allocation failed!\n");
+ return -1;
}
if (!btrfsic_is_initialized) {
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 7f390849343b..d2ef9ac2a630 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -32,6 +32,7 @@
#include <linux/writeback.h>
#include <linux/bit_spinlock.h>
#include <linux/slab.h>
+#include <linux/sched/mm.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
@@ -42,48 +43,7 @@
#include "extent_io.h"
#include "extent_map.h"
-struct compressed_bio {
- /* number of bios pending for this compressed extent */
- atomic_t pending_bios;
-
- /* the pages with the compressed data on them */
- struct page **compressed_pages;
-
- /* inode that owns this data */
- struct inode *inode;
-
- /* starting offset in the inode for our pages */
- u64 start;
-
- /* number of bytes in the inode we're working on */
- unsigned long len;
-
- /* number of bytes on disk */
- unsigned long compressed_len;
-
- /* the compression algorithm for this bio */
- int compress_type;
-
- /* number of compressed pages in the array */
- unsigned long nr_pages;
-
- /* IO errors */
- int errors;
- int mirror_num;
-
- /* for reads, this is the bio we are copying the data into */
- struct bio *orig_bio;
-
- /*
- * the start of a variable length array of checksums only
- * used by reads
- */
- u32 sums;
-};
-
-static int btrfs_decompress_bio(int type, struct page **pages_in,
- u64 disk_start, struct bio *orig_bio,
- size_t srclen);
+static int btrfs_decompress_bio(struct compressed_bio *cb);
static inline int compressed_bio_size(struct btrfs_fs_info *fs_info,
unsigned long disk_size)
@@ -94,13 +54,7 @@ static inline int compressed_bio_size(struct btrfs_fs_info *fs_info,
(DIV_ROUND_UP(disk_size, fs_info->sectorsize)) * csum_size;
}
-static struct bio *compressed_bio_alloc(struct block_device *bdev,
- u64 first_byte, gfp_t gfp_flags)
-{
- return btrfs_bio_alloc(bdev, first_byte >> 9, BIO_MAX_PAGES, gfp_flags);
-}
-
-static int check_compressed_csum(struct inode *inode,
+static int check_compressed_csum(struct btrfs_inode *inode,
struct compressed_bio *cb,
u64 disk_start)
{
@@ -111,7 +65,7 @@ static int check_compressed_csum(struct inode *inode,
u32 csum;
u32 *cb_sum = &cb->sums;
- if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)
+ if (inode->flags & BTRFS_INODE_NODATASUM)
return 0;
for (i = 0; i < cb->nr_pages; i++) {
@@ -124,10 +78,8 @@ static int check_compressed_csum(struct inode *inode,
kunmap_atomic(kaddr);
if (csum != *cb_sum) {
- btrfs_info(BTRFS_I(inode)->root->fs_info,
- "csum failed ino %llu extent %llu csum %u wanted %u mirror %d",
- btrfs_ino(inode), disk_start, csum, *cb_sum,
- cb->mirror_num);
+ btrfs_print_data_csum_error(inode, disk_start, csum,
+ *cb_sum, cb->mirror_num);
ret = -EIO;
goto fail;
}
@@ -157,17 +109,17 @@ static void end_compressed_bio_read(struct bio *bio)
unsigned long index;
int ret;
- if (bio->bi_error)
+ if (bio->bi_status)
cb->errors = 1;
/* if there are more bios still pending for this compressed
* extent, just exit
*/
- if (!atomic_dec_and_test(&cb->pending_bios))
+ if (!refcount_dec_and_test(&cb->pending_bios))
goto out;
inode = cb->inode;
- ret = check_compressed_csum(inode, cb,
+ ret = check_compressed_csum(BTRFS_I(inode), cb,
(u64)bio->bi_iter.bi_sector << 9);
if (ret)
goto csum_failed;
@@ -175,11 +127,8 @@ static void end_compressed_bio_read(struct bio *bio)
/* ok, we're the last bio for this extent, lets start
* the decompression.
*/
- ret = btrfs_decompress_bio(cb->compress_type,
- cb->compressed_pages,
- cb->start,
- cb->orig_bio,
- cb->compressed_len);
+ ret = btrfs_decompress_bio(cb);
+
csum_failed:
if (ret)
cb->errors = 1;
@@ -203,6 +152,7 @@ csum_failed:
* we have verified the checksum already, set page
* checked so the end_io handlers know about it
*/
+ ASSERT(!bio_flagged(bio, BIO_CLONED));
bio_for_each_segment_all(bvec, cb->orig_bio, i)
SetPageChecked(bvec->bv_page);
@@ -270,13 +220,13 @@ static void end_compressed_bio_write(struct bio *bio)
struct page *page;
unsigned long index;
- if (bio->bi_error)
+ if (bio->bi_status)
cb->errors = 1;
/* if there are more bios still pending for this compressed
* extent, just exit
*/
- if (!atomic_dec_and_test(&cb->pending_bios))
+ if (!refcount_dec_and_test(&cb->pending_bios))
goto out;
/* ok, we're the last bio for this extent, step one is to
@@ -289,7 +239,7 @@ static void end_compressed_bio_write(struct bio *bio)
cb->start,
cb->start + cb->len - 1,
NULL,
- bio->bi_error ? 0 : 1);
+ bio->bi_status ? 0 : 1);
cb->compressed_pages[0]->mapping = NULL;
end_compressed_writeback(inode, cb);
@@ -322,7 +272,7 @@ out:
* This also checksums the file bytes and gets things ready for
* the end io hooks.
*/
-int btrfs_submit_compressed_write(struct inode *inode, u64 start,
+blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
unsigned long len, u64 disk_start,
unsigned long compressed_len,
struct page **compressed_pages,
@@ -337,14 +287,14 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
struct page *page;
u64 first_byte = disk_start;
struct block_device *bdev;
- int ret;
+ blk_status_t ret;
int skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
WARN_ON(start & ((u64)PAGE_SIZE - 1));
cb = kmalloc(compressed_bio_size(fs_info, compressed_len), GFP_NOFS);
if (!cb)
- return -ENOMEM;
- atomic_set(&cb->pending_bios, 0);
+ return BLK_STS_RESOURCE;
+ refcount_set(&cb->pending_bios, 0);
cb->errors = 0;
cb->inode = inode;
cb->start = start;
@@ -357,30 +307,26 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
bdev = fs_info->fs_devices->latest_bdev;
- bio = compressed_bio_alloc(bdev, first_byte, GFP_NOFS);
- if (!bio) {
- kfree(cb);
- return -ENOMEM;
- }
+ bio = btrfs_bio_alloc(bdev, first_byte);
bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
bio->bi_private = cb;
bio->bi_end_io = end_compressed_bio_write;
- atomic_inc(&cb->pending_bios);
+ refcount_set(&cb->pending_bios, 1);
/* create and submit bios for the compressed pages */
bytes_left = compressed_len;
for (pg_index = 0; pg_index < cb->nr_pages; pg_index++) {
+ int submit = 0;
+
page = compressed_pages[pg_index];
page->mapping = inode->i_mapping;
if (bio->bi_iter.bi_size)
- ret = io_tree->ops->merge_bio_hook(page, 0,
+ submit = io_tree->ops->merge_bio_hook(page, 0,
PAGE_SIZE,
bio, 0);
- else
- ret = 0;
page->mapping = NULL;
- if (ret || bio_add_page(bio, page, PAGE_SIZE, 0) <
+ if (submit || bio_add_page(bio, page, PAGE_SIZE, 0) <
PAGE_SIZE) {
bio_get(bio);
@@ -390,7 +336,7 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
* we inc the count. Otherwise, the cb might get
* freed before we're done setting it up
*/
- atomic_inc(&cb->pending_bios);
+ refcount_inc(&cb->pending_bios);
ret = btrfs_bio_wq_end_io(fs_info, bio,
BTRFS_WQ_ENDIO_DATA);
BUG_ON(ret); /* -ENOMEM */
@@ -402,14 +348,13 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
ret = btrfs_map_bio(fs_info, bio, 0, 1);
if (ret) {
- bio->bi_error = ret;
+ bio->bi_status = ret;
bio_endio(bio);
}
bio_put(bio);
- bio = compressed_bio_alloc(bdev, first_byte, GFP_NOFS);
- BUG_ON(!bio);
+ bio = btrfs_bio_alloc(bdev, first_byte);
bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
bio->bi_private = cb;
bio->bi_end_io = end_compressed_bio_write;
@@ -436,7 +381,7 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
ret = btrfs_map_bio(fs_info, bio, 0, 1);
if (ret) {
- bio->bi_error = ret;
+ bio->bi_status = ret;
bio_endio(bio);
}
@@ -571,7 +516,7 @@ next:
* After the compressed pages are read, we copy the bytes into the
* bio we were passed and then call the bio end_io calls
*/
-int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
+blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
int mirror_num, unsigned long bio_flags)
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
@@ -588,7 +533,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
u64 em_len;
u64 em_start;
struct extent_map *em;
- int ret = -ENOMEM;
+ blk_status_t ret = BLK_STS_RESOURCE;
int faili = 0;
u32 *sums;
@@ -602,14 +547,14 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
PAGE_SIZE);
read_unlock(&em_tree->lock);
if (!em)
- return -EIO;
+ return BLK_STS_IOERR;
compressed_len = em->block_len;
cb = kmalloc(compressed_bio_size(fs_info, compressed_len), GFP_NOFS);
if (!cb)
goto out;
- atomic_set(&cb->pending_bios, 0);
+ refcount_set(&cb->pending_bios, 0);
cb->errors = 0;
cb->inode = inode;
cb->mirror_num = mirror_num;
@@ -640,7 +585,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
__GFP_HIGHMEM);
if (!cb->compressed_pages[pg_index]) {
faili = pg_index - 1;
- ret = -ENOMEM;
+ ret = BLK_STS_RESOURCE;
goto fail2;
}
}
@@ -652,28 +597,26 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
/* include any pages we added in add_ra-bio_pages */
cb->len = bio->bi_iter.bi_size;
- comp_bio = compressed_bio_alloc(bdev, cur_disk_byte, GFP_NOFS);
- if (!comp_bio)
- goto fail2;
+ comp_bio = btrfs_bio_alloc(bdev, cur_disk_byte);
bio_set_op_attrs (comp_bio, REQ_OP_READ, 0);
comp_bio->bi_private = cb;
comp_bio->bi_end_io = end_compressed_bio_read;
- atomic_inc(&cb->pending_bios);
+ refcount_set(&cb->pending_bios, 1);
for (pg_index = 0; pg_index < nr_pages; pg_index++) {
+ int submit = 0;
+
page = cb->compressed_pages[pg_index];
page->mapping = inode->i_mapping;
page->index = em_start >> PAGE_SHIFT;
if (comp_bio->bi_iter.bi_size)
- ret = tree->ops->merge_bio_hook(page, 0,
+ submit = tree->ops->merge_bio_hook(page, 0,
PAGE_SIZE,
comp_bio, 0);
- else
- ret = 0;
page->mapping = NULL;
- if (ret || bio_add_page(comp_bio, page, PAGE_SIZE, 0) <
+ if (submit || bio_add_page(comp_bio, page, PAGE_SIZE, 0) <
PAGE_SIZE) {
bio_get(comp_bio);
@@ -687,7 +630,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
* we inc the count. Otherwise, the cb might get
* freed before we're done setting it up
*/
- atomic_inc(&cb->pending_bios);
+ refcount_inc(&cb->pending_bios);
if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) {
ret = btrfs_lookup_bio_sums(inode, comp_bio,
@@ -699,15 +642,13 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
ret = btrfs_map_bio(fs_info, comp_bio, mirror_num, 0);
if (ret) {
- comp_bio->bi_error = ret;
+ comp_bio->bi_status = ret;
bio_endio(comp_bio);
}
bio_put(comp_bio);
- comp_bio = compressed_bio_alloc(bdev, cur_disk_byte,
- GFP_NOFS);
- BUG_ON(!comp_bio);
+ comp_bio = btrfs_bio_alloc(bdev, cur_disk_byte);
bio_set_op_attrs(comp_bio, REQ_OP_READ, 0);
comp_bio->bi_private = cb;
comp_bio->bi_end_io = end_compressed_bio_read;
@@ -728,7 +669,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
ret = btrfs_map_bio(fs_info, comp_bio, mirror_num, 0);
if (ret) {
- comp_bio->bi_error = ret;
+ comp_bio->bi_status = ret;
bio_endio(comp_bio);
}
@@ -803,6 +744,7 @@ static struct list_head *find_workspace(int type)
struct list_head *workspace;
int cpus = num_online_cpus();
int idx = type - 1;
+ unsigned nofs_flag;
struct list_head *idle_ws = &btrfs_comp_ws[idx].idle_ws;
spinlock_t *ws_lock = &btrfs_comp_ws[idx].ws_lock;
@@ -832,7 +774,15 @@ again:
atomic_inc(total_ws);
spin_unlock(ws_lock);
+ /*
+ * Allocation helpers call vmalloc that can't use GFP_NOFS, so we have
+ * to turn it off here because we might get called from the restricted
+ * context of btrfs_compress_bio/btrfs_compress_pages
+ */
+ nofs_flag = memalloc_nofs_save();
workspace = btrfs_compress_op[idx]->alloc_workspace();
+ memalloc_nofs_restore(nofs_flag);
+
if (IS_ERR(workspace)) {
atomic_dec(total_ws);
wake_up(ws_wait);
@@ -913,32 +863,28 @@ static void free_workspaces(void)
}
/*
- * given an address space and start/len, compress the bytes.
+ * Given an address space and start and length, compress the bytes into @pages
+ * that are allocated on demand.
*
- * pages are allocated to hold the compressed result and stored
- * in 'pages'
+ * @out_pages is an in/out parameter, holds maximum number of pages to allocate
+ * and returns number of actually allocated pages
*
- * out_pages is used to return the number of pages allocated. There
- * may be pages allocated even if we return an error
- *
- * total_in is used to return the number of bytes actually read. It
- * may be smaller then len if we had to exit early because we
+ * @total_in is used to return the number of bytes actually read. It
+ * may be smaller than the input length if we had to exit early because we
* ran out of room in the pages array or because we cross the
* max_out threshold.
*
- * total_out is used to return the total number of compressed bytes
+ * @total_out is an in/out parameter, must be set to the input length and will
+ * be also used to return the total number of compressed bytes
*
- * max_out tells us the max number of bytes that we're allowed to
+ * @max_out tells us the max number of bytes that we're allowed to
* stuff into pages
*/
int btrfs_compress_pages(int type, struct address_space *mapping,
- u64 start, unsigned long len,
- struct page **pages,
- unsigned long nr_dest_pages,
+ u64 start, struct page **pages,
unsigned long *out_pages,
unsigned long *total_in,
- unsigned long *total_out,
- unsigned long max_out)
+ unsigned long *total_out)
{
struct list_head *workspace;
int ret;
@@ -946,10 +892,9 @@ int btrfs_compress_pages(int type, struct address_space *mapping,
workspace = find_workspace(type);
ret = btrfs_compress_op[type-1]->compress_pages(workspace, mapping,
- start, len, pages,
- nr_dest_pages, out_pages,
- total_in, total_out,
- max_out);
+ start, pages,
+ out_pages,
+ total_in, total_out);
free_workspace(type, workspace);
return ret;
}
@@ -968,19 +913,16 @@ int btrfs_compress_pages(int type, struct address_space *mapping,
* be contiguous. They all correspond to the range of bytes covered by
* the compressed extent.
*/
-static int btrfs_decompress_bio(int type, struct page **pages_in,
- u64 disk_start, struct bio *orig_bio,
- size_t srclen)
+static int btrfs_decompress_bio(struct compressed_bio *cb)
{
struct list_head *workspace;
int ret;
+ int type = cb->compress_type;
workspace = find_workspace(type);
-
- ret = btrfs_compress_op[type-1]->decompress_bio(workspace, pages_in,
- disk_start, orig_bio,
- srclen);
+ ret = btrfs_compress_op[type - 1]->decompress_bio(workspace, cb);
free_workspace(type, workspace);
+
return ret;
}
@@ -1017,13 +959,14 @@ void btrfs_exit_compress(void)
*
* total_out is the last byte of the buffer
*/
-int btrfs_decompress_buf2page(char *buf, unsigned long buf_start,
+int btrfs_decompress_buf2page(const char *buf, unsigned long buf_start,
unsigned long total_out, u64 disk_start,
struct bio *bio)
{
unsigned long buf_offset;
unsigned long current_buf_start;
unsigned long start_byte;
+ unsigned long prev_start_byte;
unsigned long working_bytes = total_out - buf_start;
unsigned long bytes;
char *kaddr;
@@ -1071,26 +1014,34 @@ int btrfs_decompress_buf2page(char *buf, unsigned long buf_start,
if (!bio->bi_iter.bi_size)
return 0;
bvec = bio_iter_iovec(bio, bio->bi_iter);
-
+ prev_start_byte = start_byte;
start_byte = page_offset(bvec.bv_page) - disk_start;
/*
- * make sure our new page is covered by this
- * working buffer
+ * We need to make sure we're only adjusting
+ * our offset into compression working buffer when
+ * we're switching pages. Otherwise we can incorrectly
+ * keep copying when we were actually done.
*/
- if (total_out <= start_byte)
- return 1;
+ if (start_byte != prev_start_byte) {
+ /*
+ * make sure our new page is covered by this
+ * working buffer
+ */
+ if (total_out <= start_byte)
+ return 1;
- /*
- * the next page in the biovec might not be adjacent
- * to the last page, but it might still be found
- * inside this working buffer. bump our offset pointer
- */
- if (total_out > start_byte &&
- current_buf_start < start_byte) {
- buf_offset = start_byte - buf_start;
- working_bytes = total_out - start_byte;
- current_buf_start = buf_start + buf_offset;
+ /*
+ * the next page in the biovec might not be adjacent
+ * to the last page, but it might still be found
+ * inside this working buffer. bump our offset pointer
+ */
+ if (total_out > start_byte &&
+ current_buf_start < start_byte) {
+ buf_offset = start_byte - buf_start;
+ working_bytes = total_out - start_byte;
+ current_buf_start = buf_start + buf_offset;
+ }
}
}
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index 09879579fbc8..87f6d3332163 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -19,29 +19,80 @@
#ifndef __BTRFS_COMPRESSION_
#define __BTRFS_COMPRESSION_
+/*
+ * We want to make sure that amount of RAM required to uncompress an extent is
+ * reasonable, so we limit the total size in ram of a compressed extent to
+ * 128k. This is a crucial number because it also controls how easily we can
+ * spread reads across cpus for decompression.
+ *
+ * We also want to make sure the amount of IO required to do a random read is
+ * reasonably small, so we limit the size of a compressed extent to 128k.
+ */
+
+/* Maximum length of compressed data stored on disk */
+#define BTRFS_MAX_COMPRESSED (SZ_128K)
+/* Maximum size of data before compression */
+#define BTRFS_MAX_UNCOMPRESSED (SZ_128K)
+
+struct compressed_bio {
+ /* number of bios pending for this compressed extent */
+ refcount_t pending_bios;
+
+ /* the pages with the compressed data on them */
+ struct page **compressed_pages;
+
+ /* inode that owns this data */
+ struct inode *inode;
+
+ /* starting offset in the inode for our pages */
+ u64 start;
+
+ /* number of bytes in the inode we're working on */
+ unsigned long len;
+
+ /* number of bytes on disk */
+ unsigned long compressed_len;
+
+ /* the compression algorithm for this bio */
+ int compress_type;
+
+ /* number of compressed pages in the array */
+ unsigned long nr_pages;
+
+ /* IO errors */
+ int errors;
+ int mirror_num;
+
+ /* for reads, this is the bio we are copying the data into */
+ struct bio *orig_bio;
+
+ /*
+ * the start of a variable length array of checksums only
+ * used by reads
+ */
+ u32 sums;
+};
+
void btrfs_init_compress(void);
void btrfs_exit_compress(void);
int btrfs_compress_pages(int type, struct address_space *mapping,
- u64 start, unsigned long len,
- struct page **pages,
- unsigned long nr_dest_pages,
+ u64 start, struct page **pages,
unsigned long *out_pages,
unsigned long *total_in,
- unsigned long *total_out,
- unsigned long max_out);
+ unsigned long *total_out);
int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page,
unsigned long start_byte, size_t srclen, size_t destlen);
-int btrfs_decompress_buf2page(char *buf, unsigned long buf_start,
+int btrfs_decompress_buf2page(const char *buf, unsigned long buf_start,
unsigned long total_out, u64 disk_start,
struct bio *bio);
-int btrfs_submit_compressed_write(struct inode *inode, u64 start,
+blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
unsigned long len, u64 disk_start,
unsigned long compressed_len,
struct page **compressed_pages,
unsigned long nr_pages);
-int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
+blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
int mirror_num, unsigned long bio_flags);
enum btrfs_compression_type {
@@ -59,19 +110,14 @@ struct btrfs_compress_op {
int (*compress_pages)(struct list_head *workspace,
struct address_space *mapping,
- u64 start, unsigned long len,
+ u64 start,
struct page **pages,
- unsigned long nr_dest_pages,
unsigned long *out_pages,
unsigned long *total_in,
- unsigned long *total_out,
- unsigned long max_out);
+ unsigned long *total_out);
int (*decompress_bio)(struct list_head *workspace,
- struct page **pages_in,
- u64 disk_start,
- struct bio *orig_bio,
- size_t srclen);
+ struct compressed_bio *cb);
int (*decompress)(struct list_head *workspace,
unsigned char *data_in,
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index a426dc822d4d..3f4daa9d6e2c 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -19,7 +19,7 @@
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/rbtree.h>
-#include <linux/vmalloc.h>
+#include <linux/mm.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
@@ -28,9 +28,9 @@
static int split_node(struct btrfs_trans_handle *trans, struct btrfs_root
*root, struct btrfs_path *path, int level);
-static int split_leaf(struct btrfs_trans_handle *trans, struct btrfs_root
- *root, struct btrfs_key *ins_key,
- struct btrfs_path *path, int data_size, int extend);
+static int split_leaf(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+ const struct btrfs_key *ins_key, struct btrfs_path *path,
+ int data_size, int extend);
static int push_node_left(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
struct extent_buffer *dst,
@@ -426,7 +426,7 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
tm_root = &fs_info->tree_mod_log;
for (node = rb_first(tm_root); node; node = next) {
next = rb_next(node);
- tm = container_of(node, struct tree_mod_elem, node);
+ tm = rb_entry(node, struct tree_mod_elem, node);
if (tm->seq > min_seq)
continue;
rb_erase(node, tm_root);
@@ -453,14 +453,12 @@ __tree_mod_log_insert(struct btrfs_fs_info *fs_info, struct tree_mod_elem *tm)
struct rb_node *parent = NULL;
struct tree_mod_elem *cur;
- BUG_ON(!tm);
-
tm->seq = btrfs_inc_tree_mod_seq(fs_info);
tm_root = &fs_info->tree_mod_log;
new = &tm_root->rb_node;
while (*new) {
- cur = container_of(*new, struct tree_mod_elem, node);
+ cur = rb_entry(*new, struct tree_mod_elem, node);
parent = *new;
if (cur->logical < tm->logical)
new = &((*new)->rb_left);
@@ -569,7 +567,7 @@ tree_mod_log_insert_key(struct btrfs_fs_info *fs_info,
static noinline int
tree_mod_log_insert_move(struct btrfs_fs_info *fs_info,
struct extent_buffer *eb, int dst_slot, int src_slot,
- int nr_items, gfp_t flags)
+ int nr_items)
{
struct tree_mod_elem *tm = NULL;
struct tree_mod_elem **tm_list = NULL;
@@ -580,11 +578,11 @@ tree_mod_log_insert_move(struct btrfs_fs_info *fs_info,
if (!tree_mod_need_log(fs_info, eb))
return 0;
- tm_list = kcalloc(nr_items, sizeof(struct tree_mod_elem *), flags);
+ tm_list = kcalloc(nr_items, sizeof(struct tree_mod_elem *), GFP_NOFS);
if (!tm_list)
return -ENOMEM;
- tm = kzalloc(sizeof(*tm), flags);
+ tm = kzalloc(sizeof(*tm), GFP_NOFS);
if (!tm) {
ret = -ENOMEM;
goto free_tms;
@@ -598,7 +596,7 @@ tree_mod_log_insert_move(struct btrfs_fs_info *fs_info,
for (i = 0; i + dst_slot < src_slot && i < nr_items; i++) {
tm_list[i] = alloc_tree_mod_elem(eb, i + dst_slot,
- MOD_LOG_KEY_REMOVE_WHILE_MOVING, flags);
+ MOD_LOG_KEY_REMOVE_WHILE_MOVING, GFP_NOFS);
if (!tm_list[i]) {
ret = -ENOMEM;
goto free_tms;
@@ -665,7 +663,7 @@ __tree_mod_log_free_eb(struct btrfs_fs_info *fs_info,
static noinline int
tree_mod_log_insert_root(struct btrfs_fs_info *fs_info,
struct extent_buffer *old_root,
- struct extent_buffer *new_root, gfp_t flags,
+ struct extent_buffer *new_root,
int log_removal)
{
struct tree_mod_elem *tm = NULL;
@@ -680,14 +678,14 @@ tree_mod_log_insert_root(struct btrfs_fs_info *fs_info,
if (log_removal && btrfs_header_level(old_root) > 0) {
nritems = btrfs_header_nritems(old_root);
tm_list = kcalloc(nritems, sizeof(struct tree_mod_elem *),
- flags);
+ GFP_NOFS);
if (!tm_list) {
ret = -ENOMEM;
goto free_tms;
}
for (i = 0; i < nritems; i++) {
tm_list[i] = alloc_tree_mod_elem(old_root, i,
- MOD_LOG_KEY_REMOVE_WHILE_FREEING, flags);
+ MOD_LOG_KEY_REMOVE_WHILE_FREEING, GFP_NOFS);
if (!tm_list[i]) {
ret = -ENOMEM;
goto free_tms;
@@ -695,7 +693,7 @@ tree_mod_log_insert_root(struct btrfs_fs_info *fs_info,
}
}
- tm = kzalloc(sizeof(*tm), flags);
+ tm = kzalloc(sizeof(*tm), GFP_NOFS);
if (!tm) {
ret = -ENOMEM;
goto free_tms;
@@ -746,7 +744,7 @@ __tree_mod_log_search(struct btrfs_fs_info *fs_info, u64 start, u64 min_seq,
tm_root = &fs_info->tree_mod_log;
node = tm_root->rb_node;
while (node) {
- cur = container_of(node, struct tree_mod_elem, node);
+ cur = rb_entry(node, struct tree_mod_elem, node);
if (cur->logical < start) {
node = node->rb_left;
} else if (cur->logical > start) {
@@ -875,7 +873,7 @@ tree_mod_log_eb_move(struct btrfs_fs_info *fs_info, struct extent_buffer *dst,
{
int ret;
ret = tree_mod_log_insert_move(fs_info, dst, dst_offset, src_offset,
- nr_items, GFP_NOFS);
+ nr_items);
BUG_ON(ret < 0);
}
@@ -945,7 +943,7 @@ tree_mod_log_set_root_pointer(struct btrfs_root *root,
{
int ret;
ret = tree_mod_log_insert_root(root->fs_info, root->node,
- new_root_node, GFP_NOFS, log_removal);
+ new_root_node, log_removal);
BUG_ON(ret < 0);
}
@@ -1074,7 +1072,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
ret = btrfs_dec_ref(trans, root, buf, 1);
BUG_ON(ret); /* -ENOMEM */
}
- clean_tree_block(trans, fs_info, buf);
+ clean_tree_block(fs_info, buf);
*last_ref = 1;
}
return 0;
@@ -1326,7 +1324,7 @@ __tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct extent_buffer *eb,
next = rb_next(&tm->node);
if (!next)
break;
- tm = container_of(next, struct tree_mod_elem, node);
+ tm = rb_entry(next, struct tree_mod_elem, node);
if (tm->logical != first_tm->logical)
break;
}
@@ -1580,7 +1578,8 @@ static int close_blocks(u64 blocknr, u64 other, u32 blocksize)
/*
* compare two keys in a memcmp fashion
*/
-static int comp_keys(struct btrfs_disk_key *disk, struct btrfs_key *k2)
+static int comp_keys(const struct btrfs_disk_key *disk,
+ const struct btrfs_key *k2)
{
struct btrfs_key k1;
@@ -1592,7 +1591,7 @@ static int comp_keys(struct btrfs_disk_key *disk, struct btrfs_key *k2)
/*
* same as comp_keys only with two btrfs_key's
*/
-int btrfs_comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2)
+int btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2)
{
if (k1->objectid > k2->objectid)
return 1;
@@ -1732,8 +1731,8 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
* slot may point to max if the key is bigger than all of the keys
*/
static noinline int generic_bin_search(struct extent_buffer *eb,
- unsigned long p,
- int item_size, struct btrfs_key *key,
+ unsigned long p, int item_size,
+ const struct btrfs_key *key,
int max, int *slot)
{
int low = 0;
@@ -1802,7 +1801,7 @@ static noinline int generic_bin_search(struct extent_buffer *eb,
* simple bin_search frontend that does the right thing for
* leaves vs nodes
*/
-static int bin_search(struct extent_buffer *eb, struct btrfs_key *key,
+static int bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
int level, int *slot)
{
if (level == 0)
@@ -1819,7 +1818,7 @@ static int bin_search(struct extent_buffer *eb, struct btrfs_key *key,
slot);
}
-int btrfs_bin_search(struct extent_buffer *eb, struct btrfs_key *key,
+int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
int level, int *slot)
{
return bin_search(eb, key, level, slot);
@@ -1937,7 +1936,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
path->locks[level] = 0;
path->nodes[level] = NULL;
- clean_tree_block(trans, fs_info, mid);
+ clean_tree_block(fs_info, mid);
btrfs_tree_unlock(mid);
/* once for the path */
free_extent_buffer(mid);
@@ -1998,7 +1997,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
if (wret < 0 && wret != -ENOSPC)
ret = wret;
if (btrfs_header_nritems(right) == 0) {
- clean_tree_block(trans, fs_info, right);
+ clean_tree_block(fs_info, right);
btrfs_tree_unlock(right);
del_ptr(root, path, level + 1, pslot + 1);
root_sub_used(root, right->len);
@@ -2042,7 +2041,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
BUG_ON(wret == 1);
}
if (btrfs_header_nritems(mid) == 0) {
- clean_tree_block(trans, fs_info, mid);
+ clean_tree_block(fs_info, mid);
btrfs_tree_unlock(mid);
del_ptr(root, path, level + 1, pslot);
root_sub_used(root, mid->len);
@@ -2437,10 +2436,9 @@ noinline void btrfs_unlock_up_safe(struct btrfs_path *path, int level)
* reada. -EAGAIN is returned and the search must be repeated.
*/
static int
-read_block_for_search(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct btrfs_path *p,
- struct extent_buffer **eb_ret, int level, int slot,
- struct btrfs_key *key, u64 time_seq)
+read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
+ struct extent_buffer **eb_ret, int level, int slot,
+ const struct btrfs_key *key)
{
struct btrfs_fs_info *fs_info = root->fs_info;
u64 blocknr;
@@ -2587,7 +2585,7 @@ done:
}
static void key_search_validate(struct extent_buffer *b,
- struct btrfs_key *key,
+ const struct btrfs_key *key,
int level)
{
#ifdef CONFIG_BTRFS_ASSERT
@@ -2606,7 +2604,7 @@ static void key_search_validate(struct extent_buffer *b,
#endif
}
-static int key_search(struct extent_buffer *b, struct btrfs_key *key,
+static int key_search(struct extent_buffer *b, const struct btrfs_key *key,
int level, int *prev_cmp, int *slot)
{
if (*prev_cmp != 0) {
@@ -2668,9 +2666,9 @@ int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
* tree. if ins_len < 0, nodes will be merged as we walk down the tree (if
* possible)
*/
-int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
- *root, struct btrfs_key *key, struct btrfs_path *p, int
- ins_len, int cow)
+int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+ const struct btrfs_key *key, struct btrfs_path *p,
+ int ins_len, int cow)
{
struct btrfs_fs_info *fs_info = root->fs_info;
struct extent_buffer *b;
@@ -2870,8 +2868,8 @@ cow_done:
goto done;
}
- err = read_block_for_search(trans, root, p,
- &b, level, slot, key, 0);
+ err = read_block_for_search(root, p, &b, level,
+ slot, key);
if (err == -EAGAIN)
goto again;
if (err) {
@@ -2953,7 +2951,7 @@ done:
* The resulting path and return value will be set up as if we called
* btrfs_search_slot at that point in time with ins_len and cow both set to 0.
*/
-int btrfs_search_old_slot(struct btrfs_root *root, struct btrfs_key *key,
+int btrfs_search_old_slot(struct btrfs_root *root, const struct btrfs_key *key,
struct btrfs_path *p, u64 time_seq)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -3014,8 +3012,8 @@ again:
goto done;
}
- err = read_block_for_search(NULL, root, p, &b, level,
- slot, key, time_seq);
+ err = read_block_for_search(root, p, &b, level,
+ slot, key);
if (err == -EAGAIN)
goto again;
if (err) {
@@ -3067,8 +3065,9 @@ done:
* < 0 on error
*/
int btrfs_search_slot_for_read(struct btrfs_root *root,
- struct btrfs_key *key, struct btrfs_path *p,
- int find_higher, int return_any)
+ const struct btrfs_key *key,
+ struct btrfs_path *p, int find_higher,
+ int return_any)
{
int ret;
struct extent_buffer *leaf;
@@ -3166,7 +3165,7 @@ static void fixup_low_keys(struct btrfs_fs_info *fs_info,
*/
void btrfs_set_item_key_safe(struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
- struct btrfs_key *new_key)
+ const struct btrfs_key *new_key)
{
struct btrfs_disk_key disk_key;
struct extent_buffer *eb;
@@ -3594,8 +3593,7 @@ noinline int btrfs_leaf_free_space(struct btrfs_fs_info *fs_info,
* min slot controls the lowest index we're willing to push to the
* right. We'll push up to and including min_slot, but no lower
*/
-static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info,
+static noinline int __push_leaf_right(struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
int data_size, int empty,
struct extent_buffer *right,
@@ -3669,14 +3667,14 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
/* make room in the right data area */
data_end = leaf_data_end(fs_info, right);
memmove_extent_buffer(right,
- btrfs_leaf_data(right) + data_end - push_space,
- btrfs_leaf_data(right) + data_end,
+ BTRFS_LEAF_DATA_OFFSET + data_end - push_space,
+ BTRFS_LEAF_DATA_OFFSET + data_end,
BTRFS_LEAF_DATA_SIZE(fs_info) - data_end);
/* copy from the left data area */
- copy_extent_buffer(right, left, btrfs_leaf_data(right) +
+ copy_extent_buffer(right, left, BTRFS_LEAF_DATA_OFFSET +
BTRFS_LEAF_DATA_SIZE(fs_info) - push_space,
- btrfs_leaf_data(left) + leaf_data_end(fs_info, left),
+ BTRFS_LEAF_DATA_OFFSET + leaf_data_end(fs_info, left),
push_space);
memmove_extent_buffer(right, btrfs_item_nr_offset(push_items),
@@ -3704,7 +3702,7 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
if (left_nritems)
btrfs_mark_buffer_dirty(left);
else
- clean_tree_block(trans, fs_info, left);
+ clean_tree_block(fs_info, left);
btrfs_mark_buffer_dirty(right);
@@ -3716,7 +3714,7 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
if (path->slots[0] >= left_nritems) {
path->slots[0] -= left_nritems;
if (btrfs_header_nritems(path->nodes[0]) == 0)
- clean_tree_block(trans, fs_info, path->nodes[0]);
+ clean_tree_block(fs_info, path->nodes[0]);
btrfs_tree_unlock(path->nodes[0]);
free_extent_buffer(path->nodes[0]);
path->nodes[0] = right;
@@ -3809,7 +3807,7 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
return 0;
}
- return __push_leaf_right(trans, fs_info, path, min_data_size, empty,
+ return __push_leaf_right(fs_info, path, min_data_size, empty,
right, free_space, left_nritems, min_slot);
out_unlock:
btrfs_tree_unlock(right);
@@ -3825,8 +3823,7 @@ out_unlock:
* item at 'max_slot' won't be touched. Use (u32)-1 to make us do all the
* items
*/
-static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info,
+static noinline int __push_leaf_left(struct btrfs_fs_info *fs_info,
struct btrfs_path *path, int data_size,
int empty, struct extent_buffer *left,
int free_space, u32 right_nritems,
@@ -3891,9 +3888,9 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
push_space = BTRFS_LEAF_DATA_SIZE(fs_info) -
btrfs_item_offset_nr(right, push_items - 1);
- copy_extent_buffer(left, right, btrfs_leaf_data(left) +
+ copy_extent_buffer(left, right, BTRFS_LEAF_DATA_OFFSET +
leaf_data_end(fs_info, left) - push_space,
- btrfs_leaf_data(right) +
+ BTRFS_LEAF_DATA_OFFSET +
btrfs_item_offset_nr(right, push_items - 1),
push_space);
old_left_nritems = btrfs_header_nritems(left);
@@ -3920,9 +3917,9 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
if (push_items < right_nritems) {
push_space = btrfs_item_offset_nr(right, push_items - 1) -
leaf_data_end(fs_info, right);
- memmove_extent_buffer(right, btrfs_leaf_data(right) +
+ memmove_extent_buffer(right, BTRFS_LEAF_DATA_OFFSET +
BTRFS_LEAF_DATA_SIZE(fs_info) - push_space,
- btrfs_leaf_data(right) +
+ BTRFS_LEAF_DATA_OFFSET +
leaf_data_end(fs_info, right), push_space);
memmove_extent_buffer(right, btrfs_item_nr_offset(0),
@@ -3945,7 +3942,7 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
if (right_nritems)
btrfs_mark_buffer_dirty(right);
else
- clean_tree_block(trans, fs_info, right);
+ clean_tree_block(fs_info, right);
btrfs_item_key(right, &disk_key, 0);
fixup_low_keys(fs_info, path, &disk_key, 1);
@@ -4035,7 +4032,7 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
goto out;
}
- return __push_leaf_left(trans, fs_info, path, min_data_size,
+ return __push_leaf_left(fs_info, path, min_data_size,
empty, left, free_space, right_nritems,
max_slot);
out:
@@ -4072,8 +4069,8 @@ static noinline void copy_for_split(struct btrfs_trans_handle *trans,
nritems * sizeof(struct btrfs_item));
copy_extent_buffer(right, l,
- btrfs_leaf_data(right) + BTRFS_LEAF_DATA_SIZE(fs_info) -
- data_copy_size, btrfs_leaf_data(l) +
+ BTRFS_LEAF_DATA_OFFSET + BTRFS_LEAF_DATA_SIZE(fs_info) -
+ data_copy_size, BTRFS_LEAF_DATA_OFFSET +
leaf_data_end(fs_info, l), data_copy_size);
rt_data_off = BTRFS_LEAF_DATA_SIZE(fs_info) - btrfs_item_end_nr(l, mid);
@@ -4160,6 +4157,9 @@ static noinline int push_for_double_split(struct btrfs_trans_handle *trans,
/* try to push all the items before our slot into the next leaf */
slot = path->slots[0];
+ space_needed = data_size;
+ if (slot > 0)
+ space_needed -= btrfs_leaf_free_space(fs_info, path->nodes[0]);
ret = push_leaf_left(trans, root, path, 1, space_needed, 0, slot);
if (ret < 0)
return ret;
@@ -4180,7 +4180,7 @@ static noinline int push_for_double_split(struct btrfs_trans_handle *trans,
*/
static noinline int split_leaf(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct btrfs_key *ins_key,
+ const struct btrfs_key *ins_key,
struct btrfs_path *path, int data_size,
int extend)
{
@@ -4215,6 +4215,10 @@ static noinline int split_leaf(struct btrfs_trans_handle *trans,
if (wret < 0)
return wret;
if (wret) {
+ space_needed = data_size;
+ if (slot > 0)
+ space_needed -= btrfs_leaf_free_space(fs_info,
+ l);
wret = push_leaf_left(trans, root, path, space_needed,
space_needed, 0, (u32)-1);
if (wret < 0)
@@ -4412,10 +4416,9 @@ err:
return ret;
}
-static noinline int split_item(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info,
+static noinline int split_item(struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
- struct btrfs_key *new_key,
+ const struct btrfs_key *new_key,
unsigned long split_offset)
{
struct extent_buffer *leaf;
@@ -4501,7 +4504,7 @@ static noinline int split_item(struct btrfs_trans_handle *trans,
int btrfs_split_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct btrfs_key *new_key,
+ const struct btrfs_key *new_key,
unsigned long split_offset)
{
int ret;
@@ -4510,7 +4513,7 @@ int btrfs_split_item(struct btrfs_trans_handle *trans,
if (ret)
return ret;
- ret = split_item(trans, root->fs_info, path, new_key, split_offset);
+ ret = split_item(root->fs_info, path, new_key, split_offset);
return ret;
}
@@ -4525,7 +4528,7 @@ int btrfs_split_item(struct btrfs_trans_handle *trans,
int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct btrfs_key *new_key)
+ const struct btrfs_key *new_key)
{
struct extent_buffer *leaf;
int ret;
@@ -4604,8 +4607,8 @@ void btrfs_truncate_item(struct btrfs_fs_info *fs_info,
/* shift the data */
if (from_end) {
- memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
- data_end + size_diff, btrfs_leaf_data(leaf) +
+ memmove_extent_buffer(leaf, BTRFS_LEAF_DATA_OFFSET +
+ data_end + size_diff, BTRFS_LEAF_DATA_OFFSET +
data_end, old_data_start + new_size - data_end);
} else {
struct btrfs_disk_key disk_key;
@@ -4631,8 +4634,8 @@ void btrfs_truncate_item(struct btrfs_fs_info *fs_info,
}
}
- memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
- data_end + size_diff, btrfs_leaf_data(leaf) +
+ memmove_extent_buffer(leaf, BTRFS_LEAF_DATA_OFFSET +
+ data_end + size_diff, BTRFS_LEAF_DATA_OFFSET +
data_end, old_data_start - data_end);
offset = btrfs_disk_key_offset(&disk_key);
@@ -4704,8 +4707,8 @@ void btrfs_extend_item(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
}
/* shift the data */
- memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
- data_end - data_size, btrfs_leaf_data(leaf) +
+ memmove_extent_buffer(leaf, BTRFS_LEAF_DATA_OFFSET +
+ data_end - data_size, BTRFS_LEAF_DATA_OFFSET +
data_end, old_data - data_end);
data_end = old_data;
@@ -4726,7 +4729,7 @@ void btrfs_extend_item(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
* that doesn't call btrfs_search_slot
*/
void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
- struct btrfs_key *cpu_key, u32 *data_size,
+ const struct btrfs_key *cpu_key, u32 *data_size,
u32 total_data, u32 total_size, int nr)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -4787,8 +4790,8 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
(nritems - slot) * sizeof(struct btrfs_item));
/* shift the data */
- memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
- data_end - total_data, btrfs_leaf_data(leaf) +
+ memmove_extent_buffer(leaf, BTRFS_LEAF_DATA_OFFSET +
+ data_end - total_data, BTRFS_LEAF_DATA_OFFSET +
data_end, old_data - data_end);
data_end = old_data;
}
@@ -4820,7 +4823,7 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct btrfs_key *cpu_key, u32 *data_size,
+ const struct btrfs_key *cpu_key, u32 *data_size,
int nr)
{
int ret = 0;
@@ -4851,9 +4854,9 @@ int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
* Given a key and some data, insert an item into the tree.
* This does all the path init required, making room in the tree if needed.
*/
-int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root
- *root, struct btrfs_key *cpu_key, void *data, u32
- data_size)
+int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+ const struct btrfs_key *cpu_key, void *data,
+ u32 data_size)
{
int ret = 0;
struct btrfs_path *path;
@@ -4980,9 +4983,9 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
if (slot + nr != nritems) {
int data_end = leaf_data_end(fs_info, leaf);
- memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
+ memmove_extent_buffer(leaf, BTRFS_LEAF_DATA_OFFSET +
data_end + dsize,
- btrfs_leaf_data(leaf) + data_end,
+ BTRFS_LEAF_DATA_OFFSET + data_end,
last_off - data_end);
for (i = slot + nr; i < nritems; i++) {
@@ -5008,7 +5011,7 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
btrfs_set_header_level(leaf, 0);
} else {
btrfs_set_path_blocking(path);
- clean_tree_block(trans, fs_info, leaf);
+ clean_tree_block(fs_info, leaf);
btrfs_del_leaf(trans, root, path, leaf);
}
} else {
@@ -5243,7 +5246,7 @@ out:
static int tree_move_down(struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
- int *level, int root_level)
+ int *level)
{
struct extent_buffer *eb;
@@ -5258,8 +5261,7 @@ static int tree_move_down(struct btrfs_fs_info *fs_info,
return 0;
}
-static int tree_move_next_or_upnext(struct btrfs_fs_info *fs_info,
- struct btrfs_path *path,
+static int tree_move_next_or_upnext(struct btrfs_path *path,
int *level, int root_level)
{
int ret = 0;
@@ -5298,10 +5300,9 @@ static int tree_advance(struct btrfs_fs_info *fs_info,
int ret;
if (*level == 0 || !allow_down) {
- ret = tree_move_next_or_upnext(fs_info, path, level,
- root_level);
+ ret = tree_move_next_or_upnext(path, level, root_level);
} else {
- ret = tree_move_down(fs_info, path, level, root_level);
+ ret = tree_move_down(fs_info, path, level);
}
if (ret >= 0) {
if (*level == 0)
@@ -5391,13 +5392,10 @@ int btrfs_compare_trees(struct btrfs_root *left_root,
goto out;
}
- tmp_buf = kmalloc(fs_info->nodesize, GFP_KERNEL | __GFP_NOWARN);
+ tmp_buf = kvmalloc(fs_info->nodesize, GFP_KERNEL);
if (!tmp_buf) {
- tmp_buf = vmalloc(fs_info->nodesize);
- if (!tmp_buf) {
- ret = -ENOMEM;
- goto out;
- }
+ ret = -ENOMEM;
+ goto out;
}
left_path->search_commit_root = 1;
@@ -5784,8 +5782,8 @@ again:
next = c;
next_rw_lock = path->locks[level];
- ret = read_block_for_search(NULL, root, path, &next, level,
- slot, &key, 0);
+ ret = read_block_for_search(root, path, &next, level,
+ slot, &key);
if (ret == -EAGAIN)
goto again;
@@ -5834,8 +5832,8 @@ again:
if (!level)
break;
- ret = read_block_for_search(NULL, root, path, &next, level,
- 0, &key, 0);
+ ret = read_block_for_search(root, path, &next, level,
+ 0, &key);
if (ret == -EAGAIN)
goto again;
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 6a823719b6c5..3f3eb7b17cac 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -20,6 +20,7 @@
#define __BTRFS_CTREE__
#include <linux/mm.h>
+#include <linux/sched/signal.h>
#include <linux/highmem.h>
#include <linux/fs.h>
#include <linux/rwsem.h>
@@ -38,6 +39,7 @@
#include <linux/security.h>
#include <linux/sizes.h>
#include <linux/dynamic_debug.h>
+#include <linux/refcount.h>
#include "extent_io.h"
#include "extent_map.h"
#include "async-thread.h"
@@ -46,7 +48,6 @@ struct btrfs_trans_handle;
struct btrfs_transaction;
struct btrfs_pending_snapshot;
extern struct kmem_cache *btrfs_trans_handle_cachep;
-extern struct kmem_cache *btrfs_transaction_cachep;
extern struct kmem_cache *btrfs_bit_radix_cachep;
extern struct kmem_cache *btrfs_path_cachep;
extern struct kmem_cache *btrfs_free_space_cachep;
@@ -97,6 +98,14 @@ static const int btrfs_csum_sizes[] = { 4 };
#define BTRFS_MAX_EXTENT_SIZE SZ_128M
+/*
+ * Count how many BTRFS_MAX_EXTENT_SIZE cover the @size
+ */
+static inline u32 count_max_extents(u64 size)
+{
+ return div_u64(size + BTRFS_MAX_EXTENT_SIZE - 1, BTRFS_MAX_EXTENT_SIZE);
+}
+
struct btrfs_mapping_tree {
struct extent_map_tree map_tree;
};
@@ -509,7 +518,7 @@ struct btrfs_caching_control {
struct btrfs_work work;
struct btrfs_block_group_cache *block_group;
u64 progress;
- atomic_t count;
+ refcount_t count;
};
/* Once caching_thread() finds this much free space, it will wake up waiters. */
@@ -529,6 +538,14 @@ struct btrfs_io_ctl {
unsigned check_crcs:1;
};
+/*
+ * Tree to record all locked full stripes of a RAID5/6 block group
+ */
+struct btrfs_full_stripe_locks_tree {
+ struct rb_root root;
+ struct mutex lock;
+};
+
struct btrfs_block_group_cache {
struct btrfs_key key;
struct btrfs_block_group_item item;
@@ -639,6 +656,9 @@ struct btrfs_block_group_cache {
* Protected by free_space_lock.
*/
int needs_free_space;
+
+ /* Record locked full stripes for RAID5/6 block group */
+ struct btrfs_full_stripe_locks_tree full_stripe_locks_root;
};
/* delayed seq elem */
@@ -649,6 +669,8 @@ struct seq_list {
#define SEQ_LIST_INIT(name) { .list = LIST_HEAD_INIT((name).list), .seq = 0 }
+#define SEQ_LAST ((u64)-1)
+
enum btrfs_orphan_cleanup_state {
ORPHAN_CLEANUP_STARTED = 1,
ORPHAN_CLEANUP_DONE = 2,
@@ -693,6 +715,15 @@ struct btrfs_delayed_root;
#define BTRFS_FS_BTREE_ERR 11
#define BTRFS_FS_LOG1_ERR 12
#define BTRFS_FS_LOG2_ERR 13
+#define BTRFS_FS_QUOTA_OVERRIDE 14
+/* Used to record internally whether fs has been frozen */
+#define BTRFS_FS_FROZEN 15
+
+/*
+ * Indicate that a whole-filesystem exclusive operation is running
+ * (device replace, resize, device add/delete, balance)
+ */
+#define BTRFS_FS_EXCL_OP 14
struct btrfs_fs_info {
u8 fsid[BTRFS_FSID_SIZE];
@@ -720,8 +751,7 @@ struct btrfs_fs_info {
struct rb_root block_group_cache_tree;
/* keep track of unallocated space */
- spinlock_t free_chunk_lock;
- u64 free_chunk_space;
+ atomic64_t free_chunk_space;
struct extent_io_tree freed_extents[2];
struct extent_io_tree *pinned_extents;
@@ -769,17 +799,7 @@ struct btrfs_fs_info {
* so it is also safe.
*/
u64 max_inline;
- /*
- * Protected by ->chunk_mutex and sb->s_umount.
- *
- * The reason that we use two lock to protect it is because only
- * remount and mount operations can change it and these two operations
- * are under sb->s_umount, but the read side (chunk allocation) can not
- * acquire sb->s_umount or the deadlock would happen. So we use two
- * locks to protect it. On the write side, we must acquire two locks,
- * and on the read side, we just need acquire one of them.
- */
- u64 alloc_start;
+
struct btrfs_transaction *running_transaction;
wait_queue_head_t transaction_throttle;
wait_queue_head_t transaction_wait;
@@ -801,7 +821,6 @@ struct btrfs_fs_info {
struct btrfs_super_block *super_for_commit;
struct super_block *sb;
struct inode *btree_inode;
- struct backing_dev_info bdi;
struct mutex tree_log_mutex;
struct mutex transaction_kthread_mutex;
struct mutex cleaner_mutex;
@@ -1058,8 +1077,6 @@ struct btrfs_fs_info {
/* device replace state */
struct btrfs_dev_replace dev_replace;
- atomic_t mutually_exclusive_operation_running;
-
struct percpu_counter bio_counter;
wait_queue_head_t replace_wait;
@@ -1082,9 +1099,6 @@ struct btrfs_fs_info {
*/
struct list_head pinned_chunks;
- /* Used to record internally whether fs has been frozen */
- int fs_frozen;
-
/* Cached block sizes */
u32 nodesize;
u32 sectorsize;
@@ -1212,7 +1226,7 @@ struct btrfs_root {
dev_t anon_dev;
spinlock_t root_item_lock;
- atomic_t refs;
+ refcount_t refs;
struct mutex delalloc_mutex;
spinlock_t delalloc_lock;
@@ -1250,23 +1264,22 @@ struct btrfs_root {
atomic_t will_be_snapshoted;
/* For qgroup metadata space reserve */
- atomic_t qgroup_meta_rsv;
+ atomic64_t qgroup_meta_rsv;
};
+
static inline u32 btrfs_inode_sectorsize(const struct inode *inode)
{
return btrfs_sb(inode->i_sb)->sectorsize;
}
-static inline u32 __BTRFS_LEAF_DATA_SIZE(u32 blocksize)
-{
- return blocksize - sizeof(struct btrfs_header);
-}
-
static inline u32 BTRFS_LEAF_DATA_SIZE(const struct btrfs_fs_info *info)
{
- return __BTRFS_LEAF_DATA_SIZE(info->nodesize);
+
+ return info->nodesize - sizeof(struct btrfs_header);
}
+#define BTRFS_LEAF_DATA_OFFSET offsetof(struct btrfs_leaf, items)
+
static inline u32 BTRFS_MAX_ITEM_SIZE(const struct btrfs_fs_info *info)
{
return BTRFS_LEAF_DATA_SIZE(info) - sizeof(struct btrfs_item);
@@ -1528,8 +1541,27 @@ static inline void btrfs_set_##name(type *s, u##bits val) \
s->member = cpu_to_le##bits(val); \
}
+
+static inline u64 btrfs_device_total_bytes(struct extent_buffer *eb,
+ struct btrfs_dev_item *s)
+{
+ BUILD_BUG_ON(sizeof(u64) !=
+ sizeof(((struct btrfs_dev_item *)0))->total_bytes);
+ return btrfs_get_64(eb, s, offsetof(struct btrfs_dev_item,
+ total_bytes));
+}
+static inline void btrfs_set_device_total_bytes(struct extent_buffer *eb,
+ struct btrfs_dev_item *s,
+ u64 val)
+{
+ BUILD_BUG_ON(sizeof(u64) !=
+ sizeof(((struct btrfs_dev_item *)0))->total_bytes);
+ WARN_ON(!IS_ALIGNED(val, eb->fs_info->sectorsize));
+ btrfs_set_64(eb, s, offsetof(struct btrfs_dev_item, total_bytes), val);
+}
+
+
BTRFS_SETGET_FUNCS(device_type, struct btrfs_dev_item, type, 64);
-BTRFS_SETGET_FUNCS(device_total_bytes, struct btrfs_dev_item, total_bytes, 64);
BTRFS_SETGET_FUNCS(device_bytes_used, struct btrfs_dev_item, bytes_used, 64);
BTRFS_SETGET_FUNCS(device_io_align, struct btrfs_dev_item, io_align, 32);
BTRFS_SETGET_FUNCS(device_io_width, struct btrfs_dev_item, io_width, 32);
@@ -1953,7 +1985,7 @@ BTRFS_SETGET_STACK_FUNCS(disk_key_offset, struct btrfs_disk_key, offset, 64);
BTRFS_SETGET_STACK_FUNCS(disk_key_type, struct btrfs_disk_key, type, 8);
static inline void btrfs_disk_key_to_cpu(struct btrfs_key *cpu,
- struct btrfs_disk_key *disk)
+ const struct btrfs_disk_key *disk)
{
cpu->offset = le64_to_cpu(disk->offset);
cpu->type = disk->type;
@@ -1961,7 +1993,7 @@ static inline void btrfs_disk_key_to_cpu(struct btrfs_key *cpu,
}
static inline void btrfs_cpu_key_to_disk(struct btrfs_disk_key *disk,
- struct btrfs_key *cpu)
+ const struct btrfs_key *cpu)
{
disk->offset = cpu_to_le64(cpu->offset);
disk->type = cpu->type;
@@ -1993,8 +2025,7 @@ static inline void btrfs_dir_item_key_to_cpu(struct extent_buffer *eb,
btrfs_disk_key_to_cpu(key, &disk_key);
}
-
-static inline u8 btrfs_key_type(struct btrfs_key *key)
+static inline u8 btrfs_key_type(const struct btrfs_key *key)
{
return key->type;
}
@@ -2300,10 +2331,6 @@ static inline int btrfs_super_csum_size(struct btrfs_super_block *s)
return btrfs_csum_sizes[t];
}
-static inline unsigned long btrfs_leaf_data(struct extent_buffer *l)
-{
- return offsetof(struct btrfs_leaf, items);
-}
/*
* The leaf data grows from end-to-front in the node.
@@ -2514,11 +2541,11 @@ BTRFS_SETGET_STACK_FUNCS(stack_dev_replace_cursor_right,
/* helper function to cast into the data area of the leaf. */
#define btrfs_item_ptr(leaf, slot, type) \
- ((type *)(btrfs_leaf_data(leaf) + \
+ ((type *)(BTRFS_LEAF_DATA_OFFSET + \
btrfs_item_offset_nr(leaf, slot)))
#define btrfs_item_ptr_offset(leaf, slot) \
- ((unsigned long)(btrfs_leaf_data(leaf) + \
+ ((unsigned long)(BTRFS_LEAF_DATA_OFFSET + \
btrfs_item_offset_nr(leaf, slot)))
static inline bool btrfs_mixed_space_info(struct btrfs_space_info *space_info)
@@ -2539,7 +2566,7 @@ u64 btrfs_csum_bytes_to_leaves(struct btrfs_fs_info *fs_info, u64 csum_bytes);
static inline u64 btrfs_calc_trans_metadata_size(struct btrfs_fs_info *fs_info,
unsigned num_items)
{
- return fs_info->nodesize * BTRFS_MAX_LEVEL * 2 * num_items;
+ return (u64)fs_info->nodesize * BTRFS_MAX_LEVEL * 2 * num_items;
}
/*
@@ -2549,7 +2576,7 @@ static inline u64 btrfs_calc_trans_metadata_size(struct btrfs_fs_info *fs_info,
static inline u64 btrfs_calc_trunc_metadata_size(struct btrfs_fs_info *fs_info,
unsigned num_items)
{
- return fs_info->nodesize * BTRFS_MAX_LEVEL * num_items;
+ return (u64)fs_info->nodesize * BTRFS_MAX_LEVEL * num_items;
}
int btrfs_should_throttle_delayed_refs(struct btrfs_trans_handle *trans,
@@ -2577,8 +2604,7 @@ int btrfs_pin_extent_for_log_replay(struct btrfs_fs_info *fs_info,
u64 bytenr, u64 num_bytes);
int btrfs_exclude_logged_extents(struct btrfs_fs_info *fs_info,
struct extent_buffer *eb);
-int btrfs_cross_ref_exist(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+int btrfs_cross_ref_exist(struct btrfs_root *root,
u64 objectid, u64 offset, u64 bytenr);
struct btrfs_block_group_cache *btrfs_lookup_block_group(
struct btrfs_fs_info *info,
@@ -2587,10 +2613,11 @@ void btrfs_get_block_group(struct btrfs_block_group_cache *cache);
void btrfs_put_block_group(struct btrfs_block_group_cache *cache);
int get_block_group_index(struct btrfs_block_group_cache *cache);
struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, u64 parent,
- u64 root_objectid,
- struct btrfs_disk_key *key, int level,
- u64 hint, u64 empty_size);
+ struct btrfs_root *root,
+ u64 parent, u64 root_objectid,
+ const struct btrfs_disk_key *key,
+ int level, u64 hint,
+ u64 empty_size);
void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct extent_buffer *buf,
@@ -2623,8 +2650,7 @@ int btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info,
u64 start, u64 len, int delalloc);
int btrfs_free_and_pin_reserved_extent(struct btrfs_fs_info *fs_info,
u64 start, u64 len);
-void btrfs_prepare_extent_commit(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info);
+void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info);
int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info);
int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
@@ -2657,7 +2683,9 @@ void btrfs_get_block_group_trimming(struct btrfs_block_group_cache *cache);
void btrfs_put_block_group_trimming(struct btrfs_block_group_cache *cache);
void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info);
-u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data);
+u64 btrfs_data_alloc_profile(struct btrfs_fs_info *fs_info);
+u64 btrfs_metadata_alloc_profile(struct btrfs_fs_info *fs_info);
+u64 btrfs_system_alloc_profile(struct btrfs_fs_info *fs_info);
void btrfs_clear_space_info_full(struct btrfs_fs_info *info);
enum btrfs_reserve_flush_enum {
@@ -2680,28 +2708,31 @@ enum btrfs_flush_state {
COMMIT_TRANS = 6,
};
-int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len);
-int btrfs_alloc_data_chunk_ondemand(struct inode *inode, u64 bytes);
-void btrfs_free_reserved_data_space(struct inode *inode, u64 start, u64 len);
+int btrfs_alloc_data_chunk_ondemand(struct btrfs_inode *inode, u64 bytes);
+int btrfs_check_data_free_space(struct inode *inode,
+ struct extent_changeset **reserved, u64 start, u64 len);
+void btrfs_free_reserved_data_space(struct inode *inode,
+ struct extent_changeset *reserved, u64 start, u64 len);
+void btrfs_delalloc_release_space(struct inode *inode,
+ struct extent_changeset *reserved, u64 start, u64 len);
void btrfs_free_reserved_data_space_noquota(struct inode *inode, u64 start,
u64 len);
void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info);
void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans);
int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
- struct inode *inode);
-void btrfs_orphan_release_metadata(struct inode *inode);
+ struct btrfs_inode *inode);
+void btrfs_orphan_release_metadata(struct btrfs_inode *inode);
int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
struct btrfs_block_rsv *rsv,
int nitems,
u64 *qgroup_reserved, bool use_global_rsv);
void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info,
- struct btrfs_block_rsv *rsv,
- u64 qgroup_reserved);
-int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes);
-void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes);
-int btrfs_delalloc_reserve_space(struct inode *inode, u64 start, u64 len);
-void btrfs_delalloc_release_space(struct inode *inode, u64 start, u64 len);
+ struct btrfs_block_rsv *rsv);
+int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes);
+void btrfs_delalloc_release_metadata(struct btrfs_inode *inode, u64 num_bytes);
+int btrfs_delalloc_reserve_space(struct inode *inode,
+ struct extent_changeset **reserved, u64 start, u64 len);
void btrfs_init_block_rsv(struct btrfs_block_rsv *rsv, unsigned short type);
struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_fs_info *fs_info,
unsigned short type);
@@ -2724,7 +2755,7 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
void btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
struct btrfs_block_rsv *block_rsv,
u64 num_bytes);
-int btrfs_inc_block_group_ro(struct btrfs_root *root,
+int btrfs_inc_block_group_ro(struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache *cache);
void btrfs_dec_block_group_ro(struct btrfs_block_group_cache *cache);
void btrfs_put_block_group_cache(struct btrfs_fs_info *info);
@@ -2750,9 +2781,9 @@ u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
struct btrfs_fs_info *info, u64 start, u64 end);
/* ctree.c */
-int btrfs_bin_search(struct extent_buffer *eb, struct btrfs_key *key,
+int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
int level, int *slot);
-int btrfs_comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2);
+int btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2);
int btrfs_previous_item(struct btrfs_root *root,
struct btrfs_path *path, u64 min_objectid,
int type);
@@ -2760,7 +2791,7 @@ int btrfs_previous_extent_item(struct btrfs_root *root,
struct btrfs_path *path, u64 min_objectid);
void btrfs_set_item_key_safe(struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
- struct btrfs_key *new_key);
+ const struct btrfs_key *new_key);
struct extent_buffer *btrfs_root_node(struct btrfs_root *root);
struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root);
int btrfs_find_next_key(struct btrfs_root *root, struct btrfs_path *path,
@@ -2802,22 +2833,23 @@ void btrfs_truncate_item(struct btrfs_fs_info *fs_info,
int btrfs_split_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct btrfs_key *new_key,
+ const struct btrfs_key *new_key,
unsigned long split_offset);
int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct btrfs_key *new_key);
+ const struct btrfs_key *new_key);
int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
u64 inum, u64 ioff, u8 key_type, struct btrfs_key *found_key);
-int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
- *root, struct btrfs_key *key, struct btrfs_path *p, int
- ins_len, int cow);
-int btrfs_search_old_slot(struct btrfs_root *root, struct btrfs_key *key,
+int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+ const struct btrfs_key *key, struct btrfs_path *p,
+ int ins_len, int cow);
+int btrfs_search_old_slot(struct btrfs_root *root, const struct btrfs_key *key,
struct btrfs_path *p, u64 time_seq);
int btrfs_search_slot_for_read(struct btrfs_root *root,
- struct btrfs_key *key, struct btrfs_path *p,
- int find_higher, int return_any);
+ const struct btrfs_key *key,
+ struct btrfs_path *p, int find_higher,
+ int return_any);
int btrfs_realloc_node(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct extent_buffer *parent,
int start_slot, u64 *last_ret,
@@ -2840,19 +2872,20 @@ static inline int btrfs_del_item(struct btrfs_trans_handle *trans,
}
void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
- struct btrfs_key *cpu_key, u32 *data_size,
+ const struct btrfs_key *cpu_key, u32 *data_size,
u32 total_data, u32 total_size, int nr);
-int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root
- *root, struct btrfs_key *key, void *data, u32 data_size);
+int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+ const struct btrfs_key *key, void *data, u32 data_size);
int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct btrfs_key *cpu_key, u32 *data_size, int nr);
+ const struct btrfs_key *cpu_key, u32 *data_size,
+ int nr);
static inline int btrfs_insert_empty_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct btrfs_key *key,
+ const struct btrfs_key *key,
u32 data_size)
{
return btrfs_insert_empty_items(trans, root, path, key, &data_size, 1);
@@ -2941,15 +2974,15 @@ int btrfs_del_root_ref(struct btrfs_trans_handle *trans,
u64 root_id, u64 ref_id, u64 dirid, u64 *sequence,
const char *name, int name_len);
int btrfs_del_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
- struct btrfs_key *key);
-int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root
- *root, struct btrfs_key *key, struct btrfs_root_item
- *item);
+ const struct btrfs_key *key);
+int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+ const struct btrfs_key *key,
+ struct btrfs_root_item *item);
int __must_check btrfs_update_root(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_key *key,
struct btrfs_root_item *item);
-int btrfs_find_root(struct btrfs_root *root, struct btrfs_key *search_key,
+int btrfs_find_root(struct btrfs_root *root, const struct btrfs_key *search_key,
struct btrfs_path *path, struct btrfs_root_item *root_item,
struct btrfs_key *root_key);
int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info);
@@ -2975,7 +3008,7 @@ int btrfs_check_dir_item_collision(struct btrfs_root *root, u64 dir,
const char *name, int name_len);
int btrfs_insert_dir_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root, const char *name,
- int name_len, struct inode *dir,
+ int name_len, struct btrfs_inode *dir,
struct btrfs_key *location, u8 type, u64 index);
struct btrfs_dir_item *btrfs_lookup_dir_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
@@ -3007,12 +3040,14 @@ struct btrfs_dir_item *btrfs_lookup_xattr(struct btrfs_trans_handle *trans,
const char *name, u16 name_len,
int mod);
int verify_dir_item(struct btrfs_fs_info *fs_info,
- struct extent_buffer *leaf,
+ struct extent_buffer *leaf, int slot,
struct btrfs_dir_item *dir_item);
struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
const char *name,
int name_len);
+bool btrfs_is_name_len_valid(struct extent_buffer *leaf, int slot,
+ unsigned long start, u16 name_len);
/* orphan.c */
int btrfs_insert_orphan_item(struct btrfs_trans_handle *trans,
@@ -3054,8 +3089,8 @@ int btrfs_find_name_in_ext_backref(struct btrfs_path *path,
struct btrfs_dio_private;
int btrfs_del_csums(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info, u64 bytenr, u64 len);
-int btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio, u32 *dst);
-int btrfs_lookup_bio_sums_dio(struct inode *inode, struct bio *bio,
+blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio, u32 *dst);
+blk_status_t btrfs_lookup_bio_sums_dio(struct inode *inode, struct bio *bio,
u64 logical_offset);
int btrfs_insert_file_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
@@ -3070,11 +3105,11 @@ int btrfs_lookup_file_extent(struct btrfs_trans_handle *trans,
int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_ordered_sum *sums);
-int btrfs_csum_one_bio(struct inode *inode, struct bio *bio,
+blk_status_t btrfs_csum_one_bio(struct inode *inode, struct bio *bio,
u64 file_start, int contig);
int btrfs_lookup_csums_range(struct btrfs_root *root, u64 start, u64 end,
struct list_head *list, int search_commit);
-void btrfs_extent_item_to_extent_map(struct inode *inode,
+void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
const struct btrfs_path *path,
struct btrfs_file_extent_item *fi,
const bool new_inline,
@@ -3093,9 +3128,9 @@ struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode,
int delay_iput);
void btrfs_wait_and_free_delalloc_work(struct btrfs_delalloc_work *work);
-struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *page,
- size_t pg_offset, u64 start, u64 len,
- int create);
+struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
+ struct page *page, size_t pg_offset, u64 start,
+ u64 len, int create);
noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
u64 *orig_start, u64 *orig_block_len,
u64 *ram_bytes);
@@ -3116,13 +3151,13 @@ static inline void btrfs_force_ra(struct address_space *mapping,
}
struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry);
-int btrfs_set_inode_index(struct inode *dir, u64 *index);
+int btrfs_set_inode_index(struct btrfs_inode *dir, u64 *index);
int btrfs_unlink_inode(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct inode *dir, struct inode *inode,
+ struct btrfs_inode *dir, struct btrfs_inode *inode,
const char *name, int name_len);
int btrfs_add_link(struct btrfs_trans_handle *trans,
- struct inode *parent_inode, struct inode *inode,
+ struct btrfs_inode *parent_inode, struct btrfs_inode *inode,
const char *name, int name_len, int add_backref, u64 index);
int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
@@ -3147,7 +3182,8 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
int btrfs_merge_bio_hook(struct page *page, unsigned long offset,
size_t size, struct bio *bio,
unsigned long bio_flags);
-int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf);
+void btrfs_set_range_writeback(void *private_data, u64 start, u64 end);
+int btrfs_page_mkwrite(struct vm_fault *vmf);
int btrfs_readpage(struct file *file, struct page *page);
void btrfs_evict_inode(struct inode *inode);
int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc);
@@ -3159,15 +3195,16 @@ void btrfs_destroy_cachep(void);
long btrfs_ioctl_trans_end(struct file *file);
struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
struct btrfs_root *root, int *was_new);
-struct extent_map *btrfs_get_extent(struct inode *inode, struct page *page,
- size_t pg_offset, u64 start, u64 end,
- int create);
+struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
+ struct page *page, size_t pg_offset,
+ u64 start, u64 end, int create);
int btrfs_update_inode(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct inode *inode);
int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct inode *inode);
-int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode);
+int btrfs_orphan_add(struct btrfs_trans_handle *trans,
+ struct btrfs_inode *inode);
int btrfs_orphan_cleanup(struct btrfs_root *root);
void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
@@ -3208,11 +3245,11 @@ ssize_t btrfs_dedupe_file_range(struct file *src_file, u64 loff, u64 olen,
int btrfs_auto_defrag_init(void);
void btrfs_auto_defrag_exit(void);
int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
- struct inode *inode);
+ struct btrfs_inode *inode);
int btrfs_run_defrag_inodes(struct btrfs_fs_info *fs_info);
void btrfs_cleanup_defrag_inodes(struct btrfs_fs_info *fs_info);
int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync);
-void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
+void btrfs_drop_extent_cache(struct btrfs_inode *inode, u64 start, u64 end,
int skip_pinned);
extern const struct file_operations btrfs_file_operations;
int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
@@ -3226,7 +3263,7 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct inode *inode, u64 start,
u64 end, int drop_cache);
int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
- struct inode *inode, u64 start, u64 end);
+ struct btrfs_inode *inode, u64 start, u64 end);
int btrfs_release_file(struct inode *inode, struct file *file);
int btrfs_dirty_pages(struct inode *inode, struct page **pages,
size_t num_pages, loff_t pos, size_t write_bytes,
@@ -3447,7 +3484,8 @@ do { \
"BTRFS: Transaction aborted (error %d)\n", \
(errno)); \
} else { \
- pr_debug("BTRFS: Transaction aborted (error %d)\n", \
+ btrfs_debug((trans)->fs_info, \
+ "Transaction aborted (error %d)", \
(errno)); \
} \
} \
@@ -3637,6 +3675,12 @@ int btrfs_scrub_cancel_dev(struct btrfs_fs_info *info,
struct btrfs_device *dev);
int btrfs_scrub_progress(struct btrfs_fs_info *fs_info, u64 devid,
struct btrfs_scrub_progress *progress);
+static inline void btrfs_init_full_stripe_locks_tree(
+ struct btrfs_full_stripe_locks_tree *locks_root)
+{
+ locks_root->root = RB_ROOT;
+ mutex_init(&locks_root->lock);
+}
/* dev-replace.c */
void btrfs_bio_counter_inc_blocked(struct btrfs_fs_info *fs_info);
@@ -3661,8 +3705,7 @@ struct reada_control *btrfs_reada_add(struct btrfs_root *root,
struct btrfs_key *start, struct btrfs_key *end);
int btrfs_reada_wait(void *handle);
void btrfs_reada_detach(void *handle);
-int btree_readahead_hook(struct btrfs_fs_info *fs_info,
- struct extent_buffer *eb, int err);
+int btree_readahead_hook(struct extent_buffer *eb, int err);
static inline int is_fstree(u64 rootid)
{
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 80982a83c9fd..8ae409b5a61d 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -52,7 +52,7 @@ static inline void btrfs_init_delayed_node(
{
delayed_node->root = root;
delayed_node->inode_id = inode_id;
- atomic_set(&delayed_node->refs, 0);
+ refcount_set(&delayed_node->refs, 0);
delayed_node->ins_root = RB_ROOT;
delayed_node->del_root = RB_ROOT;
mutex_init(&delayed_node->mutex);
@@ -72,16 +72,16 @@ static inline int btrfs_is_continuous_delayed_item(
return 0;
}
-static struct btrfs_delayed_node *btrfs_get_delayed_node(struct inode *inode)
+static struct btrfs_delayed_node *btrfs_get_delayed_node(
+ struct btrfs_inode *btrfs_inode)
{
- struct btrfs_inode *btrfs_inode = BTRFS_I(inode);
struct btrfs_root *root = btrfs_inode->root;
- u64 ino = btrfs_ino(inode);
+ u64 ino = btrfs_ino(btrfs_inode);
struct btrfs_delayed_node *node;
- node = ACCESS_ONCE(btrfs_inode->delayed_node);
+ node = READ_ONCE(btrfs_inode->delayed_node);
if (node) {
- atomic_inc(&node->refs);
+ refcount_inc(&node->refs);
return node;
}
@@ -89,14 +89,14 @@ static struct btrfs_delayed_node *btrfs_get_delayed_node(struct inode *inode)
node = radix_tree_lookup(&root->delayed_nodes_tree, ino);
if (node) {
if (btrfs_inode->delayed_node) {
- atomic_inc(&node->refs); /* can be accessed */
+ refcount_inc(&node->refs); /* can be accessed */
BUG_ON(btrfs_inode->delayed_node != node);
spin_unlock(&root->inode_lock);
return node;
}
btrfs_inode->delayed_node = node;
/* can be accessed and cached in the inode */
- atomic_add(2, &node->refs);
+ refcount_add(2, &node->refs);
spin_unlock(&root->inode_lock);
return node;
}
@@ -107,16 +107,15 @@ static struct btrfs_delayed_node *btrfs_get_delayed_node(struct inode *inode)
/* Will return either the node or PTR_ERR(-ENOMEM) */
static struct btrfs_delayed_node *btrfs_get_or_create_delayed_node(
- struct inode *inode)
+ struct btrfs_inode *btrfs_inode)
{
struct btrfs_delayed_node *node;
- struct btrfs_inode *btrfs_inode = BTRFS_I(inode);
struct btrfs_root *root = btrfs_inode->root;
- u64 ino = btrfs_ino(inode);
+ u64 ino = btrfs_ino(btrfs_inode);
int ret;
again:
- node = btrfs_get_delayed_node(inode);
+ node = btrfs_get_delayed_node(btrfs_inode);
if (node)
return node;
@@ -126,7 +125,7 @@ again:
btrfs_init_delayed_node(node, root, ino);
/* cached in the btrfs inode and can be accessed */
- atomic_add(2, &node->refs);
+ refcount_set(&node->refs, 2);
ret = radix_tree_preload(GFP_NOFS);
if (ret) {
@@ -167,7 +166,7 @@ static void btrfs_queue_delayed_node(struct btrfs_delayed_root *root,
} else {
list_add_tail(&node->n_list, &root->node_list);
list_add_tail(&node->p_list, &root->prepare_list);
- atomic_inc(&node->refs); /* inserted into list */
+ refcount_inc(&node->refs); /* inserted into list */
root->nodes++;
set_bit(BTRFS_DELAYED_NODE_IN_LIST, &node->flags);
}
@@ -181,7 +180,7 @@ static void btrfs_dequeue_delayed_node(struct btrfs_delayed_root *root,
spin_lock(&root->lock);
if (test_bit(BTRFS_DELAYED_NODE_IN_LIST, &node->flags)) {
root->nodes--;
- atomic_dec(&node->refs); /* not in the list */
+ refcount_dec(&node->refs); /* not in the list */
list_del_init(&node->n_list);
if (!list_empty(&node->p_list))
list_del_init(&node->p_list);
@@ -202,7 +201,7 @@ static struct btrfs_delayed_node *btrfs_first_delayed_node(
p = delayed_root->node_list.next;
node = list_entry(p, struct btrfs_delayed_node, n_list);
- atomic_inc(&node->refs);
+ refcount_inc(&node->refs);
out:
spin_unlock(&delayed_root->lock);
@@ -229,7 +228,7 @@ static struct btrfs_delayed_node *btrfs_next_delayed_node(
p = node->n_list.next;
next = list_entry(p, struct btrfs_delayed_node, n_list);
- atomic_inc(&next->refs);
+ refcount_inc(&next->refs);
out:
spin_unlock(&delayed_root->lock);
@@ -254,11 +253,11 @@ static void __btrfs_release_delayed_node(
btrfs_dequeue_delayed_node(delayed_root, delayed_node);
mutex_unlock(&delayed_node->mutex);
- if (atomic_dec_and_test(&delayed_node->refs)) {
+ if (refcount_dec_and_test(&delayed_node->refs)) {
bool free = false;
struct btrfs_root *root = delayed_node->root;
spin_lock(&root->inode_lock);
- if (atomic_read(&delayed_node->refs) == 0) {
+ if (refcount_read(&delayed_node->refs) == 0) {
radix_tree_delete(&root->delayed_nodes_tree,
delayed_node->inode_id);
free = true;
@@ -287,7 +286,7 @@ static struct btrfs_delayed_node *btrfs_first_prepared_delayed_node(
p = delayed_root->prepare_list.next;
list_del_init(p);
node = list_entry(p, struct btrfs_delayed_node, p_list);
- atomic_inc(&node->refs);
+ refcount_inc(&node->refs);
out:
spin_unlock(&delayed_root->lock);
@@ -309,7 +308,7 @@ static struct btrfs_delayed_item *btrfs_alloc_delayed_item(u32 data_len)
item->ins_or_del = 0;
item->bytes_reserved = 0;
item->delayed_node = NULL;
- atomic_set(&item->refs, 1);
+ refcount_set(&item->refs, 1);
}
return item;
}
@@ -484,7 +483,7 @@ static void btrfs_release_delayed_item(struct btrfs_delayed_item *item)
{
if (item) {
__btrfs_remove_delayed_item(item);
- if (atomic_dec_and_test(&item->refs))
+ if (refcount_dec_and_test(&item->refs))
kfree(item);
}
}
@@ -574,7 +573,7 @@ static void btrfs_delayed_item_release_metadata(struct btrfs_fs_info *fs_info,
static int btrfs_delayed_inode_reserve_metadata(
struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct inode *inode,
+ struct btrfs_inode *inode,
struct btrfs_delayed_node *node)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -603,13 +602,13 @@ static int btrfs_delayed_inode_reserve_metadata(
* worth which is less likely to hurt us.
*/
if (src_rsv && src_rsv->type == BTRFS_BLOCK_RSV_DELALLOC) {
- spin_lock(&BTRFS_I(inode)->lock);
+ spin_lock(&inode->lock);
if (test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
- &BTRFS_I(inode)->runtime_flags))
+ &inode->runtime_flags))
release = true;
else
src_rsv = NULL;
- spin_unlock(&BTRFS_I(inode)->lock);
+ spin_unlock(&inode->lock);
}
/*
@@ -1196,7 +1195,7 @@ int btrfs_run_delayed_items_nr(struct btrfs_trans_handle *trans,
}
int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
- struct inode *inode)
+ struct btrfs_inode *inode)
{
struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
struct btrfs_path *path;
@@ -1233,9 +1232,9 @@ int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
return ret;
}
-int btrfs_commit_inode_delayed_inode(struct inode *inode)
+int btrfs_commit_inode_delayed_inode(struct btrfs_inode *inode)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
struct btrfs_trans_handle *trans;
struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
struct btrfs_path *path;
@@ -1288,15 +1287,15 @@ out:
return ret;
}
-void btrfs_remove_delayed_node(struct inode *inode)
+void btrfs_remove_delayed_node(struct btrfs_inode *inode)
{
struct btrfs_delayed_node *delayed_node;
- delayed_node = ACCESS_ONCE(BTRFS_I(inode)->delayed_node);
+ delayed_node = READ_ONCE(inode->delayed_node);
if (!delayed_node)
return;
- BTRFS_I(inode)->delayed_node = NULL;
+ inode->delayed_node = NULL;
btrfs_release_delayed_node(delayed_node);
}
@@ -1434,7 +1433,7 @@ void btrfs_balance_delayed_items(struct btrfs_fs_info *fs_info)
int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
const char *name, int name_len,
- struct inode *dir,
+ struct btrfs_inode *dir,
struct btrfs_disk_key *disk_key, u8 type,
u64 index)
{
@@ -1510,7 +1509,7 @@ static int btrfs_delete_delayed_insertion_item(struct btrfs_fs_info *fs_info,
int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
- struct inode *dir, u64 index)
+ struct btrfs_inode *dir, u64 index)
{
struct btrfs_delayed_node *node;
struct btrfs_delayed_item *item;
@@ -1558,7 +1557,7 @@ end:
return ret;
}
-int btrfs_inode_delayed_dir_index_count(struct inode *inode)
+int btrfs_inode_delayed_dir_index_count(struct btrfs_inode *inode)
{
struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
@@ -1575,7 +1574,7 @@ int btrfs_inode_delayed_dir_index_count(struct inode *inode)
return -EINVAL;
}
- BTRFS_I(inode)->index_cnt = delayed_node->index_cnt;
+ inode->index_cnt = delayed_node->index_cnt;
btrfs_release_delayed_node(delayed_node);
return 0;
}
@@ -1587,7 +1586,7 @@ bool btrfs_readdir_get_delayed_items(struct inode *inode,
struct btrfs_delayed_node *delayed_node;
struct btrfs_delayed_item *item;
- delayed_node = btrfs_get_delayed_node(inode);
+ delayed_node = btrfs_get_delayed_node(BTRFS_I(inode));
if (!delayed_node)
return false;
@@ -1601,14 +1600,14 @@ bool btrfs_readdir_get_delayed_items(struct inode *inode,
mutex_lock(&delayed_node->mutex);
item = __btrfs_first_delayed_insertion_item(delayed_node);
while (item) {
- atomic_inc(&item->refs);
+ refcount_inc(&item->refs);
list_add_tail(&item->readdir_list, ins_list);
item = __btrfs_next_delayed_item(item);
}
item = __btrfs_first_delayed_deletion_item(delayed_node);
while (item) {
- atomic_inc(&item->refs);
+ refcount_inc(&item->refs);
list_add_tail(&item->readdir_list, del_list);
item = __btrfs_next_delayed_item(item);
}
@@ -1622,7 +1621,7 @@ bool btrfs_readdir_get_delayed_items(struct inode *inode,
* insert/delete delayed items in this period. So we also needn't
* requeue or dequeue this delayed node.
*/
- atomic_dec(&delayed_node->refs);
+ refcount_dec(&delayed_node->refs);
return true;
}
@@ -1635,13 +1634,13 @@ void btrfs_readdir_put_delayed_items(struct inode *inode,
list_for_each_entry_safe(curr, next, ins_list, readdir_list) {
list_del(&curr->readdir_list);
- if (atomic_dec_and_test(&curr->refs))
+ if (refcount_dec_and_test(&curr->refs))
kfree(curr);
}
list_for_each_entry_safe(curr, next, del_list, readdir_list) {
list_del(&curr->readdir_list);
- if (atomic_dec_and_test(&curr->refs))
+ if (refcount_dec_and_test(&curr->refs))
kfree(curr);
}
@@ -1668,7 +1667,7 @@ int btrfs_should_delete_dir_index(struct list_head *del_list,
list_del(&curr->readdir_list);
ret = (curr->key.offset == index);
- if (atomic_dec_and_test(&curr->refs))
+ if (refcount_dec_and_test(&curr->refs))
kfree(curr);
if (ret)
@@ -1706,7 +1705,7 @@ int btrfs_readdir_delayed_dir_index(struct dir_context *ctx,
list_del(&curr->readdir_list);
if (curr->key.offset < ctx->pos) {
- if (atomic_dec_and_test(&curr->refs))
+ if (refcount_dec_and_test(&curr->refs))
kfree(curr);
continue;
}
@@ -1723,7 +1722,7 @@ int btrfs_readdir_delayed_dir_index(struct dir_context *ctx,
over = !dir_emit(ctx, name, name_len,
location.objectid, d_type);
- if (atomic_dec_and_test(&curr->refs))
+ if (refcount_dec_and_test(&curr->refs))
kfree(curr);
if (over)
@@ -1776,7 +1775,7 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev)
struct btrfs_delayed_node *delayed_node;
struct btrfs_inode_item *inode_item;
- delayed_node = btrfs_get_delayed_node(inode);
+ delayed_node = btrfs_get_delayed_node(BTRFS_I(inode));
if (!delayed_node)
return -ENOENT;
@@ -1791,7 +1790,7 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev)
i_uid_write(inode, btrfs_stack_inode_uid(inode_item));
i_gid_write(inode, btrfs_stack_inode_gid(inode_item));
- btrfs_i_size_write(inode, btrfs_stack_inode_size(inode_item));
+ btrfs_i_size_write(BTRFS_I(inode), btrfs_stack_inode_size(inode_item));
inode->i_mode = btrfs_stack_inode_mode(inode_item);
set_nlink(inode, btrfs_stack_inode_nlink(inode_item));
inode_set_bytes(inode, btrfs_stack_inode_nbytes(inode_item));
@@ -1831,7 +1830,7 @@ int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
struct btrfs_delayed_node *delayed_node;
int ret = 0;
- delayed_node = btrfs_get_or_create_delayed_node(inode);
+ delayed_node = btrfs_get_or_create_delayed_node(BTRFS_I(inode));
if (IS_ERR(delayed_node))
return PTR_ERR(delayed_node);
@@ -1841,7 +1840,7 @@ int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
goto release_node;
}
- ret = btrfs_delayed_inode_reserve_metadata(trans, root, inode,
+ ret = btrfs_delayed_inode_reserve_metadata(trans, root, BTRFS_I(inode),
delayed_node);
if (ret)
goto release_node;
@@ -1856,9 +1855,9 @@ release_node:
return ret;
}
-int btrfs_delayed_delete_inode_ref(struct inode *inode)
+int btrfs_delayed_delete_inode_ref(struct btrfs_inode *inode)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
struct btrfs_delayed_node *delayed_node;
/*
@@ -1933,7 +1932,7 @@ static void __btrfs_kill_delayed_node(struct btrfs_delayed_node *delayed_node)
mutex_unlock(&delayed_node->mutex);
}
-void btrfs_kill_delayed_inode_items(struct inode *inode)
+void btrfs_kill_delayed_inode_items(struct btrfs_inode *inode)
{
struct btrfs_delayed_node *delayed_node;
@@ -1964,7 +1963,7 @@ void btrfs_kill_all_delayed_nodes(struct btrfs_root *root)
inode_id = delayed_nodes[n - 1]->inode_id + 1;
for (i = 0; i < n; i++)
- atomic_inc(&delayed_nodes[i]->refs);
+ refcount_inc(&delayed_nodes[i]->refs);
spin_unlock(&root->inode_lock);
for (i = 0; i < n; i++) {
diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h
index 8a2bf5e3e4cf..c4189d495934 100644
--- a/fs/btrfs/delayed-inode.h
+++ b/fs/btrfs/delayed-inode.h
@@ -26,7 +26,7 @@
#include <linux/list.h>
#include <linux/wait.h>
#include <linux/atomic.h>
-
+#include <linux/refcount.h>
#include "ctree.h"
/* types of the delayed item */
@@ -67,7 +67,7 @@ struct btrfs_delayed_node {
struct rb_root del_root;
struct mutex mutex;
struct btrfs_inode_item inode_item;
- atomic_t refs;
+ refcount_t refs;
u64 index_cnt;
unsigned long flags;
int count;
@@ -80,7 +80,7 @@ struct btrfs_delayed_item {
struct list_head readdir_list; /* used for readdir items */
u64 bytes_reserved;
struct btrfs_delayed_node *delayed_node;
- atomic_t refs;
+ refcount_t refs;
int ins_or_del;
u32 data_len;
char data[0];
@@ -101,15 +101,15 @@ static inline void btrfs_init_delayed_root(
int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
const char *name, int name_len,
- struct inode *dir,
+ struct btrfs_inode *dir,
struct btrfs_disk_key *disk_key, u8 type,
u64 index);
int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
- struct inode *dir, u64 index);
+ struct btrfs_inode *dir, u64 index);
-int btrfs_inode_delayed_dir_index_count(struct inode *inode);
+int btrfs_inode_delayed_dir_index_count(struct btrfs_inode *inode);
int btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info);
@@ -119,17 +119,17 @@ int btrfs_run_delayed_items_nr(struct btrfs_trans_handle *trans,
void btrfs_balance_delayed_items(struct btrfs_fs_info *fs_info);
int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
- struct inode *inode);
+ struct btrfs_inode *inode);
/* Used for evicting the inode. */
-void btrfs_remove_delayed_node(struct inode *inode);
-void btrfs_kill_delayed_inode_items(struct inode *inode);
-int btrfs_commit_inode_delayed_inode(struct inode *inode);
+void btrfs_remove_delayed_node(struct btrfs_inode *inode);
+void btrfs_kill_delayed_inode_items(struct btrfs_inode *inode);
+int btrfs_commit_inode_delayed_inode(struct btrfs_inode *inode);
int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct inode *inode);
int btrfs_fill_inode(struct inode *inode, u32 *rdev);
-int btrfs_delayed_delete_inode_ref(struct inode *inode);
+int btrfs_delayed_delete_inode_ref(struct btrfs_inode *inode);
/* Used for drop dead root */
void btrfs_kill_all_delayed_nodes(struct btrfs_root *root);
diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c
index ef724a5fc30e..93ffa898df6d 100644
--- a/fs/btrfs/delayed-ref.c
+++ b/fs/btrfs/delayed-ref.c
@@ -164,7 +164,7 @@ int btrfs_delayed_ref_lock(struct btrfs_trans_handle *trans,
if (mutex_trylock(&head->mutex))
return 0;
- atomic_inc(&head->node.refs);
+ refcount_inc(&head->node.refs);
spin_unlock(&delayed_refs->lock);
mutex_lock(&head->mutex);
@@ -470,7 +470,8 @@ add_tail:
static noinline void
update_existing_head_ref(struct btrfs_delayed_ref_root *delayed_refs,
struct btrfs_delayed_ref_node *existing,
- struct btrfs_delayed_ref_node *update)
+ struct btrfs_delayed_ref_node *update,
+ int *old_ref_mod_ret)
{
struct btrfs_delayed_ref_head *existing_ref;
struct btrfs_delayed_ref_head *ref;
@@ -523,6 +524,8 @@ update_existing_head_ref(struct btrfs_delayed_ref_root *delayed_refs,
* currently, for refs we just added we know we're a-ok.
*/
old_ref_mod = existing_ref->total_ref_mod;
+ if (old_ref_mod_ret)
+ *old_ref_mod_ret = old_ref_mod;
existing->ref_mod += update->ref_mod;
existing_ref->total_ref_mod += update->ref_mod;
@@ -550,13 +553,15 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
struct btrfs_delayed_ref_node *ref,
struct btrfs_qgroup_extent_record *qrecord,
u64 bytenr, u64 num_bytes, u64 ref_root, u64 reserved,
- int action, int is_data)
+ int action, int is_data, int *qrecord_inserted_ret,
+ int *old_ref_mod, int *new_ref_mod)
{
struct btrfs_delayed_ref_head *existing;
struct btrfs_delayed_ref_head *head_ref = NULL;
struct btrfs_delayed_ref_root *delayed_refs;
int count_mod = 1;
int must_insert_reserved = 0;
+ int qrecord_inserted = 0;
/* If reserved is provided, it must be a data extent. */
BUG_ON(!is_data && reserved);
@@ -589,7 +594,7 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
delayed_refs = &trans->transaction->delayed_refs;
/* first set the basic ref node struct up */
- atomic_set(&ref->refs, 1);
+ refcount_set(&ref->refs, 1);
ref->bytenr = bytenr;
ref->num_bytes = num_bytes;
ref->ref_mod = count_mod;
@@ -623,6 +628,8 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
if(btrfs_qgroup_trace_extent_nolock(fs_info,
delayed_refs, qrecord))
kfree(qrecord);
+ else
+ qrecord_inserted = 1;
}
spin_lock_init(&head_ref->lock);
@@ -635,7 +642,8 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
if (existing) {
WARN_ON(ref_root && reserved && existing->qgroup_ref_root
&& existing->qgroup_reserved);
- update_existing_head_ref(delayed_refs, &existing->node, ref);
+ update_existing_head_ref(delayed_refs, &existing->node, ref,
+ old_ref_mod);
/*
* we've updated the existing ref, free the newly
* allocated ref
@@ -643,6 +651,8 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
kmem_cache_free(btrfs_delayed_ref_head_cachep, head_ref);
head_ref = existing;
} else {
+ if (old_ref_mod)
+ *old_ref_mod = 0;
if (is_data && count_mod < 0)
delayed_refs->pending_csums += num_bytes;
delayed_refs->num_heads++;
@@ -650,6 +660,10 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
atomic_inc(&delayed_refs->num_entries);
trans->delayed_ref_updates++;
}
+ if (qrecord_inserted_ret)
+ *qrecord_inserted_ret = qrecord_inserted;
+ if (new_ref_mod)
+ *new_ref_mod = head_ref->total_ref_mod;
return head_ref;
}
@@ -677,7 +691,7 @@ add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
delayed_refs = &trans->transaction->delayed_refs;
/* first set the basic ref node struct up */
- atomic_set(&ref->refs, 1);
+ refcount_set(&ref->refs, 1);
ref->bytenr = bytenr;
ref->num_bytes = num_bytes;
ref->ref_mod = 1;
@@ -734,7 +748,7 @@ add_delayed_data_ref(struct btrfs_fs_info *fs_info,
seq = atomic64_read(&fs_info->tree_mod_seq);
/* first set the basic ref node struct up */
- atomic_set(&ref->refs, 1);
+ refcount_set(&ref->refs, 1);
ref->bytenr = bytenr;
ref->num_bytes = num_bytes;
ref->ref_mod = 1;
@@ -773,12 +787,14 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
struct btrfs_trans_handle *trans,
u64 bytenr, u64 num_bytes, u64 parent,
u64 ref_root, int level, int action,
- struct btrfs_delayed_extent_op *extent_op)
+ struct btrfs_delayed_extent_op *extent_op,
+ int *old_ref_mod, int *new_ref_mod)
{
struct btrfs_delayed_tree_ref *ref;
struct btrfs_delayed_ref_head *head_ref;
struct btrfs_delayed_ref_root *delayed_refs;
struct btrfs_qgroup_extent_record *record = NULL;
+ int qrecord_inserted;
BUG_ON(extent_op && extent_op->is_data);
ref = kmem_cache_alloc(btrfs_delayed_tree_ref_cachep, GFP_NOFS);
@@ -806,12 +822,16 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
* the spin lock
*/
head_ref = add_delayed_ref_head(fs_info, trans, &head_ref->node, record,
- bytenr, num_bytes, 0, 0, action, 0);
+ bytenr, num_bytes, 0, 0, action, 0,
+ &qrecord_inserted, old_ref_mod,
+ new_ref_mod);
add_delayed_tree_ref(fs_info, trans, head_ref, &ref->node, bytenr,
num_bytes, parent, ref_root, level, action);
spin_unlock(&delayed_refs->lock);
+ if (qrecord_inserted)
+ return btrfs_qgroup_trace_extent_post(fs_info, record);
return 0;
free_head_ref:
@@ -830,14 +850,14 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
u64 bytenr, u64 num_bytes,
u64 parent, u64 ref_root,
u64 owner, u64 offset, u64 reserved, int action,
- struct btrfs_delayed_extent_op *extent_op)
+ int *old_ref_mod, int *new_ref_mod)
{
struct btrfs_delayed_data_ref *ref;
struct btrfs_delayed_ref_head *head_ref;
struct btrfs_delayed_ref_root *delayed_refs;
struct btrfs_qgroup_extent_record *record = NULL;
+ int qrecord_inserted;
- BUG_ON(extent_op && !extent_op->is_data);
ref = kmem_cache_alloc(btrfs_delayed_data_ref_cachep, GFP_NOFS);
if (!ref)
return -ENOMEM;
@@ -859,7 +879,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
}
}
- head_ref->extent_op = extent_op;
+ head_ref->extent_op = NULL;
delayed_refs = &trans->transaction->delayed_refs;
spin_lock(&delayed_refs->lock);
@@ -870,13 +890,16 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
*/
head_ref = add_delayed_ref_head(fs_info, trans, &head_ref->node, record,
bytenr, num_bytes, ref_root, reserved,
- action, 1);
+ action, 1, &qrecord_inserted,
+ old_ref_mod, new_ref_mod);
add_delayed_data_ref(fs_info, trans, head_ref, &ref->node, bytenr,
num_bytes, parent, ref_root, owner, offset,
action);
spin_unlock(&delayed_refs->lock);
+ if (qrecord_inserted)
+ return btrfs_qgroup_trace_extent_post(fs_info, record);
return 0;
}
@@ -899,7 +922,7 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
add_delayed_ref_head(fs_info, trans, &head_ref->node, NULL, bytenr,
num_bytes, 0, 0, BTRFS_UPDATE_DELAYED_HEAD,
- extent_op->is_data);
+ extent_op->is_data, NULL, NULL, NULL);
spin_unlock(&delayed_refs->lock);
return 0;
@@ -911,11 +934,8 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
* the head node if any where found, or NULL if not.
*/
struct btrfs_delayed_ref_head *
-btrfs_find_delayed_ref_head(struct btrfs_trans_handle *trans, u64 bytenr)
+btrfs_find_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_refs, u64 bytenr)
{
- struct btrfs_delayed_ref_root *delayed_refs;
-
- delayed_refs = &trans->transaction->delayed_refs;
return find_ref_head(&delayed_refs->href_root, bytenr, 0);
}
diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h
index 50947b5a9152..ce88e4ac5276 100644
--- a/fs/btrfs/delayed-ref.h
+++ b/fs/btrfs/delayed-ref.h
@@ -18,6 +18,8 @@
#ifndef __DELAYED_REF__
#define __DELAYED_REF__
+#include <linux/refcount.h>
+
/* these are the possible values of struct btrfs_delayed_ref_node->action */
#define BTRFS_ADD_DELAYED_REF 1 /* add one backref to the tree */
#define BTRFS_DROP_DELAYED_REF 2 /* delete one backref from the tree */
@@ -53,7 +55,7 @@ struct btrfs_delayed_ref_node {
u64 seq;
/* ref count on this data structure */
- atomic_t refs;
+ refcount_t refs;
/*
* how many refs is this entry adding or deleting. For
@@ -220,8 +222,8 @@ btrfs_free_delayed_extent_op(struct btrfs_delayed_extent_op *op)
static inline void btrfs_put_delayed_ref(struct btrfs_delayed_ref_node *ref)
{
- WARN_ON(atomic_read(&ref->refs) == 0);
- if (atomic_dec_and_test(&ref->refs)) {
+ WARN_ON(refcount_read(&ref->refs) == 0);
+ if (refcount_dec_and_test(&ref->refs)) {
WARN_ON(ref->in_tree);
switch (ref->type) {
case BTRFS_TREE_BLOCK_REF_KEY:
@@ -245,13 +247,14 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
struct btrfs_trans_handle *trans,
u64 bytenr, u64 num_bytes, u64 parent,
u64 ref_root, int level, int action,
- struct btrfs_delayed_extent_op *extent_op);
+ struct btrfs_delayed_extent_op *extent_op,
+ int *old_ref_mod, int *new_ref_mod);
int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
struct btrfs_trans_handle *trans,
u64 bytenr, u64 num_bytes,
u64 parent, u64 ref_root,
u64 owner, u64 offset, u64 reserved, int action,
- struct btrfs_delayed_extent_op *extent_op);
+ int *old_ref_mod, int *new_ref_mod);
int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
struct btrfs_trans_handle *trans,
u64 bytenr, u64 num_bytes,
@@ -262,7 +265,8 @@ void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
struct btrfs_delayed_ref_head *head);
struct btrfs_delayed_ref_head *
-btrfs_find_delayed_ref_head(struct btrfs_trans_handle *trans, u64 bytenr);
+btrfs_find_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_refs,
+ u64 bytenr);
int btrfs_delayed_ref_lock(struct btrfs_trans_handle *trans,
struct btrfs_delayed_ref_head *head);
static inline void btrfs_delayed_ref_unlock(struct btrfs_delayed_ref_head *head)
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index 5de280b9ad73..bee3edeea7a3 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -304,8 +304,9 @@ void btrfs_after_dev_replace_commit(struct btrfs_fs_info *fs_info)
dev_replace->cursor_left_last_write_of_item;
}
-int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, char *tgtdev_name,
- u64 srcdevid, char *srcdev_name, int read_src)
+int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
+ const char *tgtdev_name, u64 srcdevid, const char *srcdev_name,
+ int read_src)
{
struct btrfs_root *root = fs_info->dev_root;
struct btrfs_trans_handle *trans;
@@ -387,7 +388,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, char *tgtdev_name,
if (ret)
btrfs_err(fs_info, "kobj add dev failed %d", ret);
- btrfs_wait_ordered_roots(fs_info, -1, 0, (u64)-1);
+ btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
/* force writing the updated state information to disk */
trans = btrfs_start_transaction(root, 0);
@@ -506,7 +507,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return ret;
}
- btrfs_wait_ordered_roots(fs_info, -1, 0, (u64)-1);
+ btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
@@ -545,8 +546,10 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
mutex_unlock(&fs_info->chunk_mutex);
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
mutex_unlock(&uuid_mutex);
+ btrfs_rm_dev_replace_blocked(fs_info);
if (tgt_device)
btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
+ btrfs_rm_dev_replace_unblocked(fs_info);
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return scrub_ret;
@@ -664,7 +667,7 @@ void btrfs_dev_replace_status(struct btrfs_fs_info *fs_info,
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
srcdev = dev_replace->srcdev;
- args->status.progress_1000 = div_u64(dev_replace->cursor_left,
+ args->status.progress_1000 = div64_u64(dev_replace->cursor_left,
div_u64(btrfs_device_get_total_bytes(srcdev), 1000));
break;
}
@@ -783,8 +786,7 @@ int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info)
}
btrfs_dev_replace_unlock(dev_replace, 1);
- WARN_ON(atomic_xchg(
- &fs_info->mutually_exclusive_operation_running, 1));
+ WARN_ON(test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags));
task = kthread_run(btrfs_dev_replace_kthread, fs_info, "btrfs-devrepl");
return PTR_ERR_OR_ZERO(task);
}
@@ -813,7 +815,7 @@ static int btrfs_dev_replace_kthread(void *data)
(unsigned int)progress);
}
btrfs_dev_replace_continue_on_mount(fs_info);
- atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
+ clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
return 0;
}
diff --git a/fs/btrfs/dev-replace.h b/fs/btrfs/dev-replace.h
index 54ea12bda15b..f94a76844ae7 100644
--- a/fs/btrfs/dev-replace.h
+++ b/fs/btrfs/dev-replace.h
@@ -27,8 +27,9 @@ int btrfs_run_dev_replace(struct btrfs_trans_handle *trans,
void btrfs_after_dev_replace_commit(struct btrfs_fs_info *fs_info);
int btrfs_dev_replace_by_ioctl(struct btrfs_fs_info *fs_info,
struct btrfs_ioctl_dev_replace_args *args);
-int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, char *tgtdev_name,
- u64 srcdevid, char *srcdev_name, int read_src);
+int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
+ const char *tgtdev_name, u64 srcdevid, const char *srcdev_name,
+ int read_src);
void btrfs_dev_replace_status(struct btrfs_fs_info *fs_info,
struct btrfs_ioctl_dev_replace_args *args);
int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info,
diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c
index b039fe0c751a..41cb9196eaa8 100644
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -80,7 +80,8 @@ int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
struct extent_buffer *leaf;
u32 data_size;
- BUG_ON(name_len + data_len > BTRFS_MAX_XATTR_SIZE(root->fs_info));
+ if (name_len + data_len > BTRFS_MAX_XATTR_SIZE(root->fs_info))
+ return -ENOSPC;
key.objectid = objectid;
key.type = BTRFS_XATTR_ITEM_KEY;
@@ -120,7 +121,7 @@ int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
*/
int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, struct btrfs_root
*root, const char *name, int name_len,
- struct inode *dir, struct btrfs_key *location,
+ struct btrfs_inode *dir, struct btrfs_key *location,
u8 type, u64 index)
{
int ret = 0;
@@ -174,8 +175,7 @@ second_insert:
btrfs_release_path(path);
ret2 = btrfs_insert_delayed_dir_index(trans, root->fs_info, name,
- name_len, dir, &disk_key, type,
- index);
+ name_len, dir, &disk_key, type, index);
out_free:
btrfs_free_path(path);
if (ret)
@@ -395,8 +395,6 @@ struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_fs_info *fs_info,
leaf = path->nodes[0];
dir_item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_dir_item);
- if (verify_dir_item(fs_info, leaf, dir_item))
- return NULL;
total_len = btrfs_item_size_nr(leaf, path->slots[0]);
while (cur < total_len) {
@@ -405,6 +403,8 @@ struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_fs_info *fs_info,
btrfs_dir_data_len(leaf, dir_item);
name_ptr = (unsigned long)(dir_item + 1);
+ if (verify_dir_item(fs_info, leaf, path->slots[0], dir_item))
+ return NULL;
if (btrfs_dir_name_len(leaf, dir_item) == name_len &&
memcmp_extent_buffer(leaf, name, name_ptr, name_len) == 0)
return dir_item;
@@ -453,9 +453,11 @@ int btrfs_delete_one_dir_name(struct btrfs_trans_handle *trans,
int verify_dir_item(struct btrfs_fs_info *fs_info,
struct extent_buffer *leaf,
+ int slot,
struct btrfs_dir_item *dir_item)
{
u16 namelen = BTRFS_NAME_LEN;
+ int ret;
u8 type = btrfs_dir_type(leaf, dir_item);
if (type >= BTRFS_FT_MAX) {
@@ -468,10 +470,16 @@ int verify_dir_item(struct btrfs_fs_info *fs_info,
if (btrfs_dir_name_len(leaf, dir_item) > namelen) {
btrfs_crit(fs_info, "invalid dir item name len: %u",
- (unsigned)btrfs_dir_data_len(leaf, dir_item));
+ (unsigned)btrfs_dir_name_len(leaf, dir_item));
return 1;
}
+ namelen = btrfs_dir_name_len(leaf, dir_item);
+ ret = btrfs_is_name_len_valid(leaf, slot,
+ (unsigned long)(dir_item + 1), namelen);
+ if (!ret)
+ return 1;
+
/* BTRFS_MAX_XATTR_SIZE is the same for all dir items */
if ((btrfs_dir_data_len(leaf, dir_item) +
btrfs_dir_name_len(leaf, dir_item)) >
@@ -484,3 +492,67 @@ int verify_dir_item(struct btrfs_fs_info *fs_info,
return 0;
}
+
+bool btrfs_is_name_len_valid(struct extent_buffer *leaf, int slot,
+ unsigned long start, u16 name_len)
+{
+ struct btrfs_fs_info *fs_info = leaf->fs_info;
+ struct btrfs_key key;
+ u32 read_start;
+ u32 read_end;
+ u32 item_start;
+ u32 item_end;
+ u32 size;
+ bool ret = true;
+
+ ASSERT(start > BTRFS_LEAF_DATA_OFFSET);
+
+ read_start = start - BTRFS_LEAF_DATA_OFFSET;
+ read_end = read_start + name_len;
+ item_start = btrfs_item_offset_nr(leaf, slot);
+ item_end = btrfs_item_end_nr(leaf, slot);
+
+ btrfs_item_key_to_cpu(leaf, &key, slot);
+
+ switch (key.type) {
+ case BTRFS_DIR_ITEM_KEY:
+ case BTRFS_XATTR_ITEM_KEY:
+ case BTRFS_DIR_INDEX_KEY:
+ size = sizeof(struct btrfs_dir_item);
+ break;
+ case BTRFS_INODE_REF_KEY:
+ size = sizeof(struct btrfs_inode_ref);
+ break;
+ case BTRFS_INODE_EXTREF_KEY:
+ size = sizeof(struct btrfs_inode_extref);
+ break;
+ case BTRFS_ROOT_REF_KEY:
+ case BTRFS_ROOT_BACKREF_KEY:
+ size = sizeof(struct btrfs_root_ref);
+ break;
+ default:
+ ret = false;
+ goto out;
+ }
+
+ if (read_start < item_start) {
+ ret = false;
+ goto out;
+ }
+ if (read_end > item_end) {
+ ret = false;
+ goto out;
+ }
+
+ /* there shall be item(s) before name */
+ if (read_start - item_start < size) {
+ ret = false;
+ goto out;
+ }
+
+out:
+ if (!ret)
+ btrfs_crit(fs_info, "invalid dir item name len: %u",
+ (unsigned int)name_len);
+ return ret;
+}
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 18004169552c..080e2ebb8aa0 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -64,8 +64,7 @@
static const struct extent_io_ops btree_extent_io_ops;
static void end_workqueue_fn(struct btrfs_work *work);
static void free_fs_root(struct btrfs_root *root);
-static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
- int read_only);
+static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info);
static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
struct btrfs_fs_info *fs_info);
@@ -88,9 +87,8 @@ struct btrfs_end_io_wq {
bio_end_io_t *end_io;
void *private;
struct btrfs_fs_info *info;
- int error;
+ blk_status_t status;
enum btrfs_wq_endio_type metadata;
- struct list_head list;
struct btrfs_work work;
};
@@ -119,9 +117,9 @@ void btrfs_end_io_wq_exit(void)
* just before they are sent down the IO stack.
*/
struct async_submit_bio {
- struct inode *inode;
+ void *private_data;
+ struct btrfs_fs_info *fs_info;
struct bio *bio;
- struct list_head list;
extent_submit_bio_hook_t *submit_bio_start;
extent_submit_bio_hook_t *submit_bio_done;
int mirror_num;
@@ -132,7 +130,7 @@ struct async_submit_bio {
*/
u64 bio_offset;
struct btrfs_work work;
- int error;
+ blk_status_t status;
};
/*
@@ -220,12 +218,12 @@ void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb,
* extents on the btree inode are pretty simple, there's one extent
* that covers the entire device
*/
-static struct extent_map *btree_get_extent(struct inode *inode,
+static struct extent_map *btree_get_extent(struct btrfs_inode *inode,
struct page *page, size_t pg_offset, u64 start, u64 len,
int create)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct extent_map_tree *em_tree = &inode->extent_tree;
struct extent_map *em;
int ret;
@@ -266,7 +264,7 @@ out:
return em;
}
-u32 btrfs_csum_data(char *data, u32 seed, size_t len)
+u32 btrfs_csum_data(const char *data, u32 seed, size_t len)
{
return btrfs_crc32c(seed, data, len);
}
@@ -763,7 +761,7 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
err:
if (reads_done &&
test_and_clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags))
- btree_readahead_hook(fs_info, eb, ret);
+ btree_readahead_hook(eb, ret);
if (ret) {
/*
@@ -788,7 +786,7 @@ static int btree_io_failed_hook(struct page *page, int failed_mirror)
eb->read_mirror = failed_mirror;
atomic_dec(&eb->io_pages);
if (test_and_clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags))
- btree_readahead_hook(eb->fs_info, eb, -EIO);
+ btree_readahead_hook(eb, -EIO);
return -EIO; /* we fixed nothing */
}
@@ -800,7 +798,7 @@ static void end_workqueue_bio(struct bio *bio)
btrfs_work_func_t func;
fs_info = end_io_wq->info;
- end_io_wq->error = bio->bi_error;
+ end_io_wq->status = bio->bi_status;
if (bio_op(bio) == REQ_OP_WRITE) {
if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA) {
@@ -837,19 +835,19 @@ static void end_workqueue_bio(struct bio *bio)
btrfs_queue_work(wq, &end_io_wq->work);
}
-int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
+blk_status_t btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
enum btrfs_wq_endio_type metadata)
{
struct btrfs_end_io_wq *end_io_wq;
end_io_wq = kmem_cache_alloc(btrfs_end_io_wq_cache, GFP_NOFS);
if (!end_io_wq)
- return -ENOMEM;
+ return BLK_STS_RESOURCE;
end_io_wq->private = bio->bi_private;
end_io_wq->end_io = bio->bi_end_io;
end_io_wq->info = info;
- end_io_wq->error = 0;
+ end_io_wq->status = 0;
end_io_wq->bio = bio;
end_io_wq->metadata = metadata;
@@ -869,14 +867,14 @@ unsigned long btrfs_async_submit_limit(struct btrfs_fs_info *info)
static void run_one_async_start(struct btrfs_work *work)
{
struct async_submit_bio *async;
- int ret;
+ blk_status_t ret;
async = container_of(work, struct async_submit_bio, work);
- ret = async->submit_bio_start(async->inode, async->bio,
+ ret = async->submit_bio_start(async->private_data, async->bio,
async->mirror_num, async->bio_flags,
async->bio_offset);
if (ret)
- async->error = ret;
+ async->status = ret;
}
static void run_one_async_done(struct btrfs_work *work)
@@ -886,7 +884,7 @@ static void run_one_async_done(struct btrfs_work *work)
int limit;
async = container_of(work, struct async_submit_bio, work);
- fs_info = BTRFS_I(async->inode)->root->fs_info;
+ fs_info = async->fs_info;
limit = btrfs_async_submit_limit(fs_info);
limit = limit * 2 / 3;
@@ -899,13 +897,13 @@ static void run_one_async_done(struct btrfs_work *work)
wake_up(&fs_info->async_submit_wait);
/* If an error occurred we just want to clean up the bio and move on */
- if (async->error) {
- async->bio->bi_error = async->error;
+ if (async->status) {
+ async->bio->bi_status = async->status;
bio_endio(async->bio);
return;
}
- async->submit_bio_done(async->inode, async->bio, async->mirror_num,
+ async->submit_bio_done(async->private_data, async->bio, async->mirror_num,
async->bio_flags, async->bio_offset);
}
@@ -917,20 +915,20 @@ static void run_one_async_free(struct btrfs_work *work)
kfree(async);
}
-int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
- struct bio *bio, int mirror_num,
- unsigned long bio_flags,
- u64 bio_offset,
- extent_submit_bio_hook_t *submit_bio_start,
- extent_submit_bio_hook_t *submit_bio_done)
+blk_status_t btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
+ int mirror_num, unsigned long bio_flags,
+ u64 bio_offset, void *private_data,
+ extent_submit_bio_hook_t *submit_bio_start,
+ extent_submit_bio_hook_t *submit_bio_done)
{
struct async_submit_bio *async;
async = kmalloc(sizeof(*async), GFP_NOFS);
if (!async)
- return -ENOMEM;
+ return BLK_STS_RESOURCE;
- async->inode = inode;
+ async->private_data = private_data;
+ async->fs_info = fs_info;
async->bio = bio;
async->mirror_num = mirror_num;
async->submit_bio_start = submit_bio_start;
@@ -942,7 +940,7 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
async->bio_flags = bio_flags;
async->bio_offset = bio_offset;
- async->error = 0;
+ async->status = 0;
atomic_inc(&fs_info->nr_async_submits);
@@ -960,12 +958,13 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
return 0;
}
-static int btree_csum_one_bio(struct bio *bio)
+static blk_status_t btree_csum_one_bio(struct bio *bio)
{
struct bio_vec *bvec;
struct btrfs_root *root;
int i, ret = 0;
+ ASSERT(!bio_flagged(bio, BIO_CLONED));
bio_for_each_segment_all(bvec, bio, i) {
root = BTRFS_I(bvec->bv_page->mapping->host)->root;
ret = csum_dirty_buffer(root->fs_info, bvec->bv_page);
@@ -973,12 +972,12 @@ static int btree_csum_one_bio(struct bio *bio)
break;
}
- return ret;
+ return errno_to_blk_status(ret);
}
-static int __btree_submit_bio_start(struct inode *inode, struct bio *bio,
- int mirror_num, unsigned long bio_flags,
- u64 bio_offset)
+static blk_status_t __btree_submit_bio_start(void *private_data, struct bio *bio,
+ int mirror_num, unsigned long bio_flags,
+ u64 bio_offset)
{
/*
* when we're called for a write, we're already in the async
@@ -987,11 +986,12 @@ static int __btree_submit_bio_start(struct inode *inode, struct bio *bio,
return btree_csum_one_bio(bio);
}
-static int __btree_submit_bio_done(struct inode *inode, struct bio *bio,
- int mirror_num, unsigned long bio_flags,
- u64 bio_offset)
+static blk_status_t __btree_submit_bio_done(void *private_data, struct bio *bio,
+ int mirror_num, unsigned long bio_flags,
+ u64 bio_offset)
{
- int ret;
+ struct inode *inode = private_data;
+ blk_status_t ret;
/*
* when we're called for a write, we're already in the async
@@ -999,13 +999,13 @@ static int __btree_submit_bio_done(struct inode *inode, struct bio *bio,
*/
ret = btrfs_map_bio(btrfs_sb(inode->i_sb), bio, mirror_num, 1);
if (ret) {
- bio->bi_error = ret;
+ bio->bi_status = ret;
bio_endio(bio);
}
return ret;
}
-static int check_async_write(struct inode *inode, unsigned long bio_flags)
+static int check_async_write(unsigned long bio_flags)
{
if (bio_flags & EXTENT_BIO_TREE_LOG)
return 0;
@@ -1016,13 +1016,14 @@ static int check_async_write(struct inode *inode, unsigned long bio_flags)
return 1;
}
-static int btree_submit_bio_hook(struct inode *inode, struct bio *bio,
- int mirror_num, unsigned long bio_flags,
- u64 bio_offset)
+static blk_status_t btree_submit_bio_hook(void *private_data, struct bio *bio,
+ int mirror_num, unsigned long bio_flags,
+ u64 bio_offset)
{
+ struct inode *inode = private_data;
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- int async = check_async_write(inode, bio_flags);
- int ret;
+ int async = check_async_write(bio_flags);
+ blk_status_t ret;
if (bio_op(bio) != REQ_OP_WRITE) {
/*
@@ -1044,8 +1045,8 @@ static int btree_submit_bio_hook(struct inode *inode, struct bio *bio,
* kthread helpers are used to submit writes so that
* checksumming can happen in parallel across all CPUs
*/
- ret = btrfs_wq_submit_bio(fs_info, inode, bio, mirror_num, 0,
- bio_offset,
+ ret = btrfs_wq_submit_bio(fs_info, bio, mirror_num, 0,
+ bio_offset, private_data,
__btree_submit_bio_start,
__btree_submit_bio_done);
}
@@ -1055,7 +1056,7 @@ static int btree_submit_bio_hook(struct inode *inode, struct bio *bio,
return 0;
out_w_error:
- bio->bi_error = ret;
+ bio->bi_status = ret;
bio_endio(bio);
return ret;
}
@@ -1223,10 +1224,10 @@ int btrfs_write_tree_block(struct extent_buffer *buf)
buf->start + buf->len - 1);
}
-int btrfs_wait_tree_block_writeback(struct extent_buffer *buf)
+void btrfs_wait_tree_block_writeback(struct extent_buffer *buf)
{
- return filemap_fdatawait_range(buf->pages[0]->mapping,
- buf->start, buf->start + buf->len - 1);
+ filemap_fdatawait_range(buf->pages[0]->mapping,
+ buf->start, buf->start + buf->len - 1);
}
struct extent_buffer *read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr,
@@ -1248,8 +1249,7 @@ struct extent_buffer *read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr,
}
-void clean_tree_block(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info,
+void clean_tree_block(struct btrfs_fs_info *fs_info,
struct extent_buffer *buf)
{
if (btrfs_header_generation(buf) ==
@@ -1257,9 +1257,9 @@ void clean_tree_block(struct btrfs_trans_handle *trans,
btrfs_assert_tree_locked(buf);
if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &buf->bflags)) {
- __percpu_counter_add(&fs_info->dirty_metadata_bytes,
- -buf->len,
- fs_info->dirty_metadata_batch);
+ percpu_counter_add_batch(&fs_info->dirty_metadata_bytes,
+ -buf->len,
+ fs_info->dirty_metadata_batch);
/* ugh, clear_extent_buffer_dirty needs to lock the page */
btrfs_set_lock_blocking(buf);
clear_extent_buffer_dirty(buf);
@@ -1342,15 +1342,14 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
atomic_set(&root->log_writers, 0);
atomic_set(&root->log_batch, 0);
atomic_set(&root->orphan_inodes, 0);
- atomic_set(&root->refs, 1);
+ refcount_set(&root->refs, 1);
atomic_set(&root->will_be_snapshoted, 0);
- atomic_set(&root->qgroup_meta_rsv, 0);
+ atomic64_set(&root->qgroup_meta_rsv, 0);
root->log_transid = 0;
root->log_transid_committed = -1;
root->last_log_commit = 0;
if (!dummy)
- extent_io_tree_init(&root->dirty_log_pages,
- fs_info->btree_inode->i_mapping);
+ extent_io_tree_init(&root->dirty_log_pages, NULL);
memset(&root->root_key, 0, sizeof(root->root_key));
memset(&root->root_item, 0, sizeof(root->root_item));
@@ -1800,7 +1799,7 @@ static int btrfs_congested_fn(void *congested_data, int bdi_bits)
list_for_each_entry_rcu(device, &info->fs_devices->devices, dev_list) {
if (!device->bdev)
continue;
- bdi = blk_get_backing_dev_info(device->bdev);
+ bdi = device->bdev->bd_bdi;
if (bdi_congested(bdi, bdi_bits)) {
ret = 1;
break;
@@ -1810,21 +1809,6 @@ static int btrfs_congested_fn(void *congested_data, int bdi_bits)
return ret;
}
-static int setup_bdi(struct btrfs_fs_info *info, struct backing_dev_info *bdi)
-{
- int err;
-
- err = bdi_setup_and_register(bdi, "btrfs");
- if (err)
- return err;
-
- bdi->ra_pages = VM_MAX_READAHEAD * 1024 / PAGE_SIZE;
- bdi->congested_fn = btrfs_congested_fn;
- bdi->congested_data = info;
- bdi->capabilities |= BDI_CAP_CGROUP_WRITEBACK;
- return 0;
-}
-
/*
* called by the kthread helper functions to finally call the bio end_io
* functions. This is where read checksum verification actually happens
@@ -1837,7 +1821,7 @@ static void end_workqueue_fn(struct btrfs_work *work)
end_io_wq = container_of(work, struct btrfs_end_io_wq, work);
bio = end_io_wq->bio;
- bio->bi_error = end_io_wq->error;
+ bio->bi_status = end_io_wq->status;
bio->bi_private = end_io_wq->private;
bio->bi_end_io = end_io_wq->end_io;
kmem_cache_free(btrfs_end_io_wq_cache, end_io_wq);
@@ -2207,11 +2191,9 @@ static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
btrfs_destroy_workqueue(fs_info->delalloc_workers);
btrfs_destroy_workqueue(fs_info->workers);
btrfs_destroy_workqueue(fs_info->endio_workers);
- btrfs_destroy_workqueue(fs_info->endio_meta_workers);
btrfs_destroy_workqueue(fs_info->endio_raid56_workers);
btrfs_destroy_workqueue(fs_info->endio_repair_workers);
btrfs_destroy_workqueue(fs_info->rmw_workers);
- btrfs_destroy_workqueue(fs_info->endio_meta_write_workers);
btrfs_destroy_workqueue(fs_info->endio_write_workers);
btrfs_destroy_workqueue(fs_info->endio_freespace_worker);
btrfs_destroy_workqueue(fs_info->submit_workers);
@@ -2221,6 +2203,13 @@ static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
btrfs_destroy_workqueue(fs_info->flush_workers);
btrfs_destroy_workqueue(fs_info->qgroup_rescan_workers);
btrfs_destroy_workqueue(fs_info->extent_workers);
+ /*
+ * Now that all other work queues are destroyed, we can safely destroy
+ * the queues used for metadata I/O, since tasks from those other work
+ * queues can do metadata I/O operations.
+ */
+ btrfs_destroy_workqueue(fs_info->endio_meta_workers);
+ btrfs_destroy_workqueue(fs_info->endio_meta_write_workers);
}
static void free_root_extent_buffers(struct btrfs_root *root)
@@ -2321,7 +2310,7 @@ static void btrfs_init_btree_inode(struct btrfs_fs_info *fs_info)
inode->i_mapping->a_ops = &btree_aops;
RB_CLEAR_NODE(&BTRFS_I(inode)->rb_node);
- extent_io_tree_init(&BTRFS_I(inode)->io_tree, inode->i_mapping);
+ extent_io_tree_init(&BTRFS_I(inode)->io_tree, inode);
BTRFS_I(inode)->io_tree.track_uptodate = 0;
extent_map_tree_init(&BTRFS_I(inode)->extent_tree);
@@ -2598,16 +2587,10 @@ int open_ctree(struct super_block *sb,
goto fail;
}
- ret = setup_bdi(fs_info, &fs_info->bdi);
- if (ret) {
- err = ret;
- goto fail_srcu;
- }
-
ret = percpu_counter_init(&fs_info->dirty_metadata_bytes, 0, GFP_KERNEL);
if (ret) {
err = ret;
- goto fail_bdi;
+ goto fail_srcu;
}
fs_info->dirty_metadata_batch = PAGE_SIZE *
(1 + ilog2(nr_cpu_ids));
@@ -2644,7 +2627,6 @@ int open_ctree(struct super_block *sb,
spin_lock_init(&fs_info->fs_roots_radix_lock);
spin_lock_init(&fs_info->delayed_iput_lock);
spin_lock_init(&fs_info->defrag_inodes_lock);
- spin_lock_init(&fs_info->free_chunk_lock);
spin_lock_init(&fs_info->tree_mod_seq_lock);
spin_lock_init(&fs_info->super_lock);
spin_lock_init(&fs_info->qgroup_op_lock);
@@ -2680,12 +2662,11 @@ int open_ctree(struct super_block *sb,
atomic_set(&fs_info->qgroup_op_seq, 0);
atomic_set(&fs_info->reada_works_cnt, 0);
atomic64_set(&fs_info->tree_mod_seq, 0);
- fs_info->fs_frozen = 0;
fs_info->sb = sb;
fs_info->max_inline = BTRFS_DEFAULT_MAX_INLINE;
fs_info->metadata_ratio = 0;
fs_info->defrag_inodes = RB_ROOT;
- fs_info->free_chunk_space = 0;
+ atomic64_set(&fs_info->free_chunk_space, 0);
fs_info->tree_mod_log = RB_ROOT;
fs_info->commit_interval = BTRFS_DEFAULT_COMMIT_INTERVAL;
fs_info->avg_delayed_ref_runtime = NSEC_PER_SEC >> 6; /* div by 64 */
@@ -2715,7 +2696,6 @@ int open_ctree(struct super_block *sb,
sb->s_blocksize = 4096;
sb->s_blocksize_bits = blksize_bits(4096);
- sb->s_bdi = &fs_info->bdi;
btrfs_init_btree_inode(fs_info);
@@ -2723,10 +2703,8 @@ int open_ctree(struct super_block *sb,
fs_info->block_group_cache_tree = RB_ROOT;
fs_info->first_logical_byte = (u64)-1;
- extent_io_tree_init(&fs_info->freed_extents[0],
- fs_info->btree_inode->i_mapping);
- extent_io_tree_init(&fs_info->freed_extents[1],
- fs_info->btree_inode->i_mapping);
+ extent_io_tree_init(&fs_info->freed_extents[0], NULL);
+ extent_io_tree_init(&fs_info->freed_extents[1], NULL);
fs_info->pinned_extents = &fs_info->freed_extents[0];
set_bit(BTRFS_FS_BARRIER, &fs_info->flags);
@@ -2802,7 +2780,7 @@ int open_ctree(struct super_block *sb,
memcpy(fs_info->fsid, fs_info->super_copy->fsid, BTRFS_FSID_SIZE);
- ret = btrfs_check_super_valid(fs_info, sb->s_flags & MS_RDONLY);
+ ret = btrfs_check_super_valid(fs_info);
if (ret) {
btrfs_err(fs_info, "superblock contains fatal errors");
err = -EINVAL;
@@ -2912,9 +2890,12 @@ int open_ctree(struct super_block *sb,
goto fail_sb_buffer;
}
- fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super);
- fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages,
- SZ_4M / PAGE_SIZE);
+ sb->s_bdi->congested_fn = btrfs_congested_fn;
+ sb->s_bdi->congested_data = fs_info;
+ sb->s_bdi->capabilities |= BDI_CAP_CGROUP_WRITEBACK;
+ sb->s_bdi->ra_pages = VM_MAX_READAHEAD * 1024 / PAGE_SIZE;
+ sb->s_bdi->ra_pages *= btrfs_super_num_devices(disk_super);
+ sb->s_bdi->ra_pages = max(sb->s_bdi->ra_pages, SZ_4M / PAGE_SIZE);
sb->s_blocksize = sectorsize;
sb->s_blocksize_bits = blksize_bits(sectorsize);
@@ -3263,7 +3244,6 @@ fail_fsdev_sysfs:
fail_block_groups:
btrfs_put_block_group_cache(fs_info);
- btrfs_free_block_groups(fs_info);
fail_tree_roots:
free_root_pointers(fs_info, 1);
@@ -3271,6 +3251,7 @@ fail_tree_roots:
fail_sb_buffer:
btrfs_stop_all_workers(fs_info);
+ btrfs_free_block_groups(fs_info);
fail_alloc:
fail_iput:
btrfs_mapping_tree_free(&fs_info->mapping_tree);
@@ -3282,8 +3263,6 @@ fail_delalloc_bytes:
percpu_counter_destroy(&fs_info->delalloc_bytes);
fail_dirty_metadata_bytes:
percpu_counter_destroy(&fs_info->dirty_metadata_bytes);
-fail_bdi:
- bdi_destroy(&fs_info->bdi);
fail_srcu:
cleanup_srcu_struct(&fs_info->subvol_srcu);
fail:
@@ -3411,7 +3390,7 @@ struct buffer_head *btrfs_read_dev_super(struct block_device *bdev)
*/
static int write_dev_supers(struct btrfs_device *device,
struct btrfs_super_block *sb,
- int do_barriers, int wait, int max_mirrors)
+ int wait, int max_mirrors)
{
struct buffer_head *bh;
int i;
@@ -3450,7 +3429,7 @@ static int write_dev_supers(struct btrfs_device *device,
btrfs_set_super_bytenr(sb, bytenr);
crc = ~(u32)0;
- crc = btrfs_csum_data((char *)sb +
+ crc = btrfs_csum_data((const char *)sb +
BTRFS_CSUM_SIZE, crc,
BTRFS_SUPER_INFO_SIZE -
BTRFS_CSUM_SIZE);
@@ -3485,10 +3464,12 @@ static int write_dev_supers(struct btrfs_device *device,
* we fua the first super. The others we allow
* to go down lazy.
*/
- if (i == 0)
- ret = btrfsic_submit_bh(REQ_OP_WRITE, REQ_FUA, bh);
- else
+ if (i == 0) {
+ ret = btrfsic_submit_bh(REQ_OP_WRITE,
+ REQ_SYNC | REQ_FUA, bh);
+ } else {
ret = btrfsic_submit_bh(REQ_OP_WRITE, REQ_SYNC, bh);
+ }
if (ret)
errors++;
}
@@ -3501,64 +3482,61 @@ static int write_dev_supers(struct btrfs_device *device,
*/
static void btrfs_end_empty_barrier(struct bio *bio)
{
- if (bio->bi_private)
- complete(bio->bi_private);
- bio_put(bio);
+ complete(bio->bi_private);
}
/*
- * trigger flushes for one the devices. If you pass wait == 0, the flushes are
- * sent down. With wait == 1, it waits for the previous flush.
- *
- * any device where the flush fails with eopnotsupp are flagged as not-barrier
- * capable
+ * Submit a flush request to the device if it supports it. Error handling is
+ * done in the waiting counterpart.
*/
-static int write_dev_flush(struct btrfs_device *device, int wait)
+static void write_dev_flush(struct btrfs_device *device)
{
- struct bio *bio;
- int ret = 0;
+ struct request_queue *q = bdev_get_queue(device->bdev);
+ struct bio *bio = device->flush_bio;
- if (device->nobarriers)
- return 0;
+ if (!test_bit(QUEUE_FLAG_WC, &q->queue_flags))
+ return;
- if (wait) {
- bio = device->flush_bio;
- if (!bio)
- return 0;
+ bio_reset(bio);
+ bio->bi_end_io = btrfs_end_empty_barrier;
+ bio->bi_bdev = device->bdev;
+ bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_PREFLUSH;
+ init_completion(&device->flush_wait);
+ bio->bi_private = &device->flush_wait;
- wait_for_completion(&device->flush_wait);
+ submit_bio(bio);
+ device->flush_bio_sent = 1;
+}
- if (bio->bi_error) {
- ret = bio->bi_error;
- btrfs_dev_stat_inc_and_print(device,
- BTRFS_DEV_STAT_FLUSH_ERRS);
- }
+/*
+ * If the flush bio has been submitted by write_dev_flush, wait for it.
+ */
+static blk_status_t wait_dev_flush(struct btrfs_device *device)
+{
+ struct bio *bio = device->flush_bio;
- /* drop the reference from the wait == 0 run */
- bio_put(bio);
- device->flush_bio = NULL;
+ if (!device->flush_bio_sent)
+ return 0;
- return ret;
- }
+ device->flush_bio_sent = 0;
+ wait_for_completion_io(&device->flush_wait);
- /*
- * one reference for us, and we leave it for the
- * caller
- */
- device->flush_bio = NULL;
- bio = btrfs_io_bio_alloc(GFP_NOFS, 0);
- if (!bio)
- return -ENOMEM;
+ return bio->bi_status;
+}
- bio->bi_end_io = btrfs_end_empty_barrier;
- bio->bi_bdev = device->bdev;
- bio->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH;
- init_completion(&device->flush_wait);
- bio->bi_private = &device->flush_wait;
- device->flush_bio = bio;
+static int check_barrier_error(struct btrfs_fs_devices *fsdevs)
+{
+ int dev_flush_error = 0;
+ struct btrfs_device *dev;
- bio_get(bio);
- btrfsic_submit_bio(bio);
+ list_for_each_entry_rcu(dev, &fsdevs->devices, dev_list) {
+ if (!dev->bdev || dev->last_flush_error)
+ dev_flush_error++;
+ }
+
+ if (dev_flush_error >
+ fsdevs->fs_info->num_tolerated_disk_barrier_failures)
+ return -EIO;
return 0;
}
@@ -3571,25 +3549,21 @@ static int barrier_all_devices(struct btrfs_fs_info *info)
{
struct list_head *head;
struct btrfs_device *dev;
- int errors_send = 0;
int errors_wait = 0;
- int ret;
+ blk_status_t ret;
/* send down all the barriers */
head = &info->fs_devices->devices;
list_for_each_entry_rcu(dev, head, dev_list) {
if (dev->missing)
continue;
- if (!dev->bdev) {
- errors_send++;
+ if (!dev->bdev)
continue;
- }
if (!dev->in_fs_metadata || !dev->writeable)
continue;
- ret = write_dev_flush(dev, 0);
- if (ret)
- errors_send++;
+ write_dev_flush(dev);
+ dev->last_flush_error = 0;
}
/* wait for all the barriers */
@@ -3603,13 +3577,23 @@ static int barrier_all_devices(struct btrfs_fs_info *info)
if (!dev->in_fs_metadata || !dev->writeable)
continue;
- ret = write_dev_flush(dev, 1);
- if (ret)
+ ret = wait_dev_flush(dev);
+ if (ret) {
+ dev->last_flush_error = ret;
+ btrfs_dev_stat_inc_and_print(dev,
+ BTRFS_DEV_STAT_FLUSH_ERRS);
errors_wait++;
+ }
+ }
+
+ if (errors_wait) {
+ /*
+ * At some point we need the status of all disks
+ * to arrive at the volume status. So error checking
+ * is being pushed to a separate loop.
+ */
+ return check_barrier_error(info->fs_devices);
}
- if (errors_send > info->num_tolerated_disk_barrier_failures ||
- errors_wait > info->num_tolerated_disk_barrier_failures)
- return -EIO;
return 0;
}
@@ -3696,7 +3680,7 @@ int btrfs_calc_num_tolerated_disk_barrier_failures(
return num_tolerated_disk_barrier_failures;
}
-static int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
+int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
{
struct list_head *head;
struct btrfs_device *dev;
@@ -3753,7 +3737,7 @@ static int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
flags = btrfs_super_flags(sb);
btrfs_set_super_flags(sb, flags | BTRFS_HEADER_FLAG_WRITTEN);
- ret = write_dev_supers(dev, sb, do_barriers, 0, max_mirrors);
+ ret = write_dev_supers(dev, sb, 0, max_mirrors);
if (ret)
total_errors++;
}
@@ -3776,7 +3760,7 @@ static int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
if (!dev->in_fs_metadata || !dev->writeable)
continue;
- ret = write_dev_supers(dev, sb, do_barriers, 1, max_mirrors);
+ ret = write_dev_supers(dev, sb, 1, max_mirrors);
if (ret)
total_errors++;
}
@@ -3790,12 +3774,6 @@ static int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
return 0;
}
-int write_ctree_super(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info, int max_mirrors)
-{
- return write_all_supers(fs_info, max_mirrors);
-}
-
/* Drop a fs root from the radix tree and free it. */
void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
struct btrfs_root *root)
@@ -3985,8 +3963,6 @@ void close_ctree(struct btrfs_fs_info *fs_info)
btrfs_put_block_group_cache(fs_info);
- btrfs_free_block_groups(fs_info);
-
/*
* we must make sure there is not any read request to
* submit after we stopping all workers.
@@ -3994,6 +3970,8 @@ void close_ctree(struct btrfs_fs_info *fs_info)
invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
btrfs_stop_all_workers(fs_info);
+ btrfs_free_block_groups(fs_info);
+
clear_bit(BTRFS_FS_OPEN, &fs_info->flags);
free_root_pointers(fs_info, 1);
@@ -4010,7 +3988,6 @@ void close_ctree(struct btrfs_fs_info *fs_info)
percpu_counter_destroy(&fs_info->dirty_metadata_bytes);
percpu_counter_destroy(&fs_info->delalloc_bytes);
percpu_counter_destroy(&fs_info->bio_counter);
- bdi_destroy(&fs_info->bdi);
cleanup_srcu_struct(&fs_info->subvol_srcu);
btrfs_free_stripe_hash_table(fs_info);
@@ -4071,9 +4048,9 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
buf->start, transid, fs_info->generation);
was_dirty = set_extent_buffer_dirty(buf);
if (!was_dirty)
- __percpu_counter_add(&fs_info->dirty_metadata_bytes,
- buf->len,
- fs_info->dirty_metadata_batch);
+ percpu_counter_add_batch(&fs_info->dirty_metadata_bytes,
+ buf->len,
+ fs_info->dirty_metadata_batch);
#ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
if (btrfs_header_level(buf) == 0 && check_leaf(root, buf)) {
btrfs_print_leaf(fs_info, buf);
@@ -4122,8 +4099,7 @@ int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid)
return btree_read_extent_buffer_pages(fs_info, buf, parent_transid);
}
-static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
- int read_only)
+static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info)
{
struct btrfs_super_block *sb = fs_info->super_copy;
u64 nodesize = btrfs_super_nodesize(sb);
@@ -4347,7 +4323,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
head = rb_entry(node, struct btrfs_delayed_ref_head,
href_node);
if (!mutex_trylock(&head->mutex)) {
- atomic_inc(&head->node.refs);
+ refcount_inc(&head->node.refs);
spin_unlock(&delayed_refs->lock);
mutex_lock(&head->mutex);
@@ -4601,11 +4577,6 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
cur_trans->state =TRANS_STATE_COMPLETED;
wake_up(&cur_trans->commit_wait);
-
- /*
- memset(cur_trans, 0, sizeof(*cur_trans));
- kmem_cache_free(btrfs_transaction_cachep, cur_trans);
- */
}
static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info)
@@ -4619,7 +4590,7 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info)
t = list_first_entry(&fs_info->trans_list,
struct btrfs_transaction, list);
if (t->state >= TRANS_STATE_COMMIT_START) {
- atomic_inc(&t->use_count);
+ refcount_inc(&t->use_count);
spin_unlock(&fs_info->trans_lock);
btrfs_wait_for_commit(fs_info, t->transid);
btrfs_put_transaction(t);
@@ -4661,10 +4632,21 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info)
return 0;
}
+static struct btrfs_fs_info *btree_fs_info(void *private_data)
+{
+ struct inode *inode = private_data;
+ return btrfs_sb(inode->i_sb);
+}
+
static const struct extent_io_ops btree_extent_io_ops = {
- .readpage_end_io_hook = btree_readpage_end_io_hook,
- .readpage_io_failed_hook = btree_io_failed_hook,
+ /* mandatory callbacks */
.submit_bio_hook = btree_submit_bio_hook,
+ .readpage_end_io_hook = btree_readpage_end_io_hook,
/* note we're sharing with inode.c for the merge bio hook */
.merge_bio_hook = btrfs_merge_bio_hook,
+ .readpage_io_failed_hook = btree_io_failed_hook,
+ .set_range_writeback = btrfs_set_range_writeback,
+ .tree_fs_info = btree_fs_info,
+
+ /* optional callbacks */
};
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index 44dcd9af6b7c..0a634d3ffc16 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -52,14 +52,12 @@ int reada_tree_block_flagged(struct btrfs_fs_info *fs_info, u64 bytenr,
struct extent_buffer *btrfs_find_create_tree_block(
struct btrfs_fs_info *fs_info,
u64 bytenr);
-void clean_tree_block(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info, struct extent_buffer *buf);
+void clean_tree_block(struct btrfs_fs_info *fs_info, struct extent_buffer *buf);
int open_ctree(struct super_block *sb,
struct btrfs_fs_devices *fs_devices,
char *options);
void close_ctree(struct btrfs_fs_info *fs_info);
-int write_ctree_super(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info, int max_mirrors);
+int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors);
struct buffer_head *btrfs_read_dev_super(struct block_device *bdev);
int btrfs_read_dev_one_super(struct block_device *bdev, int copy_num,
struct buffer_head **bh_ret);
@@ -103,14 +101,14 @@ struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info);
*/
static inline struct btrfs_root *btrfs_grab_fs_root(struct btrfs_root *root)
{
- if (atomic_inc_not_zero(&root->refs))
+ if (refcount_inc_not_zero(&root->refs))
return root;
return NULL;
}
static inline void btrfs_put_fs_root(struct btrfs_root *root)
{
- if (atomic_dec_and_test(&root->refs))
+ if (refcount_dec_and_test(&root->refs))
kfree(root);
}
@@ -118,18 +116,18 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf);
int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid,
int atomic);
int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid);
-u32 btrfs_csum_data(char *data, u32 seed, size_t len);
+u32 btrfs_csum_data(const char *data, u32 seed, size_t len);
void btrfs_csum_final(u32 crc, u8 *result);
-int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
+blk_status_t btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
enum btrfs_wq_endio_type metadata);
-int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
- struct bio *bio, int mirror_num,
- unsigned long bio_flags, u64 bio_offset,
+blk_status_t btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
+ int mirror_num, unsigned long bio_flags,
+ u64 bio_offset, void *private_data,
extent_submit_bio_hook_t *submit_bio_start,
extent_submit_bio_hook_t *submit_bio_done);
unsigned long btrfs_async_submit_limit(struct btrfs_fs_info *info);
int btrfs_write_tree_block(struct extent_buffer *buf);
-int btrfs_wait_tree_block_writeback(struct extent_buffer *buf);
+void btrfs_wait_tree_block_writeback(struct extent_buffer *buf);
int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info);
int btrfs_add_log_tree(struct btrfs_trans_handle *trans,
diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index 340d90751263..fa66980726c9 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -30,7 +30,7 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len,
len = BTRFS_FID_SIZE_NON_CONNECTABLE;
type = FILEID_BTRFS_WITHOUT_PARENT;
- fid->objectid = btrfs_ino(inode);
+ fid->objectid = btrfs_ino(BTRFS_I(inode));
fid->root_objectid = BTRFS_I(inode)->root->objectid;
fid->gen = inode->i_generation;
@@ -166,13 +166,13 @@ static struct dentry *btrfs_get_parent(struct dentry *child)
if (!path)
return ERR_PTR(-ENOMEM);
- if (btrfs_ino(dir) == BTRFS_FIRST_FREE_OBJECTID) {
+ if (btrfs_ino(BTRFS_I(dir)) == BTRFS_FIRST_FREE_OBJECTID) {
key.objectid = root->root_key.objectid;
key.type = BTRFS_ROOT_BACKREF_KEY;
key.offset = (u64)-1;
root = fs_info->tree_root;
} else {
- key.objectid = btrfs_ino(dir);
+ key.objectid = btrfs_ino(BTRFS_I(dir));
key.type = BTRFS_INODE_REF_KEY;
key.offset = (u64)-1;
}
@@ -235,13 +235,10 @@ static int btrfs_get_name(struct dentry *parent, char *name,
int ret;
u64 ino;
- if (!dir || !inode)
- return -EINVAL;
-
if (!S_ISDIR(dir->i_mode))
return -EINVAL;
- ino = btrfs_ino(inode);
+ ino = btrfs_ino(BTRFS_I(inode));
path = btrfs_alloc_path();
if (!path)
@@ -255,7 +252,7 @@ static int btrfs_get_name(struct dentry *parent, char *name,
root = fs_info->tree_root;
} else {
key.objectid = ino;
- key.offset = btrfs_ino(dir);
+ key.offset = btrfs_ino(BTRFS_I(dir));
key.type = BTRFS_INODE_REF_KEY;
}
@@ -285,6 +282,11 @@ static int btrfs_get_name(struct dentry *parent, char *name,
name_len = btrfs_inode_ref_name_len(leaf, iref);
}
+ ret = btrfs_is_name_len_valid(leaf, path->slots[0], name_ptr, name_len);
+ if (!ret) {
+ btrfs_free_path(path);
+ return -EIO;
+ }
read_extent_buffer(leaf, name, name_ptr, name_len);
btrfs_free_path(path);
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index dcd2e798767e..375f8c728d91 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -16,6 +16,7 @@
* Boston, MA 021110-1307, USA.
*/
#include <linux/sched.h>
+#include <linux/sched/signal.h>
#include <linux/pagemap.h>
#include <linux/writeback.h>
#include <linux/blkdev.h>
@@ -96,10 +97,11 @@ static int btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache,
u64 num_bytes, int delalloc);
static int block_rsv_use_bytes(struct btrfs_block_rsv *block_rsv,
u64 num_bytes);
-static int __reserve_metadata_bytes(struct btrfs_root *root,
+static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
struct btrfs_space_info *space_info,
u64 orig_bytes,
- enum btrfs_reserve_flush_enum flush);
+ enum btrfs_reserve_flush_enum flush,
+ bool system_chunk);
static void space_info_add_new_bytes(struct btrfs_fs_info *fs_info,
struct btrfs_space_info *space_info,
u64 num_bytes);
@@ -130,6 +132,16 @@ void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
if (atomic_dec_and_test(&cache->count)) {
WARN_ON(cache->pinned > 0);
WARN_ON(cache->reserved > 0);
+
+ /*
+ * If not empty, someone is still holding mutex of
+ * full_stripe_lock, which can only be released by caller.
+ * And it will definitely cause use-after-free when caller
+ * tries to release full stripe lock.
+ *
+ * No better way to resolve, but only to warn.
+ */
+ WARN_ON(!RB_EMPTY_ROOT(&cache->full_stripe_locks_root.root));
kfree(cache->free_space_ctl);
kfree(cache);
}
@@ -315,14 +327,14 @@ get_caching_control(struct btrfs_block_group_cache *cache)
}
ctl = cache->caching_ctl;
- atomic_inc(&ctl->count);
+ refcount_inc(&ctl->count);
spin_unlock(&cache->lock);
return ctl;
}
static void put_caching_control(struct btrfs_caching_control *ctl)
{
- if (atomic_dec_and_test(&ctl->count))
+ if (refcount_dec_and_test(&ctl->count))
kfree(ctl);
}
@@ -598,7 +610,7 @@ static int cache_block_group(struct btrfs_block_group_cache *cache,
init_waitqueue_head(&caching_ctl->wait);
caching_ctl->block_group = cache;
caching_ctl->progress = cache->key.objectid;
- atomic_set(&caching_ctl->count, 1);
+ refcount_set(&caching_ctl->count, 1);
btrfs_init_work(&caching_ctl->work, btrfs_cache_helper,
caching_thread, NULL, NULL);
@@ -619,7 +631,7 @@ static int cache_block_group(struct btrfs_block_group_cache *cache,
struct btrfs_caching_control *ctl;
ctl = cache->caching_ctl;
- atomic_inc(&ctl->count);
+ refcount_inc(&ctl->count);
prepare_to_wait(&ctl->wait, &wait, TASK_UNINTERRUPTIBLE);
spin_unlock(&cache->lock);
@@ -706,7 +718,7 @@ static int cache_block_group(struct btrfs_block_group_cache *cache,
}
down_write(&fs_info->commit_root_sem);
- atomic_inc(&caching_ctl->count);
+ refcount_inc(&caching_ctl->count);
list_add_tail(&caching_ctl->list, &fs_info->caching_block_groups);
up_write(&fs_info->commit_root_sem);
@@ -755,6 +767,26 @@ static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
return NULL;
}
+static void add_pinned_bytes(struct btrfs_fs_info *fs_info, s64 num_bytes,
+ u64 owner, u64 root_objectid)
+{
+ struct btrfs_space_info *space_info;
+ u64 flags;
+
+ if (owner < BTRFS_FIRST_FREE_OBJECTID) {
+ if (root_objectid == BTRFS_CHUNK_TREE_OBJECTID)
+ flags = BTRFS_BLOCK_GROUP_SYSTEM;
+ else
+ flags = BTRFS_BLOCK_GROUP_METADATA;
+ } else {
+ flags = BTRFS_BLOCK_GROUP_DATA;
+ }
+
+ space_info = __find_space_info(fs_info, flags);
+ ASSERT(space_info);
+ percpu_counter_add(&space_info->total_bytes_pinned, num_bytes);
+}
+
/*
* after adding space to the filesystem, we need to clear the full flags
* on all the space infos.
@@ -888,10 +920,10 @@ search_again:
delayed_refs = &trans->transaction->delayed_refs;
spin_lock(&delayed_refs->lock);
- head = btrfs_find_delayed_ref_head(trans, bytenr);
+ head = btrfs_find_delayed_ref_head(delayed_refs, bytenr);
if (head) {
if (!mutex_trylock(&head->mutex)) {
- atomic_inc(&head->node.refs);
+ refcount_inc(&head->node.refs);
spin_unlock(&delayed_refs->lock);
btrfs_release_path(path);
@@ -1035,10 +1067,11 @@ out_free:
#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
static int convert_extent_item_v0(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
u64 owner, u32 extra_size)
{
+ struct btrfs_root *root = fs_info->extent_root;
struct btrfs_extent_item *item;
struct btrfs_extent_item_v0 *ei0;
struct btrfs_extent_ref_v0 *ref0;
@@ -1092,7 +1125,7 @@ static int convert_extent_item_v0(struct btrfs_trans_handle *trans,
return ret;
BUG_ON(ret); /* Corruption */
- btrfs_extend_item(root->fs_info, path, new_size);
+ btrfs_extend_item(fs_info, path, new_size);
leaf = path->nodes[0];
item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
@@ -1151,12 +1184,13 @@ static int match_extent_data_ref(struct extent_buffer *leaf,
}
static noinline int lookup_extent_data_ref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
u64 bytenr, u64 parent,
u64 root_objectid,
u64 owner, u64 offset)
{
+ struct btrfs_root *root = fs_info->extent_root;
struct btrfs_key key;
struct btrfs_extent_data_ref *ref;
struct extent_buffer *leaf;
@@ -1238,12 +1272,13 @@ fail:
}
static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
u64 bytenr, u64 parent,
u64 root_objectid, u64 owner,
u64 offset, int refs_to_add)
{
+ struct btrfs_root *root = fs_info->extent_root;
struct btrfs_key key;
struct extent_buffer *leaf;
u32 size;
@@ -1317,7 +1352,7 @@ fail:
}
static noinline int remove_extent_data_ref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
int refs_to_drop, int *last_ref)
{
@@ -1354,7 +1389,7 @@ static noinline int remove_extent_data_ref(struct btrfs_trans_handle *trans,
num_refs -= refs_to_drop;
if (num_refs == 0) {
- ret = btrfs_del_item(trans, root, path);
+ ret = btrfs_del_item(trans, fs_info->extent_root, path);
*last_ref = 1;
} else {
if (key.type == BTRFS_EXTENT_DATA_REF_KEY)
@@ -1416,11 +1451,12 @@ static noinline u32 extent_data_ref_count(struct btrfs_path *path,
}
static noinline int lookup_tree_block_ref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
u64 bytenr, u64 parent,
u64 root_objectid)
{
+ struct btrfs_root *root = fs_info->extent_root;
struct btrfs_key key;
int ret;
@@ -1449,7 +1485,7 @@ static noinline int lookup_tree_block_ref(struct btrfs_trans_handle *trans,
}
static noinline int insert_tree_block_ref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
u64 bytenr, u64 parent,
u64 root_objectid)
@@ -1466,7 +1502,8 @@ static noinline int insert_tree_block_ref(struct btrfs_trans_handle *trans,
key.offset = root_objectid;
}
- ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
+ ret = btrfs_insert_empty_item(trans, fs_info->extent_root,
+ path, &key, 0);
btrfs_release_path(path);
return ret;
}
@@ -1524,14 +1561,14 @@ static int find_next_key(struct btrfs_path *path, int level,
*/
static noinline_for_stack
int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
struct btrfs_extent_inline_ref **ref_ret,
u64 bytenr, u64 num_bytes,
u64 parent, u64 root_objectid,
u64 owner, u64 offset, int insert)
{
- struct btrfs_fs_info *fs_info = root->fs_info;
+ struct btrfs_root *root = fs_info->extent_root;
struct btrfs_key key;
struct extent_buffer *leaf;
struct btrfs_extent_item *ei;
@@ -1614,7 +1651,7 @@ again:
err = -ENOENT;
goto out;
}
- ret = convert_extent_item_v0(trans, root, path, owner,
+ ret = convert_extent_item_v0(trans, fs_info, path, owner,
extra_size);
if (ret < 0) {
err = ret;
@@ -1716,7 +1753,7 @@ out:
* helper to add new inline back ref
*/
static noinline_for_stack
-void setup_inline_extent_backref(struct btrfs_root *root,
+void setup_inline_extent_backref(struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
struct btrfs_extent_inline_ref *iref,
u64 parent, u64 root_objectid,
@@ -1739,7 +1776,7 @@ void setup_inline_extent_backref(struct btrfs_root *root,
type = extent_ref_type(parent, owner);
size = btrfs_extent_inline_ref_size(type);
- btrfs_extend_item(root->fs_info, path, size);
+ btrfs_extend_item(fs_info, path, size);
ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
refs = btrfs_extent_refs(leaf, ei);
@@ -1777,7 +1814,7 @@ void setup_inline_extent_backref(struct btrfs_root *root,
}
static int lookup_extent_backref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
struct btrfs_extent_inline_ref **ref_ret,
u64 bytenr, u64 num_bytes, u64 parent,
@@ -1785,7 +1822,7 @@ static int lookup_extent_backref(struct btrfs_trans_handle *trans,
{
int ret;
- ret = lookup_inline_extent_backref(trans, root, path, ref_ret,
+ ret = lookup_inline_extent_backref(trans, fs_info, path, ref_ret,
bytenr, num_bytes, parent,
root_objectid, owner, offset, 0);
if (ret != -ENOENT)
@@ -1795,11 +1832,12 @@ static int lookup_extent_backref(struct btrfs_trans_handle *trans,
*ref_ret = NULL;
if (owner < BTRFS_FIRST_FREE_OBJECTID) {
- ret = lookup_tree_block_ref(trans, root, path, bytenr, parent,
- root_objectid);
+ ret = lookup_tree_block_ref(trans, fs_info, path, bytenr,
+ parent, root_objectid);
} else {
- ret = lookup_extent_data_ref(trans, root, path, bytenr, parent,
- root_objectid, owner, offset);
+ ret = lookup_extent_data_ref(trans, fs_info, path, bytenr,
+ parent, root_objectid, owner,
+ offset);
}
return ret;
}
@@ -1808,7 +1846,7 @@ static int lookup_extent_backref(struct btrfs_trans_handle *trans,
* helper to update/remove inline back ref
*/
static noinline_for_stack
-void update_inline_extent_backref(struct btrfs_root *root,
+void update_inline_extent_backref(struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
struct btrfs_extent_inline_ref *iref,
int refs_to_mod,
@@ -1866,14 +1904,14 @@ void update_inline_extent_backref(struct btrfs_root *root,
memmove_extent_buffer(leaf, ptr, ptr + size,
end - ptr - size);
item_size -= size;
- btrfs_truncate_item(root->fs_info, path, item_size, 1);
+ btrfs_truncate_item(fs_info, path, item_size, 1);
}
btrfs_mark_buffer_dirty(leaf);
}
static noinline_for_stack
int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
u64 bytenr, u64 num_bytes, u64 parent,
u64 root_objectid, u64 owner,
@@ -1883,15 +1921,15 @@ int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
struct btrfs_extent_inline_ref *iref;
int ret;
- ret = lookup_inline_extent_backref(trans, root, path, &iref,
+ ret = lookup_inline_extent_backref(trans, fs_info, path, &iref,
bytenr, num_bytes, parent,
root_objectid, owner, offset, 1);
if (ret == 0) {
BUG_ON(owner < BTRFS_FIRST_FREE_OBJECTID);
- update_inline_extent_backref(root, path, iref,
+ update_inline_extent_backref(fs_info, path, iref,
refs_to_add, extent_op, NULL);
} else if (ret == -ENOENT) {
- setup_inline_extent_backref(root, path, iref, parent,
+ setup_inline_extent_backref(fs_info, path, iref, parent,
root_objectid, owner, offset,
refs_to_add, extent_op);
ret = 0;
@@ -1900,7 +1938,7 @@ int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
}
static int insert_extent_backref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
u64 bytenr, u64 parent, u64 root_objectid,
u64 owner, u64 offset, int refs_to_add)
@@ -1908,10 +1946,10 @@ static int insert_extent_backref(struct btrfs_trans_handle *trans,
int ret;
if (owner < BTRFS_FIRST_FREE_OBJECTID) {
BUG_ON(refs_to_add != 1);
- ret = insert_tree_block_ref(trans, root, path, bytenr,
+ ret = insert_tree_block_ref(trans, fs_info, path, bytenr,
parent, root_objectid);
} else {
- ret = insert_extent_data_ref(trans, root, path, bytenr,
+ ret = insert_extent_data_ref(trans, fs_info, path, bytenr,
parent, root_objectid,
owner, offset, refs_to_add);
}
@@ -1919,7 +1957,7 @@ static int insert_extent_backref(struct btrfs_trans_handle *trans,
}
static int remove_extent_backref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
struct btrfs_extent_inline_ref *iref,
int refs_to_drop, int is_data, int *last_ref)
@@ -1928,14 +1966,14 @@ static int remove_extent_backref(struct btrfs_trans_handle *trans,
BUG_ON(!is_data && refs_to_drop != 1);
if (iref) {
- update_inline_extent_backref(root, path, iref,
+ update_inline_extent_backref(fs_info, path, iref,
-refs_to_drop, NULL, last_ref);
} else if (is_data) {
- ret = remove_extent_data_ref(trans, root, path, refs_to_drop,
+ ret = remove_extent_data_ref(trans, fs_info, path, refs_to_drop,
last_ref);
} else {
*last_ref = 1;
- ret = btrfs_del_item(trans, root, path);
+ ret = btrfs_del_item(trans, fs_info->extent_root, path);
}
return ret;
}
@@ -2075,6 +2113,7 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
u64 bytenr, u64 num_bytes, u64 parent,
u64 root_objectid, u64 owner, u64 offset)
{
+ int old_ref_mod, new_ref_mod;
int ret;
BUG_ON(owner < BTRFS_FIRST_FREE_OBJECTID &&
@@ -2082,15 +2121,21 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
if (owner < BTRFS_FIRST_FREE_OBJECTID) {
ret = btrfs_add_delayed_tree_ref(fs_info, trans, bytenr,
- num_bytes,
- parent, root_objectid, (int)owner,
- BTRFS_ADD_DELAYED_REF, NULL);
+ num_bytes, parent,
+ root_objectid, (int)owner,
+ BTRFS_ADD_DELAYED_REF, NULL,
+ &old_ref_mod, &new_ref_mod);
} else {
ret = btrfs_add_delayed_data_ref(fs_info, trans, bytenr,
- num_bytes, parent, root_objectid,
- owner, offset, 0,
- BTRFS_ADD_DELAYED_REF, NULL);
+ num_bytes, parent,
+ root_objectid, owner, offset,
+ 0, BTRFS_ADD_DELAYED_REF,
+ &old_ref_mod, &new_ref_mod);
}
+
+ if (ret == 0 && old_ref_mod < 0 && new_ref_mod >= 0)
+ add_pinned_bytes(fs_info, -num_bytes, owner, root_objectid);
+
return ret;
}
@@ -2117,9 +2162,9 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
path->reada = READA_FORWARD;
path->leave_spinning = 1;
/* this will setup the path even if it fails to insert the back ref */
- ret = insert_inline_extent_backref(trans, fs_info->extent_root, path,
- bytenr, num_bytes, parent,
- root_objectid, owner, offset,
+ ret = insert_inline_extent_backref(trans, fs_info, path, bytenr,
+ num_bytes, parent, root_objectid,
+ owner, offset,
refs_to_add, extent_op);
if ((ret < 0 && ret != -EAGAIN) || !ret)
goto out;
@@ -2143,9 +2188,8 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
path->reada = READA_FORWARD;
path->leave_spinning = 1;
/* now insert the actual backref */
- ret = insert_extent_backref(trans, fs_info->extent_root,
- path, bytenr, parent, root_objectid,
- owner, offset, refs_to_add);
+ ret = insert_extent_backref(trans, fs_info, path, bytenr, parent,
+ root_objectid, owner, offset, refs_to_add);
if (ret)
btrfs_abort_transaction(trans, ret);
out:
@@ -2290,8 +2334,7 @@ again:
item_size = btrfs_item_size_nr(leaf, path->slots[0]);
#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
if (item_size < sizeof(*ei)) {
- ret = convert_extent_item_v0(trans, fs_info->extent_root,
- path, (u64)-1, 0);
+ ret = convert_extent_item_v0(trans, fs_info, path, (u64)-1, 0);
if (ret < 0) {
err = ret;
goto out;
@@ -2396,6 +2439,16 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans,
head = btrfs_delayed_node_to_head(node);
trace_run_delayed_ref_head(fs_info, node, head, node->action);
+ if (head->total_ref_mod < 0) {
+ struct btrfs_block_group_cache *cache;
+
+ cache = btrfs_lookup_block_group(fs_info, node->bytenr);
+ ASSERT(cache);
+ percpu_counter_add(&cache->space_info->total_bytes_pinned,
+ -node->num_bytes);
+ btrfs_put_block_group(cache);
+ }
+
if (insert_reserved) {
btrfs_pin_extent(fs_info, node->bytenr,
node->num_bytes, 1);
@@ -2975,7 +3028,7 @@ again:
struct btrfs_delayed_ref_node *ref;
ref = &head->node;
- atomic_inc(&ref->refs);
+ refcount_inc(&ref->refs);
spin_unlock(&delayed_refs->lock);
/*
@@ -2998,7 +3051,6 @@ again:
goto again;
}
out:
- assert_qgroups_uptodate(trans);
trans->can_flush_pending_bgs = can_flush_pending_bgs;
return 0;
}
@@ -3028,8 +3080,7 @@ int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
return ret;
}
-static noinline int check_delayed_ref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+static noinline int check_delayed_ref(struct btrfs_root *root,
struct btrfs_path *path,
u64 objectid, u64 offset, u64 bytenr)
{
@@ -3037,18 +3088,23 @@ static noinline int check_delayed_ref(struct btrfs_trans_handle *trans,
struct btrfs_delayed_ref_node *ref;
struct btrfs_delayed_data_ref *data_ref;
struct btrfs_delayed_ref_root *delayed_refs;
+ struct btrfs_transaction *cur_trans;
int ret = 0;
- delayed_refs = &trans->transaction->delayed_refs;
+ cur_trans = root->fs_info->running_transaction;
+ if (!cur_trans)
+ return 0;
+
+ delayed_refs = &cur_trans->delayed_refs;
spin_lock(&delayed_refs->lock);
- head = btrfs_find_delayed_ref_head(trans, bytenr);
+ head = btrfs_find_delayed_ref_head(delayed_refs, bytenr);
if (!head) {
spin_unlock(&delayed_refs->lock);
return 0;
}
if (!mutex_trylock(&head->mutex)) {
- atomic_inc(&head->node.refs);
+ refcount_inc(&head->node.refs);
spin_unlock(&delayed_refs->lock);
btrfs_release_path(path);
@@ -3090,8 +3146,7 @@ static noinline int check_delayed_ref(struct btrfs_trans_handle *trans,
return ret;
}
-static noinline int check_committed_ref(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+static noinline int check_committed_ref(struct btrfs_root *root,
struct btrfs_path *path,
u64 objectid, u64 offset, u64 bytenr)
{
@@ -3162,9 +3217,8 @@ out:
return ret;
}
-int btrfs_cross_ref_exist(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- u64 objectid, u64 offset, u64 bytenr)
+int btrfs_cross_ref_exist(struct btrfs_root *root, u64 objectid, u64 offset,
+ u64 bytenr)
{
struct btrfs_path *path;
int ret;
@@ -3175,12 +3229,12 @@ int btrfs_cross_ref_exist(struct btrfs_trans_handle *trans,
return -ENOENT;
do {
- ret = check_committed_ref(trans, root, path, objectid,
+ ret = check_committed_ref(root, path, objectid,
offset, bytenr);
if (ret && ret != -ENOENT)
goto out;
- ret2 = check_delayed_ref(trans, root, path, objectid,
+ ret2 = check_delayed_ref(root, path, objectid,
offset, bytenr);
} while (ret2 == -EAGAIN);
@@ -3348,6 +3402,7 @@ static int cache_save_setup(struct btrfs_block_group_cache *block_group,
struct btrfs_fs_info *fs_info = block_group->fs_info;
struct btrfs_root *root = fs_info->tree_root;
struct inode *inode = NULL;
+ struct extent_changeset *data_reserved = NULL;
u64 alloc_hint = 0;
int dcs = BTRFS_DC_ERROR;
u64 num_pages = 0;
@@ -3368,7 +3423,7 @@ static int cache_save_setup(struct btrfs_block_group_cache *block_group,
if (trans->aborted)
return 0;
again:
- inode = lookup_free_space_inode(root, block_group, path);
+ inode = lookup_free_space_inode(fs_info, block_group, path);
if (IS_ERR(inode) && PTR_ERR(inode) != -ENOENT) {
ret = PTR_ERR(inode);
btrfs_release_path(path);
@@ -3382,7 +3437,8 @@ again:
if (block_group->ro)
goto out_free;
- ret = create_free_space_inode(root, trans, block_group, path);
+ ret = create_free_space_inode(fs_info, trans, block_group,
+ path);
if (ret)
goto out_free;
goto again;
@@ -3424,7 +3480,7 @@ again:
if (ret)
goto out_put;
- ret = btrfs_truncate_free_space_cache(root, trans, NULL, inode);
+ ret = btrfs_truncate_free_space_cache(trans, NULL, inode);
if (ret)
goto out_put;
}
@@ -3435,7 +3491,8 @@ again:
/*
* don't bother trying to write stuff out _if_
* a) we're not cached,
- * b) we're with nospace_cache mount option.
+ * b) we're with nospace_cache mount option,
+ * c) we're with v2 space_cache (FREE_SPACE_TREE).
*/
dcs = BTRFS_DC_WRITTEN;
spin_unlock(&block_group->lock);
@@ -3465,7 +3522,7 @@ again:
num_pages *= 16;
num_pages *= PAGE_SIZE;
- ret = btrfs_check_data_free_space(inode, 0, num_pages);
+ ret = btrfs_check_data_free_space(inode, &data_reserved, 0, num_pages);
if (ret)
goto out_put;
@@ -3496,6 +3553,7 @@ out:
block_group->disk_cache_state = dcs;
spin_unlock(&block_group->lock);
+ extent_changeset_free(data_reserved);
return ret;
}
@@ -3906,87 +3964,83 @@ static const char *alloc_name(u64 flags)
};
}
-static int update_space_info(struct btrfs_fs_info *info, u64 flags,
- u64 total_bytes, u64 bytes_used,
- u64 bytes_readonly,
- struct btrfs_space_info **space_info)
+static int create_space_info(struct btrfs_fs_info *info, u64 flags,
+ struct btrfs_space_info **new)
{
- struct btrfs_space_info *found;
+
+ struct btrfs_space_info *space_info;
int i;
- int factor;
int ret;
- if (flags & (BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID1 |
- BTRFS_BLOCK_GROUP_RAID10))
- factor = 2;
- else
- factor = 1;
-
- found = __find_space_info(info, flags);
- if (found) {
- spin_lock(&found->lock);
- found->total_bytes += total_bytes;
- found->disk_total += total_bytes * factor;
- found->bytes_used += bytes_used;
- found->disk_used += bytes_used * factor;
- found->bytes_readonly += bytes_readonly;
- if (total_bytes > 0)
- found->full = 0;
- space_info_add_new_bytes(info, found, total_bytes -
- bytes_used - bytes_readonly);
- spin_unlock(&found->lock);
- *space_info = found;
- return 0;
- }
- found = kzalloc(sizeof(*found), GFP_NOFS);
- if (!found)
+ space_info = kzalloc(sizeof(*space_info), GFP_NOFS);
+ if (!space_info)
return -ENOMEM;
- ret = percpu_counter_init(&found->total_bytes_pinned, 0, GFP_KERNEL);
+ ret = percpu_counter_init(&space_info->total_bytes_pinned, 0,
+ GFP_KERNEL);
if (ret) {
- kfree(found);
+ kfree(space_info);
return ret;
}
for (i = 0; i < BTRFS_NR_RAID_TYPES; i++)
- INIT_LIST_HEAD(&found->block_groups[i]);
- init_rwsem(&found->groups_sem);
- spin_lock_init(&found->lock);
- found->flags = flags & BTRFS_BLOCK_GROUP_TYPE_MASK;
- found->total_bytes = total_bytes;
- found->disk_total = total_bytes * factor;
- found->bytes_used = bytes_used;
- found->disk_used = bytes_used * factor;
- found->bytes_pinned = 0;
- found->bytes_reserved = 0;
- found->bytes_readonly = bytes_readonly;
- found->bytes_may_use = 0;
- found->full = 0;
- found->max_extent_size = 0;
- found->force_alloc = CHUNK_ALLOC_NO_FORCE;
- found->chunk_alloc = 0;
- found->flush = 0;
- init_waitqueue_head(&found->wait);
- INIT_LIST_HEAD(&found->ro_bgs);
- INIT_LIST_HEAD(&found->tickets);
- INIT_LIST_HEAD(&found->priority_tickets);
-
- ret = kobject_init_and_add(&found->kobj, &space_info_ktype,
+ INIT_LIST_HEAD(&space_info->block_groups[i]);
+ init_rwsem(&space_info->groups_sem);
+ spin_lock_init(&space_info->lock);
+ space_info->flags = flags & BTRFS_BLOCK_GROUP_TYPE_MASK;
+ space_info->force_alloc = CHUNK_ALLOC_NO_FORCE;
+ init_waitqueue_head(&space_info->wait);
+ INIT_LIST_HEAD(&space_info->ro_bgs);
+ INIT_LIST_HEAD(&space_info->tickets);
+ INIT_LIST_HEAD(&space_info->priority_tickets);
+
+ ret = kobject_init_and_add(&space_info->kobj, &space_info_ktype,
info->space_info_kobj, "%s",
- alloc_name(found->flags));
+ alloc_name(space_info->flags));
if (ret) {
- kfree(found);
+ percpu_counter_destroy(&space_info->total_bytes_pinned);
+ kfree(space_info);
return ret;
}
- *space_info = found;
- list_add_rcu(&found->list, &info->space_info);
+ *new = space_info;
+ list_add_rcu(&space_info->list, &info->space_info);
if (flags & BTRFS_BLOCK_GROUP_DATA)
- info->data_sinfo = found;
+ info->data_sinfo = space_info;
return ret;
}
+static void update_space_info(struct btrfs_fs_info *info, u64 flags,
+ u64 total_bytes, u64 bytes_used,
+ u64 bytes_readonly,
+ struct btrfs_space_info **space_info)
+{
+ struct btrfs_space_info *found;
+ int factor;
+
+ if (flags & (BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID1 |
+ BTRFS_BLOCK_GROUP_RAID10))
+ factor = 2;
+ else
+ factor = 1;
+
+ found = __find_space_info(info, flags);
+ ASSERT(found);
+ spin_lock(&found->lock);
+ found->total_bytes += total_bytes;
+ found->disk_total += total_bytes * factor;
+ found->bytes_used += bytes_used;
+ found->disk_used += bytes_used * factor;
+ found->bytes_readonly += bytes_readonly;
+ if (total_bytes > 0)
+ found->full = 0;
+ space_info_add_new_bytes(info, found, total_bytes -
+ bytes_used - bytes_readonly);
+ spin_unlock(&found->lock);
+ *space_info = found;
+}
+
static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
{
u64 extra_flags = chunk_to_extended(flags) &
@@ -4102,7 +4156,7 @@ static u64 get_alloc_profile(struct btrfs_fs_info *fs_info, u64 orig_flags)
return btrfs_reduce_alloc_profile(fs_info, flags);
}
-u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data)
+static u64 get_alloc_profile_by_root(struct btrfs_root *root, int data)
{
struct btrfs_fs_info *fs_info = root->fs_info;
u64 flags;
@@ -4119,10 +4173,34 @@ u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data)
return ret;
}
-int btrfs_alloc_data_chunk_ondemand(struct inode *inode, u64 bytes)
+u64 btrfs_data_alloc_profile(struct btrfs_fs_info *fs_info)
+{
+ return get_alloc_profile(fs_info, BTRFS_BLOCK_GROUP_DATA);
+}
+
+u64 btrfs_metadata_alloc_profile(struct btrfs_fs_info *fs_info)
+{
+ return get_alloc_profile(fs_info, BTRFS_BLOCK_GROUP_METADATA);
+}
+
+u64 btrfs_system_alloc_profile(struct btrfs_fs_info *fs_info)
+{
+ return get_alloc_profile(fs_info, BTRFS_BLOCK_GROUP_SYSTEM);
+}
+
+static u64 btrfs_space_info_used(struct btrfs_space_info *s_info,
+ bool may_use_included)
+{
+ ASSERT(s_info);
+ return s_info->bytes_used + s_info->bytes_reserved +
+ s_info->bytes_pinned + s_info->bytes_readonly +
+ (may_use_included ? s_info->bytes_may_use : 0);
+}
+
+int btrfs_alloc_data_chunk_ondemand(struct btrfs_inode *inode, u64 bytes)
{
struct btrfs_space_info *data_sinfo;
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_root *root = inode->root;
struct btrfs_fs_info *fs_info = root->fs_info;
u64 used;
int ret = 0;
@@ -4144,9 +4222,7 @@ int btrfs_alloc_data_chunk_ondemand(struct inode *inode, u64 bytes)
again:
/* make sure we have enough space to handle the data first */
spin_lock(&data_sinfo->lock);
- used = data_sinfo->bytes_used + data_sinfo->bytes_reserved +
- data_sinfo->bytes_pinned + data_sinfo->bytes_readonly +
- data_sinfo->bytes_may_use;
+ used = btrfs_space_info_used(data_sinfo, true);
if (used + bytes > data_sinfo->total_bytes) {
struct btrfs_trans_handle *trans;
@@ -4161,7 +4237,7 @@ again:
data_sinfo->force_alloc = CHUNK_ALLOC_FORCE;
spin_unlock(&data_sinfo->lock);
alloc:
- alloc_target = btrfs_get_alloc_profile(root, 1);
+ alloc_target = btrfs_data_alloc_profile(fs_info);
/*
* It is ugly that we don't call nolock join
* transaction for the free space inode case here.
@@ -4212,7 +4288,7 @@ commit_trans:
if (need_commit > 0) {
btrfs_start_delalloc_roots(fs_info, 0, -1);
- btrfs_wait_ordered_roots(fs_info, -1, 0,
+ btrfs_wait_ordered_roots(fs_info, U64_MAX, 0,
(u64)-1);
}
@@ -4252,12 +4328,8 @@ commit_trans:
return ret;
}
-/*
- * New check_data_free_space() with ability for precious data reservation
- * Will replace old btrfs_check_data_free_space(), but for patch split,
- * add a new function first and then replace it.
- */
-int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len)
+int btrfs_check_data_free_space(struct inode *inode,
+ struct extent_changeset **reserved, u64 start, u64 len)
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
int ret;
@@ -4267,14 +4339,16 @@ int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len)
round_down(start, fs_info->sectorsize);
start = round_down(start, fs_info->sectorsize);
- ret = btrfs_alloc_data_chunk_ondemand(inode, len);
+ ret = btrfs_alloc_data_chunk_ondemand(BTRFS_I(inode), len);
if (ret < 0)
return ret;
/* Use new btrfs_qgroup_reserve_data to reserve precious data space. */
- ret = btrfs_qgroup_reserve_data(inode, start, len);
- if (ret)
+ ret = btrfs_qgroup_reserve_data(inode, reserved, start, len);
+ if (ret < 0)
btrfs_free_reserved_data_space_noquota(inode, start, len);
+ else
+ ret = 0;
return ret;
}
@@ -4315,7 +4389,8 @@ void btrfs_free_reserved_data_space_noquota(struct inode *inode, u64 start,
* This one will handle the per-inode data rsv map for accurate reserved
* space framework.
*/
-void btrfs_free_reserved_data_space(struct inode *inode, u64 start, u64 len)
+void btrfs_free_reserved_data_space(struct inode *inode,
+ struct extent_changeset *reserved, u64 start, u64 len)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -4325,7 +4400,7 @@ void btrfs_free_reserved_data_space(struct inode *inode, u64 start, u64 len)
start = round_down(start, root->fs_info->sectorsize);
btrfs_free_reserved_data_space_noquota(inode, start, len);
- btrfs_qgroup_free_data(inode, start, len);
+ btrfs_qgroup_free_data(inode, reserved, start, len);
}
static void force_metadata_allocation(struct btrfs_fs_info *info)
@@ -4421,9 +4496,7 @@ void check_system_chunk(struct btrfs_trans_handle *trans,
info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_SYSTEM);
spin_lock(&info->lock);
- left = info->total_bytes - info->bytes_used - info->bytes_pinned -
- info->bytes_reserved - info->bytes_readonly -
- info->bytes_may_use;
+ left = info->total_bytes - btrfs_space_info_used(info, true);
spin_unlock(&info->lock);
num_devs = get_profile_num_devs(fs_info, type);
@@ -4439,9 +4512,8 @@ void check_system_chunk(struct btrfs_trans_handle *trans,
}
if (left < thresh) {
- u64 flags;
+ u64 flags = btrfs_system_alloc_profile(fs_info);
- flags = btrfs_get_alloc_profile(fs_info->chunk_root, 0);
/*
* Ignore failure to create system chunk. We might end up not
* needing it, as we might not need to COW all nodes/leafs from
@@ -4482,10 +4554,10 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans,
space_info = __find_space_info(fs_info, flags);
if (!space_info) {
- ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
- BUG_ON(ret); /* -ENOMEM */
+ ret = create_space_info(fs_info, flags, &space_info);
+ if (ret)
+ return ret;
}
- BUG_ON(!space_info); /* Logic error */
again:
spin_lock(&space_info->lock);
@@ -4590,11 +4662,11 @@ out:
return ret;
}
-static int can_overcommit(struct btrfs_root *root,
+static int can_overcommit(struct btrfs_fs_info *fs_info,
struct btrfs_space_info *space_info, u64 bytes,
- enum btrfs_reserve_flush_enum flush)
+ enum btrfs_reserve_flush_enum flush,
+ bool system_chunk)
{
- struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
u64 profile;
u64 space_size;
@@ -4605,9 +4677,12 @@ static int can_overcommit(struct btrfs_root *root,
if (space_info->flags & BTRFS_BLOCK_GROUP_DATA)
return 0;
- profile = btrfs_get_alloc_profile(root, 0);
- used = space_info->bytes_used + space_info->bytes_reserved +
- space_info->bytes_pinned + space_info->bytes_readonly;
+ if (system_chunk)
+ profile = btrfs_system_alloc_profile(fs_info);
+ else
+ profile = btrfs_metadata_alloc_profile(fs_info);
+
+ used = btrfs_space_info_used(space_info, false);
/*
* We only want to allow over committing if we have lots of actual space
@@ -4623,9 +4698,7 @@ static int can_overcommit(struct btrfs_root *root,
used += space_info->bytes_may_use;
- spin_lock(&fs_info->free_chunk_lock);
- avail = fs_info->free_chunk_space;
- spin_unlock(&fs_info->free_chunk_lock);
+ avail = atomic64_read(&fs_info->free_chunk_space);
/*
* If we have dup, raid1 or raid10 then only half of the free
@@ -4675,14 +4748,14 @@ static void btrfs_writeback_inodes_sb_nr(struct btrfs_fs_info *fs_info,
}
}
-static inline int calc_reclaim_items_nr(struct btrfs_fs_info *fs_info,
+static inline u64 calc_reclaim_items_nr(struct btrfs_fs_info *fs_info,
u64 to_reclaim)
{
u64 bytes;
- int nr;
+ u64 nr;
bytes = btrfs_calc_trans_metadata_size(fs_info, 1);
- nr = (int)div64_u64(to_reclaim, bytes);
+ nr = div64_u64(to_reclaim, bytes);
if (!nr)
nr = 1;
return nr;
@@ -4693,24 +4766,23 @@ static inline int calc_reclaim_items_nr(struct btrfs_fs_info *fs_info,
/*
* shrink metadata reservation for delalloc
*/
-static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
- bool wait_ordered)
+static void shrink_delalloc(struct btrfs_fs_info *fs_info, u64 to_reclaim,
+ u64 orig, bool wait_ordered)
{
- struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_block_rsv *block_rsv;
struct btrfs_space_info *space_info;
struct btrfs_trans_handle *trans;
u64 delalloc_bytes;
u64 max_reclaim;
+ u64 items;
long time_left;
unsigned long nr_pages;
int loops;
- int items;
enum btrfs_reserve_flush_enum flush;
/* Calc the number of the pages we need flush for space reservation */
items = calc_reclaim_items_nr(fs_info, to_reclaim);
- to_reclaim = (u64)items * EXTENT_SIZE_PER_ITEM;
+ to_reclaim = items * EXTENT_SIZE_PER_ITEM;
trans = (struct btrfs_trans_handle *)current->journal_info;
block_rsv = &fs_info->delalloc_block_rsv;
@@ -4753,7 +4825,7 @@ skip_async:
else
flush = BTRFS_RESERVE_NO_FLUSH;
spin_lock(&space_info->lock);
- if (can_overcommit(root, space_info, orig, flush)) {
+ if (can_overcommit(fs_info, space_info, orig, flush, false)) {
spin_unlock(&space_info->lock);
break;
}
@@ -4787,11 +4859,10 @@ skip_async:
* get us somewhere and then commit the transaction if it does. Otherwise it
* will return -ENOSPC.
*/
-static int may_commit_transaction(struct btrfs_root *root,
+static int may_commit_transaction(struct btrfs_fs_info *fs_info,
struct btrfs_space_info *space_info,
u64 bytes, int force)
{
- struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_block_rsv *delayed_rsv = &fs_info->delayed_block_rsv;
struct btrfs_trans_handle *trans;
@@ -4816,14 +4887,14 @@ static int may_commit_transaction(struct btrfs_root *root,
spin_lock(&delayed_rsv->lock);
if (percpu_counter_compare(&space_info->total_bytes_pinned,
- bytes - delayed_rsv->size) >= 0) {
+ bytes - delayed_rsv->size) < 0) {
spin_unlock(&delayed_rsv->lock);
return -ENOSPC;
}
spin_unlock(&delayed_rsv->lock);
commit:
- trans = btrfs_join_transaction(root);
+ trans = btrfs_join_transaction(fs_info->extent_root);
if (IS_ERR(trans))
return -ENOSPC;
@@ -4837,11 +4908,11 @@ struct reserve_ticket {
wait_queue_head_t wait;
};
-static int flush_space(struct btrfs_root *root,
+static int flush_space(struct btrfs_fs_info *fs_info,
struct btrfs_space_info *space_info, u64 num_bytes,
u64 orig_bytes, int state)
{
- struct btrfs_fs_info *fs_info = root->fs_info;
+ struct btrfs_root *root = fs_info->extent_root;
struct btrfs_trans_handle *trans;
int nr;
int ret = 0;
@@ -4864,7 +4935,7 @@ static int flush_space(struct btrfs_root *root,
break;
case FLUSH_DELALLOC:
case FLUSH_DELALLOC_WAIT:
- shrink_delalloc(root, num_bytes * 2, orig_bytes,
+ shrink_delalloc(fs_info, num_bytes * 2, orig_bytes,
state == FLUSH_DELALLOC_WAIT);
break;
case ALLOC_CHUNK:
@@ -4874,14 +4945,15 @@ static int flush_space(struct btrfs_root *root,
break;
}
ret = do_chunk_alloc(trans, fs_info,
- btrfs_get_alloc_profile(root, 0),
+ btrfs_metadata_alloc_profile(fs_info),
CHUNK_ALLOC_NO_FORCE);
btrfs_end_transaction(trans);
if (ret > 0 || ret == -ENOSPC)
ret = 0;
break;
case COMMIT_TRANS:
- ret = may_commit_transaction(root, space_info, orig_bytes, 0);
+ ret = may_commit_transaction(fs_info, space_info,
+ orig_bytes, 0);
break;
default:
ret = -ENOSPC;
@@ -4894,8 +4966,9 @@ static int flush_space(struct btrfs_root *root,
}
static inline u64
-btrfs_calc_reclaim_metadata_size(struct btrfs_root *root,
- struct btrfs_space_info *space_info)
+btrfs_calc_reclaim_metadata_size(struct btrfs_fs_info *fs_info,
+ struct btrfs_space_info *space_info,
+ bool system_chunk)
{
struct reserve_ticket *ticket;
u64 used;
@@ -4910,14 +4983,14 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_root *root,
return to_reclaim;
to_reclaim = min_t(u64, num_online_cpus() * SZ_1M, SZ_16M);
- if (can_overcommit(root, space_info, to_reclaim,
- BTRFS_RESERVE_FLUSH_ALL))
+ if (can_overcommit(fs_info, space_info, to_reclaim,
+ BTRFS_RESERVE_FLUSH_ALL, system_chunk))
return 0;
- used = space_info->bytes_used + space_info->bytes_reserved +
- space_info->bytes_pinned + space_info->bytes_readonly +
- space_info->bytes_may_use;
- if (can_overcommit(root, space_info, SZ_1M, BTRFS_RESERVE_FLUSH_ALL))
+ used = btrfs_space_info_used(space_info, true);
+
+ if (can_overcommit(fs_info, space_info, SZ_1M,
+ BTRFS_RESERVE_FLUSH_ALL, system_chunk))
expected = div_factor_fine(space_info->total_bytes, 95);
else
expected = div_factor_fine(space_info->total_bytes, 90);
@@ -4931,17 +5004,18 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_root *root,
return to_reclaim;
}
-static inline int need_do_async_reclaim(struct btrfs_space_info *space_info,
- struct btrfs_root *root, u64 used)
+static inline int need_do_async_reclaim(struct btrfs_fs_info *fs_info,
+ struct btrfs_space_info *space_info,
+ u64 used, bool system_chunk)
{
- struct btrfs_fs_info *fs_info = root->fs_info;
u64 thresh = div_factor_fine(space_info->total_bytes, 98);
/* If we're just plain full then async reclaim just slows us down. */
if ((space_info->bytes_used + space_info->bytes_reserved) >= thresh)
return 0;
- if (!btrfs_calc_reclaim_metadata_size(root, space_info))
+ if (!btrfs_calc_reclaim_metadata_size(fs_info, space_info,
+ system_chunk))
return 0;
return (used >= thresh && !btrfs_fs_closing(fs_info) &&
@@ -4978,8 +5052,8 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
spin_lock(&space_info->lock);
- to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info->fs_root,
- space_info);
+ to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info, space_info,
+ false);
if (!to_reclaim) {
space_info->flush = 0;
spin_unlock(&space_info->lock);
@@ -4993,16 +5067,17 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
struct reserve_ticket *ticket;
int ret;
- ret = flush_space(fs_info->fs_root, space_info, to_reclaim,
- to_reclaim, flush_state);
+ ret = flush_space(fs_info, space_info, to_reclaim, to_reclaim,
+ flush_state);
spin_lock(&space_info->lock);
if (list_empty(&space_info->tickets)) {
space_info->flush = 0;
spin_unlock(&space_info->lock);
return;
}
- to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info->fs_root,
- space_info);
+ to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info,
+ space_info,
+ false);
ticket = list_first_entry(&space_info->tickets,
struct reserve_ticket, list);
if (last_tickets_id == space_info->tickets_id) {
@@ -5040,8 +5115,8 @@ static void priority_reclaim_metadata_space(struct btrfs_fs_info *fs_info,
int flush_state = FLUSH_DELAYED_ITEMS_NR;
spin_lock(&space_info->lock);
- to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info->fs_root,
- space_info);
+ to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info, space_info,
+ false);
if (!to_reclaim) {
spin_unlock(&space_info->lock);
return;
@@ -5049,8 +5124,8 @@ static void priority_reclaim_metadata_space(struct btrfs_fs_info *fs_info,
spin_unlock(&space_info->lock);
do {
- flush_space(fs_info->fs_root, space_info, to_reclaim,
- to_reclaim, flush_state);
+ flush_space(fs_info, space_info, to_reclaim, to_reclaim,
+ flush_state);
flush_state++;
spin_lock(&space_info->lock);
if (ticket->bytes == 0) {
@@ -5120,12 +5195,12 @@ static int wait_reserve_ticket(struct btrfs_fs_info *fs_info,
* regain reservations will be made and this will fail if there is not enough
* space already.
*/
-static int __reserve_metadata_bytes(struct btrfs_root *root,
+static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
struct btrfs_space_info *space_info,
u64 orig_bytes,
- enum btrfs_reserve_flush_enum flush)
+ enum btrfs_reserve_flush_enum flush,
+ bool system_chunk)
{
- struct btrfs_fs_info *fs_info = root->fs_info;
struct reserve_ticket ticket;
u64 used;
int ret = 0;
@@ -5135,9 +5210,7 @@ static int __reserve_metadata_bytes(struct btrfs_root *root,
spin_lock(&space_info->lock);
ret = -ENOSPC;
- used = space_info->bytes_used + space_info->bytes_reserved +
- space_info->bytes_pinned + space_info->bytes_readonly +
- space_info->bytes_may_use;
+ used = btrfs_space_info_used(space_info, true);
/*
* If we have enough space then hooray, make our reservation and carry
@@ -5149,7 +5222,8 @@ static int __reserve_metadata_bytes(struct btrfs_root *root,
trace_btrfs_space_reservation(fs_info, "space_info",
space_info->flags, orig_bytes, 1);
ret = 0;
- } else if (can_overcommit(root, space_info, orig_bytes, flush)) {
+ } else if (can_overcommit(fs_info, space_info, orig_bytes, flush,
+ system_chunk)) {
space_info->bytes_may_use += orig_bytes;
trace_btrfs_space_reservation(fs_info, "space_info",
space_info->flags, orig_bytes, 1);
@@ -5176,7 +5250,7 @@ static int __reserve_metadata_bytes(struct btrfs_root *root,
orig_bytes, flush,
"enospc");
queue_work(system_unbound_wq,
- &root->fs_info->async_reclaim_work);
+ &fs_info->async_reclaim_work);
}
} else {
list_add_tail(&ticket.list,
@@ -5190,7 +5264,8 @@ static int __reserve_metadata_bytes(struct btrfs_root *root,
* the async reclaim as we will panic.
*/
if (!test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags) &&
- need_do_async_reclaim(space_info, root, used) &&
+ need_do_async_reclaim(fs_info, space_info,
+ used, system_chunk) &&
!work_busy(&fs_info->async_reclaim_work)) {
trace_btrfs_trigger_flush(fs_info, space_info->flags,
orig_bytes, flush, "preempt");
@@ -5248,9 +5323,10 @@ static int reserve_metadata_bytes(struct btrfs_root *root,
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
int ret;
+ bool system_chunk = (root == fs_info->chunk_root);
- ret = __reserve_metadata_bytes(root, block_rsv->space_info, orig_bytes,
- flush);
+ ret = __reserve_metadata_bytes(fs_info, block_rsv->space_info,
+ orig_bytes, flush, system_chunk);
if (ret == -ENOSPC &&
unlikely(root->orphan_cleanup_state == ORPHAN_CLEANUP_STARTED)) {
if (block_rsv != global_rsv &&
@@ -5359,9 +5435,7 @@ static void space_info_add_old_bytes(struct btrfs_fs_info *fs_info,
* overcommit, and if we can't then we just need to free up our space
* and not satisfy any requests.
*/
- used = space_info->bytes_used + space_info->bytes_reserved +
- space_info->bytes_pinned + space_info->bytes_readonly +
- space_info->bytes_may_use;
+ used = btrfs_space_info_used(space_info, true);
if (used - num_bytes >= space_info->total_bytes)
check_overcommit = true;
again:
@@ -5373,8 +5447,7 @@ again:
* adding the ticket space would be a double count.
*/
if (check_overcommit &&
- !can_overcommit(fs_info->extent_root, space_info, 0,
- flush))
+ !can_overcommit(fs_info, space_info, 0, flush, false))
break;
if (num_bytes >= ticket->bytes) {
list_del_init(&ticket->list);
@@ -5630,9 +5703,7 @@ static void update_global_block_rsv(struct btrfs_fs_info *fs_info)
block_rsv->size = min_t(u64, num_bytes, SZ_512M);
if (block_rsv->reserved < block_rsv->size) {
- num_bytes = sinfo->bytes_used + sinfo->bytes_pinned +
- sinfo->bytes_reserved + sinfo->bytes_readonly +
- sinfo->bytes_may_use;
+ num_bytes = btrfs_space_info_used(sinfo, true);
if (sinfo->total_bytes > num_bytes) {
num_bytes = sinfo->total_bytes - num_bytes;
num_bytes = min(num_bytes,
@@ -5735,10 +5806,10 @@ void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans)
/* Can only return 0 or -ENOSPC */
int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
- struct inode *inode)
+ struct btrfs_inode *inode)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = inode->root;
/*
* We always use trans->block_rsv here as we will have reserved space
* for our orphan when starting the transaction, using get_block_rsv()
@@ -5755,19 +5826,19 @@ int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
*/
u64 num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1);
- trace_btrfs_space_reservation(fs_info, "orphan",
- btrfs_ino(inode), num_bytes, 1);
+ trace_btrfs_space_reservation(fs_info, "orphan", btrfs_ino(inode),
+ num_bytes, 1);
return btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
}
-void btrfs_orphan_release_metadata(struct inode *inode)
+void btrfs_orphan_release_metadata(struct btrfs_inode *inode)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = inode->root;
u64 num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1);
- trace_btrfs_space_reservation(fs_info, "orphan",
- btrfs_ino(inode), num_bytes, 0);
+ trace_btrfs_space_reservation(fs_info, "orphan", btrfs_ino(inode),
+ num_bytes, 0);
btrfs_block_rsv_release(fs_info, root->orphan_block_rsv, num_bytes);
}
@@ -5799,7 +5870,7 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)) {
/* One for parent inode, two for dir entries */
num_bytes = 3 * fs_info->nodesize;
- ret = btrfs_qgroup_reserve_meta(root, num_bytes);
+ ret = btrfs_qgroup_reserve_meta(root, num_bytes, true);
if (ret)
return ret;
} else {
@@ -5824,8 +5895,7 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
}
void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info,
- struct btrfs_block_rsv *rsv,
- u64 qgroup_reserved)
+ struct btrfs_block_rsv *rsv)
{
btrfs_block_rsv_release(fs_info, rsv, (u64)-1);
}
@@ -5840,35 +5910,32 @@ void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info,
* reserved extents that need to be freed. This must be called with
* BTRFS_I(inode)->lock held.
*/
-static unsigned drop_outstanding_extent(struct inode *inode, u64 num_bytes)
+static unsigned drop_outstanding_extent(struct btrfs_inode *inode,
+ u64 num_bytes)
{
unsigned drop_inode_space = 0;
unsigned dropped_extents = 0;
- unsigned num_extents = 0;
+ unsigned num_extents;
- num_extents = (unsigned)div64_u64(num_bytes +
- BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE);
+ num_extents = count_max_extents(num_bytes);
ASSERT(num_extents);
- ASSERT(BTRFS_I(inode)->outstanding_extents >= num_extents);
- BTRFS_I(inode)->outstanding_extents -= num_extents;
+ ASSERT(inode->outstanding_extents >= num_extents);
+ inode->outstanding_extents -= num_extents;
- if (BTRFS_I(inode)->outstanding_extents == 0 &&
+ if (inode->outstanding_extents == 0 &&
test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
- &BTRFS_I(inode)->runtime_flags))
+ &inode->runtime_flags))
drop_inode_space = 1;
/*
* If we have more or the same amount of outstanding extents than we have
* reserved then we need to leave the reserved extents count alone.
*/
- if (BTRFS_I(inode)->outstanding_extents >=
- BTRFS_I(inode)->reserved_extents)
+ if (inode->outstanding_extents >= inode->reserved_extents)
return drop_inode_space;
- dropped_extents = BTRFS_I(inode)->reserved_extents -
- BTRFS_I(inode)->outstanding_extents;
- BTRFS_I(inode)->reserved_extents -= dropped_extents;
+ dropped_extents = inode->reserved_extents - inode->outstanding_extents;
+ inode->reserved_extents -= dropped_extents;
return dropped_extents + drop_inode_space;
}
@@ -5890,24 +5957,21 @@ static unsigned drop_outstanding_extent(struct inode *inode, u64 num_bytes)
*
* This must be called with BTRFS_I(inode)->lock held.
*/
-static u64 calc_csum_metadata_size(struct inode *inode, u64 num_bytes,
+static u64 calc_csum_metadata_size(struct btrfs_inode *inode, u64 num_bytes,
int reserve)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
u64 old_csums, num_csums;
- if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM &&
- BTRFS_I(inode)->csum_bytes == 0)
+ if (inode->flags & BTRFS_INODE_NODATASUM && inode->csum_bytes == 0)
return 0;
- old_csums = btrfs_csum_bytes_to_leaves(fs_info,
- BTRFS_I(inode)->csum_bytes);
+ old_csums = btrfs_csum_bytes_to_leaves(fs_info, inode->csum_bytes);
if (reserve)
- BTRFS_I(inode)->csum_bytes += num_bytes;
+ inode->csum_bytes += num_bytes;
else
- BTRFS_I(inode)->csum_bytes -= num_bytes;
- num_csums = btrfs_csum_bytes_to_leaves(fs_info,
- BTRFS_I(inode)->csum_bytes);
+ inode->csum_bytes -= num_bytes;
+ num_csums = btrfs_csum_bytes_to_leaves(fs_info, inode->csum_bytes);
/* No change, no need to reserve more */
if (old_csums == num_csums)
@@ -5920,14 +5984,14 @@ static u64 calc_csum_metadata_size(struct inode *inode, u64 num_bytes,
return btrfs_calc_trans_metadata_size(fs_info, old_csums - num_csums);
}
-int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
+int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = inode->root;
struct btrfs_block_rsv *block_rsv = &fs_info->delalloc_block_rsv;
u64 to_reserve = 0;
u64 csum_bytes;
- unsigned nr_extents = 0;
+ unsigned nr_extents;
enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_FLUSH_ALL;
int ret = 0;
bool delalloc_lock = true;
@@ -5955,31 +6019,28 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
schedule_timeout(1);
if (delalloc_lock)
- mutex_lock(&BTRFS_I(inode)->delalloc_mutex);
+ mutex_lock(&inode->delalloc_mutex);
num_bytes = ALIGN(num_bytes, fs_info->sectorsize);
- spin_lock(&BTRFS_I(inode)->lock);
- nr_extents = (unsigned)div64_u64(num_bytes +
- BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE);
- BTRFS_I(inode)->outstanding_extents += nr_extents;
+ spin_lock(&inode->lock);
+ nr_extents = count_max_extents(num_bytes);
+ inode->outstanding_extents += nr_extents;
nr_extents = 0;
- if (BTRFS_I(inode)->outstanding_extents >
- BTRFS_I(inode)->reserved_extents)
- nr_extents += BTRFS_I(inode)->outstanding_extents -
- BTRFS_I(inode)->reserved_extents;
+ if (inode->outstanding_extents > inode->reserved_extents)
+ nr_extents += inode->outstanding_extents -
+ inode->reserved_extents;
/* We always want to reserve a slot for updating the inode. */
to_reserve = btrfs_calc_trans_metadata_size(fs_info, nr_extents + 1);
to_reserve += calc_csum_metadata_size(inode, num_bytes, 1);
- csum_bytes = BTRFS_I(inode)->csum_bytes;
- spin_unlock(&BTRFS_I(inode)->lock);
+ csum_bytes = inode->csum_bytes;
+ spin_unlock(&inode->lock);
if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)) {
ret = btrfs_qgroup_reserve_meta(root,
- nr_extents * fs_info->nodesize);
+ nr_extents * fs_info->nodesize, true);
if (ret)
goto out_fail;
}
@@ -5991,17 +6052,17 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
goto out_fail;
}
- spin_lock(&BTRFS_I(inode)->lock);
+ spin_lock(&inode->lock);
if (test_and_set_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
- &BTRFS_I(inode)->runtime_flags)) {
+ &inode->runtime_flags)) {
to_reserve -= btrfs_calc_trans_metadata_size(fs_info, 1);
release_extra = true;
}
- BTRFS_I(inode)->reserved_extents += nr_extents;
- spin_unlock(&BTRFS_I(inode)->lock);
+ inode->reserved_extents += nr_extents;
+ spin_unlock(&inode->lock);
if (delalloc_lock)
- mutex_unlock(&BTRFS_I(inode)->delalloc_mutex);
+ mutex_unlock(&inode->delalloc_mutex);
if (to_reserve)
trace_btrfs_space_reservation(fs_info, "delalloc",
@@ -6012,17 +6073,17 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
return 0;
out_fail:
- spin_lock(&BTRFS_I(inode)->lock);
+ spin_lock(&inode->lock);
dropped = drop_outstanding_extent(inode, num_bytes);
/*
* If the inodes csum_bytes is the same as the original
* csum_bytes then we know we haven't raced with any free()ers
* so we can just reduce our inodes csum bytes and carry on.
*/
- if (BTRFS_I(inode)->csum_bytes == csum_bytes) {
+ if (inode->csum_bytes == csum_bytes) {
calc_csum_metadata_size(inode, num_bytes, 0);
} else {
- u64 orig_csum_bytes = BTRFS_I(inode)->csum_bytes;
+ u64 orig_csum_bytes = inode->csum_bytes;
u64 bytes;
/*
@@ -6033,8 +6094,8 @@ out_fail:
* number of bytes that were freed while we were trying our
* reservation.
*/
- bytes = csum_bytes - BTRFS_I(inode)->csum_bytes;
- BTRFS_I(inode)->csum_bytes = csum_bytes;
+ bytes = csum_bytes - inode->csum_bytes;
+ inode->csum_bytes = csum_bytes;
to_free = calc_csum_metadata_size(inode, bytes, 0);
@@ -6043,7 +6104,7 @@ out_fail:
* been making this reservation and our ->csum_bytes were not
* artificially inflated.
*/
- BTRFS_I(inode)->csum_bytes = csum_bytes - num_bytes;
+ inode->csum_bytes = csum_bytes - num_bytes;
bytes = csum_bytes - orig_csum_bytes;
bytes = calc_csum_metadata_size(inode, bytes, 0);
@@ -6055,13 +6116,13 @@ out_fail:
* need to do anything, the other free-ers did the correct
* thing.
*/
- BTRFS_I(inode)->csum_bytes = orig_csum_bytes - num_bytes;
+ inode->csum_bytes = orig_csum_bytes - num_bytes;
if (bytes > to_free)
to_free = bytes - to_free;
else
to_free = 0;
}
- spin_unlock(&BTRFS_I(inode)->lock);
+ spin_unlock(&inode->lock);
if (dropped)
to_free += btrfs_calc_trans_metadata_size(fs_info, dropped);
@@ -6071,7 +6132,7 @@ out_fail:
btrfs_ino(inode), to_free, 0);
}
if (delalloc_lock)
- mutex_unlock(&BTRFS_I(inode)->delalloc_mutex);
+ mutex_unlock(&inode->delalloc_mutex);
return ret;
}
@@ -6084,27 +6145,27 @@ out_fail:
* once we complete IO for a given set of bytes to release their metadata
* reservations.
*/
-void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes)
+void btrfs_delalloc_release_metadata(struct btrfs_inode *inode, u64 num_bytes)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
u64 to_free = 0;
unsigned dropped;
num_bytes = ALIGN(num_bytes, fs_info->sectorsize);
- spin_lock(&BTRFS_I(inode)->lock);
+ spin_lock(&inode->lock);
dropped = drop_outstanding_extent(inode, num_bytes);
if (num_bytes)
to_free = calc_csum_metadata_size(inode, num_bytes, 0);
- spin_unlock(&BTRFS_I(inode)->lock);
+ spin_unlock(&inode->lock);
if (dropped > 0)
to_free += btrfs_calc_trans_metadata_size(fs_info, dropped);
if (btrfs_is_testing(fs_info))
return;
- trace_btrfs_space_reservation(fs_info, "delalloc",
- btrfs_ino(inode), to_free, 0);
+ trace_btrfs_space_reservation(fs_info, "delalloc", btrfs_ino(inode),
+ to_free, 0);
btrfs_block_rsv_release(fs_info, &fs_info->delalloc_block_rsv, to_free);
}
@@ -6115,6 +6176,8 @@ void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes)
* @inode: inode we're writing to
* @start: start range we are writing to
* @len: how long the range we are writing to
+ * @reserved: mandatory parameter, record actually reserved qgroup ranges of
+ * current reservation.
*
* This will do the following things
*
@@ -6132,16 +6195,17 @@ void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes)
* Return 0 for success
* Return <0 for error(-ENOSPC or -EQUOT)
*/
-int btrfs_delalloc_reserve_space(struct inode *inode, u64 start, u64 len)
+int btrfs_delalloc_reserve_space(struct inode *inode,
+ struct extent_changeset **reserved, u64 start, u64 len)
{
int ret;
- ret = btrfs_check_data_free_space(inode, start, len);
+ ret = btrfs_check_data_free_space(inode, reserved, start, len);
if (ret < 0)
return ret;
- ret = btrfs_delalloc_reserve_metadata(inode, len);
+ ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), len);
if (ret < 0)
- btrfs_free_reserved_data_space(inode, start, len);
+ btrfs_free_reserved_data_space(inode, *reserved, start, len);
return ret;
}
@@ -6160,10 +6224,11 @@ int btrfs_delalloc_reserve_space(struct inode *inode, u64 start, u64 len)
* list if there are no delalloc bytes left.
* Also it will handle the qgroup reserved space.
*/
-void btrfs_delalloc_release_space(struct inode *inode, u64 start, u64 len)
+void btrfs_delalloc_release_space(struct inode *inode,
+ struct extent_changeset *reserved, u64 start, u64 len)
{
- btrfs_delalloc_release_metadata(inode, len);
- btrfs_free_reserved_data_space(inode, start, len);
+ btrfs_delalloc_release_metadata(BTRFS_I(inode), len);
+ btrfs_free_reserved_data_space(inode, reserved, start, len);
}
static int update_block_group(struct btrfs_trans_handle *trans,
@@ -6239,6 +6304,8 @@ static int update_block_group(struct btrfs_trans_handle *trans,
trace_btrfs_space_reservation(info, "pinned",
cache->space_info->flags,
num_bytes, 1);
+ percpu_counter_add(&cache->space_info->total_bytes_pinned,
+ num_bytes);
set_extent_dirty(info->pinned_extents,
bytenr, bytenr + num_bytes - 1,
GFP_NOFS | __GFP_NOFAIL);
@@ -6315,6 +6382,7 @@ static int pin_down_extent(struct btrfs_fs_info *fs_info,
trace_btrfs_space_reservation(fs_info, "pinned",
cache->space_info->flags, num_bytes, 1);
+ percpu_counter_add(&cache->space_info->total_bytes_pinned, num_bytes);
set_extent_dirty(fs_info->pinned_extents, bytenr,
bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL);
return 0;
@@ -6561,8 +6629,7 @@ static int btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache,
spin_unlock(&space_info->lock);
return ret;
}
-void btrfs_prepare_extent_commit(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info)
+void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info)
{
struct btrfs_caching_control *next;
struct btrfs_caching_control *caching_ctl;
@@ -6786,27 +6853,6 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans,
return 0;
}
-static void add_pinned_bytes(struct btrfs_fs_info *fs_info, u64 num_bytes,
- u64 owner, u64 root_objectid)
-{
- struct btrfs_space_info *space_info;
- u64 flags;
-
- if (owner < BTRFS_FIRST_FREE_OBJECTID) {
- if (root_objectid == BTRFS_CHUNK_TREE_OBJECTID)
- flags = BTRFS_BLOCK_GROUP_SYSTEM;
- else
- flags = BTRFS_BLOCK_GROUP_METADATA;
- } else {
- flags = BTRFS_BLOCK_GROUP_DATA;
- }
-
- space_info = __find_space_info(fs_info, flags);
- BUG_ON(!space_info); /* Logic bug */
- percpu_counter_add(&space_info->total_bytes_pinned, num_bytes);
-}
-
-
static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *info,
struct btrfs_delayed_ref_node *node, u64 parent,
@@ -6845,7 +6891,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
if (is_data)
skinny_metadata = 0;
- ret = lookup_extent_backref(trans, extent_root, path, &iref,
+ ret = lookup_extent_backref(trans, info, path, &iref,
bytenr, num_bytes, parent,
root_objectid, owner_objectid,
owner_offset);
@@ -6877,8 +6923,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
#endif
if (!found_extent) {
BUG_ON(iref);
- ret = remove_extent_backref(trans, extent_root, path,
- NULL, refs_to_drop,
+ ret = remove_extent_backref(trans, info, path, NULL,
+ refs_to_drop,
is_data, &last_ref);
if (ret) {
btrfs_abort_transaction(trans, ret);
@@ -6953,8 +6999,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
if (item_size < sizeof(*ei)) {
BUG_ON(found_extent || extent_slot != path->slots[0]);
- ret = convert_extent_item_v0(trans, extent_root, path,
- owner_objectid, 0);
+ ret = convert_extent_item_v0(trans, info, path, owner_objectid,
+ 0);
if (ret < 0) {
btrfs_abort_transaction(trans, ret);
goto out;
@@ -7021,7 +7067,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
btrfs_mark_buffer_dirty(leaf);
}
if (found_extent) {
- ret = remove_extent_backref(trans, extent_root, path,
+ ret = remove_extent_backref(trans, info, path,
iref, refs_to_drop,
is_data, &last_ref);
if (ret) {
@@ -7029,8 +7075,6 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
goto out;
}
}
- add_pinned_bytes(info, -num_bytes, owner_objectid,
- root_objectid);
} else {
if (found_extent) {
BUG_ON(is_data && refs_to_drop !=
@@ -7095,7 +7139,7 @@ static noinline int check_ref_cleanup(struct btrfs_trans_handle *trans,
delayed_refs = &trans->transaction->delayed_refs;
spin_lock(&delayed_refs->lock);
- head = btrfs_find_delayed_ref_head(trans, bytenr);
+ head = btrfs_find_delayed_ref_head(delayed_refs, bytenr);
if (!head)
goto out_delayed_unlock;
@@ -7162,19 +7206,19 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
int ret;
if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
- ret = btrfs_add_delayed_tree_ref(fs_info, trans,
- buf->start, buf->len,
- parent,
+ int old_ref_mod, new_ref_mod;
+
+ ret = btrfs_add_delayed_tree_ref(fs_info, trans, buf->start,
+ buf->len, parent,
root->root_key.objectid,
btrfs_header_level(buf),
- BTRFS_DROP_DELAYED_REF, NULL);
+ BTRFS_DROP_DELAYED_REF, NULL,
+ &old_ref_mod, &new_ref_mod);
BUG_ON(ret); /* -ENOMEM */
+ pin = old_ref_mod >= 0 && new_ref_mod < 0;
}
- if (!last_ref)
- return;
-
- if (btrfs_header_generation(buf) == trans->transid) {
+ if (last_ref && btrfs_header_generation(buf) == trans->transid) {
struct btrfs_block_group_cache *cache;
if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
@@ -7183,6 +7227,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
goto out;
}
+ pin = 0;
cache = btrfs_lookup_block_group(fs_info, buf->start);
if (btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) {
@@ -7198,18 +7243,19 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
btrfs_free_reserved_bytes(cache, buf->len, 0);
btrfs_put_block_group(cache);
trace_btrfs_reserved_extent_free(fs_info, buf->start, buf->len);
- pin = 0;
}
out:
if (pin)
add_pinned_bytes(fs_info, buf->len, btrfs_header_level(buf),
root->root_key.objectid);
- /*
- * Deleting the buffer, clear the corrupt flag since it doesn't matter
- * anymore.
- */
- clear_bit(EXTENT_BUFFER_CORRUPT, &buf->bflags);
+ if (last_ref) {
+ /*
+ * Deleting the buffer, clear the corrupt flag since it doesn't
+ * matter anymore.
+ */
+ clear_bit(EXTENT_BUFFER_CORRUPT, &buf->bflags);
+ }
}
/* Can return -ENOMEM */
@@ -7218,12 +7264,12 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans,
u64 bytenr, u64 num_bytes, u64 parent, u64 root_objectid,
u64 owner, u64 offset)
{
+ int old_ref_mod, new_ref_mod;
int ret;
if (btrfs_is_testing(fs_info))
return 0;
- add_pinned_bytes(fs_info, num_bytes, owner, root_objectid);
/*
* tree log blocks never actually go into the extent allocation
@@ -7233,19 +7279,25 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans,
WARN_ON(owner >= BTRFS_FIRST_FREE_OBJECTID);
/* unlocks the pinned mutex */
btrfs_pin_extent(fs_info, bytenr, num_bytes, 1);
+ old_ref_mod = new_ref_mod = 0;
ret = 0;
} else if (owner < BTRFS_FIRST_FREE_OBJECTID) {
ret = btrfs_add_delayed_tree_ref(fs_info, trans, bytenr,
- num_bytes,
- parent, root_objectid, (int)owner,
- BTRFS_DROP_DELAYED_REF, NULL);
+ num_bytes, parent,
+ root_objectid, (int)owner,
+ BTRFS_DROP_DELAYED_REF, NULL,
+ &old_ref_mod, &new_ref_mod);
} else {
ret = btrfs_add_delayed_data_ref(fs_info, trans, bytenr,
- num_bytes,
- parent, root_objectid, owner,
- offset, 0,
- BTRFS_DROP_DELAYED_REF, NULL);
+ num_bytes, parent,
+ root_objectid, owner, offset,
+ 0, BTRFS_DROP_DELAYED_REF,
+ &old_ref_mod, &new_ref_mod);
}
+
+ if (ret == 0 && old_ref_mod >= 0 && new_ref_mod < 0)
+ add_pinned_bytes(fs_info, num_bytes, owner, root_objectid);
+
return ret;
}
@@ -7419,12 +7471,11 @@ btrfs_release_block_group(struct btrfs_block_group_cache *cache,
* If there is no suitable free space, we will record the max size of
* the free space extent currently.
*/
-static noinline int find_free_extent(struct btrfs_root *orig_root,
+static noinline int find_free_extent(struct btrfs_fs_info *fs_info,
u64 ram_bytes, u64 num_bytes, u64 empty_size,
u64 hint_byte, struct btrfs_key *ins,
u64 flags, int delalloc)
{
- struct btrfs_fs_info *fs_info = orig_root->fs_info;
int ret = 0;
struct btrfs_root *root = fs_info->extent_root;
struct btrfs_free_cluster *last_ptr = NULL;
@@ -7716,18 +7767,20 @@ unclustered_alloc:
last_ptr->fragmented = 1;
spin_unlock(&last_ptr->lock);
}
- spin_lock(&block_group->free_space_ctl->tree_lock);
- if (cached &&
- block_group->free_space_ctl->free_space <
- num_bytes + empty_cluster + empty_size) {
- if (block_group->free_space_ctl->free_space >
- max_extent_size)
- max_extent_size =
- block_group->free_space_ctl->free_space;
- spin_unlock(&block_group->free_space_ctl->tree_lock);
- goto loop;
+ if (cached) {
+ struct btrfs_free_space_ctl *ctl =
+ block_group->free_space_ctl;
+
+ spin_lock(&ctl->tree_lock);
+ if (ctl->free_space <
+ num_bytes + empty_cluster + empty_size) {
+ if (ctl->free_space > max_extent_size)
+ max_extent_size = ctl->free_space;
+ spin_unlock(&ctl->tree_lock);
+ goto loop;
+ }
+ spin_unlock(&ctl->tree_lock);
}
- spin_unlock(&block_group->free_space_ctl->tree_lock);
offset = btrfs_find_space_for_alloc(block_group, search_start,
num_bytes, empty_size,
@@ -7908,9 +7961,8 @@ static void dump_space_info(struct btrfs_fs_info *fs_info,
spin_lock(&info->lock);
btrfs_info(fs_info, "space_info %llu has %llu free, is %sfull",
info->flags,
- info->total_bytes - info->bytes_used - info->bytes_pinned -
- info->bytes_reserved - info->bytes_readonly -
- info->bytes_may_use, (info->full) ? "" : "not ");
+ info->total_bytes - btrfs_space_info_used(info, true),
+ info->full ? "" : "not ");
btrfs_info(fs_info,
"space_info total=%llu, used=%llu, pinned=%llu, reserved=%llu, may_use=%llu, readonly=%llu",
info->total_bytes, info->bytes_used, info->bytes_pinned,
@@ -7948,10 +8000,10 @@ int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes,
u64 flags;
int ret;
- flags = btrfs_get_alloc_profile(root, is_data);
+ flags = get_alloc_profile_by_root(root, is_data);
again:
WARN_ON(num_bytes < fs_info->sectorsize);
- ret = find_free_extent(root, ram_bytes, num_bytes, empty_size,
+ ret = find_free_extent(fs_info, ram_bytes, num_bytes, empty_size,
hint_byte, ins, flags, delalloc);
if (!ret && !is_data) {
btrfs_dec_block_group_reservations(fs_info, ins->objectid);
@@ -8192,10 +8244,9 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
BUG_ON(root_objectid == BTRFS_TREE_LOG_OBJECTID);
ret = btrfs_add_delayed_data_ref(fs_info, trans, ins->objectid,
- ins->offset, 0,
- root_objectid, owner, offset,
- ram_bytes, BTRFS_ADD_DELAYED_EXTENT,
- NULL);
+ ins->offset, 0, root_objectid, owner,
+ offset, ram_bytes,
+ BTRFS_ADD_DELAYED_EXTENT, NULL, NULL);
return ret;
}
@@ -8256,7 +8307,7 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
btrfs_set_header_generation(buf, trans->transid);
btrfs_set_buffer_lockdep_class(root->root_key.objectid, buf, level);
btrfs_tree_lock(buf);
- clean_tree_block(trans, fs_info, buf);
+ clean_tree_block(fs_info, buf);
clear_bit(EXTENT_BUFFER_STALE, &buf->bflags);
btrfs_set_lock_blocking(buf);
@@ -8351,10 +8402,11 @@ static void unuse_block_rsv(struct btrfs_fs_info *fs_info,
* returns the tree buffer or an ERR_PTR on error.
*/
struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- u64 parent, u64 root_objectid,
- struct btrfs_disk_key *key, int level,
- u64 hint, u64 empty_size)
+ struct btrfs_root *root,
+ u64 parent, u64 root_objectid,
+ const struct btrfs_disk_key *key,
+ int level, u64 hint,
+ u64 empty_size)
{
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_key ins;
@@ -8414,11 +8466,11 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
extent_op->is_data = false;
extent_op->level = level;
- ret = btrfs_add_delayed_tree_ref(fs_info, trans,
- ins.objectid, ins.offset,
- parent, root_objectid, level,
+ ret = btrfs_add_delayed_tree_ref(fs_info, trans, ins.objectid,
+ ins.offset, parent,
+ root_objectid, level,
BTRFS_ADD_DELAYED_EXTENT,
- extent_op);
+ extent_op, NULL, NULL);
if (ret)
goto out_free_delayed;
}
@@ -8876,7 +8928,7 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
btrfs_set_lock_blocking(eb);
path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING;
}
- clean_tree_block(trans, fs_info, eb);
+ clean_tree_block(fs_info, eb);
}
if (eb == root->node) {
@@ -9346,8 +9398,7 @@ static int inc_block_group_ro(struct btrfs_block_group_cache *cache, int force)
num_bytes = cache->key.offset - cache->reserved - cache->pinned -
cache->bytes_super - btrfs_block_group_used(&cache->item);
- if (sinfo->bytes_used + sinfo->bytes_reserved + sinfo->bytes_pinned +
- sinfo->bytes_may_use + sinfo->bytes_readonly + num_bytes +
+ if (btrfs_space_info_used(sinfo, true) + num_bytes +
min_allocable_bytes <= sinfo->total_bytes) {
sinfo->bytes_readonly += num_bytes;
cache->ro++;
@@ -9360,17 +9411,16 @@ out:
return ret;
}
-int btrfs_inc_block_group_ro(struct btrfs_root *root,
+int btrfs_inc_block_group_ro(struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache *cache)
{
- struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_trans_handle *trans;
u64 alloc_flags;
int ret;
again:
- trans = btrfs_join_transaction(root);
+ trans = btrfs_join_transaction(fs_info->extent_root);
if (IS_ERR(trans))
return PTR_ERR(trans);
@@ -9557,9 +9607,8 @@ int btrfs_can_relocate(struct btrfs_fs_info *fs_info, u64 bytenr)
* all of the extents from this block group. If we can, we're good
*/
if ((space_info->total_bytes != block_group->key.offset) &&
- (space_info->bytes_used + space_info->bytes_reserved +
- space_info->bytes_pinned + space_info->bytes_readonly +
- min_free < space_info->total_bytes)) {
+ (btrfs_space_info_used(space_info, false) + min_free <
+ space_info->total_bytes)) {
spin_unlock(&space_info->lock);
goto out;
}
@@ -9742,6 +9791,11 @@ void btrfs_put_block_group_cache(struct btrfs_fs_info *info)
}
}
+/*
+ * Must be called only after stopping all workers, since we could have block
+ * group caching kthreads running, and therefore they could race with us if we
+ * freed the block groups before stopping them.
+ */
int btrfs_free_block_groups(struct btrfs_fs_info *info)
{
struct btrfs_block_group_cache *block_group;
@@ -9781,9 +9835,6 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
list_del(&block_group->list);
up_write(&block_group->space_info->groups_sem);
- if (block_group->cached == BTRFS_CACHE_STARTED)
- wait_block_group_cache_done(block_group);
-
/*
* We haven't cached this block group, which means we could
* possibly have excluded extents on this block group.
@@ -9793,6 +9844,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
free_excluded_extents(info, block_group);
btrfs_remove_free_space_cache(block_group);
+ ASSERT(block_group->cached != BTRFS_CACHE_STARTED);
ASSERT(list_empty(&block_group->dirty_list));
ASSERT(list_empty(&block_group->io_list));
ASSERT(list_empty(&block_group->bg_list));
@@ -9920,6 +9972,7 @@ btrfs_create_block_group_cache(struct btrfs_fs_info *fs_info,
btrfs_init_free_space_ctl(cache);
atomic_set(&cache->trimming, 0);
mutex_init(&cache->free_space_lock);
+ btrfs_init_full_stripe_locks_tree(&cache->full_stripe_locks_root);
return cache;
}
@@ -10050,19 +10103,9 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
}
trace_btrfs_add_block_group(info, cache, 0);
- ret = update_space_info(info, cache->flags, found_key.offset,
- btrfs_block_group_used(&cache->item),
- cache->bytes_super, &space_info);
- if (ret) {
- btrfs_remove_free_space_cache(cache);
- spin_lock(&info->block_group_cache_lock);
- rb_erase(&cache->cache_node,
- &info->block_group_cache_tree);
- RB_CLEAR_NODE(&cache->cache_node);
- spin_unlock(&info->block_group_cache_lock);
- btrfs_put_block_group(cache);
- goto error;
- }
+ update_space_info(info, cache->flags, found_key.offset,
+ btrfs_block_group_used(&cache->item),
+ cache->bytes_super, &space_info);
cache->space_info = space_info;
@@ -10194,16 +10237,19 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
}
#endif
/*
- * Call to ensure the corresponding space_info object is created and
- * assigned to our block group, but don't update its counters just yet.
- * We want our bg to be added to the rbtree with its ->space_info set.
+ * Ensure the corresponding space_info object is created and
+ * assigned to our block group. We want our bg to be added to the rbtree
+ * with its ->space_info set.
*/
- ret = update_space_info(fs_info, cache->flags, 0, 0, 0,
- &cache->space_info);
- if (ret) {
- btrfs_remove_free_space_cache(cache);
- btrfs_put_block_group(cache);
- return ret;
+ cache->space_info = __find_space_info(fs_info, cache->flags);
+ if (!cache->space_info) {
+ ret = create_space_info(fs_info, cache->flags,
+ &cache->space_info);
+ if (ret) {
+ btrfs_remove_free_space_cache(cache);
+ btrfs_put_block_group(cache);
+ return ret;
+ }
}
ret = btrfs_add_block_group_cache(fs_info, cache);
@@ -10218,18 +10264,8 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
* the rbtree, update the space info's counters.
*/
trace_btrfs_add_block_group(fs_info, cache, 1);
- ret = update_space_info(fs_info, cache->flags, size, bytes_used,
+ update_space_info(fs_info, cache->flags, size, bytes_used,
cache->bytes_super, &cache->space_info);
- if (ret) {
- btrfs_remove_free_space_cache(cache);
- spin_lock(&fs_info->block_group_cache_lock);
- rb_erase(&cache->cache_node,
- &fs_info->block_group_cache_tree);
- RB_CLEAR_NODE(&cache->cache_node);
- spin_unlock(&fs_info->block_group_cache_lock);
- btrfs_put_block_group(cache);
- return ret;
- }
update_global_block_rsv(fs_info);
__link_block_group(cache->space_info, cache);
@@ -10317,7 +10353,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
* get the inode first so any iput calls done for the io_list
* aren't the final iput (no unlinks allowed now)
*/
- inode = lookup_free_space_inode(tree_root, block_group, path);
+ inode = lookup_free_space_inode(fs_info, block_group, path);
mutex_lock(&trans->transaction->cache_write_mutex);
/*
@@ -10344,7 +10380,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
mutex_unlock(&trans->transaction->cache_write_mutex);
if (!IS_ERR(inode)) {
- ret = btrfs_orphan_add(trans, inode);
+ ret = btrfs_orphan_add(trans, BTRFS_I(inode));
if (ret) {
btrfs_add_delayed_iput(inode);
goto out;
@@ -10419,7 +10455,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
&fs_info->caching_block_groups, list)
if (ctl->block_group == block_group) {
caching_ctl = ctl;
- atomic_inc(&caching_ctl->count);
+ refcount_inc(&caching_ctl->count);
break;
}
}
@@ -10777,21 +10813,21 @@ int btrfs_init_space_info(struct btrfs_fs_info *fs_info)
mixed = 1;
flags = BTRFS_BLOCK_GROUP_SYSTEM;
- ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
+ ret = create_space_info(fs_info, flags, &space_info);
if (ret)
goto out;
if (mixed) {
flags = BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA;
- ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
+ ret = create_space_info(fs_info, flags, &space_info);
} else {
flags = BTRFS_BLOCK_GROUP_METADATA;
- ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
+ ret = create_space_info(fs_info, flags, &space_info);
if (ret)
goto out;
flags = BTRFS_BLOCK_GROUP_DATA;
- ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
+ ret = create_space_info(fs_info, flags, &space_info);
}
out:
return ret;
@@ -10853,7 +10889,7 @@ static int btrfs_trim_free_extents(struct btrfs_device *device,
spin_lock(&fs_info->trans_lock);
trans = fs_info->running_transaction;
if (trans)
- atomic_inc(&trans->use_count);
+ refcount_inc(&trans->use_count);
spin_unlock(&fs_info->trans_lock);
ret = find_free_dev_extent_start(trans, device, minlen, start,
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 4ac383a3a649..0aff9b278c19 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -68,7 +68,7 @@ void btrfs_leak_debug_check(void)
pr_err("BTRFS: state leak: start %llu end %llu state %u in tree %d refs %d\n",
state->start, state->end, state->state,
extent_state_in_tree(state),
- atomic_read(&state->refs));
+ refcount_read(&state->refs));
list_del(&state->leak_list);
kmem_cache_free(extent_state_cache, state);
}
@@ -87,19 +87,9 @@ void btrfs_leak_debug_check(void)
static inline void __btrfs_debug_check_extent_io_range(const char *caller,
struct extent_io_tree *tree, u64 start, u64 end)
{
- struct inode *inode;
- u64 isize;
-
- if (!tree->mapping)
- return;
-
- inode = tree->mapping->host;
- isize = i_size_read(inode);
- if (end >= PAGE_SIZE && (end % 2) == 0 && end != isize - 1) {
- btrfs_debug_rl(BTRFS_I(inode)->root->fs_info,
- "%s: ino %llu isize %llu odd range [%llu,%llu]",
- caller, btrfs_ino(inode), isize, start, end);
- }
+ if (tree->ops && tree->ops->check_extent_io_range)
+ tree->ops->check_extent_io_range(tree->private_data, caller,
+ start, end);
}
#else
#define btrfs_leak_debug_add(new, head) do {} while (0)
@@ -144,7 +134,7 @@ static void add_extent_changeset(struct extent_state *state, unsigned bits,
if (!set && (state->state & bits) == 0)
return;
changeset->bytes_changed += state->end - state->start + 1;
- ret = ulist_add(changeset->range_changed, state->start, state->end,
+ ret = ulist_add(&changeset->range_changed, state->start, state->end,
GFP_ATOMIC);
/* ENOMEM */
BUG_ON(ret < 0);
@@ -154,9 +144,9 @@ static noinline void flush_write_bio(void *data);
static inline struct btrfs_fs_info *
tree_fs_info(struct extent_io_tree *tree)
{
- if (!tree->mapping)
- return NULL;
- return btrfs_sb(tree->mapping->host->i_sb);
+ if (tree->ops)
+ return tree->ops->tree_fs_info(tree->private_data);
+ return NULL;
}
int __init extent_io_init(void)
@@ -174,7 +164,8 @@ int __init extent_io_init(void)
goto free_state_cache;
btrfs_bioset = bioset_create(BIO_POOL_SIZE,
- offsetof(struct btrfs_io_bio, bio));
+ offsetof(struct btrfs_io_bio, bio),
+ BIOSET_NEED_BVECS);
if (!btrfs_bioset)
goto free_buffer_cache;
@@ -213,19 +204,24 @@ void extent_io_exit(void)
}
void extent_io_tree_init(struct extent_io_tree *tree,
- struct address_space *mapping)
+ void *private_data)
{
tree->state = RB_ROOT;
tree->ops = NULL;
tree->dirty_bytes = 0;
spin_lock_init(&tree->lock);
- tree->mapping = mapping;
+ tree->private_data = private_data;
}
static struct extent_state *alloc_extent_state(gfp_t mask)
{
struct extent_state *state;
+ /*
+ * The given mask might be not appropriate for the slab allocator,
+ * drop the unsupported bits
+ */
+ mask &= ~(__GFP_DMA32|__GFP_HIGHMEM);
state = kmem_cache_alloc(extent_state_cache, mask);
if (!state)
return state;
@@ -233,7 +229,7 @@ static struct extent_state *alloc_extent_state(gfp_t mask)
state->failrec = NULL;
RB_CLEAR_NODE(&state->rb_node);
btrfs_leak_debug_add(&state->leak_list, &states);
- atomic_set(&state->refs, 1);
+ refcount_set(&state->refs, 1);
init_waitqueue_head(&state->wq);
trace_alloc_extent_state(state, mask, _RET_IP_);
return state;
@@ -243,7 +239,7 @@ void free_extent_state(struct extent_state *state)
{
if (!state)
return;
- if (atomic_dec_and_test(&state->refs)) {
+ if (refcount_dec_and_test(&state->refs)) {
WARN_ON(extent_state_in_tree(state));
btrfs_leak_debug_del(&state->leak_list);
trace_free_extent_state(state, _RET_IP_);
@@ -364,8 +360,7 @@ static void merge_cb(struct extent_io_tree *tree, struct extent_state *new,
struct extent_state *other)
{
if (tree->ops && tree->ops->merge_extent_hook)
- tree->ops->merge_extent_hook(tree->mapping->host, new,
- other);
+ tree->ops->merge_extent_hook(tree->private_data, new, other);
}
/*
@@ -416,14 +411,14 @@ static void set_state_cb(struct extent_io_tree *tree,
struct extent_state *state, unsigned *bits)
{
if (tree->ops && tree->ops->set_bit_hook)
- tree->ops->set_bit_hook(tree->mapping->host, state, bits);
+ tree->ops->set_bit_hook(tree->private_data, state, bits);
}
static void clear_state_cb(struct extent_io_tree *tree,
struct extent_state *state, unsigned *bits)
{
if (tree->ops && tree->ops->clear_bit_hook)
- tree->ops->clear_bit_hook(tree->mapping->host, state, bits);
+ tree->ops->clear_bit_hook(tree->private_data, state, bits);
}
static void set_state_bits(struct extent_io_tree *tree,
@@ -472,7 +467,7 @@ static void split_cb(struct extent_io_tree *tree, struct extent_state *orig,
u64 split)
{
if (tree->ops && tree->ops->split_extent_hook)
- tree->ops->split_extent_hook(tree->mapping->host, orig, split);
+ tree->ops->split_extent_hook(tree->private_data, orig, split);
}
/*
@@ -635,7 +630,7 @@ again:
if (cached && extent_state_in_tree(cached) &&
cached->start <= start && cached->end > start) {
if (clear)
- atomic_dec(&cached->refs);
+ refcount_dec(&cached->refs);
state = cached;
goto hit_next;
}
@@ -787,7 +782,7 @@ process_node:
if (state->state & bits) {
start = state->start;
- atomic_inc(&state->refs);
+ refcount_inc(&state->refs);
wait_on_state(tree, state);
free_extent_state(state);
goto again;
@@ -828,7 +823,7 @@ static void cache_state_if_flags(struct extent_state *state,
if (cached_ptr && !(*cached_ptr)) {
if (!flags || (state->state & flags)) {
*cached_ptr = state;
- atomic_inc(&state->refs);
+ refcount_inc(&state->refs);
}
}
}
@@ -1396,17 +1391,7 @@ void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end)
*/
static void set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end)
{
- unsigned long index = start >> PAGE_SHIFT;
- unsigned long end_index = end >> PAGE_SHIFT;
- struct page *page;
-
- while (index <= end_index) {
- page = find_get_page(tree->mapping, index);
- BUG_ON(!page); /* Pages should be in the extent_io_tree */
- set_page_writeback(page);
- put_page(page);
- index++;
- }
+ tree->ops->set_range_writeback(tree->private_data, start, end);
}
/* find the first state struct with 'bits' set after 'start', and
@@ -1532,7 +1517,7 @@ static noinline u64 find_delalloc_range(struct extent_io_tree *tree,
if (!found) {
*start = state->start;
*cached_state = state;
- atomic_inc(&state->refs);
+ refcount_inc(&state->refs);
}
found++;
*end = state->end;
@@ -1549,33 +1534,24 @@ out:
return found;
}
+static int __process_pages_contig(struct address_space *mapping,
+ struct page *locked_page,
+ pgoff_t start_index, pgoff_t end_index,
+ unsigned long page_ops, pgoff_t *index_ret);
+
static noinline void __unlock_for_delalloc(struct inode *inode,
struct page *locked_page,
u64 start, u64 end)
{
- int ret;
- struct page *pages[16];
unsigned long index = start >> PAGE_SHIFT;
unsigned long end_index = end >> PAGE_SHIFT;
- unsigned long nr_pages = end_index - index + 1;
- int i;
+ ASSERT(locked_page);
if (index == locked_page->index && end_index == index)
return;
- while (nr_pages > 0) {
- ret = find_get_pages_contig(inode->i_mapping, index,
- min_t(unsigned long, nr_pages,
- ARRAY_SIZE(pages)), pages);
- for (i = 0; i < ret; i++) {
- if (pages[i] != locked_page)
- unlock_page(pages[i]);
- put_page(pages[i]);
- }
- nr_pages -= ret;
- index += ret;
- cond_resched();
- }
+ __process_pages_contig(inode->i_mapping, locked_page, index, end_index,
+ PAGE_UNLOCK, NULL);
}
static noinline int lock_delalloc_pages(struct inode *inode,
@@ -1584,59 +1560,19 @@ static noinline int lock_delalloc_pages(struct inode *inode,
u64 delalloc_end)
{
unsigned long index = delalloc_start >> PAGE_SHIFT;
- unsigned long start_index = index;
+ unsigned long index_ret = index;
unsigned long end_index = delalloc_end >> PAGE_SHIFT;
- unsigned long pages_locked = 0;
- struct page *pages[16];
- unsigned long nrpages;
int ret;
- int i;
- /* the caller is responsible for locking the start index */
+ ASSERT(locked_page);
if (index == locked_page->index && index == end_index)
return 0;
- /* skip the page at the start index */
- nrpages = end_index - index + 1;
- while (nrpages > 0) {
- ret = find_get_pages_contig(inode->i_mapping, index,
- min_t(unsigned long,
- nrpages, ARRAY_SIZE(pages)), pages);
- if (ret == 0) {
- ret = -EAGAIN;
- goto done;
- }
- /* now we have an array of pages, lock them all */
- for (i = 0; i < ret; i++) {
- /*
- * the caller is taking responsibility for
- * locked_page
- */
- if (pages[i] != locked_page) {
- lock_page(pages[i]);
- if (!PageDirty(pages[i]) ||
- pages[i]->mapping != inode->i_mapping) {
- ret = -EAGAIN;
- unlock_page(pages[i]);
- put_page(pages[i]);
- goto done;
- }
- }
- put_page(pages[i]);
- pages_locked++;
- }
- nrpages -= ret;
- index += ret;
- cond_resched();
- }
- ret = 0;
-done:
- if (ret && pages_locked) {
- __unlock_for_delalloc(inode, locked_page,
- delalloc_start,
- ((u64)(start_index + pages_locked - 1)) <<
- PAGE_SHIFT);
- }
+ ret = __process_pages_contig(inode->i_mapping, locked_page, index,
+ end_index, PAGE_LOCK, &index_ret);
+ if (ret == -EAGAIN)
+ __unlock_for_delalloc(inode, locked_page, delalloc_start,
+ (u64)index_ret << PAGE_SHIFT);
return ret;
}
@@ -1726,37 +1662,48 @@ out_failed:
return found;
}
-void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
- u64 delalloc_end, struct page *locked_page,
- unsigned clear_bits,
- unsigned long page_ops)
+static int __process_pages_contig(struct address_space *mapping,
+ struct page *locked_page,
+ pgoff_t start_index, pgoff_t end_index,
+ unsigned long page_ops, pgoff_t *index_ret)
{
- struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
- int ret;
+ unsigned long nr_pages = end_index - start_index + 1;
+ unsigned long pages_locked = 0;
+ pgoff_t index = start_index;
struct page *pages[16];
- unsigned long index = start >> PAGE_SHIFT;
- unsigned long end_index = end >> PAGE_SHIFT;
- unsigned long nr_pages = end_index - index + 1;
+ unsigned ret;
+ int err = 0;
int i;
- clear_extent_bit(tree, start, end, clear_bits, 1, 0, NULL, GFP_NOFS);
- if (page_ops == 0)
- return;
+ if (page_ops & PAGE_LOCK) {
+ ASSERT(page_ops == PAGE_LOCK);
+ ASSERT(index_ret && *index_ret == start_index);
+ }
if ((page_ops & PAGE_SET_ERROR) && nr_pages > 0)
- mapping_set_error(inode->i_mapping, -EIO);
+ mapping_set_error(mapping, -EIO);
while (nr_pages > 0) {
- ret = find_get_pages_contig(inode->i_mapping, index,
+ ret = find_get_pages_contig(mapping, index,
min_t(unsigned long,
nr_pages, ARRAY_SIZE(pages)), pages);
- for (i = 0; i < ret; i++) {
+ if (ret == 0) {
+ /*
+ * Only if we're going to lock these pages,
+ * can we find nothing at @index.
+ */
+ ASSERT(page_ops & PAGE_LOCK);
+ err = -EAGAIN;
+ goto out;
+ }
+ for (i = 0; i < ret; i++) {
if (page_ops & PAGE_SET_PRIVATE2)
SetPagePrivate2(pages[i]);
if (pages[i] == locked_page) {
put_page(pages[i]);
+ pages_locked++;
continue;
}
if (page_ops & PAGE_CLEAR_DIRTY)
@@ -1769,12 +1716,40 @@ void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
end_page_writeback(pages[i]);
if (page_ops & PAGE_UNLOCK)
unlock_page(pages[i]);
+ if (page_ops & PAGE_LOCK) {
+ lock_page(pages[i]);
+ if (!PageDirty(pages[i]) ||
+ pages[i]->mapping != mapping) {
+ unlock_page(pages[i]);
+ put_page(pages[i]);
+ err = -EAGAIN;
+ goto out;
+ }
+ }
put_page(pages[i]);
+ pages_locked++;
}
nr_pages -= ret;
index += ret;
cond_resched();
}
+out:
+ if (err && index_ret)
+ *index_ret = start_index + pages_locked - 1;
+ return err;
+}
+
+void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
+ u64 delalloc_end, struct page *locked_page,
+ unsigned clear_bits,
+ unsigned long page_ops)
+{
+ clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, clear_bits, 1, 0,
+ NULL, GFP_NOFS);
+
+ __process_pages_contig(inode->i_mapping, locked_page,
+ start >> PAGE_SHIFT, end >> PAGE_SHIFT,
+ page_ops, NULL);
}
/*
@@ -1965,11 +1940,12 @@ static void check_page_uptodate(struct extent_io_tree *tree, struct page *page)
SetPageUptodate(page);
}
-int free_io_failure(struct inode *inode, struct io_failure_record *rec)
+int free_io_failure(struct extent_io_tree *failure_tree,
+ struct extent_io_tree *io_tree,
+ struct io_failure_record *rec)
{
int ret;
int err = 0;
- struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
set_state_failrec(failure_tree, rec->start, NULL);
ret = clear_extent_bits(failure_tree, rec->start,
@@ -1978,7 +1954,7 @@ int free_io_failure(struct inode *inode, struct io_failure_record *rec)
if (ret)
err = ret;
- ret = clear_extent_bits(&BTRFS_I(inode)->io_tree, rec->start,
+ ret = clear_extent_bits(io_tree, rec->start,
rec->start + rec->len - 1,
EXTENT_DAMAGED);
if (ret && !err)
@@ -1998,28 +1974,21 @@ int free_io_failure(struct inode *inode, struct io_failure_record *rec)
* currently, there can be no more than two copies of every data bit. thus,
* exactly one rewrite is required.
*/
-int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
- struct page *page, unsigned int pg_offset, int mirror_num)
+int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
+ u64 length, u64 logical, struct page *page,
+ unsigned int pg_offset, int mirror_num)
{
- struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
struct bio *bio;
struct btrfs_device *dev;
u64 map_length = 0;
u64 sector;
struct btrfs_bio *bbio = NULL;
- struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
int ret;
ASSERT(!(fs_info->sb->s_flags & MS_RDONLY));
BUG_ON(!mirror_num);
- /* we can't repair anything in raid56 yet */
- if (btrfs_is_parity_mirror(map_tree, logical, length, mirror_num))
- return 0;
-
- bio = btrfs_io_bio_alloc(GFP_NOFS, 1);
- if (!bio)
- return -EIO;
+ bio = btrfs_io_bio_alloc(1);
bio->bi_iter.bi_size = 0;
map_length = length;
@@ -2029,17 +1998,35 @@ int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
* read repair operation.
*/
btrfs_bio_counter_inc_blocked(fs_info);
- ret = btrfs_map_block(fs_info, BTRFS_MAP_WRITE, logical,
- &map_length, &bbio, mirror_num);
- if (ret) {
- btrfs_bio_counter_dec(fs_info);
- bio_put(bio);
- return -EIO;
+ if (btrfs_is_parity_mirror(fs_info, logical, length, mirror_num)) {
+ /*
+ * Note that we don't use BTRFS_MAP_WRITE because it's supposed
+ * to update all raid stripes, but here we just want to correct
+ * bad stripe, thus BTRFS_MAP_READ is abused to only get the bad
+ * stripe's dev and sector.
+ */
+ ret = btrfs_map_block(fs_info, BTRFS_MAP_READ, logical,
+ &map_length, &bbio, 0);
+ if (ret) {
+ btrfs_bio_counter_dec(fs_info);
+ bio_put(bio);
+ return -EIO;
+ }
+ ASSERT(bbio->mirror_num == 1);
+ } else {
+ ret = btrfs_map_block(fs_info, BTRFS_MAP_WRITE, logical,
+ &map_length, &bbio, mirror_num);
+ if (ret) {
+ btrfs_bio_counter_dec(fs_info);
+ bio_put(bio);
+ return -EIO;
+ }
+ BUG_ON(mirror_num != bbio->mirror_num);
}
- BUG_ON(mirror_num != bbio->mirror_num);
- sector = bbio->stripes[mirror_num-1].physical >> 9;
+
+ sector = bbio->stripes[bbio->mirror_num - 1].physical >> 9;
bio->bi_iter.bi_sector = sector;
- dev = bbio->stripes[mirror_num-1].dev;
+ dev = bbio->stripes[bbio->mirror_num - 1].dev;
btrfs_put_bbio(bbio);
if (!dev || !dev->bdev || !dev->writeable) {
btrfs_bio_counter_dec(fs_info);
@@ -2060,7 +2047,7 @@ int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
btrfs_info_rl_in_rcu(fs_info,
"read error corrected: ino %llu off %llu (dev %s sector %llu)",
- btrfs_ino(inode), start,
+ ino, start,
rcu_str_deref(dev->name), sector);
btrfs_bio_counter_dec(fs_info);
bio_put(bio);
@@ -2080,8 +2067,7 @@ int repair_eb_io_failure(struct btrfs_fs_info *fs_info,
for (i = 0; i < num_pages; i++) {
struct page *p = eb->pages[i];
- ret = repair_io_failure(fs_info->btree_inode, start,
- PAGE_SIZE, start, p,
+ ret = repair_io_failure(fs_info, 0, start, PAGE_SIZE, start, p,
start - page_offset(p), mirror_num);
if (ret)
break;
@@ -2095,24 +2081,24 @@ int repair_eb_io_failure(struct btrfs_fs_info *fs_info,
* each time an IO finishes, we do a fast check in the IO failure tree
* to see if we need to process or clean up an io_failure_record
*/
-int clean_io_failure(struct inode *inode, u64 start, struct page *page,
- unsigned int pg_offset)
+int clean_io_failure(struct btrfs_fs_info *fs_info,
+ struct extent_io_tree *failure_tree,
+ struct extent_io_tree *io_tree, u64 start,
+ struct page *page, u64 ino, unsigned int pg_offset)
{
u64 private;
struct io_failure_record *failrec;
- struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
struct extent_state *state;
int num_copies;
int ret;
private = 0;
- ret = count_range_bits(&BTRFS_I(inode)->io_failure_tree, &private,
- (u64)-1, 1, EXTENT_DIRTY, 0);
+ ret = count_range_bits(failure_tree, &private, (u64)-1, 1,
+ EXTENT_DIRTY, 0);
if (!ret)
return 0;
- ret = get_state_failrec(&BTRFS_I(inode)->io_failure_tree, start,
- &failrec);
+ ret = get_state_failrec(failure_tree, start, &failrec);
if (ret)
return 0;
@@ -2128,25 +2114,25 @@ int clean_io_failure(struct inode *inode, u64 start, struct page *page,
if (fs_info->sb->s_flags & MS_RDONLY)
goto out;
- spin_lock(&BTRFS_I(inode)->io_tree.lock);
- state = find_first_extent_bit_state(&BTRFS_I(inode)->io_tree,
+ spin_lock(&io_tree->lock);
+ state = find_first_extent_bit_state(io_tree,
failrec->start,
EXTENT_LOCKED);
- spin_unlock(&BTRFS_I(inode)->io_tree.lock);
+ spin_unlock(&io_tree->lock);
if (state && state->start <= failrec->start &&
state->end >= failrec->start + failrec->len - 1) {
num_copies = btrfs_num_copies(fs_info, failrec->logical,
failrec->len);
if (num_copies > 1) {
- repair_io_failure(inode, start, failrec->len,
- failrec->logical, page,
- pg_offset, failrec->failed_mirror);
+ repair_io_failure(fs_info, ino, start, failrec->len,
+ failrec->logical, page, pg_offset,
+ failrec->failed_mirror);
}
}
out:
- free_io_failure(inode, failrec);
+ free_io_failure(failure_tree, io_tree, failrec);
return 0;
}
@@ -2157,9 +2143,9 @@ out:
* - under ordered extent
* - the inode is freeing
*/
-void btrfs_free_io_failure_record(struct inode *inode, u64 start, u64 end)
+void btrfs_free_io_failure_record(struct btrfs_inode *inode, u64 start, u64 end)
{
- struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
+ struct extent_io_tree *failure_tree = &inode->io_failure_tree;
struct io_failure_record *failrec;
struct extent_state *state, *next;
@@ -2272,7 +2258,7 @@ int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
return 0;
}
-int btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
+bool btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
struct io_failure_record *failrec, int failed_mirror)
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
@@ -2288,7 +2274,7 @@ int btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
btrfs_debug(fs_info,
"Check Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d",
num_copies, failrec->this_mirror, failed_mirror);
- return 0;
+ return false;
}
/*
@@ -2329,10 +2315,10 @@ int btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
btrfs_debug(fs_info,
"Check Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d",
num_copies, failrec->this_mirror, failed_mirror);
- return 0;
+ return false;
}
- return 1;
+ return true;
}
@@ -2346,10 +2332,7 @@ struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
struct btrfs_io_bio *btrfs_failed_bio;
struct btrfs_io_bio *btrfs_bio;
- bio = btrfs_io_bio_alloc(GFP_NOFS, 1);
- if (!bio)
- return NULL;
-
+ bio = btrfs_io_bio_alloc(1);
bio->bi_end_io = endio_func;
bio->bi_iter.bi_sector = failrec->logical >> 9;
bio->bi_bdev = fs_info->fs_devices->latest_bdev;
@@ -2387,8 +2370,10 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
struct io_failure_record *failrec;
struct inode *inode = page->mapping->host;
struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
+ struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
struct bio *bio;
int read_mode = 0;
+ blk_status_t status;
int ret;
BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
@@ -2397,9 +2382,9 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
if (ret)
return ret;
- ret = btrfs_check_repairable(inode, failed_bio, failrec, failed_mirror);
- if (!ret) {
- free_io_failure(inode, failrec);
+ if (!btrfs_check_repairable(inode, failed_bio, failrec,
+ failed_mirror)) {
+ free_io_failure(failure_tree, tree, failrec);
return -EIO;
}
@@ -2411,21 +2396,18 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
start - page_offset(page),
(int)phy_offset, failed_bio->bi_end_io,
NULL);
- if (!bio) {
- free_io_failure(inode, failrec);
- return -EIO;
- }
bio_set_op_attrs(bio, REQ_OP_READ, read_mode);
btrfs_debug(btrfs_sb(inode->i_sb),
"Repair Read Error: submitting new read[%#x] to this_mirror=%d, in_validation=%d",
read_mode, failrec->this_mirror, failrec->in_validation);
- ret = tree->ops->submit_bio_hook(inode, bio, failrec->this_mirror,
+ status = tree->ops->submit_bio_hook(tree->private_data, bio, failrec->this_mirror,
failrec->bio_flags, 0);
- if (ret) {
- free_io_failure(inode, failrec);
+ if (status) {
+ free_io_failure(failure_tree, tree, failrec);
bio_put(bio);
+ ret = blk_status_to_errno(status);
}
return ret;
@@ -2441,17 +2423,14 @@ void end_extent_writepage(struct page *page, int err, u64 start, u64 end)
tree = &BTRFS_I(page->mapping->host)->io_tree;
- if (tree->ops && tree->ops->writepage_end_io_hook) {
- ret = tree->ops->writepage_end_io_hook(page, start,
- end, NULL, uptodate);
- if (ret)
- uptodate = 0;
- }
+ if (tree->ops && tree->ops->writepage_end_io_hook)
+ tree->ops->writepage_end_io_hook(page, start, end, NULL,
+ uptodate);
if (!uptodate) {
ClearPageUptodate(page);
SetPageError(page);
- ret = ret < 0 ? ret : -EIO;
+ ret = err < 0 ? err : -EIO;
mapping_set_error(page->mapping, ret);
}
}
@@ -2467,11 +2446,13 @@ void end_extent_writepage(struct page *page, int err, u64 start, u64 end)
*/
static void end_bio_extent_writepage(struct bio *bio)
{
+ int error = blk_status_to_errno(bio->bi_status);
struct bio_vec *bvec;
u64 start;
u64 end;
int i;
+ ASSERT(!bio_flagged(bio, BIO_CLONED));
bio_for_each_segment_all(bvec, bio, i) {
struct page *page = bvec->bv_page;
struct inode *inode = page->mapping->host;
@@ -2496,7 +2477,7 @@ static void end_bio_extent_writepage(struct bio *bio)
start = page_offset(page);
end = start + bvec->bv_offset + bvec->bv_len - 1;
- end_extent_writepage(page, bio->bi_error, start, end);
+ end_extent_writepage(page, error, start, end);
end_page_writeback(page);
}
@@ -2529,9 +2510,9 @@ endio_readpage_release_extent(struct extent_io_tree *tree, u64 start, u64 len,
static void end_bio_extent_readpage(struct bio *bio)
{
struct bio_vec *bvec;
- int uptodate = !bio->bi_error;
+ int uptodate = !bio->bi_status;
struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
- struct extent_io_tree *tree;
+ struct extent_io_tree *tree, *failure_tree;
u64 offset = 0;
u64 start;
u64 end;
@@ -2542,6 +2523,7 @@ static void end_bio_extent_readpage(struct bio *bio)
int ret;
int i;
+ ASSERT(!bio_flagged(bio, BIO_CLONED));
bio_for_each_segment_all(bvec, bio, i) {
struct page *page = bvec->bv_page;
struct inode *inode = page->mapping->host;
@@ -2549,9 +2531,10 @@ static void end_bio_extent_readpage(struct bio *bio)
btrfs_debug(fs_info,
"end_bio_extent_readpage: bi_sector=%llu, err=%d, mirror=%u",
- (u64)bio->bi_iter.bi_sector, bio->bi_error,
+ (u64)bio->bi_iter.bi_sector, bio->bi_status,
io_bio->mirror_num);
tree = &BTRFS_I(inode)->io_tree;
+ failure_tree = &BTRFS_I(inode)->io_failure_tree;
/* We always issue full-page reads, but if some block
* in a page fails to read, blk_update_request() will
@@ -2574,42 +2557,54 @@ static void end_bio_extent_readpage(struct bio *bio)
len = bvec->bv_len;
mirror = io_bio->mirror_num;
- if (likely(uptodate && tree->ops &&
- tree->ops->readpage_end_io_hook)) {
+ if (likely(uptodate && tree->ops)) {
ret = tree->ops->readpage_end_io_hook(io_bio, offset,
page, start, end,
mirror);
if (ret)
uptodate = 0;
else
- clean_io_failure(inode, start, page, 0);
+ clean_io_failure(BTRFS_I(inode)->root->fs_info,
+ failure_tree, tree, start,
+ page,
+ btrfs_ino(BTRFS_I(inode)), 0);
}
if (likely(uptodate))
goto readpage_ok;
- if (tree->ops && tree->ops->readpage_io_failed_hook) {
+ if (tree->ops) {
ret = tree->ops->readpage_io_failed_hook(page, mirror);
- if (!ret && !bio->bi_error)
- uptodate = 1;
- } else {
+ if (ret == -EAGAIN) {
+ /*
+ * Data inode's readpage_io_failed_hook() always
+ * returns -EAGAIN.
+ *
+ * The generic bio_readpage_error handles errors
+ * the following way: If possible, new read
+ * requests are created and submitted and will
+ * end up in end_bio_extent_readpage as well (if
+ * we're lucky, not in the !uptodate case). In
+ * that case it returns 0 and we just go on with
+ * the next page in our bio. If it can't handle
+ * the error it will return -EIO and we remain
+ * responsible for that page.
+ */
+ ret = bio_readpage_error(bio, offset, page,
+ start, end, mirror);
+ if (ret == 0) {
+ uptodate = !bio->bi_status;
+ offset += len;
+ continue;
+ }
+ }
+
/*
- * The generic bio_readpage_error handles errors the
- * following way: If possible, new read requests are
- * created and submitted and will end up in
- * end_bio_extent_readpage as well (if we're lucky, not
- * in the !uptodate case). In that case it returns 0 and
- * we just go on with the next page in our bio. If it
- * can't handle the error it will return -EIO and we
- * remain responsible for that page.
+ * metadata's readpage_io_failed_hook() always returns
+ * -EIO and fixes nothing. -EIO is also returned if
+ * data inode error could not be fixed.
*/
- ret = bio_readpage_error(bio, offset, page, start, end,
- mirror);
- if (ret == 0) {
- uptodate = !bio->bi_error;
- offset += len;
- continue;
- }
+ ASSERT(ret == -EIO);
}
readpage_ok:
if (likely(uptodate)) {
@@ -2656,77 +2651,80 @@ readpage_ok:
endio_readpage_release_extent(tree, extent_start, extent_len,
uptodate);
if (io_bio->end_io)
- io_bio->end_io(io_bio, bio->bi_error);
+ io_bio->end_io(io_bio, blk_status_to_errno(bio->bi_status));
bio_put(bio);
}
/*
- * this allocates from the btrfs_bioset. We're returning a bio right now
- * but you can call btrfs_io_bio for the appropriate container_of magic
+ * Initialize the members up to but not including 'bio'. Use after allocating a
+ * new bio by bio_alloc_bioset as it does not initialize the bytes outside of
+ * 'bio' because use of __GFP_ZERO is not supported.
*/
-struct bio *
-btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
- gfp_t gfp_flags)
+static inline void btrfs_io_bio_init(struct btrfs_io_bio *btrfs_bio)
{
- struct btrfs_io_bio *btrfs_bio;
- struct bio *bio;
-
- bio = bio_alloc_bioset(gfp_flags, nr_vecs, btrfs_bioset);
+ memset(btrfs_bio, 0, offsetof(struct btrfs_io_bio, bio));
+}
- if (bio == NULL && (current->flags & PF_MEMALLOC)) {
- while (!bio && (nr_vecs /= 2)) {
- bio = bio_alloc_bioset(gfp_flags,
- nr_vecs, btrfs_bioset);
- }
- }
+/*
+ * The following helpers allocate a bio. As it's backed by a bioset, it'll
+ * never fail. We're returning a bio right now but you can call btrfs_io_bio
+ * for the appropriate container_of magic
+ */
+struct bio *btrfs_bio_alloc(struct block_device *bdev, u64 first_byte)
+{
+ struct bio *bio;
- if (bio) {
- bio->bi_bdev = bdev;
- bio->bi_iter.bi_sector = first_sector;
- btrfs_bio = btrfs_io_bio(bio);
- btrfs_bio->csum = NULL;
- btrfs_bio->csum_allocated = NULL;
- btrfs_bio->end_io = NULL;
- }
+ bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, btrfs_bioset);
+ bio->bi_bdev = bdev;
+ bio->bi_iter.bi_sector = first_byte >> 9;
+ btrfs_io_bio_init(btrfs_io_bio(bio));
return bio;
}
-struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask)
+struct bio *btrfs_bio_clone(struct bio *bio)
{
struct btrfs_io_bio *btrfs_bio;
struct bio *new;
- new = bio_clone_bioset(bio, gfp_mask, btrfs_bioset);
- if (new) {
- btrfs_bio = btrfs_io_bio(new);
- btrfs_bio->csum = NULL;
- btrfs_bio->csum_allocated = NULL;
- btrfs_bio->end_io = NULL;
- }
+ /* Bio allocation backed by a bioset does not fail */
+ new = bio_clone_fast(bio, GFP_NOFS, btrfs_bioset);
+ btrfs_bio = btrfs_io_bio(new);
+ btrfs_io_bio_init(btrfs_bio);
+ btrfs_bio->iter = bio->bi_iter;
return new;
}
-/* this also allocates from the btrfs_bioset */
-struct bio *btrfs_io_bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs)
+struct bio *btrfs_io_bio_alloc(unsigned int nr_iovecs)
{
- struct btrfs_io_bio *btrfs_bio;
struct bio *bio;
- bio = bio_alloc_bioset(gfp_mask, nr_iovecs, btrfs_bioset);
- if (bio) {
- btrfs_bio = btrfs_io_bio(bio);
- btrfs_bio->csum = NULL;
- btrfs_bio->csum_allocated = NULL;
- btrfs_bio->end_io = NULL;
- }
+ /* Bio allocation backed by a bioset does not fail */
+ bio = bio_alloc_bioset(GFP_NOFS, nr_iovecs, btrfs_bioset);
+ btrfs_io_bio_init(btrfs_io_bio(bio));
return bio;
}
+struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size)
+{
+ struct bio *bio;
+ struct btrfs_io_bio *btrfs_bio;
+
+ /* this will never fail when it's backed by a bioset */
+ bio = bio_clone_fast(orig, GFP_NOFS, btrfs_bioset);
+ ASSERT(bio);
+
+ btrfs_bio = btrfs_io_bio(bio);
+ btrfs_io_bio_init(btrfs_bio);
+
+ bio_trim(bio, offset >> 9, size >> 9);
+ btrfs_bio->iter = bio->bi_iter;
+ return bio;
+}
static int __must_check submit_one_bio(struct bio *bio, int mirror_num,
unsigned long bio_flags)
{
- int ret = 0;
+ blk_status_t ret = 0;
struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
struct page *page = bvec->bv_page;
struct extent_io_tree *tree = bio->bi_private;
@@ -2737,14 +2735,14 @@ static int __must_check submit_one_bio(struct bio *bio, int mirror_num,
bio->bi_private = NULL;
bio_get(bio);
- if (tree->ops && tree->ops->submit_bio_hook)
- ret = tree->ops->submit_bio_hook(page->mapping->host, bio,
+ if (tree->ops)
+ ret = tree->ops->submit_bio_hook(tree->private_data, bio,
mirror_num, bio_flags, start);
else
btrfsic_submit_bio(bio);
bio_put(bio);
- return ret;
+ return blk_status_to_errno(ret);
}
static int merge_bio(struct extent_io_tree *tree, struct page *page,
@@ -2752,7 +2750,7 @@ static int merge_bio(struct extent_io_tree *tree, struct page *page,
unsigned long bio_flags)
{
int ret = 0;
- if (tree->ops && tree->ops->merge_bio_hook)
+ if (tree->ops)
ret = tree->ops->merge_bio_hook(page, offset, size, bio,
bio_flags);
return ret;
@@ -2765,7 +2763,6 @@ static int submit_extent_page(int op, int op_flags, struct extent_io_tree *tree,
size_t size, unsigned long offset,
struct block_device *bdev,
struct bio **bio_ret,
- unsigned long max_pages,
bio_end_io_t end_io_func,
int mirror_num,
unsigned long prev_bio_flags,
@@ -2802,14 +2799,11 @@ static int submit_extent_page(int op, int op_flags, struct extent_io_tree *tree,
}
}
- bio = btrfs_bio_alloc(bdev, sector, BIO_MAX_PAGES,
- GFP_NOFS | __GFP_HIGH);
- if (!bio)
- return -ENOMEM;
-
+ bio = btrfs_bio_alloc(bdev, sector << 9);
bio_add_page(bio, page, page_size, offset);
bio->bi_end_io = end_io_func;
bio->bi_private = tree;
+ bio->bi_write_hint = page->mapping->host->i_write_hint;
bio_set_op_attrs(bio, op, op_flags);
if (wbc) {
wbc_init_bio(wbc, bio);
@@ -2856,7 +2850,7 @@ __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
em = *em_cached;
if (extent_map_in_tree(em) && start >= em->start &&
start < extent_map_end(em)) {
- atomic_inc(&em->refs);
+ refcount_inc(&em->refs);
return em;
}
@@ -2864,10 +2858,10 @@ __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
*em_cached = NULL;
}
- em = get_extent(inode, page, pg_offset, start, len, 0);
+ em = get_extent(BTRFS_I(inode), page, pg_offset, start, len, 0);
if (em_cached && !IS_ERR_OR_NULL(em)) {
BUG_ON(*em_cached);
- atomic_inc(&em->refs);
+ refcount_inc(&em->refs);
*em_cached = em;
}
return em;
@@ -2931,7 +2925,6 @@ static int __do_readpage(struct extent_io_tree *tree,
}
}
while (cur <= end) {
- unsigned long pnr = (last_byte >> PAGE_SHIFT) + 1;
bool force_bio_submit = false;
if (cur >= last_byte) {
@@ -3066,10 +3059,9 @@ static int __do_readpage(struct extent_io_tree *tree,
continue;
}
- pnr -= page->index;
ret = submit_extent_page(REQ_OP_READ, read_flags, tree, NULL,
page, sector, disk_io_size, pg_offset,
- bdev, bio, pnr,
+ bdev, bio,
end_bio_extent_readpage, mirror_num,
*bio_flags,
this_bio_flag,
@@ -3110,7 +3102,7 @@ static inline void __do_contiguous_readpages(struct extent_io_tree *tree,
inode = pages[0]->mapping->host;
while (1) {
lock_extent(tree, start, end);
- ordered = btrfs_lookup_ordered_range(inode, start,
+ ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start,
end - start + 1);
if (!ordered)
break;
@@ -3182,7 +3174,7 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
while (1) {
lock_extent(tree, start, end);
- ordered = btrfs_lookup_ordered_range(inode, start,
+ ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start,
PAGE_SIZE);
if (!ordered)
break;
@@ -3210,7 +3202,7 @@ int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
return ret;
}
-static void update_nr_written(struct page *page, struct writeback_control *wbc,
+static void update_nr_written(struct writeback_control *wbc,
unsigned long nr_written)
{
wbc->nr_to_write -= nr_written;
@@ -3330,7 +3322,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
u64 block_start;
u64 iosize;
sector_t sector;
- struct extent_state *cached_state = NULL;
struct extent_map *em;
struct block_device *bdev;
size_t pg_offset = 0;
@@ -3349,10 +3340,9 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
else
redirty_page_for_writepage(wbc, page);
- update_nr_written(page, wbc, nr_written);
+ update_nr_written(wbc, nr_written);
unlock_page(page);
- ret = 1;
- goto done_unlocked;
+ return 1;
}
}
@@ -3360,7 +3350,7 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
* we don't want to touch the inode after unlocking the page,
* so we update the mapping writeback index now
*/
- update_nr_written(page, wbc, nr_written + 1);
+ update_nr_written(wbc, nr_written + 1);
end = page_end;
if (i_size <= start) {
@@ -3374,7 +3364,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
while (cur <= end) {
u64 em_end;
- unsigned long max_nr;
if (cur >= i_size) {
if (tree->ops && tree->ops->writepage_end_io_hook)
@@ -3382,7 +3371,7 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
page_end, NULL, 1);
break;
}
- em = epd->get_extent(inode, page, pg_offset, cur,
+ em = epd->get_extent(BTRFS_I(inode), page, pg_offset, cur,
end - cur + 1, 1);
if (IS_ERR_OR_NULL(em)) {
SetPageError(page);
@@ -3431,8 +3420,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
continue;
}
- max_nr = (i_size >> PAGE_SHIFT) + 1;
-
set_range_writeback(tree, cur, cur + iosize - 1);
if (!PageWriteback(page)) {
btrfs_err(BTRFS_I(inode)->root->fs_info,
@@ -3442,11 +3429,14 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
ret = submit_extent_page(REQ_OP_WRITE, write_flags, tree, wbc,
page, sector, iosize, pg_offset,
- bdev, &epd->bio, max_nr,
+ bdev, &epd->bio,
end_bio_extent_writepage,
0, 0, 0, false);
- if (ret)
+ if (ret) {
SetPageError(page);
+ if (PageWriteback(page))
+ end_page_writeback(page);
+ }
cur = cur + iosize;
pg_offset += iosize;
@@ -3454,11 +3444,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
}
done:
*nr_ret = nr;
-
-done_unlocked:
-
- /* drop our reference on any cached states */
- free_extent_state(cached_state);
return ret;
}
@@ -3590,9 +3575,9 @@ lock_extent_buffer_for_io(struct extent_buffer *eb,
set_bit(EXTENT_BUFFER_WRITEBACK, &eb->bflags);
spin_unlock(&eb->refs_lock);
btrfs_set_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN);
- __percpu_counter_add(&fs_info->dirty_metadata_bytes,
- -eb->len,
- fs_info->dirty_metadata_batch);
+ percpu_counter_add_batch(&fs_info->dirty_metadata_bytes,
+ -eb->len,
+ fs_info->dirty_metadata_batch);
ret = 1;
} else {
spin_unlock(&eb->refs_lock);
@@ -3693,6 +3678,7 @@ static void end_bio_extent_buffer_writepage(struct bio *bio)
struct extent_buffer *eb;
int i, done;
+ ASSERT(!bio_flagged(bio, BIO_CLONED));
bio_for_each_segment_all(bvec, bio, i) {
struct page *page = bvec->bv_page;
@@ -3700,7 +3686,7 @@ static void end_bio_extent_buffer_writepage(struct bio *bio)
BUG_ON(!eb);
done = atomic_dec_and_test(&eb->io_pages);
- if (bio->bi_error ||
+ if (bio->bi_status ||
test_bit(EXTENT_BUFFER_WRITE_ERR, &eb->bflags)) {
ClearPageUptodate(page);
set_btree_ioerr(page);
@@ -3750,7 +3736,7 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
* header 0 1 2 .. N ... data_N .. data_2 data_1 data_0
*/
start = btrfs_item_nr_offset(nritems);
- end = btrfs_leaf_data(eb) + leaf_data_end(fs_info, eb);
+ end = BTRFS_LEAF_DATA_OFFSET + leaf_data_end(fs_info, eb);
memzero_extent_buffer(eb, start, end - start);
}
@@ -3761,20 +3747,21 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
set_page_writeback(p);
ret = submit_extent_page(REQ_OP_WRITE, write_flags, tree, wbc,
p, offset >> 9, PAGE_SIZE, 0, bdev,
- &epd->bio, -1,
+ &epd->bio,
end_bio_extent_buffer_writepage,
0, epd->bio_flags, bio_flags, false);
epd->bio_flags = bio_flags;
if (ret) {
set_btree_ioerr(p);
- end_page_writeback(p);
+ if (PageWriteback(p))
+ end_page_writeback(p);
if (atomic_sub_and_test(num_pages - i, &eb->io_pages))
end_extent_buffer_writeback(eb);
ret = -EIO;
break;
}
offset += PAGE_SIZE;
- update_nr_written(p, wbc, 1);
+ update_nr_written(wbc, 1);
unlock_page(p);
}
@@ -3926,8 +3913,7 @@ retry:
* WB_SYNC_ALL then we were called for data integrity and we must wait for
* existing IO to complete.
*/
-static int extent_write_cache_pages(struct extent_io_tree *tree,
- struct address_space *mapping,
+static int extent_write_cache_pages(struct address_space *mapping,
struct writeback_control *wbc,
writepage_t writepage, void *data,
void (*flush_fn)(void *))
@@ -4168,8 +4154,7 @@ int extent_writepages(struct extent_io_tree *tree,
.bio_flags = 0,
};
- ret = extent_write_cache_pages(tree, mapping, wbc,
- __extent_writepage, &epd,
+ ret = extent_write_cache_pages(mapping, wbc, __extent_writepage, &epd,
flush_write_bio);
flush_epd_write_bio(&epd);
return ret;
@@ -4264,8 +4249,6 @@ static int try_release_extent_state(struct extent_map_tree *map,
EXTENT_IOBITS, 0, NULL))
ret = 0;
else {
- if ((mask & GFP_NOFS) == GFP_NOFS)
- mask = GFP_NOFS;
/*
* at this point we can safely clear everything except the
* locked bit and the nodatasum bit
@@ -4354,7 +4337,7 @@ static struct extent_map *get_extent_skip_holes(struct inode *inode,
if (len == 0)
break;
len = ALIGN(len, sectorsize);
- em = get_extent(inode, NULL, 0, offset, len, 0);
+ em = get_extent(BTRFS_I(inode), NULL, 0, offset, len, 0);
if (IS_ERR_OR_NULL(em))
return em;
@@ -4373,6 +4356,119 @@ static struct extent_map *get_extent_skip_holes(struct inode *inode,
return NULL;
}
+/*
+ * To cache previous fiemap extent
+ *
+ * Will be used for merging fiemap extent
+ */
+struct fiemap_cache {
+ u64 offset;
+ u64 phys;
+ u64 len;
+ u32 flags;
+ bool cached;
+};
+
+/*
+ * Helper to submit fiemap extent.
+ *
+ * Will try to merge current fiemap extent specified by @offset, @phys,
+ * @len and @flags with cached one.
+ * And only when we fails to merge, cached one will be submitted as
+ * fiemap extent.
+ *
+ * Return value is the same as fiemap_fill_next_extent().
+ */
+static int emit_fiemap_extent(struct fiemap_extent_info *fieinfo,
+ struct fiemap_cache *cache,
+ u64 offset, u64 phys, u64 len, u32 flags)
+{
+ int ret = 0;
+
+ if (!cache->cached)
+ goto assign;
+
+ /*
+ * Sanity check, extent_fiemap() should have ensured that new
+ * fiemap extent won't overlap with cahced one.
+ * Not recoverable.
+ *
+ * NOTE: Physical address can overlap, due to compression
+ */
+ if (cache->offset + cache->len > offset) {
+ WARN_ON(1);
+ return -EINVAL;
+ }
+
+ /*
+ * Only merges fiemap extents if
+ * 1) Their logical addresses are continuous
+ *
+ * 2) Their physical addresses are continuous
+ * So truly compressed (physical size smaller than logical size)
+ * extents won't get merged with each other
+ *
+ * 3) Share same flags except FIEMAP_EXTENT_LAST
+ * So regular extent won't get merged with prealloc extent
+ */
+ if (cache->offset + cache->len == offset &&
+ cache->phys + cache->len == phys &&
+ (cache->flags & ~FIEMAP_EXTENT_LAST) ==
+ (flags & ~FIEMAP_EXTENT_LAST)) {
+ cache->len += len;
+ cache->flags |= flags;
+ goto try_submit_last;
+ }
+
+ /* Not mergeable, need to submit cached one */
+ ret = fiemap_fill_next_extent(fieinfo, cache->offset, cache->phys,
+ cache->len, cache->flags);
+ cache->cached = false;
+ if (ret)
+ return ret;
+assign:
+ cache->cached = true;
+ cache->offset = offset;
+ cache->phys = phys;
+ cache->len = len;
+ cache->flags = flags;
+try_submit_last:
+ if (cache->flags & FIEMAP_EXTENT_LAST) {
+ ret = fiemap_fill_next_extent(fieinfo, cache->offset,
+ cache->phys, cache->len, cache->flags);
+ cache->cached = false;
+ }
+ return ret;
+}
+
+/*
+ * Emit last fiemap cache
+ *
+ * The last fiemap cache may still be cached in the following case:
+ * 0 4k 8k
+ * |<- Fiemap range ->|
+ * |<------------ First extent ----------->|
+ *
+ * In this case, the first extent range will be cached but not emitted.
+ * So we must emit it before ending extent_fiemap().
+ */
+static int emit_last_fiemap_cache(struct btrfs_fs_info *fs_info,
+ struct fiemap_extent_info *fieinfo,
+ struct fiemap_cache *cache)
+{
+ int ret;
+
+ if (!cache->cached)
+ return 0;
+
+ ret = fiemap_fill_next_extent(fieinfo, cache->offset, cache->phys,
+ cache->len, cache->flags);
+ cache->cached = false;
+ if (ret > 0)
+ ret = 0;
+ return ret;
+}
+
int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
__u64 start, __u64 len, get_extent_t *get_extent)
{
@@ -4390,6 +4486,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
struct extent_state *cached_state = NULL;
struct btrfs_path *path;
struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct fiemap_cache cache = { 0 };
int end = 0;
u64 em_start = 0;
u64 em_len = 0;
@@ -4410,8 +4507,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
* lookup the last file extent. We're not using i_size here
* because there might be preallocation past i_size
*/
- ret = btrfs_lookup_file_extent(NULL, root, path, btrfs_ino(inode), -1,
- 0);
+ ret = btrfs_lookup_file_extent(NULL, root, path,
+ btrfs_ino(BTRFS_I(inode)), -1, 0);
if (ret < 0) {
btrfs_free_path(path);
return ret;
@@ -4426,7 +4523,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
found_type = found_key.type;
/* No extents, but there might be delalloc bits */
- if (found_key.objectid != btrfs_ino(inode) ||
+ if (found_key.objectid != btrfs_ino(BTRFS_I(inode)) ||
found_type != BTRFS_EXTENT_DATA_KEY) {
/* have to trust i_size as the end */
last = (u64)-1;
@@ -4535,8 +4632,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
* lookup stuff.
*/
ret = btrfs_check_shared(trans, root->fs_info,
- root->objectid,
- btrfs_ino(inode), bytenr);
+ root->objectid,
+ btrfs_ino(BTRFS_I(inode)), bytenr);
if (trans)
btrfs_end_transaction(trans);
if (ret < 0)
@@ -4569,8 +4666,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
flags |= FIEMAP_EXTENT_LAST;
end = 1;
}
- ret = fiemap_fill_next_extent(fieinfo, em_start, disko,
- em_len, flags);
+ ret = emit_fiemap_extent(fieinfo, &cache, em_start, disko,
+ em_len, flags);
if (ret) {
if (ret == 1)
ret = 0;
@@ -4578,6 +4675,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
}
}
out_free:
+ if (!ret)
+ ret = emit_last_fiemap_cache(root->fs_info, fieinfo, &cache);
free_extent_map(em);
out:
btrfs_free_path(path);
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 17f9ce479ed7..4f030912f3ef 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -2,6 +2,7 @@
#define __EXTENTIO__
#include <linux/rbtree.h>
+#include <linux/refcount.h>
#include "ulist.h"
/* bits for the extent state */
@@ -14,14 +15,17 @@
#define EXTENT_DEFRAG (1U << 6)
#define EXTENT_BOUNDARY (1U << 9)
#define EXTENT_NODATASUM (1U << 10)
-#define EXTENT_DO_ACCOUNTING (1U << 11)
+#define EXTENT_CLEAR_META_RESV (1U << 11)
#define EXTENT_FIRST_DELALLOC (1U << 12)
#define EXTENT_NEED_WAIT (1U << 13)
#define EXTENT_DAMAGED (1U << 14)
#define EXTENT_NORESERVE (1U << 15)
#define EXTENT_QGROUP_RESERVED (1U << 16)
#define EXTENT_CLEAR_DATA_RESV (1U << 17)
+#define EXTENT_DELALLOC_NEW (1U << 18)
#define EXTENT_IOBITS (EXTENT_LOCKED | EXTENT_WRITEBACK)
+#define EXTENT_DO_ACCOUNTING (EXTENT_CLEAR_META_RESV | \
+ EXTENT_CLEAR_DATA_RESV)
#define EXTENT_CTLBITS (EXTENT_DO_ACCOUNTING | EXTENT_FIRST_DELALLOC)
/*
@@ -45,13 +49,14 @@
#define EXTENT_BUFFER_IN_TREE 10
#define EXTENT_BUFFER_WRITE_ERR 11 /* write IO error */
-/* these are flags for extent_clear_unlock_delalloc */
+/* these are flags for __process_pages_contig */
#define PAGE_UNLOCK (1 << 0)
#define PAGE_CLEAR_DIRTY (1 << 1)
#define PAGE_SET_WRITEBACK (1 << 2)
#define PAGE_END_WRITEBACK (1 << 3)
#define PAGE_SET_PRIVATE2 (1 << 4)
#define PAGE_SET_ERROR (1 << 5)
+#define PAGE_LOCK (1 << 6)
/*
* page->private values. Every page that is controlled by the extent
@@ -83,41 +88,56 @@ extern void le_bitmap_clear(u8 *map, unsigned int start, int len);
struct extent_state;
struct btrfs_root;
+struct btrfs_inode;
struct btrfs_io_bio;
struct io_failure_record;
-typedef int (extent_submit_bio_hook_t)(struct inode *inode, struct bio *bio,
+typedef blk_status_t (extent_submit_bio_hook_t)(void *private_data, struct bio *bio,
int mirror_num, unsigned long bio_flags,
u64 bio_offset);
struct extent_io_ops {
- int (*fill_delalloc)(struct inode *inode, struct page *locked_page,
- u64 start, u64 end, int *page_started,
- unsigned long *nr_written);
- int (*writepage_start_hook)(struct page *page, u64 start, u64 end);
+ /*
+ * The following callbacks must be allways defined, the function
+ * pointer will be called unconditionally.
+ */
extent_submit_bio_hook_t *submit_bio_hook;
+ int (*readpage_end_io_hook)(struct btrfs_io_bio *io_bio, u64 phy_offset,
+ struct page *page, u64 start, u64 end,
+ int mirror);
int (*merge_bio_hook)(struct page *page, unsigned long offset,
size_t size, struct bio *bio,
unsigned long bio_flags);
int (*readpage_io_failed_hook)(struct page *page, int failed_mirror);
- int (*readpage_end_io_hook)(struct btrfs_io_bio *io_bio, u64 phy_offset,
- struct page *page, u64 start, u64 end,
- int mirror);
- int (*writepage_end_io_hook)(struct page *page, u64 start, u64 end,
+ struct btrfs_fs_info *(*tree_fs_info)(void *private_data);
+ void (*set_range_writeback)(void *private_data, u64 start, u64 end);
+
+ /*
+ * Optional hooks, called if the pointer is not NULL
+ */
+ int (*fill_delalloc)(void *private_data, struct page *locked_page,
+ u64 start, u64 end, int *page_started,
+ unsigned long *nr_written);
+
+ int (*writepage_start_hook)(struct page *page, u64 start, u64 end);
+ void (*writepage_end_io_hook)(struct page *page, u64 start, u64 end,
struct extent_state *state, int uptodate);
- void (*set_bit_hook)(struct inode *inode, struct extent_state *state,
+ void (*set_bit_hook)(void *private_data, struct extent_state *state,
unsigned *bits);
- void (*clear_bit_hook)(struct inode *inode, struct extent_state *state,
- unsigned *bits);
- void (*merge_extent_hook)(struct inode *inode,
+ void (*clear_bit_hook)(void *private_data,
+ struct extent_state *state,
+ unsigned *bits);
+ void (*merge_extent_hook)(void *private_data,
struct extent_state *new,
struct extent_state *other);
- void (*split_extent_hook)(struct inode *inode,
+ void (*split_extent_hook)(void *private_data,
struct extent_state *orig, u64 split);
+ void (*check_extent_io_range)(void *private_data, const char *caller,
+ u64 start, u64 end);
};
struct extent_io_tree {
struct rb_root state;
- struct address_space *mapping;
+ void *private_data;
u64 dirty_bytes;
int track_uptodate;
spinlock_t lock;
@@ -131,7 +151,7 @@ struct extent_state {
/* ADD NEW ELEMENTS AFTER THIS */
wait_queue_head_t wq;
- atomic_t refs;
+ refcount_t refs;
unsigned state;
struct io_failure_record *failrec;
@@ -189,12 +209,46 @@ struct extent_buffer {
*/
struct extent_changeset {
/* How many bytes are set/cleared in this operation */
- u64 bytes_changed;
+ unsigned int bytes_changed;
/* Changed ranges */
- struct ulist *range_changed;
+ struct ulist range_changed;
};
+static inline void extent_changeset_init(struct extent_changeset *changeset)
+{
+ changeset->bytes_changed = 0;
+ ulist_init(&changeset->range_changed);
+}
+
+static inline struct extent_changeset *extent_changeset_alloc(void)
+{
+ struct extent_changeset *ret;
+
+ ret = kmalloc(sizeof(*ret), GFP_KERNEL);
+ if (!ret)
+ return NULL;
+
+ extent_changeset_init(ret);
+ return ret;
+}
+
+static inline void extent_changeset_release(struct extent_changeset *changeset)
+{
+ if (!changeset)
+ return;
+ changeset->bytes_changed = 0;
+ ulist_release(&changeset->range_changed);
+}
+
+static inline void extent_changeset_free(struct extent_changeset *changeset)
+{
+ if (!changeset)
+ return;
+ extent_changeset_release(changeset);
+ kfree(changeset);
+}
+
static inline void extent_set_compress_type(unsigned long *bio_flags,
int compress_type)
{
@@ -208,14 +262,13 @@ static inline int extent_compress_type(unsigned long bio_flags)
struct extent_map_tree;
-typedef struct extent_map *(get_extent_t)(struct inode *inode,
+typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode,
struct page *page,
size_t pg_offset,
u64 start, u64 len,
int create);
-void extent_io_tree_init(struct extent_io_tree *tree,
- struct address_space *mapping);
+void extent_io_tree_init(struct extent_io_tree *tree, void *private_data);
int try_release_extent_mapping(struct extent_map_tree *map,
struct extent_io_tree *tree, struct page *page,
gfp_t mask);
@@ -443,19 +496,21 @@ void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
u64 delalloc_end, struct page *locked_page,
unsigned bits_to_clear,
unsigned long page_ops);
-struct bio *
-btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
- gfp_t gfp_flags);
-struct bio *btrfs_io_bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs);
-struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask);
+struct bio *btrfs_bio_alloc(struct block_device *bdev, u64 first_byte);
+struct bio *btrfs_io_bio_alloc(unsigned int nr_iovecs);
+struct bio *btrfs_bio_clone(struct bio *bio);
+struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size);
struct btrfs_fs_info;
-
-int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
- struct page *page, unsigned int pg_offset,
- int mirror_num);
-int clean_io_failure(struct inode *inode, u64 start, struct page *page,
- unsigned int pg_offset);
+struct btrfs_inode;
+
+int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
+ u64 length, u64 logical, struct page *page,
+ unsigned int pg_offset, int mirror_num);
+int clean_io_failure(struct btrfs_fs_info *fs_info,
+ struct extent_io_tree *failure_tree,
+ struct extent_io_tree *io_tree, u64 start,
+ struct page *page, u64 ino, unsigned int pg_offset);
void end_extent_writepage(struct page *page, int err, u64 start, u64 end);
int repair_eb_io_failure(struct btrfs_fs_info *fs_info,
struct extent_buffer *eb, int mirror_num);
@@ -479,16 +534,20 @@ struct io_failure_record {
int in_validation;
};
-void btrfs_free_io_failure_record(struct inode *inode, u64 start, u64 end);
+
+void btrfs_free_io_failure_record(struct btrfs_inode *inode, u64 start,
+ u64 end);
int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
struct io_failure_record **failrec_ret);
-int btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
- struct io_failure_record *failrec, int fail_mirror);
+bool btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
+ struct io_failure_record *failrec, int fail_mirror);
struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
struct io_failure_record *failrec,
struct page *page, int pg_offset, int icsum,
bio_end_io_t *endio_func, void *data);
-int free_io_failure(struct inode *inode, struct io_failure_record *rec);
+int free_io_failure(struct extent_io_tree *failure_tree,
+ struct extent_io_tree *io_tree,
+ struct io_failure_record *rec);
#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
noinline u64 find_lock_delalloc_range(struct inode *inode,
struct extent_io_tree *tree,
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 26f9ac719d20..69850155870c 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -55,7 +55,7 @@ struct extent_map *alloc_extent_map(void)
em->flags = 0;
em->compress_type = BTRFS_COMPRESS_NONE;
em->generation = 0;
- atomic_set(&em->refs, 1);
+ refcount_set(&em->refs, 1);
INIT_LIST_HEAD(&em->list);
return em;
}
@@ -71,8 +71,8 @@ void free_extent_map(struct extent_map *em)
{
if (!em)
return;
- WARN_ON(atomic_read(&em->refs) == 0);
- if (atomic_dec_and_test(&em->refs)) {
+ WARN_ON(refcount_read(&em->refs) == 0);
+ if (refcount_dec_and_test(&em->refs)) {
WARN_ON(extent_map_in_tree(em));
WARN_ON(!list_empty(&em->list));
if (test_bit(EXTENT_FLAG_FS_MAPPING, &em->flags))
@@ -322,7 +322,7 @@ static inline void setup_extent_mapping(struct extent_map_tree *tree,
struct extent_map *em,
int modified)
{
- atomic_inc(&em->refs);
+ refcount_inc(&em->refs);
em->mod_start = em->start;
em->mod_len = em->len;
@@ -381,7 +381,7 @@ __lookup_extent_mapping(struct extent_map_tree *tree,
if (strict && !(end > em->start && start < extent_map_end(em)))
return NULL;
- atomic_inc(&em->refs);
+ refcount_inc(&em->refs);
return em;
}
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index eb8b8fae036b..a67b2def5413 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -2,6 +2,7 @@
#define __EXTENTMAP__
#include <linux/rbtree.h>
+#include <linux/refcount.h>
#define EXTENT_MAP_LAST_BYTE ((u64)-4)
#define EXTENT_MAP_HOLE ((u64)-3)
@@ -41,7 +42,7 @@ struct extent_map {
*/
struct map_lookup *map_lookup;
};
- atomic_t refs;
+ refcount_t refs;
unsigned int compress_type;
struct list_head list;
};
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index e97e322c28f0..fdcb41002623 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -160,11 +160,12 @@ static void btrfs_io_bio_endio_readpage(struct btrfs_io_bio *bio, int err)
kfree(bio->csum_allocated);
}
-static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
+static blk_status_t __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
u64 logical_offset, u32 *dst, int dio)
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct bio_vec *bvec;
+ struct bio_vec bvec;
+ struct bvec_iter iter;
struct btrfs_io_bio *btrfs_bio = btrfs_io_bio(bio);
struct btrfs_csum_item *item = NULL;
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
@@ -177,12 +178,12 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
u64 page_bytes_left;
u32 diff;
int nblocks;
- int count = 0, i;
+ int count = 0;
u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
path = btrfs_alloc_path();
if (!path)
- return -ENOMEM;
+ return BLK_STS_RESOURCE;
nblocks = bio->bi_iter.bi_size >> inode->i_sb->s_blocksize_bits;
if (!dst) {
@@ -191,7 +192,7 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
csum_size, GFP_NOFS);
if (!btrfs_bio->csum_allocated) {
btrfs_free_path(path);
- return -ENOMEM;
+ return BLK_STS_RESOURCE;
}
btrfs_bio->csum = btrfs_bio->csum_allocated;
btrfs_bio->end_io = btrfs_io_bio_endio_readpage;
@@ -206,15 +207,13 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
if (bio->bi_iter.bi_size > PAGE_SIZE * 8)
path->reada = READA_FORWARD;
- WARN_ON(bio->bi_vcnt <= 0);
-
/*
* the free space stuff is only read when it hasn't been
* updated in the current transaction. So, we can safely
* read from the commit root and sidestep a nasty deadlock
* between reading the free space cache and updating the csum tree.
*/
- if (btrfs_is_free_space_inode(inode)) {
+ if (btrfs_is_free_space_inode(BTRFS_I(inode))) {
path->search_commit_root = 1;
path->skip_locking = 1;
}
@@ -223,13 +222,13 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
if (dio)
offset = logical_offset;
- bio_for_each_segment_all(bvec, bio, i) {
- page_bytes_left = bvec->bv_len;
+ bio_for_each_segment(bvec, bio, iter) {
+ page_bytes_left = bvec.bv_len;
if (count)
goto next;
if (!dio)
- offset = page_offset(bvec->bv_page) + bvec->bv_offset;
+ offset = page_offset(bvec.bv_page) + bvec.bv_offset;
count = btrfs_find_ordered_sum(inode, offset, disk_bytenr,
(u32 *)csum, nblocks);
if (count)
@@ -255,7 +254,7 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
} else {
btrfs_info_rl(fs_info,
"no csum found for inode %llu start %llu",
- btrfs_ino(inode), offset);
+ btrfs_ino(BTRFS_I(inode)), offset);
}
item = NULL;
btrfs_release_path(path);
@@ -303,12 +302,12 @@ next:
return 0;
}
-int btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio, u32 *dst)
+blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio, u32 *dst)
{
return __btrfs_lookup_bio_sums(inode, bio, 0, dst, 0);
}
-int btrfs_lookup_bio_sums_dio(struct inode *inode, struct bio *bio, u64 offset)
+blk_status_t btrfs_lookup_bio_sums_dio(struct inode *inode, struct bio *bio, u64 offset)
{
return __btrfs_lookup_bio_sums(inode, bio, offset, NULL, 1);
}
@@ -433,26 +432,26 @@ fail:
return ret;
}
-int btrfs_csum_one_bio(struct inode *inode, struct bio *bio,
+blk_status_t btrfs_csum_one_bio(struct inode *inode, struct bio *bio,
u64 file_start, int contig)
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
struct btrfs_ordered_sum *sums;
struct btrfs_ordered_extent *ordered = NULL;
char *data;
- struct bio_vec *bvec;
+ struct bvec_iter iter;
+ struct bio_vec bvec;
int index;
int nr_sectors;
- int i, j;
unsigned long total_bytes = 0;
unsigned long this_sum_bytes = 0;
+ int i;
u64 offset;
- WARN_ON(bio->bi_vcnt <= 0);
sums = kzalloc(btrfs_ordered_sum_size(fs_info, bio->bi_iter.bi_size),
GFP_NOFS);
if (!sums)
- return -ENOMEM;
+ return BLK_STS_RESOURCE;
sums->len = bio->bi_iter.bi_size;
INIT_LIST_HEAD(&sums->list);
@@ -465,19 +464,19 @@ int btrfs_csum_one_bio(struct inode *inode, struct bio *bio,
sums->bytenr = (u64)bio->bi_iter.bi_sector << 9;
index = 0;
- bio_for_each_segment_all(bvec, bio, j) {
+ bio_for_each_segment(bvec, bio, iter) {
if (!contig)
- offset = page_offset(bvec->bv_page) + bvec->bv_offset;
+ offset = page_offset(bvec.bv_page) + bvec.bv_offset;
if (!ordered) {
ordered = btrfs_lookup_ordered_extent(inode, offset);
BUG_ON(!ordered); /* Logic error */
}
- data = kmap_atomic(bvec->bv_page);
+ data = kmap_atomic(bvec.bv_page);
nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info,
- bvec->bv_len + fs_info->sectorsize
+ bvec.bv_len + fs_info->sectorsize
- 1);
for (i = 0; i < nr_sectors; i++) {
@@ -504,12 +503,12 @@ int btrfs_csum_one_bio(struct inode *inode, struct bio *bio,
+ total_bytes;
index = 0;
- data = kmap_atomic(bvec->bv_page);
+ data = kmap_atomic(bvec.bv_page);
}
sums->sums[index] = ~(u32)0;
sums->sums[index]
- = btrfs_csum_data(data + bvec->bv_offset
+ = btrfs_csum_data(data + bvec.bv_offset
+ (i * fs_info->sectorsize),
sums->sums[index],
fs_info->sectorsize);
@@ -643,7 +642,33 @@ int btrfs_del_csums(struct btrfs_trans_handle *trans,
/* delete the entire item, it is inside our range */
if (key.offset >= bytenr && csum_end <= end_byte) {
- ret = btrfs_del_item(trans, root, path);
+ int del_nr = 1;
+
+ /*
+ * Check how many csum items preceding this one in this
+ * leaf correspond to our range and then delete them all
+ * at once.
+ */
+ if (key.offset > bytenr && path->slots[0] > 0) {
+ int slot = path->slots[0] - 1;
+
+ while (slot >= 0) {
+ struct btrfs_key pk;
+
+ btrfs_item_key_to_cpu(leaf, &pk, slot);
+ if (pk.offset < bytenr ||
+ pk.type != BTRFS_EXTENT_CSUM_KEY ||
+ pk.objectid !=
+ BTRFS_EXTENT_CSUM_OBJECTID)
+ break;
+ path->slots[0] = slot;
+ del_nr++;
+ key.offset = pk.offset;
+ slot--;
+ }
+ }
+ ret = btrfs_del_items(trans, root, path,
+ path->slots[0], del_nr);
if (ret)
goto out;
if (key.offset == bytenr)
@@ -856,8 +881,8 @@ insert:
tmp = min(tmp, (next_offset - file_key.offset) >>
fs_info->sb->s_blocksize_bits);
- tmp = max((u64)1, tmp);
- tmp = min(tmp, (u64)MAX_CSUM_ITEMS(fs_info, csum_size));
+ tmp = max_t(u64, 1, tmp);
+ tmp = min_t(u64, tmp, MAX_CSUM_ITEMS(fs_info, csum_size));
ins_size = csum_size * tmp;
} else {
ins_size = csum_size;
@@ -904,14 +929,14 @@ fail_unlock:
goto out;
}
-void btrfs_extent_item_to_extent_map(struct inode *inode,
+void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
const struct btrfs_path *path,
struct btrfs_file_extent_item *fi,
const bool new_inline,
struct extent_map *em)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = inode->root;
struct extent_buffer *leaf = path->nodes[0];
const int slot = path->slots[0];
struct btrfs_key key;
@@ -976,8 +1001,8 @@ void btrfs_extent_item_to_extent_map(struct inode *inode,
}
} else {
btrfs_err(fs_info,
- "unknown file extent item type %d, inode %llu, offset %llu, root %llu",
- type, btrfs_ino(inode), extent_start,
+ "unknown file extent item type %d, inode %llu, offset %llu, "
+ "root %llu", type, btrfs_ino(inode), extent_start,
root->root_key.objectid);
}
}
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index b5c5da215d05..9e75d8a39aac 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -92,10 +92,10 @@ static int __compare_inode_defrag(struct inode_defrag *defrag1,
* If an existing record is found the defrag item you
* pass in is freed
*/
-static int __btrfs_add_inode_defrag(struct inode *inode,
+static int __btrfs_add_inode_defrag(struct btrfs_inode *inode,
struct inode_defrag *defrag)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
struct inode_defrag *entry;
struct rb_node **p;
struct rb_node *parent = NULL;
@@ -123,7 +123,7 @@ static int __btrfs_add_inode_defrag(struct inode *inode,
return -EEXIST;
}
}
- set_bit(BTRFS_INODE_IN_DEFRAG, &BTRFS_I(inode)->runtime_flags);
+ set_bit(BTRFS_INODE_IN_DEFRAG, &inode->runtime_flags);
rb_link_node(&defrag->rb_node, parent, p);
rb_insert_color(&defrag->rb_node, &fs_info->defrag_inodes);
return 0;
@@ -145,10 +145,10 @@ static inline int __need_auto_defrag(struct btrfs_fs_info *fs_info)
* enabled
*/
int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
- struct inode *inode)
+ struct btrfs_inode *inode)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = inode->root;
struct inode_defrag *defrag;
u64 transid;
int ret;
@@ -156,13 +156,13 @@ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
if (!__need_auto_defrag(fs_info))
return 0;
- if (test_bit(BTRFS_INODE_IN_DEFRAG, &BTRFS_I(inode)->runtime_flags))
+ if (test_bit(BTRFS_INODE_IN_DEFRAG, &inode->runtime_flags))
return 0;
if (trans)
transid = trans->transid;
else
- transid = BTRFS_I(inode)->root->last_trans;
+ transid = inode->root->last_trans;
defrag = kmem_cache_zalloc(btrfs_inode_defrag_cachep, GFP_NOFS);
if (!defrag)
@@ -173,7 +173,7 @@ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
defrag->root = root->root_key.objectid;
spin_lock(&fs_info->defrag_inodes_lock);
- if (!test_bit(BTRFS_INODE_IN_DEFRAG, &BTRFS_I(inode)->runtime_flags)) {
+ if (!test_bit(BTRFS_INODE_IN_DEFRAG, &inode->runtime_flags)) {
/*
* If we set IN_DEFRAG flag and evict the inode from memory,
* and then re-read this inode, this new inode doesn't have
@@ -194,10 +194,10 @@ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
* the same inode in the tree, we will merge them together (by
* __btrfs_add_inode_defrag()) and free the one that we want to requeue.
*/
-static void btrfs_requeue_inode_defrag(struct inode *inode,
+static void btrfs_requeue_inode_defrag(struct btrfs_inode *inode,
struct inode_defrag *defrag)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
int ret;
if (!__need_auto_defrag(fs_info))
@@ -334,7 +334,7 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
*/
if (num_defrag == BTRFS_DEFRAG_BATCH) {
defrag->last_offset = range.start;
- btrfs_requeue_inode_defrag(inode, defrag);
+ btrfs_requeue_inode_defrag(BTRFS_I(inode), defrag);
} else if (defrag->last_offset && !defrag->cycled) {
/*
* we didn't fill our defrag batch, but
@@ -343,7 +343,7 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
*/
defrag->last_offset = 0;
defrag->cycled = 1;
- btrfs_requeue_inode_defrag(inode, defrag);
+ btrfs_requeue_inode_defrag(BTRFS_I(inode), defrag);
} else {
kmem_cache_free(btrfs_inode_defrag_cachep, defrag);
}
@@ -529,13 +529,13 @@ int btrfs_dirty_pages(struct inode *inode, struct page **pages,
* this drops all the extents in the cache that intersect the range
* [start, end]. Existing extents are split as required.
*/
-void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
+void btrfs_drop_extent_cache(struct btrfs_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;
+ struct extent_map_tree *em_tree = &inode->extent_tree;
u64 len = end - start + 1;
u64 gen;
int ret;
@@ -702,7 +702,7 @@ int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
struct btrfs_file_extent_item *fi;
struct btrfs_key key;
struct btrfs_key new_key;
- u64 ino = btrfs_ino(inode);
+ u64 ino = btrfs_ino(BTRFS_I(inode));
u64 search_start = start;
u64 disk_bytenr = 0;
u64 num_bytes = 0;
@@ -720,7 +720,7 @@ int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
int leafs_visited = 0;
if (drop_cache)
- btrfs_drop_extent_cache(inode, start, end - 1, 0);
+ btrfs_drop_extent_cache(BTRFS_I(inode), start, end - 1, 0);
if (start >= BTRFS_I(inode)->disk_i_size && !replace_extent)
modify_tree = 0;
@@ -1082,10 +1082,10 @@ static int extent_mergeable(struct extent_buffer *leaf, int slot,
* two or three.
*/
int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
- struct inode *inode, u64 start, u64 end)
+ struct btrfs_inode *inode, u64 start, u64 end)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = inode->root;
struct extent_buffer *leaf;
struct btrfs_path *path;
struct btrfs_file_extent_item *fi;
@@ -1404,6 +1404,47 @@ fail:
}
+static int btrfs_find_new_delalloc_bytes(struct btrfs_inode *inode,
+ const u64 start,
+ const u64 len,
+ struct extent_state **cached_state)
+{
+ u64 search_start = start;
+ const u64 end = start + len - 1;
+
+ while (search_start < end) {
+ const u64 search_len = end - search_start + 1;
+ struct extent_map *em;
+ u64 em_len;
+ int ret = 0;
+
+ em = btrfs_get_extent(inode, NULL, 0, search_start,
+ search_len, 0);
+ if (IS_ERR(em))
+ return PTR_ERR(em);
+
+ if (em->block_start != EXTENT_MAP_HOLE)
+ goto next;
+
+ em_len = em->len;
+ if (em->start < search_start)
+ em_len -= search_start - em->start;
+ if (em_len > search_len)
+ em_len = search_len;
+
+ ret = set_extent_bit(&inode->io_tree, search_start,
+ search_start + em_len - 1,
+ EXTENT_DELALLOC_NEW,
+ NULL, cached_state, GFP_NOFS);
+next:
+ search_start = extent_map_end(em);
+ free_extent_map(em);
+ if (ret)
+ return ret;
+ }
+ return 0;
+}
+
/*
* This function locks the extent and properly waits for data=ordered extents
* to finish before allowing the pages to be modified if need.
@@ -1415,13 +1456,13 @@ fail:
* the other < 0 number - Something wrong happens
*/
static noinline int
-lock_and_cleanup_extent_if_need(struct inode *inode, struct page **pages,
+lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
size_t num_pages, loff_t pos,
size_t write_bytes,
u64 *lockstart, u64 *lockend,
struct extent_state **cached_state)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
u64 start_pos;
u64 last_pos;
int i;
@@ -1432,33 +1473,44 @@ lock_and_cleanup_extent_if_need(struct inode *inode, struct page **pages,
+ round_up(pos + write_bytes - start_pos,
fs_info->sectorsize) - 1;
- if (start_pos < inode->i_size) {
+ if (start_pos < inode->vfs_inode.i_size ||
+ (inode->flags & BTRFS_INODE_PREALLOC)) {
struct btrfs_ordered_extent *ordered;
- lock_extent_bits(&BTRFS_I(inode)->io_tree,
- start_pos, last_pos, cached_state);
+ unsigned int clear_bits;
+
+ lock_extent_bits(&inode->io_tree, start_pos, last_pos,
+ cached_state);
ordered = btrfs_lookup_ordered_range(inode, start_pos,
last_pos - start_pos + 1);
if (ordered &&
ordered->file_offset + ordered->len > start_pos &&
ordered->file_offset <= last_pos) {
- unlock_extent_cached(&BTRFS_I(inode)->io_tree,
- start_pos, last_pos,
- cached_state, GFP_NOFS);
+ unlock_extent_cached(&inode->io_tree, start_pos,
+ last_pos, cached_state, GFP_NOFS);
for (i = 0; i < num_pages; i++) {
unlock_page(pages[i]);
put_page(pages[i]);
}
- btrfs_start_ordered_extent(inode, ordered, 1);
+ btrfs_start_ordered_extent(&inode->vfs_inode,
+ ordered, 1);
btrfs_put_ordered_extent(ordered);
return -EAGAIN;
}
if (ordered)
btrfs_put_ordered_extent(ordered);
-
- clear_extent_bit(&BTRFS_I(inode)->io_tree, start_pos,
- last_pos, EXTENT_DIRTY | EXTENT_DELALLOC |
- EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
- 0, 0, cached_state, GFP_NOFS);
+ ret = btrfs_find_new_delalloc_bytes(inode, start_pos,
+ last_pos - start_pos + 1,
+ cached_state);
+ clear_bits = EXTENT_DIRTY | EXTENT_DELALLOC |
+ EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG;
+ if (ret)
+ clear_bits |= EXTENT_DELALLOC_NEW | EXTENT_LOCKED;
+ clear_extent_bit(&inode->io_tree, start_pos,
+ last_pos, clear_bits,
+ (clear_bits & EXTENT_LOCKED) ? 1 : 0,
+ 0, cached_state, GFP_NOFS);
+ if (ret)
+ return ret;
*lockstart = start_pos;
*lockend = last_pos;
ret = 1;
@@ -1474,11 +1526,11 @@ lock_and_cleanup_extent_if_need(struct inode *inode, struct page **pages,
return ret;
}
-static noinline int check_can_nocow(struct inode *inode, loff_t pos,
+static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos,
size_t *write_bytes)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = inode->root;
struct btrfs_ordered_extent *ordered;
u64 lockstart, lockend;
u64 num_bytes;
@@ -1493,19 +1545,20 @@ static noinline int check_can_nocow(struct inode *inode, loff_t pos,
fs_info->sectorsize) - 1;
while (1) {
- lock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend);
+ lock_extent(&inode->io_tree, lockstart, lockend);
ordered = btrfs_lookup_ordered_range(inode, lockstart,
lockend - lockstart + 1);
if (!ordered) {
break;
}
- unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend);
- btrfs_start_ordered_extent(inode, ordered, 1);
+ unlock_extent(&inode->io_tree, lockstart, lockend);
+ btrfs_start_ordered_extent(&inode->vfs_inode, ordered, 1);
btrfs_put_ordered_extent(ordered);
}
num_bytes = lockend - lockstart + 1;
- ret = can_nocow_extent(inode, lockstart, &num_bytes, NULL, NULL, NULL);
+ ret = can_nocow_extent(&inode->vfs_inode, lockstart, &num_bytes,
+ NULL, NULL, NULL);
if (ret <= 0) {
ret = 0;
btrfs_end_write_no_snapshoting(root);
@@ -1514,7 +1567,7 @@ static noinline int check_can_nocow(struct inode *inode, loff_t pos,
num_bytes - pos + lockstart);
}
- unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend);
+ unlock_extent(&inode->io_tree, lockstart, lockend);
return ret;
}
@@ -1528,6 +1581,7 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
struct btrfs_root *root = BTRFS_I(inode)->root;
struct page **pages = NULL;
struct extent_state *cached_state = NULL;
+ struct extent_changeset *data_reserved = NULL;
u64 release_bytes = 0;
u64 lockstart;
u64 lockend;
@@ -1575,11 +1629,14 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
reserve_bytes = round_up(write_bytes + sector_offset,
fs_info->sectorsize);
- ret = btrfs_check_data_free_space(inode, pos, write_bytes);
+ extent_changeset_release(data_reserved);
+ ret = btrfs_check_data_free_space(inode, &data_reserved, pos,
+ write_bytes);
if (ret < 0) {
if ((BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW |
BTRFS_INODE_PREALLOC)) &&
- check_can_nocow(inode, pos, &write_bytes) > 0) {
+ check_can_nocow(BTRFS_I(inode), pos,
+ &write_bytes) > 0) {
/*
* For nodata cow case, no need to reserve
* data space.
@@ -1599,11 +1656,13 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
}
}
- ret = btrfs_delalloc_reserve_metadata(inode, reserve_bytes);
+ ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode),
+ reserve_bytes);
if (ret) {
if (!only_release_metadata)
- btrfs_free_reserved_data_space(inode, pos,
- write_bytes);
+ btrfs_free_reserved_data_space(inode,
+ data_reserved, pos,
+ write_bytes);
else
btrfs_end_write_no_snapshoting(root);
break;
@@ -1623,9 +1682,9 @@ again:
if (ret)
break;
- ret = lock_and_cleanup_extent_if_need(inode, pages, num_pages,
- pos, write_bytes, &lockstart,
- &lockend, &cached_state);
+ ret = lock_and_cleanup_extent_if_need(BTRFS_I(inode), pages,
+ num_pages, pos, write_bytes, &lockstart,
+ &lockend, &cached_state);
if (ret < 0) {
if (ret == -EAGAIN)
goto again;
@@ -1677,7 +1736,7 @@ again:
spin_unlock(&BTRFS_I(inode)->lock);
}
if (only_release_metadata) {
- btrfs_delalloc_release_metadata(inode,
+ btrfs_delalloc_release_metadata(BTRFS_I(inode),
release_bytes);
} else {
u64 __pos;
@@ -1685,8 +1744,9 @@ again:
__pos = round_down(pos,
fs_info->sectorsize) +
(dirty_pages << PAGE_SHIFT);
- btrfs_delalloc_release_space(inode, __pos,
- release_bytes);
+ btrfs_delalloc_release_space(inode,
+ data_reserved, __pos,
+ release_bytes);
}
}
@@ -1738,14 +1798,16 @@ again:
if (release_bytes) {
if (only_release_metadata) {
btrfs_end_write_no_snapshoting(root);
- btrfs_delalloc_release_metadata(inode, release_bytes);
+ btrfs_delalloc_release_metadata(BTRFS_I(inode),
+ release_bytes);
} else {
- btrfs_delalloc_release_space(inode,
- round_down(pos, fs_info->sectorsize),
- release_bytes);
+ btrfs_delalloc_release_space(inode, data_reserved,
+ round_down(pos, fs_info->sectorsize),
+ release_bytes);
}
}
+ extent_changeset_free(data_reserved);
return num_written ? num_written : ret;
}
@@ -1820,17 +1882,36 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
bool sync = (file->f_flags & O_DSYNC) || IS_SYNC(file->f_mapping->host);
ssize_t err;
loff_t pos;
- size_t count;
+ size_t count = iov_iter_count(from);
loff_t oldsize;
int clean_page = 0;
- inode_lock(inode);
+ if (!inode_trylock(inode)) {
+ if (iocb->ki_flags & IOCB_NOWAIT)
+ return -EAGAIN;
+ inode_lock(inode);
+ }
+
err = generic_write_checks(iocb, from);
if (err <= 0) {
inode_unlock(inode);
return err;
}
+ pos = iocb->ki_pos;
+ if (iocb->ki_flags & IOCB_NOWAIT) {
+ /*
+ * We will allocate space in case nodatacow is not set,
+ * so bail
+ */
+ if (!(BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW |
+ BTRFS_INODE_PREALLOC)) ||
+ check_can_nocow(BTRFS_I(inode), pos, &count) <= 0) {
+ inode_unlock(inode);
+ return -EAGAIN;
+ }
+ }
+
current->backing_dev_info = inode_to_bdi(inode);
err = file_remove_privs(file);
if (err) {
@@ -1858,8 +1939,6 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
*/
update_time_for_write(inode);
- pos = iocb->ki_pos;
- count = iov_iter_count(from);
start_pos = round_down(pos, fs_info->sectorsize);
oldsize = i_size_read(inode);
if (start_pos > oldsize) {
@@ -1955,7 +2034,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
struct btrfs_log_ctx ctx;
- int ret = 0;
+ int ret = 0, err;
bool full_sync = 0;
u64 len;
@@ -1974,7 +2053,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
*/
ret = start_ordered_ops(inode, start, end);
if (ret)
- return ret;
+ goto out;
inode_lock(inode);
atomic_inc(&root->log_batch);
@@ -2062,7 +2141,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
* commit does not start nor waits for ordered extents to complete.
*/
smp_mb();
- if (btrfs_inode_in_log(inode, fs_info->generation) ||
+ if (btrfs_inode_in_log(BTRFS_I(inode), fs_info->generation) ||
(full_sync && BTRFS_I(inode)->last_trans <=
fs_info->last_trans_committed) ||
(!btrfs_have_ordered_extents_in_range(inode, start, len) &&
@@ -2079,10 +2158,10 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
* An ordered extent might have started before and completed
* already with io errors, in which case the inode was not
* updated and we end up here. So check the inode's mapping
- * flags for any errors that might have happened while doing
- * writeback of file data.
+ * for any errors that might have happened since we last
+ * checked called fsync.
*/
- ret = filemap_check_errors(inode->i_mapping);
+ ret = filemap_check_wb_err(inode->i_mapping, file->f_wb_err);
inode_unlock(inode);
goto out;
}
@@ -2171,6 +2250,9 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
ret = btrfs_end_transaction(trans);
}
out:
+ err = file_check_and_advance_wb_err(file);
+ if (!ret)
+ ret = err;
return ret > 0 ? -EIO : ret;
}
@@ -2193,7 +2275,7 @@ static int btrfs_file_mmap(struct file *filp, struct vm_area_struct *vma)
return 0;
}
-static int hole_mergeable(struct inode *inode, struct extent_buffer *leaf,
+static int hole_mergeable(struct btrfs_inode *inode, struct extent_buffer *leaf,
int slot, u64 start, u64 end)
{
struct btrfs_file_extent_item *fi;
@@ -2222,15 +2304,16 @@ static int hole_mergeable(struct inode *inode, struct extent_buffer *leaf,
return 0;
}
-static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode,
- struct btrfs_path *path, u64 offset, u64 end)
+static int fill_holes(struct btrfs_trans_handle *trans,
+ struct btrfs_inode *inode,
+ struct btrfs_path *path, u64 offset, u64 end)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = 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 extent_map_tree *em_tree = &inode->extent_tree;
struct btrfs_key key;
int ret;
@@ -2253,7 +2336,7 @@ static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode,
}
leaf = path->nodes[0];
- if (hole_mergeable(inode, leaf, path->slots[0]-1, offset, end)) {
+ if (hole_mergeable(inode, leaf, path->slots[0] - 1, offset, end)) {
u64 num_bytes;
path->slots[0]--;
@@ -2285,9 +2368,8 @@ static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode,
}
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);
+ 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;
@@ -2297,8 +2379,7 @@ out:
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);
+ set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags);
} else {
hole_em->start = offset;
hole_em->len = end - offset;
@@ -2321,7 +2402,7 @@ out:
free_extent_map(hole_em);
if (ret)
set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
- &BTRFS_I(inode)->runtime_flags);
+ &inode->runtime_flags);
}
return 0;
@@ -2335,17 +2416,15 @@ out:
*/
static int find_first_non_hole(struct inode *inode, u64 *start, u64 *len)
{
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
struct extent_map *em;
int ret = 0;
- em = btrfs_get_extent(inode, NULL, 0, *start, *len, 0);
- if (IS_ERR_OR_NULL(em)) {
- if (!em)
- ret = -ENOMEM;
- else
- ret = PTR_ERR(em);
- return ret;
- }
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0,
+ round_down(*start, fs_info->sectorsize),
+ round_up(*len, fs_info->sectorsize), 0);
+ if (IS_ERR(em))
+ return PTR_ERR(em);
/* Hole or vacuum extent(only exists in no-hole mode) */
if (em->block_start == EXTENT_MAP_HOLE) {
@@ -2551,8 +2630,8 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
trans->block_rsv = &fs_info->trans_block_rsv;
if (cur_offset < drop_end && cur_offset < ino_size) {
- ret = fill_holes(trans, inode, path, cur_offset,
- drop_end);
+ ret = fill_holes(trans, BTRFS_I(inode), path,
+ cur_offset, drop_end);
if (ret) {
/*
* If we failed then we didn't insert our hole
@@ -2623,7 +2702,8 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
* cur_offset == drop_end).
*/
if (cur_offset < ino_size && cur_offset < drop_end) {
- ret = fill_holes(trans, inode, path, cur_offset, drop_end);
+ ret = fill_holes(trans, BTRFS_I(inode), path,
+ cur_offset, drop_end);
if (ret) {
/* Same comment as above. */
btrfs_abort_transaction(trans, ret);
@@ -2718,6 +2798,7 @@ static long btrfs_fallocate(struct file *file, int mode,
{
struct inode *inode = file_inode(file);
struct extent_state *cached_state = NULL;
+ struct extent_changeset *data_reserved = NULL;
struct falloc_range *range;
struct falloc_range *tmp;
struct list_head reserve_list;
@@ -2748,7 +2829,8 @@ static long btrfs_fallocate(struct file *file, int mode,
*
* For qgroup space, it will be checked later.
*/
- ret = btrfs_alloc_data_chunk_ondemand(inode, alloc_end - alloc_start);
+ ret = btrfs_alloc_data_chunk_ondemand(BTRFS_I(inode),
+ alloc_end - alloc_start);
if (ret < 0)
return ret;
@@ -2828,13 +2910,10 @@ static long btrfs_fallocate(struct file *file, int mode,
/* First, check if we exceed the qgroup limit */
INIT_LIST_HEAD(&reserve_list);
while (1) {
- em = btrfs_get_extent(inode, NULL, 0, cur_offset,
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur_offset,
alloc_end - cur_offset, 0);
- if (IS_ERR_OR_NULL(em)) {
- if (!em)
- ret = -ENOMEM;
- else
- ret = PTR_ERR(em);
+ if (IS_ERR(em)) {
+ ret = PTR_ERR(em);
break;
}
last_byte = min(extent_map_end(em), alloc_end);
@@ -2849,18 +2928,20 @@ static long btrfs_fallocate(struct file *file, int mode,
free_extent_map(em);
break;
}
- ret = btrfs_qgroup_reserve_data(inode, cur_offset,
- last_byte - cur_offset);
- if (ret < 0)
+ ret = btrfs_qgroup_reserve_data(inode, &data_reserved,
+ cur_offset, last_byte - cur_offset);
+ if (ret < 0) {
+ free_extent_map(em);
break;
+ }
} else {
/*
* Do not need to reserve unwritten extent for this
* range, free reserved data space first, otherwise
* it'll result in false ENOSPC error.
*/
- btrfs_free_reserved_data_space(inode, cur_offset,
- last_byte - cur_offset);
+ btrfs_free_reserved_data_space(inode, data_reserved,
+ cur_offset, last_byte - cur_offset);
}
free_extent_map(em);
cur_offset = last_byte;
@@ -2876,11 +2957,12 @@ static long btrfs_fallocate(struct file *file, int mode,
if (!ret)
ret = btrfs_prealloc_file_range(inode, mode,
range->start,
- range->len, 1 << inode->i_blkbits,
+ range->len, i_blocksize(inode),
offset + len, &alloc_hint);
else
- btrfs_free_reserved_data_space(inode, range->start,
- range->len);
+ btrfs_free_reserved_data_space(inode,
+ data_reserved, range->start,
+ range->len);
list_del(&range->list);
kfree(range);
}
@@ -2918,8 +3000,9 @@ out:
inode_unlock(inode);
/* Let go of our reservation. */
if (ret != 0)
- btrfs_free_reserved_data_space(inode, alloc_start,
- alloc_end - cur_offset);
+ btrfs_free_reserved_data_space(inode, data_reserved,
+ alloc_start, alloc_end - cur_offset);
+ extent_changeset_free(data_reserved);
return ret;
}
@@ -2955,7 +3038,8 @@ static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
&cached_state);
while (start < inode->i_size) {
- em = btrfs_get_extent_fiemap(inode, NULL, 0, start, len, 0);
+ em = btrfs_get_extent_fiemap(BTRFS_I(inode), NULL, 0,
+ start, len, 0);
if (IS_ERR(em)) {
ret = PTR_ERR(em);
em = NULL;
@@ -3019,13 +3103,19 @@ out:
return offset;
}
+static int btrfs_file_open(struct inode *inode, struct file *filp)
+{
+ filp->f_mode |= FMODE_AIO_NOWAIT;
+ return generic_file_open(inode, filp);
+}
+
const struct file_operations btrfs_file_operations = {
.llseek = btrfs_file_llseek,
.read_iter = generic_file_read_iter,
.splice_read = generic_file_splice_read,
.write_iter = btrfs_file_write_iter,
.mmap = btrfs_file_mmap,
- .open = generic_file_open,
+ .open = btrfs_file_open,
.release = btrfs_release_file,
.fsync = btrfs_sync_file,
.fallocate = btrfs_fallocate,
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 7015892c9ee8..c5e6180cdb8c 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -18,6 +18,7 @@
#include <linux/pagemap.h>
#include <linux/sched.h>
+#include <linux/sched/signal.h>
#include <linux/slab.h>
#include <linux/math64.h>
#include <linux/ratelimit.h>
@@ -94,12 +95,11 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root,
return inode;
}
-struct inode *lookup_free_space_inode(struct btrfs_root *root,
+struct inode *lookup_free_space_inode(struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache
*block_group, struct btrfs_path *path)
{
struct inode *inode = NULL;
- struct btrfs_fs_info *fs_info = root->fs_info;
u32 flags = BTRFS_INODE_NODATASUM | BTRFS_INODE_NODATACOW;
spin_lock(&block_group->lock);
@@ -109,7 +109,7 @@ struct inode *lookup_free_space_inode(struct btrfs_root *root,
if (inode)
return inode;
- inode = __lookup_free_space_inode(root, path,
+ inode = __lookup_free_space_inode(fs_info->tree_root, path,
block_group->key.objectid);
if (IS_ERR(inode))
return inode;
@@ -192,7 +192,7 @@ static int __create_free_space_inode(struct btrfs_root *root,
return 0;
}
-int create_free_space_inode(struct btrfs_root *root,
+int create_free_space_inode(struct btrfs_fs_info *fs_info,
struct btrfs_trans_handle *trans,
struct btrfs_block_group_cache *block_group,
struct btrfs_path *path)
@@ -200,11 +200,11 @@ int create_free_space_inode(struct btrfs_root *root,
int ret;
u64 ino;
- ret = btrfs_find_free_objectid(root, &ino);
+ ret = btrfs_find_free_objectid(fs_info->tree_root, &ino);
if (ret < 0)
return ret;
- return __create_free_space_inode(root, trans, path, ino,
+ return __create_free_space_inode(fs_info->tree_root, trans, path, ino,
block_group->key.objectid);
}
@@ -227,21 +227,21 @@ int btrfs_check_trunc_cache_free_space(struct btrfs_fs_info *fs_info,
return ret;
}
-int btrfs_truncate_free_space_cache(struct btrfs_root *root,
- struct btrfs_trans_handle *trans,
+int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans,
struct btrfs_block_group_cache *block_group,
struct inode *inode)
{
+ struct btrfs_root *root = BTRFS_I(inode)->root;
int ret = 0;
- struct btrfs_path *path = btrfs_alloc_path();
bool locked = false;
- if (!path) {
- ret = -ENOMEM;
- goto fail;
- }
-
if (block_group) {
+ struct btrfs_path *path = btrfs_alloc_path();
+
+ if (!path) {
+ ret = -ENOMEM;
+ goto fail;
+ }
locked = true;
mutex_lock(&trans->transaction->cache_write_mutex);
if (!list_empty(&block_group->io_list)) {
@@ -258,10 +258,10 @@ int btrfs_truncate_free_space_cache(struct btrfs_root *root,
spin_lock(&block_group->lock);
block_group->disk_cache_state = BTRFS_DC_CLEAR;
spin_unlock(&block_group->lock);
+ btrfs_free_path(path);
}
- btrfs_free_path(path);
- btrfs_i_size_write(inode, 0);
+ btrfs_i_size_write(BTRFS_I(inode), 0);
truncate_pagecache(inode, 0);
/*
@@ -286,14 +286,14 @@ fail:
return ret;
}
-static int readahead_cache(struct inode *inode)
+static void readahead_cache(struct inode *inode)
{
struct file_ra_state *ra;
unsigned long last_index;
ra = kzalloc(sizeof(*ra), GFP_NOFS);
if (!ra)
- return -ENOMEM;
+ return;
file_ra_state_init(ra, inode->i_mapping);
last_index = (i_size_read(inode) - 1) >> PAGE_SHIFT;
@@ -301,8 +301,6 @@ static int readahead_cache(struct inode *inode)
page_cache_sync_readahead(inode->i_mapping, ra, NULL, 0, last_index);
kfree(ra);
-
- return 0;
}
static int io_ctl_init(struct btrfs_io_ctl *io_ctl, struct inode *inode,
@@ -313,7 +311,7 @@ static int io_ctl_init(struct btrfs_io_ctl *io_ctl, struct inode *inode,
num_pages = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE);
- if (btrfs_ino(inode) != BTRFS_FREE_INO_OBJECTID)
+ if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FREE_INO_OBJECTID)
check_crcs = 1;
/* Make sure we can fit our crcs into the first page */
@@ -357,7 +355,7 @@ static void io_ctl_map_page(struct btrfs_io_ctl *io_ctl, int clear)
io_ctl->orig = io_ctl->cur;
io_ctl->size = PAGE_SIZE;
if (clear)
- memset(io_ctl->cur, 0, PAGE_SIZE);
+ clear_page(io_ctl->cur);
}
static void io_ctl_drop_pages(struct btrfs_io_ctl *io_ctl)
@@ -730,9 +728,7 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
if (ret)
return ret;
- ret = readahead_cache(inode);
- if (ret)
- goto out;
+ readahead_cache(inode);
ret = io_ctl_prepare_pages(&io_ctl, inode, 1);
if (ret)
@@ -828,7 +824,6 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache *block_group)
{
struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
- struct btrfs_root *root = fs_info->tree_root;
struct inode *inode;
struct btrfs_path *path;
int ret = 0;
@@ -852,7 +847,7 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info,
path->search_commit_root = 1;
path->skip_locking = 1;
- inode = lookup_free_space_inode(root, block_group, path);
+ inode = lookup_free_space_inode(fs_info, block_group, path);
if (IS_ERR(inode)) {
btrfs_free_path(path);
return 0;
@@ -1128,8 +1123,7 @@ cleanup_bitmap_list(struct list_head *bitmap_list)
static void noinline_for_stack
cleanup_write_cache_enospc(struct inode *inode,
struct btrfs_io_ctl *io_ctl,
- struct extent_state **cached_state,
- struct list_head *bitmap_list)
+ struct extent_state **cached_state)
{
io_ctl_drop_pages(io_ctl);
unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0,
@@ -1225,8 +1219,6 @@ int btrfs_wait_cache_io(struct btrfs_trans_handle *trans,
* @ctl - the free space cache we are going to write out
* @block_group - the block_group for this cache if it belongs to a block_group
* @trans - the trans handle
- * @path - the path to use
- * @offset - the offset for the key we'll insert
*
* This function writes out a free space cache struct to disk for quick recovery
* on mount. This will return 0 if it was successful in writing the cache out,
@@ -1236,8 +1228,7 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
struct btrfs_free_space_ctl *ctl,
struct btrfs_block_group_cache *block_group,
struct btrfs_io_ctl *io_ctl,
- struct btrfs_trans_handle *trans,
- struct btrfs_path *path, u64 offset)
+ struct btrfs_trans_handle *trans)
{
struct btrfs_fs_info *fs_info = root->fs_info;
struct extent_state *cached_state = NULL;
@@ -1365,7 +1356,7 @@ out_nospc_locked:
mutex_unlock(&ctl->cache_writeout_mutex);
out_nospc:
- cleanup_write_cache_enospc(inode, io_ctl, &cached_state, &bitmap_list);
+ cleanup_write_cache_enospc(inode, io_ctl, &cached_state);
if (block_group && (block_group->flags & BTRFS_BLOCK_GROUP_DATA))
up_write(&block_group->data_rwsem);
@@ -1378,7 +1369,6 @@ int btrfs_write_out_cache(struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache *block_group,
struct btrfs_path *path)
{
- struct btrfs_root *root = fs_info->tree_root;
struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
struct inode *inode;
int ret = 0;
@@ -1390,13 +1380,12 @@ int btrfs_write_out_cache(struct btrfs_fs_info *fs_info,
}
spin_unlock(&block_group->lock);
- inode = lookup_free_space_inode(root, block_group, path);
+ inode = lookup_free_space_inode(fs_info, block_group, path);
if (IS_ERR(inode))
return 0;
- ret = __btrfs_write_out_cache(root, inode, ctl, block_group,
- &block_group->io_ctl, trans,
- path, block_group->key.objectid);
+ ret = __btrfs_write_out_cache(fs_info->tree_root, inode, ctl,
+ block_group, &block_group->io_ctl, trans);
if (ret) {
#ifdef DEBUG
btrfs_err(fs_info,
@@ -3543,8 +3532,7 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
return 0;
memset(&io_ctl, 0, sizeof(io_ctl));
- ret = __btrfs_write_out_cache(root, inode, ctl, NULL, &io_ctl,
- trans, path, 0);
+ ret = __btrfs_write_out_cache(root, inode, ctl, NULL, &io_ctl, trans);
if (!ret) {
/*
* At this point writepages() didn't error out, so our metadata
@@ -3558,7 +3546,8 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
if (ret) {
if (release_metadata)
- btrfs_delalloc_release_metadata(inode, inode->i_size);
+ btrfs_delalloc_release_metadata(BTRFS_I(inode),
+ inode->i_size);
#ifdef DEBUG
btrfs_err(fs_info,
"failed to write free ino cache for root %llu",
diff --git a/fs/btrfs/free-space-cache.h b/fs/btrfs/free-space-cache.h
index 6f3c025a2c6c..79eca4cabb1c 100644
--- a/fs/btrfs/free-space-cache.h
+++ b/fs/btrfs/free-space-cache.h
@@ -51,18 +51,17 @@ struct btrfs_free_space_op {
struct btrfs_io_ctl;
-struct inode *lookup_free_space_inode(struct btrfs_root *root,
+struct inode *lookup_free_space_inode(struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache
*block_group, struct btrfs_path *path);
-int create_free_space_inode(struct btrfs_root *root,
+int create_free_space_inode(struct btrfs_fs_info *fs_info,
struct btrfs_trans_handle *trans,
struct btrfs_block_group_cache *block_group,
struct btrfs_path *path);
int btrfs_check_trunc_cache_free_space(struct btrfs_fs_info *fs_info,
struct btrfs_block_rsv *rsv);
-int btrfs_truncate_free_space_cache(struct btrfs_root *root,
- struct btrfs_trans_handle *trans,
+int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans,
struct btrfs_block_group_cache *block_group,
struct inode *inode);
int load_free_space_cache(struct btrfs_fs_info *fs_info,
diff --git a/fs/btrfs/free-space-tree.c b/fs/btrfs/free-space-tree.c
index ff0c55337c2e..a5e34de06c2f 100644
--- a/fs/btrfs/free-space-tree.c
+++ b/fs/btrfs/free-space-tree.c
@@ -17,7 +17,7 @@
*/
#include <linux/kernel.h>
-#include <linux/vmalloc.h>
+#include <linux/sched/mm.h>
#include "ctree.h"
#include "disk-io.h"
#include "locking.h"
@@ -153,22 +153,21 @@ static inline u32 free_space_bitmap_size(u64 size, u32 sectorsize)
static u8 *alloc_bitmap(u32 bitmap_size)
{
- void *mem;
+ u8 *ret;
+ unsigned int nofs_flag;
/*
- * The allocation size varies, observed numbers were < 4K up to 16K.
- * Using vmalloc unconditionally would be too heavy, we'll try
- * contiguous allocations first.
+ * GFP_NOFS doesn't work with kvmalloc(), but we really can't recurse
+ * into the filesystem as the free space bitmap can be modified in the
+ * critical section of a transaction commit.
+ *
+ * TODO: push the memalloc_nofs_{save,restore}() to the caller where we
+ * know that recursion is unsafe.
*/
- if (bitmap_size <= PAGE_SIZE)
- return kzalloc(bitmap_size, GFP_NOFS);
-
- mem = kzalloc(bitmap_size, GFP_NOFS | __GFP_NOWARN);
- if (mem)
- return mem;
-
- return __vmalloc(bitmap_size, GFP_NOFS | __GFP_HIGHMEM | __GFP_ZERO,
- PAGE_KERNEL);
+ nofs_flag = memalloc_nofs_save();
+ ret = kvzalloc(bitmap_size, GFP_KERNEL);
+ memalloc_nofs_restore(nofs_flag);
+ return ret;
}
int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
@@ -1189,11 +1188,7 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info)
btrfs_set_fs_compat_ro(fs_info, FREE_SPACE_TREE_VALID);
clear_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &fs_info->flags);
- ret = btrfs_commit_transaction(trans);
- if (ret)
- return ret;
-
- return 0;
+ return btrfs_commit_transaction(trans);
abort:
clear_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &fs_info->flags);
@@ -1269,7 +1264,7 @@ int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info)
list_del(&free_space_root->dirty_list);
btrfs_tree_lock(free_space_root->node);
- clean_tree_block(trans, fs_info, free_space_root->node);
+ clean_tree_block(fs_info, free_space_root->node);
btrfs_tree_unlock(free_space_root->node);
btrfs_free_tree_block(trans, free_space_root, free_space_root->node,
0, 1);
@@ -1278,11 +1273,7 @@ int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info)
free_extent_buffer(free_space_root->commit_root);
kfree(free_space_root);
- ret = btrfs_commit_transaction(trans);
- if (ret)
- return ret;
-
- return 0;
+ return btrfs_commit_transaction(trans);
abort:
btrfs_abort_transaction(trans, ret);
diff --git a/fs/btrfs/hash.c b/fs/btrfs/hash.c
index a97fdc156a03..baacc1866861 100644
--- a/fs/btrfs/hash.c
+++ b/fs/btrfs/hash.c
@@ -38,6 +38,7 @@ u32 btrfs_crc32c(u32 crc, const void *address, unsigned int length)
{
SHASH_DESC_ON_STACK(shash, tfm);
u32 *ctx = (u32 *)shash_desc_ctx(shash);
+ u32 retval;
int err;
shash->tfm = tfm;
@@ -47,5 +48,7 @@ u32 btrfs_crc32c(u32 crc, const void *address, unsigned int length)
err = crypto_shash_update(shash, address, length);
BUG_ON(err);
- return *ctx;
+ retval = *ctx;
+ barrier_data(ctx);
+ return retval;
}
diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c
index 144b119ff43f..d02019747d00 100644
--- a/fs/btrfs/inode-map.c
+++ b/fs/btrfs/inode-map.c
@@ -400,6 +400,7 @@ int btrfs_save_ino_cache(struct btrfs_root *root,
struct btrfs_path *path;
struct inode *inode;
struct btrfs_block_rsv *rsv;
+ struct extent_changeset *data_reserved = NULL;
u64 num_bytes;
u64 alloc_hint = 0;
int ret;
@@ -467,7 +468,7 @@ again:
}
if (i_size_read(inode) > 0) {
- ret = btrfs_truncate_free_space_cache(root, trans, NULL, inode);
+ ret = btrfs_truncate_free_space_cache(trans, NULL, inode);
if (ret) {
if (ret != -ENOSPC)
btrfs_abort_transaction(trans, ret);
@@ -492,14 +493,14 @@ again:
/* Just to make sure we have enough space */
prealloc += 8 * PAGE_SIZE;
- ret = btrfs_delalloc_reserve_space(inode, 0, prealloc);
+ ret = btrfs_delalloc_reserve_space(inode, &data_reserved, 0, prealloc);
if (ret)
goto out_put;
ret = btrfs_prealloc_file_range_trans(inode, trans, 0, 0, prealloc,
prealloc, prealloc, &alloc_hint);
if (ret) {
- btrfs_delalloc_release_metadata(inode, prealloc);
+ btrfs_delalloc_release_metadata(BTRFS_I(inode), prealloc);
goto out_put;
}
@@ -516,6 +517,7 @@ out:
trans->bytes_reserved = num_bytes;
btrfs_free_path(path);
+ extent_changeset_free(data_reserved);
return ret;
}
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 1e861a063721..95c212037095 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -71,6 +71,7 @@ struct btrfs_dio_data {
u64 reserve;
u64 unsubmitted_oe_range_start;
u64 unsubmitted_oe_range_end;
+ int overwrite;
};
static const struct inode_operations btrfs_dir_inode_operations;
@@ -85,7 +86,6 @@ static const struct extent_io_ops btrfs_extent_io_ops;
static struct kmem_cache *btrfs_inode_cachep;
struct kmem_cache *btrfs_trans_handle_cachep;
-struct kmem_cache *btrfs_transaction_cachep;
struct kmem_cache *btrfs_path_cachep;
struct kmem_cache *btrfs_free_space_cachep;
@@ -108,11 +108,36 @@ static noinline int cow_file_range(struct inode *inode,
u64 start, u64 end, u64 delalloc_end,
int *page_started, unsigned long *nr_written,
int unlock, struct btrfs_dedupe_hash *hash);
-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, u64 ram_bytes,
- int type);
+static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len,
+ u64 orig_start, u64 block_start,
+ u64 block_len, u64 orig_block_len,
+ u64 ram_bytes, int compress_type,
+ int type);
+
+static void __endio_write_update_ordered(struct inode *inode,
+ const u64 offset, const u64 bytes,
+ const bool uptodate);
+
+/*
+ * Cleanup all submitted ordered extents in specified range to handle errors
+ * from the fill_dellaloc() callback.
+ *
+ * NOTE: caller must ensure that when an error happens, it can not call
+ * extent_clear_unlock_delalloc() to clear both the bits EXTENT_DO_ACCOUNTING
+ * and EXTENT_DELALLOC simultaneously, because that causes the reserved metadata
+ * to be released, which we want to happen only when finishing the ordered
+ * extent (btrfs_finish_ordered_io()). Also note that the caller of the
+ * fill_delalloc() callback already does proper cleanup for the first page of
+ * the range, that is, it invokes the callback writepage_end_io_hook() for the
+ * range of the first page.
+ */
+static inline void btrfs_cleanup_ordered_extents(struct inode *inode,
+ const u64 offset,
+ const u64 bytes)
+{
+ return __endio_write_update_ordered(inode, offset + PAGE_SIZE,
+ bytes - PAGE_SIZE, false);
+}
static int btrfs_dirty_inode(struct inode *inode);
@@ -152,7 +177,6 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans,
char *kaddr;
unsigned long ptr;
struct btrfs_file_extent_item *ei;
- int err = 0;
int ret;
size_t cur_size = size;
unsigned long offset;
@@ -166,7 +190,7 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans,
struct btrfs_key key;
size_t datasize;
- key.objectid = btrfs_ino(inode);
+ key.objectid = btrfs_ino(BTRFS_I(inode));
key.offset = start;
key.type = BTRFS_EXTENT_DATA_KEY;
@@ -174,10 +198,8 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans,
path->leave_spinning = 1;
ret = btrfs_insert_empty_item(trans, root, path, &key,
datasize);
- if (ret) {
- err = ret;
+ if (ret)
goto fail;
- }
}
leaf = path->nodes[0];
ei = btrfs_item_ptr(leaf, path->slots[0],
@@ -232,9 +254,8 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans,
BTRFS_I(inode)->disk_i_size = inode->i_size;
ret = btrfs_update_inode(trans, root, inode);
- return ret;
fail:
- return err;
+ return ret;
}
@@ -315,8 +336,8 @@ static noinline int cow_file_range_inline(struct btrfs_root *root,
}
set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags);
- btrfs_delalloc_release_metadata(inode, end + 1 - start);
- btrfs_drop_extent_cache(inode, start, aligned_end - 1, 0);
+ btrfs_delalloc_release_metadata(BTRFS_I(inode), end + 1 - start);
+ btrfs_drop_extent_cache(BTRFS_I(inode), start, aligned_end - 1, 0);
out:
/*
* Don't forget to free the reserved space, as for inlined extent
@@ -324,7 +345,7 @@ out:
* And at reserve time, it's always aligned to page size, so
* just free one page here.
*/
- btrfs_qgroup_free_data(inode, 0, PAGE_SIZE);
+ btrfs_qgroup_free_data(inode, NULL, 0, PAGE_SIZE);
btrfs_free_path(path);
btrfs_end_transaction(trans);
return ret;
@@ -388,6 +409,15 @@ static inline int inode_need_compress(struct inode *inode)
return 0;
}
+static inline void inode_should_defrag(struct btrfs_inode *inode,
+ u64 start, u64 end, u64 num_bytes, u64 small_write)
+{
+ /* If this is a small write inside eof, kick off a defrag */
+ if (num_bytes < small_write &&
+ (start > 0 || end + 1 < inode->disk_i_size))
+ btrfs_add_inode_defrag(NULL, inode);
+}
+
/*
* we create compressed extents in two phases. The first
* phase compresses a range of pages that have already been
@@ -420,26 +450,23 @@ static noinline void compress_file_range(struct inode *inode,
int ret = 0;
struct page **pages = NULL;
unsigned long nr_pages;
- unsigned long nr_pages_ret = 0;
unsigned long total_compressed = 0;
unsigned long total_in = 0;
- unsigned long max_compressed = SZ_128K;
- unsigned long max_uncompressed = SZ_128K;
int i;
int will_compress;
int compress_type = fs_info->compress_type;
int redirty = 0;
- /* if this is a small write inside eof, kick off a defrag */
- if ((end - start + 1) < SZ_16K &&
- (start > 0 || end + 1 < BTRFS_I(inode)->disk_i_size))
- btrfs_add_inode_defrag(NULL, inode);
+ inode_should_defrag(BTRFS_I(inode), start, end, end - start + 1,
+ SZ_16K);
actual_end = min_t(u64, isize, end + 1);
again:
will_compress = 0;
nr_pages = (end >> PAGE_SHIFT) - (start >> PAGE_SHIFT) + 1;
- nr_pages = min_t(unsigned long, nr_pages, SZ_128K / PAGE_SIZE);
+ BUILD_BUG_ON((BTRFS_MAX_COMPRESSED % PAGE_SIZE) != 0);
+ nr_pages = min_t(unsigned long, nr_pages,
+ BTRFS_MAX_COMPRESSED / PAGE_SIZE);
/*
* we don't want to send crud past the end of i_size through
@@ -464,17 +491,8 @@ again:
(start > 0 || end + 1 < BTRFS_I(inode)->disk_i_size))
goto cleanup_and_bail_uncompressed;
- /* we want to make sure that amount of ram required to uncompress
- * an extent is reasonable, so we limit the total size in ram
- * of a compressed extent to 128k. This is a crucial number
- * because it also controls how easily we can spread reads across
- * cpus for decompression.
- *
- * We also want to make sure the amount of IO required to do
- * a random read is reasonably small, so we limit the size of
- * a compressed extent to 128k.
- */
- total_compressed = min(total_compressed, max_uncompressed);
+ total_compressed = min_t(unsigned long, total_compressed,
+ BTRFS_MAX_UNCOMPRESSED);
num_bytes = ALIGN(end - start + 1, blocksize);
num_bytes = max(blocksize, num_bytes);
total_in = 0;
@@ -509,16 +527,15 @@ again:
redirty = 1;
ret = btrfs_compress_pages(compress_type,
inode->i_mapping, start,
- total_compressed, pages,
- nr_pages, &nr_pages_ret,
+ pages,
+ &nr_pages,
&total_in,
- &total_compressed,
- max_compressed);
+ &total_compressed);
if (!ret) {
unsigned long offset = total_compressed &
(PAGE_SIZE - 1);
- struct page *page = pages[nr_pages_ret - 1];
+ struct page *page = pages[nr_pages - 1];
char *kaddr;
/* zero the tail end of the last page, we might be
@@ -541,7 +558,7 @@ cont:
* to make an uncompressed inline extent.
*/
ret = cow_file_range_inline(root, inode, start, end,
- 0, 0, NULL);
+ 0, BTRFS_COMPRESS_NONE, NULL);
} else {
/* try making a compressed inline extent */
ret = cow_file_range_inline(root, inode, start, end,
@@ -550,7 +567,7 @@ cont:
}
if (ret <= 0) {
unsigned long clear_flags = EXTENT_DELALLOC |
- EXTENT_DEFRAG;
+ EXTENT_DELALLOC_NEW | EXTENT_DEFRAG;
unsigned long page_error_op;
clear_flags |= (ret < 0) ? EXTENT_DO_ACCOUNTING : 0;
@@ -568,8 +585,10 @@ cont:
PAGE_SET_WRITEBACK |
page_error_op |
PAGE_END_WRITEBACK);
- btrfs_free_reserved_data_space_noquota(inode, start,
- end - start + 1);
+ if (ret == 0)
+ btrfs_free_reserved_data_space_noquota(inode,
+ start,
+ end - start + 1);
goto free_pages_out;
}
}
@@ -584,12 +603,11 @@ cont:
/*
* one last check to make sure the compression is really a
- * win, compare the page count read with the blocks on disk
+ * win, compare the page count read with the blocks on disk,
+ * compression must free at least one sector size
*/
total_in = ALIGN(total_in, PAGE_SIZE);
- if (total_compressed >= total_in) {
- will_compress = 0;
- } else {
+ if (total_compressed + blocksize <= total_in) {
num_bytes = total_in;
*num_added += 1;
@@ -599,7 +617,7 @@ cont:
* will submit them to the elevator.
*/
add_async_extent(async_cow, start, num_bytes,
- total_compressed, pages, nr_pages_ret,
+ total_compressed, pages, nr_pages,
compress_type);
if (start + num_bytes < end) {
@@ -616,14 +634,14 @@ cont:
* the compression code ran but failed to make things smaller,
* free any pages it allocated and our page pointer array
*/
- for (i = 0; i < nr_pages_ret; i++) {
+ for (i = 0; i < nr_pages; i++) {
WARN_ON(pages[i]->mapping);
put_page(pages[i]);
}
kfree(pages);
pages = NULL;
total_compressed = 0;
- nr_pages_ret = 0;
+ nr_pages = 0;
/* flag the file so we don't compress in the future */
if (!btrfs_test_opt(fs_info, FORCE_COMPRESS) &&
@@ -652,7 +670,7 @@ cleanup_and_bail_uncompressed:
return;
free_pages_out:
- for (i = 0; i < nr_pages_ret; i++) {
+ for (i = 0; i < nr_pages; i++) {
WARN_ON(pages[i]->mapping);
put_page(pages[i]);
}
@@ -690,7 +708,6 @@ static noinline void submit_compressed_extents(struct inode *inode,
struct btrfs_key ins;
struct extent_map *em;
struct btrfs_root *root = BTRFS_I(inode)->root;
- struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
struct extent_io_tree *io_tree;
int ret = 0;
@@ -778,46 +795,19 @@ retry:
* here we're doing allocation and writeback of the
* compressed pages
*/
- btrfs_drop_extent_cache(inode, async_extent->start,
- async_extent->start +
- async_extent->ram_size - 1, 0);
-
- em = alloc_extent_map();
- if (!em) {
- ret = -ENOMEM;
- goto out_free_reserve;
- }
- 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->ram_bytes = async_extent->ram_size;
- em->bdev = 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, 1);
- write_unlock(&em_tree->lock);
- if (ret != -EEXIST) {
- free_extent_map(em);
- break;
- }
- btrfs_drop_extent_cache(inode, async_extent->start,
- async_extent->start +
- async_extent->ram_size - 1, 0);
- }
-
- if (ret)
+ em = create_io_em(inode, async_extent->start,
+ async_extent->ram_size, /* len */
+ async_extent->start, /* orig_start */
+ ins.objectid, /* block_start */
+ ins.offset, /* block_len */
+ ins.offset, /* orig_block_len */
+ async_extent->ram_size, /* ram_bytes */
+ async_extent->compress_type,
+ BTRFS_ORDERED_COMPRESSED);
+ if (IS_ERR(em))
+ /* ret value is not necessary due to void function */
goto out_free_reserve;
+ free_extent_map(em);
ret = btrfs_add_ordered_extent_compress(inode,
async_extent->start,
@@ -827,7 +817,8 @@ retry:
BTRFS_ORDERED_COMPRESSED,
async_extent->compress_type);
if (ret) {
- btrfs_drop_extent_cache(inode, async_extent->start,
+ btrfs_drop_extent_cache(BTRFS_I(inode),
+ async_extent->start,
async_extent->start +
async_extent->ram_size - 1, 0);
goto out_free_reserve;
@@ -845,13 +836,12 @@ retry:
NULL, EXTENT_LOCKED | EXTENT_DELALLOC,
PAGE_UNLOCK | PAGE_CLEAR_DIRTY |
PAGE_SET_WRITEBACK);
- ret = btrfs_submit_compressed_write(inode,
+ if (btrfs_submit_compressed_write(inode,
async_extent->start,
async_extent->ram_size,
ins.objectid,
ins.offset, async_extent->pages,
- async_extent->nr_pages);
- if (ret) {
+ async_extent->nr_pages)) {
struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
struct page *p = async_extent->pages[0];
const u64 start = async_extent->start;
@@ -882,6 +872,7 @@ out_free:
async_extent->start +
async_extent->ram_size - 1,
NULL, EXTENT_LOCKED | EXTENT_DELALLOC |
+ EXTENT_DELALLOC_NEW |
EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING,
PAGE_UNLOCK | PAGE_CLEAR_DIRTY |
PAGE_SET_WRITEBACK | PAGE_END_WRITEBACK |
@@ -948,14 +939,16 @@ static noinline int cow_file_range(struct inode *inode,
u64 num_bytes;
unsigned long ram_size;
u64 disk_num_bytes;
- u64 cur_alloc_size;
+ u64 cur_alloc_size = 0;
u64 blocksize = fs_info->sectorsize;
struct btrfs_key ins;
struct extent_map *em;
- struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
+ unsigned clear_bits;
+ unsigned long page_ops;
+ bool extent_reserved = false;
int ret = 0;
- if (btrfs_is_free_space_inode(inode)) {
+ if (btrfs_is_free_space_inode(BTRFS_I(inode))) {
WARN_ON_ONCE(1);
ret = -EINVAL;
goto out_unlock;
@@ -965,19 +958,17 @@ static noinline int cow_file_range(struct inode *inode,
num_bytes = max(blocksize, num_bytes);
disk_num_bytes = num_bytes;
- /* if this is a small write inside eof, kick off defrag */
- if (num_bytes < SZ_64K &&
- (start > 0 || end + 1 < BTRFS_I(inode)->disk_i_size))
- btrfs_add_inode_defrag(NULL, inode);
+ inode_should_defrag(BTRFS_I(inode), start, end, num_bytes, SZ_64K);
if (start == 0) {
/* lets try to make an inline extent */
- ret = cow_file_range_inline(root, inode, start, end, 0, 0,
- NULL);
+ ret = cow_file_range_inline(root, inode, start, end, 0,
+ BTRFS_COMPRESS_NONE, NULL);
if (ret == 0) {
extent_clear_unlock_delalloc(inode, start, end,
delalloc_end, NULL,
EXTENT_LOCKED | EXTENT_DELALLOC |
+ EXTENT_DELALLOC_NEW |
EXTENT_DEFRAG, PAGE_UNLOCK |
PAGE_CLEAR_DIRTY | PAGE_SET_WRITEBACK |
PAGE_END_WRITEBACK);
@@ -996,53 +987,32 @@ static noinline int cow_file_range(struct inode *inode,
btrfs_super_total_bytes(fs_info->super_copy));
alloc_hint = get_extent_allocation_hint(inode, start, num_bytes);
- btrfs_drop_extent_cache(inode, start, start + num_bytes - 1, 0);
+ btrfs_drop_extent_cache(BTRFS_I(inode), start,
+ start + num_bytes - 1, 0);
while (disk_num_bytes > 0) {
- unsigned long op;
-
cur_alloc_size = disk_num_bytes;
ret = btrfs_reserve_extent(root, cur_alloc_size, cur_alloc_size,
fs_info->sectorsize, 0, alloc_hint,
&ins, 1, 1);
if (ret < 0)
goto out_unlock;
+ cur_alloc_size = ins.offset;
+ extent_reserved = true;
- em = alloc_extent_map();
- if (!em) {
- ret = -ENOMEM;
- goto out_reserve;
- }
- em->start = start;
- 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->ram_bytes = ram_size;
- em->bdev = 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, 1);
- write_unlock(&em_tree->lock);
- if (ret != -EEXIST) {
- free_extent_map(em);
- break;
- }
- btrfs_drop_extent_cache(inode, start,
- start + ram_size - 1, 0);
- }
- if (ret)
+ em = create_io_em(inode, start, ins.offset, /* len */
+ start, /* orig_start */
+ ins.objectid, /* block_start */
+ ins.offset, /* block_len */
+ ins.offset, /* orig_block_len */
+ ram_size, /* ram_bytes */
+ BTRFS_COMPRESS_NONE, /* compress_type */
+ BTRFS_ORDERED_REGULAR /* type */);
+ if (IS_ERR(em))
goto out_reserve;
+ free_extent_map(em);
- cur_alloc_size = ins.offset;
ret = btrfs_add_ordered_extent(inode, start, ins.objectid,
ram_size, cur_alloc_size, 0);
if (ret)
@@ -1052,15 +1022,24 @@ static noinline int cow_file_range(struct inode *inode,
BTRFS_DATA_RELOC_TREE_OBJECTID) {
ret = btrfs_reloc_clone_csums(inode, start,
cur_alloc_size);
+ /*
+ * Only drop cache here, and process as normal.
+ *
+ * We must not allow extent_clear_unlock_delalloc()
+ * at out_unlock label to free meta of this ordered
+ * extent, as its meta should be freed by
+ * btrfs_finish_ordered_io().
+ *
+ * So we must continue until @start is increased to
+ * skip current ordered extent.
+ */
if (ret)
- goto out_drop_extent_cache;
+ btrfs_drop_extent_cache(BTRFS_I(inode), start,
+ start + ram_size - 1, 0);
}
btrfs_dec_block_group_reservations(fs_info, ins.objectid);
- if (disk_num_bytes < cur_alloc_size)
- break;
-
/* we're not doing compressed IO, don't unlock the first
* page (which the caller expects to stay locked), don't
* clear any dirty bits and don't set any writeback bits
@@ -1068,34 +1047,69 @@ static noinline int cow_file_range(struct inode *inode,
* Do set the Private2 bit so we know this page was properly
* setup for writepage
*/
- op = unlock ? PAGE_UNLOCK : 0;
- op |= PAGE_SET_PRIVATE2;
+ page_ops = unlock ? PAGE_UNLOCK : 0;
+ page_ops |= PAGE_SET_PRIVATE2;
extent_clear_unlock_delalloc(inode, start,
start + ram_size - 1,
delalloc_end, locked_page,
EXTENT_LOCKED | EXTENT_DELALLOC,
- op);
- disk_num_bytes -= cur_alloc_size;
+ page_ops);
+ if (disk_num_bytes < cur_alloc_size)
+ disk_num_bytes = 0;
+ else
+ disk_num_bytes -= cur_alloc_size;
num_bytes -= cur_alloc_size;
alloc_hint = ins.objectid + ins.offset;
start += cur_alloc_size;
+ extent_reserved = false;
+
+ /*
+ * btrfs_reloc_clone_csums() error, since start is increased
+ * extent_clear_unlock_delalloc() at out_unlock label won't
+ * free metadata of current ordered extent, we're OK to exit.
+ */
+ if (ret)
+ goto out_unlock;
}
out:
return ret;
out_drop_extent_cache:
- btrfs_drop_extent_cache(inode, start, start + ram_size - 1, 0);
+ btrfs_drop_extent_cache(BTRFS_I(inode), start, start + ram_size - 1, 0);
out_reserve:
btrfs_dec_block_group_reservations(fs_info, ins.objectid);
btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 1);
out_unlock:
+ clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW |
+ EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV;
+ page_ops = PAGE_UNLOCK | PAGE_CLEAR_DIRTY | PAGE_SET_WRITEBACK |
+ PAGE_END_WRITEBACK;
+ /*
+ * If we reserved an extent for our delalloc range (or a subrange) and
+ * failed to create the respective ordered extent, then it means that
+ * when we reserved the extent we decremented the extent's size from
+ * the data space_info's bytes_may_use counter and incremented the
+ * space_info's bytes_reserved counter by the same amount. We must make
+ * sure extent_clear_unlock_delalloc() does not try to decrement again
+ * the data space_info's bytes_may_use counter, therefore we do not pass
+ * it the flag EXTENT_CLEAR_DATA_RESV.
+ */
+ if (extent_reserved) {
+ extent_clear_unlock_delalloc(inode, start,
+ start + cur_alloc_size,
+ start + cur_alloc_size,
+ locked_page,
+ clear_bits,
+ page_ops);
+ start += cur_alloc_size;
+ if (start >= end)
+ goto out;
+ }
extent_clear_unlock_delalloc(inode, start, end, delalloc_end,
locked_page,
- EXTENT_LOCKED | EXTENT_DO_ACCOUNTING |
- EXTENT_DELALLOC | EXTENT_DEFRAG,
- PAGE_UNLOCK | PAGE_CLEAR_DIRTY |
- PAGE_SET_WRITEBACK | PAGE_END_WRITEBACK);
+ clear_bits | EXTENT_CLEAR_DATA_RESV,
+ page_ops);
goto out;
}
@@ -1164,7 +1178,6 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
struct btrfs_root *root = BTRFS_I(inode)->root;
unsigned long nr_pages;
u64 cur_end;
- int limit = 10 * SZ_1M;
clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, EXTENT_LOCKED,
1, 0, NULL, GFP_NOFS);
@@ -1196,12 +1209,6 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
btrfs_queue_work(fs_info->delalloc_workers, &async_cow->work);
- if (atomic_read(&fs_info->async_delalloc_pages) > limit) {
- wait_event(fs_info->async_submit_wait,
- (atomic_read(&fs_info->async_delalloc_pages) <
- limit));
- }
-
while (atomic_read(&fs_info->async_submit_draining) &&
atomic_read(&fs_info->async_delalloc_pages)) {
wait_event(fs_info->async_submit_wait,
@@ -1250,11 +1257,11 @@ static noinline int run_delalloc_nocow(struct inode *inode,
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
struct btrfs_root *root = BTRFS_I(inode)->root;
- struct btrfs_trans_handle *trans;
struct extent_buffer *leaf;
struct btrfs_path *path;
struct btrfs_file_extent_item *fi;
struct btrfs_key found_key;
+ struct extent_map *em;
u64 cow_start;
u64 cur_offset;
u64 extent_end;
@@ -1269,7 +1276,7 @@ static noinline int run_delalloc_nocow(struct inode *inode,
int nocow;
int check_prev = 1;
bool nolock;
- u64 ino = btrfs_ino(inode);
+ u64 ino = btrfs_ino(BTRFS_I(inode));
path = btrfs_alloc_path();
if (!path) {
@@ -1284,32 +1291,12 @@ static noinline int run_delalloc_nocow(struct inode *inode,
return -ENOMEM;
}
- nolock = btrfs_is_free_space_inode(inode);
-
- if (nolock)
- trans = btrfs_join_transaction_nolock(root);
- else
- trans = btrfs_join_transaction(root);
-
- if (IS_ERR(trans)) {
- extent_clear_unlock_delalloc(inode, start, end, end,
- locked_page,
- EXTENT_LOCKED | EXTENT_DELALLOC |
- EXTENT_DO_ACCOUNTING |
- EXTENT_DEFRAG, PAGE_UNLOCK |
- PAGE_CLEAR_DIRTY |
- PAGE_SET_WRITEBACK |
- PAGE_END_WRITEBACK);
- btrfs_free_path(path);
- return PTR_ERR(trans);
- }
-
- trans->block_rsv = &fs_info->delalloc_block_rsv;
+ nolock = btrfs_is_free_space_inode(BTRFS_I(inode));
cow_start = (u64)-1;
cur_offset = start;
while (1) {
- ret = btrfs_lookup_file_extent(trans, root, path, ino,
+ ret = btrfs_lookup_file_extent(NULL, root, path, ino,
cur_offset, 0);
if (ret < 0)
goto error;
@@ -1382,7 +1369,7 @@ next_slot:
goto out_check;
if (btrfs_extent_readonly(fs_info, disk_bytenr))
goto out_check;
- if (btrfs_cross_ref_exist(trans, root, ino,
+ if (btrfs_cross_ref_exist(root, ino,
found_key.offset -
extent_offset, disk_bytenr))
goto out_check;
@@ -1404,10 +1391,16 @@ next_slot:
* either valid or do not exist.
*/
if (csum_exist_in_range(fs_info, disk_bytenr,
- num_bytes))
+ num_bytes)) {
+ if (!nolock)
+ btrfs_end_write_no_snapshoting(root);
goto out_check;
- if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr))
+ }
+ if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr)) {
+ if (!nolock)
+ btrfs_end_write_no_snapshoting(root);
goto out_check;
+ }
nocow = 1;
} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
extent_end = found_key.offset +
@@ -1455,35 +1448,28 @@ out_check:
}
if (extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
- struct extent_map *em;
- struct extent_map_tree *em_tree;
- em_tree = &BTRFS_I(inode)->extent_tree;
- em = alloc_extent_map();
- BUG_ON(!em); /* -ENOMEM */
- em->start = cur_offset;
- 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->ram_bytes = ram_bytes;
- em->bdev = 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_FILLING, &em->flags);
- em->generation = -1;
- while (1) {
- write_lock(&em_tree->lock);
- ret = add_extent_mapping(em_tree, em, 1);
- write_unlock(&em_tree->lock);
- if (ret != -EEXIST) {
- free_extent_map(em);
- break;
- }
- btrfs_drop_extent_cache(inode, em->start,
- em->start + em->len - 1, 0);
+ u64 orig_start = found_key.offset - extent_offset;
+
+ em = create_io_em(inode, cur_offset, num_bytes,
+ orig_start,
+ disk_bytenr, /* block_start */
+ num_bytes, /* block_len */
+ disk_num_bytes, /* orig_block_len */
+ ram_bytes, BTRFS_COMPRESS_NONE,
+ BTRFS_ORDERED_PREALLOC);
+ if (IS_ERR(em)) {
+ if (!nolock && nocow)
+ btrfs_end_write_no_snapshoting(root);
+ if (nocow)
+ btrfs_dec_nocow_writers(fs_info,
+ disk_bytenr);
+ ret = PTR_ERR(em);
+ goto error;
}
+ free_extent_map(em);
+ }
+
+ if (extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
type = BTRFS_ORDERED_PREALLOC;
} else {
type = BTRFS_ORDERED_NOCOW;
@@ -1496,15 +1482,14 @@ out_check:
BUG_ON(ret); /* -ENOMEM */
if (root->root_key.objectid ==
- BTRFS_DATA_RELOC_TREE_OBJECTID) {
+ BTRFS_DATA_RELOC_TREE_OBJECTID)
+ /*
+ * Error handled later, as we must prevent
+ * extent_clear_unlock_delalloc() in error handler
+ * from freeing metadata of created ordered extent.
+ */
ret = btrfs_reloc_clone_csums(inode, cur_offset,
num_bytes);
- if (ret) {
- if (!nolock && nocow)
- btrfs_end_write_no_snapshoting(root);
- goto error;
- }
- }
extent_clear_unlock_delalloc(inode, cur_offset,
cur_offset + num_bytes - 1, end,
@@ -1516,6 +1501,14 @@ out_check:
if (!nolock && nocow)
btrfs_end_write_no_snapshoting(root);
cur_offset = extent_end;
+
+ /*
+ * btrfs_reloc_clone_csums() error, now we're OK to call error
+ * handler, as metadata for created ordered extent will only
+ * be freed by btrfs_finish_ordered_io().
+ */
+ if (ret)
+ goto error;
if (cur_offset > end)
break;
}
@@ -1534,10 +1527,6 @@ out_check:
}
error:
- err = btrfs_end_transaction(trans);
- if (!ret)
- ret = err;
-
if (ret && cur_offset < end)
extent_clear_unlock_delalloc(inode, cur_offset, end, end,
locked_page, EXTENT_LOCKED |
@@ -1573,10 +1562,11 @@ static inline int need_force_cow(struct inode *inode, u64 start, u64 end)
/*
* extent_io.c call back to do delayed allocation processing
*/
-static int run_delalloc_range(struct inode *inode, struct page *locked_page,
+static int run_delalloc_range(void *private_data, struct page *locked_page,
u64 start, u64 end, int *page_started,
unsigned long *nr_written)
{
+ struct inode *inode = private_data;
int ret;
int force_cow = need_force_cow(inode, start, end);
@@ -1595,12 +1585,15 @@ static int run_delalloc_range(struct inode *inode, struct page *locked_page,
ret = cow_file_range_async(inode, locked_page, start, end,
page_started, nr_written);
}
+ if (ret)
+ btrfs_cleanup_ordered_extents(inode, start, end - start + 1);
return ret;
}
-static void btrfs_split_extent_hook(struct inode *inode,
+static void btrfs_split_extent_hook(void *private_data,
struct extent_state *orig, u64 split)
{
+ struct inode *inode = private_data;
u64 size;
/* not delalloc, ignore it */
@@ -1609,7 +1602,7 @@ static void btrfs_split_extent_hook(struct inode *inode,
size = orig->end - orig->start + 1;
if (size > BTRFS_MAX_EXTENT_SIZE) {
- u64 num_extents;
+ u32 num_extents;
u64 new_size;
/*
@@ -1617,13 +1610,10 @@ static void btrfs_split_extent_hook(struct inode *inode,
* applies here, just in reverse.
*/
new_size = orig->end - split + 1;
- num_extents = div64_u64(new_size + BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE);
+ num_extents = count_max_extents(new_size);
new_size = split - orig->start;
- num_extents += div64_u64(new_size + BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE);
- if (div64_u64(size + BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE) >= num_extents)
+ num_extents += count_max_extents(new_size);
+ if (count_max_extents(size) >= num_extents)
return;
}
@@ -1638,12 +1628,13 @@ static void btrfs_split_extent_hook(struct inode *inode,
* extents, such as when we are doing sequential writes, so we can properly
* account for the metadata space we'll need.
*/
-static void btrfs_merge_extent_hook(struct inode *inode,
+static void btrfs_merge_extent_hook(void *private_data,
struct extent_state *new,
struct extent_state *other)
{
+ struct inode *inode = private_data;
u64 new_size, old_size;
- u64 num_extents;
+ u32 num_extents;
/* not delalloc, ignore it */
if (!(other->state & EXTENT_DELALLOC))
@@ -1681,14 +1672,10 @@ static void btrfs_merge_extent_hook(struct inode *inode,
* this case.
*/
old_size = other->end - other->start + 1;
- num_extents = div64_u64(old_size + BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE);
+ num_extents = count_max_extents(old_size);
old_size = new->end - new->start + 1;
- num_extents += div64_u64(old_size + BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE);
-
- if (div64_u64(new_size + BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE) >= num_extents)
+ num_extents += count_max_extents(old_size);
+ if (count_max_extents(new_size) >= num_extents)
return;
spin_lock(&BTRFS_I(inode)->lock);
@@ -1720,15 +1707,15 @@ static void btrfs_add_delalloc_inodes(struct btrfs_root *root,
}
static void btrfs_del_delalloc_inode(struct btrfs_root *root,
- struct inode *inode)
+ struct btrfs_inode *inode)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
spin_lock(&root->delalloc_lock);
- if (!list_empty(&BTRFS_I(inode)->delalloc_inodes)) {
- list_del_init(&BTRFS_I(inode)->delalloc_inodes);
+ if (!list_empty(&inode->delalloc_inodes)) {
+ list_del_init(&inode->delalloc_inodes);
clear_bit(BTRFS_INODE_IN_DELALLOC_LIST,
- &BTRFS_I(inode)->runtime_flags);
+ &inode->runtime_flags);
root->nr_delalloc_inodes--;
if (!root->nr_delalloc_inodes) {
spin_lock(&fs_info->delalloc_root_lock);
@@ -1745,9 +1732,10 @@ static void btrfs_del_delalloc_inode(struct btrfs_root *root,
* bytes in this file, and to maintain the list of inodes that
* have pending delalloc work to be done.
*/
-static void btrfs_set_bit_hook(struct inode *inode,
+static void btrfs_set_bit_hook(void *private_data,
struct extent_state *state, unsigned *bits)
{
+ struct inode *inode = private_data;
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
@@ -1761,7 +1749,7 @@ static void btrfs_set_bit_hook(struct inode *inode,
if (!(state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) {
struct btrfs_root *root = BTRFS_I(inode)->root;
u64 len = state->end + 1 - state->start;
- bool do_list = !btrfs_is_free_space_inode(inode);
+ bool do_list = !btrfs_is_free_space_inode(BTRFS_I(inode));
if (*bits & EXTENT_FIRST_DELALLOC) {
*bits &= ~EXTENT_FIRST_DELALLOC;
@@ -1775,8 +1763,8 @@ static void btrfs_set_bit_hook(struct inode *inode,
if (btrfs_is_testing(fs_info))
return;
- __percpu_counter_add(&fs_info->delalloc_bytes, len,
- fs_info->delalloc_batch);
+ percpu_counter_add_batch(&fs_info->delalloc_bytes, len,
+ fs_info->delalloc_batch);
spin_lock(&BTRFS_I(inode)->lock);
BTRFS_I(inode)->delalloc_bytes += len;
if (*bits & EXTENT_DEFRAG)
@@ -1786,24 +1774,32 @@ static void btrfs_set_bit_hook(struct inode *inode,
btrfs_add_delalloc_inodes(root, inode);
spin_unlock(&BTRFS_I(inode)->lock);
}
+
+ if (!(state->state & EXTENT_DELALLOC_NEW) &&
+ (*bits & EXTENT_DELALLOC_NEW)) {
+ spin_lock(&BTRFS_I(inode)->lock);
+ BTRFS_I(inode)->new_delalloc_bytes += state->end + 1 -
+ state->start;
+ spin_unlock(&BTRFS_I(inode)->lock);
+ }
}
/*
* extent_io.c clear_bit_hook, see set_bit_hook for why
*/
-static void btrfs_clear_bit_hook(struct inode *inode,
+static void btrfs_clear_bit_hook(void *private_data,
struct extent_state *state,
unsigned *bits)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_inode *inode = BTRFS_I((struct inode *)private_data);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
u64 len = state->end + 1 - state->start;
- u64 num_extents = div64_u64(len + BTRFS_MAX_EXTENT_SIZE -1,
- BTRFS_MAX_EXTENT_SIZE);
+ u32 num_extents = count_max_extents(len);
- spin_lock(&BTRFS_I(inode)->lock);
+ spin_lock(&inode->lock);
if ((state->state & EXTENT_DEFRAG) && (*bits & EXTENT_DEFRAG))
- BTRFS_I(inode)->defrag_bytes -= len;
- spin_unlock(&BTRFS_I(inode)->lock);
+ inode->defrag_bytes -= len;
+ spin_unlock(&inode->lock);
/*
* set_bit and clear bit hooks normally require _irqsave/restore
@@ -1811,15 +1807,15 @@ static void btrfs_clear_bit_hook(struct inode *inode,
* bit, which is only set or cleared with irqs on
*/
if ((state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) {
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_root *root = inode->root;
bool do_list = !btrfs_is_free_space_inode(inode);
if (*bits & EXTENT_FIRST_DELALLOC) {
*bits &= ~EXTENT_FIRST_DELALLOC;
- } else if (!(*bits & EXTENT_DO_ACCOUNTING)) {
- spin_lock(&BTRFS_I(inode)->lock);
- BTRFS_I(inode)->outstanding_extents -= num_extents;
- spin_unlock(&BTRFS_I(inode)->lock);
+ } else if (!(*bits & EXTENT_CLEAR_META_RESV)) {
+ spin_lock(&inode->lock);
+ inode->outstanding_extents -= num_extents;
+ spin_unlock(&inode->lock);
}
/*
@@ -1827,7 +1823,7 @@ static void btrfs_clear_bit_hook(struct inode *inode,
* don't need to call dellalloc_release_metadata if there is an
* error.
*/
- if (*bits & EXTENT_DO_ACCOUNTING &&
+ if (*bits & EXTENT_CLEAR_META_RESV &&
root != fs_info->tree_root)
btrfs_delalloc_release_metadata(inode, len);
@@ -1835,22 +1831,30 @@ static void btrfs_clear_bit_hook(struct inode *inode,
if (btrfs_is_testing(fs_info))
return;
- if (root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID
- && do_list && !(state->state & EXTENT_NORESERVE)
- && (*bits & (EXTENT_DO_ACCOUNTING |
- EXTENT_CLEAR_DATA_RESV)))
- btrfs_free_reserved_data_space_noquota(inode,
+ if (root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID &&
+ do_list && !(state->state & EXTENT_NORESERVE) &&
+ (*bits & EXTENT_CLEAR_DATA_RESV))
+ btrfs_free_reserved_data_space_noquota(
+ &inode->vfs_inode,
state->start, len);
- __percpu_counter_add(&fs_info->delalloc_bytes, -len,
- fs_info->delalloc_batch);
- spin_lock(&BTRFS_I(inode)->lock);
- BTRFS_I(inode)->delalloc_bytes -= len;
- if (do_list && BTRFS_I(inode)->delalloc_bytes == 0 &&
+ percpu_counter_add_batch(&fs_info->delalloc_bytes, -len,
+ fs_info->delalloc_batch);
+ spin_lock(&inode->lock);
+ inode->delalloc_bytes -= len;
+ if (do_list && inode->delalloc_bytes == 0 &&
test_bit(BTRFS_INODE_IN_DELALLOC_LIST,
- &BTRFS_I(inode)->runtime_flags))
+ &inode->runtime_flags))
btrfs_del_delalloc_inode(root, inode);
- spin_unlock(&BTRFS_I(inode)->lock);
+ spin_unlock(&inode->lock);
+ }
+
+ if ((state->state & EXTENT_DELALLOC_NEW) &&
+ (*bits & EXTENT_DELALLOC_NEW)) {
+ spin_lock(&inode->lock);
+ ASSERT(inode->new_delalloc_bytes >= len);
+ inode->new_delalloc_bytes -= len;
+ spin_unlock(&inode->lock);
}
}
@@ -1895,11 +1899,12 @@ int btrfs_merge_bio_hook(struct page *page, unsigned long offset,
* At IO completion time the cums attached on the ordered extent record
* are inserted into the btree
*/
-static int __btrfs_submit_bio_start(struct inode *inode, struct bio *bio,
+static blk_status_t __btrfs_submit_bio_start(void *private_data, struct bio *bio,
int mirror_num, unsigned long bio_flags,
u64 bio_offset)
{
- int ret = 0;
+ struct inode *inode = private_data;
+ blk_status_t ret = 0;
ret = btrfs_csum_one_bio(inode, bio, 0, 0);
BUG_ON(ret); /* -ENOMEM */
@@ -1914,16 +1919,17 @@ static int __btrfs_submit_bio_start(struct inode *inode, struct bio *bio,
* At IO completion time the cums attached on the ordered extent record
* are inserted into the btree
*/
-static int __btrfs_submit_bio_done(struct inode *inode, struct bio *bio,
+static blk_status_t __btrfs_submit_bio_done(void *private_data, struct bio *bio,
int mirror_num, unsigned long bio_flags,
u64 bio_offset)
{
+ struct inode *inode = private_data;
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- int ret;
+ blk_status_t ret;
ret = btrfs_map_bio(fs_info, bio, mirror_num, 1);
if (ret) {
- bio->bi_error = ret;
+ bio->bi_status = ret;
bio_endio(bio);
}
return ret;
@@ -1933,20 +1939,21 @@ static int __btrfs_submit_bio_done(struct inode *inode, struct bio *bio,
* extent_io.c submission hook. This does the right thing for csum calculation
* on write, or reading the csums from the tree before a read
*/
-static int btrfs_submit_bio_hook(struct inode *inode, struct bio *bio,
- int mirror_num, unsigned long bio_flags,
- u64 bio_offset)
+static blk_status_t btrfs_submit_bio_hook(void *private_data, struct bio *bio,
+ int mirror_num, unsigned long bio_flags,
+ u64 bio_offset)
{
+ struct inode *inode = private_data;
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
struct btrfs_root *root = BTRFS_I(inode)->root;
enum btrfs_wq_endio_type metadata = BTRFS_WQ_ENDIO_DATA;
- int ret = 0;
+ blk_status_t ret = 0;
int skip_sum;
int async = !atomic_read(&BTRFS_I(inode)->sync_writers);
skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
- if (btrfs_is_free_space_inode(inode))
+ if (btrfs_is_free_space_inode(BTRFS_I(inode)))
metadata = BTRFS_WQ_ENDIO_FREE_SPACE;
if (bio_op(bio) != REQ_OP_WRITE) {
@@ -1970,8 +1977,8 @@ static int btrfs_submit_bio_hook(struct inode *inode, struct bio *bio,
if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID)
goto mapit;
/* we're doing a write, do the async checksumming */
- ret = btrfs_wq_submit_bio(fs_info, inode, bio, mirror_num,
- bio_flags, bio_offset,
+ ret = btrfs_wq_submit_bio(fs_info, bio, mirror_num, bio_flags,
+ bio_offset, inode,
__btrfs_submit_bio_start,
__btrfs_submit_bio_done);
goto out;
@@ -1985,8 +1992,8 @@ mapit:
ret = btrfs_map_bio(fs_info, bio, mirror_num, 0);
out:
- if (ret < 0) {
- bio->bi_error = ret;
+ if (ret) {
+ bio->bi_status = ret;
bio_endio(bio);
}
return ret;
@@ -1997,8 +2004,7 @@ out:
* at IO completion time based on sums calculated at bio submission time.
*/
static noinline int add_pending_csums(struct btrfs_trans_handle *trans,
- struct inode *inode, u64 file_offset,
- struct list_head *list)
+ struct inode *inode, struct list_head *list)
{
struct btrfs_ordered_sum *sum;
@@ -2030,6 +2036,7 @@ static void btrfs_writepage_fixup_worker(struct btrfs_work *work)
struct btrfs_writepage_fixup *fixup;
struct btrfs_ordered_extent *ordered;
struct extent_state *cached_state = NULL;
+ struct extent_changeset *data_reserved = NULL;
struct page *page;
struct inode *inode;
u64 page_start;
@@ -2056,7 +2063,7 @@ again:
if (PagePrivate2(page))
goto out;
- ordered = btrfs_lookup_ordered_range(inode, page_start,
+ ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), page_start,
PAGE_SIZE);
if (ordered) {
unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start,
@@ -2067,7 +2074,7 @@ again:
goto again;
}
- ret = btrfs_delalloc_reserve_space(inode, page_start,
+ ret = btrfs_delalloc_reserve_space(inode, &data_reserved, page_start,
PAGE_SIZE);
if (ret) {
mapping_set_error(page->mapping, ret);
@@ -2087,6 +2094,7 @@ out_page:
unlock_page(page);
put_page(page);
kfree(fixup);
+ extent_changeset_free(data_reserved);
}
/*
@@ -2138,6 +2146,7 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans,
struct btrfs_path *path;
struct extent_buffer *leaf;
struct btrfs_key ins;
+ u64 qg_released;
int extent_inserted = 0;
int ret;
@@ -2161,7 +2170,7 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans,
goto out;
if (!extent_inserted) {
- ins.objectid = btrfs_ino(inode);
+ ins.objectid = btrfs_ino(BTRFS_I(inode));
ins.offset = file_pos;
ins.type = BTRFS_EXTENT_DATA_KEY;
@@ -2193,14 +2202,17 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans,
ins.objectid = disk_bytenr;
ins.offset = disk_num_bytes;
ins.type = BTRFS_EXTENT_ITEM_KEY;
- ret = btrfs_alloc_reserved_file_extent(trans, root->root_key.objectid,
- btrfs_ino(inode), file_pos,
- ram_bytes, &ins);
+
/*
* Release the reserved range from inode dirty range map, as it is
* already moved into delayed_ref_head
*/
- btrfs_qgroup_release_data(inode, file_pos, ram_bytes);
+ ret = btrfs_qgroup_release_data(inode, file_pos, ram_bytes);
+ if (ret < 0)
+ goto out;
+ qg_released = ret;
+ ret = btrfs_alloc_reserved_file_extent(trans, root->root_key.objectid,
+ btrfs_ino(BTRFS_I(inode)), file_pos, qg_released, &ins);
out:
btrfs_free_path(path);
@@ -2320,7 +2332,7 @@ static noinline int record_one_backref(u64 inum, u64 offset, u64 root_id,
u64 num_bytes;
if (BTRFS_I(inode)->root->root_key.objectid == root_id &&
- inum == btrfs_ino(inode))
+ inum == btrfs_ino(BTRFS_I(inode)))
return 0;
key.objectid = root_id;
@@ -2589,7 +2601,7 @@ static noinline int relink_extent_backref(struct btrfs_path *path,
if (ret)
goto out_free_path;
again:
- key.objectid = btrfs_ino(inode);
+ key.objectid = btrfs_ino(BTRFS_I(inode));
key.type = BTRFS_EXTENT_DATA_KEY;
key.offset = start;
@@ -2768,7 +2780,7 @@ record_old_file_extents(struct inode *inode,
if (!path)
goto out_kfree;
- key.objectid = btrfs_ino(inode);
+ key.objectid = btrfs_ino(BTRFS_I(inode));
key.type = BTRFS_EXTENT_DATA_KEY;
key.offset = new->file_pos;
@@ -2803,7 +2815,7 @@ record_old_file_extents(struct inode *inode,
btrfs_item_key_to_cpu(l, &key, slot);
- if (key.objectid != btrfs_ino(inode))
+ if (key.objectid != btrfs_ino(BTRFS_I(inode)))
break;
if (key.type != BTRFS_EXTENT_DATA_KEY)
break;
@@ -2886,17 +2898,25 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
u64 logical_len = ordered_extent->len;
bool nolock;
bool truncated = false;
+ bool range_locked = false;
+ bool clear_new_delalloc_bytes = false;
+
+ if (!test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) &&
+ !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags) &&
+ !test_bit(BTRFS_ORDERED_DIRECT, &ordered_extent->flags))
+ clear_new_delalloc_bytes = true;
- nolock = btrfs_is_free_space_inode(inode);
+ nolock = btrfs_is_free_space_inode(BTRFS_I(inode));
if (test_bit(BTRFS_ORDERED_IOERR, &ordered_extent->flags)) {
ret = -EIO;
goto out;
}
- btrfs_free_io_failure_record(inode, ordered_extent->file_offset,
- ordered_extent->file_offset +
- ordered_extent->len - 1);
+ btrfs_free_io_failure_record(BTRFS_I(inode),
+ ordered_extent->file_offset,
+ ordered_extent->file_offset +
+ ordered_extent->len - 1);
if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered_extent->flags)) {
truncated = true;
@@ -2914,7 +2934,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
* space for NOCOW range.
* As NOCOW won't cause a new delayed ref, just free the space
*/
- btrfs_qgroup_free_data(inode, ordered_extent->file_offset,
+ btrfs_qgroup_free_data(inode, NULL, ordered_extent->file_offset,
ordered_extent->len);
btrfs_ordered_update_i_size(inode, 0, ordered_extent);
if (nolock)
@@ -2933,13 +2953,14 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
goto out;
}
+ range_locked = true;
lock_extent_bits(io_tree, ordered_extent->file_offset,
ordered_extent->file_offset + ordered_extent->len - 1,
&cached_state);
ret = test_range_bit(io_tree, ordered_extent->file_offset,
ordered_extent->file_offset + ordered_extent->len - 1,
- EXTENT_DEFRAG, 1, cached_state);
+ EXTENT_DEFRAG, 0, cached_state);
if (ret) {
u64 last_snapshot = btrfs_root_last_snapshot(&root->root_item);
if (0 && last_snapshot >= BTRFS_I(inode)->generation)
@@ -2958,7 +2979,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
trans = NULL;
- goto out_unlock;
+ goto out;
}
trans->block_rsv = &fs_info->delalloc_block_rsv;
@@ -2967,7 +2988,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
compress_type = ordered_extent->compress_type;
if (test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
BUG_ON(compress_type);
- ret = btrfs_mark_extent_written(trans, inode,
+ ret = btrfs_mark_extent_written(trans, BTRFS_I(inode),
ordered_extent->file_offset,
ordered_extent->file_offset +
logical_len);
@@ -2990,26 +3011,38 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
trans->transid);
if (ret < 0) {
btrfs_abort_transaction(trans, ret);
- goto out_unlock;
+ goto out;
}
- add_pending_csums(trans, inode, ordered_extent->file_offset,
- &ordered_extent->list);
+ add_pending_csums(trans, inode, &ordered_extent->list);
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, ret);
- goto out_unlock;
+ goto out;
}
ret = 0;
-out_unlock:
- unlock_extent_cached(io_tree, ordered_extent->file_offset,
- ordered_extent->file_offset +
- ordered_extent->len - 1, &cached_state, GFP_NOFS);
out:
+ if (range_locked || clear_new_delalloc_bytes) {
+ unsigned int clear_bits = 0;
+
+ if (range_locked)
+ clear_bits |= EXTENT_LOCKED;
+ if (clear_new_delalloc_bytes)
+ clear_bits |= EXTENT_DELALLOC_NEW;
+ clear_extent_bit(&BTRFS_I(inode)->io_tree,
+ ordered_extent->file_offset,
+ ordered_extent->file_offset +
+ ordered_extent->len - 1,
+ clear_bits,
+ (clear_bits & EXTENT_LOCKED) ? 1 : 0,
+ 0, &cached_state, GFP_NOFS);
+ }
+
if (root != fs_info->tree_root)
- btrfs_delalloc_release_metadata(inode, ordered_extent->len);
+ btrfs_delalloc_release_metadata(BTRFS_I(inode),
+ ordered_extent->len);
if (trans)
btrfs_end_transaction(trans);
@@ -3024,7 +3057,7 @@ out:
clear_extent_uptodate(io_tree, start, end, NULL, GFP_NOFS);
/* Drop the cache for the part of the extent we didn't write. */
- btrfs_drop_extent_cache(inode, start, end, 0);
+ btrfs_drop_extent_cache(BTRFS_I(inode), start, end, 0);
/*
* If the ordered extent had an IOERR or something else went
@@ -3072,7 +3105,7 @@ static void finish_ordered_fn(struct btrfs_work *work)
btrfs_finish_ordered_io(ordered_extent);
}
-static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
+static void btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
struct extent_state *state, int uptodate)
{
struct inode *inode = page->mapping->host;
@@ -3086,9 +3119,9 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
ClearPagePrivate2(page);
if (!btrfs_dec_test_ordered_pending(inode, &ordered_extent, start,
end - start + 1, uptodate))
- return 0;
+ return;
- if (btrfs_is_free_space_inode(inode)) {
+ if (btrfs_is_free_space_inode(BTRFS_I(inode))) {
wq = fs_info->endio_freespace_worker;
func = btrfs_freespace_write_helper;
} else {
@@ -3099,8 +3132,6 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
btrfs_init_work(&ordered_extent->work, func, finish_ordered_fn, NULL,
NULL);
btrfs_queue_work(wq, &ordered_extent->work);
-
- return 0;
}
static int __readpage_endio_check(struct inode *inode,
@@ -3123,9 +3154,8 @@ static int __readpage_endio_check(struct inode *inode,
kunmap_atomic(kaddr);
return 0;
zeroit:
- btrfs_warn_rl(BTRFS_I(inode)->root->fs_info,
- "csum failed ino %llu off %llu csum %u expected csum %u",
- btrfs_ino(inode), start, csum, csum_expected);
+ btrfs_print_data_csum_error(BTRFS_I(inode), start, csum, csum_expected,
+ io_bio->mirror_num);
memset(kaddr + pgoff, 1, len);
flush_dcache_page(page);
kunmap_atomic(kaddr);
@@ -3263,10 +3293,11 @@ void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans,
* NOTE: caller of this function should reserve 5 units of metadata for
* this function.
*/
-int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
+int btrfs_orphan_add(struct btrfs_trans_handle *trans,
+ struct btrfs_inode *inode)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = inode->root;
struct btrfs_block_rsv *block_rsv = NULL;
int reserve = 0;
int insert = 0;
@@ -3288,7 +3319,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
}
if (!test_and_set_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
- &BTRFS_I(inode)->runtime_flags)) {
+ &inode->runtime_flags)) {
#if 0
/*
* For proper ENOSPC handling, we should do orphan
@@ -3305,7 +3336,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
}
if (!test_and_set_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
- &BTRFS_I(inode)->runtime_flags))
+ &inode->runtime_flags))
reserve = 1;
spin_unlock(&root->orphan_lock);
@@ -3316,10 +3347,10 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
if (ret) {
atomic_dec(&root->orphan_inodes);
clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
- &BTRFS_I(inode)->runtime_flags);
+ &inode->runtime_flags);
if (insert)
clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
- &BTRFS_I(inode)->runtime_flags);
+ &inode->runtime_flags);
return ret;
}
}
@@ -3331,12 +3362,12 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
atomic_dec(&root->orphan_inodes);
if (reserve) {
clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
- &BTRFS_I(inode)->runtime_flags);
+ &inode->runtime_flags);
btrfs_orphan_release_metadata(inode);
}
if (ret != -EEXIST) {
clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
- &BTRFS_I(inode)->runtime_flags);
+ &inode->runtime_flags);
btrfs_abort_transaction(trans, ret);
return ret;
}
@@ -3361,20 +3392,20 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
* item for this particular inode.
*/
static int btrfs_orphan_del(struct btrfs_trans_handle *trans,
- struct inode *inode)
+ struct btrfs_inode *inode)
{
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_root *root = inode->root;
int delete_item = 0;
int release_rsv = 0;
int ret = 0;
spin_lock(&root->orphan_lock);
if (test_and_clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
- &BTRFS_I(inode)->runtime_flags))
+ &inode->runtime_flags))
delete_item = 1;
if (test_and_clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
- &BTRFS_I(inode)->runtime_flags))
+ &inode->runtime_flags))
release_rsv = 1;
spin_unlock(&root->orphan_lock);
@@ -3548,7 +3579,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
ret = PTR_ERR(trans);
goto out;
}
- ret = btrfs_orphan_add(trans, inode);
+ ret = btrfs_orphan_add(trans, BTRFS_I(inode));
btrfs_end_transaction(trans);
if (ret) {
iput(inode);
@@ -3557,7 +3588,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
ret = btrfs_truncate(inode);
if (ret)
- btrfs_orphan_del(NULL, inode);
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
} else {
nr_unlink++;
}
@@ -3712,7 +3743,7 @@ static int btrfs_read_locked_inode(struct inode *inode)
set_nlink(inode, btrfs_inode_nlink(leaf, inode_item));
i_uid_write(inode, btrfs_inode_uid(leaf, inode_item));
i_gid_write(inode, btrfs_inode_gid(leaf, inode_item));
- btrfs_i_size_write(inode, btrfs_inode_size(leaf, inode_item));
+ btrfs_i_size_write(BTRFS_I(inode), btrfs_inode_size(leaf, inode_item));
inode->i_atime.tv_sec = btrfs_timespec_sec(leaf, &inode_item->atime);
inode->i_atime.tv_nsec = btrfs_timespec_nsec(leaf, &inode_item->atime);
@@ -3789,7 +3820,7 @@ cache_index:
goto cache_acl;
btrfs_item_key_to_cpu(leaf, &location, path->slots[0]);
- if (location.objectid != btrfs_ino(inode))
+ if (location.objectid != btrfs_ino(BTRFS_I(inode)))
goto cache_acl;
ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
@@ -3811,14 +3842,14 @@ cache_acl:
* any xattrs or acls
*/
maybe_acls = acls_after_inode_item(leaf, path->slots[0],
- btrfs_ino(inode), &first_xattr_slot);
+ btrfs_ino(BTRFS_I(inode)), &first_xattr_slot);
if (first_xattr_slot != -1) {
path->slots[0] = first_xattr_slot;
ret = btrfs_load_inode_props(inode, path);
if (ret)
btrfs_err(fs_info,
"error loading props for ino %llu (root %llu): %d",
- btrfs_ino(inode),
+ btrfs_ino(BTRFS_I(inode)),
root->root_key.objectid, ret);
}
btrfs_free_path(path);
@@ -3960,7 +3991,7 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans,
* The data relocation inode should also be directly updated
* without delay
*/
- if (!btrfs_is_free_space_inode(inode)
+ if (!btrfs_is_free_space_inode(BTRFS_I(inode))
&& root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID
&& !test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags)) {
btrfs_update_root_times(trans, root);
@@ -3993,7 +4024,8 @@ noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans,
*/
static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct inode *dir, struct inode *inode,
+ struct btrfs_inode *dir,
+ struct btrfs_inode *inode,
const char *name, int name_len)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -4040,10 +4072,10 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
* that we delay to delete it, and just do this deletion when
* we update the inode item.
*/
- if (BTRFS_I(inode)->dir_index) {
+ if (inode->dir_index) {
ret = btrfs_delayed_delete_inode_ref(inode);
if (!ret) {
- index = BTRFS_I(inode)->dir_index;
+ index = inode->dir_index;
goto skip_backref;
}
}
@@ -4064,15 +4096,15 @@ skip_backref:
goto err;
}
- ret = btrfs_del_inode_ref_in_log(trans, root, name, name_len,
- inode, dir_ino);
+ ret = btrfs_del_inode_ref_in_log(trans, root, name, name_len, inode,
+ dir_ino);
if (ret != 0 && ret != -ENOENT) {
btrfs_abort_transaction(trans, ret);
goto err;
}
- ret = btrfs_del_dir_entries_in_log(trans, root, name, name_len,
- dir, index);
+ ret = btrfs_del_dir_entries_in_log(trans, root, name, name_len, dir,
+ index);
if (ret == -ENOENT)
ret = 0;
else if (ret)
@@ -4082,26 +4114,26 @@ err:
if (ret)
goto out;
- btrfs_i_size_write(dir, dir->i_size - name_len * 2);
- inode_inc_iversion(inode);
- inode_inc_iversion(dir);
- inode->i_ctime = dir->i_mtime =
- dir->i_ctime = current_time(inode);
- ret = btrfs_update_inode(trans, root, dir);
+ btrfs_i_size_write(dir, dir->vfs_inode.i_size - name_len * 2);
+ inode_inc_iversion(&inode->vfs_inode);
+ inode_inc_iversion(&dir->vfs_inode);
+ inode->vfs_inode.i_ctime = dir->vfs_inode.i_mtime =
+ dir->vfs_inode.i_ctime = current_time(&inode->vfs_inode);
+ ret = btrfs_update_inode(trans, root, &dir->vfs_inode);
out:
return ret;
}
int btrfs_unlink_inode(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct inode *dir, struct inode *inode,
+ struct btrfs_inode *dir, struct btrfs_inode *inode,
const char *name, int name_len)
{
int ret;
ret = __btrfs_unlink_inode(trans, root, dir, inode, name, name_len);
if (!ret) {
- drop_nlink(inode);
- ret = btrfs_update_inode(trans, root, inode);
+ drop_nlink(&inode->vfs_inode);
+ ret = btrfs_update_inode(trans, root, &inode->vfs_inode);
}
return ret;
}
@@ -4139,15 +4171,17 @@ static int btrfs_unlink(struct inode *dir, struct dentry *dentry)
if (IS_ERR(trans))
return PTR_ERR(trans);
- btrfs_record_unlink_dir(trans, dir, d_inode(dentry), 0);
+ btrfs_record_unlink_dir(trans, BTRFS_I(dir), BTRFS_I(d_inode(dentry)),
+ 0);
- ret = btrfs_unlink_inode(trans, root, dir, d_inode(dentry),
- dentry->d_name.name, dentry->d_name.len);
+ ret = btrfs_unlink_inode(trans, root, BTRFS_I(dir),
+ BTRFS_I(d_inode(dentry)), dentry->d_name.name,
+ dentry->d_name.len);
if (ret)
goto out;
if (inode->i_nlink == 0) {
- ret = btrfs_orphan_add(trans, inode);
+ ret = btrfs_orphan_add(trans, BTRFS_I(inode));
if (ret)
goto out;
}
@@ -4170,7 +4204,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
struct btrfs_key key;
u64 index;
int ret;
- u64 dir_ino = btrfs_ino(dir);
+ u64 dir_ino = btrfs_ino(BTRFS_I(dir));
path = btrfs_alloc_path();
if (!path)
@@ -4222,13 +4256,13 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
}
btrfs_release_path(path);
- ret = btrfs_delete_delayed_dir_index(trans, fs_info, dir, index);
+ ret = btrfs_delete_delayed_dir_index(trans, fs_info, BTRFS_I(dir), index);
if (ret) {
btrfs_abort_transaction(trans, ret);
goto out;
}
- btrfs_i_size_write(dir, dir->i_size - name_len * 2);
+ btrfs_i_size_write(BTRFS_I(dir), dir->i_size - name_len * 2);
inode_inc_iversion(dir);
dir->i_mtime = dir->i_ctime = current_time(dir);
ret = btrfs_update_inode_fallback(trans, root, dir);
@@ -4249,14 +4283,14 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
if (inode->i_size > BTRFS_EMPTY_DIR_SIZE)
return -ENOTEMPTY;
- if (btrfs_ino(inode) == BTRFS_FIRST_FREE_OBJECTID)
+ if (btrfs_ino(BTRFS_I(inode)) == BTRFS_FIRST_FREE_OBJECTID)
return -EPERM;
trans = __unlink_start_trans(dir);
if (IS_ERR(trans))
return PTR_ERR(trans);
- if (unlikely(btrfs_ino(inode) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
+ if (unlikely(btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
err = btrfs_unlink_subvol(trans, root, dir,
BTRFS_I(inode)->location.objectid,
dentry->d_name.name,
@@ -4264,17 +4298,18 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
goto out;
}
- err = btrfs_orphan_add(trans, inode);
+ err = btrfs_orphan_add(trans, BTRFS_I(inode));
if (err)
goto out;
last_unlink_trans = BTRFS_I(inode)->last_unlink_trans;
/* now the directory is empty */
- err = btrfs_unlink_inode(trans, root, dir, d_inode(dentry),
- dentry->d_name.name, dentry->d_name.len);
+ err = btrfs_unlink_inode(trans, root, BTRFS_I(dir),
+ BTRFS_I(d_inode(dentry)), dentry->d_name.name,
+ dentry->d_name.len);
if (!err) {
- btrfs_i_size_write(inode, 0);
+ btrfs_i_size_write(BTRFS_I(inode), 0);
/*
* Propagate the last_unlink_trans value of the deleted dir to
* its parent directory. This is to prevent an unrecoverable
@@ -4398,7 +4433,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
int extent_type = -1;
int ret;
int err = 0;
- u64 ino = btrfs_ino(inode);
+ u64 ino = btrfs_ino(BTRFS_I(inode));
u64 bytes_deleted = 0;
bool be_nice = 0;
bool should_throttle = 0;
@@ -4410,7 +4445,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
* for non-free space inodes and ref cows, we want to back off from
* time to time
*/
- if (!btrfs_is_free_space_inode(inode) &&
+ if (!btrfs_is_free_space_inode(BTRFS_I(inode)) &&
test_bit(BTRFS_ROOT_REF_COWS, &root->state))
be_nice = 1;
@@ -4426,7 +4461,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
*/
if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) ||
root == fs_info->tree_root)
- btrfs_drop_extent_cache(inode, ALIGN(new_size,
+ btrfs_drop_extent_cache(BTRFS_I(inode), ALIGN(new_size,
fs_info->sectorsize),
(u64)-1, 0);
@@ -4437,7 +4472,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
* items.
*/
if (min_type == 0 && root == BTRFS_I(inode)->root)
- btrfs_kill_delayed_inode_items(inode);
+ btrfs_kill_delayed_inode_items(BTRFS_I(inode));
key.objectid = ino;
key.offset = (u64)-1;
@@ -4493,28 +4528,25 @@ search_again:
if (extent_type != BTRFS_FILE_EXTENT_INLINE) {
item_end +=
btrfs_file_extent_num_bytes(leaf, fi);
+
+ trace_btrfs_truncate_show_fi_regular(
+ BTRFS_I(inode), leaf, fi,
+ found_key.offset);
} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
item_end += btrfs_file_extent_inline_len(leaf,
path->slots[0], fi);
+
+ trace_btrfs_truncate_show_fi_inline(
+ BTRFS_I(inode), leaf, fi, path->slots[0],
+ found_key.offset);
}
item_end--;
}
if (found_type > min_type) {
del_item = 1;
} else {
- if (item_end < new_size) {
- /*
- * With NO_HOLES mode, for the following mapping
- *
- * [0-4k][hole][8k-12k]
- *
- * if truncating isize down to 6k, it ends up
- * isize being 8k.
- */
- if (btrfs_fs_incompat(root->fs_info, NO_HOLES))
- last_size = new_size;
+ if (item_end < new_size)
break;
- }
if (found_key.offset >= new_size)
del_item = 1;
else
@@ -4697,8 +4729,12 @@ out:
btrfs_abort_transaction(trans, ret);
}
error:
- if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
+ if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
+ ASSERT(last_size >= new_size);
+ if (!err && last_size > new_size)
+ last_size = new_size;
btrfs_ordered_update_i_size(inode, last_size, NULL);
+ }
btrfs_free_path(path);
@@ -4734,6 +4770,7 @@ int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len,
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
struct btrfs_ordered_extent *ordered;
struct extent_state *cached_state = NULL;
+ struct extent_changeset *data_reserved = NULL;
char *kaddr;
u32 blocksize = fs_info->sectorsize;
pgoff_t index = from >> PAGE_SHIFT;
@@ -4748,7 +4785,7 @@ int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len,
(!len || ((len & (blocksize - 1)) == 0)))
goto out;
- ret = btrfs_delalloc_reserve_space(inode,
+ ret = btrfs_delalloc_reserve_space(inode, &data_reserved,
round_down(from, blocksize), blocksize);
if (ret)
goto out;
@@ -4756,7 +4793,7 @@ int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len,
again:
page = find_or_create_page(mapping, index, mask);
if (!page) {
- btrfs_delalloc_release_space(inode,
+ btrfs_delalloc_release_space(inode, data_reserved,
round_down(from, blocksize),
blocksize);
ret = -ENOMEM;
@@ -4828,11 +4865,12 @@ again:
out_unlock:
if (ret)
- btrfs_delalloc_release_space(inode, block_start,
+ btrfs_delalloc_release_space(inode, data_reserved, block_start,
blocksize);
unlock_page(page);
put_page(page);
out:
+ extent_changeset_free(data_reserved);
return ret;
}
@@ -4870,8 +4908,8 @@ static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode,
return ret;
}
- ret = btrfs_insert_file_extent(trans, root, btrfs_ino(inode), offset,
- 0, 0, len, 0, len, 0, 0, 0);
+ ret = btrfs_insert_file_extent(trans, root, btrfs_ino(BTRFS_I(inode)),
+ offset, 0, 0, len, 0, len, 0, 0, 0);
if (ret)
btrfs_abort_transaction(trans, ret);
else
@@ -4918,7 +4956,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
lock_extent_bits(io_tree, hole_start, block_end - 1,
&cached_state);
- ordered = btrfs_lookup_ordered_range(inode, hole_start,
+ ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), hole_start,
block_end - hole_start);
if (!ordered)
break;
@@ -4930,7 +4968,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
cur_offset = hole_start;
while (1) {
- em = btrfs_get_extent(inode, NULL, 0, cur_offset,
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur_offset,
block_end - cur_offset, 0);
if (IS_ERR(em)) {
err = PTR_ERR(em);
@@ -4947,7 +4985,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
hole_size);
if (err)
break;
- btrfs_drop_extent_cache(inode, cur_offset,
+ btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset,
cur_offset + hole_size - 1, 0);
hole_em = alloc_extent_map();
if (!hole_em) {
@@ -4973,7 +5011,8 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
write_unlock(&em_tree->lock);
if (err != -EEXIST)
break;
- btrfs_drop_extent_cache(inode, cur_offset,
+ btrfs_drop_extent_cache(BTRFS_I(inode),
+ cur_offset,
cur_offset +
hole_size - 1, 0);
}
@@ -5070,7 +5109,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
* so we need to guarantee from this point on that everything
* will be consistent.
*/
- ret = btrfs_orphan_add(trans, inode);
+ ret = btrfs_orphan_add(trans, BTRFS_I(inode));
btrfs_end_transaction(trans);
if (ret)
return ret;
@@ -5079,14 +5118,21 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
truncate_setsize(inode, newsize);
/* Disable nonlocked read DIO to avoid the end less truncate */
- btrfs_inode_block_unlocked_dio(inode);
+ btrfs_inode_block_unlocked_dio(BTRFS_I(inode));
inode_dio_wait(inode);
- btrfs_inode_resume_unlocked_dio(inode);
+ btrfs_inode_resume_unlocked_dio(BTRFS_I(inode));
ret = btrfs_truncate(inode);
if (ret && inode->i_nlink) {
int err;
+ /* To get a stable disk_i_size */
+ err = btrfs_wait_ordered_range(inode, 0, (u64)-1);
+ if (err) {
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
+ return err;
+ }
+
/*
* failed to truncate, disk_i_size is only adjusted down
* as we remove extents, so it should represent the true
@@ -5095,11 +5141,11 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
*/
trans = btrfs_join_transaction(root);
if (IS_ERR(trans)) {
- btrfs_orphan_del(NULL, inode);
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
return ret;
}
i_size_write(inode, BTRFS_I(inode)->disk_i_size);
- err = btrfs_orphan_del(trans, inode);
+ err = btrfs_orphan_del(trans, BTRFS_I(inode));
if (err)
btrfs_abort_transaction(trans, err);
btrfs_end_transaction(trans);
@@ -5219,7 +5265,7 @@ static void evict_inode_truncate_pages(struct inode *inode)
* Note, end is the bytenr of last byte, so we need + 1 here.
*/
if (state->state & EXTENT_DELALLOC)
- btrfs_qgroup_free_data(inode, start, end - start + 1);
+ btrfs_qgroup_free_data(inode, NULL, start, end - start + 1);
clear_extent_bit(io_tree, start, end,
EXTENT_LOCKED | EXTENT_DIRTY |
@@ -5257,18 +5303,18 @@ void btrfs_evict_inode(struct inode *inode)
if (inode->i_nlink &&
((btrfs_root_refs(&root->root_item) != 0 &&
root->root_key.objectid != BTRFS_ROOT_TREE_OBJECTID) ||
- btrfs_is_free_space_inode(inode)))
+ btrfs_is_free_space_inode(BTRFS_I(inode))))
goto no_delete;
if (is_bad_inode(inode)) {
- btrfs_orphan_del(NULL, inode);
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
goto no_delete;
}
/* do we really want it for ->i_nlink > 0 and zero btrfs_root_refs? */
if (!special_file(inode->i_mode))
btrfs_wait_ordered_range(inode, 0, (u64)-1);
- btrfs_free_io_failure_record(inode, 0, (u64)-1);
+ btrfs_free_io_failure_record(BTRFS_I(inode), 0, (u64)-1);
if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags)) {
BUG_ON(test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
@@ -5282,22 +5328,22 @@ void btrfs_evict_inode(struct inode *inode)
goto no_delete;
}
- ret = btrfs_commit_inode_delayed_inode(inode);
+ ret = btrfs_commit_inode_delayed_inode(BTRFS_I(inode));
if (ret) {
- btrfs_orphan_del(NULL, inode);
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
goto no_delete;
}
rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP);
if (!rsv) {
- btrfs_orphan_del(NULL, inode);
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
goto no_delete;
}
rsv->size = min_size;
rsv->failfast = 1;
global_rsv = &fs_info->global_block_rsv;
- btrfs_i_size_write(inode, 0);
+ btrfs_i_size_write(BTRFS_I(inode), 0);
/*
* This is a bit simpler than btrfs_truncate since we've already
@@ -5332,14 +5378,14 @@ void btrfs_evict_inode(struct inode *inode)
btrfs_warn(fs_info,
"Could not get space for a delete, will truncate on mount %d",
ret);
- btrfs_orphan_del(NULL, inode);
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
btrfs_free_block_rsv(fs_info, rsv);
goto no_delete;
}
trans = btrfs_join_transaction(root);
if (IS_ERR(trans)) {
- btrfs_orphan_del(NULL, inode);
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
btrfs_free_block_rsv(fs_info, rsv);
goto no_delete;
}
@@ -5365,7 +5411,7 @@ void btrfs_evict_inode(struct inode *inode)
if (ret) {
ret = btrfs_commit_transaction(trans);
if (ret) {
- btrfs_orphan_del(NULL, inode);
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
btrfs_free_block_rsv(fs_info, rsv);
goto no_delete;
}
@@ -5394,20 +5440,20 @@ void btrfs_evict_inode(struct inode *inode)
*/
if (ret == 0) {
trans->block_rsv = root->orphan_block_rsv;
- btrfs_orphan_del(trans, inode);
+ btrfs_orphan_del(trans, BTRFS_I(inode));
} else {
- btrfs_orphan_del(NULL, inode);
+ btrfs_orphan_del(NULL, BTRFS_I(inode));
}
trans->block_rsv = &fs_info->trans_block_rsv;
if (!(root == fs_info->tree_root ||
root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID))
- btrfs_return_ino(root, btrfs_ino(inode));
+ btrfs_return_ino(root, btrfs_ino(BTRFS_I(inode)));
btrfs_end_transaction(trans);
btrfs_btree_balance_dirty(fs_info);
no_delete:
- btrfs_remove_delayed_node(inode);
+ btrfs_remove_delayed_node(BTRFS_I(inode));
clear_inode(inode);
}
@@ -5429,8 +5475,8 @@ static int btrfs_inode_by_name(struct inode *dir, struct dentry *dentry,
if (!path)
return -ENOMEM;
- di = btrfs_lookup_dir_item(NULL, root, path, btrfs_ino(dir), name,
- namelen, 0);
+ di = btrfs_lookup_dir_item(NULL, root, path, btrfs_ino(BTRFS_I(dir)),
+ name, namelen, 0);
if (IS_ERR(di))
ret = PTR_ERR(di);
@@ -5485,7 +5531,7 @@ static int fixup_tree_root_location(struct btrfs_fs_info *fs_info,
leaf = path->nodes[0];
ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_root_ref);
- if (btrfs_root_ref_dirid(leaf, ref) != btrfs_ino(dir) ||
+ if (btrfs_root_ref_dirid(leaf, ref) != btrfs_ino(BTRFS_I(dir)) ||
btrfs_root_ref_name_len(leaf, ref) != dentry->d_name.len)
goto out;
@@ -5520,7 +5566,7 @@ static void inode_tree_add(struct inode *inode)
struct rb_node **p;
struct rb_node *parent;
struct rb_node *new = &BTRFS_I(inode)->rb_node;
- u64 ino = btrfs_ino(inode);
+ u64 ino = btrfs_ino(BTRFS_I(inode));
if (inode_unhashed(inode))
return;
@@ -5531,9 +5577,9 @@ static void inode_tree_add(struct inode *inode)
parent = *p;
entry = rb_entry(parent, struct btrfs_inode, rb_node);
- if (ino < btrfs_ino(&entry->vfs_inode))
+ if (ino < btrfs_ino(BTRFS_I(&entry->vfs_inode)))
p = &parent->rb_left;
- else if (ino > btrfs_ino(&entry->vfs_inode))
+ else if (ino > btrfs_ino(BTRFS_I(&entry->vfs_inode)))
p = &parent->rb_right;
else {
WARN_ON(!(entry->vfs_inode.i_state &
@@ -5593,9 +5639,9 @@ again:
prev = node;
entry = rb_entry(node, struct btrfs_inode, rb_node);
- if (objectid < btrfs_ino(&entry->vfs_inode))
+ if (objectid < btrfs_ino(BTRFS_I(&entry->vfs_inode)))
node = node->rb_left;
- else if (objectid > btrfs_ino(&entry->vfs_inode))
+ else if (objectid > btrfs_ino(BTRFS_I(&entry->vfs_inode)))
node = node->rb_right;
else
break;
@@ -5603,7 +5649,7 @@ again:
if (!node) {
while (prev) {
entry = rb_entry(prev, struct btrfs_inode, rb_node);
- if (objectid <= btrfs_ino(&entry->vfs_inode)) {
+ if (objectid <= btrfs_ino(BTRFS_I(&entry->vfs_inode))) {
node = prev;
break;
}
@@ -5612,7 +5658,7 @@ again:
}
while (node) {
entry = rb_entry(node, struct btrfs_inode, rb_node);
- objectid = btrfs_ino(&entry->vfs_inode) + 1;
+ objectid = btrfs_ino(BTRFS_I(&entry->vfs_inode)) + 1;
inode = igrab(&entry->vfs_inode);
if (inode) {
spin_unlock(&root->inode_lock);
@@ -5796,7 +5842,7 @@ static int btrfs_dentry_delete(const struct dentry *dentry)
if (btrfs_root_refs(&root->root_item) == 0)
return 1;
- if (btrfs_ino(inode) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)
+ if (btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)
return 1;
}
return 0;
@@ -5832,7 +5878,6 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
struct inode *inode = file_inode(file);
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
struct btrfs_root *root = BTRFS_I(inode)->root;
- struct btrfs_item *item;
struct btrfs_dir_item *di;
struct btrfs_key key;
struct btrfs_key found_key;
@@ -5865,7 +5910,7 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
key.type = BTRFS_DIR_INDEX_KEY;
key.offset = ctx->pos;
- key.objectid = btrfs_ino(inode);
+ key.objectid = btrfs_ino(BTRFS_I(inode));
ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
if (ret < 0)
@@ -5883,7 +5928,6 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
continue;
}
- item = btrfs_item_nr(slot);
btrfs_item_key_to_cpu(leaf, &found_key, slot);
if (found_key.objectid != key.objectid)
@@ -5898,7 +5942,7 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
ctx->pos = found_key.offset;
di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
- if (verify_dir_item(fs_info, leaf, di))
+ if (verify_dir_item(fs_info, leaf, slot, di))
goto next;
name_len = btrfs_dir_name_len(leaf, di);
@@ -5974,7 +6018,8 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc)
if (test_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags))
return 0;
- if (btrfs_fs_closing(root->fs_info) && btrfs_is_free_space_inode(inode))
+ if (btrfs_fs_closing(root->fs_info) &&
+ btrfs_is_free_space_inode(BTRFS_I(inode)))
nolock = true;
if (wbc->sync_mode == WB_SYNC_ALL) {
@@ -6054,9 +6099,9 @@ static int btrfs_update_time(struct inode *inode, struct timespec *now,
* and then set the in-memory index_cnt variable to reflect
* free sequence numbers
*/
-static int btrfs_set_inode_index_count(struct inode *inode)
+static int btrfs_set_inode_index_count(struct btrfs_inode *inode)
{
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_root *root = inode->root;
struct btrfs_key key, found_key;
struct btrfs_path *path;
struct extent_buffer *leaf;
@@ -6085,7 +6130,7 @@ static int btrfs_set_inode_index_count(struct inode *inode)
* else has to start at 2
*/
if (path->slots[0] == 0) {
- BTRFS_I(inode)->index_cnt = 2;
+ inode->index_cnt = 2;
goto out;
}
@@ -6096,11 +6141,11 @@ static int btrfs_set_inode_index_count(struct inode *inode)
if (found_key.objectid != btrfs_ino(inode) ||
found_key.type != BTRFS_DIR_INDEX_KEY) {
- BTRFS_I(inode)->index_cnt = 2;
+ inode->index_cnt = 2;
goto out;
}
- BTRFS_I(inode)->index_cnt = found_key.offset + 1;
+ inode->index_cnt = found_key.offset + 1;
out:
btrfs_free_path(path);
return ret;
@@ -6110,11 +6155,11 @@ out:
* helper to find a free sequence number in a given directory. This current
* code is very simple, later versions will do smarter things in the btree
*/
-int btrfs_set_inode_index(struct inode *dir, u64 *index)
+int btrfs_set_inode_index(struct btrfs_inode *dir, u64 *index)
{
int ret = 0;
- if (BTRFS_I(dir)->index_cnt == (u64)-1) {
+ if (dir->index_cnt == (u64)-1) {
ret = btrfs_inode_delayed_dir_index_count(dir);
if (ret) {
ret = btrfs_set_inode_index_count(dir);
@@ -6123,8 +6168,8 @@ int btrfs_set_inode_index(struct inode *dir, u64 *index)
}
}
- *index = BTRFS_I(dir)->index_cnt;
- BTRFS_I(dir)->index_cnt++;
+ *index = dir->index_cnt;
+ dir->index_cnt++;
return ret;
}
@@ -6185,7 +6230,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
if (dir && name) {
trace_btrfs_inode_request(dir);
- ret = btrfs_set_inode_index(dir, index);
+ ret = btrfs_set_inode_index(BTRFS_I(dir), index);
if (ret) {
btrfs_free_path(path);
iput(inode);
@@ -6294,7 +6339,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
if (ret)
btrfs_err(fs_info,
"error inheriting props for ino %llu (root %llu): %d",
- btrfs_ino(inode), root->root_key.objectid, ret);
+ btrfs_ino(BTRFS_I(inode)), root->root_key.objectid, ret);
return inode;
@@ -6320,18 +6365,18 @@ static inline u8 btrfs_inode_type(struct inode *inode)
* inode to the parent directory.
*/
int btrfs_add_link(struct btrfs_trans_handle *trans,
- struct inode *parent_inode, struct inode *inode,
+ struct btrfs_inode *parent_inode, struct btrfs_inode *inode,
const char *name, int name_len, int add_backref, u64 index)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
int ret = 0;
struct btrfs_key key;
- struct btrfs_root *root = BTRFS_I(parent_inode)->root;
+ struct btrfs_root *root = parent_inode->root;
u64 ino = btrfs_ino(inode);
u64 parent_ino = btrfs_ino(parent_inode);
if (unlikely(ino == BTRFS_FIRST_FREE_OBJECTID)) {
- memcpy(&key, &BTRFS_I(inode)->root->root_key, sizeof(key));
+ memcpy(&key, &inode->root->root_key, sizeof(key));
} else {
key.objectid = ino;
key.type = BTRFS_INODE_ITEM_KEY;
@@ -6353,7 +6398,7 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
ret = btrfs_insert_dir_item(trans, root, name, name_len,
parent_inode, &key,
- btrfs_inode_type(inode), index);
+ btrfs_inode_type(&inode->vfs_inode), index);
if (ret == -EEXIST || ret == -EOVERFLOW)
goto fail_dir_item;
else if (ret) {
@@ -6361,12 +6406,12 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
return ret;
}
- btrfs_i_size_write(parent_inode, parent_inode->i_size +
+ btrfs_i_size_write(parent_inode, parent_inode->vfs_inode.i_size +
name_len * 2);
- inode_inc_iversion(parent_inode);
- parent_inode->i_mtime = parent_inode->i_ctime =
- current_time(parent_inode);
- ret = btrfs_update_inode(trans, root, parent_inode);
+ inode_inc_iversion(&parent_inode->vfs_inode);
+ parent_inode->vfs_inode.i_mtime = parent_inode->vfs_inode.i_ctime =
+ current_time(&parent_inode->vfs_inode);
+ ret = btrfs_update_inode(trans, root, &parent_inode->vfs_inode);
if (ret)
btrfs_abort_transaction(trans, ret);
return ret;
@@ -6390,8 +6435,8 @@ fail_dir_item:
}
static int btrfs_add_nondir(struct btrfs_trans_handle *trans,
- struct inode *dir, struct dentry *dentry,
- struct inode *inode, int backref, u64 index)
+ struct btrfs_inode *dir, struct dentry *dentry,
+ struct btrfs_inode *inode, int backref, u64 index)
{
int err = btrfs_add_link(trans, dir, inode,
dentry->d_name.name, dentry->d_name.len,
@@ -6427,8 +6472,8 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
goto out_unlock;
inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
- dentry->d_name.len, btrfs_ino(dir), objectid,
- mode, &index);
+ dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), objectid,
+ mode, &index);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
goto out_unlock;
@@ -6447,7 +6492,8 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
if (err)
goto out_unlock_inode;
- err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
+ err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode),
+ 0, index);
if (err) {
goto out_unlock_inode;
} else {
@@ -6499,8 +6545,8 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
goto out_unlock;
inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
- dentry->d_name.len, btrfs_ino(dir), objectid,
- mode, &index);
+ dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), objectid,
+ mode, &index);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
goto out_unlock;
@@ -6524,7 +6570,8 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
if (err)
goto out_unlock_inode;
- err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
+ err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode),
+ 0, index);
if (err)
goto out_unlock_inode;
@@ -6566,7 +6613,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
if (inode->i_nlink >= BTRFS_LINK_MAX)
return -EMLINK;
- err = btrfs_set_inode_index(dir, &index);
+ err = btrfs_set_inode_index(BTRFS_I(dir), &index);
if (err)
goto fail;
@@ -6590,7 +6637,8 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
ihold(inode);
set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags);
- err = btrfs_add_nondir(trans, dir, dentry, inode, 1, index);
+ err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode),
+ 1, index);
if (err) {
drop_inode = 1;
@@ -6604,12 +6652,12 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
* If new hard link count is 1, it's a file created
* with open(2) O_TMPFILE flag.
*/
- err = btrfs_orphan_del(trans, inode);
+ err = btrfs_orphan_del(trans, BTRFS_I(inode));
if (err)
goto fail;
}
d_instantiate(dentry, inode);
- btrfs_log_new_name(trans, inode, NULL, parent);
+ btrfs_log_new_name(trans, BTRFS_I(inode), NULL, parent);
}
btrfs_balance_delayed_items(fs_info);
@@ -6649,8 +6697,8 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
goto out_fail;
inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
- dentry->d_name.len, btrfs_ino(dir), objectid,
- S_IFDIR | mode, &index);
+ dentry->d_name.len, btrfs_ino(BTRFS_I(dir)), objectid,
+ S_IFDIR | mode, &index);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
goto out_fail;
@@ -6665,13 +6713,14 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
if (err)
goto out_fail_inode;
- btrfs_i_size_write(inode, 0);
+ btrfs_i_size_write(BTRFS_I(inode), 0);
err = btrfs_update_inode(trans, root, inode);
if (err)
goto out_fail_inode;
- err = btrfs_add_link(trans, dir, inode, dentry->d_name.name,
- dentry->d_name.len, 0, index);
+ err = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode),
+ dentry->d_name.name,
+ dentry->d_name.len, 0, index);
if (err)
goto out_fail_inode;
@@ -6788,6 +6837,20 @@ static noinline int uncompress_inline(struct btrfs_path *path,
max_size = min_t(unsigned long, PAGE_SIZE, max_size);
ret = btrfs_decompress(compress_type, tmp, page,
extent_offset, inline_size, max_size);
+
+ /*
+ * decompression code contains a memset to fill in any space between the end
+ * of the uncompressed data and the end of max_size in case the decompressed
+ * data ends up shorter than ram_bytes. That doesn't cover the hole between
+ * the end of an inline extent and the beginning of the next block, so we
+ * cover that region here.
+ */
+
+ if (max_size + pg_offset < PAGE_SIZE) {
+ char *map = kmap(page);
+ memset(map + pg_offset + max_size, 0, PAGE_SIZE - max_size - pg_offset);
+ kunmap(page);
+ }
kfree(tmp);
return ret;
}
@@ -6800,12 +6863,12 @@ static noinline int uncompress_inline(struct btrfs_path *path,
*
* This also copies inline extents directly into the page.
*/
-
-struct extent_map *btrfs_get_extent(struct inode *inode, struct page *page,
- size_t pg_offset, u64 start, u64 len,
- int create)
+struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
+ struct page *page,
+ size_t pg_offset, u64 start, u64 len,
+ int create)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
int ret;
int err = 0;
u64 extent_start = 0;
@@ -6813,13 +6876,13 @@ struct extent_map *btrfs_get_extent(struct inode *inode, struct page *page,
u64 objectid = btrfs_ino(inode);
u32 found_type;
struct btrfs_path *path = NULL;
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_root *root = inode->root;
struct btrfs_file_extent_item *item;
struct extent_buffer *leaf;
struct btrfs_key found_key;
struct extent_map *em = NULL;
- struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
- struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
+ struct extent_map_tree *em_tree = &inode->extent_tree;
+ struct extent_io_tree *io_tree = &inode->io_tree;
struct btrfs_trans_handle *trans = NULL;
const bool new_inline = !page || create;
@@ -6899,11 +6962,18 @@ again:
found_type == BTRFS_FILE_EXTENT_PREALLOC) {
extent_end = extent_start +
btrfs_file_extent_num_bytes(leaf, item);
+
+ trace_btrfs_get_extent_show_fi_regular(inode, leaf, item,
+ extent_start);
} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
size_t size;
size = btrfs_file_extent_inline_len(leaf, path->slots[0], item);
extent_end = ALIGN(extent_start + size,
fs_info->sectorsize);
+
+ trace_btrfs_get_extent_show_fi_inline(inode, leaf, item,
+ path->slots[0],
+ extent_start);
}
next:
if (start >= extent_end) {
@@ -6932,7 +7002,8 @@ next:
goto not_found_em;
}
- btrfs_extent_item_to_extent_map(inode, path, item, new_inline, em);
+ btrfs_extent_item_to_extent_map(inode, path, item,
+ new_inline, em);
if (found_type == BTRFS_FILE_EXTENT_REG ||
found_type == BTRFS_FILE_EXTENT_PREALLOC) {
@@ -7084,9 +7155,10 @@ out:
return em;
}
-struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *page,
- size_t pg_offset, u64 start, u64 len,
- int create)
+struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
+ struct page *page,
+ size_t pg_offset, u64 start, u64 len,
+ int create)
{
struct extent_map *em;
struct extent_map *hole_em = NULL;
@@ -7099,19 +7171,17 @@ struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *pag
em = btrfs_get_extent(inode, page, pg_offset, start, len, create);
if (IS_ERR(em))
return em;
- if (em) {
- /*
- * 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 &&
- !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
- return em;
- else
- hole_em = em;
- }
+ /*
+ * 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 &&
+ !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
+ return em;
+ else
+ hole_em = em;
/* check to see if we've wrapped (len == -1 or similar) */
end = start + len;
@@ -7123,7 +7193,7 @@ struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *pag
em = NULL;
/* ok, we didn't find anything, lets look for delalloc */
- found = count_range_bits(&BTRFS_I(inode)->io_tree, &range_start,
+ found = count_range_bits(&inode->io_tree, &range_start,
end, len, EXTENT_DELALLOC, 1);
found_end = range_start + found;
if (found_end < range_start)
@@ -7225,9 +7295,11 @@ static struct extent_map *btrfs_create_dio_extent(struct inode *inode,
int ret;
if (type != BTRFS_ORDERED_NOCOW) {
- em = create_pinned_em(inode, start, len, orig_start,
- block_start, block_len, orig_block_len,
- ram_bytes, type);
+ em = create_io_em(inode, start, len, orig_start,
+ block_start, block_len, orig_block_len,
+ ram_bytes,
+ BTRFS_COMPRESS_NONE, /* compress_type */
+ type);
if (IS_ERR(em))
goto out;
}
@@ -7236,7 +7308,7 @@ static struct extent_map *btrfs_create_dio_extent(struct inode *inode,
if (ret) {
if (em) {
free_extent_map(em);
- btrfs_drop_extent_cache(inode, start,
+ btrfs_drop_extent_cache(BTRFS_I(inode), start,
start + len - 1, 0);
}
em = ERR_PTR(ret);
@@ -7264,7 +7336,7 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode,
em = btrfs_create_dio_extent(inode, start, ins.offset, start,
ins.objectid, ins.offset, ins.offset,
- ins.offset, 0);
+ ins.offset, BTRFS_ORDERED_REGULAR);
btrfs_dec_block_group_reservations(fs_info, ins.objectid);
if (IS_ERR(em))
btrfs_free_reserved_extent(fs_info, ins.objectid,
@@ -7282,7 +7354,6 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
u64 *ram_bytes)
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_trans_handle *trans;
struct btrfs_path *path;
int ret;
struct extent_buffer *leaf;
@@ -7302,8 +7373,8 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
if (!path)
return -ENOMEM;
- ret = btrfs_lookup_file_extent(NULL, root, path, btrfs_ino(inode),
- offset, 0);
+ ret = btrfs_lookup_file_extent(NULL, root, path,
+ btrfs_ino(BTRFS_I(inode)), offset, 0);
if (ret < 0)
goto out;
@@ -7319,7 +7390,7 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
ret = 0;
leaf = path->nodes[0];
btrfs_item_key_to_cpu(leaf, &key, slot);
- if (key.objectid != btrfs_ino(inode) ||
+ if (key.objectid != btrfs_ino(BTRFS_I(inode)) ||
key.type != BTRFS_EXTENT_DATA_KEY) {
/* not our file or wrong item type, must cow */
goto out;
@@ -7385,15 +7456,9 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
* look for other files referencing this extent, if we
* find any we must cow
*/
- trans = btrfs_join_transaction(root);
- if (IS_ERR(trans)) {
- ret = 0;
- goto out;
- }
- ret = btrfs_cross_ref_exist(trans, root, btrfs_ino(inode),
+ ret = btrfs_cross_ref_exist(root, btrfs_ino(BTRFS_I(inode)),
key.offset - backref_offset, disk_bytenr);
- btrfs_end_transaction(trans);
if (ret) {
ret = 0;
goto out;
@@ -7423,11 +7488,11 @@ out:
bool btrfs_page_exists_in_range(struct inode *inode, loff_t start, loff_t end)
{
struct radix_tree_root *root = &inode->i_mapping->page_tree;
- int found = false;
+ bool found = false;
void **pagep = NULL;
struct page *page = NULL;
- int start_idx;
- int end_idx;
+ unsigned long start_idx;
+ unsigned long end_idx;
start_idx = start >> PAGE_SHIFT;
@@ -7504,7 +7569,7 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
* doing DIO to, so we need to make sure there's no ordered
* extents in this range.
*/
- ordered = btrfs_lookup_ordered_range(inode, lockstart,
+ ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), lockstart,
lockend - lockstart + 1);
/*
@@ -7570,17 +7635,23 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
return ret;
}
-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, u64 ram_bytes,
- int type)
+/* The callers of this must take lock_extent() */
+static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len,
+ u64 orig_start, u64 block_start,
+ u64 block_len, u64 orig_block_len,
+ u64 ram_bytes, int compress_type,
+ int type)
{
struct extent_map_tree *em_tree;
struct extent_map *em;
struct btrfs_root *root = BTRFS_I(inode)->root;
int ret;
+ ASSERT(type == BTRFS_ORDERED_PREALLOC ||
+ type == BTRFS_ORDERED_COMPRESSED ||
+ type == BTRFS_ORDERED_NOCOW ||
+ type == BTRFS_ORDERED_REGULAR);
+
em_tree = &BTRFS_I(inode)->extent_tree;
em = alloc_extent_map();
if (!em)
@@ -7588,8 +7659,6 @@ static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
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;
@@ -7598,15 +7667,23 @@ static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
em->ram_bytes = ram_bytes;
em->generation = -1;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
- if (type == BTRFS_ORDERED_PREALLOC)
+ if (type == BTRFS_ORDERED_PREALLOC) {
set_bit(EXTENT_FLAG_FILLING, &em->flags);
+ } else if (type == BTRFS_ORDERED_COMPRESSED) {
+ set_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
+ em->compress_type = compress_type;
+ }
do {
- btrfs_drop_extent_cache(inode, em->start,
+ btrfs_drop_extent_cache(BTRFS_I(inode), em->start,
em->start + em->len - 1, 0);
write_lock(&em_tree->lock);
ret = add_extent_mapping(em_tree, em, 1);
write_unlock(&em_tree->lock);
+ /*
+ * The caller has taken lock_extent(), who could race with us
+ * to add em?
+ */
} while (ret == -EEXIST);
if (ret) {
@@ -7614,6 +7691,7 @@ static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
return ERR_PTR(ret);
}
+ /* em got 2 refs now, callers needs to do free_extent_map once. */
return em;
}
@@ -7621,10 +7699,8 @@ static void adjust_dio_outstanding_extents(struct inode *inode,
struct btrfs_dio_data *dio_data,
const u64 len)
{
- unsigned num_extents;
+ unsigned num_extents = count_max_extents(len);
- num_extents = (unsigned) div64_u64(len + BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE);
/*
* If we have an outstanding_extents count still set then we're
* within our reservation, otherwise we need to adjust our inode
@@ -7687,7 +7763,7 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
goto err;
}
- em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len, 0);
if (IS_ERR(em)) {
ret = PTR_ERR(em);
goto unlock_err;
@@ -7804,7 +7880,7 @@ unlock:
* Need to update the i_size under the extent lock so buffered
* readers will get the updated i_size when we unlock.
*/
- if (start + len > i_size_read(inode))
+ if (!dio_data->overwrite && start + len > i_size_read(inode))
i_size_write(inode, start + len);
adjust_dio_outstanding_extents(inode, dio_data, len);
@@ -7910,9 +7986,12 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio,
bio_end_io_t *repair_endio, void *repair_arg)
{
struct io_failure_record *failrec;
+ struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
+ struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
struct bio *bio;
int isector;
int read_mode = 0;
+ int segs;
int ret;
BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
@@ -7924,23 +8003,19 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio,
ret = btrfs_check_dio_repairable(inode, failed_bio, failrec,
failed_mirror);
if (!ret) {
- free_io_failure(inode, failrec);
+ free_io_failure(failure_tree, io_tree, failrec);
return -EIO;
}
- if ((failed_bio->bi_vcnt > 1)
- || (failed_bio->bi_io_vec->bv_len
- > btrfs_inode_sectorsize(inode)))
+ segs = bio_segments(failed_bio);
+ if (segs > 1 ||
+ (failed_bio->bi_io_vec->bv_len > btrfs_inode_sectorsize(inode)))
read_mode |= REQ_FAILFAST_DEV;
isector = start - btrfs_io_bio(failed_bio)->logical;
isector >>= inode->i_sb->s_blocksize_bits;
bio = btrfs_create_repair_bio(inode, failed_bio, failrec, page,
pgoff, isector, repair_endio, repair_arg);
- if (!bio) {
- free_io_failure(inode, failrec);
- return -EIO;
- }
bio_set_op_attrs(bio, REQ_OP_READ, read_mode);
btrfs_debug(BTRFS_I(inode)->root->fs_info,
@@ -7949,7 +8024,7 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio,
ret = submit_dio_repair_bio(inode, bio, failrec->this_mirror);
if (ret) {
- free_io_failure(inode, failrec);
+ free_io_failure(failure_tree, io_tree, failrec);
bio_put(bio);
}
@@ -7966,20 +8041,25 @@ struct btrfs_retry_complete {
static void btrfs_retry_endio_nocsum(struct bio *bio)
{
struct btrfs_retry_complete *done = bio->bi_private;
- struct inode *inode;
+ struct inode *inode = done->inode;
struct bio_vec *bvec;
+ struct extent_io_tree *io_tree, *failure_tree;
int i;
- if (bio->bi_error)
+ if (bio->bi_status)
goto end;
ASSERT(bio->bi_vcnt == 1);
- inode = bio->bi_io_vec->bv_page->mapping->host;
+ io_tree = &BTRFS_I(inode)->io_tree;
+ failure_tree = &BTRFS_I(inode)->io_failure_tree;
ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(inode));
done->uptodate = 1;
+ ASSERT(!bio_flagged(bio, BIO_CLONED));
bio_for_each_segment_all(bvec, bio, i)
- clean_io_failure(done->inode, done->start, bvec->bv_page, 0);
+ clean_io_failure(BTRFS_I(inode)->root->fs_info, failure_tree,
+ io_tree, done->start, bvec->bv_page,
+ btrfs_ino(BTRFS_I(inode)), 0);
end:
complete(&done->done);
bio_put(bio);
@@ -7989,36 +8069,40 @@ static int __btrfs_correct_data_nocsum(struct inode *inode,
struct btrfs_io_bio *io_bio)
{
struct btrfs_fs_info *fs_info;
- struct bio_vec *bvec;
+ struct bio_vec bvec;
+ struct bvec_iter iter;
struct btrfs_retry_complete done;
u64 start;
unsigned int pgoff;
u32 sectorsize;
int nr_sectors;
- int i;
int ret;
+ int err = 0;
fs_info = BTRFS_I(inode)->root->fs_info;
sectorsize = fs_info->sectorsize;
start = io_bio->logical;
done.inode = inode;
+ io_bio->bio.bi_iter = io_bio->iter;
- bio_for_each_segment_all(bvec, &io_bio->bio, i) {
- nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec->bv_len);
- pgoff = bvec->bv_offset;
+ bio_for_each_segment(bvec, &io_bio->bio, iter) {
+ nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec.bv_len);
+ pgoff = bvec.bv_offset;
next_block_or_try_again:
done.uptodate = 0;
done.start = start;
init_completion(&done.done);
- ret = dio_read_error(inode, &io_bio->bio, bvec->bv_page,
+ ret = dio_read_error(inode, &io_bio->bio, bvec.bv_page,
pgoff, start, start + sectorsize - 1,
io_bio->mirror_num,
btrfs_retry_endio_nocsum, &done);
- if (ret)
- return ret;
+ if (ret) {
+ err = ret;
+ goto next;
+ }
wait_for_completion(&done.done);
@@ -8027,46 +8111,53 @@ next_block_or_try_again:
goto next_block_or_try_again;
}
+next:
start += sectorsize;
- if (nr_sectors--) {
+ nr_sectors--;
+ if (nr_sectors) {
pgoff += sectorsize;
+ ASSERT(pgoff < PAGE_SIZE);
goto next_block_or_try_again;
}
}
- return 0;
+ return err;
}
static void btrfs_retry_endio(struct bio *bio)
{
struct btrfs_retry_complete *done = bio->bi_private;
struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
- struct inode *inode;
+ struct extent_io_tree *io_tree, *failure_tree;
+ struct inode *inode = done->inode;
struct bio_vec *bvec;
- u64 start;
int uptodate;
int ret;
int i;
- if (bio->bi_error)
+ if (bio->bi_status)
goto end;
uptodate = 1;
- start = done->start;
-
ASSERT(bio->bi_vcnt == 1);
- inode = bio->bi_io_vec->bv_page->mapping->host;
- ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(inode));
+ ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(done->inode));
+ io_tree = &BTRFS_I(inode)->io_tree;
+ failure_tree = &BTRFS_I(inode)->io_failure_tree;
+
+ ASSERT(!bio_flagged(bio, BIO_CLONED));
bio_for_each_segment_all(bvec, bio, i) {
- ret = __readpage_endio_check(done->inode, io_bio, i,
- bvec->bv_page, bvec->bv_offset,
- done->start, bvec->bv_len);
+ ret = __readpage_endio_check(inode, io_bio, i, bvec->bv_page,
+ bvec->bv_offset, done->start,
+ bvec->bv_len);
if (!ret)
- clean_io_failure(done->inode, done->start,
- bvec->bv_page, bvec->bv_offset);
+ clean_io_failure(BTRFS_I(inode)->root->fs_info,
+ failure_tree, io_tree, done->start,
+ bvec->bv_page,
+ btrfs_ino(BTRFS_I(inode)),
+ bvec->bv_offset);
else
uptodate = 0;
}
@@ -8077,11 +8168,12 @@ end:
bio_put(bio);
}
-static int __btrfs_subio_endio_read(struct inode *inode,
- struct btrfs_io_bio *io_bio, int err)
+static blk_status_t __btrfs_subio_endio_read(struct inode *inode,
+ struct btrfs_io_bio *io_bio, blk_status_t err)
{
struct btrfs_fs_info *fs_info;
- struct bio_vec *bvec;
+ struct bio_vec bvec;
+ struct bvec_iter iter;
struct btrfs_retry_complete done;
u64 start;
u64 offset = 0;
@@ -8089,7 +8181,7 @@ static int __btrfs_subio_endio_read(struct inode *inode,
int nr_sectors;
unsigned int pgoff;
int csum_pos;
- int i;
+ bool uptodate = (err == 0);
int ret;
fs_info = BTRFS_I(inode)->root->fs_info;
@@ -8098,29 +8190,31 @@ static int __btrfs_subio_endio_read(struct inode *inode,
err = 0;
start = io_bio->logical;
done.inode = inode;
+ io_bio->bio.bi_iter = io_bio->iter;
- bio_for_each_segment_all(bvec, &io_bio->bio, i) {
- nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec->bv_len);
+ bio_for_each_segment(bvec, &io_bio->bio, iter) {
+ nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec.bv_len);
- pgoff = bvec->bv_offset;
+ pgoff = bvec.bv_offset;
next_block:
- csum_pos = BTRFS_BYTES_TO_BLKS(fs_info, offset);
- ret = __readpage_endio_check(inode, io_bio, csum_pos,
- bvec->bv_page, pgoff, start,
- sectorsize);
- if (likely(!ret))
- goto next;
+ if (uptodate) {
+ csum_pos = BTRFS_BYTES_TO_BLKS(fs_info, offset);
+ ret = __readpage_endio_check(inode, io_bio, csum_pos,
+ bvec.bv_page, pgoff, start, sectorsize);
+ if (likely(!ret))
+ goto next;
+ }
try_again:
done.uptodate = 0;
done.start = start;
init_completion(&done.done);
- ret = dio_read_error(inode, &io_bio->bio, bvec->bv_page,
+ ret = dio_read_error(inode, &io_bio->bio, bvec.bv_page,
pgoff, start, start + sectorsize - 1,
io_bio->mirror_num,
btrfs_retry_endio, &done);
if (ret) {
- err = ret;
+ err = errno_to_blk_status(ret);
goto next;
}
@@ -8136,8 +8230,10 @@ next:
ASSERT(nr_sectors);
- if (--nr_sectors) {
+ nr_sectors--;
+ if (nr_sectors) {
pgoff += sectorsize;
+ ASSERT(pgoff < PAGE_SIZE);
goto next_block;
}
}
@@ -8145,8 +8241,8 @@ next:
return err;
}
-static int btrfs_subio_endio_read(struct inode *inode,
- struct btrfs_io_bio *io_bio, int err)
+static blk_status_t btrfs_subio_endio_read(struct inode *inode,
+ struct btrfs_io_bio *io_bio, blk_status_t err)
{
bool skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
@@ -8166,10 +8262,13 @@ static void btrfs_endio_direct_read(struct bio *bio)
struct inode *inode = dip->inode;
struct bio *dio_bio;
struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
- int err = bio->bi_error;
+ blk_status_t err = bio->bi_status;
- if (dip->flags & BTRFS_DIO_ORIG_BIO_SUBMITTED)
+ if (dip->flags & BTRFS_DIO_ORIG_BIO_SUBMITTED) {
err = btrfs_subio_endio_read(inode, io_bio, err);
+ if (!err)
+ bio->bi_status = 0;
+ }
unlock_extent(&BTRFS_I(inode)->io_tree, dip->logical_offset,
dip->logical_offset + dip->bytes - 1);
@@ -8177,25 +8276,34 @@ static void btrfs_endio_direct_read(struct bio *bio)
kfree(dip);
- dio_bio->bi_error = bio->bi_error;
- dio_end_io(dio_bio, bio->bi_error);
+ dio_bio->bi_status = bio->bi_status;
+ dio_end_io(dio_bio);
if (io_bio->end_io)
- io_bio->end_io(io_bio, err);
+ io_bio->end_io(io_bio, blk_status_to_errno(err));
bio_put(bio);
}
-static void btrfs_endio_direct_write_update_ordered(struct inode *inode,
- const u64 offset,
- const u64 bytes,
- const int uptodate)
+static void __endio_write_update_ordered(struct inode *inode,
+ const u64 offset, const u64 bytes,
+ const bool uptodate)
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
struct btrfs_ordered_extent *ordered = NULL;
+ struct btrfs_workqueue *wq;
+ btrfs_work_func_t func;
u64 ordered_offset = offset;
u64 ordered_bytes = bytes;
int ret;
+ if (btrfs_is_free_space_inode(BTRFS_I(inode))) {
+ wq = fs_info->endio_freespace_worker;
+ func = btrfs_freespace_write_helper;
+ } else {
+ wq = fs_info->endio_write_workers;
+ func = btrfs_endio_write_helper;
+ }
+
again:
ret = btrfs_dec_test_first_ordered_pending(inode, &ordered,
&ordered_offset,
@@ -8204,9 +8312,8 @@ again:
if (!ret)
goto out_test;
- btrfs_init_work(&ordered->work, btrfs_endio_write_helper,
- finish_ordered_fn, NULL, NULL);
- btrfs_queue_work(fs_info->endio_write_workers, &ordered->work);
+ btrfs_init_work(&ordered->work, func, finish_ordered_fn, NULL, NULL);
+ btrfs_queue_work(wq, &ordered->work);
out_test:
/*
* our bio might span multiple ordered extents. If we haven't
@@ -8224,23 +8331,22 @@ static void btrfs_endio_direct_write(struct bio *bio)
struct btrfs_dio_private *dip = bio->bi_private;
struct bio *dio_bio = dip->dio_bio;
- btrfs_endio_direct_write_update_ordered(dip->inode,
- dip->logical_offset,
- dip->bytes,
- !bio->bi_error);
+ __endio_write_update_ordered(dip->inode, dip->logical_offset,
+ dip->bytes, !bio->bi_status);
kfree(dip);
- dio_bio->bi_error = bio->bi_error;
- dio_end_io(dio_bio, bio->bi_error);
+ dio_bio->bi_status = bio->bi_status;
+ dio_end_io(dio_bio);
bio_put(bio);
}
-static int __btrfs_submit_bio_start_direct_io(struct inode *inode,
+static blk_status_t __btrfs_submit_bio_start_direct_io(void *private_data,
struct bio *bio, int mirror_num,
unsigned long bio_flags, u64 offset)
{
- int ret;
+ struct inode *inode = private_data;
+ blk_status_t ret;
ret = btrfs_csum_one_bio(inode, bio, offset, 1);
BUG_ON(ret); /* -ENOMEM */
return 0;
@@ -8249,12 +8355,13 @@ static int __btrfs_submit_bio_start_direct_io(struct inode *inode,
static void btrfs_end_dio_bio(struct bio *bio)
{
struct btrfs_dio_private *dip = bio->bi_private;
- int err = bio->bi_error;
+ blk_status_t err = bio->bi_status;
if (err)
btrfs_warn(BTRFS_I(dip->inode)->root->fs_info,
"direct IO failed ino %llu rw %d,%u sector %#Lx len %u err no %d",
- btrfs_ino(dip->inode), bio_op(bio), bio->bi_opf,
+ btrfs_ino(BTRFS_I(dip->inode)), bio_op(bio),
+ bio->bi_opf,
(unsigned long long)bio->bi_iter.bi_sector,
bio->bi_iter.bi_size, err);
@@ -8278,31 +8385,21 @@ static void btrfs_end_dio_bio(struct bio *bio)
if (dip->errors) {
bio_io_error(dip->orig_bio);
} else {
- dip->dio_bio->bi_error = 0;
+ dip->dio_bio->bi_status = 0;
bio_endio(dip->orig_bio);
}
out:
bio_put(bio);
}
-static struct bio *btrfs_dio_bio_alloc(struct block_device *bdev,
- u64 first_sector, gfp_t gfp_flags)
-{
- struct bio *bio;
- bio = btrfs_bio_alloc(bdev, first_sector, BIO_MAX_PAGES, gfp_flags);
- if (bio)
- bio_associate_current(bio);
- return bio;
-}
-
-static inline int btrfs_lookup_and_bind_dio_csum(struct inode *inode,
+static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode,
struct btrfs_dio_private *dip,
struct bio *bio,
u64 file_offset)
{
struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
struct btrfs_io_bio *orig_io_bio = btrfs_io_bio(dip->orig_bio);
- int ret;
+ blk_status_t ret;
/*
* We load all the csum data we need when we submit
@@ -8333,7 +8430,7 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
struct btrfs_dio_private *dip = bio->bi_private;
bool write = bio_op(bio) == REQ_OP_WRITE;
- int ret;
+ blk_status_t ret;
if (async_submit)
async_submit = !atomic_read(&BTRFS_I(inode)->sync_writers);
@@ -8350,8 +8447,8 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
goto map;
if (write && async_submit) {
- ret = btrfs_wq_submit_bio(fs_info, inode, bio, 0, 0,
- file_offset,
+ ret = btrfs_wq_submit_bio(fs_info, bio, 0, 0,
+ file_offset, inode,
__btrfs_submit_bio_start_direct_io,
__btrfs_submit_bio_done);
goto err;
@@ -8381,103 +8478,83 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip,
{
struct inode *inode = dip->inode;
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root *root = BTRFS_I(inode)->root;
struct bio *bio;
struct bio *orig_bio = dip->orig_bio;
- struct bio_vec *bvec;
u64 start_sector = orig_bio->bi_iter.bi_sector;
u64 file_offset = dip->logical_offset;
- u64 submit_len = 0;
u64 map_length;
- u32 blocksize = fs_info->sectorsize;
int async_submit = 0;
- int nr_sectors;
+ u64 submit_len;
+ int clone_offset = 0;
+ int clone_len;
int ret;
- int i, j;
map_length = orig_bio->bi_iter.bi_size;
+ submit_len = map_length;
ret = btrfs_map_block(fs_info, btrfs_op(orig_bio), start_sector << 9,
&map_length, NULL, 0);
if (ret)
return -EIO;
- if (map_length >= orig_bio->bi_iter.bi_size) {
+ if (map_length >= submit_len) {
bio = orig_bio;
dip->flags |= BTRFS_DIO_ORIG_BIO_SUBMITTED;
goto submit;
}
/* async crcs make it difficult to collect full stripe writes. */
- if (btrfs_get_alloc_profile(root, 1) & BTRFS_BLOCK_GROUP_RAID56_MASK)
+ if (btrfs_data_alloc_profile(fs_info) & BTRFS_BLOCK_GROUP_RAID56_MASK)
async_submit = 0;
else
async_submit = 1;
- bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS);
- if (!bio)
- return -ENOMEM;
-
- bio->bi_opf = orig_bio->bi_opf;
- bio->bi_private = dip;
- bio->bi_end_io = btrfs_end_dio_bio;
- btrfs_io_bio(bio)->logical = file_offset;
+ /* bio split */
+ ASSERT(map_length <= INT_MAX);
atomic_inc(&dip->pending_bios);
+ do {
+ clone_len = min_t(int, submit_len, map_length);
- bio_for_each_segment_all(bvec, orig_bio, j) {
- nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec->bv_len);
- i = 0;
-next_block:
- if (unlikely(map_length < submit_len + blocksize ||
- bio_add_page(bio, bvec->bv_page, blocksize,
- bvec->bv_offset + (i * blocksize)) < blocksize)) {
- /*
- * inc the count before we submit the bio so
- * we know the end IO handler won't happen before
- * we inc the count. Otherwise, the dip might get freed
- * before we're done setting it up
- */
- atomic_inc(&dip->pending_bios);
- ret = __btrfs_submit_dio_bio(bio, inode,
- file_offset, skip_sum,
- async_submit);
- if (ret) {
- bio_put(bio);
- atomic_dec(&dip->pending_bios);
- goto out_err;
- }
-
- start_sector += submit_len >> 9;
- file_offset += submit_len;
+ /*
+ * This will never fail as it's passing GPF_NOFS and
+ * the allocation is backed by btrfs_bioset.
+ */
+ bio = btrfs_bio_clone_partial(orig_bio, clone_offset,
+ clone_len);
+ bio->bi_private = dip;
+ bio->bi_end_io = btrfs_end_dio_bio;
+ btrfs_io_bio(bio)->logical = file_offset;
+
+ ASSERT(submit_len >= clone_len);
+ submit_len -= clone_len;
+ if (submit_len == 0)
+ break;
- submit_len = 0;
+ /*
+ * Increase the count before we submit the bio so we know
+ * the end IO handler won't happen before we increase the
+ * count. Otherwise, the dip might get freed before we're
+ * done setting it up.
+ */
+ atomic_inc(&dip->pending_bios);
- bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev,
- start_sector, GFP_NOFS);
- if (!bio)
- goto out_err;
- bio->bi_opf = orig_bio->bi_opf;
- bio->bi_private = dip;
- bio->bi_end_io = btrfs_end_dio_bio;
- btrfs_io_bio(bio)->logical = file_offset;
+ ret = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum,
+ async_submit);
+ if (ret) {
+ bio_put(bio);
+ atomic_dec(&dip->pending_bios);
+ goto out_err;
+ }
- map_length = orig_bio->bi_iter.bi_size;
- ret = btrfs_map_block(fs_info, btrfs_op(orig_bio),
- start_sector << 9,
- &map_length, NULL, 0);
- if (ret) {
- bio_put(bio);
- goto out_err;
- }
+ clone_offset += clone_len;
+ start_sector += clone_len >> 9;
+ file_offset += clone_len;
- goto next_block;
- } else {
- submit_len += blocksize;
- if (--nr_sectors) {
- i++;
- goto next_block;
- }
- }
- }
+ map_length = submit_len;
+ ret = btrfs_map_block(fs_info, btrfs_op(orig_bio),
+ start_sector << 9, &map_length, NULL, 0);
+ if (ret)
+ goto out_err;
+ } while (submit_len > 0);
submit:
ret = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum,
@@ -8504,19 +8581,15 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode,
loff_t file_offset)
{
struct btrfs_dio_private *dip = NULL;
- struct bio *io_bio = NULL;
- struct btrfs_io_bio *btrfs_bio;
+ struct bio *bio = NULL;
+ struct btrfs_io_bio *io_bio;
int skip_sum;
bool write = (bio_op(dio_bio) == REQ_OP_WRITE);
int ret = 0;
skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
- io_bio = btrfs_bio_clone(dio_bio, GFP_NOFS);
- if (!io_bio) {
- ret = -ENOMEM;
- goto free_ordered;
- }
+ bio = btrfs_bio_clone(dio_bio);
dip = kzalloc(sizeof(*dip), GFP_NOFS);
if (!dip) {
@@ -8529,17 +8602,17 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode,
dip->logical_offset = file_offset;
dip->bytes = dio_bio->bi_iter.bi_size;
dip->disk_bytenr = (u64)dio_bio->bi_iter.bi_sector << 9;
- io_bio->bi_private = dip;
- dip->orig_bio = io_bio;
+ bio->bi_private = dip;
+ dip->orig_bio = bio;
dip->dio_bio = dio_bio;
atomic_set(&dip->pending_bios, 0);
- btrfs_bio = btrfs_io_bio(io_bio);
- btrfs_bio->logical = file_offset;
+ io_bio = btrfs_io_bio(bio);
+ io_bio->logical = file_offset;
if (write) {
- io_bio->bi_end_io = btrfs_endio_direct_write;
+ bio->bi_end_io = btrfs_endio_direct_write;
} else {
- io_bio->bi_end_io = btrfs_endio_direct_read;
+ bio->bi_end_io = btrfs_endio_direct_read;
dip->subio_endio = btrfs_subio_endio_read;
}
@@ -8562,8 +8635,8 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode,
if (!ret)
return;
- if (btrfs_bio->end_io)
- btrfs_bio->end_io(btrfs_bio, ret);
+ if (io_bio->end_io)
+ io_bio->end_io(io_bio, ret);
free_ordered:
/*
@@ -8575,35 +8648,34 @@ free_ordered:
* same as btrfs_endio_direct_[write|read] because we can't call these
* callbacks - they require an allocated dip and a clone of dio_bio.
*/
- if (io_bio && dip) {
- io_bio->bi_error = -EIO;
- bio_endio(io_bio);
+ if (bio && dip) {
+ bio_io_error(bio);
/*
- * The end io callbacks free our dip, do the final put on io_bio
+ * The end io callbacks free our dip, do the final put on bio
* and all the cleanup and final put for dio_bio (through
* dio_end_io()).
*/
dip = NULL;
- io_bio = NULL;
+ bio = NULL;
} else {
if (write)
- btrfs_endio_direct_write_update_ordered(inode,
+ __endio_write_update_ordered(inode,
file_offset,
dio_bio->bi_iter.bi_size,
- 0);
+ false);
else
unlock_extent(&BTRFS_I(inode)->io_tree, file_offset,
file_offset + dio_bio->bi_iter.bi_size - 1);
- dio_bio->bi_error = -EIO;
+ dio_bio->bi_status = BLK_STS_IOERR;
/*
* Releases and cleans up our dio_bio, no need to bio_put()
* nor bio_endio()/bio_io_error() against dio_bio.
*/
- dio_end_io(dio_bio, ret);
+ dio_end_io(dio_bio);
}
- if (io_bio)
- bio_put(io_bio);
+ if (bio)
+ bio_put(bio);
kfree(dip);
}
@@ -8647,6 +8719,7 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
struct inode *inode = file->f_mapping->host;
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
struct btrfs_dio_data dio_data = { 0 };
+ struct extent_changeset *data_reserved = NULL;
loff_t offset = iocb->ki_pos;
size_t count = 0;
int flags = 0;
@@ -8679,15 +8752,18 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
* not unlock the i_mutex at this case.
*/
if (offset + count <= inode->i_size) {
+ dio_data.overwrite = 1;
inode_unlock(inode);
relock = true;
+ } else if (iocb->ki_flags & IOCB_NOWAIT) {
+ ret = -EAGAIN;
+ goto out;
}
- ret = btrfs_delalloc_reserve_space(inode, offset, count);
+ ret = btrfs_delalloc_reserve_space(inode, &data_reserved,
+ offset, count);
if (ret)
goto out;
- dio_data.outstanding_extents = div64_u64(count +
- BTRFS_MAX_EXTENT_SIZE - 1,
- BTRFS_MAX_EXTENT_SIZE);
+ dio_data.outstanding_extents = count_max_extents(count);
/*
* We need to know how many extents we reserved so that we can
@@ -8716,8 +8792,8 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
current->journal_info = NULL;
if (ret < 0 && ret != -EIOCBQUEUED) {
if (dio_data.reserve)
- btrfs_delalloc_release_space(inode, offset,
- dio_data.reserve);
+ btrfs_delalloc_release_space(inode, data_reserved,
+ offset, dio_data.reserve);
/*
* On error we might have left some ordered extents
* without submitting corresponding bios for them, so
@@ -8726,14 +8802,14 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
*/
if (dio_data.unsubmitted_oe_range_start <
dio_data.unsubmitted_oe_range_end)
- btrfs_endio_direct_write_update_ordered(inode,
+ __endio_write_update_ordered(inode,
dio_data.unsubmitted_oe_range_start,
dio_data.unsubmitted_oe_range_end -
dio_data.unsubmitted_oe_range_start,
- 0);
+ false);
} else if (ret >= 0 && (size_t)ret < count)
- btrfs_delalloc_release_space(inode, offset,
- count - (size_t)ret);
+ btrfs_delalloc_release_space(inode, data_reserved,
+ offset, count - (size_t)ret);
}
out:
if (wakeup)
@@ -8741,6 +8817,7 @@ out:
if (relock)
inode_lock(inode);
+ extent_changeset_free(data_reserved);
return ret;
}
@@ -8831,7 +8908,7 @@ static int btrfs_releasepage(struct page *page, gfp_t gfp_flags)
{
if (PageWriteback(page) || PageDirty(page))
return 0;
- return __btrfs_releasepage(page, gfp_flags & GFP_NOFS);
+ return __btrfs_releasepage(page, gfp_flags);
}
static void btrfs_invalidatepage(struct page *page, unsigned int offset,
@@ -8866,7 +8943,7 @@ static void btrfs_invalidatepage(struct page *page, unsigned int offset,
lock_extent_bits(tree, page_start, page_end, &cached_state);
again:
start = page_start;
- ordered = btrfs_lookup_ordered_range(inode, start,
+ ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start,
page_end - start + 1);
if (ordered) {
end = min(page_end, ordered->file_offset + ordered->len - 1);
@@ -8877,6 +8954,7 @@ again:
if (!inode_evicting)
clear_extent_bit(tree, start, end,
EXTENT_DIRTY | EXTENT_DELALLOC |
+ EXTENT_DELALLOC_NEW |
EXTENT_LOCKED | EXTENT_DO_ACCOUNTING |
EXTENT_DEFRAG, 1, 0, &cached_state,
GFP_NOFS);
@@ -8930,12 +9008,12 @@ again:
* free the entire extent.
*/
if (PageDirty(page))
- btrfs_qgroup_free_data(inode, page_start, PAGE_SIZE);
+ btrfs_qgroup_free_data(inode, NULL, page_start, PAGE_SIZE);
if (!inode_evicting) {
clear_extent_bit(tree, page_start, page_end,
EXTENT_LOCKED | EXTENT_DIRTY |
- EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING |
- EXTENT_DEFRAG, 1, 1,
+ EXTENT_DELALLOC | EXTENT_DELALLOC_NEW |
+ EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 1, 1,
&cached_state, GFP_NOFS);
__btrfs_releasepage(page, GFP_NOFS);
@@ -8964,14 +9042,15 @@ again:
* beyond EOF, then the page is guaranteed safe against truncation until we
* unlock the page.
*/
-int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
+int btrfs_page_mkwrite(struct vm_fault *vmf)
{
struct page *page = vmf->page;
- struct inode *inode = file_inode(vma->vm_file);
+ struct inode *inode = file_inode(vmf->vma->vm_file);
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
struct btrfs_ordered_extent *ordered;
struct extent_state *cached_state = NULL;
+ struct extent_changeset *data_reserved = NULL;
char *kaddr;
unsigned long zero_start;
loff_t size;
@@ -8997,10 +9076,10 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
* end up waiting indefinitely to get a lock on the page currently
* being processed by btrfs_page_mkwrite() function.
*/
- ret = btrfs_delalloc_reserve_space(inode, page_start,
+ ret = btrfs_delalloc_reserve_space(inode, &data_reserved, page_start,
reserved_space);
if (!ret) {
- ret = file_update_time(vma->vm_file);
+ ret = file_update_time(vmf->vma->vm_file);
reserved = 1;
}
if (ret) {
@@ -9032,7 +9111,8 @@ again:
* we can't set the delalloc bits if there are pending ordered
* extents. Drop our locks and wait for them to finish
*/
- ordered = btrfs_lookup_ordered_range(inode, page_start, page_end);
+ ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), page_start,
+ PAGE_SIZE);
if (ordered) {
unlock_extent_cached(io_tree, page_start, page_end,
&cached_state, GFP_NOFS);
@@ -9050,17 +9130,17 @@ again:
spin_lock(&BTRFS_I(inode)->lock);
BTRFS_I(inode)->outstanding_extents++;
spin_unlock(&BTRFS_I(inode)->lock);
- btrfs_delalloc_release_space(inode, page_start,
- PAGE_SIZE - reserved_space);
+ btrfs_delalloc_release_space(inode, data_reserved,
+ page_start, PAGE_SIZE - reserved_space);
}
}
/*
- * XXX - page_mkwrite gets called every time the page is dirtied, even
- * if it was already dirty, so for space accounting reasons we need to
- * clear any delalloc bits for the range we are fixing to save. There
- * is probably a better way to do this, but for now keep consistent with
- * prepare_pages in the normal write path.
+ * page_mkwrite gets called when the page is firstly dirtied after it's
+ * faulted in, but write(2) could also dirty a page and set delalloc
+ * bits, thus in this case for space account reason, we still need to
+ * clear any delalloc bits within this page range since we have to
+ * reserve data&meta space before lock_page() (see above comments).
*/
clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, end,
EXTENT_DIRTY | EXTENT_DELALLOC |
@@ -9102,13 +9182,16 @@ again:
out_unlock:
if (!ret) {
sb_end_pagefault(inode->i_sb);
+ extent_changeset_free(data_reserved);
return VM_FAULT_LOCKED;
}
unlock_page(page);
out:
- btrfs_delalloc_release_space(inode, page_start, reserved_space);
+ btrfs_delalloc_release_space(inode, data_reserved, page_start,
+ reserved_space);
out_noreserve:
sb_end_pagefault(inode->i_sb);
+ extent_changeset_free(data_reserved);
return ret;
}
@@ -9230,7 +9313,7 @@ static int btrfs_truncate(struct inode *inode)
if (ret == 0 && inode->i_nlink > 0) {
trans->block_rsv = root->orphan_block_rsv;
- ret = btrfs_orphan_del(trans, inode);
+ ret = btrfs_orphan_del(trans, BTRFS_I(inode));
if (ret)
err = ret;
}
@@ -9275,7 +9358,7 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
inode->i_fop = &btrfs_dir_file_operations;
set_nlink(inode, 1);
- btrfs_i_size_write(inode, 0);
+ btrfs_i_size_write(BTRFS_I(inode), 0);
unlock_new_inode(inode);
err = btrfs_subvol_inherit_props(trans, new_root, parent_root);
@@ -9305,6 +9388,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
ei->last_sub_trans = 0;
ei->logged_trans = 0;
ei->delalloc_bytes = 0;
+ ei->new_delalloc_bytes = 0;
ei->defrag_bytes = 0;
ei->disk_i_size = 0;
ei->flags = 0;
@@ -9329,8 +9413,8 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
inode = &ei->vfs_inode;
extent_map_tree_init(&ei->extent_tree);
- extent_io_tree_init(&ei->io_tree, &inode->i_data);
- extent_io_tree_init(&ei->io_failure_tree, &inode->i_data);
+ extent_io_tree_init(&ei->io_tree, inode);
+ extent_io_tree_init(&ei->io_failure_tree, inode);
ei->io_tree.track_uptodate = 1;
ei->io_failure_tree.track_uptodate = 1;
atomic_set(&ei->sync_writers, 0);
@@ -9348,7 +9432,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
void btrfs_test_destroy_inode(struct inode *inode)
{
- btrfs_drop_extent_cache(inode, 0, (u64)-1, 0);
+ btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0);
kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode));
}
#endif
@@ -9370,6 +9454,7 @@ void btrfs_destroy_inode(struct inode *inode)
WARN_ON(BTRFS_I(inode)->outstanding_extents);
WARN_ON(BTRFS_I(inode)->reserved_extents);
WARN_ON(BTRFS_I(inode)->delalloc_bytes);
+ WARN_ON(BTRFS_I(inode)->new_delalloc_bytes);
WARN_ON(BTRFS_I(inode)->csum_bytes);
WARN_ON(BTRFS_I(inode)->defrag_bytes);
@@ -9384,7 +9469,7 @@ void btrfs_destroy_inode(struct inode *inode)
if (test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
&BTRFS_I(inode)->runtime_flags)) {
btrfs_info(fs_info, "inode %llu still on the orphan list",
- btrfs_ino(inode));
+ btrfs_ino(BTRFS_I(inode)));
atomic_dec(&root->orphan_inodes);
}
@@ -9403,7 +9488,7 @@ void btrfs_destroy_inode(struct inode *inode)
}
btrfs_qgroup_check_reserved_leak(inode);
inode_tree_del(inode);
- btrfs_drop_extent_cache(inode, 0, (u64)-1, 0);
+ btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0);
free:
call_rcu(&inode->i_rcu, btrfs_i_callback);
}
@@ -9438,7 +9523,6 @@ void btrfs_destroy_cachep(void)
rcu_barrier();
kmem_cache_destroy(btrfs_inode_cachep);
kmem_cache_destroy(btrfs_trans_handle_cachep);
- kmem_cache_destroy(btrfs_transaction_cachep);
kmem_cache_destroy(btrfs_path_cachep);
kmem_cache_destroy(btrfs_free_space_cachep);
}
@@ -9458,12 +9542,6 @@ int btrfs_init_cachep(void)
if (!btrfs_trans_handle_cachep)
goto fail;
- btrfs_transaction_cachep = kmem_cache_create("btrfs_transaction",
- sizeof(struct btrfs_transaction), 0,
- SLAB_TEMPORARY | SLAB_MEM_SPREAD, NULL);
- if (!btrfs_transaction_cachep)
- goto fail;
-
btrfs_path_cachep = kmem_cache_create("btrfs_path",
sizeof(struct btrfs_path), 0,
SLAB_MEM_SPREAD, NULL);
@@ -9482,18 +9560,36 @@ fail:
return -ENOMEM;
}
-static int btrfs_getattr(struct vfsmount *mnt,
- struct dentry *dentry, struct kstat *stat)
+static int btrfs_getattr(const struct path *path, struct kstat *stat,
+ u32 request_mask, unsigned int flags)
{
u64 delalloc_bytes;
- struct inode *inode = d_inode(dentry);
+ struct inode *inode = d_inode(path->dentry);
u32 blocksize = inode->i_sb->s_blocksize;
+ u32 bi_flags = BTRFS_I(inode)->flags;
+
+ stat->result_mask |= STATX_BTIME;
+ stat->btime.tv_sec = BTRFS_I(inode)->i_otime.tv_sec;
+ stat->btime.tv_nsec = BTRFS_I(inode)->i_otime.tv_nsec;
+ if (bi_flags & BTRFS_INODE_APPEND)
+ stat->attributes |= STATX_ATTR_APPEND;
+ if (bi_flags & BTRFS_INODE_COMPRESS)
+ stat->attributes |= STATX_ATTR_COMPRESSED;
+ if (bi_flags & BTRFS_INODE_IMMUTABLE)
+ stat->attributes |= STATX_ATTR_IMMUTABLE;
+ if (bi_flags & BTRFS_INODE_NODUMP)
+ stat->attributes |= STATX_ATTR_NODUMP;
+
+ stat->attributes_mask |= (STATX_ATTR_APPEND |
+ STATX_ATTR_COMPRESSED |
+ STATX_ATTR_IMMUTABLE |
+ STATX_ATTR_NODUMP);
generic_fillattr(inode, stat);
stat->dev = BTRFS_I(inode)->root->anon_dev;
spin_lock(&BTRFS_I(inode)->lock);
- delalloc_bytes = BTRFS_I(inode)->delalloc_bytes;
+ delalloc_bytes = BTRFS_I(inode)->new_delalloc_bytes;
spin_unlock(&BTRFS_I(inode)->lock);
stat->blocks = (ALIGN(inode_get_bytes(inode), blocksize) +
ALIGN(delalloc_bytes, blocksize)) >> 9;
@@ -9513,8 +9609,8 @@ static int btrfs_rename_exchange(struct inode *old_dir,
struct inode *old_inode = old_dentry->d_inode;
struct timespec ctime = current_time(old_inode);
struct dentry *parent;
- u64 old_ino = btrfs_ino(old_inode);
- u64 new_ino = btrfs_ino(new_inode);
+ u64 old_ino = btrfs_ino(BTRFS_I(old_inode));
+ u64 new_ino = btrfs_ino(BTRFS_I(new_inode));
u64 old_idx = 0;
u64 new_idx = 0;
u64 root_objectid;
@@ -9550,10 +9646,10 @@ static int btrfs_rename_exchange(struct inode *old_dir,
* We need to find a free sequence number both in the source and
* in the destination directory for the exchange.
*/
- ret = btrfs_set_inode_index(new_dir, &old_idx);
+ ret = btrfs_set_inode_index(BTRFS_I(new_dir), &old_idx);
if (ret)
goto out_fail;
- ret = btrfs_set_inode_index(old_dir, &new_idx);
+ ret = btrfs_set_inode_index(BTRFS_I(old_dir), &new_idx);
if (ret)
goto out_fail;
@@ -9571,7 +9667,8 @@ static int btrfs_rename_exchange(struct inode *old_dir,
new_dentry->d_name.name,
new_dentry->d_name.len,
old_ino,
- btrfs_ino(new_dir), old_idx);
+ btrfs_ino(BTRFS_I(new_dir)),
+ old_idx);
if (ret)
goto out_fail;
}
@@ -9587,7 +9684,8 @@ static int btrfs_rename_exchange(struct inode *old_dir,
old_dentry->d_name.name,
old_dentry->d_name.len,
new_ino,
- btrfs_ino(old_dir), new_idx);
+ btrfs_ino(BTRFS_I(old_dir)),
+ new_idx);
if (ret)
goto out_fail;
}
@@ -9603,8 +9701,10 @@ static int btrfs_rename_exchange(struct inode *old_dir,
new_inode->i_ctime = ctime;
if (old_dentry->d_parent != new_dentry->d_parent) {
- btrfs_record_unlink_dir(trans, old_dir, old_inode, 1);
- btrfs_record_unlink_dir(trans, new_dir, new_inode, 1);
+ btrfs_record_unlink_dir(trans, BTRFS_I(old_dir),
+ BTRFS_I(old_inode), 1);
+ btrfs_record_unlink_dir(trans, BTRFS_I(new_dir),
+ BTRFS_I(new_inode), 1);
}
/* src is a subvolume */
@@ -9615,8 +9715,8 @@ static int btrfs_rename_exchange(struct inode *old_dir,
old_dentry->d_name.name,
old_dentry->d_name.len);
} else { /* src is an inode */
- ret = __btrfs_unlink_inode(trans, root, old_dir,
- old_dentry->d_inode,
+ ret = __btrfs_unlink_inode(trans, root, BTRFS_I(old_dir),
+ BTRFS_I(old_dentry->d_inode),
old_dentry->d_name.name,
old_dentry->d_name.len);
if (!ret)
@@ -9635,8 +9735,8 @@ static int btrfs_rename_exchange(struct inode *old_dir,
new_dentry->d_name.name,
new_dentry->d_name.len);
} else { /* dest is an inode */
- ret = __btrfs_unlink_inode(trans, dest, new_dir,
- new_dentry->d_inode,
+ ret = __btrfs_unlink_inode(trans, dest, BTRFS_I(new_dir),
+ BTRFS_I(new_dentry->d_inode),
new_dentry->d_name.name,
new_dentry->d_name.len);
if (!ret)
@@ -9647,7 +9747,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
goto out_fail;
}
- ret = btrfs_add_link(trans, new_dir, old_inode,
+ ret = btrfs_add_link(trans, BTRFS_I(new_dir), BTRFS_I(old_inode),
new_dentry->d_name.name,
new_dentry->d_name.len, 0, old_idx);
if (ret) {
@@ -9655,7 +9755,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
goto out_fail;
}
- ret = btrfs_add_link(trans, old_dir, new_inode,
+ ret = btrfs_add_link(trans, BTRFS_I(old_dir), BTRFS_I(new_inode),
old_dentry->d_name.name,
old_dentry->d_name.len, 0, new_idx);
if (ret) {
@@ -9670,13 +9770,15 @@ static int btrfs_rename_exchange(struct inode *old_dir,
if (root_log_pinned) {
parent = new_dentry->d_parent;
- btrfs_log_new_name(trans, old_inode, old_dir, parent);
+ btrfs_log_new_name(trans, BTRFS_I(old_inode), BTRFS_I(old_dir),
+ parent);
btrfs_end_log_trans(root);
root_log_pinned = false;
}
if (dest_log_pinned) {
parent = old_dentry->d_parent;
- btrfs_log_new_name(trans, new_inode, new_dir, parent);
+ btrfs_log_new_name(trans, BTRFS_I(new_inode), BTRFS_I(new_dir),
+ parent);
btrfs_end_log_trans(dest);
dest_log_pinned = false;
}
@@ -9693,11 +9795,11 @@ out_fail:
* allow the tasks to sync it.
*/
if (ret && (root_log_pinned || dest_log_pinned)) {
- if (btrfs_inode_in_log(old_dir, fs_info->generation) ||
- btrfs_inode_in_log(new_dir, fs_info->generation) ||
- btrfs_inode_in_log(old_inode, fs_info->generation) ||
+ if (btrfs_inode_in_log(BTRFS_I(old_dir), fs_info->generation) ||
+ btrfs_inode_in_log(BTRFS_I(new_dir), fs_info->generation) ||
+ btrfs_inode_in_log(BTRFS_I(old_inode), fs_info->generation) ||
(new_inode &&
- btrfs_inode_in_log(new_inode, fs_info->generation)))
+ btrfs_inode_in_log(BTRFS_I(new_inode), fs_info->generation)))
btrfs_set_log_full_commit(fs_info, trans);
if (root_log_pinned) {
@@ -9736,7 +9838,7 @@ static int btrfs_whiteout_for_rename(struct btrfs_trans_handle *trans,
inode = btrfs_new_inode(trans, root, dir,
dentry->d_name.name,
dentry->d_name.len,
- btrfs_ino(dir),
+ btrfs_ino(BTRFS_I(dir)),
objectid,
S_IFCHR | WHITEOUT_MODE,
&index);
@@ -9755,8 +9857,8 @@ static int btrfs_whiteout_for_rename(struct btrfs_trans_handle *trans,
if (ret)
goto out;
- ret = btrfs_add_nondir(trans, dir, dentry,
- inode, 0, index);
+ ret = btrfs_add_nondir(trans, BTRFS_I(dir), dentry,
+ BTRFS_I(inode), 0, index);
if (ret)
goto out;
@@ -9784,10 +9886,10 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
u64 index = 0;
u64 root_objectid;
int ret;
- u64 old_ino = btrfs_ino(old_inode);
+ u64 old_ino = btrfs_ino(BTRFS_I(old_inode));
bool log_pinned = false;
- if (btrfs_ino(new_dir) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)
+ if (btrfs_ino(BTRFS_I(new_dir)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)
return -EPERM;
/* we only allow rename subvolume link between subvolumes */
@@ -9795,7 +9897,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
return -EXDEV;
if (old_ino == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID ||
- (new_inode && btrfs_ino(new_inode) == BTRFS_FIRST_FREE_OBJECTID))
+ (new_inode && btrfs_ino(BTRFS_I(new_inode)) == BTRFS_FIRST_FREE_OBJECTID))
return -ENOTEMPTY;
if (S_ISDIR(old_inode->i_mode) && new_inode &&
@@ -9855,7 +9957,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
if (dest != root)
btrfs_record_root_in_trans(trans, dest);
- ret = btrfs_set_inode_index(new_dir, &index);
+ ret = btrfs_set_inode_index(BTRFS_I(new_dir), &index);
if (ret)
goto out_fail;
@@ -9870,7 +9972,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
new_dentry->d_name.name,
new_dentry->d_name.len,
old_ino,
- btrfs_ino(new_dir), index);
+ btrfs_ino(BTRFS_I(new_dir)), index);
if (ret)
goto out_fail;
}
@@ -9883,7 +9985,8 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
old_inode->i_ctime = current_time(old_dir);
if (old_dentry->d_parent != new_dentry->d_parent)
- btrfs_record_unlink_dir(trans, old_dir, old_inode, 1);
+ btrfs_record_unlink_dir(trans, BTRFS_I(old_dir),
+ BTRFS_I(old_inode), 1);
if (unlikely(old_ino == BTRFS_FIRST_FREE_OBJECTID)) {
root_objectid = BTRFS_I(old_inode)->root->root_key.objectid;
@@ -9891,8 +9994,8 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
old_dentry->d_name.name,
old_dentry->d_name.len);
} else {
- ret = __btrfs_unlink_inode(trans, root, old_dir,
- d_inode(old_dentry),
+ ret = __btrfs_unlink_inode(trans, root, BTRFS_I(old_dir),
+ BTRFS_I(d_inode(old_dentry)),
old_dentry->d_name.name,
old_dentry->d_name.len);
if (!ret)
@@ -9906,7 +10009,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
if (new_inode) {
inode_inc_iversion(new_inode);
new_inode->i_ctime = current_time(new_inode);
- if (unlikely(btrfs_ino(new_inode) ==
+ if (unlikely(btrfs_ino(BTRFS_I(new_inode)) ==
BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
root_objectid = BTRFS_I(new_inode)->location.objectid;
ret = btrfs_unlink_subvol(trans, dest, new_dir,
@@ -9915,20 +10018,21 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
new_dentry->d_name.len);
BUG_ON(new_inode->i_nlink == 0);
} else {
- ret = btrfs_unlink_inode(trans, dest, new_dir,
- d_inode(new_dentry),
+ ret = btrfs_unlink_inode(trans, dest, BTRFS_I(new_dir),
+ BTRFS_I(d_inode(new_dentry)),
new_dentry->d_name.name,
new_dentry->d_name.len);
}
if (!ret && new_inode->i_nlink == 0)
- ret = btrfs_orphan_add(trans, d_inode(new_dentry));
+ ret = btrfs_orphan_add(trans,
+ BTRFS_I(d_inode(new_dentry)));
if (ret) {
btrfs_abort_transaction(trans, ret);
goto out_fail;
}
}
- ret = btrfs_add_link(trans, new_dir, old_inode,
+ ret = btrfs_add_link(trans, BTRFS_I(new_dir), BTRFS_I(old_inode),
new_dentry->d_name.name,
new_dentry->d_name.len, 0, index);
if (ret) {
@@ -9942,7 +10046,8 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
if (log_pinned) {
struct dentry *parent = new_dentry->d_parent;
- btrfs_log_new_name(trans, old_inode, old_dir, parent);
+ btrfs_log_new_name(trans, BTRFS_I(old_inode), BTRFS_I(old_dir),
+ parent);
btrfs_end_log_trans(root);
log_pinned = false;
}
@@ -9969,11 +10074,11 @@ out_fail:
* allow the tasks to sync it.
*/
if (ret && log_pinned) {
- if (btrfs_inode_in_log(old_dir, fs_info->generation) ||
- btrfs_inode_in_log(new_dir, fs_info->generation) ||
- btrfs_inode_in_log(old_inode, fs_info->generation) ||
+ if (btrfs_inode_in_log(BTRFS_I(old_dir), fs_info->generation) ||
+ btrfs_inode_in_log(BTRFS_I(new_dir), fs_info->generation) ||
+ btrfs_inode_in_log(BTRFS_I(old_inode), fs_info->generation) ||
(new_inode &&
- btrfs_inode_in_log(new_inode, fs_info->generation)))
+ btrfs_inode_in_log(BTRFS_I(new_inode), fs_info->generation)))
btrfs_set_log_full_commit(fs_info, trans);
btrfs_end_log_trans(root);
@@ -10237,8 +10342,8 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
goto out_unlock;
inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
- dentry->d_name.len, btrfs_ino(dir), objectid,
- S_IFLNK|S_IRWXUGO, &index);
+ dentry->d_name.len, btrfs_ino(BTRFS_I(dir)),
+ objectid, S_IFLNK|S_IRWXUGO, &index);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
goto out_unlock;
@@ -10264,7 +10369,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
err = -ENOMEM;
goto out_unlock_inode;
}
- key.objectid = btrfs_ino(inode);
+ key.objectid = btrfs_ino(BTRFS_I(inode));
key.offset = 0;
key.type = BTRFS_EXTENT_DATA_KEY;
datasize = btrfs_file_extent_calc_inline_size(name_len);
@@ -10294,7 +10399,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
inode_nohighmem(inode);
inode->i_mapping->a_ops = &btrfs_symlink_aops;
inode_set_bytes(inode, name_len);
- btrfs_i_size_write(inode, name_len);
+ btrfs_i_size_write(BTRFS_I(inode), name_len);
err = btrfs_update_inode(trans, root, inode);
/*
* Last step, add directory indexes for our symlink inode. This is the
@@ -10302,7 +10407,8 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
* elsewhere above.
*/
if (!err)
- err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
+ err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry,
+ BTRFS_I(inode), 0, index);
if (err) {
drop_inode = 1;
goto out_unlock_inode;
@@ -10388,7 +10494,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
break;
}
- btrfs_drop_extent_cache(inode, cur_offset,
+ btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset,
cur_offset + ins.offset -1, 0);
em = alloc_extent_map();
@@ -10415,7 +10521,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
write_unlock(&em_tree->lock);
if (ret != -EEXIST)
break;
- btrfs_drop_extent_cache(inode, cur_offset,
+ btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset,
cur_offset + ins.offset - 1,
0);
}
@@ -10452,7 +10558,7 @@ next:
btrfs_end_transaction(trans);
}
if (cur_offset < end)
- btrfs_free_reserved_data_space(inode, cur_offset,
+ btrfs_free_reserved_data_space(inode, NULL, cur_offset,
end - cur_offset + 1);
return ret;
}
@@ -10517,7 +10623,7 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
goto out;
inode = btrfs_new_inode(trans, root, dir, NULL, 0,
- btrfs_ino(dir), objectid, mode, &index);
+ btrfs_ino(BTRFS_I(dir)), objectid, mode, &index);
if (IS_ERR(inode)) {
ret = PTR_ERR(inode);
inode = NULL;
@@ -10537,7 +10643,7 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
ret = btrfs_update_inode(trans, root, inode);
if (ret)
goto out_inode;
- ret = btrfs_orphan_add(trans, inode);
+ ret = btrfs_orphan_add(trans, BTRFS_I(inode));
if (ret)
goto out_inode;
@@ -10567,6 +10673,48 @@ out_inode:
}
+__attribute__((const))
+static int btrfs_readpage_io_failed_hook(struct page *page, int failed_mirror)
+{
+ return -EAGAIN;
+}
+
+static struct btrfs_fs_info *iotree_fs_info(void *private_data)
+{
+ struct inode *inode = private_data;
+ return btrfs_sb(inode->i_sb);
+}
+
+static void btrfs_check_extent_io_range(void *private_data, const char *caller,
+ u64 start, u64 end)
+{
+ struct inode *inode = private_data;
+ u64 isize;
+
+ isize = i_size_read(inode);
+ if (end >= PAGE_SIZE && (end % 2) == 0 && end != isize - 1) {
+ btrfs_debug_rl(BTRFS_I(inode)->root->fs_info,
+ "%s: ino %llu isize %llu odd range [%llu,%llu]",
+ caller, btrfs_ino(BTRFS_I(inode)), isize, start, end);
+ }
+}
+
+void btrfs_set_range_writeback(void *private_data, u64 start, u64 end)
+{
+ struct inode *inode = private_data;
+ unsigned long index = start >> PAGE_SHIFT;
+ unsigned long end_index = end >> PAGE_SHIFT;
+ struct page *page;
+
+ while (index <= end_index) {
+ page = find_get_page(inode->i_mapping, index);
+ ASSERT(page); /* Pages should be in the extent_io_tree */
+ set_page_writeback(page);
+ put_page(page);
+ index++;
+ }
+}
+
static const struct inode_operations btrfs_dir_inode_operations = {
.getattr = btrfs_getattr,
.lookup = btrfs_lookup,
@@ -10605,16 +10753,23 @@ static const struct file_operations btrfs_dir_file_operations = {
};
static const struct extent_io_ops btrfs_extent_io_ops = {
- .fill_delalloc = run_delalloc_range,
+ /* mandatory callbacks */
.submit_bio_hook = btrfs_submit_bio_hook,
- .merge_bio_hook = btrfs_merge_bio_hook,
.readpage_end_io_hook = btrfs_readpage_end_io_hook,
+ .merge_bio_hook = btrfs_merge_bio_hook,
+ .readpage_io_failed_hook = btrfs_readpage_io_failed_hook,
+ .tree_fs_info = iotree_fs_info,
+ .set_range_writeback = btrfs_set_range_writeback,
+
+ /* optional callbacks */
+ .fill_delalloc = run_delalloc_range,
.writepage_end_io_hook = btrfs_writepage_end_io_hook,
.writepage_start_hook = btrfs_writepage_start_hook,
.set_bit_hook = btrfs_set_bit_hook,
.clear_bit_hook = btrfs_clear_bit_hook,
.merge_extent_hook = btrfs_merge_extent_hook,
.split_extent_hook = btrfs_split_extent_hook,
+ .check_extent_io_range = btrfs_check_extent_io_range,
};
/*
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 33f967d30b2a..fa1b78cf25f6 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -37,7 +37,7 @@
#include <linux/bit_spinlock.h>
#include <linux/security.h>
#include <linux/xattr.h>
-#include <linux/vmalloc.h>
+#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/blkdev.h>
#include <linux/uuid.h>
@@ -395,7 +395,7 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
q = bdev_get_queue(device->bdev);
if (blk_queue_discard(q)) {
num_devices++;
- minlen = min((u64)q->limits.discard_granularity,
+ minlen = min_t(u64, q->limits.discard_granularity,
minlen);
}
}
@@ -434,7 +434,7 @@ int btrfs_is_empty_uuid(u8 *uuid)
static noinline int create_subvol(struct inode *dir,
struct dentry *dentry,
- char *name, int namelen,
+ const char *name, int namelen,
u64 *async_transid,
struct btrfs_qgroup_inherit *inherit)
{
@@ -487,8 +487,7 @@ static noinline int create_subvol(struct inode *dir,
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
- btrfs_subvolume_release_metadata(fs_info, &block_rsv,
- qgroup_reserved);
+ btrfs_subvolume_release_metadata(fs_info, &block_rsv);
goto fail_free;
}
trans->block_rsv = &block_rsv;
@@ -581,27 +580,27 @@ static noinline int create_subvol(struct inode *dir,
/*
* insert the directory item
*/
- ret = btrfs_set_inode_index(dir, &index);
+ ret = btrfs_set_inode_index(BTRFS_I(dir), &index);
if (ret) {
btrfs_abort_transaction(trans, ret);
goto fail;
}
ret = btrfs_insert_dir_item(trans, root,
- name, namelen, dir, &key,
+ name, namelen, BTRFS_I(dir), &key,
BTRFS_FT_DIR, index);
if (ret) {
btrfs_abort_transaction(trans, ret);
goto fail;
}
- btrfs_i_size_write(dir, dir->i_size + namelen * 2);
+ btrfs_i_size_write(BTRFS_I(dir), dir->i_size + namelen * 2);
ret = btrfs_update_inode(trans, root, dir);
BUG_ON(ret);
ret = btrfs_add_root_ref(trans, fs_info,
objectid, root->root_key.objectid,
- btrfs_ino(dir), index, name, namelen);
+ btrfs_ino(BTRFS_I(dir)), index, name, namelen);
BUG_ON(ret);
ret = btrfs_uuid_tree_add(trans, fs_info, root_item->uuid,
@@ -613,7 +612,7 @@ fail:
kfree(root_item);
trans->block_rsv = NULL;
trans->bytes_reserved = 0;
- btrfs_subvolume_release_metadata(fs_info, &block_rsv, qgroup_reserved);
+ btrfs_subvolume_release_metadata(fs_info, &block_rsv);
if (async_transid) {
*async_transid = trans->transid;
@@ -657,7 +656,7 @@ static void btrfs_wait_for_no_snapshoting_writes(struct btrfs_root *root)
}
static int create_snapshot(struct btrfs_root *root, struct inode *dir,
- struct dentry *dentry, char *name, int namelen,
+ struct dentry *dentry,
u64 *async_transid, bool readonly,
struct btrfs_qgroup_inherit *inherit)
{
@@ -670,12 +669,12 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
return -EINVAL;
- pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
+ pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_KERNEL);
if (!pending_snapshot)
return -ENOMEM;
pending_snapshot->root_item = kzalloc(sizeof(struct btrfs_root_item),
- GFP_NOFS);
+ GFP_KERNEL);
pending_snapshot->path = btrfs_alloc_path();
if (!pending_snapshot->root_item || !pending_snapshot->path) {
ret = -ENOMEM;
@@ -690,7 +689,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
if (ret)
goto dec_and_free;
- btrfs_wait_ordered_extents(root, -1, 0, (u64)-1);
+ btrfs_wait_ordered_extents(root, U64_MAX, 0, (u64)-1);
btrfs_init_block_rsv(&pending_snapshot->block_rsv,
BTRFS_BLOCK_RSV_TEMP);
@@ -753,9 +752,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
d_instantiate(dentry, inode);
ret = 0;
fail:
- btrfs_subvolume_release_metadata(fs_info,
- &pending_snapshot->block_rsv,
- pending_snapshot->qgroup_reserved);
+ btrfs_subvolume_release_metadata(fs_info, &pending_snapshot->block_rsv);
dec_and_free:
if (atomic_dec_and_test(&root->will_be_snapshoted))
wake_up_atomic_t(&root->will_be_snapshoted);
@@ -835,7 +832,7 @@ static inline int btrfs_may_create(struct inode *dir, struct dentry *child)
* inside this filesystem so it's quite a bit simpler.
*/
static noinline int btrfs_mksubvol(const struct path *parent,
- char *name, int namelen,
+ const char *name, int namelen,
struct btrfs_root *snap_src,
u64 *async_transid, bool readonly,
struct btrfs_qgroup_inherit *inherit)
@@ -874,7 +871,7 @@ static noinline int btrfs_mksubvol(const struct path *parent,
goto out_up_read;
if (snap_src) {
- error = create_snapshot(snap_src, dir, dentry, name, namelen,
+ error = create_snapshot(snap_src, dir, dentry,
async_transid, readonly, inherit);
} else {
error = create_subvol(dir, dentry, name, namelen,
@@ -941,7 +938,7 @@ static int find_new_extents(struct btrfs_root *root,
struct btrfs_file_extent_item *extent;
int type;
int ret;
- u64 ino = btrfs_ino(inode);
+ u64 ino = btrfs_ino(BTRFS_I(inode));
path = btrfs_alloc_path();
if (!path)
@@ -1012,7 +1009,7 @@ static struct extent_map *defrag_lookup_extent(struct inode *inode, u64 start)
/* get the big lock and read metadata off disk */
lock_extent_bits(io_tree, start, end, &cached);
- em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len, 0);
unlock_extent_cached(io_tree, start, end, &cached, GFP_NOFS);
if (IS_ERR(em))
@@ -1130,6 +1127,7 @@ static int cluster_pages_for_defrag(struct inode *inode,
struct btrfs_ordered_extent *ordered;
struct extent_state *cached_state = NULL;
struct extent_io_tree *tree;
+ struct extent_changeset *data_reserved = NULL;
gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping);
file_end = (isize - 1) >> PAGE_SHIFT;
@@ -1138,7 +1136,7 @@ static int cluster_pages_for_defrag(struct inode *inode,
page_cnt = min_t(u64, (u64)num_pages, (u64)file_end - start_index + 1);
- ret = btrfs_delalloc_reserve_space(inode,
+ ret = btrfs_delalloc_reserve_space(inode, &data_reserved,
start_index << PAGE_SHIFT,
page_cnt << PAGE_SHIFT);
if (ret)
@@ -1229,7 +1227,7 @@ again:
spin_lock(&BTRFS_I(inode)->lock);
BTRFS_I(inode)->outstanding_extents++;
spin_unlock(&BTRFS_I(inode)->lock);
- btrfs_delalloc_release_space(inode,
+ btrfs_delalloc_release_space(inode, data_reserved,
start_index << PAGE_SHIFT,
(page_cnt - i_done) << PAGE_SHIFT);
}
@@ -1250,15 +1248,17 @@ again:
unlock_page(pages[i]);
put_page(pages[i]);
}
+ extent_changeset_free(data_reserved);
return i_done;
out:
for (i = 0; i < i_done; i++) {
unlock_page(pages[i]);
put_page(pages[i]);
}
- btrfs_delalloc_release_space(inode,
+ btrfs_delalloc_release_space(inode, data_reserved,
start_index << PAGE_SHIFT,
page_cnt << PAGE_SHIFT);
+ extent_changeset_free(data_reserved);
return ret;
}
@@ -1507,7 +1507,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
if (ret)
return ret;
- if (atomic_xchg(&fs_info->mutually_exclusive_operation_running, 1)) {
+ if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
mnt_drop_write_file(file);
return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
}
@@ -1622,13 +1622,13 @@ out_free:
kfree(vol_args);
out:
mutex_unlock(&fs_info->volume_mutex);
- atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
+ clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
mnt_drop_write_file(file);
return ret;
}
static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
- char *name, unsigned long fd, int subvol,
+ const char *name, unsigned long fd, int subvol,
u64 *transid, bool readonly,
struct btrfs_qgroup_inherit *inherit)
{
@@ -1780,7 +1780,7 @@ static noinline int btrfs_ioctl_subvol_getflags(struct file *file,
int ret = 0;
u64 flags = 0;
- if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID)
+ if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID)
return -EINVAL;
down_read(&fs_info->subvol_sem);
@@ -1812,7 +1812,7 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file,
if (ret)
goto out;
- if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
+ if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) {
ret = -EINVAL;
goto out_drop_write;
}
@@ -2446,7 +2446,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
if (err)
goto out_dput;
- if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
+ if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) {
err = -EINVAL;
goto out_dput;
}
@@ -2497,7 +2497,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
trans->block_rsv = &block_rsv;
trans->bytes_reserved = block_rsv.size;
- btrfs_record_snapshot_destroy(trans, dir);
+ btrfs_record_snapshot_destroy(trans, BTRFS_I(dir));
ret = btrfs_unlink_subvol(trans, root, dir,
dest->root_key.objectid,
@@ -2555,7 +2555,7 @@ out_end_trans:
err = ret;
inode->i_flags |= S_DEAD;
out_release:
- btrfs_subvolume_release_metadata(fs_info, &block_rsv, qgroup_reserved);
+ btrfs_subvolume_release_metadata(fs_info, &block_rsv);
out_up_write:
up_write(&fs_info->subvol_sem);
if (err) {
@@ -2613,9 +2613,6 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
goto out;
}
ret = btrfs_defrag_root(root);
- if (ret)
- goto out;
- ret = btrfs_defrag_root(root->fs_info->extent_root);
break;
case S_IFREG:
if (!(file->f_mode & FMODE_WRITE)) {
@@ -2667,7 +2664,7 @@ static long btrfs_ioctl_add_dev(struct btrfs_fs_info *fs_info, void __user *arg)
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- if (atomic_xchg(&fs_info->mutually_exclusive_operation_running, 1))
+ if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags))
return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
mutex_lock(&fs_info->volume_mutex);
@@ -2686,7 +2683,7 @@ static long btrfs_ioctl_add_dev(struct btrfs_fs_info *fs_info, void __user *arg)
kfree(vol_args);
out:
mutex_unlock(&fs_info->volume_mutex);
- atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
+ clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
return ret;
}
@@ -2714,7 +2711,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
if (vol_args->flags & ~BTRFS_VOL_ARG_V2_FLAGS_SUPPORTED)
return -EOPNOTSUPP;
- if (atomic_xchg(&fs_info->mutually_exclusive_operation_running, 1)) {
+ if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
goto out;
}
@@ -2727,7 +2724,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
ret = btrfs_rm_device(fs_info, vol_args->name, 0);
}
mutex_unlock(&fs_info->volume_mutex);
- atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
+ clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
if (!ret) {
if (vol_args->flags & BTRFS_DEVICE_SPEC_BY_ID)
@@ -2758,7 +2755,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
if (ret)
return ret;
- if (atomic_xchg(&fs_info->mutually_exclusive_operation_running, 1)) {
+ if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
goto out_drop_write;
}
@@ -2778,7 +2775,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
btrfs_info(fs_info, "disk deleted %s", vol_args->name);
kfree(vol_args);
out:
- atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
+ clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
out_drop_write:
mnt_drop_write_file(file);
@@ -3047,11 +3044,21 @@ static int btrfs_cmp_data_prepare(struct inode *src, u64 loff,
cmp->src_pages = src_pgarr;
cmp->dst_pages = dst_pgarr;
- ret = gather_extent_pages(src, cmp->src_pages, cmp->num_pages, loff);
+ /*
+ * If deduping ranges in the same inode, locking rules make it mandatory
+ * to always lock pages in ascending order to avoid deadlocks with
+ * concurrent tasks (such as starting writeback/delalloc).
+ */
+ if (src == dst && dst_loff < loff) {
+ swap(src_pgarr, dst_pgarr);
+ swap(loff, dst_loff);
+ }
+
+ ret = gather_extent_pages(src, src_pgarr, cmp->num_pages, loff);
if (ret)
goto out;
- ret = gather_extent_pages(dst, cmp->dst_pages, cmp->num_pages, dst_loff);
+ ret = gather_extent_pages(dst, dst_pgarr, cmp->num_pages, dst_loff);
out:
if (ret)
@@ -3059,8 +3066,7 @@ out:
return 0;
}
-static int btrfs_cmp_data(struct inode *src, u64 loff, struct inode *dst,
- u64 dst_loff, u64 len, struct cmp_pages *cmp)
+static int btrfs_cmp_data(u64 len, struct cmp_pages *cmp)
{
int ret = 0;
int i;
@@ -3128,26 +3134,27 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
int ret;
u64 len = olen;
struct cmp_pages cmp;
- int same_inode = 0;
+ bool same_inode = (src == dst);
u64 same_lock_start = 0;
u64 same_lock_len = 0;
- if (src == dst)
- same_inode = 1;
-
if (len == 0)
return 0;
- if (same_inode) {
+ if (same_inode)
inode_lock(src);
+ else
+ btrfs_double_inode_lock(src, dst);
- ret = extent_same_check_offsets(src, loff, &len, olen);
- if (ret)
- goto out_unlock;
- ret = extent_same_check_offsets(src, dst_loff, &len, olen);
- if (ret)
- goto out_unlock;
+ ret = extent_same_check_offsets(src, loff, &len, olen);
+ if (ret)
+ goto out_unlock;
+ ret = extent_same_check_offsets(dst, dst_loff, &len, olen);
+ if (ret)
+ goto out_unlock;
+
+ if (same_inode) {
/*
* Single inode case wants the same checks, except we
* don't want our length pushed out past i_size as
@@ -3175,16 +3182,6 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
same_lock_start = min_t(u64, loff, dst_loff);
same_lock_len = max_t(u64, loff, dst_loff) + len - same_lock_start;
- } else {
- btrfs_double_inode_lock(src, dst);
-
- ret = extent_same_check_offsets(src, loff, &len, olen);
- if (ret)
- goto out_unlock;
-
- ret = extent_same_check_offsets(dst, dst_loff, &len, olen);
- if (ret)
- goto out_unlock;
}
/* don't make the dst file partly checksummed */
@@ -3236,7 +3233,7 @@ again:
}
/* pass original length for comparison so we stay within i_size */
- ret = btrfs_cmp_data(src, loff, dst, dst_loff, olen, &cmp);
+ ret = btrfs_cmp_data(olen, &cmp);
if (ret == 0)
ret = btrfs_clone(src, dst, loff, olen, len, dst_loff, 1);
@@ -3304,7 +3301,7 @@ static int clone_finish_inode_update(struct btrfs_trans_handle *trans,
if (endoff > destoff + olen)
endoff = destoff + olen;
if (endoff > inode->i_size)
- btrfs_i_size_write(inode, endoff);
+ btrfs_i_size_write(BTRFS_I(inode), endoff);
ret = btrfs_update_inode(trans, root, inode);
if (ret) {
@@ -3317,20 +3314,19 @@ out:
return ret;
}
-static void clone_update_extent_map(struct inode *inode,
+static void clone_update_extent_map(struct btrfs_inode *inode,
const struct btrfs_trans_handle *trans,
const struct btrfs_path *path,
const u64 hole_offset,
const u64 hole_len)
{
- struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
+ struct extent_map_tree *em_tree = &inode->extent_tree;
struct extent_map *em;
int ret;
em = alloc_extent_map();
if (!em) {
- set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
- &BTRFS_I(inode)->runtime_flags);
+ set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags);
return;
}
@@ -3344,7 +3340,7 @@ static void clone_update_extent_map(struct inode *inode,
if (btrfs_file_extent_type(path->nodes[0], fi) ==
BTRFS_FILE_EXTENT_INLINE)
set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
- &BTRFS_I(inode)->runtime_flags);
+ &inode->runtime_flags);
} else {
em->start = hole_offset;
em->len = hole_len;
@@ -3370,8 +3366,7 @@ static void clone_update_extent_map(struct inode *inode,
}
if (ret)
- set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
- &BTRFS_I(inode)->runtime_flags);
+ set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags);
}
/*
@@ -3399,8 +3394,7 @@ static void clone_update_extent_map(struct inode *inode,
* data into the destination inode's inline extent if the later is greater then
* the former.
*/
-static int clone_copy_inline_extent(struct inode *src,
- struct inode *dst,
+static int clone_copy_inline_extent(struct inode *dst,
struct btrfs_trans_handle *trans,
struct btrfs_path *path,
struct btrfs_key *new_key,
@@ -3420,7 +3414,7 @@ static int clone_copy_inline_extent(struct inode *src,
if (new_key->offset > 0)
return -EOPNOTSUPP;
- key.objectid = btrfs_ino(dst);
+ key.objectid = btrfs_ino(BTRFS_I(dst));
key.type = BTRFS_EXTENT_DATA_KEY;
key.offset = 0;
ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
@@ -3435,7 +3429,7 @@ static int clone_copy_inline_extent(struct inode *src,
goto copy_inline_extent;
}
btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
- if (key.objectid == btrfs_ino(dst) &&
+ if (key.objectid == btrfs_ino(BTRFS_I(dst)) &&
key.type == BTRFS_EXTENT_DATA_KEY) {
ASSERT(key.offset > 0);
return -EOPNOTSUPP;
@@ -3469,7 +3463,7 @@ static int clone_copy_inline_extent(struct inode *src,
} else if (ret == 0) {
btrfs_item_key_to_cpu(path->nodes[0], &key,
path->slots[0]);
- if (key.objectid == btrfs_ino(dst) &&
+ if (key.objectid == btrfs_ino(BTRFS_I(dst)) &&
key.type == BTRFS_EXTENT_DATA_KEY)
return -EOPNOTSUPP;
}
@@ -3548,12 +3542,9 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
u64 last_dest_end = destoff;
ret = -ENOMEM;
- buf = kmalloc(fs_info->nodesize, GFP_KERNEL | __GFP_NOWARN);
- if (!buf) {
- buf = vmalloc(fs_info->nodesize);
- if (!buf)
- return ret;
- }
+ buf = kvmalloc(fs_info->nodesize, GFP_KERNEL);
+ if (!buf)
+ return ret;
path = btrfs_alloc_path();
if (!path) {
@@ -3563,7 +3554,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
path->reada = READA_FORWARD;
/* clone data */
- key.objectid = btrfs_ino(src);
+ key.objectid = btrfs_ino(BTRFS_I(src));
key.type = BTRFS_EXTENT_DATA_KEY;
key.offset = off;
@@ -3606,7 +3597,7 @@ process_slot:
btrfs_item_key_to_cpu(leaf, &key, slot);
if (key.type > BTRFS_EXTENT_DATA_KEY ||
- key.objectid != btrfs_ino(src))
+ key.objectid != btrfs_ino(BTRFS_I(src)))
break;
if (key.type == BTRFS_EXTENT_DATA_KEY) {
@@ -3659,7 +3650,7 @@ process_slot:
path->leave_spinning = 0;
memcpy(&new_key, &key, sizeof(new_key));
- new_key.objectid = btrfs_ino(inode);
+ new_key.objectid = btrfs_ino(BTRFS_I(inode));
if (off <= key.offset)
new_key.offset = key.offset + destoff - off;
else
@@ -3749,7 +3740,7 @@ process_slot:
fs_info,
disko, diskl, 0,
root->root_key.objectid,
- btrfs_ino(inode),
+ btrfs_ino(BTRFS_I(inode)),
new_key.offset - datao);
if (ret) {
btrfs_abort_transaction(trans,
@@ -3779,7 +3770,7 @@ process_slot:
size -= skip + trim;
datal -= skip + trim;
- ret = clone_copy_inline_extent(src, inode,
+ ret = clone_copy_inline_extent(inode,
trans, path,
&new_key,
drop_start,
@@ -3798,11 +3789,12 @@ process_slot:
/* If we have an implicit hole (NO_HOLES feature). */
if (drop_start < new_key.offset)
- clone_update_extent_map(inode, trans,
+ clone_update_extent_map(BTRFS_I(inode), trans,
NULL, drop_start,
new_key.offset - drop_start);
- clone_update_extent_map(inode, trans, path, 0, 0);
+ clone_update_extent_map(BTRFS_I(inode), trans,
+ path, 0, 0);
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(path);
@@ -3852,8 +3844,9 @@ process_slot:
btrfs_end_transaction(trans);
goto out;
}
- clone_update_extent_map(inode, trans, NULL, last_dest_end,
- destoff + len - last_dest_end);
+ clone_update_extent_map(BTRFS_I(inode), trans, NULL,
+ last_dest_end,
+ destoff + len - last_dest_end);
ret = clone_finish_inode_update(trans, inode, destoff + len,
destoff, olen, no_time_update);
}
@@ -4449,13 +4442,11 @@ static long btrfs_ioctl_dev_replace(struct btrfs_fs_info *fs_info,
ret = -EROFS;
goto out;
}
- if (atomic_xchg(
- &fs_info->mutually_exclusive_operation_running, 1)) {
+ if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
} else {
ret = btrfs_dev_replace_by_ioctl(fs_info, p);
- atomic_set(
- &fs_info->mutually_exclusive_operation_running, 0);
+ clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
}
break;
case BTRFS_IOCTL_DEV_REPLACE_CMD_STATUS:
@@ -4600,7 +4591,7 @@ static long btrfs_ioctl_logical_to_ino(struct btrfs_fs_info *fs_info,
out:
btrfs_free_path(path);
- vfree(inodes);
+ kvfree(inodes);
kfree(loi);
return ret;
@@ -4650,7 +4641,7 @@ static long btrfs_ioctl_balance(struct file *file, void __user *arg)
return ret;
again:
- if (!atomic_xchg(&fs_info->mutually_exclusive_operation_running, 1)) {
+ if (!test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
mutex_lock(&fs_info->volume_mutex);
mutex_lock(&fs_info->balance_mutex);
need_unlock = true;
@@ -4696,7 +4687,7 @@ again:
}
locked:
- BUG_ON(!atomic_read(&fs_info->mutually_exclusive_operation_running));
+ BUG_ON(!test_bit(BTRFS_FS_EXCL_OP, &fs_info->flags));
if (arg) {
bargs = memdup_user(arg, sizeof(*bargs));
@@ -4752,11 +4743,10 @@ locked:
do_balance:
/*
- * Ownership of bctl and mutually_exclusive_operation_running
+ * Ownership of bctl and filesystem flag BTRFS_FS_EXCL_OP
* goes to to btrfs_balance. bctl is freed in __cancel_balance,
* or, if restriper was paused all the way until unmount, in
- * free_fs_info. mutually_exclusive_operation_running is
- * cleared in __cancel_balance.
+ * free_fs_info. The flag is cleared in __cancel_balance.
*/
need_unlock = false;
@@ -4776,7 +4766,7 @@ out_unlock:
mutex_unlock(&fs_info->balance_mutex);
mutex_unlock(&fs_info->volume_mutex);
if (need_unlock)
- atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
+ clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
out:
mnt_drop_write_file(file);
return ret;
@@ -4910,7 +4900,6 @@ static long btrfs_ioctl_qgroup_assign(struct file *file, void __user *arg)
goto out;
}
- /* FIXME: check if the IDs really exist */
if (sa->assign) {
ret = btrfs_add_qgroup_relation(trans, fs_info,
sa->src, sa->dst);
@@ -4969,7 +4958,6 @@ static long btrfs_ioctl_qgroup_create(struct file *file, void __user *arg)
goto out;
}
- /* FIXME: check if the IDs really exist */
if (sa->create) {
ret = btrfs_create_qgroup(trans, fs_info, sa->qgroupid);
} else {
@@ -5023,7 +5011,6 @@ static long btrfs_ioctl_qgroup_limit(struct file *file, void __user *arg)
qgroupid = root->root_key.objectid;
}
- /* FIXME: check if the IDs really exist */
ret = btrfs_limit_qgroup(trans, fs_info, qgroupid, &sa->lim);
err = btrfs_end_transaction(trans);
@@ -5129,7 +5116,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file,
down_write(&fs_info->subvol_sem);
- if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
+ if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) {
ret = -EINVAL;
goto out;
}
@@ -5653,6 +5640,10 @@ long btrfs_ioctl(struct file *file, unsigned int
#ifdef CONFIG_COMPAT
long btrfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
+ /*
+ * These all access 32-bit values anyway so no further
+ * handling is necessary.
+ */
switch (cmd) {
case FS_IOC32_GETFLAGS:
cmd = FS_IOC_GETFLAGS;
@@ -5663,8 +5654,6 @@ long btrfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
case FS_IOC32_GETVERSION:
cmd = FS_IOC_GETVERSION;
break;
- default:
- return -ENOIOCTLCMD;
}
return btrfs_ioctl(file, cmd, (unsigned long) compat_ptr(arg));
diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c
index 45d26980caf9..d433e75d489a 100644
--- a/fs/btrfs/lzo.c
+++ b/fs/btrfs/lzo.c
@@ -18,13 +18,14 @@
#include <linux/kernel.h>
#include <linux/slab.h>
-#include <linux/vmalloc.h>
+#include <linux/mm.h>
#include <linux/init.h>
#include <linux/err.h>
#include <linux/sched.h>
#include <linux/pagemap.h>
#include <linux/bio.h>
#include <linux/lzo.h>
+#include <linux/refcount.h>
#include "compression.h"
#define LZO_LEN 4
@@ -40,9 +41,9 @@ static void lzo_free_workspace(struct list_head *ws)
{
struct workspace *workspace = list_entry(ws, struct workspace, list);
- vfree(workspace->buf);
- vfree(workspace->cbuf);
- vfree(workspace->mem);
+ kvfree(workspace->buf);
+ kvfree(workspace->cbuf);
+ kvfree(workspace->mem);
kfree(workspace);
}
@@ -50,13 +51,13 @@ static struct list_head *lzo_alloc_workspace(void)
{
struct workspace *workspace;
- workspace = kzalloc(sizeof(*workspace), GFP_NOFS);
+ workspace = kzalloc(sizeof(*workspace), GFP_KERNEL);
if (!workspace)
return ERR_PTR(-ENOMEM);
- workspace->mem = vmalloc(LZO1X_MEM_COMPRESS);
- workspace->buf = vmalloc(lzo1x_worst_compress(PAGE_SIZE));
- workspace->cbuf = vmalloc(lzo1x_worst_compress(PAGE_SIZE));
+ workspace->mem = kvmalloc(LZO1X_MEM_COMPRESS, GFP_KERNEL);
+ workspace->buf = kvmalloc(lzo1x_worst_compress(PAGE_SIZE), GFP_KERNEL);
+ workspace->cbuf = kvmalloc(lzo1x_worst_compress(PAGE_SIZE), GFP_KERNEL);
if (!workspace->mem || !workspace->buf || !workspace->cbuf)
goto fail;
@@ -76,7 +77,7 @@ static inline void write_compress_length(char *buf, size_t len)
memcpy(buf, &dlen, LZO_LEN);
}
-static inline size_t read_compress_length(char *buf)
+static inline size_t read_compress_length(const char *buf)
{
__le32 dlen;
@@ -86,13 +87,11 @@ static inline size_t read_compress_length(char *buf)
static int lzo_compress_pages(struct list_head *ws,
struct address_space *mapping,
- u64 start, unsigned long len,
+ u64 start,
struct page **pages,
- unsigned long nr_dest_pages,
unsigned long *out_pages,
unsigned long *total_in,
- unsigned long *total_out,
- unsigned long max_out)
+ unsigned long *total_out)
{
struct workspace *workspace = list_entry(ws, struct workspace, list);
int ret = 0;
@@ -102,7 +101,9 @@ static int lzo_compress_pages(struct list_head *ws,
struct page *in_page = NULL;
struct page *out_page = NULL;
unsigned long bytes_left;
-
+ unsigned long len = *total_out;
+ unsigned long nr_dest_pages = *out_pages;
+ const unsigned long max_out = nr_dest_pages * PAGE_SIZE;
size_t in_len;
size_t out_len;
char *buf;
@@ -141,7 +142,7 @@ static int lzo_compress_pages(struct list_head *ws,
ret = lzo1x_1_compress(data_in, in_len, workspace->cbuf,
&out_len, workspace->mem);
if (ret != LZO_E_OK) {
- pr_debug("BTRFS: deflate in loop returned %d\n",
+ pr_debug("BTRFS: lzo in loop returned %d\n",
ret);
ret = -EIO;
goto out;
@@ -229,8 +230,10 @@ static int lzo_compress_pages(struct list_head *ws,
in_len = min(bytes_left, PAGE_SIZE);
}
- if (tot_out > tot_in)
+ if (tot_out >= tot_in) {
+ ret = -E2BIG;
goto out;
+ }
/* store the size of all chunks of compressed data */
cpage_out = kmap(pages[0]);
@@ -254,16 +257,13 @@ out:
return ret;
}
-static int lzo_decompress_bio(struct list_head *ws,
- struct page **pages_in,
- u64 disk_start,
- struct bio *orig_bio,
- size_t srclen)
+static int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
{
struct workspace *workspace = list_entry(ws, struct workspace, list);
int ret = 0, ret2;
char *data_in;
unsigned long page_in_index = 0;
+ size_t srclen = cb->compressed_len;
unsigned long total_pages_in = DIV_ROUND_UP(srclen, PAGE_SIZE);
unsigned long buf_start;
unsigned long buf_offset = 0;
@@ -278,6 +278,9 @@ static int lzo_decompress_bio(struct list_head *ws,
unsigned long tot_len;
char *buf;
bool may_late_unmap, need_unmap;
+ struct page **pages_in = cb->compressed_pages;
+ u64 disk_start = cb->start;
+ struct bio *orig_bio = cb->orig_bio;
data_in = kmap(pages_in[0]);
tot_len = read_compress_length(data_in);
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 041c3326d109..a3aca495e33e 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -212,7 +212,7 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
set_bit(BTRFS_ORDERED_DIRECT, &entry->flags);
/* one ref for the tree */
- atomic_set(&entry->refs, 1);
+ refcount_set(&entry->refs, 1);
init_waitqueue_head(&entry->wait);
INIT_LIST_HEAD(&entry->list);
INIT_LIST_HEAD(&entry->root_extent_list);
@@ -358,7 +358,7 @@ int btrfs_dec_test_first_ordered_pending(struct inode *inode,
out:
if (!ret && cached && entry) {
*cached = entry;
- atomic_inc(&entry->refs);
+ refcount_inc(&entry->refs);
}
spin_unlock_irqrestore(&tree->lock, flags);
return ret == 0;
@@ -425,14 +425,14 @@ have_entry:
out:
if (!ret && cached && entry) {
*cached = entry;
- atomic_inc(&entry->refs);
+ refcount_inc(&entry->refs);
}
spin_unlock_irqrestore(&tree->lock, flags);
return ret == 0;
}
/* Needs to either be called under a log transaction or the log_mutex */
-void btrfs_get_logged_extents(struct inode *inode,
+void btrfs_get_logged_extents(struct btrfs_inode *inode,
struct list_head *logged_list,
const loff_t start,
const loff_t end)
@@ -442,7 +442,7 @@ void btrfs_get_logged_extents(struct inode *inode,
struct rb_node *n;
struct rb_node *prev;
- tree = &BTRFS_I(inode)->ordered_tree;
+ tree = &inode->ordered_tree;
spin_lock_irq(&tree->lock);
n = __tree_search(&tree->tree, end, &prev);
if (!n)
@@ -456,7 +456,7 @@ void btrfs_get_logged_extents(struct inode *inode,
if (test_and_set_bit(BTRFS_ORDERED_LOGGED, &ordered->flags))
continue;
list_add(&ordered->log_list, logged_list);
- atomic_inc(&ordered->refs);
+ refcount_inc(&ordered->refs);
}
spin_unlock_irq(&tree->lock);
}
@@ -565,7 +565,7 @@ void btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry)
trace_btrfs_ordered_extent_put(entry->inode, entry);
- if (atomic_dec_and_test(&entry->refs)) {
+ if (refcount_dec_and_test(&entry->refs)) {
ASSERT(list_empty(&entry->log_list));
ASSERT(list_empty(&entry->trans_list));
ASSERT(list_empty(&entry->root_extent_list));
@@ -623,7 +623,7 @@ void btrfs_remove_ordered_extent(struct inode *inode,
spin_lock(&fs_info->trans_lock);
trans = fs_info->running_transaction;
if (trans)
- atomic_inc(&trans->use_count);
+ refcount_inc(&trans->use_count);
spin_unlock(&fs_info->trans_lock);
ASSERT(trans);
@@ -663,7 +663,7 @@ static void btrfs_run_ordered_extent_work(struct btrfs_work *work)
* wait for all the ordered extents in a root. This is done when balancing
* space between drives.
*/
-int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
+u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr,
const u64 range_start, const u64 range_len)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -671,7 +671,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
LIST_HEAD(skipped);
LIST_HEAD(works);
struct btrfs_ordered_extent *ordered, *next;
- int count = 0;
+ u64 count = 0;
const u64 range_end = range_start + range_len;
mutex_lock(&root->ordered_extent_mutex);
@@ -690,7 +690,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
list_move_tail(&ordered->root_extent_list,
&root->ordered_extents);
- atomic_inc(&ordered->refs);
+ refcount_inc(&ordered->refs);
spin_unlock(&root->ordered_extent_lock);
btrfs_init_work(&ordered->flush_work,
@@ -701,7 +701,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
cond_resched();
spin_lock(&root->ordered_extent_lock);
- if (nr != -1)
+ if (nr != U64_MAX)
nr--;
count++;
}
@@ -720,13 +720,13 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
return count;
}
-int btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
- const u64 range_start, const u64 range_len)
+u64 btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
+ const u64 range_start, const u64 range_len)
{
struct btrfs_root *root;
struct list_head splice;
- int done;
- int total_done = 0;
+ u64 total_done = 0;
+ u64 done;
INIT_LIST_HEAD(&splice);
@@ -748,9 +748,8 @@ int btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
total_done += done;
spin_lock(&fs_info->ordered_root_lock);
- if (nr != -1) {
+ if (nr != U64_MAX) {
nr -= done;
- WARN_ON(nr < 0);
}
}
list_splice_tail(&splice, &fs_info->ordered_roots);
@@ -870,7 +869,7 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct inode *inode,
if (!offset_in_entry(entry, file_offset))
entry = NULL;
if (entry)
- atomic_inc(&entry->refs);
+ refcount_inc(&entry->refs);
out:
spin_unlock_irq(&tree->lock);
return entry;
@@ -879,15 +878,14 @@ out:
/* Since the DIO code tries to lock a wide area we need to look for any ordered
* extents that exist in the range, rather than just the start of the range.
*/
-struct btrfs_ordered_extent *btrfs_lookup_ordered_range(struct inode *inode,
- u64 file_offset,
- u64 len)
+struct btrfs_ordered_extent *btrfs_lookup_ordered_range(
+ struct btrfs_inode *inode, u64 file_offset, u64 len)
{
struct btrfs_ordered_inode_tree *tree;
struct rb_node *node;
struct btrfs_ordered_extent *entry = NULL;
- tree = &BTRFS_I(inode)->ordered_tree;
+ tree = &inode->ordered_tree;
spin_lock_irq(&tree->lock);
node = tree_search(tree, file_offset);
if (!node) {
@@ -912,7 +910,7 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_range(struct inode *inode,
}
out:
if (entry)
- atomic_inc(&entry->refs);
+ refcount_inc(&entry->refs);
spin_unlock_irq(&tree->lock);
return entry;
}
@@ -923,7 +921,7 @@ bool btrfs_have_ordered_extents_in_range(struct inode *inode,
{
struct btrfs_ordered_extent *oe;
- oe = btrfs_lookup_ordered_range(inode, file_offset, len);
+ oe = btrfs_lookup_ordered_range(BTRFS_I(inode), file_offset, len);
if (oe) {
btrfs_put_ordered_extent(oe);
return true;
@@ -949,7 +947,7 @@ btrfs_lookup_first_ordered_extent(struct inode *inode, u64 file_offset)
goto out;
entry = rb_entry(node, struct btrfs_ordered_extent, rb_node);
- atomic_inc(&entry->refs);
+ refcount_inc(&entry->refs);
out:
spin_unlock_irq(&tree->lock);
return entry;
@@ -984,8 +982,18 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
}
disk_i_size = BTRFS_I(inode)->disk_i_size;
- /* truncate file */
- if (disk_i_size > i_size) {
+ /*
+ * truncate file.
+ * If ordered is not NULL, then this is called from endio and
+ * disk_i_size will be updated by either truncate itself or any
+ * in-flight IOs which are inside the disk_i_size.
+ *
+ * Because btrfs_setsize() may set i_size with disk_i_size if truncate
+ * fails somehow, we need to make sure we have a precise disk_i_size by
+ * updating it as usual.
+ *
+ */
+ if (!ordered && disk_i_size > i_size) {
BTRFS_I(inode)->disk_i_size = orig_offset;
ret = 0;
goto out;
@@ -1032,25 +1040,22 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
/* We treat this entry as if it doesn't exist */
if (test_bit(BTRFS_ORDERED_UPDATED_ISIZE, &test->flags))
continue;
- if (test->file_offset + test->len <= disk_i_size)
+
+ if (entry_end(test) <= disk_i_size)
break;
if (test->file_offset >= i_size)
break;
- if (entry_end(test) > disk_i_size) {
- /*
- * we don't update disk_i_size now, so record this
- * undealt i_size. Or we will not know the real
- * i_size.
- */
- if (test->outstanding_isize < offset)
- test->outstanding_isize = offset;
- if (ordered &&
- ordered->outstanding_isize >
- test->outstanding_isize)
- test->outstanding_isize =
- ordered->outstanding_isize;
- goto out;
- }
+
+ /*
+ * We don't update disk_i_size now, so record this undealt
+ * i_size. Or we will not know the real i_size.
+ */
+ if (test->outstanding_isize < offset)
+ test->outstanding_isize = offset;
+ if (ordered &&
+ ordered->outstanding_isize > test->outstanding_isize)
+ test->outstanding_isize = ordered->outstanding_isize;
+ goto out;
}
new_i_size = min_t(u64, offset, i_size);
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index 5f2b0ca28705..56c4c0ee6381 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -75,6 +75,8 @@ struct btrfs_ordered_sum {
* in the logging code. */
#define BTRFS_ORDERED_PENDING 11 /* We are waiting for this ordered extent to
* complete in the current transaction. */
+#define BTRFS_ORDERED_REGULAR 12 /* Regular IO for COW */
+
struct btrfs_ordered_extent {
/* logical offset in the file */
u64 file_offset;
@@ -111,7 +113,7 @@ struct btrfs_ordered_extent {
int compress_type;
/* reference count */
- atomic_t refs;
+ refcount_t refs;
/* the inode we belong to */
struct inode *inode;
@@ -187,9 +189,10 @@ void btrfs_start_ordered_extent(struct inode *inode,
int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len);
struct btrfs_ordered_extent *
btrfs_lookup_first_ordered_extent(struct inode * inode, u64 file_offset);
-struct btrfs_ordered_extent *btrfs_lookup_ordered_range(struct inode *inode,
- u64 file_offset,
- u64 len);
+struct btrfs_ordered_extent *btrfs_lookup_ordered_range(
+ struct btrfs_inode *inode,
+ u64 file_offset,
+ u64 len);
bool btrfs_have_ordered_extents_in_range(struct inode *inode,
u64 file_offset,
u64 len);
@@ -197,11 +200,11 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
struct btrfs_ordered_extent *ordered);
int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
u32 *sum, int len);
-int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
+u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr,
const u64 range_start, const u64 range_len);
-int btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
+u64 btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
const u64 range_start, const u64 range_len);
-void btrfs_get_logged_extents(struct inode *inode,
+void btrfs_get_logged_extents(struct btrfs_inode *inode,
struct list_head *logged_list,
const loff_t start,
const loff_t end);
diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c
index cdafbf92ef0c..fcae61e175f3 100644
--- a/fs/btrfs/print-tree.c
+++ b/fs/btrfs/print-tree.c
@@ -261,8 +261,11 @@ void btrfs_print_leaf(struct btrfs_fs_info *fs_info, struct extent_buffer *l)
case BTRFS_BLOCK_GROUP_ITEM_KEY:
bi = btrfs_item_ptr(l, i,
struct btrfs_block_group_item);
- pr_info("\t\tblock group used %llu\n",
- btrfs_disk_block_group_used(l, bi));
+ pr_info(
+ "\t\tblock group used %llu chunk_objectid %llu flags %llu\n",
+ btrfs_disk_block_group_used(l, bi),
+ btrfs_disk_block_group_chunk_objectid(l, bi),
+ btrfs_disk_block_group_flags(l, bi));
break;
case BTRFS_CHUNK_ITEM_KEY:
print_chunk(l, btrfs_item_ptr(l, i,
diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c
index f2621e330954..4b23ae5d0e5c 100644
--- a/fs/btrfs/props.c
+++ b/fs/btrfs/props.c
@@ -164,6 +164,7 @@ static int iterate_object_props(struct btrfs_root *root,
size_t),
void *ctx)
{
+ struct btrfs_fs_info *fs_info = root->fs_info;
int ret;
char *name_buf = NULL;
char *value_buf = NULL;
@@ -214,6 +215,12 @@ static int iterate_object_props(struct btrfs_root *root,
name_ptr = (unsigned long)(di + 1);
data_ptr = name_ptr + name_len;
+ if (verify_dir_item(fs_info, leaf,
+ path->slots[0], di)) {
+ ret = -EIO;
+ goto out;
+ }
+
if (name_len <= XATTR_BTRFS_PREFIX_LEN ||
memcmp_extent_buffer(leaf, XATTR_BTRFS_PREFIX,
name_ptr,
@@ -279,7 +286,7 @@ static void inode_prop_iterator(void *ctx,
if (unlikely(ret))
btrfs_warn(root->fs_info,
"error applying prop %s to ino %llu (root %llu): %d",
- handler->xattr_name, btrfs_ino(inode),
+ handler->xattr_name, btrfs_ino(BTRFS_I(inode)),
root->root_key.objectid, ret);
else
set_bit(BTRFS_INODE_HAS_PROPS, &BTRFS_I(inode)->runtime_flags);
@@ -288,7 +295,7 @@ static void inode_prop_iterator(void *ctx,
int btrfs_load_inode_props(struct inode *inode, struct btrfs_path *path)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
- u64 ino = btrfs_ino(inode);
+ u64 ino = btrfs_ino(BTRFS_I(inode));
int ret;
ret = iterate_object_props(root, path, ino, inode_prop_iterator, inode);
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 662821f1252c..4ce351efe281 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -47,50 +47,6 @@
* - check all ioctl parameters
*/
-/*
- * one struct for each qgroup, organized in fs_info->qgroup_tree.
- */
-struct btrfs_qgroup {
- u64 qgroupid;
-
- /*
- * state
- */
- u64 rfer; /* referenced */
- u64 rfer_cmpr; /* referenced compressed */
- u64 excl; /* exclusive */
- u64 excl_cmpr; /* exclusive compressed */
-
- /*
- * limits
- */
- u64 lim_flags; /* which limits are set */
- u64 max_rfer;
- u64 max_excl;
- u64 rsv_rfer;
- u64 rsv_excl;
-
- /*
- * reservation tracking
- */
- u64 reserved;
-
- /*
- * lists
- */
- struct list_head groups; /* groups this group is member of */
- struct list_head members; /* groups that are members of this group */
- struct list_head dirty; /* dirty groups */
- struct rb_node node; /* tree of qgroups */
-
- /*
- * temp variables for accounting operations
- * Refer to qgroup_shared_accounting() for details.
- */
- u64 old_refcnt;
- u64 new_refcnt;
-};
-
static void btrfs_qgroup_update_old_refcnt(struct btrfs_qgroup *qg, u64 seq,
int mod)
{
@@ -319,7 +275,7 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info)
if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
return 0;
- fs_info->qgroup_ulist = ulist_alloc(GFP_NOFS);
+ fs_info->qgroup_ulist = ulist_alloc(GFP_KERNEL);
if (!fs_info->qgroup_ulist) {
ret = -ENOMEM;
goto out;
@@ -876,7 +832,7 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
goto out;
}
- fs_info->qgroup_ulist = ulist_alloc(GFP_NOFS);
+ fs_info->qgroup_ulist = ulist_alloc(GFP_KERNEL);
if (!fs_info->qgroup_ulist) {
ret = -ENOMEM;
goto out;
@@ -1019,7 +975,7 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
list_del(&quota_root->dirty_list);
btrfs_tree_lock(quota_root->node);
- clean_tree_block(trans, fs_info, quota_root->node);
+ clean_tree_block(fs_info, quota_root->node);
btrfs_tree_unlock(quota_root->node);
btrfs_free_tree_block(trans, quota_root, quota_root->node, 0, 1);
@@ -1038,6 +994,18 @@ static void qgroup_dirty(struct btrfs_fs_info *fs_info,
list_add(&qgroup->dirty, &fs_info->dirty_qgroups);
}
+static void report_reserved_underflow(struct btrfs_fs_info *fs_info,
+ struct btrfs_qgroup *qgroup,
+ u64 num_bytes)
+{
+#ifdef CONFIG_BTRFS_DEBUG
+ WARN_ON(qgroup->reserved < num_bytes);
+ btrfs_debug(fs_info,
+ "qgroup %llu reserved space underflow, have: %llu, to free: %llu",
+ qgroup->qgroupid, qgroup->reserved, num_bytes);
+#endif
+ qgroup->reserved = 0;
+}
/*
* The easy accounting, if we are adding/removing the only ref for an extent
* then this qgroup and all of the parent qgroups get their reference and
@@ -1065,8 +1033,13 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info,
WARN_ON(sign < 0 && qgroup->excl < num_bytes);
qgroup->excl += sign * num_bytes;
qgroup->excl_cmpr += sign * num_bytes;
- if (sign > 0)
- qgroup->reserved -= num_bytes;
+ if (sign > 0) {
+ trace_qgroup_update_reserve(fs_info, qgroup, -(s64)num_bytes);
+ if (qgroup->reserved < num_bytes)
+ report_reserved_underflow(fs_info, qgroup, num_bytes);
+ else
+ qgroup->reserved -= num_bytes;
+ }
qgroup_dirty(fs_info, qgroup);
@@ -1086,8 +1059,15 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info,
qgroup->rfer_cmpr += sign * num_bytes;
WARN_ON(sign < 0 && qgroup->excl < num_bytes);
qgroup->excl += sign * num_bytes;
- if (sign > 0)
- qgroup->reserved -= num_bytes;
+ if (sign > 0) {
+ trace_qgroup_update_reserve(fs_info, qgroup,
+ -(s64)num_bytes);
+ if (qgroup->reserved < num_bytes)
+ report_reserved_underflow(fs_info, qgroup,
+ num_bytes);
+ else
+ qgroup->reserved -= num_bytes;
+ }
qgroup->excl_cmpr += sign * num_bytes;
qgroup_dirty(fs_info, qgroup);
@@ -1156,7 +1136,7 @@ int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
if (btrfs_qgroup_level(src) >= btrfs_qgroup_level(dst))
return -EINVAL;
- tmp = ulist_alloc(GFP_NOFS);
+ tmp = ulist_alloc(GFP_KERNEL);
if (!tmp)
return -ENOMEM;
@@ -1205,7 +1185,7 @@ out:
return ret;
}
-int __del_qgroup_relation(struct btrfs_trans_handle *trans,
+static int __del_qgroup_relation(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info, u64 src, u64 dst)
{
struct btrfs_root *quota_root;
@@ -1216,7 +1196,7 @@ int __del_qgroup_relation(struct btrfs_trans_handle *trans,
int ret = 0;
int err;
- tmp = ulist_alloc(GFP_NOFS);
+ tmp = ulist_alloc(GFP_KERNEL);
if (!tmp)
return -ENOMEM;
@@ -1426,37 +1406,6 @@ out:
return ret;
}
-int btrfs_qgroup_prepare_account_extents(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info)
-{
- struct btrfs_qgroup_extent_record *record;
- struct btrfs_delayed_ref_root *delayed_refs;
- struct rb_node *node;
- u64 qgroup_to_skip;
- int ret = 0;
-
- delayed_refs = &trans->transaction->delayed_refs;
- qgroup_to_skip = delayed_refs->qgroup_to_skip;
-
- /*
- * No need to do lock, since this function will only be called in
- * btrfs_commit_transaction().
- */
- node = rb_first(&delayed_refs->dirty_extent_root);
- while (node) {
- record = rb_entry(node, struct btrfs_qgroup_extent_record,
- node);
- ret = btrfs_find_all_roots(NULL, fs_info, record->bytenr, 0,
- &record->old_roots);
- if (ret < 0)
- break;
- if (qgroup_to_skip)
- ulist_del(record->old_roots, qgroup_to_skip, 0);
- node = rb_next(node);
- }
- return ret;
-}
-
int btrfs_qgroup_trace_extent_nolock(struct btrfs_fs_info *fs_info,
struct btrfs_delayed_ref_root *delayed_refs,
struct btrfs_qgroup_extent_record *record)
@@ -1486,6 +1435,28 @@ int btrfs_qgroup_trace_extent_nolock(struct btrfs_fs_info *fs_info,
return 0;
}
+int btrfs_qgroup_trace_extent_post(struct btrfs_fs_info *fs_info,
+ struct btrfs_qgroup_extent_record *qrecord)
+{
+ struct ulist *old_root;
+ u64 bytenr = qrecord->bytenr;
+ int ret;
+
+ ret = btrfs_find_all_roots(NULL, fs_info, bytenr, 0, &old_root);
+ if (ret < 0)
+ return ret;
+
+ /*
+ * Here we don't need to get the lock of
+ * trans->transaction->delayed_refs, since inserted qrecord won't
+ * be deleted, only qrecord->node may be modified (new qrecord insert)
+ *
+ * So modifying qrecord->old_roots is safe here
+ */
+ qrecord->old_roots = old_root;
+ return 0;
+}
+
int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes,
gfp_t gfp_flag)
@@ -1511,9 +1482,11 @@ int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans,
spin_lock(&delayed_refs->lock);
ret = btrfs_qgroup_trace_extent_nolock(fs_info, delayed_refs, record);
spin_unlock(&delayed_refs->lock);
- if (ret > 0)
+ if (ret > 0) {
kfree(record);
- return 0;
+ return 0;
+ }
+ return btrfs_qgroup_trace_extent_post(fs_info, record);
}
int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
@@ -1554,6 +1527,7 @@ int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
if (ret)
return ret;
}
+ cond_resched();
return 0;
}
@@ -1571,8 +1545,7 @@ int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
* If we increment the root nodes slot counter past the number of
* elements, 1 is returned to signal completion of the search.
*/
-static int adjust_slots_upwards(struct btrfs_root *root,
- struct btrfs_path *path, int root_level)
+static int adjust_slots_upwards(struct btrfs_path *path, int root_level)
{
int level = 0;
int nr, slot;
@@ -1713,7 +1686,7 @@ walk_down:
goto out;
/* Nonzero return here means we completed our search */
- ret = adjust_slots_upwards(root, path, root_level);
+ ret = adjust_slots_upwards(path, root_level);
if (ret)
break;
@@ -1914,6 +1887,35 @@ static int qgroup_update_counters(struct btrfs_fs_info *fs_info,
return 0;
}
+/*
+ * Check if the @roots potentially is a list of fs tree roots
+ *
+ * Return 0 for definitely not a fs/subvol tree roots ulist
+ * Return 1 for possible fs/subvol tree roots in the list (considering an empty
+ * one as well)
+ */
+static int maybe_fs_roots(struct ulist *roots)
+{
+ struct ulist_node *unode;
+ struct ulist_iterator uiter;
+
+ /* Empty one, still possible for fs roots */
+ if (!roots || roots->nnodes == 0)
+ return 1;
+
+ ULIST_ITER_INIT(&uiter);
+ unode = ulist_next(roots, &uiter);
+ if (!unode)
+ return 1;
+
+ /*
+ * If it contains fs tree roots, then it must belong to fs/subvol
+ * trees.
+ * If it contains a non-fs tree, it won't be shared with fs/subvol trees.
+ */
+ return is_fstree(unode->val);
+}
+
int
btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
@@ -1927,13 +1929,24 @@ btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans,
u64 nr_old_roots = 0;
int ret = 0;
- if (new_roots)
+ if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
+ return 0;
+
+ if (new_roots) {
+ if (!maybe_fs_roots(new_roots))
+ goto out_free;
nr_new_roots = new_roots->nnodes;
- if (old_roots)
+ }
+ if (old_roots) {
+ if (!maybe_fs_roots(old_roots))
+ goto out_free;
nr_old_roots = old_roots->nnodes;
+ }
- if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
+ /* Quick exit, either not fs tree roots, or won't affect any qgroup */
+ if (nr_old_roots == 0 && nr_new_roots == 0)
goto out_free;
+
BUG_ON(!fs_info->quota_root);
trace_btrfs_qgroup_account_extent(fs_info, bytenr, num_bytes,
@@ -2012,16 +2025,32 @@ int btrfs_qgroup_account_extents(struct btrfs_trans_handle *trans,
if (!ret) {
/*
- * Use (u64)-1 as time_seq to do special search, which
+ * Old roots should be searched when inserting qgroup
+ * extent record
+ */
+ if (WARN_ON(!record->old_roots)) {
+ /* Search commit root to find old_roots */
+ ret = btrfs_find_all_roots(NULL, fs_info,
+ record->bytenr, 0,
+ &record->old_roots);
+ if (ret < 0)
+ goto cleanup;
+ }
+
+ /*
+ * Use SEQ_LAST as time_seq to do special search, which
* doesn't lock tree or delayed_refs and search current
* root. It's safe inside commit_transaction().
*/
ret = btrfs_find_all_roots(trans, fs_info,
- record->bytenr, (u64)-1, &new_roots);
+ record->bytenr, SEQ_LAST, &new_roots);
if (ret < 0)
goto cleanup;
- if (qgroup_to_skip)
+ if (qgroup_to_skip) {
ulist_del(new_roots, qgroup_to_skip, 0);
+ ulist_del(record->old_roots, qgroup_to_skip,
+ 0);
+ }
ret = btrfs_qgroup_account_extent(trans, fs_info,
record->bytenr, record->num_bytes,
record->old_roots, new_roots);
@@ -2170,9 +2199,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
goto out;
}
- rcu_read_lock();
level_size = fs_info->nodesize;
- rcu_read_unlock();
}
/*
@@ -2306,13 +2333,27 @@ out:
return ret;
}
-static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
+static bool qgroup_check_limits(const struct btrfs_qgroup *qg, u64 num_bytes)
+{
+ if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_RFER) &&
+ qg->reserved + (s64)qg->rfer + num_bytes > qg->max_rfer)
+ return false;
+
+ if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_EXCL) &&
+ qg->reserved + (s64)qg->excl + num_bytes > qg->max_excl)
+ return false;
+
+ return true;
+}
+
+static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes, bool enforce)
{
struct btrfs_root *quota_root;
struct btrfs_qgroup *qgroup;
struct btrfs_fs_info *fs_info = root->fs_info;
u64 ref_root = root->root_key.objectid;
int ret = 0;
+ int retried = 0;
struct ulist_node *unode;
struct ulist_iterator uiter;
@@ -2322,6 +2363,11 @@ static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
if (num_bytes == 0)
return 0;
+ if (test_bit(BTRFS_FS_QUOTA_OVERRIDE, &fs_info->flags) &&
+ capable(CAP_SYS_RESOURCE))
+ enforce = false;
+
+retry:
spin_lock(&fs_info->qgroup_lock);
quota_root = fs_info->quota_root;
if (!quota_root)
@@ -2347,16 +2393,28 @@ static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
qg = unode_aux_to_qgroup(unode);
- if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_RFER) &&
- qg->reserved + (s64)qg->rfer + num_bytes >
- qg->max_rfer) {
- ret = -EDQUOT;
- goto out;
- }
+ if (enforce && !qgroup_check_limits(qg, num_bytes)) {
+ /*
+ * Commit the tree and retry, since we may have
+ * deletions which would free up space.
+ */
+ if (!retried && qg->reserved > 0) {
+ struct btrfs_trans_handle *trans;
- if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_EXCL) &&
- qg->reserved + (s64)qg->excl + num_bytes >
- qg->max_excl) {
+ spin_unlock(&fs_info->qgroup_lock);
+ ret = btrfs_start_delalloc_inodes(root, 0);
+ if (ret)
+ return ret;
+ btrfs_wait_ordered_extents(root, U64_MAX, 0, (u64)-1);
+ trans = btrfs_join_transaction(root);
+ if (IS_ERR(trans))
+ return PTR_ERR(trans);
+ ret = btrfs_commit_transaction(trans);
+ if (ret)
+ return ret;
+ retried++;
+ goto retry;
+ }
ret = -EDQUOT;
goto out;
}
@@ -2379,6 +2437,7 @@ static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
qg = unode_aux_to_qgroup(unode);
+ trace_qgroup_update_reserve(fs_info, qg, num_bytes);
qg->reserved += num_bytes;
}
@@ -2424,7 +2483,11 @@ void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
qg = unode_aux_to_qgroup(unode);
- qg->reserved -= num_bytes;
+ trace_qgroup_update_reserve(fs_info, qg, -(s64)num_bytes);
+ if (qg->reserved < num_bytes)
+ report_reserved_underflow(fs_info, qg, num_bytes);
+ else
+ qg->reserved -= num_bytes;
list_for_each_entry(glist, &qg->groups, next_group) {
ret = ulist_add(fs_info->qgroup_ulist,
@@ -2439,23 +2502,6 @@ out:
spin_unlock(&fs_info->qgroup_lock);
}
-static inline void qgroup_free(struct btrfs_root *root, u64 num_bytes)
-{
- return btrfs_qgroup_free_refroot(root->fs_info, root->objectid,
- num_bytes);
-}
-void assert_qgroups_uptodate(struct btrfs_trans_handle *trans)
-{
- if (list_empty(&trans->qgroup_ref_list) && !trans->delayed_ref_elem.seq)
- return;
- btrfs_err(trans->fs_info,
- "qgroups not uptodate in trans handle %p: list is%s empty, seq is %#x.%x",
- trans, list_empty(&trans->qgroup_ref_list) ? "" : " not",
- (u32)(trans->delayed_ref_elem.seq >> 32),
- (u32)trans->delayed_ref_elem.seq);
- BUG();
-}
-
/*
* returns < 0 on error, 0 when more leafs are to be scanned.
* returns 1 when done.
@@ -2789,71 +2835,146 @@ btrfs_qgroup_rescan_resume(struct btrfs_fs_info *fs_info)
* Return <0 for error (including -EQUOT)
*
* NOTE: this function may sleep for memory allocation.
+ * if btrfs_qgroup_reserve_data() is called multiple times with
+ * same @reserved, caller must ensure when error happens it's OK
+ * to free *ALL* reserved space.
*/
-int btrfs_qgroup_reserve_data(struct inode *inode, u64 start, u64 len)
+int btrfs_qgroup_reserve_data(struct inode *inode,
+ struct extent_changeset **reserved_ret, u64 start,
+ u64 len)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
- struct extent_changeset changeset;
struct ulist_node *unode;
struct ulist_iterator uiter;
+ struct extent_changeset *reserved;
+ u64 orig_reserved;
+ u64 to_reserve;
int ret;
if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags) ||
!is_fstree(root->objectid) || len == 0)
return 0;
- changeset.bytes_changed = 0;
- changeset.range_changed = ulist_alloc(GFP_NOFS);
+ /* @reserved parameter is mandatory for qgroup */
+ if (WARN_ON(!reserved_ret))
+ return -EINVAL;
+ if (!*reserved_ret) {
+ *reserved_ret = extent_changeset_alloc();
+ if (!*reserved_ret)
+ return -ENOMEM;
+ }
+ reserved = *reserved_ret;
+ /* Record already reserved space */
+ orig_reserved = reserved->bytes_changed;
ret = set_record_extent_bits(&BTRFS_I(inode)->io_tree, start,
- start + len -1, EXTENT_QGROUP_RESERVED, &changeset);
+ start + len -1, EXTENT_QGROUP_RESERVED, reserved);
+
+ /* Newly reserved space */
+ to_reserve = reserved->bytes_changed - orig_reserved;
trace_btrfs_qgroup_reserve_data(inode, start, len,
- changeset.bytes_changed,
- QGROUP_RESERVE);
+ to_reserve, QGROUP_RESERVE);
if (ret < 0)
goto cleanup;
- ret = qgroup_reserve(root, changeset.bytes_changed);
+ ret = qgroup_reserve(root, to_reserve, true);
if (ret < 0)
goto cleanup;
- ulist_free(changeset.range_changed);
return ret;
cleanup:
- /* cleanup already reserved ranges */
+ /* cleanup *ALL* already reserved ranges */
ULIST_ITER_INIT(&uiter);
- while ((unode = ulist_next(changeset.range_changed, &uiter)))
+ while ((unode = ulist_next(&reserved->range_changed, &uiter)))
clear_extent_bit(&BTRFS_I(inode)->io_tree, unode->val,
unode->aux, EXTENT_QGROUP_RESERVED, 0, 0, NULL,
GFP_NOFS);
- ulist_free(changeset.range_changed);
+ extent_changeset_release(reserved);
return ret;
}
-static int __btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len,
- int free)
+/* Free ranges specified by @reserved, normally in error path */
+static int qgroup_free_reserved_data(struct inode *inode,
+ struct extent_changeset *reserved, u64 start, u64 len)
{
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct ulist_node *unode;
+ struct ulist_iterator uiter;
struct extent_changeset changeset;
- int trace_op = QGROUP_RELEASE;
+ int freed = 0;
int ret;
- changeset.bytes_changed = 0;
- changeset.range_changed = ulist_alloc(GFP_NOFS);
- if (!changeset.range_changed)
- return -ENOMEM;
+ extent_changeset_init(&changeset);
+ len = round_up(start + len, root->fs_info->sectorsize);
+ start = round_down(start, root->fs_info->sectorsize);
+ ULIST_ITER_INIT(&uiter);
+ while ((unode = ulist_next(&reserved->range_changed, &uiter))) {
+ u64 range_start = unode->val;
+ /* unode->aux is the inclusive end */
+ u64 range_len = unode->aux - range_start + 1;
+ u64 free_start;
+ u64 free_len;
+
+ extent_changeset_release(&changeset);
+
+ /* Only free range in range [start, start + len) */
+ if (range_start >= start + len ||
+ range_start + range_len <= start)
+ continue;
+ free_start = max(range_start, start);
+ free_len = min(start + len, range_start + range_len) -
+ free_start;
+ /*
+ * TODO: To also modify reserved->ranges_reserved to reflect
+ * the modification.
+ *
+ * However as long as we free qgroup reserved according to
+ * EXTENT_QGROUP_RESERVED, we won't double free.
+ * So not need to rush.
+ */
+ ret = clear_record_extent_bits(&BTRFS_I(inode)->io_failure_tree,
+ free_start, free_start + free_len - 1,
+ EXTENT_QGROUP_RESERVED, &changeset);
+ if (ret < 0)
+ goto out;
+ freed += changeset.bytes_changed;
+ }
+ btrfs_qgroup_free_refroot(root->fs_info, root->objectid, freed);
+ ret = freed;
+out:
+ extent_changeset_release(&changeset);
+ return ret;
+}
+
+static int __btrfs_qgroup_release_data(struct inode *inode,
+ struct extent_changeset *reserved, u64 start, u64 len,
+ int free)
+{
+ struct extent_changeset changeset;
+ int trace_op = QGROUP_RELEASE;
+ int ret;
+
+ /* In release case, we shouldn't have @reserved */
+ WARN_ON(!free && reserved);
+ if (free && reserved)
+ return qgroup_free_reserved_data(inode, reserved, start, len);
+ extent_changeset_init(&changeset);
ret = clear_record_extent_bits(&BTRFS_I(inode)->io_tree, start,
start + len -1, EXTENT_QGROUP_RESERVED, &changeset);
if (ret < 0)
goto out;
- if (free) {
- qgroup_free(BTRFS_I(inode)->root, changeset.bytes_changed);
+ if (free)
trace_op = QGROUP_FREE;
- }
trace_btrfs_qgroup_release_data(inode, start, len,
changeset.bytes_changed, trace_op);
+ if (free)
+ btrfs_qgroup_free_refroot(BTRFS_I(inode)->root->fs_info,
+ BTRFS_I(inode)->root->objectid,
+ changeset.bytes_changed);
+ ret = changeset.bytes_changed;
out:
- ulist_free(changeset.range_changed);
+ extent_changeset_release(&changeset);
return ret;
}
@@ -2862,14 +2983,17 @@ out:
*
* Should be called when a range of pages get invalidated before reaching disk.
* Or for error cleanup case.
+ * if @reserved is given, only reserved range in [@start, @start + @len) will
+ * be freed.
*
* For data written to disk, use btrfs_qgroup_release_data().
*
* NOTE: This function may sleep for memory allocation.
*/
-int btrfs_qgroup_free_data(struct inode *inode, u64 start, u64 len)
+int btrfs_qgroup_free_data(struct inode *inode,
+ struct extent_changeset *reserved, u64 start, u64 len)
{
- return __btrfs_qgroup_release_data(inode, start, len, 1);
+ return __btrfs_qgroup_release_data(inode, reserved, start, len, 1);
}
/*
@@ -2889,10 +3013,11 @@ int btrfs_qgroup_free_data(struct inode *inode, u64 start, u64 len)
*/
int btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len)
{
- return __btrfs_qgroup_release_data(inode, start, len, 0);
+ return __btrfs_qgroup_release_data(inode, NULL, start, len, 0);
}
-int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes)
+int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes,
+ bool enforce)
{
struct btrfs_fs_info *fs_info = root->fs_info;
int ret;
@@ -2902,26 +3027,28 @@ int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes)
return 0;
BUG_ON(num_bytes != round_down(num_bytes, fs_info->nodesize));
- ret = qgroup_reserve(root, num_bytes);
+ trace_qgroup_meta_reserve(root, (s64)num_bytes);
+ ret = qgroup_reserve(root, num_bytes, enforce);
if (ret < 0)
return ret;
- atomic_add(num_bytes, &root->qgroup_meta_rsv);
+ atomic64_add(num_bytes, &root->qgroup_meta_rsv);
return ret;
}
void btrfs_qgroup_free_meta_all(struct btrfs_root *root)
{
struct btrfs_fs_info *fs_info = root->fs_info;
- int reserved;
+ u64 reserved;
if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) ||
!is_fstree(root->objectid))
return;
- reserved = atomic_xchg(&root->qgroup_meta_rsv, 0);
+ reserved = atomic64_xchg(&root->qgroup_meta_rsv, 0);
if (reserved == 0)
return;
- qgroup_free(root, reserved);
+ trace_qgroup_meta_reserve(root, -(s64)reserved);
+ btrfs_qgroup_free_refroot(fs_info, root->objectid, reserved);
}
void btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes)
@@ -2933,9 +3060,10 @@ void btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes)
return;
BUG_ON(num_bytes != round_down(num_bytes, fs_info->nodesize));
- WARN_ON(atomic_read(&root->qgroup_meta_rsv) < num_bytes);
- atomic_sub(num_bytes, &root->qgroup_meta_rsv);
- qgroup_free(root, num_bytes);
+ WARN_ON(atomic64_read(&root->qgroup_meta_rsv) < num_bytes);
+ atomic64_sub(num_bytes, &root->qgroup_meta_rsv);
+ trace_qgroup_meta_reserve(root, -(s64)num_bytes);
+ btrfs_qgroup_free_refroot(fs_info, root->objectid, num_bytes);
}
/*
@@ -2949,23 +3077,22 @@ void btrfs_qgroup_check_reserved_leak(struct inode *inode)
struct ulist_iterator iter;
int ret;
- changeset.bytes_changed = 0;
- changeset.range_changed = ulist_alloc(GFP_NOFS);
- if (WARN_ON(!changeset.range_changed))
- return;
-
+ extent_changeset_init(&changeset);
ret = clear_record_extent_bits(&BTRFS_I(inode)->io_tree, 0, (u64)-1,
EXTENT_QGROUP_RESERVED, &changeset);
WARN_ON(ret < 0);
if (WARN_ON(changeset.bytes_changed)) {
ULIST_ITER_INIT(&iter);
- while ((unode = ulist_next(changeset.range_changed, &iter))) {
+ while ((unode = ulist_next(&changeset.range_changed, &iter))) {
btrfs_warn(BTRFS_I(inode)->root->fs_info,
"leaking qgroup reserved space, ino: %lu, start: %llu, end: %llu",
inode->i_ino, unode->val, unode->aux);
}
- qgroup_free(BTRFS_I(inode)->root, changeset.bytes_changed);
+ btrfs_qgroup_free_refroot(BTRFS_I(inode)->root->fs_info,
+ BTRFS_I(inode)->root->objectid,
+ changeset.bytes_changed);
+
}
- ulist_free(changeset.range_changed);
+ extent_changeset_release(&changeset);
}
diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h
index 416ae8e1d23c..d9984e87cddf 100644
--- a/fs/btrfs/qgroup.h
+++ b/fs/btrfs/qgroup.h
@@ -62,6 +62,50 @@ struct btrfs_qgroup_extent_record {
};
/*
+ * one struct for each qgroup, organized in fs_info->qgroup_tree.
+ */
+struct btrfs_qgroup {
+ u64 qgroupid;
+
+ /*
+ * state
+ */
+ u64 rfer; /* referenced */
+ u64 rfer_cmpr; /* referenced compressed */
+ u64 excl; /* exclusive */
+ u64 excl_cmpr; /* exclusive compressed */
+
+ /*
+ * limits
+ */
+ u64 lim_flags; /* which limits are set */
+ u64 max_rfer;
+ u64 max_excl;
+ u64 rsv_rfer;
+ u64 rsv_excl;
+
+ /*
+ * reservation tracking
+ */
+ u64 reserved;
+
+ /*
+ * lists
+ */
+ struct list_head groups; /* groups this group is member of */
+ struct list_head members; /* groups that are members of this group */
+ struct list_head dirty; /* dirty groups */
+ struct rb_node node; /* tree of qgroups */
+
+ /*
+ * temp variables for accounting operations
+ * Refer to qgroup_shared_accounting() for details.
+ */
+ u64 old_refcnt;
+ u64 new_refcnt;
+};
+
+/*
* For qgroup event trace points only
*/
#define QGROUP_RESERVE (1<<0)
@@ -90,13 +134,13 @@ int btrfs_limit_qgroup(struct btrfs_trans_handle *trans,
int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info);
void btrfs_free_qgroup_config(struct btrfs_fs_info *fs_info);
struct btrfs_delayed_extent_op;
-int btrfs_qgroup_prepare_account_extents(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info);
+
/*
* Inform qgroup to trace one dirty extent, its info is recorded in @record.
- * So qgroup can account it at commit trans time.
+ * So qgroup can account it at transaction committing time.
*
- * No lock version, caller must acquire delayed ref lock and allocate memory.
+ * No lock version, caller must acquire delayed ref lock and allocated memory,
+ * then call btrfs_qgroup_trace_extent_post() after exiting lock context.
*
* Return 0 for success insert
* Return >0 for existing record, caller can free @record safely.
@@ -108,11 +152,37 @@ int btrfs_qgroup_trace_extent_nolock(
struct btrfs_qgroup_extent_record *record);
/*
+ * Post handler after qgroup_trace_extent_nolock().
+ *
+ * NOTE: Current qgroup does the expensive backref walk at transaction
+ * committing time with TRANS_STATE_COMMIT_DOING, this blocks incoming
+ * new transaction.
+ * This is designed to allow btrfs_find_all_roots() to get correct new_roots
+ * result.
+ *
+ * However for old_roots there is no need to do backref walk at that time,
+ * since we search commit roots to walk backref and result will always be
+ * correct.
+ *
+ * Due to the nature of no lock version, we can't do backref there.
+ * So we must call btrfs_qgroup_trace_extent_post() after exiting
+ * spinlock context.
+ *
+ * TODO: If we can fix and prove btrfs_find_all_roots() can get correct result
+ * using current root, then we can move all expensive backref walk out of
+ * transaction committing, but not now as qgroup accounting will be wrong again.
+ */
+int btrfs_qgroup_trace_extent_post(struct btrfs_fs_info *fs_info,
+ struct btrfs_qgroup_extent_record *qrecord);
+
+/*
* Inform qgroup to trace one dirty extent, specified by @bytenr and
* @num_bytes.
* So qgroup can account it at commit trans time.
*
- * Better encapsulated version.
+ * Better encapsulated version, with memory allocation and backref walk for
+ * commit roots.
+ * So this can sleep.
*
* Return 0 if the operation is done.
* Return <0 for error, like memory allocation failure or invalid parameter
@@ -159,17 +229,12 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
struct btrfs_qgroup_inherit *inherit);
void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
u64 ref_root, u64 num_bytes);
-/*
- * TODO: Add proper trace point for it, as btrfs_qgroup_free() is
- * called by everywhere, can't provide good trace for delayed ref case.
- */
static inline void btrfs_qgroup_free_delayed_ref(struct btrfs_fs_info *fs_info,
u64 ref_root, u64 num_bytes)
{
- btrfs_qgroup_free_refroot(fs_info, ref_root, num_bytes);
trace_btrfs_qgroup_free_delayed_ref(fs_info, ref_root, num_bytes);
+ btrfs_qgroup_free_refroot(fs_info, ref_root, num_bytes);
}
-void assert_qgroups_uptodate(struct btrfs_trans_handle *trans);
#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
int btrfs_verify_qgroup_counts(struct btrfs_fs_info *fs_info, u64 qgroupid,
@@ -177,11 +242,14 @@ int btrfs_verify_qgroup_counts(struct btrfs_fs_info *fs_info, u64 qgroupid,
#endif
/* New io_tree based accurate qgroup reserve API */
-int btrfs_qgroup_reserve_data(struct inode *inode, u64 start, u64 len);
+int btrfs_qgroup_reserve_data(struct inode *inode,
+ struct extent_changeset **reserved, u64 start, u64 len);
int btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len);
-int btrfs_qgroup_free_data(struct inode *inode, u64 start, u64 len);
+int btrfs_qgroup_free_data(struct inode *inode,
+ struct extent_changeset *reserved, u64 start, u64 len);
-int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes);
+int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes,
+ bool enforce);
void btrfs_qgroup_free_meta_all(struct btrfs_root *root);
void btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes);
void btrfs_qgroup_check_reserved_leak(struct inode *inode);
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index d2a9a1ee5361..208638384cd2 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -31,7 +31,7 @@
#include <linux/hash.h>
#include <linux/list_sort.h>
#include <linux/raid/xor.h>
-#include <linux/vmalloc.h>
+#include <linux/mm.h>
#include <asm/div64.h>
#include "ctree.h"
#include "extent_map.h"
@@ -149,7 +149,7 @@ struct btrfs_raid_bio {
int generic_bio_cnt;
- atomic_t refs;
+ refcount_t refs;
atomic_t stripes_pending;
@@ -218,12 +218,9 @@ int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info)
* of a failing mount.
*/
table_size = sizeof(*table) + sizeof(*h) * num_entries;
- table = kzalloc(table_size, GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT);
- if (!table) {
- table = vzalloc(table_size);
- if (!table)
- return -ENOMEM;
- }
+ table = kvzalloc(table_size, GFP_KERNEL);
+ if (!table)
+ return -ENOMEM;
spin_lock_init(&table->cache_lock);
INIT_LIST_HEAD(&table->stripe_cache);
@@ -389,7 +386,7 @@ static void __remove_rbio_from_cache(struct btrfs_raid_bio *rbio)
if (bio_list_empty(&rbio->bio_list)) {
if (!list_empty(&rbio->hash_list)) {
list_del_init(&rbio->hash_list);
- atomic_dec(&rbio->refs);
+ refcount_dec(&rbio->refs);
BUG_ON(!list_empty(&rbio->plug_list));
}
}
@@ -480,7 +477,7 @@ static void cache_rbio(struct btrfs_raid_bio *rbio)
/* bump our ref if we were not in the list before */
if (!test_and_set_bit(RBIO_CACHE_BIT, &rbio->flags))
- atomic_inc(&rbio->refs);
+ refcount_inc(&rbio->refs);
if (!list_empty(&rbio->stripe_cache)){
list_move(&rbio->stripe_cache, &table->stripe_cache);
@@ -677,11 +674,9 @@ static noinline int lock_stripe_add(struct btrfs_raid_bio *rbio)
struct btrfs_raid_bio *freeit = NULL;
struct btrfs_raid_bio *cache_drop = NULL;
int ret = 0;
- int walk = 0;
spin_lock_irqsave(&h->lock, flags);
list_for_each_entry(cur, &h->hash_list, hash_list) {
- walk++;
if (cur->bbio->raid_map[0] == rbio->bbio->raid_map[0]) {
spin_lock(&cur->bio_list_lock);
@@ -691,7 +686,7 @@ static noinline int lock_stripe_add(struct btrfs_raid_bio *rbio)
test_bit(RBIO_CACHE_BIT, &cur->flags) &&
!test_bit(RBIO_RMW_LOCKED_BIT, &cur->flags)) {
list_del_init(&cur->hash_list);
- atomic_dec(&cur->refs);
+ refcount_dec(&cur->refs);
steal_rbio(cur, rbio);
cache_drop = cur;
@@ -740,7 +735,7 @@ static noinline int lock_stripe_add(struct btrfs_raid_bio *rbio)
}
}
lockit:
- atomic_inc(&rbio->refs);
+ refcount_inc(&rbio->refs);
list_add(&rbio->hash_list, &h->hash_list);
out:
spin_unlock_irqrestore(&h->lock, flags);
@@ -786,7 +781,7 @@ static noinline void unlock_stripe(struct btrfs_raid_bio *rbio)
}
list_del_init(&rbio->hash_list);
- atomic_dec(&rbio->refs);
+ refcount_dec(&rbio->refs);
/*
* we use the plug list to hold all the rbios
@@ -803,7 +798,7 @@ static noinline void unlock_stripe(struct btrfs_raid_bio *rbio)
list_del_init(&rbio->plug_list);
list_add(&next->hash_list, &h->hash_list);
- atomic_inc(&next->refs);
+ refcount_inc(&next->refs);
spin_unlock(&rbio->bio_list_lock);
spin_unlock_irqrestore(&h->lock, flags);
@@ -845,8 +840,7 @@ static void __free_raid_bio(struct btrfs_raid_bio *rbio)
{
int i;
- WARN_ON(atomic_read(&rbio->refs) < 0);
- if (!atomic_dec_and_test(&rbio->refs))
+ if (!refcount_dec_and_test(&rbio->refs))
return;
WARN_ON(!list_empty(&rbio->stripe_cache));
@@ -874,7 +868,7 @@ static void free_raid_bio(struct btrfs_raid_bio *rbio)
* this frees the rbio and runs through all the bios in the
* bio_list and calls end_io on them
*/
-static void rbio_orig_end_io(struct btrfs_raid_bio *rbio, int err)
+static void rbio_orig_end_io(struct btrfs_raid_bio *rbio, blk_status_t err)
{
struct bio *cur = bio_list_get(&rbio->bio_list);
struct bio *next;
@@ -887,7 +881,7 @@ static void rbio_orig_end_io(struct btrfs_raid_bio *rbio, int err)
while (cur) {
next = cur->bi_next;
cur->bi_next = NULL;
- cur->bi_error = err;
+ cur->bi_status = err;
bio_endio(cur);
cur = next;
}
@@ -900,7 +894,7 @@ static void rbio_orig_end_io(struct btrfs_raid_bio *rbio, int err)
static void raid_write_end_io(struct bio *bio)
{
struct btrfs_raid_bio *rbio = bio->bi_private;
- int err = bio->bi_error;
+ blk_status_t err = bio->bi_status;
int max_errors;
if (err)
@@ -917,7 +911,7 @@ static void raid_write_end_io(struct bio *bio)
max_errors = (rbio->operation == BTRFS_RBIO_PARITY_SCRUB) ?
0 : rbio->bbio->max_errors;
if (atomic_read(&rbio->error) > max_errors)
- err = -EIO;
+ err = BLK_STS_IOERR;
rbio_orig_end_io(rbio, err);
}
@@ -999,7 +993,7 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info,
rbio->stripe_npages = stripe_npages;
rbio->faila = -1;
rbio->failb = -1;
- atomic_set(&rbio->refs, 1);
+ refcount_set(&rbio->refs, 1);
atomic_set(&rbio->error, 0);
atomic_set(&rbio->stripes_pending, 0);
@@ -1095,7 +1089,7 @@ static int rbio_add_io_page(struct btrfs_raid_bio *rbio,
* devices or if they are not contiguous
*/
if (last_end == disk_start && stripe->dev->bdev &&
- !last->bi_error &&
+ !last->bi_status &&
last->bi_bdev == stripe->dev->bdev) {
ret = bio_add_page(last, page, PAGE_SIZE, 0);
if (ret == PAGE_SIZE)
@@ -1104,10 +1098,7 @@ static int rbio_add_io_page(struct btrfs_raid_bio *rbio,
}
/* put a new bio on the list */
- bio = btrfs_io_bio_alloc(GFP_NOFS, bio_max_len >> PAGE_SHIFT?:1);
- if (!bio)
- return -ENOMEM;
-
+ bio = btrfs_io_bio_alloc(bio_max_len >> PAGE_SHIFT ?: 1);
bio->bi_iter.bi_size = 0;
bio->bi_bdev = stripe->dev->bdev;
bio->bi_iter.bi_sector = disk_start >> 9;
@@ -1145,20 +1136,27 @@ static void validate_rbio_for_rmw(struct btrfs_raid_bio *rbio)
static void index_rbio_pages(struct btrfs_raid_bio *rbio)
{
struct bio *bio;
- struct bio_vec *bvec;
u64 start;
unsigned long stripe_offset;
unsigned long page_index;
- int i;
spin_lock_irq(&rbio->bio_list_lock);
bio_list_for_each(bio, &rbio->bio_list) {
+ struct bio_vec bvec;
+ struct bvec_iter iter;
+ int i = 0;
+
start = (u64)bio->bi_iter.bi_sector << 9;
stripe_offset = start - rbio->bbio->raid_map[0];
page_index = stripe_offset >> PAGE_SHIFT;
- bio_for_each_segment_all(bvec, bio, i)
- rbio->bio_pages[page_index + i] = bvec->bv_page;
+ if (bio_flagged(bio, BIO_CLONED))
+ bio->bi_iter = btrfs_io_bio(bio)->iter;
+
+ bio_for_each_segment(bvec, bio, iter) {
+ rbio->bio_pages[page_index + i] = bvec.bv_page;
+ i++;
+ }
}
spin_unlock_irq(&rbio->bio_list_lock);
}
@@ -1432,11 +1430,14 @@ static int fail_bio_stripe(struct btrfs_raid_bio *rbio,
*/
static void set_bio_pages_uptodate(struct bio *bio)
{
- struct bio_vec *bvec;
- int i;
+ struct bio_vec bvec;
+ struct bvec_iter iter;
+
+ if (bio_flagged(bio, BIO_CLONED))
+ bio->bi_iter = btrfs_io_bio(bio)->iter;
- bio_for_each_segment_all(bvec, bio, i)
- SetPageUptodate(bvec->bv_page);
+ bio_for_each_segment(bvec, bio, iter)
+ SetPageUptodate(bvec.bv_page);
}
/*
@@ -1451,7 +1452,7 @@ static void raid_rmw_end_io(struct bio *bio)
{
struct btrfs_raid_bio *rbio = bio->bi_private;
- if (bio->bi_error)
+ if (bio->bi_status)
fail_bio_stripe(rbio, bio);
else
set_bio_pages_uptodate(bio);
@@ -1994,7 +1995,7 @@ static void raid_recover_end_io(struct bio *bio)
* we only read stripe pages off the disk, set them
* up to date if there were no errors
*/
- if (bio->bi_error)
+ if (bio->bi_status)
fail_bio_stripe(rbio, bio);
else
set_bio_pages_uptodate(bio);
@@ -2120,6 +2121,11 @@ int raid56_parity_recover(struct btrfs_fs_info *fs_info, struct bio *bio,
struct btrfs_raid_bio *rbio;
int ret;
+ if (generic_io) {
+ ASSERT(bbio->mirror_num == mirror_num);
+ btrfs_io_bio(bio)->mirror_num = mirror_num;
+ }
+
rbio = alloc_rbio(fs_info, bbio, stripe_len);
if (IS_ERR(rbio)) {
if (generic_io)
@@ -2196,6 +2202,8 @@ static void read_rebuild_work(struct btrfs_work *work)
/*
* The following code is used to scrub/replace the parity stripe
*
+ * Caller must have already increased bio_counter for getting @bbio.
+ *
* Note: We need make sure all the pages that add into the scrub/replace
* raid bio are correct and not be changed during the scrub/replace. That
* is those pages just hold metadata or file data with checksum.
@@ -2233,6 +2241,12 @@ raid56_parity_alloc_scrub_rbio(struct btrfs_fs_info *fs_info, struct bio *bio,
ASSERT(rbio->stripe_npages == stripe_nsectors);
bitmap_copy(rbio->dbitmap, dbitmap, stripe_nsectors);
+ /*
+ * We have already increased bio_counter when getting bbio, record it
+ * so we can free it at rbio_orig_end_io().
+ */
+ rbio->generic_bio_cnt = 1;
+
return rbio;
}
@@ -2520,7 +2534,7 @@ static void raid56_parity_scrub_end_io(struct bio *bio)
{
struct btrfs_raid_bio *rbio = bio->bi_private;
- if (bio->bi_error)
+ if (bio->bi_status)
fail_bio_stripe(rbio, bio);
else
set_bio_pages_uptodate(bio);
@@ -2675,6 +2689,12 @@ raid56_alloc_missing_rbio(struct btrfs_fs_info *fs_info, struct bio *bio,
return NULL;
}
+ /*
+ * When we get bbio, we have already increased bio_counter, record it
+ * so we can free it at rbio_orig_end_io()
+ */
+ rbio->generic_bio_cnt = 1;
+
return rbio;
}
diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c
index e88bca87f5d2..ab852b8e3e37 100644
--- a/fs/btrfs/reada.c
+++ b/fs/btrfs/reada.c
@@ -66,7 +66,6 @@ struct reada_extctl {
struct reada_extent {
u64 logical;
struct btrfs_key top;
- int err;
struct list_head extctl;
int refcnt;
spinlock_t lock;
@@ -209,9 +208,9 @@ cleanup:
return;
}
-int btree_readahead_hook(struct btrfs_fs_info *fs_info,
- struct extent_buffer *eb, int err)
+int btree_readahead_hook(struct extent_buffer *eb, int err)
{
+ struct btrfs_fs_info *fs_info = eb->fs_info;
int ret = 0;
struct reada_extent *re;
@@ -235,10 +234,10 @@ start_machine:
return ret;
}
-static struct reada_zone *reada_find_zone(struct btrfs_fs_info *fs_info,
- struct btrfs_device *dev, u64 logical,
+static struct reada_zone *reada_find_zone(struct btrfs_device *dev, u64 logical,
struct btrfs_bio *bbio)
{
+ struct btrfs_fs_info *fs_info = dev->fs_info;
int ret;
struct reada_zone *zone;
struct btrfs_block_group_cache *cache = NULL;
@@ -270,6 +269,12 @@ static struct reada_zone *reada_find_zone(struct btrfs_fs_info *fs_info,
if (!zone)
return NULL;
+ ret = radix_tree_preload(GFP_KERNEL);
+ if (ret) {
+ kfree(zone);
+ return NULL;
+ }
+
zone->start = start;
zone->end = end;
INIT_LIST_HEAD(&zone->list);
@@ -299,6 +304,7 @@ static struct reada_zone *reada_find_zone(struct btrfs_fs_info *fs_info,
zone = NULL;
}
spin_unlock(&fs_info->reada_lock);
+ radix_tree_preload_end();
return zone;
}
@@ -313,7 +319,6 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info,
struct btrfs_bio *bbio = NULL;
struct btrfs_device *dev;
struct btrfs_device *prev_dev;
- u32 blocksize;
u64 length;
int real_stripes;
int nzones = 0;
@@ -334,7 +339,6 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info,
if (!re)
return NULL;
- blocksize = fs_info->nodesize;
re->logical = logical;
re->top = *top;
INIT_LIST_HEAD(&re->extctl);
@@ -344,10 +348,10 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info,
/*
* map block
*/
- length = blocksize;
+ length = fs_info->nodesize;
ret = btrfs_map_block(fs_info, BTRFS_MAP_GET_READ_MIRRORS, logical,
&length, &bbio, 0);
- if (ret || !bbio || length < blocksize)
+ if (ret || !bbio || length < fs_info->nodesize)
goto error;
if (bbio->num_stripes > BTRFS_MAX_MIRRORS) {
@@ -367,7 +371,7 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info,
if (!dev->bdev)
continue;
- zone = reada_find_zone(fs_info, dev, logical, bbio);
+ zone = reada_find_zone(dev, logical, bbio);
if (!zone)
continue;
@@ -386,6 +390,10 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info,
goto error;
}
+ ret = radix_tree_preload(GFP_KERNEL);
+ if (ret)
+ goto error;
+
/* insert extent in reada_tree + all per-device trees, all or nothing */
btrfs_dev_replace_lock(&fs_info->dev_replace, 0);
spin_lock(&fs_info->reada_lock);
@@ -395,13 +403,16 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info,
re_exist->refcnt++;
spin_unlock(&fs_info->reada_lock);
btrfs_dev_replace_unlock(&fs_info->dev_replace, 0);
+ radix_tree_preload_end();
goto error;
}
if (ret) {
spin_unlock(&fs_info->reada_lock);
btrfs_dev_replace_unlock(&fs_info->dev_replace, 0);
+ radix_tree_preload_end();
goto error;
}
+ radix_tree_preload_end();
prev_dev = NULL;
dev_replace_is_ongoing = btrfs_dev_replace_is_ongoing(
&fs_info->dev_replace);
@@ -639,9 +650,9 @@ static int reada_pick_zone(struct btrfs_device *dev)
return 1;
}
-static int reada_start_machine_dev(struct btrfs_fs_info *fs_info,
- struct btrfs_device *dev)
+static int reada_start_machine_dev(struct btrfs_device *dev)
{
+ struct btrfs_fs_info *fs_info = dev->fs_info;
struct reada_extent *re = NULL;
int mirror_num = 0;
struct extent_buffer *eb = NULL;
@@ -754,8 +765,7 @@ static void __reada_start_machine(struct btrfs_fs_info *fs_info)
list_for_each_entry(device, &fs_devices->devices, dev_list) {
if (atomic_read(&device->reada_in_flight) <
MAX_IN_FLIGHT)
- enqueued += reada_start_machine_dev(fs_info,
- device);
+ enqueued += reada_start_machine_dev(device);
}
mutex_unlock(&fs_devices->device_list_mutex);
total += enqueued;
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 379711048fb0..65661d1aae4e 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1548,9 +1548,9 @@ again:
prev = node;
entry = rb_entry(node, struct btrfs_inode, rb_node);
- if (objectid < btrfs_ino(&entry->vfs_inode))
+ if (objectid < btrfs_ino(entry))
node = node->rb_left;
- else if (objectid > btrfs_ino(&entry->vfs_inode))
+ else if (objectid > btrfs_ino(entry))
node = node->rb_right;
else
break;
@@ -1558,7 +1558,7 @@ again:
if (!node) {
while (prev) {
entry = rb_entry(prev, struct btrfs_inode, rb_node);
- if (objectid <= btrfs_ino(&entry->vfs_inode)) {
+ if (objectid <= btrfs_ino(entry)) {
node = prev;
break;
}
@@ -1573,7 +1573,7 @@ again:
return inode;
}
- objectid = btrfs_ino(&entry->vfs_inode) + 1;
+ objectid = btrfs_ino(entry) + 1;
if (cond_resched_lock(&root->inode_lock))
goto again;
@@ -1609,8 +1609,8 @@ static int get_new_location(struct inode *reloc_inode, u64 *new_bytenr,
return -ENOMEM;
bytenr -= BTRFS_I(reloc_inode)->index_cnt;
- ret = btrfs_lookup_file_extent(NULL, root, path, btrfs_ino(reloc_inode),
- bytenr, 0);
+ ret = btrfs_lookup_file_extent(NULL, root, path,
+ btrfs_ino(BTRFS_I(reloc_inode)), bytenr, 0);
if (ret < 0)
goto out;
if (ret > 0) {
@@ -1698,11 +1698,11 @@ int replace_file_extents(struct btrfs_trans_handle *trans,
if (first) {
inode = find_next_inode(root, key.objectid);
first = 0;
- } else if (inode && btrfs_ino(inode) < key.objectid) {
+ } else if (inode && btrfs_ino(BTRFS_I(inode)) < key.objectid) {
btrfs_add_delayed_iput(inode);
inode = find_next_inode(root, key.objectid);
}
- if (inode && btrfs_ino(inode) == key.objectid) {
+ if (inode && btrfs_ino(BTRFS_I(inode)) == key.objectid) {
end = key.offset +
btrfs_file_extent_num_bytes(leaf, fi);
WARN_ON(!IS_ALIGNED(key.offset,
@@ -1714,8 +1714,8 @@ int replace_file_extents(struct btrfs_trans_handle *trans,
if (!ret)
continue;
- btrfs_drop_extent_cache(inode, key.offset, end,
- 1);
+ btrfs_drop_extent_cache(BTRFS_I(inode),
+ key.offset, end, 1);
unlock_extent(&BTRFS_I(inode)->io_tree,
key.offset, end);
}
@@ -2088,7 +2088,7 @@ static int invalidate_extent_cache(struct btrfs_root *root,
inode = find_next_inode(root, objectid);
if (!inode)
break;
- ino = btrfs_ino(inode);
+ ino = btrfs_ino(BTRFS_I(inode));
if (ino > max_key->objectid) {
iput(inode);
@@ -2130,7 +2130,7 @@ static int invalidate_extent_cache(struct btrfs_root *root,
/* the lock_extent waits for readpage to complete */
lock_extent(&BTRFS_I(inode)->io_tree, start, end);
- btrfs_drop_extent_cache(inode, start, end, 1);
+ btrfs_drop_extent_cache(BTRFS_I(inode), start, end, 1);
unlock_extent(&BTRFS_I(inode)->io_tree, start, end);
}
return 0;
@@ -3093,11 +3093,12 @@ int prealloc_file_extent_cluster(struct inode *inode,
u64 prealloc_start = cluster->start - offset;
u64 prealloc_end = cluster->end - offset;
u64 cur_offset;
+ struct extent_changeset *data_reserved = NULL;
BUG_ON(cluster->start != cluster->boundary[0]);
inode_lock(inode);
- ret = btrfs_check_data_free_space(inode, prealloc_start,
+ ret = btrfs_check_data_free_space(inode, &data_reserved, prealloc_start,
prealloc_end + 1 - prealloc_start);
if (ret)
goto out;
@@ -3113,8 +3114,8 @@ int prealloc_file_extent_cluster(struct inode *inode,
lock_extent(&BTRFS_I(inode)->io_tree, start, end);
num_bytes = end + 1 - start;
if (cur_offset < start)
- btrfs_free_reserved_data_space(inode, cur_offset,
- start - cur_offset);
+ btrfs_free_reserved_data_space(inode, data_reserved,
+ cur_offset, start - cur_offset);
ret = btrfs_prealloc_file_range(inode, 0, start,
num_bytes, num_bytes,
end + 1, &alloc_hint);
@@ -3125,10 +3126,11 @@ int prealloc_file_extent_cluster(struct inode *inode,
nr++;
}
if (cur_offset < prealloc_end)
- btrfs_free_reserved_data_space(inode, cur_offset,
- prealloc_end + 1 - cur_offset);
+ btrfs_free_reserved_data_space(inode, data_reserved,
+ cur_offset, prealloc_end + 1 - cur_offset);
out:
inode_unlock(inode);
+ extent_changeset_free(data_reserved);
return ret;
}
@@ -3161,7 +3163,7 @@ int setup_extent_mapping(struct inode *inode, u64 start, u64 end,
free_extent_map(em);
break;
}
- btrfs_drop_extent_cache(inode, start, end, 0);
+ btrfs_drop_extent_cache(BTRFS_I(inode), start, end, 0);
}
unlock_extent(&BTRFS_I(inode)->io_tree, start, end);
return ret;
@@ -3203,7 +3205,8 @@ static int relocate_file_extent_cluster(struct inode *inode,
index = (cluster->start - offset) >> PAGE_SHIFT;
last_index = (cluster->end - offset) >> PAGE_SHIFT;
while (index <= last_index) {
- ret = btrfs_delalloc_reserve_metadata(inode, PAGE_SIZE);
+ ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode),
+ PAGE_SIZE);
if (ret)
goto out;
@@ -3215,7 +3218,7 @@ static int relocate_file_extent_cluster(struct inode *inode,
page = find_or_create_page(inode->i_mapping, index,
mask);
if (!page) {
- btrfs_delalloc_release_metadata(inode,
+ btrfs_delalloc_release_metadata(BTRFS_I(inode),
PAGE_SIZE);
ret = -ENOMEM;
goto out;
@@ -3234,7 +3237,7 @@ static int relocate_file_extent_cluster(struct inode *inode,
if (!PageUptodate(page)) {
unlock_page(page);
put_page(page);
- btrfs_delalloc_release_metadata(inode,
+ btrfs_delalloc_release_metadata(BTRFS_I(inode),
PAGE_SIZE);
ret = -EIO;
goto out;
@@ -3543,7 +3546,7 @@ truncate:
goto out;
}
- ret = btrfs_truncate_free_space_cache(root, trans, block_group, inode);
+ ret = btrfs_truncate_free_space_cache(trans, block_group, inode);
btrfs_end_transaction(trans);
btrfs_btree_balance_dirty(fs_info);
@@ -4245,7 +4248,7 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info,
BUG_ON(IS_ERR(inode) || is_bad_inode(inode));
BTRFS_I(inode)->index_cnt = group->key.objectid;
- err = btrfs_orphan_add(trans, inode);
+ err = btrfs_orphan_add(trans, BTRFS_I(inode));
out:
btrfs_end_transaction(trans);
btrfs_btree_balance_dirty(fs_info);
@@ -4268,8 +4271,7 @@ static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info)
INIT_LIST_HEAD(&rc->reloc_roots);
backref_cache_init(&rc->backref_cache);
mapping_tree_init(&rc->reloc_root_tree);
- extent_io_tree_init(&rc->processed_blocks,
- fs_info->btree_inode->i_mapping);
+ extent_io_tree_init(&rc->processed_blocks, NULL);
return rc;
}
@@ -4334,7 +4336,7 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
rc->block_group = btrfs_lookup_block_group(fs_info, group_start);
BUG_ON(!rc->block_group);
- ret = btrfs_inc_block_group_ro(extent_root, rc->block_group);
+ ret = btrfs_inc_block_group_ro(fs_info, rc->block_group);
if (ret) {
err = ret;
goto out;
@@ -4347,8 +4349,7 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
goto out;
}
- inode = lookup_free_space_inode(fs_info->tree_root, rc->block_group,
- path);
+ inode = lookup_free_space_inode(fs_info, rc->block_group, path);
btrfs_free_path(path);
if (!IS_ERR(inode))
@@ -4372,7 +4373,7 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
btrfs_wait_block_group_reservations(rc->block_group);
btrfs_wait_nocow_writers(rc->block_group);
- btrfs_wait_ordered_roots(fs_info, -1,
+ btrfs_wait_ordered_roots(fs_info, U64_MAX,
rc->block_group->key.objectid,
rc->block_group->key.offset);
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index 4c6735491ee0..460db0cb2d07 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -74,7 +74,7 @@ static void btrfs_read_root_item(struct extent_buffer *eb, int slot,
*
* If we find something return 0, otherwise > 0, < 0 on error.
*/
-int btrfs_find_root(struct btrfs_root *root, struct btrfs_key *search_key,
+int btrfs_find_root(struct btrfs_root *root, const struct btrfs_key *search_key,
struct btrfs_path *path, struct btrfs_root_item *root_item,
struct btrfs_key *root_key)
{
@@ -207,7 +207,7 @@ out:
}
int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
- struct btrfs_key *key, struct btrfs_root_item *item)
+ const struct btrfs_key *key, struct btrfs_root_item *item)
{
/*
* Make sure generation v1 and v2 match. See update_root for details.
@@ -337,7 +337,7 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
/* drop the root item for 'key' from 'root' */
int btrfs_del_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
- struct btrfs_key *key)
+ const struct btrfs_key *key)
{
struct btrfs_path *path;
int ret;
@@ -390,6 +390,13 @@ again:
WARN_ON(btrfs_root_ref_dirid(leaf, ref) != dirid);
WARN_ON(btrfs_root_ref_name_len(leaf, ref) != name_len);
ptr = (unsigned long)(ref + 1);
+ ret = btrfs_is_name_len_valid(leaf, path->slots[0], ptr,
+ name_len);
+ if (!ret) {
+ err = -EIO;
+ goto out;
+ }
+
WARN_ON(memcmp_extent_buffer(leaf, name, ptr, name_len));
*sequence = btrfs_root_ref_sequence(leaf, ref);
@@ -501,8 +508,9 @@ void btrfs_update_root_times(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
struct btrfs_root_item *item = &root->root_item;
- struct timespec ct = current_fs_time(root->fs_info->sb);
+ struct timespec ct;
+ ktime_get_real_ts(&ct);
spin_lock(&root->root_item_lock);
btrfs_set_root_ctransid(item, trans->transid);
btrfs_set_stack_timespec_sec(&item->ctime, ct.tv_sec);
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 9a94670536a6..6f1e4c984b94 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -18,6 +18,7 @@
#include <linux/blkdev.h>
#include <linux/ratelimit.h>
+#include <linux/sched/mm.h>
#include "ctree.h"
#include "volumes.h"
#include "disk-io.h"
@@ -64,7 +65,7 @@ struct scrub_ctx;
#define SCRUB_MAX_PAGES_PER_BLOCK 16 /* 64k per node/leaf/sector */
struct scrub_recover {
- atomic_t refs;
+ refcount_t refs;
struct btrfs_bio *bbio;
u64 map_length;
};
@@ -95,7 +96,7 @@ struct scrub_bio {
struct scrub_ctx *sctx;
struct btrfs_device *dev;
struct bio *bio;
- int err;
+ blk_status_t status;
u64 logical;
u64 physical;
#if SCRUB_PAGES_PER_WR_BIO >= SCRUB_PAGES_PER_RD_BIO
@@ -112,7 +113,7 @@ struct scrub_block {
struct scrub_page *pagev[SCRUB_MAX_PAGES_PER_BLOCK];
int page_count;
atomic_t outstanding_pages;
- atomic_t refs; /* free mem on transition to zero */
+ refcount_t refs; /* free mem on transition to zero */
struct scrub_ctx *sctx;
struct scrub_parity *sparity;
struct {
@@ -140,9 +141,9 @@ struct scrub_parity {
int nsectors;
- int stripe_len;
+ u64 stripe_len;
- atomic_t refs;
+ refcount_t refs;
struct list_head spages;
@@ -161,14 +162,6 @@ struct scrub_parity {
unsigned long bitmap[0];
};
-struct scrub_wr_ctx {
- struct scrub_bio *wr_curr_bio;
- struct btrfs_device *tgtdev;
- int pages_per_wr_bio; /* <= SCRUB_PAGES_PER_WR_BIO */
- atomic_t flush_all_writes;
- struct mutex wr_lock;
-};
-
struct scrub_ctx {
struct scrub_bio *bios[SCRUB_BIOS_PER_SCTX];
struct btrfs_fs_info *fs_info;
@@ -183,11 +176,14 @@ struct scrub_ctx {
atomic_t cancel_req;
int readonly;
int pages_per_rd_bio;
- u32 sectorsize;
- u32 nodesize;
int is_dev_replace;
- struct scrub_wr_ctx wr_ctx;
+
+ struct scrub_bio *wr_curr_bio;
+ struct mutex wr_lock;
+ int pages_per_wr_bio; /* <= SCRUB_PAGES_PER_WR_BIO */
+ atomic_t flush_all_writes;
+ struct btrfs_device *wr_tgtdev;
/*
* statistics
@@ -202,7 +198,7 @@ struct scrub_ctx {
* doesn't free the scrub context before or while the workers are
* doing the wakeup() call.
*/
- atomic_t refs;
+ refcount_t refs;
};
struct scrub_fixup_nodatasum {
@@ -240,6 +236,13 @@ struct scrub_warning {
struct btrfs_device *dev;
};
+struct full_stripe_lock {
+ struct rb_node node;
+ u64 logical;
+ u64 refs;
+ struct mutex mutex;
+};
+
static void scrub_pending_bio_inc(struct scrub_ctx *sctx);
static void scrub_pending_bio_dec(struct scrub_ctx *sctx);
static void scrub_pending_trans_workers_inc(struct scrub_ctx *sctx);
@@ -282,12 +285,6 @@ static void scrub_remap_extent(struct btrfs_fs_info *fs_info,
u64 *extent_physical,
struct btrfs_device **extent_dev,
int *extent_mirror_num);
-static int scrub_setup_wr_ctx(struct scrub_ctx *sctx,
- struct scrub_wr_ctx *wr_ctx,
- struct btrfs_fs_info *fs_info,
- struct btrfs_device *dev,
- int is_dev_replace);
-static void scrub_free_wr_ctx(struct scrub_wr_ctx *wr_ctx);
static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx,
struct scrub_page *spage);
static void scrub_wr_submit(struct scrub_ctx *sctx);
@@ -307,7 +304,7 @@ static void scrub_put_ctx(struct scrub_ctx *sctx);
static void scrub_pending_bio_inc(struct scrub_ctx *sctx)
{
- atomic_inc(&sctx->refs);
+ refcount_inc(&sctx->refs);
atomic_inc(&sctx->bios_in_flight);
}
@@ -351,6 +348,222 @@ static void scrub_blocked_if_needed(struct btrfs_fs_info *fs_info)
}
/*
+ * Insert new full stripe lock into full stripe locks tree
+ *
+ * Return pointer to existing or newly inserted full_stripe_lock structure if
+ * everything works well.
+ * Return ERR_PTR(-ENOMEM) if we failed to allocate memory
+ *
+ * NOTE: caller must hold full_stripe_locks_root->lock before calling this
+ * function
+ */
+static struct full_stripe_lock *insert_full_stripe_lock(
+ struct btrfs_full_stripe_locks_tree *locks_root,
+ u64 fstripe_logical)
+{
+ struct rb_node **p;
+ struct rb_node *parent = NULL;
+ struct full_stripe_lock *entry;
+ struct full_stripe_lock *ret;
+
+ WARN_ON(!mutex_is_locked(&locks_root->lock));
+
+ p = &locks_root->root.rb_node;
+ while (*p) {
+ parent = *p;
+ entry = rb_entry(parent, struct full_stripe_lock, node);
+ if (fstripe_logical < entry->logical) {
+ p = &(*p)->rb_left;
+ } else if (fstripe_logical > entry->logical) {
+ p = &(*p)->rb_right;
+ } else {
+ entry->refs++;
+ return entry;
+ }
+ }
+
+ /* Insert new lock */
+ ret = kmalloc(sizeof(*ret), GFP_KERNEL);
+ if (!ret)
+ return ERR_PTR(-ENOMEM);
+ ret->logical = fstripe_logical;
+ ret->refs = 1;
+ mutex_init(&ret->mutex);
+
+ rb_link_node(&ret->node, parent, p);
+ rb_insert_color(&ret->node, &locks_root->root);
+ return ret;
+}
+
+/*
+ * Search for a full stripe lock of a block group
+ *
+ * Return pointer to existing full stripe lock if found
+ * Return NULL if not found
+ */
+static struct full_stripe_lock *search_full_stripe_lock(
+ struct btrfs_full_stripe_locks_tree *locks_root,
+ u64 fstripe_logical)
+{
+ struct rb_node *node;
+ struct full_stripe_lock *entry;
+
+ WARN_ON(!mutex_is_locked(&locks_root->lock));
+
+ node = locks_root->root.rb_node;
+ while (node) {
+ entry = rb_entry(node, struct full_stripe_lock, node);
+ if (fstripe_logical < entry->logical)
+ node = node->rb_left;
+ else if (fstripe_logical > entry->logical)
+ node = node->rb_right;
+ else
+ return entry;
+ }
+ return NULL;
+}
+
+/*
+ * Helper to get full stripe logical from a normal bytenr.
+ *
+ * Caller must ensure @cache is a RAID56 block group.
+ */
+static u64 get_full_stripe_logical(struct btrfs_block_group_cache *cache,
+ u64 bytenr)
+{
+ u64 ret;
+
+ /*
+ * Due to chunk item size limit, full stripe length should not be
+ * larger than U32_MAX. Just a sanity check here.
+ */
+ WARN_ON_ONCE(cache->full_stripe_len >= U32_MAX);
+
+ /*
+ * round_down() can only handle power of 2, while RAID56 full
+ * stripe length can be 64KiB * n, so we need to manually round down.
+ */
+ ret = div64_u64(bytenr - cache->key.objectid, cache->full_stripe_len) *
+ cache->full_stripe_len + cache->key.objectid;
+ return ret;
+}
+
+/*
+ * Lock a full stripe to avoid concurrency of recovery and read
+ *
+ * It's only used for profiles with parities (RAID5/6), for other profiles it
+ * does nothing.
+ *
+ * Return 0 if we locked full stripe covering @bytenr, with a mutex held.
+ * So caller must call unlock_full_stripe() at the same context.
+ *
+ * Return <0 if encounters error.
+ */
+static int lock_full_stripe(struct btrfs_fs_info *fs_info, u64 bytenr,
+ bool *locked_ret)
+{
+ struct btrfs_block_group_cache *bg_cache;
+ struct btrfs_full_stripe_locks_tree *locks_root;
+ struct full_stripe_lock *existing;
+ u64 fstripe_start;
+ int ret = 0;
+
+ *locked_ret = false;
+ bg_cache = btrfs_lookup_block_group(fs_info, bytenr);
+ if (!bg_cache) {
+ ASSERT(0);
+ return -ENOENT;
+ }
+
+ /* Profiles not based on parity don't need full stripe lock */
+ if (!(bg_cache->flags & BTRFS_BLOCK_GROUP_RAID56_MASK))
+ goto out;
+ locks_root = &bg_cache->full_stripe_locks_root;
+
+ fstripe_start = get_full_stripe_logical(bg_cache, bytenr);
+
+ /* Now insert the full stripe lock */
+ mutex_lock(&locks_root->lock);
+ existing = insert_full_stripe_lock(locks_root, fstripe_start);
+ mutex_unlock(&locks_root->lock);
+ if (IS_ERR(existing)) {
+ ret = PTR_ERR(existing);
+ goto out;
+ }
+ mutex_lock(&existing->mutex);
+ *locked_ret = true;
+out:
+ btrfs_put_block_group(bg_cache);
+ return ret;
+}
+
+/*
+ * Unlock a full stripe.
+ *
+ * NOTE: Caller must ensure it's the same context calling corresponding
+ * lock_full_stripe().
+ *
+ * Return 0 if we unlock full stripe without problem.
+ * Return <0 for error
+ */
+static int unlock_full_stripe(struct btrfs_fs_info *fs_info, u64 bytenr,
+ bool locked)
+{
+ struct btrfs_block_group_cache *bg_cache;
+ struct btrfs_full_stripe_locks_tree *locks_root;
+ struct full_stripe_lock *fstripe_lock;
+ u64 fstripe_start;
+ bool freeit = false;
+ int ret = 0;
+
+ /* If we didn't acquire full stripe lock, no need to continue */
+ if (!locked)
+ return 0;
+
+ bg_cache = btrfs_lookup_block_group(fs_info, bytenr);
+ if (!bg_cache) {
+ ASSERT(0);
+ return -ENOENT;
+ }
+ if (!(bg_cache->flags & BTRFS_BLOCK_GROUP_RAID56_MASK))
+ goto out;
+
+ locks_root = &bg_cache->full_stripe_locks_root;
+ fstripe_start = get_full_stripe_logical(bg_cache, bytenr);
+
+ mutex_lock(&locks_root->lock);
+ fstripe_lock = search_full_stripe_lock(locks_root, fstripe_start);
+ /* Unpaired unlock_full_stripe() detected */
+ if (!fstripe_lock) {
+ WARN_ON(1);
+ ret = -ENOENT;
+ mutex_unlock(&locks_root->lock);
+ goto out;
+ }
+
+ if (fstripe_lock->refs == 0) {
+ WARN_ON(1);
+ btrfs_warn(fs_info, "full stripe lock at %llu refcount underflow",
+ fstripe_lock->logical);
+ } else {
+ fstripe_lock->refs--;
+ }
+
+ if (fstripe_lock->refs == 0) {
+ rb_erase(&fstripe_lock->node, &locks_root->root);
+ freeit = true;
+ }
+ mutex_unlock(&locks_root->lock);
+
+ mutex_unlock(&fstripe_lock->mutex);
+ if (freeit)
+ kfree(fstripe_lock);
+out:
+ btrfs_put_block_group(bg_cache);
+ return ret;
+}
+
+/*
* used for workers that require transaction commits (i.e., for the
* NOCOW case)
*/
@@ -358,7 +571,7 @@ static void scrub_pending_trans_workers_inc(struct scrub_ctx *sctx)
{
struct btrfs_fs_info *fs_info = sctx->fs_info;
- atomic_inc(&sctx->refs);
+ refcount_inc(&sctx->refs);
/*
* increment scrubs_running to prevent cancel requests from
* completing as long as a worker is running. we must also
@@ -422,8 +635,6 @@ static noinline_for_stack void scrub_free_ctx(struct scrub_ctx *sctx)
if (!sctx)
return;
- scrub_free_wr_ctx(&sctx->wr_ctx);
-
/* this can happen when scrub is cancelled */
if (sctx->curr != -1) {
struct scrub_bio *sbio = sctx->bios[sctx->curr];
@@ -443,13 +654,14 @@ static noinline_for_stack void scrub_free_ctx(struct scrub_ctx *sctx)
kfree(sbio);
}
+ kfree(sctx->wr_curr_bio);
scrub_free_csums(sctx);
kfree(sctx);
}
static void scrub_put_ctx(struct scrub_ctx *sctx)
{
- if (atomic_dec_and_test(&sctx->refs))
+ if (refcount_dec_and_test(&sctx->refs))
scrub_free_ctx(sctx);
}
@@ -459,12 +671,11 @@ struct scrub_ctx *scrub_setup_ctx(struct btrfs_device *dev, int is_dev_replace)
struct scrub_ctx *sctx;
int i;
struct btrfs_fs_info *fs_info = dev->fs_info;
- int ret;
sctx = kzalloc(sizeof(*sctx), GFP_KERNEL);
if (!sctx)
goto nomem;
- atomic_set(&sctx->refs, 1);
+ refcount_set(&sctx->refs, 1);
sctx->is_dev_replace = is_dev_replace;
sctx->pages_per_rd_bio = SCRUB_PAGES_PER_RD_BIO;
sctx->curr = -1;
@@ -489,8 +700,6 @@ struct scrub_ctx *scrub_setup_ctx(struct btrfs_device *dev, int is_dev_replace)
sctx->bios[i]->next_free = -1;
}
sctx->first_free = 0;
- sctx->nodesize = fs_info->nodesize;
- sctx->sectorsize = fs_info->sectorsize;
atomic_set(&sctx->bios_in_flight, 0);
atomic_set(&sctx->workers_pending, 0);
atomic_set(&sctx->cancel_req, 0);
@@ -501,12 +710,16 @@ struct scrub_ctx *scrub_setup_ctx(struct btrfs_device *dev, int is_dev_replace)
spin_lock_init(&sctx->stat_lock);
init_waitqueue_head(&sctx->list_wait);
- ret = scrub_setup_wr_ctx(sctx, &sctx->wr_ctx, fs_info,
- fs_info->dev_replace.tgtdev, is_dev_replace);
- if (ret) {
- scrub_free_ctx(sctx);
- return ERR_PTR(ret);
+ WARN_ON(sctx->wr_curr_bio != NULL);
+ mutex_init(&sctx->wr_lock);
+ sctx->wr_curr_bio = NULL;
+ if (is_dev_replace) {
+ WARN_ON(!fs_info->dev_replace.tgtdev);
+ sctx->pages_per_wr_bio = SCRUB_PAGES_PER_WR_BIO;
+ sctx->wr_tgtdev = fs_info->dev_replace.tgtdev;
+ atomic_set(&sctx->flush_all_writes, 0);
}
+
return sctx;
nomem:
@@ -521,6 +734,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
u32 nlink;
int ret;
int i;
+ unsigned nofs_flag;
struct extent_buffer *eb;
struct btrfs_inode_item *inode_item;
struct scrub_warning *swarn = warn_ctx;
@@ -559,7 +773,14 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
nlink = btrfs_inode_nlink(eb, inode_item);
btrfs_release_path(swarn->path);
+ /*
+ * init_path might indirectly call vmalloc, or use GFP_KERNEL. Scrub
+ * uses GFP_NOFS in this context, so we keep it consistent but it does
+ * not seem to be strictly necessary.
+ */
+ nofs_flag = memalloc_nofs_save();
ipath = init_ipath(4096, local_root, swarn->path);
+ memalloc_nofs_restore(nofs_flag);
if (IS_ERR(ipath)) {
ret = PTR_ERR(ipath);
ipath = NULL;
@@ -733,7 +954,7 @@ static int scrub_fixup_readpage(u64 inum, u64 offset, u64 root, void *fixup_ctx)
ret = -EIO;
goto out;
}
- ret = repair_io_failure(inode, offset, PAGE_SIZE,
+ ret = repair_io_failure(fs_info, inum, offset, PAGE_SIZE,
fixup->logical, page,
offset - page_offset(page),
fixup->mirror_num);
@@ -859,12 +1080,14 @@ out:
static inline void scrub_get_recover(struct scrub_recover *recover)
{
- atomic_inc(&recover->refs);
+ refcount_inc(&recover->refs);
}
-static inline void scrub_put_recover(struct scrub_recover *recover)
+static inline void scrub_put_recover(struct btrfs_fs_info *fs_info,
+ struct scrub_recover *recover)
{
- if (atomic_dec_and_test(&recover->refs)) {
+ if (refcount_dec_and_test(&recover->refs)) {
+ btrfs_bio_counter_dec(fs_info);
btrfs_put_bbio(recover->bbio);
kfree(recover);
}
@@ -894,6 +1117,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
int mirror_index;
int page_num;
int success;
+ bool full_stripe_locked;
static DEFINE_RATELIMIT_STATE(_rs, DEFAULT_RATELIMIT_INTERVAL,
DEFAULT_RATELIMIT_BURST);
@@ -919,6 +1143,24 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
have_csum = sblock_to_check->pagev[0]->have_csum;
dev = sblock_to_check->pagev[0]->dev;
+ /*
+ * For RAID5/6, race can happen for a different device scrub thread.
+ * For data corruption, Parity and Data threads will both try
+ * to recovery the data.
+ * Race can lead to doubly added csum error, or even unrecoverable
+ * error.
+ */
+ ret = lock_full_stripe(fs_info, logical, &full_stripe_locked);
+ if (ret < 0) {
+ spin_lock(&sctx->stat_lock);
+ if (ret == -ENOMEM)
+ sctx->stat.malloc_errors++;
+ sctx->stat.read_errors++;
+ sctx->stat.uncorrectable_errors++;
+ spin_unlock(&sctx->stat_lock);
+ return ret;
+ }
+
if (sctx->is_dev_replace && !is_metadata && !have_csum) {
sblocks_for_recheck = NULL;
goto nodatasum_case;
@@ -1243,7 +1485,7 @@ out:
sblock->pagev[page_index]->sblock = NULL;
recover = sblock->pagev[page_index]->recover;
if (recover) {
- scrub_put_recover(recover);
+ scrub_put_recover(fs_info, recover);
sblock->pagev[page_index]->recover =
NULL;
}
@@ -1253,6 +1495,9 @@ out:
kfree(sblocks_for_recheck);
}
+ ret = unlock_full_stripe(fs_info, logical, full_stripe_locked);
+ if (ret < 0)
+ return ret;
return 0;
}
@@ -1332,20 +1577,23 @@ static int scrub_setup_recheck_block(struct scrub_block *original_sblock,
* with a length of PAGE_SIZE, each returned stripe
* represents one mirror
*/
+ btrfs_bio_counter_inc_blocked(fs_info);
ret = btrfs_map_sblock(fs_info, BTRFS_MAP_GET_READ_MIRRORS,
- logical, &mapped_length, &bbio, 0, 1);
+ logical, &mapped_length, &bbio);
if (ret || !bbio || mapped_length < sublen) {
btrfs_put_bbio(bbio);
+ btrfs_bio_counter_dec(fs_info);
return -EIO;
}
recover = kzalloc(sizeof(struct scrub_recover), GFP_NOFS);
if (!recover) {
btrfs_put_bbio(bbio);
+ btrfs_bio_counter_dec(fs_info);
return -ENOMEM;
}
- atomic_set(&recover->refs, 1);
+ refcount_set(&recover->refs, 1);
recover->bbio = bbio;
recover->map_length = mapped_length;
@@ -1367,7 +1615,7 @@ leave_nomem:
spin_lock(&sctx->stat_lock);
sctx->stat.malloc_errors++;
spin_unlock(&sctx->stat_lock);
- scrub_put_recover(recover);
+ scrub_put_recover(fs_info, recover);
return -ENOMEM;
}
scrub_page_get(page);
@@ -1409,7 +1657,7 @@ leave_nomem:
scrub_get_recover(recover);
page->recover = recover;
}
- scrub_put_recover(recover);
+ scrub_put_recover(fs_info, recover);
length -= sublen;
logical += sublen;
page_index++;
@@ -1420,14 +1668,14 @@ leave_nomem:
struct scrub_bio_ret {
struct completion event;
- int error;
+ blk_status_t status;
};
static void scrub_bio_wait_endio(struct bio *bio)
{
struct scrub_bio_ret *ret = bio->bi_private;
- ret->error = bio->bi_error;
+ ret->status = bio->bi_status;
complete(&ret->event);
}
@@ -1445,7 +1693,7 @@ static int scrub_submit_raid56_bio_wait(struct btrfs_fs_info *fs_info,
int ret;
init_completion(&done.event);
- done.error = 0;
+ done.status = 0;
bio->bi_iter.bi_sector = page->logical >> 9;
bio->bi_private = &done;
bio->bi_end_io = scrub_bio_wait_endio;
@@ -1457,7 +1705,7 @@ static int scrub_submit_raid56_bio_wait(struct btrfs_fs_info *fs_info,
return ret;
wait_for_completion(&done.event);
- if (done.error)
+ if (done.status)
return -EIO;
return 0;
@@ -1489,24 +1737,23 @@ static void scrub_recheck_block(struct btrfs_fs_info *fs_info,
}
WARN_ON(!page->page);
- bio = btrfs_io_bio_alloc(GFP_NOFS, 1);
- if (!bio) {
- page->io_error = 1;
- sblock->no_io_error_seen = 0;
- continue;
- }
+ bio = btrfs_io_bio_alloc(1);
bio->bi_bdev = page->dev->bdev;
bio_add_page(bio, page->page, PAGE_SIZE, 0);
if (!retry_failed_mirror && scrub_is_page_on_raid56(page)) {
- if (scrub_submit_raid56_bio_wait(fs_info, bio, page))
+ if (scrub_submit_raid56_bio_wait(fs_info, bio, page)) {
+ page->io_error = 1;
sblock->no_io_error_seen = 0;
+ }
} else {
bio->bi_iter.bi_sector = page->physical >> 9;
bio_set_op_attrs(bio, REQ_OP_READ, 0);
- if (btrfsic_submit_bio_wait(bio))
+ if (btrfsic_submit_bio_wait(bio)) {
+ page->io_error = 1;
sblock->no_io_error_seen = 0;
+ }
}
bio_put(bio);
@@ -1578,9 +1825,7 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
return -EIO;
}
- bio = btrfs_io_bio_alloc(GFP_NOFS, 1);
- if (!bio)
- return -EIO;
+ bio = btrfs_io_bio_alloc(1);
bio->bi_bdev = page_bad->dev->bdev;
bio->bi_iter.bi_sector = page_bad->physical >> 9;
bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
@@ -1636,7 +1881,7 @@ static int scrub_write_page_to_dev_replace(struct scrub_block *sblock,
if (spage->io_error) {
void *mapped_buffer = kmap_atomic(spage->page);
- memset(mapped_buffer, 0, PAGE_SIZE);
+ clear_page(mapped_buffer);
flush_dcache_page(spage->page);
kunmap_atomic(mapped_buffer);
}
@@ -1646,37 +1891,31 @@ static int scrub_write_page_to_dev_replace(struct scrub_block *sblock,
static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx,
struct scrub_page *spage)
{
- struct scrub_wr_ctx *wr_ctx = &sctx->wr_ctx;
struct scrub_bio *sbio;
int ret;
- mutex_lock(&wr_ctx->wr_lock);
+ mutex_lock(&sctx->wr_lock);
again:
- if (!wr_ctx->wr_curr_bio) {
- wr_ctx->wr_curr_bio = kzalloc(sizeof(*wr_ctx->wr_curr_bio),
+ if (!sctx->wr_curr_bio) {
+ sctx->wr_curr_bio = kzalloc(sizeof(*sctx->wr_curr_bio),
GFP_KERNEL);
- if (!wr_ctx->wr_curr_bio) {
- mutex_unlock(&wr_ctx->wr_lock);
+ if (!sctx->wr_curr_bio) {
+ mutex_unlock(&sctx->wr_lock);
return -ENOMEM;
}
- wr_ctx->wr_curr_bio->sctx = sctx;
- wr_ctx->wr_curr_bio->page_count = 0;
+ sctx->wr_curr_bio->sctx = sctx;
+ sctx->wr_curr_bio->page_count = 0;
}
- sbio = wr_ctx->wr_curr_bio;
+ sbio = sctx->wr_curr_bio;
if (sbio->page_count == 0) {
struct bio *bio;
sbio->physical = spage->physical_for_dev_replace;
sbio->logical = spage->logical;
- sbio->dev = wr_ctx->tgtdev;
+ sbio->dev = sctx->wr_tgtdev;
bio = sbio->bio;
if (!bio) {
- bio = btrfs_io_bio_alloc(GFP_KERNEL,
- wr_ctx->pages_per_wr_bio);
- if (!bio) {
- mutex_unlock(&wr_ctx->wr_lock);
- return -ENOMEM;
- }
+ bio = btrfs_io_bio_alloc(sctx->pages_per_wr_bio);
sbio->bio = bio;
}
@@ -1685,7 +1924,7 @@ again:
bio->bi_bdev = sbio->dev->bdev;
bio->bi_iter.bi_sector = sbio->physical >> 9;
bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
- sbio->err = 0;
+ sbio->status = 0;
} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
spage->physical_for_dev_replace ||
sbio->logical + sbio->page_count * PAGE_SIZE !=
@@ -1699,7 +1938,7 @@ again:
if (sbio->page_count < 1) {
bio_put(sbio->bio);
sbio->bio = NULL;
- mutex_unlock(&wr_ctx->wr_lock);
+ mutex_unlock(&sctx->wr_lock);
return -EIO;
}
scrub_wr_submit(sctx);
@@ -1709,23 +1948,22 @@ again:
sbio->pagev[sbio->page_count] = spage;
scrub_page_get(spage);
sbio->page_count++;
- if (sbio->page_count == wr_ctx->pages_per_wr_bio)
+ if (sbio->page_count == sctx->pages_per_wr_bio)
scrub_wr_submit(sctx);
- mutex_unlock(&wr_ctx->wr_lock);
+ mutex_unlock(&sctx->wr_lock);
return 0;
}
static void scrub_wr_submit(struct scrub_ctx *sctx)
{
- struct scrub_wr_ctx *wr_ctx = &sctx->wr_ctx;
struct scrub_bio *sbio;
- if (!wr_ctx->wr_curr_bio)
+ if (!sctx->wr_curr_bio)
return;
- sbio = wr_ctx->wr_curr_bio;
- wr_ctx->wr_curr_bio = NULL;
+ sbio = sctx->wr_curr_bio;
+ sctx->wr_curr_bio = NULL;
WARN_ON(!sbio->bio->bi_bdev);
scrub_pending_bio_inc(sctx);
/* process all writes in a single worker thread. Then the block layer
@@ -1740,7 +1978,7 @@ static void scrub_wr_bio_end_io(struct bio *bio)
struct scrub_bio *sbio = bio->bi_private;
struct btrfs_fs_info *fs_info = sbio->dev->fs_info;
- sbio->err = bio->bi_error;
+ sbio->status = bio->bi_status;
sbio->bio = bio;
btrfs_init_work(&sbio->work, btrfs_scrubwrc_helper,
@@ -1755,7 +1993,7 @@ static void scrub_wr_bio_end_io_worker(struct btrfs_work *work)
int i;
WARN_ON(sbio->page_count > SCRUB_PAGES_PER_WR_BIO);
- if (sbio->err) {
+ if (sbio->status) {
struct btrfs_dev_replace *dev_replace =
&sbio->sctx->fs_info->dev_replace;
@@ -1829,7 +2067,7 @@ static int scrub_checksum_data(struct scrub_block *sblock)
page = sblock->pagev[0]->page;
buffer = kmap_atomic(page);
- len = sctx->sectorsize;
+ len = sctx->fs_info->sectorsize;
index = 0;
for (;;) {
u64 l = min_t(u64, len, PAGE_SIZE);
@@ -1894,7 +2132,7 @@ static int scrub_checksum_tree_block(struct scrub_block *sblock)
BTRFS_UUID_SIZE))
sblock->header_error = 1;
- len = sctx->nodesize - BTRFS_CSUM_SIZE;
+ len = sctx->fs_info->nodesize - BTRFS_CSUM_SIZE;
mapped_size = PAGE_SIZE - BTRFS_CSUM_SIZE;
p = ((u8 *)mapped_buffer) + BTRFS_CSUM_SIZE;
index = 0;
@@ -2000,12 +2238,12 @@ static int scrub_checksum_super(struct scrub_block *sblock)
static void scrub_block_get(struct scrub_block *sblock)
{
- atomic_inc(&sblock->refs);
+ refcount_inc(&sblock->refs);
}
static void scrub_block_put(struct scrub_block *sblock)
{
- if (atomic_dec_and_test(&sblock->refs)) {
+ if (refcount_dec_and_test(&sblock->refs)) {
int i;
if (sblock->sparity)
@@ -2077,10 +2315,7 @@ again:
sbio->dev = spage->dev;
bio = sbio->bio;
if (!bio) {
- bio = btrfs_io_bio_alloc(GFP_KERNEL,
- sctx->pages_per_rd_bio);
- if (!bio)
- return -ENOMEM;
+ bio = btrfs_io_bio_alloc(sctx->pages_per_rd_bio);
sbio->bio = bio;
}
@@ -2089,7 +2324,7 @@ again:
bio->bi_bdev = sbio->dev->bdev;
bio->bi_iter.bi_sector = sbio->physical >> 9;
bio_set_op_attrs(bio, REQ_OP_READ, 0);
- sbio->err = 0;
+ sbio->status = 0;
} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
spage->physical ||
sbio->logical + sbio->page_count * PAGE_SIZE !=
@@ -2125,7 +2360,7 @@ static void scrub_missing_raid56_end_io(struct bio *bio)
struct scrub_block *sblock = bio->bi_private;
struct btrfs_fs_info *fs_info = sblock->sctx->fs_info;
- if (bio->bi_error)
+ if (bio->bi_status)
sblock->no_io_error_seen = 0;
bio_put(bio);
@@ -2168,10 +2403,10 @@ static void scrub_missing_raid56_worker(struct btrfs_work *work)
scrub_block_put(sblock);
if (sctx->is_dev_replace &&
- atomic_read(&sctx->wr_ctx.flush_all_writes)) {
- mutex_lock(&sctx->wr_ctx.wr_lock);
+ atomic_read(&sctx->flush_all_writes)) {
+ mutex_lock(&sctx->wr_lock);
scrub_wr_submit(sctx);
- mutex_unlock(&sctx->wr_ctx.wr_lock);
+ mutex_unlock(&sctx->wr_lock);
}
scrub_pending_bio_dec(sctx);
@@ -2189,8 +2424,9 @@ static void scrub_missing_raid56_pages(struct scrub_block *sblock)
int ret;
int i;
+ btrfs_bio_counter_inc_blocked(fs_info);
ret = btrfs_map_sblock(fs_info, BTRFS_MAP_GET_READ_MIRRORS, logical,
- &length, &bbio, 0, 1);
+ &length, &bbio);
if (ret || !bbio || !bbio->raid_map)
goto bbio_out;
@@ -2205,10 +2441,7 @@ static void scrub_missing_raid56_pages(struct scrub_block *sblock)
goto bbio_out;
}
- bio = btrfs_io_bio_alloc(GFP_NOFS, 0);
- if (!bio)
- goto bbio_out;
-
+ bio = btrfs_io_bio_alloc(0);
bio->bi_iter.bi_sector = logical >> 9;
bio->bi_private = sblock;
bio->bi_end_io = scrub_missing_raid56_end_io;
@@ -2233,6 +2466,7 @@ static void scrub_missing_raid56_pages(struct scrub_block *sblock)
rbio_out:
bio_put(bio);
bbio_out:
+ btrfs_bio_counter_dec(fs_info);
btrfs_put_bbio(bbio);
spin_lock(&sctx->stat_lock);
sctx->stat.malloc_errors++;
@@ -2257,7 +2491,7 @@ static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
/* one ref inside this function, plus one for each page added to
* a bio later on */
- atomic_set(&sblock->refs, 1);
+ refcount_set(&sblock->refs, 1);
sblock->sctx = sctx;
sblock->no_io_error_seen = 1;
@@ -2334,7 +2568,7 @@ static void scrub_bio_end_io(struct bio *bio)
struct scrub_bio *sbio = bio->bi_private;
struct btrfs_fs_info *fs_info = sbio->dev->fs_info;
- sbio->err = bio->bi_error;
+ sbio->status = bio->bi_status;
sbio->bio = bio;
btrfs_queue_work(fs_info->scrub_workers, &sbio->work);
@@ -2347,7 +2581,7 @@ static void scrub_bio_end_io_worker(struct btrfs_work *work)
int i;
BUG_ON(sbio->page_count > SCRUB_PAGES_PER_RD_BIO);
- if (sbio->err) {
+ if (sbio->status) {
for (i = 0; i < sbio->page_count; i++) {
struct scrub_page *spage = sbio->pagev[i];
@@ -2374,10 +2608,10 @@ static void scrub_bio_end_io_worker(struct btrfs_work *work)
spin_unlock(&sctx->list_lock);
if (sctx->is_dev_replace &&
- atomic_read(&sctx->wr_ctx.flush_all_writes)) {
- mutex_lock(&sctx->wr_ctx.wr_lock);
+ atomic_read(&sctx->flush_all_writes)) {
+ mutex_lock(&sctx->wr_lock);
scrub_wr_submit(sctx);
- mutex_unlock(&sctx->wr_ctx.wr_lock);
+ mutex_unlock(&sctx->wr_lock);
}
scrub_pending_bio_dec(sctx);
@@ -2387,7 +2621,7 @@ static inline void __scrub_mark_bitmap(struct scrub_parity *sparity,
unsigned long *bitmap,
u64 start, u64 len)
{
- u32 offset;
+ u64 offset;
int nsectors;
int sectorsize = sparity->sctx->fs_info->sectorsize;
@@ -2397,8 +2631,8 @@ static inline void __scrub_mark_bitmap(struct scrub_parity *sparity,
}
start -= sparity->logic_start;
- start = div_u64_rem(start, sparity->stripe_len, &offset);
- offset /= sectorsize;
+ start = div64_u64_rem(start, sparity->stripe_len, &offset);
+ offset = div_u64(offset, sectorsize);
nsectors = (int)len / sectorsize;
if (offset + nsectors <= sparity->nsectors) {
@@ -2472,8 +2706,8 @@ static int scrub_find_csum(struct scrub_ctx *sctx, u64 logical, u8 *csum)
if (!sum)
return 0;
- index = ((u32)(logical - sum->bytenr)) / sctx->sectorsize;
- num_sectors = sum->len / sctx->sectorsize;
+ index = ((u32)(logical - sum->bytenr)) / sctx->fs_info->sectorsize;
+ num_sectors = sum->len / sctx->fs_info->sectorsize;
memcpy(csum, sum->sums + index, sctx->csum_size);
if (index == num_sectors - 1) {
list_del(&sum->list);
@@ -2492,19 +2726,19 @@ static int scrub_extent(struct scrub_ctx *sctx, u64 logical, u64 len,
u32 blocksize;
if (flags & BTRFS_EXTENT_FLAG_DATA) {
- blocksize = sctx->sectorsize;
+ blocksize = sctx->fs_info->sectorsize;
spin_lock(&sctx->stat_lock);
sctx->stat.data_extents_scrubbed++;
sctx->stat.data_bytes_scrubbed += len;
spin_unlock(&sctx->stat_lock);
} else if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
- blocksize = sctx->nodesize;
+ blocksize = sctx->fs_info->nodesize;
spin_lock(&sctx->stat_lock);
sctx->stat.tree_extents_scrubbed++;
sctx->stat.tree_bytes_scrubbed += len;
spin_unlock(&sctx->stat_lock);
} else {
- blocksize = sctx->sectorsize;
+ blocksize = sctx->fs_info->sectorsize;
WARN_ON(1);
}
@@ -2557,7 +2791,7 @@ static int scrub_pages_for_parity(struct scrub_parity *sparity,
/* one ref inside this function, plus one for each page added to
* a bio later on */
- atomic_set(&sblock->refs, 1);
+ refcount_set(&sblock->refs, 1);
sblock->sctx = sctx;
sblock->no_io_error_seen = 1;
sblock->sparity = sparity;
@@ -2638,11 +2872,11 @@ static int scrub_extent_for_parity(struct scrub_parity *sparity,
}
if (flags & BTRFS_EXTENT_FLAG_DATA) {
- blocksize = sctx->sectorsize;
+ blocksize = sctx->fs_info->sectorsize;
} else if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
- blocksize = sctx->nodesize;
+ blocksize = sctx->fs_info->nodesize;
} else {
- blocksize = sctx->sectorsize;
+ blocksize = sctx->fs_info->sectorsize;
WARN_ON(1);
}
@@ -2696,7 +2930,7 @@ static int get_raid56_logic_offset(u64 physical, int num,
for (i = 0; i < nr_data_stripes(map); i++) {
*offset = last_offset + i * map->stripe_len;
- stripe_nr = div_u64(*offset, map->stripe_len);
+ stripe_nr = div64_u64(*offset, map->stripe_len);
stripe_nr = div_u64(stripe_nr, nr_data_stripes(map));
/* Work out the disk rotation on this stripe-set */
@@ -2750,7 +2984,7 @@ static void scrub_parity_bio_endio(struct bio *bio)
struct scrub_parity *sparity = (struct scrub_parity *)bio->bi_private;
struct btrfs_fs_info *fs_info = sparity->sctx->fs_info;
- if (bio->bi_error)
+ if (bio->bi_status)
bitmap_or(sparity->ebitmap, sparity->ebitmap, sparity->dbitmap,
sparity->nsectors);
@@ -2767,7 +3001,6 @@ static void scrub_parity_check_and_repair(struct scrub_parity *sparity)
struct btrfs_fs_info *fs_info = sctx->fs_info;
struct bio *bio;
struct btrfs_raid_bio *rbio;
- struct scrub_page *spage;
struct btrfs_bio *bbio = NULL;
u64 length;
int ret;
@@ -2777,15 +3010,14 @@ static void scrub_parity_check_and_repair(struct scrub_parity *sparity)
goto out;
length = sparity->logic_end - sparity->logic_start;
+
+ btrfs_bio_counter_inc_blocked(fs_info);
ret = btrfs_map_sblock(fs_info, BTRFS_MAP_WRITE, sparity->logic_start,
- &length, &bbio, 0, 1);
+ &length, &bbio);
if (ret || !bbio || !bbio->raid_map)
goto bbio_out;
- bio = btrfs_io_bio_alloc(GFP_NOFS, 0);
- if (!bio)
- goto bbio_out;
-
+ bio = btrfs_io_bio_alloc(0);
bio->bi_iter.bi_sector = sparity->logic_start >> 9;
bio->bi_private = sparity;
bio->bi_end_io = scrub_parity_bio_endio;
@@ -2797,9 +3029,6 @@ static void scrub_parity_check_and_repair(struct scrub_parity *sparity)
if (!rbio)
goto rbio_out;
- list_for_each_entry(spage, &sparity->spages, list)
- raid56_add_scrub_pages(rbio, spage->page, spage->logical);
-
scrub_pending_bio_inc(sctx);
raid56_parity_submit_scrub_rbio(rbio);
return;
@@ -2807,6 +3036,7 @@ static void scrub_parity_check_and_repair(struct scrub_parity *sparity)
rbio_out:
bio_put(bio);
bbio_out:
+ btrfs_bio_counter_dec(fs_info);
btrfs_put_bbio(bbio);
bitmap_or(sparity->ebitmap, sparity->ebitmap, sparity->dbitmap,
sparity->nsectors);
@@ -2824,12 +3054,12 @@ static inline int scrub_calc_parity_bitmap_len(int nsectors)
static void scrub_parity_get(struct scrub_parity *sparity)
{
- atomic_inc(&sparity->refs);
+ refcount_inc(&sparity->refs);
}
static void scrub_parity_put(struct scrub_parity *sparity)
{
- if (!atomic_dec_and_test(&sparity->refs))
+ if (!refcount_dec_and_test(&sparity->refs))
return;
scrub_parity_check_and_repair(sparity);
@@ -2881,7 +3111,7 @@ static noinline_for_stack int scrub_raid56_parity(struct scrub_ctx *sctx,
sparity->scrub_dev = sdev;
sparity->logic_start = logic_start;
sparity->logic_end = logic_end;
- atomic_set(&sparity->refs, 1);
+ refcount_set(&sparity->refs, 1);
INIT_LIST_HEAD(&sparity->spages);
sparity->dbitmap = sparity->bitmap;
sparity->ebitmap = (void *)sparity->bitmap + bitmap_len;
@@ -3052,9 +3282,9 @@ out:
logic_end - logic_start);
scrub_parity_put(sparity);
scrub_submit(sctx);
- mutex_lock(&sctx->wr_ctx.wr_lock);
+ mutex_lock(&sctx->wr_lock);
scrub_wr_submit(sctx);
- mutex_unlock(&sctx->wr_ctx.wr_lock);
+ mutex_unlock(&sctx->wr_lock);
btrfs_release_path(path);
return ret < 0 ? ret : 0;
@@ -3100,7 +3330,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
physical = map->stripes[num].physical;
offset = 0;
- nstripes = div_u64(length, map->stripe_len);
+ nstripes = div64_u64(length, map->stripe_len);
if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
offset = map->stripe_len * num;
increment = map->stripe_len * map->num_stripes;
@@ -3210,14 +3440,14 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
*/
if (atomic_read(&fs_info->scrub_pause_req)) {
/* push queued extents */
- atomic_set(&sctx->wr_ctx.flush_all_writes, 1);
+ atomic_set(&sctx->flush_all_writes, 1);
scrub_submit(sctx);
- mutex_lock(&sctx->wr_ctx.wr_lock);
+ mutex_lock(&sctx->wr_lock);
scrub_wr_submit(sctx);
- mutex_unlock(&sctx->wr_ctx.wr_lock);
+ mutex_unlock(&sctx->wr_lock);
wait_event(sctx->list_wait,
atomic_read(&sctx->bios_in_flight) == 0);
- atomic_set(&sctx->wr_ctx.flush_all_writes, 0);
+ atomic_set(&sctx->flush_all_writes, 0);
scrub_blocked_if_needed(fs_info);
}
@@ -3424,9 +3654,9 @@ skip:
out:
/* push queued extents */
scrub_submit(sctx);
- mutex_lock(&sctx->wr_ctx.wr_lock);
+ mutex_lock(&sctx->wr_lock);
scrub_wr_submit(sctx);
- mutex_unlock(&sctx->wr_ctx.wr_lock);
+ mutex_unlock(&sctx->wr_lock);
blk_finish_plug(&plug);
btrfs_free_path(path);
@@ -3584,7 +3814,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
* -> btrfs_scrub_pause()
*/
scrub_pause_on(fs_info);
- ret = btrfs_inc_block_group_ro(root, cache);
+ ret = btrfs_inc_block_group_ro(fs_info, cache);
if (!ret && is_dev_replace) {
/*
* If we are doing a device replace wait for any tasks
@@ -3606,7 +3836,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
*/
btrfs_wait_block_group_reservations(cache);
btrfs_wait_nocow_writers(cache);
- ret = btrfs_wait_ordered_roots(fs_info, -1,
+ ret = btrfs_wait_ordered_roots(fs_info, U64_MAX,
cache->key.objectid,
cache->key.offset);
if (ret > 0) {
@@ -3663,11 +3893,11 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
* write requests are really completed when bios_in_flight
* changes to 0.
*/
- atomic_set(&sctx->wr_ctx.flush_all_writes, 1);
+ atomic_set(&sctx->flush_all_writes, 1);
scrub_submit(sctx);
- mutex_lock(&sctx->wr_ctx.wr_lock);
+ mutex_lock(&sctx->wr_lock);
scrub_wr_submit(sctx);
- mutex_unlock(&sctx->wr_ctx.wr_lock);
+ mutex_unlock(&sctx->wr_lock);
wait_event(sctx->list_wait,
atomic_read(&sctx->bios_in_flight) == 0);
@@ -3681,7 +3911,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
*/
wait_event(sctx->list_wait,
atomic_read(&sctx->workers_pending) == 0);
- atomic_set(&sctx->wr_ctx.flush_all_writes, 0);
+ atomic_set(&sctx->flush_all_writes, 0);
scrub_pause_off(fs_info);
@@ -4084,34 +4314,6 @@ static void scrub_remap_extent(struct btrfs_fs_info *fs_info,
btrfs_put_bbio(bbio);
}
-static int scrub_setup_wr_ctx(struct scrub_ctx *sctx,
- struct scrub_wr_ctx *wr_ctx,
- struct btrfs_fs_info *fs_info,
- struct btrfs_device *dev,
- int is_dev_replace)
-{
- WARN_ON(wr_ctx->wr_curr_bio != NULL);
-
- mutex_init(&wr_ctx->wr_lock);
- wr_ctx->wr_curr_bio = NULL;
- if (!is_dev_replace)
- return 0;
-
- WARN_ON(!dev->bdev);
- wr_ctx->pages_per_wr_bio = SCRUB_PAGES_PER_WR_BIO;
- wr_ctx->tgtdev = dev;
- atomic_set(&wr_ctx->flush_all_writes, 0);
- return 0;
-}
-
-static void scrub_free_wr_ctx(struct scrub_wr_ctx *wr_ctx)
-{
- mutex_lock(&wr_ctx->wr_lock);
- kfree(wr_ctx->wr_curr_bio);
- wr_ctx->wr_curr_bio = NULL;
- mutex_unlock(&wr_ctx->wr_lock);
-}
-
static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
int mirror_num, u64 physical_for_dev_replace)
{
@@ -4240,7 +4442,7 @@ out:
scrub_pending_trans_workers_dec(sctx);
}
-static int check_extent_to_block(struct inode *inode, u64 start, u64 len,
+static int check_extent_to_block(struct btrfs_inode *inode, u64 start, u64 len,
u64 logical)
{
struct extent_state *cached_state = NULL;
@@ -4250,7 +4452,7 @@ static int check_extent_to_block(struct inode *inode, u64 start, u64 len,
u64 lockstart = start, lockend = start + len - 1;
int ret = 0;
- io_tree = &BTRFS_I(inode)->io_tree;
+ io_tree = &inode->io_tree;
lock_extent_bits(io_tree, lockstart, lockend, &cached_state);
ordered = btrfs_lookup_ordered_range(inode, lockstart, len);
@@ -4329,7 +4531,8 @@ static int copy_nocow_pages_for_inode(u64 inum, u64 offset, u64 root,
io_tree = &BTRFS_I(inode)->io_tree;
nocow_ctx_logical = nocow_ctx->logical;
- ret = check_extent_to_block(inode, offset, len, nocow_ctx_logical);
+ ret = check_extent_to_block(BTRFS_I(inode), offset, len,
+ nocow_ctx_logical);
if (ret) {
ret = ret > 0 ? 0 : ret;
goto out;
@@ -4376,7 +4579,7 @@ again:
}
}
- ret = check_extent_to_block(inode, offset, len,
+ ret = check_extent_to_block(BTRFS_I(inode), offset, len,
nocow_ctx_logical);
if (ret) {
ret = ret > 0 ? 0 : ret;
@@ -4413,7 +4616,7 @@ static int write_page_nocow(struct scrub_ctx *sctx,
struct btrfs_device *dev;
int ret;
- dev = sctx->wr_ctx.tgtdev;
+ dev = sctx->wr_tgtdev;
if (!dev)
return -EIO;
if (!dev->bdev) {
@@ -4421,13 +4624,7 @@ static int write_page_nocow(struct scrub_ctx *sctx,
"scrub write_page_nocow(bdev == NULL) is unexpected");
return -EIO;
}
- bio = btrfs_io_bio_alloc(GFP_NOFS, 1);
- if (!bio) {
- spin_lock(&sctx->stat_lock);
- sctx->stat.malloc_errors++;
- spin_unlock(&sctx->stat_lock);
- return -ENOMEM;
- }
+ bio = btrfs_io_bio_alloc(1);
bio->bi_iter.bi_size = 0;
bio->bi_iter.bi_sector = physical_for_dev_replace >> 9;
bio->bi_bdev = dev->bdev;
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index d145ce804620..b082210df9c8 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -1069,6 +1069,12 @@ static int iterate_dir_item(struct btrfs_root *root, struct btrfs_path *path,
}
}
+ ret = btrfs_is_name_len_valid(eb, path->slots[0],
+ (unsigned long)(di + 1), name_len + data_len);
+ if (!ret) {
+ ret = -EIO;
+ goto out;
+ }
if (name_len + data_len > buf_len) {
buf_len = name_len + data_len;
if (is_vmalloc_addr(buf)) {
@@ -1083,7 +1089,7 @@ static int iterate_dir_item(struct btrfs_root *root, struct btrfs_path *path,
buf = tmp;
}
if (!buf) {
- buf = vmalloc(buf_len);
+ buf = kvmalloc(buf_len, GFP_KERNEL);
if (!buf) {
ret = -ENOMEM;
goto out;
@@ -1681,6 +1687,9 @@ static int is_inode_existent(struct send_ctx *sctx, u64 ino, u64 gen)
{
int ret;
+ if (ino == BTRFS_FIRST_FREE_OBJECTID)
+ return 1;
+
ret = get_cur_inode_state(sctx, ino, gen);
if (ret < 0)
goto out;
@@ -1847,7 +1856,7 @@ out:
*/
static int will_overwrite_ref(struct send_ctx *sctx, u64 dir, u64 dir_gen,
const char *name, int name_len,
- u64 *who_ino, u64 *who_gen)
+ u64 *who_ino, u64 *who_gen, u64 *who_mode)
{
int ret = 0;
u64 gen;
@@ -1866,7 +1875,7 @@ static int will_overwrite_ref(struct send_ctx *sctx, u64 dir, u64 dir_gen,
* not deleted and then re-created, if it was then we have no overwrite
* and we can just unlink this entry.
*/
- if (sctx->parent_root) {
+ if (sctx->parent_root && dir != BTRFS_FIRST_FREE_OBJECTID) {
ret = get_inode_info(sctx->parent_root, dir, NULL, &gen, NULL,
NULL, NULL, NULL);
if (ret < 0 && ret != -ENOENT)
@@ -1896,7 +1905,7 @@ static int will_overwrite_ref(struct send_ctx *sctx, u64 dir, u64 dir_gen,
if (other_inode > sctx->send_progress ||
is_waiting_for_move(sctx, other_inode)) {
ret = get_inode_info(sctx->parent_root, other_inode, NULL,
- who_gen, NULL, NULL, NULL, NULL);
+ who_gen, who_mode, NULL, NULL, NULL);
if (ret < 0)
goto out;
@@ -1934,6 +1943,19 @@ static int did_overwrite_ref(struct send_ctx *sctx,
if (ret <= 0)
goto out;
+ if (dir != BTRFS_FIRST_FREE_OBJECTID) {
+ ret = get_inode_info(sctx->send_root, dir, NULL, &gen, NULL,
+ NULL, NULL, NULL);
+ if (ret < 0 && ret != -ENOENT)
+ goto out;
+ if (ret) {
+ ret = 0;
+ goto out;
+ }
+ if (gen != dir_gen)
+ goto out;
+ }
+
/* check if the ref was overwritten by another ref */
ret = lookup_dir_item_inode(sctx->send_root, dir, name, name_len,
&ow_inode, &other_type);
@@ -2753,15 +2775,20 @@ out:
struct recorded_ref {
struct list_head list;
- char *dir_path;
char *name;
struct fs_path *full_path;
u64 dir;
u64 dir_gen;
- int dir_path_len;
int name_len;
};
+static void set_ref_path(struct recorded_ref *ref, struct fs_path *path)
+{
+ ref->full_path = path;
+ ref->name = (char *)kbasename(ref->full_path->start);
+ ref->name_len = ref->full_path->end - ref->name;
+}
+
/*
* We need to process new refs before deleted refs, but compare_tree gives us
* everything mixed. So we first record all refs and later process them.
@@ -2778,17 +2805,7 @@ static int __record_ref(struct list_head *head, u64 dir,
ref->dir = dir;
ref->dir_gen = dir_gen;
- ref->full_path = path;
-
- ref->name = (char *)kbasename(ref->full_path->start);
- ref->name_len = ref->full_path->end - ref->name;
- ref->dir_path = ref->full_path->start;
- if (ref->name == ref->full_path->start)
- ref->dir_path_len = 0;
- else
- ref->dir_path_len = ref->full_path->end -
- ref->full_path->start - 1 - ref->name_len;
-
+ set_ref_path(ref, path);
list_add_tail(&ref->list, head);
return 0;
}
@@ -3530,9 +3547,17 @@ static int is_ancestor(struct btrfs_root *root,
struct fs_path *fs_path)
{
u64 ino = ino2;
+ bool free_path = false;
+ int ret = 0;
+
+ if (!fs_path) {
+ fs_path = fs_path_alloc();
+ if (!fs_path)
+ return -ENOMEM;
+ free_path = true;
+ }
while (ino > BTRFS_FIRST_FREE_OBJECTID) {
- int ret;
u64 parent;
u64 parent_gen;
@@ -3541,13 +3566,18 @@ static int is_ancestor(struct btrfs_root *root,
if (ret < 0) {
if (ret == -ENOENT && ino == ino2)
ret = 0;
- return ret;
+ goto out;
+ }
+ if (parent == ino1) {
+ ret = parent_gen == ino1_gen ? 1 : 0;
+ goto out;
}
- if (parent == ino1)
- return parent_gen == ino1_gen ? 1 : 0;
ino = parent;
}
- return 0;
+ out:
+ if (free_path)
+ fs_path_free(fs_path);
+ return ret;
}
static int wait_for_parent_move(struct send_ctx *sctx,
@@ -3556,6 +3586,7 @@ static int wait_for_parent_move(struct send_ctx *sctx,
{
int ret = 0;
u64 ino = parent_ref->dir;
+ u64 ino_gen = parent_ref->dir_gen;
u64 parent_ino_before, parent_ino_after;
struct fs_path *path_before = NULL;
struct fs_path *path_after = NULL;
@@ -3576,6 +3607,8 @@ static int wait_for_parent_move(struct send_ctx *sctx,
* at get_cur_path()).
*/
while (ino > BTRFS_FIRST_FREE_OBJECTID) {
+ u64 parent_ino_after_gen;
+
if (is_waiting_for_move(sctx, ino)) {
/*
* If the current inode is an ancestor of ino in the
@@ -3598,7 +3631,7 @@ static int wait_for_parent_move(struct send_ctx *sctx,
fs_path_reset(path_after);
ret = get_first_ref(sctx->send_root, ino, &parent_ino_after,
- NULL, path_after);
+ &parent_ino_after_gen, path_after);
if (ret < 0)
goto out;
ret = get_first_ref(sctx->parent_root, ino, &parent_ino_before,
@@ -3615,10 +3648,20 @@ static int wait_for_parent_move(struct send_ctx *sctx,
if (ino > sctx->cur_ino &&
(parent_ino_before != parent_ino_after || len1 != len2 ||
memcmp(path_before->start, path_after->start, len1))) {
- ret = 1;
- break;
+ u64 parent_ino_gen;
+
+ ret = get_inode_info(sctx->parent_root, ino, NULL,
+ &parent_ino_gen, NULL, NULL, NULL,
+ NULL);
+ if (ret < 0)
+ goto out;
+ if (ino_gen == parent_ino_gen) {
+ ret = 1;
+ break;
+ }
}
ino = parent_ino_after;
+ ino_gen = parent_ino_after_gen;
}
out:
@@ -3640,6 +3683,36 @@ out:
return ret;
}
+static int update_ref_path(struct send_ctx *sctx, struct recorded_ref *ref)
+{
+ int ret;
+ struct fs_path *new_path;
+
+ /*
+ * Our reference's name member points to its full_path member string, so
+ * we use here a new path.
+ */
+ new_path = fs_path_alloc();
+ if (!new_path)
+ return -ENOMEM;
+
+ ret = get_cur_path(sctx, ref->dir, ref->dir_gen, new_path);
+ if (ret < 0) {
+ fs_path_free(new_path);
+ return ret;
+ }
+ ret = fs_path_add(new_path, ref->name, ref->name_len);
+ if (ret < 0) {
+ fs_path_free(new_path);
+ return ret;
+ }
+
+ fs_path_free(ref->full_path);
+ set_ref_path(ref, new_path);
+
+ return 0;
+}
+
/*
* This does all the move/link/unlink/rmdir magic.
*/
@@ -3653,10 +3726,13 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
struct fs_path *valid_path = NULL;
u64 ow_inode = 0;
u64 ow_gen;
+ u64 ow_mode;
int did_overwrite = 0;
int is_orphan = 0;
u64 last_dir_ino_rm = 0;
bool can_rename = true;
+ bool orphanized_dir = false;
+ bool orphanized_ancestor = false;
btrfs_debug(fs_info, "process_recorded_refs %llu", sctx->cur_ino);
@@ -3754,7 +3830,7 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
*/
ret = will_overwrite_ref(sctx, cur->dir, cur->dir_gen,
cur->name, cur->name_len,
- &ow_inode, &ow_gen);
+ &ow_inode, &ow_gen, &ow_mode);
if (ret < 0)
goto out;
if (ret) {
@@ -3771,6 +3847,8 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
cur->full_path);
if (ret < 0)
goto out;
+ if (S_ISDIR(ow_mode))
+ orphanized_dir = true;
/*
* If ow_inode has its rename operation delayed
@@ -3808,9 +3886,16 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
* might contain the pre-orphanization name of
* ow_inode, which is no longer valid.
*/
- fs_path_reset(valid_path);
- ret = get_cur_path(sctx, sctx->cur_ino,
- sctx->cur_inode_gen, valid_path);
+ ret = is_ancestor(sctx->parent_root,
+ ow_inode, ow_gen,
+ sctx->cur_ino, NULL);
+ if (ret > 0) {
+ orphanized_ancestor = true;
+ fs_path_reset(valid_path);
+ ret = get_cur_path(sctx, sctx->cur_ino,
+ sctx->cur_inode_gen,
+ valid_path);
+ }
if (ret < 0)
goto out;
} else {
@@ -3869,6 +3954,18 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
if (ret < 0)
goto out;
} else {
+ /*
+ * We might have previously orphanized an inode
+ * which is an ancestor of our current inode,
+ * so our reference's full path, which was
+ * computed before any such orphanizations, must
+ * be updated.
+ */
+ if (orphanized_dir) {
+ ret = update_ref_path(sctx, cur);
+ if (ret < 0)
+ goto out;
+ }
ret = send_link(sctx, cur->full_path,
valid_path);
if (ret < 0)
@@ -3931,6 +4028,18 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
if (ret < 0)
goto out;
if (!ret) {
+ /*
+ * If we orphanized any ancestor before, we need
+ * to recompute the full path for deleted names,
+ * since any such path was computed before we
+ * processed any references and orphanized any
+ * ancestor inode.
+ */
+ if (orphanized_ancestor) {
+ ret = update_ref_path(sctx, cur);
+ if (ret < 0)
+ goto out;
+ }
ret = send_unlink(sctx, cur->full_path);
if (ret < 0)
goto out;
@@ -5155,15 +5264,18 @@ static int is_extent_unchanged(struct send_ctx *sctx,
while (key.offset < ekey->offset + left_len) {
ei = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item);
right_type = btrfs_file_extent_type(eb, ei);
- if (right_type != BTRFS_FILE_EXTENT_REG) {
+ if (right_type != BTRFS_FILE_EXTENT_REG &&
+ right_type != BTRFS_FILE_EXTENT_INLINE) {
ret = 0;
goto out;
}
- right_disknr = btrfs_file_extent_disk_bytenr(eb, ei);
- right_len = btrfs_file_extent_num_bytes(eb, ei);
- right_offset = btrfs_file_extent_offset(eb, ei);
- right_gen = btrfs_file_extent_generation(eb, ei);
+ if (right_type == BTRFS_FILE_EXTENT_INLINE) {
+ right_len = btrfs_file_extent_inline_len(eb, slot, ei);
+ right_len = PAGE_ALIGN(right_len);
+ } else {
+ right_len = btrfs_file_extent_num_bytes(eb, ei);
+ }
/*
* Are we at extent 8? If yes, we know the extent is changed.
@@ -5175,6 +5287,23 @@ static int is_extent_unchanged(struct send_ctx *sctx,
goto out;
}
+ /*
+ * We just wanted to see if when we have an inline extent, what
+ * follows it is a regular extent (wanted to check the above
+ * condition for inline extents too). This should normally not
+ * happen but it's possible for example when we have an inline
+ * compressed extent representing data with a size matching
+ * the page size (currently the same as sector size).
+ */
+ if (right_type == BTRFS_FILE_EXTENT_INLINE) {
+ ret = 0;
+ goto out;
+ }
+
+ right_disknr = btrfs_file_extent_disk_bytenr(eb, ei);
+ right_offset = btrfs_file_extent_offset(eb, ei);
+ right_gen = btrfs_file_extent_generation(eb, ei);
+
left_offset_fixed = left_offset;
if (key.offset < ekey->offset) {
/* Fix the right offset for 2a and 7. */
@@ -5277,6 +5406,81 @@ out:
return ret;
}
+static int range_is_hole_in_parent(struct send_ctx *sctx,
+ const u64 start,
+ const u64 end)
+{
+ struct btrfs_path *path;
+ struct btrfs_key key;
+ struct btrfs_root *root = sctx->parent_root;
+ u64 search_start = start;
+ int ret;
+
+ path = alloc_path_for_send();
+ if (!path)
+ return -ENOMEM;
+
+ key.objectid = sctx->cur_ino;
+ key.type = BTRFS_EXTENT_DATA_KEY;
+ key.offset = search_start;
+ ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+ if (ret < 0)
+ goto out;
+ if (ret > 0 && path->slots[0] > 0)
+ path->slots[0]--;
+
+ while (search_start < end) {
+ struct extent_buffer *leaf = path->nodes[0];
+ int slot = path->slots[0];
+ struct btrfs_file_extent_item *fi;
+ u64 extent_end;
+
+ if (slot >= btrfs_header_nritems(leaf)) {
+ ret = btrfs_next_leaf(root, path);
+ if (ret < 0)
+ goto out;
+ else if (ret > 0)
+ break;
+ continue;
+ }
+
+ btrfs_item_key_to_cpu(leaf, &key, slot);
+ if (key.objectid < sctx->cur_ino ||
+ key.type < BTRFS_EXTENT_DATA_KEY)
+ goto next;
+ if (key.objectid > sctx->cur_ino ||
+ key.type > BTRFS_EXTENT_DATA_KEY ||
+ key.offset >= end)
+ break;
+
+ fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
+ if (btrfs_file_extent_type(leaf, fi) ==
+ BTRFS_FILE_EXTENT_INLINE) {
+ u64 size = btrfs_file_extent_inline_len(leaf, slot, fi);
+
+ extent_end = ALIGN(key.offset + size,
+ root->fs_info->sectorsize);
+ } else {
+ extent_end = key.offset +
+ btrfs_file_extent_num_bytes(leaf, fi);
+ }
+ if (extent_end <= start)
+ goto next;
+ if (btrfs_file_extent_disk_bytenr(leaf, fi) == 0) {
+ search_start = extent_end;
+ goto next;
+ }
+ ret = 0;
+ goto out;
+next:
+ path->slots[0]++;
+ }
+ ret = 1;
+out:
+ btrfs_free_path(path);
+ return ret;
+}
+
static int maybe_send_hole(struct send_ctx *sctx, struct btrfs_path *path,
struct btrfs_key *key)
{
@@ -5321,8 +5525,17 @@ static int maybe_send_hole(struct send_ctx *sctx, struct btrfs_path *path,
return ret;
}
- if (sctx->cur_inode_last_extent < key->offset)
- ret = send_hole(sctx, key->offset);
+ if (sctx->cur_inode_last_extent < key->offset) {
+ ret = range_is_hole_in_parent(sctx,
+ sctx->cur_inode_last_extent,
+ key->offset);
+ if (ret < 0)
+ return ret;
+ else if (ret == 0)
+ ret = send_hole(sctx, key->offset);
+ else
+ ret = 0;
+ }
sctx->cur_inode_last_extent = extent_end;
return ret;
}
@@ -6192,8 +6405,13 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
goto out;
}
+ /*
+ * Check that we don't overflow at later allocations, we request
+ * clone_sources_count + 1 items, and compare to unsigned long inside
+ * access_ok.
+ */
if (arg->clone_sources_count >
- ULLONG_MAX / sizeof(*arg->clone_sources)) {
+ ULONG_MAX / sizeof(struct clone_root) - 1) {
ret = -EINVAL;
goto out;
}
@@ -6242,22 +6460,16 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
sctx->clone_roots_cnt = arg->clone_sources_count;
sctx->send_max_size = BTRFS_SEND_BUF_SIZE;
- sctx->send_buf = kmalloc(sctx->send_max_size, GFP_KERNEL | __GFP_NOWARN);
+ sctx->send_buf = kvmalloc(sctx->send_max_size, GFP_KERNEL);
if (!sctx->send_buf) {
- sctx->send_buf = vmalloc(sctx->send_max_size);
- if (!sctx->send_buf) {
- ret = -ENOMEM;
- goto out;
- }
+ ret = -ENOMEM;
+ goto out;
}
- sctx->read_buf = kmalloc(BTRFS_SEND_READ_SIZE, GFP_KERNEL | __GFP_NOWARN);
+ sctx->read_buf = kvmalloc(BTRFS_SEND_READ_SIZE, GFP_KERNEL);
if (!sctx->read_buf) {
- sctx->read_buf = vmalloc(BTRFS_SEND_READ_SIZE);
- if (!sctx->read_buf) {
- ret = -ENOMEM;
- goto out;
- }
+ ret = -ENOMEM;
+ goto out;
}
sctx->pending_dir_moves = RB_ROOT;
@@ -6266,25 +6478,19 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
alloc_size = sizeof(struct clone_root) * (arg->clone_sources_count + 1);
- sctx->clone_roots = kzalloc(alloc_size, GFP_KERNEL | __GFP_NOWARN);
+ sctx->clone_roots = kzalloc(alloc_size, GFP_KERNEL);
if (!sctx->clone_roots) {
- sctx->clone_roots = vzalloc(alloc_size);
- if (!sctx->clone_roots) {
- ret = -ENOMEM;
- goto out;
- }
+ ret = -ENOMEM;
+ goto out;
}
alloc_size = arg->clone_sources_count * sizeof(*arg->clone_sources);
if (arg->clone_sources_count) {
- clone_sources_tmp = kmalloc(alloc_size, GFP_KERNEL | __GFP_NOWARN);
+ clone_sources_tmp = kvmalloc(alloc_size, GFP_KERNEL);
if (!clone_sources_tmp) {
- clone_sources_tmp = vmalloc(alloc_size);
- if (!clone_sources_tmp) {
- ret = -ENOMEM;
- goto out;
- }
+ ret = -ENOMEM;
+ goto out;
}
ret = copy_from_user(clone_sources_tmp, arg->clone_sources,
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index b5ae7d3d1896..12540b6104b5 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -265,7 +265,7 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
function, line, errstr);
return;
}
- ACCESS_ONCE(trans->transaction->aborted) = errno;
+ WRITE_ONCE(trans->transaction->aborted, errno);
/* Wake up anybody who may be waiting on this transaction */
wake_up(&fs_info->transaction_wait);
wake_up(&fs_info->transaction_blocked_wait);
@@ -549,16 +549,19 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
case Opt_ssd:
btrfs_set_and_info(info, SSD,
"use ssd allocation scheme");
+ btrfs_clear_opt(info->mount_opt, NOSSD);
break;
case Opt_ssd_spread:
btrfs_set_and_info(info, SSD_SPREAD,
"use spread ssd allocation scheme");
btrfs_set_opt(info->mount_opt, SSD);
+ btrfs_clear_opt(info->mount_opt, NOSSD);
break;
case Opt_nossd:
btrfs_set_and_info(info, NOSSD,
"not using ssd allocation scheme");
btrfs_clear_opt(info->mount_opt, SSD);
+ btrfs_clear_opt(info->mount_opt, SSD_SPREAD);
break;
case Opt_barrier:
btrfs_clear_and_info(info, NOBARRIER,
@@ -598,18 +601,8 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
}
break;
case Opt_alloc_start:
- num = match_strdup(&args[0]);
- if (num) {
- mutex_lock(&info->chunk_mutex);
- info->alloc_start = memparse(num, NULL);
- mutex_unlock(&info->chunk_mutex);
- kfree(num);
- btrfs_info(info, "allocations start at %llu",
- info->alloc_start);
- } else {
- ret = -ENOMEM;
- goto out;
- }
+ btrfs_info(info,
+ "option alloc_start is obsolete, ignored");
break;
case Opt_acl:
#ifdef CONFIG_BTRFS_FS_POSIX_ACL
@@ -1114,7 +1107,7 @@ static int get_default_subvol_objectid(struct btrfs_fs_info *fs_info, u64 *objec
static int btrfs_fill_super(struct super_block *sb,
struct btrfs_fs_devices *fs_devices,
- void *data, int silent)
+ void *data)
{
struct inode *inode;
struct btrfs_fs_info *fs_info = btrfs_sb(sb);
@@ -1133,6 +1126,13 @@ static int btrfs_fill_super(struct super_block *sb,
#endif
sb->s_flags |= MS_I_VERSION;
sb->s_iflags |= SB_I_CGROUPWB;
+
+ err = super_setup_bdi(sb);
+ if (err) {
+ btrfs_err(fs_info, "super_setup_bdi failed");
+ return err;
+ }
+
err = open_ctree(sb, fs_devices, (char *)data);
if (err) {
btrfs_err(fs_info, "open_ctree failed");
@@ -1154,7 +1154,6 @@ static int btrfs_fill_super(struct super_block *sb,
goto fail_close;
}
- save_mount_options(sb, data);
cleancache_init_fs(sb);
sb->s_flags |= MS_ACTIVE;
return 0;
@@ -1177,7 +1176,7 @@ int btrfs_sync_fs(struct super_block *sb, int wait)
return 0;
}
- btrfs_wait_ordered_roots(fs_info, -1, 0, (u64)-1);
+ btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
trans = btrfs_attach_transaction_barrier(root);
if (IS_ERR(trans)) {
@@ -1222,8 +1221,6 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
seq_puts(seq, ",nobarrier");
if (info->max_inline != BTRFS_DEFAULT_MAX_INLINE)
seq_printf(seq, ",max_inline=%llu", info->max_inline);
- if (info->alloc_start != 0)
- seq_printf(seq, ",alloc_start=%llu", info->alloc_start);
if (info->thread_pool_size != min_t(unsigned long,
num_online_cpus() + 2, 8))
seq_printf(seq, ",thread_pool=%d", info->thread_pool_size);
@@ -1611,8 +1608,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
} else {
snprintf(s->s_id, sizeof(s->s_id), "%pg", bdev);
btrfs_sb(s)->bdev_holder = fs_type;
- error = btrfs_fill_super(s, fs_devices, data,
- flags & MS_SILENT ? 1 : 0);
+ error = btrfs_fill_super(s, fs_devices, data);
}
if (error) {
deactivate_locked_super(s);
@@ -1707,7 +1703,6 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
unsigned long old_opts = fs_info->mount_opt;
unsigned long old_compress_type = fs_info->compress_type;
u64 old_max_inline = fs_info->max_inline;
- u64 old_alloc_start = fs_info->alloc_start;
int old_thread_pool_size = fs_info->thread_pool_size;
unsigned int old_metadata_ratio = fs_info->metadata_ratio;
int ret;
@@ -1786,8 +1781,7 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
}
if (fs_info->fs_devices->missing_devices >
- fs_info->num_tolerated_disk_barrier_failures &&
- !(*flags & MS_RDONLY)) {
+ fs_info->num_tolerated_disk_barrier_failures) {
btrfs_warn(fs_info,
"too many missing devices, writeable remount is not allowed");
ret = -EACCES;
@@ -1847,9 +1841,6 @@ restore:
fs_info->mount_opt = old_opts;
fs_info->compress_type = old_compress_type;
fs_info->max_inline = old_max_inline;
- mutex_lock(&fs_info->chunk_mutex);
- fs_info->alloc_start = old_alloc_start;
- mutex_unlock(&fs_info->chunk_mutex);
btrfs_resize_thread_pool(fs_info,
old_thread_pool_size, fs_info->thread_pool_size);
fs_info->metadata_ratio = old_metadata_ratio;
@@ -1890,18 +1881,15 @@ static inline void btrfs_descending_sort_devices(
static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
u64 *free_bytes)
{
- struct btrfs_root *root = fs_info->tree_root;
struct btrfs_device_info *devices_info;
struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
struct btrfs_device *device;
u64 skip_space;
u64 type;
u64 avail_space;
- u64 used_space;
u64 min_stripe_size;
int min_stripes = 1, num_stripes = 1;
int i = 0, nr_devices;
- int ret;
/*
* We aren't under the device list lock, so this is racy-ish, but good
@@ -1919,12 +1907,12 @@ static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
}
devices_info = kmalloc_array(nr_devices, sizeof(*devices_info),
- GFP_NOFS);
+ GFP_KERNEL);
if (!devices_info)
return -ENOMEM;
/* calc min stripe number for data space allocation */
- type = btrfs_get_alloc_profile(root, 1);
+ type = btrfs_data_alloc_profile(fs_info);
if (type & BTRFS_BLOCK_GROUP_RAID0) {
min_stripes = 2;
num_stripes = nr_devices;
@@ -1941,8 +1929,6 @@ static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
else
min_stripe_size = BTRFS_STRIPE_LEN;
- if (fs_info->alloc_start)
- mutex_lock(&fs_devices->device_list_mutex);
rcu_read_lock();
list_for_each_entry_rcu(device, &fs_devices->devices, dev_list) {
if (!device->in_fs_metadata || !device->bdev ||
@@ -1965,34 +1951,6 @@ static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
*/
skip_space = SZ_1M;
- /* user can set the offset in fs_info->alloc_start. */
- if (fs_info->alloc_start &&
- fs_info->alloc_start + BTRFS_STRIPE_LEN <=
- device->total_bytes) {
- rcu_read_unlock();
- skip_space = max(fs_info->alloc_start, skip_space);
-
- /*
- * btrfs can not use the free space in
- * [0, skip_space - 1], we must subtract it from the
- * total. In order to implement it, we account the used
- * space in this range first.
- */
- ret = btrfs_account_dev_extents_size(device, 0,
- skip_space - 1,
- &used_space);
- if (ret) {
- kfree(devices_info);
- mutex_unlock(&fs_devices->device_list_mutex);
- return ret;
- }
-
- rcu_read_lock();
-
- /* calc the free space in [0, skip_space - 1] */
- skip_space -= used_space;
- }
-
/*
* we can use the free space in [0, skip_space - 1], subtract
* it from the total.
@@ -2011,8 +1969,6 @@ static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
i++;
}
rcu_read_unlock();
- if (fs_info->alloc_start)
- mutex_unlock(&fs_devices->device_list_mutex);
nr_devices = i;
@@ -2049,10 +2005,9 @@ static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
* multiplier to scale the sizes.
*
* Unused device space usage is based on simulating the chunk allocator
- * algorithm that respects the device sizes, order of allocations and the
- * 'alloc_start' value, this is a close approximation of the actual use but
- * there are other factors that may change the result (like a new metadata
- * chunk).
+ * algorithm that respects the device sizes and order of allocations. This is
+ * a close approximation of the actual use but there are other factors that may
+ * change the result (like a new metadata chunk).
*
* If metadata is exhausted, f_bavail will be 0.
*/
@@ -2235,7 +2190,7 @@ static int btrfs_freeze(struct super_block *sb)
struct btrfs_fs_info *fs_info = btrfs_sb(sb);
struct btrfs_root *root = fs_info->tree_root;
- fs_info->fs_frozen = 1;
+ set_bit(BTRFS_FS_FROZEN, &fs_info->flags);
/*
* We don't need a barrier here, we'll wait for any transaction that
* could be in progress on other threads (and do delayed iputs that
@@ -2254,7 +2209,9 @@ static int btrfs_freeze(struct super_block *sb)
static int btrfs_unfreeze(struct super_block *sb)
{
- btrfs_sb(sb)->fs_frozen = 0;
+ struct btrfs_fs_info *fs_info = btrfs_sb(sb);
+
+ clear_bit(BTRFS_FS_FROZEN, &fs_info->flags);
return 0;
}
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index 1f157fba8940..c2d5f3580b4c 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -447,11 +447,52 @@ static ssize_t btrfs_clone_alignment_show(struct kobject *kobj,
BTRFS_ATTR(clone_alignment, btrfs_clone_alignment_show);
+static ssize_t quota_override_show(struct kobject *kobj,
+ struct kobj_attribute *a, char *buf)
+{
+ struct btrfs_fs_info *fs_info = to_fs_info(kobj);
+ int quota_override;
+
+ quota_override = test_bit(BTRFS_FS_QUOTA_OVERRIDE, &fs_info->flags);
+ return snprintf(buf, PAGE_SIZE, "%d\n", quota_override);
+}
+
+static ssize_t quota_override_store(struct kobject *kobj,
+ struct kobj_attribute *a,
+ const char *buf, size_t len)
+{
+ struct btrfs_fs_info *fs_info = to_fs_info(kobj);
+ unsigned long knob;
+ int err;
+
+ if (!fs_info)
+ return -EPERM;
+
+ if (!capable(CAP_SYS_RESOURCE))
+ return -EPERM;
+
+ err = kstrtoul(buf, 10, &knob);
+ if (err)
+ return err;
+ if (knob > 1)
+ return -EINVAL;
+
+ if (knob)
+ set_bit(BTRFS_FS_QUOTA_OVERRIDE, &fs_info->flags);
+ else
+ clear_bit(BTRFS_FS_QUOTA_OVERRIDE, &fs_info->flags);
+
+ return len;
+}
+
+BTRFS_ATTR_RW(quota_override, quota_override_show, quota_override_store);
+
static const struct attribute *btrfs_attrs[] = {
BTRFS_ATTR_PTR(label),
BTRFS_ATTR_PTR(nodesize),
BTRFS_ATTR_PTR(sectorsize),
BTRFS_ATTR_PTR(clone_alignment),
+ BTRFS_ATTR_PTR(quota_override),
NULL,
};
diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c
index ea272432c930..b18ab8f327a5 100644
--- a/fs/btrfs/tests/btrfs-tests.c
+++ b/fs/btrfs/tests/btrfs-tests.c
@@ -237,7 +237,6 @@ void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans)
{
memset(trans, 0, sizeof(*trans));
trans->transid = 1;
- INIT_LIST_HEAD(&trans->qgroup_ref_list);
trans->type = __TRANS_DUMMY;
}
diff --git a/fs/btrfs/tests/extent-io-tests.c b/fs/btrfs/tests/extent-io-tests.c
index 133753232a94..d06b1c931d05 100644
--- a/fs/btrfs/tests/extent-io-tests.c
+++ b/fs/btrfs/tests/extent-io-tests.c
@@ -87,7 +87,7 @@ static int test_find_delalloc(u32 sectorsize)
return -ENOMEM;
}
- extent_io_tree_init(&tmp, &inode->i_data);
+ extent_io_tree_init(&tmp, inode);
/*
* First go through and create and mark all of our pages dirty, we pin
diff --git a/fs/btrfs/tests/inode-tests.c b/fs/btrfs/tests/inode-tests.c
index 4d0f038e14f1..8c91d03cc82d 100644
--- a/fs/btrfs/tests/inode-tests.c
+++ b/fs/btrfs/tests/inode-tests.c
@@ -278,7 +278,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
/* First with no extents */
BTRFS_I(inode)->root = root;
- em = btrfs_get_extent(inode, NULL, 0, 0, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, sectorsize, 0);
if (IS_ERR(em)) {
em = NULL;
test_msg("Got an error when we shouldn't have\n");
@@ -293,7 +293,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
goto out;
}
free_extent_map(em);
- btrfs_drop_extent_cache(inode, 0, (u64)-1, 0);
+ btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0);
/*
* All of the magic numbers are based on the mapping setup in
@@ -302,7 +302,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
*/
setup_file_extents(root, sectorsize);
- em = btrfs_get_extent(inode, NULL, 0, 0, (u64)-1, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, (u64)-1, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -323,7 +323,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -350,7 +350,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -372,7 +372,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
free_extent_map(em);
/* Regular extent */
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -399,7 +399,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
free_extent_map(em);
/* The next 3 are split extents */
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -428,7 +428,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -450,7 +450,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -484,7 +484,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
free_extent_map(em);
/* Prealloc extent */
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -513,7 +513,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
free_extent_map(em);
/* The next 3 are a half written prealloc extent */
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -543,7 +543,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -576,7 +576,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -611,7 +611,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
free_extent_map(em);
/* Now for the compressed extent */
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -645,7 +645,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
free_extent_map(em);
/* Split compressed extent */
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -680,7 +680,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -707,7 +707,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -742,7 +742,8 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
free_extent_map(em);
/* A hole between regular extents but no hole extent */
- em = btrfs_get_extent(inode, NULL, 0, offset + 6, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset + 6,
+ sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -769,7 +770,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, 4096 * 1024, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, 4096 * 1024, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -802,7 +803,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
offset = em->start + em->len;
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -885,7 +886,7 @@ static int test_hole_first(u32 sectorsize, u32 nodesize)
insert_inode_item_key(root);
insert_extent(root, sectorsize, sectorsize, sectorsize, 0, sectorsize,
sectorsize, BTRFS_FILE_EXTENT_REG, 0, 1);
- em = btrfs_get_extent(inode, NULL, 0, 0, 2 * sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, 2 * sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
@@ -907,7 +908,8 @@ static int test_hole_first(u32 sectorsize, u32 nodesize)
}
free_extent_map(em);
- em = btrfs_get_extent(inode, NULL, 0, sectorsize, 2 * sectorsize, 0);
+ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, sectorsize,
+ 2 * sectorsize, 0);
if (IS_ERR(em)) {
test_msg("Got an error when we shouldn't have\n");
goto out;
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 0e0508f488b2..f615d59b0489 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -60,8 +60,8 @@ static const unsigned int btrfs_blocked_trans_types[TRANS_STATE_MAX] = {
void btrfs_put_transaction(struct btrfs_transaction *transaction)
{
- WARN_ON(atomic_read(&transaction->use_count) == 0);
- if (atomic_dec_and_test(&transaction->use_count)) {
+ WARN_ON(refcount_read(&transaction->use_count) == 0);
+ if (refcount_dec_and_test(&transaction->use_count)) {
BUG_ON(!list_empty(&transaction->list));
WARN_ON(!RB_EMPTY_ROOT(&transaction->delayed_refs.href_root));
if (transaction->delayed_refs.pending_csums)
@@ -93,7 +93,7 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction)
btrfs_put_block_group_trimming(cache);
btrfs_put_block_group(cache);
}
- kmem_cache_free(btrfs_transaction_cachep, transaction);
+ kfree(transaction);
}
}
@@ -207,7 +207,7 @@ loop:
spin_unlock(&fs_info->trans_lock);
return -EBUSY;
}
- atomic_inc(&cur_trans->use_count);
+ refcount_inc(&cur_trans->use_count);
atomic_inc(&cur_trans->num_writers);
extwriter_counter_inc(cur_trans, type);
spin_unlock(&fs_info->trans_lock);
@@ -228,7 +228,7 @@ loop:
*/
BUG_ON(type == TRANS_JOIN_NOLOCK);
- cur_trans = kmem_cache_alloc(btrfs_transaction_cachep, GFP_NOFS);
+ cur_trans = kmalloc(sizeof(*cur_trans), GFP_NOFS);
if (!cur_trans)
return -ENOMEM;
@@ -238,11 +238,11 @@ loop:
* someone started a transaction after we unlocked. Make sure
* to redo the checks above
*/
- kmem_cache_free(btrfs_transaction_cachep, cur_trans);
+ kfree(cur_trans);
goto loop;
} else if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
spin_unlock(&fs_info->trans_lock);
- kmem_cache_free(btrfs_transaction_cachep, cur_trans);
+ kfree(cur_trans);
return -EROFS;
}
@@ -257,7 +257,7 @@ loop:
* One for this trans handle, one so it will live on until we
* commit the transaction.
*/
- atomic_set(&cur_trans->use_count, 2);
+ refcount_set(&cur_trans->use_count, 2);
atomic_set(&cur_trans->pending_ordered, 0);
cur_trans->flags = 0;
cur_trans->start_time = get_seconds();
@@ -294,7 +294,7 @@ loop:
spin_lock_init(&cur_trans->dropped_roots_lock);
list_add_tail(&cur_trans->list, &fs_info->trans_list);
extent_io_tree_init(&cur_trans->dirty_pages,
- fs_info->btree_inode->i_mapping);
+ fs_info->btree_inode);
fs_info->generation++;
cur_trans->transid = fs_info->generation;
fs_info->running_transaction = cur_trans;
@@ -432,7 +432,7 @@ static void wait_current_trans(struct btrfs_fs_info *fs_info)
spin_lock(&fs_info->trans_lock);
cur_trans = fs_info->running_transaction;
if (cur_trans && is_transaction_blocked(cur_trans)) {
- atomic_inc(&cur_trans->use_count);
+ refcount_inc(&cur_trans->use_count);
spin_unlock(&fs_info->trans_lock);
wait_event(fs_info->transaction_wait,
@@ -474,7 +474,8 @@ static inline bool need_reserve_reloc_root(struct btrfs_root *root)
static struct btrfs_trans_handle *
start_transaction(struct btrfs_root *root, unsigned int num_items,
- unsigned int type, enum btrfs_reserve_flush_enum flush)
+ unsigned int type, enum btrfs_reserve_flush_enum flush,
+ bool enforce_qgroups)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -505,9 +506,10 @@ start_transaction(struct btrfs_root *root, unsigned int num_items,
* Do the reservation before we join the transaction so we can do all
* the appropriate flushing if need be.
*/
- if (num_items > 0 && root != fs_info->chunk_root) {
+ if (num_items && root != fs_info->chunk_root) {
qgroup_reserved = num_items * fs_info->nodesize;
- ret = btrfs_qgroup_reserve_meta(root, qgroup_reserved);
+ ret = btrfs_qgroup_reserve_meta(root, qgroup_reserved,
+ enforce_qgroups);
if (ret)
return ERR_PTR(ret);
@@ -570,7 +572,6 @@ again:
h->type = type;
h->can_flush_pending_bgs = true;
- INIT_LIST_HEAD(&h->qgroup_ref_list);
INIT_LIST_HEAD(&h->new_bgs);
smp_mb();
@@ -613,8 +614,9 @@ struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root,
unsigned int num_items)
{
return start_transaction(root, num_items, TRANS_START,
- BTRFS_RESERVE_FLUSH_ALL);
+ BTRFS_RESERVE_FLUSH_ALL, true);
}
+
struct btrfs_trans_handle *btrfs_start_transaction_fallback_global_rsv(
struct btrfs_root *root,
unsigned int num_items,
@@ -625,7 +627,14 @@ struct btrfs_trans_handle *btrfs_start_transaction_fallback_global_rsv(
u64 num_bytes;
int ret;
- trans = btrfs_start_transaction(root, num_items);
+ /*
+ * We have two callers: unlink and block group removal. The
+ * former should succeed even if we will temporarily exceed
+ * quota and the latter operates on the extent root so
+ * qgroup enforcement is ignored anyway.
+ */
+ trans = start_transaction(root, num_items, TRANS_START,
+ BTRFS_RESERVE_FLUSH_ALL, false);
if (!IS_ERR(trans) || PTR_ERR(trans) != -ENOSPC)
return trans;
@@ -654,25 +663,25 @@ struct btrfs_trans_handle *btrfs_start_transaction_lflush(
unsigned int num_items)
{
return start_transaction(root, num_items, TRANS_START,
- BTRFS_RESERVE_FLUSH_LIMIT);
+ BTRFS_RESERVE_FLUSH_LIMIT, true);
}
struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root)
{
- return start_transaction(root, 0, TRANS_JOIN,
- BTRFS_RESERVE_NO_FLUSH);
+ return start_transaction(root, 0, TRANS_JOIN, BTRFS_RESERVE_NO_FLUSH,
+ true);
}
struct btrfs_trans_handle *btrfs_join_transaction_nolock(struct btrfs_root *root)
{
return start_transaction(root, 0, TRANS_JOIN_NOLOCK,
- BTRFS_RESERVE_NO_FLUSH);
+ BTRFS_RESERVE_NO_FLUSH, true);
}
struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *root)
{
return start_transaction(root, 0, TRANS_USERSPACE,
- BTRFS_RESERVE_NO_FLUSH);
+ BTRFS_RESERVE_NO_FLUSH, true);
}
/*
@@ -691,7 +700,7 @@ struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *root
struct btrfs_trans_handle *btrfs_attach_transaction(struct btrfs_root *root)
{
return start_transaction(root, 0, TRANS_ATTACH,
- BTRFS_RESERVE_NO_FLUSH);
+ BTRFS_RESERVE_NO_FLUSH, true);
}
/*
@@ -707,7 +716,7 @@ btrfs_attach_transaction_barrier(struct btrfs_root *root)
struct btrfs_trans_handle *trans;
trans = start_transaction(root, 0, TRANS_ATTACH,
- BTRFS_RESERVE_NO_FLUSH);
+ BTRFS_RESERVE_NO_FLUSH, true);
if (IS_ERR(trans) && PTR_ERR(trans) == -ENOENT)
btrfs_wait_for_commit(root->fs_info, 0);
@@ -734,7 +743,7 @@ int btrfs_wait_for_commit(struct btrfs_fs_info *fs_info, u64 transid)
list_for_each_entry(t, &fs_info->trans_list, list) {
if (t->transid == transid) {
cur_trans = t;
- atomic_inc(&cur_trans->use_count);
+ refcount_inc(&cur_trans->use_count);
ret = 0;
break;
}
@@ -763,7 +772,7 @@ int btrfs_wait_for_commit(struct btrfs_fs_info *fs_info, u64 transid)
if (t->state == TRANS_STATE_COMPLETED)
break;
cur_trans = t;
- atomic_inc(&cur_trans->use_count);
+ refcount_inc(&cur_trans->use_count);
break;
}
}
@@ -866,14 +875,14 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
if (lock && !atomic_read(&info->open_ioctl_trans) &&
should_end_transaction(trans) &&
- ACCESS_ONCE(cur_trans->state) == TRANS_STATE_RUNNING) {
+ READ_ONCE(cur_trans->state) == TRANS_STATE_RUNNING) {
spin_lock(&info->trans_lock);
if (cur_trans->state == TRANS_STATE_RUNNING)
cur_trans->state = TRANS_STATE_BLOCKED;
spin_unlock(&info->trans_lock);
}
- if (lock && ACCESS_ONCE(cur_trans->state) == TRANS_STATE_BLOCKED) {
+ if (lock && READ_ONCE(cur_trans->state) == TRANS_STATE_BLOCKED) {
if (throttle)
return btrfs_commit_transaction(trans);
else
@@ -907,7 +916,6 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
wake_up_process(info->transaction_kthread);
err = -EIO;
}
- assert_qgroups_uptodate(trans);
kmem_cache_free(btrfs_trans_handle_cachep, trans);
if (must_run_delayed_refs) {
@@ -1354,12 +1362,8 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,
* enabled. If this check races with the ioctl, rescan will
* kick in anyway.
*/
- mutex_lock(&fs_info->qgroup_ioctl_lock);
- if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)) {
- mutex_unlock(&fs_info->qgroup_ioctl_lock);
+ if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
return 0;
- }
- mutex_unlock(&fs_info->qgroup_ioctl_lock);
/*
* We are going to commit transaction, see btrfs_commit_transaction()
@@ -1370,9 +1374,6 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,
ret = commit_fs_roots(trans, fs_info);
if (ret)
goto out;
- ret = btrfs_qgroup_prepare_account_extents(trans, fs_info);
- if (ret < 0)
- goto out;
ret = btrfs_qgroup_account_extents(trans, fs_info);
if (ret < 0)
goto out;
@@ -1499,12 +1500,12 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
/*
* insert the directory item
*/
- ret = btrfs_set_inode_index(parent_inode, &index);
+ ret = btrfs_set_inode_index(BTRFS_I(parent_inode), &index);
BUG_ON(ret); /* -ENOMEM */
/* check if there is a file/dir which has the same name. */
dir_item = btrfs_lookup_dir_item(NULL, parent_root, path,
- btrfs_ino(parent_inode),
+ btrfs_ino(BTRFS_I(parent_inode)),
dentry->d_name.name,
dentry->d_name.len, 0);
if (dir_item != NULL && !IS_ERR(dir_item)) {
@@ -1598,7 +1599,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
*/
ret = btrfs_add_root_ref(trans, fs_info, objectid,
parent_root->root_key.objectid,
- btrfs_ino(parent_inode), index,
+ btrfs_ino(BTRFS_I(parent_inode)), index,
dentry->d_name.name, dentry->d_name.len);
if (ret) {
btrfs_abort_transaction(trans, ret);
@@ -1638,7 +1639,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
ret = btrfs_insert_dir_item(trans, parent_root,
dentry->d_name.name, dentry->d_name.len,
- parent_inode, &key,
+ BTRFS_I(parent_inode), &key,
BTRFS_FT_DIR, index);
/* We have check then name at the beginning, so it is impossible. */
BUG_ON(ret == -EEXIST || ret == -EOVERFLOW);
@@ -1647,7 +1648,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
goto fail;
}
- btrfs_i_size_write(parent_inode, parent_inode->i_size +
+ btrfs_i_size_write(BTRFS_I(parent_inode), parent_inode->i_size +
dentry->d_name.len * 2);
parent_inode->i_mtime = parent_inode->i_ctime =
current_time(parent_inode);
@@ -1833,7 +1834,7 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans,
/* take transaction reference */
cur_trans = trans->transaction;
- atomic_inc(&cur_trans->use_count);
+ refcount_inc(&cur_trans->use_count);
btrfs_end_transaction(trans);
@@ -1922,7 +1923,7 @@ static inline int btrfs_start_delalloc_flush(struct btrfs_fs_info *fs_info)
static inline void btrfs_wait_delalloc_flush(struct btrfs_fs_info *fs_info)
{
if (btrfs_test_opt(fs_info, FLUSHONCOMMIT))
- btrfs_wait_ordered_roots(fs_info, -1, 0, (u64)-1);
+ btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
}
static inline void
@@ -1940,7 +1941,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
int ret;
/* Stop the commit early if ->aborted is set */
- if (unlikely(ACCESS_ONCE(cur_trans->aborted))) {
+ if (unlikely(READ_ONCE(cur_trans->aborted))) {
ret = cur_trans->aborted;
btrfs_end_transaction(trans);
return ret;
@@ -2009,7 +2010,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
spin_lock(&fs_info->trans_lock);
if (cur_trans->state >= TRANS_STATE_COMMIT_START) {
spin_unlock(&fs_info->trans_lock);
- atomic_inc(&cur_trans->use_count);
+ refcount_inc(&cur_trans->use_count);
ret = btrfs_end_transaction(trans);
wait_for_commit(cur_trans);
@@ -2029,7 +2030,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
prev_trans = list_entry(cur_trans->list.prev,
struct btrfs_transaction, list);
if (prev_trans->state != TRANS_STATE_COMPLETED) {
- atomic_inc(&prev_trans->use_count);
+ refcount_inc(&prev_trans->use_count);
spin_unlock(&fs_info->trans_lock);
wait_for_commit(prev_trans);
@@ -2080,7 +2081,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
atomic_read(&cur_trans->num_writers) == 1);
/* ->aborted might be set after the previous check, so check it */
- if (unlikely(ACCESS_ONCE(cur_trans->aborted))) {
+ if (unlikely(READ_ONCE(cur_trans->aborted))) {
ret = cur_trans->aborted;
goto scrub_continue;
}
@@ -2124,13 +2125,6 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
goto scrub_continue;
}
- /* Reocrd old roots for later qgroup accounting */
- ret = btrfs_qgroup_prepare_account_extents(trans, fs_info);
- if (ret) {
- mutex_unlock(&fs_info->reloc_mutex);
- goto scrub_continue;
- }
-
/*
* make sure none of the code above managed to slip in a
* delayed item
@@ -2173,6 +2167,17 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
btrfs_free_log_root_tree(trans, fs_info);
/*
+ * commit_fs_roots() can call btrfs_save_ino_cache(), which generates
+ * new delayed refs. Must handle them or qgroup can be wrong.
+ */
+ ret = btrfs_run_delayed_refs(trans, fs_info, (unsigned long)-1);
+ if (ret) {
+ mutex_unlock(&fs_info->tree_log_mutex);
+ mutex_unlock(&fs_info->reloc_mutex);
+ goto scrub_continue;
+ }
+
+ /*
* Since fs roots are all committed, we can get a quite accurate
* new_roots. So let's do quota accounting.
*/
@@ -2194,14 +2199,14 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
* The tasks which save the space cache and inode cache may also
* update ->aborted, check it.
*/
- if (unlikely(ACCESS_ONCE(cur_trans->aborted))) {
+ if (unlikely(READ_ONCE(cur_trans->aborted))) {
ret = cur_trans->aborted;
mutex_unlock(&fs_info->tree_log_mutex);
mutex_unlock(&fs_info->reloc_mutex);
goto scrub_continue;
}
- btrfs_prepare_extent_commit(trans, fs_info);
+ btrfs_prepare_extent_commit(fs_info);
cur_trans = fs_info->running_transaction;
@@ -2217,7 +2222,6 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
switch_commit_roots(cur_trans, fs_info);
- assert_qgroups_uptodate(trans);
ASSERT(list_empty(&cur_trans->dirty_bgs));
ASSERT(list_empty(&cur_trans->io_bgs));
update_super_roots(fs_info);
@@ -2251,7 +2255,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
goto scrub_continue;
}
- ret = write_ctree_super(trans, fs_info, 0);
+ ret = write_all_supers(fs_info, 0);
if (ret) {
mutex_unlock(&fs_info->tree_log_mutex);
goto scrub_continue;
@@ -2300,7 +2304,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
* it'll result in deadlock about SB_FREEZE_FS.
*/
if (current != fs_info->transaction_kthread &&
- current != fs_info->cleaner_kthread && !fs_info->fs_frozen)
+ current != fs_info->cleaner_kthread &&
+ !test_bit(BTRFS_FS_FROZEN, &fs_info->flags))
btrfs_run_delayed_iputs(fs_info);
return ret;
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 5dfb5590fff6..c55e44560103 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -18,6 +18,8 @@
#ifndef __BTRFS_TRANSACTION__
#define __BTRFS_TRANSACTION__
+
+#include <linux/refcount.h>
#include "btrfs_inode.h"
#include "delayed-ref.h"
#include "ctree.h"
@@ -49,7 +51,7 @@ struct btrfs_transaction {
* transaction can end
*/
atomic_t num_writers;
- atomic_t use_count;
+ refcount_t use_count;
atomic_t pending_ordered;
unsigned long flags;
@@ -125,8 +127,6 @@ struct btrfs_trans_handle {
unsigned int type;
struct btrfs_root *root;
struct btrfs_fs_info *fs_info;
- struct seq_list delayed_ref_elem;
- struct list_head qgroup_ref_list;
struct list_head new_bgs;
};
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index eeffff84f280..f20ef211a73d 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -97,7 +97,7 @@
#define LOG_WALK_REPLAY_ALL 3
static int btrfs_log_inode(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode,
+ struct btrfs_root *root, struct btrfs_inode *inode,
int inode_only,
const loff_t start,
const loff_t end,
@@ -631,8 +631,8 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
* file. This must be done before the btrfs_drop_extents run
* so we don't try to drop this extent.
*/
- ret = btrfs_lookup_file_extent(trans, root, path, btrfs_ino(inode),
- start, 0);
+ ret = btrfs_lookup_file_extent(trans, root, path,
+ btrfs_ino(BTRFS_I(inode)), start, 0);
if (ret == 0 &&
(found_type == BTRFS_FILE_EXTENT_REG ||
@@ -673,6 +673,10 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
unsigned long dest_offset;
struct btrfs_key ins;
+ if (btrfs_file_extent_disk_bytenr(eb, item) == 0 &&
+ btrfs_fs_incompat(fs_info, NO_HOLES))
+ goto update_inode;
+
ret = btrfs_insert_empty_item(trans, root, path, key,
sizeof(*item));
if (ret)
@@ -825,6 +829,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
}
inode_add_bytes(inode, nbytes);
+update_inode:
ret = btrfs_update_inode(trans, root, inode);
out:
if (inode)
@@ -843,7 +848,7 @@ out:
static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct inode *dir,
+ struct btrfs_inode *dir,
struct btrfs_dir_item *di)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -875,7 +880,8 @@ static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans,
if (ret)
goto out;
- ret = btrfs_unlink_inode(trans, root, dir, inode, name, name_len);
+ ret = btrfs_unlink_inode(trans, root, dir, BTRFS_I(inode), name,
+ name_len);
if (ret)
goto out;
else
@@ -991,8 +997,8 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
struct btrfs_root *log_root,
- struct inode *dir, struct inode *inode,
- struct extent_buffer *eb,
+ struct btrfs_inode *dir,
+ struct btrfs_inode *inode,
u64 inode_objectid, u64 parent_objectid,
u64 ref_index, char *name, int namelen,
int *search_done)
@@ -1047,12 +1053,11 @@ again:
parent_objectid,
victim_name,
victim_name_len)) {
- inc_nlink(inode);
+ inc_nlink(&inode->vfs_inode);
btrfs_release_path(path);
- ret = btrfs_unlink_inode(trans, root, dir,
- inode, victim_name,
- victim_name_len);
+ ret = btrfs_unlink_inode(trans, root, dir, inode,
+ victim_name, victim_name_len);
kfree(victim_name);
if (ret)
return ret;
@@ -1115,16 +1120,16 @@ again:
victim_name_len)) {
ret = -ENOENT;
victim_parent = read_one_inode(root,
- parent_objectid);
+ parent_objectid);
if (victim_parent) {
- inc_nlink(inode);
+ inc_nlink(&inode->vfs_inode);
btrfs_release_path(path);
ret = btrfs_unlink_inode(trans, root,
- victim_parent,
- inode,
- victim_name,
- victim_name_len);
+ BTRFS_I(victim_parent),
+ inode,
+ victim_name,
+ victim_name_len);
if (!ret)
ret = btrfs_run_delayed_items(
trans,
@@ -1170,15 +1175,19 @@ next:
return 0;
}
-static int extref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr,
- u32 *namelen, char **name, u64 *index,
- u64 *parent_objectid)
+static int extref_get_fields(struct extent_buffer *eb, int slot,
+ unsigned long ref_ptr, u32 *namelen, char **name,
+ u64 *index, u64 *parent_objectid)
{
struct btrfs_inode_extref *extref;
extref = (struct btrfs_inode_extref *)ref_ptr;
*namelen = btrfs_inode_extref_name_len(eb, extref);
+ if (!btrfs_is_name_len_valid(eb, slot, (unsigned long)&extref->name,
+ *namelen))
+ return -EIO;
+
*name = kmalloc(*namelen, GFP_NOFS);
if (*name == NULL)
return -ENOMEM;
@@ -1193,14 +1202,19 @@ static int extref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr,
return 0;
}
-static int ref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr,
- u32 *namelen, char **name, u64 *index)
+static int ref_get_fields(struct extent_buffer *eb, int slot,
+ unsigned long ref_ptr, u32 *namelen, char **name,
+ u64 *index)
{
struct btrfs_inode_ref *ref;
ref = (struct btrfs_inode_ref *)ref_ptr;
*namelen = btrfs_inode_ref_name_len(eb, ref);
+ if (!btrfs_is_name_len_valid(eb, slot, (unsigned long)(ref + 1),
+ *namelen))
+ return -EIO;
+
*name = kmalloc(*namelen, GFP_NOFS);
if (*name == NULL)
return -ENOMEM;
@@ -1275,8 +1289,8 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
while (ref_ptr < ref_end) {
if (log_ref_ver) {
- ret = extref_get_fields(eb, ref_ptr, &namelen, &name,
- &ref_index, &parent_objectid);
+ ret = extref_get_fields(eb, slot, ref_ptr, &namelen,
+ &name, &ref_index, &parent_objectid);
/*
* parent object can change from one array
* item to another.
@@ -1288,15 +1302,16 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
goto out;
}
} else {
- ret = ref_get_fields(eb, ref_ptr, &namelen, &name,
- &ref_index);
+ ret = ref_get_fields(eb, slot, ref_ptr, &namelen,
+ &name, &ref_index);
}
if (ret)
goto out;
/* if we already have a perfect match, we're done */
- if (!inode_in_dir(root, path, btrfs_ino(dir), btrfs_ino(inode),
- ref_index, name, namelen)) {
+ if (!inode_in_dir(root, path, btrfs_ino(BTRFS_I(dir)),
+ btrfs_ino(BTRFS_I(inode)), ref_index,
+ name, namelen)) {
/*
* look for a conflicting back reference in the
* metadata. if we find one we have to unlink that name
@@ -1307,7 +1322,8 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
if (!search_done) {
ret = __add_inode_ref(trans, root, path, log,
- dir, inode, eb,
+ BTRFS_I(dir),
+ BTRFS_I(inode),
inode_objectid,
parent_objectid,
ref_index, name, namelen,
@@ -1320,8 +1336,9 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
}
/* insert our name */
- ret = btrfs_add_link(trans, dir, inode, name, namelen,
- 0, ref_index);
+ ret = btrfs_add_link(trans, BTRFS_I(dir),
+ BTRFS_I(inode),
+ name, namelen, 0, ref_index);
if (ret)
goto out;
@@ -1360,7 +1377,7 @@ static int insert_orphan_item(struct btrfs_trans_handle *trans,
}
static int count_inode_extrefs(struct btrfs_root *root,
- struct inode *inode, struct btrfs_path *path)
+ struct btrfs_inode *inode, struct btrfs_path *path)
{
int ret = 0;
int name_len;
@@ -1404,7 +1421,7 @@ static int count_inode_extrefs(struct btrfs_root *root,
}
static int count_inode_refs(struct btrfs_root *root,
- struct inode *inode, struct btrfs_path *path)
+ struct btrfs_inode *inode, struct btrfs_path *path)
{
int ret;
struct btrfs_key key;
@@ -1477,19 +1494,19 @@ static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans,
struct btrfs_path *path;
int ret;
u64 nlink = 0;
- u64 ino = btrfs_ino(inode);
+ u64 ino = btrfs_ino(BTRFS_I(inode));
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
- ret = count_inode_refs(root, inode, path);
+ ret = count_inode_refs(root, BTRFS_I(inode), path);
if (ret < 0)
goto out;
nlink = ret;
- ret = count_inode_extrefs(root, inode, path);
+ ret = count_inode_extrefs(root, BTRFS_I(inode), path);
if (ret < 0)
goto out;
@@ -1639,7 +1656,8 @@ static noinline int insert_one_name(struct btrfs_trans_handle *trans,
return -EIO;
}
- ret = btrfs_add_link(trans, dir, inode, name, name_len, 1, index);
+ ret = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode), name,
+ name_len, 1, index);
/* FIXME, put inode into FIXUP list */
@@ -1769,7 +1787,7 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
if (!exists)
goto out;
- ret = drop_one_dir_item(trans, root, path, dir, dst_di);
+ ret = drop_one_dir_item(trans, root, path, BTRFS_I(dir), dst_di);
if (ret)
goto out;
@@ -1778,7 +1796,7 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
out:
btrfs_release_path(path);
if (!ret && update_size) {
- btrfs_i_size_write(dir, dir->i_size + name_len * 2);
+ btrfs_i_size_write(BTRFS_I(dir), dir->i_size + name_len * 2);
ret = btrfs_update_inode(trans, root, dir);
}
kfree(name);
@@ -1832,7 +1850,7 @@ static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans,
ptr_end = ptr + item_size;
while (ptr < ptr_end) {
di = (struct btrfs_dir_item *)ptr;
- if (verify_dir_item(fs_info, eb, di))
+ if (verify_dir_item(fs_info, eb, slot, di))
return -EIO;
name_len = btrfs_dir_name_len(eb, di);
ret = replay_one_name(trans, root, path, eb, di, key);
@@ -2008,7 +2026,7 @@ again:
ptr_end = ptr + item_size;
while (ptr < ptr_end) {
di = (struct btrfs_dir_item *)ptr;
- if (verify_dir_item(fs_info, eb, di)) {
+ if (verify_dir_item(fs_info, eb, slot, di)) {
ret = -EIO;
goto out;
}
@@ -2052,8 +2070,8 @@ again:
}
inc_nlink(inode);
- ret = btrfs_unlink_inode(trans, root, dir, inode,
- name, name_len);
+ ret = btrfs_unlink_inode(trans, root, BTRFS_I(dir),
+ BTRFS_I(inode), name, name_len);
if (!ret)
ret = btrfs_run_delayed_items(trans, fs_info);
kfree(name);
@@ -2093,6 +2111,7 @@ static int replay_xattr_deletes(struct btrfs_trans_handle *trans,
struct btrfs_path *path,
const u64 ino)
{
+ struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_key search_key;
struct btrfs_path *log_path;
int i;
@@ -2134,6 +2153,12 @@ process_leaf:
u32 this_len = sizeof(*di) + name_len + data_len;
char *name;
+ ret = verify_dir_item(fs_info, path->nodes[0],
+ path->slots[0], di);
+ if (ret) {
+ ret = -EIO;
+ goto out;
+ }
name = kmalloc(name_len, GFP_NOFS);
if (!name) {
ret = -ENOMEM;
@@ -2469,7 +2494,7 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
if (trans) {
btrfs_tree_lock(next);
btrfs_set_lock_blocking(next);
- clean_tree_block(trans, fs_info, next);
+ clean_tree_block(fs_info, next);
btrfs_wait_tree_block_writeback(next);
btrfs_tree_unlock(next);
}
@@ -2549,7 +2574,7 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans,
if (trans) {
btrfs_tree_lock(next);
btrfs_set_lock_blocking(next);
- clean_tree_block(trans, fs_info, next);
+ clean_tree_block(fs_info, next);
btrfs_wait_tree_block_writeback(next);
btrfs_tree_unlock(next);
}
@@ -2627,7 +2652,7 @@ static int walk_log_tree(struct btrfs_trans_handle *trans,
if (trans) {
btrfs_tree_lock(next);
btrfs_set_lock_blocking(next);
- clean_tree_block(trans, fs_info, next);
+ clean_tree_block(fs_info, next);
btrfs_wait_tree_block_writeback(next);
btrfs_tree_unlock(next);
}
@@ -2958,7 +2983,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
* the running transaction open, so a full commit can't hop
* in and cause problems either.
*/
- ret = write_ctree_super(trans, fs_info, 1);
+ ret = write_all_supers(fs_info, 1);
if (ret) {
btrfs_set_log_full_commit(fs_info, trans);
btrfs_abort_transaction(trans, ret);
@@ -3084,7 +3109,7 @@ int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans,
int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
const char *name, int name_len,
- struct inode *dir, u64 index)
+ struct btrfs_inode *dir, u64 index)
{
struct btrfs_root *log;
struct btrfs_dir_item *di;
@@ -3094,14 +3119,14 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
int bytes_del = 0;
u64 dir_ino = btrfs_ino(dir);
- if (BTRFS_I(dir)->logged_trans < trans->transid)
+ if (dir->logged_trans < trans->transid)
return 0;
ret = join_running_log_trans(root);
if (ret)
return 0;
- mutex_lock(&BTRFS_I(dir)->log_mutex);
+ mutex_lock(&dir->log_mutex);
log = root->log_root;
path = btrfs_alloc_path();
@@ -3176,7 +3201,7 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
fail:
btrfs_free_path(path);
out_unlock:
- mutex_unlock(&BTRFS_I(dir)->log_mutex);
+ mutex_unlock(&dir->log_mutex);
if (ret == -ENOSPC) {
btrfs_set_log_full_commit(root->fs_info, trans);
ret = 0;
@@ -3192,25 +3217,25 @@ out_unlock:
int btrfs_del_inode_ref_in_log(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
const char *name, int name_len,
- struct inode *inode, u64 dirid)
+ struct btrfs_inode *inode, u64 dirid)
{
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_root *log;
u64 index;
int ret;
- if (BTRFS_I(inode)->logged_trans < trans->transid)
+ if (inode->logged_trans < trans->transid)
return 0;
ret = join_running_log_trans(root);
if (ret)
return 0;
log = root->log_root;
- mutex_lock(&BTRFS_I(inode)->log_mutex);
+ mutex_lock(&inode->log_mutex);
ret = btrfs_del_inode_ref(trans, log, name, name_len, btrfs_ino(inode),
dirid, &index);
- mutex_unlock(&BTRFS_I(inode)->log_mutex);
+ mutex_unlock(&inode->log_mutex);
if (ret == -ENOSPC) {
btrfs_set_log_full_commit(fs_info, trans);
ret = 0;
@@ -3260,7 +3285,7 @@ static noinline int insert_dir_log_key(struct btrfs_trans_handle *trans,
* to replay anything deleted before the fsync
*/
static noinline int log_dir_items(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode,
+ struct btrfs_root *root, struct btrfs_inode *inode,
struct btrfs_path *path,
struct btrfs_path *dst_path, int key_type,
struct btrfs_log_ctx *ctx,
@@ -3450,7 +3475,7 @@ done:
* key logged by this transaction.
*/
static noinline int log_directory_changes(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode,
+ struct btrfs_root *root, struct btrfs_inode *inode,
struct btrfs_path *path,
struct btrfs_path *dst_path,
struct btrfs_log_ctx *ctx)
@@ -3464,9 +3489,8 @@ again:
min_key = 0;
max_key = 0;
while (1) {
- ret = log_dir_items(trans, root, inode, path,
- dst_path, key_type, ctx, min_key,
- &max_key);
+ ret = log_dir_items(trans, root, inode, path, dst_path, key_type,
+ ctx, min_key, &max_key);
if (ret)
return ret;
if (max_key == (u64)-1)
@@ -3595,34 +3619,34 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
static int log_inode_item(struct btrfs_trans_handle *trans,
struct btrfs_root *log, struct btrfs_path *path,
- struct inode *inode)
+ struct btrfs_inode *inode)
{
struct btrfs_inode_item *inode_item;
int ret;
ret = btrfs_insert_empty_item(trans, log, path,
- &BTRFS_I(inode)->location,
- sizeof(*inode_item));
+ &inode->location, sizeof(*inode_item));
if (ret && ret != -EEXIST)
return ret;
inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
struct btrfs_inode_item);
- fill_inode_item(trans, path->nodes[0], inode_item, inode, 0, 0);
+ fill_inode_item(trans, path->nodes[0], inode_item, &inode->vfs_inode,
+ 0, 0);
btrfs_release_path(path);
return 0;
}
static noinline int copy_items(struct btrfs_trans_handle *trans,
- struct inode *inode,
+ struct btrfs_inode *inode,
struct btrfs_path *dst_path,
struct btrfs_path *src_path, u64 *last_extent,
int start_slot, int nr, int inode_only,
u64 logged_isize)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
unsigned long src_offset;
unsigned long dst_offset;
- struct btrfs_root *log = BTRFS_I(inode)->root->log_root;
+ struct btrfs_root *log = inode->root->log_root;
struct btrfs_file_extent_item *extent;
struct btrfs_inode_item *inode_item;
struct extent_buffer *src = src_path->nodes[0];
@@ -3633,7 +3657,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
char *ins_data;
int i;
struct list_head ordered_sums;
- int skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
+ int skip_csum = inode->flags & BTRFS_INODE_NODATASUM;
bool has_extents = false;
bool need_find_last_extent = true;
bool done = false;
@@ -3675,7 +3699,8 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
dst_path->slots[0],
struct btrfs_inode_item);
fill_inode_item(trans, dst_path->nodes[0], inode_item,
- inode, inode_only == LOG_INODE_EXISTS,
+ &inode->vfs_inode,
+ inode_only == LOG_INODE_EXISTS,
logged_isize);
} else {
copy_extent_buffer(dst_path->nodes[0], src, dst_offset,
@@ -3783,7 +3808,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
if (need_find_last_extent) {
u64 len;
- ret = btrfs_prev_leaf(BTRFS_I(inode)->root, src_path);
+ ret = btrfs_prev_leaf(inode->root, src_path);
if (ret < 0)
return ret;
if (ret)
@@ -3825,8 +3850,8 @@ fill_holes:
if (need_find_last_extent) {
/* btrfs_prev_leaf could return 1 without releasing the path */
btrfs_release_path(src_path);
- ret = btrfs_search_slot(NULL, BTRFS_I(inode)->root, &first_key,
- src_path, 0, 0);
+ ret = btrfs_search_slot(NULL, inode->root, &first_key,
+ src_path, 0, 0);
if (ret < 0)
return ret;
ASSERT(ret == 0);
@@ -3846,7 +3871,7 @@ fill_holes:
u64 extent_end;
if (i >= btrfs_header_nritems(src_path->nodes[0])) {
- ret = btrfs_next_leaf(BTRFS_I(inode)->root, src_path);
+ ret = btrfs_next_leaf(inode->root, src_path);
if (ret < 0)
return ret;
ASSERT(ret == 0);
@@ -3881,8 +3906,7 @@ fill_holes:
offset = *last_extent;
len = key.offset - *last_extent;
ret = btrfs_insert_file_extent(trans, log, btrfs_ino(inode),
- offset, 0, 0, len, 0, len, 0,
- 0, 0);
+ offset, 0, 0, len, 0, len, 0, 0, 0);
if (ret)
break;
*last_extent = extent_end;
@@ -4055,7 +4079,7 @@ static int wait_ordered_extents(struct btrfs_trans_handle *trans,
}
static int log_one_extent(struct btrfs_trans_handle *trans,
- struct inode *inode, struct btrfs_root *root,
+ struct btrfs_inode *inode, struct btrfs_root *root,
const struct extent_map *em,
struct btrfs_path *path,
const struct list_head *logged_list,
@@ -4072,8 +4096,8 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
int extent_inserted = 0;
bool ordered_io_err = false;
- ret = wait_ordered_extents(trans, inode, root, em, logged_list,
- &ordered_io_err);
+ ret = wait_ordered_extents(trans, &inode->vfs_inode, root, em,
+ logged_list, &ordered_io_err);
if (ret)
return ret;
@@ -4084,7 +4108,7 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
btrfs_init_map_token(&token);
- ret = __btrfs_drop_extents(trans, log, inode, path, em->start,
+ ret = __btrfs_drop_extents(trans, log, &inode->vfs_inode, path, em->start,
em->start + em->len, NULL, 0, 1,
sizeof(*fi), &extent_inserted);
if (ret)
@@ -4150,7 +4174,7 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct inode *inode,
+ struct btrfs_inode *inode,
struct btrfs_path *path,
struct list_head *logged_list,
struct btrfs_log_ctx *ctx,
@@ -4159,14 +4183,14 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
{
struct extent_map *em, *n;
struct list_head extents;
- struct extent_map_tree *tree = &BTRFS_I(inode)->extent_tree;
+ struct extent_map_tree *tree = &inode->extent_tree;
u64 test_gen;
int ret = 0;
int num = 0;
INIT_LIST_HEAD(&extents);
- down_write(&BTRFS_I(inode)->dio_sem);
+ down_write(&inode->dio_sem);
write_lock(&tree->lock);
test_gen = root->fs_info->last_trans_committed;
@@ -4188,7 +4212,7 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
if (em->generation <= test_gen)
continue;
/* Need a ref to keep it from getting evicted from cache */
- atomic_inc(&em->refs);
+ refcount_inc(&em->refs);
set_bit(EXTENT_FLAG_LOGGING, &em->flags);
list_add_tail(&em->list, &extents);
num++;
@@ -4206,7 +4230,7 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
* without writing to the log tree and the fsync must report the
* file data write error and not commit the current transaction.
*/
- ret = filemap_check_errors(inode->i_mapping);
+ ret = filemap_check_errors(inode->vfs_inode.i_mapping);
if (ret)
ctx->io_err = ret;
process:
@@ -4235,13 +4259,13 @@ process:
}
WARN_ON(!list_empty(&extents));
write_unlock(&tree->lock);
- up_write(&BTRFS_I(inode)->dio_sem);
+ up_write(&inode->dio_sem);
btrfs_release_path(path);
return ret;
}
-static int logged_inode_size(struct btrfs_root *log, struct inode *inode,
+static int logged_inode_size(struct btrfs_root *log, struct btrfs_inode *inode,
struct btrfs_path *path, u64 *size_ret)
{
struct btrfs_key key;
@@ -4279,7 +4303,7 @@ static int logged_inode_size(struct btrfs_root *log, struct inode *inode,
*/
static int btrfs_log_all_xattrs(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct inode *inode,
+ struct btrfs_inode *inode,
struct btrfs_path *path,
struct btrfs_path *dst_path)
{
@@ -4374,7 +4398,7 @@ static int btrfs_log_all_xattrs(struct btrfs_trans_handle *trans,
*/
static int btrfs_log_trailing_hole(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct inode *inode,
+ struct btrfs_inode *inode,
struct btrfs_path *path)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -4385,7 +4409,7 @@ static int btrfs_log_trailing_hole(struct btrfs_trans_handle *trans,
struct extent_buffer *leaf;
struct btrfs_root *log = root->log_root;
const u64 ino = btrfs_ino(inode);
- const u64 i_size = i_size_read(inode);
+ const u64 i_size = i_size_read(&inode->vfs_inode);
if (!btrfs_fs_incompat(fs_info, NO_HOLES))
return 0;
@@ -4495,7 +4519,7 @@ static int btrfs_log_trailing_hole(struct btrfs_trans_handle *trans,
static int btrfs_check_ref_name_override(struct extent_buffer *eb,
const int slot,
const struct btrfs_key *key,
- struct inode *inode,
+ struct btrfs_inode *inode,
u64 *other_ino)
{
int ret;
@@ -4538,6 +4562,12 @@ static int btrfs_check_ref_name_override(struct extent_buffer *eb,
this_len = sizeof(*extref) + this_name_len;
}
+ ret = btrfs_is_name_len_valid(eb, slot, name_ptr,
+ this_name_len);
+ if (!ret) {
+ ret = -EIO;
+ goto out;
+ }
if (this_name_len > name_len) {
char *new_name;
@@ -4551,9 +4581,8 @@ static int btrfs_check_ref_name_override(struct extent_buffer *eb,
}
read_extent_buffer(eb, name, name_ptr, this_name_len);
- di = btrfs_lookup_dir_item(NULL, BTRFS_I(inode)->root,
- search_path, parent,
- name, this_name_len, 0);
+ di = btrfs_lookup_dir_item(NULL, inode->root, search_path,
+ parent, name, this_name_len, 0);
if (di && !IS_ERR(di)) {
struct btrfs_key di_key;
@@ -4596,7 +4625,7 @@ out:
* This handles both files and directories.
*/
static int btrfs_log_inode(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode,
+ struct btrfs_root *root, struct btrfs_inode *inode,
int inode_only,
const loff_t start,
const loff_t end,
@@ -4618,7 +4647,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
int ins_nr;
bool fast_search = false;
u64 ino = btrfs_ino(inode);
- struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
+ struct extent_map_tree *em_tree = &inode->extent_tree;
u64 logged_isize = 0;
bool need_log_inode_item = true;
@@ -4639,9 +4668,9 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
/* today the code can only do partial logging of directories */
- if (S_ISDIR(inode->i_mode) ||
+ if (S_ISDIR(inode->vfs_inode.i_mode) ||
(!test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
- &BTRFS_I(inode)->runtime_flags) &&
+ &inode->runtime_flags) &&
inode_only >= LOG_INODE_EXISTS))
max_key.type = BTRFS_XATTR_ITEM_KEY;
else
@@ -4654,8 +4683,8 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
* order for the log replay code to mark inodes for link count
* fixup (create temporary BTRFS_TREE_LOG_FIXUP_OBJECTID items).
*/
- if (S_ISDIR(inode->i_mode) ||
- BTRFS_I(inode)->generation > fs_info->last_trans_committed)
+ if (S_ISDIR(inode->vfs_inode.i_mode) ||
+ inode->generation > fs_info->last_trans_committed)
ret = btrfs_commit_inode_delayed_items(trans, inode);
else
ret = btrfs_commit_inode_delayed_inode(inode);
@@ -4668,17 +4697,16 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
if (inode_only == LOG_OTHER_INODE) {
inode_only = LOG_INODE_EXISTS;
- mutex_lock_nested(&BTRFS_I(inode)->log_mutex,
- SINGLE_DEPTH_NESTING);
+ mutex_lock_nested(&inode->log_mutex, SINGLE_DEPTH_NESTING);
} else {
- mutex_lock(&BTRFS_I(inode)->log_mutex);
+ mutex_lock(&inode->log_mutex);
}
/*
* a brute force approach to making sure we get the most uptodate
* copies of everything.
*/
- if (S_ISDIR(inode->i_mode)) {
+ if (S_ISDIR(inode->vfs_inode.i_mode)) {
int max_key_type = BTRFS_DIR_LOG_INDEX_KEY;
if (inode_only == LOG_INODE_EXISTS)
@@ -4699,31 +4727,30 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
* (zeroes), as if an expanding truncate happened,
* instead of getting a file of 4Kb only.
*/
- err = logged_inode_size(log, inode, path,
- &logged_isize);
+ err = logged_inode_size(log, inode, path, &logged_isize);
if (err)
goto out_unlock;
}
if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
- &BTRFS_I(inode)->runtime_flags)) {
+ &inode->runtime_flags)) {
if (inode_only == LOG_INODE_EXISTS) {
max_key.type = BTRFS_XATTR_ITEM_KEY;
ret = drop_objectid_items(trans, log, path, ino,
max_key.type);
} else {
clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
- &BTRFS_I(inode)->runtime_flags);
+ &inode->runtime_flags);
clear_bit(BTRFS_INODE_COPY_EVERYTHING,
- &BTRFS_I(inode)->runtime_flags);
+ &inode->runtime_flags);
while(1) {
ret = btrfs_truncate_inode_items(trans,
- log, inode, 0, 0);
+ log, &inode->vfs_inode, 0, 0);
if (ret != -EAGAIN)
break;
}
}
} else if (test_and_clear_bit(BTRFS_INODE_COPY_EVERYTHING,
- &BTRFS_I(inode)->runtime_flags) ||
+ &inode->runtime_flags) ||
inode_only == LOG_INODE_EXISTS) {
if (inode_only == LOG_INODE_ALL)
fast_search = true;
@@ -4764,18 +4791,17 @@ again:
if ((min_key.type == BTRFS_INODE_REF_KEY ||
min_key.type == BTRFS_INODE_EXTREF_KEY) &&
- BTRFS_I(inode)->generation == trans->transid) {
+ inode->generation == trans->transid) {
u64 other_ino = 0;
ret = btrfs_check_ref_name_override(path->nodes[0],
- path->slots[0],
- &min_key, inode,
- &other_ino);
+ path->slots[0], &min_key, inode,
+ &other_ino);
if (ret < 0) {
err = ret;
goto out_unlock;
} else if (ret > 0 && ctx &&
- other_ino != btrfs_ino(ctx->inode)) {
+ other_ino != btrfs_ino(BTRFS_I(ctx->inode))) {
struct btrfs_key inode_key;
struct inode *other_inode;
@@ -4823,9 +4849,10 @@ again:
* update the log with the new name before we
* unpin it.
*/
- err = btrfs_log_inode(trans, root, other_inode,
- LOG_OTHER_INODE,
- 0, LLONG_MAX, ctx);
+ err = btrfs_log_inode(trans, root,
+ BTRFS_I(other_inode),
+ LOG_OTHER_INODE, 0, LLONG_MAX,
+ ctx);
iput(other_inode);
if (err)
goto out_unlock;
@@ -4979,25 +5006,25 @@ log_extents:
write_unlock(&em_tree->lock);
}
- if (inode_only == LOG_INODE_ALL && S_ISDIR(inode->i_mode)) {
+ if (inode_only == LOG_INODE_ALL && S_ISDIR(inode->vfs_inode.i_mode)) {
ret = log_directory_changes(trans, root, inode, path, dst_path,
- ctx);
+ ctx);
if (ret) {
err = ret;
goto out_unlock;
}
}
- spin_lock(&BTRFS_I(inode)->lock);
- BTRFS_I(inode)->logged_trans = trans->transid;
- BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->last_sub_trans;
- spin_unlock(&BTRFS_I(inode)->lock);
+ spin_lock(&inode->lock);
+ inode->logged_trans = trans->transid;
+ inode->last_log_commit = inode->last_sub_trans;
+ spin_unlock(&inode->lock);
out_unlock:
if (unlikely(err))
btrfs_put_logged_extents(&logged_list);
else
btrfs_submit_logged_extents(&logged_list, log);
- mutex_unlock(&BTRFS_I(inode)->log_mutex);
+ mutex_unlock(&inode->log_mutex);
btrfs_free_path(path);
btrfs_free_path(dst_path);
@@ -5021,13 +5048,13 @@ out_unlock:
* we logged the inode or it might have also done the unlink).
*/
static bool btrfs_must_commit_transaction(struct btrfs_trans_handle *trans,
- struct inode *inode)
+ struct btrfs_inode *inode)
{
- struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+ struct btrfs_fs_info *fs_info = inode->root->fs_info;
bool ret = false;
- mutex_lock(&BTRFS_I(inode)->log_mutex);
- if (BTRFS_I(inode)->last_unlink_trans > fs_info->last_trans_committed) {
+ mutex_lock(&inode->log_mutex);
+ if (inode->last_unlink_trans > fs_info->last_trans_committed) {
/*
* Make sure any commits to the log are forced to be full
* commits.
@@ -5035,7 +5062,7 @@ static bool btrfs_must_commit_transaction(struct btrfs_trans_handle *trans,
btrfs_set_log_full_commit(fs_info, trans);
ret = true;
}
- mutex_unlock(&BTRFS_I(inode)->log_mutex);
+ mutex_unlock(&inode->log_mutex);
return ret;
}
@@ -5047,14 +5074,14 @@ static bool btrfs_must_commit_transaction(struct btrfs_trans_handle *trans,
* a full commit is required.
*/
static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
- struct inode *inode,
+ struct btrfs_inode *inode,
struct dentry *parent,
struct super_block *sb,
u64 last_committed)
{
int ret = 0;
struct dentry *old_parent = NULL;
- struct inode *orig_inode = inode;
+ struct btrfs_inode *orig_inode = inode;
/*
* for regular files, if its inode is already on disk, we don't
@@ -5062,15 +5089,15 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
* we can use the last_unlink_trans field to record renames
* and other fun in this file.
*/
- if (S_ISREG(inode->i_mode) &&
- BTRFS_I(inode)->generation <= last_committed &&
- BTRFS_I(inode)->last_unlink_trans <= last_committed)
- goto out;
+ if (S_ISREG(inode->vfs_inode.i_mode) &&
+ inode->generation <= last_committed &&
+ inode->last_unlink_trans <= last_committed)
+ goto out;
- if (!S_ISDIR(inode->i_mode)) {
+ if (!S_ISDIR(inode->vfs_inode.i_mode)) {
if (!parent || d_really_is_negative(parent) || sb != parent->d_sb)
goto out;
- inode = d_inode(parent);
+ inode = BTRFS_I(d_inode(parent));
}
while (1) {
@@ -5081,7 +5108,7 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
* think this inode has already been logged.
*/
if (inode != orig_inode)
- BTRFS_I(inode)->logged_trans = trans->transid;
+ inode->logged_trans = trans->transid;
smp_mb();
if (btrfs_must_commit_transaction(trans, inode)) {
@@ -5093,7 +5120,7 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
break;
if (IS_ROOT(parent)) {
- inode = d_inode(parent);
+ inode = BTRFS_I(d_inode(parent));
if (btrfs_must_commit_transaction(trans, inode))
ret = 1;
break;
@@ -5102,7 +5129,7 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
parent = dget_parent(parent);
dput(old_parent);
old_parent = parent;
- inode = d_inode(parent);
+ inode = BTRFS_I(d_inode(parent));
}
dput(old_parent);
@@ -5159,7 +5186,7 @@ struct btrfs_dir_list {
*/
static int log_new_dir_dentries(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
- struct inode *start_inode,
+ struct btrfs_inode *start_inode,
struct btrfs_log_ctx *ctx)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -5237,7 +5264,7 @@ process_leaf:
goto next_dir_inode;
}
- if (btrfs_inode_in_log(di_inode, trans->transid)) {
+ if (btrfs_inode_in_log(BTRFS_I(di_inode), trans->transid)) {
iput(di_inode);
break;
}
@@ -5245,10 +5272,10 @@ process_leaf:
ctx->log_new_dentries = false;
if (type == BTRFS_FT_DIR || type == BTRFS_FT_SYMLINK)
log_mode = LOG_INODE_ALL;
- ret = btrfs_log_inode(trans, root, di_inode,
+ ret = btrfs_log_inode(trans, root, BTRFS_I(di_inode),
log_mode, 0, LLONG_MAX, ctx);
if (!ret &&
- btrfs_must_commit_transaction(trans, di_inode))
+ btrfs_must_commit_transaction(trans, BTRFS_I(di_inode)))
ret = 1;
iput(di_inode);
if (ret)
@@ -5289,14 +5316,14 @@ next_dir_inode:
}
static int btrfs_log_all_parents(struct btrfs_trans_handle *trans,
- struct inode *inode,
+ struct btrfs_inode *inode,
struct btrfs_log_ctx *ctx)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
int ret;
struct btrfs_path *path;
struct btrfs_key key;
- struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_root *root = inode->root;
const u64 ino = btrfs_ino(inode);
path = btrfs_alloc_path();
@@ -5365,14 +5392,14 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans,
if (ctx)
ctx->log_new_dentries = false;
- ret = btrfs_log_inode(trans, root, dir_inode,
+ ret = btrfs_log_inode(trans, root, BTRFS_I(dir_inode),
LOG_INODE_ALL, 0, LLONG_MAX, ctx);
if (!ret &&
- btrfs_must_commit_transaction(trans, dir_inode))
+ btrfs_must_commit_transaction(trans, BTRFS_I(dir_inode)))
ret = 1;
if (!ret && ctx && ctx->log_new_dentries)
ret = log_new_dir_dentries(trans, root,
- dir_inode, ctx);
+ BTRFS_I(dir_inode), ctx);
iput(dir_inode);
if (ret)
goto out;
@@ -5392,7 +5419,8 @@ out:
* the last committed transaction
*/
static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode,
+ struct btrfs_root *root,
+ struct btrfs_inode *inode,
struct dentry *parent,
const loff_t start,
const loff_t end,
@@ -5406,9 +5434,9 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
int ret = 0;
u64 last_committed = fs_info->last_trans_committed;
bool log_dentries = false;
- struct inode *orig_inode = inode;
+ struct btrfs_inode *orig_inode = inode;
- sb = inode->i_sb;
+ sb = inode->vfs_inode.i_sb;
if (btrfs_test_opt(fs_info, NOTREELOG)) {
ret = 1;
@@ -5425,14 +5453,13 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
goto end_no_trans;
}
- if (root != BTRFS_I(inode)->root ||
- btrfs_root_refs(&root->root_item) == 0) {
+ if (root != inode->root || btrfs_root_refs(&root->root_item) == 0) {
ret = 1;
goto end_no_trans;
}
- ret = check_parent_dirs_for_sync(trans, inode, parent,
- sb, last_committed);
+ ret = check_parent_dirs_for_sync(trans, inode, parent, sb,
+ last_committed);
if (ret)
goto end_no_trans;
@@ -5455,14 +5482,14 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
* we can use the last_unlink_trans field to record renames
* and other fun in this file.
*/
- if (S_ISREG(inode->i_mode) &&
- BTRFS_I(inode)->generation <= last_committed &&
- BTRFS_I(inode)->last_unlink_trans <= last_committed) {
+ if (S_ISREG(inode->vfs_inode.i_mode) &&
+ inode->generation <= last_committed &&
+ inode->last_unlink_trans <= last_committed) {
ret = 0;
goto end_trans;
}
- if (S_ISDIR(inode->i_mode) && ctx && ctx->log_new_dentries)
+ if (S_ISDIR(inode->vfs_inode.i_mode) && ctx && ctx->log_new_dentries)
log_dentries = true;
/*
@@ -5506,7 +5533,7 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
* but the file inode does not have a matching BTRFS_INODE_REF_KEY item
* and has a link count of 2.
*/
- if (BTRFS_I(inode)->last_unlink_trans > last_committed) {
+ if (inode->last_unlink_trans > last_committed) {
ret = btrfs_log_all_parents(trans, orig_inode, ctx);
if (ret)
goto end_trans;
@@ -5516,14 +5543,13 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
if (!parent || d_really_is_negative(parent) || sb != parent->d_sb)
break;
- inode = d_inode(parent);
- if (root != BTRFS_I(inode)->root)
+ inode = BTRFS_I(d_inode(parent));
+ if (root != inode->root)
break;
- if (BTRFS_I(inode)->generation > last_committed) {
+ if (inode->generation > last_committed) {
ret = btrfs_log_inode(trans, root, inode,
- LOG_INODE_EXISTS,
- 0, LLONG_MAX, ctx);
+ LOG_INODE_EXISTS, 0, LLONG_MAX, ctx);
if (ret)
goto end_trans;
}
@@ -5567,8 +5593,8 @@ int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
struct dentry *parent = dget_parent(dentry);
int ret;
- ret = btrfs_log_inode_parent(trans, root, d_inode(dentry), parent,
- start, end, 0, ctx);
+ ret = btrfs_log_inode_parent(trans, root, BTRFS_I(d_inode(dentry)),
+ parent, start, end, 0, ctx);
dput(parent);
return ret;
@@ -5730,7 +5756,7 @@ error:
* inodes, etc) are done.
*/
void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
- struct inode *dir, struct inode *inode,
+ struct btrfs_inode *dir, struct btrfs_inode *inode,
int for_rename)
{
/*
@@ -5743,23 +5769,23 @@ void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
* into the file. When the file is logged we check it and
* don't log the parents if the file is fully on disk.
*/
- mutex_lock(&BTRFS_I(inode)->log_mutex);
- BTRFS_I(inode)->last_unlink_trans = trans->transid;
- mutex_unlock(&BTRFS_I(inode)->log_mutex);
+ mutex_lock(&inode->log_mutex);
+ inode->last_unlink_trans = trans->transid;
+ mutex_unlock(&inode->log_mutex);
/*
* if this directory was already logged any new
* names for this file/dir will get recorded
*/
smp_mb();
- if (BTRFS_I(dir)->logged_trans == trans->transid)
+ if (dir->logged_trans == trans->transid)
return;
/*
* if the inode we're about to unlink was logged,
* the log will be properly updated for any new names
*/
- if (BTRFS_I(inode)->logged_trans == trans->transid)
+ if (inode->logged_trans == trans->transid)
return;
/*
@@ -5776,9 +5802,9 @@ void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
return;
record:
- mutex_lock(&BTRFS_I(dir)->log_mutex);
- BTRFS_I(dir)->last_unlink_trans = trans->transid;
- mutex_unlock(&BTRFS_I(dir)->log_mutex);
+ mutex_lock(&dir->log_mutex);
+ dir->last_unlink_trans = trans->transid;
+ mutex_unlock(&dir->log_mutex);
}
/*
@@ -5794,11 +5820,11 @@ record:
* parent root and tree of tree roots trees, etc) are done.
*/
void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans,
- struct inode *dir)
+ struct btrfs_inode *dir)
{
- mutex_lock(&BTRFS_I(dir)->log_mutex);
- BTRFS_I(dir)->last_unlink_trans = trans->transid;
- mutex_unlock(&BTRFS_I(dir)->log_mutex);
+ mutex_lock(&dir->log_mutex);
+ dir->last_unlink_trans = trans->transid;
+ mutex_unlock(&dir->log_mutex);
}
/*
@@ -5809,27 +5835,25 @@ void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans,
* full transaction commit is required.
*/
int btrfs_log_new_name(struct btrfs_trans_handle *trans,
- struct inode *inode, struct inode *old_dir,
+ struct btrfs_inode *inode, struct btrfs_inode *old_dir,
struct dentry *parent)
{
- struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_root * root = BTRFS_I(inode)->root;
+ struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
+ struct btrfs_root *root = inode->root;
/*
* this will force the logging code to walk the dentry chain
* up for the file
*/
- if (S_ISREG(inode->i_mode))
- BTRFS_I(inode)->last_unlink_trans = trans->transid;
+ if (S_ISREG(inode->vfs_inode.i_mode))
+ inode->last_unlink_trans = trans->transid;
/*
* if this inode hasn't been logged and directory we're renaming it
* from hasn't been logged, we don't need to log it
*/
- if (BTRFS_I(inode)->logged_trans <=
- fs_info->last_trans_committed &&
- (!old_dir || BTRFS_I(old_dir)->logged_trans <=
- fs_info->last_trans_committed))
+ if (inode->logged_trans <= fs_info->last_trans_committed &&
+ (!old_dir || old_dir->logged_trans <= fs_info->last_trans_committed))
return 0;
return btrfs_log_inode_parent(trans, root, inode, parent, 0,
diff --git a/fs/btrfs/tree-log.h b/fs/btrfs/tree-log.h
index ab858e31ccbc..483027f9a7f4 100644
--- a/fs/btrfs/tree-log.h
+++ b/fs/btrfs/tree-log.h
@@ -48,13 +48,13 @@ static inline void btrfs_init_log_ctx(struct btrfs_log_ctx *ctx,
static inline void btrfs_set_log_full_commit(struct btrfs_fs_info *fs_info,
struct btrfs_trans_handle *trans)
{
- ACCESS_ONCE(fs_info->last_trans_log_full_commit) = trans->transid;
+ WRITE_ONCE(fs_info->last_trans_log_full_commit, trans->transid);
}
static inline int btrfs_need_log_full_commit(struct btrfs_fs_info *fs_info,
struct btrfs_trans_handle *trans)
{
- return ACCESS_ONCE(fs_info->last_trans_log_full_commit) ==
+ return READ_ONCE(fs_info->last_trans_log_full_commit) ==
trans->transid;
}
@@ -72,19 +72,19 @@ int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
const char *name, int name_len,
- struct inode *dir, u64 index);
+ struct btrfs_inode *dir, u64 index);
int btrfs_del_inode_ref_in_log(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
const char *name, int name_len,
- struct inode *inode, u64 dirid);
+ struct btrfs_inode *inode, u64 dirid);
void btrfs_end_log_trans(struct btrfs_root *root);
int btrfs_pin_log_trans(struct btrfs_root *root);
void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
- struct inode *dir, struct inode *inode,
+ struct btrfs_inode *dir, struct btrfs_inode *inode,
int for_rename);
void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans,
- struct inode *dir);
+ struct btrfs_inode *dir);
int btrfs_log_new_name(struct btrfs_trans_handle *trans,
- struct inode *inode, struct inode *old_dir,
+ struct btrfs_inode *inode, struct btrfs_inode *old_dir,
struct dentry *parent);
#endif
diff --git a/fs/btrfs/ulist.c b/fs/btrfs/ulist.c
index b1434bb57e36..d8edf164f81c 100644
--- a/fs/btrfs/ulist.c
+++ b/fs/btrfs/ulist.c
@@ -52,13 +52,13 @@ void ulist_init(struct ulist *ulist)
}
/**
- * ulist_fini - free up additionally allocated memory for the ulist
+ * ulist_release - free up additionally allocated memory for the ulist
* @ulist: the ulist from which to free the additional memory
*
* This is useful in cases where the base 'struct ulist' has been statically
* allocated.
*/
-static void ulist_fini(struct ulist *ulist)
+void ulist_release(struct ulist *ulist)
{
struct ulist_node *node;
struct ulist_node *next;
@@ -79,7 +79,7 @@ static void ulist_fini(struct ulist *ulist)
*/
void ulist_reinit(struct ulist *ulist)
{
- ulist_fini(ulist);
+ ulist_release(ulist);
ulist_init(ulist);
}
@@ -105,13 +105,13 @@ struct ulist *ulist_alloc(gfp_t gfp_mask)
* ulist_free - free dynamically allocated ulist
* @ulist: ulist to free
*
- * It is not necessary to call ulist_fini before.
+ * It is not necessary to call ulist_release before.
*/
void ulist_free(struct ulist *ulist)
{
if (!ulist)
return;
- ulist_fini(ulist);
+ ulist_release(ulist);
kfree(ulist);
}
diff --git a/fs/btrfs/ulist.h b/fs/btrfs/ulist.h
index a01a2c45825f..53c913632733 100644
--- a/fs/btrfs/ulist.h
+++ b/fs/btrfs/ulist.h
@@ -19,9 +19,6 @@
*
*/
struct ulist_iterator {
-#ifdef CONFIG_BTRFS_DEBUG
- int i;
-#endif
struct list_head *cur_list; /* hint to start search */
};
@@ -32,10 +29,6 @@ struct ulist_node {
u64 val; /* value to store */
u64 aux; /* auxiliary value saved along with the val */
-#ifdef CONFIG_BTRFS_DEBUG
- int seqnum; /* sequence number this node is added */
-#endif
-
struct list_head list; /* used to link node */
struct rb_node rb_node; /* used to speed up search */
};
@@ -51,6 +44,7 @@ struct ulist {
};
void ulist_init(struct ulist *ulist);
+void ulist_release(struct ulist *ulist);
void ulist_reinit(struct ulist *ulist);
struct ulist *ulist_alloc(gfp_t gfp_mask);
void ulist_free(struct ulist *ulist);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 3c3c69c0eee4..5eb7217738ed 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -134,12 +134,16 @@ const int btrfs_raid_mindev_error[BTRFS_NR_RAID_TYPES] = {
};
static int init_first_rw_device(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info,
- struct btrfs_device *device);
+ struct btrfs_fs_info *fs_info);
static int btrfs_relocate_sys_chunks(struct btrfs_fs_info *fs_info);
static void __btrfs_reset_dev_stats(struct btrfs_device *dev);
static void btrfs_dev_stat_print_on_error(struct btrfs_device *dev);
static void btrfs_dev_stat_print_on_load(struct btrfs_device *device);
+static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
+ enum btrfs_map_op op,
+ u64 logical, u64 *length,
+ struct btrfs_bio **bbio_ret,
+ int mirror_num, int need_raid_map);
DEFINE_MUTEX(uuid_mutex);
static LIST_HEAD(fs_uuids);
@@ -238,6 +242,17 @@ static struct btrfs_device *__alloc_device(void)
if (!dev)
return ERR_PTR(-ENOMEM);
+ /*
+ * Preallocate a bio that's always going to be used for flushing device
+ * barriers and matches the device lifespan
+ */
+ dev->flush_bio = bio_alloc_bioset(GFP_KERNEL, 0, NULL);
+ if (!dev->flush_bio) {
+ kfree(dev);
+ return ERR_PTR(-ENOMEM);
+ }
+ bio_get(dev->flush_bio);
+
INIT_LIST_HEAD(&dev->dev_list);
INIT_LIST_HEAD(&dev->dev_alloc_list);
INIT_LIST_HEAD(&dev->resized_list);
@@ -366,7 +381,7 @@ static noinline void run_scheduled_bios(struct btrfs_device *device)
*/
blk_start_plug(&plug);
- bdi = blk_get_backing_dev_info(device->bdev);
+ bdi = device->bdev->bd_bdi;
limit = btrfs_async_submit_limit(fs_info);
limit = limit * 2 / 3;
@@ -834,6 +849,7 @@ static void __free_device(struct work_struct *work)
device = container_of(work, struct btrfs_device, rcu_work);
rcu_string_free(device->name);
+ bio_put(device->flush_bio);
kfree(device);
}
@@ -1009,14 +1025,13 @@ static int __btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
q = bdev_get_queue(bdev);
if (blk_queue_discard(q))
device->can_discard = 1;
+ if (!blk_queue_nonrot(q))
+ fs_devices->rotating = 1;
device->bdev = bdev;
device->in_fs_metadata = 0;
device->mode = flags;
- if (!blk_queue_nonrot(bdev_get_queue(bdev)))
- fs_devices->rotating = 1;
-
fs_devices->open_devices++;
if (device->writeable &&
device->devid != BTRFS_DEV_REPLACE_DEVID) {
@@ -1350,15 +1365,13 @@ int find_free_dev_extent_start(struct btrfs_transaction *transaction,
int ret;
int slot;
struct extent_buffer *l;
- u64 min_search_start;
/*
* We don't want to overwrite the superblock on the drive nor any area
* used by the boot loader (grub for example), so we make sure to start
* at an offset of at least 1MB.
*/
- min_search_start = max(fs_info->alloc_start, 1024ull * 1024);
- search_start = max(search_start, min_search_start);
+ search_start = max_t(u64, search_start, SZ_1M);
path = btrfs_alloc_path();
if (!path)
@@ -1726,7 +1739,7 @@ out:
* Function to update ctime/mtime for a given device path.
* Mainly used for ctime/mtime based probe like libblkid.
*/
-static void update_dev_time(char *path_name)
+static void update_dev_time(const char *path_name)
{
struct file *filp;
@@ -1852,7 +1865,8 @@ void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
fs_info->fs_devices->latest_bdev = next_device->bdev;
}
-int btrfs_rm_device(struct btrfs_fs_info *fs_info, char *device_path, u64 devid)
+int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
+ u64 devid)
{
struct btrfs_device *device;
struct btrfs_fs_devices *cur_devices;
@@ -2092,7 +2106,7 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
}
static int btrfs_find_device_by_path(struct btrfs_fs_info *fs_info,
- char *device_path,
+ const char *device_path,
struct btrfs_device **device)
{
int ret = 0;
@@ -2119,7 +2133,7 @@ static int btrfs_find_device_by_path(struct btrfs_fs_info *fs_info,
}
int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
- char *device_path,
+ const char *device_path,
struct btrfs_device **device)
{
*device = NULL;
@@ -2152,7 +2166,8 @@ int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
* Lookup a device given by device id, or the path if the id is 0.
*/
int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid,
- char *devpath, struct btrfs_device **device)
+ const char *devpath,
+ struct btrfs_device **device)
{
int ret;
@@ -2308,7 +2323,7 @@ error:
return ret;
}
-int btrfs_init_new_device(struct btrfs_fs_info *fs_info, char *device_path)
+int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path)
{
struct btrfs_root *root = fs_info->dev_root;
struct request_queue *q;
@@ -2382,7 +2397,8 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, char *device_path)
device->io_width = fs_info->sectorsize;
device->io_align = fs_info->sectorsize;
device->sector_size = fs_info->sectorsize;
- device->total_bytes = i_size_read(bdev->bd_inode);
+ device->total_bytes = round_down(i_size_read(bdev->bd_inode),
+ fs_info->sectorsize);
device->disk_total_bytes = device->total_bytes;
device->commit_total_bytes = device->total_bytes;
device->fs_info = fs_info;
@@ -2412,16 +2428,14 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, char *device_path)
fs_info->fs_devices->total_devices++;
fs_info->fs_devices->total_rw_bytes += device->total_bytes;
- spin_lock(&fs_info->free_chunk_lock);
- fs_info->free_chunk_space += device->total_bytes;
- spin_unlock(&fs_info->free_chunk_lock);
+ atomic64_add(device->total_bytes, &fs_info->free_chunk_space);
- if (!blk_queue_nonrot(bdev_get_queue(bdev)))
+ if (!blk_queue_nonrot(q))
fs_info->fs_devices->rotating = 1;
tmp = btrfs_super_total_bytes(fs_info->super_copy);
btrfs_set_super_total_bytes(fs_info->super_copy,
- tmp + device->total_bytes);
+ round_down(tmp + device->total_bytes, fs_info->sectorsize));
tmp = btrfs_super_num_devices(fs_info->super_copy);
btrfs_set_super_num_devices(fs_info->super_copy, tmp + 1);
@@ -2440,7 +2454,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, char *device_path)
if (seeding_dev) {
mutex_lock(&fs_info->chunk_mutex);
- ret = init_first_rw_device(trans, fs_info, device);
+ ret = init_first_rw_device(trans, fs_info);
mutex_unlock(&fs_info->chunk_mutex);
if (ret) {
btrfs_abort_transaction(trans, ret);
@@ -2516,7 +2530,7 @@ error:
}
int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
- char *device_path,
+ const char *device_path,
struct btrfs_device *srcdev,
struct btrfs_device **device_out)
{
@@ -2569,7 +2583,7 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
goto error;
}
- name = rcu_string_strdup(device_path, GFP_NOFS);
+ name = rcu_string_strdup(device_path, GFP_KERNEL);
if (!name) {
kfree(device);
ret = -ENOMEM;
@@ -2684,6 +2698,8 @@ int btrfs_grow_device(struct btrfs_trans_handle *trans,
if (!device->writeable)
return -EACCES;
+ new_size = round_down(new_size, fs_info->sectorsize);
+
mutex_lock(&fs_info->chunk_mutex);
old_total = btrfs_super_total_bytes(super_copy);
diff = new_size - device->total_bytes;
@@ -2696,7 +2712,8 @@ int btrfs_grow_device(struct btrfs_trans_handle *trans,
fs_devices = fs_info->fs_devices;
- btrfs_set_super_total_bytes(super_copy, old_total + diff);
+ btrfs_set_super_total_bytes(super_copy,
+ round_down(old_total + diff, fs_info->sectorsize));
device->fs_devices->total_rw_bytes += diff;
btrfs_device_set_total_bytes(device, new_size);
@@ -2794,10 +2811,38 @@ static int btrfs_del_sys_chunk(struct btrfs_fs_info *fs_info,
return ret;
}
+static struct extent_map *get_chunk_map(struct btrfs_fs_info *fs_info,
+ u64 logical, u64 length)
+{
+ struct extent_map_tree *em_tree;
+ struct extent_map *em;
+
+ em_tree = &fs_info->mapping_tree.map_tree;
+ read_lock(&em_tree->lock);
+ em = lookup_extent_mapping(em_tree, logical, length);
+ read_unlock(&em_tree->lock);
+
+ if (!em) {
+ btrfs_crit(fs_info, "unable to find logical %llu length %llu",
+ logical, length);
+ return ERR_PTR(-EINVAL);
+ }
+
+ if (em->start > logical || em->start + em->len < logical) {
+ btrfs_crit(fs_info,
+ "found a bad mapping, wanted %llu-%llu, found %llu-%llu",
+ logical, length, em->start, em->start + em->len);
+ free_extent_map(em);
+ return ERR_PTR(-EINVAL);
+ }
+
+ /* callers are responsible for dropping em's ref. */
+ return em;
+}
+
int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info, u64 chunk_offset)
{
- struct extent_map_tree *em_tree;
struct extent_map *em;
struct map_lookup *map;
u64 dev_extent_len = 0;
@@ -2805,23 +2850,15 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
int i, ret = 0;
struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
- em_tree = &fs_info->mapping_tree.map_tree;
-
- read_lock(&em_tree->lock);
- em = lookup_extent_mapping(em_tree, chunk_offset, 1);
- read_unlock(&em_tree->lock);
-
- if (!em || em->start > chunk_offset ||
- em->start + em->len < chunk_offset) {
+ em = get_chunk_map(fs_info, chunk_offset, 1);
+ if (IS_ERR(em)) {
/*
* This is a logic error, but we don't want to just rely on the
* user having built with ASSERT enabled, so if ASSERT doesn't
* do anything we still error out.
*/
ASSERT(0);
- if (em)
- free_extent_map(em);
- return -EINVAL;
+ return PTR_ERR(em);
}
map = em->map_lookup;
mutex_lock(&fs_info->chunk_mutex);
@@ -2849,9 +2886,7 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
mutex_lock(&fs_info->chunk_mutex);
btrfs_device_set_bytes_used(device,
device->bytes_used - dev_extent_len);
- spin_lock(&fs_info->free_chunk_lock);
- fs_info->free_chunk_space += dev_extent_len;
- spin_unlock(&fs_info->free_chunk_lock);
+ atomic64_add(dev_extent_len, &fs_info->free_chunk_space);
btrfs_clear_space_info_full(fs_info);
mutex_unlock(&fs_info->chunk_mutex);
}
@@ -3735,7 +3770,7 @@ static void __cancel_balance(struct btrfs_fs_info *fs_info)
if (ret)
btrfs_handle_fs_error(fs_info, ret, NULL);
- atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
+ clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
}
/* Non-zero return value signifies invalidity */
@@ -3754,6 +3789,7 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
struct btrfs_ioctl_balance_args *bargs)
{
struct btrfs_fs_info *fs_info = bctl->fs_info;
+ u64 meta_target, data_target;
u64 allowed;
int mixed = 0;
int ret;
@@ -3850,11 +3886,16 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
}
} while (read_seqretry(&fs_info->profiles_lock, seq));
- if (btrfs_get_num_tolerated_disk_barrier_failures(bctl->meta.target) <
- btrfs_get_num_tolerated_disk_barrier_failures(bctl->data.target)) {
+ /* if we're not converting, the target field is uninitialized */
+ meta_target = (bctl->meta.flags & BTRFS_BALANCE_ARGS_CONVERT) ?
+ bctl->meta.target : fs_info->avail_metadata_alloc_bits;
+ data_target = (bctl->data.flags & BTRFS_BALANCE_ARGS_CONVERT) ?
+ bctl->data.target : fs_info->avail_data_alloc_bits;
+ if (btrfs_get_num_tolerated_disk_barrier_failures(meta_target) <
+ btrfs_get_num_tolerated_disk_barrier_failures(data_target)) {
btrfs_warn(fs_info,
"metadata profile 0x%llx has lower redundancy than data profile 0x%llx",
- bctl->meta.target, bctl->data.target);
+ meta_target, data_target);
}
if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) {
@@ -3909,7 +3950,7 @@ out:
__cancel_balance(fs_info);
else {
kfree(bctl);
- atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
+ clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
}
return ret;
}
@@ -3999,7 +4040,7 @@ int btrfs_recover_balance(struct btrfs_fs_info *fs_info)
btrfs_balance_sys(leaf, item, &disk_bargs);
btrfs_disk_balance_args_to_cpu(&bctl->sys, &disk_bargs);
- WARN_ON(atomic_xchg(&fs_info->mutually_exclusive_operation_running, 1));
+ WARN_ON(test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags));
mutex_lock(&fs_info->volume_mutex);
mutex_lock(&fs_info->balance_mutex);
@@ -4362,7 +4403,10 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
struct btrfs_super_block *super_copy = fs_info->super_copy;
u64 old_total = btrfs_super_total_bytes(super_copy);
u64 old_size = btrfs_device_get_total_bytes(device);
- u64 diff = old_size - new_size;
+ u64 diff;
+
+ new_size = round_down(new_size, fs_info->sectorsize);
+ diff = old_size - new_size;
if (device->is_tgtdev_for_dev_replace)
return -EINVAL;
@@ -4378,9 +4422,7 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
btrfs_device_set_total_bytes(device, new_size);
if (device->writeable) {
device->fs_devices->total_rw_bytes -= diff;
- spin_lock(&fs_info->free_chunk_lock);
- fs_info->free_chunk_space -= diff;
- spin_unlock(&fs_info->free_chunk_lock);
+ atomic64_sub(diff, &fs_info->free_chunk_space);
}
mutex_unlock(&fs_info->chunk_mutex);
@@ -4491,7 +4533,8 @@ again:
&fs_info->fs_devices->resized_devices);
WARN_ON(diff > old_total);
- btrfs_set_super_total_bytes(super_copy, old_total - diff);
+ btrfs_set_super_total_bytes(super_copy,
+ round_down(old_total - diff, fs_info->sectorsize));
mutex_unlock(&fs_info->chunk_mutex);
/* Now btrfs_update_device() will change the on-disk size. */
@@ -4504,9 +4547,7 @@ done:
btrfs_device_set_total_bytes(device, old_size);
if (device->writeable)
device->fs_devices->total_rw_bytes += diff;
- spin_lock(&fs_info->free_chunk_lock);
- fs_info->free_chunk_space += diff;
- spin_unlock(&fs_info->free_chunk_lock);
+ atomic64_add(diff, &fs_info->free_chunk_space);
mutex_unlock(&fs_info->chunk_mutex);
}
return ret;
@@ -4584,8 +4625,7 @@ static void check_raid56_incompat_flag(struct btrfs_fs_info *info, u64 type)
/ sizeof(struct btrfs_stripe) + 1)
static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info, u64 start,
- u64 type)
+ u64 start, u64 type)
{
struct btrfs_fs_info *info = trans->fs_info;
struct btrfs_fs_devices *fs_devices = info->fs_devices;
@@ -4785,7 +4825,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
stripe_size = div_u64(stripe_size, dev_stripes);
/* align to BTRFS_STRIPE_LEN */
- stripe_size = div_u64(stripe_size, raid_stripe_len);
+ stripe_size = div64_u64(stripe_size, raid_stripe_len);
stripe_size *= raid_stripe_len;
map = kmalloc(map_lookup_size(num_stripes), GFP_NOFS);
@@ -4833,7 +4873,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
ret = add_extent_mapping(em_tree, em, 0);
if (!ret) {
list_add_tail(&em->list, &trans->transaction->pending_chunks);
- atomic_inc(&em->refs);
+ refcount_inc(&em->refs);
}
write_unlock(&em_tree->lock);
if (ret) {
@@ -4852,9 +4892,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
btrfs_device_set_bytes_used(map->stripes[i].dev, num_bytes);
}
- spin_lock(&info->free_chunk_lock);
- info->free_chunk_space -= (stripe_size * map->num_stripes);
- spin_unlock(&info->free_chunk_lock);
+ atomic64_sub(stripe_size * map->num_stripes, &info->free_chunk_space);
free_extent_map(em);
check_raid56_incompat_flag(info, type);
@@ -4888,7 +4926,6 @@ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
struct btrfs_device *device;
struct btrfs_chunk *chunk;
struct btrfs_stripe *stripe;
- struct extent_map_tree *em_tree;
struct extent_map *em;
struct map_lookup *map;
size_t item_size;
@@ -4897,24 +4934,9 @@ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
int i = 0;
int ret = 0;
- em_tree = &fs_info->mapping_tree.map_tree;
- read_lock(&em_tree->lock);
- em = lookup_extent_mapping(em_tree, chunk_offset, chunk_size);
- read_unlock(&em_tree->lock);
-
- if (!em) {
- btrfs_crit(fs_info, "unable to find logical %Lu len %Lu",
- chunk_offset, chunk_size);
- return -EINVAL;
- }
-
- if (em->start != chunk_offset || em->len != chunk_size) {
- btrfs_crit(fs_info,
- "found a bad mapping, wanted %Lu-%Lu, found %Lu-%Lu",
- chunk_offset, chunk_size, em->start, em->len);
- free_extent_map(em);
- return -EINVAL;
- }
+ em = get_chunk_map(fs_info, chunk_offset, chunk_size);
+ if (IS_ERR(em))
+ return PTR_ERR(em);
map = em->map_lookup;
item_size = btrfs_chunk_item_size(map->num_stripes);
@@ -5009,29 +5031,26 @@ int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
ASSERT(mutex_is_locked(&fs_info->chunk_mutex));
chunk_offset = find_next_chunk(fs_info);
- return __btrfs_alloc_chunk(trans, fs_info, chunk_offset, type);
+ return __btrfs_alloc_chunk(trans, chunk_offset, type);
}
static noinline int init_first_rw_device(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info,
- struct btrfs_device *device)
+ struct btrfs_fs_info *fs_info)
{
- struct btrfs_root *extent_root = fs_info->extent_root;
u64 chunk_offset;
u64 sys_chunk_offset;
u64 alloc_profile;
int ret;
chunk_offset = find_next_chunk(fs_info);
- alloc_profile = btrfs_get_alloc_profile(extent_root, 0);
- ret = __btrfs_alloc_chunk(trans, fs_info, chunk_offset, alloc_profile);
+ alloc_profile = btrfs_metadata_alloc_profile(fs_info);
+ ret = __btrfs_alloc_chunk(trans, chunk_offset, alloc_profile);
if (ret)
return ret;
sys_chunk_offset = find_next_chunk(fs_info);
- alloc_profile = btrfs_get_alloc_profile(fs_info->chunk_root, 0);
- ret = __btrfs_alloc_chunk(trans, fs_info, sys_chunk_offset,
- alloc_profile);
+ alloc_profile = btrfs_system_alloc_profile(fs_info);
+ ret = __btrfs_alloc_chunk(trans, sys_chunk_offset, alloc_profile);
return ret;
}
@@ -5057,15 +5076,12 @@ int btrfs_chunk_readonly(struct btrfs_fs_info *fs_info, u64 chunk_offset)
{
struct extent_map *em;
struct map_lookup *map;
- struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
int readonly = 0;
int miss_ndevs = 0;
int i;
- read_lock(&map_tree->map_tree.lock);
- em = lookup_extent_mapping(&map_tree->map_tree, chunk_offset, 1);
- read_unlock(&map_tree->map_tree.lock);
- if (!em)
+ em = get_chunk_map(fs_info, chunk_offset, 1);
+ if (IS_ERR(em))
return 1;
map = em->map_lookup;
@@ -5119,34 +5135,19 @@ void btrfs_mapping_tree_free(struct btrfs_mapping_tree *tree)
int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
{
- struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
struct extent_map *em;
struct map_lookup *map;
- struct extent_map_tree *em_tree = &map_tree->map_tree;
int ret;
- read_lock(&em_tree->lock);
- em = lookup_extent_mapping(em_tree, logical, len);
- read_unlock(&em_tree->lock);
-
- /*
- * We could return errors for these cases, but that could get ugly and
- * we'd probably do the same thing which is just not do anything else
- * and exit, so return 1 so the callers don't try to use other copies.
- */
- if (!em) {
- btrfs_crit(fs_info, "No mapping for %Lu-%Lu", logical,
- logical+len);
- return 1;
- }
-
- if (em->start > logical || em->start + em->len < logical) {
- btrfs_crit(fs_info, "Invalid mapping for %Lu-%Lu, got %Lu-%Lu",
- logical, logical+len, em->start,
- em->start + em->len);
- free_extent_map(em);
+ em = get_chunk_map(fs_info, logical, len);
+ if (IS_ERR(em))
+ /*
+ * We could return errors for these cases, but that could get
+ * ugly and we'd probably do the same thing which is just not do
+ * anything else and exit, so return 1 so the callers don't try
+ * to use other copies.
+ */
return 1;
- }
map = em->map_lookup;
if (map->type & (BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID1))
@@ -5162,7 +5163,8 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
free_extent_map(em);
btrfs_dev_replace_lock(&fs_info->dev_replace, 0);
- if (btrfs_dev_replace_is_ongoing(&fs_info->dev_replace))
+ if (btrfs_dev_replace_is_ongoing(&fs_info->dev_replace) &&
+ fs_info->dev_replace.tgtdev)
ret++;
btrfs_dev_replace_unlock(&fs_info->dev_replace, 0);
@@ -5175,15 +5177,11 @@ unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
{
struct extent_map *em;
struct map_lookup *map;
- struct extent_map_tree *em_tree = &map_tree->map_tree;
unsigned long len = fs_info->sectorsize;
- read_lock(&em_tree->lock);
- em = lookup_extent_mapping(em_tree, logical, len);
- read_unlock(&em_tree->lock);
- BUG_ON(!em);
+ em = get_chunk_map(fs_info, logical, len);
+ WARN_ON(IS_ERR(em));
- BUG_ON(em->start > logical || em->start + em->len < logical);
map = em->map_lookup;
if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
len = map->stripe_len * nr_data_stripes(map);
@@ -5191,20 +5189,16 @@ unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
return len;
}
-int btrfs_is_parity_mirror(struct btrfs_mapping_tree *map_tree,
+int btrfs_is_parity_mirror(struct btrfs_fs_info *fs_info,
u64 logical, u64 len, int mirror_num)
{
struct extent_map *em;
struct map_lookup *map;
- struct extent_map_tree *em_tree = &map_tree->map_tree;
int ret = 0;
- read_lock(&em_tree->lock);
- em = lookup_extent_mapping(em_tree, logical, len);
- read_unlock(&em_tree->lock);
- BUG_ON(!em);
+ em = get_chunk_map(fs_info, logical, len);
+ WARN_ON(IS_ERR(em));
- BUG_ON(em->start > logical || em->start + em->len < logical);
map = em->map_lookup;
if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
ret = 1;
@@ -5297,25 +5291,353 @@ static struct btrfs_bio *alloc_btrfs_bio(int total_stripes, int real_stripes)
GFP_NOFS|__GFP_NOFAIL);
atomic_set(&bbio->error, 0);
- atomic_set(&bbio->refs, 1);
+ refcount_set(&bbio->refs, 1);
return bbio;
}
void btrfs_get_bbio(struct btrfs_bio *bbio)
{
- WARN_ON(!atomic_read(&bbio->refs));
- atomic_inc(&bbio->refs);
+ WARN_ON(!refcount_read(&bbio->refs));
+ refcount_inc(&bbio->refs);
}
void btrfs_put_bbio(struct btrfs_bio *bbio)
{
if (!bbio)
return;
- if (atomic_dec_and_test(&bbio->refs))
+ if (refcount_dec_and_test(&bbio->refs))
kfree(bbio);
}
+/* can REQ_OP_DISCARD be sent with other REQ like REQ_OP_WRITE? */
+/*
+ * Please note that, discard won't be sent to target device of device
+ * replace.
+ */
+static int __btrfs_map_block_for_discard(struct btrfs_fs_info *fs_info,
+ u64 logical, u64 length,
+ struct btrfs_bio **bbio_ret)
+{
+ struct extent_map *em;
+ struct map_lookup *map;
+ struct btrfs_bio *bbio;
+ u64 offset;
+ u64 stripe_nr;
+ u64 stripe_nr_end;
+ u64 stripe_end_offset;
+ u64 stripe_cnt;
+ u64 stripe_len;
+ u64 stripe_offset;
+ u64 num_stripes;
+ u32 stripe_index;
+ u32 factor = 0;
+ u32 sub_stripes = 0;
+ u64 stripes_per_dev = 0;
+ u32 remaining_stripes = 0;
+ u32 last_stripe = 0;
+ int ret = 0;
+ int i;
+
+ /* discard always return a bbio */
+ ASSERT(bbio_ret);
+
+ em = get_chunk_map(fs_info, logical, length);
+ if (IS_ERR(em))
+ return PTR_ERR(em);
+
+ map = em->map_lookup;
+ /* we don't discard raid56 yet */
+ if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
+ ret = -EOPNOTSUPP;
+ goto out;
+ }
+
+ offset = logical - em->start;
+ length = min_t(u64, em->len - offset, length);
+
+ stripe_len = map->stripe_len;
+ /*
+ * stripe_nr counts the total number of stripes we have to stride
+ * to get to this block
+ */
+ stripe_nr = div64_u64(offset, stripe_len);
+
+ /* stripe_offset is the offset of this block in its stripe */
+ stripe_offset = offset - stripe_nr * stripe_len;
+
+ stripe_nr_end = round_up(offset + length, map->stripe_len);
+ stripe_nr_end = div64_u64(stripe_nr_end, map->stripe_len);
+ stripe_cnt = stripe_nr_end - stripe_nr;
+ stripe_end_offset = stripe_nr_end * map->stripe_len -
+ (offset + length);
+ /*
+ * after this, stripe_nr is the number of stripes on this
+ * device we have to walk to find the data, and stripe_index is
+ * the number of our device in the stripe array
+ */
+ num_stripes = 1;
+ stripe_index = 0;
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID0 |
+ BTRFS_BLOCK_GROUP_RAID10)) {
+ if (map->type & BTRFS_BLOCK_GROUP_RAID0)
+ sub_stripes = 1;
+ else
+ sub_stripes = map->sub_stripes;
+
+ factor = map->num_stripes / sub_stripes;
+ num_stripes = min_t(u64, map->num_stripes,
+ sub_stripes * stripe_cnt);
+ stripe_nr = div_u64_rem(stripe_nr, factor, &stripe_index);
+ stripe_index *= sub_stripes;
+ stripes_per_dev = div_u64_rem(stripe_cnt, factor,
+ &remaining_stripes);
+ div_u64_rem(stripe_nr_end - 1, factor, &last_stripe);
+ last_stripe *= sub_stripes;
+ } else if (map->type & (BTRFS_BLOCK_GROUP_RAID1 |
+ BTRFS_BLOCK_GROUP_DUP)) {
+ num_stripes = map->num_stripes;
+ } else {
+ stripe_nr = div_u64_rem(stripe_nr, map->num_stripes,
+ &stripe_index);
+ }
+
+ bbio = alloc_btrfs_bio(num_stripes, 0);
+ if (!bbio) {
+ ret = -ENOMEM;
+ goto out;
+ }
+
+ for (i = 0; i < num_stripes; i++) {
+ bbio->stripes[i].physical =
+ map->stripes[stripe_index].physical +
+ stripe_offset + stripe_nr * map->stripe_len;
+ bbio->stripes[i].dev = map->stripes[stripe_index].dev;
+
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID0 |
+ BTRFS_BLOCK_GROUP_RAID10)) {
+ bbio->stripes[i].length = stripes_per_dev *
+ map->stripe_len;
+
+ if (i / sub_stripes < remaining_stripes)
+ bbio->stripes[i].length +=
+ map->stripe_len;
+
+ /*
+ * Special for the first stripe and
+ * the last stripe:
+ *
+ * |-------|...|-------|
+ * |----------|
+ * off end_off
+ */
+ if (i < sub_stripes)
+ bbio->stripes[i].length -=
+ stripe_offset;
+
+ if (stripe_index >= last_stripe &&
+ stripe_index <= (last_stripe +
+ sub_stripes - 1))
+ bbio->stripes[i].length -=
+ stripe_end_offset;
+
+ if (i == sub_stripes - 1)
+ stripe_offset = 0;
+ } else {
+ bbio->stripes[i].length = length;
+ }
+
+ stripe_index++;
+ if (stripe_index == map->num_stripes) {
+ stripe_index = 0;
+ stripe_nr++;
+ }
+ }
+
+ *bbio_ret = bbio;
+ bbio->map_type = map->type;
+ bbio->num_stripes = num_stripes;
+out:
+ free_extent_map(em);
+ return ret;
+}
+
+/*
+ * In dev-replace case, for repair case (that's the only case where the mirror
+ * is selected explicitly when calling btrfs_map_block), blocks left of the
+ * left cursor can also be read from the target drive.
+ *
+ * For REQ_GET_READ_MIRRORS, the target drive is added as the last one to the
+ * array of stripes.
+ * For READ, it also needs to be supported using the same mirror number.
+ *
+ * If the requested block is not left of the left cursor, EIO is returned. This
+ * can happen because btrfs_num_copies() returns one more in the dev-replace
+ * case.
+ */
+static int get_extra_mirror_from_replace(struct btrfs_fs_info *fs_info,
+ u64 logical, u64 length,
+ u64 srcdev_devid, int *mirror_num,
+ u64 *physical)
+{
+ struct btrfs_bio *bbio = NULL;
+ int num_stripes;
+ int index_srcdev = 0;
+ int found = 0;
+ u64 physical_of_found = 0;
+ int i;
+ int ret = 0;
+
+ ret = __btrfs_map_block(fs_info, BTRFS_MAP_GET_READ_MIRRORS,
+ logical, &length, &bbio, 0, 0);
+ if (ret) {
+ ASSERT(bbio == NULL);
+ return ret;
+ }
+
+ num_stripes = bbio->num_stripes;
+ if (*mirror_num > num_stripes) {
+ /*
+ * BTRFS_MAP_GET_READ_MIRRORS does not contain this mirror,
+ * that means that the requested area is not left of the left
+ * cursor
+ */
+ btrfs_put_bbio(bbio);
+ return -EIO;
+ }
+
+ /*
+ * process the rest of the function using the mirror_num of the source
+ * drive. Therefore look it up first. At the end, patch the device
+ * pointer to the one of the target drive.
+ */
+ for (i = 0; i < num_stripes; i++) {
+ if (bbio->stripes[i].dev->devid != srcdev_devid)
+ continue;
+
+ /*
+ * In case of DUP, in order to keep it simple, only add the
+ * mirror with the lowest physical address
+ */
+ if (found &&
+ physical_of_found <= bbio->stripes[i].physical)
+ continue;
+
+ index_srcdev = i;
+ found = 1;
+ physical_of_found = bbio->stripes[i].physical;
+ }
+
+ btrfs_put_bbio(bbio);
+
+ ASSERT(found);
+ if (!found)
+ return -EIO;
+
+ *mirror_num = index_srcdev + 1;
+ *physical = physical_of_found;
+ return ret;
+}
+
+static void handle_ops_on_dev_replace(enum btrfs_map_op op,
+ struct btrfs_bio **bbio_ret,
+ struct btrfs_dev_replace *dev_replace,
+ int *num_stripes_ret, int *max_errors_ret)
+{
+ struct btrfs_bio *bbio = *bbio_ret;
+ u64 srcdev_devid = dev_replace->srcdev->devid;
+ int tgtdev_indexes = 0;
+ int num_stripes = *num_stripes_ret;
+ int max_errors = *max_errors_ret;
+ int i;
+
+ if (op == BTRFS_MAP_WRITE) {
+ int index_where_to_add;
+
+ /*
+ * duplicate the write operations while the dev replace
+ * procedure is running. Since the copying of the old disk to
+ * the new disk takes place at run time while the filesystem is
+ * mounted writable, the regular write operations to the old
+ * disk have to be duplicated to go to the new disk as well.
+ *
+ * Note that device->missing is handled by the caller, and that
+ * the write to the old disk is already set up in the stripes
+ * array.
+ */
+ index_where_to_add = num_stripes;
+ for (i = 0; i < num_stripes; i++) {
+ if (bbio->stripes[i].dev->devid == srcdev_devid) {
+ /* write to new disk, too */
+ struct btrfs_bio_stripe *new =
+ bbio->stripes + index_where_to_add;
+ struct btrfs_bio_stripe *old =
+ bbio->stripes + i;
+
+ new->physical = old->physical;
+ new->length = old->length;
+ new->dev = dev_replace->tgtdev;
+ bbio->tgtdev_map[i] = index_where_to_add;
+ index_where_to_add++;
+ max_errors++;
+ tgtdev_indexes++;
+ }
+ }
+ num_stripes = index_where_to_add;
+ } else if (op == BTRFS_MAP_GET_READ_MIRRORS) {
+ int index_srcdev = 0;
+ int found = 0;
+ u64 physical_of_found = 0;
+
+ /*
+ * During the dev-replace procedure, the target drive can also
+ * be used to read data in case it is needed to repair a corrupt
+ * block elsewhere. This is possible if the requested area is
+ * left of the left cursor. In this area, the target drive is a
+ * full copy of the source drive.
+ */
+ for (i = 0; i < num_stripes; i++) {
+ if (bbio->stripes[i].dev->devid == srcdev_devid) {
+ /*
+ * In case of DUP, in order to keep it simple,
+ * only add the mirror with the lowest physical
+ * address
+ */
+ if (found &&
+ physical_of_found <=
+ bbio->stripes[i].physical)
+ continue;
+ index_srcdev = i;
+ found = 1;
+ physical_of_found = bbio->stripes[i].physical;
+ }
+ }
+ if (found) {
+ struct btrfs_bio_stripe *tgtdev_stripe =
+ bbio->stripes + num_stripes;
+
+ tgtdev_stripe->physical = physical_of_found;
+ tgtdev_stripe->length =
+ bbio->stripes[index_srcdev].length;
+ tgtdev_stripe->dev = dev_replace->tgtdev;
+ bbio->tgtdev_map[index_srcdev] = num_stripes;
+
+ tgtdev_indexes++;
+ num_stripes++;
+ }
+ }
+
+ *num_stripes_ret = num_stripes;
+ *max_errors_ret = max_errors;
+ bbio->num_tgtdevs = tgtdev_indexes;
+ *bbio_ret = bbio;
+}
+
+static bool need_full_stripe(enum btrfs_map_op op)
+{
+ return (op == BTRFS_MAP_WRITE || op == BTRFS_MAP_GET_READ_MIRRORS);
+}
+
static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
enum btrfs_map_op op,
u64 logical, u64 *length,
@@ -5324,14 +5646,9 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
{
struct extent_map *em;
struct map_lookup *map;
- struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
- struct extent_map_tree *em_tree = &map_tree->map_tree;
u64 offset;
u64 stripe_offset;
- u64 stripe_end_offset;
u64 stripe_nr;
- u64 stripe_nr_orig;
- u64 stripe_nr_end;
u64 stripe_len;
u32 stripe_index;
int i;
@@ -5347,23 +5664,13 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
u64 physical_to_patch_in_first_stripe = 0;
u64 raid56_full_stripe_start = (u64)-1;
- read_lock(&em_tree->lock);
- em = lookup_extent_mapping(em_tree, logical, *length);
- read_unlock(&em_tree->lock);
-
- if (!em) {
- btrfs_crit(fs_info, "unable to find logical %llu len %llu",
- logical, *length);
- return -EINVAL;
- }
+ if (op == BTRFS_MAP_DISCARD)
+ return __btrfs_map_block_for_discard(fs_info, logical,
+ *length, bbio_ret);
- if (em->start > logical || em->start + em->len < logical) {
- btrfs_crit(fs_info,
- "found a bad mapping, wanted %Lu, found %Lu-%Lu",
- logical, em->start, em->start + em->len);
- free_extent_map(em);
- return -EINVAL;
- }
+ em = get_chunk_map(fs_info, logical, *length);
+ if (IS_ERR(em))
+ return PTR_ERR(em);
map = em->map_lookup;
offset = logical - em->start;
@@ -5402,14 +5709,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
raid56_full_stripe_start *= full_stripe_len;
}
- if (op == BTRFS_MAP_DISCARD) {
- /* we don't discard raid56 yet */
- if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
- ret = -EOPNOTSUPP;
- goto out;
- }
- *length = min_t(u64, em->len - offset, *length);
- } else if (map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
+ if (map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
u64 max_len;
/* For writes to RAID[56], allow a full stripeset across all disks.
For other RAID types and for RAID[56] reads, just allow a single
@@ -5440,105 +5740,28 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
btrfs_dev_replace_set_lock_blocking(dev_replace);
if (dev_replace_is_ongoing && mirror_num == map->num_stripes + 1 &&
- op != BTRFS_MAP_WRITE && op != BTRFS_MAP_DISCARD &&
- op != BTRFS_MAP_GET_READ_MIRRORS && dev_replace->tgtdev != NULL) {
- /*
- * in dev-replace case, for repair case (that's the only
- * case where the mirror is selected explicitly when
- * calling btrfs_map_block), blocks left of the left cursor
- * can also be read from the target drive.
- * For REQ_GET_READ_MIRRORS, the target drive is added as
- * the last one to the array of stripes. For READ, it also
- * needs to be supported using the same mirror number.
- * If the requested block is not left of the left cursor,
- * EIO is returned. This can happen because btrfs_num_copies()
- * returns one more in the dev-replace case.
- */
- u64 tmp_length = *length;
- struct btrfs_bio *tmp_bbio = NULL;
- int tmp_num_stripes;
- u64 srcdev_devid = dev_replace->srcdev->devid;
- int index_srcdev = 0;
- int found = 0;
- u64 physical_of_found = 0;
-
- ret = __btrfs_map_block(fs_info, BTRFS_MAP_GET_READ_MIRRORS,
- logical, &tmp_length, &tmp_bbio, 0, 0);
- if (ret) {
- WARN_ON(tmp_bbio != NULL);
- goto out;
- }
-
- tmp_num_stripes = tmp_bbio->num_stripes;
- if (mirror_num > tmp_num_stripes) {
- /*
- * BTRFS_MAP_GET_READ_MIRRORS does not contain this
- * mirror, that means that the requested area
- * is not left of the left cursor
- */
- ret = -EIO;
- btrfs_put_bbio(tmp_bbio);
- goto out;
- }
-
- /*
- * process the rest of the function using the mirror_num
- * of the source drive. Therefore look it up first.
- * At the end, patch the device pointer to the one of the
- * target drive.
- */
- for (i = 0; i < tmp_num_stripes; i++) {
- if (tmp_bbio->stripes[i].dev->devid != srcdev_devid)
- continue;
-
- /*
- * In case of DUP, in order to keep it simple, only add
- * the mirror with the lowest physical address
- */
- if (found &&
- physical_of_found <= tmp_bbio->stripes[i].physical)
- continue;
-
- index_srcdev = i;
- found = 1;
- physical_of_found = tmp_bbio->stripes[i].physical;
- }
-
- btrfs_put_bbio(tmp_bbio);
-
- if (!found) {
- WARN_ON(1);
- ret = -EIO;
+ !need_full_stripe(op) && dev_replace->tgtdev != NULL) {
+ ret = get_extra_mirror_from_replace(fs_info, logical, *length,
+ dev_replace->srcdev->devid,
+ &mirror_num,
+ &physical_to_patch_in_first_stripe);
+ if (ret)
goto out;
- }
-
- mirror_num = index_srcdev + 1;
- patch_the_first_stripe_for_dev_replace = 1;
- physical_to_patch_in_first_stripe = physical_of_found;
+ else
+ patch_the_first_stripe_for_dev_replace = 1;
} else if (mirror_num > map->num_stripes) {
mirror_num = 0;
}
num_stripes = 1;
stripe_index = 0;
- stripe_nr_orig = stripe_nr;
- stripe_nr_end = ALIGN(offset + *length, map->stripe_len);
- stripe_nr_end = div_u64(stripe_nr_end, map->stripe_len);
- stripe_end_offset = stripe_nr_end * map->stripe_len -
- (offset + *length);
-
if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
- if (op == BTRFS_MAP_DISCARD)
- num_stripes = min_t(u64, map->num_stripes,
- stripe_nr_end - stripe_nr_orig);
stripe_nr = div_u64_rem(stripe_nr, map->num_stripes,
&stripe_index);
- if (op != BTRFS_MAP_WRITE && op != BTRFS_MAP_DISCARD &&
- op != BTRFS_MAP_GET_READ_MIRRORS)
+ if (op != BTRFS_MAP_WRITE && op != BTRFS_MAP_GET_READ_MIRRORS)
mirror_num = 1;
} else if (map->type & BTRFS_BLOCK_GROUP_RAID1) {
- if (op == BTRFS_MAP_WRITE || op == BTRFS_MAP_DISCARD ||
- op == BTRFS_MAP_GET_READ_MIRRORS)
+ if (op == BTRFS_MAP_WRITE || op == BTRFS_MAP_GET_READ_MIRRORS)
num_stripes = map->num_stripes;
else if (mirror_num)
stripe_index = mirror_num - 1;
@@ -5551,8 +5774,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
}
} else if (map->type & BTRFS_BLOCK_GROUP_DUP) {
- if (op == BTRFS_MAP_WRITE || op == BTRFS_MAP_DISCARD ||
- op == BTRFS_MAP_GET_READ_MIRRORS) {
+ if (op == BTRFS_MAP_WRITE || op == BTRFS_MAP_GET_READ_MIRRORS) {
num_stripes = map->num_stripes;
} else if (mirror_num) {
stripe_index = mirror_num - 1;
@@ -5568,10 +5790,6 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
if (op == BTRFS_MAP_WRITE || op == BTRFS_MAP_GET_READ_MIRRORS)
num_stripes = map->sub_stripes;
- else if (op == BTRFS_MAP_DISCARD)
- num_stripes = min_t(u64, map->sub_stripes *
- (stripe_nr_end - stripe_nr_orig),
- map->num_stripes);
else if (mirror_num)
stripe_index += mirror_num - 1;
else {
@@ -5589,7 +5807,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
(op == BTRFS_MAP_WRITE || op == BTRFS_MAP_GET_READ_MIRRORS ||
mirror_num > 1)) {
/* push stripe_nr back to the start of the full stripe */
- stripe_nr = div_u64(raid56_full_stripe_start,
+ stripe_nr = div64_u64(raid56_full_stripe_start,
stripe_len * nr_data_stripes(map));
/* RAID[56] write or recovery. Return all stripes */
@@ -5614,8 +5832,9 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
/* We distribute the parity blocks across stripes */
div_u64_rem(stripe_nr + stripe_index, map->num_stripes,
&stripe_index);
- if ((op != BTRFS_MAP_WRITE && op != BTRFS_MAP_DISCARD &&
- op != BTRFS_MAP_GET_READ_MIRRORS) && mirror_num <= 1)
+ if ((op != BTRFS_MAP_WRITE &&
+ op != BTRFS_MAP_GET_READ_MIRRORS) &&
+ mirror_num <= 1)
mirror_num = 1;
}
} else {
@@ -5637,8 +5856,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
}
num_alloc_stripes = num_stripes;
- if (dev_replace_is_ongoing) {
- if (op == BTRFS_MAP_WRITE || op == BTRFS_MAP_DISCARD)
+ if (dev_replace_is_ongoing && dev_replace->tgtdev != NULL) {
+ if (op == BTRFS_MAP_WRITE)
num_alloc_stripes <<= 1;
if (op == BTRFS_MAP_GET_READ_MIRRORS)
num_alloc_stripes++;
@@ -5650,14 +5869,12 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
ret = -ENOMEM;
goto out;
}
- if (dev_replace_is_ongoing)
+ if (dev_replace_is_ongoing && dev_replace->tgtdev != NULL)
bbio->tgtdev_map = (int *)(bbio->stripes + num_alloc_stripes);
/* build raid_map */
- if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK &&
- need_raid_map &&
- ((op == BTRFS_MAP_WRITE || op == BTRFS_MAP_GET_READ_MIRRORS) ||
- mirror_num > 1)) {
+ if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK && need_raid_map &&
+ (need_full_stripe(op) || mirror_num > 1)) {
u64 tmp;
unsigned rot;
@@ -5681,173 +5898,27 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
RAID6_Q_STRIPE;
}
- if (op == BTRFS_MAP_DISCARD) {
- u32 factor = 0;
- u32 sub_stripes = 0;
- u64 stripes_per_dev = 0;
- u32 remaining_stripes = 0;
- u32 last_stripe = 0;
- if (map->type &
- (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID10)) {
- if (map->type & BTRFS_BLOCK_GROUP_RAID0)
- sub_stripes = 1;
- else
- sub_stripes = map->sub_stripes;
-
- factor = map->num_stripes / sub_stripes;
- stripes_per_dev = div_u64_rem(stripe_nr_end -
- stripe_nr_orig,
- factor,
- &remaining_stripes);
- div_u64_rem(stripe_nr_end - 1, factor, &last_stripe);
- last_stripe *= sub_stripes;
- }
-
- for (i = 0; i < num_stripes; i++) {
- bbio->stripes[i].physical =
- map->stripes[stripe_index].physical +
- stripe_offset + stripe_nr * map->stripe_len;
- bbio->stripes[i].dev = map->stripes[stripe_index].dev;
-
- if (map->type & (BTRFS_BLOCK_GROUP_RAID0 |
- BTRFS_BLOCK_GROUP_RAID10)) {
- bbio->stripes[i].length = stripes_per_dev *
- map->stripe_len;
-
- if (i / sub_stripes < remaining_stripes)
- bbio->stripes[i].length +=
- map->stripe_len;
-
- /*
- * Special for the first stripe and
- * the last stripe:
- *
- * |-------|...|-------|
- * |----------|
- * off end_off
- */
- if (i < sub_stripes)
- bbio->stripes[i].length -=
- stripe_offset;
-
- if (stripe_index >= last_stripe &&
- stripe_index <= (last_stripe +
- sub_stripes - 1))
- bbio->stripes[i].length -=
- stripe_end_offset;
-
- if (i == sub_stripes - 1)
- stripe_offset = 0;
- } else
- bbio->stripes[i].length = *length;
-
- stripe_index++;
- if (stripe_index == map->num_stripes) {
- /* This could only happen for RAID0/10 */
- stripe_index = 0;
- stripe_nr++;
- }
- }
- } else {
- for (i = 0; i < num_stripes; i++) {
- bbio->stripes[i].physical =
- map->stripes[stripe_index].physical +
- stripe_offset +
- stripe_nr * map->stripe_len;
- bbio->stripes[i].dev =
- map->stripes[stripe_index].dev;
- stripe_index++;
- }
+ for (i = 0; i < num_stripes; i++) {
+ bbio->stripes[i].physical =
+ map->stripes[stripe_index].physical +
+ stripe_offset +
+ stripe_nr * map->stripe_len;
+ bbio->stripes[i].dev =
+ map->stripes[stripe_index].dev;
+ stripe_index++;
}
- if (op == BTRFS_MAP_WRITE || op == BTRFS_MAP_GET_READ_MIRRORS)
+ if (need_full_stripe(op))
max_errors = btrfs_chunk_max_errors(map);
if (bbio->raid_map)
sort_parity_stripes(bbio, num_stripes);
- tgtdev_indexes = 0;
- if (dev_replace_is_ongoing &&
- (op == BTRFS_MAP_WRITE || op == BTRFS_MAP_DISCARD) &&
- dev_replace->tgtdev != NULL) {
- int index_where_to_add;
- u64 srcdev_devid = dev_replace->srcdev->devid;
-
- /*
- * duplicate the write operations while the dev replace
- * procedure is running. Since the copying of the old disk
- * to the new disk takes place at run time while the
- * filesystem is mounted writable, the regular write
- * operations to the old disk have to be duplicated to go
- * to the new disk as well.
- * Note that device->missing is handled by the caller, and
- * that the write to the old disk is already set up in the
- * stripes array.
- */
- index_where_to_add = num_stripes;
- for (i = 0; i < num_stripes; i++) {
- if (bbio->stripes[i].dev->devid == srcdev_devid) {
- /* write to new disk, too */
- struct btrfs_bio_stripe *new =
- bbio->stripes + index_where_to_add;
- struct btrfs_bio_stripe *old =
- bbio->stripes + i;
-
- new->physical = old->physical;
- new->length = old->length;
- new->dev = dev_replace->tgtdev;
- bbio->tgtdev_map[i] = index_where_to_add;
- index_where_to_add++;
- max_errors++;
- tgtdev_indexes++;
- }
- }
- num_stripes = index_where_to_add;
- } else if (dev_replace_is_ongoing &&
- op == BTRFS_MAP_GET_READ_MIRRORS &&
- dev_replace->tgtdev != NULL) {
- u64 srcdev_devid = dev_replace->srcdev->devid;
- int index_srcdev = 0;
- int found = 0;
- u64 physical_of_found = 0;
-
- /*
- * During the dev-replace procedure, the target drive can
- * also be used to read data in case it is needed to repair
- * a corrupt block elsewhere. This is possible if the
- * requested area is left of the left cursor. In this area,
- * the target drive is a full copy of the source drive.
- */
- for (i = 0; i < num_stripes; i++) {
- if (bbio->stripes[i].dev->devid == srcdev_devid) {
- /*
- * In case of DUP, in order to keep it
- * simple, only add the mirror with the
- * lowest physical address
- */
- if (found &&
- physical_of_found <=
- bbio->stripes[i].physical)
- continue;
- index_srcdev = i;
- found = 1;
- physical_of_found = bbio->stripes[i].physical;
- }
- }
- if (found) {
- struct btrfs_bio_stripe *tgtdev_stripe =
- bbio->stripes + num_stripes;
-
- tgtdev_stripe->physical = physical_of_found;
- tgtdev_stripe->length =
- bbio->stripes[index_srcdev].length;
- tgtdev_stripe->dev = dev_replace->tgtdev;
- bbio->tgtdev_map[index_srcdev] = num_stripes;
-
- tgtdev_indexes++;
- num_stripes++;
- }
+ if (dev_replace_is_ongoing && dev_replace->tgtdev != NULL &&
+ need_full_stripe(op)) {
+ handle_ops_on_dev_replace(op, &bbio, dev_replace, &num_stripes,
+ &max_errors);
}
*bbio_ret = bbio;
@@ -5855,7 +5926,6 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
bbio->num_stripes = num_stripes;
bbio->max_errors = max_errors;
bbio->mirror_num = mirror_num;
- bbio->num_tgtdevs = tgtdev_indexes;
/*
* this is the case that REQ_READ && dev_replace_is_ongoing &&
@@ -5888,19 +5958,15 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
/* For Scrub/replace */
int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
u64 logical, u64 *length,
- struct btrfs_bio **bbio_ret, int mirror_num,
- int need_raid_map)
+ struct btrfs_bio **bbio_ret)
{
- return __btrfs_map_block(fs_info, op, logical, length, bbio_ret,
- mirror_num, need_raid_map);
+ return __btrfs_map_block(fs_info, op, logical, length, bbio_ret, 0, 1);
}
int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
u64 chunk_start, u64 physical, u64 devid,
u64 **logical, int *naddrs, int *stripe_len)
{
- struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
- struct extent_map_tree *em_tree = &map_tree->map_tree;
struct extent_map *em;
struct map_lookup *map;
u64 *buf;
@@ -5910,24 +5976,11 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
u64 rmap_len;
int i, j, nr = 0;
- read_lock(&em_tree->lock);
- em = lookup_extent_mapping(em_tree, chunk_start, 1);
- read_unlock(&em_tree->lock);
-
- if (!em) {
- btrfs_err(fs_info, "couldn't find em for chunk %Lu",
- chunk_start);
+ em = get_chunk_map(fs_info, chunk_start, 1);
+ if (IS_ERR(em))
return -EIO;
- }
- if (em->start != chunk_start) {
- btrfs_err(fs_info, "bad chunk start, em=%Lu, wanted=%Lu",
- em->start, chunk_start);
- free_extent_map(em);
- return -EIO;
- }
map = em->map_lookup;
-
length = em->len;
rmap_len = map->stripe_len;
@@ -5951,7 +6004,7 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
continue;
stripe_nr = physical - map->stripes[i].physical;
- stripe_nr = div_u64(stripe_nr, map->stripe_len);
+ stripe_nr = div64_u64(stripe_nr, map->stripe_len);
if (map->type & BTRFS_BLOCK_GROUP_RAID10) {
stripe_nr = stripe_nr * map->num_stripes + i;
@@ -5996,9 +6049,10 @@ static void btrfs_end_bio(struct bio *bio)
struct btrfs_bio *bbio = bio->bi_private;
int is_orig_bio = 0;
- if (bio->bi_error) {
+ if (bio->bi_status) {
atomic_inc(&bbio->error);
- if (bio->bi_error == -EIO || bio->bi_error == -EREMOTEIO) {
+ if (bio->bi_status == BLK_STS_IOERR ||
+ bio->bi_status == BLK_STS_TARGET) {
unsigned int stripe_index =
btrfs_io_bio(bio)->stripe_index;
struct btrfs_device *dev;
@@ -6036,13 +6090,13 @@ static void btrfs_end_bio(struct bio *bio)
* beyond the tolerance of the btrfs bio
*/
if (atomic_read(&bbio->error) > bbio->max_errors) {
- bio->bi_error = -EIO;
+ bio->bi_status = BLK_STS_IOERR;
} else {
/*
* this bio is actually up to date, we didn't
* go over the max number of errors
*/
- bio->bi_error = 0;
+ bio->bi_status = 0;
}
btrfs_end_bbio(bbio, bio);
@@ -6153,7 +6207,7 @@ static void bbio_error(struct btrfs_bio *bbio, struct bio *bio, u64 logical)
btrfs_io_bio(bio)->mirror_num = bbio->mirror_num;
bio->bi_iter.bi_sector = logical >> 9;
- bio->bi_error = -EIO;
+ bio->bi_status = BLK_STS_IOERR;
btrfs_end_bbio(bbio, bio);
}
}
@@ -6215,15 +6269,14 @@ int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
for (dev_nr = 0; dev_nr < total_devs; dev_nr++) {
dev = bbio->stripes[dev_nr].dev;
if (!dev || !dev->bdev ||
- (bio_op(bio) == REQ_OP_WRITE && !dev->writeable)) {
+ (bio_op(first_bio) == REQ_OP_WRITE && !dev->writeable)) {
bbio_error(bbio, first_bio, logical);
continue;
}
- if (dev_nr < total_devs - 1) {
- bio = btrfs_bio_clone(first_bio, GFP_NOFS);
- BUG_ON(!bio); /* -ENOMEM */
- } else
+ if (dev_nr < total_devs - 1)
+ bio = btrfs_bio_clone(first_bio);
+ else
bio = first_bio;
submit_stripe_bio(bbio, bio, bbio->stripes[dev_nr].physical,
@@ -6638,10 +6691,8 @@ static int read_one_dev(struct btrfs_fs_info *fs_info,
device->in_fs_metadata = 1;
if (device->writeable && !device->is_tgtdev_for_dev_replace) {
device->fs_devices->total_rw_bytes += device->total_bytes;
- spin_lock(&fs_info->free_chunk_lock);
- fs_info->free_chunk_space += device->total_bytes -
- device->bytes_used;
- spin_unlock(&fs_info->free_chunk_lock);
+ atomic64_add(device->total_bytes - device->bytes_used,
+ &fs_info->free_chunk_space);
}
ret = 0;
return ret;
@@ -6958,7 +7009,8 @@ static int update_dev_stat_item(struct btrfs_trans_handle *trans,
key.offset = device->devid;
path = btrfs_alloc_path();
- BUG_ON(!path);
+ if (!path)
+ return -ENOMEM;
ret = btrfs_search_slot(trans, dev_root, &key, path, -1, 1);
if (ret < 0) {
btrfs_warn_in_rcu(fs_info,
@@ -7106,7 +7158,7 @@ int btrfs_get_dev_stats(struct btrfs_fs_info *fs_info,
return 0;
}
-void btrfs_scratch_superblocks(struct block_device *bdev, char *device_path)
+void btrfs_scratch_superblocks(struct block_device *bdev, const char *device_path)
{
struct buffer_head *bh;
struct btrfs_super_block *disk_super;
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index 24ba6bc3ec34..6f45fd60d15a 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -74,6 +74,8 @@ struct btrfs_device {
int missing;
int can_discard;
int is_tgtdev_for_dev_replace;
+ int last_flush_error;
+ int flush_bio_sent;
#ifdef __BTRFS_NEED_DEVICE_DATA_ORDERED
seqcount_t data_seqcount;
@@ -123,7 +125,6 @@ struct btrfs_device {
struct list_head resized_list;
/* for sending down flush barriers */
- int nobarriers;
struct bio *flush_bio;
struct completion flush_wait;
@@ -280,6 +281,11 @@ struct btrfs_io_bio {
u8 csum_inline[BTRFS_BIO_INLINE_CSUM_SIZE];
u8 *csum_allocated;
btrfs_io_bio_end_io_t *end_io;
+ struct bvec_iter iter;
+ /*
+ * This member must come last, bio_alloc_bioset will allocate enough
+ * bytes for entire btrfs_io_bio but relies on bio being last.
+ */
struct bio bio;
};
@@ -298,7 +304,7 @@ struct btrfs_bio;
typedef void (btrfs_bio_end_io_t) (struct btrfs_bio *bio, int err);
struct btrfs_bio {
- atomic_t refs;
+ refcount_t refs;
atomic_t stripes_pending;
struct btrfs_fs_info *fs_info;
u64 map_type; /* get from map_lookup->type */
@@ -400,8 +406,7 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
struct btrfs_bio **bbio_ret, int mirror_num);
int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
u64 logical, u64 *length,
- struct btrfs_bio **bbio_ret, int mirror_num,
- int need_raid_map);
+ struct btrfs_bio **bbio_ret);
int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
u64 chunk_start, u64 physical, u64 devid,
u64 **logical, int *naddrs, int *stripe_len);
@@ -422,16 +427,16 @@ void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
struct btrfs_device *device, struct btrfs_device *this_dev);
int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
- char *device_path,
+ const char *device_path,
struct btrfs_device **device);
int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid,
- char *devpath,
+ const char *devpath,
struct btrfs_device **device);
struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
const u64 *devid,
const u8 *uuid);
int btrfs_rm_device(struct btrfs_fs_info *fs_info,
- char *device_path, u64 devid);
+ const char *device_path, u64 devid);
void btrfs_cleanup_fs_uuids(void);
int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len);
int btrfs_grow_device(struct btrfs_trans_handle *trans,
@@ -439,9 +444,9 @@ int btrfs_grow_device(struct btrfs_trans_handle *trans,
struct btrfs_device *btrfs_find_device(struct btrfs_fs_info *fs_info, u64 devid,
u8 *uuid, u8 *fsid);
int btrfs_shrink_device(struct btrfs_device *device, u64 new_size);
-int btrfs_init_new_device(struct btrfs_fs_info *fs_info, char *path);
+int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *path);
int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
- char *device_path,
+ const char *device_path,
struct btrfs_device *srcdev,
struct btrfs_device **device_out);
int btrfs_balance(struct btrfs_balance_control *bctl,
@@ -474,8 +479,8 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
struct btrfs_device *tgtdev);
void btrfs_init_dev_replace_tgtdev_for_resume(struct btrfs_fs_info *fs_info,
struct btrfs_device *tgtdev);
-void btrfs_scratch_superblocks(struct block_device *bdev, char *device_path);
-int btrfs_is_parity_mirror(struct btrfs_mapping_tree *map_tree,
+void btrfs_scratch_superblocks(struct block_device *bdev, const char *device_path);
+int btrfs_is_parity_mirror(struct btrfs_fs_info *fs_info,
u64 logical, u64 len, int mirror_num);
unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
struct btrfs_mapping_tree *map_tree,
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index 9621c7f2503e..2c7e53f9ff1b 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -47,8 +47,8 @@ ssize_t __btrfs_getxattr(struct inode *inode, const char *name,
return -ENOMEM;
/* lookup the xattr by name */
- di = btrfs_lookup_xattr(NULL, root, path, btrfs_ino(inode), name,
- strlen(name), 0);
+ di = btrfs_lookup_xattr(NULL, root, path, btrfs_ino(BTRFS_I(inode)),
+ name, strlen(name), 0);
if (!di) {
ret = -ENODATA;
goto out;
@@ -108,8 +108,8 @@ static int do_setxattr(struct btrfs_trans_handle *trans,
path->skip_release_on_error = 1;
if (!value) {
- di = btrfs_lookup_xattr(trans, root, path, btrfs_ino(inode),
- name, name_len, -1);
+ di = btrfs_lookup_xattr(trans, root, path,
+ btrfs_ino(BTRFS_I(inode)), name, name_len, -1);
if (!di && (flags & XATTR_REPLACE))
ret = -ENODATA;
else if (IS_ERR(di))
@@ -128,8 +128,8 @@ static int do_setxattr(struct btrfs_trans_handle *trans,
*/
if (flags & XATTR_REPLACE) {
ASSERT(inode_is_locked(inode));
- di = btrfs_lookup_xattr(NULL, root, path, btrfs_ino(inode),
- name, name_len, 0);
+ di = btrfs_lookup_xattr(NULL, root, path,
+ btrfs_ino(BTRFS_I(inode)), name, name_len, 0);
if (!di)
ret = -ENODATA;
else if (IS_ERR(di))
@@ -140,7 +140,7 @@ static int do_setxattr(struct btrfs_trans_handle *trans,
di = NULL;
}
- ret = btrfs_insert_xattr_item(trans, root, path, btrfs_ino(inode),
+ ret = btrfs_insert_xattr_item(trans, root, path, btrfs_ino(BTRFS_I(inode)),
name, name_len, value, size);
if (ret == -EOVERFLOW) {
/*
@@ -278,7 +278,7 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
* NOTE: we set key.offset = 0; because we want to start with the
* first xattr that we find and walk forward
*/
- key.objectid = btrfs_ino(inode);
+ key.objectid = btrfs_ino(BTRFS_I(inode));
key.type = BTRFS_XATTR_ITEM_KEY;
key.offset = 0;
@@ -336,7 +336,7 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
u32 this_len = sizeof(*di) + name_len + data_len;
unsigned long name_ptr = (unsigned long)(di + 1);
- if (verify_dir_item(fs_info, leaf, di)) {
+ if (verify_dir_item(fs_info, leaf, slot, di)) {
ret = -EIO;
goto err;
}
diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c
index da497f184ff4..c248f9286366 100644
--- a/fs/btrfs/zlib.c
+++ b/fs/btrfs/zlib.c
@@ -24,12 +24,13 @@
#include <linux/slab.h>
#include <linux/zlib.h>
#include <linux/zutil.h>
-#include <linux/vmalloc.h>
+#include <linux/mm.h>
#include <linux/init.h>
#include <linux/err.h>
#include <linux/sched.h>
#include <linux/pagemap.h>
#include <linux/bio.h>
+#include <linux/refcount.h>
#include "compression.h"
struct workspace {
@@ -42,7 +43,7 @@ static void zlib_free_workspace(struct list_head *ws)
{
struct workspace *workspace = list_entry(ws, struct workspace, list);
- vfree(workspace->strm.workspace);
+ kvfree(workspace->strm.workspace);
kfree(workspace->buf);
kfree(workspace);
}
@@ -52,14 +53,14 @@ static struct list_head *zlib_alloc_workspace(void)
struct workspace *workspace;
int workspacesize;
- workspace = kzalloc(sizeof(*workspace), GFP_NOFS);
+ workspace = kzalloc(sizeof(*workspace), GFP_KERNEL);
if (!workspace)
return ERR_PTR(-ENOMEM);
workspacesize = max(zlib_deflate_workspacesize(MAX_WBITS, MAX_MEM_LEVEL),
zlib_inflate_workspacesize());
- workspace->strm.workspace = vmalloc(workspacesize);
- workspace->buf = kmalloc(PAGE_SIZE, GFP_NOFS);
+ workspace->strm.workspace = kvmalloc(workspacesize, GFP_KERNEL);
+ workspace->buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
if (!workspace->strm.workspace || !workspace->buf)
goto fail;
@@ -73,13 +74,11 @@ fail:
static int zlib_compress_pages(struct list_head *ws,
struct address_space *mapping,
- u64 start, unsigned long len,
+ u64 start,
struct page **pages,
- unsigned long nr_dest_pages,
unsigned long *out_pages,
unsigned long *total_in,
- unsigned long *total_out,
- unsigned long max_out)
+ unsigned long *total_out)
{
struct workspace *workspace = list_entry(ws, struct workspace, list);
int ret;
@@ -89,6 +88,9 @@ static int zlib_compress_pages(struct list_head *ws,
struct page *in_page = NULL;
struct page *out_page = NULL;
unsigned long bytes_left;
+ unsigned long len = *total_out;
+ unsigned long nr_dest_pages = *out_pages;
+ const unsigned long max_out = nr_dest_pages * PAGE_SIZE;
*out_pages = 0;
*total_out = 0;
@@ -210,10 +212,7 @@ out:
return ret;
}
-static int zlib_decompress_bio(struct list_head *ws, struct page **pages_in,
- u64 disk_start,
- struct bio *orig_bio,
- size_t srclen)
+static int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
{
struct workspace *workspace = list_entry(ws, struct workspace, list);
int ret = 0, ret2;
@@ -221,8 +220,12 @@ static int zlib_decompress_bio(struct list_head *ws, struct page **pages_in,
char *data_in;
size_t total_out = 0;
unsigned long page_in_index = 0;
+ size_t srclen = cb->compressed_len;
unsigned long total_pages_in = DIV_ROUND_UP(srclen, PAGE_SIZE);
unsigned long buf_start;
+ struct page **pages_in = cb->compressed_pages;
+ u64 disk_start = cb->start;
+ struct bio *orig_bio = cb->orig_bio;
data_in = kmap(pages_in[page_in_index]);
workspace->strm.next_in = data_in;